This article has Open Peer Review reports available.
Early utilization of hypertonic peritoneal dialysate and subsequent risks of non-traumatic amputation among peritoneal dialysis patients: a nationwide retrospective longitudinal study
© Lin et al.; licensee BioMed Central Ltd. 2013
Received: 8 November 2012
Accepted: 13 June 2013
Published: 20 June 2013
The hemodialysis (HD) population has a particularly high incidence of amputation, which is likely associated with decreased tissue oxygenation during HD. However, information about the risk factors leading to amputation in peritoneal dialysis (PD) patients is limited. Here, we have investigated the association between the use of hypertonic peritoneal dialysate (HPD) and subsequent amputation in PD patients.
Based on the data from the Taiwan National Health Insurance research database, this observational cohort study enrolled 203 PD patients who had received HPD early during treatment and had not undergone amputation and 296 PD controls who had not undergone amputation. Subjects were followed through until the end of 2009 and the event rates of new non-traumatic amputation were compared between groups.
The incidence of amputation was 3 times higher for the HPD cohort than for the comparison cohort (23.68 vs. 8.01 per 1000 person-years). The hazard ratio (HR) for this group, estimated using a multivariable Cox model, was 2.48 (95% confidence interval [CI] = 1.06–5.79). The HR for patients with both diabetes and early adoption of HPD increased to 44.34 (95% CI = 5.51-357.03), compared to non-HPD non-diabetic PD controls.
Early utilization of HPD in PD patients is associated with increasing risk of amputation; this risk considerably increases for those with concomitant diabetes.
Non-traumatic extremity amputation, usually related to diabetes, places a considerable burden on individuals, families, and health care finances . Renal failure and dialysis initiation are recognized risk factors for non-traumatic amputation [2, 3]. Compared with the general population, dialysis patients experience higher amputation incidence and have much higher post-amputation mortality rates . O’Hare et al reported that hemodialysis (HD) patients who were male, diabetic, had been previously diagnosed with peripheral vascular disease (PVD), had high mean systolic blood pressure, or elevated phosphate levels had increased risks of future amputation . Only a few studies have investigated the risk factors of amputation in peritoneal dialysis (PD) patients, and these have had limited sample size [6, 7]. Pliakogiannis et al studied 71 diabetic patients receiving PD and noted that low levels of albumin, peripheral neuropathy, and coronary artery disease were associated with foot lesions . In the clinic, we noted that PD patients who required hypertonic peritoneal dialysis (HPD) solution soon after the initiation of PD were more likely to require amputation. We hypothesize that early HPD use is related to future amputation in PD patients. To test this hypothesis, we conducted a retrospective cohort analysis to assess the incidence of amputation among PD patients and whether this incidence was affected by HPD utilization within the first 6 months of initiating PD.
The present study was based on data obtained from the National Health Insurance Research Database (NHIRD) of the National Health Research Institute. The universal National Health Insurance program, launched in March 1995, provides affordable health care to nearly 99% of the population and is contracted with 97% of clinics and hospitals throughout the nation. From the NHIRD, we extracted a Longitudinal Health Insurance Database (LHID), which contained one million beneficiaries, randomly selected from 1996-2000 insurers. The LHID database contains comprehensive annual health care data for all beneficiaries including birth date, gender, diagnostic codes, prescriptions, procedures, and surgeries. The disease diagnosis was coded with International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) form outpatient, inpatient and catastrophic illness file. Personal information was de-identified before the release of the research database, thus this study was exempt from approval by the Institutional Review Board.
