Obstetric patients in a surgical intensive care unit: prognostic factors and outcome
K. MJAHED, D. HAMOUDI, S. SALMI, & L. BARROU
Department of Anaesthesia and Intensive Care Unit, Ibn Rochd University Hospital, Casablanca, Morocco
Summary
The objective of this study was to assess the incidence, prognostic factors and the outcome of obstetric patients admitted in a surgical intensive care unit (SICU) during the ante-partum or postpartum period (within 6 weeks of delivery). Between 1995 and 2002, the patients transferred from the department of obstetrics were retrospectively included into the study.
Demographics included: obstetric data, medical and surgical histories, diagnosis, simplified acute physiology score (SAPS II), acute physiology and chronic health evaluation system APACHE II score; and the occurrence of organ failure, therapeutic interventions, length of stay in the SICU and outcome were recorded. During the study period, 364 obstetric patients were admitted to the SICU. Obstetric admissions to the SICU represented 0.6% of all deliveries and the SICU utilisation rate was 14.96%. The main indications for admission were eclampsia (70.6%) and postpartum haemorrhage (16.2%). The overall mortality rate was 16.7% (n¼61). In a logistic regression model, risk factors for death included organ system failure (odds ratio (OR)¼3.95 confidence interval (CI) [1.84 – 8.48], bilirubin 412 mg/l (OR¼1.017 CI [1.00 – 1.03]), and prolonged prothrombin time (OR¼0.97 CI [0.95 – 0.99]). Median length of stay was longer in non- survivors (6.5+7.3 vs 5.5+4.6 days). Maternal condition on admission and associated complications are the major determinant of maternal outcome.
Introduction
Critical illness requiring admission to a surgical intensive care unit (SICU) is a relatively uncommon complication of pregnancy, accounting for less than 3% of SICU admissions in developed countries (Baskett and Sternadel 1998; Gilbert et al. 2003; Hazelgrove et al. 2001; Mahutte et al. 1999;
Lapinsky et al. 1997; Zeeman et al. 2003). Several studies have reported SICU use and outcome around the time of childbirth in western countries, but relatively few studies have been published in developing countries. The need for SICU for obstetrics patients use ranges between 10% and 20% of all SICU admissions in developing countries and mortality is very high (Karnad et al. 2004; Platteau et al.
1997).
In Morocco, maternal mortality rates have decreased significantly since 1960, but women continue to die during pregnancy and in the postpartum period, despite improve- ment in antenatal care.
In the present study, we have tried to identify the incidence, prognostic factors and outcome of the critically ill obstetric patients admitted to SICU.
Material and methods
The study was conducted in the multidisciplinary SICU at Ibn Rochd Hospital University, a teaching hospital in Casablanca, Morocco. The records of all obstetric patients admitted to the SICU from January 1995 to December 2002 were reviewed. The SICU is a 12-bed unit caring for a variety of patients including, trauma, general surgery, neurosurgery and obstetric patients. For this study,
obstetric admissions were considered to include women from 14 weeks’ gestation to 6 weeks’ postpartum. Data collected included the reason for admission, maternal age, gestational age, parity, mode of delivery, time before admission (time between admission on obstetric unit and admission in ICU), duration of SICU stay, Glasgow Coma Scale score at admission to the SICU, clinical and laboratory findings and outcome (death or discharge from SICU and the hospital). The severity of the cases was measured at the admission to the SICU by the simplified acute physiology score (SAPS II) (Le Gall et al. 1993), the acute physiology and chronic health evaluation system, APACHE II (Knaus et al. 1985b) and the organ system failure as defined according to Knaus (Knaus et al. 1985a).
