Objective To review the clinical characteristics and outcomes of criti
cally ill obstetrical patients and to determine the outcome predictabi
lity using the Acute Physiology and Chronic Health Evaluation (APACHE)
II scoring system. Methods A retrospective data collection of all obs
tetrical patients (n=49) admitted for more than 24 hours to the Intens
ive Care Unit at Kwong Wah Hospital from 1988 to 1995 was conducted. D
emographics, obstetric data, preexistent medical problems, diagnosis,
days staying in the ICU and ICU related data were recorded for each pa
tient. Results Obstetric admissions to ICU during the study period rep
resented 0.12% of all deliveries during this period. There was a predo
minance of postpartum admissions and obstetric diagnosis responsible f
or the patients' critical illness. Massive postpartum haemorrhage was
the single most common cause of ICU admission, representing 53.0% of a
ll patients. Preeclampsia and (14.3%), anaesthesia related complicatio
ns and medical diseases complicating pregnancy (14.3%) were the other
common disease categories for ICU admission. Two cases (4.1%) of surgi
cal disease complicating pregnancy were admitted. The maternal mortali
ty rate was 5.1 deaths per 100 000 total births, or 2 maternal deaths
in 39 354 total deliveries in this study period. All deaths were due t
o nonobstetric causes. The perinatal mortality rate was 10.0% (5 cases
) in this study group. Conclusions When applying the Acute Physiology
and Chronic Health Evaluation (APACHE) II scoring system in predicting
the final outcome in this group of obstetric patients, we found that
our obstetric patients requiring intensive care had a better outcome t
han predicted, as expressed by a low mortality ratio (0.25).