- Email: [email protected]
Home nocturnal hemodialysis in children
HOME NOCTURNAL HEMODIALYSIS IN CHILDREN DENIS F. GEARY, MB, MRCP(UK), FRCP(C), E. PIVA, RN, J. TYRRELL, RN, M. J. GAJARIA, BSC, CDP, G. PICONE, MSW, L. E. KEATING, BA, MSC, AND E. A. HARVEY, MD, FRCP(C)
Objective To describe the effect of home nocturnal hemodialysis (NHD) in North American children. Study design Four teenagers underwent NHD for 8 hours, 6 to 7 nights/week, using either central venous lines or fistulae for periods of 6 to 12 months. Blood flow approximated 200 mL/min, and dialysate flow was 300 mL/min; the dialysate contained potassium and phosphate. The procedure was remotely monitored.
Results The children had unrestricted diets and fluid allowance and did not require phosphate binders. Persistent relative hypotension developed in 2 of 4 children. Weekly Kt/V urea values were consistently >10; other biochemical measures varied. Quality of life and school attendance improved in 3 of 4 children. The workload and reported emotional burden of NHD was substantial. No significant complications occurred. Dialysate losses of calcium, phosphate and carnitine required supplementation. The annual cost per patient was $64,000 Canadian, which represented a 27% savings compared with thrice weekly incenter hemodialysis. Conclusions NHD is feasible in selected children, allows free dietary and fluid intake, and improves patient wellbeing. The burden on the family is substantial, and NHD requires support of a dedicated multidisciplinary team. (J Pediatr 2005; 147:383-7) emodialysis (HD) was first introduced as a maintenance therapy for children with chronic renal failure in 1968.1 Approximately a decade later, continuous ambulatory peritoneal dialysis was introduced.2 Since then, peritoneal dialysis (PD) has become the most widely used dialysis modality for North American children.3 The use of hemodialysis as a home nocturnal treatment (NHD) was developed for adults in 1994,4,5 and the use of this treatment has expanded in the adult renal failure population, in whom it reportedly improves dialysis delivery and patient wellbeing, at reduced cost.6-12 We describe our experience with NHD in 4 children at The Hospital for Sick Children in Toronto, including clinical, psychosocial, and economic outcomes.
H
PATIENTS Patient 1 Patient 1 was a 13-year-old male with end-stage renal disease caused by focal segmental glomerulosclerosis who started PD at age 4 years, and subsequently underwent bilateral nephrectomy and renal transplantation at age 8. At age 11, because of recurrent disease in the graft and chronic allograft nephropathy, he underwent graft nephrectomy and restarted PD. He had recurrent peritonitis and 2 episodes of pancreatitis, which necessitated a switch to HD. On HD, he remained malnourished, had multiple central venous access clots and infections, was markedly hypertensive and repeatedly fluid overloaded, and was unable to comply with dietary and medication prescriptions. In June 2002, he started NHD.
Patient 2 Focal segmental glomerulosclerosis developed in this 15.5-year-old female patient at age 5 years, and she started maintenance PD 3 years later. In 1996, she received a cadaveric AV CVL HD
Arteriovenous Central venous lines Hemodialysis
NHD PD
Nocturnal hemodialysis Peritoneal dialysis
From the Division of Nephrology, Department of Paediatrics, The Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada. Supported by a grant from the CHANGE foundation. Submitted for publication Nov 10, 2004; last revision received Apr 7, 2005; accepted Apr 14, 2005. Reprint requests: Denis F. Geary, MB, MRCP(UK), FRCP(C), Chief, Division of Nephrology, The Hospital for Sick Children, 555 University Ave, Toronto, ON M5G 1X8. E-mail: denis. [email protected]. 0022-3476/$ - see front matter Copyright ª 2005 Elsevier Inc. All rights reserved. 10.1016/j.jpeds.2005.04.034
383
renal transplant, which was complicated by recurrence of her original disease, development of seizures and facial palsy, chronic allograft nephropathy, and malnutrition. In 1998, she restarted cycling PD and had a gastrostomy tube inserted for supplemental feeds. She was referred to our hospital in 2003 because of increasing morbidity associated with ultrafiltration failure, despite approximately 20 hours daily of cycling PD. Because of her distance from the hospital, the family elected to start NHD.
