Utility of the Dipstick Micraltest II^TM in the screening of microalbuminuria of diabetes mellitus type 2 and essential hypertension

Utilidad de la tira reactiva Micraltest II^TMen el escrutinio de microalbuminuria en diabetes mellitus tipo 2 e hipertensión esencial

Laura Cortés–Sanabria,* Héctor R. Martínez–Ramírez,** José L. Hernández,* Enrique Rojas–Campos,*** José L. Canales–Muñoz,*** Alfonso M. Cueto–Manzano***

* Coordinación de Salud Pública, Delegación Jalisco.

** Departamento de Educación e Investigación en Salud, UMF No. 34.

Correspondence and reprint request:
Laura Cortés–Sanabria, M.D.
Coordinación de Salud Pública Delegación Jalisco, IMSS.
Av. Circunvalación No. 553, Col. Independencia.
Phone: 52 (33) 3609–4406 Fax: 52 (33) 3133–5090
E–mail: laura.cortes@imss.gob.mx

Recibido el 3 de mayo de 2005. ]]> Aceptado el 1 de febrero de 2006.

ABSTRACT

Background. In Mexico, diabetes mellitus type 2 and hypertension are leading causes of end–stage renal disease. Diagnosis of early renal damage with detection of microalbuminuria (microAlbU) is fundamental for treatment and prevention, and so avoiding the catastrophes of renal failure. For screening purposes, several simplified tests, including dipstick methods, fulfill the accuracy requirements for microAlbU detection compared with gold standards; however, no study has established the reliability of such tests in our setting.

Aim. To evaluate the utility of micraltest II^TM as a screening test for microAlbU compared with nephelometry in patients with diabetes mellitus type 2 and non–diabetic patients with essential hypertension.

Patients and methods. Patients with diabetes mellitus type 2 as well as patients with essential hypertension of any age, sex and time of evolution, attending to three primary health–care units (UMF No. 3, 92 and 93, Guadalajara, Jalisco) were included. Patients with transitory albuminuria, secondary hypertension and serum creatinine > 2 mg/dL were excluded. Micraltest II^TM was performed in the first morning urine sample, and nephelometry was performed in a 24–h urine collection. Diagnostic accuracy of the dipstick test was then determined.

Results. 245 patients were studied: 71 (29%) were diabetics without hypertension, 95 (39%) were diabetics with hypertension, and 79 (32%) had only essential hypertension. In diabetic patients, micraltest II^TM sensitivity was 83%, specificity 96%, and positive and negative predictive values were 95% and 88%, respectively. Correlation between nephelometry and micraltest II^TM results was 0.81 (p < 0.001). The best cut–off point for microAlbU was 30.5 mg/L, and area under the curve (± SEM) was 0.91 ± 0.03 (confidence interval 95%: 0.85–0.96). In non–diabetic patients with essential hypertension, micraltest II^TM sensitivity was 75%, specificity 95%, and positive and negative predictive values were 43% and 99%, respectively. Correlation between nephelometry and micraltest II^TM results was 0.43 (p < 0.001). The best cut–off point for microAlbU was 28.2 mg/L, and area under the curve was 0.85 ± 0.13 (0.60–1.10).

Conclusion. Micraltest II^TM dispstick is a rapid, valid and reliable method for albuminuria screening in patients with diabetes mellitus type 2 and in those non–diabetic patients with essential hypertension in our setting.

Key words. Microalbuminuria. Micraltest^TM. Nephelometry. Diabetes mellitus 2. Essential Hypertension. Early nephropathy.

Antecedentes. En México, la diabetes mellitus tipo 2 y la hipertensión son las principales causas de insuficiencia renal crónica terminal. El diagnóstico temprano con detección de microalbuminuria (microAlbU) es fundamental para el tratamiento y prevención, y así evitar las catástrofes de la falla renal. Con el fin de tamizaje, varias pruebas simples, incluyendo las tiras reactivas, cumplen con los requerimientos de exactitud para detección de microAlbU comparados con esténdares de oro; sin embargo, ningún estudio ha establecido la confiabilidad de dichos métodos en nuestro medio.

Objetivo. Evaluar la utilidad del micraltest II^TM como prueba de tamizaje para microAlbU comparada con nefelometría en pacientes con diabetes mellitus tipo 2 y pacientes no diabáticos con hipertensión arterial esencial.

