Muhammad Yakoob Ahmedani ( Baqai Institute of Diabetology and Endocrinology, Baqai Medical University, Karachi, Pakistan. )
Muhammad Zafar Iqbal Hydrie ( Baqai Institute of Diabetology and Endocrinology, Baqai Medical University, Karachi, Pakistan. )
Azhar Iqbal ( Baqai Institute of Diabetology and Endocrinology, Baqai Medical University, Karachi, Pakistan. )
Asma Gul ( Baqai Institute of Diabetology and Endocrinology, Baqai Medical University, Karachi, Pakistan. )
Waheed Baig Mirza ( Baqai Institute of Diabetology and Endocrinology, Baqai Medical University, Karachi, Pakistan. )
Objective: To determine the frequency of microalbuminuria and its associations in type-2 diabetic subjects attending diabetes centers/clinics across Karachi, Pakistan.
Methods: Two thousand one hundred subjects with type-2 diabetes were screened for microalbuminuria using Micral test strip II. A single screening test was performed in 25 diabetes centers/units in different districts of Karachi from January 2003 to December 2003.
Results: The overall prevalence of microalbuminuria was 34%. Mean age of subjects was 53.1 years + 11.9 years, mean BMI was 25.8 ± 4.1 and mean duration of diabetes was 8.8 ± 5.21 years. Fifty seven percent were males and 43% females. Sixty two percent of the subjects had a systolic blood pressure >130 mmHg. Forty five percent had a family history of diabetes and 5% had a family history of hypertension. Univariate analyses demonstrated significant associations between microalbuminuria and age, duration of diabetes, male gender, smoking status, microvascular and macrovascular complications, hypertension, high triglycerides, high serum LDL, low serum HDL, and high fasting & random blood sugars. When adjusted for the effects of other variables in the model, age, diastolic blood pressure, serum LDL and retinopathy were found to be significantly associated with microalbuminuria.
Conclusion: The prevalence of microalbuminuria in type 2 diabetic subjects in this cross-sectional multicentre study across Karachi was 34% and this was significantly related to age, diastolic blood pressure, serum LDL and retinopathy (JPMA 55:382;2005).
Diabetic nephropathy is characterized by proteinuria and is the leading cause of end-stage renal disease world wide. It constitutes the major work load of dialysis centers. The estimated cost for dialysis per diabetic subject in Pakistan is around $30000/year.1 Diabetic subjects on dialysis and transplant recipients also have higher morbidity and mortality rates than their non diabetic counterparts.2 Progression to established diabetic nephropathy occurs through several stages. Microalbuminuria defined as urinary albumin excretion rate of 20-200 µg/min or urinary protein excretion rate of 30-300 µg/min predicts future development of overt nephropathy.3 As microalbuminuria can be reversed and the future development of overt diabetic nephropathy significantly reduced, screening for microalbuminuria and timely therapeutic intervention has become standard of care world wide. A cross sectional multi-centered study was conducted in Karachi to establish the frequency of microalbuminuria in Type-2 diabetic subjects.
Patients and Methods
Two thousand one hundred subjects with type-2 diabetes were screened for microalbuminuria on their scheduled visits to the outpatient department of 25 diabetic centers/clinics using a single Micral test strip II test between January 2003 and December 2003.
Known type 2 diabetic subjects more than 18 years of age were included in the study. Subjects were excluded from the study if they came to the clinic after vigorous exercise, had any serious illness such as history of heart failure, Urinary Tract Infections or were known patients of nephropathy.
Clinical details of each subject was recorded in a specified proforma especially designed for this study. This included height, weight and body mass index. Type-2 diabetes was diagnosed based on the WHO study group report criteria.
Subjects were identified as hypertensive if they were on antihypertensive medication or if they had a systolic blood pressure more than 130mmHg or diastolic blood pressure more than 85 mmHg.
A subject was labeled as smoker if he/she was smoking actively or was an active smoker in the last 6 months.
The fundus was examined using Vista 20 direct ophthalmoscope by a diabetologist. The retinopathy was taken as positive if there was evidence of either microdots, hard exudates, soft exudates, new vessels or maculopathy.
Peripheral neuropathy was defined as absent touch or vibratory sensations of the feet. Touch was assessed by 10 gm monofilament and vibration sensation by 128 Hz tuning fork.