This 16-year-old boy developed Wegener’s granulomatosis with crescentic glomerulonephritis at age 15 years. He progressed to end-stage renal disease and started PD. The family requested a switch to NHD after 6 months because of peritonitis and severe hypertension.
Patient 4 Hemolytic uremic syndrome developed in this 12.5year-old girl with end-stage renal disease and severe hypertension at age 3 years, requiring bilateral nephrectomies. In 2001, she underwent a living-donor kidney transplant, but chronic allograft nephropathy and end-stage renal disease developed. Because of geographic location and fears of increasing obesity if she was treated with PD, the family elected to start NHD.
METHODS The technical considerations, including home renovations, dialysis machines and water purification units, the possible need for water softeners, and the remote monitoring process, were previously reported for our program and for adults.13,14 Vascular access was provided initially with central venous lines (CVL) in all 4 patients and, subsequently, arteriovenous (AV) fistulae in 2 patients. During NHD, central venous lines are held in place with a locking device described by Pierratos.7 Access through AV fistulas has been with both single-needle and 2-needle techniques; we used a ‘‘buttonhole’’ technique by repeated insertion of the access needles at the same location to form a needle-track,15 which facilitated access by the care provider at home. The parent supervisor, patient, or both was taught this technique for access of a fistulae by the teaching nurse. Dialysate flows of 200 to 300 mL/min were used, with blood flows of 150 to 200 mL/min. Dialysis was provided for approximately 8 hours nightly, 6 to 7 times per week, usually while the patient was sleeping. The dialysate contained potassium at a concentration of 2 to 3 mEq/L. Calcium content of the dialysate was adjusted between 3.0 and 3.5 mEq/L, because chronic negative calcium balance may develop with large ultrafiltration volumes. Phosphate was added to the acid dialysate concentrate as a sodium phosphate (Fleet) enema, which provided a concentration approximating 3 mg/dL. Geary et al
The NHD program was a collaborative effort between the nurses, technologist, social workers, dieticians, and medical staff. The nursing commitment was substantial and demanded expertise in HD and family education. Collaboration with Fresenius Medical Care greatly facilitated the process because of their staff expertise and provision of teaching materials.
Cost Analysis
Patient 3
384
Staff Requirements
Estimation of program costs, expressed in Canadian (Cd) dollars, included: home renovations, including plumbing, electrical, cost of water testing, and installation of phone lines, were measured as actual costs and amortized over 1 year. Costs for the 6-week training period were itemized prospectively and amortized over 1 year. Equipment, including dialysis machines, water softeners, reverse osmosis machines, computer hardware, and home centrifuge, were itemized individually and amortized over 5 years. Dialysis disposables/surgical supplies were individually itemized and included as actual costs. Staff costs were prospectively itemized and are presented as the mean for the 4 patients; nursing costs for ongoing patient care were calculated with a nurse-to-patient ratio of 1:8, with training costs extra. Program development costs (including initial supplies, teaching materials, staff time) are provided separately. No costs are included for physicians or medication.
