HELLP syndrome

Study objectives: To determine the incidence of systemic inflammatory response syndrome (SIRS) and organ failure and to describe the outcomes in critically ill obstetric patients who have been treated in medical ICUs.

Design: Retrospective review.

Setting: A multidisciplinary ICU at a tertiary-care institution.

Methods: We collected data on 74 obstetric patients who were admitted consecutively to the ICU from January 1991 through December 1998. Acute physiology and chronic health evaluation (APACHE) II scores were calculated. A p value < 0.05 was considered to be significant.

Measurements and results: Fifty-eight percent of patients were admitted to the ICU postpartum. Their mean ([+ or -] SD) age was 25.9 [+ or -] 7.0 years, and 64% were African American and 34% were white. Fifty percent had preexisting medical conditions. Their mean APACHE II score was 14.0 [+ or -] 5.9, and their predicted mortality rate was 17.6%. The most common reason for admission was respiratory insufficiency. Preeclampsia was present in 38% of patients, and hemolytic anemia, elevated liver enzymes, and low platelet count syndrome were present in 7% of patients. SIRS developed in 59% of patients. Patients with SIRS had longer ICU stays (p = 0.0008). Organ failure developed in 65% of patients, and ARDS developed in 15% of patients. Invasive mechanical ventilation was required in 45% of patients, and pulmonary artery catheterization was required in 35% of patients. The in-hospital mortality rate was 2.7%. There were five spontaneous abortions and eight perinatal deaths.

Conclusions: The most common reason for admission to the ICU of critically ill obstetric patients was respiratory failure. Despite the severity of illness and the development of SIRS and organ failure in most patients, the mortality rate was low. (CHEST 2001; 120:1271-1277)

Key words: APACHE; critical illness; eclampsia; hemodynamics; ICUs; multiple organ failure; obstetrics; respiratory insufficiency; sepsis

Abbreviations: APACHE = acute physiology and chronic health evaluation; HELLP = hemolytic anemia, elevated liver enzymes, and low platelet count; SIRS = systemic inflammatory response syndrome

ICU admission is required for about 0.1 to 0.9% of pregnant patients. (1-5) Altered maternal physiology, the presence of a fetus, and diseases specific to pregnancy pose various challenges in providing care to critically ill obstetric patients. (6) The reported mortality rate of critically ill obstetric patients admitted to the ICU ranges from 0 to 36%. (2,5,7) The incidence and impact of systemic inflammatory response syndrome (SIRS) and the development of multiple organ failure in critically ill obstetric patients have not been well described. We undertook this retrospective study to determine the incidence of SIRS and organ failure and to describe the clinical course and outcome of critically ill obstetric patients who have been treated in our ICU.

MATERIALS AND METHODS

We reviewed the medical records of obstetric patients who were admitted consecutively to the medical ICU of the University Medical Center (Jacksonville, FL) from January 1991 through December 1998. University Medical Center is a 528-bed, tertiary-care, urban university hospital, serving predominantly the indigent population of northeast Florida. The medical ICU at University Medical Center is a 16-bed unit providing care to critically ill medical and nontrauma neurosurgical and obstetric patients. A team of intensivists and maternal-fetal specialists, along with house staff, treats all critically ill obstetric patients admitted to the medical ICU. Maternal-fetal specialists are available in the hospital 24 h per day and make several daily rounds in the ICU when there are obstetric patients. Intensivists are available in the ICU for most of the day, and there are members of the resident house staff available 24 h each day. Although the maternal-fetal specialists focus on the obstetric complications and the intensivists focus on the other complications, medical decisions are made after discussion among team members whenever necessary.

During the study period, 78 patients were admitted to the medical ICU. The medical records of 4 patients could not be obtained, leaving the records of 74 patients available for the study. The data collected included age, race, underlying chronic disease, duration of pregnancy on admission to the hospital, reason for medical ICU admission, pregnancy-related and other medical diagnoses, use of mechanical ventilation and pulmonary artery catheters, length of medical ICU and hospital stay, development of ARDS, development of SIRS, development of organ failure, maternal mortality, and pregnancy loss.

