Prevalence of Early Stages of Chronic Kidney Disease in Healthy Army Personnel

Abstract

Introduction

Chronic kidney disease (CKD) is associated with significant morbidity and mortality. Screening and detection of early stages of CKD can help institute interventions that may delay the progression of the disease. One aim was to study the prevalence of early stages of CKD in the Army.

Methods

A cross-sectional study of Army Personnel in an Army cantt in Central India was carried out. All participants filled a structured questionnaire and anthropometric data was collected. Investigative profile included routine urine exam, semi-quantitative microalbuminuria (MAU), serum creatinine, lipid profile and fasting blood glucose. Glomerular Filteration rate (eGFR) was calculated using the Modification of Diet in Renal Diseases (MDRD) study equation.

Result

A total of 1920 subjects were examined with 731 (38.07%) from Arms and 1189 (61.93%) from Services. 348 were excluded and of the remaining 1572 subjects, 141 (8.97 %) had MAU and 157 (9.99 %) had deranged Albumin Creatinine Ratio (ACR). Mean eGFR by MDRD equation was 102 ± 25.84 ml/min/1.73m². Early CKD was seen in 150 (9.54 %) with 84 (5.34 %) in stage I CKD, 55 (3.5%) in stage II and 11 (0.7%) in stage III. Multiple logistic regression showed BMI > 23, the presence of DM and HTN were independent risk factors for CKD.

Conclusion

9.54% of healthy army personnel were found to have early stages of CKD. Institution of screening programs can result in early detection of CKD.

Introduction

Changing demographics, increasing affluence and sedentary lifestyles have led to increasing prevalence of non-communicable diseases like diabetes mellitus (DM), obesity, hypertension (HTN), cardiovascular disease (CYD) and chronic kidney disease (CKD) even in developing countries like India. It is estimated that 4 out of 5 chronic disease deaths now occur in low and middle income countries [1]. The actual prevalence is likely to be much more as this represents only hospital admission data. CKD is important among this group because, apart from its own morbidity, mortality and high risk for progression to End Stage Renal Disease (ESRD) requiring expensive renal replacement therapy (RRT), it is also the most important independent risk factor for CVD [2]. In fact, of all the known risk factors for CVD, CKD poses the biggest threat with CKD patients being up to 100 times more vulnerable to CV events [3]. The seventh report of the Joint National Committee on prevention, detection, evaluation and treatment of high blood pressure (JNC-7) guidelines recommend that Glomerular Filtration Rate (GFR) below 60 ml/min or the presence of microalbuminuria be considered cardiovascular disease equivalents [4]. Data from the United States population studies like the Third National Health and Nutrition Examination Survey (NHANES III) and the Kidney Early Evaluation Program (KEEP) shows that 11 to 15.6% of adult US population has chronic kidney disease [5, 6]. However, there is limited data from India and none from the Armed forces on the prevalence of CKD. Based on reports from tertiary care centres, prevalence of ESRD and CKD in India is estimated to be 100 per million population (PMP) and 0.79 to 1.39% respectively [7]. However, a recent study on the prevalence of low GFR in a North Indian population has shown that 4.2% of the population had GFR<60 ml/min/1.73m² [8] which is equivalent to stage III CKD and above, thereby suggesting that the actual prevalence of CKD may be much higher than previously estimated. Studies have shown that if detected early, then control of glycaemia [9], blood pressure [10, 11] and diet can reduce albumin excretion and reduce risk of progression to CKD. Unfortunately, most of CKD patients are referred to a physician/nephrologist only at a time when there is manifest CV disease or dialysis requirement. Armed forces personnel are authorised comprehensive free medical care for life and the cost of renal replacement therapy (RRT) forms a significant chunk of the expenditure of Armed Forces Medical Services. Therefore, screening for early stages of CKD may prove to be cost effective in the long run considering the high cost of RRT. The aim of the present study was to screen apparently healthy army personnel for early stages of CKD.

Materials and Methods

Study Design: This cross-sectional study was carried out in an Army cantt in Agra city from Feb 2009 to Aug 2009. All service personnel in the station were surveyed as part of a “Comprehensive Health Survey for Detection of Life Style Diseases” at the local Military Hospital. The entire posted strength of two units of Arms and three units of Services was incorporated into the study. The Hospital Ethical Committee approved the Study protocol.

