Invasive group A Streptococcus infections in Gothenburg,

Abstract

Invasive group A Streptococcus infections in Gothenburg, Sahlgrenska University Hospital, January 2008 – June 2013 ABSTRACT Background: Invasive group A Streptococcal infection (iGAS) is a severe disease, often requiring admission to an Intensive Care Unit (ICU). Incidence of iGAS in Sweden has markedly increased during the last 30 years. The reason for this is not fully understood. Aim: First, to do a survey on local epidemiology, clinical manifestations and treatment for iGAS. Second, to investigate whether CRP and S-­‐lactate can be used as early predictors for intensive care and death, and if there is a correlation between CRP and S-­‐ lactate. Method: A retrospective medical record review of adults treated for iGAS at Sahlgrenska University Hospital, Gothenburg between 1st January 2008 and 30th of June 2013. Descriptive analysis was used for the first aim. For the second we used Fisher’s Exact Test for S-­‐lactate, and visualization with boxplot for CRP, as predictive values for ICU and death. Pearson’s analysis for correlation was performed correlation between CRP and S-­‐lactate. Results: 146 patients (60 female, 86 male, median age 61 years) were identified from medical records. Local incidence followed the same trend as the national. Incidence was highest in November -­‐ May. The main verified underlying focuses were skin and soft tissue (48%) and lung (14%). In only 4 % the throat was a suspected focus and in 14% of the patients focus was unknown. Antibiotics were given both as mono therapy and combined therapy. The most common antibiotic treatment before culture was cephalosporins and PcG and aminoglycosides in descending order. After etiological diagnosis was confirmed, the most used antibiotics were instead PcG, PcV and clindamycin. Twentyeight % of the patients (n=41, median age 55 y.) was admitted to ICU, and 49% (n=20) of those had no major comorbidity. Over all mortality rate within 1 month after hospital discharge was 12% (n=17/146, median age 79 y.). Four of those had no major comorbidity. The total number of patients with S-­‐lactate sample taken was 102. Patients with initial S-­‐lactate < 2.0 mmol/l were never admitted to ICU, however, 7% of them died. S-­‐lactate 2.0 - 3.9 mmol/l led to ICU in 39% and mortality was 2%. Among patients with S-­‐lactate ≥ 4.0 mmol/l, 56% were admitted to ICU and 34% died. CRP was found not to be a predictor for intensive care and death. There was no significant correlation between initial CRP concentrations and S-lactate. Conclusions: We saw a rising trend of iGAS 2008 – 2013, with incidence peaks during winter and spring, and the main focuses were soft tissue and lung. Elderly and patients with major comorbidities were in the majority over all and in all subgroups, but also young and healthy were affected. Initial S-­‐lactate, but not CRP concentrations, was a good predictor for disease severity.