2024 Volume 47 Issue 7 Pages 1301-1306
Cefcapene pivoxil hydrochloride is an antibiotic often used by women who are or may be pregnant. However, the safety of exposure to it during the first trimester of pregnancy has not been assessed. In this study, we aimed to clarify the effects of exposure during the first trimester of pregnancy on maternal and fetal outcomes. Data were obtained from pregnant women who were counseled on drug use during pregnancy at two Japanese facilities from April 1988 to December 2017. The incidence of major malformations in singleton pregnancy was compared between neonates born to women who took cefcapene pivoxil hydrochloride (n = 270) and control drugs (n = 1594) during their first trimester. The adjusted odds ratio of the incidence of major malformations was calculated using multivariate logistic regression analysis adjusted for smoking during pregnancy and maternal age. The incidence of major malformations was 2.6% in the cefcapene pivoxil hydrochloride group and 1.8% in the control group. There were no significant differences in the incidence between the cefcapene pivoxil hydrochloride and control groups (adjusted odds ratio: 1.48 [95% confidence interval: 0.64–3.42], p = 0.36). This prospective cohort study showed that exposure to cefcapene pivoxil hydrochloride during the first trimester of pregnancy was not associated with increased risk of major malformations in infants. Our findings will help healthcare providers in choosing appropriate medicines.
Cefcapene pivoxil hydrochloride (CFPN-PI) is a third-generation cephalosporin antibiotic that was approved for use in Japan in 1997. It inhibits the synthesis of bacterial cell walls and exhibits a broad spectrum of antibacterial activity against Gram-positive bacteria such as methicillin-sensitive Staphylococcus aureus and Streptococcus pyogenes, and Gram-negative bacteria such as Serratia spp. and Proteus spp.1,2) CFPN-PI also shows potent antibacterial activity against penicillin-resistant Streptococcus pneumoniae and ampicillin-resistant Haemophilus influenzae, because of its resistance to the β-lactamase produced by these bacteria.3–5) It is therefore used to treat a variety of infections including cystitis, pharyngitis and sinusitis.6–8) CFPN-PI is widely used in patients of all ages including children, because the risk of anaphylaxis has been reported to be lower with third-generation cephalosporins than with other penicillins.9,10)
Women often suffer from urinary tract infections (UTI) during sexually active periods and pregnancy, with an estimated 2–7% of pregnant women suffering from asymptomatic bacteriuria (ASB).11,12) CFPN-PI is recommended as a first-line drug for UTI such as cystitis and ASB in pregnant women by the guide to clinical management of infectious diseases published by the Japanese Association for Infectious Diseases and the Japanese Society of Chemotherapy.13) However, its safety has not been assessed. Pregnant women are generally recommended to avoid exposure to various chemicals including drugs during the first trimester, when fetal morphological abnormalities may occur.14) Czeizel et al. reported that six cephalosporins surveyed such as cephalexin, cefuroxime, cefaclor, cefamandole, cefotaxime and cefoperazone, and cephalexin, cefuroxime statistically showed no teratogenicity in neonates born to women who took them during the first trimester of pregnancy.15) However, they did not assess teratogenicity of CFPN-PI because this drug is used primarily in Japan and Southeast Asia. To the best of our knowledge, there have been only two reports assessing the effects of CFPN-PI exposure during pregnancy on maternal and neonatal toxicity. Nasu et al. reported that long-term treatment with CFPN-PI in pregnant women during gestational weeks 28 to 39 induced hypocarnitinemia in both mothers and naonates.16) Narimatsu and Tamura reported that no congenital anomalies were found in neonates born to women treated with CFPN-PI for premature rupture of membranes after gestational week 36.17) However, there are no reports assessing the incidence of teratogenicity in neonates born to women who took CFPN-PI during the first trimester of pregnancy, when teratogenicity can be induced by drugs.
In Japan, Toranomon Hospital and the Japan Drug Information Institute in Pregnancy at the National Center for Child Health and Development collect information on the type of drugs and duration of exposure, based on the medication history of pregnant women. Previous studies using the combined database of these facilities showed that exposure to domperidone18) or leukotriene receptor antagonists19) during the first trimester of pregnancy was not a risk factor for major malformations in neonates. In this study, we assessed the safety of CFPN-PI by using the combined database to compare the incidence of major malformations in neonates born to women who took CFPN-PI and those who took control drugs during the first trimester of pregnancy.
The data collection was carried out as in previous studies.18,19) Data were obtained from interviews conducted as counseling on drug use during pregnancy provided to Japanese women at Toranomon Hospital and the Japan Drug Information Institute in Pregnancy at the National Center for Child Health and Development. Informed consent for participation in this study was obtained in writing from the individuals at the time of the consultation. All data including pregnancy backgrounds and outcomes were collected from the individual by mail.
ParticipantsPregnant women who took CFPN-PI (CFPN-PI group) or control drugs (control group) during the first trimester of pregnancy were enrolled. The control drugs were selected from medicines that were considered to be non-teratogenic in a previous report by Habermann et al.20) They included acetaminophen, antihistamines reported to have no increased risk of teratogenicity, antibacterial agents such as penicillin, histamine H2-receptor blockers, digestive enzyme preparations, and topical agents such as eye drops and ointments. First trimester exposure was defined as taking the drug between 28 and 97 d after the first day of the last menstrual period (gestational weeks 4 to 13). The exclusion criteria were: (1) a refusal to participate in the study, (2) counseling only for medication for the woman’s partner, (3) no response to the questionnaire on pregnancy outcomes, (4) multiple episodes of counseling during the same pregnancy, (5) did not take either CFPN-PI or the control drugs, (6) taking drugs reported to be teratogenic,21) and (7) no information on medication.
Primary EndpointThe primary endpoint in this study was the incidence of major malformations in singleton births. To assess the teratogenic risk of CFPN-PI, the incidence of major malformations was compared between the CFPN-PI and control groups. Major malformations were defined using guidelines from the European Surveillance of Congenital Abnomalies (EUROCAT),22) which includes essential epidemiological information on congenital anomalies among approximately 1.5 million births per year in Europe. Major congenital abnormalities not included in EUROCAT were diagnosed by a specialist in congenital abnormalities. The incidence of major malformations was analyzed in live-born and single-birth neonates without identified chromosomal abnormalities, and calculated by dividing the number of neonates with major malformations by the total number of neonates in each group.
Secondary EndpointsSecondary endpoints were live birth, stillbirth, miscarriage, abortion, and gestational age. Stillbirth was defined as fetal death after 22 weeks of gestation. Miscarriage was defined as fetal loss before 22 weeks of gestation. Preterm birth was defined as birth after 22 weeks gestation but before 37 weeks.
Statistical AnalysisData on maternal characteristics are shown as mean ± standard deviation (S.D.) or number, n (%). The characteristics were compared between the CFPN-PI and control groups and analyzed using Student’s t-test for continuous variables and Fisher’s exact test for categorical variables. Differences in proportion of major malformations and secondary endpoints were compared between the CFPN-PI and control groups using Fisher’s exact test. Univariable logistic regression analysis was used to calculate the crude odds ratios (ORs) and 95% confidence interval (95% CI) of major malformations and ventricular septal defects in the CFPN-PI group against the control group. Adjusted ORs were also calculated using multivariable logistic regression analysis adjusted for smoking during pregnancy and maternal age, which have been reported to affect the incidence of major malformations.23) In this analysis, the logarithmic odds for the incidence of major malformations as an objective variable could be described by the explanatory variables such as maternal age, smoking habit and taking CFPN-PI with their partial regression coefficients. The partial regression coefficients for taking CFPN-PI could be made the influence of other explanatory variables on the logarithmic odds for incidence of major malformations negligibly small. Because the adjusted OR for taking CFPN-PI was estimated using partial regression coefficients, it can eliminate the influence of maternal age and smoking habit. These analyses excluded missing values, and differences with p-values <0.05 were considered statistically significant. Data analyses used JMP pro version 17.0.0 (SAS Institute, Cary, NC, U.S.A.). The validity of sample size was evaluated by Power and Sample Size Calculation version 3.1.6 (Vanderbilt University, Nashville, TN, U.S.A.). The sample size used could detect a 2.75-fold increase in the incidence of major malformations, relative control group (1.8%), with 80% statistical power.
Ethics StatementThis study was approved by the ethics committee of Hiroshima University (Approval No. E2017-1152) and the National Center for Child Health and Development (Approval No. 1502) in compliance with the Declaration of Helsinki and current legal regulations in Japan. Informed consent was obtained from all participants. Information collected from the questionnaires was entered into the database and de-identified by an information manager. The researchers were therefore unable to identify individuals from the analyzed data.
The flow of participant enrollment was the same as in previous studies.19,20) From the two databases, 13599 cases that met the inclusion criteria and did not meet the exclusion criteria were combined for analysis (Fig. 1). Of these, 11410 were excluded because they either did not use control drugs or also used teratogenic drugs. In total, 2189 cases were included in the analysis, of which 307 were in the CFPN-PI group and 1882 in the control group.
Maternal characteristics are shown in Table 1. The mean age of mothers at the time of consultation was 30.5 ± 4.8 years in the CFPN-PI group and 30.7 ± 4.7 years in the control group. There was no significant difference in age distribution between the two groups. There were also no significant differences in alcohol intake during pregnancy between the groups. However, the proportion of women who stopped smoking after becoming pregnant was higher in the CFPN-PI group (15.0%, 46/307) than the control group (10.6%, 200/1882). There were no significant differences in the proportions of different gravidity, parity and abortions between the two groups.
| CFPN-PI (n = 307) | Controls (n = 1882) | p-Value | |
|---|---|---|---|
| Maternal age (years), n (%)a) | 30.5 ± 4.8 | 30.7 ± 4.7 | 0.662 |
| ≧35 | 61 (19.9) | 409 (21.7) | 0.500 |
| 30–34 | 122 (39.7) | 679 (36.1) | 0.226 |
| <30 | 124 (40.4) | 791 (42.0) | 0.618 |
| Alcohol intake during pregnancy, n (%)b) | |||
| No use | 160 (52.1) | 978 (52.0) | 0.248 |
| Stop before pregnancy | 11 (3.6) | 53 (2.8) | 0.592 |
| Stop after pregnancy | 117 (38.1) | 620 (32.9) | 0.359 |
| Ongoing | 4 (1.3) | 20 (1.1) | 0.773 |
| Smoking during pregnancy, n (%)c) | |||
| No habit | 228 (74.3) | 1400 (74.4) | 0.107 |
| Stop before pregnancy | 5 (1.6) | 49 (2.6) | 0.330 |
| Stop after pregnancy | 46 (15.0) | 200 (10.6) | 0.067 |
| Ongoing | 15 (4.9) | 66 (3.5) | 0.334 |
| Gravidity, n (%)d) | |||
| 0 | 137 (44.7) | 758 (40.3) | 0.168 |
| ≧1 | 169 (55.0) | 1113 (59.1) | — |
| Parity, n (%)e) | |||
| 0 | 168 (54.7) | 926 (49.2) | 0.084 |
| ≧1 | 138 (45.0) | 945 (50.2) | — |
| Previous miscarriages, n (%) | 23 (7.5) | 103 (5.5) | 0.185 |
| Previous abortions, n (%) | 13 (4.2) | 99 (5.3) | 0.576 |
Data are shown as mean ± S.D. or n (%). p < 0.05 was considered statistically significant. S.D., standard deviations of the mean. CFPN-PI, cefcapene pivoxil hydrochloride. The missing values were as follows; a) CFPN-PI (n = 0) and Controls (n = 3); b) CFPN-PI (n = 15) and Controls (n = 211); c) CFPN-PI (n = 13) and Controls (n = 167); d) CFPN-PI (n = 1) and Controls (n = 11); e) CFPN-PI (n = 1) and Controls (n = 11).
