Study design and participants
Totally eight diabetes-specific local clinics in Southern Taiwan and the diabetic department of Kaohsiung E-Da Hospital, Taiwan jointed in this cross-sectional study. A Total of 2482 older outpatient subjects with T2DM were enrolled. The inclusion criteria aged ≥65 years, clinically diagnosed with T2DM between January 2006 and October 2021, relatively healthy without acute illness and without evidence suggesting the possibility of a non-diabetic renal disease (included primary glomerular diseases, drug-induced nephropathy, reflux nephropathy, nephrolithiasis, polycystic kidney and renal-related infectious diseases). The exclusion criteria were patients: 1) aged <65 years; 2) with type 1 diabetes; 3) with history with cancer, liver or urologic diseases; 4) who had been hospitalized for any reasons within 3 months prior to enrollment; 5) with recently use of allopurinol or uricosuric agents for gouty arthritis; 6) who underwent contrast examinations during the follow-up period; and 7) who could not provide complete demographics and personal medical informatio. In addition, to avoid the potential development/presence of primary glomerular diseases, we also excluded patients with persistent hematuria with and without urinary casts.
The diagnosis of T2DM was based on the World Health Organization criteria [16]. All of the patients were followed up in accordance with the diabetes comprehensive management program suggested by the Taiwan National Health Insurance at 3-month intervals. On each follow up visiting, standardized physical examinations, biochemical measurements after fasting, measurements of urine albumin and creatinine were performed. All participants received standard treatment based on recent updated diabetes, hypertension, and dyslipidemia management guidelines. Our study protocol and procedures has been approved by E-Da Hospital Institutional Review Board with certificate number EMRP-108-111 and EMRP-109-109 and the Ethics Committees of Pingtung Christian Hospital with an approval certificate on 16 December 2005.
Key measures
IC was determined using the ICOPE (WHO) screening tools, including six functional assessments of the following five domains: locomotion, cognition, vitality, sensory (visual and sensory), and psychological symptoms [17]. If subject was unable to complete five chair rises within 14 seconds, limited locomotion mobility was defined. If the patients gave an inappropriate answer to either of two questions on orientation in time and space, or could not recall the three words they were asked to remember , impaired cognitive dysfunction was defined. If subject suffered from weight loss greater than 3 kg over 3 months or with the loss of appetite, malnutrition was defined. If subject had any eye problems such as difficulty in seeing far, reading, eye diseases, or current ophthalmic medical treatment, a visual impairment was defined. If subject failed to hear whispers in the whisper test, a hearing loss was defined. If subject bothered by feeling down, feeling depressed or hopeless, or having little interest or pleasure in doing things over the previous 2 weeks, a depressive symptoms were suggested. Finally, impairment in each item was scored as one point, and IC score was defined as the sum of the six functional assessments, with a higher score indicating greater functional impairment.
Obesity was defined according to the Ministry of Health and Welfare, Taiwan,
criteria instead of the WHO criteria, as it has been suggested that the WHO body mass index (BMI) cut-off point for obesity (≥30 kg/m2) may be too high for Asians, thereby underestimating associated health risks [18, 19]. Accordingly, we defined underweight as BMI below 18.5 kg/m2, normal weight as between 18.5≤ BMI <24 kg/m2, overweight as 24≤ BMI <27 kg/m2, mild obesity as 27≤ BMI <30 kg/m2, moderate obesity as 30≤ BMI <35 kg/m2, and severe obesity as BMI greater than 35 kg/m2 [20].
Renal function (estimated glomerular filtration rate (eGFR)) was estimated using the CKD-EPI two-concentration race equation [21]: GFR = 141 × min(Scr /κ, 1)α × max(Scr /κ, 1)-1.209 × 0.993Age × 1.018 [if female] × 1.159 [if black], where Scr is serum creatinine (mg/dL), κ is 0.7 if females and 0.9 if males, α is -0.329 if females and -0.411 if males, min indicates the minimum of Scr/κ or 1, and max indicates the maximum of Scr/κ or 1. Albuminuria was defined by the albumin-to-creatinine ratio (UACR) from spot urine. The presence of albuminuria was defined by at least two measurements of UACR >30 mg/g in a 6-month period during follow-up. The CKD risk estimation using the combination of eGFR and albuminuria categories which suggested by the KDIGO 2012 guidelines was used in our study : low risk (eGFR ≥60 mL/min/1.73 m2 and UACR <30 mg/g), moderately increased risk (eGFR >60 mL/min/1.73 m2 and 30< UACR <300 mg/g, or 45< eGFR <60 mL/min/1.73 m2 and 30< UACR <300 mg/g), high risk (30< eGFR <60 mL/min/1.73 m2 and UACR >300 mg/g, or eGFR >60 mL/min/1.73 m2 and UACR >300 mg/g), and very high risk (15< eGFR <60 mL/min/1.73 m2 and UACR >300 mg/g, or eGFR <15 mL/min/1.73 m2 and UACR >300 mg/g) [22].
