Prevalence of microalbuminuria in Saudi Arabians with non-insulin-dependent diabetes mellitus: a clinic-based study

Diabetes Research and Clinical Practice 26 (1994) 1IS-120

ELSEVIER

Prevalence of microalbuminuria in Saudi Arabians with non-insulin-dependent diabetes mellitus: a clinic-based study Aus A. Alzaid*“, S. Sobkib, V. De Silvab aDepartmeni of Medicine, Armed Forces Hospital, P.O. Box 27446, Riyadh 11417, Saudi Arabia ‘Department of Biochemistry, Armed Forces Hospital, Riyadb, Saudi Arabia

Received 22 March 1994; revision received 17 June; accepted 25 July 1994

Abstract

prevalence of diabetic nephropathy varies in different racial groups, being especially high in communities that have abandoned an active traditional living and embraced a modem but sedentary life-style. As a new and rapidly developing country, Saudi Arabia has witnessed impressive changes in socioeconomic growth and development and concurrently, a disturbing trend in non-insulin-dependent diabetes mellitus (NIDDM). These observations therefore prompted us to investigate the prevalence of microalbuminuria among Saudi Arabians with NIDDM. Two hundred and eleven patients attending a large Diabetic Clinic in Riyadh were screened for microalbuminuria (30-300 mg/24 h). Twenty-seven subjects had clinical proteinuria (dipstick-positive) and were excluded, leaving 184 cases for analysis. Seventy-six subjects (76/184,41.3%) had microalbuminuria. These subjects had higher fasting plasma glucose concentrations (P = 0.002) and greater body mass index (P = 0.049) than subjects with normal albumin excretion rate (< 30 mg/24 h). There were no significant differences between subjects with and without microabuminuria with regards to fasting total plasma cholesterol and triglycerides concentrations, frequency of hypertension, duration of diabetes or type of therapy for diabetes. In multivariate analysis, glycaemia (P
in Saudi Arabia must now await further population-based studies. Keywords:

Microalbuminuria;

Saudi Arabia; Type 2 diabetes; Renal disease; Diabetic complications

1. Introduction Microalbuminuria is an independent risk factor for premature death, mainly from cardiovascular disease in people with non-insulin-dependent * Corresponding author. 0 1994 Elsevier Science Ireland 0168-8227/94/307.00 SSDI 0168-8227(94)00986-5

diabetes mellitus (NIDDM) [l-4]. However, the frequency and severity of albuminuria in diabetic patients may vary in different populations [5-71. Saudi Arabia has a relatively high incidence of NIDDM [8], a disease also perceived to be rising in incidence as a consequence of the sudden affluence and urbanization which have taken place

Ltd. All rights reserved

116

A.A. Alzaid et al. /Diabetes

over the past four decades [9]. This study therefore, set out to investigate the prevalence of microalbuminuria and factors which might be related to its development in Saudi Arabians with NIDDM. 2. Methods Over a 6-month interval, all willing patients with NIDDM who were attending a Diabetic Clinic at the Armed Forces Hospital, one of the main hospitals in Riyadh, were asked to submit to screening for microalbuminuria. The diagnosis of NIDDM was based on age at onset of diabetes > 30 years and lack of requirement for insulin therapy at the time of diagnosis. Following regular assessment at the clinic, each potential screenee was given verbal and written instructions on the timing and use of containers for 24 h urine collection. Two hundred and eleven patients successfully submitted appropriate urine specimens. Of these, 27 subjects had overt proteinuria as indicated by positive dipsticktesting (Combur-8, Boehringer Mannheim, Germany). These subjects were excluded from this report. The remaining 184 cases were analyzed for the presence of microalbuminuria which was defined as urinary albumin excretion rate of 30-300 mg/24 h. Albumin excretion rates were also determined in a group of non-diabetic subjects. This ‘reference’ group comprised 39 healthy (i.e. with no known or documented history of major past or present illness) individuals (24 males and 15 females, age (mean f S.D.) 47 f 3 years) who were attending primary health care clinics in the same hospital for insignificant health related problems. Specifically, none of the subjects in this group had a personal history of diabetes, hypertension or renal disease and all had normal fasting plasma glucose concentrations. Urine samples were collected in preservativefree plastic containers. Total volume (mean f SD.) of urine per specimen collected was 1721 ml ( f 768). Aliquots from each sample were stored at 4°C and analysis was performed within 1 week in all cases. Microalbuminuria was determined by an ‘in-house’ immunoturbidimetric method [lo] which had a detection limit of 5 mg/l, intra-assay

