Postoperative complications and risk factors among women who underwent caesarean delivery from Northern Tanzania: a hospital-based analytical cross-sectional study

Background: Caesarean section is one of the most performed surgeries across the whole globe. However, just like any major surgery, CS is associated with complications like haemorrhage, surgical site infection, Venous thromboembolism and anaesthesia related complications. These associated morbidity and mortality have been there despite improvements in blood transfusion services, the use of prophylactic antibiotics, anaesthesia and surgical techniques. This study aims to determine the rate of caesarean section complications, characterize their timing and identify risk factors to maternal complications among women delivered at KCMC hospital. Method: A cross-sectional analytical study of 386 women who delivered by CS at Kilimanjaro Christian Medical Center(KCMC) hospital from October 2018 to May 2019 was done. Data were obtained from the delivery register, patients, clinical and anaesthesia notes using a structured pre-tested questionnaire. Social-demographic characteristic, past medical, obstetric, surgical and family social history were sought from every study participant. Prevalence of complication was computed as a proportion of women with one or more complications out of all the women delivered during the study period. STATA version 13 was used for multivariable analysis to determine independent risk factors for the most prevalent complications. Odds ratio and 95% CI has been used to make an association. Whereby P-value <0.05 is considered statistically signicant. Results: Out of 386 deliveries,106(27.5%) had one or more complications. The most common immediate complications were anaesthesia related, blood transfusion , postpartum hemorrhage(PPH), hysterectomy and ICU admission. While most common early complications (24 -72 hours of the operation) were puerperal sepsis, anaesthesia related complications, blood transfusion and ICU admission. Longer duration of surgery signicantly associated with all the complications (aOR 95% CI 1.02-8.50), Grand multiparity was signicantly associated with blood transfusion (aOR 7.0, 95% CI and PPH (aOR 6.4, 95%CI 1.5-24.24) while pre-operative anaemia was signicantly associated with blood transfusion (aOR


Background
Cesarean section is a surgical procedure performed when a vaginal delivery is not possible or safe, or when the health of the mother or the baby is at risk. Currently, 18.6% of all the deliveries worldwide are through Caesarean section. The lowest rates of CS are being seen in the less developed regions and highest rates seen in the highly developed regions at the rates of 6% and 27.2% respectively.(1) This is irrespective of the World health organization (WHO) recommendation of CS rate of about 10-15% which seems to be the optimal range for targeted provision of this life-saving intervention. (2) The rate of caesarean section deliveries has been on the rise yearly irrespective of the lack of evidence supporting maternal and perinatal bene ts. According to Global, Regional and National estimates on the increasing trends of CS from 1990-2014, global average CS rate increased by 12.4% (from 6.7%-19.1%) with an annual increase rate of 4.4% (3) Caesarean section is a life saving surgical procedure that is mostly encountered in obstetric practices and it is one of the signalling functions of a functioning Comprehensive Emergency Obstetric Care (CEmOC),CS can be done as a planned or emergency deliveryHowever, "Every effort should be made to provide caesarean sections to women in need, rather than striving to achieve a speci c rate" (4) This marked increase in the rate of primary elective CS without medical indication and has stimulated debates in the medical community over the past decade whereby, this trend has been based on the evidence that a planned CS reduces the incidences of maternal pelvic disorders that include: faecal,urine incontinence and pelvic organ prolapse, intrapartum hypoxia, intrapartum death, birth trauma, avoidance of anxiety, pain that is associated with vaginal delivery and fear of litigation (5,6).
The most common indications for caesarean section include failure to obstructed labour,fetal distress, fetal malpresentation, cord prolapse, fetal macrosomia, antepartum hemorrhage, uterine rupture (7,8).Other indications are Some fetal bleeding disorders,prior uterine scar that entered the endometrial cavity, a maternal infection that has a signi cant risk for vertical transmission during labour and delivery, maternal request due to fear of labour and available evidence that delivery by caesarean section decreases risks for pelvic organ prolapse, faecal and urine incontinence (5,9).
Severe maternal morbidity and mortality are mostly encountered among women who underwent an emergency caesarean section as compared to the planned caesarean section.(9,10) Preeclampsia, eclampsia, increasing maternal age, preterm delivery especially when the gestational age is <30 weeks, previous abdominal surgery (other than CS), type 1 diabetes, multiple pregnancy and uterine rupture have also been documented as risk factors to caesarean section related complications (11,12). Maternal obesity (BMI >35kg/mm 2 ) has also been found to be highly associated with wound complications whereby vertical skin incisions have 7 times higher chance of wound infection and wound separation compared to transverse skin incision (13).
Maternal morbidity, mortality and long-term sequel post caesarean section are di cult to deny, related complications encountered are blood transfusion, endometritis, super cial surgical site infection, haemorrhage, injury to the pelvic organs, thromboembolic disorders and anaesthesia related complications (5,14,15).
In long term complications, caesarean section delivery is associated with abnormal placentation, uterine rupture in the subsequent pregnancy, adhesions, unexplained stillbirth, preterm birth especially when the previous CS was done during the second stage of labour and scar complications which includes postmenstrual spotting, endometriosis, uterine scar pregnancy, numbness and pain (16,17).
After exclusion of pre-existing morbidity, CS births have a 4-5 higher fold of morbidity and mortality as compared to vaginal delivery (18,19) Maternal mortality (MM) that is related to CS has dropped from 90% in the year 1800s to about 0.1% in 1950. Currently, MM that is related to CS is about 6-13/100,000 deliveries in the high resource countries and 7.6/1,000 deliveries in low and middle-income countries respectively, the highest-burden being reported from SSA (10.9/1,000 deaths. This drop has been explained by improvements in surgical and anaesthesia techniques, antibiotic prophylaxis and availability of blood transfusion services (19,20).
This study aims to determine the rate of caesarean section complications, characterize their timing and identify the risk factors to maternal complications among women delivered at KCMC hospital.