Data included the need for mechanical ventilation, dialysis and blood transfusion. The syndrome of haemolytic anaemia, elevated liver enzymes, and low platelet count (HELLP) was identified by the presence of haemolysis, abnormal peripheral smear, increased bilirubin (412 mg/l) and increased lactic dehydrogenase (LDH 4600 U/l), elevated liver enzymes defined as increased serum aspartate aminotransferase (ASAT) level 470 IU/l, alanine amino- transferase (ALAT) level 460 IU/l and low platelets defined as platelet5100,000 mm3. Diagnosis of dissemi- nated intravascular coagulation (DIC) included: low platelets (100,000 mm73), fibrinogen 5300 mg/dl, ele- vated fibrin degradation products and prolonged prothrombin time550%. Diagnosis of pulmonary oedema included acute onset of shortness of breath, hypoxia and radiologic evidence of pulmonary oedema. Adult respira- tory distress syndrome (ARDS) was defined as hypoxaemic respiratory failure (arterial oxygen pressure 560 mmHg
Correspondence: K. Mjahed, 125 rue Larache, Hay Essalam CIL, Casablanca 20200, Morocco. E-mail: kmjahed@yahoo.fr ISSN 0144-3615 print/ISSN 1364-6893 onlineÓ2006 Informa UK Ltd.
DOI: 10.1080/01443610600720188 J Obstet Gynaecol Downloaded from informahealthcare.com by University of Melbourne on 06/05/13 For personal use only.
with a fractional concentration of oxygen in the inspired gas of at least 0.60) requiring mechanical ventilation, in which the chest roentgenogram showed bilateral alveolar infil- trates without cardiac aetiology. Two subsets of patients were identified according to survivors or non- survivors.
Comparisons between groups were made with the unpaired t-test for continuous variables and thew2-test for categorical data with calculation of the relative risk and its 95%
confidence interval (CI). A value of p50.05 was con- sidered significant. Stepwise logistic regression was used to explore the effect of several variables on risk factors for mortality.
Results
Between 1 January 1995 and 31 December 2002, 364 obstetric patients were admitted to the SICU. Over the respective study period, our institution had 58,708 deliveries giving SICU utilisation rates of 0.6%. Obstetric patients accounted for 14.96% of total SICU admissions in the same period (5,446). There were 305 (84%) obstetric patients admitted directly from the labour ward or obstetric theatre and 59 (16.2%) transferred from another hospital.
Time before admission in SICU was 20+46 h (range 1 – 96 h). The mean age of these patients was 28+7 years (range 16 – 47). The mean duration of stay in SICU was 5.7+5 days (range 1 – 33 days). The mean gestational age at delivery was 35+5 weeks. A total of 246 (67%) patients had a caesarean section. There were 194 primiparae and 170 multiparae.
A total of 380 babies were born to these women with a perinatal mortality rate of 32%. The median APACHE II score was 12+5 and SAPS II score was 31+13.
The failure of one or more organs occurred in 203 patients (56%). One organ failed in 140 patients, two failed in 53 patients, three failed in nine patients, and four failed in one patient. Demographic data are reported in Table I.
The majority of patients (344) 94.5% were admitted postpartum. Eclampsia and postpartum haemorrhage accounted for 70.6% and 16.2%, respectively, of the admissions (Table II). Indications for transfer to the SICU in patients affected by severe pre-eclampsia (n¼6) or eclampsia (n¼251) were neurological complications in 25 cases (haemorrhagic stroken¼13, ischaemic stroken¼11 and one case of cortical blindness), pneumonia aspiration in 11 cases (woman in coma), ARDS in 10 cases, pulmonary oedema in nine cases, HELLP syndrome in 17 cases, severe intravascular coagulopathy in 15 cases, acute renal failure in 28 cases, abruptio placenta in 11 cases and two cases of subcapsular liver haematoma. Many of them were associated (Table III).
The causes of haemorrhage were, ruptured uterus (n¼12), uterine atony (n¼9), placenta praevia or accreta (n¼8), abruptio placenta (n¼4), lacerations (n¼5), retained products of conception (n¼3) and disseminated intravascular coagulopathy (n¼20). There were 18 cases of haemorrhage with no documented aetiology for the bleeding. Thirty one required hysterectomies. Nine patients died because of haemorrhagic shock.
Six patients were admitted for cardiac disease. Four had rheumatic valvular disease, one had cardiomyopathy and one had endocarditis. There were no deaths among this group. There were 14 cases of sepsis, eight peritonitis (four pelvic sepsis) and four meningitis (one case was herpetic meningitis which resulted in maternal death).