Psychosocial and Clinical Evaluations Psychosocial evaluations were performed by a social worker (G.P.) on the basis of comments provided by the children, their parents, or both, an analysis of the number of clinic visits for 6 months preceding and subsequently during each 6 months while undergoing NHD, and a record of school attendance. Quality of life was estimated using the PedsQL version 4.0 generic core scale.16 This 23-item self-report questionnaire includes 8 items to determine physical functioning; psychosocial functioning is evaluated by 5 questions each related to emotional, social, and school-functioning. Responses to all questions are scored to provide a total of 0 to 100, with higher scores indicating a better health-related quality of life. A large number of clinical and biochemical tests were monitored to evaluate efficacy of dialysis and clinical wellbeing, and results are reported in tabular form. Single pool Kt/V urea was estimated with the Daugirdas logarithmic formula.17 Clinical and biochemical outcomes were collected at baseline and 3, 6, and 12 months. Results are presented in descriptive form, because only 4 patients are included. Comparative statistical analysis is not provided. The NHD program was initiated for clinical care reasons. However, because data were collected prospectively and there was intent to publish our experience, Research Ethics Board approval was obtained before patients started the program. Also, because these patients are a small group that might be identifiable, written consent was obtained to include the data in the case reports. The Journal of Pediatrics September 2005
RESULTS Patient Selection The 4 families who met eligibility criteria agreed to participate in the program and completed the training requirements. No families have been denied access to this program. Clinical and biochemical outcomes are reported at 3 and 6 months for all patients; 12-month data are available for patients 1 and 2. Patient 1 received NHD for 1 year before switching to a hybrid form of dialysis of NHD Sunday to Wednesday nights inclusive, with an in-center HD on Friday for respite purposes. There have been no dropouts from the program, and no patient deaths.
Technical and Vascular Access Evaluations TECHNICAL. Few technical problems were encountered. Remote monitoring was used for all procedures, but no phone calls to patient homes were required because of a failure to respond to an alarm in timely manner. One family was called when an air detect alarm was noted by the monitor. VASCULAR ACCESS. Episodes of CVL dysfunction were treated with interdialytic instillation of intraluminal tissue plasminogen activator. One CVL infection (Staphyloccoccus aureas) was documented in patient 1 when he fell in a lake at camp, and 1 tunnel infection was observed in patient 2 (diphtheroid organism). The frequency of these episodes was not considered different than in our general HD population. The use of the ‘‘buttonhole’’ technique to access a fistula repeatedly through the same site was associated with the development of an aneurysm at the site, but to date this has not interfered with its continued use. No line disconnections were reported. STAFF REQUIREMENTS. Patients and their parents were trained for 6 weeks: 3 in-center 6-hour sessions per week for 5 weeks, followed by 1 week (3-5 sessions) of nights at the hospital, where the parents are required to perform all duties; the nurse is also available overnight. In addition, parents were required to write 7 tests, with a passing grade considered to be >80% for each examination. Ongoing monitoring of safety and efficacy includes remote monitoring, weekly blood tests, and monthly clinic visits. COST EVALUATION. The costs for development of the NHD program were Cd $4200 (approximately $3400 in US dollars). Patient/family training cost averaged Cd $9000 for the 6 weeks, as aforementioned. The costs of the training are amortized over 1 year and included in the overall annual cost estimate of Cd $64000.