Pacientes y métodos. Se incluyeron pacientes con diabetes mellitus tipo 2, así como pacientes con hipertensión arterial esencial de cualquiera de los dos sexos, sexo y tiempo de evolución que atendían a tres unidades de Medicina Familiar (UMF No. 3, 92 y 93, Guadalajara, Jalisco). Se excluyeron pacientes con albuminuria transitoria, hipertensión secundaria y creatinina sárica > 2 mg/dL. El micraltest II^TM se realizó en la primera muestra matutina de orina, y la nefelometría en recolecciones de orina de 24 horas. La exactitud diagnóstica de la tira reactiva fue luego determinada.

Resultados. Doscientos cuarenta y cinco pacientes fueron estudiados: 71 (29%) eran diabáticos sin hipertensión, 95 (39%) eran diabáticos con hipertensión, y 79 (32%) tenían sólo hipertensión arterial esencial. En los pacientes diabáticos, el micraltest II^TM tuvo una sensibilidad de 83%, especificidad de 96%, y valores predictivos positivo y negativo de 95% y 88%, respectivamente. La correlación entre la nefelometría y el micraltest II^TM fue 0.81 (p < 0.001). El mejor punto de corte para la detección de microAlbU fue 30.5 mg/L, y el área bajo la curva (± EE) fue 0.91 ± 0.03 (intervalo de confianza 95%: 0.85–0.96). En los pacientes no diabáticos con hipertensión esencial, el micraltest II^TM tuvo una sensibilidad de 75%, especificidad de 95%, y valores predictivos positivo y negativo de 43 y 99%, respectivamente. La correlación entre los resultados de nefelometría y micraltest II^TM fue 0.43 (p < 0.001). El mejor punto de corte para microAlbU fue 28.2 mg/L, y el área bajo la curva fue 0.85 ± 0.13 (intervalo de confianza 95%:0.60–1.10).

Conclusión. La tira reactiva micraltest II^TM es un método rápido, válido y confiable para el tamizaje de albuminuria en pacientes con diabetes mellitus tipo 2 y pacientes no diabáticos con hipertensión arterial esencial en nuestro medio.

Palabras clave. Microalbuminuria. Micraltest^TM. Nefelometría. Diabetes Mellitus 2. Hipertensión arterial esencial. Nefropatía temprana.

INTRODUCTION

Albumin excretion rate (AER) has been widely advocated as a means for diagnosis of early renal damage in diabetes and essential hypertension.^1,2 Microalbuminuria (microAlbU), defined as an AER of 20–200 mg/minute, is associated with development of clinical proteinuria, chronic renal failure, and premature cardiovascular mortality in patients with diabetes mellitus type 1 and type 2.^3,4 Annual microAlbU screening in diabetic patients is currently recommended by several national and international organizations.^5–7 In non–diabetic subjects with essential hypertension, elevated AER have been found to be a strong predictor for cardiovascular disease and nephropathy.⁸

In Mexican population, diabetes mellitus type 2 and hypertension are highly prevalent^12,13 and, as in many parts of the world, they are leading causes of end–stage renal disease.^14,15 Detection of early stages of renal disease is fundamental for establishing treatment and prevention measures in order to avoid the clinical and economical catastrophes of end–stage renal disease. Several studies about prevalence of microAlbU in diabetic and hypertensive patients have been performed in our country;¹⁶"¹⁸ most of them were based on the determination of microAlbU using dipstick tests. However, to the best of our knowledge, no study has established the reliability of dipstick methods compared with gold standards for AER in our setting. Therefore, the aim of this study was to evaluate the utility of micraltest II^TM as a screening test for microAlbU compared with nephelometry in patients with diabetes mellitus type 2 and essential hypertension.

PATIENTS AND METHODS

This is an observational, cross–sectional, and comparative study on a diagnostic test. Patients with previous diagnosis of diabetes mellitus type 2 (according with the American Diabetes Association criteria¹⁹) with or without hypertension, as well as patients with essential hypertension without diabetes mellitus (according with the Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure²⁰), of any age, sex and time since diagnosis were included. Patients were randomly selected from the total diabetes mellitus type 2 and essential hypertension population attending to three primary health–care units (Unidades de Medicina Familiar No. 3, 92 and 93) in the metropolitan area of the city of Guadalajara, Jalisco. Patients were excluded if they had cardiac failure or renal tract disease, acute febrile illnesses, urinary tract infection, hematuria (women were not examined during menstruation), abnormal urinary sediment, any level of proteinuria in urinaly–sis, and serum creatinine > 2 mg/dL.

