Caesarean Section rate amongst Obstetricians at a tertiary-care hospital of Karachi

Objective: To analyze the Caesarean Section (CS) rate among different groups of consultants dealing with Nulliparous Term Singleton pregnancy with Vertex (NTSV) presentation delivering at a tertiary care hospital of Karachi over four months. Methods: This is a secondary analysis of a retrospective data that analyzed factors affecting the CS-rate of NTSV patients. Patients with CS (n=106) were taken as cases and vaginal deliveries (n=106) as controls. This was an unmatched retrospective case-control study. Results: Mean age of patients was 26.6(SD: 4.2) years. Mean gestational-age was 38.6(SD: 1.0) weeks. Likelihood of CS was slightly less in patients who were attended by feto-maternal consultants(OR:0.81 CI:0.38-1.07) and was slightly more in patients managed by non-full-time faculty (OR:1.04 CI:0.59-1.85). Odds of CS was highest amongst consultants having average monthly volumes of 21-30 patients/month (OR:1.069 CI:0.48-2.34). However none of the above findings were statistically significant. A non-significant increase in risk of CS was observed with increase in experience of physicians (p=0.787). Conclusion: The results did not show statistically significant difference in CS rate among different groups of Obstetricians. This might indicate that managing labour according to standard guidelines can eliminate physicians’ bias. This can be further evaluated with larger multicenter prospective studies.


INTRODUCTION
Increasing Caesarean Section (CS) rates are of great concern all over the world. 1 This imparts increased risk to mothers' health and life when compared to vaginal deliveries, with no added benefits to fetus. Literature indicates different reasons of rising CS rate. These include increasing rates of labor inductions 2 , inclination of patients for CS, lesser chances of litigation, easy scheduling, and subjectivity of providers' decision during labour. [3][4][5] In order to modify the practices and other factors affecting the mode of delivery (MOD) many centers use CS rate as a quality indicator especially among Nulliparous at Term and Singleton pregnancies with Vertex presentation (NTSV). 6 The rate of CS among NTSV varies widely, from 10.3% to 34.2%. 3,[7][8][9] NTSV being low risk is most eligible for vaginal birth. MOD among this group is not only a major determinant of the future obstetrics course, 10 but also reflects the true statistic of CS rate. Indications like major placenta previa, cord prolapse and malpresentation are non-modifiable and are according to standard practices. Variations in practices are seen in conditions like dysfunctional labour and non-reassuring fetal status. 6 This variation is minimized in developed countries by implementation of guidelines to standardize the clinical practices. 11 In Pakistan there is no system to gather or monitor national data nor do we have standard guidelines for management of labour. Practices vary among institutes and among consultants. No local data is available but it is observed that very few institutes have set protocols for labour management and even fewer have a system of implementation and audits. The decisions of labour induction and augmentation, use of analgesia during labour, instrumentation and CS are mostly subjective and based on consultants' discretion. There are few determinants affecting the CS rate like Obstetrician's experience; volume of patients they were dealing with; working as fulltime or non-full-time faculty; and dealing with high or low-risk pregnancies.
The rationale of this analysis was to assess if different characteristics of consultant obstetricians are affecting MOD among NTSV at our hospital. This may help to identify practices among different groups of consultants which can eventually be used to reduce CS rate by modifying these practices. The objective of this study was to analyze the CS rate of different groups of consultants dealing with NTSV women delivering at a tertiary care hospital of Karachi over four months.

