Diabetic Retinopathy is A Poor Predictor of Type of Nephropathy in Proteinuric Type 2 Diabetic Patients

Original Article Diabetic Retinopathy is A Poor Predictor of Type of Nephropathy in Proteinuric Type 2 Diabetic Patients J Prakash*, M Lodha**, SK Singh***, Rubina Vohra+, R Raja**, Usha++ Abstract Background : Both diabetic nephropathy (DN) and nondiabetic nephropathy (NDN) are reported to occur in patients with type 2 diabetes mellitus (DM). The precise diagnosis of the type of nephropathy has obvious clinical and prognostic implication. The aim of the study was to evaluate the histologic spectrum of nephropathy in proteinuric type 2 diabetic patients and to find the correlation between type of nephropathy and diabetic retinopathy (DR). Methods : Twenty eight proteinuric type 2 diabetic patients were included in the study. Five patients (ADPKD 3 and chronic pyelonephritis 2) were excluded from biopsy. Percutaneous renal biopsy was carried in remaining 23 patients. Results : There was a preponderance of male (75%) and majority of the patients were in the age group of 30-78 years. Duration of diabetes ranged between 4 months to 25 years with mean ± SD of 10.53±7.62 years. The presenting features were nephrotic syndrome 14 (60.9%), non-nephrotic proteinuria 9 (39.1%) and impaired renal function in 19 (82.6%) patients. Renal biopsy in 23 cases revealed; isolated diabetic nephropathy 13 (56.2%), NDN 7 (13.43%) and 3 (13%) patients had NDN superimposed on diabetic nephropathy. Membranous nephropathy (2), focal segmental (2), mesangiocapillary glomerulonephritis (1) were the nondiabetic glomerular disease in our type 2 diabetic patients. Chronic pyelonephritis and ischemic interstial nephropathy was the predominant tubulointerstial lesion in this study. Diabetic retinopathy (DR) observed in 12 (75%) patients with biopsy proven DN and absent in 4 (25%) patients with DN. The distribution of renal lesions in patients with DR (n=15) showed DN in 9 (60%), NDN 3 (20%) and remaining 3 patients had combined lesions. Renal biopsy in 8 patients without DR showed typical DN in 4 (50%) and NDN in 4 (50%) patients. Conclusion : This study demonstrates presence of both glomerular and tubulointerstitial lesions unrelated to diabetes (NDN) in proteinuric type 2 diabetic patients. Further presence or absence of DR was a poor predictor of diabetic nephropathy because DN was noted in 50% of patients without DR and 40% of patients with DR had non-diabetic nephropathy either alone or in combination with DN. INTRODUCTION Diabetic nephropathy is one of the major microvascular complications of diabetes mellitus. It contributes significantly to the mortality and morbidity of diabetic patients. Diabetic nephropathy is the leading cause of the ESRD worldwide. But diabetic patients are not immune to the renal disease other than diabetic nephropathy. Instead, it may be possible that the abnormal diabetic kidney is more susceptible to different type of glomerulonephritis. Renal disease other *Professor of Nephrology; **Junior Resident in Medicine; ***Professor of Endocrinology; +Senior Resident in Nephrology; ++Professor of Pathology, Department of Nephrology; Institute of Medical Sciences, Banaras Hindu University, Varanasi - 221 005. Received : 18.8.2006; Revised : 17.4.2007; Accepted : 2.5.2007 than diabetic nephropathy (nondiabetic renal disease) can occur in type 2 diabetic patients. A wide spectrum of nondiabetic nephropathy (NDN) including both glomerular and tubulointerstitial lesions are reported in patients with type 2 DM and their precise diagnosis requires histological examination. The exact incidence of such NDN is not known. The reported incidence of NDN ranged from 23-54% in proteinuric type 2 DM patients. 1-3 Biopsy studies suggest that 25-50% of patient with type 2 diabetes has glomerular lesions unrelated to or in addition to diabetic nephropathy. 3-5 This wide variation in incidence of NDN may be related to selection bias and variability in biopsy criteria for the diagnosis of nondiabetic nephropathy. The aim and objective of the present study were to examine the spectrum of nephropathy in proteinuric type 2 diabetic patients on renal biopsy. Further this study was aimed 412 www.japi.org JAPI VOL. 55 JUNE 2007

