Systematic Reviews to Inform Practice, May/June 2025

Postpartum hemorrhage (PPH) is defined as a blood loss of 500 mL or more within 24 hours of birth.1 PPH occurs in approximately 1% to 3% of births in the United States.2 Globally, 20% of pregnancy-related deaths are attributed to PPH,1 and in the United States, approximately 11% of pregnancy-related deaths are caused by PPH.3 A factor that may influence the incidence of PPH is the length and course of the third stage of labor, defined as the period from the birth of the newborn to the complete expulsion of the placenta and membranes.4 The expected length of the third stage is generally accepted as 30 minutes.5 In the case of a prolonged third stage, the recommended management is usually manual removal of the placenta.6, 7 Recently, some clinicians and researchers have begun to consider reducing the length of time for expectant management of the third stage of labor and have suggested that manual removal of the placenta is indicated after 20 minutes.8, 9 de Vries10 and colleagues conducted a systematic review and meta-analysis to evaluate the association between the length of the third stage of labor after a vaginal birth and adverse maternal outcomes. They also sought to identify whether the risk of adverse maternal outcomes can be reduced by performing manual removal of the placenta earlier than current clinical practice. The authors conducted a search of 7 databases and screened the reference lists of national guidelines from high-income countries regarding the prevention of PPH. Randomized trials, comparative studies, and prospective and retrospective cohort trials that were published between January 1, 2000, and June 13, 2023, in English, French, German, Italian, and Dutch languages were eligible for inclusion in the review. The primary outcome of interest was adverse maternal outcomes, defined as any complication occurring to the birthing person during or after birth, such as PPH, blood transfusion, intensive care unit admission, and peripartum hysterectomy. The secondary outcome of interest was the risk of adverse outcome as it related to the length of the third stage of labor. The researchers identified 16 articles that addressed the relationships between the length of the third stage of labor and maternal outcomes and 3 articles that evaluated the association between the timing of the manual removal of the placenta and adverse maternal outcomes. One article measured both outcomes of interest, so a total of 18 articles were included in the full review. Included studies were conducted in the United States, Australia, Denmark, the Netherlands, Israel, Switzerland, Sweden, Japan, and Egypt, and one study was conducted across multiple countries; the authors do not indicate how many of the 18 studies were conducted in each nation. The risk of bias was rated as low in 1, moderate in 7, and high in 10 studies. Studies that presented an odds ratio (OR) for the likelihood of PPH were included in the meta-analysis, regardless of bias rating. To examine the association between the length of the third stage of labor and the likelihood of PPH among birthing people with a mixed risk profile when the third stage of labor was measured after 15 minutes, 4 studies were included in the meta-analysis, with a total of 39,324 events or births evaluated. The likelihood of PPH was 5.55 times higher (OR, 5.55; 95% CI, 1.74-17.72) among people who had a third stage of labor that lasted for 15 minutes or longer than in those whose third stage was less than 15 minutes. Another meta-analysis considered hemorrhage in birthing people with a mixed risk profile when the third stage of labor was measured after 30 minutes (5 studies, N = 21,966 events) and found the odds of PPH was significantly higher among those whose third stage was 30 minutes or longer than in those whose third stage was less than 30 minutes (OR, 3.12; 95% CI, 1.83- 5.30). The authors included a variety of additional analyses, some of which do not meet the expectations for meta-analysis due to the number of included studies or the heterogeneity of the outcome measures. The authors also note that the definition of PPH varied among the included studies, with some using a definition of 1000 mL and some using a definition of 500 mL. This is a significant difference when PPH is the key outcome of the analyses. They were not able to reach conclusions about the timing of manual removal of the placenta and the risk of adverse outcomes. The authors of this systematic review and meta-analysis use the findings of their analyses to propose that the third stage of labor should be redefined as lasting less than 15 minutes, although they note that the associations identified here between the length of third stage and PPH do not imply causality. Additionally, they did not find evidence that manual removal of the placenta, the primary intervention to shorten the third stage of labor, was an effective intervention for reducing adverse maternal outcomes. Although it is likely that when a birthing person experiences a longer third stage of labor, they are more likely to experience a hemorrhage, this may be due to subsequent interventions that occur once a prolonged third stage is diagnosed. The findings here do not indicate that a change in clinical practice is indicated, although the eye-catching title published in one of the leading journals of obstetric practice in the United States does have the potential to influence practice. Until further evidence demonstrates that an intervention that can safely shorten the third stage of labor and reduce the risk of PPH, clinicians will be best served by adhering to current practice guidelines for the management of the third stage of labor. Chronic pelvic pain (CPP) is a complex diagnosis with a complex collection of symptoms that cause of physical and psychological suffering.1, 2, 3 Although conditions such as endometriosis, uterine fibroids, interstitial cystitis, and pelvic congestion syndrome are commonly recognized as causes of CPP, many cases of CPP have no