Fluconazole and Other Azole Embryopathy

Phenocopy of Antley-Bixler syndrome and the reproductive risks of azole antifungals

Literature and regulatory information updated: 29 October 2026.

0. Initial Reference Table

1. Introduction

1.1. Fluconazole and its entry into clinical practice

Fluconazole is a bis-triazole antifungal whose action depends on inhibition of fungal sterol 14-α-demethylase, encoded by the CYP51 gene, the enzyme required for conversion of lanosterol into ergosterol and, through that route, for the integrity of the fungal membrane [26]. The molecule was authorised in the United States on 29 January 1990, following the original application submitted by Pfizer under the trade name Diflucan® [20]; in Portugal, the 50 mg and 150 mg capsules obtained marketing authorisation on 24 and 17 November 1988, with renewal on 21 January 2011 [17]. Near-complete oral bioavailability, a long half-life and good penetration into the central nervous system made the drug the mainstay of treatment for mucocutaneous candidiasis, cryptococcal meningitis and coccidioidomycosis, and at the same time a medicine widely used in a young, fertile population: vulvovaginal candidiasis affects about 10% of pregnant women, with a risk two to ten times that of non-pregnant women [26].

1.2. The first reports of malformations

The first published case of congenital malformations in a child born to a woman treated with fluconazole was described by Lee and colleagues in 1992 [27]. In 1996, Pursley and colleagues assembled three infants born to women who had received fluconazole during or beyond the first trimester: all had congenital anomalies, no other drug was implicated and only one of the three survived; the anomalies were mainly craniofacial, skeletal — with thin, wavy ribs and ossification defects — and cardiac, and corresponded to what had already been observed in developmental toxicology studies in animals, in which fluconazole produced craniofacial ossification defects, thin wavy ribs and renal pelvis anomalies [28]. One of the three infants had previously been classified as having Antley-Bixler syndrome; the timing of the exposure and the similarity to the other two led the authors to reinterpret the deformities as a teratogenic effect of the drug. Aleck and Bartley added a fourth patient in 1997 with a multiple malformation syndrome after fluconazole use in pregnancy [29].

1.3. The definition of a recognisable phenotype

Lopez-Rangel and Van Allen published the fifth case in 2005 and, with it, the first explicit attempt to delineate an identifiable embryopathy [21]. They reviewed the four previous reports and identified as common features the multiple synostoses — craniosynostosis and synostoses of the limbs and digits —, congenital cardiac defects, skeletal anomalies and a recognisable facial dysmorphism, all associated with continuous use of 400 to 800 mg daily. They concluded that prenatal exposure to high-dose fluconazole has an identifiable phenotype, distinct both from Antley-Bixler syndrome and from the classic syndromic craniosynostoses.

1.4. Regulatory intervention

On 3 August 2011, the United States Food and Drug Administration (FDA) issued a safety communication stating that chronic, high doses of 400 to 800 mg daily may be associated with a rare and distinct set of malformations in children whose mothers were treated in the first trimester, and that the risk did not appear to be associated with the single low 150 mg dose used in vaginal candidiasis [13]. The pregnancy category of fluconazole changed from C to D for indications other than vaginal candidiasis, the single 150 mg dose remaining category C. The agency summarised the data in five reports: four with chronic high-dose intravenous fluconazole for coccidioidal meningitis and one in a woman with human immunodeficiency virus (HIV) infection treated orally with a high dose for vaginal candidiasis; all the cases shared features of Antley-Bixler syndrome, which is autosomal recessive. The agency considered that chronic high-dose fluconazole may be teratogenic in humans when used in the first trimester, that the magnitude of that risk is unknown and that the five reports suggest a possible dose-threshold effect.

Five years later, on 26 April 2016, the FDA announced that it was evaluating a Danish study concluding that there was a possible increased risk of spontaneous abortion with oral fluconazole, mostly as one or two 150 mg doses, and advised cautious prescribing while the review was under way, recalling that the Centers for Disease Control and Prevention (CDC) guidance recommends only topical antifungals for vulvovaginal candidiasis in pregnancy [30]. In the European Union, the Pharmacovigilance Risk Assessment Committee (PRAC) concluded, in the periodic safety procedure closed in 2020, that a causal relationship between fluconazole and congenital malformations is at least a reasonable possibility, and repeated the conclusion for adverse pregnancy outcomes in general in the 2023 procedure, with mandatory implementation in the Member States by 22 February 2024 [31, 32].

1.5. Definition of the entities covered

This article covers three distinct realities that the literature and clinical practice tend to conflate. The first is fluconazole embryopathy in the strict sense: a rare, recognisable malformation pattern, documented in five children, associated with continuous exposure to 400–800 mg daily throughout most or all of the first trimester, generally in the treatment of severe endemic mycoses. The second is the set of reproductive risks — spontaneous abortion and small increases in musculoskeletal and possibly cardiac malformations — studied in cohorts of hundreds of thousands of pregnancies exposed to cumulative doses of 150 to 450 mg, the doses used for vulvovaginal candidiasis. The third is the behaviour of the remaining azoles, systemic and topical, for which there are convergent animal data, scant human data and heterogeneous regulatory decisions. To these is added the question of the differential diagnosis with the genetic phenocopies, in particular P450 oxidoreductase deficiency and the syndromic craniosynostoses.

2. Molecular Genetics and Pathophysiology

2.1. Nature of the exposure and absence of transmission

Fluconazole embryopathy has no causal gene and no pattern of inheritance: it is an exogenous embryopathy, determined by the dose, the duration and the gestational timing of the exposure. The genetic interest of the condition is twofold. On the one hand, the molecular target of the drug belongs to a metabolic pathway whose monogenic disorders reproduce the phenotype, which makes this embryopathy one of the classic examples of a pharmacological phenocopy of a genetic disease [33]. On the other, maternal and fetal genetic variation in the metabolising enzymes conditions the effective exposure of the embryo, as discussed below.

2.2. Inhibition of sterol 14-α-demethylase

Fluconazole inhibits fungal CYP51, but the homologous human enzyme, encoded by CYP51A1 at 7q21.2, catalyses the step that removes the 14-α-methyl group from lanosterol in cholesterol biosynthesis [1]. The human gene spans 22 kb and ten exons, and the recombinant enzyme converts 24,25-dihydrolanosterol into its 8,14-diene metabolite in the presence of NADPH and P450 oxidoreductase; lipid deprivation in the culture medium of adrenocortical cells induces CYP51 expression, which is suppressed by 25-hydroxycholesterol [1]. Inhibition of this enzyme in the embryo interrupts de novo cholesterol synthesis downstream of lanosterol, in a tissue that depends on that synthesis of its own: cholesterol is indispensable for membrane formation, for hedgehog signalling, which requires palmitoylation and cholesterol modification of the ligand, and for local steroid production [26].

The most direct evidence comes from the mouse model. Inactivation of Cyp51 in the mouse is lethal around embryonic day 15 and reproduces several prenatal features of Antley-Bixler syndrome: facial hypoplasia, brachycephaly, bowed limb bones with abnormal joints and camptodactyly, accompanied by left ventricular hypoplasia, ventricular septal defects, epicardial defects, defective formation of the cardiac vessels and pericardial haemorrhage [1, 34]. Embryos lacking Cyp51 have no functional mRNA and no detectable protein, markedly accumulate the two substrates of the enzyme — lanosterol and 24,25-dihydrolanosterol — and show no downstream cholesterol precursors, which confirms complete blockade of synthesis; ten cholesterogenesis genes are overexpressed, with the exception of 7-dehydrocholesterol reductase. Lethality is attributed to cardiac failure, and the downstream molecular alterations lie in the retinoic acid and sonic hedgehog pathways [34].

The same biochemical blockade has been documented in human cells from patients with an Antley-Bixler phenotype. In cholesterol-free medium, used to stimulate endogenous synthesis, lymphoblastoid lines from a patient with ambiguous genitalia and no FGFR2 variant accumulated greatly increased amounts of lanosterol and dihydrolanosterol, without analysis of CYP51A1 revealing an evident pathogenic variant [4]. In P450 oxidoreductase deficiency, the same subtle sterol abnormality is compatible with partial blockade at the level of CYP51: serum cholesterol concentrations are grossly normal, but lanosterol and dihydrolanosterol accumulate when cells are grown in cholesterol-poor medium, and the amniotic fluid may contain di- and trimethylated sterols, a finding not exclusive to this disease [12].

2.3. The retinoic acid hypothesis

The second pathway implicated is retinoic acid catabolism. The enzymes CYP26A1, CYP26B1 and CYP26C1 degrade all-trans retinoic acid, keeping it within the narrow window compatible with morphogenesis; CYP26B1, encoded at 2p13.2, converts all-trans retinoic acid into more polar metabolites, this being its preferred substrate [2]. In ICR mice given a single teratogenic dose of 700 mg/kg of fluconazole on the eighth day of gestation, expression of Cyp26a1 and Cyp26b1 in embryos collected at 12, 24 and 48 hours was increased, with no change in Cyp26c1 [35]. Disruption of retinoid catabolism is the proposed explanation for the common spectrum of facial, axial skeletal and limb defects that azoles — medical and agricultural — produce in animals [36]. The argument is strengthened by the observation that the malformation spectrum of a single high dose of ketoconazole in the rat partly overlaps that of vitamin A, with some identical malformations and coinciding windows of sensitivity, others identical with different windows and others exclusive to each agent, such as the discontinuous ribs of ketoconazole and the neural tube defects of retinol palmitate [37].

The human translation of this pathway is documented in genetics: biallelic variants in CYP26B1 cause an autosomal recessive entity with radiohumeral fusions and other skeletal and craniofacial anomalies, with a phenotype resembling that of Antley-Bixler [7]. In the original description, three siblings of first-cousin parents had occipital encephalocele, radiohumeral fusions, oligodactyly, advanced bone maturation and calvarial mineralisation defects — the proband died in utero at 35 weeks and the two subsequent pregnancies were terminated at 15 and 12 weeks after ultrasound diagnosis — and a Turkish child of consanguineous parents had coronal and lambdoid craniosynostosis, a large sagittal skull defect, limited elbow extension and arachnodactyly, dying at 5 months of age; no variants were found in FGFR2, POR or FGF9 [7]. This observation led to the proposal that disturbance of retinoic acid metabolism, and not only of cholesterogenesis, is essential to the skeletal phenotype of P450 oxidoreductase deficiency [38].

2.4. Aromatase inhibition and other mechanisms

Azoles also inhibit aromatase (CYP19A1), which converts C19 androgens into C18 oestrogens, an activity critical for maintenance of early and mid pregnancy and for the regulation of labour. In an assay with recombinant enzyme and a fluorimetric substrate, concordant with the classic assay in placental microsomes, inhibitory potency varied by several orders of magnitude between azoles, and the plasma or tissue concentrations achieved with oral fluconazole and ketoconazole and with topical econazole, bifonazole, clotrimazole, miconazole and sulconazole lay at or above the respective half-maximal inhibitory concentrations [39]. The United States label for fluconazole explicitly attributes the effects observed in the rat to inhibition of oestrogen synthesis and to the known effects of reduced oestrogen on gestation, organogenesis and parturition [20].

2.5. The genetic phenocopy: P450 oxidoreductase deficiency

The link between the drug and the genetic disease was established in 2004, when variants in POR — the gene encoding P450 oxidoreductase, the obligate electron donor of all microsomal P450 enzymes — were identified in a woman with amenorrhoea and in three children with Antley-Bixler syndrome; the authors noted that POR variants equally affect the P450 enzymes that metabolise drugs, which explained the association of the syndrome with maternal fluconazole intake [40]. The OMIM entry for the form with genital anomalies and disordered steroidogenesis frames the same reasoning the other way round: without sufficient POR activity, substances normally converted into innocuous products by the hepatic P450 enzymes may reach teratogenic concentrations, and fluconazole is given as an example of that interaction [4].

P450 oxidoreductase, encoded at 7q11.23, accepts a pair of electrons from NADPH through its FAD group, transfers them to FMN and delivers them to the roughly fifty human microsomal P450 enzymes, including those of cholesterogenesis (squalene epoxidase and CYP51A1), those of steroidogenesis (CYP17A1, CYP21A2, CYP19A1 and CYP3A7) and those of retinoic acid catabolism (CYP26A1 and CYP26B1) [3, 38, 41]. Biallelic deficiency therefore produces a biochemical picture of apparent combined 17-α-hydroxylase and 21-hydroxylase deficiency, a form of congenital adrenal hyperplasia in which, uniquely, skeletal anomalies may occur [12]. The urinary steroid profile is consistent and distinctive: in the nine patients studied by Shackleton in 2004, the raised ratio between the metabolites of the adrenal precursors pregnenolone and progesterone and the metabolites of cortisol was the most characteristic feature, and pregnanediol was proposed as a marker analyte because it is essentially absent from the urine of normal individuals and of those with other defects of steroid biosynthesis [4].

