MYH9-Related Disorders

Macrothrombocytopenia with granulocyte inclusions, with or without nephropathy, sensorineural hearing loss and cataracts (May-Hegglin anomaly and Sebastian, Fechtner and Epstein syndromes; DFNA17)

Literature and regulatory information updated: 1 November 2026.

0. Initial Reference Table

1. Introduction

1.1. Definition

MYH9-related disease (MYH9-RD) is an autosomal dominant disorder caused by heterozygous variants in the MYH9 gene, which encodes the heavy chain of non-muscle myosin IIA (NMMIIA), also called myosin-9 [2]. Three haematological abnormalities are present in all patients from birth: platelet macrocytosis, defined in GeneReviews as more than 40% of platelets with a diameter greater than 3.9 µm; thrombocytopenia, with a count below 150 × 10⁹/L; and aggregates of the MYH9 protein in the cytoplasm of neutrophils [2]. Over a lifetime, most patients develop one or more extra-haematological manifestations, namely sensorineural hearing loss, proteinuric glomerular nephropathy, presenile cataracts and elevated liver enzymes [2, 34]. GeneReviews includes within MYH9-RD all carriers of a pathogenic variant with congenital macrothrombocytopenia and myosin-9 aggregates in neutrophils, with or without extra-haematological manifestations [2]. There are no consensus clinical diagnostic criteria, and the diagnosis is established by the combination of suggestive findings with a heterozygous pathogenic variant in MYH9 [2].

1.2. Historical descriptions

In 1909, Richard May described inclusions in the peripheral blood granulocytes of an asymptomatic 24-year-old woman, and in 1945 Robert Hegglin observed, in two generations of a family, the association of thrombocytopenia, giant platelets and leucocyte inclusions, which came to be called May-Hegglin anomaly [1, 35]. In 1962, Oski and colleagues described the anomaly in a mother and her two children and noted that, of 24 published cases, only 9 had thrombocytopenia [1]. In 1972, Epstein and colleagues described two unrelated families with macrothrombocytopathy, nephritis and hearing loss, in which the kidney disease was indistinguishable from that of X-linked Alport syndrome, except for being more severe in women, and in which the platelets had a prolonged bleeding time and reduced aggregation to collagen and adrenaline [1, 36]. In 1985, Peterson and colleagues described a four-generation family in which 8 of 17 members had nephritis, high-tone hearing loss, congenital cataracts, macrothrombocytopenia and inclusions in neutrophils and eosinophils, in variable combinations; all affected adults had cataracts, the kidney disease ranged from microscopic haematuria to end-stage kidney disease (ESKD) with dialysis and transplantation, and platelet aggregation was normal [37]. The name “Fechtner syndrome” presumably derives from that family's surname [1]. In 1990, Greinacher and colleagues named Sebastian platelet syndrome an inherited macrothrombocytopenia with neutrophil inclusions similar to those of Fechtner syndrome, but without hearing loss, cataracts or nephritis [38]. In 1997–2000, Lalwani and colleagues mapped to chromosome 22 an autosomal dominant nonsyndromic hearing loss with cochleosaccular degeneration, designated DFNA17 [3, 9].

1.3. Molecular discovery and nosological unification

Between 1999 and 2000, linkage studies in Japanese, Israeli and other families placed the loci for May-Hegglin anomaly, Fechtner syndrome and Epstein syndrome in overlapping regions of 22q12.3-q13 [1]. In 2000, the May-Hegglin/Fechtner Consortium identified six MYH9 variants in seven probands with May-Hegglin anomaly, Fechtner syndrome or Sebastian syndrome, and Kelley and colleagues identified variants in the same gene in ten families with May-Hegglin anomaly, including one de novo variant [39, 40]. In the same year, the p.(Arg705His) variant was associated with DFNA17 [9]. In 2001, Heath and colleagues showed that MYH9 variants also cause Epstein syndrome and Alport syndrome with macrothrombocytopenia, and that substitutions of arginine 702 occurred only in the forms with nephritis and hearing loss [41]. In 2003, Seri and colleagues re-evaluated 19 families and demonstrated an abnormal distribution of myosin in the leucocytes of all patients, even in those without visible Döhle-like bodies, high-tone hearing loss in 83% and cataracts in 23% of patients referred as May-Hegglin anomaly or Sebastian syndrome, and renal abnormalities in these patients as well; they concluded that the four syndromes represent a single disease with a continuous clinical spectrum and proposed the designation “MYH9-related disease” [42]. In 2015, Verver and colleagues identified p.(Arg705His) in two families whose affected members had, in addition to hearing loss, thrombocytopenia, giant platelets, leucocyte inclusions and elevated liver enzymes, and proposed that DFNA17 should no longer be a separate entity [26].

1.4. Current nomenclature and classifications in use

OMIM adopts the designation MATINS (macrothrombocytopenia and granulocyte inclusions with or without nephritis or sensorineural hearing loss; #155100), lists as alternative titles May-Hegglin anomaly, Sebastian, Epstein and Fechtner syndromes, Alport syndrome with macrothrombocytopenia and macrothrombocytopenia with progressive sensorineural hearing loss, and includes the disease in the series of platelet-type bleeding disorders as BDPLT6 [1]. OMIM keeps DFNA17 in a separate entry (#603622) [3]. Orphanet uses the designation MYH9-related syndromic thrombocytopenia (ORPHA:182050) and retains the four classical syndromes as clinical subtypes [13–17]. There are therefore two positions on DFNA17: GeneReviews and Verver include it within the MYH9-RD spectrum [2, 26], whereas OMIM keeps it as an entity and a 2024 review excludes it from MYH9-RD on the grounds that it is a form without haematological manifestations [3, 43]. The Japanese and Korean hearing loss cohorts of 2025–2026 show carriers of MYH9 variants with hearing loss and absent or undocumented thrombocytopenia or macrothrombocytopenia [20, 44]. Table 1.1 summarises the correspondence between the former designations and the manifestations.

Table 1.1. Historical designations of MYH9-RD and the manifestations that defined them (adapted from GeneReviews, after Seri 2003) [2]

The distinction between May-Hegglin anomaly and Sebastian syndrome rested on ultrastructural differences in the inclusions: those of May-Hegglin anomaly have parallel filaments of about 10 nm, and those of Fechtner syndrome and Sebastian syndrome are aggregates of ribosomes and segments of rough endoplasmic reticulum, without these filaments [1, 37, 38]. Epstein syndrome was defined by the absence of inclusions on conventional staining, but immunofluorescence demonstrates myosin-9 aggregates in these patients as well; Orphanet refers to it as “Alport syndrome with macrothrombocytopenia” and to Fechtner syndrome as “Alport syndrome with leucocyte inclusions and macrothrombocytopenia” [16, 17, 42, 45]. The designation “Epstein syndrome” is still used in the Japanese literature for the severe subtype associated with the high-risk variants p.(Arg702Cys), p.(Arg702His) and p.(Ser96Leu) [46].

1.5. Scope

This article covers MYH9-RD in all its presentations, DFNA17 and the allelic skin disorder known as MALTA syndrome. Bernard-Soulier syndrome and the other inherited thrombocytopenias with giant platelets, Alport syndrome, genetic steroid-resistant nephrotic syndrome, Wiskott-Aldrich syndrome, 22q11.2 deletion, Jacobsen syndrome, FLNA-related disorders and nonsyndromic genetic hearing loss appear here only as differential diagnoses. DIAPH1-related disease (DFNA1 with macrothrombocytopenia), the closest differential of MYH9-RD with hearing loss, is covered in full in Section 6.6. Somatic alterations of MYH9 in tumours, studied as evidence of a tumour-suppressor role of the gene, are outside the scope [34].

2. Molecular Genetics and Pathophysiology

2.1. The gene and the protein

MYH9 is located at 22q12.3, spans more than 106 kb and has 41 exons; the first exon is untranslated and the initiation codon is in exon 2, so exon numbering may vary between laboratories [2, 34]. The gene has the features of a housekeeping gene, with no TATA box and a high GC content, and has two enhancer regions in intron 1 [34]. The protein, with 1960 amino acids and about 230 kDa, is the heavy chain of class II non-muscle myosin, isoform A; in mammals there are three isoforms, IIA, IIB and IIC, encoded by MYH9, MYH10 and MYH14 [34]. NMMIIA is a hexamer formed by a homodimer of heavy chains, two 20 kDa regulatory light chains, which control activity, and two 17 kDa essential light chains, which stabilise the heavy chain [34]. Each heavy chain has a head domain, which includes the globular motor domain (with the SH3-like motif, the upper and lower 50 kDa subdomains and the converter) and the neck with two IQ motifs, and an α-helical coiled-coil tail domain ending in a 34-residue non-helical tailpiece (NHT) with a phosphorylation site (Ser1943) [34]. Exons 2 to 19 encode the motor domain, exon 20 the neck, exons 21 to 40 the coiled-coil tail and exon 41 the non-helical tailpiece [34, 47]. The molecule alternates between a folded inactive conformation (10S) and an extended conformation (6S), which assembles into bipolar filaments capable of moving actin [34]. NMMIIA generates the mechanical force required for cytokinesis, migration, adhesion, polarisation, maintenance of cell shape and secretion, and is the only myosin present in platelets [2, 7, 47]. The murine orthologue Myh9, on chromosome 15, has the same organisation and 97.1% protein identity, which makes the mouse a suitable model of the disease [34].

2.2. Molecular mechanism

MYH9-RD is autosomal dominant, and since myosin functions as a dimer of two heavy chains, the most likely mechanism is dominant-negative: the mutant chains are incorporated into the hexamer and disrupt its function [2, 7]. This interpretation is consistent with the rarity of loss-of-function variants outside the last exon and with the low haploinsufficiency score assigned to the gene [7]. In neutrophils, the mutant protein aggregates with the normal protein; an antibody directed at the C-terminal segment detects the aggregates even in patients with truncating variants that remove that epitope, which demonstrates dimerisation of the normal protein with the mutant [45]. The aggregates are ribonucleoprotein complexes containing NMMIIA, MYH9 messenger RNA and ribosomes [43].

The relative severity of head and tail variants has a partial structural explanation. Motor domain variants reduce ATPase activity: p.(Arg702Cys) retains 25% of maximal MgATPase activity and moves actin at half the normal velocity, whereas p.(Asn93Lys) retains only 4% and does not move filaments, with a tendency to aggregate [48]. Pecci and colleagues proposed that residue 702, in the SH1 helix, stabilises the connection between that helix and the relay loop, so that its substitution uncouples ATP hydrolysis from force production, whereas variants at the interface between the SH3 motif and the motor domain reduce ATPase activity but may maintain tension [23]. In the tail, Asp1424 occupies position c of the heptad and its substitution by histidine, which replaces a negative charge with a positive one, may destabilise the inactive 10S conformation; Arg1165 lies near the first fold of that conformation [23]. In 2026, cellular studies showed that p.(Asn93Lys), described as motor-inhibiting, has a small effect on the motor in vitro but increases filament stability and binding to the chaperone UNC45a, forming amorphous aggregates that sequester the normal protein and NMMIIB, whereas the aggregates of p.(Glu1841Lys) do not include the normal protein; the authors propose this difference as the molecular basis of the relative severity [49]. A variant described in 2026, p.(Arg424Gly), increases ATP turnover and motor activity and was presented as the first gain-of-function variant of MYH9 [50].

2.3. Pathophysiology by tissue

Platelets. Macrothrombocytopenia results from a defect in the final stages of platelet biogenesis [2]. NMMIIA is not required for megakaryocyte differentiation or maturation, and megakaryocytes are present in normal or increased numbers, but it is essential for the extension of proplatelets, the cytoplasmic extensions that cross the wall of the marrow sinusoids and release platelets from their tips [2, 34]. Megakaryocytes from patients and murine models form few proplatelets, with reduced branching and large-diameter tips, which decreases the number of platelets released and increases their size, owing to the lack of the constrictive actomyosin force that separates and delimits platelets [34]. NMMIIA also mediates the suppression of proplatelet formation when the megakaryocyte adheres to type I collagen through integrin α2β1, a mechanism that prevents release before the megakaryocyte reaches the sinusoids; this control is altered in MYH9-RD and favours ectopic release in the marrow [34]. The variants also reduce megakaryocyte migration towards the vessels [2, 51]. Bleeding does not depend on the count alone: in murine models with the variants equivalent to NM_002473.6:c.2104C>T p.(Arg702Cys), NM_002473.6:c.4270G>A p.(Asp1424Asn) and NM_002473.6:c.5521G>A p.(Glu1841Lys), clot retraction is reduced and thrombi are unstable even after adjustment for the count, which points to a deficit in the force generated by platelets [43, 52].

Kidney. Renal injury results mainly from podocyte dysfunction [2]. Knock-in models develop focal segmental glomerulosclerosis (FSGS) with podocyte foot process effacement and loss of slit diaphragms, and podocyte-restricted ablation of NMMIIA reproduces the phenotype; mutant podocytes have a reorganised actomyosin cytoskeleton and increased motility [34]. Myosin IIA is also expressed in mesangial and tubular cells, and angiotensin II decreases MYH9 expression in podocytes through NOX4-mediated oxidative stress and TRPC6 activation, which supports the hypothesis that activation of the renin-angiotensin system acts as a trigger for progression [19].

Inner ear. NMMIIA is located in the hair cells of the organ of Corti, mainly in the stereocilia, in the spiral ligament and in the spiral limbus; by analogy with the hearing losses caused by other myosins (VI, VIIa and XVA), it is assumed that the variants compromise the structural integrity of the stereocilia [34]. In the original DFNA17 family, temporal bone histopathology showed cochleosaccular degeneration (Scheibe dysplasia), which affects the membranous cochlea and the saccule and spares the bony labyrinth, the utricle and the semicircular canals [3].

Other tissues. The pathogenesis of the cataracts and of the elevated liver enzymes is unknown [34]. The erythrocytes of three patients had less actin, α-spectrin and tropomodulin and a more porous membrane skeleton, more deformable under hyperosmotic pressure [53]. Defects in neutrophil motility have been described, with no established clinical consequence [19].

2.4. Variant spectrum

By 2018 more than 80 different variants had been described [34]. Most are single amino acid substitutions in the head domain or in the coiled-coil tail, or small in-frame deletions or insertions, concentrated in exons 2, 17, 25, 26, 27, 31 and 39 [2]. Nonsense and frameshift variants affect exclusively the last coding exon (exon 41) and alter a variable portion of the non-helical tailpiece; together with intron 40 variants, they account for about 20% of families [2, 34]. About 70% of families have variants in only six residues: Ser96 and Arg702 in the head, Arg1165, Asp1424 and Glu1841 in the coiled-coil tail and Arg1933 in the non-helical tailpiece [2, 34]. In the Italian cohort of 255 patients from 121 families, 28% of families had head variants (20% at Arg702 and 9 families at the SH3/motor domain interface, with residues Trp33, Val34, Asn93, Ala95 and Ser96) and 72% tail variants (45% coiled-coil substitutions, 23% truncating exon 41 variants, with p.(Arg1933Ter) in 16% of families) [23]. In the French cohort of 109 patients, 35.8% had variants in the motor domain and 63.3% in the tail; 21 of the 43 variants were novel and the first splicing variant, c.5765+2T>A, was described [47]. By 2014, 83 different alterations had been published [47].

Large deletions are rare. Deletions and duplications of 21 or 42 bp in exons 21, 24 and 25, frequent in the repetitive sequence region of exon 25, are detected by sequencing; a 1220-nucleotide deletion with breakpoints in the neighbouring introns removes one exon in frame (exon 25 in the numbering of the original article, exon 26 in that of GeneReviews) and was only detected by reverse transcription followed by polymerase chain reaction (RT-PCR) of neutrophil RNA and by long-range PCR [2, 34, 54]. In 2026, minigene assays confirmed a splicing effect for c.3838-2A>G (in-frame skipping of exon 29) and for c.5765+2T>A and c.5765+2T>G (inclusion of a 50 bp cryptic exon with stop codons in the last exon, which escape nonsense-mediated decay); seven other intronic variants or single-base deletions annotated as possibly pathogenic did not alter splicing and were reclassified as likely benign or of uncertain significance [55].

Table 2.1 presents the reference variants, described according to Human Genome Variation Society (HGVS) nomenclature and normalised on the MANE transcript NM_002473.6 with VariantValidator; the c. and p. numbering is identical to that of version NM_002473.4 used in the literature [2, 8].

Table 2.1. Reference MYH9 variants (NM_002473.6; NP_002464.1; GRCh38) [2, 8, 9, 23, 50, 55]

2.5. Genotype–phenotype correlations

The position of the variant determines the platelet count and the risk of late manifestations. In the 2008 Italian registry, with 108 patients from 50 families, all carriers of motor domain variants had severe thrombocytopenia and developed nephritis and hearing loss before the age of 40, whereas carriers of tail variants had higher counts and lower risk [56]. With 255 patients and five more years of follow-up, Pecci and colleagues refined the model in 2014 and defined the course associated with seven genotypes, which account for about 85% of cases [23]. Patients with head variants have mean counts of about 30 × 10⁹/L, compared with about 80 × 10⁹/L for tail variants, larger platelets and more bleeding [34]. In the French cohort, motor domain variants were associated with more patients with bleeding (65% vs 33.9%), a lower median count (32 vs 58 × 10⁹/L), a higher percentage of giant platelets (26% vs 14.5%), more extra-haematological manifestations (57.9% vs 30.6%) and earlier nephropathy (median 18.5 vs 40 years) [47].

Table 2.2. Prognostic groups by genotype in the Italian registry (255 patients) [2, 23, 57]

The same residue may have different consequences depending on the substituting amino acid: none of the 13 patients with p.(Asp1424Asn) developed nephropathy, unlike carriers of p.(Asp1424His) [23]. After the onset of proteinuria, progression to ESKD was much faster in patients with Arg702 variants (age at ESKD 24 years, standard deviation 9) than in patients with coiled-coil substitutions (42 years, standard deviation 23); fewer than half of the latter progressed to ESKD, and complete sequencing of nine patients with ESKD and coiled-coil variants did not identify an explanatory second MYH9 variant, which suggests the involvement of other genetic or environmental factors [23]. GeneReviews identifies no genotype–phenotype correlation for elevated liver enzymes [2]. The French series recorded no extra-haematological manifestations in any carrier of p.(Arg1165Cys) and showed hearing loss in two patients with exon 41 or intron 40 variants, which illustrates the limits of the model in small cohorts [47].

The morphology of the inclusions on immunofluorescence also relates to the genotype: type I inclusions, one or two per cell, large and intense, are associated with variants of the distal tail and the non-helical tailpiece (exons 39 to 41), and types II and III, numerous and small, with variants of the head and of the tail up to exon 31 [43, 47].

2.6. Penetrance, expressivity and mosaicism

Penetrance is complete for platelet macrocytosis with giant platelets and for myosin-9 aggregates in neutrophils; thrombocytopenia is congenital in almost all patients, with very few above 150 × 10⁹/L [2]. The expressivity of hearing loss, nephropathy, cataracts and elevated liver enzymes varies in age at onset and severity, and a sibling who inherits the variant may have a phenotype different from that of the proband [2]. Somatic mosaicism produces attenuated phenotypes: in the first case described, the father of a boy with typical May-Hegglin anomaly had a normal count, mixed normal and giant platelets, inclusions in 14% of neutrophils and the variant in only 6% of leucocyte DNA, with a similar proportion in the oral mucosa and hair bulbs, indicating a mutation prior to gastrulation [58]. In another family, the parent had 24% of neutrophils with aggregates and a normal count, and in yet another the father of a proband with a severe syndromic phenotype had only mild macrothrombocytopenia due to somatic mosaicism [2, 59, 60]. Germline mosaicism has also been described, with two affected siblings born to unaffected parents [2, 61].

2.7. Allelic disorders and variants of difficult interpretation

DFNA17, caused by p.(Arg705His), was described in two families as autosomal dominant nonsyndromic hearing loss and was later associated with the haematological manifestations of MYH9-RD [9, 26]. In 2026, in the Japanese hearing loss cohort, 7 of 18 assessable carriers of MYH9 variants had thrombocytopenia and 8 of 13 macrothrombocytopenia, and in a Korean series none of the MYH9 carriers had syndromic manifestations, one of them with a slightly elevated mean platelet volume [20, 44]. The 2018 review accepts that some variants described in nonsyndromic hearing loss may correspond to MYH9-RD with predominant hearing loss, but refers to families in which haematological, renal and lens abnormalities were excluded [34].

MALTA syndrome (MYH9-associated elastin aggregation) is characterised by skin lesions, such as non-malignant sweat duct proliferation, atrophoderma vermiculata, multiple syringomas and milia, with irregular distribution of elastic fibres in the dermis; in-frame deletions and missense and frameshift variants have been described in five unrelated families, with dominant inheritance [2, 62]. In 2026, a man of about 20 years was described with bilateral facial lesions and benign deep syringoid ductal proliferation, compound heterozygous for p.(Gly455Ser) and p.(Ile1626Val) in cis, inherited from one parent, and p.(Arg1497Gln), inherited from the other; his brother, with the same variants, was asymptomatic [4]. The authors interpret this family as an autosomal recessive form with incomplete penetrance, which places MALTA among the conditions with two reported inheritance patterns [4]. The diagnosis distinguishes it from microcystic adnexal carcinoma and avoids aggressive surgery [4].

Atypical MYH9-RD and variants of difficult interpretation have been described. A family with p.(Arg424Gly) had proteinuric nephropathy, elevated liver enzymes and intermittent thrombocytopenia, but no Döhle-like bodies or aggregates [50]. p.(Ile1816Val), previously associated with Epstein syndrome, was found in a 16-year-old boy with membranous nephropathy and focal sclerosis, a normal platelet count, no inclusions and normal immunofluorescence in neutrophils and podocytes; the variant is more frequent in East Asia and predicted to be benign, and the case shows the risk of attributing disease to rare variants without a supporting phenotype [63]. p.(Leu46Phe), recorded as a polymorphism with a frequency of 2 to 4 per thousand, was the only variant in a French patient with macrothrombocytopenia and early hearing loss, and did not worsen the phenotype when combined with other variants [47].

