To the best of our knowledge, this study was the first to comprehensively investigate the prevalence and severity of constipation in relation to the mode of delivery with a validated defecation questionnaire [11]. We showed with multivariable analysis, in which we adjusted for cofounders contributing to constipation, that the prevalence of constipation was higher in women in the vaginal delivery group than in women in the CS group. Furthermore, we showed that the prevalence of specific defecation problems, such as difficulties emptying bowels, straining, feeling of incomplete defecation, not able to defecate daily despite the urge to empty bowels, and manually assisted attempts at defecating, was higher in women of the vaginal delivery group than in those in the CS group. Moreover, we found two additional factors associated with constipation: age and tearing of the perineum.

Our study showed that approximately a quarter of parous women experienced functional constipation, corresponding to previous studies [13, 14]. Mainly due to the different definitions of constipation, the documented prevalence of constipation in parous women ranges widely, from 2.6% to 26.9% [13]. The prevalence of constipation is higher in women than in men [15]. Obstetric injury of pelvic floor musculature and the anal sphincter complex after vaginal delivery may influence the onset of constipation and dyssynergic defecation [16, 17].

Dyssynergic defecation is the inability to coordinate the abdominal and pelvic floor muscles to evacuate stools, which could lead to constipation [18]. There is evidence that dyssynergic defecation, as determined by anorectal physiological testing, is the major pathophysiological derangement in constipation after vaginal delivery [16]. The study by Park and colleagues suggests that dyssynergic defecation might be associated with abnormal anal sphincter muscle function during vaginal delivery and seems to increase the risk of obstetric anal sphincter injuries [16]. Their study did not allow for a comparison between vaginal delivery and CS, possibly because of the small sample size [16]. One could suggest that women with a strong dyssynergic pelvic floor might have a less adaptive pelvic floor during labour and are more likely to be referred for CS. Shi and colleagues found that constipated pregnant women had a CS more often than non-constipated women but could not identify constipation as a risk factor for CS [19]. Our results, however, showed a lower prevalence of constipation in women in the CS group. Previously, pelvic floor dysfunction was shown to be associated with vaginal delivery, which partially explains our findings regarding constipation, even though most studies failed to distinguish between the types of pelvic floor dysfunction [3, 20].

In the vaginal delivery group, we found a higher prevalence of straining, a symptom that correlates with dyssynergic defecation. Besides pelvic floor injury due to childbirth, straining due to constipation can be harmful to the pelvic floor and can contribute to changes in anorectal physiology, which influences defecation habits [21]. Our study design, however, prevented us from knowing whether this was already present before childbirth or before the onset of constipation. Nevertheless, we also observed an increased prevalence of dyssynergic defecation symptoms in women from the vaginal delivery group, such as manually assisted defecation and inability to defecate daily despite the urge to empty bowels, which emphasised the extent of the constipation problem. Furthermore, we found that women in the vaginal delivery group have a significantly higher Renzi score, although not clinically relevant, as the scores were both below 9, which is the cut-off score for obstructed defecation syndrome [12]. This outcome was in line with that of another study [3]. It has also been reported that CS protects against pelvic floor dysfunction [20], except for faecal incontinence [3].

With regard to childbirth injuries causing dyssynergic defecation, Marchand and colleagues showed that anal sphincter tears occurred 2.94 times more often in women who reported dyssynergic defecation [18]. Interestingly, we found that a perineal tear was associated with constipation. However, we could not distinguish the extent of perineal tear in our questionnaire as it was a self-administered questionnaire. As far as we know, however, no studies demonstrated a relationship between constipation and perineal tears.

Constipation is one of the most commonly reported physical health problems after childbirth. The impact of constipation on the quality of life may not be underestimated as it has a major impact on patient’s well-being [1, 5, 22].

Surprisingly, existing research has predominantly focused on faecal incontinence rather than on constipation, even though the effect of constipation on the quality of life is larger [23]. Besides influencing the mother's physical health, it is known that constipation can also disturb the physical health of newborn infants and the attachment process between mothers and their infants [24]. Therefore, it is necessary to be aware of the risk of constipation and its consequences, not only during pregnancy but also after childbirth, especially as previous research conducted among a general Dutch population revealed that 49% of individuals experiencing constipation had never discussed their discomfort with anyone [2]. This underscores the prevailing social discomfort in addressing constipation openly, highlighting the possible undertreatment of constipation by patients and health care providers. Proactively addressing the issue and offering appropriate interventions, with special awareness to women with a history of a vaginal delivery, might prevent the worsening of neglected constipation.

With our study, we cannot conclude any causality between constipation and mode of delivery. However, this was an observational study, and we have shown a higher prevalence of constipation in the vaginal delivery group. Therefore, a prospective study would be invaluable to investigate the causality of constipation in relation to the mode of delivery and perineal trauma.

We examined a Dutch hospital population, which possibly led to a selection bias towards women with higher-risk pregnancies. In the Netherlands, women with a medical indication give birth in a hospital under the supervision of a gynaecologist. Indications include complicated pregnancy or childbirth, wish for pain medication during childbirth by healthy women, the medical condition of the mother or whether the infant may be expected to require medical attention. We know that the CS rate of our study is lower than expected for the Dutch population. However, this can be explained owing to the exclusion of women with a history of both a vaginal delivery and a CS [25]. Currently, approximately 26.9% of Dutch women give birth at home or in the hospital with the help of a midwife [25]. In our study, all women had delivered in the hospital at least once, under the supervision of the gynaecologist. This implies that the health of a part of our sample might be impaired, which may bias our findings. However, the prevalence of constipation was 24.9% in our cohort, which is in accordance with other literature [2].

Another potential limitation is that the general analyses of obstetrical information were composed of the self-reports of participating women, which could lead to memory bias. The medical information was anonymised for the researchers. Consequently, more detailed medical information, such as the degree of perineal tears, could not be obtained. There is, however, sufficient evidence that medical reports corresponded with women’s own reports [26]. Therefore, we believe that the subjective self-reports of the parous women apply to this study. Furthermore, self-reporting is consistent and valid if symptoms are still present at the time of answering the questionnaire [27]. Also, it is unknown whether the higher prevalence of constipation resulted from vaginal delivery, because the situation before childbirth was unknown to us. Finally, the usage of forceps is rare in the Netherlands, and therefore, we did not distinguish between a forceps and a vacuum in our article.

The main strength of this study is the large sample size. To the best of our knowledge, this was the first study to examine the prevalence and severity of constipation in a vaginal delivery group and a CS group using a validated questionnaire. In addition, the DeFeC questionnaire is validated and extensively qualified, and contains questions regarding faecal and bowel function-related symptoms [11]. This allows us to determine functional constipation and other relevant, in this case, gynaecological, medical history [11]. This enabled us to adjust for bowel function-related diseases in our multivariable analysis, which could influence bowel function. Finally, the questionnaire is self administered, possibly reducing the embarrassment the respondents may have felt regarding their faecal health.