Customize your JAMA Network experience by selecting one or more topics from the list below.
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
Although treatment with oral fluconazole during pregnancy is generally discouraged,1 approximately 4% of pregnant women in the United States use fluconazole.2 There are concerns that fluconazole use may be associated with stillbirth, particularly in doses above those commonly used for the treatment of vaginal candidiasis (150 mg, administered once or twice). In a nationwide Danish study,3 any fluconazole exposure was not associated with an increased risk of stillbirth, but the estimate had limited precision (hazard ratio [HR], 1.32 [95% CI, 0.82-2.14]); exposure to doses above 300 mg was associated with stillbirth (HR, 4.10 [95% CI, 1.89-8.90]).
We investigated if fluconazole use during pregnancy is associated with stillbirth and neonatal death.
Nationwide register data were used to identify all pregnancies with singleton live births and stillbirths in Sweden (July 2006-December 2014) and Norway (January 2005-December 2015). We excluded pregnancies with missing maternal personal identification number, missing or implausible gestational age, nonresidence in the country, prescription for fluconazole within 28 days before conception, and prescription for any nonfluconazole oral azole antifungal between 28 days before conception and delivery. The study was approved by the respective research ethics committees. Informed consent was waived.
The primary outcomes were stillbirth (fetal loss after 22 completed weeks; except during July 2006-June 2008 in Sweden, when it was defined as after 28 completed weeks) and neonatal death (0-27 days after live birth) associated with any fluconazole exposure at any time during pregnancy, as defined by filled prescriptions. Secondary analyses investigated outcomes by fluconazole dose.
Using logistic regression, propensity scores were estimated in each country data set. Fluconazole-exposed and unexposed pregnancies were matched (1:10) on age and propensity scores. A distinct matched cohort was created for analyses of stillbirth (based on live births and stillbirths) and neonatal death (based on live births). The cohort for analysis of stillbirth also included gestational day at fluconazole exposure as a matching criterion.3 Following matching, data from the 2 countries were pooled for analysis. Sensitivity analyses restricted to Sweden were conducted including a broader set of covariates in the propensity score. Cox and Poisson regression were used to estimate HRs for stillbirth and risk ratios (RRs) for neonatal death, respectively (SAS [SAS Institute], version 9.4).
From a cohort of 1 485 316 pregnancies (852 959 in Sweden and 632 357 in Norway), 10 669 exposed and 106 690 unexposed pregnancies were included in the matched analysis of stillbirth, and 10 640 exposed and 106 387 unexposed pregnancies in the matched analysis of neonatal death. Baseline characteristics were well balanced between groups (Table 1).
There were 2.7 stillbirths per 1000 exposed pregnancies and 3.6 per 1000 unexposed pregnancies (HR, 0.76 [95% CI, 0.52-1.10]), and 1.2 neonatal deaths per 1000 exposed pregnancies and 1.7 per 1000 unexposed pregnancies (RR, 0.73 [95% CI, 0.42-1.29]; Table 2). Results were similar for doses of 300 mg or less and for more than 300 mg. Sensitivity analyses with a broader set of covariates in the propensity score were consistent with the primary analyses.
In this cohort study, fluconazole use in pregnancy was not associated with significantly increased risks of stillbirth or neonatal death. The outcome of neonatal death has not been reported previously, to our knowledge. An increased risk of stillbirth suggested by the Danish study3 for any fluconazole exposure or for doses more than 300 mg was not confirmed. This study included twice the number of fluconazole-exposed pregnancies from 2 countries, although the number exposed to higher doses was still small. However, in both studies, CIs were wide and, for any exposure, neither result was statistically significant. The previous result may have been due to chance.
The possibility of confounding cannot be excluded; of concern would be unmeasured confounders that could bias results toward no increased risk. Filled prescriptions were used to define drug exposure; any nonuse of fluconazole would bias results toward the null.
Although the data on fluconzazole use in pregnancy suggest no increased risk of stillbirth, additional studies should be conducted and the collective body of data3-5 scrutinized by drug authorities before recommendations to guide clinical decision making are made, and weighed against the benefits of therapy.
Accepted for Publication: April 23, 2018.
Corresponding Author: Björn Pasternak, MD, PhD, Clinical Epidemiology Division T2, Department of Medicine Solna, Karolinska Institutet, 17176 Stockholm, Sweden (firstname.lastname@example.org).
Author Contributions: Dr Pasternak had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Pasternak.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Wintzell.
Obtained funding: Pasternak.
Administrative, technical, or material support: Engeland, Neovius, Stephansson.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This study was supported by the Thrasher Research Fund, the Magnus Bergvall Foundation, and the Karolinska Institutet Research Foundation. Drs Pasternak and Stephansson were supported by investigator grants from the Strategic Research Area Epidemiology program at Karolinska Institutet.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Create a personal account or sign in to: