Preterm birth is one of the major causes of perinatal morbidity and mortality [1, 2]. Despite the improvements in neonatal intensive care, they are at increased risk of long-term neurodevelopmental disabilities and major complications [3]. Because uterine contractions are the most frequently recognized symptoms of preterm labor, inhibition of contractions with tocolytic agents to prolong pregnancy is the main target of the treatment in women with preterm labor.

Common agents that are used for the inhibition of uterine contractility are beta-agonists, prostaglandin synthase inhibitors, magnesium sulfate, oxytocin receptor antagonists and calcium channel blockers. Due to the major side effects of beta-agonists, calcium channel blockers (CCB) have been utilized as effective tocolytic agents owing to their safety profile. Among calcium channel blockers, nifedipine is the most frequently used agent due to the availability of oral formulation, and lower cost in addition to the minimal maternal adverse effects associated with its use [4, 5, 6]. A Cochrane review including 12 randomized controlled trials concluded that CCBs, particularly nifedipine, reduce the risk of delivery within 7 days of treatment and delivery before 34 weeks of gestation with reduced fetal complications such as respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis when compared to other tocolytic agents [7, 8].

Nifedipine shows its effect by reducing uterine contractility and causing vasodilatation on vascular smooth muscles. From this point of view, we hypothesized that nifedipine might have an adverse effect on the amount of postpartum hemorrhage due to its vasodilatory and contractility properties. The objective of this study is to examine whether or not the exposure to CCBs before delivery is associated with an increased risk of postpartum hemorrhage (PPH) in patients presenting with preterm labor.

There were 246 patients in the control group and 240 patients in the study group. Demographic variables are shown in Table 1. No significant difference was observed in terms of age, BMI (body mass index), gravidity, parity, live birth numbers, weeks of gestation and previous delivery methods between the control group and nifedipine group. However, the presence of PPROM in the nifedipine group was significantly lower than that of the control group (p 0.001). Related to PPROM, we had a chorioamnionitis incidence of 11.2% (n = 31) in the control group and 10.8% (n = 29) in the nifedipine group (p = 0.888). There were no significant differences between the groups in terms of cesarean section indications, labor durations and mean newborn weight. Besides, we had in both groups statistically insignificant 2 cases complicated with atony, which were managed medically or with Bakri balloon insertion.

Both pre- and postpartum hemoglobin and hematocrit levels were similar in the two groups. However, both hemoglobin level and hematocrit decreased significantly in the two groups from the pre- to postpartum period (Table 2). Postpartum platelet count was significantly lower than the prepartum platelet count in the nifedipine group, but was still in normal range.

The nifedipine dose administered at the last 24 hours of the treatment was significantly correlated with the change in the hemoglobin (r = 0.176, p = 0.006) and hematocrit levels (r = 0.139, p = 0.030) form the pre- to the postpartum period (Table 3, Fig. 1).

Fig. 1.

Correlation graphic for related nifedipin use 24 hours before delivery.

No significant correlations were observed between the duration of the labor and medication for 72 hours, medication for 1-week, total nifedipine dosage, nifedipine dosage for the last 24 hours, total nifedipine use in days and time between nifedipine stop to birth (p $>$ 0.05) (Table 4). The only statistic that is different in Table 4 is the p value = 0.028.

As shown in Table 5, we conducted subgroup analyses for various aspects of nifedipine medications, including the sum of nifedipine medications taken over 72 hours, the sum of medications taken over 1 week, the overall total nifedipine dose, and the total nifedipine dose for the last 24 hours before delivery.

Table 5 provides a comprehensive overview of the measures of nifedipine use, including median values, interquartile ranges (IQRs) and minimum–maximum ranges for values for which min–max values were not available for some values within the nifedipine group.

Postpartum hemorrhage is one of the major causes of maternal morbidity and mortality worldwide. That’s why modern medicine tries to find out possible agents for the treatment of postpartum hemorrhage to diminish the blood loss used in the course of labor. Calcium channel blockers are one of these agents frequently used during pregnancy for tocolysis during preterm labor course, in addition to its antihypertensive properties [12, 13, 14, 15, 16].

Calcium channel blockers show their effects by inhibiting the influx of calcium into peripheral arterial smooth muscle cells and uterine smooth muscle cells, which further leads to vasodilatation and relaxation in myometrium [12]. While this effect is beneficial in the management of preterm labor or gestational hypertension; however, following delivery, this effect becomes dangerous as effective uterine contractions are required to compress the uterine vasculature [13]. In the absence of effective contractions, the uterus becomes relaxed and excessive bleeding resulting in postpartum hemorrhage and maternal mortality may occur. Because of the relaxation effect on smooth muscles and uterine muscles, several authors have suggested that exposure to CCBs before delivery may increase the risk of uterine atony and PPH [17, 18].

The use of CCB prior to delivery may theoretically increase the risk of PPH by causing uterine relaxation has not been studied thoroughly. The study by Yang et al. [19] which investigated the amount of blood loss during labor in hypertensive patients taking CCBs found a significant increase in postpartum hemorrhage in patients receiving CCB’s during the active stages of labor compared to subjects in whom CCBs were stopped before the labor. In accordance with Yang’s study, we found a positive correlation between the PPH and the nifedipine dosage received 24 hours before the labor, although there were no significant correlations between the PPH and the nifedipine dosage received 72 hours before the labor. In this cohort study, we did not observe a significant difference in terms of pre- and postpartum hemoglobin and hematocrit levels between control and nifedipine groups; however, when we analyzed the PPH risk according to the time of CCB use in subgroups, usage before the last 24 hours of delivery is likely related to an increase in PPH. A positive correlation is observed between difference in hemoglobin and hematocrit levels from the pre- to the postpartum period and nifedipine dosage for the last 24 hours but not for 72 hours or the dosage received for the last 1 week.

A possible explanation for this might be that CCBs have short half-life and the drug may have been cleared from the blood quickly until delivery, so that their effect easily dissolves in myometrium. However, when nifedipine is received close to the labor, there is not enough time for the drug to be cleared off. The risk of PPH associated with CCBs may results from the direct action on the myometrium. It appears that subjects are vulnerable to PPH promoted by the nifedipine immediately before the delivery. Another large scaled study, which was conducted by Bateman et al. [20] confirms our results. They compared the risk of postpartum hemorrhage in hypertensive patients receiving CCBs and other antihypertensive drugs (methyldopa or labetalol). They found no differences in the risk of PPH between two groups. The possible explanation for the lack of a difference in the PPH rate between the subjects receiving CCBs and other antihypertensive drugs is that Bateman and colleagues included outpatient subjects, and CCBs were probably stopped long before the delivery so that the unfavorable effects of the CCBs on the PPH was not observed in that study.

The main difference of our study from the previous studies is that the study population included in our study was consisting of subjects with preterm labor in contrast to the previous studies conducted on subjects with gestational hypertension. To the best of our knowledge, this is the first study investigating the obstetric adverse effects of (i.e., bleeding effect) CCBs when used for the management of preterm delivery. According to preterm delivery protocols, after loading dose, CCB’s may be given up to 8 times a day in the first 24 hours, and then the dosage is decreased gradually until corticosteroid medication is completed. Because the doses are higher than used in hypertensive states (4 or 6 times a day), the uterine relaxation may be exaggerated because of higher plasma levels soon before delivery.

The lack of direct measurement of the bleeding during and after the labor would provide more accurate information regarding the impact of nifedipine on postpartum hemorrhage. However, data concerning the blood loss was not available in many of the patients. Therefore, we had to use the pre-and postpartum hemoglobin and hematocrit levels to estimate the amount of the postpartum blood loss in the two groups. These results therefore need to be interpreted with caution.

In conclusion, our study showed that nifedipine use for tocolytic purpose may increase hempglobin/hematocrit levels and platelate counts during delivery among women with preterm labor, but this was in an acceptable amount and statistically insignificant. However, other studies may have totally different results, which shows us that this topic needs further investigation. Although not to conclude with our study, obstetricians should be aware with the use of nifedipin and be prepared for greater blood loss and postpartum hemorrhage to prevent maternal morbidity and even mortality [21].

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

HK and AG designed the research study. HK and NK performed the research. SSC, ZAcar, AAA and BU provided help and advice on topic and methods. NK, ZAyt and AG analyzed the data. NK and AG wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.

Kanuni Sultan Süleyman Training and Research Hospital review board, ethics committee or ethical review board approved the study on 19/07/2016 with the number 26817412. All human subjects provided written informed consent with guarantees of confidentiality.

The authors would like to forward sincere thanks to a large team who work together included technical help, writing assistance and departmental head that only provided general support.

This research received no external funding.

The authors declare no conflict of interest.