ORIGINAL ARTICLE

Insulin Resistance as One of Indicators for Metabolic Syndrome and Its Associated Factors in Indonesian Elderly Arya G. Roosheroe, Siti Setiati, Rahmi Istanti Department of Internal Medicine, Faculty of Medicine, University of Indonesia – Cipto Mangunkusumo Hospital. Jl. Diponegoro no. 71, Jakarta Pusat 10430, Indonesia. Correspondence mail: [email protected] ABSTRAK Tujuan: mendapatkan faktor-faktor yang berhubungan dengan resistensi insulin pada usia lanjut di Indonesia. Metode: penelitian dengan disain cross sectional dilakukan di Poliklinik Usia Lanjut RSCM Jakarta dengan jumlah sampel 172 usia lanjut. Data yang dikumpulkan meliputi karakteristik subyek (usia, jenis kelamin), indeks massa tubuh, lemak tubuh total, lemak subkutan perifer, lemak subkutan trunkal, lingkar pinggang, asupan karbohidrat dan serat, aktivitas fisik, dan konsentrasi 25(OH)D. Besar sampel dihitung dengan rumus besar sample untuk uji hipotesis beda 2 proporsi dan untuk uji hipotesis beda rerata pada 2 kelompok independent. Tingkat kepercayaan yang digunakan 95% dan kekuatan uji 80%. Analisis chisquare dan t-test independent digunakan sebagai analisis bivariat. Analisis statistik regresi logistik digunakan untuk melihat variabel yang paling mempengaruhi resistensi insulin. Batas kemaknaan yang digunakan adalah p 100% needs)

84 (48.84)

Fiber intake -- Less ( 100% needs)

14 (8.14)

Insulin resistance -- No

129 (75.00)

-- Yes

43 (25.00)

Mean values of some numerical variables are shown on Table 2. Results of bivariate analysis using Chi-square test showed the association between insulin resistance and the affecting factors as shown on Table 3. We found that the variables such as age (OR 0.44; 95% CI 0.22-0.89) and body mass index (OR 4.58; 95% CI 2.08-10.09) were significantly associated to insulin resistance. The results of mean difference analysis using unpaired t-test, which demonstrated the mean differences of numeric variables between subjects with insulin resistance and those without insulin resistance, are shown on Table 4. We found significantly higher mean values (p0.05).

201

Arya G. Roosheroe

Acta Med Indones-Indones J Intern Med

Table 2. Mean values of body mass index, fat mass, fat thickness, waist circumference, physical activity, intake, and HOMA-IR Variables

Mean (sd)

Min

Max

Body mass index

23.09 (0.19)

12.46

39.42

Fat mass (%)

29.61 (10.06)

3.00

65.8

Fat mass (Kg)

17.16(1.17)

0.39

84.00

Fat thickness beneath triceps skin

19.94 (5.45)

4.00

36.00

Fat thickness beneath thigh skin

17.29 (4.35)

6.00

30.00

Fat thickness beneath subscapular skin

17.45 (5.34)

5.00

31.50

Fat thickness beneath suprailiac skin

20.98 (5.63)

5.00

35.00

Fat thickness beneath abdominal skin

26.00 (5.99)

7.50

45.00

Peripheral subcutaneous fat thickness

37,24 (9.54)

10.00

65.00

Trunk subcutaneous fat thickness

64.43 (15,61)

17.50

107.00

Waist circumference

84.59 (11.22)

57.5

118.00

4.55 (1.78)

-0.40

8.70

Carbohydrate intake

237.75 (74.81)

7.14

435.13

Fiber intake

21.92 (33.89)

2.17

275.14

1.32 (2.53)

0.26

28.50

Physical activity

HOMA-IR

Table 3. Association between age, body mass index, 25(OH)D concentration, carbohydrate intake, fiber intake and insulin resistance Variables

Insulin Resistance

OR (95% CI)

p

74 (57.36)

0.44

0.024

27 (62.79)

55 (42.64)

(0.22 – 0.89)

-- Overweight-obesity

33 (76.74)

54 (41.86)

4.58

-- Underweight -normal

10 (23.26)

75 (58.14)

(2.08 – 10.09)

39 (90.69)

114 (88.37)

1.28

4 (9.30)

15 (11.63)

(0.40 – 4.09)

-- >100% needs

20 (46.51)

64 (49.61)

0.78

-- 70 years

16 (37.21)

-- 60-70 years

Age

Body mass index 0.000

25(OH)D concentration -- Deficiency (50)

0.67

Carbohydrate intake 0.73

Fiber intake

The results of logistic regression analysis (Table 5) indicated that only the variables of trunk and peripheral subcutaneous fat were significantly affecting insulin resistance in elderly, with OR 1.09 (95% CI 1.05-1.15) and OR 0.93 (95% CI 0.87-0.99), respectively. DISCUSSION

In our study, the number of elderly subjects with nutritional status of overweight and obesity

202

0.34

were more prevalent than underweight subjects. The phenomena might be caused by nutritional transition in Indonesia. As a developing country, Indonesia has quite dramatic changes in agepopulation structures, accompanied with changes in dietary style, habit and physical activity. As a result, this will cause increased risk of obesity and chronic disease. Altered dietary habit includes diet changes from high fiber and low fat diet to high fat, high carbohydrate and low fiber diet.

