Survival outcomes in older patients with different stages of acute kidney injury defined by the addition of urine output criteria.

Abstract

To the Editor: Globally, acute kidney injury (AKI) is a common life-threatening condition. AKI is more prevalent in elderly patients, with increasing mortality rates. Compared with younger patients, older patients with AKI have a higher risk of persistent injury and poorer prognoses.[1] Older individuals often have comorbidities, increasing their susceptibility to AKI. Kidney Disease: Improving Global Outcomes (KDIGO) criteria define and stage AKI based on the higher stage of AKI defined by either serum creatinine (SCr) level or urine output (UO).[2] However, the record of UO is difficult to access in clinic, the data on the value of UO combined with SCr for AKI diagnosis or prognosis in elderly patients are lacking. Therefore, the objectives of this study were to investigate whether UO would influence the classification of AKI and improve the accuracy of prediction of the clinical prognosis in elderly patients. We performed a retrospective cohort study to assess the survival outcomes of AKI in older patients with different stages of AKI based on SCr alone or the combination criteria. The study included male patients aged >65 years who presented to the Second Medical Center (Geriatric Department) of the Chinese People's Liberation Army General Hospital with AKI between January 1, 2008, and December 31, 2018. Patients with either dialysis-dependent end-stage kidney failure or incomplete medical history were excluded from the study. Recruited patients were followed up for >90 days after the AKI occurence to determine the clinical outcome of mortality. This study was registered in ChiCTR (http://www.chictr.org.cn/, ChiCTR2200055518) and was approved by the Ethics Committee of Chinese People's Liberation Army General Hospital (No. S2022–342-01). The informed consent of the present study was waived owing to the retrospective nature. The SCr diagnostic criteria defined in the KDIGO AKI guidelines were adopted to recruit patients; the criteria were as follows: (1) increase in SCr >26.5 μmol/L within 48 h or (2) increase in SCr exceeding 1.5-fold above the baseline value within 7 days. The severity of AKI was classified by the KDIGO staging criteria using the SCr criteria alone or a combination of SCr or UO criteria. AKI was staged into three stages for severity. Categorical variables were presented as as a percentage (%) and were compared using the chi-squared test. The Kaplan-Meier method and log-rank test were used to compare the overall survival rates among patients with different stages of AKI based on different classification criteria. Data were obtained from hospital electronic health records. Age, comorbidities, and accompanying conditions (such as mechanical ventilation, proteinuria) at AKI onset were included in univariable logistic regression analysis. Variables with P < 0.05 in the univariable regression analysis were included in the multivariable binary logistic regression equation as covariates to estimate odds ratios (ORs) of AKI stages for mortality. Statistical analyses were performed using SPSS version 22.0, for Windows (SPSS Inc., Chicago, IL, USA). A total of 1509 eligible patients who experienced AKI were included in the final cohort. Based on SCr criteria alone, patients were categorized as follows: stage 1 AKI (n = 1314; 87.1%), stage 2 AKI (n = 151; 10.0%), and stage 3 AKI (n = 44; 2.9%). According to the KDIGO criteria, the distribution of AKI was as follows: stage 1, 1023 (67.8%); stage 2, 290 (19.2%); and stage 3, 196 (13.0%). Overall, 556 patients (36.8%) died within 90 days. Compared with survival group, non-survival group had higher proportion of severe AKI (stage 3), either defined by the SCr criteria alone or the KDIGO criteria (SCr criteria: 4.1% [23/556] vs. 2.2% [21/953], P < 0001; KDIGO criteria: 27.2% [151/556] vs. 4.7% [45/953], P < 0001). The 90-day survival rate decreased from 65.7% (863/1314) in those with stage 1 to 45.7% (69/151) and 47.7% (21/44) for those with stages 2 and 3 AKI, respectively, based on SCr criteria. Kaplan-Meier curves for 90-day cumulative survival showed a significantly lower survival rate in stage 2 or stage 3 compared to stage 1 when categorized by SCr criteria alone (stage 2 vs. stage 1, hazard ratio [HR] = 1.92, 95% CI [1.42–2.58], P < 0.001; stage 3 vs. stage 1, HR = 1.72, 95% CI [1.02–2.90], P = 0.009) [Figure 1A]. However, there was no