Study design and area
This was a hospital-based analytical cross-sectional study conducted from October 2018 to May 2019, in the department of Obstetrics and Gynaecology at Kilimanjaro Christian Medical Centre (KCMC).
KCMC is a zonal consultant and a teaching hospital located in Moshi municipality. It serves as a catchment area for facilities in Kilimanjaro, Manyara, Tanga and Arusha regions in Northern Tanzania and some districts in Kenya. The average delivery per year is about 4000

Study population
Pregnant women with an intrauterine pregnancy at a gestation age of ≥28 weeks, delivered by CS (both elective and emergency) at KCMC hospital and have consented to participate in the study were involved. Women who presented with uterine rupture, or where classical incision, a vertical incision that is con ned to the lower segment, inverted "T" incision and inverted "J" incision was used in the index delivery were excluded from the study.

Sample size and sampling technique
Systematic sampling technique was used in order to get a required sample size of 385 participants.
Data collection method, tools and procedure Face to face interview was conducted by the principal investigator or by a trained research assistant who is uent in both English and Swahili. These interviews were conducted in the ward where these patients were admitted after a caesarean section delivery.
List of women delivered by CS was obtained from the delivery register that is available in the department of obstetrics and gynecology. Detailed information about the research and its purpose were explained to each client who met the inclusion criteria. Then consent was sought from every client before being enrolled in the study. However, the rights of an individual client to be part of the study or refuse to be part of the study was observed. At the enrolment, social-demographic characteristic, detailed past medical, obstetrics, surgical and family social history were sought from every study participant.
Other variables of interest like examination ndings, pre-operative diagnosis, indication for CS, duration of surgery, type of anaesthesia used, presence of any incidental intra-operative ndings or adverse events and the postoperative diagnosis was retrieved from the patient's les.
A data collection form was used to extract all relevant information about the study, variables from patients and their hospital le records before they are discharged from the hospital. This information was later transferred into a computer database containing all corresponding variables needed for the study. The decision to perform Caesarean section was done by an attending Registrar, resident or a specialist.
Enrolled participants were visited daily until the 3 rd day to see if they developed any of the CS-related complications.

