Bloodstream infections caused by gram negative bacteria in geriatric patients: Epidemiology and risk factors for mortality

Bloodstream infections (BSI) are an important cause of morbidity and mortality in geriatric patients. This retrospective study aimed to examine the risk factors for epidemiology and mortality in patients aged 65 and over who were admitted to the hospital between 1 January 2017 and 31 December 2022 and developed bloodstream infections caused by gram-negative bacteria. Patients aged 65 and over who were hospitalized and who developed a bloodstream infection when admitted to the hospital or during their follow-up, who had gram-negative bacterial growth in their blood samples, who had a first bacteremia attack, and who had a single bacterial growth were included in the study. Of the 110 patients in the study, 58 (52.7%) were male and 52 (47.3%) were female. Their ages ranged from 65 to 95, and the average age was 74.2 years. Except for one patient, the patients had a history of at least one physician-diagnosed chronic disease. 70 (63.6%) of the infections were healthcare-related. The most common detectable source of infection was the urinary system (43.6%). Fever response developed in 82 (74.5%) of the patients, 35 (31.8%) had hypotension in the first 2 days, and 26 (23.6%) required vasopressor agents in the first 2 days. The average serum C-reactive protein (CRP) level of the patients when first studied was 153.1 mg/L, while the average procalcitonin level was 5.1 ng/ml. The three most frequently growing bacteria in blood culture were Escherichia coli, Pseudomonas aeruginosa, and Klebsiella pneumoniae, respectively. In the study, carbapenem resistance was detected in 17 patients (15.5%), while extended-spectrum beta-lactamase (ESBL) production was present in 37 (51.4%) of the patients with growth belonging to the Enterobacterales family. Appropriate empiric antimicrobial treatment was started in 66 (60.0%) of the patients, and antibiotics were changed in 76 (69.1%). At the 14-day clinical endpoint, 29 patients (26.4%) died, and at the 28-day clinical endpoint, 37 patients (33.6%) died. Mortality was higher in those of male gender, those followed in the intensive care unit, and those with thrombocytopenia (p<0.05). Mortality was also higher in patients who developed hypotension, required vasopressor agents, developed acute renal failure, increased serum total bilirubin levels, and developed hyponatremia (p<0.05). The increase in mortality was found to be statistically significant in patients with carbapenem-resistant growth, Pitt bacteremia score ≥4, qPitt bacteremia score ≥2, and patients with high procalcitonin, ferritin or lactate levels at the onset of BSI (p<0.05). According to the results of multiple regression analysis, mortality was approximately seventeen times higher in patients with a Pitt bacteremia score of four and above, and approximately six times higher in patients with with hyperbilirubinemia. A significant portion of elderly patients who develop BSI progress to sepsis or septic shock, and the excess of comorbid conditions, bacterial resistance and weakening of the immune system increase mortality. It is thought that the use of procalcitonin at the beginning is superior to the use of CRP in the follow-up of patients with BSI and the selection of antimicrobial treatment, and the increase in ferritin level can be used as a prognostic marker. PBS and qPitt scores may also be supportive in patient management and predicting mortality in gram-negative bacteremias that develop in geriatric patients. Keywords: Geriatrics, gram-negative bacteria, epidemiology, mortality