Mortality and quality of life in the five years after severe sepsis.

INTRODUCTION: Severe sepsis is associated with high levels of morbidity and mortality, placing a high burden on healthcare resources. We aimed to study outcomes in the five years after severe sepsis. METHODS: This was a cohort study using data from a prospective audit in 26 adult ICUs in Scotland. Mortality was measured using clinical databases and quality of life using Short Form 36 (SF-36) at 3.5 and 5 years after severe sepsis. RESULTS: A total of 439 patients were recruited with a 58% mortality at 3.5 years and 61% mortality at 5 years. A total of 85 and 67 patients responded at 3.5 and 5 years follow-up, respectively. SF-36 physical component score (PCS) was low compared to population controls at 3.5 years (mean 41.8 (SD 11.8)) and at 5 years (mean 44.8 (SD 12.7)). SF-36 mental component score (MCS) was slightly lower than population controls at 3.5 years (mean 47.7 (SD 14.6)) and at 5 years after severe sepsis (mean 48.8 (SD 12.6)). The majority of patients were satisfied with their current quality of life (QOL) (80%) and all patients would be willing to be treated in an ICU again if they become critically ill despite many having unpleasant memories (19%) and recall (29%) of ICU events. CONCLUSIONS: Patients with severe sepsis have a high ongoing mortality after severe sepsis. They also have a significantly lower physical QOL compared to population norms but mental QOL scores were only slightly below population norms up to five years after severe sepsis. All survivors would be willing to be treated in an ICU again if critically ill. Mortality and QOL outcomes were broadly similar to other critically ill cohorts throughout the five years of follow-up.

Anemia during and at discharge from intensive care: the impact of restrictive blood transfusion practice.

OBJECTIVE: To document the prevalence of anemia among patients admitted to intensive care (ICU) and, among survivors, at ICU discharge when restrictive transfusion practice was used. DESIGN: This was an observational cohort study. SETTING: Ten of the 26 general ICUs in Scotland. PATIENTS AND PARTICIPANTS: One thousand twenty-three sequential ICU admissions over 100 days, representing 44% of all ICU admissions in Scotland during the study period, studied daily from admission to discharge or death in the ICU. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: The median transfusion trigger used, in the absence of bleeding, was 78 g/l (interquartile range 73-84); <2% of transfusion triggers were above the upper limit of the national transfusion trigger guideline (100 g/l). Overall, 25% of admissions had a hemoglobin concentration <90 g/l at ICU admission. Seven hundred sixty-six patients admitted survived to ICU discharge. Among these, the prevalence of anemia (male <130 g/l; female <115 g/l) at ICU discharge was 87.0 (95% CI: 83.6 to 89.9)% for males and 79.6 (74.8 to 83.7)% for females. Of the male survivors 24.1 (20.3 to 28.3)% and of the female 27.9 (23.4 to 33.2)% had a hemoglobin <90 g/l at ICU discharge. The prevalence was similar for patients with and without pre-existing ischemic heart disease. Logistic regression found independent associations between having a hemoglobin concentration <90 g/l at ICU discharge and the first measured hemoglobin in ICU, the presence of acute renal failure and thrombocytopenia during ICU stay. CONCLUSIONS: Anemia is highly prevalent in ICUs that use restrictive transfusion triggers. The impact of anemia on functional recovery after intensive care requires investigation.

Red cell requirements for intensive care units adhering to evidence-based transfusion guidelines.

BACKGROUND: Anemia commonly complicates critical illness. Restrictive transfusion triggers are appropriate in this setting, but no large studies have measured red cell (RBC) requirements for intensive care patients when evidence-based transfusion guidelines are followed consistently. STUDY DESIGN AND METHODS: Data were recorded daily for 1023 of 1042 sequential admissions to 10 intensive care units (ICUs) over 100 days. The sample comprised 44 percent of all ICU admissions in Scotland during this period. RBC transfusions and the occurrence of clinically significant hemorrhage were recorded for every ICU day. Transfusion episodes were classified as either associated with or not associated with hemorrhage. Measures of RBC use were derived for the cohort and for Scotland with national audit data. RESULTS: A total of 39.5 percent (95% confidence interval [CI], 36.5%-42.5%) of admissions received transfusions. Eighteen percent of admissions received at least one transfusion associated with hemorrhage and 26 percent received at least one transfusion not associated with hemorrhage. The median (interquartile range) transfusion trigger in the absence of hemorrhage was 78 (73-78) g/L. The overall mean RBC use was 1.87 (95% CI, 1.79-1.96) units per admission or 0.34 (95% CI, 0.33-0.36) units per ICU-day. Forty-seven percent of RBCs administered were not associated with clinically significant hemorrhage. Mean RBC requirements for intensive care in Scotland were estimated to be 3950 (95% CI, 3780-4140) per million-adult-population per year. This represented 7 to 8 percent of the Scottish blood supply. CONCLUSIONS: Despite evidence-based transfusion practice, 40 percent of ICU patients receive transfusions, which account for 7 to 8 percent of the national blood supply.