Study population
We conducted a retrospective observational cohort study of 88 elderly patients who were admitted with decompensated heart failure between September 2023 to April 2024 in Lagan Valley Hospital – Lisburn and Ulster Hospital Dundonald, Northern Ireland. These two major tertiary care and secondary care hospitals, respectively, cover the entire southeastern trust of Norther Ireland, which represents most of the elderly population in the country. Both hospitals have specialized geriatric protocols for elderly care with onsite availability of cardiac services. We assessed 120 patients with suspected heart failure over eight months. The cases were randomly selected as snapshots covering each month. Among these 120 patients, 88 were confirmed to have heart failure based on clinical symptoms, echocardiography findings, and NT-proBNP findings on admission.
We assessed 120 patients with suspected heart failure over eight months. The cases were randomly selected as snapshots covering each month. Among these 120 patients, 88 were confirmed to have heart failure based on clinical symptoms, echocardiography findings, and NT-proBNP findings on admission. Verbal informed consent was obtained from each patient for echocardiography, blood testing, and clinical history and follow-up, as well as for anonymized patient information sharing.
Baseline measurements
In all patients, a thorough medical history was recorded, including details of any previous myocardial infarction or revascularization, angina pectoris, arterial hypertension, suspected congestive heart failure (defined by symptoms of shortness of breath or leg edema), previous stroke or transient ischemic attacks, diabetes, atrial fibrillation, and any malignancy. This information was obtained from medical records, directly from the patients, or both. Clinical frailty was evaluated for each patient using the Rockwood Clinical Frailty Scale.
Echocardiography findings were evaluated for all the identified cases. Ejection fraction and presence of any regional wall motion abnormalities, valve pathologies, and LV thrombus was documented. NT-proBNP has been measured for majority of the patients as a part of the heart failure work up during hospital admission.
There were no exclusion criteria. The inclusion criteria as follows: >65 years of age or a Rockwood Clinical Frailty Scale score of > 4, established acute hospital admission with heart failure, and an NT-proBNP value of > 400 ng/L.
Follow up
Heart failure medications prescribed upon discharge were documented and categorized as beta blockers, diuretics, dapagliflozin, digoxin, antiplatelets, or anticoagulants. All 88 patients were followed up for 12 weeks after initial hospital discharge. Follow-up pathways were traced through electronic data records, including follow up with any cardiologist, heart failure nurse, or community-based heart failure care provider. Any medication changes or additions during follow up were recorded. Patients who received repeat NT-proBNP were also recorded. Follow up was monitored in terms of two major end points: heart failure–related death and re-admission due to heart failure.
Data analysis
All patients were sorted into two main cohorts: a heart failure with reduced ejection fraction (HFrEF/EF > 60%) cohort and a heart failure with preserved ejection fraction (HFpEF/EF > 40%) cohort. Each cohort was again categorized into three main sub-cohorts according to NT-proBNP value on admission: <400 ng/L, 400–2000 ng/L, and > 2000ng/L.
In total, six sub-cohorts were established, and data were analyzed for each sub-cohort. Average age, gender distribution, clinical frailty score, presenting symptoms, and past medical history were listed under demographic distribution data table. Follow-up details, heart failure admissions, and deaths were listed under morbidity and mortality data tables separately.
NT-proBNP values and end points
The same cohorts were further followed up for 12–15 weeks to monitor the two above-mentioned end points. In addition, whether patients received outpatient heart failure follow-up (e.g., with heart failure nurse, cardiac hub, or cardiology specialist) was also monitored. Repeat BNP values during follow-ups were compared with each patient’s initial BNP value and documented as either a “reduction” or “increase.” These parameters were again correlated with the aforementioned end points.( Tables 3 and 4 )
Furthermore, we observed whether there is an increased readmissions or deaths with rising NT-proBNP levels or whether there is decreased readmissions or deaths with declining NT-proBNP values
Categorical variables are reported as frequencies and percentages. Normally distributed continuous variables are presented as means ± standard deviations, whereas non-normally distributed continuous variables are presented medians and interquartile ranges. Student’s t-test or the Mann–Whitney U test for continuous variables, and the chi-squared or Fisher’s exact test for categorical variables were used comparison between groups, as appropriate. When needed, the variables were transformed for further analysis.