Losses of 1,25- and 24,25-dihydroxycholecalciferol in the peritoneal fluid of patients treated with continuous ambulatory peritoneal dialysis

We measured peritoneal losses of the active vitamin D metabolites 1,25(OH)2D3 and 24,25(OH)2D3 in patients receiving continuous ambulatory peritoneal dialysis (CAPD). The serum concentration of 24,25(OH)2D3 was considerably lower than in hemodialysis patients. The serum concentration of 1,25(OH)2D3 was undetectable and rose to levels similar to those in hemodialysis patients only after loading with much higher oral doses of 1-alpha-vitamin D3 than those received by hemodialysis patients. Losses of both metabolites in peritoneal fluid were considerable, averaging approximately 6-8% of the plasma pool per day. These losses lead to low serum levels of these active vitamin D metabolites in CAPD patients, which may be an important factor in exacerbating renal osteodystrophy. Our results indicate the need for increased replacement doses of vitamin D metabolites in CAPD patients.     

Opportunities to improve the care of patients with kidney transplant failure

BACKGROUND: Patients initiating dialysis after a failed kidney transplant are a subgroup of chronic kidney disease (CKD) patients who have not been characterized. Exposure to immunosuppressive medications differentiates these patients from the general incident end-stage renal disease (ESRD) population. METHODS: Data from the Health Care Financing Administration 2728 form were used to determine the hematocrit, erythropoietin use, serum albumin and glomerular filtration rate (GFR) among 4643 patients with failed kidney transplants who initiated dialysis between April 1995 and December 1998. RESULTS: At dialysis initiation, the mean hematocrit, serum albumin and GFR were 27.5%, 3.3 g/dL, and 8.4 mL/min/1.73 m2, respectively, and only 35% of patients had received erythropoietin. In a multivariate analysis, patients <45 years, females, patients of non-White race, non-diabetic patients, hemodialysis patients and patients with a lower serum albumin had a higher odds for hematocrit <30%. Erythropoietin use was associated with female gender, White race, increased time since transplantation, being employed, peritoneal dialysis, and higher serum albumin. Patients >/=45 years and patients with diabetes or congestive heart failure had higher odds for hypoalbuminemia, while employed patients, peritoneal dialysis patients, patients with higher hematocrit and patients who had received erythropoietin had lower odds of hypoalbuminemia. CONCLUSIONS: Despite being known to specialty physicians, patients with failed kidney transplants initiate dialysis at levels of hematocrit, serum albumin, and GFR that may be suboptimal and similar to those in the general incident ESRD population. Socioeconomic factors remain important barriers to the provision of CKD care even among these patients with established specialty physician contact. Improved CKD care could improve the outcomes of this unique subgroup of CKD patients.     

Amyloid urinary-tract calculi in patients on chronic dialysis

Urinary calculi found in 4 patients on chronic hemodialysis or continuous ambulatory peritoneal dialysis (CAPD) were identified as protein calculi by infrared spectroscopic analysis. Positive Congo red staining and immunological assessment revealed that the calculi were composed of amyloid protein derived from beta 2-microglobulin. A comparison of the patients who excreted calculi with 10 patients on chronic dialysis without urinary calculi showed no significant differences in the urinary and serum levels of beta 2-microglobulin. The mechanism of amyloid calculus formation may involve factors independent of the concentration of beta 2-microglobulin in urine or serum. Urinary calculi found in patients on chronic hemodialysis or CAPD were composed of amyloid protein derived from beta 2-microglobulin.     

A study of beta 2-microglobulin skin deposits in dialyzed patients and healthy controls

This study reports on beta 2-microglobulin (beta 2M) deposits in the skin of 12 uremic patients and three kidney transplant recipients compared with eight healthy controls. Uremic patients were treated by hemodialysis (HD), hemofiltration (HF), hemodiafiltration (HDF), or continuous ambulatory peritoneal dialysis (CAPD) for a period lasting from 1 to 19 years. Congo red staining of the skin was negative in patients and controls. However, immunofluorescent staining with an anti-beta 2-microglobulin monoclonal antibody was positive in the skin of all patients and of six of the eight controls. Beta 2M skin deposition is more intense in patients than in controls and increases with patient age and the duration of dialysis. A stron correlation is observed between the extent of skin beta 2M deposits and clinical manifestations due to beta 2M deposits in internal organs. However, no correlation is found between beta 2M skin deposits and sex or beta 2M serum levels.     