Laboratory measurements
Routine tests including a clinical examination, recent medication side effect assessment, body weight, blood pressure, urinary sediment and urinalysis, complete blood count, serum chemistry, and HbA1c concentrations were performed during each regular visits. The urinary albumin concentration was measured after overnight fasting by immunoturbidimetry (Beckman Instruments, Galway, Ireland). The detection limit was 2 mg/L, with the interassay and intraassay coefficients of variance <8%. In the study admission, the patients were defined as normoalbuminuric if they had a UACR <30 mg/g in at least two consecutive overnight urine collections. If the patient had first UACR measurement >30 mg/g, a repeat urine test will be asked and checked to confirm the diagnosis of albuminuria within 3 to 6 months in the follow-up period later. If the urine specimen showed the presence of urinary infections, the specimen will not be used and a new sample was collected after antibiotics treatment. To exclude primary renal diseases, abnormal urinary sediment should not be noted in the urine specimen (presence of any protein, red blood cells, hemoglobin, white blood cells, nitrites or casts ). Serum creatinine was measured by the Jaffe method. Serum HbA1C, total cholesterol, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), triglycerides, hemoglobin, creatinine, and glucose were determined using a parallel-multichannel analyzer (Hitachi 7170A, Tokyo, Japan) by standard commercial methods after an overnight fast as in our previous report [23].
Variables
All participants completed a standard questionnaire that assessed age, gender, cigarette use, history of disease (T2DM, diabetes duration, hyperlipidemia, hypertension, heart disease, and cancer) in face-to-face interviews with trained interviewers. Subject’s blood pressure was measured by trained clinical assistants with digital automatic blood pressure monitor (model HEM-907; Omron, Omron, Japan) after resting for 5 minutes. Hypertension was defined as a systolic blood pressure (SBP) ≥140 mmHg, a diastolic blood pressure (DBP) ≥90 mmHg, or if the patient was recent using antihypertensive medication. Anthropometric parameters including BMI (kg/m2) were measured. Hyperlipidemia was defined according to the ATP III criteria as following: triglycerides ≥150 mg/dl, and/or HDL-C <35 mg/dl in men or <39 mg/dl in women, and/or total cholesterol ≥200 mg/dl, and/or LDL-C ≥130 mg/dl, or those undergoing treatment for lipid disorders.
Statistical analysis
Data normality was analyzed using the Kolmogorov-Smirnov test. Continuous, normally distributed variables are presented as mean ± SD, and non-normally distributed variables as median (interquartile range). Categorical variables are presented as frequencies and/or percentages. Baseline characteristics were compared between groups using one-way analysis of variance (ANOVA) for normally distributed variables. The chi-square test was used to compare categorical variables.
Logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the risk of CKD in each IC score, compared with an IC score of 0 as the reference. To test linear risk trends, a tertiles as a continuous variable in the regression models was used. Biochemical parameters according to IC score at baseline were tested for trends.
ORs and corresponding 95% CIs were calculated using univariate and multivariate logistic regression models to evaluate the relationships between IC scores and the risk of CKD. A p value <0.05 was considered to be statistically significant. JMP version 7.0 for Windows (SAS Institute, Cary, NC, USA) was used in our analysis.
An Excel sheet provided by Andersson and co-authors [24] was used into the database and compute the relevant indicators of interactions. Using a logistic regression model, a value was obtained and taken as the estimated additive interaction between IC score and obesity status. The interaction based on the additive model was determined using the following indexes: the relative excess risk of interaction (RERI), attributable proportion of interaction (API), synergy index (SI), measure of multiplicative interaction for risk ratios [25] and their 95% CIs using the delta method [26]. The RERI refers to the excess risk due to the interaction relative to the risk without exposure. The API is the attributable proportion of disease caused by the interaction in subjects with both exposures. The SI refers to the excess risk from both exposures when there is a biological interaction due to the risk from both exposures without interaction. The RERI has been showed to be the best measure of interaction using a proportional hazards model [27]. If the RERI and AP are equal to 0, absence of additive interactions was defined [28]. Finally, an indicative biological interaction is considered when RERI >0, AP >0, S >1, or a measure of multiplicative interaction for risk ratios >1.