ARDS was defined in accordance with the definition established by the American-European Consensus Conference on ARDS. (8) SIRS, sepsis, severe sepsis, and septic shock were defined in accordance with definitions given by the American College of Chest Physicians and the Society of Critical Care Medicine. (9) The syndrome of hemolytic anemia, elevated liver enzymes, and low platelet count (HELLP) was identified by the presence of hemolysis (ie, abnormal peripheral smear, total bilirubin level of > 1.2 mg/dL, and lactate dehydrogenase level of > 600 U/L), elevated liver enzymes (ie, transaminase level of > 70 U/L and lactate dehydrogenase level of > 600 U/L), and thrombocytopenia (ie, platelet count, < 100,000/[mm.sup.3]). (10) Thrombotic thrombocytopenic purpura was identified by the presence of four of the following five criteria: microangiopathic hemolytic anemia with negative Coombs test result; platelet count, < 100,000/[mm.sup.3]; CNS abnormalities; fever; and renal dysfunction. Disseminated intravascular coagulation was defined as the presence of prolonged prothrombin time, decreased fibrinogen level, elevated D-dimer level, and thrombocytopenia. Cardiovascular, respiratory, hematologic, renal, and CNS failures were defined according to Knaus et al. (11) Liver failure was defined as a serum bilirubin level of [greater than or equal to] 6 mg/dL and a prothrombin time of [greater than or equal to] 4 s longer than the control. GI failure was defined as GI bleeding, intestinal obstruction, or pancreatitis preventing enteral feeding for at least 24 h or until death. The acute physiology and chronic health evaluation (APACHE) II scores and predicted mortality rates were calculated as described in the literature. (12) The standardized mortality ratio was determined as the ratio of actual mortality to predicted mortality. Pregnancy loss at < 20 weeks of gestation was defined as spontaneous abortion. Pregnancy loss at [greater than or equal to] 20 weeks of gestation and neonatal deaths up to 28 days after birth were considered to be perinatal deaths.

Computer software (StatView, version 5.0; SAS Institute; Cary, NC) was used for statistical analysis. Means were expressed with their SDs. Comparisons between groups were made using Student's t test, Mann-Whitney U test, [chi square] test, and Fisher's Exact Test. A p value of < 0.05 was considered to be significant.

RESULTS

Of the 74 patients, 31 (42%) were admitted to the medical ICU antepartum and 43 (58%) were admitted postpartum. Of the 31 patients admitted to the medical ICU antepartum, 12 delivered during their ICU stay and 19 were discharged from ICU undelivered. On admission to the hospital, the mean ([+ or -] SD) duration of pregnancy was 29.2 [+ or -] 9.1 weeks, gravidity was 2.7 [+ or -] 1.7 (median, 2; range, 1 to 7), and parity was 1.1 [+ or -] 1.2 (median, 1; range, 0 to 5). Data on age, race, and chronic medical conditions are given in Table 1. APACHE II scores were as follows: 0 to 5, 1 patient; 6 to 10, 23 patients; 11 to 15, 29 patients (including 1 nonsurvivor); 16 to 20, 8 patients; 21 to 25, 10 patients; and > 25, 3 patients (including 1 nonsurvivor). The mean ([+ or -] SD) APACHE II score was 14.0 [+ or -] 5.9 (range, 5 to 33). The predicted mortality rate was 17.6%.

The reasons for ICU admission are given in Table 2, with the most common reason being respiratory insufficiency. Pregnancy-related complications and other complications diagnosed during a patient's hospitalization are listed in Table 3. The miscellaneous pregnancy-related complications not listed in Table 3 were septic abortion, cephalopelvic disproportion, ruptured ectopic pregnancy, acute fatty liver of pregnancy, and thrombotic thrombocytopenic purpura, with each complication occurring in one patient. The miscellaneous non-pregnancy-related complications not listed in Table 3 were anaphylaxis, cardiogenic shock, liver hematoma, sickle chest syndrome, varicella pneumonia, supraventricular tachycardia, and pericarditis, with each complication occurring in one patient.