2.8. Modifiers and MYH9 associations outside MYH9-RD

In 2008, common variants in the MYH9 region were associated with non-diabetic kidney disease and FSGS in African Americans; in 2010 this association was found to be due to the G1 and G2 haplotypes of the neighbouring gene APOL1, in linkage disequilibrium [19, 34, 48]. Some MYH9 polymorphisms have been associated with diabetic kidney disease in European and Asian populations, for example in the Canary Islands and in Han Chinese, without it being known whether they tag variants of the APOL1–APOL6 region [19, 34]. In murine models, loss of Myh9 favours squamous cell carcinomas and lobular breast carcinoma, and 5% of head and neck squamous cell carcinomas have somatic MYH9 variants; no increased cancer risk has been demonstrated in patients with MYH9-RD [19, 34].

3. Clinical Spectrum and Manifestations by System

3.1. Overview

MYH9-RD presents at birth with haematological abnormalities only, and the clinical course depends on the non-congenital manifestations that appear over decades: in some patients, mild asymptomatic thrombocytopenia is the only manifestation throughout life, while others develop syndromic pictures with spontaneous bleeding, ESKD, hearing loss and cataracts [23]. Table 3.1 summarises the frequency of the manifestations in the Italian Registry, the main natural history cohort.

Table 3.1. Frequency of the manifestations of MYH9-RD (Italian Registry; Pecci 2012 and 2014, summarised in GeneReviews) [2, 23, 64]

The 2014 Italian cohort included all consecutive patients in the Italian Registry up to March 2012: 255 patients from 121 families, 134 male and 121 female, with a mean age of 33 years (standard deviation 21) [23]. The French cohort, recruited over 8 years from 295 patients with inherited macrothrombocytopenia, brought together 109 patients (76 probands and 33 relatives; 74 adults and 35 children, seven under 1 year of age), 91.7% of European origin [47]. These series are observational and do not stratify the renal course by the use of nephroprotective drugs, so the progression rates describe the course observed in reference centres and not the untreated natural history.

The 2021 French national protocol (PNDS) of the HAS, written by the CRPP experts on the basis of French experience and the literature, gives rounded frequencies that differ from those of the Italian Registry: extra-haematological manifestations in about 40% of patients, absent at birth and appearing in childhood or mainly in adulthood, hearing loss in about 50%, glomerulopathy in about 30%, cataract in about 5% and a doubling of aminotransferases in about 50% [33]. These values are overall proportions without an explicit denominator or age of observation and are closer to the French series than to the GeneReviews lifetime risk estimates (hearing loss in 80–85%), which mainly reflects the difference between point prevalence and cumulative risk [2, 33, 47].

3.2. Haematological manifestations

Platelets. Giant platelets, larger than erythrocytes, are always present on the smear [2]. The mean platelet diameter was 4.5 µm (95% confidence interval, 95% CI, 4.2–4.8) in 125 patients, compared with 2.6 µm (95% CI 2.4–2.7) in 55 controls [2]. Platelet size depends on the position of the variant: in the same multicentre Pavia study, with centralised blinded measurement by image analysis of about 200 platelets per patient, the 25 carriers of head domain variants had a mean diameter of 5.64 µm (95% CI 4.8–6.4), 75.3% of platelets above 3.9 µm and a mean count of 19 × 10⁹/L, whereas the 100 carriers of tail variants had 4.28 µm (95% CI 4.0–4.6), 54.0% large platelets and a mean count of 39 × 10⁹/L; only patients with head variants had a mean diameter more than twice that of controls [65]. In the French series, the median proportion of giant platelets was 17%, of intermediate platelets 44% and of normal platelets 33%, and the percentage of giant platelets increased with the severity of thrombocytopenia [47]. Thrombocytopenia ranges from mild to severe and usually remains stable throughout each patient's life [2]. Automated counters fail to recognise the larger platelets and underestimate count and size: in the Italian registry, the median automated count was 31 × 10⁹/L (interquartile range, IQR, 16–53) in the 255 patients, and in the 167 with a phase-contrast microscopic count the median microscopic count was 70 × 10⁹/L (IQR 40–95), compared with 26 × 10⁹/L (IQR 13–46) on the counter; 6 patients (4%) had a normal microscopic count [23]. In the French series, the mean count was 53 × 10⁹/L (from less than 10 to 136), with no patient above 150 × 10⁹/L; 7.4% had 10 × 10⁹/L or less, 44.8% between 11 and 50, 37.3% between 51 and 99 and 10.3% above 100 [47]. Thrombocytopenia may be intermittent: in a French nephrology series, counts ranged between 4,000 and 350,000/µL [19]. Platelet function is usually normal [2, 34], although the original descriptions of Epstein syndrome report reduced adhesion and aggregation [1].

Neutrophils. Döhle-like bodies are faint, slightly basophilic cytoplasmic inclusions, similar to the Döhle bodies of infection, visible on May-Grünwald-Giemsa staining in 42–84% of patients and, in those patients, in 15–100% of neutrophils [2, 34]. In the French series, they were seen in 92 of 99 stained smears, with a mean of 84.6% positive leucocytes (5–100%) [47]. Unlike toxic Döhle bodies, which are small, irregular and peripheral and accompany toxic granulation, the inclusions of May-Hegglin anomaly are larger, spindle-shaped, located anywhere in the cytoplasm and, on electron microscopy, contain parallel filaments [1]. Immunofluorescence shows the aggregates in all neutrophils of all patients [34].

Bleeding. The severity of spontaneous bleeding correlates with the platelet count [2]. On the World Health Organization (WHO) scale, applied to 183 Italian patients, 57% had no bleeding, 15% had grade 1 (skin only), 19% grade 2 (mild blood loss) and 9% grade 3 (blood loss requiring transfusion); patients with grade 3 had a median microscopic count of 38 × 10⁹/L and an automated count of 10 × 10⁹/L [23]. About 28–30% of patients have spontaneous mucosal bleeding, mainly menorrhagia, epistaxis and gingival bleeding, and life-threatening haemorrhages are rare [2, 34]. In the French series, 41 of 91 patients (45%) had bleeding, with a median count of 32 compared with 58 × 10⁹/L in those who did not bleed; below 50 × 10⁹/L there were significantly more patients with bleeding, and platelet size was not related to bleeding [47]. Bleeding was the first manifestation in 29.7% of cases, almost always in childhood and before the age of 5; exceptions recorded were a post-traumatic extradural haematoma at 5 years, bleeding after tonsillectomy at 4 years, a neonatal parietal haematoma, a meningeal haemorrhage at birth and another neonatal haemorrhage [47]. The risk is greater after haemostatic challenges, such as surgery, childbirth, trauma or antiplatelet drugs [2].

According to the French PNDS, the count is usually between 30 and 100 × 10⁹/L, spontaneous bleeding occurs mainly below 20 × 10⁹/L, marked bleeding is concentrated in patients with fewer than 50 × 10⁹/L without being systematic in them, and bleeding severity does not vary with age [33]. The most common spontaneous bleeding events are bruising (petechial purpura is rare), epistaxis, gingival bleeding and menorrhagia, and provoked bleeding follows trauma, general or gynaecological and obstetric surgery and dental extractions [33]. The count varies between affected members of a family but remains constant in the same patient, with fluctuations in acquired situations: viral infections may worsen thrombocytopenia and inflammation may increase the count [33].

3.3. Sensorineural hearing loss

Hearing loss is the most frequent non-congenital manifestation [34]. In the Italian registry, it was present in 114 of 237 patients with an audiogram (48%) and in 58% of families; 90% of patients with an abnormal audiogram had clinically relevant loss; the mean age at onset was 31 years (standard deviation 19), evenly distributed between the first and sixth decades (36% before 20 years, 33% between 20 and 40, 31% after 40), with a rate of 1.71 per 100 person-years (95% CI 1.39–2.11) [23]. The hearing loss is bilateral, begins at high and mid frequencies and, in advanced forms, affects all frequencies; forms with onset in childhood or adolescence usually progress to severe to profound hearing loss [34, 57]. In a study of 115 audiograms from 63 patients in 45 families, mild hearing loss had a flat configuration and severe hearing loss a down-sloping configuration; the variants p.(Arg702Cys), p.(Arg702His) and p.(Arg1165Leu) were associated with the most progressive hearing loss, p.(Arg1165Cys) with mild non-progressive hearing loss, p.(Glu1841Lys) with mild progression and non-helical tailpiece variants with no substantial progression [57]. In the French series, hearing loss affected 34.7% of assessable patients, began earlier than nephropathy (median 16 vs 27 years), was isolated in 54% of cases and required hearing aids in 13 of 35 [47]. In a 2026 Japanese cohort, some patients lost about 50 dB in five years, sometimes asymmetrically, and all progressed, at different rates, to bilateral profound hearing loss, faster with head variants [20]. In the implanted patients of an international series, the mean age at onset of hearing loss was 16.6 years, language development was normal and hearing loss lasted on average 24.2 years before implantation [66].

The French PNDS describes a hearing loss initially limited to the high frequencies and considers that onset in childhood or adolescence generally heralds severe hearing loss in the first decades of life [33].

In DFNA17, the original American family had onset at 10 years in the high frequencies and moderate to severe hearing loss in the third decade; in an Australian family, onset ranged from 6 years to the middle of the second decade, with initial high-frequency loss that flattened and became severe to profound in the second or third decade; in a Brazilian family, ten members had progressive bilateral hearing loss at all frequencies, with onset between the first and fifth decades [3, 67].

3.4. Nephropathy

Nephropathy presents with proteinuria and microscopic haematuria, but the haematuria may result from the thrombocytopenia, so proteinuria is the most reliable indicator of glomerular involvement [2]. In the Italian registry, 61 of 247 patients (25%) developed proteinuric nephropathy, defined as 24-hour proteinuria greater than 0.5 g on two tests three months apart; the mean age at onset was 27 years (standard deviation 14), 72% were diagnosed before 35 years, and the rate was 0.77 per 100 person-years (95% CI 0.56–1.08) [23]. After a median follow-up of 36 months, 40 of the 61 (66%) had chronic renal failure, at a mean age of 31 years, and 26 (43%) had ESKD with dialysis, at a mean age of 30 years; the rate of progression to ESKD was 6.79 per 100 person-years (95% CI 4.32–9.53) [23]. GeneReviews rounds the proportion with chronic kidney disease to 64% [2]. In some patients, proteinuria appears later or progresses slowly [2, 34].

The French PNDS considers the renal prognosis difficult to predict and lists heavy proteinuria, arterial hypertension and certain variants as unfavourable factors; when focal segmental hyalinosis (FSGS) occurs with thrombocytopenia, MYH9-RD should be considered in order to avoid inappropriate treatment [33].

Carriers of Arg702 variants develop renal injury before 45 years and progress to ESKD about five years after the onset of proteinuria [23]. In a Japanese series of nine patients with Arg702 variants, proteinuria and/or haematuria began in early childhood and renal function deteriorated rapidly in adolescence [68]. In a 2025 Japanese national survey, among 177 patients with MYH9-RD, 40 had the severe subtype known as Epstein syndrome, with the variants p.(Arg702Cys), p.(Arg702His) or p.(Ser96Leu); more than half had renal failure, hearing loss and proteinuria by 19 years and 80% by 32 years, and 95% of the variants were de novo [46]. In a French series of patients referred to nephrology (indirect attribution to Tabibzadeh and colleagues), the median age at referral was 30 years, about 20% were already in ESKD and a further 50% progressed; proteinuria was present in 85% at presentation, could reach nephrotic levels and cause nephrotic syndrome, and haematuria was present in only about 50%; in one family, age at ESKD ranged between 24 and 78 years [19].

Renal biopsy is rarely performed because of the thrombocytopenia. In a Japanese national survey with nine biopsies, all showed mesangial matrix expansion and segmental podocyte foot process effacement; 78% had aggregated dense material in podocytes and endothelial swelling, 56% electron-dense deposits and alterations of the glomerular basement membrane (thinning, irregular thickening, splitting of the lamina densa, without the basket-weave appearance of Alport syndrome), 33% immunoglobulin or complement deposits and 22% FSGS [69]. In Sekine's series, histology was consistent with FSGS and podocyte immunostaining for NMMIIA was decreased [68].

3.5. Cataracts

Cataracts affected 43 of 235 Italian patients (18%), were bilateral in 72% and began at a mean age of 37 years (standard deviation 30), with four congenital cases and a rate of 0.57 per 100 person-years (95% CI 0.41–0.82) [23]. There was no difference between head and tail variants; the incidence was lower with non-helical tailpiece variants and higher with p.(Asp1424His) [23]. In the French series, cataracts were rare (4.9%) [47].

The French PNDS describes them as bilateral cerulean cataracts (with bluish opacities), in about 5% of patients, with mean onset between the second and fourth decades and congenital forms described [33]. In the original Fechtner syndrome, all affected adults had cataracts, described as congenital [37].

3.6. Liver

Chronic or intermittent elevation of aminotransferases, sometimes with elevated gamma-glutamyltransferase (GGT), is frequent: in the Italian registry, 38 of 75 patients (50.7%) had elevated ALT and/or AST and 17 of 63 (27.0%) elevated GGT, with values of 1.9 to 2.7 times the upper limit, and the prevalence was significantly higher than in patients with immune thrombocytopenia, in patients with other inherited thrombocytopenias and in an Italian general population (odds ratios of 8.2 to 24.7) [64]. Follow-up, ultrasound, elastography and one biopsy showed no significant structural damage or progression to liver failure, and some patients normalised their values [2, 64, 70].

3.7. Other reported associations

Isolated cases have been described of acute myeloid leukaemia and myelodysplastic syndrome, more difficult to diagnose because of the abnormal baseline marrow, extramedullary haematopoiesis mistaken for a tumour, congenital haemangioma and nonsyndromic orofacial cleft; it is not known whether these associations exceed chance [19]. Also published are a family with spastic paraplegia, a newborn with Epstein syndrome and bladder exstrophy and, in 2025, the second case of littoral cell angioma of the spleen in a patient with Epstein syndrome, operated on for bleeding after trauma [1, 71]. A 2026 Italian case describes haemorrhagic and thrombotic manifestations in the same patient, a reminder that inherited thrombocytopenia does not protect against thrombosis [72, 73].

3.8. Course by age

In the newborn and child, the disease manifests as thrombocytopenia with giant platelets, often discovered by chance: in the French series, 70.3% of patients were identified in childhood (half before 5 years), 45.9% through a routine count before surgery or during an infection, 24.4% through family history and 29.7% through bleeding [47]. Hearing loss may begin in childhood in the high-risk genotypes, and in Epstein syndrome proteinuria appears in early childhood [46, 68]. Adolescence and young adulthood concentrate the onset of nephropathy (72% before 35 years) and much of the hearing loss [23]. Cataracts and elevated liver enzymes appear mainly in adulthood, and forms with tail variants may remain throughout life with isolated thrombocytopenia or with mild late-onset hearing loss [23, 34].

4. Clinical Recognition and Warning Signs

4.1. When to suspect

MYH9-RD should be considered in the presence of thrombocytopenia with large or giant platelets, particularly when it is congenital or long-standing, shows dominant familial transmission or is associated with sensorineural hearing loss, proteinuria, early cataracts or unexplained elevation of liver enzymes [2, 43]. Suspicion is strengthened when the smear shows Döhle-like bodies in neutrophils [43]. The absence of a family history does not exclude the diagnosis, because about 35% of probands are simplex cases and because relatives with mild forms go unnoticed [2]. Table 4.1 brings together the settings in which suspicion should be raised.

Table 4.1. Clinical and laboratory settings that should lead to consideration of MYH9-RD [2, 19, 43, 74, 75]

The French PNDS adds to the initial examination the measurement of blood pressure and the search, in children, for indirect signs of incipient hearing loss, such as delayed language acquisition, school difficulties and behavioural changes (a restless child or, conversely, a withdrawn one), and recommends ophthalmological examination from the age of 20, since cataract appears on average between 20 and 40 years [33]. The assessment includes the impact of blood loss (pallor, tachycardia, functional systolic murmur, iron deficiency and anaemia) [33]. In the family enquiry, in addition to thrombocytopenia and bleeding, the age at onset of hearing loss, kidney disease and cataract in relatives is of interest, because early onset of these manifestations together with thrombocytopenia is suggestive [33].

4.2. Misdiagnosis as ITP

The most frequent and most harmful error is the diagnosis of ITP. In the Italian Registry, about 60% of index cases had been diagnosed with ITP, 30% received inappropriate treatments and 12% underwent unnecessary splenectomy [2, 34]. In the French series, 29 of 70 patients (41.4%) had that diagnosis, 25 received corticosteroids and 19 intravenous immunoglobulin without a significant increase in the count, and 10 underwent splenectomy, the youngest at 22 months [47]. In a Chinese multicentre series of 40 patients initially diagnosed with ITP, all received ITP treatments; the median diagnostic delay was 9 years in adults and 0.2 years in children, the median duration of those treatments 1.5 and 0.1 years, and extra-haematological manifestations appeared only in adults (10 of 21) [76]. In an Australian cohort studied with a next-generation sequencing (NGS) panel, only 1 of 17 patients with MYH9-RD had been referred with that suspicion [75]. The factors leading to error are the lack of a family history, mild bleeding, the failure of counters in counting and in MPV, and the difficulty of seeing the inclusions on Romanowsky stains [1, 75].

4.3. Active screening strategies

A 2024–2025 Chinese prospective study applied a platelet size trigger to patients with persistent unexplained thrombocytopenia (more than three months, below 150 × 10⁹/L): MPV of at least 14.55 fL, platelet large cell ratio (P-LCR) of at least 58%, or grade 2+ or 3+ giant platelets on the smear (at least 10% of platelets larger than an erythrocyte); positive patients underwent a 308-gene NGS panel and confirmation by immunofluorescence [74]. Of the 42 patients screened, 7 (16.7%) had MYH9-RD, 6 of them with a previous diagnosis of ITP and 5 with ITP treatments, and all were diagnosed in less than a month [74]. The thresholds depend on the analyser (Mindray BC-6800) and are not transferable without local validation [74].

4.4. Emergency signs

Situations requiring immediate assessment are active bleeding not controlled by local measures, bleeding at a critical site (intracranial, ocular, airway), organ-threatening bleeding, the need for urgent surgery with severe thrombocytopenia and labour in a pregnant woman with severe thrombocytopenia [2, 77]. In the French series, severe bleeding was concentrated in childhood and the neonatal period, including meningeal haemorrhage at birth and post-traumatic extradural haematoma [47]. Rapid deterioration of renal function, such as bleeding precipitating acute kidney injury as described in the French nephrology series, and sudden or rapidly progressive hearing loss also warrant prompt referral [19, 20]. There is no newborn screening for MYH9-RD.

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Criteria

There are no consensus clinical diagnostic criteria [2]. The diagnosis is established in a proband with suggestive findings and a heterozygous pathogenic variant in MYH9 [2]. The 2018 review by the authors of the Italian Registry considers that, in the presence of typical NMMIIA inclusions on immunofluorescence, identification of the variant is not strictly necessary for the diagnosis, but is important for prognosis and for family studies [34]. The 2025 German S2k guideline for the diagnosis of inherited platelet disorders, published in a journal by the Society of Thrombosis and Haemostasis Research (GTH), recommends with strong consensus the use of molecular genetics in diagnosis (95%) and, with consensus, to identify heterozygous carriers (78%) [25].

The French PNDS bases the laboratory diagnosis on a triad present from birth (thrombocytopenia below 150 × 10⁹/L, large and giant platelets and basophilic inclusions in neutrophils) and considers identification of the variant by MYH9 sequencing indispensable to formalise the diagnosis, for family studies and for genetic counselling, a more demanding position than that of the 2018 review [33]. It recommends that the clinical and laboratory diagnosis be established in a specialised haemostasis clinic [33].

5.2. Clinical assessment and full blood count

Assessment begins with the bleeding history using a standardised questionnaire: the ISTH-BAT (bleeding assessment tool of the International Society on Thrombosis and Haemostasis) in adults and the pedISTH-BAT in children, recommended by the German guideline (strong consensus, 96%, for the use of a questionnaire; consensus, 86%, for the choice of instruments) [2, 25]. The ISTH-BAT was validated in 1098 people from 17 countries, 482 of whom had inherited platelet disorders (286 with defects in number), and discriminates well between patients and controls [78]. The German guideline also recommends, with strong consensus (100%), a full blood count with platelet size distribution, MPV and immature platelet fraction (IPF), interpreted according to the analyser method, and microscopic examination of a stained smear [25]. In the case described in 2025, the IPF was 64% and 70% on two assessments; the literature cited by the authors reports a mean IPF of 48% in MYH9-RD compared with 13% in ITP (indirect attribution) [79]. The German guideline algorithm includes the exclusion of plasma coagulopathies before the platelet work-up [25].

The French PNDS does not recommend platelet function studies (closure time on the PFA, Platelet Function Analyser, and aggregometry) for diagnosis or follow-up, because they are difficult to interpret in the presence of thrombocytopenia and platelet function is normal, and considers bone marrow aspiration unnecessary for the diagnosis [33].

5.3. Platelet counting and sizing

Impedance and optical scatter counters do not distinguish giant platelets from erythrocytes of similar size and underestimate the count, MPV and diameter [2, 43]. The most reliable method is chamber counting with phase-contrast microscopy [2]. In a 2026 case, a pregnant woman had 5 × 10³/µL by impedance and 19 × 10³/µL by fluorescence counting on the same analyser, with the smear excluding clumps and showing giant platelets; the fluorescence count was considered the most plausible estimate [80]. On the smear, the count can be estimated by multiplying the mean number of platelets per oil-immersion field by 10 × 10³/µL [79]. In the Portuguese surgical series, counts in macrothrombocytopenia were verified by semi-quantitative estimation on the smear or by flow cytometry [81].