An anaesthetic complication was the indication for transfer to the SICU in only two cases (difficult intubation and aspiration). Endotracheal intubation and mechanical Table I. Demographic data. Comparison between survivors and non-survivors
Survivors (n¼303)
Non-survivors
(n¼61) p
Age (year) 28+7 31+6 0.004
Temperature8C 37.4+0.8 37.6+1.18 NS Blood pressure (mmHg)
Systolic 158+43 150+53 NS
Diastolic 94+25 88+29 NS
Mean 116+29 112+34 NS
Heart rate min71 113+22 114+21 NS
Organ system failure 0.6+0.7 1.6+0.7 0.001 Glasgow Coma Scale 10.5+3.1 7.7+3.6 0.001
Eclampsia (%) 215 (71) 36 (59) NS
Postpartum haemorrhage (%)
48 (16) 11 (18) NS
Primiparity (%) 171 (56.4%) 25 (41%) 0.02 Gestational age (weeks) 35+5 34+7 NS Time before admission
in SICU (h)
14+21 50+99 0.001 Transfer from
another centre (%)
47 (15.5) 12 (19) NS
Need for dialysis 8 (2.6) 10 (16.4) 0.001
Caesarean section 197 (65) 49 (80) NS
Perinatal mortality rate (%)
91 (30) 25 (41) NS
Length of stay (days) 5.5+4.6 6.5+7.3 NS Values are mean+SD or percent; SICU, surgical intensive care unit; NS, not significant.
Table II. Primary diagnosis of all patients admitted to the surgical intensive care unit (SICU)
Cases (%) Death
Eclampsia/pre-eclampsia 257 (70.6) 36
Postpartum haemorrhage 59 (16.2) 11
Sepsis 14 (3.84) 6
Septic abortion 2 2
Peritonitis 8 4
Meningitis 4 1
Neurological disorder 11 (3.02) 4
Epilepsy 2 0
Deep cerebral thrombosis 4 0
Cerebral aneurysm 1 1
Arteriovenous malformation 1 1
Brain tumour 2 2
Guillain-Barre´ syndrome 1 0
Liver disease 10 (2.7) 2
Acute fatty liver of pregnancy 4 0
Hepatitis 6 2
Cardiac disease 6 (1.6) 0
Cardiomyopathy 1 0
Valvular heart disease 4 0
Endocarditis 1 0
Other 7 (1.92) 2
Post-anaesthetic complication 2 0
Amniotic fluid embolism 2 1
Asthma 1 0
Trauma 1 0
Pulmonary embolism 1 1
Total 364 61
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Melbourne on 06/05/13 For personal use only.
ventilation were required in 237 (65%) cases and ARDS developed in six patients. The duration of ventilation was 58+85 h. Blood products were transfused in 195 patients (54%). A total of 99 received red cells (27%), 75 fresh frozen plasma (21%) and 120 platelets (33%). There were 61 maternal deaths among the 364 cases, for an overall mortality rate of 16.7%. The mean time interval between hospitalisation and maternal death was 7.1+9 days with range of 1 – 48 days and 50% of maternal deaths occurred within 48 h of admission.
On univariate analysis, factors that were significantly associated with higher mortality rate were maternal age, severity of illness at the time of admission as assessed by the APACHE II (16.3+4.9 vs 11+4.9), SAPS II (40.3+12.9 vs 28.5+11.9) or OSF (1.6+0.7 vs 0.6+0.7) score, poor neurological status assessed by Glasgow Coma Scale and late transfer to ICU (Table I).
Mortality increased with increased number of organ failure (Figure 1). The mortality rate was 1.2% for no organ failure, 18.6% for one organ failure, 50.9% for two, 55.6%
for three and 100% for four organ failure. Laboratory disturbances and the need for dialysis or long-term mechanical ventilation were associated with high mortality (Tables I and IV).
However, the observed mortality rate was much lower than the deaths predicted by the APACHE II and the SAPS II equation. The standardised mortality ratio (SMR, observed deaths/predicted deaths) was 0.87 for APACHE II and 0.70 for SAPS II for all patients admitted in ICU (Table V).
On multivariate analysis, maternal mortality rate was correlated with multiorgan system failure pro- longed prothrombin time and high levels of bilirubin (Table VI).
