Original Articles

Test Dipstick for Determination of Urinary , and Protein/Creatinine Ratio

Jiunn-Min Wang1,2, Chi-Yu Lin1, Feng-An Tsai1, Jing-Yi Chen1, Yu-Chen Koa1

1Department of Medical Laboratory, Taichung Veterans General Hospital, Taichung 40705, Taiwan 2Department of Medical Technology, Central Taiwan University of Science and Technology, Taichung 40601, Taiwan

A slight increase in urinary level could be an advanced indicator of early stage nephropa- thy. The aim of this study was to detect the presence of albumin adjusted by creatinine using a dip- stick to test samples from 301 patients with nephropathy. The dipstick method can also be used to calculate the protein-to-creatinine ratio (P/C ratio) as a valuable index for eliminating false positive or false negative protein reports, which are often caused by abnormal sample concentration variation. The results showed the test data for the dipstick show good agreement with the quantita- tive measurements of protein, creatinine and P/C ratio. The P/C ratio obtained by the dipstick showed a sensitivity of 98% compared with the protein-only results of 80% in diagnosis of clinical proteinuria/. In conclusion, the dipstick is a reliable and practical screening tool for diagnosing the early stage of nephropathy. Key words: Creatinine, protein/creatinine ratio, Clinitek Atlas PRO 12 Reagent Pak

samples. One of the main factors in variation lies in the concentrated or diluted samples that lead to urine over- Introduction estimation or underestimation. In order to solve this problem, 24-hour samples can be used to reduce the It has recently been shown that even small increases in variation in the urine concentration [7]. However, in- urinary protein or albumin excretion are early predictors convenience is encountered and re-testing increases the of failure and end-stage renal disease(ESRD) cost. [1,2]. Early intervention has been demonstrated to retard Previous studies had revealed urinary creatinine can the development or rate of progression of renal disease. be used to correct the variation occurring in concentration Patients with and/or are the pri- of random urine samples because the daily excretion of mary risk groups [3-5]. Proteinuria is a major symptom creatinine is consistent [8-11]. The protein-to-creatinine of renal diseases and the excretive amounts of protein in ratio (P/C ratio) can be used to correct the false positive the urine could be an index for evaluating the pathologi- or false negative protein results caused by the variation cal stage of renal diseases [6]. Current urinary protein in urine concentration. Calculation of the P/C ratio may testing is performed by using a urine dipstick in routine be a more objective method for estimating urinary pro- urine examinations. The principle behind the reaction is tein excretion. However, most previous studies used the “protein error of indicators”. Under a fixed pH, indi- quantitative methods to calculate the P/C ratio. The cators will change color when protein exists, and the novel urine dipstick, Clinitek Atlas PRO12 Reagent Pak concentration of urinary protein can be calculated based (PRO12), has the added features of protein-low and on the difference of color. However, the test results of creatinine test pads. The protein-low test pad has im- urinary protein are often affected by variation in urine proved specificity to albumin and the sensitivity has

Received: December 9, 2008 Revised: March 9, 2009 Accepted: April 6, 2009 Address for correspondence: Jiunn-Min Wang, Department of Medical Laboratory, Taichung Veterans General Hospital, No. 160, Sec. 3, Chung-Kang Rd., Taichung 40705 Taiwan, R.O.C. TEL:+886423592525, FAX:+886423741267, E-mail: [email protected] J Biomed Lab Sci 2009 Vol 21 No 1 23