Cesarean section was performed emergently in 32 patients and was performed electively in 2 patients. Respiratory failure developed in 19 of the 32 patients (59%) who underwent emergency cesarean section compared with 16 of the 42 patients (38%) who did not undergo emergency cesarean section (p = 0.1138). Pulmonary edema developed in 10 of the 32 patients (31%) who underwent emergency cesarean section compared with 8 of the 42 patients (19%) who did not undergo emergency cesarean section (p = 0.3479).

SIRS developed in 44 of the 74 patients (59%). Seven patients had two or more episodes of SIRS after the resolution of the first episode. Twelve of the 18 patients (67%) who were admitted to the ICU for pulmonary edema developed SIRS compared with 32 of the 56 patients (57%) who were admitted for other reasons (p = 0.6569). The median APACHE II score was 1,5 in patients with SIRS compared with 10 in patients without SIRS (p < 0.0001). The median length of hospital stay in patients with SIRS was 11 days compared with 6 days in patients without SIRS (p = 0.0038). The median length of ICU stay in patients with SIRS was 4 days compared with 2 days in patients without SIRS (p = 0.0008). The failure of one or more organs occurred in 36 patients (82%) with SIRS compared with 12 patients (40%) without SIRS (p = 0.0006). SIRS was present in all 11 patients with ARDS compared with 33 of the 63 patients (52%) without ARDS (p = 0.0022). Two patients with SIRS died, and there were no deaths in patients without SIRS (p = 0.5113).

The causes of SIRS were sepsis (39 patients), surgery (7 patients), pancreatitis (3 patients), thrombotic thrombocytopenic purpura (1 patient), and undetermined (3 patients). Two patients had septic shock, and 18 patients had severe sepsis. The causes of sepsis were pneumonia (15 patients), urinary tract infection (7 patients), endometritis (5 patients), chorioamnionitis (5 patients), infection associated with the intravascular line (2 patients), pericarditis (1 patient), disseminated herpes (1 patient), empyema (1 patient), and undetermined (2 patients). Eight of the pathogens identified as causing pneumonia were Gram-negative rods, including Pseudomonas aeruginosa (three patients) and Klebsiella pneumoniae (two patients). The most common cause of urinary tract infection was Escherichia coli (three patients).

The failure of one or more organs occurred in 48 patients (65%). One organ failed in 24 patients, two failed in 13 patients, three failed in 8 patients, and four failed in 3 patients. Organ failures were of the following types: respiratory (35 patients); hematologic (21 patients); cardiovascular (17 patients); renal (7 patients); liver (6 patients); G1 (6 patients); and neurologic (1 patient). The mean ([+ or -] SD) number of organ failures was 1.6 [+ or -] 1.2 (median, 1 organ failure) in patients with SIRS compared with 0.5 [+ or -] 0.7 (median, 0 organ failures) in patients without SIRS (p < 0.0001).

In the first 24 h of ICU stay, the lowest ratio of arterial oxygen tension to fraction of inspired oxygen concentration was [less than or equal to] 200 mm Hg in 26 patients and between 200 and 300 mm Hg in 11 patients. ARDS developed in 11 patients (15%). ARDS was associated with pancreatitis in 1 patient and sepsis in 10 patients. Only one patient with ARDS died. The median length of ICU stay in patients with ARDS was 28 days compared with 2 days in patients without ARDS (p < 0.0001). The median length of hospital stay in patients with ARDS was 39 days compared with 8 days in patients without ARDS (p < 0.0001).

Invasive mechanical ventilation was required in 33 patients (45%) for 12.0 [+ or -] 20.4 days (median, 3 days). Noninvasive mechanical ventilation was used in one patient. Pulmonary artery catheterization was required in 26 patients (35%). The pulmonary artery findings were indicative of hyperdynamic hemodynamics in 15 patients, cardiogenic shock in 3 patients, and fluid overload in 5 patients. Acute hemodialysis and exchange transfusion were used in two patients each.