clearly defined cause and, therefore, no clearly defined treatment. Studies citing prevalence rates vary in their reporting of estimates, ranging from less than 5% to nearly 30%;2 these studies often fall short of representing the full scope of what contributes to patient suffering with CPP.2 Attempts to understand CPP are limited by differing definitions of CPP, the number of conditions that can manifest as CPP, and delays in diagnosis or underreporting.1, 2 Starzec-Proserpio et al4 conducted a systematic review and meta-analysis of 38 randomized clinical trials (RCTs) to determine the effectiveness of nonpharmacological conservative therapies on CPP without clearly defined pathology. To be included in the systematic review, RCTs had to investigate active nonpharmacologic conservative therapy as a primary independent variable and have undergone publication in a peer-reviewed journal. Studies that were excluded consisted of those that investigated endometriosis, pelvic girdle pain, genitourinary syndrome of menopause, irritable bowel syndrome, and Hunner-type interstitial cystitis as primary inclusion criteria. Data were collected on pain indicators; sexual, physical, and psychological function measure; health-related quality of life (HRQOL); pelvic symptom severity; pelvic floor muscle function and morphometry; and perceived improvement. The authors also included adverse events and study attrition in their analysis. Sampling for the selected studies included adult women with pelvic pain with a duration of at least 3 months and lacking a defined pathology or known disease. When study participants had co-occurring conditions, the study was included in the review as long as CPP was the primary outcome and focus of the study. The 38 RCTs were published between 2001 and 2023 and contributed data on 2168 women, with individual studies ranging in sample size from 15 to 212 participants. They represented a variety of geographical regions: North America (n = 14; 37%), Europe (n = 12; 32%), South America (n = 4; 11%), the Middle East (n = 4; 11%), Asia (n = 2; 5%), Oceania (n = 1; 2.5%), and Africa (n = 1; 2.5%). The nonpharmacological therapies studied in the RCTs comprising the systematic review included acupuncture (n = 5), psychological interventions such as cognitive behavioral therapy and mindfulness (n = 10), tissue-based therapies such as massage, pelvic floor biofeedback, and myofascial techniques (n = 18). Multimodal physical therapy (n = 7) was also examined and included interventions such as education, pelvic floor exercises, and self-management strategies within the scope of physical therapy. Control conditions were inert treatment (no treatment, waiting list, placebo, usual care, and leaflet conditions) and nonconservative treatments (pharmacologic or surgical treatment). Pharmacologic and surgical comparators included amitriptyline, topical lidocaine or corticosteroid, anesthesia injections, and vestibulectomy. Outcomes of interest included pain indicators, sexual functioning, psychological function, HRQOL, physical function, and pelvic symptom severity. The majority of RCTs examined vulvodynia (n = 16) and CPP (n = 13); other diagnoses included interstitial cystitis/bladder pain syndrome (n = 5), dyspareunia (n = 2), genito-pelvic pain disorder (n = 1), and urogynecological pain (n = 1). All 38 studies reported on pain outcomes. Those that assessed multimodal physical therapy (n = 7; 476 participants) showed significantly lower pain intensity immediately after the intervention (standard mean difference [SMD], −1.69; 95% CI, −2.54 to −0.85). Five studies (381 participants) reported intermediate-term effects from data collected between 12 and 36 weeks after intervention, with continued lower pain intensity over time (SMD, −1.82; 95% CI, −3.13 to −0.52). After the authors transposed the pain outcome data to a standard 0 to 10 pain scale, a larger improvement in pain intensity was reported for both short-term (SMD, −2.87; 95%, CI −4.32 to −1.45) and intermediate-term measures (SMD, 3.09; 95% CI −5.32 to −0.88). For acupuncture and psychological interventions, effects on pain intensity were not significant. Starzec-Proserpio et al4 concluded that the data on tissue-based monotherapies were too heterogenous and inconsistent to perform a meta-analysis on pain outcomes. Similarly, a meta-analysis could only be performed for psychological approaches because data for multimodal physical therapy and acupuncture were deemed as too heterogenous for a meta-analysis. Small, short-term improvements in sexual functioning after psychological interventions were detected in the data from 4 studies (511 participants; SMD of −0.28; 95% CI −0.52 to −0.04). Once data were transposed to a single validated instrument (the Female Sexual Functioning Index), a larger short-term improvement in sexual functioning was detected (SMD −1.95; 95% CI −3.63 to −0.28). RCTs examining effects of tissue-based approaches (n = 4) on sexual function did not demonstrate significant differences between groups. Although data about effects of multimodal physical therapy and acupuncture interventions were not included in the meta-analysis, overall findings showed a trend toward some short-term improvements in sexual functioning. Meta-analyses were not performed for psychological function (n = 18), HRQOL (n = 9), physical function (n = 5), and pelvic symptom severity (n = 6) for the previously stated reasons. When examining overall trends, there was improvement in some aspects of psychological function and HRQOL after multimodal physical therapy and tissue-based interventions. Limitations of this meta-analysis included differences in strategies within the categories of outcomes that were studied. This presented a barrier to conducting a meta-analysis of data for several outcomes. Also, assessing various subtypes of CPP limited the number of RCTs available to examine each condition individually, especially because some RCTs did not specify the subtype of CPP that was being studied. The authors attempted to address this limitation in