Critical Appraisal of Obstetric Hysterectomy from a North Indian Tertiary Care Health Facility over a Fifteen Year Period

Introduction

Obstetric hysterectomy (OH) is described as extirpation of the uterus, during cesarean section or following vaginal delivery, or within the puerperium period. It was originally devised in 1871 as a surgical attempt to manage life threatening obstetric haemorrhage and sepsis, pioneered by Joseph Cavallini, Horatio Storer, Edward Porro, Lawson Tait etc., [1, 2, 3]. It can be either emergent or elective. Emergency OH has been customarily performed in the setting of inexorable and life-threatening obstetric hemorrhage [2, 3, 4, 5, 6]. However, several indications like uterine rupture, morbidly adherent placenta, and uterine sepsis [6, 7, 8].

Clinical proficiency and dexterities are imperative in making a prompt decision to operate before the patient’s condition deteriorates, after meticulously weighing it as a last resort to save a mother’s life, with sacrificing mother’s reproductive capability. This minimises morbidity and mortality, as conclusion from the past pollsters have highlighted that morbidity is often associated with the conditions leading to OH, and not with the procedure itself [8, 9, 10, 11]. Apposite timing and meticulous care may trim down or ward off numerous maternal complications [12, 13, 14, 15].

In modern times, improved obstetric management, availability of new-fangled drugs like misoprostol, neoteric antibiotics and ready availability of blood products transfusion has brought down the incidence of emergency OH due to obstetric haemorrhage and obstructed labour, especially in developed nations [7, 8, 9, 10, 11, 12, 13, 14, 15]. However, incidence of OH due to morbidly adherent placenta is on a rise [1, 16, 17, 18, 19, 20, 21, 22]. Thus, indirectly incidence of OH is an indicator of obstetric care dispensed in a health care facility, and in turn community [23, 24, 25, 26, 27].

Keeping this background in mind, the study was designed to critically appraise the incidence and causes of obstetric hysterectomies in a well-equipped 1500 bedded postgraduate tertiary care and teaching government hospital in the national capital and to provide an insight into the trends of etiology/ morbid factors associated with OH over two decades and scope for improvements in maternal health services, if any. This would further aid in formulating strategies for eminent reduction of maternal morbidity and mortality in the facility and the nation.

Materials and Methods

The case records of all the patients undergoing obstetric hysterectomy were studied in detail over a period of one year from January 2016 to December 2016 in department of Obstetrics and Gynecology, VMMC and Safdarjung Hospital, New Delhi. Hysterectomy done for any indication during pregnancy, labour and puerperium was included. Also, those done for complications following pregnancy termination, such as perforation and sepsis.

Each case record was analysed in detail with respect to demographic data (age, parity, booked or emergency referred case), detailed obstetric history, indications leading to hysterectomy, type of surgical procedure performed, risk factors, postoperative maternal morbidity and mortality and perinatal outcome.

The data thus obtained was compared to similar data collected seven years back in the year 2008 and fifteen years back in 2001 in Safdarjung hospital to study and compare the pattern of changing trends in the incidence and causative factors over the period.

Data was decoded from case proformas and analysis was done after applying appropriate statistical tests.

Results

There were 51 cases of OH amongst 27127 deliveries in 2016, 22 cases were observed amongst 25535 deliveries in 2008 and 29 cases of OH amongst 19092 deliveries in the year 2001. An incidence of 1.8/10000 deliveries in 2016, 0.8/1000 deliveries in 2008 whilst 1.5/1000 deliveries in 2001(Table 1). Most of the obstetric hysterectomies in the year 2016 followed caesarean deliveries (82.35%). This was slightly higher than that observed in 2008(81.82%), and was almost double than that observed in 2001(44.83%) (Table 1).

