Letters to the Editor
To the Editor:
There has been a steady increase in the proportion of deaths associated with renal disease among HIV-infected individuals.1 Many factors might be responsible for this increase, including older age and high prevalence of comorbid conditions. Proteinuria is a common feature of many of the prevalent kidney diseases. In a study of HIV-infected women, those with significant proteinuria (dipstick >1+) before highly active antiretroviral therapy (HAART) had more than double the hazard of death after HAART initiation.2 Another found that those with >2+ protein had a 50% increased risk of hospitalization and 40% increased risk of cardiovascular conditions.3 Moreover, when compared with the general population, those with HIV infection have a greater prevalence of low-level proteinuria, with 11% having microalbuminuria.4
The Infectious Diseases Society of America (IDSA) guidelines on the management of chronic kidney disease in HIV recommend use of a urine dipstick, with a threshold for detection of renal disease at 1+ proteinuria.5 Evidence suggests that dipsticks may have poor validity, however.6-7 Because dipsticks measure albumin concentration, urine dilution greatly affects the reading. In addition, some forms of kidney disease are associated with nonalbumin (tubular) proteinuria.
Because early detection of proteinuria may be important for prevention of poor outcomes, the accuracy of screening tests plays a central role in the clinical care of those at high risk for kidney disease. The purpose of the current study is to compare urine dipstick assays with a quantitative test in HIV-infected patients undergoing testing for proteinuria.
We identified all HIV-positive patients in the Johns Hopkins Nephrology Clinic who underwent a urine dipstick measurement and a urine protein-to-creatinine (P/C) ratio within the same 48-hour period between January 1995 and November 2005. Quantitative measurements reported as grams of protein/grams of creatinine (gp/gc) were calculated from spot urine collections as an estimate of grams of protein over 24 hours. The spot P/C ratio is recommended by the IDSA guidelines in those with ≥1+ on dipstick measurement.5 Urinalyses before December 2001 were performed with the Clinitek (Bayer, Elkhart, IN) automated assay, and those after that date used the Atlas (Bayer) automated assay. The study was approved by the Johns Hopkins School of Medicine Institutional Review Board.
We evaluated the ranges of quantitative proteinuria corresponding to each dipstick result. The sensitivity of a dipstick result of ≥1+ to detect a P/C ratio of 0.30 to 0.99 gp/gc was calculated. A threshold of 1+ was chosen based on the IDSA guidelines.5 A quantitative threshold of 0.3 gp/gc was chosen because proteinuria >0.3 g over 24 hours is typically considered abnormal. Analyses were done with STATA 9.0 (Stata Corporation, College Station, TX).
A total of 365 matched urine dipsticks and P/C ratios were assessed in 166 patients. Participants were 59% male, 94% African American, and had a mean age of 45 years (range: 25 to 71 years). HIV-associated nephropathy was present in 14%, hepatitis C virus in 47%, and diabetes mellitus in 10%. Of the 365 paired specimens, 324 (88%) were obtained on the same day.
All paired dipstick and quantitative results are displayed in Table 1. As noted, 8 (12.5%) of 64 patients (95% confidence interval [CI]: 5.6% to 23.2%) with urine P/C ratios <0.3 gp/gc had positive urine dipstick results (false-positive result: ≥1+). Furthermore, 13 (21.0%) of 62 patients (95% CI: 11.7% to 31.2%) with abnormal low-level quantitative results (0.3 to 0.99 gp/gc) had negative urine dipstick results (false-negative result: <1+). The overall sensitivity of a ≥1+ dipstick result to detect abnormal proteinuria (≥0.30 gp/gc) was 94.0% (283 of 301 dipstick results, 95% CI: 90.7% to 96.5%). The diagnostic accuracy of various urine dipstick thresholds to detect significant low-grade (0.3 to 0.99 gp/gc) and abnormal (≥0.3 gp/gc) proteinuria is also summarized in Table 1.