Rex Clin. Pratt. 26 (1994) 115-120

and inter-assay coefficients of variation (C.V.) of 4% and 4.6%, respectively. Biochemical indices (fasting plasma glucose, fasting total plasma cholesterol and triglycerides) measured during the study period (79%) were included in the final analysis: plasma glucose concentrations were measured by a hexokinase method (Boehringer, Mannheim, Germany). The normal range for Saudi Arabians is 3.3-6.l/mmol [ 111. Laboratory measurement of glycosylated haemoglobin (HbAr,) was not available in our hospital at the time of the study. Plasma cholesterol and triglycerides were measured as per standard method using American Monitor ‘parallel’ analyzer (American Monitor Corporation, Indianapolis, USA). Intra-assay C.V. for cholesterol was 2.4% at 3.75/mmol and 1.5% at 5.21/mmol and for triglycerides, 0.9% at l.Wmmol. Inter-assay C.V. for triglycerides was 4.2% at 0.86/mmol and 4.3% at 2.25lmmol. Clinical hypertension was considered present if blood pressure measurement recorded during study period was equal to or greater than 160195 mmHg (i.e. systolic r 160 mmHg and/or diastolic 2 95 mmHg) and/or if there was a documented history (past or present) of hypertension or use of antihypertensive therapy. Body mass index (BMI) was calculated and expressed as ks/m2.

Fig. I. Distribution rates of albumin excretion (mg/24 h) in the diabetic subjects (n = 184).

A.A. Aizaid et al. /Diabetes Rex Clin. Pratt. 26 (1994) 115-120

117

3. Results Seventy-six diabetic subjects (76/184, 41.3%) had microalbuminuria. This prevalence rate was markedly higher (P = 0.007) than that observed in the non-diabetic ‘reference’ group (5139, 12.8%). The distribution rates of albumin excretion for the diabetic and non-diabetic subjects are shown in Figs. 1 and 2, respectively. Of note, albumin excretion rates in a majority of the diabetic patients with microalbuminuria fell in the range between 30 and 100 mg/24 h, and among the non-diabetic group, none had exceeded 100 mg/24 h. Table 1 shows the clinical characteristics of diabetic subjects with and without microalbuminuria. Subjects with microalbuminuria had significantly higher fasting plasma glucose and body mass index. There were no significant differences in age, gender, lipid concentrations, prevalence of hypertension, duration of diabetes or type of diabetic therapy between subjects with and without microalbuminuria. Table 2 shows linear correlations between log albumin excretion (treated as a continuous variable) and other variables. Albumin excretion was significantly related to fasting glucose and duration of diabetes. In multivariate analysis how-

Fig. 2. Distribution rates of albumin excretion (mgG!4h) in the non-diabetic subjects (n = 39).

2.1. Statistical

analysis

Differences between groups were tested for by the Student’s t-test for continuous variables and the chi-square test for dichotomous variables. Albumin excretion rates were logarithmically transformed before analysis. Individual correlations were assessed by linear regression and multiple analysis by multivariate analysis, with log albumin excretion as the dependent variable.

Table 1 Clinical and biochemical characteristics (mean f SD.) of diabetic subjects with and without microalbuminuria Characteristic

n

P

Microalbuminuria Present

Absent

76 51

108 47

0.845

Male (%) Age (years) BMI (kg/m’) Duration of diabetes (years) Fasting glucose (Immol) Total cholesterol (/mmol) Triglycerides (Anrnol) Hypertension (%)

55 f 9.2 30.4 f 5.4

56 l 9.6 28.4 + 5.6

0.230 0.049

10.7 f 13.4 l 5.5 * 2.2 * 31

9.2 + 11.1 l 5.8 zt 2.2 f 37

0.073 0.002 0.153 0.761 0.831

Treatment Diet (%) Oral agents (%) Insulin (%)

