Summary

Peripartum hysterectomy, or the surgical removal of the uterus after the birth of a baby, can be a life-saving procedure where other measures have failed to stop haemorrhage after giving birth. It is associated with considerable morbidity for mothers and their babies, and brings an end to a woman's fertility.

Although peripartum hysterectomy is rare (fewer than 300 per year), internationally, incidence appears to be increasing, and literature suggests that this may be linked to increasing maternal age and an increase in the rate of caesarean sections. However, to date, the national rate of peripartum hysterectomy in Australia has not been reported.

This paper forms part of the work of the National Maternity Data Development Project to develop national data standards for prioritised maternal morbidity items, such as peripartum hysterectomy. National hospitalisation data were used to derive the estimated rate of peripartum hysterectomy in Australia in recent years and examine the rates of diagnoses for particular conditions contributing to peripartum hysterectomy.

Using 11 years of data from the National Hospital Morbidity Database (NHMD) for the period 1 July 2003 to 30 June 2014, 2,781 records were extracted for episodes of care for women undergoing hysterectomy procedures that had a principal diagnosis of an obstetric code or a code indicating hospitalisation for circumstances related to reproduction ('pregnancy-associated hysterectomy'). Approximately 90% of these hysterectomies were estimated to have occurred in the peripartum period, while about 10% occurred in early pregnancy (before 20 weeks' gestation). Between 2003-04 and 2013-14, there were 16 deaths (0.6%) recorded among women undergoing peripartum hysterectomy, and 93 fetal deaths (3.7%).

The overall numbers of peripartum hysterectomies increased by 37% between 2003-04 and

2013-14; however, the rate remained relatively stable over this period (about 0.79 per 1,000 mothers giving birth). Older mothers, mothers who were Indigenous, and mothers with multiple births (for example, twins) were over-represented in women who had peripartum hysterectomy, compared with the total maternal population.

Women who had pregnancy-associated hysterectomies often also had a diagnosis of postpartum haemorrhage (69%), placenta accreta/percreta/increta (46%) or placenta praevia (34%). Overall, these associations remained unchanged across the 11 years.

The findings of the analysis are consistent with studies showing that peripartum hysterectomy is strongly associated with caesarean section delivery. Over the 11-year period, 0.6 per 1,000 women giving birth had a caesarean section and a peripartum hysterectomy in the same hospital admission. Caesarean section delivery ending with hysterectomy also occurred together with postpartum haemorrhage in 1.41 per 1,000, with placenta accreta/percreta/increta in 0.99 per 1,000, and with placenta praevia in 0.97 per 1,000 mothers giving birth by caesarean section.

Nationally consistent and regularly collected information on peripartum hysterectomy and its indications would allow further investigation and monitoring of its patterns and trends, and its association with caesarean section. Further, linking hospital and perinatal data would enable peripartum hysterectomy to be examined in the context of the mother's demographics, risk factors, medical history, and pregnancy and birth.