Survival Advantage of Normal Weight in Peritoneal Dialysis Patients

Abstract

Introduction: A high body mass index (BMI) is a positive predictor of outcome in hemodialysis. But reports for peritoneal dialysis (PD) have been less numerous. The aim of the present investigation was to study the association between BMI and survival among PD patients and to discuss the main risk factors affecting survival. Materials and methods: A total of 159 patients who received PD from 1 January 2006 to 31 December 2010 at the Department of Nephrology in the Third Affiliated Hospital of Soochow University were enrolled in the study. Blood samples and baseline characteristics of the study cohort were obtained at the start of PD. Patient survival status was recorded through 31 December 2010. Results: Patients were stratified into two groups as normal weight (BMI, 18.5–24.9) and overweight (BMI, 25.0–29.9). Kaplan–Meier survival curve revealed that the normal weight patients had survival advantage over overweight patients (p < 0.01, by log-rank test). Cox proportional hazard models revealed that BMI, age, diabetes mellitus, coronary vascular disease, congestive heart failure and lipoprotein(a) (Lp(a)) were significant risk factors associated with all-cause mortality (p < 0.05). After adjustment for these covariates, survival was consistently higher for normal weight patients (p < 0.01). Furthermore, the study demonstrated that normal weight patients had lower serum Lp(a) (p < 0.05), C-reactive protein (CRP) (p < 0.05), and peritonitis rate (p < 0.05) compared with overweight patients. Conclusion: The results indicated that normal BMI at the commencement of PD had significant survival advantage in our study. The mechanisms for this might be related to lower cardiovascular risk, less chronic inflammation, and peritonitis prevalence.

• body mass index
• peritoneal dialysis
• survival
• risk factors

INTRODUCTION

In the general population, overweight or obesity is a significant risk factor for morbidity and mortality.¹ Being overweight or obese is associated with all traditional cardiovascular risk factors and with nearly all inflammatory and lipid biomarkers.² However, in people with various conditions, overweight may be associated with higher survival rate. Several studies have identified a high body mass index (BMI) as a strong positive predictor of outcome in hemodialysis (HD) patients.^3–5 There are also several studies which have analyzed the association between BMI and survival in peritoneal dialysis (PD) patients from different perspectives and using different patient populations.^6–9 But data on the influence of BMI on outcome in PD patients are more controversial. And the mechanisms by which being overweight may confer a survival advantage in dialysis patients are presently uncertain. It is well known that the common causes of death in PD patients are cardiovascular disease,¹⁰ chronic inflammation,¹¹ malnutrition,¹² and peritonitis.¹³ The aim of this research was to study the association between BMI and mortality and to evaluate these risk factors in the PD population according to BMI.

METHODS

Subjects

The cohort study was conducted at the department of Nephrology, the Third Affiliated Hospital of Soochow University, Changzhou, China. All patients who received PD at least 1 month from 1 January 2006 to 31 December 2010 were enrolled in the study.

Measurements

BMI was calculated as a patient’s dry weight (empty of dialysis fluid) in kilograms divided by the square of the

patient’s height in meters at the beginning of PD. BMI was categorized into four groups: underweight (BMI, <18.5), normal weight (BMI, 18.5–24.9), overweight (BMI, 25.0–29.9), and obese (BMI, ≥30.0).¹⁴ The baseline characteristics and blood samples were obtained before PD. Biochemical parameters of cardiovascular risk [serum total cholesterol, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, apolipoprotein A-I (ApoA-I), apolipoprotein B-100 (ApoB-100), lipoprotein(a) (Lp(a)), hemoglobin, calcium, phosphorus], chronic inflammation [serum C-reactive protein (CRP), fibrinogen], and nutrition (serum albumin, prealbumin) were assessed. The corrected serum calcium level was calculated as: calcium (mg/dL) + 0.8 [4.0 – albumin (g/dL)]. The rate of peritonitis was recorded. The initial event was defined as the starting of PD. Mortality during the 5 years was the end point. Patients were censored when they were transferred to HD, transplantation or lost to follow-ups. Patient survival status was complete through 31 December 2010. The study was approved by the local ethics committee.