Discussion
According to World Health Organization estimates, maternal mortality rates have decreased significantly in Morocco in the last 2 decades, but women continue to die during pregnancy and in the postpartum period (AbouZhar and Wardlaw 2001). Although maternal mortality rates have fallen, there were 220 maternal deaths per 100,000 live births (WHO 2002).
The cases reported in this study represent 0.62% of all deliveries performed during the same period and an SICU utilisation rate of 16.7%. This high incidence of SICU Table III. Complications during eclampsia (251) and pre-
eclampsia (6)
Survivors (n¼221)
Death (n¼36) p Cerebrovascular accident
Haemorrhagic stroke 2 11 0.001
Ischaemic stroke 7 4 0.05
Aspiration 6 5 0.002
Pulmonary oedema 7 2 NS
Adult respiratory distress syndrome
5 6 0.000
HELLP syndrome 10 7 0.001
Disseminated intravascular coagulation
7 8 0.000
Acute renal failure 16 12 0.001
Abruptio placenta 10 1 NS
Subcapsular liver haematoma
1 1 NS
Figure 1. Number of organs failed between survivors and non- survivors.p¼0.001.
Table IV. Laboratory findings Survivors
(n¼303)
Non survivors
(n¼61) p
Creatinine mg/l 12+7 23+22 0.001
Bilirubin mg/l 16+38 34+62 0.01
Haemoglobin g/dl 9.9+2.4 8.8+3.2 0.004 Platelets 109/l 167+104 144+131 NS Fibrinogen g/dl 3.23+1.55 3.15+2.4 NS Prothrombin time (%) 82+30 63+34 0.001 ASAT/ALAT (U/dl) 157+287/ 370+529/ 0.001/
90+181 245+334 0.001 Leucocytes 103/l 17.9+9 17.6+10 NS Haematocrit (%) 29.4+7.7 26.2+9.1 0.006 Values are mean+SD or percent.
Table V. Predicted mortality rates and SMR for APACHE II and SAPS II scores
Patients n
Predicted mortality risk for APACHE II
score (%)
Predicted mortality risk for SAPS II
score (%)
Survivors 303 14.9 13.7
Non-survivors 61 26.8 29.8
SMR 0.87 0.70
Total 364 16.8 16.3
SMR, standardised mortality ratio.
Table VI. Multivariate predictors of mortality in ICU
Variable Beta Odds ratio 95% CI p
Constant 72.180 – – –
OSF 1.375 3.954 1.84 – 8.48 0.001
Bilirubin 0.016 1.017 1.00 – 1.03 0.030
Prothrombin time 70.023 0.977 0.95 – 0.99 0.031 J Obstet Gynaecol Downloaded from informahealthcare.com by University of Melbourne on 06/05/13 For personal use only.
admission is also typical of others series from developing countries.
Unlike the developed world, obstetric patients account for a significant number of admissions to the intensive care unit in developing countries. In countries like South Africa, Platteau and colleagues (1997) found that obstetric patients accounted for 0.61 of all deliveries and 13.6% of ICU admissions. Karnad et al. (2004) found that the ICU admission rate was 546 admissions per 100,000 deliveries in public hospitals in India.
Lower frequencies were reported in Canada, USA and Europe, where the rate of transfer varies between 0.07%
and 0.5% (Baskett and Sternadel 1998; Hazelgrove et al.
2001; Mahutte et al. 1999; Lapinsky et al. 1997).
Consistent with findings in other studies, most of our patients were admitted to the SICU in the postpartum period. Several studies on obstetric ICU patients have been published, but most reports are small descriptive analyses, limited to 100 subjects or fewer with low rate of mortality (Baskett and Sternadel 1998; Cohen et al. 2000; Lapinsky et al. 1997). The strength of this study is that it is one of the largest reviews of maternal deaths at a single institution. If we consider only studies with a large number of patients (more than 100 obstetric patients), the mortality rate varies from 2.3% in Canada (Mahutte et al. 1999); 0.22 – 3.5% in the USA (Baskett and Sternadel 1998; Zeeman et al.
2003); 2.1 – 3.3% in the UK (Hazelgrove et al. 2001) and 7.5% in Spain (Olara et al. 2002).
