Reference: Ann Intern Med. 2020 Sep 15;173(6):426-435
A spot albumin/creatinine ratio (ACR) has largely replaced 24-hour urine collection as standard of care for finding diabetic patients who would benefit from treatment beyond glucose control to prevent kidney disease. ACE/ARB medications have demonstrated reductions in all-cause mortality among diabetic patients with ≥ 30 mg/g of albumin (microalbuminuria), and more recently, sodium-glucose transporter 2 (SGLT2) inhibitors such as canagliflozin have been shown to reduce renal events among patients with > 300 mg/g (macroalbuminuria). ACR can be costly and often not readily available at the point of care. This meta-analysis of individual-level data from 33 cohort studies of varying characteristics and sizes examined the predictive and prognostic values of the standard dipstick and protein to creatinine ratio (PCR) compared to ACR.
The investigators included cohorts of ≥ 200 participants from a variety of settings, all of whom had measured ACR and dipstick protein or PCR on the same day. A total of 33 cohorts with over 900,000 participants were included in the analysis (mean age 61 years, diabetes in 56%, hypertension in 72%, and chronic kidney disease [CKD] stage III or worse in 21%). Less than 5% of participants were Black. The relationships between ACR and both PCR and dipstick protein were modeled within each cohort, and in a series of statistical feats, complex equations were generated to convert both PCR and dipstick protein to estimated ACR for the purpose of predicting the progression of kidney disease. In over a million samples from diabetic participants, dipstick protein of ≥ 2+ had 78% sensitivity and 97% specificity for ≥ 300 mg/g albumin. This model predicted ACR values for trace protein on dipstick to be 25 mg/g (95% prediction interval [PI] 8-80 mg/g), 1+ to be 67 mg/g (95% PI 21-207 mg/g), and 2+ to be 337 mg/g (95% PI 132-860 mg/g). Pooled meta-analysis found that dipstick ≥ 1+ protein had a positive predictive value (PPV) of 86% and a negative predictive value (NPV) of 80% for ACR ≥ 30 mg/g. For ACR ≥ 300 mg/g, 1+ protein on dipstick had a PPV of 38% and a NPV of 99%, while 2+ protein had a PPV of 71% and a NPV of 98%.
The main message from this analysis is that the absence of ≥ 2+ protein on urine dipstick appears to effectively rule out macroalbuminuria. The ACR threshold of ≥ 30 mg/g has been used to indicate the need for ACE/ARB initiation, and now a threshold of 300 mg/g is recommended to initiate SGLT2 inhibitor therapy in diabetic patients. It’s important to keep in mind, however, that a test should only be ordered (despite mandates for performance measures) if it has the potential to make a diagnosis, change management, and change clinical outcomes. So perhaps the bigger question is, should we be checking for albuminuria at all for diabetic patients already on ACE/ARB medications and SGLT2 inhibitors? As stewards of healthcare systems striving to reduce overuse in the absence of impact on clinical outcomes, using dipstick analysis is perhaps at best a surrogate for a surrogate. Nevertheless, a cheaper but equivalent test seems to be an obvious choice when testing is mandated or actually indicated for management decisions.
For more information, see the topic Diabetic Kidney Disease in DynaMed.
DynaMed EBM Focus Editorial Team
This EBM Focus was written by Carina Brown, MD, Assistant Professor at Cone Health Family Medicine Residency and Katharine DeGeorge, MD, MS, Associate Professor of Family Medicine at the University of Virginia and Clinical Editor at DynaMed. Edited by Alan Ehrlich, MD, Executive Editor at DynaMed and Associate Professor in Family Medicine at the University of Massachusetts Medical School, and Dan Randall, MD, Deputy Editor for Internal Medicine at DynaMed.