The French PNDS describes the laboratory pathway: impedance counters, which are single-parameter, do not separate large platelets from erythrocytes; analyser flags should lead the biologist to suitable methods, such as optical counting by high- and low-angle light diffraction, flow cytometry with a platelet marker (anti-CD61) or counting in a Malassez chamber; and the MPV may not be calculated because of platelet anisocytosis [33]. The IPF, proportional to RNA content and therefore to platelet size, usually exceeds 40% in MYH9-RD, compared with about 10% in ITP, and is also very high in Bernard-Soulier syndrome [33]. The combination of thrombocytopenia, even moderate, with a very high MPV should lead to a search for Döhle-like bodies in a haematology laboratory, with confirmation in a specialised laboratory [33].

Measurement of platelet diameter on the smear distinguishes MYH9-RD from ITP. In a study of 376 patients with the 19 inherited thrombocytopenias then known and 87 patients with ITP, the inherited forms were classified into four groups (giant platelets, large platelets, normal or slightly increased platelets and normal or slightly decreased platelets) and thresholds with good accuracy were defined [65]. According to GeneReviews, a mean diameter greater than 3.7 µm distinguishes MYH9-RD from ITP with a sensitivity of 86% and a specificity of 87%, and more than 40% of platelets with a diameter greater than 3.9 µm (about half the diameter of an erythrocyte) with a sensitivity of 85% and a specificity of 87% [2]. In the original article, these thresholds (3.74 µm and 39.8%) were calculated by ROC analysis to separate the giant platelet group, which brings together MYH9-RD with head or tail variants and biallelic Bernard-Soulier syndrome, not only from ITP but also from all the other inherited thrombocytopenias studied; combining the two criteria raised the sensitivity to 98% (95% CI 93–99) and the specificity to 99% (95% CI 97–100) [65]. Monoallelic Bernard-Soulier syndrome, almost always due to the Bolzano variant of GP1BA, fell outside this group, because its mean diameter (3.39 µm) overlapped with that of ITP (3.11 µm) [65]. Visual estimation, by a blinded observer, of the percentage of platelets larger than half an erythrocyte correlated with image analysis (Pearson's r 0.95), which allows the percentage criterion to be applied without software [65]. The limitations of the study are the Italian origin of 80% of the patients, measurement in a single laboratory and counting with automated counters, which overestimate the severity of thrombocytopenia in macrothrombocytopenias [65]. A 2024 review cites another Pavia study, in which a mean diameter greater than 3.3 µm distinguished MYH9-RD and Bernard-Soulier syndrome from ITP with a sensitivity of 89% and a specificity of 88%, and an MPV greater than 12.4 fL with 83% and 89% (indirect attribution) [43]. Noris and colleagues confirm the diameter values of that study and add that, in another population, the same threshold of 3.3 µm had only 75% sensitivity and 67% specificity [65]. These thresholds derive from different populations and methods and are not interchangeable. In ITP, giant platelets do not usually exceed 10%, whereas in MYH9-RD they exceed 40% [79].

The French PNDS recommends measuring the mean platelet diameter in at least 100 platelets on a fresh blood smear collected in EDTA, a technique reserved for experienced laboratories and biologists, with a sensitivity and specificity of 87% for separating inherited thrombocytopenias with giant platelets (MYH9-RD and biallelic Bernard-Soulier syndrome) from ITP and the other inherited thrombocytopenias [33]. In estimation by comparison with the erythrocyte, a macroplatelet measures more than half the diameter of a normal erythrocyte and a giant platelet more than one erythrocyte; in MYH9-RD, the mean proportion is 44% macroplatelets and 17% giant platelets, never below 3% giant platelets, and the proportion of giant platelets increases as the count falls [33].

5.4. Neutrophil inclusions: conventional staining and immunofluorescence

On May-Grünwald-Giemsa staining, Döhle-like bodies are visible in only 42–84% of patients, because they may be very faint or small [2]. Immunofluorescence for myosin-9 on the smear shows aggregates in all patients, from birth and throughout life, and has been validated as a diagnostic test with sensitivity and specificity close to 100% [2, 45, 82]. In the prospective validation study, 118 consecutive patients with a strongly suggestive presentation underwent immunofluorescence and sequencing: 82 had aggregates, of whom 80 (98%) had an MYH9 variant, and none of the 36 without aggregates had a variant, which corresponded to a sensitivity of 100% and a specificity of 95% [83]. In normal neutrophils, myosin-9 is evenly distributed, and a normal localisation practically excludes the disease [2, 84]. Immunofluorescence patterns are classified as type I (one or two large, intense inclusions), type II (up to 20 small dots) and type III (speckled pattern with more than 20 aggregates smaller than 0.5 µm); types II and III are often missed on conventional staining [43, 45].

The method works on air-dried smears prepared with less than 100 µL of blood and sent by post to reference laboratories, which makes it suitable for children and for regions without access to specialised laboratories: in an international four-centre programme, it allowed the cause to be characterised in 877 of 3217 patients (27%) with suspected inherited platelet disorder [82]. The German guideline recommends, with strong consensus, preparing smears only from fresh blood collected in ethylenediaminetetraacetic acid (EDTA) at room temperature and always including a control in the staining (consensus, 82%) [25].

Immunofluorescence has limitations. In the French series, 73 of 79 smears were positive; six patients with confirmed variants (p.(Glu894Lys), p.(Arg718Trp) in two, p.(Gln372Arg), p.(Arg702His) and p.(Leu46Phe)) had no detectable inclusions, and in one of them electron microscopy showed the typical structures [47]. The authors attributed the difficulty with Arg702 variants to the low polyadenylated RNA content of the inclusions and cited inadequate pre-analytical and permeabilisation conditions as causes of false negatives [47]. The gain-of-function variant p.(Arg424Gly) produced neither Döhle-like bodies nor aggregates [50].

5.5. Molecular tests

MYH9 sequencing detects a pathogenic variant in 98% of probands [2]. Large deletions and duplications are exceptional: those of 21 or 42 bp in exons 21, 24 and 25 are detected by sequencing, and the 1220-nucleotide deletion that removes one exon in frame was only identified by RT-PCR and long-range PCR, so targeted deletion/duplication analysis is justified when the clinical picture is strong and sequencing is negative [2, 54]. GeneReviews accepts single-gene sequencing or a multigene panel that includes the genes of the differential diagnosis, and reserves exome or genome sequencing for patients in whom MYH9-RD was not considered [2]. A 2024 review advises against direct Sanger sequencing, because of the size of the gene (41 exons, about 107 kb), and recommends an NGS panel, exome or genome as first line, with Sanger confirmation [43]. The German guideline notes that, in inherited thrombocytopenias, functional tests are of limited value and early genetic testing facilitates diagnosis; extracted DNA is stable and can be sent to specialised laboratories [25]. In panels applied to uncharacterised presumably inherited thrombocytopenias, MYH9-RD was the cause in 12.4% of patients in Australia and New Zealand (15 of 121) and in 25.8% of cases with a molecular diagnosis in a Romanian paediatric cohort (8 of 31) [75, 85].

The report should state the gene, the versioned transcript (MANE is NM_002473.6, with c. numbering identical to that of NM_002473.4 used in the cohorts), the HGVS nomenclature, the zygosity and the class according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) [8, 25]. The variant should be interpreted in the context of the affected domain, because the prognosis depends on it (section 2.5). A heterozygous variant of uncertain significance (VUS) neither establishes nor excludes the diagnosis, and immunofluorescence for aggregates helps assess its pathogenicity [2, 82]. This is how a VUS was confirmed in two relatives in the Australian cohort and two novel coiled-coil variants were classified as likely pathogenic in a Chinese cohort [74, 75]. Conversely, rare variants with no haematological phenotype or aggregates, such as p.(Ile1816Val), should not be accepted as causal without proof [63]. For intronic variants, the minigene assay and SpliceAI prediction were concordant [55]. ClinGen has no actionability assertion for MYH9, which reflects the state of curation and not the clinical utility of the diagnosis [7].

5.6. Diagnostic algorithm

Table 5.1 proposes a sequence based on the sources consulted.

Table 5.1. Diagnostic sequence when MYH9-RD is suspected [2, 25, 43, 82, 83]

The French PNDS algorithm, adapted from Balduini and colleagues (2013), directs syndromic forms by the history and clinical findings and non-syndromic forms by platelet size and the smear; it uses as complementary tests myosin-9 immunofluorescence in leucocytes and quantification of the glycoproteins GPIb-IX-V and GPIIb-IIIa by flow cytometry (which discriminates Bernard-Soulier syndrome), and concludes with sequencing of MYH9 or of an NGS panel of all genes for constitutional thrombocytopenias, according to the discriminating laboratory findings [33].

5.7. Assessment after diagnosis

Table 5.2. Recommended assessments after diagnosis (GeneReviews) [2]

5.8. Prenatal and preimplantation diagnosis

Once the pathogenic variant has been identified in an affected relative, prenatal diagnosis and preimplantation genetic testing are possible [2]. The thrombocytopenia of an affected newborn is similar to that of the mother [77].

In France, a request for prenatal diagnosis is discussed in a Multidisciplinary Prenatal Diagnosis Centre (CPDPN) with the opinion of the disease specialists, and preimplantation diagnosis requires the same conditions: a known causal variant and an indication validated by a CPDPN [33]. Reproductive counselling is covered in section 8.

6. Differential Diagnosis

6.1. Principles

The differential diagnosis includes acquired thrombocytopenias, mainly ITP, the other inherited thrombocytopenias with large platelets and the inherited type IV collagen nephropathies [2]. Inherited thrombocytopenias are divided by platelet size into forms with giant, large, normal or slightly increased and normal or slightly decreased platelets [65]. All congenital macrothrombocytopenias are very rare [2]. Table 6.1 compares the main entities.

Table 6.1. Differential diagnosis of MYH9-RD [2, 34, 44, 86–88]

The abbreviation RT denotes thrombocytopenia related to the gene indicated (related thrombocytopenia).

6.2. Immune thrombocytopenia

Distinguishing ITP is the most important task, because the error leads to ineffective and potentially harmful immunosuppressants and splenectomy [2]. When the family history is absent or uncertain, microscopic examination of the smear is the simplest tool: platelets are significantly larger in MYH9-RD, and the diameter thresholds (Section 5.3) distinguish the two conditions with a sensitivity and specificity of 85–87% [2]. Myosin-9 immunofluorescence and molecular testing confirm MYH9-RD [2]. Any patient labelled as ITP refractory to corticosteroids or immunoglobulin, particularly with large platelets or a family history, should be re-evaluated before splenectomy is considered [7]. ITP may coexist with an inherited thrombocytopenia, but repeated lack of response to immunological treatments is a warning sign [75, 76]. Exploratory studies of plasma proteomics and of membrane glycoproteins and antiplatelet antibodies have identified candidate markers to separate paediatric MYH9-RD from ITP, as yet without external validation [89, 90].

Since ITP is a diagnosis of exclusion, with no routine immunological test to confirm it, the French PNDS, adapting the criteria of the 2017 ITP PNDS, lists the features that should point towards MYH9-RD: a family history of thrombocytopenia or haematological disease; diagnosis before 2 years of age; moderate thrombocytopenia or thrombocytopenia discovered by chance with no known previous count; a syndromic context in the patient or the family (kidney disease, unexplained hearing loss, early cataract); immediate refractoriness to corticosteroids and polyvalent immunoglobulin; and elevated MPV (when measurable) and IPF, a high percentage of very large platelets (often above 50%) and Döhle-like bodies [33]. The same protocol requires a high degree of suspicion before any indication for splenectomy for refractory ITP and recalls that the chance association of thrombocytopenia with hearing loss, cataract or kidney disease is possible [33]. In the neonatal period, the differential includes fetomaternal alloimmune thrombocytopenia, and persistence of thrombocytopenia with a high percentage of large platelets favours the constitutional form [33].

6.3. Other inherited macrothrombocytopenias

Biallelic Bernard-Soulier syndrome has platelets that may be as large as in MYH9-RD, but no extra-haematological manifestations; the monoallelic form is not syndromic either [2]. DIAPH1-related disease combines macrothrombocytopenia and progressive sensorineural hearing loss from childhood, which makes it the closest differential of MYH9-RD with hearing loss; myosin-9 aggregates and the absence of nephropathy point the way [2]. Section 6.6 describes this disease in detail. Syndromic forms (11q23 deletions, CDC42, FLNA, SRC) are distinguished by their extra-haematological findings, and nonsyndromic forms (ACTN1, FLI1, GFI1B, ITGA2B/ITGB3, SLFN14, TUBB1) require genetics, although the smear points towards some of them, such as the anisopoikilocytosis of GFI1B-RT and the grey platelets of the NBEAL2 syndrome [2]. Immunofluorescence on smears also identifies biallelic Bernard-Soulier syndrome, Glanzmann thrombasthenia, grey platelet syndrome and the macrothrombocytopenias due to GATA1, GFI1B, TUBB1 and FLNA [82]. These entities include Bernard-Soulier syndrome and the thrombocytopenias with giant platelets, Jacobsen syndrome, FLNA-related disorders and the 22q11.2 deletion, which may cause macrothrombocytopenia through haploinsufficiency of GP1BB. Wiskott-Aldrich syndrome is distinguished by its small platelets.

Sitosterolaemia deserves attention because the haematological abnormalities, with stomatocytes, thrombocytopenia and giant platelets, are invariable and may be the only manifestation in the early stages; plant sterols and low-density lipoprotein (LDL) cholesterol are elevated and there are xanthomas and premature atherosclerosis, and the disease is treated with a diet low in plant sterols and ezetimibe [86]. Some congenital disorders of glycosylation, such as SLC35A1-CDG, and disorders of vesicular trafficking and of the cytoskeleton include macrothrombocytopenia, and Saudubray's book classifies MATINS among the disorders of cellular trafficking due to a cytoskeletal defect, alongside Hermansky-Pudlak syndrome and Wiskott-Aldrich syndrome [87, 88].

6.4. Inherited nephropathies

Alport syndrome, caused by variants in COL4A3, COL4A4 or COL4A5 with X-linked, autosomal dominant or autosomal recessive inheritance, ranges from isolated non-progressive haematuria to progressive nephropathy with proteinuria and ESKD, and is frequently associated with sensorineural hearing loss and characteristic ocular abnormalities; it has no platelet abnormalities, so a nephropathy with macrothrombocytopenia should lead to strong consideration of MYH9-RD [2]. The historical designation “Alport syndrome with macrothrombocytopenia” and OMIM's term “nephritis” suggest haematuria, but in MYH9-RD proteinuria is the dominant manifestation and haematuria is present in only about half of patients [19]. Renal biopsy in MYH9-RD does not show the basket-weave appearance of the basement membrane typical of Alport syndrome, although there may be thinning, irregular thickening and splitting [69]. The combination of ITP and Alport syndrome as misdiagnoses in the same patient has been described [19]. MYH9-RD should be included in the differential of proteinuric nephropathies at any age, particularly with FSGS or mesangial proliferation without immune deposits [19].

6.5. Sensorineural hearing loss

DFNA17 and the forms of MYH9-RD with predominant hearing loss enter the differential of progressive autosomal dominant high-frequency hearing loss. Hearing loss due to MYH14 has an overlapping audiological phenotype, with a lower proportion of severe to profound cases (14.3% vs 41.7% in a Korean series), and the full blood count helps to separate them [44]. The GeneReviews on genetic hearing loss includes MYH9 among the genes of autosomal dominant nonsyndromic hearing loss [18].

6.6. DIAPH1-related disease (DFNA1 with macrothrombocytopenia)

6.6.1. Definition and history

DIAPH1-related disease (D-RD, DIAPH1-related disorder) is an autosomal dominant disorder that combines early-onset progressive bilateral sensorineural hearing loss, macrothrombocytopenia and, in many patients, mild neutropenia [5, 6]. The DFNA1 locus was the first locus for autosomal dominant nonsyndromic hearing loss to be mapped, in a large family from Costa Rica with fully penetrant progressive sensorineural hearing loss linked to 5q31; in 1997, Lynch and colleagues associated it with a truncating variant in the human homologue of the Drosophila diaphanous gene [91]. For almost two decades DFNA1 was considered an isolated hearing loss. In 2016, Stritt and colleagues, in the British bleeding and platelet disorders consortium BRIDGE-BPD, identified by exome sequencing the variant p.(Arg1213Ter) in two unrelated families with macrothrombocytopenia and sensorineural hearing loss, and in the same year Ueyama and colleagues described the same variant, under different numbering, in two Japanese families with hearing loss [21, 92]. In 2017, Neuhaus and colleagues confirmed the association with thrombocytopenia in two German families and proposed reclassifying DFNA1 as a syndrome, with the recommendation to include a full blood count in the initial assessment of all autosomal dominant hearing loss [27]. The series by Westbury and colleagues (2018), with 16 patients from 5 families assessed in the United Kingdom, Spain and Portugal (Coimbra), added neutropenia to the phenotype and described the first use of a thrombopoietin receptor agonist [6]. OMIM retains the designation “autosomal dominant deafness 1, with or without thrombocytopenia” (124900), and Orphanet classifies it as DIAPH1-related sensorineural hearing loss–thrombocytopenia syndrome (ORPHA:494444) [5, 93].

6.6.2. Gene, mechanism and variant spectrum

DIAPH1 (5q31.3; OMIM 602121; MANE transcript NM_005219.5) encodes DIA1 (mDia1), a diaphanous-related formin that acts as an effector of the GTPase RhoA, elongates unbranched actin filaments and stabilises microtubules [10, 21, 92]. At rest, the protein is autoinhibited by the binding between the N-terminal diaphanous inhibitory domain (DID) and the C-terminal diaphanous autoregulatory domain (DAD); binding of activated RhoA disrupts this interaction [6, 94]. The D-RD variants truncate the protein within or near the DAD and remove the basic RRKR motif (residues 1213–1216 in the MANE transcript numbering) and, in the exon 27 deletion, also the MDxLLExL motif, which prevents autoinhibition and makes DIA1 constitutively active; this is a gain-of-function mechanism, the opposite of that of the recessive loss-of-function disease [6, 21]. In megakaryocytes cultured from peripheral blood CD34+ cells, constitutive activation results in abnormal cell clustering, reduced proplatelet formation and disorganised cortical actin; platelets contain more filamentous actin and more stable microtubules, and expression of the truncated protein in cell lines reproduces these changes [92]. In the cochlea, DIA1 is located in the inner pillar cells, at the base of the outer hair cells and in the spiral ganglion neurons [27]. Transgenic mice expressing truncated DIA1 develop progressive hearing loss, beginning at high frequencies, with loss of outer hair cells in the basal turn and short, fused, elongated or sparse stereocilia, whereas the Dia1-null mouse hears normally, which confirms that the hearing loss does not result from loss of function [21]. In a knock-in model published in 2025, mice homozygous for the truncated allele had low-frequency hearing loss from 3 months, with progressive worsening, and thrombocytopenia only at 12 months; heterozygotes had no abnormalities during the period studied [24].

Table 6.2 brings together the variants with a described clinical consequence, normalised on the MANE transcript. The recurrent variant p.(Arg1213Ter) has been described in European, Japanese, Australian and Iranian families, and its historical designation p.R1204X results from the numbering of the alternative transcript NM_001079812, nine amino acids shorter [8, 21, 93, 95, 96].

Table 6.2. DIAPH1 variants associated with D-RD (transcript NM_005219.5) [6, 8, 10, 21, 27, 91, 92, 94, 97]

The genotype–phenotype correlation rests on few families. DAD truncations are associated with hearing loss and haematological abnormalities, and in Westbury's series the family with the greatest loss of the DAD (exon 27 deletion, which removes both regulatory motifs) had the lowest counts and the most bleeding symptoms [6]. The p.(Ala265Ser) variant in the DID reduces autoinhibition and increases actin polymerisation but, unlike DAD variants, does not disturb cytoskeletal assembly during cell division, and carriers have hearing loss only; Kim and colleagues propose that the degree of disturbance of cell division determines the presence of haematological involvement [94]. Conversely, the absence of thrombocytopenia on a single assessment does not exclude D-RD: in a Japanese cohort of 8 patients with p.(Arg1213Ter), hearing loss preceded thrombocytopenia in 3, and macrothrombocytopenia is progressive with age [24, 95]. Hearing loss was present in all carriers assessed in the Costa Rican family and in Westbury's series [6, 91].

Allelic disorders. Biallelic loss-of-function variants cause a distinct disease with autosomal recessive inheritance, seizures, cortical blindness and microcephaly syndrome (SCBMS; OMIM 616632; ORPHA:477814) [12, 98, 99]. It was described in 2015 in a consanguineous family with homozygous NM_005219.5:c.2332C>T p.(Gln778Ter), with microcephaly, severe visual impairment, intellectual disability and short stature, and later with NM_005219.5:c.2769del p.(Phe923LeufsTer4) and NM_005219.5:c.3145C>T p.(Arg1049Ter) in Omani families, with postnatal microcephaly, early-onset epilepsy, severe visual impairment and bronchiectasis [8, 98, 99]. In 5 Finnish patients homozygous for the splicing variant NM_005219.5:c.684+1G>A, enriched in the Finnish population, and in 2 Omani patients, SCBMS was accompanied by combined immunodeficiency due to defective lymphocyte maturation and mitochondrial dysfunction, and 3 patients developed B-cell lymphoma [8, 11]. The largest series, of 53 patients with biallelic variants published in 2026, described developmental delay, intellectual disability, progressive microcephaly, cortical visual impairment or blindness, epilepsy and brain abnormalities with occipital predominance; almost half had infections, mainly respiratory, and although most had normal lymphocyte subsets and immunoglobulins, T-cell receptor excision circles (TREC) and naive T lymphocytes were consistently low, which suggests the possibility of detection by TREC-based newborn screening for immunodeficiencies [100]. The heterozygous parents in these families do not have hearing loss [99]; the only published exception is a woman heterozygous for p.(Arg1049Ter) with sensorineural hearing loss beginning in the fifth decade, whose causal relationship the authors present as a hypothesis [101]. Rare heterozygous DIAPH1 variants, two of them de novo, were also identified in 6 of 108 non-Asian patients with sporadic moyamoya disease, with severe bilateral disease and transfusion-dependent thrombocytopenia; this is an association from genetic association studies, as yet without replication in independent cohorts in the sources consulted [102].

