Proportion, Precipitators and Prognostic Marker of 30-Day Heart Failure Readmission at MOI Teaching and Referral Hospital, Eldoret, Kenya

Background: Heart failure (HF) affects 26 million people globally. It is associated with a high 30-day readmission rate due to multiple cardiovascular and non-cardiovascular precipitants. Readmissions are associated with high mortality, which may be predicted by pre-discharge N-terminal pro brain natriuretic peptide (NT-proBNP). There is no data on HF readmission rates, its precipitators and use of NT-proBNP as a prognostic marker at Moi Teaching and Referral Hospital (MTRH) in Western Kenya. Objective: To determine the 30-day proportion of HF readmission, the precipitators of HF and the association between NT-proBNP and 30 days’ readmission. Methods:This was a six-month prospective cohort study where we carried out a census and recruited adult participants admitted with HF at Moi Teaching and Referral Hospital. At discharge from hospital, an interviewer administered questionnaire was administered and blood samples for NT-proBNP drawn. Upon readmission, precipitators for HF were identified and compliance to therapy assessed using European heart failure Self Care Behavioral Scale. Continuous variables were summarized using median and IQR, categorical variables using frequencies and percentages. Associations were determined using Chi square, Fishers exact and Wilcoxon rank test. A p-value of 4137 pg/ml, which implied poor prognosis. Of 17 readmitted patients, 12 participants (12.8%) were readmitted within 30 days at MTRH. Six (6.4%) participants were lost to follow up. Median time to readmission was 14 days, (IQR 7, 26). Pneumonia (55%) was the commonest precipitator of HF readmissions, followed by arrhythmias (atrial fibrillation) in 5 (42%), anemia 4 (33%), noncompliance 4 (33%), acute kidney injury 2 (16.6%) and no identified precipitator 1 (8.3%). There was no association between NT-proBNP and readmission (p =0.584) or NT-pro BNP and survival (p=0.773). Readmission was associated with a high mortality (p=0.008) with 50% of readmitted participants dying during the readmission period. The total mortality of both readmitted and non-readmitted participants was 16% at the end of the 6 months’ study period. Conclusions:In this cohort of participants with HF the proportion of 30 days readmission was high. Infections, mainly pneumonia was the commonest precipitator of readmission. Discharge NT-proBNP did not predict likelihood of 30 days’ readmission. Mortality was higher among participants readmitted within 30 days. Recommendations: Measures like pneumoccal vaccinations should be implemented to prevent pneumonia. Early appointments to cardiac clinic (less than 2 weeks post discharge) should be given to screen for precipitators and reduce HF readmissions. A follow up study to assess % change in NT pro BNP in relation to 30 days readmission.