| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical |
Division of Nephrology, Department of Internal Medicine, China Medical University Hospital, Taichung, Taiwan
Correspondence to: C.C. Huang, Division of Nephrology, Department of Internal Medicine, China Medical University Hospital, 2 Yu-Der Road, Taichung, Taiwan. cch659{at}yahoo.com.tw
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Background: Peripheral artery disease (PAD) is highly
prevalent among patients in end-stage renal disease. The ankle–brachial
index (ABI) is believed to be highly correlated with the subclinical PAD of
lower extremities but little is known about the associated risk factors and
outcome for PAD and ABI in patients on peritoneal dialysis (PD). Methods: We performed a cohort study of 153 patients
from a single center receiving stable PD for more than 3 months. These
patients were screened for subclinical PAD using the ABI measurement. The ABI
was measured and a ratio of <0.9 was considered abnormal. Clinical outcomes
included actuarial patient and technique survival in this study. Results: 30 patients were classified into a subclinical
PAD group. The prevalence of PAD (subclinical and overt) in our PD center was
19.61% (30/153). Advanced age, preexisting diabetes, preexisting
cardiovascular and/or cerebrovascular disease (CVD), lower renal Kt/V urea,
lower renal creatinine clearance (WCrCl), lower serum albumin level, and
higher serum triglyceride level were risk factors for PAD in our PD center.
Bivariate analysis showed that ABI was positively correlated with residual
renal Kt/V urea and WCrCl, but was not correlated with peritoneal Kt/V urea
and WCrCl. Patient and technique survival rates were significantly lower in
the low ABI group than in the normal ABI group. Conclusions: ABI is highly correlated with advanced
age, preexisting diabetes, preexisting CVD, serum albumin, serum triglyceride,
and residual renal clearance in PD patients. Also, lower ABI is independently
associated with a high risk of patient mortality and PD technique
failure.
KEY WORDS: Ankle–brachial index; peripheral artery disease; risk factors; outcome.
Peritoneal dialysis (PD) is one of the main renal replacement therapies for patients with end-stage renal disease (ESRD), and those patients are at higher risk of peripheral artery disease (PAD) than the general population. However, little is known about the associated risk factors and outcome for PAD in patients on PD treatment. Clinically, the ankle–brachial blood pressure index (ABI) has been widely used to screen for subclinical PAD and is believed to be highly correlated with PAD of lower extremities (1). A recent study revealed that ABI is an independent and potential predictor for all-cause and cardiovascular mortality in hemodialysis (HD) patients (2). Thus, early detection of overt and subclinical PAD and associated risk factors in order to further modify the risks and prevent disease is important in patients with ESRD. Previous studies have shown that the prevalence of overt and subclinical PAD is different between patients undergoing HD and the general population (3,4), but the information on prevalence and clinical correlates for PAD in PD patients is limited. The relationships between known promoters of subclinical PAD in patients receiving PD have not been determined. In the present study we aimed to investigate the prevalence of PAD and major risk factors in ESRD patients receiving PD therapy. The second aim was to elucidate the relationship between ABI and both technique and patient survival in patients undergoing PD.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
MEASURING ABI, PERITONEAL EQUILIBRATION TEST (PET), WEEKLY CREATININE CLEARANCE (WCrCl), AND Kt/V UREA
Measurements of ABI were performed in supine position; blood pressure was
measured in the left arm (brachial artery) and both ankles (posterior tibial
arteries). Two readings were recorded 10 minutes apart. The mean of these two
readings taken in the ankles and arm was used to calculate ABI. The ABI was
calculated by dividing systolic blood pressure, measured at either the right
or the left ankle, by systolic blood pressure, measured at the arm. An ABI
value greater than 0.9 was defined as normal. Subclinical PAD was defined as
an ABI value less than 0.90 in either extremity. History was also taken for
patients with ABI less than 0.90 to find whether they had signs and symptoms
compatible with intermittent claudication, defined as pain on walking that
disappeared on resting. Weekly creatinine clearance (WCrCl; liters per week),
Kt/V urea, and PET of dialysis were performed using a 2.27% dextrose solution
for a 2-L volume exchange instilled in the peritoneal cavity. The WCrCl value
was normalized to 1.73 m2 body surface area. Patients were excluded
if there was a possibility of peritonitis (based on clinical and laboratory
manifestations) when the peritoneal function test was performed. On the day of
investigation, 24-hour collection of the dialysis fluid and urine was
performed to calculate dialysis adequacy. The ratio of glucose concentration
in the dialysate after 4 hours to initial concentration (D4/D0) was reported.
