Dr. Ifeoluwa Stowe is a nephrology fellow at Beth Israel Deaconess Medical Center/Harvard Medical School. She was born and raised in Nigeria where she completed medical school at the University of Lagos. She thereafter completed her Internal Medicine Residency and Nephrology Fellowship at the Lagos University Teaching Hospital; serving as the Chief Resident of the Department of Medicine for 2 years while doing her nephrology fellowship. She moved to the United States in 2021 for Internal Medicine residency training at Baton Rouge General Medical Center, Louisiana. She is interested in clinician education, mentorship and leadership and is actively involved in resident education; presently pursuing the clinician educator track at BIDMC. Her academic interests include acute kidney injury, acid-base disorders and onconephrology. Follow her @ifeolaoye. Dr Stowe is a 2025-26 AJKD Editorial Intern.

 

Acute declines in kidney function commonly occur in patients with acute heart failure (AHF). This is a consequence of various mechanisms including increased renal venous congestion, reduced arterial perfusion (low cardiac output), intense neurohormonal activation and oxidative stress, and inflammation (Figure). Declines in kidney function also occur in the context of treatment of AHF and overall are associated with increased risk of mortality; however, there has been some evidence that this risk is reduced if occurring within the context of decongestion, and so little declines can be tolerated. But some questions remain:

• What is the magnitude of decline that can be tolerated? Almost everyone defines acute kidney injury according to the Kidney Disease Improving Global Outcomes (KDIGO) guidelines definition of a ≥ 3 mg/dl rise in creatinine. More recently, a post hoc analysis of a randomized control trial inferred that eGFR declines up to 20%; and perhaps up to 40% in the setting of decongestion are associated with a lower risk of mortality.
• How do we correctly determine that this decline is a consequence of the therapeutic intervention and is truly benign?
• Does the timing of this decline influence cardiovascular (CV) and kidney outcomes? This question was raised by a position statement of the Heart Failure Collaboratory-Academic Research Consortium (HF-ARC).

In the AJKD article by McCallum et al, the authors retrospectively studied acute declines in kidney function of 3,931 patients with AHF using data from the Efficacy of Vasopressin Antagonism in Heart Failure Outcome Study with Tolvaptan (EVEREST) trial. Acute decline in kidney function was defined as an increase in serum creatinine by ≥0.3 mg/dl or >50% increase in serum creatinine per KDIGO definitions or a percent change in creatinine as a continuous variable at 3 days, 7 days, and 14 days post randomization. The authors evaluated the association between the decline in kidney function with all-cause mortality and incident estimated glomerular filtration rate (eGFR) <30 ml/min/1.73 m² (the primary outcomes). The secondary outcomes were a composite of CV mortality or first HF rehospitalization, and >40% eGFR decline. They found that AKI at 3 days post randomization showed no association with the composite CV outcome (HR=0.96 [0.89, 1.05]) or all-cause mortality (HR=0.98 [95% CI 0.88, 1.09] per 30% creatinine increase) whereas acute declines in kidney function at 7 days were associated with the composite CV outcome (HR=1.10 [1.03, 1.18]) and mortality (HR=1.19 [1.10, 1.30] per 30% creatinine increase). In addition, AKI at 14 days was associated with composite CV outcomes (HR=1.15 [1.08, 1.23]) and mortality (HR=1.27 [1.16, 1.38] per 30% creatinine increase).

When AKI was defined as an increase in serum creatinine by ≥0.3 mg/dl and the kidney outcome was incident eGFR<30 ml/min/1.73 m², AKI at 3 days (HR=2.41[1.90,3.05]), 7 days (HR=3.13[2.55,3.83]) and 14 days (HR=3.06[2.47,3.80]) were all associated with this adverse kidney outcome. Furthermore, AKI (when defined as a ≥0.3 mg/dl rise in creatinine) at 3 days (HR=2.33[1,89,2.87]), 7 days (HR=3.50[2.93,4.18]) and 14 days (HR=3.31[2.71,4.03]) remained associated with the secondary adverse kidney outcome defined as a >40% decline in eGFR. In addition, these associations with the primary and secondary kidney outcomes persisted when AKI was defined as a >50% increase in serum creatinine or a percent change in creatinine as a continuous variable.

It is reasonable to hypothesize that early AKI was associated with a more favorable cardiac and overall outcome because it likely occurred in the context of decongestion. However, the authors did not adjust for markers of decongestion; their rationale was to enable standardization of trials and avoid complicated pre-specified variables. In practice, creatinine changes are not interpreted in isolation, especially in patients with AHF where rise in creatinine is ‘permissible’ during decongestion. Therefore, future research may need to reconcile the need for universal endpoints with the reality of clinical practice.

The variables adjusted for included medications that are of therapeutic benefit in AHF and are also associated with hemodynamic declines in eGFR, such as angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs). It would be interesting to see the results of newer trials incorporating more recent therapeutic medications introduced after the EVEREST trial recruitment period, such as the sodium-glucose cotransporter-2 inhibitors (SGLT2i). The exclusion of patients with creatinine ≥3.5 mg/dl (which is not unique to this trial) limits generalizability to those with more advanced kidney disease.

AKI occurring at 7 and 14 days was associated with an increased risk of mortality and adverse cardiovascular outcomes. This increased risk was consistent across all outcomes regardless of the definition of the acute decline in kidney function used. This again raises the question of the magnitude of AKI that can be considered benign. Although it is difficult to translate this into clinical practice, it would ultimately be relevant to have more clinical trials define AKI as a continuous exposure using percentage creatinine change, in addition to the already existent guideline-based definitions, as this may lead to further refinement of how we define AKI in the future.

It remains important to note that acute declines in kidney function occurring at all prespecified time points were associated with adverse kidney outcomes. It is well known that chronic kidney disease is in itself is associated with worse CV outcomes and mortality, so it is perplexing that the patients with AKI at 3 days who had adverse kidney outcomes did not have an increased risk of CV outcomes in this study. This could perhaps be explained by the follow-up duration of the study (median follow-up was 9.9 months); possibly if followed longer, that subset of patients may have developed adverse CV outcomes eventually. In addition, although CKD is associated with worse CV outcomes, this does not apply to every patient and we do not know if these patients had other factors that helped mitigate their risk, but it is hard to invoke this sentiment in this set of patients who already have a diagnosis of AHF. As a nephrologist, this prompts the question: when am I truly comfortable with an acute kidney injury in a patient with AHF exacerbation? The answer, in this particular setting, is never.

Despite being uncomfortable with acute declines in kidney function in this setting regardless of timing, I would gladly decongest early and tolerate initial modest declines in eGFR for the CV and mortality benefits. I am delighted that this study has shed some light on possible safety end points for future trials of AHF therapies, and I look forward to the results of other trials that will hopefully add further to this conversation.

-Post prepared by Ifeoluwa Stowe