Determination of microAlbU

Before determination of AER, recommendation was made about adequate fluid intake and avoiding excessive physical exercise three days previous to urine sampling. Results obtained by micraltest II^TM dipstick (Roche Diagnostics GmbH, Germany) were compared with nephelometry (Behring Nephelometer Analyzer II, Behring Diagnostics GmbH, Germany) which was considered as the gold standard in this study. Patients were instructed to collect a 24–h urine sample, but they were asked to collect separately the last voiding sample (first sample of the next morning). In this first voiding urine sample, a micraltest II^TM dipstick was submerged during five seconds, and a reading was performed one minute later by only one investigator (LCS). The result of this assay was read in a semiquantitative manner as 0 mg/L, 20 mg/L, 50 mg/L or 100 mg/L. A reading of 0 mg/L was interpreted as negative, and a reading of 20 mg/L or more was considered positive (microAlbU is defined as 20–200 mg/L by the micraltest II^TM). After the micraltest II^TM reading, the first voiding urine sample was mixed with the rest of the 24–h urine collection for the nephelometry measurement, which had a detection inferior limit of 2.3 µ/mL and a coefficient of variation of 2.1%. Normal AER by this method is defined as < 20 µg/ min, microAlbU is 20–200 µg/min, and macroalbuminuria is > 200 µg/min. To compare nephelometry results with those of micraltest II^TM, the latter values were converted from µg/min to µg/L. All the nephelometry measurements were performed in the Central Laboratory of the Hospital de Especialidades, CMNO, IMSS, by only one person blinded to the micraltest II^TM results.

Statistical analysis

RESULTS

Two hundred forty–five patients attending to primary health–care units were studied; 71 (29%) had diabetes mellitus type 2 without hypertension, 95 (39%) had diabetes mellitus type 2 with hypertension, 79 (32%) had only essential hypertension (without diabetes). Demographic and clinical data of patients are shown in table 1.

From the whole sample of patients, 162 (69%) had no albuminuria, 51 (21%) had microAlbU, and 25 (10%) had macroalbuminuria; median urinary AER was 7.3 (3.6–31.8) µg/min.

Results in patients with diabetes mellitus type 2

Sensitivity, specificity, positive and negative predictive values, and exactitude of micraltest II^TM, and prevalence of albuminuria in diabetic patients are shown in table 2. Validity results of the micraltest II^TM according with the time of evolution of diabetes are shown in table 3; sensitivity and positive predictive value seemed to improve with longer duration of diabetes (which increases the prevalence of albuminuria), however, specificity and exactitude of the test were roughly the same throughout this time. Correlation between nephelometry and micraltest II^TM measurements was 0.81 (p < 0.0001). Comparison of urinary album concentration determined by micraltest II^TM and nephelometry is shown in table 4.

Figure 1 shows the ROC curve results for nephelometry and micraltest II^TM for albuminuria in diabetic patients. When the performance of micraltest II^TM as a diagnostic test for microAlbU was analyzed, the calculated mean area under the ROC curve (± SEM) was 0.91 ± 0.03 (confidence interval 95%: 0.85–0.96), and the corresponding best cut–off point value was 30.5 mg/L.

Results in patients with essential hypertension

Sensitivity, specificity, positive and negative predictive values, and exactitude of micraltest II^TM, and prevalence of albuminuria in patients with essential hypertension are shown in table 5. Validity results of the micraltest II^TM according with the time of evolution of hypertension are shown in table 6; sensitivity, specificity, positive predictive value and exactitude seemed to improve with longer duration of hypertension. No calculations could be performed with a duration of this disease > 10 years as all the 23 cases present with this evolution of hypertension were true negative cases (no cases of albuminuria). Essential hypertensive patients with longer duration of hypertension displayed a non–significant trend to use more antihypertensive drugs (< 5 years: 1.41 ± 0.63; 6–10 years: 1.70 ± 0.77; >10 years: 1.78 ± 0.74), and to have lower diastolic blood pressure (<5 years: 81 ± 10 mm Hg; 6–10 years: 82 ± 8 mm Hg; >10 years: 78 ± 10 mm Hg) than the others. Angiotensin converting enzyme inhibitors and an–giotensin receptor blockers were not different between patients with different duration of hypertension. Correlation between nephelometry and micraltest II^TM measurements was 0.43 (p < 0.0001). Comparison of urinary album concentration determined by micraltest II^TM and nephelometry is shown in table 7. Figure 2 shows the ROC curve results for nephelometry and micraltest II^TM for albuminuria in patients with essential hypertension. When the performance of micraltest II^TM as a diagnostic test for microAlbU was analyzed, the calculated mean area under the ROC curve was 0.85 ± 0.13 (0.60–1.10), and the corresponding best cut–off point value was 28.2 mg/L.