59 30

I1

5.9 4.9 1.1 1.4

7 64 29

5.2 4.3 1.4 1.5

0.397 0.557 0.951

118

A.A. Alzaid et al. / Diabeles Res. Clin. Pratt. 26 (1994) 115-120

Table 2 Dnivariate analysis between albumin excretion and other variables Variable

rs

P

Age

-0.026 0.182 0.065 0.252 -0.110 -0.042

0.724 0.014 0.448 0.002 0.195 0.625

Diabetes duration Body mass index Fasting plasma glucose Total cholesterol Triglycerides

ever, fasting plasma glucose remained highly significantly related to albuminuria (Table 3). Duration of diabetes was of borderline significance. 4. Discussion We show in this report that the prevalence of microalbuminuria is relatively high (41.3%) in a clinic-based sample of Saudi Arabians with NIDDM. Also, we report herein a close but modest relationship between fasting plasma glucose concentration and daily albumin excretion rate in this population. Non-insulin-dependent diabetes mellitus is a common disease in Saudi Arabia [8,9]. Even more worrisome is the growing concern that the incidence of NIDDM may be insidiously rising as the country continues to absorb the rapid growth and development of recent decades [9]. Microalbuminuria has been recently shown to be an independent risk factor for overall mortality (mainly from cardiovascular disease) in patients with NIDDM [l-4]. Thus, it was pertinent to investigate the Table 3 Multivariate analysis to identify variables with independent association to albumin excretion Variable

P

Age

0.944

Body mass index Fasting plasma glucose Diabetes duration Hypertension

0.374 0.0008 0.0522 0.126

prevalence of this early but apparently ominous complication in our diabetic population. Our results indicate that the prevalence of microalbuminuria is at least 3-fold higher in the diabetic than the non-diabetic individuals. It is also noticeably higher than the prevalence rates generally reported (7-35%) for Caucasians with NIDDM [1,3,5, 12-151. Several important factors however, may have contributed to this apparently high prevalence rate and so deserve consideration. First, it is possible that for many of our patients, the diagnosis of diabetes may have been made later than usual. In fact, while it is generally reported that hyperglycaemia might be present for a period of 4-7 years before clinical diagnosis is made [ 161,we would suggest that this latent period might be even longer for Saudi diabetic patients in whom the symptoms of diabetes are usually typically absent even in the face of moderate/severe hyperglycaemia [ 171. Secondly, we did not gather information on the diabetic subjects who did not take part in the study either because they simply declined to participate or because they failed to submit the necessary urine samples. Furthermore, the hospital where the current investigation was performed is considered a major regional and national referral centre. Therefore, some selection bias may have been inevitable [ 181. Finally, estimation of albuminuria reported in the present work was based on a single 24 h urine collection and measurement of daily urinary creatinine (to assess adequacy of urine collection) was not sought. While it has been suggested that assessment of creatinine excretion may not necessarily be a reliable index of the completeness of urine collection [ 191, we acknowledge that such measurement as well as use of multiple urine samples would be required in future studies to confirm these preliminary findings. Nevertheless, the high frequency of microalbuminuria observed in this study might be reflective of genuine ethnic differences in predisposition to diabetic nephropathy. In this regard, it is of special interest to note that similar high prevalence rates of microalbuminuria have been observed in NIDDM patients of non-Caucasian origin including Pima Indians, Mexican Americans and Indian immigrants in Britain [5-71. Indeed, it is tempting to ascribe the high frequency of

