Temporal trends in first rehospitalisation for heart failure between 1997 and 2017. trends in hospitalization characteristics and prognostic impact

Background

Neurohormonal blockade/modulation has improved life expectancy in HF, but it is unknown whether the time between a first hospitalisation for HF and the first rehospitalisation has increased.

Purpose

To investigate temporal trends of time between first hospitalisation for HF and first rehospitalisation for HF.

Methods

Using nationwide registers, 43,176 patients with a first hospitalization of HF during 1997–2017 were included. Temporal trends in a) time to first rehospitalisation for HF and b) mortality and rehospitalisation for HF after first rehospitalisation for HF were investigated, as was length of stay and inpatient mortality during first rehospitalisation. Rehospitalisation within 30 days was considered the same event as the index diagnosis.

Results

Between 1997 and 2017, the time between first hospitalisation for HF and first rehospitalisation increased from a median of 351 days to 464 days. During first rehospitalisation for HF, the median length of stay decreased from 6 days to 3 days, in-hospital death decreased from 10% to 6%, and the 90-day absolute risk of death after rehospitalisation decreased from 25% to 19% (Figure 1). After the first rehospitalisation for HF, the 1-year age-standardized mortality and HF rehospitalisation rates decreased, during 1997–2017, from 43 to 30/100-person and 35 to 25/100-person, respectively (Figure 2). Reductions in 5-year mortality and HF hospitalization were also observed. After discharge, an increasing number of patients were started on ACE-I or ARB (1997–2017: 73.7%–79.4%), beta-blockers (1997–2017: 33.7%–83.0%), and MRAs (1997–2017: 43.1%–49.7%) within 90 days after hospitalisation.

Conclusions

Between 1997 and 2017, the median time from first hospitalisation to first rehospitalisation for HF increased by 100 days, while the length of stay and in-hospital death during hospitalisation for HF decreased markedly. 1-year and 5-year mortality and HF rehospitalisation rates after first rehospitalisation for HF also decreased between 1997 and 2017.

Funding Acknowledgement

Type of funding sources: None.

Prevalence of diabetic nephropathy and subsequent cardiovascular outcomes: a nationwide study of 74,014 patients with type 2 diabetes

Background

Nonsteroidal, selective mineralocorticoid receptor antagonists (MRAs) have been shown to reduce cardiovascular events among patients with diabetic nephropathy, a clinical syndrome characterized by persistent albuminuria. However, prior reports have suggested a substantial underuse of MRAs among eligible patients. In real-life, it is unknown how many have urinary excretion of albumin measured, and the cardiovascular outcomes among type 2 diabetes (T2D) patients with versus without albuminuria is relatively unknown.

Purpose

To identify candidates eligible for treatment with MRAs and compare the risk of cardiovascular events in real-life T2D patients with albuminuria versus without albuminuria.

Methods

Using the Danish nationwide registers, we identified all patients ≥18 years old with prevalent T2D at index date 1st January 2015 with an albumin-creatinine ratio (ACR) and a creatinine level measured within 365 days prior to the index date. For each patient, the last ACR and estimated glomerular filtration rate (eGFR) registered prior to the index date were used. eGFR was calculated using the CKD-EPI formula. The patients were separated into two groups consisting of patients with T2D with ACR ≥30 mg/g (albuminuria) or ACR<30 mg/g (no albuminuria), respectively. Outcomes for both groups were analyzed as time-to-event as a composite cardiovascular outcome of heart failure (HF), myocardial infarction (MI), stroke, and all-cause death and each component of the composite endpoint was analyzed individually. In both groups, we estimated the 4-year absolute risk of the cardiovascular outcome and the risk of experiencing HF, MI, and stroke.

Results

A total of 158,904 patients with T2D were identified and of those 74,014 patients (47%) had urinary ACR analyzed within the last year. Of those who had ACR analyzed 29,581 (40%) patients had albuminuria. The absolute 4-year risk of the composite cardiovascular outcome in patients with albuminuria and no albuminuria was 28.6% (95% confidence interval (CI): 28.1–29.1) versus 18.7% (95% CI: 18.4–19.1), respectively. The absolute 4-year risk for the individual components were HF 7.0% (95% CI: 6.7–7.3), MI 4.4% (95% CI: 4.2–4.6), and stroke 7.6% (95% CI: 7.3–7.9) for patients with albuminuria, versus HF 4.0% (95% CI: 3.8–4.2), MI 3.2% (95% CI: 3.1–3.4), and stroke 5.5% (95% CI: 5.3–5.7) for patients with no albuminuria.

Conclusion

In Denmark, only 47% of patients with T2D had ACR analyzed within a year indicating lack of adherence to the national T2D guidelines. A total of 40% of those who had albuminuria analyzed were eligible for treatment with MRAs based on trial inclusion criteria, and the absolute 4-year risk of experiencing a cardiovascular event was significant. However, patients with no albuminuria still displayed a substantial risk of experiencing a composite event of HF, MI, stroke, and all-cause death. Our analyses may have implications for implementation of new MRAs in patients with T2D.

Funding Acknowledgement

Type of funding sources: None.