DISCUSSION

Our results show that micraltest II^TM dipstick is a valid and reliable test for albuminuria (microAlbU and macroalbuminuria) detection in patients with diabetes mellitus type 2 and essential hypertension in our setting.

Office test for detecting abnormal albuminuria should be simple in use, robust, quick, and inexpensive. Several office tests seem to fulfill the requirements of adequate sensitivity, specificity and reproducibility;^21–23 however, it is always useful to validate new tests when they are to be used in populations different from those originally probed. Some authors have compared the micraltest II^TM dipstick with gold standards as radioimmunoassay or immunoturbidimetry, obtaining in general, similar results in both diabetics and in hypertensive non–diabetic patients.^{10,11,21–23} Fewer data have been reported with the mployment of nephelometry as the gold standard, in which micraltest II^TM has shown a sensitivity of 84–97%, specificity of 72–98%, positive predictive value of 77–84%, and negative predictive value of 93–94%.^24–27 The previous results are in general comparable with ours, although Poulsen, et al.,²⁶ displayed a trend to a higher sensitivity and lower specificity. As expected, prevalence of albuminuria was higher in diabetics than in those non–diabetic patients with essential hypertension in the present study; such a finding influenced on the better negative predictive value and the poorer positive predictive value of micraltest II^TM observed in patients with essential hypertension compared with diabetics. In our hands, the best cut–off point value was 30.5 mg/L in diabetics and 28.2 mg/L in patients with essential hypertension. These latter findings may explain why the highest proportion of false positives with micraltest II^TM in both patient subpopulations was observed at the 20 mg/L level of micraltest II^TM; in other words, it could be possible that a higher proportion of albuminuria cases were precisely diagnosed if the dipstick test cut–off value was 30 mg/L instead of 20 mg/L.

On the other hand, to the best of our knowledge, no study using test–strip for microAlbU in Mexico or Latin America had been previously published. Calzada–León, et al.,²⁸ reported results of the colorimetric bromphenolblue test compared with radioimmunoassay in 82 Mexican children; these authors found a sensitivity of 60%, specificity of 83%, positive predictive value of 33%, and negative predictive value of 94%. The higher precision of the test evaluated in the present study compared with that of Calzada–León, etal.,²⁸ confirms the advantage of the antibody–based screening methods compared to older colorimetry methods based on bromphenol dye.²⁹ MicroAlbU may predict later development of nephropathy in diabetics,³⁰ and is associated with other complications, as retinopathy, cardiovascular risk factors and coronary heart disease both in diabetics and non–diabetic patients.^31–33 On the other hand, renoprotection (i.e. general, dietary and pharmacological interventions to halt or at least to slow the progression of renal function decline) is now the cornerstone of the management of patients with renal disease, and it is mostly effective when is implemented at early stages.^5,6,34,35 In many parts of the world and our country,^14,15 diabetes mellitus type 2 and hypertension are the leading causes of end–stage renal disease. Thus, it is crucial to detect early nephropathy in these high risk populations. Results of the current study demonstrate a high proportion of diabetic and hypertensive patients with early nephropathy in our setting, which is in agreement with previous reports.^16–18 The latter urges to the implementation of measures of secondary prevention and treatment to avoid further progression of the kidney function deterioration in this kind of patients. This in turn, may be useful to halt the increasing frequency of end–stage renal disease, which constitutes one of the greatest socioeconomic and public health problems in our country.

In conclusion, results of the present study show that micraltest II^TM dispstick is a rapid, valid and reliable method for the albuminuria screening in patients with diabetes mellitus type 2 and in those non–diabetic patients with essential hypertension in our setting. Routine dipstick screening for microalbuminuria might be recommended in patients at risk for renal disease attending to primary health–care units.

