Predictability of adverse outcomes in hypertensive disorders of pregnancy: a multicenter prospective cohort study.

OBJECTIVES: To explore variables associated with adverse maternal/fetal/neonatal outcomes among pregnant/postpartum patients admitted to ICU for hypertensive disorders of pregnancy (HDP). METHODS: Multicenter, prospective, national cohort study. RESULTS: Variables independently associated with maternal/fetal/neonatal mortality among 172 patients were as follows: Acute Physiology and Chronic Health Evaluation-II (APACHE-II)(OR1.20[1.06-1.35]), gestational age (OR0.698[0.59-0.82]) and aspartate aminotransferase (AST)(OR1.004[1.001-1.006]). Positive likelihood ratio for headache, epigastric pain, and visual disturbances to predict composite adverse outcomes were 1.23(1.16-1.30), 0.76(0.59-1.02), and 1.1(0.98-1.2), respectively. CONCLUSIONS: Maternal/fetal mortality due to HDP was independently associated with severity of illness on admission, gestational age, and elevated AST. Accuracy of clinical symptoms to predict composite adverse outcomes was low.

Effects of short-term hyperoxia on sytemic hemodynamics, oxygen transport, and microcirculation: An observational study in patients with septic shock and healthy volunteers.

PURPOSE: To characterize the microvascular effects of a brief period of hyperoxia, in patients with septic shock and in healthy volunteers. MATERIALS AND METHODS: In 20 patients with septic shock, we assessed systemic hemodynamics, sublingual microcirculation by SDF-videomicroscopy, and skin perfusion by capillary refill time (CRT), central-peripheral temperature (ΔT°), and perfusion index. Measurements were performed at baseline and after 5 min of inspired oxygen fraction of 1.00. Additionally, we studied 8 healthy volunteers, in whom hyperoxia was prolonged to 30 min. RESULTS: In septic patients, hyperoxia increased mean arterial pressure and systemic vascular resistance, but cardiac output remained unchanged. The only significant change in sublingual microcirculation was a decreased heterogeneity flow index (1.03 [1.01-1.07] vs 1.01 [0.34-1.05], P = .002). Perfused vascular density (13.1 [12.0-15.0] vs 14.0 [12.2-14.8] mm/mm2, P = .21) and the other sublingual microvascular variables were unmodified. CRT and ΔT° did not change but perfusion index slightly decreased. In healthy volunteers, sublingual microcirculation and skin perfusion were stable. CONCLUSIONS: Short-term hyperoxia induced systemic cardiovascular changes but was not associated with noticeable derangement in sublingual microcirculation and skin perfusion. Nevertheless, longer exposures to hyperoxia might have produced different results.

Poor agreement in the calculation of venoarterial PCO2 to arteriovenous O2 content difference ratio using central and mixed venous blood samples in septic patients.

PURPOSE: Central venous minus arterial PCO2 to arterial minus central venous O2 content difference ratio (Pcv-aCO2/Ca-cvO2) has been proposed as a clinical surrogate for respiratory quotient. Our goal was to assess its interchangeability with mixed venous minus arterial PCO2 to arterial minus mixed venous O2 content difference ratio (Pmv-aCO2/Ca-mvO2). MATERIALS AND METHODS: This is a subanalysis of a previously published study. We studied 23 septic patients who had an indwelling Swan-Ganz catheter. The agreement between Pcv-aCO2/Ca-cvO2 and Pmv-aCO2/Ca-mvO2 was evaluated by Bland and Altman analysis. We also performed linear regression analysis with Pmv-aCO2/Ca-mvO2 as the dependent variable. RESULTS: 95% limits of agreement between Pcv-aCO2/Ca-cvO2 and Pmv-aCO2/Ca-mvO2 were 1.48. Pmv-aCO2/Ca-mvO2 was significantly correlated with hemoglobin and lactate (R2 = 0.48 and 0.31, respectively, P < 0.01 for both). CONCLUSIONS: In this study, Pcv-aCO2/Ca-cvO2 and Pmv-aCO2/Ca-mvO2 were not interchangeable. In addition, Pmv-aCO2/Ca-mvO2 is a composite variable, which depends on several determinants. Values of Pcv-aCO2/Ca-cvO2 should be cautiously interpreted in the assessment of critically ill patients.

Relationship of at Admission Lactate, Unmeasured Anions, and Chloride to the Outcome of Critically Ill Patients.