6.6.3. Clinical manifestations

Hearing loss is the most constant manifestation. In Westbury's series, bilateral sensorineural hearing loss was detected at newborn hearing screening or in early childhood in all 16 patients and progressed during childhood; 12 required hearing aids and 1 had a cochlear implant with a good outcome [6]. The audiometric profile varies between families: in the Costa Rican family the loss began at low frequencies, at around 10 years, and became profound at all frequencies in adulthood; in a large Japanese family it began at high frequencies in early childhood and progressed to severe hearing loss at all frequencies; in the German families it was congenital, with rapid progression and no low-frequency predominance; and in one of Ueyama's Japanese families the hearing loss worsened slowly from childhood and the patient was only referred at 48 years of age [21, 27, 95]. GeneReviews classifies DIAPH1 among the genes of autosomal dominant nonsyndromic hearing loss with low- or mid-frequency loss, a profile that corresponds to the original family [18]. In a Spanish series, the two patients with pathogenic or likely pathogenic variants had early-onset progressive bilateral hearing loss and vestibular abnormalities [22].

The haematological abnormalities are mild. All 16 of Westbury's patients had thrombocytopenia on at least one measurement, with a median count of 111 × 10⁹/L (range 13–209 × 10⁹/L), and large platelets, with a median mean platelet volume of 12.7 fL (9.3–19.8 fL); 11 had neutropenia on at least one occasion, with a median of 1.33 × 10⁹/L (0.50–4.30 × 10⁹/L) [6]. The smear showed no other morphological abnormalities, in particular no neutrophil inclusions, and neutrophil adhesion, degranulation, reactive oxygen species production and extracellular trap formation were normal, as were immunoglobulins, lymphocyte subsets and the red cell series; bone marrow biopsies showed megakaryocytes in normal numbers, but small and with hypolobulated nuclei, and in one patient reduced granulopoiesis [6]. Six of the 16 patients reported bleeding, predominantly mucocutaneous and mild, 3 had received prophylactic platelet transfusions before surgery or childbirth, and the 3 patients in one of the families had recurrent respiratory or skin infections [6]. In the large Japanese family no patient had a bleeding tendency, and mean platelet volume and platelet distribution width increased with age, indicating a progressive macrothrombocytopenia [95]. The Australian series confirmed the mild asymptomatic neutropenia and described impaired formation of procoagulant platelets and possible dental anomalies [96]. No series described nephropathy, cataracts or elevated liver enzymes, findings that distinguish D-RD from MYH9-RD [6, 103].

6.6.4. Diagnosis

There are no formal diagnostic criteria [96]. Suspicion arises in the presence of early-onset autosomal dominant sensorineural hearing loss with macrothrombocytopenia, particularly if there is mild neutropenia, no neutrophil inclusions and normal myosin-9 immunofluorescence [6, 103]. Because the thrombocytopenia is mild and may appear after the hearing loss, a full blood count with smear is part of the initial assessment of all autosomal dominant hearing loss, and a normal full blood count in childhood does not exclude the diagnosis [24, 27]. DIAPH1 is included in genetic hearing loss panels and in high-throughput sequencing panels for inherited platelet disorders [6, 18]. Panel or exome sequencing detects point variants and small deletions of the DAD, but the inversion in Westbury's family C, with intronic breakpoints, was identified by genome sequencing and confirmed by the absence of exon 27 in platelet complementary DNA, which shows that a negative panel does not exclude the disease when the phenotype is suggestive [6]. Interpretation depends on the position of the variant: a heterozygous truncation near the DAD, at the C-terminal end, is compatible with the dominant gain-of-function disease, whereas a heterozygous N-terminal or central truncation generally corresponds to a recessive SCBMS allele with no established hearing consequence [99, 101]. The report should state the transcript NM_005219.5, because the numbering of transcript NM_001079812 shifts the protein position by 9 residues [8, 93]. Novel variants found in patients with hearing loss and large platelets may remain classified as of uncertain significance until there is family segregation and functional study, as in a Saudi paediatric case with NM_005219.5:c.3633_3636del [104].

6.6.5. Treatment

Hearing loss is treated with hearing aids and, in severe to profound hearing loss, with a cochlear implant, with good results in the cases described [6, 22]. Avoidance of repeated exposure to loud noise is recommended for all people with genetic hearing loss [18]. The bleeding tendency is usually mild, and general measures and treatment of bleeding follow the principles for inherited thrombocytopenias described in Sections 7.2 and 7.3, with platelet transfusion reserved for severe bleeding or high-risk procedures, because repeated transfusions may cause anti-HLA alloimmunisation and refractoriness, as in one of Westbury's patients [6]. There is no therapy targeting the constitutive activation of DIA1 nor gene therapy in clinical development in the sources consulted, and no treatment of the mild neutropenia has been described [6, 96].

Eltrombopag was studied by Westbury and colleagues. In platelets from patients with the three variants, clinically relevant concentrations of eltrombopag stimulated the JAK2/STAT5 pathway of the thrombopoietin receptor as weakly as in controls, with no signs of abnormal platelet activation, and in cultured megakaryocytes eltrombopag reduced cell clustering and partially increased proplatelet formation, with a better effect in combination with thrombopoietin [6]. In the only published clinical use, a patient with the exon 27 deletion, anti-HLA alloantibodies and platelet refractoriness received eltrombopag 50 mg once daily from 29 days before a hip arthroplasty, increased to 75 mg daily from day 9 before until the day before surgery; the count rose from 29 × 10⁹/L to 72 × 10⁹/L, surgery was performed with tranexamic acid for 72 hours and without platelet transfusion, the estimated blood loss was 500 mL and there were no adverse effects [6]. This regimen coincides with the one studied in MYH9-RD (50 mg daily, 75 mg daily if insufficient, for 3 to 6 weeks) and is off-label; the European and American labels, for ITP, state 50 mg daily in adults and children aged 6 years or older, 25 mg daily in patients of East or Southeast Asian ancestry and a maximum of 75 mg daily (Table 7.3) [105–107]. The evidence in D-RD is limited to one patient, and the eltrombopag safety precautions described in Section 7.4, including hepatic and ophthalmological monitoring, apply in the same way [6, 106].

6.6.6. Genetic counselling

D-RD is transmitted in an autosomal dominant manner, and each child of a patient has a 50% probability of inheriting the variant [6, 27]. Haematological expressivity varies between families and with age, so relatives at risk should have a hearing assessment and full blood count and, when the familial variant is known, targeted testing [24, 95]. The sources consulted do not quantify the proportion of de novo variants. Prenatal diagnosis and preimplantation genetic diagnosis are technically possible once the familial variant has been identified, on the same terms as in Section 8.4. In families with SCBMS, the recurrence risk for each pregnancy of a carrier couple is 25%, and heterozygous carriers of loss-of-function alleles generally do not have hearing loss [99, 101].

6.6.7. Epidemiology and surveillance

Orphanet classifies D-RD as a disease with a prevalence below 1 per 1,000,000 and records, in its estimate, 8 cases or families [5]. Among 1,120 Japanese probands with bilateral nonsyndromic sensorineural hearing loss, the variant p.(Arg1213Ter) was found in 2 families (0.18%) [21]. In a tertiary centre in northern Spain, 4 of 385 patients (1%) with bilateral sensorineural hearing loss of unknown cause had DIAPH1 variants, and 2 (0.5%) had pathogenic or likely pathogenic variants associated with macrothrombocytopenia and neutropenia [22]. These proportions refer to series of patients with hearing loss and do not measure population prevalence. Outside these series, families have been published from the United Kingdom, Spain, Portugal, Germany, Australia, Japan, Korea, China, Iran, Saudi Arabia and the United States [6, 27, 92–97, 103, 104]. No prevalence data from Portugal or Brazil were found in the sources consulted.

There is no published surveillance protocol for D-RD. Table 6.3 adapts to D-RD the GeneReviews intervals for progressive genetic hearing loss and the haematological surveillance of MYH9-RD (Table 11.1), stating the extrapolation.

Table 6.3. Proposed surveillance in D-RD (extrapolated) [2, 6, 18, 24]

6.6.8. Published case

A patient from family C of Westbury's series, carrying the inversion that removes exon 27 of DIAPH1 (p.(Glu1192_Gln1220del)), had sensorineural hearing loss since childhood, macrothrombocytopenia and, owing to previous transfusions, multiple anti-HLA IgG alloantibodies with platelet refractoriness [6]. When he needed a right hip arthroplasty, unrelated to the disease, he started eltrombopag 50 mg daily 29 days before surgery, increased to 75 mg daily in the preceding 9 days; the platelet count, measured by fluorescence on a Sysmex XN analyser, rose from 29 × 10⁹/L to 72 × 10⁹/L on the day before surgery (course shown in Figure 2E of the original article) [6]. He received tranexamic acid for 72 hours from the start of surgery, received no platelets, haemostasis was satisfactory, with an estimated blood loss of 500 mL, similar to that of controls undergoing equivalent surgery, and there were no adverse effects [6]. The case supports the use of a short course of eltrombopag to avoid transfusion in an alloimmunised patient, but it is an isolated report and does not establish routine practice.

7. Treatments and Risk Management Strategies

7.1. Principles

The only guideline specific to MYH9-RD is the French National Diagnosis and Care Protocol (PNDS, Protocole National de Diagnostic et de Soins) of the Haute Autorité de Santé (HAS), published in November 2021 and written by consensus of experts from the CRPP and the MHEMO network [33]; the other formal references are the GeneReviews assessment, treatment and surveillance tables, the ISTH-BAT instrument validated for inherited platelet disorders and the multicentre study of surgery in inherited platelet disorders (SPATA) [2, 7, 108]. There is no therapy that modifies the course of the disease and no randomised controlled trial in MYH9-RD [7]. Follow-up is multidisciplinary and involves haematology or internal medicine with experience in haemostasis, nephrology, otorhinolaryngology and ophthalmology [2]. In patients without spontaneous bleeding, management is expectant, with regular full blood counts and physical examination, and interventions are reserved for haemostatic challenges [43]. Correct diagnosis is in itself a therapeutic measure, because it avoids corticosteroids, immunoglobulin, rituximab and splenectomy, which do not raise the count in these patients [2, 47, 79].

7.2. General measures and prevention of bleeding

Thrombocytopenia cannot be prevented [2]. General measures include educating the patient and doctors about drugs that affect platelet function, regular dental care and good oral hygiene to prevent gingival bleeding, and oral contraceptives to prevent or control menorrhagia, weighing the thrombotic risk of oestrogens [2]. Patients with severe thrombocytopenia and a significant bleeding tendency should avoid activities with a high risk of trauma, such as contact sports [2]. Table 7.1 lists the agents and circumstances to avoid or to use with caution.

Table 7.1. Agents and circumstances to avoid or to use only after assessment of risk and benefit (GeneReviews) [2]

Inherited thrombocytopenia does not protect against thrombosis. A 2026 commentary stresses that standard antithrombotic prophylaxis should not be omitted solely because of inherited thrombocytopenia, and that perioperative raising of the count with thrombopoietin receptor agonists may allow it to be given safely [73]. A 2026 Italian case describes bleeding and thrombosis in the same patient with MYH9-RD [72].

Vaccination, physical activity and drugs. According to the French PNDS, MYH9-RD does not contraindicate any vaccine, at any age, and the national schedule should be followed; vaccines may be given subcutaneously or intramuscularly with a fine 25 G needle, in the anterolateral thigh before 3 years of age and in the deltoid from 3 years, followed by local compression for 10 minutes [33]. Physical activity is recommended and adapted to the bleeding risk: above 50 × 10⁹/L there is no restriction for sports without a risk of trauma, below 50 × 10⁹/L contact and combat sports are discouraged, and for risk activities (cycling, horse riding, scooters, motorcycling, skiing) a helmet and protective equipment are advised; the decision is discussed with the reference centre, and the protocol does not consider an adapted physical activity professional indispensable [33]. Any prescription of anti-inflammatory, anticoagulant or antiplatelet drugs should be reported to the reference physician, so that risk activities can be adapted and the full blood count and iron monitored more closely [33].

Iron deficiency. Chronic bleeding, especially menorrhagia, causes often insidious iron deficiency, which should be looked for at every visit [33]. The French PNDS recommends, when intestinal absorption is preserved, oral iron salts at a dose of 325 mg per day in adults (a value given without specifying whether it refers to the salt or to elemental iron) and 3 to 6 mg/kg per day in children, for 3 to 6 months, to correct the anaemia and replenish stores; intravenous iron, preferably in hospital, is reserved for intolerance of or poor response to oral iron, the need for rapid correction and malabsorption [33].

7.3. Treatment of bleeding

Local measures. These are the first line for most mucocutaneous bleeding and often control mild to moderate bleeding: nasal packing or endoscopic cauterisation for epistaxis, suturing of accidental or surgical wounds, compression with gauze soaked in tranexamic acid for superficial wounds and tranexamic acid mouthwashes for gingival bleeding [2, 43].

The French PNDS details these measures: ice after trauma; for bruises and subcutaneous haematomas, pentosan polysulfate (Hemoclar®, France) or 10% arnica ointments; for superficial wounds, calcium alginate dressings or gauze soaked in tranexamic acid (Exacyl®, France); for epistaxis, as first line, packs soaked in saline kept at 4 °C or in tranexamic acid (oral formulation), calcium alginate haemostatic wicks or nasal haemostatic ointment, with hydration of the nasal mucosa to prevent recurrence; and, for oral bleeding, gentle mouthwashes with salt water, with chlorhexidine antiseptic (Paroex®, France) or with tranexamic acid for 2 to 3 minutes [33]. In the emergency setting, management depends on the type of bleeding, the count (measured by an appropriate method, because some analysers underestimate it and lead to wrong decisions), the history and the response to previous treatments; the care and emergency card issued at diagnosis allows the necessary blood components and drugs to be ordered, and the initial tests include haematology, haemostasis and pre-transfusion immunohaematology with screening for anti-HLA and antiplatelet (anti-HPA, human platelet antigen) antibodies if transfusion is anticipated [33].

Antifibrinolytics. Tranexamic acid and ε-aminocaproic acid are used systemically for mild to moderate mucocutaneous bleeding and, empirically, as prophylaxis for surgery and other haemostatic challenges, mainly low-risk ones [2]. Apart from menorrhagia, for which the French PNDS sets an oral dose (Section 7.10), the sources consulted do not give a specific dose for MYH9-RD; in the cochlear implant series, a patient with moderate thrombocytopenia received tranexamic acid 1 g intravenously before surgery and 1500 mg per day orally for 7 days, without bleeding, in a centre that uses it routinely in this setting [66]. In the SPATA study, antifibrinolytics alone were associated with excessive surgical bleeding in 17.6% of procedures (odds ratio, OR, 0.31 relative to no treatment), and the adverse effects reported were allergic or anaphylactic reactions and headache [108]. In murine models of MYH9-RD, tranexamic acid restored clot retraction and reduced bleeding [52]. Because tranexamic acid is renally eliminated, the French PNDS requires dose reduction and adjustment in moderate renal impairment, owing to the risk of accumulation, and considers it contraindicated in severe renal impairment, a restriction of particular weight in a disease in which a quarter of patients develop nephropathy [33]. The protocol accepts tranexamic acid and desmopressin for acute bleeding, outside the marketing authorisation and without precise recommendations [33].

Desmopressin. Desmopressin (1-deamino-8-D-arginine vasopressin, DDAVP), a selective vasopressin V2 receptor agonist that increases von Willebrand factor and platelet procoagulant activity, shortened the bleeding time in some patients with MYH9-RD, and uncomplicated surgery after desmopressin prophylaxis has been published; clinical efficacy studies in this disease are lacking [2, 43]. In SPATA, desmopressin, alone or with antifibrinolytics, was the prophylaxis associated with the least bleeding (8.1% of procedures; OR 0.13), mainly in platelet function defects, but it was used in only 1.1% of procedures in patients with defects in number [108]. The adverse effects described include headache, nausea and hypotension and, less frequently, hyponatraemia and renal dysfunction [108]. In a 70-year-old woman with May-Hegglin anomaly undergoing coronary artery bypass surgery, persistent bleeding after heparin reversal, with normal thromboelastography but platelet inhibition on platelet mapping, stopped with platelets and desmopressin [109]. The French PNDS acknowledges that experience with desmopressin (Minirin®) in surgery for patients with MYH9-RD is anecdotal, but accepts it, with a low level of evidence and on the opinion of many experts, as an alternative to transfusion in moderate-risk surgery, with platelets available in theatre; it does not consider a prior response test indispensable when factor VIII and von Willebrand factor are normal [33]. There are no clinical data on recombinant activated factor VII (eptacog alfa, NovoSeven®) in surgical prevention, and all these off-label prescriptions should be framed by an opinion from the CRPP or a CRC-MHC [33].

Platelet transfusion. Transfusion transiently raises the count and is effective in stopping bleeding, but exposes the patient to acute reactions, transmission of infections and alloimmunisation with refractoriness to subsequent transfusions [2]. It is therefore limited to active bleeding that cannot be controlled by other means, life- or organ-threatening bleeding and bleeding at critical sites, and is used as prophylaxis for surgery and childbirth when necessary; whenever possible, platelets from HLA-matched donors should be used [2]. When these are not available, the alternatives to reduce alloimmunisation are leucoreduction, single-donor apheresis platelets and ABO-compatible or crossmatch-compatible platelets [43]. A 2024 review cites, as general thresholds for prophylactic transfusion, 10 × 10⁹/L or 20 × 10⁹/L in the presence of risk factors and, before surgery, 50 × 10⁹/L for most major surgery and 100 × 10⁹/L for critical sites (indirect attribution) [43]. In kidney transplantation, transfusion may induce donor-specific anti-HLA antibodies, and there are reports of severe perioperative bleeding despite adequate transfusions [110]. In SPATA, 97.1% of transfusions were of fresh random-donor platelets (34.2% HLA-matched); in defects in number, the median dose was 2.5 units, given a median of 1 hour before the procedure, and transfusion alone was not associated with less bleeding, probably because it was reserved for the most severe cases [108]. The French PNDS considers transfusion of pooled platelet concentrates the undisputed treatment of severe bleeding in the emergency setting, with duration adjusted to the clinical picture and course, and finds no grounds for preferring one type of concentrate; HLA-matched platelets are discussed in the presence of anti-HLA antibodies and suspected refractoriness [33].

7.4. Thrombopoietin receptor agonists

Rationale. In MYH9-RD, as in other thrombocytopenias due to defective platelet formation, signalling between thrombopoietin and its receptor MPL is intact and megakaryocyte maturation is preserved, which makes these patients candidates for thrombopoietin receptor agonists (TPO-RAs) [28, 73]. Eltrombopag (Revolade® in the European Union, Portugal and Brazil; Promacta® in the United States) is an oral, non-peptide small molecule that mimics the action of thrombopoietin [2, 106, 107]. Its use in MYH9-RD is off-label: no TPO-RA is approved for inherited thrombocytopenias, and it should be prescribed and consented to as the use of a drug with phase II evidence [7, 73]. The French PNDS considers the rationale for these drugs strong in MYH9-RD and eltrombopag the most studied; since the count only rises at least one week after initiation, and generally from the second week, treatment can only be offered for scheduled procedures, with serial counts to determine how long the count remains above target, and it has the advantage of avoiding anti-HLA alloimmunisation [33].

Clinical trials. The first prospective trial (NCT01133860) treated 12 adults with MYH9-RD and a count below 50 × 10⁹/L with eltrombopag 50 mg daily for 3 weeks; patients who reached more than 150 × 10⁹/L stopped, those between 100 and 150 × 10⁹/L continued the dose for another 3 weeks and those remaining below 100 × 10⁹/L increased to 75 mg daily for 3 weeks [105]. There was a major response (at least 100 × 10⁹/L or three times baseline) in 8 patients and a minor response (at least double) in 3; bleeding disappeared in 8 of the 10 patients with baseline bleeding and there were mild adverse effects in 2 [105]. The results posted on the registry add that the trial recruited in Italy, between January 2009 and January 2010, patients aged 16 years or older with a mean count below 50 × 10⁹/L in the previous year (mean age 35.0 years, standard deviation 14.3; 7 women and 5 men; one participant aged 18 years or younger), excluding diseases with thromboembolic risk, thrombosis in the previous year, antiplatelet agents or anticoagulants, pregnancy or breastfeeding and creatinine of 20 mg/L or more [111]. The count was made by phase-contrast microscopy, all 12 patients completed treatment, the overall response was 91.6% (95% CI 61.5–99.7), there were no serious adverse effects, the mild effects were headache in 2 patients and dry mouth in 1, lasting 3 to 4 days at the start of treatment, and in vitro platelet aggregation was normal in 5 of the 7 patients who reached 100 × 10⁹/L [111]. The second trial (NCT02422394), an open-label phase II dose-escalation trial, took place in five Italian centres between April 2015 and May 2017 and included 24 patients with five inherited thrombocytopenias, nine of whom had MYH9-RD [28]. In part 1, patients received 50 mg daily for 3 weeks, stopping if the count exceeded 100 × 10⁹/L on day 21 and increasing to 75 mg daily for a further 3 weeks in the others [28]. Twenty-one of 23 evaluable patients responded (91.3%; 95% CI 72.0–98.9), 11 with a major response and 10 with a minor response, with a mean increase of 64.5 × 10⁹/L (95% CI 43.7–85.3) [28]. All patients with MYH9-RD responded, 7 of 9 (78%) exceeded 100 × 10⁹/L, the mean increase was 98.1 × 10⁹/L, and all those with a major response achieved it with 50 mg daily [28]. Spontaneous bleeding remitted in 10 of the 12 patients who had it at entry [28]. In part 2, four patients with spontaneous mucosal bleeding of WHO grade 2 or 3, two with MYH9-RD, received eltrombopag for 16 weeks, starting at 25 mg daily with adjustment every 4 weeks; the three who completed the period had stable remission of mucosal bleeding, two with 25 mg and two with 50 mg daily, and the two patients with MYH9-RD, who had grade 3 bleeding, improved their quality of life [28]. Thirty days after stopping, the mean count returned to 47.0 × 10⁹/L, similar to baseline, and the bleeding tendency returned to its initial pattern [28].