The delivery outcomes are summarized in Table 2. The proportion of live births was 92.8% (285/307) in the CFPN-PI group and 94.2% (1772/1882) in the control group. The incidence of preterm birth was 4.2% in the CFPN-PI group and 4.1% in the control group. There was one stillbirth (0.3%), 15 miscarriages (4.9%) and six abortions (2.0%) in the CFPN-PI group, and seven stillbirths (0.4%), 82 miscarriages (4.4%) and 20 abortions (1.1%) in the control group. There were no significant differences in incidence of any of these outcomes between the two groups.
| CFPN-PI (n = 307) | Controls (n = 1882) | p-Value | |
|---|---|---|---|
| Outcome, n (%) | |||
| Live births | 285 (92.8) | 1772 (94.2) | 0.363 |
| Stillbirths | 1 (0.3) | 7 (0.4) | 1.000 |
| Miscarriages | 15 (4.9) | 82 (4.4) | 0.654 |
| Abortions | 6 (2.0) | 20 (1.1) | 0.248 |
| Others | 0 (0) | 1 (0.1) | — |
| Number of infants, n (%)a) | |||
| 1 | 296 (96.4) | 1830 (97.2) | 0.701 |
| ≧2 | 2 (0.7) | 11 (0.6) | — |
| Detail in live births | CFPN-PI (n = 285) | controls (n = 1772) | p-Value |
| Gestational age, n (%)b) | |||
| Full-term | 273 (95.8) | 1471 (83.0) | 0.878 |
| Preterm (<37 weeks) | 12 (4.2) | 73 (4.1) | — |
Data are shown as n (%). p < 0.05 was considered statistically significant. CFPN-PI, cefcapene pivoxil hydrochloride. The missing values were as follows; a) CFPN-PI (n = 9) and Controls (n = 41), b) CFPN-PI (n = 0) and Controls (n = 228).
The incidence of major malformations in singleton births was 2.6% (7/270) in the CFPN-PI group and 1.8% (28/1594) in the control group. No significant increase in major malformations was observed in CFPN-PI group (p = 0.33). In the CFPN-PI group, malformations seen were ventricular septal defect (n = 4), a combination of ventricular septal defect and Wolff–Parkinson–White syndrome (n = 1), tetralogy of Fallot (n = 1) and a combination of pulmonary valve stenosis, peripheral pulmonary stenosis and patent foramen ovale (n = 1). In the control group, malformations were ventricular septal defect (n = 7), polydactyly (n = 3), hydronephrosis (n = 2), anorectal malformation (n = 1) and other serious conditions (n = 15) (Table 3).
