Antibiotics During Pregnancy May Increase Miscarriage Risk

Norra MacReady

May 01, 2017

Use of certain antibiotics early in pregnancy is associated with an increased risk for spontaneous abortion, the authors of a new study report.

Macrolides (except erythromycin), quinolones, tetracyclines, sulfonamides, and metronidazole all were associated with a greater risk, compared with penicillins, cephalosporins, or no antibiotic exposure at all, Flory T. Muanda, MD, and colleagues write in an article published in the May 1 issue of CMAJ.

The findings should be considered when guidelines are updated regarding treatment of infection during pregnancy, according to the authors.

This study was "well-conducted and highlights some of the potential negative consequences of using antibiotics in this population, especially if they're being prescribed inappropriately," Jason G. Newland, MD, MEd, associate professor of pediatrics, Washington University School of Medicine, St. Louis, and a spokesperson for the Infectious Diseases Society of America, told Medscape Medical News. He was not involved in the research.

Large Cohort Study

To assess the potential effect of antibiotics on miscarriage risk, Dr Muanda, from the Faculty of Pharmacy, Université de Montréal, Quebec, Canada, and colleagues analyzed data from the Quebec Pregnancy Cohort on pregnancies that occurred between January 1998 and December 2009. The cohort is ongoing and includes information on all pregnancies among women covered by the Quebec Public Prescription Drug Insurance Plan.

Eligible patients were aged 15 to 45 years on the first day of gestation and continuously insured for at least 1 year before and during their pregnancy. Women who experienced a clinically detected spontaneous abortion before gestational week 20 were considered cases, with the calendar date of the spontaneous abortion designated the index date. For each case, the authors identified 10 control patients matched by age, year of pregnancy, and gestational date within 3 days.

Antibiotic exposure was defined as "having filled at least 1 prescription for any type of antibiotic either between the first day of gestation and the index date, or before pregnancy but with a duration that overlapped the first day of gestation," the authors explain.

The study included cephalosporins, macrolides, penicillins, quinolones, sulfonamides, tetracyclines, other antibacterial agents, antiprotozoals, and urinary anti-infectives.

Overall, 182,369 pregnancies met the inclusion criteria; of those, 8702 (4.7%) ended in spontaneous abortion, at a mean gestational age of 14.1 weeks (standard deviation [SD], 3.1 weeks; median, 14 weeks). The matched control group consisted of 87,020 pregnancies.

Antibiotic exposure occurred in 12,446 (13%) of those pregnancies, including 1428 that ended in spontaneous abortion (16.4% of all pregnancies ending in spontaneous abortion). Among the control patients, 11,018 (12.6% of all controls) were exposed to antibiotics (P < .001 compared with the cases).

Women who experienced a spontaneous abortion had a mean age of 28.7 years (SD, ¬±6.3 years) compared with women in the control group, who had a mean age of 27.8 years (SD, ±5.5 years' P < .001). They were also more likely to have been diagnosed with illnesses such as depression and asthma, to have visited a hospital or emergency department within the year before pregnancy, and to report more use of medications overall.

In an effort to account for these and other potential confounders, the authors analyzed a long list of covariates, including those associated with socioeconomic status, comorbidities, and demographics.


After adjustment for possible confounding factors, the increased risk associated with a particular antibiotic, compared with no antibiotics, is shown in the table. No increased risk was associated with nitrofurantoin, erythromycin, penicillins, or cephalosporins.

Table. Risk Associated With Antibiotics

Antibiotic Number of Exposed Cases Adjusted Odds Ratio (95% Confidence Interval)
Azithromycin 110 1.65 (1.34 - 2.02)
Clarithromycin 111 2.35 (1.90 - 2.91)
Tetracyclines 67 2.59 (1.97 - 3.41)
Doxycycline 36 2.81 (1.93 - 4.10)
Minocycline 21 2.48 (1.54 - 4.00)
Quinolones 160 2.72 (2.27 - 3.27)
Ciprofloxacin 114 2.45 (1.98 - 3.03)
Norfloxacin 8 4.81 (2.05 - 11.26)
Levofloxacin 14 3.28 (1.73 - 6.24)
Sulfonamides 30 2.01 (1.36 - 2.97)
Metronidazole 53 1.70 (1.27 - 2.26)

To account for the underlying risk associated with an infection, the authors repeated the analysis, using penicillins and cephalosporins, which have the most safety data available, as the reference group. The results remain largely unchanged, they report.

They also conducted two subgroup analyses consisting only of pregnancies with urinary tract or respiratory tract infections. In the urinary tract infection group, the adjusted odds ratio associated with quinolone exposure, compared with penicillin use, was 8.73 (95% confidence interval [CI], 3.08-24.77; 17 exposed cases).

In the respiratory tract infection group, macrolide use was associated with a trend toward increased risk for spontaneous abortion compared with penicillin, but it was not statistically significant (adjusted odds ratio, 1.89; 95% CI, 0.97 - 3.69; 17 exposed cases).

The authors did not observe an increased risk for spontaneous abortion associated with nitrofurantoin exposure, a finding that "supports its use as an alternative to trimethoprim–sulfamethoxazole for the treatment of urinary tract infection during pregnancy."

Putting It in Perspective

In some instances, these findings support data from other studies, the authors point out. The class effect observed of tetracyclines and quinolones "supports current guidelines used in obstetrics that do not recommend use of these drugs in early pregnancy." Their finding that metronidazole was associated with a 70% increase in the risk for spontaneous abortion is similar to that of a study among Medicaid patients showing a 67% increased risk.

The authors acknowledge that confounding by infection severity is a potential study limitation. "However, we adjusted for several documented proxies for infection severity, such as prior exposure to antibiotics, comorbidities, hospital-based diagnosis of maternal infections and prior hospital admissions," they write.

Other, unmeasured confounders such as tobacco or alcohol use and folic acid intake may account for the findings. However, the authors state, they used two active comparator groups to attenuate this effect, so "unmeasured confounding, if present, would not fully explain this finding."

Dr Newland told Medscape Medical News that he agrees with the authors' interpretation of their findings. "I thought the authors did an excellent job in their discussion in addressing some of the potential confounders, and I don't think they oversell their claim about the adverse consequences of using antibiotics," he said.

"They acknowledge that there may be confounding factors they can't account for in a trial like this; the only way you can know for sure is to do a large, randomized controlled trial, which we're just not going to do."

Antibiotics "are life-saving drugs, but there are negative consequences when we don't use them appropriately, and they are avoidable," he added.

"A study like this may encourage clinicians to pause and consider the impact of using antibiotics in certain situations, and remind people that there may be negative consequences to these medications, which we have used very liberally at times."

One of the study authors is a consultant for plaintiffs in litigation involving antidepressants and birth defects. Dr Newland has a grant for antibiotic stewardship from Merck. The study authors and Dr Newland have disclosed no other relevant financial relationships.

CMAJ. 2017;189:E625-E633. Full text

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.