Psychosocial and Clinical Evaluations Variability in many of these outcomes is thought to reflect the diversity, complexity, and multiplicity of the problems that were present before initiation of NHD in these patients. Consistent findings among the 4 patients included normal plasma calcium and phosphate values, increased singleHome Nocturnal Hemodialysis In Children
pool Kt/V urea measurements and high b2-microglobulin levels (Table I; available online at www.jpeds.com). Bone density remained poor in patients 2 and 4, and deteriorated in patients 1 and 3, which suggests the need to avoid sustained negative calcium and phosphate balance. Both free and total plasma carnitine values were below reference range in all patients, until supplemented. The clinical significance of this finding is uncertain, although it might represent a deficiency caused by solute clearance. Many inconsistencies were also observed in outcome measures. The rise in PTH levels in patient 1 between 6 and 12 months could be caused by frequent interruption to his NHD schedule and line function problems during this time (Table I). This patient had a fistula created and transferred to hybrid dialysis after 12 months for respite. Increased CRP levels were also noted in 3 of the 4 patients; the significance of this is unknown. Information on medications for blood pressure, anemia (erythropoietic therapy), phosphate binders and carnitine supplements are presented in Table II (available online at www.jpeds.com). Although all the children had unrestricted diets and fluid intakes, weight gain was variable. Requests for accurate documentation of dietary intake were not regularly adhered to, so the effect of frequent dialysis on nutritional state remains unclear. However, for patient 2, who was completely G-tube dependant at the start of NHD, it was documented that her oral intake increased to 25% to 50% of her dietary requirements. Also, when weight loss has occurred, it is unclear whether this is in part caused by increased exercise/activity. Normalized protein catabolic rate was not measured. Blood pressure control was also variable. Patients 2 and 3, with native kidneys in situ, required anti-hypertensive medication, whereas relative hypotension developed and persisted in patients 1 and 4. These latter 2 anephric children received Midodrine before dialysis to prevent exacerbation of their hypotension. Quality of life (Table I; online at www.jpeds.com) physical and psychological scores improved in 3 of 4 patients. Patient 1, who had a decreased quality of life score between 6 and 12 months, required respite relief and started hybrid dialysis at 12 months. School attendance, as a reflection of psychosocial behavior, improved substantially in 3 children, and teacher reports included improved interaction with peers and improved performance overall. Patient 3 had persistent and significant numbers of days missed from school related to complications of his underlying Wegener’s granulomatosis and the associated immunosuppressant treatment. Specific comments from the children and parents included: ‘‘Because this program has changed my life completely, by making the quality of my life ten times better.’’; ‘‘I feel like a normal kid’’; ‘‘She can’t remember the last time she felt so good’’; ‘‘I used to push her around the block in a wheelchair, and now she is canvassing doorto-door to raise money for her school.’’ Negative comments about NHD reflected the intensity of the workload and parental anxieties related to the 385
complexity of care provided at home. The mother of patient 1 was psychologically and emotionally worn out, until respite was provided when a hybrid form of dialysis was instituted. Others described disruption for other family members and difficulty establishing a new routine within the home.
DISCUSSION This report describes the successful implementation of NHD for children. Although introduced approximately 10 years ago for adults, the only previous pediatric experience with NHD, published in abstract form from Sweden,12 described 4 children with ages similar to our own patients. They used CVL as blood access, whereas we have also successfully used AV fistulas. Overall, their brief report suggests that outcomes are comparable to our own, and that NHD provides an improvement in patient well-being compared with other dialysis modalities.12 Consequently, we recommend that this modality be considered for all teenagers, and some younger school-age children starting dialysis. The success of NHD is perhaps best exemplified by the continuation of