Ischemic heart disease was considered to be present if there is history of exertional chest pain, with unequivocal T wave changes in the ECG or if there was a typical history of chest pain documented by previous hospitalization records with ECG changes suggestive of myocardial infarction.
Subjects with absent dorsalis pedis or posterior tibial pulses with or without history of intermittent claudication were taken as having peripheral vascular disease (P.V.D.).
Fasting lipid profile, HbA1c, serum creatinine and urine D/R if done within 6 months of starting the study were noted in the proforma.
Microalbuminuria was defined as a urinary albumin >50 mg/l. Microalbuminuria was checked by using semi-quantitative dry immuno chemical screening strips. (Micral II® test strips) (Roche diagnostic GmbH Mannheim Germany).
Descriptive statistics were compared for continuous variables, and percentages for categorical variables. Association between outcome variable and independent variables were sought using chi-square test. To observe the individual effects of each exposure variable, potential confounders were simultaneously controlled by means of multiple logistic regression and ORs with 95% Cls were computed.
The baseline characteristics of the subjects are summarized in Table 1. Mean age of subjects was 53.1 years +11.9 years. Mean Body Mass Index (BMI) was 25.79 ± 4.05 kg/m². The mean duration of diabetes was 8.8 ± 5.21 years. Fifty seven percent of the subjects were males while 43% were females, 62% of the subjects had systolic blood pressure more than 130 mmHg. Forty five percent subjects had a family history of diabetes while 5% had a family history of hypertension.
In our study we found that microalbuminuria was more frequent in males (37.1% vs. 29.9%) as compared to females.
Univariate analyses of the factors associated with microalbuminuria are summarized in Table 2. Strong associations were found for microalbuminuria with age, duration of diabetes, male gender, smoking status, microvascular complications including retinopathy, macrovascular complications, hypertension, dyslipidemia, HbA1c, fasting and random blood glucose. The microalbuminuria positive group was older and had a longer duration of diabetes compared to the microalbuminuria negative group (p<0.001). The microalbuminuria positive group had a higher BMI as compared to the microalbuminuria negative group (p<0.001).
|Table 1. Baseline characteristics of patients. |
|Variables ||n (%) |
|Gender (n=2186) || |
| Male ||1237 (56.6) |
| Female ||949 (43.4) |
|Smoking Status (n=1277) || |
| Yes ||377 (29.5) |
| No ||900 (70.5) |
|Family history of diabetes || |
| Yes ||981 (44.6) |
| No ||1220 (55.4) |
|Family history of hypertension || |
| Yes ||110 (5.0) |
| No ||2091 (95.0) |
|Family history of chronic diseases || |
| Yes ||13 (0.6) |
| No ||2188 (99.4) |
|Family history of renal diseases || |
| Yes ||7 (0.3) |
| No ||2194 (99.7) |
|Hypertension (n=1226) || |
| Yes ||685 (55.9) |
| No ||541 (44.1) |
|Systolic Blood Pressure (n=1101) || |
| <130 ||416 (37.8) |
| >130 ||685 (62.2) |
|Diastolic Blood Pressure (n=1100) || |
| <85 ||527(47.9) |
| >85 ||573(52.1) |
| *Age (in years) ||53.10 + 11.90 |
| *Duration of diabetes (in years) ||8.88 + 5.21 |
| *BMI ||25.79 + 4.05 |
|* Mean ± SD is reported for continuous variables; BMI: Body Mass Index |
|Table 2. Unadjusted and adjusted odds ratio for microalbuminuria in patients. |
|Variables ||Microalbuminuria Absent n (%) ||Microalbuminuria Present n (%) ||Unadjusted Or for Microalbuminuria (95% CI) ||*Adjusted OR for Microalbuminuria (95% CI) |