the definition of CPP subtypes that comprised their inclusion and exclusion criteria for the meta-analysis. Overall, the meta-analysis demonstrated benefit from a variety of nonpharmacological therapies and strengthens the call for additional analyses to more closely examine approaches to improving outcomes in patients with CPP. Although clinical practice guidelines for the management of CPP can cause uncertainty about decision-making, acknowledging that patients’ symptomatic experiences are multifaceted can prepare both parties to anticipate multifaceted therapeutic approaches. As demonstrated by Starzec-Proserpio et al,4 single interventions cannot adequately target the myriad impacts on physical, psychological, and sexual well-being in their counseling with patients.5 In their analysis, the focus on CPP without a clear cause validates the uncertainty that both patients and providers may experience while formulating an effective treatment plan. It is notable that multimodal physical therapies revealed the most consistent improvements, supporting the need for comprehensive approaches that address a variety of pathways. Whether a patient's CPP is with or without a clearly defined cause, pharmacologic and surgical management options are limited and may be associated with a variety of unwanted effects, making exploring nonpharmacological interventions early in the diagnostic process a key component of care. Gynecologic clinicians should implement ongoing assessment of the impact of CPP and the effects of interventions during follow-up visits by administering validated tools that assess pain severity, sexual function, quality of life, and psychological coping. Because clinical practice guidelines may not adequately feature conservative treatment approaches, clinicians should consider incorporating more than one clinical practice guideline when planning care for patients with CPP.5 In many inpatient settings in the United States and other Western nations, laboring patients are not permitted access to food or oral intake beyond water or ice chips. This practice grew out of concern for the risk of aspiration pneumonia and its associated sequelae and has been common practice since the 1940s.1, 2 Because of changes in contemporary anesthesia practices, the prevalence of anesthesia induced aspiration pneumonia is estimated at a rate of less than one in one million.3 The most recent American College of Obstetricians and Gynecologists practice guidelines regarding management of labor endorse the use of oral hydration to meet hydration and caloric needs.4 The American College of Nurse-Midwives, Midwives Alliance of North America, and National Association of Certified Professional Midwives consensus statement on supporting healthy and normal physiologic childbirth also endorses the encouragement of both food and drink in labor.5 Some authors have hypothesized that restricting oral intake in labor may have an impact on the length of labor and mode of birth.2, 6 Therefore, Fagundes and colleagues conducted a systematic review to evaluate the effect of food and drink as compared with conventional management on the duration of labor and perinatal complications.7 The authors searched 4 databases and included articles published from inception through December 2023. Randomized control trials that compared oral intake with fasting on the duration of labor and that were published in English, Spanish, or Portuguese were eligible for inclusion in the review. Secondary outcomes that were also evaluated were the use of epidural anesthesia, the mode of birth, 5-minute Apgar score less than 7, regurgitation, nausea and vomiting, and Mendelson Syndrome. Six articles were identified as meeting the inclusion criteria. The included studies were conducted in the Netherlands (n = 2), Iran (n = 2), the United States (n = 1), and the United Kingdom (n = 1), with a total of 3333 participants. Studies were evaluated for bias using the Cochrane Handbook of Systematic Review of Interventions as a guide;8 3 were found to have a low risk of bias, and 3 were considered to have an unknown risk of bias. The oral intake that was studied varied across the included studies; 2 used date syrup, 2 used carbohydrate drinks, 1 used a protein-rich drink, and 1 used unrestricted food intake with low fat and residue content. In all included studies, participants in the control group were allowed only water or ice chips. Among these studies, only one demonstrated statistically significant different lengths of labor between the intervention group and the control group. In this study, participants who received date syrup or saffron-honey syrup had reduced lengths of first, second, and third stages of labor when compared with the control group (P < .01). Among secondary outcomes, there were no statistically significant differences between the intervention groups and the control groups in the incidence of instrumental births (n = 4 studies). There were also no statistically significant differences in the rate of epidural analgesia (n = 3 studies), 5-minute Apgar scores (n = 2 studies), or rates of vomiting (n = 3 studies). One included study demonstrated an increase in cesarean birth in the oral intake group (intervention n = 21, control n = 7, risk ratio, 2.91; 95% CI, 1.29-6.54). The other 5 studies demonstrated no difference in rates of cesarean birth between the intervention and control groups. The findings of this systematic review of randomized control trials that compared oral intake versus fasting in labor demonstrated very few statistically significant differences between the 2 groups. A less restrictive diet did not have a negative influence on outcomes, except for a higher rate of cesarean birth in one small study. Taken as a whole, this review helps build the case for the safety of oral intake in labor. Further research using larger sample sizes would help to advance support for the practice. For clinicians who want to support patients’ comfort in labor, oral intake appears to be safe and is in line with current practice guidelines.