A.A. Alzaid et al. /Diabetes Rex Clin. Pratt. 26 (1994) 115-120

microalbuminuria (and perhaps that of NIDDM in general) to the remarkable change in lifestyle that these populations have undergone in recent decades. Given the conclusive findings recently reported on the relationship between metabolic control and complications in insulin-dependent diabetic patients [20], it was perhaps not surprising to find glycaemia as the strongest factor related to albumin excretion in this study. Furthermore, initial fasting hyperglycaemia has been previously reported to be a strong predicting factor for the development of persistent proteinuria in patients with NIDDM 1211.However, our findings should be interpreted with caution. Fasting plasma glucose concentration per se is by no means a reliable index of chronic glycaemic control. Moreover, we did not demonstrate a cause and effect relationship between albuminuria and glycaemic control in Saudi diabetic patients in this cross-sectional study; rather, our findings merely indicated a close association between the two. A disturbing observation to note however, is the scale and magnitude of fasting hyperglycaemia encountered amongst some Saudi diabetic patients (61% of patients in the present study had fasting glucose > 10 mM). In this study, we failed to observe a significant relationship between albumin excretion rate and conventional risk factors such as age, fasting lipids, obesity and hypertension. The reasons behind this are not clear. In the case of hypertension however, it is possible that because we used a high cutoff level for identifying cases with hypertension and/or the fact that hypertension was considered merely as a categorical variable, may have contributed to the lack of association between hypertension and albuminuria found in this study. The association observed between albuminuria and the number of years since diagnosis of diabetes was of borderline significance. The fact that non-insulindependent diabetes may be silent for many years before clinical recognition [16] may explain why previous studies have reported both significant and non-significant associations between albumin excretion and known duration of the disease [ 13,2 1,221.It is also important to point out that existing evidence for associations between recognized cardiovascular risk factors and albuminuria in

119

patients with NIDDM has not always been uniform [3,23-271. This may, at least in part, be the result of differences in the methods employed for patient selection and/or laboratory analysis of microalbuminuria. In summary, the findings presented in this report indicate that microalbuminuria is relatively common in a clinic-based sample of Saudi patients with NIDDM. If the prognostic value of microalbuminuria as a predictor of cardiovascular disease [l-4] were to be also true for Saudi diabetic patients (though this has not yet been proven) and, if the current trend of NIDDM continues to rise, NIDDM could soon emerge as a growing health care problem in Saudi Arabia. However, we must stress herein that the clinic population described in this report may not necessarily represent the diabetic population at large in Saudi Arabia. Thus, large population-based studies are eagerly awaited to define the true scale and significance of albuminuria in this population. Acknowledgements We wish to thank A. Al-Dehaimi, Dr. A. AlHomaidi and the staff at Primary-Care and Diabetes Clinics (Riyadh Armed Forces Hospital) for their help with this study. References 111Mogensen, C.E. (1984) Microalbuminuria predicts clinical proteinuria and early mortality in maturity-onset diabetes. N. Engl. J. Med. 310, 356-360. 121Jarrett, R.J., Viberti, G.C., Argyropoulos, A., Hill, R.D., Mahmud, U. and Murrells, T.J. (1983) Microalbuminuria predicts mortality in non-insulin-dependent diabetes. Diabetic Med. I, 17-19. [31 Schmitz, A. and Vaeth, M. (1988) Microalbuminuria: a major risk factor in non-insulin-dependent diabetes. A IO-year follow-up study of 503 patients. Diabetic Med. I, 126-134. [41 Neil, M., Hawkins, M., Potok, M., Thorogood, M., Cohen, D. and Mann, J. (1993) A prospective population-based study of microalbuminuria as a predictor of mortality in NIDDM. Diabetes Care 16,996- 1003. ISI Allawi, J., Rao, P.V., Gilbert, R. et al. (1988) Microalbuminuria in non-insulin-dependent diabetes: its prevalence in Indian compared to Europid patients. Br. Med. J. 296, 462-464. WI Nelson, R.G., Kunzelman, C.L., Petitt, D.J., Saad, M.F.,

120

[7]

[8]

[9]

[IO]

[II]

[12]

[13]

[14]

[lS]

[16]