|Age >50 yrs (n=1396) ||826 (59.5) ||570 (78.0) ||2.41 (1.97 - 2.96) ||10.50 (4.59 - 24.16) |
|Male ||778 (53.9) ||459 (61.8) ||1.38 (1.15 - 1.65) ||†NS |
|Duration of diabetes (n=1320) || || || |
|1-5 years ||265 (37.3) ||96 (15.8) || |
|6-8 years ||201 (28.3) ||144 (23.6) ||1.98 (1.44 - 2.71) ||NS |
|9-11 years ||140 (19.7) ||165 (27.1) ||3.25 (2.35 - 4.50) |
|>11 years ||105 (14.8) ||204 (33.5) ||5.36 (3.85 - 7.47) |
|Smoker ||167 (23.4) ||210 (35.6) ||1.72 (1.35 - 2.19) ||NS |
|Retinopathy (n=1246) ||192 (29.2) ||284 (48.3) ||2.27 (1.80 - 2.86) ||6.39 (2.04 - 19.96) |
|Neuropathy (n=1184) ||215 (33.2) ||262 (48.8) ||1.91 (1.51 - 2.42) ||NS |
|Cerebrovascular (n=1168) ||92 (14.5) ||130 (24.3) ||1.89 (1.40 - 2.54) ||NS |
|Coronary vessels (n=1127) ||131 (21.8) ||196 (37.3) ||2.14 (1.65 - 2.78) ||NS |
|Peripheral Vessels (n=1029) ||66 (11.6) ||131 (28.4) ||3.00 (2.17 - 4.17) ||NS |
|Hypertension ||263 (40.6) ||422 (73.0) ||3.96 (3.11 - 5.04) ||NS |
|Systolic Blood Pressure (mmHg) || || || ||NS |
|<120 ||203 (34.6) ||83 (16.1) || |
|120<130 ||99 (16.9) ||31 (6.0) ||0.77 (0.48 - 1.23) |
|130<160 ||257 (43.8) ||316 (61.5) ||3.00 (2.22 - 4.08) |
|>160 ||28 (4.8) ||84 (16.3) ||7.34 (4.46 - 12.08) |
|Diastolic Blood Pressure (mmHg) || || || |
|<70 ||231 (39.4) ||75 (14.6) || |
|71-85 ||168 (28.6) ||53 (10.3) ||0.97 (0.65 - 1.46) ||1.12 (0.31 - 4.01) |
|85-90 ||135 (23.0) ||235 (45.8) ||5.36 (3.83 - 7.50) ||6.95 (1.96 - 24.63) |
|>90 ||53 (9.0) ||150 (29.2) ||8.72 (5.8 - 13.10) ||14.42 (3.79 - 54.75) |
|TGL >150 mg/dl ||76 (67.3) ||192 (93.7) ||7.19 (3.62 - 14.29) ||NS |
|HDL <40 (M) and > 50 (F) mg/dl ||79 (81.4) ||178 (91.3) ||2.39 (1.17 - 4.87) ||NS |
|LDL >100 mg/dl ||66 (69.5) ||178 (93.2) ||6.01 (2.95 - 12.27) ||9.19 (2.88 - 29.37) |
|Fasting Blood Sugar >110 mg/dl ||341 (93.2) ||362 (97.3) ||2.65 (1.26 - 5.60) ||NS |
|Random Blood Sugar >200 mg/dl ||222 (65.1) ||272 (72.3) ||1.40 (1.02 - 1.93) ||NS |
|BMI >23 ||919 (78.7) ||558 (86.6) ||1.76 (1.34 - 2.29) ||NS |
|*Serum Cholesterol ||192.5 ± 41.76 ||237.44 ± 46.09 ||1.02 (1.01 - 1.03) ||NS |
|*HBA1c ||7.94 ± 1.57 ||8.18 ± 0.94 ||1.20 (0.93 - 1.55) ||NS |
|Sample size is different for all variables in univariate analysis. Sample size for final multivariate model was 242. *Mean ±SD is reported.|
†Not Significant in final model
BMI - Body Mass Index
The microalbuminuria positive group had a higher systolic and diastolic pressure compared to the microalbuminuria negative group (p<0.001). The microalbuminuria positive group had a more deranged lipid profile with higher serum total cholesterol, triglycerides, LDL cholesterol and lower HDL levels compared to the microalbuminuria negative group.
When adjusted for the effects of other variables in the model, age was found to be significantly associated with microalbuminuria. (AOR: 9.67; 95% CI: 4.16 - 22.49).
High diastolic blood pressure was also significantly associated with microalbuminuria (AOR: for DBP>90 mmHg: 14.42, 95% CI: 3.79-54.75).
Serum LDL was significantly associated with microalbuminuria. The odds of subjects having serum LDL>100 mg/dl among the microalbuminuria group was 8.41 times the odds of serum LDL >100 mg/dl among subjects who did not have microalbuminuria (AOR: 9.19, 95% CI: 2.88-29.37).
Finally, the odds of subjects having retinopathy among the microalbuminuria group was 5.92 times greater than subjects who did not have microalbuminuria (AOR: 5.92, 95% CI: 1.88 -18.69). The goodness of fit test demonstrated good fit (p=0.485).