Although the maternal mortality rate of 16.7% was very high compared with the 0.5 – 4% mortality rate in some studies from Western countries, it is better than the 21 – 28% mortality rate reported from India (Karnad et al.
2004), or the 21.6% from South Africa (Platteau et al.
1997). Because of a shortage of ICU beds in other district hospitals, tertiary hospitals with ICUs receive a large number of transfers from neighboring hospitals. These transfers account for 16.2% of all obstetric ICU admissions similar to countries like Turkey, where transfers account for 15% (Demirkiran et al. 2003).
The two main reasons for ICU admission were eclampsia and postpartum haemorrhage, accounting for 316 of the 364 cases (86.8%). Hypertensive disorders of pregnancy are an important cause of maternal mortality throughout the world and the majority of these deaths are associated with eclampsia. It is a common problem in developing countries because illiteracy, lack of health awareness, education and poverty prevent women from seeking medical advice during pregnancy. In our country, only 60% are delivered in public and private hospitals. The rest have no access to obstetric care. As a result, most pre- eclampsia cases remain unrecognised until severe compli- cations such as eclampsia occur. Although hypertension is the third major cause of maternal death in developed countries, it is the major cause in our hospital. This is because postpartum haemorrhage and sepsis can be managed efficiently in the hospital if the patients can get there early. The majority of eclamptic patients who died arrived at the hospital at a later stage of the disease with serious complications.
Development of health awareness and implementation of antenatal care for all pregnant women may reduce the incidence of eclampsia. The Magpie trial showed that preventive magnesium sulphate for all severe cases of pre- eclampsia can reduce the risk of development of eclampsia and maternal death by 58% and 45%, respectively (Altman
et al. 2002). Unfortunately, magnesium sulphate is not available in our country. Obstetricians generally use antic- onvulsants, such as diazepam or midazolam to treat and prevent convulsions during eclampsia. The challenge in our country will be to ensure the free availability of this treatment. It is a relatively cheap, relatively easily available drug and it has been demonstrated that it can be used with safety. Some authors reported a mortality rate of 33.3% in complicated eclampsia and none in uncomplicated eclamp- sia (Sawhney et al. 2000). In the present analysis 28% of the eclamptic patients were complicated by a number of associated factors. Associated complications like acute renal failure, HELLP syndrome, abruptio placenta, DIC and cerebrovascular accidents (CVA) contributed to high maternal mortality rate. Acute renal failure was associated with obstetric causes of haemorrhagic shock, eclampsia, HELLP syndrome, DIC and abruptio placenta. The patterns of the renal failure and clinical courses vary with respect to the initiating cause. Septic shock occurs most commonly in the peripartum period often following chorioamnionitis, postpartum endometritis, septic abortion and urinary tract infection. Postoperative peritonitis following hysterectomy, herpetic meningitis and septic abortion were the causes of death in six patients. Cardiac diseases are a complicating factor in approximately 1 – 4%
of all pregnancies (Bhatla et al. 2003). They account for 1.6% of obstetric admissions to our SICU and the rheumatic valvular disease is the most common cause.
This may be due to the higher incidence of chronic rheumatic disease in Morocco. Severe postpartum hae- morrhage is still a major hazard to the pregnant woman and it was the second reason for ICU admission, whereas its represents the most frequent cause of admission in developed countries (Gilbert et al. 2003; Hazelgrove et al. 2001). Blood products were transfused in 54% of patients and were associated with a poor outcome. Severe liver dysfunction such as HELLP syndrome, acute fatty liver of pregnancy and acute fulminant hepatitis are responsible for a poor maternal and perinatal outcome.
High level of bilirubin was an indirect witness of liver dysfunction and was associated in multivariable analysis with high mortality rate. HELLP syndrome complicates eclampsia and significantly increases the mortality rate. We have had only one case of thromboembolic disease.
Perhaps many deaths due to thromboembolism occur after discharge from hospital and do not therefore appear in hospital review of obstetric unit.