Test Dipstick for Urinary Protein, Creatinine and P/C Ratio been increased from 15mg/dL to 8mg/dL, which also is automatically calculated by the analyzer based on the leads to the display of “trace” results which are replaced light absorbent readings of the protein and creatinine test by the indication “15mg/dL”. Creatinine can also be pads. used to calculate the P/C ratio. Results were analyzed by computing the percentage In this study, we examined urine samples from the of same level agreement,＋/－one-level agreement, sen- outpatient department and evaluated the comparability of sitivity, specificity and Cohen’s Kappa coefficient. protein, creatinine, and P/C ratio data obtained from the PRO12 dipstick using quantitative chemical methods. The results showed that the novel dipstick can eliminate Results false negative and false positive reports and provide more accurate and objective data for clinical physicians The correlation of the protein test results between the in screening and diagnosis of renal disease. dipsticks of the Roche Urisys 2400 and the quantitative methods was shown in Table 1. The rate of complete coincidence was 77.7%. Rates of 77.42%, 94.92%, and Materials and Methods 87.71% were observed for the agreement rate with the cutoff point between negative and positive results being We collected 301 random urine samples from the outpa- “trace” for the dipsticks and “25 mg/dl” for the quantita- tient department of Taichung Veterans General Hospital, tive methods. Taichung, Taiwan. All samples were examined by two The correlation of the protein test results between models of automatic urine chemistry analyzers (Roche the PRO12 dipsticks and the quantitative methods was Urisys 2400 with the dipsticks,Urisys 2400 cassette and shown in Table 2. The rate of complete coincidence was Siemens/Bayer Clinitek Atlas with the dipsticks, Clinitek 69.44%. Rates of 93.88%, 78.57%, and 86.05% were Atlas PRO12 Reagent Pak). The following quantitative observed for the agreement rate with the cutoff point methods were also performed by a Hitachi chemistry between negative and positive results being “15 mg/dL” analyzer for each urinary sample: total protein concen- for the Pro12 test strips and “8 mg/dL” for the quantita- tration in urine using the total protein dye-binding assay tive methods. The sensitivity of the PRO12 dipsticks, but (pyrogallol red), urinary albumin concentration using the the specificity showed a minor decline. immuno-turbidimetric method, and urinary creatinine The correlation of the creatinine test results between concentration using Jaffe’s method. the PRO12 dipsticks and the quantitative methods was The Clinitek Atlas PRO12 Reagent Pak is enhanced shown in Table 3. The rate of complete coincidence was with a new protein-low pad which uses bis(3’,3”-diiodo- 52.16%. Agreement was 95.68%, if we extended the 4’,4”-dihydroxy-5’,5”-dinitrophenyl)-3,4,5,6-tetrabromo- comparison range to +/- one level. sulfonephthalein (DIDNTB) dye for the detection of Table 4 shows the correlation of the P/C ratio results albumin at > 8 mg/dL. Also present on the dipstick is an between the PRO12 dipsticks and the quantitative original pad (protein-high pad) that is based on the pro- methods. Only 286 samples were compared with the P/C tein error of indicator method (tetrabromophenol blue ratio results of the quantitative methods. (There were 39 (TBPB)) for detection of protein at > 30 mg/dl, and a over-diluted samples, which were re-collected.) The rate pad for detection of creatinine that uses the peroxidase of complete coincidence was 67.44%. Rates of 98.64%, activity of the copper-creatinine complex. The P/C ratio 76.98%, and 88.11% were observed for the agreement

Table 1. Correlation between the results of the quantitative method and the protein test portion of the Roche Urisys 2400 with the dipsticks. Total protein concentration in urine (mg/dL)

<25 25-49.9 50-112.4 112.5-325 >=325 Neg* 168 23 4 1 0 25* 9 29 9 2 0 75* 0 4 32 5 1 150* 0 0 3 3 2 500* 0 0 0 4 2 *: Roche Urisys 2400 with the dipsticks (mg/dL); Neg, negative result.

24 J Biomed Lab Sci 2009 Vol 21 No 1

Table 2. Correlation between the results of the quantitative methods and the protein test portion of the PRO12 dipsticks. Urinary albumin conc. (mg/dL) Urinary total protein conc. (mg/dL)

<8 8-29.9 30-64.9 65-199.9 ≥200 Neg* 121 9 0 0 0 15* 24 10 0 0 0 30* 9 18 40 1 0 100* 0 0 24 31 3 ≥300* 0 0 0 4 7 *: PRO12 test pad (mg/dL); Neg, negative result.

Table 3. Correlation between the results obtained from the quantitative methods and the creatinine test portion of the test pad. Urinary creatinine concentration (mg/dL)

<25 25-74.9 75-149.9 150-249.9 ≥250 Neg* 9 27 3 0 0 50* 1 39 30 1 0 100* 1 7 63 31 0 200* 0 1 12 38 18 ≥300* 1 0 6 5 8 *: PRO12 test pad, protein test for creatinine (mg/dL).

