Relative risk of postpartum maternal morbidity among cesarean-delivered women: A multisite prospective cohort study in a large district, India.

Globally maternal mortality is declining. Some states in India have achieved the target mentioned in Sustainable Development Goals. Hence, maternal morbidity needs more attention. Maternal death is the tip of the iceberg; for every maternal death, 20–30 women suffer from morbidity, but it may be underestimated.[ 1 2 ] Precise estimates and consequences of postpartum maternal morbidity are obscure. World Health Organization’s (WHO) definition of maternal morbidity is conceptual and difficult to measure.[ 3 ] International Classification of Diseases-10 has rejuvenated the “late maternal death” term, extending the period to 1 year. However, WHO did not incorporate similar changes into maternal morbidity. WHO, the Government of India, etc., has given guidelines for severe acute maternal morbidity, but not for mild/moderate morbidities. Morbidity studies are comparatively rare. Maternal morbidity’s sequelae are often permanent and undermine women’s normal functioning.[ 4 ] These sequelae can affect women’s physical, mental, or sexual health.[ 5 ] Antenatal morbidities are noticed more often than postnatal because antenatal visits are more frequent. Cesarean sections generate considerable maternal mortality and morbidity. The risk for hysterectomy, bleeding, infection, thrombosis, and postpartum depression are higher after a cesarean than vaginal birth. Although WHO justifies a maximum of 15% of cesarean sections globally, including in India, the proportion of cesarean sections is increasing. Secondly, postpartum morbidities are usually mild to moderate and remain neglected. Women tend to ignore the complaints and may not go to the hospitals. The primary caregivers, unless ask for the complaints, the morbidity remains un-noticed. Most Indian studies are single-center and small. Hence, ascertainment of postpartum risk associated with cesarean section needs exploration. To calculate the relative risk of maternal morbidities occurring 6 months postpartum among cesarean-delivered women compared to vaginally delivered in Pune District, India. To calculate the relative risk of maternal morbidities occurring 6 months postpartum among cesarean-delivered women compared to vaginally delivered in Pune District, India.

Among cesarean-delivered women in selected hospitals, many were excluded due to their residence, many were unwilling to come for frequent follow-ups, and some were due to language problems. At any follow-up visit, the proportion of loss to follow-up was <10% [ Figure 2 ]. The mean age was 24.36 ± 3.87 years and ± 23.96 years of cesarean-delivered women and vaginally delivered, respectively ( t = 2.93; P = 0.0035). The proportion of women having education 12 th standard and higher was 46% among cesarean delivered and 38% among vaginally delivered (χ 2 = 21.18; P < 0.0001). Among cesarean-delivered women, the proportion of yellow ration cards (95%) was lesser than vaginally delivered (97%) (χ 2 = 6.33; P = 0.012). The parity and occupation distributions were similar in both groups. We observed major intra-operative complications in about 2% of women in the cesarean section group. Only about one-fifth of cesarean sections were planned. Common indications were previous lower segment cesarean section (LSCS), fetal distress, oligohydramnios, non-progress of labor, and mal-presentation. Obstetricians performed episiotomy on more than half the number of women. About 2% of the vaginally delivered women had excessive vaginal blood loss (>500 ml) and about 1% of the cesarean-delivered women had it (χ 2 = 9.32; P = 0.0023). However, the total blood loss, including during cesarean section and vagina, was equal in both groups. Table 1 gives details of severe acute maternal morbidity and its consequences. The proportion of mild or moderate complications during hospital stay was about 4% among cesarean delivered and 1%among vaginally delivered women (χ 2 = 12.64; P = 0.0004). The common complications, irrespective of mode of the delivery, were thrombophlebitis, breast engorgement, fever, and wound gapping requiring re-suturing. The morbidities/problems varied from 84% at the first visit among cesarean-delivered women to 62% at the third visit among vaginally delivered women. Nevertheless, in general, the cesarean-delivered group of women always had more problems. Apart from the complaints during first follow-up visits, among non-diabetic women, about 8% required blood sugar estimation. About 9% of women had moderate or severe hypertension in the cesarean group, while only 6% had moderate or severe hypertension among the vaginally delivered (χ 2 = 11.67; P = 0.0006). Anemia was similar in both groups. Physical examination and interrogation noticed that some women had calf pain, and a few had hemorrhoids regardless of the mode of delivery. About 2% vaginally delivered women complained of jaundice in the postpartum period and only 0.3% from the cesarean group (χ 2 = 18.40; P < 0.0001). On examination, the proportion of surgical site infection was about 1% and 0.1% among cesarean and vaginal delivered groups, respectively (Fisher’s exact test; P = 0.002). At the second follow-up, moderate and severe anemia was present in more than 50% of women among vaginally delivered women and about 45% of among cesarean-delivered women (χ 2 = 4.79; P = 0.03). Table 2 gives relative risks of the morbidity indicative symptoms at each follow-up visit. Italic and