The study population included all healthy adults above 20 years of age. Exclusion criteria were self-reported or documented cases of DM, HTN, CVD and renal disease. The participants were asked to report after overnight fast to the hospital where after obtaining informed consent, they were administered a pre-structured, standardized questionnaire that covered demographic data, risk factors, past, present, personal, family and medical history and lifestyle habits. Anthropometric parameters were assessed using standardized techniques. Body weight and height were measured in light clothes without footwear on a dedicated calibrated weighing scale and stand alone stadiometer. For waist and hip circumference, a standardized clinician's tape measure was used to measure the widest part of the hips and then the narrowest part of the waist at or above the umbilicus. Every participant underwent sitting right arm blood pressure measurement after 10-15 min rest by two separate nurses at an interval of atleast 10 minutes apart and the mean of two measurements was used for all analysis. All blood pressure readings were taken using two dedicated calibrated mercury sphygmomanometers. A spot, early morning, clean-catch, mid-stream urine sample was obtained from the patients. Assessment for proteinuria, haematuria, and leucocyturia was done using dipsticks (M10SG Multisticks®, Seimens® Corp, India) and the results were read on Siemens Clinitek® 50 semiautomated urine analyzer (Seimens Corp, India). Semi-quantitative MAU and albumin: creatinine ratio was obtained using Clinitek® Microalbumin strips read on a Siemens Clinitek® 50 semiautomated urine analyzer (Seimens® Corp, India).

A 5 ml sample of fasting venous blood was taken for assessing biochemical variables which included creatinine, lipid profile and Fasting Plasma Glucose (FPG). Serum creatinine (S.cr), total cholesterol (TC), HDL cholesterol, triglycerides (TG) and FPG were measured on Erba Chem 5 biochemistry analyzer (Transasia® Corp, India) using standardized kits from Seimens Diagnostics, India. Serum Creatinine was estimated by modified Jaffe's kinetic assay (Seimens® Corp, India). Control samples were included in each run and daily quality checks were carried out for all analytes. All samples were tested on the day of collection in a single lab for the entire duration of the study.

Assessment Criteria

Obesity

BMI was calculated as weight in kilograms divided by height in meters squared. WHR was calculated as waist circumference divided by hip circumference in cms. Overweight and obesity on BMI were defined as >23 and >25 respectively. Truncal obesity by WHR was defined as >0.9 for men and >0.8 for women. These definitions were based on Association of Physicians of India consensus statement for obesity for Asian Indian population [12].

Hypertension

Hypertension was defined as per JNC-7 guidelines [4]. Systolic blood pressure (SBP) ≥ 140 mmHg or diastolic blood pressure (DBP) ≥ 90 mmHg was considered HTN.

Diabetes Mellitus

Subjects with FPG>126 mg/dL were considered to have DM [13].

Chronic Kidney Disease

GFR was calculated using Modification of Diet in Renal Diseases (MDRD) Study equation (186 × (serum creatinine)-^1.154 × (age)-^0.203x (0.742 if female) ml/min/1.73m². The estimated GFR (eGFR) was used for classification of CKD stages. CKD definitions were based on K/DOQI guidelines [14]. Microalbuminuria (MAU) was taken as indicator of renal damage in those with eGFR >60 ml/min/1.73m². As only healthy individuals were included in the study with an aim to identify early kidney disease, we classified patients from stage I to stage III. Stage I was defined as eGFR>90 ml/min/1.73m² and MAU, stage II as eGFR of 60-90 ml/min/1.73m² with MAU and stage III as eGFR of 30-59 ml/min/1.73m². Patients with stage 4/5 (ESRD) were not included in the CKD analysis.

Statistical Analysis

All data was tabulated on excel worksheets and analysed using statistical software SPSS v15.0. Chi square test was used for comparing the prevalence rates between the sub groups and for determining the relationship between various risk factors and MAU at 95% CI. Fisher exact results (2-tailed) were applied where necessary. Multiple logistic regression analysis was carried out to determine the independent association of various risk factors on MAU. A p value of < 0.05 was taken as statistically significant.