2.6. Dose, duration and timing of exposure

The experimental data indicate a strictly dose-dependent mechanism and a narrow time window. In ICR mice given a single oral dose of 700 mg/kg, gestational day 10 was the period of maximum sensitivity to induction of cleft palate, the predominant effect, with 50% of fetuses affected; on days 8, 9, 11 and 12 the frequencies were, respectively, 12%, 21%, 28.7% and 2.7% [42]. Skeletal examination revealed middle ear anomalies in 15% of fetuses exposed on day 8, chiefly a dysmorphic tympanic ring and absence of the incus, and humeral shortening in 22% of those exposed on day 10. In the dose-response study on day 10, the lowest observed adverse effect level for cleft palate was 175 mg/kg, with 7.6% of fetuses affected [42]. In whole mouse embryo culture, concentrations of 50 and 75 µg/mL produced second branchial arch anomalies in 83.3% and 91.6% of conceptuses, respectively [43].

The United States label summarises the regulatory studies: in rats, doses of 80 to 320 mg/kg, corresponding to about two to eight times the clinical dose of 400 mg on a body surface area comparison, increased embryolethality and produced wavy ribs, cleft palate and abnormal craniofacial ossification; doses of 25 and 50 mg/kg caused anatomical variants, such as supernumerary ribs and renal pelvis dilatation, and delays in ossification; no fetal effects were seen at 5 to 10 mg/kg; in rabbits, abortions occurred at 75 mg/kg, about four times the clinical dose of 400 mg, with no adverse fetal effects [20].

One interaction finding deserves note for its practical relevance. In post-implantation rat embryo culture exposed for 48 hours from day 9.5, a stage equivalent to days 23 to 31 after fertilisation in humans, fluconazole at concentrations of 62.5 to 500 µM produced concentration-dependent facial effects, significantly aggravated by co-exposure to 17 mM ethanol, corresponding to the legal blood alcohol limit for driving in several countries; the same was seen with valproate [44]. This is preclinical evidence, obtained in an alternative culture model, but it suggests that moderate alcohol consumption may increase the teratogenic potency of the drug.

3. Clinical Spectrum and System-Specific Manifestations

3.1. The phenotype of high-dose exposure

The pattern described by the regulators on the basis of the five published cases comprises brachycephaly, abnormal facies, abnormal calvarial development, oral clefting — of the lip or the palate —, bowing of the femurs, thin ribs and long bones, muscle weakness with joint deformities, corresponding to arthrogryposis, and congenital heart disease [13, 20]. The summary of product characteristics in force in Portugal describes, for cases with 400 to 800 mg daily for three months or more in the treatment of coccidioidomycosis, brachycephaly, ear dysplasia, giant anterior fontanelles, femoral bowing and radiohumeral synostosis, adding that the causal relationship between fluconazole and these defects is uncertain [17]. The description in the Oxford Desk Reference: Clinical Genetics and Genomics is concordant: craniosynostosis, orbital hypoplasia, humeroradial synostosis and femoral bowing, with or without cleft palate, cardiovascular malformations and joint contractures, after continuous exposure to more than 400 mg daily in the first trimester [25].

By system, the presentation is organised as follows. The skull shows brachycephaly and craniosynostosis, with documented involvement of the coronal and lambdoid sutures, defects of calvarial mineralisation or development and a wide anterior fontanelle [17, 21]. The face shows midface hypoplasia with a shallow orbital region, hypoplastic supraorbital ridges, hypertelorism and mild ptosis, together with ear dysplasia [17, 21]. The palate may be cleft [13]. The limb skeleton shows radiohumeral or radioulnar synostosis, metacarpophalangeal and proximal interphalangeal symphalangism, femoral bowing and joint contractures with arthrogryposis [13, 21]. The chest has thin, sometimes wavy ribs, with ossification defects [28]. The heart may have congenital malformations, although the published cases do not allow a preferential type to be specified [13].

3.2. The features that separate the toxic phenotype from the genetic one

Comparison with P450 oxidoreductase deficiency is clinically decisive, because the search for potentially fatal adrenal insufficiency depends on it. The GeneReviews chapter on that deficiency notes that, in the five individuals with an Antley-Bixler-like phenotype due to early prenatal exposure to high-dose oral fluconazole, frontal bossing, choanal stenosis or atresia, genital anomalies and camptodactyly were not observed, all of them common features in the genetic disease [12]. The authors of the fifth case likewise state that the child had neither the features of Antley-Bixler nor those of other craniosynostosis syndromes [21]. Conversely, the reclassification of one of Pursley's children, initially labelled as Antley-Bixler syndrome, shows that the overlap is sufficient to generate attribution errors when the exposure history is not taken [28].

3.3. The risks associated with low and cumulative doses

The second clinical reality has no phenotype: it has risk estimates. The largest published cohort used national United States Medicaid data between 2000 and 2014 and compared 37,650 pregnancies exposed to oral fluconazole in the first trimester with 82,090 exposed to topical azoles, out of a total of 1,969,954 pregnancies [14]. The risk of musculoskeletal malformations was 52.1 per 10,000 pregnancies exposed to fluconazole (95% confidence interval 44.8 to 59.3) against 37.3 per 10,000 in those exposed to topical azoles (33.1 to 41.4), with an adjusted relative risk of 1.30 (1.09 to 1.56) and a risk difference of 12.15 per 10,000 (3.69 to 20.60). Conotruncal malformations and oral clefts showed no increase: adjusted relative risk of 1.04 (0.70 to 1.55) and 0.91 (0.61 to 1.35), respectively. The analysis by cumulative first-trimester dose, in which 65.8% of exposed women received 150 mg, 27.7% more than 150 up to 450 mg and 5.6% more than 450 mg, with a maximum of 6,000 mg, showed an adjusted relative risk for musculoskeletal malformations of 1.29 (1.05 to 1.58) with 150 mg, 1.24 (0.93 to 1.66) with more than 150 up to 450 mg and 1.98 (1.23 to 3.17) with more than 450 mg; for conotruncal malformations in the highest dose group, the relative risk was 2.30 (0.93 to 5.65), whose upper limit corresponds to 34 additional cases per 10,000 exposed pregnancies. Malformations as a whole and cardiovascular malformations as a whole did not increase: 1.00 (0.94 to 1.06) and 0.98 (0.87 to 1.10) [14].

The Danish national cohort of live births between 1996 and 2011 compared 7,352 pregnancies exposed in the first trimester — 56% with 150 mg and 31% with 300 mg cumulative — with 968,236 unexposed [16]. The prevalence of malformations was 2.86% against 2.60%, with an adjusted prevalence ratio of 1.06 (0.92 to 1.21). Of the fifteen defects previously associated with azoles, fourteen showed no increase, including craniosynostosis, cleft palate, cleft lip, limb defects, polydactyly, syndactyly, diaphragmatic hernia, cardiac defects as a whole and ventricular septal defects; there were no cases of other craniofacial defects, middle ear defects or pulmonary artery hypoplasia among exposed pregnancies. Tetralogy of Fallot was the exception: seven cases among exposed pregnancies (prevalence 0.10%) against 287 among unexposed (0.03%), with an adjusted prevalence ratio of 3.16 (1.49 to 6.71). Hypoplastic left heart had a prevalence ratio of 2.03, not significant, based on three exposed cases [16].

The older cohorts, of much smaller size, detected no signal. An Italian prospective study compared 226 pregnancies exposed in the first trimester with 452 exposed to non-teratogenic agents: 22 abortions, one stillbirth and seven children with congenital anomalies, with prevalences and neonatal growth parameters similar to those of the reference group, but with a fivefold higher frequency of elective terminations of pregnancy in the exposed group [45]. In North Jutland, between 1991 and 1996, 121 women exposed in the first trimester had 3.3% malformations (four cases) against 5.2% in 13,327 controls without any prescription, with an odds ratio of 0.65 (0.24 to 1.77); preterm birth and low birth weight likewise did not increase [46]. A general practice database analysis compared 234 women exposed to fluconazole, 492 to a topical azole and 88 to another oral azole with 1,629 unexposed in the first trimester: the relative risks of congenital anomaly were 1.1 (0.4 to 3.3), 0.6 (0.2 to 1.6) and 2.1 (0.7 to 6.8), respectively [47]. A later Danish cohort, with 1,079 women exposed in the first trimester — 74% with 150 mg, 22% with 300 mg, 2% with 350 mg and 2% with 600 mg — against 170,453 unexposed, found malformations in 4.1% against 3.6%, with an adjusted prevalence ratio of 1.0 (0.8 to 1.4) [48]. A British prescription-event monitoring study, with usable information from 15,015 questionnaires, followed 289 women exposed before or during pregnancy and identified no harmful effect, although the drug was at that time contraindicated for vaginal candidiasis in pregnancy [49].

3.4. Spontaneous abortion, stillbirth and neonatal death

The study that changed the regulatory perception of the risk was the Danish national cohort from 1997 to 2013, with 1,405,663 pregnancies [18]. Among 3,315 women exposed to oral fluconazole between weeks 7 and 22, 147 had a spontaneous abortion, against 563 among 13,246 unexposed women matched by propensity score, maternal age, calendar year and gestational age, with a hazard ratio of 1.48 (1.23 to 1.77). Using women exposed to topical intravaginal azoles as the comparator, to test confounding by indication, 130 of 2,823 women exposed to fluconazole had an abortion against 118 of 2,823 exposed to topical agents, with a hazard ratio of 1.62 (1.26 to 2.07). Stillbirth did not increase significantly: 21 in 5,382 women exposed from week 7 to birth against 77 in 21,506 unexposed, hazard ratio 1.32 (0.82 to 2.14), and 1.18 (0.64 to 2.16) against topical azoles [18].

Three subsequent studies tested these results. In the records of the Quebec Pregnancy Cohort, between 1998 and 2015, with 441,949 pregnancies, any exposure in early gestation was associated with an increased risk of spontaneous abortion: adjusted odds ratio 2.23 (1.96 to 2.54) for 345 cases exposed to a low dose, 150 mg or less, and 3.20 (2.73 to 3.75) for 249 cases exposed to a dose above 150 mg [26]. First-trimester exposure did not increase major congenital malformations as a whole, but the high dose was associated with cardiac septal closure anomalies, with an adjusted odds ratio of 1.81 (1.04 to 3.14) and thirteen exposed cases; 913 pregnancies were exposed to a low dose and 400 to a high dose in the first trimester. No association was found with stillbirth. The authors declared as limitations the absence of information on smoking, alcohol consumption and over-the-counter folic acid, the lack of microbiological confirmation of candidiasis and the restriction of the database to the 36% of pregnant women covered by the public drug insurance plan; the baseline prevalence of major malformations in the study was 8.6%, higher than the usual 3% to 5%, a difference they attributed to the founder effect of the Quebec population and which, not differing between groups, would not invalidate the comparisons [26].

The Nordic study combined the registers of Sweden, from July 2006 to December 2014, and of Norway, from January 2005 to December 2015, in a total of 1,485,316 pregnancies, with 10,669 exposed matched to 106,690 unexposed in a one-to-ten ratio [50]. There were 2.7 stillbirths per 1,000 exposed pregnancies against 3.6 per 1,000 unexposed, with a hazard ratio of 0.76 (0.52 to 1.10), and 1.2 neonatal deaths per 1,000 against 1.7 per 1,000, with a relative risk of 0.73 (0.42 to 1.29). The results were similar at doses of 300 mg or less, with a hazard ratio of 0.80 (0.50 to 1.28) in 6,739 exposed women, and above 300 mg, with 0.70 (0.39 to 1.26) in 4,543 exposed women. The authors observed that the increased risk of stillbirth suggested by the Danish study for doses above 300 mg, with a hazard ratio of 4.10 (1.89 to 8.90), was not confirmed, and that the earlier result may have been a chance finding, although in both studies the confidence intervals were wide [50].

3.5. The case-control studies and specific defects

The United States National Birth Defects Prevention Study, with expected delivery dates between 1997 and 2011, collected information by interview on fluconazole use in early pregnancy in 31,645 case mothers and 11,612 control mothers; only 44 case mothers and 6 control mothers reported exposure [51]. Six exposed children had cleft lip with cleft palate, four had an atrial septal defect and three had, one each, hypospadias, tetralogy of Fallot, dextro-transposition of the great arteries and pulmonary valve stenosis. The significant associations were with cleft lip with cleft palate, odds ratio 5.53 (1.68 to 18.24), and with dextro-transposition of the great arteries, 7.56 (1.22 to 35.45). The authors themselves stressed that, despite the size of the study, exposure was rare and that the findings were consistent with the case reports but not with the recent epidemiological studies [51].

The more recent analyses of the same study broadened the exposure to all antifungals. For non-cardiac defects, among 19,624 cases and 11,427 controls, 257 cases (1.3%) and 123 controls (1.1%) reported an antifungal in the first trimester, about two thirds of them topical; oral antifungals were reported by 10.1% of exposed cases and 4.9% of exposed controls [52]. The strongest associations were with encephalocele, odds ratio 4.9 (2.3 to 9.3) with nine exposed cases, and with Dandy-Walker malformation, 4.0 (1.6 to 8.4) with six cases; the remaining estimates were modestly raised with imprecise intervals, and miconazole, fluconazole and clotrimazole were the most frequently reported drugs. For cardiac defects, among 11,653 cases and 11,427 controls, six of the twelve estimable cardiac defects had an adjusted odds ratio above 1.5 — tetralogy of Fallot, double-outlet right ventricle with transposition, atrioventricular septal defect, hypoplastic left heart, pulmonary atresia and muscular ventricular septal defect — but the confidence intervals widely included the null, so that the authors concluded that the results do not support a large increase in risk, without excluding small increases for specific cardiac defects [53].