REFERENCES

1. Mogensen CE. Microalbuminuria predicts clinical proteinuria and early mortality in maturity–onset diabetes. N Engl J Med 1984; 310: 356–60.        [ Links ]

2. Ruilope LM, Rodicio JL. The kidney and arterial hypertension. Nephrol Dial Transplant 2001; 16: 50–2.        [ Links ]

3. Messent JW, Elliot TG, Hill RG, Jarrett RJ, Keen H, Viberti GC. Prognostic significance of microalbuminuria in insulin–dependent diabetes mellitus: a twenty–three–year follow–up study. Kidney Int 1992; 41: 836–9.        [ Links ]

4. Mogensen CE. Natural history of cardiovascular and renal disease in patients with type 2 diabetes: effect of therapeutic interventions and risk modification. Am J Cardiol 1998; 82: 4R–7R.        [ Links ]

5. The National Kidney Foundation –K/DOQI guidelines, http://www.kidney.org        [ Links ]

6. Rossert JA, Wauters JP. Recommendations for the screening and management of patients with chronic kidney disease. Nephrol Dial Transplant 2002; 17(1): 19–28.        [ Links ]

7. Keane WF, Eknoyan G. Proteinuria, albuminuria, risk, assessment, detection, elimination (PARADE): A position paper of National Kidney Foundation. Am J Kidney Dis 1999; 33: 1004–10.        [ Links ]

8. Yudkin JS, Forrest RD, Jackson CA. Microalbuminuria as predictor of vascular disease in non–diabetic subjects. Lancet 1998; 2: 530–3.        [ Links ]

9. Ismail N, Becker B, Strzelczyk P, Ritz E. Renal disease and hypertension in non insulin–dependent diabetes mellitus. Kidney Int 1999; 55: 1–28.        [ Links ]

10. Kutter D, Thoma J, Kremer A, Hansen S, Carl R. Screening for oligoalbuminuria by micral–test II a new inmunological test strip. Eur J Clin Chem Clin Biochem 1995; 33: 243–5.        [ Links ]

11. Tiu SC, Lee SS, Cheng MW. Comparison of six commercial techniques in the measurement of microalbuminuria in diabetic patients. Diabetes Care 1993; 16: 616–20.        [ Links ]

12. Velazquez–Monroy O, Rosas Peralta M, Lara Esqueda A, et al. Prevalence and interrelations of noncommunicable chronic diseases and cardiovascular risk factors in México. Arch Cardiol Mex 2003; 73: 62–77.        [ Links ]

13. Velazquez–Monroy O, Rosas Peralta M, Lara Esqueda A, et al. Hipertensión arterial en México: Resultados de la Encuesta Nacional de Salud (ENSA) 2000. Arch Cardiol Mex 2002; 72: 71–84.        [ Links ]

14. United States Renal Data System: http://www.usrds.org        [ Links ]

15. Breien AH, García BH, García GG, Gómez NB, Hernández RI, Lomelí AM, Monteón RF, Palomeque M, Ruiz MN. Epidemiología de la Insuficiencia Renal Crónica en Jalisco. Boletín del Colegio Jalisciense de Nefrología, A.C. 2001; 5: 6–7.        [ Links ]

16. Martínez–Ramírez HR, Jalomo–Martínez B, Cortés–Sanabria L, Rojas–Campos E, Barragán G, Alfaro G, Cueto Manzano AM. Renal function preservation in type 2 diabetes mellitus patients with early nephropathy: A comparative prospective cohort study between primary health–care doctors and nephrologist. Am J Kidney Dis 2006 (in press).        [ Links ]

17. Cueto–Manzano AM, Cortés–Sanabria L, Martínez–Ramírez HR, Rojas–Campos E, Barragán G, Alfaro G, Flores J, et al. Detection of early nephropathy in Mexican type 2 diabetes mellitus patients. Kidney Int 2005; 68 (Suppl. 97): S40–S45.        [ Links ]

18. González VC, Stern MP, Arredondo PB, Martínez DS, Islas AS, Revilla C, González VM, Rivera MD. Nephropathy in low income diabetics: The Mexico City diabetes study. Arch Med Res 1996; 27: 367–73.        [ Links ]

19. American Diabetes Association: Clinical Practice Recommendations 2001. Diabetes Care 2001; 24(1): S69–S72.        [ Links ]

20. Chobanian AV, Barkis HL, Cushman WC, Green LA, Izzo JL, Jones DW, et al. Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension 2003; 42: 1206–52.        [ Links ]

