OBJECTIVE â To investigate whether microalbuminuria is associated with markers of the ... Malay NIDDM subjects had an association of microalbuminuria with.
E p i d e m i o I o g y / H e a 11h N A L
Se r v ice s/ P sy ch o so cia I
R e s e a r c h
A R T I C L E
Ethnic Differences in Correlates of Microalbuminuria in NIDDM The role of the acute-phase response MARTIN B. MATTOCK, PHD A. ZAINI, MD JOHN C. PICKUP, FRCPATH
NIAZ ISLAM, MB FAHRAT KAZMI, MB GARY D. CHUSNEY, BSC
OBJECTIVE — To investigate whether microalbuminuria is associated with markers of the acute-phase response in NIDDM and whether there are ethnic differences in this association among the three main racial groups in Malaysia. RESEARCH DESIGN A N D M E T H O D S — NIDDM patients of Chinese, Indian, and Malay origin attending a diabetic clinic in Kuala Lumpur, Malaysia, were matched for age, sex, diabetes duration, and glycemic control (n = 34 in each group). Urinary albumin-to-creatinine ratio was measured in an early morning urine sample. Biochemical measurements included markers of the acute-phase response: serum sialic acid, triglyceride, and (lowered) HDL cholesterol. RESULTS — The frequency of microalbuminuria did not differ among the Chinese, Indian, and Malay patients (44, 41, and 47%, respectively). In Chinese patients, those with microalbuminuria had evidence of an augmented acute-phase response, with higher serum sialic acid and triglyceride and lower HDL cholesterol levels; and urinary albumin-to-creatinine ratio was correlated with serum sialic acid and triglyceride. The acute-phase response markers were not different in Indians, with microalbuminuria being high in even the normoalbuminuric Indians; only the mean arterial blood pressure was correlated with urinary albumin-to-creatinine ratio in the Indians. Malay NIDDM subjects had an association of microalbuminuria with acute-phase markers, but this was weaker than in the Chinese subjects. CONCLUSIONS — Microalbuminuria is associated with an acute-phase response in Chinese NIDDM patients in Malaysia, as previously found in Caucasian NIDDM subjects. Elevated urinary albumin excretion has different correlates in other racial groups, such as those originating from the Indian subcontinent. The acute-phase response may have an etiological role in microalbuminuria.
A
major reason for recent renewed interest in markers of the acute-phase response, such as the serum sialic acid and C-reactive protein concentration, is that elevated levels in the general population are powerful predictors of cardiovascular disease (1-3). We have found that the serum sialic acid concentration is strongly associated with coronary heart disease in men with NIDDM (4) and is also elevated in NIDDM and IDDM subjects with microal-
buminuria (4,5). The serum sialic acid level is further elevated in those diabetic patients with clinical proteinuria (5,6), which may reflect the increased risk of cardiovascular mortality associated with micro- and macroproteinuria. This raises the possibility that increased urinary albumin excretion may be part of, or at least associated with, the acute-phase response. In support of this, we found (7) that in Caucasian NIDDM subjects of Euro-
From the Division of Chemical Pathology (N.I., EK., G.D.C., M.B.M., J.C.P), United Medical and Dental Schools, Guy's Hospital, London, U.K.; and the Division of Endocrinology and Metabolic Medicine (A.Z.), University Hospital, Kuala Lumpur, Malaysia. Address correspondence and reprint requests to Dr. John Pickup, Division of Chemical Pathology, UMDS, Guy's Hospital, London SE1 9RT, U.K. Received for publication 29 July 1997 and accepted in revised form 14 November 1997. Abbreviations: CV, coefficient of variation.
DIABETES CARE, VOLUME 21, NUMBER 3, MARCH
1998
pean origin, high urinary albumin excretion rates and many components of metabolic syndrome X segregated in these subjects with high serum levels of markers of the acute-phase or stress response, including a 1-acid glycoprotein, C-reactive protein, serum amyloid A, sialic acid, cortisol, and interleukin-6 (the main cytokine mediator of the stress response). Little is known about ethnic differences in the association of microalbuminuria with the acute-phase response and other biochemical abnormalities. Such knowledge is of potential importance because of the marked differences in cardiovascular mortality in diabetic subjects from different countries and in diabetic patients of different ethnic origin within the same country (8). The Malaysian peninsula/Singapore is one such geographical area, where among the three main ethnic groups—Chinese, Indian, and Malay—there is a high frequency of ischemic heart disease in the diabetic and general population originating from the Indian subcontinent (9,10). Before performing a large-scale epidemiological survey of NIDDM patients in Malaysia, we sought a priori evidence for ethnic differences in the biochemical correlates of albumin excretion, particularly an association with the acute-phase response, in a matched-patient study in which the independent effect of ethnicity was evaluated by controlling for the quality of glycemic control and the characteristics of the diabetes in the three ethnic groups of NIDDM patients. RESEARCH D E S I G N A N D M E T H O D S — All NIDDM patients attending the Diabetic Clinic over a 2month period in 1995 at the University Hospital, Kuala Lumpur, Malaysia, were entered into the study. NIDDM was defined as diagnosis after the age of 30 years without insulin in the first year of treatment and without history of ketosis. From the study population, individuals were matched in 34 successive groups of three, one each of Chinese, Indian, or Malay origin, with respect to age, sex, known duration of diabetes, and glycated hemoglobin percentage. Table 1 confirms that, because of this
