#5460 acute kidney injury duration and 20-year risk of chronic kidney disease and cardiovascular disease: a population-based cohort study

Abstract Background and Aims Acute kidney injury (AKI) is associated with increased risks of chronic kidney disease (CKD) and cardiovascular disease (CVD). The duration of AKI may be an important clinical marker of individuals at high risk of adverse outcomes; however, the potential links between AKI duration and CKD and CVD remain unresolved. Therefore, we examined the associations between AKI duration and CKD and CVD in a population-based setting. Method Using population-based plasma creatinine (pCr) data, we identified individuals with first-time AKI in Denmark from 1 January 1990 to 31 December 2018. AKI was defined in accordance with the Kidney Disease: Improving Global Outcomes (KDIGO) creatinine criteria. To allow for determining AKI duration, analyses were restricted to individuals with an assessment of baseline kidney function by pCr within the prior year and evaluation of AKI duration by pCr within seven days after AKI onset. In accordance with the Acute Disease Quality Initiative (ADQI) 16 Workgroup criteria, AKI duration was categorized as “rapid reversal” if a pCr test within two days after AKI onset did not fulfill the AKI definition, as “persistent” if a pCr test between two and seven days after onset did not fulfill the AKI definition, and otherwise as “acute kidney disease” (AKD). CKD was defined by in- or outpatient hospital diagnoses codes or as ≥2 outpatient eGFR <60 ml/min/1.73 m2 separated by a least 90 days. CVD was defined by diagnosis codes and evaluated overall and as individual conditions including atrial fibrillation and flutter, ischemic heart disease, heart failure, stroke, and hypertension. When examining CKD, overall CVD, and specific CVDs only individuals without the condition were included in the analyses. Twenty-year cumulative risks and hazard ratios (HRs) were computed and compared across the three AKI duration groups. Results We identified 165,334 individuals with first-time AKI and with an assessment of baseline kidney function and AKI duration. Among these 36,514 (22%) had rapid reversal AKI, 22,619 (13%) had persistent AKI, and 110,499 (65%) had AKD. The median age was 72 years (interquartile range (IQR)), 62-80) and 52% were females. The most common comorbidities were CKD (37%) and hypertension (43%). Among the 106,916 individuals without prevalent CKD, the 20-year risks of CKD were 18.9% (95% confidence interval (CI), 18.2-19.6) for rapid reversal AKI, 23.2% (95% CI, 22.1-24.2) for persistent AKI, and 22.0% (95% CI, 21.6-22.5) for AKD. The adjusted HRs for CKD were 1.23 (1.17-1.30) for persistent AKI and 1.33 (1.28-1.39) for AKD when compared with rapid reversal AKI. Among the 59,949 individuals without prevalent CVD, the overall 20-year risks of CVD were 36.0% (95% CI, 34.0-38.0) for rapid reversal AKI, 33.2% (95% CI, 31.6-34.9) for persistent AKI, and 32.2% (95% CI, 31.4-33.0) for AKD. This corresponded to adjusted HRs of 1.02 (95% CI, 0.96-1.09) for persistent AKI and 0.97 (95% CI, 0.93-1.02) for AKD when compared with rapid reversal AKI. Findings were consistent across outcomes of atrial fibrillation or flutter, stroke, and hypertension. For ischemic heart disease and heart failure, persistent AKI was associated with adjusted HRs of 1.10 (95% CI, 1.01-1.20) and 1.13 (95% CI, 1.05-1.20), respectively, and AKD with adjusted HRs of 1.08 (95% CI, 1.02-1.15) and 1.08 (95% CI, 1.02-1.13), respectively, compared with rapid reversal AKI. Conclusion In conclusion, longer AKI durations were associated with increased long-term rates of CKD. AKI duration was not associated with rates of overall CVD; however, rates of ischemic heart disease and heart failure increased with longer AKI durations. The distinct increase in rates of CKD and specific CVDs with longer AKI durations illustrates the potential for using AKI duration as a risk marker when planning nephrology follow-up after AKI. Further studies examining whether interventions to shorten AKI duration prevent outcomes are warranted.