Diabetic nephropathy is the most frequent cause of end stage renal disease. Microalbuminuria is the first clinical detectable sign of involvement of the kidney. It affects between 20-40% of subjects 10-15 years after the onset of diabetes. Once microalbuminuria is present, it progresses over 5-10 years to proteinuria in 20-50% subjects. With microalbuminuria, the decline in renal functions varies but average reduction in glomerular filtration is around 10-12 ml/min/year.4 Progression to end stage renal disease is accelerated by hypertension. The process of renal involvement is step wise and microalbuminuria also referred to as incipient nephropathy, is potentially reversible.5 Microalbuminuria is also strongly associated with traditional cardiovascular risk factors and cardiovascular complications. Mogensen et al in 1984 reported a significant increase in cardiovascular and total mortality in subjects with type 2 diabetes who had microalbuminuria.6 Dineen and Gerstein drew similar conclusions from a meta-analysis of 11 longitudinal studies.7 Thus it is important to detect and treat incipient diabetic nephropathy. Screening for microalbuminuria can be performed by four methods: Measurement of albumin to creatinine ratio in random urine sample; 24 hour and timed overnight urine collection for protein; Microalbuminuria by using micral dip stick. Standard assays to check urinary microalbumin are not generally available in laboratories in Pakistan. An easy and accurate method of detection of microalbuminuria is albumin to creatinine ratio in a spot urine sample. Screening for microalbuminuria with micral II strips is relatively cheap, fast and has an acceptable sensitivity of 96.7% with a specificity of 71%.8 All positive tests should be rechecked and confirmed by more specific tests. In our study we did the first initial screening of urine for microalbuminuria excluding causes like exercise, U.T.Is, marked hypertension and hyperglycemia, heart failure, and febrile illness. Micral test II is an optically-read immunoassay specifically for detection of microalbuminuria and the use of these strips has been widely advocated.8
Several studies have shown the overall high sensitivity (b/w 79-99%) of Micral strips but lower specificity (67-87%) with higher negative predictive values than positive predictive values.4-9 Due to natural variations in albumin excretion it is generally recommended to carry out this test 2-3 times to confirm microalbuminuria, hence specificity of the test may not be so crucial as a subsequent test may give a true negative result. Several studies comparing Micral test II and laboratory methods of detecting albuminuria have concluded that it could be used as a screening tool but not as diagnostic tool.
In our study, frequency of microalbuminuria was found to be 34% which is similar to that reported from other Asian countries.10 Epidemiological studies report prevalence of microalbuminuria in Type 2 diabetes ranging between 8% to 32%.11,12 This variation may be due to different criteria used for defining the condition, the stage of the disease, method of assessment and ethnicity.13 Microalbuminuria was more frequent in males (37.1% vs. 29.9%) as compared to females which has been observed in other studies.14,15 The microalbuminuria positive group was older and had a longer duration of diabetes compared to the microalbuminuria negative group in agreement with other studies.14,16 The microalbuminuria positive group had a higher BMI as compared to the microalbuminuria negative group.
A cross-sectional community-based prevalence study with a large sample size of 7841 subjects showed that several CVD risk factors including BMI and W.H.R were associated with urinary albumin excretion. It was shown that obese subjects with central fat distribution had high risk of microalbuminuria independent of blood pressure and plasma glucose. It was also seen that for a given increment in age, BMI and plasma glucose the increment in U.A.E (Urinary albumin excretion) was greater in males than in female subjects.17 However, a study from Africa showed no relation of microalbuminuria to BMI.18 There is strong evidence that central obesity is related to insulin resistance.19 Several studies have shown an association between microalbuminuria and insulin resistance20 and the W.H.O definition of metabolic syndrome lists microalbuminuria as one of the important components of the syndrome.