3.6. The synthesis of the meta-analyses

The 2021 meta-analysis concluded that fluconazole should be considered a human teratogen and prescribed with caution to pregnant women and women of childbearing age [55], a position contested in a commentary published in the same journal [57]. The most recent systematic review, with a registered protocol and assessment of risk of bias and certainty of evidence, reached a different reading: the association with major malformations disappears when adjusted estimates are combined, persisting only in a sensitivity analysis restricted to studies with a valid definition of major malformation and for doses above 150 mg; the association with spontaneous abortion persists [56]. The difference between the two conclusions arises from the choice between crude and adjusted estimates and from the handling of confounding by indication, not from new data.

3.7. Patterns of use and population exposure

Exposure is common and, in the case of topical agents, increasing. In the United States, the proportion of pregnancies with oral fluconazole in the first trimester rose from 1.1% in 2000 to 2.6% in 2014 [14], and about 4% of pregnant women are estimated to use the drug at some point [50]. In a population-based study with 1,524 control participants and deliveries between 2014 and 2019, 9.8% reported a fungal infection and 8.6% antifungal use in the first trimester, with 52.7% using prescription products and 45.0% over-the-counter products, and 72.5% using topical formulations; the use of topical and over-the-counter products increased significantly over the period, while that of oral and prescription products decreased modestly [15].

4. Clinical Recognition and Warning Signs

4.1. Before conception and in pre-conception counselling

The highest-risk situation is the woman of childbearing age on prolonged treatment with high-dose fluconazole, typically for coccidioidal meningitis, cryptococcosis, histoplasmosis or refractory candidiasis in the setting of immunosuppression. The same scenario arises in women with hereditary chronic mucocutaneous candidiasis, in whom most need a prolonged oral azole, with fluconazole as first line. The first four published cases arose precisely in this setting, with chronic intravenous fluconazole for coccidioidal meningitis, and the fifth in a woman with HIV infection on prolonged oral treatment [13]. The United States label recommends that effective contraception be considered in women of childbearing age treated with 400 to 800 mg daily, maintained during treatment and for about one week, corresponding to five to six half-lives, after the last dose [20]. The summary of product characteristics in Portugal adds that, after single-dose treatment, a washout interval of one week before conceiving is recommended [17].

4.2. After an exposure has already occurred

When exposure has already occurred, three elements determine the assessment: the daily dose, the duration and the gestational age. Exposure to a single 150 mg dose for vaginal candidiasis, even if repeated, does not justify alarm about a recognisable malformation pattern: the available data from human studies do not suggest an increased risk of anomalies with this dose, and appropriate counselling is reassuring about the overall risk of major malformations, although with honest mention of the spontaneous abortion signal [20, 58]. Continuous exposure to 400 mg daily or more throughout most of the first trimester raises the question of embryopathy, and justifies a morphology ultrasound scan directed at the skull, the face, the limbs, the chest and the heart, as well as referral to a maternal-fetal medicine clinic. For the assessment of ongoing exposures, national teratology information services are the appropriate resource, such as the UK Teratology Information Service, which provides free information to professionals in the United Kingdom [25].

4.3. In the newborn

Recognition in the newborn starts from a combination of findings that, in isolation, have many causes: brachycephaly or craniosynostosis with midface hypoplasia, limited elbow extension due to radiohumeral synostosis, femoral bowing, multiple joint contractures, thin ribs on the chest radiograph, cleft palate and heart disease. Faced with this combination, the obstetric history must be reviewed explicitly for maternal drug intake: the classic approach to craniosynostosis includes, in the history, the question of valproate, associated with metopic synostosis, and of fluconazole, associated with the Antley-Bixler phenotype, besides breech presentation, multiple pregnancy, oligohydramnios, uterine constraint, maternal hyperthyroidism and paternal age [59]. The same pointer appears in the approach to bowed limbs, where the phenocopies of Antley-Bixler syndrome listed are the heterozygous FGFR2 variant and intrauterine exposure to fluconazole [60].

The most important warning sign is not a malformation. Until P450 oxidoreductase deficiency has been excluded, a newborn with an Antley-Bixler phenotype must be assessed for adrenal insufficiency, because partial cortisol deficiency manifests only under stress and is the feature that kills if it is not recognised [41]. The explicit recommendation is to perform adrenal function studies, with a cosyntropin test, regardless of the presence or absence of genital anomalies [12]. Confirmation of fluconazole exposure therefore does not remove the need for endocrine assessment; it removes it only when biochemical and, if necessary, molecular study has ruled out the genetic deficiency. The immediate clinical emergency, however, is the airway: in Antley-Bixler syndrome, obstruction at several levels ranges from severe nasal congestion to fatal obstructive apnoea in the first months if the airway is not aggressively managed, with frequent need for tracheostomy and gastrostomy [61].

4.4. Newborn screening

There is no newborn screening for drug embryopathies, nor is one conceivable, given that the diagnosis is one of exposure. It is nonetheless important to know that newborn screening may bring the genetic phenocopy to medical attention: mild cases of P450 oxidoreductase deficiency may be detected by mass screening programmes for congenital adrenal hyperplasia [38].

4.5. In the child and the adult

The child with established embryopathy subsequently presents with the functional consequences of the malformations: raised intracranial pressure from multisuture craniosynostosis, conductive hearing loss, functional limitation of the upper limbs from radiohumeral synostosis, sequelae of the heart disease and language difficulties secondary to the hearing loss and to prolonged hospital admissions. In P450 oxidoreductase deficiency, of which the picture is a phenocopy, the delays described are chiefly in speech and fine motor skills, attributed to conductive hearing loss, to the skeletal anomalies, to multiple surgical procedures under anaesthesia and to prolonged immobility, and early, effective management of upper airway obstruction, craniosynostosis, hydrocephalus and hearing loss is presented as a condition for good cognitive development [12].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. Absence of formal criteria

There are no published diagnostic criteria for fluconazole embryopathy, nor a code of its own in the nosologies: the entity falls within the Orphanet group of toxic or drug-related embryofetopathies, which has no associated definition [8]. The diagnosis is therefore an inference that requires four elements: documentation of high-dose, prolonged exposure in the first trimester, a compatible malformation pattern, reasonable exclusion of the genetic phenocopies and the absence of another explanation. The European regulatory authority itself considers that the causal relationship between the drug and adverse pregnancy outcomes is at least a reasonable possibility, a formulation that gauges well the degree of certainty available [32], and the summary of product characteristics in Portugal states that the causal relationship between fluconazole and the congenital defects described is uncertain [17].

5.2. Documentation of the exposure

Reconstruction of the exposure must record the name and the formulation, the route, the daily dose, the cumulative dose, the start and end dates in relation to the date of the last menstrual period and the indication. This precision is not bureaucratic: the distinction between a cumulative dose of 150 mg and treatment with 400 to 800 mg daily for three months separates two incomparable risk situations [13, 17]. In the epidemiological studies, exposure was almost always defined by dispensed prescriptions, which does not guarantee that the drug was taken and, conversely, does not capture hospital treatment or over-the-counter products; these limitations, declared by the authors, also apply to clinical practice, in which the interview is the only source for medicines bought without a prescription [14, 26, 62].

5.3. Exclusion of the genetic phenocopies

The assessment of a child with an Antley-Bixler-like phenotype follows the same sequence, whether or not there is a history of exposure. The urinary steroid profile by gas chromatography coupled to mass spectrometry is the central biochemical test: in P450 oxidoreductase deficiency it documents the combined impairment of 17-α-hydroxylase and 21-hydroxylase activities, and the pattern was abnormal in all genetically confirmed patients who underwent it, although not all published cases have been studied in this way [12]. The most characteristic feature is the raised ratio between the metabolites of pregnenolone and progesterone and the metabolites of cortisol, with pregnanediol as the marker analyte [4]. The Oxford Desk Reference: Clinical Genetics and Genomics likewise notes that patients with POR variants have a unique urinary steroid profile [63]. Confirmation is made by identification of biallelic variants in POR, desirable in all cases and mandatory when clinical and laboratory findings are inconclusive [12].

When the skeletal picture includes craniosynostosis and radiohumeral synostosis without the remaining features, and above all without altered steroidogenesis, the most likely hypothesis is the heterozygous FGFR2 variant [63]. Endocrine or molecular assessment is mandatory in these patients to exclude potentially fatal adrenal insufficiency [38]. Faced with occipital encephalocele, radiohumeral fusions and oligodactyly without steroid abnormality, CYP26B1 deficiency should be considered [7, 12].

5.4. Prenatal diagnosis

Prenatal diagnosis of the embryopathy is ultrasonographic and non-specific: the search is for skull shape, femoral biometry and morphology, fixed limb postures, the palate, the facial profile and cardiac anatomy. For the genetic phenocopy there is, by contrast, a validated biochemical method. Unconjugated oestriol in maternal serum is low or undetectable in pregnancies with a fetus affected by P450 oxidoreductase deficiency, a non-specific finding that also occurs in Smith-Lemli-Opitz syndrome, in steroid sulfatase deficiency, in aromatase deficiency and in some severe forms of congenital adrenal hypoplasia [12]. More useful is the steroid profile in maternal urine by gas chromatography and mass spectrometry: in a prospective study of fetuses and newborns with a confirmed genetic diagnosis, an abnormal, distinctive and uniform profile was observed from week 12 onwards, with raised levels of the pregnenolone metabolite epiallopregnanediol and of androgen metabolites of fetal origin and low levels of oestriol [12]. Monitoring of maternal urinary steroid excretion from twelve weeks is presented as non-invasive prenatal diagnosis for couples at risk [61]. Maternal virilisation during a pregnancy with an affected fetus is documented and may be the first clue [41, 61].

5.5. Limits of the tests

No laboratory test demonstrates retrospectively that a malformation was caused by fluconazole. Causal inference rests on the concordance between the exposure window, the dose and the phenotype, and is always probabilistic; the five published cases suggest a threshold effect, but the magnitude of the risk is unknown [13]. Conversely, a normal urinary steroid profile and negative molecular study of POR and FGFR2 do not prove a toxic aetiology: they make it the most reasonable explanation if the exposure is documented. Molecular study has limits of its own: sequencing does not detect deep intronic variants or rearrangements that are not covered, variants of uncertain significance require segregation study, and somatic mosaicism reduces the sensitivity of peripheral blood samples. The laboratory report must state the gene, the transcript with its version, the HGVS nomenclature at DNA and protein level, the zygosity and the classification according to the criteria of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.

In risk communication, one statistical distinction is essential. The relative increases described in the low-dose cohorts apply to rare outcomes, and their absolute translation is small: the 30% increase in the risk of musculoskeletal malformations corresponds to about 12 additional cases per 10,000 exposed pregnancies, and the European regulatory text converts the same data into approximately one additional case per 1,000 women treated with cumulative doses up to 450 mg and about four additional cases per 1,000 with doses above 450 mg [14, 17, 32]. Confusing relative risk with absolute risk is the commonest source of unwarranted anxiety in these consultations.

5.6. Practical algorithm

Faced with a newborn or fetus with craniosynostosis associated with radiohumeral synostosis, femoral bowing or multiple contractures, the sequence is as follows. First, stabilise the airway and assess adrenal function, because both threats are immediate. Second, take the history of maternal exposure, with doses and dates. Third, obtain a urinary steroid profile by gas chromatography and mass spectrometry and, in parallel, skeletal radiographs, echocardiogram, hearing assessment and renal ultrasound. Fourth, study POR and, if the steroid profile is normal and the skeletal features suggest it, FGFR2, FGFR3 and CYP26B1. Fifth, only then attribute the picture to the exposure, and do so with the reservation that the evidence permits.

6. Differential Diagnosis

6.1. Comparative table

6.2. The differential that must be resolved first

In practice, the differential that determines immediate decisions is the one separating the embryopathy from P450 oxidoreductase deficiency, for three reasons. The first is adrenal insufficiency: partial, unmasked only by stress, and lethal if not recognised, it requires replacement treatment and stress-dose cover [41]. The second is genetic counselling: the recurrence risk is 25% per pregnancy in an autosomal recessive disease and nil, without renewed exposure, in an embryopathy. The third is long-term surveillance, which in the genetic deficiency includes adrenal function, blood pressure, pubertal development and ovarian cysts. A detailed drug history is therefore the first step of the differential, and not an adjunct; the GeneReviews review itself stresses that the pregnancy history is essential for identifying the exposure [12].

6.3. The concept of phenocopy applied to the embryopathies

The fluconazole case belongs to a set of embryopathies for which genetic phenocopies have been identified: phenocopies are described for the embryopathies and fetopathies associated with warfarin, leflunomide, mycophenolate mofetil, fluconazole, thalidomide and angiotensin-converting enzyme inhibitors, caused, in most cases, by variants in genes encoding the principal targets of the teratogen or proteins belonging to the same molecular pathways [33]. The usefulness of this reading is twofold: it allows teratogenic mechanisms to be inferred from monogenic diseases and, in the reverse clinical direction, it recalls that a malformation phenotype recognisable as a "syndrome" may have an exogenous cause. Antley-Bixler syndrome with P450 oxidoreductase deficiency and the syndromic craniosynostoses, including Crouzon and Apert syndromes, are here merely differentiated.