bold numbers in the table indicate infection. Relative risk of complaints among women who underwent cesarean and vaginal delivery, Pune district, India, 2017-18 Bold and Italic indicate infection; * Actual number of episiotomies plus perineal tear repair plus an extension of episiotomy incision was 921. # Discharge and/or induration. RR=Relative risk Table 3 gives adjusted relative risk. Surgical site complaints had a higher adjusted relative risk among cesarean-delivered women at the first two follow-up visits. Breast engorgement/mastitis, urinary incontinence, and fecal incontinence had significantly higher adjusted relative risk among cesarean-delivered women at 6-month visits. Adjusted relative risk of complaints among women who underwent cesarean and vaginal delivery, Pune district, India, 2017–18 * Actual number of episiotomies plus perineal tear repair plus an extension of episiotomy incision was 921; # Discharge and/or induration The differences between the two groups regarding self-care and baby-related activities were inconsistent during follow-up. Overall, about 10% of women resumed family-related activities at 4 weeks. However, the proportion was higher among vaginally delivered women than cesarean-delivered women (χ 2 = 23.60; P < 0.001). At 6 weeks, the overall proportion of resuming family-related work was about 25–30%. However, similar to 4 weeks the proportion was higher among vaginally delivered women (χ 2 = 13.23; P = 0.003). Figure 3 gives the details of components of family-related activities. Resumption of family activities (%) after delivery, Pune District, India, 2017-19

It was a large multisite study, and the women were followed for 6 months. The authors did not include any non-teaching private hospital for three reasons; their contribution to the total number of cesareans was meager, their large number would have comprised the quality of supervision, and their cooperation was uncertain. The authors considered women’s complaints as morbidities, which may be subjective, but many studies had similar inclusion. It is well known that many patients avail services from the private sector. The enlisted complaints can be asked/sought in non-hospital settings also. The relative risk of severe acute maternal morbidity in the form of blood/products transfusion and admission to ICU/HDU among women who undergone cesarean section was 2.59 and 4.33%, respectively. During follow-up to 6 months, at some point, the adjusted relative risk of surgical site problems, lower abdominal pain, breast engorgement/mastitis, urinary incontinence, and, among cesarean-delivered women, was higher. WHO has expressed an urgent need to pay serious attention to maternal morbidity. Although maternal morbidities during hospitalization have declined, the rate of LSCS has increased. The global estimated number of episodes from five leading direct obstetric causes is high. Our focus was beyond near-miss cases like one study,[ 2 ] virtually the present study was overarching. Community-based studies reported morbidity ranging from 50 to 80%.[ 7 8 ] In India, national-level surveys report higher delivery complications than postnatal complications. The morbidity includes all complications and their sequelae.[ 9 ] High morbidity is known in India.[ 8 10 ] In the present study, the order of the common six morbidities in both groups for the first two follow-up visits was the same. At the third follow-up, after 6 months, many symptoms vanished. The reason for the overall large proportion may be proactively seeking information about several complaints. The authors observed that many women complain of weakness, which may be a reflection of anemia, which in the present study was about 50%. National surveys have also reported a similar magnitude.[ 11 ] Secondly, many have low backache, which may be due to micronutrient deficiencies, especially vitamin D. Breastfeeding women are known to have vitamin D deficiency. However, in the present study, treatment seeking for various reasons was low. Common maternal morbidities are postpartum hemorrhage, pre-eclampsia, eclampsia, severe abortion complications, and puerperal sepsis. The proportions of various morbidities during different phases may be as follows: antenatal 36.44%, intra-natal 7.55%, and postnatal 26.44%.[ 8 ] Usually, among non-severe, direct morbidities are more than indirect morbidities. A systematic review infers that postpartum hemorrhage is the commonest among direct morbidities, followed by pre-eclampsia, abortion complications, eclampsia, and gestational diabetes, whereas indirect morbidities are mental health and infections.[ 12 ] The contribution from cesarean deliveries to severe maternal morbidity may be about 37%.[ 13 ] After cesarean febrile morbidity, intra-operative surgical complications, excessive blood loss, endometritis, wound infection, anemia, and urinary tract infection (UTI) are common.[ 14 15 ] The postnatal complications (24.2%) are almost double the intra-natal complications (12.5%), and both are more in an emergency cesarean.[ 14 ] We witnessed only 1.86% of intra-operative complications during cesarean sections. While in hospital, common postnatal morbidities may be episiotomy, pelvic trauma, maternal infection, postpartum hemorrhage, or laceration. During later period, based on responses from women common morbidities with wide ranges include fever,[ 8 16 17 ] excessive bleeding/Post-postpartum hemorrhage (PPH),[ 7 17 ] backache,[ 8 16 18 ] lower abdominal pain,[ 16 ] severe anemia,[ 7 19 ] weakness, delayed milk secretion, pain in stitches[ 8 ], perineal pain,[ 16 18 ] severe pre-eclampsia,[ 19 ] single infection,[ 7 ] foul-smelling discharge, burning micturition/UTI,[ 18 20 ] genital infection, perineal tear, prolapse,[ 20 ] and bleeding per rectum.