The microalbuminuria positive group had a higher systolic and diastolic blood pressure compared to the microalbuminuria negative group (p<0.001) which has been observed by others.21
The microalbuminuria positive group had a more deranged lipid profile with higher serum total cholesterol, triglycerides, LDL cholesterol and lower HDL compared to the microalbuminuria negative group. Significant lipids abnormalities including high VLDL, LDL and triglycerides and low levels of HDL have also been reported in the literature in subjects with microalbuminuria.22 In a prospective observational study by Gall et al base-line cholesterol was found to be an independent risk factor for the development of microalbuminuria.23
There were more smokers in the microalbuminuria positive group as compared to the microalbuminuria negative group. Nilsson et al have recently shown an association of smoking with increased HbA1c values and microalbuminuria in diabetic subjects.24
Poor glycaemic control has been shown to be an independent risk factor for microalbuminuria.10,24,25 Univariate analysis showed a significant association of microalbuminuria with Fasting Blood Sugars and Random Blood Sugars but not with HbA1c. A possible explanation could be that both microalbuminuria positive and negative groups were on treatment and the mean HbA1c of both the groups was similar; Secondly, glycated hemoglobin is estimated and checked by different methods by different laboratories; hence lack of standardization may be one of the reasons for this finding. Moreover, the last HbA1c values that were available at the time of screening may not be a true representation of overall glycaemic control of that particular individual.
Microalbuminuria positive group also had more microvascular complications like retinopathy and neuropathy than the microalbuminuria negative group. This association of microalbuminuria with retinopathy and neuropathy has been reported by other groups.26,27
Macrovascular disease including coronary, cerebral and peripheral vascular disease was found to be more prevalent in the microalbuminuria positive group similar to the findings in other studies.28
In conclusion the prevalence of microalbuminuria in type 2 diabetes mellitus in this first ever cross-sectional multicentre study across Karachi is 34%. The risk factors are similar to those reported from other Asian countries. Because of the adverse impact of proteinuria on survival in subjects with type 2 diabetes, screening and intervention programs should be implemented early at the stage of microalbuminuria and risk factors should be treated aggressively.
We would like to acknowledge the guidance and support of Prof. Rubina Hakeem and Mr. Imran Waheed of BIDE research department for the study.
BIDE research department acknowledges the support of Roche Pakistan Ltd. and PharmEvo (Pvt.) Ltd. for making this study possible.
We would like to thank the following doctors for their active participation in our study.
Dr. Anees Memon, Dr. Aqeel Baig, Dr. Arshad Sattar, Dr. Asher Fawad, Dr. Aslam Pervaiz, Dr. Azeem Rajput, Dr. Fareeduddin Sheikh, Dr. Fahimuddin, Dr. Fareeduddin, Dr. Inam Ashraf, Dr. Jasmeen Tara, Dr. Jawed, Dr. Kamal Mustafa, Dr. Khursheed Anwar, Dr. Khalil Pathan, Dr. Naseeruddin, Dr. Obaid Hashmi, Dr. Osama Rasheed, Dr. Rashid Jameel, Dr. S.M.Zafar, Dr. Saleem Ahmed, Dr. Shakeel Ahmed, Dr. Veru Mal, Dr. Wajahat, Dr. Zameer Ali Siddiqi.
Special thanks to Prof. Masood Hameed Khan, Prof. Zaman Sheikh and Dr. Rayaz A. Malik for their active participation and patronage.
1. Naqvi SAJ. Nephrology services in Pakistan; Nephrol Dial Transplant 2000;15:769-71.
2. Nicholos RL. Diabetic kidney disease: Preventing dialysis and transplantation Clinical Diabetes 2003;21:55-62.
3. Graziella B, Franco M. Progression to overt nephropathy in type-2 diabetes. The Casale Monferrato study. Diabetes Care 2003;26:2150-55.
4. Ritz E, Orth SR. Nephropathy is patients with type 2 diabetes mellitus N Engl J Med 1999;341:1127-33.
5. Bruce AP, Linda HF, Kristen HS. Regression of microalbuminuria in type 1 Diabetes. NEJM 2003;348:2285-93.
6. Mogensen CE. Microalbuminuria predicts clinical proteinuria and early mortality in maturity-onset diabetes N Engl J Med 1984;310:356-60.
7. Dinneen SF, Gerstein HC. The association of microalbuminuria and mortality in non-insulin dependent diabetes mellitus. A systematic overview of the literature. Arch intern Med 1997;157:1423-8.
8. Mogensen CE, Viberti GC, Peheim E, Kutter D, Hasslacher C, Hoffmann W, et al. Multi-centre Evaluation of the micral test II test strip an immunological Rapid test for the detection of microalbuminuria. Diabetes care 1997;20:1642-6.
9. Fernandez Fernandez I, Paez Pito JM, Hermosin Bonot, Vazquez Garijo P, Ortiz Camonez MA, Tarilonte Deglado MA. Rapid screening test evaluation for microalbuminuria in Diabètes mellitus. Acta Diabetol 1998;35:199-202.
