IntroductionCulture-negative sepsis is a common but relatively understudied condition. The aim of this study was to compare the characteristics and outcomes of culture-negative versus culture-positive severe sepsis.MethodsThis was a prospective observational cohort study of 1001 patients who were admitted to the medical intensive care unit (ICU) of a university hospital from 2004 to 2009 with severe sepsis. Patients with documented fungal, viral, and parasitic infections were excluded.ResultsThere were 415 culture-negative patients (41.5%) and 586 culture-positive patients (58.5%). Gram-positive bacteria were isolated in 257 patients, and gram-negative bacteria in 390 patients. Culture-negative patients were more often women and had fewer comorbidities, less tachycardia, higher blood pressure, lower procalcitonin levels, lower Acute Physiology and Chronic Health Evaluation II (median 25.0 (interquartile range 19.0 to 32.0) versus 27.0 (21.0 to 33.0), P = 0.001) and Sequential Organ Failure Assessment scores, less cardiovascular, central nervous system, and coagulation failures, and less need for vasoactive agents than culture-positive patients. The lungs were a more common site of infection, while urinary tract, soft tissue and skin infections, infective endocarditis and primary bacteremia were less common in culture-negative than in culture-positive patients. Culture-negative patients had a shorter duration of hospital stay (12 days (7.0 to 21.0) versus 15.0 (7.0 to27.0), P = 0.02) and lower ICU mortality than culture-positive patients. Hospital mortality was lower in the culture-negative group (35.9%) than in the culture-positive group (44.0%, P = 0.01), the culture-positive subgroup, which received early appropriate antibiotics (41.9%, P = 0.11), and the culture-positive subgroup, which did not (55.5%, P < 0.001). After adjusting for covariates, culture positivity was not independently associated with mortality on multivariable analysis.ConclusionsSignificant differences between culture-negative and culture-positive sepsis are identified, with the former group having fewer comorbidities, milder severity of illness, shorter hospitalizations, and lower mortality.

IntroductionAnaemia and the associated need for packed red blood cell (PRBC) transfusions are common in patients admitted to the intensive care unit (ICU). Among many causes, blood losses from repeated diagnostic tests are contributory.MethodsThis is a before and after study in a medical ICU of a university hospital. We used a closed blood conservation device (Venous Arterial blood Management Protection, VAMP, Edwards Lifesciences, Irvine, CA, USA) to decrease PRBC transfusion requirements. We included all adult (≥18 years) patients admitted to the ICU with indwelling arterial catheters, who were expected to stay more than 24 hours and were not admitted for active gastrointestinal or any other bleeding. We collected data for six months without VAMP (control group) immediately followed by nine months (active group) with VAMP. A restrictive transfusion strategy in which clinicians were strongly discouraged from any routine transfusions when haemoglobin (Hb) levels were above 7.5 g/dL was adopted during both periods.ResultsEighty (mean age 61.6 years, 49 male) and 170 patients (mean age 60.5 years, 101 male) were included in the control and active groups respectively. The groups were comparable for age, gender, Acute Physiology and Chronic Health Evaluation (APACHE) II score, need for renal replacement therapy, length of stay, and Hb levels on discharge and at transfusion. The control group had higher Hb levels on admission (12.4 ± 2.5 vs. 11.58 ± 2.8 gm/dL, P = 0.02). Use of a blood conservation device was significantly associated with decreased requirements for PRBC transfusion (control group 0.131 unit vs. active group 0.068 unit PRBC/patient/day, P = 0.02) on multiple linear regression analysis. The control group also had a greater decline in Hb levels (2.13 ± 2.32 vs. 1.44 ± 2.08 gm/dL, P = 0.02) at discharge.ConclusionsThe use of a blood conservation device is associated with 1) reduced PRBC transfusion requirements and 2) a smaller decrease in Hb levels in the ICU.