to find how diabetic retinopathy which is a very strong predictor of DN in type 1 DM, correlates with type of nephropathy in type 2 DM patients. METHODS This prospective study was carried at department of Nephrology, University Hospital, Institute of Medical Science, Banaras Hindu University, Varanasi, INDIA from August 2003 to April 2005. Type 2DM was defined as per ADA 1987 guidelines for diagnosis and classification of diabetes mellitus. 6 Type 2 diabetic patients with persistent proteinuria of 500 mg per day with or without renal functional impairment were included in this study. Patients with gross or microscopic haematuria, active urinary sediments, rapid decline in renal function and with any contraindication for renal biopsy were excluded from the study. We selected 28 proteinuric type 2 diabetic patients. The duration of diabetes varies from 4 months to 25 years with mean ± SD of 10.53±7.62 years. Twenty four hour urinary protein excretion ranged between 0.52-10.4 gm with mean 3.42 ±2.87 gm. The serum creatinine was between 0.3-8.0mg/dl with mean of 3.09± 2.07 mg/dl. Primary renal diseases (ADPKD 3; chronic pyelonephritis 2) were noted in 5 patients and they were not included in the study. Remaining 23 patients underwent percutaneous renal biopsy after obtaining written consent. Diabetic retinopathy (DR) was assessed by formal ophthalmological examination in all patients by ophthalmologist. Urine analysis, 24 hour urinary excretion of protein, ultrasonography of kidneys, haematological and biochemical investigations were carried in all patients. Renal biopsy was performed using trucut biopsy needle. The biopsy material was processed and studied under light microscope using haematoxylin and eosin (H&E), periodic acid Schiff (PAS) and acid fuschsin organ G (AFOG) stains. Diabetic (DG) was diagnosed on the basis of mesangial expansion, with or without Kimmelstial-Wilson nodule formation. 7 Special attention was paid to presence of the following diabetic lesions: either of nodular or diffuse types, interstial fibrosis with infiltration by mononuclear inflammatory cells, hyalinization of the renal arteriolar, particularly efferent. Nondiabetic glomerular diseases was classified into well known categories depending on their standard characteristic finding on light microscope. RESULTS The main clinical characterstics of patients are shown in Table 1. Their (male 15; female 8) age ranged between 30-78 years. Nineteen patients had impaired renal function and edema was present in all cases. Diabetic retinopathy was seen in 15 and absent in 8 patients. Hypertension was absent in 12(52.25%) patients and only 11(47.8%) cases had systemic hypertension. Nonnephrotic range proteinuria was observed in 9(39%) patients and remaining 14 (60.9%) cases had nephrotic range proteinuria. Renal histopathology in 23 patients revealed: Diabetic nephropathy in 13 (56.52%) nondiabetic nephropathy 07 (30.4%) cases and 3 (13%) patients had NDN superimposed on diabetic nephropathy. Thus, diabetic nephropathy was noted in 16 patients.of the 16 patients with diabetic nephropathy 10(62.5%) had diffuse (DG), 3(18.75%) patients had nodular (Kimmelsteil-Wilson lesion nodule)and remaining 3(18.15%) had combined diffuse and nodular. Of the 10 patients with NDN, glomerular and tubulointerstitial lesions were found in equal number (5) of cases. Membranous nephropathy (2), focal segmental (2) and mesangiocapillary GN (1) were the nondiabetic glomerular disease in our type 2 diabetic patients. In