Table 4. Correlation of the P/C ratio between the results from the quantitative method and the PRO12 dipsticks. P:C(quantification) (mg/gCr)

Normal dilution <80 80-299.9 300-499.9 ≥500 Normal dilution* 0 14 1 0 0 Normal* 107 2 0 0 150* 26 28 4 0 300* 6 20 12 8 >500* 0 4 13 56 *: Roche Urisys 2400 with the dipsticks (mg/dL); Neg, negative result.

Table 5. Comparison of protein test results from the conventional dipsticks and the P/C ratio from the quantitative methods. Negative (<80) Positive (≥80) Negative * (-) 156 40 Positive * (≥25) 11 94 *: Roche Urisys 2400 with the dipsticks (mg/dL); Neg, negative result.

rate with the cutoff point between negative and positive methods. Rates of 70.15% and 93.41% were observed results being “150 mg/gCr” for the Pro12 dipsticks and for the agreement rate with the cutoff point between “80 mg/gCr” for the quantitative methods. negative and positive results being “> 25 mg/dL” for the Table 5 shows the correlation between the protein Roche test strips and “80 mg/gCr” for the quantitative test results obtained with the conventional dipsticks and methods. the P/C ratio results obtained with the quantitative

J Biomed Lab Sci 2009 Vol 21 No 1 25

Test Dipstick for Urinary Protein, Creatinine and P/C Ratio

all correlation showed that the new creatinine test pad is reliable. Discussion In comparison with the conventional dipsticks, the PRO12 dipsticks can provide both more sensitive protein In Taiwan, there are more than 46,000 dialysis patients and P/C ratio results for evaluating clinical proteinuria. and the number is increasing. Urine protein examination Furthermore, the P/C ratio can act as an indicator to de- is the first step in screening for renal diseases. However, tect over-diluted samples which have less than 10 mg/dL the urine dipstick currently used is semi-quantitative. of urinary creatinine. When negative protein results are Also, compared with the quantitative method the dip- found in over-diluted samples, the samples should be stick has sensitivity limitations (about 15-30 mg/dL). re-collected and re-examined to ensure accurate results. The urinary protein results can also be affected by varia- About 10% of routine urine screening samples were di- tions in sample concentrations. Thus, physicians cannot luted samples. That also means without the P/C ratio, the distinguish between pathological proteinuria and false diagnosis of proteinuria will be underestimated because positives caused by concentrated samples when the uri- of false negative results. In this study, we ruled out the nary protein report shows “trace”. Many studies have diluted samples and only compared the P/C ratio data of shown measurement of urinary creatinine could be a the other 286 samples between the PRO12 dipsticks and good way to correct the variation in sample concentra- the quantitative methods. tion. The protein-to-creatinine ratio (P/C ratio) can be has been defined as a urinary al- used to correct the false protein results caused by the bumin-to-creatinine ratio (A/C ratio) in the range of physiological differences in urine concentration. Calcu- 30~300 mg/gCr. Many studies have revealed that micro- lating the P/C ratio is a more objective method for as- albuminuria can be an early warning sign of renal dis- sessing urinary protein excretion. eases. The high-risk population includes hypertensive In this study, we evaluated the application of the and diabetic patients, so we suggest regular monitoring PRO12 dipstick [12]. The PRO12 is enhanced with a of urinary microalbumin by using the A/C ratio data. In protein-low test pad, which has higher specificity and this study, we used 80 mg/gCr as the cutoff point for affinity for detecting lower concentrations of albumin. comparisons with the quantitative results [13]. The re- The protein-low pad has improved specificity to urinary sults showed that the coincidence rate was 88.11%. The albumin and the sensitivity has been increased to 8~15 false negative rate was only 1.3%, which indicated the mg/dL. The original protein test pad on the PRO12 dip- P/C ratio could decrease the chance of false negatives. stick is named the “protein-high pad” and it detects the We also compared the results with those obtained with concentration of total urinary protein. The PRO12 test the Roche Urisys 2400 conventional dipsticks, the strips have an added feature, a creatinine test, that can original method we used for routine urine tests. The sen- detect urinary creatinine and it also acts as a corrective sitivity was 98.64% and specificity was 76.98%, which factor for urine concentration variations. We compared means 30% of early nephropathy cases could potentially both the conventional dipsticks and the PRO12 with the be misdiagnosed due to negative protein reports. The quantitative methods. The sensitivity of the PRO12 for data shows the P/C ratio of PRO12 could be an effective urinary protein tests was 93.88%, but the sensitivity of diagnostic tool for the monitoring of early stage renal the conventional dipsticks was only 77.42%. Further- diseases, especially for diabetes, hypertension and eld- more, the complete coincidence rate of PRO12 urinary erly health screening. protein tests reached 86.05%. Since conventional dipsticks have sensitivity and We also compared urinary creatinine results be- specificity limitations, the urinary protein results could tween the PRO12 dipsticks and the quantitative methods. be false positive or false negative. If the false results are The complete coincidence rate was 52.16%. By extend- not detected, the consequences can be serious due to ing the comparison range to one upper or one lower re- delays in diagnosis and treatment. However, it is costly port unit, the overall coincidence rate reached 95.68%. to retest false negative data. The results of this study The principle of the creatinine test in the PRO12 dip- show that the new urine dipsticks, which have the added sticks is based on the peroxidase-like activity of the features of the creatinine test and P/C ratio, increase both copper creatinine complex, and vitamin C or blood cells the sensitivity and specificity of the protein test. These in urine samples can interfere with the reaction. This improvements could effectively eliminate most false could be the reason why we did not get a high correla- negative results. We also found that using both the pro- tion using the quantitative methods. However, the over- tein and P/C ratio results to evaluate clinical proteinuria