Table 7.2. Prospective trials of eltrombopag in MYH9-RD [28, 105, 111]

The two trials are small, open-label, without a control group and with a laboratory outcome, and do not allow eltrombopag to be compared with transfusion or the reduction in severe bleeding events to be estimated [28, 105]. No negative, suspended or withdrawn trials in MYH9-RD were identified; a French trial of perioperative eltrombopag in inherited thrombocytopenias (NCT03638817, Toulouse, 13 patients aged 6 to 75 years, 2019–2023, with eltrombopag at ITP doses from 4 weeks until 2 days before the procedure and transfusion if the count was below 80 × 10⁹/L) is completed without results posted on the registry [111].

Preparation for surgery. In a single-centre retrospective series, 11 consecutive surgical procedures in patients with MYH9-RD, severe thrombocytopenia and high bleeding risk were prepared with eltrombopag, and in 10 the surgery proceeded without bleeding, other complications or transfusion [2, 112]. In 2026, the Centro Hospitalar Universitário de Santo António (Porto) published 15 procedures in 9 patients with inherited thrombocytopenias, three of whom had MYH9-RD, prepared with eltrombopag 50 mg daily started on average 24 days before and stopped 2 to 3 days after the procedure, with approval from the pharmacy and therapeutics committee and the ethics committee for off-label use [81]. There was a favourable response in 14 of 15 procedures and a major response in 10, with a median count of 45 × 10³/µL at baseline and 103 × 10³/µL on the day of surgery, without bleeding, transfusions, hepatic abnormalities or thromboses and with one case of headache [81]. In the three patients with MYH9-RD, the count rose from 28 to 68 × 10³/µL before a hernia repair at 70 years, from 42 to 160 × 10³/µL before a total knee arthroplasty at 46 years and from 46 to 80 × 10³/µL, with only 13 days of treatment, before a hernia repair at 12 years [81]. The commentary accompanying the series counted 17 published procedures with preoperative eltrombopag in MYH9-RD and concluded that TPO-RAs should be considered a preferred option for elective surgical preparation in many patients with inherited thrombocytopenias, started early enough to allow escalation from 50 to 75 mg daily [73]. GeneReviews accepts a short course of eltrombopag to raise the count transiently before elective surgery or invasive procedures [2]. In DIAPH1-related disease, the same regimen (50 mg daily, increased to 75 mg) allowed a hip arthroplasty without transfusion in an alloimmunised patient (Section 6.6.5) [6].

Kidney transplantation and prolonged use. In 2025 the first living-donor kidney transplant performed without platelet transfusion thanks to eltrombopag was published: a 19-year-old woman with p.(Arg702Cys) started the drug 6 weeks before, and the count rose from 10 × 10³/µL to 156 × 10³/µL, without major bleeding [110]. The variant was de novo; macrothrombocytopenia was detected at birth, proteinuria appeared at 2 years, a renal cyst at 3 years with spontaneous intracystic haemorrhage at 12, sensorineural hearing loss (about 40 dB) at 17 and ESKD at 19 years, without cataracts [110]. The regimen, approved by the institutional off-label use committee, began with 50 mg daily and was increased to 75 mg daily 34 days before the transplant; the count only reached 30 × 10³/µL until it was found that eltrombopag was being taken at the same time as the other medicines, and after separating the doses it reached 161 × 10³/µL on admission and 156 × 10³/µL on the day of surgery, above the target of 100 × 10³/µL [110]. Intraoperative blood loss was 146 mL, there was no platelet transfusion (red cell transfusion was the centre's routine), the drug was continued until postoperative day 18 without hepatotoxicity and graft creatinine was about 1.5 mg/dL at 15 months [110]. On postoperative day 1 there was a seizure attributed to hypertensive encephalopathy (blood pressure 165/95 mmHg, trough tacrolimus 12.9 ng/mL), with full recovery; magnetic resonance imaging showed small old infarcts in the frontal white matter, which the authors did not attribute to eltrombopag, but the patient was also taking daprodustat, an HIF prolyl hydroxylase inhibitor, and the authors recommend attention to the thromboembolic risk [110]. The same authors report that, in a Japanese national survey, 6 of 7 kidney transplant recipients with MYH9-RD required red cell transfusion for perioperative bleeding, and that a paediatric case of unrelated-donor transplantation with eltrombopag reduced, without eliminating, platelet transfusions (indirect attribution) [110]. In 2026 a 52-year-old man was described with severe thrombocytopenia, bleeding, hearing loss, cataracts and ESKD with kidney transplantation, in whom prolonged off-label eltrombopag raised the count to normal and reduced bleeding [113].

Reference doses and comparison with the label. The reference dose in MYH9-RD is that of the trials: 50 mg daily, increased to 75 mg daily if the count is insufficient after 3 weeks, in a course of 3 to 6 weeks; in prolonged use, 25 to 50 mg daily [28, 105]. For surgical preparation, starting about 3 to 4 weeks beforehand allowed the peak to be reached, which occurred on average 20 days after initiation [73, 81]. The European and American labels, which cover only ITP and other acquired thrombocytopenias, provide for the same starting dose of 50 mg daily in adults and children aged 6 years or older, 25 mg daily between 1 and 5 years, reduction to 25 mg daily in patients of East or Southeast Asian ancestry (and, in the American label, in hepatic impairment, with 12.5 mg when both coexist) and a maximum of 75 mg daily [106, 107]. The MYH9-RD sources do not describe doses in young children; in patients of East Asian ancestry, the only regimen described is that of the Japanese transplant report, which started at 50 mg daily and increased to 75 mg daily, above the label starting dose of 25 mg for this ancestry, without hepatotoxicity, but with only one patient treated [106, 107, 110]. Eltrombopag is taken fasting or with a low-calcium meal (up to 50 mg), at least 2 hours before or 4 hours after antacids, calcium-rich foods and supplements containing polyvalent cations [107].

Table 7.3. Eltrombopag: dose studied in MYH9-RD and label doses (comparison) [28, 81, 105–107, 110]

Safety. In the MYH9-RD trials, eltrombopag was well tolerated: in the second trial there were 7 grade 1 adverse effects in 5 of 24 patients, mainly transient headache or bone pain in the first 2 to 3 days (17%), and a rise in creatinine attributed to urinary retention [28]. Eltrombopag is associated with cataracts in ITP; in controlled trials in adults with ITP, cataracts appeared or worsened in 7% of patients on eltrombopag and 7% on placebo, and in 11% in the extension, and in the hepatitis C pool in 8% vs 5% [106, 107]. Since MYH9-RD predisposes to cataracts, it is relevant that no patient in the Italian trial, including three with baseline cataracts and two treated for 16 weeks, had onset or worsening of cataracts [28]. The American label has a boxed warning for severe hepatotoxicity and recommends ALT, AST and bilirubin before initiation, every 2 weeks during titration and monthly thereafter; the drug inhibits UGT1A1 and OATP1B1 and may cause indirect hyperbilirubinaemia [107]. The European SmPC recommends a full blood count with smear weekly until stable and then monthly, regular liver tests and routine ophthalmic monitoring, and reports portal venous system thromboembolism in 4% of patients with chronic liver disease treated with 75 mg daily for 2 weeks before invasive procedures, mainly with counts above 200,000/µL [106]. The drug is intensely coloured and may interfere with laboratory assays of bilirubin, creatinine and proteins [107]. In MYH9-RD, baseline elevation of liver enzymes, present in half of patients, should be recorded before treatment [64]. The concern about clonal evolution associated with prolonged TPO-RA use applies mainly to inherited thrombocytopenias with predisposition to myeloid neoplasms, such as ANKRD26-related thrombocytopenia, and not to MYH9-RD [28, 73].

Other TPO-RAs. Romiplostim, a recombinant protein given subcutaneously, was used in the surgical preparation of a patient with MYH9-RD, and in a patient described by Rabbolini and colleagues, treated for 36 months, starting at 1 µg/kg per week with weekly increases of 1 µg/kg, the count tripled in 5 weeks from 7 × 10⁹/L and remained on average at 33 × 10⁹/L with a mean dose of 6.3 µg/kg per week (indirect attribution) [34, 43, 81]. The American label of romiplostim (Nplate®), for ITP, sets the maximum dose at 10 µg/kg per week [114]. There are no trials of romiplostim in MYH9-RD [43]. Avatrombopag, an oral TPO-RA without food restrictions and without significant hepatotoxicity, approved for ITP and for thrombocytopenia of chronic liver disease, has been described in a single case of MYH9-RD initially diagnosed as ITP, with a response after failure of eltrombopag (indirect attribution) [115]. Lack of response to one TPO-RA does not exclude a response to another [73].

7.5. Regulatory status

Table 7.4. Regulatory status of eltrombopag [106, 107, 116, 117]

7.6. Surgery and invasive procedures

The need for prophylaxis before surgery or invasive procedures is decided by the type of procedure, the patient's bleeding history and the prior platelet count, and may include platelet transfusion, a short course of eltrombopag and empirical use of antifibrinolytics or desmopressin [2]. The SPATA study analysed 829 procedures in 423 patients with inherited platelet disorders from 49 centres in 17 countries, 75 of whom had MYH9-RD, the most frequent form among the defects in number [43, 108]. In defects in number, the median count at surgery was 56 × 10⁹/L, prophylaxis was used in 20.6% of procedures and there was excessive bleeding in 13.4% (15% in major surgery, 10.7% in minor invasive procedures and 10.1% in dental procedures); in MYH9-RD, excessive bleeding occurred in 15.5% of procedures [108]. Bleeding was more frequent with a WHO bleeding score of 2 or more (more than four times), with an ISTH-BAT of 6 or more, after bleeding in a previous procedure and, in defects in number, with a count below 68 × 10⁹/L and age above 70 years; cardiovascular (47.1%) and urological (34.2%) surgery carried the highest risk, but there was also bleeding in minor procedures, such as gastroscopy with biopsy [108]. Prophylaxis was associated with less bleeding in function defects, but not in defects in number, which may reflect the lower absolute risk or less intensive prophylaxis; three patients with MYH9-RD had an unfavourable course of bleeding [108]. In the cochlear implant series, four patients with severe thrombocytopenia received 1 to 2 apheresis platelet concentrates, one received tranexamic acid and four with automated counts of 70 to 129 × 10⁹/L received no prophylaxis, with no relevant postoperative bleeding [66].

The French PNDS organises the preparation of scheduled procedures in a prior consultation, in which the strategy is explained and a hospital is chosen that can count platelets by an appropriate method and supply blood components in an emergency, and gives the teams a written protocol with the haemostatic treatment and anaesthetic precautions [33]. The risk depends on the count, the bleeding history, comorbidities, previous transfusions and the type of procedure; for a low-risk procedure in a patient with few symptoms, the most frequent situation, no preoperative treatment is needed [33]. Preoperative transfusion is indicated when the recent count, measured by an appropriate technique, is below the threshold for the procedure: 50 × 10⁹/L for minor surgery, 80 × 10⁹/L for major surgery and 100 × 10⁹/L for surgery with an immediate life- or function-threatening bleeding risk, such as neurosurgery [33]. If the baseline count is close to the threshold and the procedure carries no vital or functional risk and allows bleeding to be detected quickly, the concentrates are prepared and immediately available, without systematic transfusion; in other cases, they are given in the 2 hours preceding the procedure [33]. By analogy with aplasia, in patients at risk of prior alloimmunisation (pregnancy, transfusion), anti-HLA and anti-HPA antibodies are screened for before transfusion and, if positive, the HLA-A and HLA-B phenotype and the HPA genotype are determined, so that compatible platelets can be supplied in case of ineffectiveness; the transfusion yield assesses the effectiveness of each transfusion [33]. Biological glues and haemostatic dressings reinforce local haemostasis [33].

Postoperatively, MYH9-RD does not exempt the patient from thromboprophylaxis when indicated, which agrees with the 2026 commentary [33, 73]. After transfusion, the French PNDS considers indispensable the transfusion yield in the first 24 hours, follow-up of platelet kinetics until return to baseline and later screening for anti-HLA and anti-HPA antibodies, to identify refractory states [33].

7.7. Nephropathy

Angiotensin-converting enzyme (ACE) inhibitors and angiotensin receptor blockers (ARBs), started early, may reduce or induce remission of proteinuria and thus delay the progression of kidney disease, according to retrospective observations [2, 68, 118]. Long-term benefit on progression has not been demonstrated, and there are reports of failure of this treatment [19, 34]. The 2019 nephrology review proposes that renin-angiotensin system blockade protects against triggers of progression in the less severe variants, whereas some variants would be too severe to respond [19]. In high-risk genotypes, particularly Arg702, screening for proteinuria should be especially close so that treatment can be started as early as possible [23]. The sources consulted do not describe specific doses of these drugs in MYH9-RD, nor trials with other nephroprotective agents. In ESKD, dialysis and kidney transplantation are the only options [2]. In the French series, the six patients with ESKD were transplanted [47]. In one case, preparation for transplantation with eltrombopag made it possible to avoid platelet transfusion, which may induce donor-specific anti-HLA antibodies [110]. Thrombocytopenia is a bleeding risk factor at the start of dialysis and in transplantation [110].

The French PNDS recommends screening all patients for proteinuria and haematuria, entrusting the nephropathy to a nephrologist, preferably in a rare kidney disease service, adjusting the frequency of follow-up to severity and aiming to reduce proteinuria and achieve optimal blood pressure control [33]. It sets the safety threshold for renal biopsy at 50 × 10⁹/L, in line with HAS recommendations and in the absence of other bleeding risk factors, with platelet transfusion, desmopressin or eltrombopag if necessary, and cites literature in favour of maintaining the count at 100 × 10⁹/L during the 7 to 10 days following kidney transplantation [33]. Recombinant activated factor VII has not been used to prevent bleeding from biopsy or transplantation in MYH9-RD, but it has been effective as rescue treatment for bleeding after biopsy and in transplantation in other settings, and may be considered in the presence of antiplatelet antibodies or a lack of concentrates [33].

7.8. Hearing loss

Hearing aids are indicated for clinically significant hearing loss [2]. Cochlear implantation is effective in severe to profound hearing loss: in an international series of 10 patients implanted between 1987 and 2009 in 8 centres, 9 benefited, 8 with practically normal hearing and verbal communication and 1 with a slightly poorer result, attributed to the long duration of severe hearing loss; there were no bleeding complications attributable to the disease, and the complications recorded were a small wound haematoma and a perilymph leak at the cochleostomy [66]. The authors recommend offering the implant as soon as the candidacy criteria are met [66]. In a 2026 Japanese cohort, nine patients received cochlear implants and, in the six with data, the mean threshold improved from 108.1 dB to 30.5 dB, with a mean monosyllable recognition of 75% [20]. In DFNA17, five members of an Australian family had excellent results, unlike the poor result in one patient from the original American family, which suggests the importance of early intervention [3, 29]. In unfavourable genotypes, closer audiological follow-up and earlier intervention are recommended [57]. Loud noise and ototoxic drugs should be avoided (Table 7.1) [2].

The French PNDS indicates hearing aids as the most frequent measure in less severe forms, describes excellent results of cochlear implantation in most patients with severe hearing loss and provides for haemostatic treatment immediately before implantation, according to the count [33].

7.9. Cataracts and liver

Cataracts are treated surgically when indicated, and glucocorticoids should be used only after weighing the risk [2]. Elevated liver enzymes have no specific treatment, but require caution with hepatotoxic drugs and the exclusion of other causes of liver injury [2].

The French PNDS advocates an annual ophthalmological check-up, with the final frequency defined by the ophthalmologist, and surgical treatment of cataract; it considers that elevated aminotransferases do not appear to have unfavourable prognostic value and may normalise spontaneously [33].

7.10. Pregnancy, delivery and neonatal period

GeneReviews recommends managing delivery as in other thrombocytopenias, because platelet function is usually normal, and generally indicates a count of at least 50 × 10⁹/L for delivery; severe thrombocytopenia and a history of severe bleeding are associated with more delivery-related bleeding [2]. In a retrospective multicentre study of 339 pregnancies in 181 women with 13 inherited thrombocytopenias, neither thrombocytopenia nor bleeding worsened in pregnancy, miscarriages, fetal bleeding and preterm births did not differ from those of the healthy population, and delivery-related maternal bleeding occurred in 6.8–14.2% depending on the definition, with no maternal deaths or hysterectomies; bleeding requiring transfusion was more frequent with a history of severe bleeding and a count below 50 × 10⁹/L at delivery [77]. The study, promoted by the thrombocytopenia working group of the European Hematology Association, brought together 45 centres from 20 countries; MYH9-RD was accepted with positive immunofluorescence in 12 women without molecular study, and 57 of the 181 women had had an initial misdiagnosis of ITP, 44 with unnecessary treatments, including 15 splenectomies [77]. Spontaneous miscarriages occurred in 34 pregnancies (10.1%; 95% CI 7.2–14.0), a proportion comparable to the 10.2% of an American population, and the 30 preterm births (9.9%; 95% CI 6.8–13.7) were all at 36 or 37 weeks [77]. The count at delivery, available in 188 cases, averaged 65 × 10⁹/L, compared with 56 × 10⁹/L before pregnancy [77]. There were 116 caesarean sections in 303 deliveries; of the 289 with data on analgesia, 80 (27.7%) had general anaesthesia and 46 (15.9%) spinal or epidural anaesthesia, with no bleeding complications attributed to these procedures [77]. Prophylactic platelet transfusions were given before 46 of 301 evaluable deliveries and, more rarely, corticosteroids (9), desmopressin (5), tranexamic acid (2) and recombinant activated factor VII (1), across all forms of thrombocytopenia together [77]. Excessive maternal bleeding of any type occurred in 42 deliveries (14.2%; 95% CI 10.4–18.7) and bleeding requiring transfusion in 20 (6.8%; 95% CI 4.2–10.2); 15 deliveries (5.1%) required red cell concentrate, compared with 0.5–1.2% in women without thrombocytopenia [77]. Bleeding requiring transfusion was associated with previous bleeding of WHO grade 3 or 4 (OR 5.32; 95% CI 1.22–23.11) and of grade 4 (OR 24.50; 95% CI 4.75–126.41), and ROC analysis identified as optimal thresholds 44 × 10⁹/L before pregnancy and 50 × 10⁹/L at delivery, below which the risk was 3.92 (95% CI 1.43–10.70) and 7.61 (95% CI 1.55–37.60) times higher [77]. Bleeding requiring transfusion was not less frequent in caesarean sections than in vaginal deliveries, nor in women transfused prophylactically, but the latter had lower counts at delivery (40 vs 69 × 10⁹/L), which precludes conclusions about the efficacy of prophylaxis [77]. The absence of differences between diseases is interpretable only for the forms with more deliveries, among them MYH9-RD [77]. The thrombocytopenia of affected newborns was similar to that of their mothers (mean 69 vs 58 × 10⁹/L in 126 pairs); 7 of 156 affected newborns had delivery-related bleeding, 5 with petechiae and 2 with fatal cerebral haemorrhage [77]. The two newborns who died were born vaginally to mothers with MYH9-RD and counts of 12 and 16 × 10⁹/L; one had the same variant as the mother and 7 × 10⁹/L, and the other died shortly after birth without a count or test [77]. The authors found no intrauterine bleeding in the 156 affected fetuses, and the limitations of the study are the retrospective design, missing data, comparison with the general population and overestimation of thrombocytopenia in macrothrombocytopenias by automated counters [77]. According to a 2026 review, these two deaths occurred in children of mothers with severe thrombocytopenia due to MYH9-RD born vaginally, but the number is too small to conclude that vaginal delivery increases the risk of neonatal intracranial haemorrhage [119]. GeneReviews considers newborns delivered vaginally to mothers with severe thrombocytopenia to be at increased risk [2].

Threshold recommendations diverge. GeneReviews indicates 50 × 10⁹/L for delivery [2]; the 2026 review reserves treatment during pregnancy for counts below 20 × 10⁹/L or active bleeding and refers to transfusion to at least 70 × 10⁹/L when neuraxial anaesthesia is anticipated [119]; a 2024 review cites thresholds of 50 × 10⁹/L for vaginal delivery or caesarean section and 70 × 10⁹/L for epidural anaesthesia (indirect attribution) [43]. A 2025 systematic review identified 96 pregnancies in 56 mothers with May-Hegglin anomaly: 51% vaginal deliveries and 43.8% caesarean sections, 85.4% uneventful, only one postpartum haemorrhage probably attributable to the disease, no epidural or spinal haematoma in women with neuraxial anaesthesia and 85.4% of the 103 newborns without complications [120]. In inherited platelet disorders in general, delivery in a centre with a multidisciplinary team is recommended, avoiding vacuum extraction and reserving forceps for urgent situations [119]. During pregnancy, counts should be monitored with serial full blood counts, and the automated count may greatly underestimate the real count, as shown by a pregnant woman with 5 × 10³/µL by impedance and 19 × 10³/µL by fluorescence and another, previously misdiagnosed with ITP and treated with corticosteroids, immunoglobulin and a TPO-RA, who had an uncomplicated caesarean section under general anaesthesia [80, 119, 121]. TPO-RAs have not been studied in pregnancy, which was an exclusion criterion in the trials, and the eltrombopag label considers the data insufficient and advises against breastfeeding [107, 119]. There are no reports of desmopressin at delivery in women with MYH9-RD [119].

Menorrhagia and preconception (French PNDS). The severity of menorrhagia is assessed with the Higham score, and its management follows the usual recommendations; non-steroidal anti-inflammatory drugs often used in self-medication (such as mefenamic acid and flurbiprofen) and acetylsalicylic acid are contraindicated in heavy bleeding [33]. First-line treatment is oral tranexamic acid at a dose of 1 g three times daily, from the first day and throughout menstruation, at least for the first 3 days, reduced to 20 mg/kg per day in 2 doses in very young girls, with no specific paediatric maximum in the protocol and, by extrapolation from the adult dose, not exceeding 3 g per day; hormonal treatment, alone or in combination, is decided jointly by gynaecology, endocrinology and haemostasis, after a gynaecological cause has been excluded [33]. Since, rarely, stopping hormonal treatment in order to conceive causes bleeding to recur, a fertility assessment may be offered before stopping, and simple ovarian stimulation with intrauterine insemination may be discussed in a multidisciplinary team [33].