Intraperitoneal production of erythropoietin with continuous ambulatory peritoneal dialysis

Higher hematocrit and serum erythropoietin (EPO) levels have previously been shown in end-stage renal disease patients treated with continuous ambulatory peritoneal dialysis (CAPD) compared with hemodialysis. We investigated whether EPO was produced intraperitoneally in CAPD patients. EPO concentration was 3.5±0.3 mU/ml by radioimmunoassay in 26 samples of peritoneal dialysis effluent obtained from 15 CAPD patients. EPO was not detectable in the fresh unused dialysate. No correlation was observed between EPO levels in the serum and dialysis effluent. Peritoneal macrophages were isolated from the dialysis effluent of 9 CAPD patients after an overnight dwell. The culture supernatant obtained after 24 h of in vitro culture of a million cells yielded EPO of 3.5±0.3 mU/ml. Our study demonstrated that peritoneal macrophages from CAPD patients produce EPO on in vitro stimulation, and EPO is present in the dialysis effluent of CAPD patients.     

Clinical, radiological and serum amyloid P component scintigraphic features of beta2-microglobulin amyloidosis associated with continuous ambulatory peritoneal dialysis.

BACKGROUND: Beta2-Microglobulin (beta2M) amyloidosis occurs in patients with end-stage renal failure (ESRF) who undergo long-term continuous ambulatory peritoneal dialysis (CAPD), but its prevalence in patients treated exclusively by CAPD is unknown. In addition, its features may differ from those of haemodialysis-associated beta2M amyloidosis because CAPD is more biocompatible. METHODS: We performed serum amyloid P component (SAP) scintigraphy, a specific technique for imaging amyloid deposits, in 13 consecutive patients with ESRF who had been dialysed for >5 years, at least 80% of the time by CAPD. Clinical and radiological features of beta2M amyloidosis were sought and compared with the results of SAP scintigraphy. RESULTS: SAP scans showed articular amyloid deposits in seven patients, all of whom had evidence of carpal tunnel syndrome and four of whom had arthralgia characteristic of dialysis amyloidosis. Typical radiographic bone cysts were present in only one case who had been dialysed for >17 years. The remaining six patients had no clinical, radiological or scintigraphic evidence of beta2M amyloidosis. CONCLUSIONS: The prevalence of beta2M amyloidosis in this study was comparable with that in reported haemodialysis populations. Many of the amyloid deposits demonstrated by SAP scintigraphy were not associated with symptoms, but larger and longer term studies are required to determine whether CAPD favourably influences their clinical expression.     

Serum erythropoietin levels and hematocrit in end-stage renal disease: influence of the mode of dialysis

Serum erythropoietin (Ep) levels were measured by radioimmunoassay in 70 patients with end-stage renal disease (ESRD) to evaluate the influence of the mode of dialysis on the relationship between serum Ep levels and the severity of anemia. Thirty-five patients were on hemodialysis (HD), seven were on intermittent peritoneal dialysis (IPD), and 28 were on continuous ambulatory peritoneal dialysis (CAPD). Compared to HD, CAPD patients had higher serum Ep (CAPD), 46.1 +/- 13.4 v HD, 16.9 +/- 2.2 mU/mL) and hematocrit (CAPD, 33.9 +/- 2.5 v HD, 24.8 +/- 1.4%; P less than 0.05). The Ep and Hct values for IPD patients were intermediate between the other two groups. Serum Ep levels were higher in CAPD patients in the first 4 weeks of initiation of CAPD (144 +/- 35 mU/mL, n = 6) than later (39 +/- 6.4 mU/mL, n = 24). A significant fluctuation in serum Ep and Hct values was noted in patients on all three modes of dialysis, when multiple samples were obtained at different time intervals. There was a weak correlation between serum Ep and Hct in the three groups of dialysis patients; r = 0.36, P less than 0.005. The data suggest that CAPD provides a better biochemical milieu for Ep production and responsiveness than HD treatment of ESRD.     