Statistical Analysis

Analyses were conducted with the SPSS (version 13.0) (SPSS Inc., Chicago, IL, USA). All data were first checked for normality of distribution using the Kolmogorov–Smirnov test of normality. The normal distribution data were represented as mean ± standard deviation. The nonnormal distribution data were represented as median (minimum–maximum). The Kruskal–Wallis nonparametric test was used for nonnormally distributed continues variables. Student’s t test was used to test for differences between groups for normally distributed continuous variables. Chi-square test was used to evaluate difference in prevalence. Kaplan–Meier analysis was used to plot survival by categories of BMI. Differences in the survival curves were evaluated using the log-rank test. An initial univariate Cox regression analysis was performed to compare all the potential risk factors with mortality. Then the significant variables (p < 0.1) on univariate analysis were entered into multivariate Cox regression analysis. Simple linear regression analysis and multiple linear regression analysis were used to find the associations between potential risk factors and BMI. A p < 0.05 was taken as statistically significant.

RESULTS

Baseline Characteristics of Patients

A total of 159 patients were enrolled in the study. There were 4 underweight, 83 normal weights, 69 overweight, and 3 obese patients. Because of the limited number, underweight and obese patients were excluded from the study. The detailed baseline characteristics of all patients at the start of PD are shown in Table 1.

Table 1. Baseline characteristics of all patients.

Variable	Data
Age, years	51.8 ± 15.1
Sex, M/F	68/84
BMI	23.5 (18.5–29.8)
Dialysis duration, years	2.1 ± 1.1
Current smoking	19.7%
Primary cause of ESRD
Chronic glomerulonephritis	89 (58.6%)
Diabetic nephropathy	22 (14.5%)
Hypertensive nephrosclerosis	11 (7.2%)
Lupus nephritis	5 (3.3%)
Unknown	15 (9.9%)
Other	10 (6.6%)
Other comorbidities
Hypertension	98 (64.5%)
Diabetes mellitus	28 (18.4%)
Coronary vascular disease	17 (11.2%)
Congestive heart failure	16 (10.5%)
Cerebrovascular disease	11 (7.2%)
Peripheral vascular disease	2 (1.3%)
Arrhythmia	13 (8.6%)
Chronic obstructive pulmonary disease	3 (2.0%)

Note: BMI, body mass index; ESRD, end-stage renal disease.

Figure 1. Kaplan–Meier curves of survival time for normal weight and overweight PD patients (p < 0.01, by log-rank test).

Table 2. Survival rates between normal weight and overweight groups.

BMI categories	1-year	2-year	3-year	4-year
Normal weight (%)	96.3	87.7	72.5	43.1
Overweight (%)	95.3	73.0	45.5	17.4

Note: BMI, body mass index.

Table 3. Cox regression analysis of risk factors associated with all-cause mortality.

Variables	Univariate		Multivariate
	Relative risk (95% CI)	p (<0.1 to enter)	Relative risk (95% CI)	p-Value
Age (years)	1.029 (1.011, 1.048)	0.002	1.028 (1.009, 1.048)	0.005
BMI (≥25 vs. <25)	2.482 (1.431, 4.306)	0.001	2.714 (1.494, 4.930)	0.001
Diabetes mellitus (yes)	2.295 (1.238, 4.251)	0.008	2.126 (1.131, 4.000)	0.019
Coronary vascular disease (yes)	2.492 (1.276, 4.870)	0.008	4.359 (2.079, 9.137)	<0.001
Congestive heart failure (yes)	1.984 (1.038, 3.790)	0.038	2.454 (1.247, 4.827)	0.009
LDL cholesterol, mg/dL	1.010 (1.000, 1.021)	0.056	–	–
ApoB-100, mg/dL	1.008 (1.000, 1.016)	0.047	–	–
Lp(a), mg/dL	1.102 (1.040, 1.167)	0.001	1.066 (1.004, 1.132)	0.036

Note: ApoB-100, apolipoprotein B-100; BMI, body mass index; CI, confidence interval; LDL, low-density lipoprotein; Lp(a), lipoprotein(a).