Pregnancy follow-up (French PNDS). Management depends on the count, the bleeding history and the presence of anti-HLA and anti-HPA antibodies in women previously transfused; the multidisciplinary protocol (obstetrics, anaesthesia and reference centre) is established from the second trimester, in case of preterm delivery, and updated throughout pregnancy [33]. The count is monitored monthly and more frequently from 34 weeks of gestation; since the count changes little in pregnancy, a sudden fall should prompt a search for a superimposed obstetric disease, and iron deficiency and anaemia should be treated [33]. In forms with bleeding risk, delivery should take place in a level 2 or 3 maternity unit with maternal and neonatal intensive care, a haemostasis laboratory and a blood service capable of rapidly supplying components, which may be stored in the delivery suite [33]. In the case of an interrupted pregnancy, spontaneous or voluntary, medical evacuation is contraindicated because of the bleeding risk and possible nephropathy, and surgical aspiration with haemostatic cover and ultrasound confirmation of uterine emptiness is preferred; in ectopic pregnancy, methotrexate is contraindicated and treatment is surgical [33].

Delivery (French PNDS). MYH9-RD does not increase fetal loss, prematurity or obstetric complications, but exposes the mother to postpartum haemorrhage; the risk is considered low above 50 × 10⁹/L, and below that value transfusion of concentrates immediately before delivery should be considered to exceed 50 × 10⁹/L, which, in an induced labour, means transfusing close to delivery and not at the start of induction [33]. Since the fetus has a 50% probability of having thrombocytopenia, fetal scalp blood sampling (pH or lactate) and vacuum extraction are formally contraindicated, and forceps and rotational manoeuvres are used with caution [33]. In patients with moderate bleeding risk, caesarean section is not systematic and its obstetric indications are unchanged; spontaneous labour is favoured, induction by cervical ripening with an unfavourable cervix and episiotomy are avoided, the third stage is actively managed with oxytocin, uterine revision should be easy and perineal injuries are repaired immediately [33]. Caesarean section is preceded by transfusion according to the count, with meticulous surgical haemostasis, without intraperitoneal or subcutaneous drainage and without compressive dressings, which prevent monitoring of the abdominal wall [33]. TPO-RAs in pregnancy are exceptional, decided case by case in patients with obstetric bleeding risk or a very low count, starting at the beginning of the third trimester and with the opinion of the reference centre [33].

Neuraxial anaesthesia (French PNDS). The decision is individual and shared between the reference centre, anaesthesia and obstetrics; the consensus thresholds of the French Society of Anaesthesia and Intensive Care (SFAR) may be adopted, 75 × 10⁹/L for epidural and 50 × 10⁹/L for spinal anaesthesia, and below them the decision depends on the benefit–risk balance [33]. Removal of the epidural catheter carries a risk similar to that of the puncture and requires a prior count, repeated also if blood loss exceeds the usual [33]. When neuraxial anaesthesia is contraindicated by severe thrombocytopenia that cannot be corrected, peripheral blocks and intramuscular analgesic injections are also contraindicated, and the alternatives are intravenous analgesia for vaginal delivery and general anaesthesia for caesarean section [33].

Alloimmunisation (French PNDS). Before any transfusion, anti-HLA and anti-HPA antibodies are screened for in women with previous pregnancy or transfusion [33]. Anti-HPA antibodies can cross the placenta and cause fetal and neonatal thrombocytopenia and bleeding, so they warrant more frequent obstetric ultrasound from the beginning of the second trimester and may lead, as in fetomaternal alloimmune thrombocytopenia, to intravenous immunoglobulin at a dose of 1 g/kg once weekly (a single weekly dose, which is also the maximum provided for) from 18 weeks of gestation, with or without corticosteroid therapy at 0.5 mg/kg per day from 30–32 weeks (fixed dose, with no escalation or different maximum stated in the protocol) [33]. In fetomaternal alloimmune thrombocytopenia, the 2015 HAS recommendation stratifies the start by risk: in very high or high risk situations, immunoglobulin 1 g/kg per week from 16 weeks, with prednisone 0.5 mg/kg per day from 32 weeks or earlier; in standard risk, immunoglobulin 1 g/kg per week from 20–22 weeks, with prednisone from 32 weeks [122].

Postpartum (French PNDS). The most severe forms are monitored clinically for 6 weeks, and the patient should be informed that late postpartum haemorrhage may occur up to 6 weeks; the discharge prescription includes tranexamic acid (1 g three times daily), a full blood count with iron studies one week after discharge and the contact details of the reference centre, and tranexamic acid is also prescribed for the first menstruation after delivery, particularly if there was menorrhagia [33]. Hormonal treatment should be started or resumed early by the gynaecology team, and thromboprophylaxis, when indicated, is decided case by case according to thrombotic risk factors and the count [33].

Newborn (French PNDS). A count at birth is recommended, preferably on cord blood; a count below 50 × 10⁹/L is confirmed on a peripheral venous sample with a smear, and the count is repeated on day 3 or 5, before discharge from the maternity unit, and later [33]. The severity of thrombocytopenia in an affected newborn is usually similar to that of the affected parent [33]. Below 50 × 10⁹/L, the paediatrician looks for mucocutaneous and cerebral signs of bleeding; below 20–30 × 10⁹/L or with bleeding, whatever the count, platelet transfusion is recommended, unless the reference centre advises otherwise [33]. The dose written in the PNDS, 0.2 × 10¹¹ platelets per 5 to 7 kg of body weight, not exceeding 3 to 4 × 10¹¹ platelets and 15 mL/kg, does not correspond to the 2015 French HAS recommendation on platelet transfusion, which sets for the newborn 0.1 to 0.2 × 10¹¹ platelets per kg of body weight, in a usual volume of 15 mL/kg not exceeding 20 mL/kg, and for older children and adults 0.5 to 0.7 × 10¹¹ platelets per 10 kg; the reference dose is that of the 2015 HAS recommendation [122]. The diagnosis is recorded in the child health record before discharge, acetylsalicylic acid and platelet function inhibitors are contraindicated, and any invasive procedure, even minor such as circumcision, requires specialist advice if the count is below 20–30 × 10⁹/L or there is bleeding [33]. In a newborn without a family history, due to a de novo variant, persistent thrombocytopenia without an evident cause, with large platelets and basophilic inclusions, should lead to referral to a reference or competence centre [33].

7.11. Children, adolescents and transition

The diagnosis is often made in childhood through an incidental count, and the neonatal period and the first years concentrate the rare severe bleeding episodes [47]. In children, correct diagnosis avoids immunosuppression and splenectomy, which in the French series was performed in a child as young as 22 months [47, 85]. High-risk genotypes, such as those of Epstein syndrome, require renal and hearing surveillance from childhood, because more than half of these patients have renal failure, hearing loss and proteinuria by 19 years [46]. The Porto series included a 12-year-old child treated with eltrombopag 50 mg daily before a hernia repair [81]. In adolescent girls, menorrhagia should be prevented with hormonal contraceptives [2, 43]. The French PNDS sets out the principles of paediatric follow-up and transition [33]. School entry should not be delayed; an individualised reception plan (PAI, projet d’accueil individualisé), drawn up at the request of the family or the school with the specialised centre, frames life in the community and the conditions for school sport [33]. Follow-up takes place in a specialised centre, initially once a year, with the paediatrician or general practitioner, to establish a follow-up plan with the parents, provide written information and detect extra-haematological manifestations [33]. Therapeutic education begins early, is adapted to age, is aimed at the patient and carers and is part of every consultation [33]. Transition to adult medicine is phased, with specific consultations and the participation of the patient, the parents and the paediatric and adult teams, in a period in which adherence to follow-up and treatment tends to decrease; in girls, a gynaecology consultation is recommended from the first signs of puberty, to treat menstrual bleeding if necessary, and the adolescent should be informed that the disease, well tolerated at the end of adolescence, may become complicated in the following years by extra-haematological manifestations [33].

7.12. Recent developments (2025–2026)

Table 7.5 summarises the publications of clinical relevance from 2025 to October 2026, most from the last 12 months, with the level of evidence of each. No regulatory authorisation, CHMP opinion or marketing authorisation application for MYH9-RD was identified, nor any active registered clinical trial in the disease [111].

Table 7.5. Recent publications of clinical relevance [4, 20, 46, 49, 50, 52, 55, 69, 73, 74, 81, 89, 110, 113, 115]

7.13. Guidelines and divergences

Table 7.6. Normative sources and divergences [2, 25, 43, 73, 108, 119]

The French PNDS diverges from the other sources on several points [33]. For neonatal transfusion, the dose written in the PNDS (0.2 × 10¹¹ platelets per 5–7 kg, up to 15 mL/kg) is lower than that of the 2015 HAS recommendation (0.1–0.2 × 10¹¹ platelets per kg, about 15 mL/kg and up to 20 mL/kg), which is the reference; for antenatal immunoglobulin in the presence of anti-HPA antibodies, the PNDS's single start at 18 weeks simplifies the risk stratification of the same recommendation (16 or 20–22 weeks) [33, 122]. In surgery, it adds an intermediate threshold of 80 × 10⁹/L for major surgery, where the 2024 review cites 50 × 10⁹/L for most major surgery, and keeps 100 × 10⁹/L for critical sites [33, 43]. For neuraxial anaesthesia, it separates epidural (75 × 10⁹/L) from spinal anaesthesia (50 × 10⁹/L), compared with a single threshold of 70 × 10⁹/L in the 2024 and 2026 reviews [33, 43, 119]. For delivery, the threshold of 50 × 10⁹/L coincides with that of GeneReviews and with the optimal threshold of the multicentre pregnancy study, whereas the 2026 review treats during pregnancy only below 20 × 10⁹/L or with bleeding [2, 33, 77, 119]. For hearing surveillance, it proposes audiometry from 3 years of age, every 2 to 3 years in childhood and adolescence, every 5 to 10 years in adults without hearing loss and annually after the diagnosis of hearing loss, whereas GeneReviews indicates 3 years for all and the 2024 review 3 to 5 years [2, 33, 43]. For ophthalmology, it recommends examination from 20 years of age with annual check-ups, compared with an examination at diagnosis and every 3 years in GeneReviews [2, 33]. For diagnosis, it considers identification of the variant indispensable, whereas the 2018 review accepts the diagnosis with typical immunofluorescence [33, 34]. In pregnancy, it exceptionally accepts TPO-RAs at the start of the third trimester, for which there are no studies and which the label considers to lack sufficient data [33, 107, 119]. The frequencies it gives (hearing loss in about 50%, nephropathy in about 30%, cataract in about 5%, about 30% de novo cases) differ from those of GeneReviews (80–85% over a lifetime, 25%, 18–20% and about 35%) [2, 33]. The points of convergence are the lack of value of platelet function tests, shared by the German guideline, and the refusal to omit thromboprophylaxis [25, 33, 73]. As for doses, the French PNDS is the only source with doses for menorrhagia (tranexamic acid), iron deficiency, neonatal transfusion and alloimmunisation; for eltrombopag it gives no dose, and the reference remains that of the trials (Table 7.3) [28, 33].

8. Reproductive Options and Family Counselling

8.1. Inheritance and risk to relatives

MYH9-RD is transmitted in an autosomal dominant manner [1, 2]. About 65% of probands have an affected parent and about 35% are simplex cases, mostly due to a de novo variant [2]. The proportion of de novo variants varies between series: 20–35% in the European cohorts, more than half in the Chinese series of 40 patients (52.5%) and 95% in Japanese patients with the severe subtype known as Epstein syndrome [1, 46, 74, 76]. The French PNDS indicates about 30% de novo cases [33]. Each child of a patient has a 50% probability of inheriting the variant, regardless of sex [2]. The risk to the proband's siblings is 50% if a parent is affected or a carrier, and a sibling who inherits the variant may have a phenotype different from that of the proband; if the variant is not detected in the leucocyte DNA of either parent, or if the parents have a normal haematological study, the risk is low but higher than that of the population, owing to the possibility of germline mosaicism [2]. The risk to other relatives depends on the status of the proband's parents [2].

In DIAPH1-related disease, inheritance is also autosomal dominant, with a 50% risk for each child; biallelic loss-of-function variants in the same gene cause SCBMS, which is recessive, with a 25% risk for each pregnancy of a carrier couple (Section 6.6.6) [6, 99].

8.2. Parental testing and mosaicism

When neither parent is known to be affected, molecular testing for the variant identified in the proband is recommended and, if the variant is not known or somatic mosaicism is suspected in a parent with a mild phenotype, a haematological study (platelet count and size and distribution of myosin-9 in neutrophils) [2]. If the variant is not present in either parent, the possibilities are a de novo variant (confirmed if biological parentage has been verified; “presumed de novo” if not) or germline mosaicism, or somatic and germline mosaicism, in a parent [2]. Leucocyte DNA may not detect all somatic mosaicism and does not detect a variant confined to the germ cells [2]. Mosaic individuals may have minimal phenotypes: apparently healthy parents with 14% or 24% of neutrophils with aggregates and normal counts, and a father with only mild macrothrombocytopenia [2, 58–60]. An apparently negative family history can only be accepted after molecular testing of both parents and/or an adequate haematological study, because relatives with mild forms often go unnoticed [2].

8.3. Cascade testing

Clarifying the status of all first-degree relatives is recommended, to establish appropriate follow-up, treatment and surveillance and awareness of the agents and circumstances to avoid [2]. The study includes a smear with a search for platelet macrocytosis, a platelet count and immunofluorescence for myosin-9 aggregates and, if the familial variant is known, molecular testing [2]. The German guideline recommends molecular genetics to identify heterozygous carriers of inherited platelet disorders (consensus, 78%) [25]. Because the thrombocytopenia is congenital and the disease has immediate implications (for example, avoiding acetylsalicylic acid and preventing misdiagnosis as ITP), testing of children at risk has direct clinical benefit and may begin with the haematological study, which requires only a full blood count and smear [2]. Identification of the variant allows the prognostic group to be assigned to each relative, which guides the frequency of renal surveillance [2, 23]. The French PNDS considers segregation analysis necessary at least in first-degree relatives, with platelet count and size measurement [33].

8.4. Reproductive planning

The appropriate time to determine genetic risk and discuss prenatal diagnosis and preimplantation genetic testing is before pregnancy, and genetic counselling should be offered to young adults who are affected or at risk [2]. Once the familial variant has been identified, prenatal diagnosis and preimplantation genetic testing are possible; views on their use may differ between professionals and families, and the decision is personal [2]. In France, the request is discussed in a CPDPN (Section 5.8), after a genetic counselling consultation that explains the inheritance, the recurrence risk and the extra-haematological manifestations expected with the identified variant [33]. Counselling should take into account the great intrafamilial variability of the manifestations and the prognostic value of the genotype: a child who inherits an Arg702 variant will, with high probability, have severe nephropathy and hearing loss before 40 years, whereas a child with a truncating variant of the non-helical tailpiece will probably have thrombocytopenia as the only relevant manifestation [23]. DNA banking of affected patients may be considered [2]. In the pregnancy of an affected woman, the fetus has a 50% probability of being affected and, if so, will have thrombocytopenia similar to that of the mother, which weighs on delivery planning (Section 7.10) [77]. In MALTA syndrome, the family with compound heterozygosity and an asymptomatic brother carrying the same variants shows that counselling should consider recessive inheritance and incomplete penetrance [4].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Prevalence

MYH9-RD is considered the most frequent form of inherited thrombocytopenia, but it is rare [34]. The Italian Registry, with 225 patients, indicates a prevalence in Italy of at least 3.75 per 1,000,000; this is a minimum, because mild forms are discovered by chance and severe forms are mistaken for other diseases [2]. Orphanet classifies the prevalence in the class of 1–9 per 1,000,000, with mean values of 0.29 per 100,000 in Italy and 0.3 per 100,000 in Europe, derived from the same registry [13]. Two independent estimates point to higher values: in the ExAC (Exome Aggregation Consortium) exome database, six MYH9 loss-of-function variants in about 120,000 alleles correspond to an allele frequency of about 1 in 20,000, and in the Greifswald region (Germany) five families were identified in a population of about 250,000 inhabitants, which, with at least two affected individuals per family, gives about 1 in 25,000 [19]. The two approaches measure different things (carriers of variants in the population and diagnosed cases), are not reconciled in the literature and should not be combined into an average [7]. The prevalence of inherited thrombocytopenias as a whole was estimated in Italy at 2.7 per 100,000, from a registry of patients with thrombocytopenia [28]. GeneReviews finds no evidence of variation in prevalence between populations [2].

9.2. Data by region

Europe. Natural history data come mainly from the Italian Registry (255 patients from 121 families up to 2012), the French network (109 patients, 91.7% of European origin, with Maghrebi, Kanak and Malagasy patients) and a Romanian paediatric cohort, in which MYH9 was the most frequent gene among inherited thrombocytopenias with a molecular diagnosis (8 of 31) [23, 47, 85]. In three carriers of p.(Glu1841Lys), two with Fechtner syndrome and one with May-Hegglin anomaly, a shared haplotype suggests a common ancestor [41].

Portugal. The sources consulted provide no Portuguese epidemiological data. The only Portuguese series identified is that of the Congenital Coagulopathies Centre of the Centro Hospitalar Universitário de Santo António, in Porto, with three patients with MYH9-RD prepared with eltrombopag for surgery [81]. Eltrombopag is available in Portugal as Revolade® and as several generics [116].

Brazil. A Brazilian family with DFNA17 has been described in which ten members had progressive autosomal dominant hearing loss at all frequencies, caused by p.(Arg705His) [67]. A group from Pernambuco and São Paulo published in 2026 a description of elliptical giant platelets in MYH9-RD [123]. Eltrombopag olamine is registered with ANVISA as Revolade®, generics and a similar product [117]. The sources consulted provide no Brazilian prevalence data.

North America and Oceania. DFNA17 has been described in a five-generation American family and in an Australian family of Anglo-Celtic origin [3]. In a diagnostic programme in Australia and New Zealand, MYH9-RD was the cause in 12.4% of uncharacterised presumably inherited thrombocytopenias, and almost no case had been suspected [75].

Asia. In Japan, a national survey identified 177 patients with MYH9-RD, 40 with the severe subtype [46], and in a genetic hearing loss screening programme the disease or DFNA17 accounted for 0.11% of probands [20]. In China, recent series show a higher proportion of de novo variants and long diagnostic delays in adults [74, 76]. In Korea, a series of hearing loss due to MYH9 found no syndromic manifestations [44].

9.3. Special populations and gaps

Women of childbearing age, pregnant women, children with high-risk genotypes, kidney transplant candidates and patients needing surgery are the groups with specific needs, covered in sections 7 and 11. There are no data allowing the prevalence in the Portuguese or Brazilian population to be estimated, and the absence of published cases does not mean absence of the disease. The genetic estimate based on ExAC counts only loss-of-function variants, whereas most pathogenic variants are missense, which limits its interpretation [19].

9.4. Reference networks and patient associations

In the European Union, the reference network for rare haematological diseases is ERN-EuroBloodNet, which in 2026 devoted a series of webinars to inherited platelet disorders [30]. Nephropathy falls under ERKNet and cataracts under ERN-EYE [31, 32]. The Italian MYH9-RD Registry, based at the IRCCS Policlinico San Matteo in Pavia, enrols patients and supports diagnosis [2, 124]. In the United States, the Platelet Disorder Support Association supports patients with platelet disorders, and GeneReviews also lists associations for deaf people, kidney patients and eye diseases [2, 125].

In France, diagnosis and treatment are coordinated by the Reference Centre for Constitutional Platelet Disorders (CRPP), with its coordinating site in Marseille (Hôpital de la Timone) and constituent sites in Bordeaux, Paris (Hôpital Armand-Trousseau) and Toulouse, and by 30 resource and competence centres for constitutional bleeding disorders (CRC-MHC), including Réunion and Martinique, brought together in the MHEMO rare disease health network (filière), created in December 2017 [33]. The Association Française des Hémophiles (AFH), with a committee dedicated to platelet disorders, informs and supports patients, takes part in therapeutic education and organises meetings between patients and professionals [33]. At diagnosis, the specialist issues the Ministry of Health care and emergency card, with the type of disease, the emergency treatments and the team's contact details, and possibly a follow-up booklet [33]. MYH9-RD is on the list of long-term conditions (ALD 30) with exemption from patient co-payment; up to 16 years of age the request may be made by the specialist and thereafter by the general practitioner, and in mildly symptomatic forms, without chronic treatment or extra-haematological manifestations, the declaration is decided with the patient [33]. The general practitioner is responsible for renewing haemostatic treatments, screening for iron deficiency, keeping the vaccination schedule up to date, education about bleeding risk and emergency measures, coordinating the multidisciplinary follow-up of extra-haematological manifestations, psychological support and the PAI [33].

10. Illustrative Clinical Case

The case was published in 2025 by Nukala and colleagues [79]. A young woman born in the Caribbean was diagnosed with ITP in childhood and treated successively with corticosteroids, rituximab, chemotherapeutic agents and splenectomy, without response. In childhood, sensorineural hearing loss appeared, attributed by her doctors to toxicity of the drugs used for ITP. In adolescence, by then in the United States, she developed arterial hypertension and, months later, was admitted for macroscopic haematuria after streptococcal pharyngitis: she had 34 × 10⁹/L platelets, an immature platelet fraction of 64% (normal 0.9–7.9%), proteinuria greater than 400 mg/dL on a spot sample, creatinine of 2.16 mg/dL and a glomerular filtration rate of 37 mL/min, with normal antinuclear antibodies, antineutrophil cytoplasmic antibodies and complement fractions C3 and C4 [79]. Renal biopsy showed 14 glomeruli with global or near-global sclerosis and tubular atrophy with interstitial fibrosis in 65–70% of the cortex; on electron microscopy, the basement membrane was of normal thickness and podocyte foot process effacement involved 65% of the capillary surface, with microvillous transformation, and the diagnosis was secondary FSGS with collapsing features. Five months later she started haemodialysis [79].

