Risk factors and outcome of community onset Pseudomonas aeruginosa bacteraemia in two Malaysian district specialist hospitals

INTRODUCTION: Despite the ever-growing number of community onset (CO) Pseudomonas aeruginosa (P. aeruginosa) bacteraemia, there is a dearth of district hospital-based research examining this significant infection, which is associated with high mortality. The objectives of this study were as following: (1) to determine the risk factors of CO P. aeruginosa bacteraemia, (2) to compare the 30-day mortality rate between P. aeruginosa and Escherichia coli bacteraemia and (3) to identify the predictors of 30-day mortality for CO gram negative bacteraemia.

METHODS: This is a retrospective case control study in Hospital Seri Manjung and Hospital Teluk Intan, Perak, Malaysia. P. aeruginosa bacteraemia cases that occurred between 1st January 2015 to 31st December 2019 were included, whilst E. coli bacteraemia cases that occurred within the same period were recruited successively until 1:2 case control ratio was achieved. Subjects below 12-year-old and those with polymicrobial bacteraemia were excluded. Demographic, clinical and treatment data were collected using pre-tested data collection forms by trained investigators.

RESULTS: A total of 61 patients with P. aeruginosa bacteraemia and 122 patients with E. coli bacteraemia were included. Recent admission in the earlier three months, regular haemodialysis, immunosuppressive therapy in the past 30 days, chronic wound/pressure sore at presentation and indwelling urinary catheter at presentation were identified as independent predictors of CO pseudomonal bacteraemia. Whilst older age was identified as a negative predictor of CO Pseudomonal bacteraemia (all p<0.05). The 30-day mortality rate was 34.4% in subjects with P. aeruginosa bacteraemia and 27.0% in those with E. coli bacteraemia (p=0.302). Predictors of 30-day mortality for community onset gram negative bacteraemia were as follow: older age, underlying solid tumours, neutropaenia at presentation, in-patient mechanical ventilation, and inpatient nasogastric tube insertion. Unexpectedly, receipt of inappropriate empirical antibiotics which was switched later (delayed and non-delayed switching) was identified as the negative predictors of mortality (all p<0.05).

CONCLUSION: It is prudent to restrict the usage of empirical anti-pseudomonal antibiotics among individuals at risk as liberal usage of broad-spectrum antibiotics engenders emergence of drug resistant organism, particularly in district setting where community onset pseudomonal bacteraemia remains scarce. Subjects with elevated risk of mortality should receive early escalation of care as per sepsis management guidelines.