7. Treatments and Risk Management Strategies

7.1. General principle

There is no treatment directed at fluconazole embryopathy. Treatment is, on the one hand, prevention — choosing the antifungal and the route that treat the mother without exposing the embryo in the critical window — and, on the other, surgical correction and rehabilitation of the malformations already established. The rule running through all the guidelines is the same: in vulvovaginal candidiasis, topical treatment is first line in pregnancy; in systemic mycoses in the first trimester, amphotericin B replaces the azoles. The decision is never between treating and not treating: untreated maternal infection has consequences of its own, and the mortality of cryptococcal meningitis in pregnancy approaches 25% [23].

7.2. Vulvovaginal candidiasis in pregnancy

The 2021 guidance of the Centers for Disease Control and Prevention is explicit: in pregnancy, only topical azole treatments, applied for seven days, are recommended; epidemiological studies indicate that a single 150 mg dose of fluconazole may be associated with spontaneous abortion and congenital anomalies, and it should therefore not be used [22]. The topical regimens available, in the formulation and duration stated in that guidance, are the following: clotrimazole 1% cream, 5 g intravaginally, daily for seven to fourteen days, or 2% cream for three days; miconazole 2% cream for seven days, 4% cream for three days, 100 mg pessary for seven days, 200 mg pessary for three days or 1,200 mg pessary as a single application; tioconazole 6.5% ointment, 5 g as a single application; butoconazole 2% bioadhesive cream as a single application; terconazole 0.4% cream for seven days, 0.8% cream for three days or 80 mg pessary for three days [22]. The same source warns that the creams and pessaries in these regimens are oil-based and may weaken latex condoms and diaphragms, and that the seven-day duration is the one recommended in pregnancy.

Outside pregnancy, the reference dose is fluconazole 150 mg as a single oral dose; in recurrent candidiasis, defined by three or more symptomatic episodes in less than a year, induction with 100, 150 or 200 mg every three days for a total of three doses, on days 1, 4 and 7, is recommended, followed by maintenance with 100, 150 or 200 mg weekly for six months; in severe disease, with extensive erythema, oedema, excoriation and fissuring, seven to fourteen days of a topical azole or two oral 150 mg doses separated by 72 hours; in candidiasis due to non-albicans species, seven to fourteen days of an azole other than fluconazole and, in the event of recurrence, boric acid 600 mg in a gelatin capsule vaginally, once daily for three weeks, with clinical and mycological eradication rates of about 70% [22]. None of these regimens applies to pregnant women.

The summary of product characteristics in Portugal authorises fluconazole for acute or recurrent vaginal candidiasis "when local treatment is not appropriate", with 150 mg as a single dose in the acute form and, in the recurrent form with four or more episodes a year, 150 mg every three days on days 1, 4 and 7, followed by 150 mg weekly for six months as maintenance [17]. The divergence between the CDC recommendation, which excludes fluconazole in pregnancy, and the European label, which permits it when strictly necessary, is resolved in favour of the guideline: the clinical standard in the pregnant woman with vulvovaginal candidiasis is topical treatment.

7.3. Systemic mycoses: coccidioidomycosis

The 2016 Infectious Diseases Society of America guideline for coccidioidomycosis devotes a section to pregnancy and is the most detailed of the available sources, with explicit grading [19]. The development of symptomatic coccidioidomycosis during pregnancy should prompt consideration of starting antifungal treatment (strong recommendation, moderate evidence). In an initial non-meningeal infection diagnosed in the first trimester, intravenous amphotericin B is recommended (strong, moderate); the alternatives are withholding treatment with close monitoring (weak, low) or an azole after informing the mother of the potential teratogenicity (weak, low); after the first trimester, an azole such as fluconazole or itraconazole may be considered (strong, low); a final alternative is to continue intravenous amphotericin B throughout pregnancy (weak, moderate). In coccidioidal meningitis diagnosed in the first trimester, intrathecal amphotericin B is recommended (strong, moderate), switching to an azole after the first trimester or in cases diagnosed after that stage (strong, low).

In women with a history of coccidioidomycosis who are not on treatment, the risk of reactivation is low and an antifungal is not recommended, but close follow-up should be undertaken, with coccidioidal serology at the first visit and every six to twelve weeks throughout pregnancy (strong, moderate). In women with non-meningeal disease on an azole who become pregnant with the infection in remission, the azole may be stopped, with clinical and serological monitoring every four to six weeks to detect reactivation (weak, low); if remission is not clear, the alternative is to stop the azole and start intravenous amphotericin B in the first trimester, returning to the azole afterwards (strong, low). In the pregnant woman with coccidioidal meningitis already on an azole, azole treatment should be interrupted during the first trimester to avoid the teratogenic risk (strong, moderate), starting intrathecal amphotericin B, especially if there are meningeal signs and symptoms (strong, moderate), with resumption of the azole in the second trimester (weak, low) or continuation of intrathecal amphotericin throughout gestation (weak, low); as alternatives, continuing the azole throughout gestation after the mother accepts that strategy informed of the risks and benefits (weak, low), or stopping the azole, monitoring closely in the first trimester and resuming in the second or third trimester (weak, very low evidence), a strategy that some experts do not recommend because of the risk of relapse [19]. A febrile pulmonary illness in a pregnant woman living in an endemic region, or with a compatible travel history, should be investigated for active coccidioidomycosis with chest radiography, serology and cultures (strong, moderate).

The same guideline justifies these recommendations by the available evidence: at least five cases of congenital craniosynostosis and skeletal anomalies have been observed in children of mothers who received high-dose fluconazole in early pregnancy; the FDA warning of August 2011 does not extend to treatment after the first trimester; the animal data indicate that anomalies may occur with all azoles, in particular voriconazole, and arise in early gestation, so that azole treatment after the first trimester is "probably safe"; amphotericin B has shown safety and efficacy in coccidioidomycosis during pregnancy, with the limitation of the intravenous route and of significant toxicity [19]. Outside pregnancy, when treatment is started in an adult, the recommendation is an orally absorbed azole, such as fluconazole, at a daily dose of 400 mg or more (strong, low) [19].

For the newborn, the same guideline recommends not performing coccidioidal serological tests during the first three months of life, interpreting positive tests in the first year with caution (strong, moderate); empirical treatment with fluconazole 6 to 12 mg/kg per day in infants with suspected coccidioidomycosis, continued until the diagnosis is excluded (strong, low); and avoiding breastfeeding in mothers on azoles other than fluconazole (strong, moderate) [19]. The risk of severe disease is highest in the third trimester and in the period immediately after delivery; there are numerous reports of placental infiltration by spherules without subsequent neonatal infection, which suggests that transplacental infection is rare, while aspiration of amniotic fluid or infected vaginal secretions is the presumed mechanism of transmission; positive serology in the newborn is passive, may not indicate active infection and may persist until the first year of life [19].

7.4. Systemic mycoses: cryptococcosis, invasive candidiasis and histoplasmosis

In cryptococcosis, the 2010 Infectious Diseases Society of America guideline recommends, for disseminated and central nervous system disease in pregnancy, amphotericin B deoxycholate or a lipid formulation, with or without flucytosine (evidence B-II), noting that flucytosine is category C and that its use must be weighed in terms of benefit and risk; fluconazole should be started after delivery, exposure in the first trimester being avoided, and in the second and third trimesters its use is judged against the need for continued antifungal exposure (B-III) [23]. Amphotericin B deoxycholate is the preferred initial regimen for cryptococcal meningoencephalitis, disseminated disease and severe pulmonary cryptococcosis in pregnancy, with FDA category B for use in pregnancy; it distributes well into cord serum, amniotic fluid and the placenta, with cord concentrations between one third and about 100% of maternal serum concentrations, and long clinical experience in coccidioidomycosis has documented neither teratogenicity nor undue toxicity from electrolyte imbalance or renal dysfunction in the mother or the fetus [23]. Flucytosine in combination with a polyene has been used chiefly in the second and third trimesters, with no adverse neonatal outcomes described, but the literature is limited to case reports and small series. In pulmonary cryptococcosis limited to the lungs, without disseminated or central nervous system disease and without significant symptoms, deferring antifungal treatment until after delivery may be considered, with careful observation; fluconazole is excreted in breast milk at concentrations similar to those in plasma [23]. About 45% of the cases of cryptococcosis in pregnancy described in the literature presented with symptomatic disease in the third trimester or postpartum, which the guideline relates to the rapid reversal of the physiological immunosuppression of gestation and to immune reconstitution phenomena [23].

In invasive candidiasis, the 2016 guideline of the same society states that amphotericin B is the treatment of choice in pregnancy; fluconazole, itraconazole, posaconazole and isavuconazole should be avoided, especially in the first trimester, because of the possibility of congenital defects; voriconazole is contraindicated in pregnancy because of fetal anomalies observed in animals; data on the echinocandins are scant, so that their use should be cautious; and flucytosine is contraindicated because of fetal anomalies observed in animals [24]. The global guideline on candidiasis published in 2025 by the European Confederation of Medical Mycology, in cooperation with the International Society for Human and Animal Mycology and the American Society for Microbiology, establishes the echinocandins, including rezafungin, as first line in candidaemia and in all forms of invasive candidiasis except those of the central nervous system and the eye, with liposomal amphotericin B as the alternative and fluconazole only marginally recommended for the initial treatment of candidaemia, because of failure rates and increasing resistance; ibrexafungerp and oteseconazole are presented as new options in vulvovaginal candidiasis [64]. Specific recommendations for pregnancy do not appear in the main text of that document.

In histoplasmosis, the 2025 update of the Infectious Diseases Society of America guideline, published on 10 July 2025, suggests, using GRADE methodology, not routinely treating immunocompetent adults and children with mild acute pulmonary histoplasmosis (conditional recommendation, very low certainty) and leaves the decision whether to treat the moderate form to an assessment of the severity and duration of signs and symptoms (conditional, very low certainty) [65]. When treatment is indicated, itraconazole is the preferred drug, in the original capsules or solution at 200 mg three times a day for three days and then 200 mg twice a day for six to twelve weeks in adults, or 5 mg/kg per dose, up to a maximum of 200 mg per dose, three times a day for three days and then 5 mg/kg per dose twice a day, not exceeding 400 mg daily, in children; the increased-bioavailability formulation, available only in capsules and approved for adults, is given as 130 mg three times a day for three days and then 130 mg twice a day [65]. Monitoring of serum concentrations is recommended, with a trough value of the itraconazole component above 1 mg/L and below 3 to 4 mg/L by a chromatographic method; in recent studies, about 20% of patients needed a dose adjustment for sub- or supratherapeutic levels and about 28% had adverse effects [65]. As regards pregnancy, treatment should be considered only after careful weighing of benefits and risks, ideally in consultation with a maternal-fetal medicine specialist and an infectious diseases physician, given the rarity, complexity and variability of these cases; if treatment is necessary, azoles should be avoided in the first trimester where possible, liposomal amphotericin B being used instead [65]. The text of the guideline states that the use of itraconazole is contraindicated in the first trimester of pregnancy, which requires the involvement of those specialties and recourse to alternative agents such as amphotericin.

A 2026 review of histoplasmosis in pregnancy brings together the corresponding practice: amphotericin B is the only antifungal recommended in the first trimester, owing to the known teratogenicity of the azoles, and is the only agent approved for severe or disseminated pulmonary histoplasmosis in that period, the risks of prolonged azole treatment in the subsequent trimesters and in the breastfed infant being unknown [66]. The same review describes, for liposomal amphotericin B, doses of 3 to 5 mg/kg per day in adult women, with no maximum dose defined in the source, with a regimen of 4 mg/kg three times a week in a pregnant woman with recurrent pulmonary histoplasmosis and of 2 to 10 mg/kg two to three times a week in heterogeneous invasive fungal infections, noting transplacental passage with cord blood to maternal serum ratios of 0.38 to 1, the long terminal half-life of about five days and the dose-dependent risk of nephrotoxicity; in the case reported, the patient received liposomal amphotericin B for 24 weeks of gestation before switching to an oral azole, with delivery at 37 weeks of a healthy child [66]. For itraconazole, the same source records placental passage in animal studies, teratogenicity in animal models and an increased risk of fetal loss in the first trimester in human studies, the high protein binding that may increase free concentrations in pregnancy, the variable absorption and the non-interchangeability of capsules and solution; for voriconazole, teratogenicity in rodents at doses equivalent to 0.3 times the recommended human maintenance dose, with skeletal and visceral anomalies and embryofetal toxicity in rabbits, and one report of exposure in the second and third trimesters for refractory invasive aspergillosis with neutropenia, without adverse outcomes at birth or at six months; for posaconazole and isavuconazole, embryotoxicity and teratogenicity in rodents and rabbits [66]. The voriconazole exposure report referred to is that of a 28-year-old woman, with no past history, who at 16 weeks developed fever, agranulocytosis and hepatic cytolysis from aplastic anaemia and, on the twelfth day of neutropenia, invasive aspergillosis with maxillary sinusitis, periorbital cellulitis and a pulmonary nodule; fetal assessments, including the karyotype, were normal, and delivery occurred at 35 weeks, with an appropriate weight of 2.550 kg, an Apgar score of 10 at one and five minutes, and normal clinical examination, neurological development and growth at six months [67].