26 J Biomed Lab Sci 2009 Vol 21 No 1

was better than using the protein-only results. If we use 3. Sarafidis PA, Lasaridis AN, Nilsson PM, et al: The effect the routine urine tests to identify patients with potential of rosiglitazone on urine albumin excretion in patients early stage renal diseases, treatment could lower the with type 2 diabetes mellitus and hypertension. Am J Hypertens 2005;18:227-34. chances of serious nephropathy. With the current preva- 4. Couchoud C, Pozet N, Labeeuw M, et al: Screening lence of renal diseases increasing, the PRO12 test strip early renal failure:cut-off values for creatinine as an could be used as a reliable and effective tool for early indicator of renal impairment. Kidney Int 1999;55:1878-84. screening. 5. Leung YY, Szeto CC, Tam LS, et al: Urine protein-to- According to recent data, the population with dia- creatine ratio in an untimed urine collection is a reliable betic nephropathy is expanding year by year and is also measure of proteinuria in . Rheumatol- the most rapidly growing disease population preceding ogy 2007;46(4):649-52. end-stage renal disease( ESRD). In Taiwan, the medical 6. Artunc FH, Fischer IU, Risler T, et al: Improved estima- tion of GFR by serum cystatin C in patients undergoing costs for dialysis, which is often required for end-stage cardiac catheterization. Int J Cardiol 2005;102(2):173-8. renal disease (ESRD) patients, have reached almost 28 7. Wann S and Kuo TS: The Correlation Between 24 - Hour billion New Taiwan Dollars (NT$) per year. The new - Urine Protein and Spot Urine Protein/Creatinine Ration dipstick is convenient, rapid, accurate, reliable and eco- and the Inconstancy of Daily Urine Protein and Creatinine nomical. It can effectively help prevent chronic renal Excretion. TZU-Chi Med J 1992;4:100-5. diseases by early detection and also save medical re- 8. Al RA, Baykal C, Karacay O, et al: Random Urine Pro- sources. tein-Creatinine Ratio to Predict Proteinuria in New-Onset In conclusion, we suggest that physicians use the Mild Hypertension in Late . Obstet Gynecol 2004;104:367-71. P/C ratio or A/C ratio to monitor the potential risks of 9. Waugh J, Bell SC, Kilby MD, et al: Urinary microalbu- renal diseases in outpatients, hypertensive, and diabetic min/creatinine ratios: in uncomplicated patients. The P/C ratio data, which is easily obtained by pregnancy. Clin Sci 2003;104(2):103-7. dipsticks, should be included in the routine urine data 10. Wallace JF, Pugia MJ, Lott JA, et al: Mutisite Evaluation and could improve the early diagnosis and monitoring of of a New Dipstick for Albumin, Protein, and Creatinine. J renal diseases. Decelerating the prevalence of end-stage Clin Lab Anal 2001;15:231-5. renal disease (ESRD) could effectively relieve the bur- 11. Pugia MJ, Wallace JF, Lott JA, et al: Albuminuria and Proteinuria in Hospitalized Patients as Measuued by den on the national health insurance program in Taiwan. Quantitative and Dispstick Methods. J Clin Lab Anal 2001;15:295-300. 12. Watanabe M, Funabiki K, Tsuge T, et al: Using Protein/ References Creatinine Ratios in Random Urine. J Clin Lab Anal 2005;19:160-6. 1. Birmingham DJ, Rovin BH, Shidham G, et al: Spot urine 13. Shihabi ZK, Sheehan M, Hinsdale M, et al: Clinical protein/creatinine ratios are unreliable estimates nephri- Evaluation of a New Test for Proteinuria on the Clinitek R tis flares. Kidney Int 2007;72(7):865-70. Atlas ○ System. AACC Poster No. 511, presented at 2. Ruggenenti P, Perna A, Mosconi L, et al:Urinary protein AACC/CSCC 2001 National Meting, Chicago, IL, August excretion rate is the best independent predictor of ESRF 2,2001. in non-duabetic proteinuric chronic nephropathies. Kid- ney Int 1998;53:1209-16.