A.A. Alzaid et al. /Diabetes

Bennett, P.H. and Knowlet, WC. (1989) Albuminuria in Type 2 (non-insulin-dependent) diabetes mellitus and impaired glucose tolerance in Pima Indians. Diabetologia 32, 870-876. Haffner, S.M., Mitchell, B.D., Pugh, J.A. et al. (1989) Proteinuria in Mexican Americans and Non-Hispanic whites with NIDDM. Diabetes Care 12, 530-536. Fatani, H.H., Mira, S.A. and El-Zubair, A.G. (1987) The prevalence of diabetes mellitus in urban Saudi Arabia. Diabetes Care 10, 180-183. Abu-Zeid, H.A.H. and Al-Kassab, A.K. (1992) Prevalence and health care features of hyperglycemia in semiurban-rural communities in Southern Saudi Arabia. Diabetes Care 15, 484-489. Lloyd, D.R., Hindle, E.J., Marples, J. and Gatt, J.A. (1987) Urinary albumin measurement by Immunoturbidity. Ann. Clin. B&hem. 24, 209-210. El-Ha&, M.A.F., Al-Faleh, F.Z., Al-Mofleh, LA., Warsy, A.S. and Al-Askah, A.K. (1982) Establishment of normal ‘reference’ ranges of biochemical parameters for healthy Saudi Arabs. Trop. Geogr. Med. 34 323-333. Gatling, W., Knight, C., Mullee, M.A. and Hill, R.D. (1988) Microalbuminuria in diabetes: a population study of the prevalence and an assessment of three screening tests. Diabetic Med. 5, 343-347. Marshall, S.M. and Albcrti, K.G.M.M. (1989) Comparison of the prevalence and associated features of albnormal albumin excretion in insulin-dependent and non-insulin-dependent diabetes. Q. J. Med. 70, 61-71. Klein, R., Klein, B.E.K. and Moss, S.E. (1993) Prevalence of microalbuminuria in older-onset diabetes. Diabetes Care 16, 1325- 1330. Garancini, P., Gallius, G., Calori, B., Micossi, P. and Poxxa, G. (1988) Microalbuminuria and its associated risk factors in a representative sample of Italian Type II diabetics. J. Diabet. Compl. 2, 12-15. Harris, M.I., Klein, R., Welbom, T.A. and Knuiman, M.W. (1992) Onset of NIDDM occurs at least 4-7 year before clinical diagnosis. Diabetes Care 15, 815-819.

Rex Clin. Pratt. 26 (1994) 115-120

1171 Kingston, M. and Skoog, WC. (1986) Diabetes in Saudi Arabia. Saudi Med. J. 7, 130-142. [IS] Melton, L.J., Ckhi, J.W., Palumbo, P.J. and Chu, C.P. (1984) Referral bias in diabetes research. Diabetes Care 7, 12-18. [I91 Edwards, O.M., Bayliss, R.I.S. and Millin, S. (1969) Urinary creatinine excretion as an index of the completeness of 24-h urine collections. Lancet ii, 1165- 1166. [20] The Diabetes Control and Complications Trial Research Group (1993) The effect of intensive treatment of diabetes on the development and progression of longterm complications in insulin-dependent diabetes mellitus. N. Engl. J. Med. 329, 987-997. [2 I] Ballard, D.J., Humphrey, L.L., Melton, L.J. et al. (1988) Epidemiology of persistent proteinuria in type II diabetes mellitus. Diabetes 37, 405-412. [22] McKenna, M.J., Arias, A., Feldkamp, C.S. and Whitehouse, F.W. (1991) Microalbuminuria in clinical practice. Arch. Intern. Med. 151, 1745-1747. [23] Allawi, I. and Jarrett, R.J. (1990) Microalbuminuria and cardiovascular risk factors in Type 2 diabetes mellitus. Diabetic Med. 7, 115-118. [24] Haffner, S.M., Morales, P.A., Gruber, M.K., Haxuda, H.P. and Stem, M.P. (1993) Cardiovascular risk factors in non-insulin-dependent diabetic subjects with microalbuminuria. Arterioscler. Thromb. 13, 205-210. [25] Mattock, M.B., Keen, H. and Viberti, G.C. (1988) Coronary heart disease and urinary albumin excretion rate in Type 2 (non-insulin-dependent) diabetic patients. Diabetologia 31, 82-87. [26] Jerums, G., Cooper, M.E., Seeman, E., Murray, R.M. and McNeil, J.J. (1988) Comparison of early renal dysfunction in type 1 and type 2 diabetes: differing associations with blood pressure and glycemic control. Diabetes Res. Clin. Pratt. 4, 133-141. [27] Seghieri, G., Alviggi, L., Caselli, P. et al. (1990) Serum lipids and lipoproteins in Type 2 diabetic patients with persistent microalbuminuria. Diabetic Med. 7, 810-814.