For the 74 patients, the length of stay in the medical ICU was 7.4 [+ or -] 14.4 days (median, 2.5 days) and the length of stay in the hospital was 14.7 [+ or -] 16.0 days (median, 9.5 days). Two patients died, one of cardiogenic shock and one of pulmonary embolism, yielding an in-hospital mortality rate of 2.7%. The standardized mortality ratio was 0.15.

There were five spontaneous abortions and eight perinatal deaths. Fifteen patients were discharged from the hospital without having delivered, and 46 patients had neonates who survived to be discharged from the hospital. There were no significant differences in the presence of chronic medical conditions, illicit drug use, need for mechanical ventilation, development of SIRS and ARDS, number of organ failures, and length of hospital stay between patients with and without spontaneous abortions or perinatal deaths.

DISCUSSION

In this retrospective study, we investigated the clinical course and outcome of critically ill obstetric patients treated in an ICU. Most of the patients (64%) were African American, and 50% of the patients had underlying chronic medical conditions. Eighty percent of the patients were admitted to the

ICU for respiratory insufficiency or hemodynamic instability. SIRS, the most common cause of which was sepsis, developed in most patients and was associated with a longer hospital stay. The failure of one or more organs occurred in 65% of the patients. Respiratory failure was the most common organ failure. Endotracheal intubation and mechanical ventilation were required in 45% of patients, and ARDS developed in 15% of the patients. Despite the severity of illness on admission to the ICU and the development of multiple complications, the mortality rate was only 2.7%.

This study included critically ill obstetric patients who were treated in a multispecialty ICU of a tertiary-care, university-affiliated hospital. Because of the unique characteristics of obstetric patients, some hospitals have created maternal-fetal ICUs to monitor and treat critically ill obstetric patients. (3,13) However, most hospitals lack the volume to justify such units. (14) Most critically ill obstetric patients require a multispecialty team approach in traditional ICU settings, especially when multiple organ failure and ARDS develop.

Obstetric patients with preexisting medical problems are more likely to require intensive-care support than those without preexisting medical conditions. (15) Our finding that 50% of all patients had underlying chronic medical conditions is similar to a finding by Collop and Sahn. (16) This prevalence of chronic medical conditions is higher than expected for this young population. Because our hospital is located in the inner city, the prevalence of illicit drug use, particularly the use of crack cocaine, is high and represents the largest single underlying medical condition in this patient population. The fourth most common major underlying chronic medical condition in our patients was hemoglobinopathy, reflecting the large proportion of African Americans in the study.

Consistent with findings in other studies, (1,4) most of our patients were admitted to the ICU postpartum. The two most common reasons for ICU admission were respiratory failure and hemodynamic instability, accounting for 59 of the 74 cases (80%). These findings are similar to those of Lapinsky et al. (2) Pulmonary edema was the most common cause of the respiratory failure in the patients that we studied. This finding is consistent with that of a previous large study (17) in which pulmonary edema developed in 0.5% of obstetric patients, 45% of whom required admission to the ICU and 15% of whom required endotracheal intubation and positive-pressure ventilation. The prevalence of SIRS in patients admitted for pulmonary edema was not significantly higher than the prevalence of SIRS in patients admitted for other reasons in the present study. Perioperative fluid overload, instead of systemic inflammation, contributed to a greater frequency of pulmonary edema. This may explain the lack of association between SIRS and pulmonary edema in the patients we studied.

Hemodynamic instability is more common in obstetric patients admitted to the ICU postpartum than in patients admitted antepartum. (4) In our study, 84% of the patients with hemodynamic instability were admitted to the ICU postpartum. Although hemorrhage was the reason for ICU admission in < 10% of our patients, postpartum hemorrhage was the reason for ICU admission in more than half of the patients studied in Hong Kong by Tang et al. (1) In some studies, the most frequent reason for ICU admission was hypertensive disease of pregnancy. (3,18) Mabie and Sibai (3) found that 46% of the patients in an obstetric ICU were admitted for hypertension, 44% were admitted for medical procedures, and 10% were admitted for hemorrhages. The differences in criteria for ICU admission may account for these differences among studies.