nondiabetic tubulointerstitial lesions (n=5); 3(60%) cases chronic pyelonephritis and 2 (40%) patients showed evidence of ischemic interstitial nephritis (Table 2). Nine (60%) patients with diabetic retinopathy had isolated DN on renal biopsy. Nondiabetic nephropathy was noted in 3 (20%) patients and other 3 cases had NDN superimposed on DN in presence of diabetic retinopathy. Thus 40% of proteinuric type 2 diabetic patients had NDN even in presence of diabetic retinopathy. We also observed chance of getting diabetic and nondiabetic nephropathy is equal in patients without diabetic retinopathy. Taken together it was observed that lesions of DN can occur in proteinuric type 2 diabetic patients even in absence of DR and patients with DR may have non- diabetic nephropathy on renal biopsy. DISCUSSION Type 2 DM is an emerging global epidemic and its incidence has increased dramatically in the last decade. Diabetic nephropathy (DR) occurs in about one third of patients with either type 1 or type 2 diabetes and adds enormously to the morbidity, mortality and cost of treatment. It is diagnosed clinically on the basis of persistent proteinuria (> 500 mg/d), hypertension and presence of concomitant DR in patient with long duration of DM. In addition, to classical diabetic kidney disease, patient with type 2 DM can also develop nephropathy unrelated to diabetes known as nondiabetic nephropathy (NDN). The early appearance of overt proteinuria (< 5 yrs duration of DM), rapid decline in renal functions, massive proteinuria with normal renal function, impaired renal function without significant proteinuria, or active urinary sediments are inconsistent with natural history of DN. 1 A wide spectrum of NDN can occur in patient with type 2 DM including; membranous glomerulonephritis, FSGS, MPGN, MCN, RPGN chronic tubulointerstial nephritis and amyloidosis. 3,4,8,9 Renal biopsy is required to confirm various types of histological lesions in proteinuric type 2 JAPI VOL. 55 JUNE 2007 www.japi.org 413

Table 1 : Clinical characteristic of proteinuric type 2 diabetic patients Sr. Age/sex Duration of Hypertension Diabetic Serum 24 hr urinary Renal biopsy No. diabetes retinopathy creatinine (mg %) protein 1. 60/M 4 yrs *NPDR 7.2 650 mg Diffuse 2. 58/M 18 yrs Absent 1.3 10.4 gm Membranous glomerulonephritis 3. 54/M 20 yrs + **PDR 3 6.8 gm Membranous glomerulonephritis with nodular sclerosis 4. 35/F 5 yrs + *NPDR 0.6 2 gm Diffuse diabetic 5. 45/M 3 yrs + Absent 1.5 6.5 gm Focal segmental 6. 65/M 15 yrs + **PDR 5.6 5.5 Nodular and diffuse with pyelonephritis 7. 62/M 4 yrs *NPDR 1.1 10 gm Diffuse and nodular 8. 48 /M 10 yrs *NPDR 5.3 800 mg Nodular sclerosis 9. 45/M 13 yrs Absent 8 2.3 gm Diffuse 10. 50/M 2 months + Absent 3.8 1.2 gm GBM thickening with interstitial fibrosis with history of diabetic nephropathy 11. 55/M 1 yr *NPDR 3.4 608 mg Diffuse diabetic with chronic pyelonephritis 12. 30/M 15 yrs + **PDR 1.8 4 gm Diffuse diabetic 13. 52/M 2 yrs *NPDR 2.5 5.6 gm Nodular sclerosis 14. 52/F 8 yrs Absent 1.4 2.25 gm Diffuse 15. 70/M 25 yrs + **PDR 3.1 3.87 gm Diffuse diabetic 16. 53/M 4 yrs + *NPDR 1.5 4.5 gm Diffuse mesangiocapillary GN with pyelonephritis 17. 57/M 6 yrs + *NPDR 5 5 gm Diffuse 18. 47/F 20 yrs + Absent 0.3 2.6 gm Diffuse and nodular sclerosis 19. 70/F 2 months Absent 2.4 3.8 gm Focal segmental 20. 44/F 4 yrs *NPDR 2.6 960 mg Chronic interstitial nephritis with ischemic changes 21. 56/F 7 yrs + *NPDR 2.6 5.2 gm Diffuse 22. 44/F 8 yrs Absent 2.8 640 mg Diffuse and nodular 23. 75/F 1 month Absent Absent 1.6 6.2 gm Focal segmental *NPDR - Nonproliferative diabetic retinopathy. **PDR - Proliferative diabetic retinopathy. diabetic patients and precise diagnosis of these diseases has obvious prognostic and therapeutic implication. 8-10 The exact incidence of NDN is not known in type 2 diabetic patients.however, reported incidence of NDN ranged from 23-54.3%. 