Endocrine Abstracts

Background

Type 2 Diabetes Mellitus (T2DM) and Heart Failure (HF) are two directly related diseases associated with considerable morbimortality. The prevalence of HF in T2DM populations is up to 20%, 4 times higher than the general population. Additionally, health-related quality of life, hospitalizations for HF (HHF) and mortality risk are more common when both are present. Hence, HHF risk assessment in T2DM patients may lead to prompt preventive interventions.

Aim

To assess the stratification accuracy of the Thrombolysis In Myocardial Infarction (TIMI) Risk Score for Heart Failure in Diabetes (TRS-HFDM) for predicting hospital admissions for HF in a T2DM adult outpatient population.

Material and methods

We conducted a cross-sectional study in T2DM adult outpatients between December 2015 and December 2020. The TRS-HFDM, a novel, practical and validated risk score, was applied for prediction of hospital admissions for HF. It comprises prior HF (yes = 1), history of atrial fibrillation (yes = 1), CAD (yes = 1), estimated glomerular filtration rate (yes = 1), and urine albumin-to-creatinine ratio (microalbuminuria [> 30 mg/g] = 1; macroalbuminuria [> 300 mg/g] = 2). The TRS-HFDM ranges from 0 to 7 points, where 0, 1, 2 and ≥ 3 points are graded as ’low risk’, ’intermediate risk’, ’high risk’ and ’very high risk’. ROC curve analysis was used to find the most appropriate TRS-HFDM cutoff point for the highest possible combination of sensitivity and specificity in predicting the primary endpoints: emergency room (ER) visits and HHF. Mann-Whitney U test was performed to compare TRS-HDFM scores between patients with and without the aforementioned outcomes. Binary logistic regression was applied to characterize the relationship between these variables.

Results

This study encompassed 353 T2DM patients, 197 women (55.8%), with a mean age (SD) of 73 (9) years. HF was present in 109 (30.9%) patients. 48 (13.6%) individuals had ≥ 1 ED visits and 36 (10.2%) ≥ 1 HHF. A TRS-HFDM cutoff of ≥ 3 showed the best combination of sensitivity/specificity in predicting ER visits (97.9% and 76.1%, respectively) and HHF (100% and 73.5%, respectively). There was a statistically significant increase in scores between patients who had those outcomes and those who did not (ER visit: OR = 2.835, P < 0.001; HHF: OR = 2.886, P < 0.001).

Discussion and conclusions

Our analysis proposes a cutoff of ≥ 3 to identify T2DM outpatients at risk for hospital admissions for HF. Greater absolute reductions are expected if early patient-tailored protective interventions take place in higher scores. Therefore, this simple tool may help in diminishing morbimortality in T2DM outpatients.