Based on inpatient and outpatient claims, we identified patients who had been first diagnosed with end-stage renal disease (ESRD; ICD-9-CM code 585) and were receiving PD during 1998–2009. Among these new PD patients, subjects prescribed HPD (i.e., 7.5% icodextrin solution or 4.25% dextrose solution) within the first 6 months of initiating PD were identified as the HPD cohort. Those new PD patients who did not received HPD within the first 6 months of initiating PD were considered as the comparison cohort. The index date was the date of initiating PD in the control and in the study cohort. The following patients were excluded: (1) patients who underwent amputation before the index date, (2) patients who experienced severe illness but did not have a record of PD, and (3) patients who had received HD or had undergone renal transplantation before the index date. We followed the cohorts until diagnosis and surgery for the first amputation, either major or minor, (ICD-9-CM 785.4 and 440.24; ICD-9-CM 84.10-84.17) been made, withdrawal from insurance, loss to follow-up, or December 31, 2009, whichever was the latest.
Baseline comorbidities thought to be associated with subsequent amputation were also analyzed. These included diabetes (ICD-9 code 250), hypertension (ICD-9 codes 401-405), ischemic heart disease (ICD-9 codes 410-414, A270, and A279), previous foot ulcers (ICD-9 codes 707.1-707.9), diabetic neuropathy (ICD-9 codes 353.5, 357.2, 354.0-355.9, and 337.1), peripheral vascular disease (PVD, ICD-9-CM codes 443.89 443.9), hyperparathyroidism (ICD-9-CM codes 252.0), heart failure (ICD-9-CM codes 428), diabetes duration, and dialysis vintage.
The differences in demographic characteristics and comorbidities between the study and comparison cohort were examined by using Chi-square test for categorical variables and t-test for continuous variables. We used the Kaplan-Meier method to estimate the amputation-free rates and the log-rank test to examine the statistical significance of the differences between the study groups. Cox proportion hazard model was used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for the risk of subsequent amputation. We further analyzed the interaction between early use of hypertonic dialysate and diabetes. We performed all statistical analysis using SAS 9.1 statistical software (SAS Institute, Inc., Cary, NC, USA) and drew the cumulative incidence curve using R software (R Foundation for Statistical Computing, Vienna, Austria). The differences were considered significant if a two-sided P value was less than or equal to 0.05.
Baseline demographic status and comorbidity compared between hypertonic solution cohort and comparison cohort
N = 296 (%)
N = 203 (%)
Age, mean (SD) years
Ischemic heart disease
DM foot ulcer
Peripheral vascular disease
Follow-up duration, mean (SD)
DM duration, mean (SD)
Incidence of amputation and multivariate Cox proportional hazards regression analysis measured hazard ratio for PD patients using hypertonic solution
Crude HR (95% CI)
Hazard ratio (95% CI)
The interaction between diabetes and hypertonic solution for amputation risk
Crude HR (95%CI)
Adjusted HR (95%CI)
Discussion and conclusions
In this nationwide retrospective cohort study, we have illustrated the relationship between early utilization of HPD and subsequent amputation in PD patients. The use of HPD within the first 6 months after initiation of PD was associated with a 2.48-fold increase in the HR of new amputation compared to that for the comparison cohort. Furthermore, HPD cohort who had concomitant DM carried the highest risk, with 44.34 times the incidence of amputation seen in non-HPD non-diabetic patients.
Recent commencement of dialysis therapy has been recognized as an influential risk factor for lower limb amputation in the DM population . Several researchers have studied the pathophysiological changes that occur during hemodialysis, which may contribute to these limb-threatening conditions [8–12]. Systemic hypoxemia, microcirculatory hypoperfusion, and decreased transcutaneous oxygen tension of the lower limbs can occur during and after HD; all these factors could lead to limb ischemia and amputation [9–11]. Although there are no comparable data on PD, it is possible that HPD could produce similar effects. For instance, HPD might create a more rapid fluid shift into the peritoneal cavity, reducing microcirculatory blood flow and tissue oxygen tension. Furthermore, elevated blood pH levels could cause a shift in the oxygen dissociation curve, leading to tissue hypoxia . Furthermore, the use of human recombinant erythropoietin and possible vasoconstriction of microcirculation during ultrafiltration might cause rheological changes and impair microvascular perfusion [6, 13]. Moreover, PD patients placed on HPD have clinical signs of fluid overload, such as hypertension and tissue edema, which may worsen tissue oxygenation status. Li et al. found that PD patients have higher incidence rates of mesenteric ischemia than HD patients, indicating that PD therapy might contribute to the advancement of microvascular disease . They hypothesized that chronic exposure to high glucose levels caused micro- and macrovascular damage and contributed to advancing atherosclerosis, which could lead to organ ischemia.