Respiratory insufficiency and hemodynamic instability are the two main reasons for admission of obstetric patients to the ICU. In the present study, pulmonary edema and pneumonia were the two most common causes of respiratory insufficiency, and postpartum hemorrhage and sepsis were the two most common causes of hemodynamic instability. To decrease the ICU admission rate of obstetric patients, clinicians should try to prevent the development of pulmonary edema, pneumonia, postpartum hemorrhage, and sepsis. Judicious use of IV fluid, infection control measures, and prevention and timely control of postpartum hemorrhage are likely to help achieve these objectives. Appropriate antibiotic use, surgical control of infection sources, and timely resuscitation before the development of organ damage are also essential to decrease the number of obstetric admissions to the ICU.

Preeclampsia is the most common obstetric complication in critically ill obstetric patients. (2,4,19) A recent study (20) of 105 patients admitted to the ICU in Durban, South Africa, with a diagnosis of eclampsia showed the mortality rate to be 10.5%. Other complications include HELLP syndrome, disseminated intravascular coagulation, and thrombotic thrombocytopenic purpura. In the present study, preeclampsia or eclampsia was present in 38% of the patients and HELLP syndrome was present in 7% of the patients.

SIRS develops in most hospitalized patients. (21) The causes of SIRS include infection, trauma, pancreatitis, and burns. (9) SIRS has a noninfectious cause in only 5% of patients. (22) Critically ill patients with SIRS are more likely to have longer hospital stays, multiple organ failures, and higher mortality rates than those without SIRS. (23) Consistent with these reports, SIRS developed in 59% of our patients, was caused by sepsis in most, and was associated with a longer hospital stay and the development of organ failure. The higher prevalence of SIRS-associated ARDS and organ failure in the present study highlights the importance of preventing and treating SIRS to improve the morbidity of critically ill obstetric patients.

Organ failure in critically ill patients is associated with increased mortality. (11) In our study, the failure of one or more organs developed in 65% of the patients. Respiratory failure was most common, and invasive positive-pressure ventilation was required in 45%. Previous studies have shown that mechanical ventilation is required in 12 to 55% of obstetric patients who are admitted to the ICU. (1-3,16) Although young age and lack of many underlying diseases in critically ill obstetric patients with ARDS should be associated with a good prognosis, the reported mortality rate is not low. (24) Kilpatrick and Matthay (4) found that acute lung injury was present in 25% of the obstetric patients admitted to the ICU and was associated with a mortality rate of 25%. In the study by Collop and Sahn, (16) all four of the obstetric patients who died had ARDS, as did 10 of the 21 patients in the study by Kirshon et al. (13) Because of the limitation in our data collection, we were unable to determine the prevalence of acute lung injury in our patients. Although the mortality rate was not high in our patients, ARDS developed in 15% and was associated with prolonged ICU and hospital stays.

In the last 2 decades, different prognostic systems have been developed to predict the outcome of critically ill patients admitted to ICUs. (12,25-27) APACHE II is the most widely cited. However, physiologic changes during pregnancy, such as higher respiratory and heart rates and lower hematocrit and creatinine levels, can lead to higher APACHE II scores, causing falsely elevated predicted mortality rates. The APACHE, mortality probability models, and simplified acute physiology score prognostic systems have been used to assess the severity of illness and to predict mortality in critically ill obstetric patients, with conflicting results. (1,2,7,28,29) Koch et al (7) found that actual mortality was higher than the mortality predicted by APACHE II. el-Solh and Grant (28) found there was no significant difference between the observed mortality and the mortality predicted by the APACHE II, mortality probability models, and simplified acute physiology score systems. (28) Lewinsohn et al (29) found that actual mortality was lower than the APACHE II predicted mortality. In a study from Saudi Arabia, (30) the APACHE II, but not the mortality probability models or simplified acute physiology score system, overestimated the mortality rate of obstetric patients requiring intensive care. In our study, APACHE II overestimated the actual mortality. Although the APACHE II prognostic system is the most widely cited in the critical-care literature, its original database, which was developed about 20 years ago, is unlikely to have included enough obstetric patients for accurate outcome prediction. Another limitation of the APACHE II prognostic system is its inability to predict mortality in a patient group such as ours because it was developed based on data from mixed ICU patient populations. Moreover, differences in access to health care, ICU admission criteria, and disease severity, in addition to the small number of patients with low mortality in the different studies, make comparisons difficult.