7.5. Reference fluconazole dosing and comparison of the labels

The doses that matter in this article are those defining high and low exposure. At the reference dose in the guidelines for systemic indications, fluconazole usually lies between 200 and 800 mg daily, that is, within or above the range associated with the malformation phenotype. The summary of product characteristics in Portugal indicates, in cryptococcosis, a loading dose of 400 mg on day 1 followed by 200 to 400 mg once daily, usually for at least six to eight weeks, with the possibility of increasing to 800 mg in life-threatening infections, and maintenance with 200 mg daily indefinitely in patients at high risk of relapse; in coccidioidomycosis, 200 to 400 mg once daily for eleven to twenty-four months or longer, with 800 mg daily to be considered in certain infections, particularly meningeal disease; in invasive candidiasis, a loading dose of 800 mg on day 1 followed by 400 mg once daily; in oesophageal candidiasis, 100 to 200 mg daily or 200 mg three times a week; in acute vaginal candidiasis, 150 mg as a single dose; and, in dermatomycoses, 150 mg weekly or 50 mg daily for two to four weeks, tinea pedis possibly requiring up to six weeks [17]. The United States label is concordant in broad terms: 150 mg as a single dose in vaginal candidiasis; 200 mg on the first day followed by 100 mg daily in oropharyngeal candidiasis, for at least two weeks, and in oesophageal candidiasis, with doses up to 400 mg daily and a minimum duration of three weeks; 400 mg on the first day followed by 200 to 400 mg daily in cryptococcal meningitis, for ten to twelve weeks after cerebrospinal fluid culture becomes negative, with 200 mg daily for suppression of relapse in patients with AIDS; and 400 mg daily in prophylaxis of candidiasis in bone marrow transplant recipients [20]. In children, the same label indicates 6 mg/kg on the first day followed by 3 mg/kg daily in oropharyngeal candidiasis from 6 months of age, up to 12 mg/kg per day in oesophageal candidiasis, and, for systemic Candida infections, a loading dose of 25 mg/kg on the first day, not exceeding 800 mg, followed by 12 mg/kg daily, not exceeding 400 mg, from 3 months of age, with 12 mg/kg at a postnatal age under 3 months and a gestational age of 30 weeks or more and 9 mg/kg when the gestational age is under 30 weeks [20].

The relevant difference between the two jurisdictions lies not in the doses but in the language about pregnancy. The United States label states that use in pregnancy should be avoided, except in patients with severe or potentially life-threatening fungal infections, in whom fluconazole may be used if the anticipated benefit outweighs the possible risk to the fetus, and that effective contraceptive measures should be considered in women of childbearing age treated with 400 to 800 mg daily [20]. The European text, harmonised by the decision of the Member States coordination group in November 2023 and applicable since 22 February 2024, is more restrictive: fluconazole at standard doses and in short-duration treatments should not be used in pregnancy unless strictly necessary, and fluconazole at high doses or in prolonged treatments should not be used during pregnancy except in situations of potentially life-threatening infections [17, 32].

7.6. Comparative regulatory table

7.7. The other azoles: dosing, risk and regulatory status

7.8. Non-azole alternatives

Amphotericin B is the agent that the guidelines propose in place of the azoles in the first trimester, in the three major systemic mycoses [19, 23, 24, 65]. The deoxycholate formulation has FDA category B for use in pregnancy and distributes into the fetal compartment, with cord blood concentrations between one third and about 100% of maternal serum concentrations [23]; the liposomal formulation is used at 3 to 5 mg/kg per day in adult women (with no maximum dose defined in the source), with reports of intermittent regimens of 4 mg/kg three times a week in pregnancy, and shows transplacental passage with ratios of 0.38 to 1 [66]. The limitations are the exclusively intravenous route, infusion-related reactions, electrolyte disturbances and dose-dependent nephrotoxicity, which requires aggressive electrolyte supplementation and frequent laboratory monitoring [66]. In candidaemia, the echinocandins are first line in the general population, but data in pregnancy are scant and their use is presented as cautious [24, 64]. In vulvovaginal candidiasis due to non-albicans species and in refractory recurrent disease, vaginal boric acid is an option outside pregnancy [22].

7.9. Agents and situations to avoid

The rule that follows from combining the sources is simple: avoid any systemic azole in the first trimester, use a topical azole when the indication is mucocutaneous, and reserve high-dose fluconazole for life-threatening infections when there is no alternative. Voriconazole and oteseconazole are the two agents with the strongest restriction — contraindication in pregnancy in the first case and contraindication also in women of reproductive potential in the second [24, 68]. Flucytosine is contraindicated in pregnancy by the invasive candidiasis guideline, although the cryptococcosis guideline allows it in combination with a polyene, weighing benefit and risk [23, 24]. Co-exposure to alcohol deserves mention in counselling: in an embryo culture model, ethanol at a concentration corresponding to the legal driving limit increased the potency of fluconazole in inducing craniofacial defects [44].

7.10. Ongoing pregnancy, delivery and lactation

When exposure has already occurred, there is no intervention that reverses the risk; management consists of fetal morphological assessment, documentation of the exposure and counselling proportionate to the dose. If antifungal treatment must continue, the switch to amphotericin B during the first trimester, with resumption of the azole afterwards, is the strategy recommended in coccidioidomycosis and histoplasmosis [19, 65]. In the peripartum period, the dominant consideration is maternal disease, not the exposure: in cryptococcosis, about 45% of published cases manifested in the third trimester or postpartum, with a risk of immune reconstitution, and in coccidioidomycosis the risk of severe disease is highest in that same phase [19, 23]. In lactation, a single 150 mg dose of fluconazole is compatible with breastfeeding according to the European label, which does not recommend it after repeated use or a high dose [17]; the United States label quantifies infant exposure as about 13% of the paediatric maintenance dose and cites a survey of 96 women treated with 150 mg on alternate days, with a mean of 7.3 capsules and a range of one to 29, for lactation-associated breast candidiasis, without serious adverse reactions in the infants [20]. The corresponding counselling review considers the data reassuring and sees no need to interrupt breastfeeding, including with a regimen of a 400 mg loading dose followed by 100 mg twice daily for at least two weeks [79]. For the remaining azoles, breastfeeding must be stopped according to the European texts, and the coccidioidomycosis guideline recommends not breastfeeding while on azoles other than fluconazole [19, 72, 74, 76].

7.11. Treatment of established malformations

The management of the affected child is the management of the malformations present, and it is multidisciplinary. The neonatal priority is the airway, because of obstruction at several levels that may culminate in severe obstructive apnoea, with frequent need for tracheostomy and gastrostomy until definitive surgery [61]. Next comes craniosynostosis, with neurosurgical assessment of intracranial pressure and cranioplastic correction, and hydrocephalus, which in P450 oxidoreductase deficiency required ventriculoperitoneal shunting in several patients [12]. Joint contractures improve with age and with passive mobilisation exercises, and there is no fracture tendency after the neonatal period [61]. Radiohumeral synostosis fixes the elbow in flexion and requires occupational therapy assessment directed at function. Heart disease is treated according to the defect. Cleft palate follows the usual protocol of surgical correction and support for feeding and speech. Conductive hearing loss, frequent when there is stenosis of the ear canals, requires early audiological assessment and amplification. In the genetic deficiency, the specific measures include glucocorticoid replacement with stress-dose cover, sex steroid replacement and genital surgery where indicated, with follow-up in a tertiary paediatric endocrinology service [41].

7.12. Developments of the past twelve months

Four developments belong to the period of 2025 and 2026. The first is regulatory and already described: the mandatory implementation, by 22 February 2024, of the harmonised European text affirming the increased risk of miscarriage and quantifying the musculoskeletal risk [32]; its practical effect is now seen in the national product information, including the Portuguese text [17]. The second is the publication, on 10 July 2025, of the update of the Infectious Diseases Society of America guideline on histoplasmosis, which explicitly includes pregnant people within its scope and enshrines the rule of avoiding azoles in the first trimester, using liposomal amphotericin B [65]. The third is the 2025 global guideline on candidiasis, which repositions fluconazole as an only marginally recommended option in the initial treatment of candidaemia and introduces ibrexafungerp and oteseconazole in vulvovaginal candidiasis — with the caveat that oteseconazole is contraindicated in women of reproductive potential [64, 68]. The fourth is the meta-analysis published in December 2024, with a registered protocol and GRADE assessment, which finds no significant association between first-trimester fluconazole and major malformations when adjusted estimates are combined, while maintaining the association with spontaneous abortion and a signal for doses above 150 mg in a sensitivity analysis [56]. None of these developments is a clinical trial: randomised trials of azole exposure in pregnancy do not exist, nor are they ethically conceivable, and the evidence base remains observational, with confounding by indication as the central limitation.

7.13. Studies supporting the recommendations and their limits

Reading these studies together calls for three cautions. First, estimates obtained with different comparators are not directly comparable: using unexposed women, women exposed to topical azoles or women exposed to antibiotics as the reference changes the result, as is seen in the case of transposition of the great arteries, increased against topical azoles and not against unexposed women in the same study, because of an unexpectedly low baseline risk in the topical group [14]. Second, relative risks must be read together with the absolute rate: a relative risk of 1.30 on a rate of 37 per 10,000 produces about twelve additional cases per 10,000 [14]. Third, several findings rest on very small numbers of exposed cases — seven for tetralogy of Fallot, thirteen for septal anomalies, three and six in the case-control studies — and have not been replicated, so that the chance hypothesis, with multiple testing, remains plausible; the authors of the most recent analyses themselves note that their results do not support a large increase in risk, without excluding small increases [16, 26, 51, 53]. Studies restricted to live births do not capture the severe malformations that end in abortion, stillbirth or termination, a bias that the authors quantified as minimal for conotruncal malformations and clefts and slightly favouring a higher risk for musculoskeletal ones [14].

8. Reproductive Options and Family Counselling

8.1. Recurrence risk

Fluconazole embryopathy is not transmitted. The risk of another affected child is the risk of a new equivalent exposure in the same gestational window, and not a mendelian risk. This simple statement has one condition: it requires that the genetic phenocopy has been excluded. If the picture results from P450 oxidoreductase deficiency, the recurrence risk is 25% per pregnancy, since this is an autosomal recessive disease with heterozygous carrier parents [4, 12]; if it results from an FGFR2 variant, the risk depends on the origin of the variant, and with a de novo variant it is substantially below 1%, with a known association with advanced paternal age [5, 63]. Ambiguity in the exposure history is therefore sufficient reason to complete the genetic study before giving a reproductive prognosis.

8.2. Planning before conception in a woman on an azole

Management differs with the indication and the dose. In the woman of childbearing age on high-dose fluconazole for a systemic mycosis, effective contraception is recommended during treatment and for about one week — five to six half-lives — after the last dose [20]; the European text applies the same one-week washout interval after a single dose before conceiving, and allows contraception to be considered for longer treatment courses [17]. For itraconazole prescribed for onychomycosis, the requirement is stricter: highly effective contraception during treatment and for the two subsequent months, with treatment started on the second or third day after the onset of menstruation [69]. For oteseconazole, the exposure window of about 690 days makes mitigation impracticable, and the drug is contraindicated in women of reproductive potential [68]. In women with non-meningeal coccidioidomycosis in remission on an azole who are planning pregnancy, the guideline allows the azole to be stopped with clinical and serological monitoring every four to six weeks [19].

8.3. Counselling after an exposure

Counselling must begin with quantification. For the single 150 mg dose, the message is that short-duration, low-dose exposure, such as that indicated for vaginal candidiasis, is not expected to increase the overall risk of major malformations [58]; the signal to communicate is that of spontaneous abortion, with a hazard ratio between 1.48 and 2.23 depending on the study, and that of the small increase in musculoskeletal malformations, of the order of one additional case per 1,000 women treated with cumulative doses up to 450 mg [17, 18, 26]. For cumulative doses above 450 mg, the musculoskeletal increase is about four additional cases per 1,000 [17]. For continuous exposure to 400 to 800 mg daily throughout most of the first trimester, there is no quantified estimate: there are five published cases, the magnitude of the risk is declared unknown by the regulatory authority and management consists of targeted ultrasound surveillance with individualised counselling [13].

An exposure that has already occurred is not in itself an indication to terminate the pregnancy. It is worth recalling that, in the Italian prospective study, exposed women had a fivefold higher frequency of elective terminations without the outcomes at birth differing from the reference group [45]: overestimation of the risk had consequences of its own. The editorial commentary accompanying publication of the Quebec study summarises the advisable approach: in women with disseminated fungal disease, all treatment risks must be considered in the context of the mother's health, and, if treatment is indicated, there should be a risk-benefit discussion, but appropriate treatment should not be deferred [62].

