New scores predict short- and long-term mortality in patients with sepsis

Reuters Health - 04/02/2022 - For patients hospitalized with bloodstream infection, a pair of new scores can help doctors predict how likely they are to survive 14 days and six months, respectively, researchers in Germany report.

"The use of the BLOOMY (Bloodstream Infection due to Multidrug-resistant Organisms: Multicenter Study on Risk Factors and Clinical Outcomes) scores to identify populations at high risk of mortality could facilitate the development of therapeutic and management protocols for in-hospital patients at high risk of short-term mortality, and also decision making at hospital discharge with tailored follow-up of patients at high risk of long-term mortality," they write in The Lancet Infectious Diseases.

"The scores could also have a pivotal role in improving patient selection in clinical trials and epidemiological studies that assess the burden of sepsis," they add.

To develop and validate the two scores, Dr. Evelina Tacconelli of the DZIF German Centre for Infection Research in Braunschweig and her colleagues recruited 2,568 hospitalized adults at six German university hospitals. In-hospital mortality rate was 23.8% and six-month mortality rate was 41.6%.

Predictors of 14-day mortality and six-month mortality (C statistic 0.873 and 0.807, respectively) included age, platelet and leukocyte counts, C-reactive protein concentrations, malignancy, causative pathogen, BMI and in-hospital acquisition. For 14-day mortality, predictors also included mental status, hypotension, and mechanical ventilation on day 3. By contrast, for 6-month mortality, additional predictors included focus of infection, in-hospital complications and end-of-treatment glomerular filtration rate.

The research team collected microbiological, laboratory, clinical, treatment, and survival data on day 0 and day 3, then weekly from day 7. They followed participants for 6 months and included all who were alive on day 3 in their analysis, using logistic regression and Cox proportional hazards models to develop predictive scores. They validated the data with the C statistic and determined predictive accuracy using sensitivity, specificity, and predictive values.

The BLOOMY scores were validated in 1,023 patients with bloodstream infections. They predicted 65 (61%) of 106 deaths at 14 days and (70%) of 143 deaths at six months, with a sensitivity of 61.3%, a specificity of 86.4%, a positive predictive value of 37.6% and a negative predictive value of 94.4% for 14-day mortality, and a sensitivity of 69.9%, a specificity of 66.4%, a positive predictive value of 40.8% and a negative predictive value of 87.0% for six-month mortality.

"To our knowledge, the BLOOMY cohort is the largest prospective cohort that includes ICU (intensive-care unit) and non-ICU patients, and which is not restricted by severity of disease, which helps to avoid miscalibration and thus reduce the risk of selection bias and information bias," the authors write. "The BLOOMY scores were developed on the basis of readily available and well defined objective criteria that can be used with ease, at the patient's bedside, using cutoffs defined by health-care organisations and resources."

In an accompanying editorial, Drs. Mathias W. Pletz and colleagues of Jena University Hospital, also in Germany, write, "The study authors have provided important data that might help to stimulate further research in bloodstream infections, not only to assess mortality in the short term, but also mortality in the long term."

"Further studies in infectious diseases need to systematically capture long-term mortality risk and causes of death after hospital discharge," they add. "Only then can the scientific infectious diseases community identify the underlying pathomechanisms and use them to develop new diagnostic and therapeutic strategies that will cure, in the long term, patients with bloodstream infections. The development of the BLOOMY scores is an important step on this path."

SOURCE: https://bit.ly/3L7bmOs and https://bit.ly/3s8TJFa Lancet Infectious Diseases, online January 19, 2022.

By Reuters Staff



Posted on 4 February 20228 February 2022