| Crude OR (95% CI) | p-Value | Adjusted OR (95% CI) | p-Value | ||
|---|---|---|---|---|---|
| Exposure* | Controls | 1 | 1 | ||
| CFPN-PI | 1.49 (0.64–3.44) | 0.35 | 1.48 (0.64–3.42) | 0.36 | |
| Maternal age (years) | <30 | 1 | 1 | ||
| 30–34 | 2.02 (0.89–4.61) | 0.09 | 2.05 (0.90–4.68) | 0.09 | |
| ≧35 | 2.28 (0.92–5.65) | 0.08 | 2.30 (0.93–5.71) | 0.07 | |
| Smoking during pregnancy | No | 1 | 1 | ||
| Yes | 1.46 (0.34–6.19) | 0.61 | 1.58 (0.37–6.75) | 0.54 |
*The incidence of major malformations in singletons was 2.6% (7/270) in the CFPN-PI group and 1.8% (28/1594) in the control group. No significant difference in major malformations was observed between the two groups (p = 0.33). In the CFPN-PI group, malformations were ventricular septal defect (4); a combination of ventricular septal defect and Wolff–Parkinson–White syndrome (1); tetralogy of Fallot (1); and a combination of pulmonary valve stenosis, peripheral pulmonary stenosis and patent foramen ovale (1). In the control group, malformations were ventricular septal defect (7); polydactyly (3); hydronephrosis (2); anorectal malformation (1); cleft lip and palate (1); cleft lip (1); pulmonary stenosis (1); congenital clubfoot (1); esophageal atresia (1); peripheral pulmonary stenosis (1); endocardial cushion defect (1); complete transposition of great arteries (1); double outlet right ventricle (1); adhesion of scrotum and penis (1); ventricular and atrial septal defect (1); a combination of single ventricle and pulmonary stenosis (1); a combination of ventricular septal defect, pulmonary stenosis and syndactyly (1); a combination of coarctation of the aorta and ventricular septal defect (1); a combination of atresia of ileum and meconium peritonitis (1); and a combination of pneumothorax and peripheral pulmonary stenosis (1). Data are shown as n (%). p < 0.05 was considered statistically significant. Adjusted odds ratio (OR) and 95% confidence interval (CI) were adjusted for all other variables in the table. CFPN-PI, cefcapene pivoxil hydrochloride; OR, odds ratio; CI, confidence interval.
Univariable logistic regression analysis showed that exposure to CFPN-PI was not a risk factor for major malformations (OR: 1.49 [95% CI: 0.64–3.44], p = 0.35; Table 3). We did not analyze whether alcohol intake was a risk factor for the incidence of malformation because there were no women with ongoing alcohol intake among those in the CFPN-PI group whose neonates had a malformation. Multivariable logistic regression analysis adjusted for maternal age and smoking during pregnancy also showed that exposure to CFPN-PI was not associated with the incidence of major malformations (adjusted OR: 1.48 [95% CI: 0.64–3.42], p = 0.36). On the other hand, the incidence of ventricular septal defects including complications in major malformations was significantly higher in the CFPN-PI group (1.85%, 5/270) than in the control group (0.56%, 9/1594) (adjusted OR: 3.38 [95% CI: 1.12–10.2], p = 0.03).
CFPN-PI is used by people of all ages including women who are or may be pregnant. However, whether CFPN-PI is safe for mothers and their fetuses during the first trimester of pregnancy remains unclear. In this study, we showed that the use of CFPN-PI during the first trimester of pregnancy did not increase the incidence of major malformations in singleton births compared with the use of other non-teratogenic drugs (Table 3). Sensitivity analysis also showed that the use of CFPN-PI during the first trimester of pregnancy was not a risk factor for increased incidence of major malformations. We found that the use of CFPN-PI during the first trimester of pregnancy did not affect pregnancy outcomes such as neonatal live births and the incidence of stillbirths, miscarriages and abortions (Table 2).
In this study, the most common malformations were ventricular septal defect, a combination of ventricular septal defect and Wolff-Parkinson-White syndrome, tetralogy of Fallot and a combination of pulmonary valve stenosis, peripheral pulmonary stenosis and patent foramen ovale (Table 3). Of these, ventricular septal defects were the most common, and the incidence was significantly higher in the CFPN-PI group than in the control group. Miyake reported that ventricular septal defects occur at a frequency of 0.38% (3.85 per 1000) in live-born neonates worldwide.24) In addition, the International Clearinghouse for Birth Defects Surveillance and Research reported that it accounted for approximately 10.9% of all congenital anomalies in Japan.25) Previous reports have shown that exposure to cephalosporins during pregnancy might cause higher incidence of ventricular septal defects and anal congenital anomalies.16,26–28) These findings suggest that CFPN-PI may also be a risk factor for incidence of ventricular septal defects. However, data on the incidence of ventricular septal defects may be biased by the number of tests performed because they are easy to detect using echocardiography.24) In this study, we did not analyze the frequency of echocardiography during pregnancy testing. To assess the difference in the incidence of ventricular septal defects between the CFPN-PI (1.85%) and the control groups (0.56%) with 80% power, at least 500 cases each in both groups are required. Further studies are necessary to clarify the causal relationship between CFPN-PI use and ventricular septal defects.