QT dispersion in hemodialysis and CAPD patients

Prolongation of repolarization dispersion (QT interval dispersion) measured from the 12-lead surface ECG has been associated with sudden cardiac death and ventricular tachyarrhythmias in a variety of cardiac disorders. The aim of our study was to assess the effects of hemodialysis (HD) and continuous ambulatory peritoneal dialysis (CAPD) on QT dispersion in end-stage renal disease patients. 20 chronic HD patients (mean age 57.75 +/- 13.79 years) and 20 CAPD patients (mean age 50.79 +/- 14.94 years) who had no complaints and symptoms of cardiac arrhythmias as well as 20 healthy volunteers (mean age 48.74 +/- 10.88 years) underwent ECG testing. All HD patients were on bicarbonate three times weekly with cuprophane capillaries. 12-lead ECGs were recorded on the day after HD. The CAPD patients were on a standard CAPD program (four times daily with 2,000 cm(3) peritoneal fluid). ECGs were recorded when the patients were receiving their regular standard CAPD program. All ECGs were analyzed manually by one observer. There were no statistically significant differences in dialysis duration, blood urea nitrogen, creatinine, sodium, calcium, and parathormone values between the HD and CAPD patients. The serum potassium values were significantly higher in HD patients when compared to CAPD patients. There was no difference in the mean of maximal QT among all three groups. The rate of QT interval dispersions was significantly higher in HD and CAPD patients as compared with healthy controls (p < 0.05). There was no statistically significant difference in the QT dispersion rates between HD and CAPD patients. In conclusion, there is a tendency to cardiac arrhythmias in HD patients during the postdialysis period. Although CAPD patients are receiving dialysis daily, they also have higher rates of QT dispersions and accordingly a tendency to arrhythmias.     

Leptin in CAPD patients: serum concentrations and peritoneal loss.

BACKGROUND: To determine whether serum leptin concentrations in patients undergoing continuous ambulatory peritoneal dialysis (CAPD) are influenced by peritoneal loss of leptin and to compare serum leptin levels of normal subjects with those of patients receiving renal replacement therapy such as haemodialysis (HD), CAPD, or kidney transplantation. SUBJECTS AND METHODS: Eighty-four individuals were investigated: six females and 14 males on standard CAPD; 13 females and 13 males on chronic HD; 10 female and eight male kidney transplant recipients, and 10 female and 10 male subjects as controls. Morning serum, 8-h and 24-h samples of peritoneal fluid concentrated to 6-20-fold by Centricon 3 (cutoff 3000 daltons), and 24-h urinary concentrations of leptin were measured with commercial RIA (Linco Research, Inc., USA). Venous blood and peritoneal fluid samples of albumin, beta2-microglobulin, glucose, urea, and creatinine were determined by standard laboratory techniques. Serum insulin levels were measured by radioimmunoassay. RESULTS: Patients (men and women) on CAPD and after kidney transplantation exhibited significantly higher serum concentrations of leptin and leptin/BMI ratios than control subjects. These increased values did not reach statistical significance in HD patients. Serum leptin concentrations were correlated very significantly with BMI in all cases (r=0.380, P<0.001). Moreover, in CAPD patients (r=0.630, P<0.007) and in HD patients (r=0.668, P<0.005), but not in kidney transplant recipients or control subjects, significant correlations were observed between serum leptin and insulin concentrations. Residual renal function (RRF) in the range 0-12.8 ml/min and serum beta2-microglobulin levels in the range 7.9-47.1 mg/l did not influence serum leptin levels in CAPD and HD patients. As expected, leptin was detected in the peritoneal fluid of CAPD patients. Twenty-four-hour peritoneal loss (30.95+/-21.05 ng/min) and 24-h peritoneal clearance (0.01+/-0.01 ml/kg/min) of leptin account for only 3.9% of estimated whole-body leptin production rate and 0.7% of leptin clearance from plasma respectively. Twenty-four-hour urinary losses of leptin in CAPD patients were negligible, accounting for 5.6+/-1.8% (range 0.3-15.2%) of total (peritoneal and urinary) loss of this hormone. CONCLUSIONS: These findings suggest that serum leptin levels are not affected by continuous peritoneal loss of leptin during CAPD and that insulin resistance and hyperinsulinaemia contribute to elevated serum leptin concentrations in CAPD and HD patients. The aetiology of increased serum leptin levels in kidney transplant recipients is probably different from that in dialysis patients.     