De nition of variables
The main outcome variables included intra-operative complications like visceral injury, high spinal, nausea and vomiting together with post-operative complications like post-partum hemorrhage (PPH), hysterectomy, Surgical site infection (SSI), peritonitis, re-laparotomy, endometritis, shock, blood transfusion, post-spinal headache, thromboembolism, ICU admission and death.
Immediate complications have been de ned as a disease or health condition developing in the course of caesarean delivery. This unfavourable evolution must occur within 24 hours of the caesarean delivery, while Early complications have been de ned as a disease or health condition developing in the course of caesarean delivery between 24-72 hours of the caesarean delivery.
Haemoglobin < 11 g/dl or haematocrit <30% was considered as mild Anaemia, 7-9.9 g/dl moderate anaemia and Hb <7 g/dl to be severe anaemia (21).These values were obtained from Full blood picture results that were done just before the operation for emergency CS and one day before delivery for elective CS.
The duration of surgery (skin incision to the last stitch) was regarded as 'prolonged' when it lasted more than one hour (60 minutes).
Blood loss was estimated by counting the number of soaked gauzes, abdominal packs and blood stained on the theatre linens and drapes. The decision to transfuse intra-operatively was made by the attending anaesthetist/anaesthesiologist based on pre-operative haemoglobin/ haematocrit levels, intra-operative blood loss and clinical status of the patient. While the decision to transfuse post-operatively was made by the attending doctor. The units of blood and blood products transfused were recorded in a special blood transfusion services form.

Data analysis
Data cleaning and error range check to enhance validity and consistency was observed before data analysis. Descriptive statistics were summarized in frequencies and percentages. Bar graphs and pie charts are used to present postoperative complications. Continuous variables have been presented in mean or median.
The risk factors whose independence turned out to be signi cant by the univariate logistic analysis were analysed further by multivariate logistic regression analysis. Multivariate analysis was done for the most prevalent complications, which were anaesthesia related complications, blood transfusion and PPH to control for confounders. Odds ratio and 95% CI has been used to make an association. A p-value of less than 0.05(two-tailed) was considered statistically signi cant. Data analysis was performed by using STATA version 13.0 (Copyright 1985-2017 Stata Corp LLC).

Results
A total of 2310 deliveries occurred at KCMC during the study period of which 1138 had Caesarean section. This makes a caesarian section rate of 49.1% Among those delivered by CS, ve women were excluded because uterine rupture was found intraoperatively, thus 386 participants were enrolled in the study. All of them agreed and signed their consent; this gives a participation rate of 100%.refer gure 1.
Baseline characteristics of the participant.