METHODS
This is a secondary analysis of data that has analyzed factors affecting the MOD in NTSV pregnancies. 12 It was unmatched retrospective casecontrol study, conducted at a tertiary care hospital in Karachi, Pakistan and had received exemption status from institutional ethical review committee. NTSV patients admitted between March 1 st , 2014 and June 30 th 2014, for management of labor and delivery were included. Patients delivered through planned caesarean were excluded.
Subjects were identified through computer based medical record system. Data was collected on predesigned proforma (which was validated by three subject specialists). Medical record charts were reviewed. Patients who were delivered by CS were compared with those delivered vaginally (both spontaneous and operative vaginal deliveries). The sample size was calculated with help of WHO sample size calculator and a total of 212 patients were included (106 in each arm), through purposive sampling.
At our center, admitting consultant is the main stakeholder and decision maker in patient care during and after delivery. Labour management is consistent with departmental guidelines. These guidelines are updated according to evidence-based practice, periodically. Continuous Cardiotocography is used to monitor fetal wellbeing and standardized partogram is used to monitor the progress of labour. Details of labour and delivery are entered into a computerized database by the resident who has attended delivery.
Differences in practice were analyzed on basis of following characteristic of consultants:

• Experience of consultant in years:
The experience of a consultant was calculated from the year of passing the post-graduate exam.

• Average volume of consultants per month:
The

RESULTS
Total of 212 patients were enrolled. This included 106 patients delivered through CS, taken as cases compared with equal number of vaginal deliveries, taken as controls. Patients' ages ranged from 18 to 39 years, with mean of 26.6 (SD: 4.2) years. Mean gestational age of this study group was 38.6 (SD: 1.0) weeks (Table-I).
As seen in   than those having >30 deliveries per month on an average. However, these associations were not significant (p-value 0.696). The CS rate in patients managed by NFT was slightly higher when compared with those delivered vaginally (OR1.04 CI: 0.59-1.85). This was however not statistically significant (p value: 0.883).
The Odds of CS increased slightly with increase in experience of physicians. Although statistically, this was not significant (p-value: 0.787). Univariate correlation did not show any differences in practices of different groups therefore we are not reporting multivariate analysis.

DISCUSSION
Our study showed similar results of CS rate among different groups of consultants when dealing with NTSV patients. This was a retrospective study and it was assumed that the available data was correct. Moreover the confounding factor of different teams involved in the management of cases was overlooked, assuming that the input of instructors' and residents' equally affected the results of all consultants.
The results of our study were similar to Manohar S et al., and showed that rates of normal and interventional deliveries were not different among different consultants. 11 That study compared vaginal delivery with interventional (instrumental and caesarean) deliveries while our study grouped normal and instrumental deliveries as vaginal delivery to compare it with abdominal delivery.
As the feto-maternal specialists deal with high risk patients having medical or surgical comorbids, the risk of CS is expected to be higher in this group as compared to other consultants. 6 The CS rate, in our study was slightly lesser among high-risk group of consultants though this was not statistically significant. This may be because, we selected term pregnancies and several high risk patients may need delivery before 37 completed weeks.
Consultants with higher volumes may be more comfortable in managing normal labour and have been reported to have lesser number of CS as compared to obstetrician with low volumes. 6,13 In our study CS rates among all volume quartiles were almost constant. As also seen by Clapp MA et al. the experience of obstetricians had no effect on the MOD in our patients. 13 There is a general impression in lay public that tendency of labour-induction, expedition of deliveries and low threshold for CS is higher when labour is managed by NFT consultants.
This opinion is based on the premise that these obstetricians deal with laboring patients at multiple places simultaneously and therefore spend lesser time at one centre. 6 This study showed a non-significant increase in CS rate in patients of NFT faculty and was concordant with the finding of McClelland S et al. 6 The practices of different obstetrician groups and possibility of CS among their patients were found to be similar in this study. This may be because labour is managed according to standard guidelines in our institute with regular audits and monitoring of dash-board indicators. 11 We therefore recommend future prospective studies to identify the causes of increased caesarean section rates in our institute. However this study may contribute to local data regarding different practices among obstetricians, but being a secondary analysis of a retrospective data from a single institute it has its limitation.

CONCLUSION
The results did not show statistically significant difference in CS rate among different groups of Obstetricians. This might indicate that managing labour according to standard guidelines can eliminate physicians' bias. This can be further evaluated with larger multicenter prospective studies.