ICU mortality rate should be interpreted only after taking into account the severity of illness. Several studies have evaluated the accuracy of the APACHE II scoring system in obstetric patients (Lapinsky et al. 1997; Karnad et al. 2004). In most studies, the mortality rate was significantly lower than that predicted by the APACHE II score. This finding has been reported in developed countries with low ICU mortality. Typically, the APACHE II scores are low in obstetric patients admitted to the ICU, ranging from 6.8 to 13 in many studies (Lapinsky et al.
1997). In this study, the APACHE II and SAPS II score were high for the non-survivors group. These findings indicate a high severity of illness and can explain the high mortality rate in our unit. It is not surprising that some controversies exist concerning the applicability of the APACHE II and SAPS II scores as a predictor of illness severity in the pregnant patient. Many of the abnormalities commonly associated with complicated pregnancies, J Obstet Gynaecol Downloaded from informahealthcare.com by University of Melbourne on 06/05/13 For personal use only.
especially the pregnancy-induced hypertension diseases, such as HELLP syndrome or haemostatic disorders, are not evaluated by these scoring systems, thereby leading to a possible underestimation of severity. Lewinsohn and colleagues (1994) found a poor correlation between predicted and actual mortality using the APACHE II score, whereas Scarpinato (1998) observed a better out- come than that predicted by the SAPS II score. In contrast, el-Solh and Grant (1996) found that the APACHE II and other systems in current use could predict severity of illness and ICU outcome in critically ill obstetric patients.
Multiple-organ failure was a strong predictor of mortality.
Survivors had fewer failures of organ systems (0.6+0.7) compared with those who died (1.57+0.69). In this study, 74% of the patients were admitted for eclampsia and neurologic disorders. This is not surprising as neurologic condition of the patient and its assessment with the Glasgow score or the organ failure were the strongest predictors of outcome. The percentage of patients requiring mechanical ventilation was also higher in this study as was the duration of ventilatory support. In a developed country, the proportion of patients needed for ventilatory support in an ICU has varied from 39 – 42%
and the duration is shorter (Koeberle et al. 2000). Some diagnoses, such as CVA, septic shock or haemorrhagic shock are by definition very severe, and therefore, it is not surprising that the proportion of the deaths with these conditions is greater. Cerebral pathology is the principal cause of death in 57% of eclamptic patients. It is often associated with haemostatic disorders such as DIC, thrombocytopenia and HELLP syndrome. Thus it is not surprising to find that bilirubin and prolonged prothrom- bin time were associated with a poor outcome on multivariate analysis. The severity of illness is not the only reason for the high mortality rate. Another cause was the poor quality of antenatal care received by the patients and also the referral nature of our teaching hospital offering tertiary level care. The late referral and the poor transport facilities were contributors of high maternal mortality. The delay before admission to the ICU is longer in the non-survivors group. The surgical ICU is geographically separated from the obstetric unit and is managed primarily by critical care specialists with assistance from the obstetric team. The median ICU length of stay in the current study was 5 days. It is longer than other studies but similar to a study with patients with same severity of illness (Karnad et al.
2004). Prolonged ICU stay reflects the severity of the complication.
Conclusion
At our institution, the major causes of maternal death have not changed over time, and a better understanding of these causes may lead to more effective prevention efforts.
Prenatal care, early detection and careful monitoring and treatment during pregnancy can prevent some of the pre- eclampsia eclampsia-related complications. Because of the unique characteristics of obstetric patients, since January 2003, we have a dedicated obstetric ICU within the labour and delivery area. Maternal condition on admission and associated complications are the major determinants of the outcome in patients admitted to the ICU. The mortality could be lower if the most serious cases had been treated sooner on the labour ward. Recognition of the severity of
the patient’s condition followed by rapid referral to an ICU is important to reduce maternal mortality. Efforts should be made to improve patients’ access and compliance with prenatal care.
References
AbouZhar C, Wardlaw T. 2001. Maternal mortality at the end of decade signs of progress. Bulletin of the World Health Organization 79:561 – 568.
Altman D, Carroli G, Duley L, Moodley J, Neilson J, Smith D.
2002. Magpie Trial Collaboration Group. Do women with pre- eclampsia, and their babies, benefit from magnesium sulphate?
The Magpie Trial: a randomised placebo-controlled trial.
Lancet 359:1877 – 1890.