The mortality rates of critically ill obstetric patients admitted to the ICU range from 0 to 36%. (1,2,4,7,16,19,28,29,31,32) The heterogeneity of the studies and the variations in disease severity may explain part of these differences in mortality rates. The mean ([+ or -] SD) APACHE II score of patients in the study by Lapinsky et al (2) was 6.8 [+ or -] 4.2 compared with 14.0 [+ or -] 5.9 in our study, which may explain the differences in mortality rate and length of hospital stay between the two studies. In the present study, the actual mortality rate was lower than predicted. In response to the observation of higher-than-predicted mortality in this population, (7) we instituted the described team approach in our care of critically ill obstetric patients. This may have contributed to a reduction in observed mortality. It is possible that the high variability of mortality rates of such patients correlates not only with the severity of underlying disease but also with clinical recognition of the unique needs of this patient population. However, we recognize that the present study design does not allow us to determine whether a team approach would improve outcomes.

Our study has four limitations. First, the data were collected retrospectively. Second, the sample size was small, and the mortality rate was only 2.7%, which makes it difficult to determine the prognostic factors and differences between survivors and nonsurvivors. Third, we are unable to identify preventable conditions that may have led to ICU admission because of the lack of prenatal data and a control population of obstetric patients who did not require ICU admission. Finally, because the study was performed in an inner city tertiary-care medical center, the findings may not apply to other patient populations.

In conclusion, we have described the clinical course and outcome of critically ill obstetric patients admitted to the ICU. These patients often have preexisting medical conditions. The most common reason for ICU admission is respiratory failure. Despite the high severity of illness and the development of organ failure, the mortality rate of critically ill obstetric patients is low.

REFERENCES

(1) Tang LC, Kwok AC, Wong AY, et al. Critical care in obstetrical patients: an eight-year review. Chin Med J (Engl) 1997; 110:936-941

(2) Lapinsky SE, Kruczynski K, Seaward GR, et al. Critical care management of the obstetric patients. Can J Anaesth 1997; 44:325-329

(3) Mabie WC, Sibai BM. Treatment in an obstetric intensive care unit. Am J Obstet Gynecol 1990; 162:1-4

(4) Kilpatrick SJ, Matthay MA. Obstetric patients requiring critical care: a five-year review. Chest 1992; 101:1407-1412

(5) Scarpinato L, Rampoola T, Bradenburg S. Analysis of critically ill obstetric patients requiring invasive hemodynamic monitoring or transfer to med/surg ICU at a 314 bed community hospital over a five year period [abstract]. Intensive Care Med 1995; 21(suppl):S92

(6) Lapinsky SE, Kruczynski K, Slutsky AS. Critical care in the pregnant patient. Am J Respir Crit Care Med 1995; 152:427-455

(7) Koch KA, Rodeffer HD, Sanchez-Ramos L. Critically ill obstetrical patients: outcome and predictability [abstract]. Crit Care Med 1988; 16:409

(8) Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS: definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Bespir Crit Care Med 1994; 149:818-824

(9) Bone RC, Balk BA, Cerra FB, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis: the ACCP/SCCM Consensus Conference Committee; American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101:1644-1655

(10) Pousti TJ, Tominaga GT, Scannel G. Help for the HELLP syndrome. Intensive Care World 1994; 11:62-64

(11) Knaus WA, Draper EA, Wagner DP, et al. Prognosis in acute organ-system failure. Ann Surg 1985; 202:685-693

(12) Knaus WA, Draper EA, Wagner DP, et al. APACHE II: a severity of disease classification system. Crit Care Med 1985; 13:818-829

(13) Kirshon B, Hinkley CM, Cotton DB, et al. Maternal mortality in a maternal-fetal medicine intensive care unit. J Reprod Med 1990; 35:25-28