significant difference between stage 2 and stage 3 (stage 2 vs. stage 3, HR = 0.90, 95% CI [0.50–1.62], P = 0.659). In comparison, 90-day cumulative survival rate decreased gradually with the severity of AKI based on the combination criteria (stage 2 vs. stage 1, HR=2.05, 95% CI [1.64–2.54], P < 0.001; stage 3 vs. stage 1, HR = 5.08, 95% CI [3.74–6.91], P < 0.001; stage 3 vs. stage 2, HR = 2.49, 95% CI [1.75–3.53], P < 0.001) [Figure 1B]. The 90-day cumulative survival rate decreased from 73.6% (753/1023) in those with stage 1 AKI to 53.64% (155/290) and 23.0% (45/196) in those with stages 2 and 3 AKI, respectively, based on the combination criteria.Figure 1: Associations between AKI stages and 90-day cumulative survival. (A) Kaplan-Meier survival curves for 90-day cumulative survival of older AKI patients based on SCr criteria only. (B) Kaplan-Meier survival curves for 90-day cumulative survival of older AKI patients based on the combination of SCr and UO. AKI: Acute kidney injury; CI: Confidence interval; KDIGO: Kidney Disease: Improving Global Outcomes; SCr: Serum creatinine; UO: Urine output.Logistic regression analysis was performed to further evaluate the ORs of AKI stages for mortality. Age, baseline estimated glomerular filtration rate, hypertension, hyperlipidemia, myocardial infarction, atrial fibrillation, chronic heart failure, chronic obstructive pulmonary disease, tumor, anemia, mechanical ventilation, and proteinuria were significant in the univariable logistic regression analysis and included in the multivariable binary logistic regression equation as covariates. According to the multivariable binary logistic regression, the mortality risk of patients with stage 3 AKI and stage 1 when defined by the SCr criteria alone was not statistically significant in the adjusted model (adjusted OR = 1.65, 95% CI [0.80–3.41], P = 0.179), while the adjusted mortality risk in patients with stage 3 AKI were significantky greater compared with those with stage 1 AKI when defined by the combination criteria (adjusted OR = 7.03, 95% CI [4.72–10.49], P < 0.001) [Supplementary Table 1, https://links.lww.com/CM9/B438]. Large studies have shown that even small increases in SCr levels are independently associated with increased mortality. Compared to patients without AKI, the odds of mortality were progressively higher for patients at higher AKI stages based on SCr alone.[3] However, in this study, older patients with stage 3 AKI defined by SCr alone did not exhibit worse survival than those with stage 2 AKI. In older individuals, SCr is not considered an ideal biomarker for AKI stage. Our findings support this claim and further indicate the limited usefulness of SCr for classification and prognosis prediction. The rise in SCr is often blunted in elder population due to reduced muscle mass.[4] Older patients with AKI may have a higher incidence of pre-renal, nephrotoxic, and obstructive issues.[5] Moreover, older patients are more prone to decreased renal perfusion due to heart failure, dehydration, or depletion of effective circulating volume. These risk factors can directly result in oliguria rather than increase SCr. The addition of UO criteria to SCr criteria leads to an earlier recognition of AKI in elderly patients. In this study, AKI stages based on the combination of UO and SCr criteria have a better discriminative ability of mortality than that based on SCr alone. These findings confirm the absolute necessity to diagnose and stage AKI using criteria combined with SCr and UO. This study has a few limitations. First, this was a single-center retrospective study, and so the results may not be generalizable to older patients with AKI in other locations. Second, our analysis is based on a male veteran cohort, which also limits the generalizability of our findings. Studies in older women with AKI may yield different results. Finally, the outcomes of AKI could be confounded by different causes, which were not individually discussed. In conclusion, survival outcomes did not significantly worsen with more advanced AKI stage when defined by SCr alone in older male patients. With the addition of UO to the SCr criteria, mortality increased gradually with severity of AKI across all stages. It is necessity to diagnose and stage AKI using criteria combined with SCr and UO in older male AKI patients. Funding This work was supported by a grant from the Special Scientific Research Project of Military Health Care (No. 21BJZ17) Conflicts of interest None.