Vol 44 • Number 3 • July 2012

Insulin Resistance as One of Indicators for Metabolic Syndrome

Table 4. Mean differences on body fat and physical activity of subjects with insulin resistance and subjects without insulin resistance Variables

Insulin Resistance

p value

Yes (Mean (sd))

No (Mean (sd))

Fat mass (kg)

20.89 (4.57)

15.99 (0.99)

0.007

Fat mass percentage (%)

33.50 (11.91)

28.32 (9.05)

0.011

Waist circumference (cm)

91.30 (10.59)

82.36 (10.55)

0.000

Peripheral subcutaneous fat thickness (mm)

40.15 (10.07)

36.27 (9.19)

0.029

Trunk subcutaneous fat thickness (mm)

72.94 (14.45)

61.59 (14.99)

0.00

Physical activity (metz)

4.55 (1.66)

4.55 (1.83)

0.988

Table 5. Results of logistic regression on insulin resistance and independent variables Variables

OR

95% CI

P-value

Trunk subcutaneous fat

1.09

1.05 – 1.15

0.000

Peripheral subcutaneous fat

0.93

0.87 – 0.99

0.048

Moreover, we also found that the majority of subjects (48.84%) consumed carbohydrates more than 100% nutritional requirements and 83.7% subjects had low fiber intake, 80% less than required. Total body fat, subcutaneous fat in triceps, subscapular, suprailiac, abdomen and peripheral fat found in our study is almost similar to those values reported by Dwimartutie, in her study of 55 outpatient elderly. However, BMI, abdominal and trunk subcutaneous fat found in this study is much higher. 11 HOMA-IR concentration value in our study was not normal, therefore, a transformation was conducted. We found that the mean value of HOMA-IR was 1.32. This value is lower than those values reported by Motta, which were 1.66 for 1549 elderly, and Kurniadhi, who reported the value of 1.71 for 110 elderly subjects.12-14 The condition of insulin resistance in our study was diagnosed based on HOMA-IR >75 percentile study population, and the value was

2.67. This value is comparable to the value found by Nasution in her study with 92 elderly women in nursing homes in Jakarta.15 In addition, another study reported lower HOMA-IR value for insulin resistance, which was 2.48.19 A study conducted by Lee, et al with 976 subjects aged 30-79 years old in Korea reported a comparable HOMA-IR cut-off limit for insulin resistance, which was 2.34.14 The differences in HOMA-IR cut off limit to diagnose insulin resistance, which may be different among various populations might be affected by sex, differences in body fat distribution, age, and race (ethnicity). The prevalence of insulin resistance in our study was 25%. A study conducted by Nasution also found similar prevalence of insulin resistance (25%).15 The prevalence of insulin resistance was increasing with age.15,16 The increase of prevalence reached its peak by the age of 80 years and followed by decreased prevalence.16 Botnia et al. found increased insulin resistance in older subjects.22 It has been reported that insulin resistance increased with age. The results of bivariate analysis in our study showed significant association between age and insulin resistance (p0.05). The absent of such difference may occur since the measurement method, six minute walking test, may not represent the actual activity of elderly. Exercise in elderly will increase glucose metabolism and prevents insulin resistance; however, the mechanism is still unclear. In elderly, aerobic exercise has been known to increase functional capacity and decrease diabetes risk. In addition, elderly individuals are able to adapt to exercise increment, which may lead to improved insulin performance.3 It has been said that the risk of diabetes and insulin resistance increased with higher body fat, which is calculated by using the body mass index. Although the correlation is associated with the measurement of total body fat tissue, which is measured by BMI, but some studies also show that not all fat tissue contributes equally in increased diabetes risk. Central fat depot, i.e. intra-abdominal or visceral fat, including mesenteric and intra-abdominal omentum, has greater association with insulin resistance compared to peripheral fat depot, such as gluteal or subcutaneous fat or total body fat.30 The results of bivariate analysis found that there were significant mean differences in all body fat variable (p