J Biomed Lab Sci 2009 Vol 21 No 1 27

原Test Dipstick 著 for Urinary Protein, Creatinine and P/C Ratio

新式尿液常規試紙檢測尿蛋白、尿肌酸酐及尿蛋白/肌酸酐比值之 可用性與臨床檢驗價值研究

王俊民 1,2 林啟畬 1 蔡奉安 1 陳靜宜 1 高玉真 1

1 台中榮民總醫院醫學檢驗部 2 中台科技大學

許多文獻顯示，尿液中少量白蛋白的增加可以用來評估早期腎病變及阻止糖尿病或高血壓的高危 險族群進展成末期腎病變。然而因目前使用之常規尿液試紙敏感度限制，特別是當尿蛋白濃度介 於+/-(trace)的時候，容易因為尿液濃度之稀釋或是濃縮的情況而造成錯誤的報告。檢驗室因此需 要以其他檢測方式進行確認。本實驗中，我們對門診 301位病人之隨機尿液檢體，測試了新的尿液 檢驗試紙，新增了白蛋白 (albumin)和肌酐酸 (creatinine)項目，使原來尿蛋白檢查的檢驗結果加上 新的albumin pad得到較精準的 protein結果；測試Creatinine的結果則以排除尿液濃度造成的錯誤 判斷並提供計算 protein /creatinine ratio，藉此可減少尿液檢測偽陽或 僞陰性發生，得到較精準之 P/C Ratio結果參考。我們比較試紙法所得到的尿蛋白及尿肌酐酸及 P/C ratio與定量法之間的相關 性，發現兩者間具有很高的相關性，尤其在 P/C ratio的項目中，試紙法與定量法比較，具有 98%以 上的敏感性，相較於傳統僅以尿蛋白結果作判讀的敏感度約 80%，有大幅上升。研究顯示，新試 紙為更可信的早期診斷初期腎臟病變篩檢疾病之工具。

關鍵詞：肌酸酐、尿蛋白 /肌酸酐比值、尿液試紙

收稿日期：97 年 4 月 1 日 修稿日期：97 年 10 月 21 日 接受日期：98 年 3 月 25 日 通訊作者：王俊民 台中榮民總醫院醫學檢驗部 40705 台中市中港路 3 段 160 號 電話：+886 4 23592525 傳真：+886 4 23741267 電子郵件：[email protected]

28 J Biomed Lab Sci 2009 Vol 21 No 1