1-3 Primary renal disease was observed in 29 of 79 (24%) diabetic patients admitted for evaluation of renal failure in a recent study. 11 Biopsy studies suggests that 25-50% of patients with type 2 diabetes have a glomerular lesions unrelated to or in addition to diabetic nephropathy. 3,4,12 Histopathological examination of renal tissue revealed a wide spectrum of renal lesion including both diabetic and nondiabetic nephropathy in present study. We observed DN in 16 (69.57%) patients (isolated DN 13(56.52%) and 3 (13.05%) patients had NDN superimposed on DN). Nodular was noted in 6 (37.5%) patients of biopsy proven diabetic nephropathy and in 3/6 patients nodular was associated with diffuse. In one patient nodular sclerotic lesion was coexistent with membranous glomerulopathy. This patient had diabetes for 20 years and presented with feature of nephrotic syndrome with massive proteinuria (> 5 gm/day) and nonproliferative diabetic retinopathy. Diffuse is the most common renal lesion in DN and nearly 1/3 of patients have nodular. We noted diffuse in 10 (62.5%) and 6 (37.5%) cases had nodular. Thus our observation was similar to other studies. Non diabetic nephropathy (NDN) was observed in 10(43.7%) cases in our study. Seven patients had isolated non-diabetic lesions and 3 cases had NDN superimposed on diabetic nephropathy. The spectrum 414 www.japi.org JAPI VOL. 55 JUNE 2007

Table 2 : Renal pathology in type 2 DM proteinuric patients Proteinuric type 2 DM patients (n=23) Diabetic nephropathy - 13 Nondiabetic nephropathy - 07 Nondiabetic nephropathy superimposed - 03 on diabetic nephropathy Presence of DR (n=15) Diabetic nephropathy - 09 Nondiabetic nephropathy - 03 Nondiabetic nephropathy superimposed - 03 on diabetic nephropathy Absence of DR (n=8) Diabetic nephropathy - 04 Nondiabetic nephropathy - 04 Nature of DN (n=16) Diffuse glomerulosclerosi - 10 Nodular - 03 Combined DG + Nodular lesion - 03 Nature of NDN (n=10) Membranous nephropathy - 02 Focal segmental GN - 02 Mesangiocapillary GN - 01 Chronic pyelonephritis - 03 Ischemic interstitial nephritis - 02 DN; diabetic nephropathy, NDN; nondiabetic nephropathy, DR; diabetic retinopathy, DG; diffuse, GN; glomerulopathy, DM; diabetes mellitus. of nondiabetic nephropathy including both glomerular and tubulointerstial lesions are described in diabetic patients in various studies. In our study 5 (50%) cases had glomerular lesion and 5 (50%) patients had tubulointerstial lesion. We observed membranous nephropathy (2), focal segmental (2) and mesangiocapillary glomerulonephritis in one patient. However, sarcoidosis, amyloidosis and lupus nephritis were not observed by us as a cause of proteinuria in our patients. Membranous nephropathy was reported in 20-31.25% of type 2 proteinuric diabetic. 13,14 Membranous GN was the most common non diabetic glomerular disease in type 2 proteinuric diabetic patients. 15 Isolated idiopathic membranous nephropathy without diabetic was reported in two proteinuric type 2 diabetic patients in a recent study. 16 In our study two patients were found to have membranous nephropathy, one had isolated membranous nephropathy and another had membranous nephropathy superimposed on diabetic nodular sclerosis. Focal segmental was found in 7% of patients in our previous study. 13 We found two patients with focal segmental in the present study. Both patients presented with nephrotic range proteinuria of 6.2-6.5 gm/day with relatively preserved renal function (serum creatinine 1.5-1.6 mg %). We noted only one patient with mesangiocapillary glomerulonephritis in our study, who was known diabetic for 4 years had nonproliferative diabetic retinopathy and presented with nephrotic syndrome (proteinuria = 4.5 gm/day) with relatively preserved renal function (serum creatinine = 1.5 mg/dl). We noted primary renal diseases (ADPKD3, chronic pyelonephritis 3 and ischemic nephropathy 2) were cause of chronic renal failure in our type2 diabetic patients. One study reported that 20% of type 2 diabetic patient had ischemic nephropathy secondary to atherosclerosis. 