The effect of renal replacement therapies on serum gastrointestinal system hormones

BACKGROUND: The kidney is a major site for the inactivation, degradation, and clearance of a variety of peptide hormones. It has been shown that the uremia increases or decreases gastrointestinal system (GIS) hormones. Moreover, studies investigating the serum GIS hormones levels in chronic renal failure (CRF) were conducted mainly in a particular period of the renal replacement therapy, and the changes caused by continuous ambulatory peritoneal dialysis (CAPD) and hemodialysis (HD) could not be fully demonstrated. In this study, we investigated the effect of CAPD and HD on serum GIS hormones (amylase, lipase, trypsinogen, and gastrin) levels in CRF patients who were diagnosed for the first time. METHODS: Serum amylase, lipase, trypsinogen, and gastrin levels were measured in 36 patients who were just diagnosed with CRF, 22 patients with CAPD and 14 patients with HD. GIS hormones of these patients were measured before treatment and three months from the beginning of CAPD and HD treatment. As the control group, 20 normal healthy cases with well-matched age and gender were used. RESULTS: The mean serum amylase, lipase, secretin, and gastrin levels were found meaningfully decreased according to the beginning values at third months of the CAPD and HD treatment. However, they were higher than control group. CONCLUSION: In patients receiving CAPD or HD as renal replacement therapy, GIS hormone levels were found to be lower, albeit higher than the healthy control group.     

Deposition and removal of cutaneous beta 2-microglobulin.

These studies were designed to track the cutaneous deposition of beta 2-microglobulin (beta 2M) in patients on chronic hemodialysis, patients with chronic renal insufficiency and patients with successful renal transplants. Immunoperoxidase staining of skin biopsies from dialysis patients demonstrated significantly increased amounts of beta 2M compared to controls (p < 0.01). There was a strong positive correlation between the skin beta 2M content and the years of dialysis treatment. Renal transplant recipients had decreased skin content of beta 2M compared to hemodialysis patients. There was no difference in the skin beta 2M content in patients with chronic renal insufficiency not on hemodialysis and controls. No dialysis patient had amyloid in the skin by Congo red stain. We conclude that beta 2M accumulates in the skin of patients on chronic hemodialysis. This beta 2M is not in the form of amyloid. Successful renal transplantation allows for the removal of beta 2M from the skin indicating that beta 2M not in the form of amyloid can be mobilized from tissue sites.     

Identical decline of residual renal function in high-flux biocompatible hemodialysis and CAPD.

BACKGROUND: Patients on conventional hemodialysis lose residual renal function more rapidly than patients on continuous ambulatory peritoneal dialysis (CAPD). The effect of dialysis using synthetic membranes and ultrapure water is less clear. METHODS: The decline of urea clearance was compared in a cohort of 475 incident end-stage renal failure patients who received treatment with CAPD (N=175) or hemodialysis (HD) utilizing high-flux polysulphone membranes, ultrapure water, and bicarbonate as the buffer (N=300). RESULTS: CAPD patients were significantly younger, fitter (lower comorbidity severity score), less dependent (higher Karnofsky performance score) and less likely to have presented late than HD patients. There was no difference in the mean urea clearance in each group at dialysis initiation, or at any 6-month time point during the ensuing 48 months. This was true even after exclusion of patients who had died in the first year after initiation, those transferred to another dialysis modality, or those who had been transplanted. Only age and chronic interstitial disease predicted retention of urea clearance at one year. The rate of decline of urea clearance was similar in pre- and post-dialysis initiation phases, though there may have been a step-decline of about 2 mL/min at initiation, which requires further investigation. CONCLUSIONS: In hemodialysis using high-flux biocompatible membranes and ultrapure water, residual renal function declines at a rate indistinguishable from that in CAPD. This may have important implications, since preservation of residual renal function has major benefits and is a valid therapeutic goal.     

Circulating form of beta-2-microglobulin in dialysis patients

The circulating profile of beta-2-microglobulin (beta 2M) was determined in 8 end-stage renal disease patients on long-term dialysis (6 on hemodialysis, 2 on CAPD) by measuring beta 2M in different fraction after molecular sieve separation of their sera. Four patients had carpal tunnel syndrome with demonstrated amyloid in excised wrist tissues of which 2 were positive for beta 2M. In all patients despite very high blood levels (34.3-63.1 mg/l), beta 2M eluted exclusively as a single peak in the molecular weight region of about 12,000 daltons on a calibrated Sephacryl S-200 column. Recoveries from within the peak accounted for 96% of the applied beta 2M serum concentrations. These results were confirmed by molecular sieve separation of the enriched beta 2M-containing fractions by high-pressure liquid chromatography. We conclude that immunoreactive beta 2M in dialysis patients circulates as an intact monomer without evidence for the formation of aggregates or fragments. The pathogenesis of tissue deposition of this low-molecular-weight protein and its polymerisation to form a specific amyloid remains to be defined.     

