Only 20 cases of non-obstetric (fortuitous) deaths were reported, including eight cases of motor vehicle accident, four of assault, two cases of trauma, two of suicide, three cases of herbal medicine ingestion and one "other" cause of death. These cases comprise only 3.4% of all notified maternal deaths. Due to the small numbers, meaningful conclusions are not possible. It is recommended that the notification of maternal deaths be extended to include mortuaries since many such cases, currently presenting at mortuaries, are not reported.
There were 18 cases of death due to unknown cause (3.1% of all notifications). In many instances, the cause of death could not be established due to incomplete or missing documentation. Limited reporting certainly played a major role.
Key Recommendations
Disease or trauma, not necessarily associated with pregnancy, may result in the death of a number of pregnant women. These deaths are referred to as fortuitous or non-obstetrical causes of maternal mortality. Motor vehicle accidents, assault and other trauma are amongst the most important causes of non-obstetrical deaths in South Africa. Other causes include suicide and the intake of dangerous herbal medicines. Also included in this group are deaths of unknown cause. During 1998, 20 non-obstetrical deaths (3.4% of all notified obstetrical deaths) and 18 cases of unknown cause of death (3.1% of all notified cases) were reported.
No significant demographic analysis could be done due to the small numbers of cases. However, age, parity and level of care are given in Tables 11.2 to 11.4.
It would seem that a significantly greater proportion of non-obstetrical deaths occurred in tertiary care institutions, whilst deaths of unknown cause more frequently occurred in primary care centres.
Amongst the non-obstetrical deaths, the most important are cerebral complications, hypovolaemic shock and cardiac failure or arrest. These causes are largely due to trauma. Not surprisingly, where the primary cause of death was unknown, the final or contributory cause was also unknown (80% of cases).
The small numbers and many unknown factors make the discussion of avoidable factors irrelevant in both fortuitous and unknown cases. The following comments made by the assessors are, however, important.
Case 1: The question was raised whether it serves a useful purpose to do a caesarean section in the trauma theatre on a terminally ill patient (head injuries after MVA) without first ascertaining whether the baby is still alive. In this case a fresh stillborn was delivered.
Comment: In such cases, a caesarean section should only be done on live fetuses. However, sometimes the caesarean section might save the mother's life and in such a case the operation should be done irrespective of the fetus' condition.
Case 2: A patient with a haemoglobin level of 9g% was given a blood transfusion but her vital signs were poorly monitored. Staff response to warning signs of possible blood reaction was unsatisfactory.
Comment: Blood transfusions must be given according to a specific protocol and patients must be monitored closely during the transfusion.
Case 3: A patient was admitted to hospital a week prior to her death with a severe headache. A history of domestic violence was not noted and a differential diagnosis of headache not made. She was discharged the following day, but presented in a coma one week later and subsequently died of a subdural haematoma. Bruising of the scalp was noted at the autopsy examination.
Comment: Any unusual complaint by a patient must be taken seriously. Sudden severe headache is certainly one necessitating thorough investigation.
Most of these cases were classified as such because of an absence of notes or other information. Clinical notes (clinical evaluations and observations) were incomplete, not done or missing. In cases where the patient was declared dead on arrival, information regarding the events leading to death, were just not available.
Trauma is the main reason for fortuitous deaths. Given that only institutionalised deaths were reported, it is not surprising to find that the majority of the women died in tertiary hospitals. This, however, is probably only the tip of the iceberg. Violence and trauma to women, including pregnant women, may be a much greater problem in our society than these numbers suggest.
This survey is hospital-/clinic-based, implying that most deaths occurring outside these institutions were not reported. Of particular importance is the role mortuaries may play in identifying pregnant women who have died of trauma or suicide. These institutions should play an integral role in the notification of maternal deaths.
Only 18 deaths of unknown cause were noted, contributing to only 3.1% of all maternal deaths in this report. Although limited reporting may have been a major contributing factor to this small number of cases, these low figures are reassuring.
After 3 deaths probably due to the ingestion of traditional herbal medicine, the community at large must be informed of the use of these herbal medications.
A small number of fortuitous deaths and deaths of unknown cause have been reported. Limited reporting may be important and in this respect the system of notification should be re-evaluated. A first step should be the inclusion of mortuaries as an official source of data.