The Kt/V urea and WCrCl values in this study were the residual renal
clearance, PD clearance, or the summary of renal and peritoneal clearances, as
appropriate.
COLLECTION OF CLINICAL DATA
Data on the age, sex, smoking history, comorbid conditions, underlying
disease, medication, and history of dialysis were obtained from the PD
records. Blood was drawn during the morning after an overnight fast of at
least 12 hours. Serum ferritin level was quantitatively measured by the
Immulite 2000 Analyzer (PIL2KFE-11; Siemens Healthcare Diagnostics, Deerfield,
IL, USA). Intact parathyroid hormone (PTH) levels were measured by
immunoradiometric assay using the Allegro Intact PTH assay kit (Nichols
Institute Diagnostics, San Juan Capistrano, CA, USA). All the biochemistries
were determined with an automated clinical chemistry analyzer. Patients'
clinical and laboratory data before dialysis and after initiation of PD were
collected retrospectively.
The patient group designated cardiovascular and/or cerebrovascular disease (CVD) was defined as those patients with a history of angina, coronary arterial disease, myocardial infarction, abnormal angiography results, transient ischemic attack, and/or cerebrovascular accident. During the study period, blood pressure was recorded in the supine position using a mercury sphygmomanometer. We defined blood pressure as the average value of two readings taken successively and separately. Hypertension was defined as a systolic blood pressure of 140 mmHg or greater, diastolic blood pressure of 90 mmHg or greater, by physician diagnosis, or by use of antihypertensive medication. Diabetes was defined by physician diagnosis, use of diabetes medication, fasting glucose >126 mg/dL, or non-fasting glucose >200 mg/dL. Individuals were also classified as nonsmokers (never smoked) or ever smokers (ex-smokers and current smokers). The ever smoker group (20 patients) had an 18.64 ± 15.32 pack/year history and the ex-smokers (7 patients) had a history of cessation of 7.32 ± 4.22 years in our patients. Medication history was reviewed and defined as patients taking medication for more than 3 months before measurement of ABI.
Clinical outcomes of this study included actuarial patient and technique survival. Transplantation and lost to follow-up but not conversion to HD were counted as censored observations for patient survival estimates. Lost to follow-up was counted as a censored observation for technique survival estimation.
STATISTICAL ANALYSIS
Data are expressed as mean ± standard deviation (SD). Continuous
data were compared by unpaired Student's t-test. Categorical variables were
compared by chisquare (with Yate's correction) or Fisher's exact test, as
appropriate. To analyze the correlation between two indices, Pearson's
correlation coefficient was used. Multiple linear regression analysis with
forward stepwise selection was performed to identify factors independently
correlated with ABI. The Cox proportional hazards model was used for
statistical analysis of patient and technique survival. Patient and technique
survival curves were estimated using the Kaplan–Meier technique and
tested using the log-rank test. All tests were two-tailed and a p
value of 0.05 or less was considered statistically significant. All
calculations were performed with SPSS version 10.0 (SPSS Inc., Chicago, IL,
USA).
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
In univariate analysis, the associated risk factors for PAD were advanced age (64.3 ± 11.2 vs 52.1 ± 13.8 years, p < 0.001), preexisting diabetes mellitus (33.3% vs 16.3%, p = 0.038), preexisting CVD (40.0% vs 20.3%, p = 0.044), lower total Kt/V (1.88 ± 0.29 vs 2.13 ± 0.33, p < 0.001), lower renal Kt/V (0.07 ± 0.16 vs 0.20 ± 0.31, p = 0.024), lower total WCrCl (48.85 ± 10.13 vs 57.01 ± 13.44 L/week/1.73 m2, p = 0.002), lower renal WCrCl (3.08 ± 5.51 vs 9.82 ± 15.16 L/week/1.73 m2, p = 0.018), lower serum albumin level (3.49 ± 0.37 vs 3.64 ± 0.35 g/dL, p = 0.036), and higher serum triglyceride (291.07 ± 225.08 vs 209.84 ± 140.98 mg/dL, p = 0.015). There were no statistical differences in gender, duration of PD, smoking habit, current hypertension status, intact PTH, peritoneal weekly Kt/V urea, peritoneal WCrCl, hematocrit, serum total calcium, serum phosphorus, total cholesterol level, serum ferritin level, or medication history (statin, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and antiplatelet agent) between subclinical PAD and non-subclinical PAD patients.
