Pattern of severe maternal morbidity in a tertiary hospital of Delhi,India: a pilot study

Introduction

Maternal mortality has been the indicator of measurement of maternal health. Over the last decade, identification of severe maternal morbidity has emerged as a compliment or alternative to investigation of maternal deaths. ^1,2 Severe maternal morbidity also known as ‘near miss’ is defined as a very ill pregnant or recently delivered woman who would have died had it not been that luck and good care was on her side. A review of the causes of severe maternal morbidity might lead to an awareness of the potential problems and enable patients to receive treatment sooner.

Recent research has suggested that severe maternal morbidity may be a better indicator of the quality and effectiveness of obstetric care than mortality alone. It also has been shown to identify priorities in maternal care more rapidly. ^3,4 However, there is no uniform criteria for the identification of possible cases. Identification is complex: some studies use disease-specific definitions, some suggest management-specific criteria (e.g. hysterectomy or admission to ICU) and some describe organ system dysfunction or failure-based criteria. ^5–7

There have been few published studies of obstetric morbidity and only one brief communication on ‘near miss’ or severe maternal morbidity from India. ^8–10 Our study was conducted in order to study the pattern of severe maternal morbidity in a tertiary care hospital.

Materials and methods

The study was conducted in a tertiary care teaching hospital situated in east Delhi. The hospital caters for a large, underprivileged population in east Delhi as well as the adjoining areas of Uttar Pradesh. Women admitted to the department of obstetrics and gynaecology who fulfilled the definition of severe maternal morbidity conditions were included in the study.

The definition of severe maternal morbidity was based on clinical criteria and medical and obstetric interventions that are routinely measurable. The definitions were finalized after searching the literature and in consultation with the obstetricians. The principal direct obstetric causes were defined in advance as:

severe haemorrhage (antepartum, intrapartum or postpartum) leading to a blood transfusion; hospitalization for longer than four days; caesarean section; hysterectomy or death;

Data were collected from 1 July to 31 August 2005. For each case included, a pretested open-ended proforma was completed. We recorded information about: the sociodemographic profile (Table 1); obstetric history; problems and care during in the current pregnancy; details of any current illness; treatment sought at all levels prior to admission; the outcome in terms of the mode of delivery; hysterectomy and fetal and maternal outcome.

Results

The present study was a pilot study conducted for a period of two months to observe the profile of severe maternal morbidity cases in a teaching hospital in India. Data on 65 cases of predefined conditions were obtained and there were 1900 deliveries during the study period. The incidence of severe maternal morbidity was 3.3/100 deliveries.

A total of 63 women fulfilling the definitions of severe obstetric morbidity conditions were included for analysis. The mean age of the patients was 26.3 ± 5 years. More than half (55.5%) were uneducated while 4.8% were graduates. The educational status of husbands was higher – as 22% were uneducated while almost half were educated up to secondary level. Almost half (45.4%) of the cases were from outside Delhi; the median distance travelled was 10 km. Only 38.1% of the cases were registered. The majority were antenatal admissions (68.3%): 31.7% were postdelivery or postabortion. The proportion of postdelivery or postabortion cases was higher (87.5%) in cases of sepsis compared to other diagnosis where the majority of admissions were antenatal. A history of contact with a birth attendant (skilled or unskilled) was reported in 56.7%.

The conditions recorded were: eclampsia or severe pre-eclampsia (34%), haemorrhage (34%), sepsis (12%) obstructed labour (9.5%) and others 14.5%. Fifteen women had eclampsia; seven, severe pre-eclampsia; 10, postpartum haemorrhage; 12, antepartum haemorrhage; eight, sepsis; five, uterine rupture and there were four cases of obstructed labour. Fourteen (22.2%) patients had severe anaemia; four haemorrhaged; four had a hypertensive disorder; three had sepsis. Other conditions recorded included medical conditions such as severe anaemia, jaundice and renal disease. Two women with severe anaemia had signs of congestive heart failure.

We studied the profiles of patients in relation to their place of residence. The proportion of postdelivery or abortion was significantly greater among those from outside Delhi (P < 0.05). The proportion of registered patients was similar in those from outside or from within Delhi. However, a larger number of patients with history of contact with a birth attendant were from outside Delhi (P < 0.05) (Table 2).

Table 3 shows the cause of severe maternal morbidity in relation to the outcome. Only 43.5% were normal vaginal deliveries, 55.5% were delivered by caesarean section. The caesarean section rate was highest (83.3%) among those who had obstructed labour and lowest in those with sepsis (25.0%). Hysterectomy was performed in 14.3%: the proportion of hysterectomies was higher in those who had obstructed labour and those who had a haemorrhage. The fetal outcome for 60.3% of patients was a live birth: the live birth rate for ‘other conditions was the lowest at only 33.3%.