In our study, diabetes was associated with an RR of 19.33 for amputation in PD patients. This result is in accordance with those of previous studies that found that diabetes was the strongest risk factor for amputation in HD patients [5, 15–17]. Furthermore, our results showed that diabetic patients tended to initiate HPD earlier than those without diabetes. The risk of amputation with early utilization of HPD and concomitant diabetes was 44.34 times greater than that for PD patients who were neither diabetic nor given HPD therapy. Since diabetics are at greater risk for early ultrafiltration failure, microvasculopathy, and autonomic neuropathy [18–20], early utilization of high-strength dialysate might lead to longer glucose exposure, increased glucose toxicity to the vessels, and finally, limb ischemia, than in non-diabetics. Therefore, we hypothesize that early HPD utilization in diabetic PD patients promotes atherogenesis and/or microvascular hypoperfusion.
Among patients younger than 70, the rate of amputation for both cohorts increased with age. This is likely due to an increased risk of death with surgery for this population, leading to a lower probability of amputation for patients over 70. Eagger et al. found a positive association between increasing age and amputation in the ESRD population . Combes et al. observed that while older age was associated with a decreased probability of amputation in diabetic HD patients, it actually increased the likelihood of amputation among non-diabetics . Pilakogiannis et al reported that age is not related to developing PVD, but no further information is available about the association between age and amputation in PD patients . The inconsistency of these results across studies likely reflects the diverse medical care and mortality rates of different renal populations.
McMurray et al showed that providing diabetic education and care to HD and PD patients significantly improves foot care and reduces the need for amputation . For instance, introduction of the Staged Diabetes Management clinical pathways for dialysis patients have shown to lower the amputation rate considerably . Chiropody been reported could prevent amputation in diabetic PD patients . Thus, amputation in dialysis patients might be preventable in many cases when comprehensive care is provided.
Our study has several strengths. First, it utilized the NHIRD database. This large population database with comprehensive electronic medical records provided complete data about the incidence of amputation as well as important information including that pertaining to age, sex, types of dialysis solution, dialysis vintage, and comorbidities. Second, we excluded subjects with previous amputations and included only PD patients who received long-term follow up care. Excluding patients with past amputation might eliminate possible bias with regard to analyzing future amputation and provide a clearer interpretation of the effect of HPD on limb ischemia.
The several limitations of our study should also be mentioned. First, the number of subjects included was relatively small, leading to large confidence intervals. Furthermore, it is possible that some patients had subclinical PVD. However, we also considered comorbidities such as foot ulcers, heart failure, and cardiovascular risk factors associated with PVD, suggesting that any effect of undiagnosed PVD on our results was likely to be subtle. We also did not have precise information about smoking status and the calcium and phosphate levels, which might be associated with the risk of future amputation. However, Pliakogiannis et al reported that time average Kt/v, creatinine clearance, serum calcium levels, calcium and phosphate production, and intact parathyroid hormone level are not associated with amputation in PD patients . We also lacked information about the subjects’ scores on baseline measures of circulatory status, such as the ankle-brachial index (ABI) . Given that ABI is not correlated with the severity of peripheral arterial disease among dialysis patients and would be falsely elevated by arterial calcification, it may be appropriate to overlook ABI in the current study.