Bivariate analysis showed that ABI was negatively correlated with age and serum triglyceride level (p < 0.001 and p = 0.005 respectively), and positively correlated with serum albumin level, peritoneal D4/D0 data, renal Kt/V, and renal WCrCl (p = 0.004, p = 0.047, p = 0.016, and p = 0.006, respectively), but was not correlated with duration of PD, peritoneal weekly Kt/V, or peritoneal WCrCl (Table 2).
|
A multiple linear regression model to identify factors independently correlated with ABI is shown in Table 3. Age and renal WCrCl were independently related to ABI (p < 0.001 and p = 0.006 respectively).
|
ACTUARIAL PATIENT SURVIVAL
During the study period, 28 patients were changed to HD, 1 received renal
transplants, and 3 transferred to other PD centers. During the study period,
there were 27 deaths, 3 of which occurred after the change to HD and were
counted as events during survival analysis. The causes of death were
cardiovascular diseases (10 patients), peritonitis (6 patients),
non-peritonitis infections (7 patients), malignancy (1 patient), hollow organ
rupture (2 patients), and acute pancreatitis (1 patient). The remaining 97
patients were administratively censored on 31 December 2007.
Figure 1 shows a comparison of
the patient survival curves of PD patients with ABI <0.9 and PD patients
with ABI 
0.9. The curve shows a statistically significant difference
between the two groups (p < 0.0001). Actuarial patient survival at
30 months was 78.9%. Patient survival in the group with ABI <0.9 was lower
compared with those with ABI 0.9 (43.5% vs 86.7%).
|
0.9
(33.3% vs 77.0%, p < 0.0001;
Figure 2).
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Limited data have been reported comparing the cardiovascular risk factors between HD and PD. Some authors report a worse cardiovascular outcome in PD compared with HD patients (8,9), but debates on such findings are ongoing (10). An association between PAD and duration of dialysis was found in a previous study dealing with PAD risk factors in HD patients (3). In accordance with this report, a relationship that might associate either the HD process or the uremic milieu with increased atherosclerotic disease burden has been suggested. Unlike the previous study, this was not the case in the ESRD patients undergoing PD therapy in our study.
Hyperhomocysteinemia is an independent risk factor for atherosclerotic complications in patients with ESRD. Moustapha et al. (11) found that hyperhomocysteinemia is more prevalent and intense in HD patients compared to PD patients. It is possible that, in our study, the lack of association between duration of PD and PAD could reflect the impact of lower serum homocysteine levels in PD patients.
Serum albumin level may not be a valid nutritional marker as it is affected by inflammation and external losses, but a low serum albumin level is generally accepted as an indicator of inflammation in ESRD patients (12). Strong associations between malnutrition, inflammation, and atherosclerosis in ESRD patients suggest the presence the malnutrition–inflammation–atherosclerosis (MIA) syndrome (13). Our results showed that lower serum albumin levels are associated with PAD and a significantly lower ABI. We had no direct evidence to link hypoalbuminemia and the inflammatory process in our patients, but we believe that inflammation might play a key role in the influence of the prevalence of PAD in PD patients.
Neither serum total cholesterol level nor hypertension was associated with PAD in our patients, which is similar to findings in the HD subjects in the HEMO Study (6). Smoking was a major risk factor for the development of peripheral diseases and smoking patients had more frequent complications of their atherosclerotic disease at other vessel sites (14). However, Schillinger et al. showed that smoking was associated with a reduced rate of intermediate-term restenosis after lower-limb endovascular intervention (15). Our data suggest that smoking was not associated with subclinical PAD in PD patients. The difference may be due partly to the smoking habits of our population. There was a relatively smaller proportion of diabetic patients who were current or ex-smokers compared with patients without diabetes. This may introduce confounding. Due to the smaller sample size of our population, further analysis of the correlation between smoking and subclinical PAD in the diabetic population was difficult. In accordance with the results of our study, we assumed that various chemicals contained in cigarettes (nicotine, acetaldehyde, acrolein, etc.) that had been shown to cause necrosis of vascular smooth muscle cells reduced neointima formation (16) and resulted in the observed relationship in our study.