Months later, at an emergency visit for fluid overload, the counter reported 13 × 10⁹/L platelets. The smear showed more than 50% giant platelets and 6 to 8 platelets per oil-immersion field, corresponding to a true count of at least 60 × 10⁹/L, faint cytoplasmic inclusions in neutrophils and Pappenheimer bodies, characteristic of asplenia (Figure 2 and Figure 3 of the article, available at https://pmc.ncbi.nlm.nih.gov/articles/PMC12198488/figure/F2/ and https://pmc.ncbi.nlm.nih.gov/articles/PMC12198488/figure/F3/) [79]. There was no family history of haematological or kidney disease, and no petechiae, organomegaly or bleeding from the dialysis catheter. The combination of giant platelets, leucocyte inclusions, early renal failure and early hearing loss led to next-generation sequencing, which identified the heterozygous pathogenic variant MYH9 c.2105G>A, p.(Arg702His) [79]. The differential considered ITP, Alport syndrome, grey platelet syndrome, thrombotic thrombocytopenic purpura, haemolytic uraemic syndrome and systemic lupus erythematosus, excluded on clinical and laboratory grounds [79].

The planned corticosteroids and immunoglobulin were abandoned. The patient and her mother received counselling on the nature of the disease and on testing relatives with thrombocytopenia; she was informed that desmopressin, tranexamic acid and platelet transfusion could be used before invasive procedures, cochlear implantation and ophthalmological surveillance were emphasised and she was encouraged to consider kidney transplantation [79]. The case brings together the errors that the series describe: the diagnosis of ITP despite giant platelets, immunosuppressive treatments and splenectomy without benefit, attribution of the hearing loss to another cause and delay in diagnosing the nephropathy [2, 47]. The course corresponds to that predicted for Arg702 variants, with hearing loss and ESKD in the second decade [23, 46]. The report does not describe molecular testing of the parents, so a de novo origin was not demonstrated [79]. The management proposed in the report is general and does not replace the individual assessment of bleeding risk described in section 7.

11. Prognosis and Long-Term Follow-up

11.1. Prognosis

The prognosis depends on the genotype and on the prevention of haemorrhagic and iatrogenic complications. Life expectancy has not been studied in the cohorts consulted. Thrombocytopenia remains stable throughout life and life-threatening haemorrhages are rare, but excessive surgical bleeding occurred in 15.5% of procedures in patients with MYH9-RD [2, 108]. The main cause of severe morbidity is nephropathy: among patients with nephropathy in the Italian registry, 66% had chronic renal failure and 43% ESKD after a median follow-up of 36 months, with a rate of progression to ESKD of 6.79 per 100 person-years, and, among those who reached ESKD, the age was 24 years (standard deviation 9) with Arg702 variants and 42 years (standard deviation 23) with coiled-coil variants [23]. Hearing loss interferes with daily activities in 90% of patients with an abnormal audiogram, but cochlear implantation restores verbal communication in most patients with profound hearing loss [23, 66]. Elevated liver enzymes did not progress to liver failure in any patient described [2, 64]. Patients with truncating variants of the non-helical tailpiece, p.(Asp1424Asn) or p.(Glu1841Lys) have a high probability of keeping thrombocytopenia as their only relevant manifestation [2, 23].

11.2. Surveillance

Table 11.1 reproduces the GeneReviews surveillance protocol, with the action thresholds described in the sources. Haematological surveillance is aimed mainly at patients with moderate or severe thrombocytopenia, and organ screening at patients not yet being treated for that manifestation [2].

Table 11.1. Surveillance in MYH9-RD [2, 23, 28, 43, 57, 77, 108]

The intervals in the table coincide with those of section 7: haematological and renal surveillance at least annually (six-monthly renal in high-risk genotypes) and hearing, ophthalmological and hepatic surveillance every 3 years according to GeneReviews, with the alternative of 3 to 5 years proposed in 2024 [2, 43]. The French PNDS lengthens the hearing intervals in adults without hearing loss (5 to 10 years), shortens them in childhood (2 to 3 years) and after the diagnosis of hearing loss (annually) and makes the ophthalmological check-up annual from the age of 20; it sets no interval for proteinuria screening, which it requires in all patients during regular follow-up [33]. In DIAPH1-related disease, without nephropathy or cataracts, surveillance is audiological and haematological, with audiological assessment at least annually because the hearing loss is progressive (Table 6.3) [6, 18].

12. Summary and Final Considerations

MYH9-related disease brings together in a single entity May-Hegglin anomaly, Sebastian, Fechtner and Epstein syndromes and, for most authors, DFNA17. It is caused by heterozygous MYH9 variants with a dominant-negative effect on non-muscle myosin IIA, and about one third of cases arise without a family history. All patients have, from birth, giant platelets and myosin-9 aggregates in neutrophils, and almost all have thrombocytopenia; over a lifetime, most develop sensorineural hearing loss, half benign elevation of liver enzymes, a quarter proteinuric nephropathy that often progresses to end-stage kidney disease and about one fifth cataracts. The genotype predicts the course: variants of arginine 702 are associated with severe thrombocytopenia and with nephropathy and hearing loss before 40 years, and truncating tail variants usually leave thrombocytopenia as the only manifestation.

Diagnosis depends on counting and sizing platelets under the microscope, because counters underestimate count and size, on looking for myosin-9 aggregates by immunofluorescence, which has a sensitivity close to 100% and can be performed on smears sent by post, and on sequencing MYH9, preferably in an inherited thrombocytopenia panel. The most frequent error is the diagnosis of immune thrombocytopenia, which has led to ineffective corticosteroids, immunoglobulin and splenectomy in high proportions of patients in Italy, France and China.

Treatment is supportive and multidisciplinary: local measures and antifibrinolytics for mild bleeding, HLA-matched platelet transfusion limited to severe bleeding and to prophylaxis when necessary, and off-label eltrombopag, at 50 mg daily increased to 75 mg if necessary for 3 to 6 weeks, to prepare for surgery and control bleeding, with responses in almost all patients studied and phase II evidence. Proteinuria should be looked for at least annually, or every six months in high-risk genotypes, so that renin-angiotensin system blockade can be started early; cochlear implantation restores hearing in most patients with profound hearing loss; dialysis and kidney transplantation are the options in end-stage disease. Pregnancy requires delivery planning according to the count. The 2021 French PNDS, the only specific guideline, adds its own surgical and anaesthetic thresholds, doses for menorrhagia, iron deficiency and neonatal transfusion, follow-up of pregnancy and the newborn and the organisation of care in reference and competence centres. Genetic counselling rests on the 50% risk to offspring, the possibility of parental mosaicism and the prognostic value of the genotype, which also guides the surveillance of relatives.

DIAPH1-related disease is the main differential of MYH9-RD with hearing loss: it is transmitted in an autosomal dominant manner by gain-of-function variants that truncate the autoregulatory domain of the formin DIA1, and combines congenital or early-onset progressive sensorineural hearing loss, mild progressive macrothrombocytopenia, frequent mild neutropenia and mild bleeding, without neutrophil inclusions, nephropathy or cataracts. The full blood count should be part of the assessment of all autosomal dominant hearing loss, normal myosin-9 immunofluorescence and genetics separate the two diseases, and short-term eltrombopag allowed surgery without transfusion in one patient [6, 27].

