Does Kidney Disease Without AKI Warrant More Attention?

Tejas P. Desai, MD


September 24, 2019

Previously, I discussed a new predictive model for acute kidney injury (AKI). This month, I'd like to continue with the AKI theme and highlight a possibly new and concerning finding: acute kidney diseases and disorders without AKI (AKD without AKI). In JAMA Network Open, James and colleagues[1] suggest that a "subclinical" state of AKI exists and portends a worse prognosis than initially anticipated.

The authors contend that a substantial portion of patients transiently develop acute kidney damage that falls below the threshold for AKI (as defined by Kidney Disease Improving Global Outcomes [KDIGO]). Because these patients are not considered to have sustained any meaningful kidney damage, they are not monitored as closely as those who have full-blown AKI. The Canadian researchers hypothesize that these "normal" patients actually have a worse prognosis and should have closer nephrology follow-up than is typical.

To evaluate their suspicion, the authors performed a retrospective review of over 1 million patients in Alberta who had an estimated glomerular filtration rate > 15 mL/min/1.73 m2 and at least one serum creatinine measurement in 2008. Patients were split into one of five categories: CKD, AKI, AKD without AKI, CKD plus AKD without AKI, or no kidney disease. The graphic below shows the definitions of each category.

The data set was analyzed over the next 8.5 years to calculate the risk for one of four outcomes: time to all-cause mortality, development of CKD, worsening CKD, or development of end-stage kidney disease requiring dialysis. Several confounding variables were also measured, including age, sex, socioeconomic status, presence of hypertension, Aboriginal ethnicity, and Charlson comorbidities. A multivariable-adjusted hazard ratio (HR) was calculated for each of the four outcomes, but correction for multiple-hypothesis testing was not performed, and the P value for significance was set at < .05.

Download the PDF.

Patients with AKD without AKI had a greater likelihood of all four outcomes than patients with true normal kidney function. The risk for end-stage kidney disease requiring dialysis was significantly elevated (HR, 8.56; 95% confidence interval [CI], 7.32-10.01), followed by CKD (HR, 3.17; 95% CI, 3.1-3.23) and time to death (HR, 1.42; 95% CI, 1.39-1.45). For patients with CKD at baseline, the presence of concomitant AKD without AKI resulted in a similar likelihood of CKD progression as in those with CKD and concomitant AKI at baseline (HR, 1.38 for both; 95% CI, 1.33-1.44 for both).


As a retrospective study, the analysis is limited. Although this paper suggests that even the smallest amount of kidney damage could portend a poor prognosis, I would like to see a reevaluation of the data using a multiple-hypothesis testing correction factor. If the results remain significant, then the next step would be a more robust investigational study design. In the interim, should we look more closely at patients classified as having AKD without AKI?

I found this an interesting read, but not one that will change my practice in the near future. Feel free to disagree in the comments section below.

Follow Tejas P. Desai, MD, on Twitter: @nephondemand

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.