the procedure in all patients. No patients have died, although the severity of uremia and its complications when starting NHD was extreme for patients 1 and 2. The Hemo Study for hemodialysis patients18 and the ADEMEX Study of peritoneal dialysis patients19 suggested that increasing the clearance of urea as measured with Kt/V urea, beyond certain limits, does not produce further improvement in patient outcomes. However, we do not believe that the results of these studies are applicable to NHD. Rather, the Hemo Study results suggest that for intermittent therapies, there are limitations to what can be achieved; the Hemo Study can be interpreted to suggest that more frequent dialysis, allowing for less interdialytic accumulation of toxins, may be the only remaining way to improve outcomes for patients undergoing HD. Data in small numbers of patients receiving NHD suggest that long-term survival is improved when compared with conventional dialysis modalities.8 As an alternative to NHD, the introduction of short daily dialysis at home may be considered. Short daily dialysis has the advantage that it can be provided during the daylight hours when the family is awake, and therefore remote monitoring will not be required. However, the amount of dialysis provided using a short daily regimen is less than that achieved by overnight dialysis, and because the set-up and tear-down time and supplies needed are the same for each modality, short daily dialysis is less cost-effective. The introduction of NHD to a patient’s home is a huge undertaking. However, the improvement in clinical status, school attendance and performance, which we have noted, and the small but consistent improvement in quality of life reported by our patients suggest that the outcomes justify the psychosocial burden that accompanies NHD. Significant concerns were expressed about the introduction of nocturnal cycling PD to patientsÕ homes.20 The same concerns can be expressed about the introduction of NHD. However, because 386
Geary et al
most children with end-stage renal disease are now treated with overnight cycling PD,3 it is reasonable to consider that with further experience and familiarity, NHD will become increasingly accepted if the improvements in patient well being that we report are substantiated by others. Remote monitoring of NHD was provided for all our patients. This provides a safety net and a psychological security blanket for patients and caregivers undertaking NHD. We recommend that some form of monitoring be in place for other patients starting this procedure, despite the fact that home monitoring is no longer recommended for all adult patients.8,21 However, it is also important to inform families that remote monitoring may be insufficient if a major dialysisrelated complication occurs. Our overall costs were estimated at Cd $63,670 per patient annually, which compares favorably with our in-center HD costs, both currently (Cd $88,000) and a decade ago (Cd $76,000).22 This represents a savings of 27% for each patient receiving NHD, compared with thrice-weekly in-center HD, and exceeds the savings reported in adult NHD.9,10 Although the cost of supplies is double that of conventional HD, savings still result because of the reduction in staff use. The savings are greater in children than adults because the nurse-to-patient ratio in the pediatric HD unit exceeds that of an adult unit. When the patients remain on NHD >1 year, our cost estimates would be lower because we amortized the costs of patient training, and home renovations over 1 year. If the equipment (eg, reverse osmosis and dialysis machines) is used for longer than 5 years, further savings would be realized. Finally, our estimate of appropriate nurse-to-patient ratio of 1:8 was arbitrary. Adult centers report a nurse-to-patient ratio of 1:20 for their home HD programs, but we feel that ratio will not be achievable for children because of the increased psychosocial component of pediatric care. The possibility that morbidity might result from excess dialysis must be considered for children receiving NHD. Phosphate and calcium losses in the dialysate have been documented,8 and supplements of each were added to the dialysate in our patients to prevent hypocalcemia and hypophosphatemia. Calcium deficiency may be compounded by the fact that, unlike patients receiving conventional dialysis, patients receiving NHD do not take calcium-containing phosphate binders. Sustained negative calcium balance, although not documented by us, potentially may adversely affect bones. A deficiency of