8.4. Prenatal and preimplantation diagnosis

There is no prenatal diagnosis for an embryopathy: there is ultrasound surveillance. The morphology scan must examine the shape and mineralisation of the skull, the sutures, the facial profile, the palate, the biometry and curvature of the femurs, the limb postures and the cardiac anatomy, repeated according to the findings. For the genetic phenocopies, the options are different and must be explained: in a couple with a previous child with molecularly confirmed P450 oxidoreductase deficiency, prenatal diagnosis by analysis of the known variants and preimplantation genetic diagnosis are possible, and there is also the non-invasive biochemical method of the steroid profile in maternal urine from week 12 [12, 61]. Low unconjugated oestriol in maternal serum may be the first clue, but it is non-specific [12].

8.5. Family and sibling counselling

In an embryopathy, siblings have no increased risk and no indication for testing. If, by contrast, the final diagnosis is P450 oxidoreductase deficiency, it is recommended that apparently asymptomatic siblings, older and younger, be assessed so that treatment can start early [41]. Information about fertility must be honest: in the genetic disease, there are no reports of reproduction by affected individuals and the prevalence of infertility remains uncertain [12]; in the embryopathy, the fertility of the affected child has not been studied.

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency of the embryopathy

There is no incidence estimate for fluconazole embryopathy, and it is not possible to construct one: the known numerator is five children published up to 2011, and the denominator — the number of pregnancies exposed to 400–800 mg daily throughout most of the first trimester — has never been counted [13]. The reports suggest a dose-threshold effect, which is compatible with the rarity of the phenotype in a setting where exposure to low doses is very common [13]. In the national cohorts, high-dose exposures are a minority: 5.6% of exposed pregnancies in the Medicaid cohort received more than 450 mg cumulatively in the first trimester, with a maximum of 6,000 mg [14], and 2% of the 1,079 women in the 2008 Danish cohort received 600 mg [48]. None of those studies identified cases of the full malformation phenotype, which illustrates the difference between an isolated high cumulative exposure and continuous treatment lasting three months.

9.2. Frequency of exposure

Exposure is common and geographically heterogeneous in the way it is measured. In the United States, 1.9% of 1,969,954 Medicaid-insured pregnancies between 2000 and 2014 had a prescription for oral fluconazole in the first trimester, and 4.2% had a topical azole; the proportion with fluconazole rose from 1.1% in 2000 to 2.6% in 2014 [14]. About 4% of American pregnant women are estimated to use fluconazole at some point in gestation [50]. In a population-based study with deliveries between 2014 and 2019, 9.8% of 1,524 participants reported a fungal infection and 8.6% antifungal use in the first trimester, with a predominance of topical formulations (72.5%) and a significant increase in the use of topical and over-the-counter products over the period [15]. In Denmark, 7,352 pregnancies out of a total of about 975,000 live births between 1996 and 2011 had a first-trimester prescription, corresponding to about 0.75% [16]; in Quebec, 913 pregnancies were exposed to a low dose and 400 to a high dose in the first trimester among 226,599 analysed for malformations [26]. The population base of vulvovaginal candidiasis explains these figures: it affects up to 75% of women at least once in their lives and about 10% of pregnant women, with a risk two to ten times that of non-pregnant women [26, 64].

9.3. Geography of the high-dose indications

The geographical distribution of the embryopathy follows that of the endemic mycoses requiring prolonged high-dose fluconazole. Four of the five published cases occurred during treatment of coccidioidal meningitis [13], a disease caused by Coccidioides immitis and C. posadasii, endemic in the arid soils of the southwestern United States, Mexico and parts of Central and South America, with rising incidence and recognised expansion of its distribution into new areas such as eastern Washington, Oregon and Utah; about 60% of infections are asymptomatic, roughly a third cause pulmonary disease and 0.5% to 2% disseminate, more frequently in people with cellular immunodeficiency — including pregnancy — and in individuals of African and Filipino ancestry [82]. In Europe, the disease is imported: a systematic review identified 197 cases published between 1948 and 2022 in twenty countries, with the largest number in Spain (96 cases), Russia and the former Soviet Union (35) and France (20), with 98 cases attributed to migration, 39 to travel and 5 to donor transmission; the sites of infection were pulmonary in 39%, central nervous system in 10% and osteoarticular in 1%, with the site not stated in 36%; fluconazole was first line in 27% of cases and amphotericin B in 19%, with overall mortality of 7% and stable disease in 6%, figures that rise to 24% and 14% in immunocompromised patients [83]. The risk of severe and disseminated disease is greater in pregnancy, especially in the third trimester, and the management of the exposed newborn is the subject of risk-stratification proposals published in 2026 [19, 84].

9.4. Populations studied and gaps

The cohorts underpinning the risk assessment come from a limited set of countries: Denmark, Sweden, Norway, Canada, the United States, Italy, Hungary, the United Kingdom and France. This concentration has consequences. The Quebec database has a baseline prevalence of major malformations of 8.6%, higher than the 3% to 5% usually described, a difference the authors attribute to the founder effect of the population and which, not varying between comparison groups, does not invalidate the relative estimates but limits the transfer of the absolute values [26]. The Medicaid cohort has a younger, racially diverse population with a high burden of disability, characteristics that the authors consider unlikely to affect the biological relationships studied [14]. The Nordic cohorts cover populations with universal access and complete registration, which reduces selection bias but also restricts the variability of exposure. No Portuguese or Brazilian data on gestational azole exposure or on associated outcomes were identified in the sources consulted; this absence reflects the sources examined and does not demonstrate that no published data exist.

9.5. Special populations

Three groups deserve particular attention. The first is women with HIV infection or other immunosuppression, who receive high-dose or prolonged fluconazole for refractory candidiasis, cryptococcosis or prophylaxis, and whose risk is illustrated by the fifth published case [13, 21]; vaginal Candida colonisation and symptomatic candidiasis are more frequent in these women and correlate with the severity of immunosuppression, but treatment of vulvovaginal candidiasis should not differ from that in women without HIV, and prolonged prophylaxis with fluconazole 200 mg weekly, although effective in reducing colonisation and symptoms, is not recommended in the absence of complicated disease [22]. The second is women who live in or travel to regions endemic for mycoses, in whom early diagnosis of the infection allows treatment to be planned and the azole to be avoided in the critical window [19]. The third is women of childbearing age treated for onychomycosis or dermatomycosis, benign indications in which the use of a systemic azole in pregnancy has no justification and is explicitly excluded in the labels [69].

9.6. Networks and resources

There is no European reference network dedicated to embryopathies. Follow-up of affected children falls within the European reference networks for congenital anomalies and developmental disorders, with relevance for craniofacial malformations and for multiple anomalies with developmental delay. For the assessment of ongoing exposures and for counselling, national teratology information services are the practical resource, such as the UK Teratology Information Service, which provides information on the toxicity of drugs and chemicals in pregnancy free of charge to health professionals in the United Kingdom [25]; in Europe and the Americas these services are organised in collaborative networks that, as the Italian prospective study and the itraconazole study show, also generate the prospective cohort data most useful for rare exposures [45, 56, 81].

10. Illustrative Clinical Case

The case described in greatest detail in the literature is the fifth in the series of fluconazole embryopathy, published by Lopez-Rangel and Van Allen in 2005 [21]. It concerns a male infant, assessed at 9 months of age, born to a 30-year-old woman from a 37-week pregnancy. The pregnancy was complicated by maternal human immunodeficiency virus infection and by multiple drug exposures: fluconazole 400 mg daily until the fifth month and again from the sixth month until term, efavirenz, nevirapine, methadone, dapsone, pentamidine and trimethoprim-sulfamethoxazole.

At birth, the child had seizures related to neonatal abstinence syndrome and multiple congenital anomalies were documented. On examination at 9 months, he had craniosynostosis secondary to closure of the coronal and lambdoid sutures, a shallow orbital region, hypoplastic supraorbital ridges, hypertelorism and mild ptosis. In the limbs, he had radioulnar synostosis and metacarpophalangeal and proximal interphalangeal symphalangism of the second to fifth digits, bilaterally.

The authors compared the picture with the four previous cases and identified as common features the multiple synostoses — cranial and digital —, congenital cardiac defects, skeletal anomalies and a recognisable facial dysmorphism. They concluded that the combination of cranial synostosis, multiple symphalangism and long bone anomalies was typical of the published cases of fluconazole embryopathy, that the child had no evidence of embryopathy attributable to efavirenz and that he did not have the features of Antley-Bixler syndrome or of other craniosynostosis syndromes; their reading was that prenatal exposure to fluconazole has an identifiable phenotype [21].

Three observations are needed to read this case rigorously. The first is that the causal attribution is inferential: the pregnancy proceeded with simultaneous exposure to seven drugs, and exclusion of a contribution from the others rests on the absence of their respective known malformation patterns, not on positive proof. The second is that the exposure was 400 mg daily practically throughout gestation, including the whole period of organogenesis — exactly the temporal and dosing pattern that the regulatory authority associated with the phenotype and that does not correspond to the exposures studied in the large cohorts [13]. The third is that endocrine assessment and molecular study of POR do not appear in the available description: exclusion of P450 oxidoreductase deficiency rested on the clinical features that were absent, an approach that today would be complemented by a urinary steroid profile and molecular study [12]. For the same reasons, the management in this report does not represent current routine: the management recommended today for a woman needing high-dose fluconazole in the first trimester is replacement with amphotericin B, in line with the guidelines for systemic mycoses [19, 23, 65].

From a documentary standpoint, one access limitation should be recorded: the full text of the report, including the figures of the patient, was not accessible during the preparation of this article, so the description above is based on the published structured abstract, which contains the clinical data transcribed above. Nor was it possible to obtain the full text of the 1996 series of three infants, whose abstract documents that all had congenital anomalies, that no other drug was implicated, that only one survived and that the anomalies were chiefly craniofacial, skeletal — with thin, wavy ribs and ossification defects — and cardiac, and that one of the children had initially been described as having Antley-Bixler syndrome [28].

11. Prognosis and Long-Term Follow-Up

11.1. Prognosis of children exposed to high doses

Prognostic data are scant and derive from the five published cases. In the 1996 series of three infants, only one of the three survived [28]. The 2005 case was assessed at 9 months, with survival and stable malformations, the available description not documenting later development [21]. There are no follow-up series, nor data on cognitive development, final height, lung function or outcomes in adulthood. This absence must be stated as such in counselling: individual prognosis is estimated from the malformations present, and not from data specific to the embryopathy.

Prognosis by analogy, to be used with due reservation, comes from the genetic disease that the picture reproduces. In the Antley-Bixler phenotype, respiratory compromise from upper airway obstruction at multiple levels ranges from severe nasal congestion to severe obstructive apnoea, and may lead to death in the first months of life if the airway is not aggressively managed; survivors frequently require tracheostomy and gastrostomy until definitive surgery becomes possible; despite the difficulty of assessing gross and fine motor function because of the joint contractures, the prognosis for cognitive development is reasonably good once the difficult perinatal period has been overcome, and contractures improve with age and with passive mobilisation, without a fracture tendency after the neonatal period [61]. In P450 oxidoreductase deficiency, prognosis is determined chiefly by the severity of the skeletal and craniofacial malformations; fetal death is described in cases with very severe malformations; in mild to moderate cases the prognosis is guarded in infancy and improves with age; early death from respiratory complications is a concern, but with careful airway management many children survive with a reasonably good prognosis; prospective follow-up studies describing the natural course are lacking [12].

11.2. Prognosis of pregnancies exposed to low doses

For low-dose exposures, prognosis is measured in probabilities of outcomes, already presented. In summary: the overall risk of major malformations is not increased when adjusted estimates are considered [56]; there is a small increase in musculoskeletal malformations, of the order of one additional case per 1,000 with cumulative doses up to 450 mg and four cases per 1,000 above that value [14, 17]; the risk of spontaneous abortion is increased, with hazard ratios between 1.48 and 3.20 depending on the study and the dose [18, 26, 56]; and stillbirth and neonatal death show no increase in the largest study devoted to those outcomes [50]. The child born without malformations after low-dose exposure requires no specific surveillance beyond the usual.

11.3. Surveillance table

The following table applies to the child with a compatible malformation phenotype, regardless of whether the final aetiology is toxic or genetic, while this has not been established; the rows marked as specific to P450 oxidoreductase deficiency cease to apply once that diagnosis is excluded.

The intervals given coincide with those stated in section 7, where the same interventions were described from the therapeutic side: the airway is the neonatal priority, adrenal function is assessed at diagnosis and repeated according to endocrine judgement, and cranioplasty and ventriculoperitoneal shunting are indicated by clinical signs and not by a fixed schedule.

11.4. Transition to adult care

The child with corrected malformations moves to adult care with three needs: cardiology follow-up of the operated heart disease, functional assessment of the orthopaedic limitations and, if the final diagnosis is the genetic deficiency, lifelong endocrine follow-up with education about the stress dose of glucocorticoid. In P450 oxidoreductase deficiency, surveillance is carried out by a tertiary paediatric endocrinology service throughout childhood, to monitor development and adjust steroid replacement, with periodic formal developmental assessment in experienced centres [41]. Information about fertility must be given with the reservation that the literature imposes: there are no reports of reproduction by individuals with P450 oxidoreductase deficiency and the prevalence of infertility remains uncertain [12].