BMI and Survival

Figure 1 shows a Kaplan–Meier plot of survival time by categorization of BMI for PD patients. As shown, patients in the normal weight had better survival during the course of the study compared with overweight patients (p < 0.01, by log-rank test). The cumulative survival rates of both groups are shown in Table 2. Univariate Cox regression analysis indicated that survival was influenced by age, BMI, diabetes mellitus, coronary vascular disease, congestive heart failure, LDL cholesterol, ApoB-100, and Lp(a) (p < 0.1). However, multivariate Cox regression analysis indicated that survival was influenced by age, BMI, diabetes mellitus, coronary vascular disease, congestive heart failure, and Lp(a) (p < 0.05). The results are detailed in Table 3. After adjustment for these covariates, survival was consistently higher for normal weight patients (p < 0.01), as shown in Figure 2.

BMI and Risk Factors

The clinical and laboratory characteristics of patients in different groups are detailed in the Table 4. At baseline, no significant differences were seen between groups in age, sex, dialysis duration, and current smoking. The prevalence of cardiovascular conditions, including coronary vascular disease, congestive heart failure, cerebrovascular disease, and peripheral vascular disease had also no significant differences. There were no significant differences in the initial levels of creatinine and urea nitrogen. As expected, BMI was significant different between groups (p < 0.01).

The serum levels of Lp(a) were significantly higher in overweight patients than in normal weight patients (p = 0.032). At the same time, the levels of CRP were higher in overweight patients than in normal weight patients (p = 0.003). But the levels of albumin were significantly lower in normal weight patients compared with overweight patients (p = 0.049). The prevalence of peritonitis was obviously higher in overweight patients (p = 0.047). However, the serum levels of total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides, ApoA-I, ApoB-100, hemoglobin, corrected calcium, fibrinogen, and prealbumin were similar in normal weight and overweight patients. The prevalence of hypertension and diabetes mellitus was also comparable in the two BMI groups.

Figure 2. Adjusted Cox survival curves for normal weight and overweight PD patients (p < 0.01). All models adjusted for age, diabetes mellitus, coronary vascular disease, congestive heart failure. and Lp(a).

Simple linear regression analysis and stepwise multiple linear regression analysis found that reciprocal CRP, albumin and phosphorus were significantly associated with BMI (p < 0.05).

CONCLUSION

This study indicated that normal weight at the commencement of PD was associated with significantly better survival in our center, even after adjustment for a set of clinical and biochemical parameters. These findings were in accord with previous reports by Abbott et al.¹⁵ and McDonald et al.,¹⁶ but contrast with other results, which demonstrated either neutral^6,7 or beneficial^8,9 effects of large BMI on PD outcomes. For Asians, Johansen et al.¹⁷ did not show a positive association between BMI and survival among patients beginning dialysis. But Yen et al.¹⁸ showed a survival advantage of high BMI in Taiwanese patients undergoing maintenance HD. In this study, Cox proportional hazard models revealed that age, BMI, diabetes mellitus, coronary vascular disease, congestive heart failure, and Lp(a) were significant risk factors associated with all-cause mortality.

Table 4. Comparison of clinical and laboratory characteristics between normal weight and overweight groups.