13. References

1. Online Mendelian Inheritance in Man (OMIM). #155100 Macrothrombocytopenia and granulocyte inclusions with or without nephritis or sensorineural hearing loss; MATINS [Internet]. Baltimore: Johns Hopkins University; 2026 [edition of 21/04/2026]. Available from: https://omim.org/entry/155100
2. Savoia A, Pecci A. MYH9-Related Disease. 2008 Nov 20 [updated 2021 Feb 18]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK2689/
3. Online Mendelian Inheritance in Man (OMIM). #603622 Deafness, autosomal dominant 17; DFNA17 [Internet]. Baltimore: Johns Hopkins University; 2018 [edition of 14/06/2018]. Available from: https://omim.org/entry/603622
4. Hanlon E, Brown KL, Michel HM, Roby C, Urbano MG, Mahmutovic D, et al. Autosomal recessive MALTA syndrome with compound heterozygous MYH9 variants: Structural modeling reveals disrupted myosin-9 assembly. JID Innov. 2026;6(5):100507. doi:10.1016/j.xjidi.2026.100507. PMID: 42604168. PMC13476708.
5. Orphanet. DIAPH1-related sensorineural hearing loss-thrombocytopenia syndrome. ORPHA:494444 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/494444
6. Westbury SK, Downes K, Burney C, Lozano ML, Obaji SG, Toh CH, et al. Phenotype description and response to thrombopoietin receptor agonist in DIAPH1-related disorder. Blood Adv. 2018;2(18):2341-6. doi:10.1182/bloodadvances.2018020370. PMID: 30232087. PMC6156883.
7. GeneGuidance. MYH9 — gene and condition curation [local geneguidance-FULL database, extraction of 2026-09-26].
8. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348. PMID: 28967166. PMC5765404. Service: https://variantvalidator.org (accessed 31/10/2026).
9. Lalwani AK, Goldstein JA, Kelley MJ, Luxford W, Castelein CM, Mhatre AN. Human nonsyndromic hereditary deafness DFNA17 is due to a mutation in nonmuscle myosin MYH9. Am J Hum Genet. 2000;67(5):1121-8. doi:10.1016/S0002-9297(07)62942-5. PMID: 11023810. PMC1288554.
10. National Center for Biotechnology Information. ClinVar: pathogenic and likely pathogenic variants of DIAPH1 (NM_005219.5) [Internet]. Bethesda (MD): National Library of Medicine; 2026 [accessed 01/11/2026]. Available from: https://www.ncbi.nlm.nih.gov/clinvar/?term=DIAPH1%5Bgene%5D
11. Kaustio M, Nayebzadeh N, Hinttala R, Tapiainen T, Åström P, Mamia K, et al. Loss of DIAPH1 causes SCBMS, combined immunodeficiency, and mitochondrial dysfunction. J Allergy Clin Immunol. 2021;148(2):599-611. doi:10.1016/j.jaci.2020.12.656. PMID: 33662367.
12. Orphanet. Progressive microcephaly-seizures-cortical blindness-developmental delay syndrome. ORPHA:477814 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/477814
13. Orphanet. MYH9-related syndromic thrombocytopenia. ORPHA:182050 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/182050
14. Orphanet. May-Hegglin thrombocytopenia. ORPHA:850 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/850
15. Orphanet. Sebastian syndrome. ORPHA:807 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/807
16. Orphanet. Fechtner syndrome. ORPHA:1984 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/1984
17. Orphanet. Epstein syndrome. ORPHA:1019 [Internet]. Paris: INSERM; 2026. Available from: https://www.orpha.net/en/disease/detail/1019
18. Shearer AE, Hildebrand MS, Odell AM, Smith RJH. Genetic Hearing Loss Overview. 1999 Feb 14 [updated 2023 Apr 6]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1434/
19. Fernandez-Prado R, Carriazo-Julio SM, Torra R, Ortiz A, Perez-Gomez MV. MYH9-related disease: it does exist, may be more frequent than you think and requires specific therapy. Clin Kidney J. 2019;12(4):488-493. doi:10.1093/ckj/sfz103. PMID: 31384439. PMC6671427.
20. Goto S, Sasaki A, Nishio SY, Shinkawa C, Oda K, Wada T, et al. The Clinical Details of MYH9-Related Disease and DFNA17 in a Large Japanese Hearing Loss Cohort. Genes (Basel). 2026;17(2). doi:10.3390/genes17020154. PMID: 41751538. PMC12941242.
21. Ueyama T, Ninoyu Y, Nishio SY, Miyoshi T, Torii H, Nishimura K, et al. Constitutive activation of DIA1 (DIAPH1) via C-terminal truncation causes human sensorineural hearing loss. EMBO Mol Med. 2016;8(11):1310-24. doi:10.15252/emmm.201606609. PMID: 27707755. PMC5090661.
22. González-Aguado R, Veiga-Alonso A, Onecha-De La Fuente E, Morales-Angulo C. Exploring DIAPH1 gene mutations in patients with sensorineural hearing loss of unknown etiology in Northern Spain. Acta Otorrinolaringol Esp (Engl Ed). 2025;76(5):512269. doi:10.1016/j.otoeng.2025.512269. PMID: 40651568.
23. Pecci A, Klersy C, Gresele P, Lee KJ, De Rocco D, Bozzi V, et al. MYH9-related disease: a novel prognostic model to predict the clinical evolution of the disease based on genotype-phenotype correlations. Hum Mutat. 2014;35(2):236-47. doi:10.1002/humu.22476. PMID: 24186861. PMC6233870.
24. Kurasawa S, Ganaha A, Ayabe S, Yoshiki A, Kawama F, Kitayama S, et al. Hearing loss occurs prior to thrombocytopenia in both mice and humans with DFNA1. FASEB J. 2025;39(2):e70309. doi:10.1096/fj.202402118R. PMID: 39831886.
25. Gebetsberger J, Knöfler R, Streif W. "Diagnosis of Inherited Platelet Disorders": Update of the Interdisciplinary S2k-Guideline of the Permanent Pediatric Commission of the Society of Thrombosis and Haemostasis Research (GTH e.V.). Hamostaseologie. 2025;45(4):347-354. doi:10.1055/a-2628-5488. PMID: 40795927.
26. Verver E, Pecci A, De Rocco D, Ryhänen S, Barozzi S, Kunst H, et al. R705H mutation of MYH9 is associated with MYH9-related disease and not only with non-syndromic deafness DFNA17. Clin Genet. 2015;88(1):85-9. doi:10.1111/cge.12438. PMID: 24890873.
27. Neuhaus C, Lang-Roth R, Zimmermann U, Heller R, Eisenberger T, Weikert M, et al. Extension of the clinical and molecular phenotype of DIAPH1-associated autosomal dominant hearing loss (DFNA1). Clin Genet. 2017;91(6):892-901. doi:10.1111/cge.12915. PMID: 27808407.
28. Zaninetti C, Gresele P, Bertomoro A, Klersy C, De Candia E, Veneri D, et al. Eltrombopag for the treatment of inherited thrombocytopenias: a phase II clinical trial. Haematologica. 2020;105(3):820-828. doi:10.3324/haematol.2019.223966. PMID: 31273088. PMC7049343.
29. Hildebrand MS, de Silva MG, Gardner RJ, Rose E, de Graaf CA, Bahlo M, et al. Cochlear implants for DFNA17 deafness. Laryngoscope. 2006;116(12):2211-5. doi:10.1097/01.mlg.0000242089.72880.f8. PMID: 17146397.
30. ERN-EuroBloodNet — European Reference Network on Rare Hematological Diseases [Internet]. 2026 [accessed 31/10/2026]. Available from: https://eurobloodnet.eu/
31. ERKNet — European Rare Kidney Disease Reference Network [Internet]. 2026 [accessed 31/10/2026]. Available from: https://www.erknet.org/
32. ERN-EYE — European Reference Network for Rare Eye Diseases [Internet]. 2026 [accessed 31/10/2026]. Available from: https://www.ern-eye.eu/
33. Haute Autorité de Santé. Protocole National de Diagnostic et de Soins (PNDS): Syndrome MYH9. Saint-Denis La Plaine: HAS; 17 Nov 2021 (text of October 2021; Centre de Référence des Pathologies Plaquettaires Constitutionnelles, filière MHEMO). [https://www.has-sante.fr/jcms/p_3298133](https://www.has-sante.fr/jcms/p_3298133). Accessed 6 Nov 2026.
34. Pecci A, Ma X, Savoia A, Adelstein RS. MYH9: Structure, functions and role of non-muscle myosin IIA in human disease. Gene. 2018;664:152-167. doi:10.1016/j.gene.2018.04.048. PMID: 29679756. PMC5970098.
35. Hegglin R. Simultaneous constitutional changes in neutrophils and platelets [in German]. Helv Med Acta. 1945;12:439. PMID: 21009925.
36. Epstein CJ, Sahud MA, Piel CF, Goodman JR, Bernfield MR, Kushner JH, et al. Hereditary macrothrombocytopathia, nephritis and deafness. Am J Med. 1972;52(3):299-310. doi:10.1016/0002-9343(72)90017-4. PMID: 5011389.
37. Peterson LC, Rao KV, Crosson JT, White JG. Fechtner syndrome--a variant of Alport's syndrome with leukocyte inclusions and macrothrombocytopenia. Blood. 1985;65(2):397-406. doi:10.1182/blood.v65.2.397.bloodjournal652397. PMID: 2981587.
38. Greinacher A, Nieuwenhuis HK, White JG. Sebastian platelet syndrome: a new variant of hereditary macrothrombocytopenia with leukocyte inclusions. Blut. 1990;61(5):282-8. doi:10.1007/BF01732878. PMID: 2176899.
39. Seri M, Cusano R, Gangarossa S, Caridi G, Bordo D, Lo Nigro C, et al. Mutations in MYH9 result in the May-Hegglin anomaly, and Fechtner and Sebastian syndromes. The May-Heggllin/Fechtner Syndrome Consortium. Nat Genet. 2000;26(1):103-5. doi:10.1038/79063. PMID: 10973259.
40. Kelley MJ, Jawien W, Ortel TL, Korczak JF. Mutation of MYH9, encoding non-muscle myosin heavy chain A, in May-Hegglin anomaly. Nat Genet. 2000;26(1):106-8. doi:10.1038/79069. PMID: 10973260.
41. Heath KE, Campos-Barros A, Toren A, Rozenfeld-Granot G, Carlsson LE, Savige J, et al. Nonmuscle myosin heavy chain IIA mutations define a spectrum of autosomal dominant macrothrombocytopenias: May-Hegglin anomaly and Fechtner, Sebastian, Epstein, and Alport-like syndromes. Am J Hum Genet. 2001;69(5):1033-45. doi:10.1086/324267. PMID: 11590545. PMC1274350.
42. Seri M, Pecci A, Di Bari F, Cusano R, Savino M, Panza E, et al. MYH9-related disease: May-Hegglin anomaly, Sebastian syndrome, Fechtner syndrome, and Epstein syndrome are not distinct entities but represent a variable expression of a single illness. Medicine (Baltimore). 2003;82(3):203-15. doi:10.1097/01.md.0000076006.64510.5c. PMID: 12792306.
43. Shen K, Chen T, Xiao M. MYH9-related inherited thrombocytopenia: the genetic spectrum, underlying mechanisms, clinical phenotypes, diagnosis, and management approaches. Res Pract Thromb Haemost. 2024;8(6):102552. doi:10.1016/j.rpth.2024.102552. PMID: 39309229. PMC11415342.
44. Bae SH, Joo SY, Jang SH, Kim SH, Choi JY, Won D, et al. The audiological phenotype of patients with a variant in MYH9 and MYH14 genes. Sci Rep. 2025;15(1):22324. doi:10.1038/s41598-025-08801-w. PMID: 40596561. PMC12219520.
45. Kunishima S, Matsushita T, Kojima T, Sako M, Kimura F, Jo EK, et al. Immunofluorescence analysis of neutrophil nonmuscle myosin heavy chain-A in MYH9 disorders: association of subcellular localization with MYH9 mutations. Lab Invest. 2003;83(1):115-22. doi:10.1097/01.lab.0000050960.48774.17. PMID: 12533692.
46. Shinno K, Kunishima S, Sakamoto A, Uchiyama T, Ishiguro A. Natural history of the severe subtype of MYH9-related disease (Epstein syndrome). Eur J Intern Med. 2025;138:96-100. doi:10.1016/j.ejim.2025.05.019. PMID: 40441958.
47. Saposnik B, Binard S, Fenneteau O, Nurden A, Nurden P, Hurtaud-Roux MF, et al. Mutation spectrum and genotype-phenotype correlations in a large French cohort of MYH9-Related Disorders. Mol Genet Genomic Med. 2014;2(4):297-312. doi:10.1002/mgg3.68. PMID: 25077172. PMC4113270.
48. Online Mendelian Inheritance in Man (OMIM). *160775 Myosin, heavy chain 9, nonmuscle; MYH9 [Internet]. Baltimore: Johns Hopkins University; 2026. Available from: https://omim.org/entry/160775
49. Llorente-González C, Mustafina K, Asensio-Juárez G, Garrido-Casado M, Talayero VC, Pérez-Díaz R, et al. MYH9 mutations differentially stabilize non-muscle myosin II filaments and induce distinct cellular aggregation phenotypes. Cell Mol Life Sci. 2026;83(1). doi:10.1007/s00018-026-06119-y. PMID: 41760963. PMC12961093.
50. Pollinger L, Greve JN, Grosch M, Kaliman S, Abuhattum S, Kräter M, et al. MYH9 Variant p.(Arg424Gly) Alters Nonmuscle Myosin IIA Contraction, Causing Atypical MYH9-related Disease. Kidney Int Rep. 2026;11(4):106343. doi:10.1016/j.ekir.2026.106343. PMID: 41799362. PMC12966671.
51. Pal K, Nowak R, Billington N, Liu R, Ghosh A, Sellers JR, et al. Megakaryocyte migration defects due to nonmuscle myosin IIA mutations underlie thrombocytopenia in MYH9-related disease. Blood. 2020;135(21):1887-1898. doi:10.1182/blood.2019003064. PMID: 32315395. PMC7243143.
52. Öftering P, Baumann J, Beck S, Greinacher A, Nieswandt B, Bender M. Blocking platelet glycoprotein V cleavage reduces bleeding in mouse models of MYH9-related disease. Haematologica. 2025;110(12):3099-3103. doi:10.3324/haematol.2025.287542. PMID: 40501415. PMC12666290.
53. Sun S, Pan J, Zhang N, Jin X, Zhu TN, Kang Z, et al. The membrane skeleton density of red blood cells in MYH9-related disease patients is decreased. Biosci Rep. 2026;46(10). doi:10.1042/BSR20260386. PMID: 42751867. PMC13623819.
54. Kunishima S, Matsushita T, Hamaguchi M, Saito H. Identification and characterization of the first large deletion of the MYH9 gene associated with MYH9 disorders. Eur J Haematol. 2008;80(6):540-4. doi:10.1111/j.1600-0609.2008.01046.x. PMID: 18284620.
55. Inoue S, Nagano C, Matsuo M, Aoyama S, Kimura Y, Inoki Y, et al. Clinical and functional evaluation of non-missense MYH9 variants in MYH9-related disease. Clin Exp Nephrol. 2026;30(9):1324-1331. doi:10.1007/s10157-026-02893-y. PMID: 42189408.
56. Pecci A, Panza E, Pujol-Moix N, Klersy C, Di Bari F, Bozzi V, et al. Position of nonmuscle myosin heavy chain IIA (NMMHC-IIA) mutations predicts the natural history of MYH9-related disease. Hum Mutat. 2008;29(3):409-17. doi:10.1002/humu.20661. PMID: 18059020.
57. Verver EJ, Topsakal V, Kunst HP, Huygen PL, Heller PG, Pujol-Moix N, et al. Nonmuscle Myosin Heavy Chain IIA Mutation Predicts Severity and Progression of Sensorineural Hearing Loss in Patients With MYH9-Related Disease. Ear Hear. 2016;37(1):112-20. doi:10.1097/AUD.0000000000000198. PMID: 26226608.
58. Kunishima S, Matsushita T, Yoshihara T, Nakase Y, Yokoi K, Hamaguchi M, et al. First description of somatic mosaicism in MYH9 disorders. Br J Haematol. 2005;128(3):360-5. doi:10.1111/j.1365-2141.2004.05323.x. PMID: 15667538.
59. Kunishima S, Kitamura K, Matsumoto T, Sekine T, Saito H. Somatic mosaicism in MYH9 disorders: the need to carefully evaluate apparently healthy parents. Br J Haematol. 2014;165(6):885-7. doi:10.1111/bjh.12797. PMID: 24611568.
60. Gresele P, De Rocco D, Bury L, Fierro T, Mezzasoma AM, Pecci A, et al. Apparent genotype-phenotype mismatch in a patient with MYH9-related disease: when the exception proves the rule. Thromb Haemost. 2013;110(3):618-20. doi:10.1160/TH13-02-0175. PMID: 23925420.
61. Kunishima S, Takaki K, Ito Y, Saito H. Germinal mosaicism in MYH9 disorders: a family with two affected siblings of normal parents. Br J Haematol. 2009;145(2):260-2. doi:10.1111/j.1365-2141.2009.07584.x. PMID: 19208103.
62. Fewings E, Ziemer M, Hörtnagel K, Reicherter K, Larionov A, Redman J, et al. Malta (MYH9 Associated Elastin Aggregation) Syndrome: Germline Variants in MYH9 Cause Rare Sweat Duct Proliferations and Irregular Elastin Aggregations. J Invest Dermatol. 2019;139(10):2238-2241.e6. doi:10.1016/j.jid.2019.03.1151. PMID: 31125547.
63. Konomoto T, Wakamatsu F, Sakaguchi H, Kurogi J, Tanaka E, Moritake H. Re-evaluating the MYH9 p.I1816V variant in a patient with atypical clinical presentation. Pediatr Nephrol. 2026;41(4):993-997. doi:10.1007/s00467-025-07059-8. PMID: 41243005. PMC12953253.
64. Pecci A, Biino G, Fierro T, Bozzi V, Mezzasoma A, Noris P, et al. Alteration of liver enzymes is a feature of the MYH9-related disease syndrome. PLoS One. 2012;7(4):e35986. doi:10.1371/journal.pone.0035986. PMID: 22558294. PMC3338476.
65. Noris P, Biino G, Pecci A, Civaschi E, Savoia A, Seri M, et al. Platelet diameters in inherited thrombocytopenias: analysis of 376 patients with all known disorders. Blood. 2014;124(6):e4-e10. doi:10.1182/blood-2014-03-564328. PMID: 24990887. PMC4126341.
66. Pecci A, Verver EJ, Schlegel N, Canzi P, Boccio CM, Platokouki H, et al. Cochlear implantation is safe and effective in patients with MYH9-related disease. Orphanet J Rare Dis. 2014;9:100. doi:10.1186/1750-1172-9-100. PMID: 24980457. PMC4105151.
67. Dantas VG, Lezirovitz K, Yamamoto GL, Moura de Souza CF, Ferreira SG, Mingroni-Netto RC. c.G2114A MYH9 mutation (DFNA17) causes non-syndromic autosomal dominant hearing loss in a Brazilian family. Genet Mol Biol. 2014;37(4):616-21. doi:10.1590/S1415-47572014005000025. PMID: 25505834. PMC4261959.
68. Sekine T, Konno M, Sasaki S, Moritani S, Miura T, Wong WS, et al. Patients with Epstein-Fechtner syndromes owing to MYH9 R702 mutations develop progressive proteinuric renal disease. Kidney Int. 2010;78(2):207-14. doi:10.1038/ki.2010.21. PMID: 20200500.
69. Nakatani R, Miura K, Shirai Y, Ohtsuka Y, Ohwada Y, Konomoto T, et al. Kidney pathological findings of MYH9-related disease: a cross-sectional nationwide survey in Japan. Pediatr Nephrol. 2025;40(10):3201-3209. doi:10.1007/s00467-025-06802-5. PMID: 40372468.
70. Favier R, DiFeo A, Hezard N, Fabre M, Bedossa P, Martignetti JA. A new feature of the MYH9-related syndrome: chronic transaminase elevation. Hepatology. 2013;57(3):1288-9. doi:10.1002/hep.25913. PMID: 22806255.
71. Spinelli A, Parrens M, Valmary-Degano S. Littoral cell angioma of the spleen and Epstein syndrome [in French]. Ann Pathol. 2025;45(5):438-442. doi:10.1016/j.annpat.2025.03.003. PMID: 40753045.
72. Pannunzio A, Palumbo IM, Baldacci E, Faleschini M, Ammeti D, Ligia S, et al. A Long Misdiagnosed MYH9-Related Disease With Both Bleeding and Thrombotic Manifestations: A Case Report. Haemophilia. 2026. doi:10.1111/hae.70384. PMID: 42631536.
73. Zaninetti C. Thrombopoietin receptor agonists in inherited thrombocytopenias: an evolving mosaic of evidence. Res Pract Thromb Haemost. 2026;10(6):106907. doi:10.1016/j.rpth.2026.106907. PMID: 42703607. PMC13546829.
74. Chen W, Arif AR, Ding Y, Wang L, Cai L, Arif M, et al. A proactive large platelet-driven screening strategy for the early recognition of MYH9-related disease: a Chinese prospective cohort study. Res Pract Thromb Haemost. 2026;10(5):106824. doi:10.1016/j.rpth.2026.106824. PMID: 42598527. PMC13471207.
75. Rabbolini DJ, Chun Y, Latimer M, Kunishima S, Fixter K, Valecha B, et al. Diagnosis and treatment of MYH9-RD in an Australasian cohort with thrombocytopenia. Platelets. 2018;29(8):793-800. doi:10.1080/09537104.2017.1356920. PMID: 29090586.
76. Arif AR, Zhou H, Fang Y, Cheng Y, Ye J, Chen W, et al. Clinical and genetic characteristics of 40 patients with nonmuscle myosin heavy chain 9-related disease (MYH9-RD) misdiagnosed as immune thrombocytopenia: a retrospective analysis in China. J Thromb Haemost. 2025;23(3):1043-1051. doi:10.1016/j.jtha.2024.12.001. PMID: 39672236.
77. Noris P, Schlegel N, Klersy C, Heller PG, Civaschi E, Pujol-Moix N, et al. Analysis of 339 pregnancies in 181 women with 13 different forms of inherited thrombocytopenia. Haematologica. 2014;99(8):1387-94. doi:10.3324/haematol.2014.105924. PMID: 24763399. PMC4116839.
78. Gresele P, Orsini S, Noris P, Falcinelli E, Alessi MC, Bury L, et al. Validation of the ISTH/SSC bleeding assessment tool for inherited platelet disorders: A communication from the Platelet Physiology SSC. J Thromb Haemost. 2020;18(3):732-739. doi:10.1111/jth.14683. PMID: 31750621.
79. Nukala S, Aparna F, Feliciano M, Withey JL. Thrombocytopenia, renal failure and hearing loss in a young patient: MYH9-related disorder. BMJ Case Rep. 2025;18(6). doi:10.1136/bcr-2024-262915. PMID: 40555542. PMC12198488.
80. De Girolamo M, Marrocchi A, Vitone F, Scapaticci M, Bovara M, Papillo TP, et al. Analytical discrepancy in platelet counting in MYH9-related disease: a case report. Lab Med. 2026;57(6). doi:10.1093/labmed/lmag078. PMID: 42880724. PMC13671753.
81. Oliveira M, Coutinho M, Cruz E, Morais S. Eltrombopag use in surgical procedures in inherited thrombocytopenia: expanding the role of thrombopoietin-receptor agonists. Res Pract Thromb Haemost. 2026;10(5):106861. doi:10.1016/j.rpth.2026.106861. PMID: 42620598. PMC13485643.
82. Greinacher A, Pecci A, Kunishima S, Althaus K, Nurden P, Balduini CL, et al. Diagnosis of inherited platelet disorders on a blood smear: a tool to facilitate worldwide diagnosis of platelet disorders. J Thromb Haemost. 2017;15(7):1511-1521. doi:10.1111/jth.13729. PMID: 28457011.
83. Savoia A, De Rocco D, Panza E, Bozzi V, Scandellari R, Loffredo G, et al. Heavy chain myosin 9-related disease (MYH9 -RD): neutrophil inclusions of myosin-9 as a pathognomonic sign of the disorder. Thromb Haemost. 2010;103(4):826-32. doi:10.1160/TH09-08-0593. PMID: 20174760.
84. Kitamura K, Yoshida K, Shiraishi Y, Chiba K, Tanaka H, Furukawa K, et al. Normal neutrophil myosin IIA localization in an immunofluorescence analysis can rule out MYH9 disorders. J Thromb Haemost. 2013;11(11):2071-3. doi:10.1111/jth.12406. PMID: 24106837.
85. Obrisca R, Serbanica A, Marcu A, Bica A, Jercan C, Avramescu I, et al. Clinical and Molecular Spectrum of MYH9-Thrombocytopenia: Insights from a Single Centric Pediatric Cohort. Children (Basel). 2025;12(11). doi:10.3390/children12111563. PMID: 41300679. PMC12651388.
86. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 36, 36.1 Disorders of Low Density Lipoprotein Metabolism (sitosterolemia).
87. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 43, 43.1 Congenital Disorders of Protein N-Glycosylation.
88. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 44, Disorders of Cellular Trafficking, 44.1.3 Other Types of Cellular Trafficking (Table 44.2).
89. Luo X, Cao K, Wang R, Wu W, Huang T, Zhang Y, et al. Novel biomarkers for the differential diagnosis of paediatric MYH9-related diseases and immune thrombocytopenia revealed by plasma proteomics. Br J Haematol. 2026;209(4):1598-1609. doi:10.1111/bjh.70720. PMID: 42509019.
90. Wang L, Cao K, Chen S, Fu X, Chen Y, Luo X. Analysis of Platelet Membrane Glycoproteins and The Specific Antibodies in MYH9-Related Diseases. Int J Gen Med. 2026;19:538585. doi:10.2147/IJGM.S538585. PMID: 41938364. PMC13045853.
91. Lynch ED, Lee MK, Morrow JE, Welcsh PL, León PE, King MC. Nonsyndromic deafness DFNA1 associated with mutation of a human homolog of the Drosophila gene diaphanous. Science. 1997;278(5341):1315-8. doi:10.1126/science.278.5341.1315. PMID: 9360932.
92. Stritt S, Nurden P, Turro E, Greene D, Jansen SB, Westbury SK, et al. A gain-of-function variant in DIAPH1 causes dominant macrothrombocytopenia and hearing loss. Blood. 2016;127(23):2903-14. doi:10.1182/blood-2015-10-675629. PMID: 26912466.
93. Mohseni M, Mohammadi Y, Zare Ashrafi F, Ghodratpour F, Jalalvand K, Arzhangi S, et al. An Extended Iranian Family with Autosomal Dominant Non-syndromic Hearing Loss Associated with A Nonsense Mutation in the DIAPH1 Gene. Arch Iran Med. 2023;26(3):176-80. doi:10.34172/aim.2023.27. PMID: 37543941. PMC10685723.
94. Kim BJ, Ueyama T, Miyoshi T, Lee S, Han JH, Park HR, et al. Differential disruption of autoinhibition and defect in assembly of cytoskeleton during cell division decide the fate of human DIAPH1-related cytoskeletopathy. J Med Genet. 2019;56(12):818-27. doi:10.1136/jmedgenet-2019-106282. PMID: 31473629.
95. Ganaha A, Kaname T, Shinjou A, Chinen Y, Yanagi K, Higa T, et al. Progressive macrothrombocytopenia and hearing loss in a large family with DIAPH1 related disease. Am J Med Genet A. 2017;173(10):2826-30. doi:10.1002/ajmg.a.38411. PMID: 28815995.
96. Rabbolini D, Liang HPH, Morel-Kopp MC, Connor D, Whittaker S, Dunkley S, et al. Building platelet phenotypes: Diaphanous-related formin 1 (DIAPH1)-related disorder. Platelets. 2022;33(3):432-42. doi:10.1080/09537104.2021.1937593. PMID: 34223798.
97. Feng Q, Jiang L, Zhang S, He C, Mei L, Liu Y. A novel frameshift mutation in the DIAPH1 gene causes a Chinese family autosomal dominant nonsyndromic hearing loss: Mutation in DIAPH1 causes hearing loss. Gene. 2025;936:149088. doi:10.1016/j.gene.2024.149088. PMID: 39542281.
98. Ercan-Sencicek AG, Jambi S, Franjic D, Nishimura S, Li M, El-Fishawy P, et al. Homozygous loss of DIAPH1 is a novel cause of microcephaly in humans. Eur J Hum Genet. 2015;23(2):165-72. doi:10.1038/ejhg.2014.82. PMID: 24781755. PMC4297910.
99. Al-Maawali A, Barry BJ, Rajab A, El-Quessny M, Seman A, Coury SN, et al. Novel loss-of-function variants in DIAPH1 associated with syndromic microcephaly, blindness, and early onset seizures. Am J Med Genet A. 2016;170A(2):435-40. doi:10.1002/ajmg.a.37422. PMID: 26463574. PMC5315085.
100. Galassi Deforie V, Maroofian R, Karagoz I, Godwin A, Al Sheikh E, Gestri G, et al. Recessive loss of DIAPH1 function causes a progressive neurodevelopmental syndrome with variable immunological involvement. Genet Med. 2026;28(5):102551. doi:10.1016/j.gim.2026.102551. PMID: 41860019.
101. Kim BJ, Miyoshi T, Chaudhry T, Friedman TB, Choi BY, Ueyama T. Late-onset hearing loss case associated with a heterozygous truncating variant of DIAPH1. Clin Genet. 2022;101(4):466-71. doi:10.1111/cge.14115. PMID: 35060117. PMC8981108.
102. Kundishora AJ, Peters ST, Pinard A, Duran D, Panchagnula S, Barak T, et al. DIAPH1 Variants in Non-East Asian Patients With Sporadic Moyamoya Disease. JAMA Neurol. 2021;78(8):993-1003. doi:10.1001/jamaneurol.2021.1681. PMID: 34125151. PMC8204259.
103. Karki NR, Ajebo G, Savage N, Kutlar A. DIAPH1 Mutation as a Novel Cause of Autosomal Dominant Macrothrombocytopenia and Hearing Loss. Acta Haematol. 2021;144(1):91-4. doi:10.1159/000506727. PMID: 32594080.
104. Alasmari BG, Alpakra M, Hassanien SS, Elmugadam AA, Elzubair L, Suliman EA, et al. A Novel Variant in the DIAPH1 Gene Causing Macrothrombocytopenia and Non-syndromic Hearing Loss in a Pediatric Saudi Girl. Cureus. 2024;16(5):e61044. doi:10.7759/cureus.61044. PMID: 38915998. PMC11195521.
105. Pecci A, Gresele P, Klersy C, Savoia A, Noris P, Fierro T, et al. Eltrombopag for the treatment of the inherited thrombocytopenia deriving from MYH9 mutations. Blood. 2010;116(26):5832-7. doi:10.1182/blood-2010-08-304725. PMID: 20844233.
106. European Medicines Agency. Revolade (eltrombopag): EPAR — product information (summary of product characteristics, PT) [Internet]. Amsterdam: EMA; authorisation of 11/03/2010, revision 39. EMEA/H/C/001110. Available from: https://www.ema.europa.eu/pt/documents/product-information/revolade-epar-product-information_pt.pdf
107. Novartis Pharmaceuticals Corporation. PROMACTA (eltrombopag) tablets / for oral suspension. Prescribing information [Internet]. Silver Spring (MD): U.S. Food and Drug Administration; revision in force on 18/12/2025. NDA 022291. Available from: https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7714a0ed-34bb-46e6-a0a5-b363908b22c2
108. Orsini S, Noris P, Bury L, Heller PG, Santoro C, Kadir RA, et al. Bleeding risk of surgery and its prevention in patients with inherited platelet disorders. Haematologica. 2017;102(7):1192-1203. doi:10.3324/haematol.2016.160754. PMID: 28385783. PMC5566025.
109. Bond JA, Statler A, Bozek J, Abbas Khan MA, Ellison MB. May-Hegglin Anomaly (MHA): A Rare Platelet Disorder Presenting for Coronary Artery Bypass Grafting, Managed With Post-bypass Administration of Platelets and Desmopressin. Cureus. 2025;17(9):e93595. doi:10.7759/cureus.93595. PMID: 41035814. PMC12483478.
110. Kato A, Shirai Y, Kanda S, Kunishima S, Shimizu T, Ishida H, et al. Successful kidney transplantation using eltrombopag in a patient with MYH9-related disease. CEN Case Rep. 2025;14(6):833-836. doi:10.1007/s13730-025-01032-y. PMID: 40889065. PMC12605837.
111. ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine; records NCT01133860, NCT02422394, NCT03638817, NCT00925236 [accessed 31/10/2026]. Available from: https://clinicaltrials.gov
112. Zaninetti C, Barozzi S, Bozzi V, Gresele P, Balduini CL, Pecci A. Eltrombopag in preparation for surgery in patients with severe MYH9-related thrombocytopenia. Am J Hematol. 2019;94(8):E199-E201. doi:10.1002/ajh.25500. PMID: 31034630.
113. Limoli MC, Santuccio G, Markovic U, Grasso S, Di Raimondo F, Giuffrida G. From delayed recognition to long-term disease control: Eltrombopag in severe MYH9-related disease. Transfus Apher Sci. 2026;65(6):104539. doi:10.1016/j.transci.2026.104539. PMID: 42837890.
114. U.S. Food and Drug Administration. Nplate (romiplostim) for injection, for subcutaneous use. Prescribing information [Internet]. Thousand Oaks (CA): Amgen; revision of 06/2026. Available from: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=BasicSearch.process (search: Nplate)
115. de Oliveira M, Morais S. Advancing thrombopoietin receptor agonist therapy in inherited thrombocytopenia: new insights into avatrombopag use. Res Pract Thromb Haemost. 2026;10(6):106955. doi:10.1016/j.rpth.2026.106955. PMID: 42835379. PMC13635719.
116. INFARMED. Infomed — search “Eltrombopag” (authorised and marketed medicines) [Internet]. Lisbon: INFARMED; 2026 [accessed 31/10/2026]. Available from: https://extranet.infarmed.pt/INFOMED-fo/
117. Agência Nacional de Vigilância Sanitária (ANVISA). Open data — registered medicines (eltrombopag olamine) [Internet]. Brasília: ANVISA; 2026 [accessed 31/10/2026]. Available from: https://dados.anvisa.gov.br/dados/DADOS_ABERTOS_MEDICAMENTOS.csv
118. Pecci A, Granata A, Fiore CE, Balduini CL. Renin-angiotensin system blockade is effective in reducing proteinuria of patients with progressive nephropathy caused by MYH9 mutations (Fechtner-Epstein syndrome). Nephrol Dial Transplant. 2008;23(8):2690-2. doi:10.1093/ndt/gfn277. PMID: 18503011.
119. Zibara V, Machin N. Inherited Platelet Disorders During Pregnancy and Delivery: Overview of Management Strategies and Emerging Therapeutic Considerations. Hematol Rep. 2026;18(2). doi:10.3390/hematolrep18020016. PMID: 41874099. PMC13010773.
120. Ng YJ, Lee JY, Foo C, Lee WK. Elective Caesarean Delivery in a Patient With May-Hegglin Anomaly: Its Concerns and Review of the Literature. Cureus. 2025;17(6):e85549. doi:10.7759/cureus.85549. PMID: 40630371. PMC12235332.
121. Mastanzade M, Koç A. Pregnancy in May-Hegglin Anomaly: Diagnostic Challenges and Management Considerations. Case Rep Hematol. 2025;2025:4997232. doi:10.1155/crh/4997232. PMID: 40689043. PMC12271717.
122. Haute Autorité de Santé; Agence nationale de sécurité du médicament et des produits de santé. Transfusion de plaquettes : produits, indications. Recommandation de bonne pratique – argumentaire scientifique. Saint-Denis La Plaine: HAS; 2015 Oct. Available from: https://www.has-sante.fr/upload/docs/application/pdf/2015-11/argumentaire_-_transfusions_de_plaquettes.pdf
123. Melo MAW, Silveira CM, Ribeiro MM, Magalhães ERO, Arcanjo GS. Giant Elliptical Platelets as a Cytological Component in MYH9-Related Disease. Int J Lab Hematol. 2026. doi:10.1111/ijlh.70238. PMID: 42680207.
124. Registro Italiano della Malattia MYH9-correlata [Internet]. Pavia: IRCCS Policlinico San Matteo; 2026 [accessed 31/10/2026]. Available from: http://www.registromyh9.org/
125. Platelet Disorder Support Association [Internet]. 2026 [accessed 31/10/2026]. Available from: https://www.pdsa.org/