In conclusion, along with diabetes, early utilization of HPD is associated with the subsequent risk of amputation in PD patients. We suggest that PD patients be provided intensive education on foot protection and screening for evidence of limb ischemia, especially those who received HPD early. Further studies are required to determine the efficacy and cost effectiveness of integrated and multidisciplinary foot care in PD patients.
This study was supported by a grant from China Medical University Hospital (DMR-100-172) in Taichung, Taiwan.
- Boulton AJM, Vileikyte L, Ragnarson-Tennvall G, Apelqvist J: The global burden of diabetic foot disease. Lancet. 2005, 366: 1719-1724. 10.1016/S0140-6736(05)67698-2.View ArticlePubMedGoogle Scholar
- McGrath N, Curran B: Recent commencement of dialysis is a risk factor for lower-extremity amputation in a high-risk diabetic population. Diabetes Care. 2000, 23: 432-433. 10.2337/diacare.23.3.432.View ArticlePubMedGoogle Scholar
- Game FL, Chipchase SY, Hubbard R, Burden RP, Jeffcoate WJ: Temporal association between the incidence of foot ulceration and the start of dialysis in diabetes mellitus. NDT. 2006, 21: 3207-3210. 10.1093/ndt/gfl427.PubMedGoogle Scholar
- Eggers PW, Gohdes D, Pugh J: Nontraumatic lower extremity amputations in the Medicare end-stage renal disease population. Kidney Int. 1999, 56: 1524-1533. 10.1046/j.1523-1755.1999.00668.x.View ArticlePubMedGoogle Scholar
- O'Hare AM, Bacchetti P, Segal M, Hsu C, Johansen KL: Factors associated with future amputation among patients undergoing hemodialysis: Results from the Dialysis Morbidity and Mortality Study Waves 3 and 4. Am J Kidney Dis. 2003, 41 (1): 162-170. 10.1053/ajkd.2003.50000.View ArticlePubMedGoogle Scholar
- Wakeen M, Zimmerman SW: Association between human recombinant EPO and peripheral vascular disease in diabetic patients receiving peritoneal dialysis. Am J Kidney Dis. 1998, 32: 488-493. 10.1053/ajkd.1998.v32.pm9740167.View ArticlePubMedGoogle Scholar
- Pliakogiannis T, Bailey S, Cherukuri S, Taskapan H, Ahmad M, Oliver T, Bargman J, Oreopoulos D: Vascular complications of the lower extremities in diabetic patients on peritoneal dialysis. Clin Nephrol. 2008, 69: 361-367.View ArticlePubMedGoogle Scholar
- Torrance J, Milne F, Hurwitz S, Zwi S, Rabkin R: Changes in oxygen delivery during hemodialysis. Clin Nephrol. 1975, 3: 54-59.PubMedGoogle Scholar
- Aurigemma NM, Feldman NT, Gottlieb M, Ingram RH, Lazarus JM, Lowrie EG: Arterial oxygenation during hemodialysis. NEJM. 1977, 297: 871-873. 10.1056/NEJM197710202971607.View ArticlePubMedGoogle Scholar
- Bemelmans RHH, Boerma EC, Barendregt J, Ince C, Rommes JH, Spronk PE: Changes in the volume status of haemodialysis patients are reflected in sublingual microvascular perfusion. NDT. 2009, 24: 3487-3492. 10.1093/ndt/gfp267.PubMedGoogle Scholar
- Hinchliffe RJ, Kirk B, Bhattacharjee D, Roe S, Jeffcoate W, Game F: The effect of haemodialysis on transcutaneous oxygen tension in patients with diabetes—a pilot study. NDT. 2006, 21: 1981-1983. 10.1093/ndt/gfl241.PubMedGoogle Scholar