21. Pegorado A, Singh A, Bakir A, Arruda JAL, Dunea G. Simplified screening for microalbuminuria. Ann Intern Med 1997; 127: 817–19.        [ Links ]

22. Gerber LM, Johnston K, Alderman HM. Assessment of a new dipstick test in screening for microalbuminuria in patients with hypertension. Am J Hypertension 1998; 11: 1321–7.        [ Links ]

23. Gilbert R, Akdeniz A, Jerums G. Detection of microalbuminuria in diabetic patients by urinary dipstick. Diabetes Res Clin Practice 1997; 35: 57–60.        [ Links ]

24. Minetti EE, Cozzi MG, Granata S, Guidi E. Accuracy of the urinary albumin tritator stick "Micral–Test" in kidney disease patients. Nephrol Dial Transplant 1997; 12: 78–80.        [ Links ]

25. Fernández F, Pérez PJM, Hermosín BT, Vazquez GP, Ortiz CMA, Tarilonte DMA. Rapid screening test evaluation for microalbuminuria in diabetes mellitus. Acta Diabetol 1998; 35: 199–202.        [ Links ]

26. Pouslen PL, Mogensen CE. Evaluation of a new semiquantitative stick for microalbuminuria. Diabetes Care 1995: 18: 732–3.        [ Links ]

27. Mogensen CE, Viberti GC, Peheim E, Kutter D, Hasslacher C, Hofmann W, et al. Multicenter evaluation of the micraltest II test strip, an inmunologic rapid test for the detection of microalbuminuria. Diabetes Care 1997; 20: 1642–6.        [ Links ]

28. Calzada–León R, Altamirano–Bustamante N, Robles–Valdés C, Franco–Rodríguez A, Franco–Betancur H, Jimánez C, et al. Sensibilidad y especificidad de la determinación cuantitativa de microalbuminuria para el diagnóstico de nefropatía diabática. Bol Med Hosp Infant Mex 1994; 51: 174–8.        [ Links ]

29. Poulsen PL. Office tests for microalbuminuria. In: Mogensen CE (Ed.). The kidney and hypertension in diabetes mellitus. Third edition. Kluwer Academic Publishers, Norwell; 1997, pp. 127–34.        [ Links ]

30. Mogensen CE. Microalbuminuria as a predictor of clinical diabetic nephropathy. Kidney Int 1987; 31: 673–89.        [ Links ]

31. Keane WF. Proteinuria: Its clinical importance and role in progressive renal disease. Am J Kidney Dis 2000; 35(1): S97–S105.        [ Links ]

32. Pedrinelli R. Microalbuminuria in essential hypertension. A marker of systemic vascular damage. Nephrol Dial Transplant 1997; 12: 379–98.        [ Links ]

33. Weinstock BW, Keane WF. Proteinuria and cardiovascular disease. Am J Kidney Dis 2001; 38 (1): S8–S13.        [ Links ]

34. Hebert LA, Wilmer WA, Falkenhain ME, Ladson–Wofford SE, Nahman NS, Rovin BH. Renoprotection: One or many therapies? Kidney Int 2001; 59: 1211–26.        [ Links ]

35. Jungers P. Late referral: loss of chance for the patient, loss of money for society. Nephrol Dial Transplant 2002; 17: 371–5.        [ Links ] ]]> 1984 310 310 356-60 2001 16 16 50-2 1992 41 41 836-9 1998 82 82 4R-7R The National Kidney Foundation 2002 17 1 1 19-28 1999 33 33 1004-10 1998 2 2 530-3 1999 55 55 1-28 1995 33 33 243-5 1993 16 16 616-20 2003 73 73 62-77 2002 72 72 71-84 United States Renal Data System 2001 5 5 6-7 2006 2005 68^s97 68^s97 97 S40-S45 1996 27 27 367-73 American Diabetes Association 2001 24 1 1 S69-S72 2003 42 42 1206-52 1997 127 127 817-19 1998 11 11 1321-7 1997 35 35 57-60 1997 12 12 78-80 1998 35 35 199-202 1995 18 18 732-3 1997 20 20 1642-6 1994 51 51 174-8 1997 Third edition 127-34 1987 31 31 673-89 2000 35 1 1 S97-S105 1997 12 12 379-98 2001 38 1 1 S8-S13 2002 17 17 371-5