17 In a study of 142 patients with diabetic entering renal replacement therapy (RRT) reflex nephropathy in 13, glomerulonephritis in 11, polycystic kidney in four, analgesic nephropathy in three and renovascular occlusion in one case was noted. 18 Thus primary renal diseases are important cases of chronic renal failure in type 2 diabetic patients. In 8 proteinuric type 2 diabetic patients without diabetic retinopathy, biopsy showed typical diabetic nephropathy in 4(50%) and occurrence of non diabetic nephropathy in another 4(50%) patients. These patterns reflects chance of getting diabetic and non diabetic renal disease are nearly equal in type 2 diabetic patients in absence of diabetic retinopathy(dr). Gall MA et al also noted occurrence of diabetic and non diabetic renal disease in equal number of their type 2 diabetic patients who lacked retinopathy. 19 In our study, fifteen proteinuric type 2 diabetic patients with diabetic retinopathy underwent renal biopsy. Result of the biopsy revealed diabetic nephropathy in 9(60%) patients, isolated non- diabetic nephropathy in 3(20%) and combined diabetic and non diabetic nephropathy in remaining 3(20%) patients. In our previous publication 31% of patients with non-diabetic renal diseases had background diabetic retinopathy. 13 Several studies have shown that 90-95% of type 1 diabetics with diabetic nephropathy have retinopathy in contrast 50-75% of type 2 DM with nephropathy have retinopathy. Parving et al noted 41% of proteinuric type 2 DM patients with DN lacked diabetic retinopathy. 4 There was no evidence of diabetic retinopathy in one third of type 2 diabetic patients with proven nephropathy. 19 We also observed four patients with biopsy proven diabetic nephropathy did not have evidence of diabetic retinopathy. Biopsy study in 93 proteinuric patients with type 2 diabetes mellitus without retinopathy showed diabetic nephropathy in 69% cases and 31% had either nondiabetic glomerulopathy (13%) or normal glomerular structure (18%). 20 In another recent study renal biopsy in 76 type 2 diabetic patients without retinopathy revealed typical diabetic nephropathy in 17(22%), combined diabetic and nondiabetic glomerulopathy in 21(28%) and 37(49%) patients had only nondiabetic glomerulopathy. 21 Thus, 20-40% of type 2 diabetic patients with biopsy proven diabetic nephropathy did not have evidence of diabetic retinopathy. The renal retinal relationship in type 2 diabetic patients may not be helpful for clinical diagnosis of diabetic nephropathy and lack of retinopathy is a poor predictor of nondiabetic kidney disease in type 2 diabetic patients. Therefore presence or absence of retinopathy did not prove to be significant enough in distinguishing diabetic JAPI VOL. 55 JUNE 2007 www.japi.org 415

and NDN in type 2 diabetic patients. Diabetic retinopathy in proteinuric type 2 diabetic patients may favour diabetic nephropathy but does not exclude non-diabetic renal disease. The present work demonstrates that presence of diabetic retinopathy alone does not favour the diagnosis of diabetic nephropathy in proteinuric type 2 diabetic patients. The renal biopsy is necessary for precise diagnosis of diabetic and non diabetic renal lesions in proteinuric type 2 diabetic patients even in the presence of diabetic retinopathy. However, given the risk of renal biopsy, it is neither ethical nor necessary to perform renal biopsies in all proteinuric diabetic patients. In summary, presence of diabetic retinopathy in proteinuric type 2 diabetic patients does not predict the nature of nephropathy. Diabetic nephropathy can occur in absence of DR and patients with DR may have both diabetic and nondiabetic nephropathy. REFERENCES 1. Nuhad Ismail, Bryan, Piotr Strzelczyk, Eberhard Ritz. Renal disease and hypertension in non-insulin dependent diabetes mellitus. Kidney International 1999;55:1-28. 