Abstract: FR-OR25
Dipstick Urinalysis Protein and Specific Gravity Can Identify Patients With Early CKD Who Lack a Quantified Proteinuria Measurement
Session Information
- CKD: Clinical Outcomes, Trials, Mechanistic Insights
November 04, 2022 | Location: W307, Orange County Convention Center‚ West Building
Abstract Time: 05:06 PM - 05:15 PM
Category: CKD (Non-Dialysis)
- 2201 CKD (Non-Dialysis): Epidemiology‚ Risk Factors‚ and Prevention
Authors
- McAdams, Meredith C., The University of Texas Southwestern Medical Center, Dallas, Texas, United States
- Li, Michael M., The University of Texas Southwestern Medical Center, Dallas, Texas, United States
- Xu, Pin, The University of Texas Southwestern Medical Center, Dallas, Texas, United States
- Willett, Duwayne L., The University of Texas Southwestern Medical Center, Dallas, Texas, United States
- Hedayati, Susan, The University of Texas Southwestern Medical Center, Dallas, Texas, United States
Background
Urine albumin-to-creatinine ratio (UACR) >30 mg/g or protein-to-creatinine ratio (UPCR) >0.15 g/g are the gold standard cutoffs for diagnosing early stages of CKD for patients with preserved estimated glomerular filtration rates (eGFR) >60 mL/min, but these tests are not commonly obtained. Dipstick urinalysis protein (DSP) is a widely-available and routinely obtained test that provides a semi-qualitative measurement of albuminuria but is affected by urine specific gravity (SG).
Methods
We conducted an analysis using the EPIC electronic health records of 11,229 patients with a UPCR or UACR value obtained on the same day as a urinalysis. Prognostic utility of various DSP cutoffs (negative/trace, 30, 100, 300, and ≥500 mg/dL) were compared to clinically significant proteinuria (UACR >30 mg/g or UPCR >0.15 g/g). Predictive models for proteinuria were built using DSP, adjusted for SG and assessed with receiver operating characteristic (ROC) curves comparing areas under the curve (AUC). 10-fold cross-validation was performed.
Results
Of 11,229 included, 4073 (36%) had clinically significant proteinuria based on gold standard definitions. 54.0% were female, 27.5% had diabetes mellitus, 51.3% hypertension, and 11.9% cardiovascular disease. A DSP of 30 mg/dL had an odds ratio (OR) (95% CI) of 5.84 (5.22, 6.52) and 9.92 (8.73, 11.27), when adjusted for SG, to predict proteinuria. A DSP of 100 mg/dL had an OR of 44.63 (95% CI 36.11, 55.15) and 77.66 (95% CI 61.67, 97.80), when adjusted for SG. A DSP value ≥300 mg/dL had an OR of 278.37 (95% CI 115.00, 673.81) and 495.91 (95% CI 202.03, 1, 217.26) in the SG adjusted model. Addition of SG to DSP improved the AUC to 0.824 (95% CI 0.815, 0.833) from 0.765 (95%CI, 0.756, 0.773) for DSP alone, P<.001. The final model including SG had a specificity of 93%, positive predictive value of 83%, negative predictive value of 80%, and positive and negative likelihood ratios of 9.52 and 0.43. Optimal predicted SG cutoffs to use for each DSP value were negative/trace=1.001; 30 mg/dL=1.024; 100 mg/dL=1.044; ≥300 mg/dL=1.063.
Conclusion
Combining DSP and SG from dipstick urinalysis can accurately determine clinically significant proteinuria to identify patients with early stages of CKD in clinical practice.