The proportion of women who quit smoking after becoming pregnant was higher in the CFPN-PI group (15%) than the control group (10.6%) (Table 1). Some women continued to smoke during pregnancy although there were no significant differences for this between the CFPN-PI and the control groups. Smoking habits during pregnancy have been reported to increase the incidence of low weight in neonates at birth, which is associated with preterm birth29,30) and major cleft lip and palate malformations.31,32) However, there were no cases of cleft lip and palate in in the CFPN-PI group in this study, and incidence was only 0.11% (2/1882) in the control group, consistent with a previous report that the incidence of cleft lip and palate was less than 0.1–0.2%.33) Our sensitivity analysis showed that smoking during pregnancy was not a risk factor for increased incidence of major malformations (Table 3). The significant difference in the proportion of women who quit smoking after becoming pregnant therefore did not affect the incidence of major malformations.
This study had some limitations. The first is the possible presence of potential confounding factors. The risk of malformations associated with CFPN-PI use was analyzed after adjusting for smoking habits and maternal age using multivariable regression analysis. However, other potentially confounding factors were not adjusted. These include folic acid intake34,35) and treatment of gestational diabetes.36) Thus, it is possible that some pregnant women could have prevented the occurrence of malformations by these factors. We also did not adjust for the condition for which CFPN-PI was prescribed or its severity, both of which can affect the dose and course duration of CFPN-PI. We were also unable to analyze whether alcohol intake was a risk factor for the incidence of malformation in the multivariable logistic regression analysis because there were no women reporting ongoing alcohol intake among those in the CFPN-PI group whose neonates were born with malformations. Furthermore, we collected data on the ongoing or stop of alcohol intake before and after the detection of pregnancy. Because pregnancies are often detected after the absolute sensitive period, the risk of alcohol intake immediately before the detection of pregnancy for the incidence of malformations cannot be ruled out. In this study, we could not assess it because data on alcohol intake immediately before the detection of pregnancy might be included in both ‘stop alcohol intake before pregnancy’ and ‘stop alcohol intake after pregnancy’ groups. Maternal background including a history of teratogenic infections such as cytomegalovirus and rubella,37) was also not confirmed. The safety of CFPN-PI in racial groups other than Japanese people is unclear because this study included only Japanese women. It has been reported that long-term treatment with CFPN-PI in pregnant women during gestational weeks 28 to 39 induced hypocarnitinemia and its hypoglycemic symptom in both mothers and naonates.16) In this study, we could not obtain the cases of neonatal abnormalities with hypocarnitinemia and its hypoglycemic symptom. To clarify the incidence of hypocarnitinemia, laboratory data on liver function, blood levels of electrolyte, carnitine and sugar should be collected in future.
In conclusion, we found that exposure to CFPN-PI during the first trimester of pregnancy was not associated with the incidence of major malformations. This information is clinically important and will be helpful to healthcare providers concerned about selecting CFPN-PI because of insufficient data.
We thank Melissa Leffler, MBA, for editing a draft of this manuscript.
Y.M., T.T., A.M. and H.M. conceived and designed the study. K.H. and N.Y. contributed to data collection. M.G. managed the collected data. Y.K. and R.I. analyzed the data. Y.K. and Y.M. wrote the manuscript. T.Y., T.T. and H.M. revised the manuscript. All authors read and approved the final manuscript.
The authors declare no conflict of interest.