Substantial glucose loading (150 g – 300 g per day), which is associated with the use of standard glucose-based dialysis solutions, imposes a carbohydrate load that could lead to longer-term metabolic complications such as hypertriglyceridemia (17,18). Modification of the PD procedure with the use of non glucose-containing solution (icodextrin) to decrease serum triglyceride levels has been reported (19,20). Whether therapy with non glucose-containing solution is effective at reducing the progression of subclinical PAD in PD patients is still unknown. Moreover, Hasimu et al. published data with an emphasis on the need for medical management (statin, angiotensin-converting enzyme inhibitor or angiotensin receptor blocker, and antiplatelet agent) for PAD patients (21). The benefit of those medications was not shown in our results and it is probable that the follow-up period in our study was not adequate. A randomized prospective control study with modifications to minimize biases is worthy of further investigation.
Up to now, there is no accepted opinion on whether dialysis adequacy is associated with PAD and/or ABI values in HD patients. The published information of the USRDS Dialysis Morbidity and Mortality Study (DMMS) shows negative associations between Kt/V urea and PAD in HD patients in Waves 3 and 4, but not in Wave 1 (3). A report from Taiwan showed no correlation between Kt/V urea and subclinical PAD in HD patients (4). To our knowledge, whether dialysis adequacy is associated with overt or subclinical PAD in PD patients remains unanswered. It seems that a small solute clearance by RRF plays an important role in influencing the ABI values with a more positive correlation compared to that of peritoneal clearance. Residual renal function of PD patients has recently been found to be an important predictor of many clinical outcomes. Low RRF is thought to be associated with poor nutritional status, inflammation, and higher overall mortality via enhancement of atherosclerosis (22,23). Our finding also revealed a positive correlation between the small-solute clearance by RRF and ABI values in PD patients.
Low ABI has been reported to be independently associated with a high risk of all-cause and cardiovascular mortality in high-risk populations (2,24). In the current study, we performed a prospective analysis of PD patients. We found that a low ABI can be a predictor of both patient and technique survival in patients undergoing PD. Multivariate analysis revealed that the role of a low ABI was independent of hypoalbuminemia in affecting mortality in PD patients. In our study, 22 patients with PD peritonitis were associated with PD technique failure. A recent report from South Korea found that reduced RRF is a peritonitis risk in patients on continuous ambulatory PD (25). It is possible that patients with better preserved RRF might have better immune defense mechanisms along with a better preserved nutritional status. As previously discussed in our report, a low ABI is associated with poorer RRF. These findings together probably suggest that a lower ABI value is associated with the development of technique failure in our patients.
There were several limitations in this study. First, this study did not examine the relationships between possible risk factors and de novo development of PAD and ABI values. Second, some clinical and laboratory parameters were collected retrospectively; therefore, some inflammation markers (C-reactive protein, etc.) were not completely collected for analysis. Third, data on some novel cardiovascular risk factors [lipoprotein(a), homocysteine level, body-mass index, metabolic syndrome, etc.] that are associated with PAD in the HD and general populations were unavailable due to lack of data. Thus, the association of such novel parameters in influencing overt PAD and ABI values could not be examined. In spite of these limitations, we believe that the results of this study are worthy of further investigation with prospective randomized trials with modifications to minimize biases.
In conclusion, our data show that PAD is common in patients undergoing PD. Advanced age, preexisting CVD, preexisting diabetes, hypoalbuminemia, hypertriglyceridemia, and low residual renal clearance appear to play important roles in mediating PAD. A low ABI is associated with poor prognosis of patient and technique survival in the PD population. A further study will be needed to prospectively examine whether variables discovered in our study are also associated with the de novo development of PAD and the influence of the ABI value in PD patients.
|
FOOTNOTES |
|---|
Received 14 December 2007; accepted 24 April 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