12. Summary and Concluding Remarks

Fluconazole embryopathy is an entity with two faces that should be kept separate. The first is a rare, recognisable malformation pattern, documented in five children exposed to 400 to 800 mg daily throughout most or all of the first trimester, generally during treatment of coccidioidal meningitis or of refractory candidiasis in immunosuppression, and comprising brachycephaly and craniosynostosis, abnormal calvarial development, midface hypoplasia, cleft palate, radiohumeral or radioulnar synostosis, symphalangism, femoral bowing, thin ribs and long bones, arthrogryposis and congenital heart disease. The second is the set of risks quantified in the large cohorts for the doses used in vulvovaginal candidiasis: an increased risk of spontaneous abortion, consistent across three independent populations, and a small increase in musculoskeletal malformations, of the order of one additional case per 1,000 women treated with cumulative doses up to 450 mg and four cases per 1,000 above that value, without an increase in oral clefts and with cardiac signals that are inconsistent between studies. Conflating these two realities leads to two symmetrical errors: alarming out of all proportion a woman who took a 150 mg dose, and underestimating the risk of prolonged high-dose treatment.

The genetic interest of the condition lies in the mechanism and in the differential diagnosis. Fluconazole inhibits sterol 14-α-demethylase, whose human counterpart, CYP51A1, catalyses an obligatory step of cholesterol synthesis; inactivation of the homologous gene in the mouse reproduces features of Antley-Bixler syndrome and is lethal through cardiac failure, with alterations in the retinoic acid and hedgehog pathways. Disruption of retinoic acid catabolism, demonstrated by overexpression of CYP26A1 and CYP26B1 in exposed embryos and corroborated by the existence of an autosomal recessive disease due to CYP26B1 variants with a similar phenotype, is the second pathway implicated, to which aromatase inhibition is added. The result is one of the clearest examples of a pharmacological phenocopy of a monogenic disease: the same pathway, struck by a biallelic POR variant or by pharmacological inhibition, produces a superimposable phenotype, to the point that one of the children in the original series was initially classified as having Antley-Bixler syndrome.

From this follows the clinical priority. Faced with a newborn with craniosynostosis, radiohumeral synostosis, femoral bowing and contractures, the maternal drug history is part of the diagnosis, not an adjunct; and, until P450 oxidoreductase deficiency has been excluded by urinary steroid profile and molecular study, assessment of adrenal function is mandatory, because partial cortisol deficiency manifests only under stress and is the potentially lethal element of the picture. The airway is the other immediate emergency. The recurrence risk depends entirely on this distinction: nil without renewed exposure in the embryopathy, 25% per pregnancy in P450 oxidoreductase deficiency, below 1% with a de novo FGFR2 variant.

Prevention is consensual and simple to state. In vulvovaginal candidiasis in pregnancy, treatment is a topical azole for seven days, and oral fluconazole should not be used. In systemic mycoses diagnosed in the first trimester, amphotericin B replaces the azoles, intravenously in non-meningeal disease and intrathecally in coccidioidal meningitis, the azole being resumed after the first trimester; the coccidioidomycosis guideline grades each alternative, and the 2025 update on histoplasmosis restates the same rule. In parallel, effective contraception and a washout interval of about one week after the last dose are recommended in women treated with 400 to 800 mg daily. What is not acceptable is deferring treatment of an invasive mycosis out of teratogenic fear: the mortality of cryptococcal meningitis in pregnancy approaches 25%, and maternal health is a condition of fetal health.

Three substantive limitations remain. There is no incidence estimate for the embryopathy, because the denominator of pregnancies exposed to prolonged high doses has never been counted, and the regulatory authority declares the magnitude of the risk to be unknown. There are no long-term follow-up data on affected children, so that prognosis is estimated from the malformations present and, by guarded analogy, from the corresponding genetic disease. And the evidence base for the low doses is entirely observational, with confounding by indication as the central limitation: the divergence between the meta-analyses that regard fluconazole as a human teratogen and the one that finds no association with major malformations after adjustment reflects methodological choices, not new data. The randomised trials that would settle the question do not exist and are not ethically conceivable, which makes precision in documenting the exposure — daily dose, cumulative dose, duration and gestational age — the most useful instrument available to the clinician, both for individual counselling and for future epidemiology.