10. Varghese A, Deepa R, Rema M, Mohan V. Prevalence of microalbuminuria in type 2 diabetes mellitus at a diabetes centre in southern India. Postgrad Med J 2001;77:399-402.
11. Gupta DK, Verma LK, Khosla PK, Dash SC. The prevalence of microalbuminuria in diabetes, a study from north India. Diabetes Res Clin Pract 1991;12:125-8.
12. Allawi J, Rao PV, Gillbert R, Scott G, Jarrett RJ, Keen H, et al. Microalbuminuria in non-insulin-dependent diabetes: its prevalence in India compared with Europid patients. B.M.J 1988;296:462-4.
13. Burden AC, McNally PG, Feehally J, Walls J. Increased incidence of end stage renal failure secondary to diabetes mellitus in Asian ethnic groups in the United Kingdom. Diabetes Med 1992;9:641-5.
14. John L, Rao PS, Kanagasabapathy AS. Prevalence of diabetic nephropathy in non-insulin dependant diabetes. Indian J Med Res 1991;94:24-9.
15. Mather HM, Chaturvedi N, Kheley AM. Comparison of prevalence and risk factors of microalbuminuria in south Asians and Europeans with type 2 diabetes mellitus. Diabet Med 1998;15:672-7.
16. Hashim R, Rehman K, Ahmed TA, Mushtaq S, Lubna Z, Attique M. Microalbuminuria and associated risk factors in Type 2 diabetes. JCPSP 2004;14:84-7.
17. Verhave JC, Hillege HL, Burgerhaf JG, Navis G, deZeeuwD, de jong PE; Prevend Study Group Cardiovascular risk factors are differently associated with urinary albumin excretion in men and women. J Am Soc Nephrol 2003;14:1330-5.
18. Erasmus RT, Okesina B. Microalbuminuria in a clinic population of type 2 Melanesian diabetics: relationship to glycaemic control, blood pressure and duration of diabetes. Cent Afr J Med 1999;45:291-4.
19. Haffiner SM, Stern MP, Hazuda HP, Pugh J, Patterson JK. Do upper-body and centralized adiposity measure different aspects of regional body-fat distribution? Relationship to non-insulin-dependent diabetes mellitus, lipids and lipoproteins. Diabetes 1987;36;43-51.
20. Mykkanen L, Haffner SM, Kuuissto J, Pyorala K, Laakso M. Microalbuminuria precedes the development of NIDDM. Diabetes 1994;43:552-7.
21. Hirano T. Lipoprotein abnormalities in diabetic nephropathy. Kidney Int, 1999;56 (suppl. 71), S22-4.
22. Gall MA, Hougaarod P, Borch-Johnsen K, Parving HH. Risk factors for development of incipient and over diabetic nephropathy in patients with non-insulin dependent diabetes mellitus prospective observation study. BMJ 1997;314:783-8.
23. Nilsson PM, Gudbjorns dottir S, Eliasson B. Smoking is associated with increased HbA1c values and microalbuminuria in patients with diabetes data from the National Diabetes Register in Sweden. Diabetes & Metabolism 2004;30:261-8.
24. Haffiner SM, Morales PA, Gruber MK, Hazuda HP, Stern MP. Cardio vascular risk factors in non-insulin dependent diabetic subjects with microalbuminuria. Arterioscler Thromb 1993;13:205-10.
25. Meigs JB, D'Agostino RB, Nathan DM, Rifai N, Wilson PWF. Framinsham Offspring Study. Longitudinal association of glycamia & microalbuminuria Diabetes Care 2002 25:977-83.
26. Bell DS, Ketchum CH, Robinson CA. Microalbuminuria associated with diabetic neuropathy. Diabetes Care 1992;15:528-31.
27. Borch-Johnson-K, Feldt-Rasmussen, Strandgrand S, Schroll M, Jensen JS. Urinary albumin Excretion: An independent predictor of Ischemic heart disease. Arterioscler Thomb Vasc. Biol 1999;19;1992-7.
28. Yuyun MF, Khaw KT, Luben R, Welch A, Binghani S, Day NE, et al. A prospective study of microalbuminuria and incident coronary heart disease and its prognostic significance in a British population, the EPIC-Norfolk study. AMJ Epidemol 2004;159;284-93.