Quality of life in end-stage renal disease: a reexamination

A self-administered questionnaire assessing both objective and subjective quality of life was completed by 489 end-stage renal disease (ESRD) patients in a representative sample of an entire network. Patients differed in both objective and subjective quality of life when examined as a function of treatment modality. The quality of life is similar for successful transplant and home hemodialysis patients; these patients appear to fare better than other treatment groups on both objective and subjective measures. Patients receiving staff-assisted center hemodialysis and continuous ambulatory peritoneal dialysis (CAPD) report markedly diminished quality of life; these decrements remained after statistically controlling for nontreatment variables. Diminished quality of life was most pronounced in dialysis patients who had experienced failed transplants. All treatment groups showed some objective losses, especially loss of employment, but patients in the best rehabilitated treatment groups showed near-normal subjective quality of life. The results confirm previous reports that the subjective quality of life of ESRD patients can be nearly normal despite objective losses, but demonstrate that inadequate definition of treatment groups has led to misperceptions about the impact of transplant failure.     

Serum hyaluronic acid and procollagen III amino terminal propeptide in chronic renal failure

Elevated serum concentrations of hyaluronic acid (HA) and procollagen III amino terminal propeptide (PIIINP) have been found in various diseases characterized by altered metabolism of collagen. In the present study, their serum levels were measured in 105 renal patients and 22 normal controls. Median HA concentrations were 23 micrograms/l in controls, 47 micrograms/l in patients with chronic renal failure (CRF, not on dialysis; p less than 0.001), 75 micrograms/l on CAPD (p less than 0.001) vs. controls, p = 0.045 vs. CRF), and 167 micrograms/l on hemodialysis (p less than 0.001 vs. controls, CRF, and CAPD), respectively. The values correlated positively with age but not with renal function or the type of renal disease. In hemodialysis patients, HA correlated with the duration of renal replacement therapy and serum beta 2-microglobulin but not with serum alkaline phosphatase or C-terminal parathormone. Serum HA did not change significantly during hemodialysis treatment and was independent of the type of dialyzer membrane material. Median PIIINP values were 2.7 micrograms/l in controls, 4.4 micrograms/l in patients with CRF (p less than 0.001), 6.9 micrograms/l on CAPD (p less than 0.001 vs. controls, p = 0.022 vs. CRF), and 8.6 micrograms/l on hemodialysis (p = 0.001 vs. controls, NS vs. CRF or CAPD). Values correlated with HA only in patients on CAPD but they did not correlate with age, renal function or duration of renal replacement therapy. It is concluded that renal failure, especially long-term dialysis treatment, is associated with elevated serum concentrations of HA and--to a minor degree--PIINP. Thus, they may be a sign of altered connective tissue metabolism in patients on long-term dialysis.     

Autonomic function in uremic patients treated by hemodialysis or CAPD and in transplant patients

Tests of autonomic nerve function have been performed in patients receiving dialysis and following transplantation. These tests, consisting of blood pressure and heart rate response to standard stimuli were carried out in four groups of subjects: 10 patients were receiving continuous ambulatory peritoneal dialysis (CAPD), 12 hemodialysis, 11 had functioning transplants and there were 12 healthy subjects. The heart rate responses to the Valsalva maneuver, deep breathing and standing were equally reduced in the hemodialysis and CAPD patients while the responses in the transplant patients were not significantly different from those in the control subjects. Peroneal nerve conduction velocity was used to measure the degree of peripheral neuropathy and while in most patients there was a positive relationship with the autonomic studies, in five patients the results were clearly dissociated. Plasma parathyroid hormone (PTH) levels were elevated in both the dialysis groups but not in the transplanted patients and there was no correlation between the PTH levels and any of the neurological tests. This study has confirmed the presence of autonomic neuropathy in dialysed uremic patients and has demonstrated almost complete resolution of this following transplantation but has failed to show any benefit of CAPD over hemodialysis in protecting against neuropathy.     