2. Balakuntalum S Kasinath, Salim K Mujais, Benjamin H Spargo, Adrian I Iatz. Non diabetic renal disease in patients with Diabetes mellitus. The Am J Medicine1983;75:613-17. 3. Gambara V, Mecca G, Remuzzi G, Bertani. Heterogeneous nature of renal lesions in type 11 diabetes. J Am Soc Nephrol 1993;3:1458-66. 4. Parving HH, Gall MA, Skott P, Jorgenson HE, Jorgenson F, Nielsen B, Larsen S. Prevalence and causes of albuminuria in non-insulin-dependent diabetic patients. Kidney International 1992;41:758-62. 5. Lipkin GW, Yeates C, Howic A, Micheal J, And D, Richards NT. More than one third of type 2 diabetic with renal disease do not have diabetic retinopathy: A prospective study (Abstract). J Am Soc Nephrol 1994;5:379-83. 6. American Diabetes Association: Report of the expert committee on the diagnosis and classification of Diabetes mellitus. Diabetic Care 1997;20:1183-97. 7. Richards NT, Greaves SJ, Lee AJ, Howie D, Adu JM. Increaed prevalence of renal biopsy findings other than diabetic glomerulopathy in type II Diabetes mellitus. Nephrol Dial Transplant 1992;7:397-99. 8. Lee EY, Chung CH, Choi SO. Non diabetic renal diseases in patients with non-insuloin dependent diabetes mellitus. Yonsei Med J 1999;40:21-6. 9. John GT, Date A, Korula A, Jeyaseelan L, Shastry JCM, Jacob CK. Nondiabetic renal disease in noninsulin-dependent diabetics in a south Indian hospital. Nephron 1994;67:441-3. 10. Venkateswara K, Crosson JT. Idiopathic membranous glomerulonephritis in diabetic patients: Report of three cases and review of literature. Arch Intrn Med 1980;140:634-7. 11. Lippert J, Ritz E, Schwarzbeck A, Schneidner P. The rising tide of end stage renal failure from diabetic nephropathy type 2.An epidemological analysis. Nephrol Dial Transplant 1995;10:462-7. 12. Amoah E, Glickman JL, Malchoff CD, Sturgill BC, Kaiser DL, Bolton WK. Clinical identification of non-diabetic renal disease in diabetic patients with type 1 and type 2 disease presenting with renal dysfunction. Am J Nephrol 1988;8:204-11. 13. Prakash J, Sen D, Usha, Kumar NS. Non diabetic renal disease in patients with type2 diabetes mellitus. J Assoc Physician India 2001;49:415-20. 14. Premalatha G, Vidhya K, Deepa R, Ravi Kumar R, Rema M, Mohan V. Prevelance of NDRD in type2 DM patients in a diabetes centre in southern India. J Assoc Physician India 2002:50:1135-1139. 15. Gill HS, Dash SC, Dinda AK. Clinico-pathological study of non diabetic glomerular disease in NIDDM. Indian Journal of Nephrol 1997;7:123-8. 16. Tarrass F, Anabi A, Zamad M, Ramdani B, et al. Idiopathic membranous glomerulonephritis in patients with type 2 diabetes mellitus. Hong Kong J Nephrol 2005;7:34-7. 17. Cordonnier D. Glomerular involvement in type 2 diabetes-is it all diabetic? Nephrol Dial Transplant 1996;11:936-38. 18. Koch M, Thomas B, Tschoc W, Ritz. Survival and predictors of death in dialysed diabetic patients. Diabetologia 1993;36:1113-7. 19. Gall MA, Rossing P, Skott P, et al. Prevelance of micro and macroalbuminuria, arterial hypertension, retinopathy and large vessel disease in European type 2(NIDDM) diabetic patients. Diabetologia 1991;34:653-8. 20. Christen PK, Larsen S, Horn T, Obsen S, Parving HH. Causes of albuminuria in patients with type 2 diabetes without diabetic retinopathy. Kid Int 2000;58:1719-31. 21. Kveder R, Kajtna-Koselj M, Rott T, Bren AF. Nephrotic syndrome in patients with diabetes mellitus is not always associated with diabetic nephropathy. Nephrol Dial Transplant 2001:16[Suppl 6]:86-7. Announcement HIV Congress 2008 Date : 21st to 23rd March, 2008 Venue : Taj Lands End Hotel, Mumbai. Contact : JK Maniar, Organising Chairperson Email : jkmaniar@vsnl.com 416 www.japi.org JAPI VOL. 55 JUNE 2007