13. References

1. Online Mendelian Inheritance in Man, OMIM®. MIM number: 601637. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/601637 (accessed 29 Oct 2026).
2. Online Mendelian Inheritance in Man, OMIM®. MIM number: 605207. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/605207 (accessed 29 Oct 2026).
3. Online Mendelian Inheritance in Man, OMIM®. MIM number: 124015. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/124015 (accessed 29 Oct 2026).
4. Online Mendelian Inheritance in Man, OMIM®. MIM number: 201750. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/201750 (accessed 29 Oct 2026).
5. Online Mendelian Inheritance in Man, OMIM®. MIM number: 207410. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/207410 (accessed 29 Oct 2026).
6. Online Mendelian Inheritance in Man, OMIM®. MIM number: 613571. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/613571 (accessed 29 Oct 2026).
7. Online Mendelian Inheritance in Man, OMIM®. MIM number: 614416. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/614416 (accessed 29 Oct 2026).
8. Orphanet. Toxic or drug-related embryofetopathy (ORPHA:251529). Available from: https://www.orpha.net/en/disease/detail/251529
9. Orphanet. Antley-Bixler syndrome (ORPHA:83). Available from: https://www.orpha.net/en/disease/detail/83
10. Orphanet. Antley-Bixler syndrome with genital anomaly and disorder of steroidogenesis (ORPHA:63269). Available from: https://www.orpha.net/en/disease/detail/63269
11. Orphanet. Antley-Bixler syndrome without genital anomaly or disorder of steroidogenesis (ORPHA:596008). Available from: https://www.orpha.net/en/disease/detail/596008
12. Idkowiak J, Cragun D, Hopkin RJ, Arlt W. Cytochrome P450 Oxidoreductase Deficiency. 2005 Sep 8 [updated 2017 Aug 3]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1419/
13. U.S. Food and Drug Administration. FDA Drug Safety Communication: Use of long-term, high-dose Diflucan (fluconazole) during pregnancy may be associated with birth defects in infants [Internet]. Silver Spring (MD): FDA; 2011 Aug 3. Available from: https://www.fda.gov/Drugs/DrugSafety/ucm266030.htm (archived copy: https://web.archive.org/web/2019/https://www.fda.gov/Drugs/DrugSafety/ucm266030.htm)
14. Zhu Y, Bateman BT, Gray KJ, Hernandez-Diaz S, Mogun H, Straub L, et al. Oral fluconazole use in the first trimester and risk of congenital malformations: population based cohort study. BMJ. 2020;369:m1494. doi:10.1136/bmj.m1494. PMID: 32434758. PMCID: PMC7237981.
15. Papadopoulos EA, Howley MM, Fisher SC, Werler MM, Romitti PA, Browne ML. Patterns of Yeast Infections and Antifungal Use in the First Trimester of Pregnancy: Findings from the Birth Defects Study to Evaluate Pregnancy exposureS, 2014-2019. J Womens Health (Larchmt). 2025;34(10):1232-1239. doi:10.1177/15409996251363394. PMID: 40719452. PMCID: PMC13098719.
16. Mølgaard-Nielsen D, Pasternak B, Hviid A. Use of oral fluconazole during pregnancy and the risk of birth defects. N Engl J Med. 2013;369(9):830-839. doi:10.1056/nejmoa1301066. PMID: 23984730.
17. INFARMED. Diflucan 50 mg e 150 mg cápsulas (fluconazol) — Resumo das Características do Medicamento. Laboratórios Pfizer, Lda. Available from: https://extranet.infarmed.pt/INFOMED-fo/download-ficheiro.xhtml?med_id=2576&tipo_doc=rcm
18. Mølgaard-Nielsen D, Svanström H, Melbye M, Hviid A, Pasternak B. Association Between Use of Oral Fluconazole During Pregnancy and Risk of Spontaneous Abortion and Stillbirth. JAMA. 2016;315(1):58-67. doi:10.1001/jama.2015.17844. PMID: 26746458.
19. Galgiani JN, Ampel NM, Blair JE, Catanzaro A, Geertsma F, Hoover SE, et al. 2016 Infectious Diseases Society of America (IDSA) Clinical Practice Guideline for the Treatment of Coccidioidomycosis. Clin Infect Dis. 2016;63(6):e112-46. doi:10.1093/cid/ciw360. PMID: 27470238.
20. Pfizer (Roerig). DIFLUCAN (fluconazole) tablets and oral suspension; DIFLUCAN in sodium chloride injection. Prescribing information. NDA 019949/020090. Revised 2026 Apr. Available from: https://dailymed.nlm.nih.gov/dailymed/lookup.cfm?setid=f694c617-3383-416c-91b6-b94fda371204
21. Lopez-Rangel E, Van Allen MI. Prenatal exposure to fluconazole: an identifiable dysmorphic phenotype. Birth Defects Res A Clin Mol Teratol. 2005;73(11):919-923. doi:10.1002/bdra.20189. PMID: 16265639.
22. Workowski KA, Bachmann LH, Chan PA, Johnston CM, Muzny CA, Park I, et al. Sexually Transmitted Infections Treatment Guidelines, 2021. MMWR Recomm Rep. 2021;70(4):1-187. doi:10.15585/mmwr.rr7004a1. PMID: 34292926. PMCID: PMC8344968.
23. Perfect JR, Dismukes WE, Dromer F, Goldman DL, Graybill JR, Hamill RJ, et al. Clinical practice guidelines for the management of cryptococcal disease: 2010 update by the infectious diseases society of america. Clin Infect Dis. 2010;50(3):291-322. doi:10.1086/649858. PMID: 20047480. PMCID: PMC5826644.
24. Pappas PG, Kauffman CA, Andes DR, Clancy CJ, Marr KA, Ostrosky-Zeichner L, et al. Clinical Practice Guideline for the Management of Candidiasis: 2016 Update by the Infectious Diseases Society of America. Clin Infect Dis. 2016;62(4):e1-50. doi:10.1093/cid/civ933. PMID: 26679628. PMCID: PMC4725385.
25. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 718.
26. Bérard A, Sheehy O, Zhao JP, Gorgui J, Bernatsky S, de Moura CS, et al. Associations between low- and high-dose oral fluconazole and pregnancy outcomes: 3 nested case-control studies. CMAJ. 2019;191(7):E179-E187. doi:10.1503/cmaj.180963. PMID: 30782643. PMCID: PMC6379167.
27. Lee BE, Feinberg M, Abraham JJ, Murthy AR. Congenital malformations in an infant born to a woman treated with fluconazole. Pediatr Infect Dis J. 1992;11(12):1062-1064. PMID: 1461702.
28. Pursley TJ, Blomquist IK, Abraham J, Andersen HF, Bartley JA. Fluconazole-induced congenital anomalies in three infants. Clin Infect Dis. 1996;22(2):336-340. doi:10.1093/clinids/22.2.336. PMID: 8838193.
29. Aleck KA, Bartley DL. Multiple malformation syndrome following fluconazole use in pregnancy: report of an additional patient. Am J Med Genet. 1997;72(3):253-256. doi:10.1002/(sici)1096-8628(19971031)72:3<253::aid-ajmg1>3.0.co;2-s. PMID: 9332650.
30. U.S. Food and Drug Administration. FDA Drug Safety Communication: FDA to review study examining use of oral fluconazole (Diflucan) in pregnancy [Internet]. Silver Spring (MD): FDA; 2016 Apr 26. Available from: https://www.fda.gov/Drugs/DrugSafety/ucm497482.htm (archived copy: https://web.archive.org/web/2019/https://www.fda.gov/Drugs/DrugSafety/ucm497482.htm)
31. European Medicines Agency, CMDh. Fluconazole: scientific conclusions and grounds for variation, amendments to the product information and timetable for implementation (PSUSA/00001404/202003). Amsterdam: EMA; 2020. Available from: https://www.ema.europa.eu/en/documents/psusa/fluconazole-cmdh-scientific-conclusions-and-grounds-variation-amendments-product-information-and-timetable-implementation-psusa00001404202003_en.pdf
32. European Medicines Agency, CMDh. Fluconazole: scientific conclusions and grounds for variation, amendments to the product information and timetable for implementation (PSUSA/00001404/202303). Amsterdam: EMA; 2023. Available from: https://www.ema.europa.eu/en/documents/psusa/fluconazole-cmdh-scientific-conclusions-and-grounds-variation-amendments-product-information-and-timetable-implementation-psusa-00001404-202303_en.pdf
33. Cassina M, Cagnoli GA, Zuccarello D, Di Gianantonio E, Clementi M. Human teratogens and genetic phenocopies. Understanding pathogenesis through human genes mutation. Eur J Med Genet. 2017;60(1):22-31. doi:10.1016/j.ejmg.2016.09.011. PMID: 27639441.
34. Keber R, Motaln H, Wagner KD, Debeljak N, Rassoulzadegan M, Ačimovič J, et al. Mouse knockout of the cholesterogenic cytochrome P450 lanosterol 14alpha-demethylase (Cyp51) resembles Antley-Bixler syndrome. J Biol Chem. 2011;286(33):29086-29097. doi:10.1074/jbc.m111.253245. PMID: 21705796. PMCID: PMC3190716.
35. Tiboni GM, Marotta F, Carletti E. Fluconazole alters CYP26 gene expression in mouse embryos. Reprod Toxicol. 2009;27(2):199-202. doi:10.1016/j.reprotox.2009.01.001. PMID: 19429397.
36. Giavini E, Menegola E. Are azole fungicides a teratogenic risk for human conceptus?. Toxicol Lett. 2010;198(2):106-111. doi:10.1016/j.toxlet.2010.07.005. PMID: 20633616.
37. Mineshima H, Fukuta T, Kato E, Uchida K, Aoki T, Matsuno Y, et al. Malformation spectrum induced by ketoconazole after single administration to pregnant rats during the critical period - comparison with vitamin A-induced malformation spectrum. J Appl Toxicol. 2012;32(2):98-107. doi:10.1002/jat.1636. PMID: 21319175.
38. Spranger JW, Brill PW, Hall C, Nishimura G, Superti-Furga A, Unger S. Bone Dysplasias: An Atlas of Genetic Disorders of Skeletal Development. 4th ed. Oxford University Press; 2018. p. 837–838.
39. Kragie L, Turner SD, Patten CJ, Crespi CL, Stresser DM. Assessing pregnancy risks of azole antifungals using a high throughput aromatase inhibition assay. Endocr Res. 2002;28(3):129-140. doi:10.1081/erc-120015045. PMID: 12489563.
40. Flück CE, Tajima T, Pandey AV, Arlt W, Okuhara K, Verge CF, et al. Mutant P450 oxidoreductase causes disordered steroidogenesis with and without Antley-Bixler syndrome. Nat Genet. 2004;36(3):228-230. doi:10.1038/ng1300. PMID: 14758361.
41. GeneGuidance. POR — cytochrome P450 oxidoreductase (accessed 2026-09-26).
42. Tiboni GM, Giampietro F. Murine teratology of fluconazole: evaluation of developmental phase specificity and dose dependence. Pediatr Res. 2005;58(1):94-99. doi:10.1203/01.pdr.0000166754.24957.73. PMID: 15901894.
43. Tiboni GM. Second branchial arch anomalies induced by fluconazole, a bis-triazole antifungal agent, in cultured mouse embryos. Res Commun Chem Pathol Pharmacol. 1993;79(3):381-384. PMID: 8480084.
44. Metruccio F, Battistoni M, Di Renzo F, Moretto A, Menegola E. Moderate alcohol consumption during pregnancy increases potency of two different drugs (the antifungal fluconazole and the antiepileptic valproate) in inducing craniofacial defects: prediction by the in vitro rat whole embryo culture. Arch Toxicol. 2023;97(2):619-629. doi:10.1007/s00204-022-03410-2. PMID: 36385218. PMCID: PMC9859839.
45. Mastroiacovo P, Mazzone T, Botto LD, Serafini MA, Finardi A, Caramelli L, et al. Prospective assessment of pregnancy outcomes after first-trimester exposure to fluconazole. Am J Obstet Gynecol. 1996;175(6):1645-1650. doi:10.1016/s0002-9378(96)70119-9. PMID: 8987954.
46. Sorensen HT, Nielsen GL, Olesen C, Larsen H, Steffensen FH, Schønheyder HC, et al. Risk of malformations and other outcomes in children exposed to fluconazole in utero. Br J Clin Pharmacol. 1999;48(2):234-238. doi:10.1046/j.1365-2125.1999.00989.x. PMID: 10417502. PMCID: PMC2014300.
47. Jick SS. Pregnancy outcomes after maternal exposure to fluconazole. Pharmacotherapy. 1999;19(2):221-222. doi:10.1592/phco.19.3.221.30919. PMID: 10030772.
48. Nørgaard M, Pedersen L, Gislum M, Erichsen R, Søgaard KK, Schønheyder HC, et al. Maternal use of fluconazole and risk of congenital malformations: a Danish population-based cohort study. J Antimicrob Chemother. 2008;62(1):172-176. doi:10.1093/jac/dkn157. PMID: 18400803.
49. Inman W, Pearce G, Wilton L. Safety of fluconazole in the treatment of vaginal candidiasis. A prescription-event monitoring study, with special reference to the outcome of pregnancy. Eur J Clin Pharmacol. 1994;46(2):115-118. doi:10.1007/bf00199872. PMID: 8039528.
50. Pasternak B, Wintzell V, Furu K, Engeland A, Neovius M, Stephansson O. Oral Fluconazole in Pregnancy and Risk of Stillbirth and Neonatal Death. JAMA. 2018;319(22):2333-2335. doi:10.1001/jama.2018.6237. PMID: 29896619. PMCID: PMC6583485.
51. Howley MM, Carter TC, Browne ML, Romitti PA, Cunniff CM, Druschel CM. Fluconazole use and birth defects in the National Birth Defects Prevention Study. Am J Obstet Gynecol. 2016;214(5):657.e1-9. doi:10.1016/j.ajog.2015.11.022. PMID: 26640069. PMCID: PMC10041360.
52. Papadopoulos EA, Howley MM, Fisher SC, Van Zutphen AR, Werler MM, Romitti PA, et al. Antifungal medication use during pregnancy and the risk of selected major birth defects in the National Birth Defects Prevention Study, 1997-2011. Pharmacoepidemiol Drug Saf. 2024;33(1):e5741. doi:10.1002/pds.5741. PMID: 38112229. PMCID: PMC13137076.
53. Papadopoulos EA, Howley MM, Fisher SC, Van Zutphen AR, Werler MM, Romitti PA, et al. Antifungal medication use during early pregnancy and the risk of congenital heart defects in the National Birth Defects Prevention Study, 1997-2011. Birth Defects Res. 2024;116(2):e2308. doi:10.1002/bdr2.2308. PMID: 38343154. PMCID: PMC13094731.
54. Alsaad AM, Kaplan YC, Koren G. Exposure to fluconazole and risk of congenital malformations in the offspring: A systematic review and meta-analysis. Reprod Toxicol. 2015;52:78-82. doi:10.1016/j.reprotox.2015.02.009. PMID: 25724389.
55. Budani MC, Fensore S, Di Marzio M, Tiboni GM. Maternal use of fluconazole and congenital malformations in the progeny: A meta-analysis of the literature. Reprod Toxicol. 2021;100:42-51. doi:10.1016/j.reprotox.2020.12.018. PMID: 33383164.
56. Latour M, Vauzelle C, Elefant E, Tubach F, Padberg S, Martin B, et al. Risk of congenital malformations and miscarriages following maternal use of oral fluconazole during the first trimester of pregnancy: a systematic review and meta-analysis. Eur J Epidemiol. 2024;39(12):1325-1340. doi:10.1007/s10654-024-01177-7. PMID: 39658659.
57. Marin B, Vauzelle C, Coulm B, Beghin D, Latour M, Tubach F, et al. Comment on "Maternal use of fluconazole and congenital malformations in the progeny: A meta-analysis of the literature". Reprod Toxicol. 2021;104:168. doi:10.1016/j.reprotox.2021.05.003. PMID: 34098045.
58. Kaplan YC, Koren G, Bozzo P. Fluconazole exposure during pregnancy. Can Fam Physician. 2015;61(8):685-686. PMID: 26505064. PMCID: PMC4541431.
59. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 378–380.
60. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 711.
61. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 596–597.
62. Paquette VC, Elwood C. The safety of oral fluconazole therapy in pregnancy. CMAJ. 2019;191(7):E177-E178. doi:10.1503/cmaj.190079. PMID: 30782642. PMCID: PMC6379176.
63. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 380.
64. Cornely OA, Sprute R, Bassetti M, Chen SC, Groll AH, Kurzai O, et al. Global guideline for the diagnosis and management of candidiasis: an initiative of the ECMM in cooperation with ISHAM and ASM. Lancet Infect Dis. 2025;25(5):e280-e293. doi:10.1016/s1473-3099(24)00749-7. PMID: 39956121.
65. Pappas P, Lentz RJ, Stover KR, Wiederhold NP, Ardura MI, Baddley JW, et al. 2025 Clinical Practice Guideline Update by the Infectious Diseases Society of America on Histoplasmosis: Treatment of Mild or Moderate Acute Pulmonary Histoplasmosis in Adults, Children, and Pregnant People. Clin Infect Dis. 2026;81(Suppl 3):i39-i45. doi:10.1093/cid/ciaf258. PMID: 40637160.
66. Kavalier M, Ross J, Bastug K, Bahr NC, Barros N, Semler M, et al. Histoplasmosis in Pregnancy: A Review and Patient Perspective. Open Forum Infect Dis. 2026;13(9):ofag546. doi:10.1093/ofid/ofag546. PMID: 42761582. PMCID: PMC13586759.
67. Shoai Tehrani M, Sicre de Fontbrune F, Roth P, Allisy C, Bougnoux ME, Hermine O, et al. Case report of exposure to voriconazole in the second and third trimesters of pregnancy. Antimicrob Agents Chemother. 2013;57(2):1094-1095. doi:10.1128/aac.00899-12. PMID: 23165463. PMCID: PMC3553730.
68. Mycovia Pharmaceuticals. VIVJOA (oteseconazole) capsules. Prescribing information. NDA 215888. Revised 2026 Jan. Available from: https://dailymed.nlm.nih.gov/
69. Janssen. SPORANOX (itraconazole) capsules. Prescribing information. NDA 020083. Revised 2026 Jul. Available from: https://dailymed.nlm.nih.gov/
70. Mayne Pharma. TOLSURA (itraconazole) capsules. Prescribing information. NDA 208901. Revised 2025 Apr. Available from: https://dailymed.nlm.nih.gov/
71. Pfizer. VFEND (voriconazole). Prescribing information. NDA 021267. Revised 2026 Mar. Available from: https://dailymed.nlm.nih.gov/
72. European Medicines Agency. VFEND (voriconazole): EPAR — Product information. Available from: https://www.ema.europa.eu/en/documents/product-information/vfend-epar-product-information_en.pdf
73. Merck Sharp & Dohme. NOXAFIL (posaconazole). Prescribing information. Revised 2026 Feb. Available from: https://dailymed.nlm.nih.gov/
74. European Medicines Agency. Noxafil (posaconazole): EPAR — Product information. Available from: https://www.ema.europa.eu/en/documents/product-information/noxafil-epar-product-information_en.pdf
75. Astellas. CRESEMBA (isavuconazonium sulfate). Prescribing information. NDA 207500/207501. Revised 2025 Apr. Available from: https://dailymed.nlm.nih.gov/
76. European Medicines Agency. Cresemba (isavuconazole): EPAR — Product information. Available from: https://www.ema.europa.eu/en/documents/product-information/cresemba-epar-product-information_en.pdf
77. Ketoconazole shampoo 2%. Prescribing information. ANDA 218498. Revised 2026 Jan. Available from: https://dailymed.nlm.nih.gov/
78. Czeizel AE, Tóth M, Rockenbauer M. No teratogenic effect after clotrimazole therapy during pregnancy. Epidemiology. 1999;10(4):437-440. doi:10.1097/00001648-199907000-00013. PMID: 10401880.
79. Kaplan YC, Koren G, Ito S, Bozzo P. Fluconazole use during breastfeeding. Can Fam Physician. 2015;61(10):875-876. PMID: 26759844. PMCID: PMC4607332.
80. Bar-Oz B, Moretti ME, Bishai R, Mareels G, Van Tittelboom T, Verspeelt J, et al. Pregnancy outcome after in utero exposure to itraconazole: a prospective cohort study. Am J Obstet Gynecol. 2000;183(3):617-620. doi:10.1067/mob.2000.105962. PMID: 10992182.
81. De Santis M, Di Gianantonio E, Cesari E, Ambrosini G, Straface G, Clementi M. First-trimester itraconazole exposure and pregnancy outcome: a prospective cohort study of women contacting teratology information services in Italy. Drug Saf. 2009;32(3):239-244. doi:10.2165/00002018-200932030-00006. PMID: 19338381.
82. Crum NF. Coccidioidomycosis: A Contemporary Review. Infect Dis Ther. 2022;11(2):713-742. doi:10.1007/s40121-022-00606-y. PMID: 35233706. PMCID: PMC8887663.
83. Shastri P, Bains G, Henriksen AS, Karas A. Coccidioidomycosis in Europe: a systematic literature review of epidemiology, treatment and outcomes. J Antimicrob Chemother. 2026;81(1):dkaf407. doi:10.1093/jac/dkaf407. PMID: 41206478. PMCID: PMC12802897.
84. Darmawan DO, Mathew R, Pasca AM, Aziz N, Zhang X, Kappagoda S, et al. Neonatal management and outcomes in the setting of disseminated coccidioidomycosis in pregnancy. J Perinatol. 2026;46(5):692-699. doi:10.1038/s41372-026-02659-0. PMID: 42026156.