Analysing the trends of risk factors over these years, it was found that important ones contributing to rupture uterus were dehiscence of previous uterine scar (45.45% in 2008 and 66.66% in 2016) and obstructed labour (45.45% in 2008 and 33.33% in 2016). Morbidly adherent placenta was the second most common cause for obstetric hysterectomy in 2008(22.7%) and 2016 (23.52%) (Table 3). Furthermore, all the cases of obstructed labour that required emergency obstetric Majority of our patients in both the years underwent total hysterectomies (Table 4). Nevertheless, the percentage of subtotal hysterectomies increased in the 2008; and furthermore in 2016 indicating a trend towards more conservative surgical approach.

hysterectomy were referred from outside and that too in late stage in labour in moribund condition with little scope for conservative management.

Most of the women had postoperative shock, pyrexia, paralytic ileus, and wound infection through all these years. Mortality was 17.64% in 2016 and 18.28% in 2008 as compared to 20.68% in 2001.

Discussion

Incidence of emergency OH in the present study was 1.8 % (2016) which is considerably higher when compared to conclusions from past investigators [1, 2, 3, 4, 5, 28, 29, 30, 31, 32]. Marked differences in various researches conducted across different countries till date may be consequent to variable cesarean delivery rates, definitions, backdrops (population contrasted with facility based), and accessibility of additional treatments [1, 2, 3, 10, 11, 12, 13, 14, 15]. The incidence in this study is computed for cases presenting over one year at a stretch at three different time periods, whereas other reviews have given incidence over a period of three or more years [2, 3, 4, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 28, 29, 30, 31, 32]. Moreover, our hospital is one of the largest tertiary care referral center catering to patients all across northern India; as such most of the women were referred from outside peripheral hospitals in moribund condition after complications occurred.

Comparing the trends of the incidence of OH over similar data collected over a period of 14 years from the same hospital, a significant drop in incidence of OH was observed in 2008, when contrasted with that of 2001. This can be attributed to increasing health awareness and practical impact of novel health initiatives across the country, particularly creating a roadmap to predict and prevent, and eventually combat haemorrhage in obstetrics [1, 2, 3]. Nevertheless, the upsurge of OH in 2016 could be ascribed to increasing cesarean delivery rates globally which may lead to a higher incidence of placental pathology [1, 3, 4].

the fact that postpartum hemorrhage snowballs with high parity, a risk factor for uterine atony [1, 2, 3, 4, 13, 19]. Another justification for reduced risk of hysterectomy in nullipara is absence of prior caesarean delivery [1, 4]. In comparison with previous research studies [1, 3, 4, 5, 13, 14, 15, 18, 19, 20], a relatively younger women in the present study echoes a younger age at matrimonial alliance and conception in India. Another striking observation in our data was that most of them were un-booked with no prior antenatal supervision, hailing mostly from lower socioeconomic background. This majorly reflects towards the population base the hospital caters to. It also exposes the felt need to segregate truly unregistered women from referred patients having received antenatal care outside the center of study, to gauge the penetration of essential obstetric services in the country, and propose effective interventions to increase the gamut of the same [2, 4].

Table 5 describes a comparative evaluation of incidence of OH and its risk factors by various researchers in recent times.

Reviewing the causes for hysterectomy, PPH (62%) [Especially atonic PPH] emerged as the most common indication for obstetric hysterectomy in 2001. This was similar to the inferences of Agashe (60%) and Marathe and Kant, et al. However rupture uterus outdid other indications in 2008 (50%) and 2016 (35.29%). This was akin to conclusions by many recent authors including Mantri, et al. (60%) and Ambiye and Venkatraman (67.8%) [2, 3, 4, 5, 6, 29, 30, 31, 32].

A significant decrease in incidence of PPH as a cause of obstetric hysterectomy in 2016 and 2008, when compared to 2001 shows that there is a considerable improvement in medical management available for PPH. Readily available oxytocics, prostaglandins and transfusion facilities in institutional set up, coupled with contemporary interventions (B-Lynch sutures, Cho sutures, Hayman technique and arterial embolization) contribute towards this effect [1, 2, 3, 7, 8, 9, 10, 11, 12, 13]. It also emphasizes the widespread emergency and essential obstetric services imparted even in the primary centres.