The results of our study suggest that the urine dipstick may not be an adequate tool for screening proteinuria in HIV-positive patients. At clinically relevant lower levels of proteinuria (0.3 to 1.0 gp/gc), a threshold of ≥1+ proteinuria had a false-negative rate of 21.0%. Therefore, as many as 1 in 5 patients with proteinuria in this range may not be recognized and may be subject to delayed workup and treatment. Furthermore, we found that 12.5% of those with a 1+ dipstick result had <0.3 gp/gc. In sum, despite recommendations to use urine dipsticks for screening of renal disease in the HIV-positive population,5 the poor validity of qualitative testing prevents any single cutoff from serving as a reliable marker for predicting proteinuria at ranges where HIV providers may consider further workup of renal involvement.
Reports in hypertensive and diabetic renal disease suggest that urine dipstick tests have poor sensitivity in early disease.6,7 The results of the current study add to the growing body of evidence that questions the appropriateness of the urine dipstick as a screening test for patients in whom low-level proteinuria is a risk factor for poor outcomes.
An important weakness of the data is the use of different samples to compare tests. We limited the period between the assays to 2 days, however. Another limitation is that nephrology patients were studied, who are more likely to have higher levels of proteinuria (>1 gp/gc). This selection bias might diminish the proportion of patients with low-level proteinuria, with a corresponding bias toward higher sensitivity of >1+ dipstick results when assessing all levels of proteinuria >0.3 gp/gc. We attempted to minimize this bias by focusing on the 0.3- to 1.0-gp/gc range, however, which corresponds to clinically relevant early stages of disease and may better reflect levels in the HIV-infected population who undergo screening.
The early and accurate assessment of proteinuria is a critical aspect in management of kidney disease in HIV-infected patients. Because kidney disease may present with low levels of proteinuria, it is important to use a highly sensitive diagnostic test. Our results suggest that qualitative methods fail to attain adequate precision. Therefore, it may be prudent to revisit guidelines that recommend using the dipstick as a screening tool. Further investigations of proteinuria measurements, including comparison of random and 24-hour protein measurements, may help to determine the best method of screening and monitoring renal disease in HIV-infected patients.
The authors are grateful to Dr. Samir Gupta for his feedback during the preparation of this manuscript.
Mark J. Siedner, MD, MPH*
Mohamed G. Atta, MD, MPH*
Gregory M. Lucas, MD, PhD*
Mark A. Perazella, MD†
Derek M. Fine, MD*
*Johns Hopkins University School of Medicine Baltimore, MD
†Yale University School of Medicine New Haven, CT
1. Selik RM, Byers RH Jr, Dworkin MS. Trends in diseases reported on U.S. death certificates that mentioned HIV infection, 1987-1999. J Acquir Immune Defic Syndr
2. Szczech LA, Hoover DR, Feldman JG, et al. Association between renal disease and outcomes among HIV-infected women receiving or not receiving antiretroviral therapy. Clin Infect Dis
3. Gardner LI, Klein RS, Szczech LA, et al. Rates and risk factors for condition-specific hospitalizations in HIV-infected and uninfected women. J Acquir Immune Defic Syndr
4. Szczech LA, Grunfeld C, Scherzer R, et al. Microalbuminuria in HIV infection. AIDS
5. Gupta SK, Eustace JA, Winston JA, et al. Guidelines for the management of chronic kidney disease in HIV-infected patients: recommendations of the HIV Medicine Association of the Infectious Diseases Society of America. Clin Infect Dis
6. Garcia C, Bordier L, Burnat P, et al. Urinary dipsticks must not be used to detect diabetes-induced incipient nephropathy [in French]. Presse Med
© 2008 Lippincott Williams & Wilkins, Inc.
7. Zeller A, Sigle JP, Battegay E, et al. Value of a standard urinary dipstick test for detecting microalbuminuria in patients with newly diagnosed hypertension. Swiss Med Wkly