Baskett TF, Sternadel J. 1998. Maternal intensive care and near- miss mortality in obstetrics. British Journal of Obstetrics and Gynaecology 105:981 – 984.
Bhatla N, Lal S, Behera G, Kriplani A, Mittal S, Agarwal N et al.
2003. Cardiac disease in pregnancy. International Journal of Gynecology and Obstetrics 82:153 – 159.
Cohen J, Singer P, Kogan A, Hod M, Bar J. 2000. Course and outcome of obstetric patients in a general intensive care unit. Acta Obstetricia and Gynecologica Scandinavica 79:846 – 850.
Demirkiran O, Dikmen Y, Utku T, Urkmez S. 2003. Critically ill obstetric patients in the intensive care unit. International Journal of Obstetric Anesthesia 12:266 – 270.
el-Solh AA, Grant BJ. 1996. A comparison of severity of illness scoring systems for critically ill obstetric patients. Chest 110:1299 – 1304.
Gilbert TT, Smulian JC, Martin AA, Ananth CV, Scorza W, Scardella AT. 2003. Obstetric admissions to the intensive care unit: outcomes and severity of illness. Obstetrics and Gynecol- ogy 102:897 – 903.
Hazelgrove JF, Price C, Pappachan VJ, Smith GB. 2001. Multi- center study of obstetric admissions to 14 intensive care units in southern England. Critical Care Medicine 29:770 – 775.
Karnad DR, Lapsia V, Krishnan A. 2004. Prognostic factors in obstetric patients admitted to an Indian intensive care unit.
Critical Care Medicine 32:1294 – 1299.
Knaus WA, Draper EA, Wagner DP, Zimmerman JE. 1985a.
Prognosis in acute organ-system failure. Annals of Surgery 202:685 – 693.
Knaus WA, Draper EA, Wagner DP, Zimmerman JE. 1985b.
APACHE II: A severity of disease classification system. Critical Care Medicine 1985 13:818 – 829.
Koeberle P, Levy A, Surcin S, Bartholin F, Clement G, Bachour K et al. 2000. Severe obstetric complications necessitating hospitalization and intensive care: a ten year retrospective study.
Annales Franc¸aises d’Anesthe´sie-re´animation 19:445 – 451.
Lapinsky SE, Kruczynski K, Seaward GR, Farine D, Grossman RF. 1997. Critical care management of the obstetric patient.
Canadian Journal of Anaesthesia 44:325 – 329.
Le Gall JR, Lemeshow S, Saulnier F. 1993. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. Journal of the American Medical Association 270:2957 – 2963.
Lewinsohn G, Herman A, Leonov Y, Klinowski E. 1994. Critically ill obstetrical patients: outcome and predictability. Critical Care Medicine 22:1412 – 1414.
Mahutte NG, Murphy-Kaulbeck L, Le Q, Solomon J, Benjamin A, Boyd ME. 1999. Obstetric admissions to the intensive care unit.
Obstetrics and Gynecology 94:263 – 266.
Olarra J, Longarela AM, Suarez L, Palacio FJ. 2002. Critically ill obstetric patients treated in an ICU. Chest 121:2077 – 2078.
Platteau P, Engelhardt T, Moodley J, Muckart DJ. 1997. Obstetric and gynaecological patients in an intensive care unit: a 1 year review. Tropical Doctor 27:202 – 206.
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Melbourne on 06/05/13 For personal use only.
Sawhney H, Aggarwal N, Biswas R, Vasishta K, Gopalan S. 2000.
Maternal mortality associated with eclampsia and severe preeclampsia of pregnancy. Journal of Obstetrics and Gynecol- ogy Research 26:351 – 356.
Scarpinato L. 1998. Obstetric critical care. Critical Care Medicine 26:616 – 618.
WHO. 2002. Maternal mortality in 2000: estimates developed by WHO, UNICEF and UNFPA. Geneva: World Health Organi- zation.
Zeeman GG, Wendel GD, Cunningham FG. 2003. A blueprint for obstetric care. American Journal of Obstetrics and Gynecology 188:532 – 536.
J Obstet Gynaecol Downloaded from informahealthcare.com by University of Melbourne on 06/05/13 For personal use only.