[ 18 ] Almost all studies noted a better quality of life among women after vaginal deliveries than cesarean deliveries, particularly among primiparas[ 21 24 ] but not in multipara.[ 21 ] The present study confirmed the higher relative risk of morbidities among cesarean-delivered women at some point of follow-up, as shown in Table 2 . Most of the studies did not calculate morbidities’ relative risk or odds ratio. The risk for various morbidities may be almost two to three times among cesarean deliveries than vaginal.[ 18 25 ] Like other studies, the present study also pointed out two to four times higher risk of severe morbidities among the cesarean group than vaginal.[ 19 25 ] An elective cesarean section carries an increased risk of maternal mortality and several morbidities and increases further for cesarean without proper indication. All morbidities are more in emergency cesarean than elective.[ 6 ] Even in a prospective study, most complications were more in cesarean than vaginal and more in an emergency than elective. However, we did not find any difference between an emergency and elective cesarean section. The primary postpartum hemorrhage is more minor among cesarean-delivered women than vaginally delivered, but about secondary postpartum hemorrhage, there is uncertainty. The risk of secondary hemorrhage may be higher among cesarean-delivered women than virginally delivered.[ 26 27 ] But in one study, contrary results were observed.[ 28 ] A review study has reported higher hemorrhage rates among cesarean deliveries than vaginal[ 29 ]; notably, an emergency cesarean may have more severe hemorrhage than an elective.[ 6 30 ] Sometimes, hemorrhage may be extensive, requiring hysterectomy.[ 25 ] But one study has not observed any difference.[ 31 ] One large study found that cesarean sections had a reduced risk of severe postpartum hemorrhage.[ 32 ] Our findings suggest that fever and burning micturition may be non-specific indicators of infection as they continued for more than 6 months. In the past, a high proportion of postpartum fever in field-level surveys was a common observation. Several studies have shown that puerperal infection,[ 25 29 30 33 34 ] surgical wound infection,[ 6 25 33 ] endometriosis,[ 34 ] pneumonia,[ 34 ] and UTI[ 35 ] are more among women undergone cesarean delivery. Usually, emergency cesarean leads to more complications than elective, including severe infections.[ 30 ] Similarly, UTI,[ 35 ] wound infection, and puerperal/pelvic infection are more in an emergency cesarean than vaginally delivered.[ 6 ] But in one study the infection rate was similar among cesarean-delivered and vaginally delivered women.[ 31 ] Conversely, UTIs may be more among vaginally delivered women,[ 20 ] which we also observed in the first follow-up. Anemia, anesthetic complications, cardiac arrest, internal artery ligation, thromboembolism, bladder injury, rupture of the anal sphincter, hysterectomy, and obstetric hematomas are also more common among cesarean-delivered women than vaginal ones.[ 6 25 29 33 ] Usually, they are observed more among emergency cesarean-delivered women than elective. The present study reports a higher incidence of urinary incontinence among them. Only the prolapse occurs more frequently in vaginal than cesarean deliveries.[ 20 ] The present study substantiates the higher occurrence of severe acute maternal morbidities among the cesarean group. Blood transfusion is more frequently required in cesarean-delivered women,[ 34 ] they are more often admitted to ICU. Almost all morbidities are more among cesarean-delivered women than vaginal ones except hemorrhage requiring blood transfusion.[ 25 ] Hospital stay is significantly longer among cesarean-delivered women than vaginally.[ 25 31 ] This is due to the operative procedure and more morbidities and leads to higher expenditure. Besides maternal medical morbidities, women suffer from many social consequences like decreased social support, change in relationships, child-caring problems, family structure, domestic violence, and economic effects like productivity loss.[ 5 ] The present study noted differences in the resumption of family-related activities. We did not include any non-teaching private hospitals and did not differentiate between follow-up at the original institute, any other institute, camp, home visit, or telephonic. We did not include postpartum depression. Inclusion of the women’s complaints may be subjective bias. The overall results may not apply to the general population, as the proportion of cesarean section is not 50%.

Many women have some postpartum health problems, particularly cesarean-delivered women. The obstetricians should rationalize the cesarean section rate. Health care workers should insist on cesarean-delivered women for regular follow-up and assess for pain, induration/discharge at the surgical site, complications like urinary incontinence, and breast engorgement/mastitis. The United Nations Population Fund supported this work through the State Health System Resource Centre, Government of Maharashtra. There is no grant number. It was a Memorandum of Understanding between State Health System Resource Centre, Pune, and Bharati Vidyapeeth Deemed University Medical College, Pune, 26/05/2017. There are no conflicts of interest.