(14) McCormack DM. Care of the obstetrical patient in the traditional intensive care unit. Crit Care Nurs Q 1998; 21:1-11

(15) Bouvier-Colle MH, Varnoux N, Salanave B, et al. Case-control study of risk factors for obstetric patients' admission to intensive care units. Eur J Obstet Gynecol Reprod Biol 1997; 74:173-177

(16) Collop NA, Sahn SA. Critical illness in pregnancy: an analysis of 20 patients admitted to a medical intensive care unit. Chest 1993; 103:1548-1552

(17) DiFederico EM, Burlingame JM, Kilpatrick SJ, et al. Pulmonary edema in obstetric patients is rapidly resolved except in the presence of infection or of nitroglycerin tocolysis after open fetal surgery. Am J Obstet Gynecol 1998; 179:925-933

(18) Graham SG, Luxton MC. The requirement for intensive care support for the pregnant population. Anaesthesia 1989; 44: 581-584

(19) Platteau P, Engelhardt T, Moodley J, et al. Obstetric and gynaecological patients in an intensive care unit: a 1 year review. Trop Doct 1997; 27:202-206

(20) Bhagwanjee S, Paruk F, Moodley J, et al. Intensive care unit morbidity and mortality from eclampsia: an evaluation of the acute physiology and chronic health evaluation II score and the Glasgow coma scale score. Crit Care Med 2000; 28:120-124

(21) Rangel-Frausto MS, Pittet D, Costigan M, et al. The natural history of the systemic inflammatory response syndrome (SIRS): a prospective study. JAMA 1995; 273:117-123

(22) Brun-Buisson C, Doyon F, Carlet J, et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults: a multicenter prospective study in intensive care units; French ICU Group for Severe Sepsis. JAMA 1995; 274:968-974

(23) Afessa B. Systemic inflammatory response syndrome in patients hospitalized for gastrointestinal bleeding. Crit Care Med 1999; 27:554-557

(24) Hankins GD, Nolan TE. Adult respiratory distress syndrome in obstetrics. Obstet Gynecol Clin North Am 1991; 18:273-287

(25) Knaus WA, Wagner DP, Draper EA, et al. The APACHE III prognostic system: risk prediction of hospital mortality for critically ill hospitalized adults. Chest 1991; 100:1619-1636

(26) Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270:2957-2963

(27) Lemeshow S, Teres D, Klar J, et al. Mortality probability models (MPM II) based on an international cohort of intensive care unit patients. JAMA 1993; 270:2478-2486

(28) el-Solh AA, Grant BJ. A comparison of severity of illness scoring systems for critically ill obstetric patients. Chest 1996; 110:1299-1304

(29) Lewinsohn G, Herman A, Leonov Y, et al. Critically ill obstetrical patients: outcomes and predictability. Crit Care Med 1994; 22:1412-1414

(30) Arabi Y, Goraj R, Horsfall D, et al. Scoring systems in obstetric patients requiring intensive care in Saudi Arabia: a ten-year review [abstract]. Chest 1999; 116(suppl):286S

(31) Scarpinato L. Critically ill obstetric patients: outcomes and predictability utilizing the new simplified acute physiology score (SAPS II) in a 314 bed community hospital [abstract]. Intensive Care Med 1995; 21(suppl):S105

(32) Hogg B, Hauth JC, Kimberlin D, et al. Intensive care unit utilization during pregnancy [abstract]. Obstet Gynecol 2000; 95(suppl):S62

Bekele Afessa, MD, FCCP; Bethany Green, DO ([dagger]); Isaac Delke, MD; and Kathryn Koch, MD, FCCP

* From the Department of Internal Medicine (Drs. Afessa, Green, and Koch) and the Department of Obstetrics and Gynecology (Dr. Delke), University of Florida Health Science Center, Jacksonville, FL.

([dagger]) Dr. Green is currently at the University of Washington, Seattle, WA.

Correspondence to: Bekele Afessa, MD, FCCP, Division of Pulmonary and Critical Care Medicine and Internal Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905

COPYRIGHT 2001 American College of Chest Physicians
COPYRIGHT 2001 Gale Group