- Wathen RL, Ward RA, Harding GB, Meyer LC: Acid-base and metabolic responses to anion infusion in the anesthetized dog. Kidney Int. 1982, 21: 592-599. 10.1038/ki.1982.66.View ArticlePubMedGoogle Scholar
- Sotirakopoulos N, Tsitsios T, Stambolidou M, Athanasiou G, Peiou M, Kokkinou V, Mavromatidis K: The red blood cell deformability in patients suffering from end stage renal failure on hemodialysis or continuous ambulatory peritoneal dialysis. Ren Fail. 2004, 26: 179-183. 10.1081/JDI-120038517.View ArticlePubMedGoogle Scholar
- Li SY, Chen YT, Chen TJ, Tsai LW, Yang WC, Chen TW: Mesenteric Ischemia in Patients with End-Stage Renal Disease: A Nationwide Longitudinal Study. Am J Nephrol. 2012, 35: 491-497. 10.1159/000338451.View ArticlePubMedGoogle Scholar
- Combe C, Albert JM, Bragg-Gresham JL, Andreucci VE, Disney A, Fukuhara S, Goodkin DA, Gillespie BW, Saito A, Jadoul M: The burden of amputation among hemodialysis patients in the Dialysis Outcomes and Practice Patterns Study (DOPPS). Am J Kidney Dis. 2009, 54: 680-692. 10.1053/j.ajkd.2009.04.035.View ArticlePubMedGoogle Scholar
- Speckman RA, Frankenfield DL, Roman SH, Eggers PW, Bedinger MR, Rocco MV, McClellan WM: Diabetes is the strongest risk factor for lower-extremity amputation in new hemodialysis patients. Diabetes Care. 2004, 27: 2198-2203. 10.2337/diacare.27.9.2198.View ArticlePubMedGoogle Scholar
- Perales MCS, Cortés MJG, Utiel FJB, Viedma G, Gil J, del Barrio PP, Hinojosa JB, Liébana A, Bañasco VP: Incidence and risk factors for non-traumatic amputation of lower limbs in patients on hemodialysis. Nefrologia. 2005, 25: 399-406.Google Scholar
- Selgas R, Bajo MA, Castro MJ, del Peso G, Aguilera A, Fernandez-Perpen A, Cirugeda A, Sanchez-Tomero JA: Risk factors responsible for ultrafiltration failure in early stages of peritoneal dialysis. Peritl Dial Int. 2000, 20: 631-636.Google Scholar
- Said G, Goulon-Goeau C, Slama G, Tchobroutsky G: Severe early-onset polyneuropathy in insulin-dependent diabetes mellitus. NEJM. 1992, 326: 1257-1263. 10.1056/NEJM199205073261905.View ArticlePubMedGoogle Scholar
- Cooper ME, Bonnet F, Oldfield M, Jandeleit-Dahm K: Mechanisms of diabetic vasculopathy: an overview. Am J Hypertens. 2001, 14: 475-486. 10.1016/S0895-7061(00)01323-6.View ArticlePubMedGoogle Scholar
- McMurray SD, Johnson G, Davis S, McDougall K: Diabetes education and care management significantly improve patient outcomes in the dialysis unit. Am J Kidney Dis. 2002, 40: 566-575. 10.1053/ajkd.2002.34915.View ArticlePubMedGoogle Scholar
- Rith-Najarian S, Gohdes D: Preventing amputations among patients with diabetes on dialysis. Diabetes Care. 2000, 23: 1445-1456. 10.2337/diacare.23.9.1445.View ArticlePubMedGoogle Scholar
- Lipscombe J, Jassal SV, Bailey S, Bargman JM, Vas S, Oreopoulos DG: Chiropody may prevent amputations in diabetic patients on peritoneal dialysis. Peritl Dial Int. 2003, 23: 255-259.Google Scholar
- Adragao T, Pires A, Branco P, Castro R, Oliveira A, Nogueira C, Bordalo J, Curto JD, Prata MM: Ankle—brachial index, vascular calcifications and mortality in dialysis patients. NDT. 2012, 27: 318-325. 10.1093/ndt/gfr233.PubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2369/14/128/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.