Sepsis registry in a tertiary care hospital – A 9 month observational study

Abstract

Background: Adult sepsis in India poses a major challenge for clinicians and hospitals given the complete lack of local and national data of its public health magnitude. Severe sepsis is often associated with high mortality and morbidity leading to increased cost of care for patients and institutions. To better respond to the clinical and financial cost of sepsis care in our tertiary care hospital we developed a sepsis registry based on the Surviving Sepsis Campaign. Methods & Materials: A computer based real time sepsis registry was developed to collect data of all sepsis patients presenting to the ED. This registry included demographic information, prehospital care, clinical information and patient care details as per the Surviving Sepsis Campaign guidelines. This database was aligned with our hospital laboratory information system. Frequent reviews of the data collection process, quality and completeness were done to optimize the registry. Results: Out of a total 301 patients, 66% (199/301) were males. 44% (131/301) were between 61-80 yrs while 34% (102/301) were between 41-60yrs of age. 40% (120/301) of the cases were transfers from nearby hospitals to our tertiary care. Further categorization revealed 46% cases were 'sepsis', 14% 'septic shock' and 40% were 'severe sepsis'. The SOFA score on admission of the 65% of the cases were < 9 while 19% had a SOFA score of 9-10 and 16% had a score more than 11. 51% (153/301) had an average length of stay < 7 days. The primary focus of infection were pneumonia (39%), UTI (38%) and skin and soft tissue infections (16%). 22% of the cases had concomitant bacteremia. Compliance with the 3hr bundle was present in 83% of the cases. Overall mortality was 28% (86/301) with male gender and a SOFA score of > 9 significantly associated with fatality (p<0.05). Lack of compliance with 3 hr bundle and lactate levels > 2.5 mmol/L were also associated with mortality (p<0.05). 71% of the registry cases were culture positive, of which 33% had a polymicrobial infection. Conclusion: This sepsis registry is proving to be a key data source for defining the burden of the disease in our community.