There were two maternal deaths caused by postpartum haemorrhage, leading to disseminated intravascular coagulation in one and severe anaemia leading to pulmonary oedema in the other.

Discussion

This study was conducted in order to observe the pattern of severe maternal morbidity conditions seen in a tertiary hospital in Delhi. The incidence of severe maternal morbidity was 3.3/100 deliveries which is comparable to the figure of 4–8% reported for hospitals in resource poor settings in the World Health Organization's (WHO) systematic review of severe acute maternal morbidity. ⁵ It was lower than that observed in a Nigerian study where almost one in six women had suffered a life-saving emergency. ¹¹ Another study from West Africa by Oladapo et al. ¹² reported a figure of 6.6/100 live births. The difference may be due to the variety of definitions used. Oladapo et al. ¹² included caesarean sections performed for other reasons in their severe maternal morbidity figures. Fillipi et al. ⁷ reported a wide variation in the incidence of near miss – from 1% to almost 25% of all deliveries. A much lower incidence of severe obstetric conditions has been reported from European countries. ^13,14

We used disease-specific criteria including the clinician's evaluation as well as threshold levels for the degree of severity. This definition was used as it is easy to interpret and cases can be identified retrospectively. However, subjectivity is inherent in the identification of cases and definition may not be straightforward in all cases as they also depends on the nature and organization of the health system in the country from which the report originates.

The organ system-based criteria, which are very labour intensive, have been used in some studies which have reported a prevalence of 0.38–1.9%. ^3,13 Others have used admissions to the intensive care unit as a definition of severe obstetric morbidity. However, this definition has a limited role in settings such as ours where intensive care beds are a scarce resource and facilities in the intensive care units differ. ⁴ Lower rates (0.01–2.99%) have been reported in studies using these criteria.

Common causes of severe maternal morbidity were haemorrhage, hypertensive disorders, sepsis and obstructed labour. The proportion of these conditions is similar to those reported from other developing countries. ^15–17 However, the study from West Africa reported a higher proportion of dystocia (30%) as the second most common cause after haemorrhage. ⁶ Studies from Europe have reported haemorrhage and hypertensive disorders as the most common causes in their region. ^13,14 We had difficulty in classifying cases with more than one condition, especially when severe anaemia, which was observed in a very high proportion (22.2%) of cases, was accompanied by the other causes of severe maternal morbidity. The WHO systemic review on causes of maternal mortality has shown that anaemia is the cause in 12.8% of maternal deaths in Asia, 3.7% in Africa and none in the developed countries. Studies from our country have reported it to be an important cause. However, its role as a contributor to maternal mortality and morbidity may be underestimated. ^18–21 Only the study by Oladapo et al. ¹² classified severe anaemia separately – they reported it as a cause in about 11% of ‘near miss’ events and 20% of maternal deaths. Other studies based on the disease-specific criteria of severe maternal mortality have not categorized severe anaemia as a separate condition. In this report, the presence of severe anaemia in 22% cases in our study may indicate that severe anaemia should be listed as an independent cause for severe maternal morbidity.

The profile of our cases showed that 61.9% were not registered for antenatal care and 45.4% were from outside Delhi and had travelled a median distance of 10 km. Those from outside Delhi were more likely to be referred after delivery and most of them had contact with a birth attendant before reporting to the hospital. Studies have highlighted the fact that those least advantaged are most likely to experience severe maternal morbidity. ¹⁶ It has also been reported that a diagnosis of severe maternal morbidity cases was associated with referral. ^6,11 An audit of severe maternal morbidity in Uganda reported that nearly half the patients were at home when the events occurred and in more than half the cases, patient-related factors were responsible for the delay in seeking care. ¹⁷ Referral is a complex variable that involves components such as: the decision to seek care; the perception of the risk by the woman, the family members and the health personnel; and the financial and geographical accessibility. ¹⁸

A limitation of this report is the short duration of the study period in which the small sample was studied. It is proposed to study the assessment of standard of care for severe maternal morbidity at all levels of health care in Delhi.

Our study highlights the fact that severe maternal morbidity cases place a significant burden on health resources and reflects the quality of health care available in our country. Haemorrhage, eclampsia and severe pre-eclampsia, sepsis, obstructed labour and severe anaemia are clearly causes for severe maternal morbidity that should be identified and treated without delay in order to improve feto-maternal outcomes.