Mendu ML, Lundquist A, Aizer AA, Leaf DE, Robinson E, Steele DJR, Waikar SS. The Usefulness of Diagnostic Testing in the Initial Evaluation of Chronic Kidney Disease. JAMA Intern Med. 2015;175(5):853–856. doi:10.1001/jamainternmed.2015.17
Chronic kidney disease (CKD) affects approximately 13% of adults in the United States and is associated with significant morbidity, mortality, and costs.1- 3 There is a broad differential for CKD, including diabetes mellitus, hypertension, glomerulonephritis, tubulointerstitial disease, urologic causes, and unknown causes.2 To our knowledge, a comprehensive assessment of the tests used in CKD evaluation has not been conducted. We determined how often laboratory and imaging tests were obtained in the initial evaluation of CKD and whether these tests affected diagnosis and/or management.
We conducted a retrospective cohort study of patients referred for initial evaluation of CKD from January 1, 2010, to January 1, 2013, to nephrology clinics affiliated with Brigham and Women’s Hospital and Massachusetts General Hospital in Boston, Massachusetts; 1487 patients were included (Table 1). Partners Institutional Review Board approved the study and waived the need for informed consent. Electronic medical records were abstracted. We used methods to ensure the validity and reliability of data, including review of 10 initial medical records by 2 of us (M.L.M. and S.S.W.) to refine criteria.4 Tests obtained at another clinic before the nephrology clinic visit were documented.
We reviewed nephrology progress notes to ascertain the presumed cause of CKD and whether a test had been documented to affect the diagnosis and/or management. A test was considered to have affected diagnosis and/or management if it was specifically stated to have contributed to, confirmed, or established the underlying diagnosis of and/or any management decision related to CKD. This definition included documentation of negative and positive test results and diagnoses related to CKD. A second reviewer (E.R.) blindly abstracted a random sample of 36 patients’ records (2.4% of patients). The degree of interrater agreement, assessed by the prevalence-adjusted, bias-adjusted κ statistic,5,6 was a mean (SE) of 0.89 (0.02).
Among the 1487 patients included, common comorbidities were hypertension (79.0%) and diabetes (58.4%), and CKD stages were 3b (39.5%) and 3a (28.7%) (Table 1). Frequently obtained tests included measurement of calcium (94.8%), hemoglobin (84.0%), phosphate (83.5%), urine sediment (74.8%), and parathyroid hormone (74.1%) levels; urine dipstick for blood (69.9%) and protein (69.7%); serum protein electrophoresis (68.1%); and renal ultrasonography (67.7%) (Table 2). Determination of the hemoglobin A1c level, urine total protein to creatinine ratio, and urine microalbumin to creatinine ratio had relatively high yields, affecting diagnosis in 15.4%, 14.1%, and 13.0% of the patients and management in 10.1%, 13.7%, and 13.3%, respectively. Serum protein electrophoresis and renal ultrasonography, although frequently performed, had much lower yields, affecting diagnosis in 1.4% and 5.9% and management in 1.7% and 3.3% of the patients, respectively. Results of tests to detect antineutrophil cytoplasmic antibody and antiglomerular basement membrane antibody did not affect the diagnosis or management in any patients.
In this analysis of patients undergoing initial evaluation of CKD, we found that many tests are obtained frequently despite low rates of effect on diagnosis and management. Certain tests, such as serum protein electrophoresis and screening for antinuclear antibody, C3, C4, hepatitis C, hepatitis B, and antineutrophil cytoplasmic antibody, were obtained often (13.4%-68.1%) despite infrequently affecting diagnosis or management (0-1.7%). In contrast, hemoglobin A1c and urine protein quantification tests affected the diagnosis and management in 13.0% to 15.4% of the patients. These findings are limited by the retrospective study design, subjective nature of evaluating clinical usefulness, potential underestimation of the benefit of negative test results, and representation from only 2 academic medical centers in the northeastern United States. Further investigation incorporating community-based patients and identifying subgroups benefiting from more extensive evaluation is needed. However, this study suggests that reflexively ordering several tests for CKD evaluation and management may be unnecessary. An evidence-based, targeted approach based on pretest probabilities of disease for diagnosis and management may be more efficient and reduce costs.
Corresponding Author: Mallika L. Mendu, MD, MBA, Division of Renal Medicine, Brigham and Women’s Hospital, Harvard Medical School, 75 Francis St, MRB-4, Boston, MA 02115 (firstname.lastname@example.org).
Published Online: March 2, 2015. doi:10.1001/jamainternmed.2015.17.
Author Contributions: Drs Mendu and Waikar had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Mendu, Aizer, Steele, Waikar.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Steele, Mendu.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Mendu, Aizer.
Administrative, technical, or material support: Mendu, Leaf, Waikar.
Study supervision: Robinson, Steele, Waikar.
Conflict of Interest Disclosures: Dr Waikar served as a consultant to Abbvie, CVS Caremark, Harvard Clinical Research Institute, and Takeda; provided expert testimony or consultation for litigation related to nephrogenic systemic fibrosis (GE Healthcare) and mercury exposure; and has received grants from the National Institute of Diabetes and Digestive Kidney Diseases, Genzyme, Merck, Otsuka, Pfizer, and Satellite Healthcare. No other conflicts are reported.