The Effect of Renal Replacement Therapies on Serum Gastrointestinal System Hormones

Background. The kidney is a major site for the inactivation, degradation, and clearance of a variety of peptide hormones. It has been shown that the uremia increases or decreases gastrointestinal system (GIS) hormones. Moreover, studies investigating the serum GIS hormones levels in chronic renal failure (CRF) were conducted mainly in a particular period of the renal replacement therapy, and the changes caused by continuous ambulatory peritoneal dialysis (CAPD) and hemodialysis (HD) could not be fully demonstrated. In this study, we investigated the effect of CAPD and HD on serum GIS hormones (amylase, lipase, trypsinogen, and gastrin) levels in CRF patients who were diagnosed for the first time. Methods. Serum amylase, lipase, trypsinogen, and gastrin levels were measured in 36 patients who were just diagnosed with CRF, 22 patients with CAPD and 14 patients with HD. GIS hormones of these patients were measured before treatment and three months from the beginning of CAPD and HD treatment. As the control group, 20 normal healthy cases with well-matched age and gender were used. Results. The mean serum amylase, lipase, secretin, and gastrin levels were found meaningfully decreased according to the beginning values at third months of the CAPD and HD treatment. However, they were higher than control group. Conclusion. In patients receiving CAPD or HD as renal replacement therapy, GIS hormone levels were found to be lower, albeit higher than the healthy control group.     

Exposure of patients to phthalates from polyvinyl chloride tubes and bags during dialysis

The exposure of patients to the plasticizer di-(2-ethylhexyl)phthalate (DEHP) from polyvinyl chloride (PVC) tubes and bags during dialysis has been investigated. In vitro studies of the migration of DEHP from hemodialysis tubes into plasma revealed a migration coefficient of 7.7 micrograms/ml/h. An artificial kidney did not influence the plasma concentration of DEHP. The calculated exposure from a single hemodialysis can be as much as the total annual exposure using continuous ambulatory peritoneal dialysis (CAPD). CAPD fluids in PVC bags contain low concentrations of DEHP and its hydrolysis product mono-(2-ethylhexyl)phthalate (MEHP). These phthalates were analyzed in serum from patients receiving both CAPD and hemodialysis using gas chromatography. A group of patients not yet on dialysis treatment was used as control. In no case could MEHP be detected. The DEHP concentration was 0.8-4.2 micrograms/ml serum in the 17 hemodialysis patients after dialysis and 0.1-0.9 microgram/ml in 4 of the CAPD patients. In 3 of the CAPD and all of the predialysis patients, DEHP could not be detected (less than 0.1 microgram/ml).     

Glucose concentration in the dialysate does not contribute to lipid profiles in patients undergoing CAPD

The aim of this study was to investigate lipid profiles in patients with end-stage renal disease receiving hemodialysis (HD), continuous ambulatory peritoneal dialysis (CAPD), or no dialysis (nondialytic treatment group, NT), and to analyze the association between dyslipidemia in CAPD patients with glucose-containing dialysate dosages. Lipid profiles were determined in 64 NT patients, 62 HD patients, and 180 CAPD patients at a single time point. NT patients' samples were collected following hospitalization due to renal failure. HD and CAPD patients' samples were collected after 3 months of dialysis. The association between lipid profiles of 180 CAPD patients and glucose-containing dialysate was analyzed using Pearson methods; 76.56% of NT patients, 66.13% of HD patients, and 72.22% of CAPD patients had dyslipidemia. Compared with NT patients, CAPD patients had significantly altered levels of cholesterol, triglycerides, high-density lipoprotein, apolipoproein (Apo)-A1, and Apo-E (p < 0.05), but unchanged levels of low-density lipoprotein or Apo-B. There was no correlation between the three different concentrations of glucose in the dialysate with the lipid profile of CAPD patients. We concluded that patients on CAPD exhibit dyslipidemia, and that different concentrations of glucose in the dialysate do not affect lipid profiles in these patients.     

Role of erythropoietin in the reversal of anemia of renal failure with continuous ambulatory peritoneal dialysis

We report serum erythropoietin levels in a patient who showed significant improvement in hematocrit when switched from hemodialysis to continuous ambulatory peritoneal dialysis (CAPD) treatment. This 22-year-old woman had severe anemia and low serum immunoreactive erythropoietin levels (8.0 +/- 1.2 mU/ml; n = 5) while on hemodialysis for 7 years. Serum erythropoietin levels were 80 and 177 mU/ml, 2 and 3 weeks, respectively, after starting CAPD. This was followed by an increase in reticulocyte count from 3.9 to 22% and hematocrit from 19 to 48%. The serum erythropoietin concentration obtained on CAPD treatment (62.7 +/- 15.2 mU/ml; n = 9) was significantly higher than that obtained on hemodialysis. Our findings indicate that CAPD facilitates increased erythropoietin production compared to hemodialysis and that the anemia of uremia may reverse if sufficient erythropoietin is available.