But, high incidence of rupture uterus as a major cause of obstetric hysterectomy in 2008 and 2016, even in this present era of modern obstetrics, reflects poorly on the peripherally available health services in terms of clinical proficiency and adept labour monitoring services, especially during trials of labour after caesarean delivery, and inadequate and delayed referral services [1, 2, 3, 4]. Lack of awareness about regular antenatal visits among women, and deficient number of skilled birth attendants in the local villages having erudition about timely intervention, especially transferring for institutional deliveries in time, also contributes to associated calamitous obstetric complications [27, 28, 29, 30, 31, 32].

Evaluating the risk factors, the eminent ones were dehiscence of previous uterine scar and obstructed labour. The association between prior history of cesarean delivery and hysterectomy is consistent with previous studies [1, 2, 3, 4]. It has been recently concluded that nations with high cesarean delivery rates had higher hysterectomy prevalence statistics compared with those with lower rates [1, 4]. Also, compared with vaginal delivery in unscarred uteri, vaginal birth following cesarean delivery and repeat cesarean delivery carry higher risks of hysterectomy [1, 2, 3, 4, 5, 6].

Morbidly adherent placenta was the second most common cause for obstetric hysterectomy in 2008(22.7%) and 2016 (23.52%). Furthermore, all the cases of obstructed labour that required emergency obstetric hysterectomy were referred from outside and that too in late stage in labour in moribund condition with little scope left for conservative management.

Postoperative shock, pyrexia, paralytic ileus, and wound infection were common complications observed through all the years. Surgical complications may be strengthened by placental pathology altering anatomy of lower uterine segment and pelvis and amplified blood supply to pelvic organs during pregnancy [2, 3, 4, 5, 6, 19, 20, 21, 22, 23, 24, 25]. Prolonged labour, intrauterine manipulation and dormant sepsis probably explain these complications. These could be prevented by early referral of such cases to well- equipped centres which can treat emergency obstetric cases promptly and efficiently [2, 3, 4, 5, 28, 29, 30, 31, 32]. The mortality amongst our patients was 17.64% in 2016 and 18.28% in 2008 as compared to 20.68% in 2001. This was akin to various other studies have reported mortality in the range of 9.3% to 32% [2, 3, 4, 5, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]. A decline in mortality in OH patients over these years is certainly a pat on the back on the available clinical expertise and proficient management offered at the present facility. Also deployment of innovative strategies including community based assessment of communication and transportation system, and inculcation of modern technologies in the hospital health delivery system like use of mobile phones would reduce delayed decision-making and transportation during childbirth. At the same time, these results encourage all for provision of a continuum of expert care to a woman from the start of the pregnancy all the way through her post-partum period, for preventing the preventable, as much as we can, for saving her life, and in the pursuit of achieving the sustainable health development in our country.

Conclusion

Incidence of EOH has come down over the years, and PPH is no more the major contributing factor for the same, which suggests improvement in institutional intrapartum management. Increase in caesarean deliveries in the present era has led to increased incidence of previous scar dehiscence and adherent placenta. Thus, caesarean delivery should judiciously be performed. A high percentage of obstructed labor cases, which are referred late to institutions, still contribute to OH, reverberating a scope for much needed improvement in the community health services. Awareness regarding the need for proper antenatal and intra-natal care and availability of adequate and prompt referral services are cardinal prerequisites to avoid this life saving, but undesired procedure. Community edification towards institutional deliveries or by trained dais would prevent many such emergencies. The proportion of maternal deaths associated with OH

and its indications can be prevented by skilled maternal care, active management of labor, early recognition of complications, timely referral, and effortlessly available conveyance and blood transfusion amenities. Also, well- trained and experienced obstetricians are vital in taking a prompt decision which would go a long way in saving mothers’ lives.