It was a prospective cohort study. We selected all non-teaching hospitals from the public sector, performing five or more cesarean sections per month, one government teaching hospital, and one private teaching hospital. Figure 1 gives the geographical locations of the participant hospitals. Blocks and participant hospitals, Pune District, India, 2017-18 Pune District is on the western side of India, having a population of 9,429,408 (Census, 2011) scattered in 13 blocks. The study duration was from July 1, 2017 to December 31, 2018. The authors enrolled women from September 1, 2017 to March 31, 2018. We considered the cesarean section as exposure. We enrolled all women at the delivery time and completed the initial assessment. The women were requested to come for follow-up at 4 weeks, 6 weeks, and 6 months after delivery. The first follow-up started on October 16, 2017, and the third follow-up lasted until December 31, 2018. The authors trained a team of an obstetrician, in-charge nurses of the postnatal ward, and unit-wise post-graduate senior residents in a teaching hospital. One member of the team collected the information by interviewing the women. The authors collected the forms from rural areas monthly or during a visit and weekly from corporation areas. A research coordinator (a public health specialist) and two medico-social workers under the authors’ supervision monitored the data collection. The study enlisted all cesarean-delivered women in these hospitals. Women who were residents of the Pune district and willing for follow-up visits were eligible. We excluded critically ill women, women having a major psychiatric illness, and women who do not understand Marathi, Hindi, or English. For each cesarean delivered, one age (±2.5 years) and parity matched vaginally delivered woman was enrolled in the comparison group. The research coordinator and medico-social workers ensured a high follow-up by telephonic contact with women. Auxiliary Nurse Midwives visited the homes of defaulter women and referred them to the hospitals. Authors conducted special camps to persuade women to attend the hospitals. Women not attending even the camps and not available at home were interviewed on the phone by coordinator/medico-social workers. Figure 2 gives the details of followed participants. Follow-up of delivered women in Pune District, India, 2017-18 Mode of delivery was the independent variable. The study also recorded age, parity, education, occupation, and income. In Maharashtra, families with a yearly income ≤₹15,000 are provided yellow ration cards, ₹15,001–99,999 are provided orange ration cards, and having an income of ₹100,000 or more are offered white ration cards. We considered ration cards as a proxy for income. Families with an income of less than ₹15,000 are considered below the poverty line and are entitled to free services. We considered any fever, surgical wound infection, vaginal discharge, and burning micturition indicators of infection. We also recorded hospital stays and the direct cost of hospitalization. Admission to the intensive care unit (ICU) or high dependency unit (HDU), blood/blood products infusion, anesthetic complication, and emergency hysterectomy constituted severe acute maternal morbidity.[ 1 ] After discussion with senior faculty from Obstetrics and Gynecology Department, the authors decided to ask specifically about the common 16 symptoms/complaints apart from usual interrogation and examination [ Table 1 ]. The obstetricians examined the women and measured their hemoglobin (at 4 and 6 weeks). Another dependent variable was the resumption of daily activities, categorized into two groups; personal/baby-related activities like bathing, feeding, nursing, etc., and family-related activities like cooking, cleaning the house, washing clothes, and usual outdoor working. Severe acute maternal morbidity during hospitalization, Pune District, India, 2017-18 * Only 241 in cesarean and 242 vaginal. LSCS=lower segment cesarean section; CI=confidence interval; ICU=intensive care unit; HDU=high dependency unit; SD=standard deviation The authors prepared four forms; one for baseline information and three for follow-up visits. The forms were validated by experts and translated into the local language. The calculated sample size was 3,102 (1,551 in each group), intentionally considering the low prevalence of puerperal fever/pelvic infection after cesarean section as 1.5 and 0.5% after vaginal delivery[ 6 ] with 80% power and 95% confidence. We used the Statistical Package for Social Sciences (Version 25) and STATA (15.1). The P < 0.05 was considered significant. The number of women lost to follow-up, non-responses, and illegible handwriting was clubbed together and not considered for analysis. The Chi-square test, t-test, and Fisher’s exact test were used for morbidities during hospitalization and women’s examinations. We calculated the relative risk with a 95% confidence interval for post-discharge symptoms. We used a generalized linear regression model with a log link and binomial distribution to calculate the adjusted relative risk and for variables (including demographics) found to have significant relative risk in bivariate analysis. The Institutional Ethical Committee (DCGI Regd. No. ECR/313/MH/2013/RR-16) provided ethical approval for this study. The authors obtained informed written consent from all the women for participation and publication.