	Normal weight	Overweight	p-Value
Number	83	69
Body mass index	21.0 (18.5–24.7)	26.5 (25.0–29.8)	0.001
Age (years)	50.6 ± 15.6	53.4 ± 14.5	0.256
Sex (M/F)	38/45	30/39	0.776
Dialysis duration (years)	2.3 ± 1.2	2.0 ± 1.0	0.094
Current smoking	19 (22.4%)	11 (16.4%)	0.361
Coronary vascular disease	10 (12.0%)	7 (10.1%)	0.711
Congestive heart failure	8 (9.6%)	8 (11.6%)	0.696
Cerebrovascular disease	5 (6.0%)	6 (8.7%)	0.527
Peripheral vascular disease	1 (1.2%)	1 (1.4%)	0.895
Arrhythmia	8 (9.6%)	5 (7.2%)	0.600
Chronic obstructive pulmonary disease	1 (1.2%)	2 (2.9%)	0.469
Creatinine, mg/dL	8.5 ± 2.3	9.1 ± 2.2	0.101
BUN, mg/dL	44.0 ± 15.1	47.1 ± 15.5	0.219
Cardiovascular risk
Total cholesterol, mg/dL	185.8 ± 48.2	191.1 ± 52.2	0.519
LDL cholesterol, mg/dL	86.3 ± 26.3	91.1 ± 29.4	0.287
HDL cholesterol, mg/dL	42.3 ± 11.2	40.1 ± 12.7	0.266
Triglycerides, mg/dL	146.4 (58.5–409.2)	164.1 (44.3–485.4)	0.406
ApoA-I, mg/dL	122.1 ± 13.0	123.6 ± 15.1	0.526
ApoB-100, mg/dL	95.2 ± 31.1	98.0 ± 28.2	0.572
Lp(a), mg/dL	8.6 ± 4.2	10.2 ± 5.0	0.032
Hemoglobin, g/dL	10.0 ± 2.2	9.6 ± 1.8	0.217
Corrected calcium, mg/dL	9.8 ± 1.4	9.6 ± 0.8	0.337
Phosphorus, mg/dL	4.9 ± 1.6	4.1 ± 1.3	0.004
Hypertension	52 (62.7%)	46 (66.7%)	0.606
Diabetes mellitus	13 (15.7%)	15 (21.7%)	0.336
Chronic inflammation
CRP, mg/L	4.7 (1.5–25.6)	7.0 (0.9–27.4)	0.003
Fibrinogen, mg/dL	385.3 ± 90.9	407.7 ± 118.2	0.189
Nutrition
Serum albumin, g/dL	3.3 ± 0.5	3.4 ± 0.5	0.049
Serum prealbumin, mg/L	312.9 ± 45.3	320.4 ± 44.2	0.305
Peritonitis rate	13 (15.7%)	20 (29.0%)	0.047

Notes: *p < 0.05. ApoA-I, apolipoprotein A-I; ApoB-100, apolipoprotein B-100; BUN, blood urea nitrogen; CRP, C-reactive protein; HDL, high-density lipoprotein; LDL, low-density lipoprotein; Lp(a), lipoprotein(a).

What are the possible mechanisms leading to the survival advantage of normal weight? Indeed, few studies have evaluated the risk factors in PD patients according to BMI. Previous study showed that abnormal concentrations of Lp(a) in patients with end-stage renal disease might be the cause of the high risk of atherosclerosis.¹⁹ In this study, the serum levels of Lp(a) were significantly higher in overweight patients, which may be related to high cardiovascular disease, therefore contributed to increased mortality in overweight patients. The research also showed that serum levels of CRP were significantly higher in overweight patients than in normal weight patients, which was in accord with the results from de Araujo Antunes et al.²⁰ At the same time, linear regression analysis found that CRP was a significant risk factor associated with BMI. An elevated level of CRP, which was a marker of inflammation, was a risk factor for morbidity and mortality in the general population²¹; and in PD patients, CRP was strongly associated with cardiovascular disease.^22,23 Therefore, we confirm that high levels of CRP in overweight patients are associated with bad outcomes. Ohkuma et al. further demonstrated that Lp(a) and CRP levels have an important role in carotid atherosclerosis in PD patients.²⁴ Moreover, the incidence of peritonitis was elevated obviously in overweight group. It is well known that peritonitis was a risk factor for mortality in PD patients.¹³ Therefore, we believe that poor survival of overweight patients is due to cardiovascular risk, chronic inflammation, and high incidence of peritonitis.

Despite our efforts to provide unbiased comparisons, there are limitations to the present study. First, underweight and obese patients were not included in the study because of the very limited number. This could lead to omissions of the population. Second, it is possible that not all patients were at their true dry weight before PD. This could lead to bias comparisons between groups. Third, this study was limited to only one center. It could lead to selection bias. Despite its limitations, this study suggests a protective effect in PD patients with normal BMI. High levels of serum Lp(a) and CRP at the start of PD may contribute to increased mortality in overweight patients. In addition, peritonitis during the course of PD is related to poor outcome in overweight patients. Further studies are needed to evaluate whether these differences lead to the poor survival of overweight patients observed in PD patients.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.