Results

A total of 1920 subjects with a mean age of 34.72 ± 7.57 were examined of which 731 (38.07%) were from Arms and 1189 (61.93%) were from services. By BMI criteria, 660 (34.38%) were found to be overweight (BMI > 23) and 317 (16.51%) were found to be obese (BMI > 30); where as 716 (37.29%) subjects had WHR > 0.9. Hypertension was seen in 163 (8.49%) subjects and 21 subjects (1.09%) were found to be diabetic. For further analysis of evidence of CKD, 76 subjects were excluded due to pre-existing disease and 272 patients could not undergo test for MAU due to logistic reasons and were thus not included. A total of 1572 subjects were thereby evaluated for CKD of which 141 (8.97%) subjects had MAU and 157 (9.99%) had deranged ACR. Mean eGFR was 102 ± 25.84 ml/min/1.73m². Early CKD was seen in 150 (9.54%) subjects, of which stage I CKD were 84 (5.34%), stage II were 55 (3.5%) and stage III were 11 (0.7%). The results of general characteristics, MAU and CKD stages for the sub groups of arms and services is summarised in Table 1.

Table 1.

Comparison of studied characteristics between subgroups of Arms and Services

General characteristics
Characteristic studied	Arms (n=731)	Services (n=1189)	Total (n=1920)
Age (mean ± SD)	31.51 ± 5.91	36.70 ± 7.8	34.72 ± 7.57
Overweight (BMI >23)	178 (24.35%)	482 (40.54%)	660 (34.38%)
Obesity (BMI >25)	95 (13.00%)	222 (18.67%)	317 (16.51%)
Abnormal WHR	197 (26.95%)	519 (43.65%)	716 (37.29%)
Hypertension	39 (5.34%)	124 (10.43%)	163 (8.49%)
Diabetes (FPG >126mg/dL)	7 (0.96%)	14 (1.18%)	21 (1.09%)
Hypercholesterolemia (Total S. Cholesterol >200mg/dl)	75 (10.26%)	114 (9.59%)	189 (9.84%)
Evaluation for CKD
Characteristic studied	Arms (n=555)	Services (n=1017)	Total (n=1572)
Microalbuminuria (>30 mg/L)	41 (7.39%)	100 (9.83%)	141 (8.97%)
Albumin: creatinine ratio (30-300 mg/g)	46 (8.29%)	113 (11.11%)	157 (9.99%)
GFR (Mean ± SD)	102 ± 26.03	103 ± 25.74	102 ± 25.84
CKD Stages	66 (8.28%)	104 (10.12%)	150 (9.54%)
a) Stage I	21 (3.78%)	63 (6.09%)	84 (5.34%)
b) Stage II	20 (3.6%)	35 (3.44%)	55 (3.5%)
c) Stage III	5 (0.9%)	6 (0.59%)	11 (0.7%)

Risk factors for microalbuminuria (and CKD by implication) were evaluated for the entire population as well as the sub groups of arms and services using Chi square test. The results are shown in Table 2.

Table 2.

Evaluation of the risk factors for microalbuminuria (MAU) in sub groups of Arms and Services

Risk factor	Arms (n=555)		p value	Services (n=1017)		p value	Total (n=1572)		p value
	MAU +*	MAU -		MAU +	MAU -		MAU+	MAU-
BMI >23	22	201	NS**	69	550	NS	91	751	< 0.01
Abnormal WHR	14	153	NS	56	402	<0.05	70	555	< 0.05
Hypertension	4	29	NS	20	92	<0.005	24	121	< 0.001
Diabetes	1	3	NS	6	6	<0.001***	7	9	< 0.001***
Hypercholesterolemia	3	60	NS	15	83	<0.001	18	143	NS