Discussion
During the study period 49.1% of women were delivered by caesarean section, out of these 27.5% had CS related complications. The most common immediate complications (with 24 hours of CS delivery) were anaesthesia-related complications, blood transfusion and PPH. While the most common "early complications" (within 24 to 72 hours of the operation); were puerperal sepsis, anaesthesia-related complications and blood transfusion.Longer duration of surgery was signi cantly associated with all the complications.
In this study, the rate of caesarean section related complications among women delivered by CS at KCMC hospital was 27.5%, these ndings were the same as the ndings from Finland (27% ) and Norway (24.1%) (11,22). This can be explained by the fact that all these studies were done in large obstetric units in referral centers which receives high risk and complicated patients and the same duration of follow-up was used. The rate of complications that were observed in this study is in contrast to the rate that was observed in India(55.1%), Israel(5.7%) and Nigeria(13.3%) (9,23,24). This can be explained by differences in study designs and methods that were used, for example, a retrospective cross-sectional study that was done in large tertiary hospital in Nigeria, included all CS delivery that were only done during the day where most of the team members are available and very energetic, showed a far much lower complication rate. However, this diversity implies that just like any surgery, CS delivery is prone to complications and this cuts across high income to low-income countries Most common complications that were encountered within the rst 24 hours of CS ("immediate complications") were found to be Anaesthesia related complications, blood transfusion and PPH. These complications were more common in the emergency CS group compared to the elective CS group. These ndings were supported by studies from Israel (24) and Nepal and Saudi Arabia (25,26). These similarities can be explained by the fact that all these studies were done in referral and tertiary hospitals which receives women from lower facilities with complications like prolonged labour which makes them prone to hemorrhage intra operatively and post operatively.
This study showed that puerperal sepsis, anaesthesia-related complications and blood transfusion were the most common "early complications" encountered after caesarean section. These ndings were in line with studies from Israel where endometritis was found to be the most prevalent complication, reported to occur on day two and three. (24). This ndings emphasises the bene t of providing prophylactic antibiotic to every parturient that delivers by CS since Caesarean delivery is a known risk factor to endometritis, it also emphasize that CS should be done only when indicated.
Longer duration of surgery (>60 minutes) was found to associated with all complications, which was similar to ndings from cohort studies that were done in Norway and Nigeria (27,28). However, ndings from this study contrast with ndings from Finland, India, Norway, Israel and South Africa which showed emergency CS, cervical dilatation, maternal characteristics and fetal factor were associated with all complications. (11,22,24,29).However, this serves as a continuous alarm to physicians to modify the already known risk factors before surgery to minimize incidences of complications while optimizing good maternal-fetal outcomes.
Pre-operative anaemia was also found to be signi cantly related to a blood transfusion during or after surgery, this nding is similar to ndings from Nigeria, India and U.S.A(30-32) This is because caesarean section is associated with signi cant blood loss as compared to vaginal delivery. Hence small amounts of blood loss intra-operatively in women with preoperative anaemia can produce symptoms and signs of anaemia which will necessitate blood transfusion. Therefore, care providers should make every effort to optimize the haemoglobin levels of every pregnant woman before delivery.
Grand multiparity was also found to be positively and signi cantly associated with blood transfusion intra and/or postoperatively. This nding was similar to ndings from Nigeria and Saudi Arabia (26,33). This can be explained by the fact that grand multiparity has been known to be a risk factor for preoperative anaemia, abnormal placentation and post-partum haemorrhage(PPH) due to uterine atony. All these are a signi cant predictor for blood transfusion. However, literature to contrast these ndings was not found. Hence continuous education on family planning must be done to reduce risks that are associated with grand-multiparity.

Strengths and limitations
This study has highlighted the magnitude of CS delivery-related complications at KCMC hospital. It also highlights the risk factors for the most prevalent complications.
However, this study was prone to recall bias, especially when participants were supposed to provide information on their past medical and surgical history.
As well there may be under-reporting of the caesarean section related complications since participants were not followed up for other possible early complications following hospital discharge. This is because most of the time our patients get discharged from the hospital on the third-day post caesarean section if there are no complications.

Conclusion
Just like any surgery, caesarean section delivery is never without complications. During the study period, 27.5% of the study participants had one or more CS-related complications. With most of the complications being reported in the rst 24 hours of the operation where anaesthesia related complication was the leading complication followed by blood transfusion and haemorrhage. Puerperal sepsis, Anaesthesia related complications and Blood transfusion were the most common complication after 24 hours of the operation. Longer duration of surgery was signi cantly associated with all the complications. Grand multiparity and pre-operative anaemia were signi cantly associated with blood transfusion while grand multiparity was signi cantly associated with PPH. Therefore, CS should be performed when there is strong obstetric indication, health facilities and practitioners should aim at reducing the risks like anemia and grand-multiparity during pre and antenatal visits. Committee before the initiation of the study. Written informed consent was obtained and con dentiality was ensured. Finger print was used for illiterate women. Codes and hospital numbers were used instead of patient's names. It was emphasized that participation in the study was voluntary and that the woman could withdrawal from the study at any time if she wished.

Consent for publication
Not applicable.

Availability of data and materials
Data set generated and obtained from the analysis of this study are not publicly available to ensure con dentiality, however, it can be obtained from the corresponding author on a reasonable request.

Competing interest
There is no competing interest in this study.  Early postoperative complications following caesarean section (N=386)