385
Microalbuminuria and acute-phase response
Indians with and without microalbuminuria did not differ. In the Indian subjects with normoalbuminuria, the serum sialic Indian Chinese Malay acid and triglyceride levels were elevated and the HDL cholesterol reduced to the n 34 34 34 values similar to those found in Chinese Sex (M/F) 14/20 14/20 14/20 subjects with microalbuminuria. In the Age (years) 57.2 ±9.3 56.0 ±11.0 53.0 ±10.0 Malay NIDDM subjects, variables were Diabetes duration (years) 14.0 (5-30) 12.0 (1-28) 11.5(1-25) similar, except that serum triglyceride was Glycated hemoglobin (%) 8.3 ±2.2 8.5 ±2.4 8.4 + 2.2 higher in the microalbuminuric patients. BMI (kg/m2) 24.1 ±4.4 23.1 ±2.9 26.3 ±3.5* Table 4 shows the correlation of uriData are means ± SD or median (range). *P < 0.001 vs. Chinese and P < 0.02 vs. Indian. nary albumin-to-creatinine ratio with clinical and biochemical variables. When matching procedure, the NIDDM patients quent paired comparisons between ethnic analyzed as the three ethnic groups comhad comparable clinical features and groups or between those with and without bined, urinary albumin-to-creatinine ratio glycemic control. microalbuminuria were by the t test or the was significantly related to mean arterial A venous blood sample was taken and Mann-Whitney 17 test for skewed data. The blood pressure, glycated hemoglobin, the serum stored at — 70°Cforlater analysis X2 test was used to compare the frequencies serum sialic acid, and triglyceride concenof lipids and sialic acid. Height and weight of normo-, micro-, or macroalbuminuria in tration and negatively associated with were recorded in indoor clothing without the ethnic groups. Correlations were sought serum HDL cholesterol concentration. In Chinese subjects, there were strong associshoes and BMI calculated as weight (in kilo- using Spearman's method. ations of urinary albumin-to-creatinine grams) divided by the square of height (in meters). Arterial blood pressure was meas- RESULTS — Table 2 compares the fre- ratio with sialic acid and triglyceride conured with a standard sphygmomanometer quency of normo-, micro-, and macroalbu- centration, while in the Indians, only mean after at least 10 min of rest and mean arter- minuria in the Chinese, Indian, and Malay arterial pressure approached a significant ial pressure calculated as one-third systolic NIDDM patients. There was no significant association. In the Malays, as well as glyblood pressure minus one-third diastolic difference in the percentage of diabetic cated hemoglobin, serum sialic acid and blood pressure plus diastolic blood pres- patients with these grades of albumin triglyceride concentrations were related to sure. An early morning urine sample was excretion among the ethnic groups. The the urinary albumin-to-creatinine ratio, collected, and stored at — 70°C for later median urinary albumin-to-creatinine ratio although the last two less strongly than in measurement of the urinary albumin-to- was also similar for each racial group with the Chinese patients. Diabetes duration creatinine ratio. Microalbuminuria was normoalbuminuria, as was the median also approached significance as a correlate denned as a urinary albumin-to-creatinine ratio in the microalbuminuric subjects in the Malays, as did a low BMI. Figure 1 shows scatter diagrams of the relationship ratio of 2.8-40 mg/mmol (women) and (Table 3). between serum sialic acid concentration 1.9-28 mg/mmol (men) (11). Values above Table 3 compares in the three ethnic or below these ranges were denned as groups the clinical and biochemical fea- (as an example of a marker of the acutemacroalbuminuria or normoalbuminuria, tures of those with and without microalbu- phase response) and urinary albumin-torespectively (11). minuria; macroalbuminuric subjects were creatinine ratio in the three ethnic groups. Total serum sialic acid was measured by not included in this analysis because of the a coupled enzymatic method (Boehringer small numbers affected. In the Chinese CONCLUSIONS— This study shows Mannheim, Lewes, Sussex, U.K.) adapted NIDDM subjects, indicators of the acute- that, although the prevalence of microalbuby us (12) for use on microtiter plates phase response were related to microalbu- minuria was similar in the three ethnic (interassay coefficient of variation [CV] 6%). minuria, i.e., the serum sialic acid and groups, there are ethnic differences in the Total serum cholesterol and triglyceride triglyceride concentrations were signifi- correlates of microalbuminuria in these were assayed by enzymatic methods cantly higher and the HDL cholesterol groups of NIDDM subjects in Malaysia. In (Boehringer Mannheim), using a Cobas-Bio significantly lower in those with microal- Chinese patients, there was evidence that an analyzer (Roche, Welwyn Garden City, buminuria than in those with a normal uri- acute-phase response was associated with Herts, U.K.). HDL cholesterol was meas- nary albumin excretion rate. In contrast, microalbuminuria: elevated serum sialic ured by a dextran sulphate-Mg2+ precipi- the clinical and biochemical features in acid and triglyceride and lowered HDL chotation method (13). Glycated hemoglobin was measured by an affinity chromatography method (IMX, Abbott Laboratories, Table 2—Percentage of patients in each ethnic group who were categorized as having normo-, Abbott Park, IL; interassay CV 6.5%). Uri- micro-, or macroalbuminuria nary albumin was measured by an automated immunoturbidimetric method, Ethnic group Microalbuminuria Macroalbuminuria Normoalbuminuria interassay CV