^*MAU >30 mg/L taken as MAU +,

^**NS – Not significant (p >0.05),

^***Fisher exact results applied

To evaluate the independent association of the risk factors of CKD, multiple logistic regression was carried out for the total population as well as the sub groups. A backward stepwise multiple logistic regression analysis with backward selection of factors that might independently be associated with development of microalbuminuria was performed on a number of predictors. The full model included BMI > 23, WHR > 0.9, TC > 200 and presence of DM, HTN. The final regression model and results is given at Table 3. In the sub group of Arms, no significant association could be found for any of the factors with MAU. In the sub group of services, a significant association was found with HTN (OR=2.053; 95% CI 1.187-3.552; p=0.010) and DM (OR=8.451; 95% CI 2.622-27.235; p=0.0001). For the total population, a significant association was found with BMI (OR=1.54; 95% CI 1.067-2.223; p=0.021), DM (OR=7.403; 95% CI 2.665-20.564; p=0.0001) and HTN (OR=l .914; 95% CI 1.172-3.126; p=0.010).

Table 3.

Multivariate Analysis of Predictors for MAU using stepwise Logistic Regression

Variable	ß coefficient	Odds ratio (OR)	95% CI for OR		Sig (p)
Arms*
BMI >23	0.561	1.752	0.898	3.418	0.100
WHR >0.9	0.009	1.009	0.499	2.038	0.981
DM	1.511	4.529	0.435	47.145	0.206
HTN	0.559	1.749	0.558	5.484	0.338
TC >200	− 0.694	0.500	0.146	1.715	0.270
Services**
HTN	0.720	2.053	1.187	3.552	0.010
DM	2.134	8.451	2.622	27.235	0.000
Total Population***
BMI >23	0.492	1.540	1.067	2.223	0.021
HTN	0.649	1.914	1.172	3.126	0.010
DM	2.002	7.403	2.665	20.564	0.000

^*Backward stepwise analysis did not change the results as none of the predictors were significantly associated with MAU.

^**BMI, WHR and TC were found to be not significantly associated with MAU and were removed by the model from the equation (based on the LHR test).

^***WHR and TC were found to be not significantly associated with MAU and were removed by the model from the equation (based on the LHR test).

Discussion

In the absence of a renal registry or screening programs there is no data on the real burden of CKD in India. Current estimates of burden of CKD in India based on data from tertiary care centres are 0.79 to 1.39% [7]. However, studies from other parts of the world have shown much higher prevalence. The NHANES shows that 13.07% of US population has CKD [15]. Tillin et al [16] found 8.7% MAU in ethnically South-Asian population of Britain who were predominantly first generation migrants of Indian origin and in Singapore, Nang et al [17] found 27% CKD in ethnic Indians. A community based study around Delhi using eGFR by MDRD and CG equations found at least 4.2% of population had low eGFR (<60 ml/min/1.73m²) [8] equivalent to stage III or greater CKD, thereby suggesting that the actual prevalence of CKD may be much higher. Our study shows that 9.54% of apparently healthy army personnel have early stages of CKD. Although our sample is highly selective and not representative of the Indian population, it does show that burden of CKD even in a low risk healthy population who are under regular preventive medical surveillance is much higher than previous estimates. We have excluded high risk groups like known DM and HTN in our sample and hence the actual prevalence is likely to be much higher even in the army. There is, therefore, a definitive need for further community based studies on the prevalence of CKD both in the army and in the general population in India.

