To the Editor:
In 2010, we reported higher risk of cryptorchidism among boys prenatally exposed to maternal use of acetaminophen.1 Another paper was published shortly thereafter also showing associations of prenatal exposure to the cyclooxygenase (COX) inhibitors, acetaminophen, acetylsalicylic acid, and ibuprofen with cryptorchidism in boys.2 Experimental data showed reduced testicular prostaglandin and testosterone production and reduced anogenital distance in prenatally exposed rats.2 Increased apoptosis in testis tissue and reduced testosterone levels in serum from experimental cryptorchid mice exposed to COX-2 inhibitors have also been reported.3 Yet another study has demonstrated that weak analgesic COX inhibitors alongside numerous environmental contaminants inhibit prostaglandin synthesis in mouse Sertoli cells and in fetal rat testes.4
The Danish and European Medicines Agencies have evaluated the evidence from the epidemiologic reports,1,2 and have requested additional data analyses. We acknowledge their wish to further scrutinize the data, given the public health concern that has been raised. Pregnant women frequently use weak analgesics—particularly drugs containing acetaminophen—and such drugs are sold over-the-counter in most countries in Europe and in the United States.1,5,6
In our study, 47,400 women were interviewed during and after pregnancy, and their liveborn sons were followed for a diagnosis of cryptorchidism and for a corrective orchiopexy.1 We estimated hazard ratios (HRs) using Cox regression models, adjusting for potential confounders. Analyses were performed according to weeks of acetaminophen use during the entire pregnancy, and during gestational weeks 8-14 (a suggested susceptible window of “male programming”). The combined use of acetaminophen, ibuprofen, and acetylsalicylic acid was also evaluated (Table, Model A). The study was criticized for not presenting estimates adjusted for indications of weak analgesics use, and for not taking maternal alcohol consumption into account, (although alcohol's confounding potential is questionable.7) Estimates from Model B (Table) demonstrate that adjustment for the aforementioned factors changed the estimates only minimally. We then excluded women without a report of pain during pregnancy (Model C), which also had limited impact on the estimates. Finally, we divided the cohort into women reporting muscle or joint disease or fever or inflammation/infection during pregnancy (Model D) and women reporting none of these specific indications of analgesics use (Model E). This stratification showed subtle signs of effect modification judged by changes in the point estimates. We observed slightly higher HRs with cumulative acetaminophen exposure during the entire pregnancy, but paradoxically slightly lower HRs during weeks 8-14, among women reporting no disease to trigger weak analgesic use (Model E). The opposite pattern was apparent among women reporting muscle or joint disease or signs of infection or inflammation (Model D). For combined exposures to the weak analgesics, slightly higher HRs were observed among women reporting no disease (Model E).
Recent studies have strengthened the biologic plausibility of the suggested association.2-4 If inhibition of prostaglandin synthesis is the main causal mechanism, we expect adverse effects from most pharmaceutical COX inhibitors. Studies in humans are scant and results inconsistent. Strong associations between prenatal exposure to several weak analgesics and cryptorchidism was observed in Denmark, but not in Finland.2 Our own study reported moderate effects of acetaminophen exposure, and little evidence of effects from ibuprofen and acetylsalicylic acid.1 The additional analyses presented here give little reason to modify our original conclusion, given uncertainties due to reduced sample sizes in the stratified analyses and the role of chance. It is clearly premature to draw firm conclusions on this important question and we encourage others with data to study the hypothesis.
The financial support for the Danish National Birth Cohort was acknowledged in the original publication. Aarhus University, Faculty of Health Sciences, supported Morten Søndergaard Jensen with a PhD scholarship.
Morten Søndergaard Jensen
Tine Brink Henriksen
Perinatal Epidemiology Research Unit, Department of Pediatrics
Aarhus University Hospital
Arizona Respiratory Center
University of Arizona
Ane Marie Thulstrup
Department of Occupational Medicine Aarhus University Hospital
Henrik Toft Sørensen
Department of Clinical Epidemiology Aarhus University Hospital
Jens Peter Bonde
Department of Occupational and Environmental Medicine
Bispebjerg Hospital, University of Copenhagen
Department of Epidemiology
The Institute of Public Health
University of Aarhus
1. Jensen MS, Rebordosa C, Thulstrup AM, et al. Maternal use of acetaminophen, ibuprofen and acetylsalicylic acid during pregnancy and risk of cryptorchidism. Epidemiology.
2. Kristensen DM, Hass U, Lesne L, et al. Intrauterine exposure to mild analgesics is a risk factor for development of male reproductive disorders in human and rat. Hum Reprod.
3. Kubota H, Sasaki S, Kubota Y, et al. Cyclooxygenase-2 protects germ cells against spermatogenesis disturbance in experimental cryptorchidism model mice. J Androl.
4. Kristensen DM, Skalkam ML, Audouze K, et al. Many putative endocrine disruptors inhibit prostaglandin synthesis. Environ Health Perspect.
5. Headley J, Northstone K, Simmons H, Golding J; ALSPAC Study Team. Medication use during pregnancy: data from the Avon Longitudinal Study of Parents and Children. Eur J Clin Pharmacol.
6. Werler MM, Mitchell AA, Hernandez-Diaz S, Honein MA. Use of over-the-counter medications during pregnancy. Am J Obstet Gynecol.
7. Jensen MS, Bonde JP, Olsen J. Comment on maternal alcohol consumption-Re: “Cryptorchidism and hypospadias as a sign of testicular dysgenesis syndrome (TDS): Environmental connection”. Birth Defects Res A Clin Mol Teratol.