OBJECTIVES: To investigate the association between the concentration of the causative anions responsible for the main types of metabolic acidosis and the outcome. DESIGN: Prospective observational study. SETTING: Teaching ICU. PATIENTS: All patients admitted from January 2006 to December 2014. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Four thousand nine hundred one patients were admitted throughout the study period; 1,609 met criteria for metabolic acidosis and 145 had normal acid-base values. The association between at admission lactate, unmeasured anions, and chloride concentration with outcome was assessed by multivariate analysis in the whole cohort and in patients with metabolic acidosis. We also compared the mortality of patients with lactic, unmeasured anions, and hyperchloremic metabolic acidosis with that of patients without acid-base disorders. In the whole population, increased lactate and unmeasured anions were independently associated with increased mortality, even after adjusting for potential confounders (odds ratio [95% CI], 1.14 (1.08-1.20); p < 0.0001 and 1.04 (1.02-1.06); p < 0.0001, respectively). In patients with metabolic acidosis, the results were similar. Patients with lactic and unmeasured anions acidosis, but not those with hyperchloremic acidosis, had an increased mortality compared to patients without alterations (17.7%, 12.7%, 4.9%, and 5.8%, respectively; p < 0.05). CONCLUSIONS: In this large cohort of critically ill patients, increased concentrations of lactate and unmeasured anions, but not chloride, were associated with increased mortality. In addition, increased unmeasured anions were the leading cause of metabolic acidosis.

Dissociation between sublingual and gut microcirculation in the response to a fluid challenge in postoperative patients with abdominal sepsis.

BACKGROUND: This study was performed to compare intestinal and sublingual microcirculation and their response to a fluid challenge. METHODS: Twenty-two septic patients in the first postoperative day of an intestinal surgery, in which an ostomy had been constructed, were evaluated both before and 20 min after a challenge of 10 mL/kg of 6% hydroxyethylstarch 130/0.4. We measured systemic hemodynamics and sublingual and intestinal microcirculation. Correlations between variables were determined through the Pearson test. RESULTS: Fluid administration increased the cardiac index (2.6 ± 0.5 vs. 3.3 ± 1.0 L/min/m(2), P < 0.01) and mean arterial blood pressure (68 ± 11 vs. 82 ± 12 mm Hg, P < 0.0001). The sublingual but not the intestinal red blood cell (RBC) velocity increased (912 ± 270 vs. 1,064 ± 200 µm/s, P < 0.002 and 679 ± 379 vs. 747 ± 419 µm/s, P = 0.12, respectively). The sublingual and intestinal perfused vascular density (PVD) did not change significantly (15.2 ± 2.9 vs. 16.1 ± 1.2 mm/mm(2) and 12.3 ± 6.7 vs. 13.0 ± 6.7 mm/mm(2)). We found no correlation between the basal sublingual and intestinal RBC velocities or between their changes in response to the fluid challenge. The individual changes in sublingual RBC velocity correlated with those in cardiac index and basal RBC velocity. Individual changes in intestinal RBC velocity did not correlate with either the cardiac index modifications or the basal RBC velocity. The same pattern was observed with the sublingual and the intestinal PVDs. The sublingual RBC velocities and PVDs were similar between survivors and nonsurvivors. But the intestinal RBC velocities and PVDs were lower in nonsurvivors. CONCLUSIONS: In this series of postoperative septic patients, we found a dissociation between sublingual and intestinal microcirculation. The improvement in the sublingual microcirculation after fluid challenge was dependent on the basal state and the increase in cardiac output. In contrast, the intestinal microcirculation behaved as an isolated territory.

Clinical characteristics and outcomes of obstetric patients admitted to the intensive care unit.

OBJECTIVE: To identify the reasons for admitting pregnant women to intensive care units (ICUs) in Buenos Aires, Argentina. METHODS: The admission diagnoses of pregnant women hospitalized over 2 years at 4 ICUs were retrospectively studied. RESULTS: During the studied period, 242 (3.9%) of the 6271 ICU patients were pregnant women, for an incidence of 8.1 per 1000 deliveries. The main reasons for admitting them at ICUs were hypertensive disorders, followed by postpartum hemorrhage and sepsis. More than a third (39.7%) was in a first pregnancy. The main nonobstetric reason for admission was pneumonia. The median pregnancy duration on admission was 36 weeks (range, 33-38 weeks) but it was less than 34 weeks for 66 (27.2%) of the women, 12.4% on whom required ventilation. Mortality was highest among those admitted for nonobstetric reasons (13.3% vs 0.5%; P<0.05). The median stay for obstetric or nonobstetric conditions was 2 versus 5 days (range, 2-3 days vs 2-8 days) (P<0.001). CONCLUSION: Postpartum hemorrhage and hypertensive disorders were the most common reasons for admitting pregnant women to an ICU, followed by sepsis. Nonobstetric causes of admission were associated with higher morbidity and mortality rates.