In India, the cost of RRT is prohibitive and was estimated in 2002 to be around Rs. 4.25 lakhs (USD 8500) per year per patient for haemodialysis and Rs. 4.95 lakhs (USD 9900) per year per patient for continuous ambulatory peritoneal dialysis [18]. Although the cost of RRT has reduced to about a third since then, it still remains very expensive. The army provides comprehensive free lifelong medical facilities to its personnel and there are currently approx 300 serving pers on RRT in the army which costs approx Rs. 4.46 crores per annum on RRT for these patients. This does not include the cost due to other medical expenses, complications, co-morbidities and loss of manhours/productivity of the individual. Screening for CKD is increasingly being implemented in the developed countries. While screening for CKD is accepted practice in high risk groups like those with HTN or DM [19, 20], generalised population based screening is also gaining advocacy. Recent evidence from a survey in Norway on 65604 people indicates that a high-risk screening model would identify less than half of those with CKD [21]. Furthermore, it is known from several epidemiologic studies that for every patient with known hypertension or diabetes, there is one individual in the population for whom this diagnosis is not yet made but who already can have considerable associated end-organ damage [22, 23, 24, 25]. Patients with diabetes or hypertension are already on medication and given advice on measures that can reduce the risk of progression to ESRD. Consequently, the number of individuals who are identified by targeted screening and for whom such screening results in a change of medical treatment may be limited. Atthobari et al [25] showed that screening of an adult population for MAU and subsequent treatment of individuals with positive screening results was cost-effective when calculated to prevent cardiovascular end points. Klebe et al [26] calculated that the costs of implementing guidelines for CKD in the United Kingdom could be recouped by delaying dialysis requirement by just one year for one individual per 10,000 patients treated. Palmer et al [27] calculated that the earlier the individual at risk is detected and preventive treatment started, the more cost-effective such treatment will be. The army already has in place annual and periodic medical examinations where most parameters required for such screening are part of the protocol. Addition of a single test for MAU and use of a GFR estimating equation can help in screening for early stages of CKD also. The cost of the semi-quantitative test for MAU used in this study is approx Rs. 80 per patient. Therefore, instituting screening for CKD in the armed forces can be implemented easily by dovetailing it with the current medical examination guidelines and can prove to be cost effective in the end.

We have used a single spot semi-quantitative method for determination of MAU. However, ideally two out of three samples in a three month period should be positive to establish chronicity for a diagnosis of CKD as per KDOQI guidelines [14]. Studies have shown that spot urine albumin concentration and ACR as screening tests are appropriate for application in Indo-Asian population [28] and has been routinely used in epidemiological studies including the initial NHANES cohort and the Singapore Prospective Study Programme [17]. While quantitative estimation of MAU is more accurate, semi-quantitative tests are equally good and more cost effective for screening programs in differentiating between normo-albuminuria, microalbuminuria and macroalbuminiuria. The Clinitek® Microalbumin system that was used in this study provides a reliable means to screen for MAU. When considered as a two-class test for albumin with a cut off of 30 mg/L the Clinitek system gave a sensitivity of 95.4% with specificity of 78.9% and a positive predictive value of 87.4% [29].

In our study, we found that 16.51 and 37.29% of subjects have generalised obesity and abdominal obesity respectively. Deepa et al [30] found 43.2 and 35.1% generalised obesity and abdominal obesity respectively in an urban South Indian population. Similarly, Prabhakaran et al [31] found 35 and 43% with generalised obesity and abdominal obesity in an industrial population. These studies used higher BMI cut offs of 25 and 30 as compared to the much more stringent criteria of BMI of 23 and 25 that has been used in this study while the WHR and waist circumference criteria are similar to those used in our study. While prevalence of obesity/overweight even by these stringent BMI criteria is much lower in our study as is to be expected from a healthy military population, abdominal obesity by WHR is significant at 37.29% and similar to the 35-43% in the general population. This highlights, therefore, that while the stress on “ideal weight for age” in the army has resulted in better control of BMI, abdominal obesity and its associated risk is still prevalent.

Examination of various risk factors for CKD and MAU showed positive correlations for BMI > 23, HTN and FPG >126mg/dL. Singh et al [8] have observed positive association with BMI and renal impairment. Sanches et al [32] have shown correlation with WC and WHR with reduced GFR. Our study has shown similar correlations of these factors with CKD. DM is a known risk factor for MAU and CKD and hence a positive correlation is expected.

In conclusion our study shows 9.54% of apparently healthy army personnel have early stages of CKD and 37.29% have abdominal obesity. Use of a simple semi-quantitative test for microalbuminuria dovetailed into the existing periodic medical examinations may help in identification of these cases. Institution of medical and lifestyle measures to prevent progression to ESRD can prove cost effective in the long run. Presence of DM, HTN and BMI > 23 were found to be independent risk factors for CKD.

Conflicts of Interest

None identified.

Intellectual Contribution of Authors

Study Concept: Maj Gen PP Varma, sm

Drafting & Manuscript Revision: Lt Col DK Raman, Maj Gen PP Varma, sm

Statistical Analysis: Lt Col TS Ramakrishnan

Study Supervision: Maj Gen PP Varma, SM, Lt Col DK Raman, Lt Col Pragnya Singh