carnitine, requiring supplementation, was documented in our patients. The potential for carnitine deficiency with frequent dialysis has been addressed in the report of the Carnitine Consensus Conference,23 although this concern was not mentioned in previous reports of NHD in adults. Prolonged deficiency of carnitine may result in muscle weakness, cardiomyopathy, intradialytic hypotension, and anemia and should be considered in patients receiving NHD who manifest these symptoms. The authors wish to thank Dr Andreas Pierratos from the Humber River Regional Hospital in Toronto and Michaelene Ouwendyk from Fresenius Medical Care for their assistance in the development of our Home Nocturnal Hemodialysis program. The Journal of Pediatrics September 2005
REFERENCES 1. Fine RN, De Palma JR, Lieberman E, Donnell GN, Gordon A, Maxwell MH. Extended hemodialysis in children with chronic renal failure. J Pediatr 1968;73:706-13. 2. Irwin MA, Balfe JW, Hardy B. Continuous ambulatory peritoneal dialysis in pediatrics. J Am Assoc Nephrol Nurses Tech 1981;8:11-3. 3. Leonard MB, Donaldson LA, Ho M, Geary DF. A prospective cohort study of incident maintenance dialysis in children: A NAPRTC study. Kidney Int 2003;63:744-55. 4. Uldall R, Francouer R, Ouwendyk M, Wallace L, Langos V, Ecclestone A, et al. Simplified nocturnal home hemodialysis (SNHHD): a new approach to renal replacement therapy. J Am Soc Nephrol 1994;5:428. 5. Uldall R, Ouwendyk M, Francouer R, Wallace L, Sit W, Vas S, et al. Slow nocturnal home hemodialysis at the Wellesley Hospital. Adv Ren Replace Ther 1996;2:133-6. 6. Lockridge RS Jr, Spencer M, Craft V, Pipkin M, Campbell D, McPhatter L, et al. Nocturnal home hemodialysis in North America. Adv Ren Replace Ther 2001;8:250-6. 7. Pierratos A. Nocturnal home haemodialysis: an update on a five-year experience. Nephrol Dial Transplant 1999;14:2835-40. 8. Pierratos A. Daily nocturnal hemodialysis. Kidney Int 2004;65:1975-86. 9. McFarlane PA, Bayoumi AM, Pierratos A, Redelmeier DA. The quality of life and cost utility of home nocturnal and conventional in-center hemodialysis. Kidney Int 2003;64:1004-11. 10. Kroeker A, Clark WF, Heidenheim P, Kuenzig L, Rosemary L, Meyette M, et al. An operating cost comparison between conventional and home quotidian hemodialysis. Am J Kidney Dis 2003;42:S49-55. 11. Chan CT, Floras JS, Miller JA, Richardson RM, Pierratos A. Regression of left ventricular hypertrophy after conversion to nocturnal hemodialysis. Kidney Int 2002;61:2235-9. 12. Simonsen O. Slow nocturnal dialysis as a rescue treatment for children and young patients with end-stage renal failure. J Am Soc Nephrol 2000;11:327A.
Home Nocturnal Hemodialysis In Children
13. Geary DF, Piva E, Gajaria M, Tyrrell J, Picone G, Harvey EA. Development of a nocturnal home hemodialysis (NHHD) program for children. Semin Dial 2004;17:115-7. 14. Mehrabian S, Morgan D, Schlaeper C, Kortas C, Lindsay RM. Equipment and water treatment considerations for the provision of quotidian home hemodialysis. Am J Kidney Dis 2003;42:S66-70. 15. Twardowski Z. Constant site (buttonhole) method of needle insertion for hemodialysis. Dial Transplant 1995;24. 16. Varni JW, Seid M, Rode CA. The PedsQL: Measurement model for the pediatric quality of life inventory. Med Care 1999;37:126-39. 17. Daugirdas JT. Second generation logarithmic estimates of single pool variable volume Kt/V: an analysis of error. J Am Soc Nephrol 1993;4:1205-13. 18. Eknoyan G, Beck GJ, Cheung AK, Daugirdas JT, Greene T, Kusek JW, et al. Efect of dialysis dose and membrane flux in maintenance. New Engl J Med 2002;347:2010-9. 19. Paniagua R, Amato D, Vonesh E, Correa-Rotter R, Ramos A, Moran J, et al. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J Am Soc Nephrol 2002;13:1307-20. 20. Fine RN. Peritoneal dialysis update. J Pediatr 1982;100:1-7. 21. Heidenheim AP, Leitch R, Kortas C, Lindsay RM. Patient monitoring in the London Daily Nocturnal Hemodialysis Study. Am J Kidney Dis 2003; 42:S61-5. 22. Coyte PC, Young LG, Tipper BL, Mitchell VM, Stoffman PR, Willumsen J, et al. An economic evaluation of hospital-based hemodialysis and home-based peritoneal dialysis for pediatric patients. Am J Kidney Dis 1996;27:557-65. 23. Eknoyan G, Latos DL, Lindberg J. Practice recommendations for the usoe of L-Carnitine in dialysis-related carnitine disorder. Am J Kidney Dis 2003;41:868-76.
387
Objective To describe the effect of home nocturnal hemodialysis (NHD) in North American children. Study design Four teenagers underwent NHD for 8 hours, 6 to 7 nights/week, using either central venous lines or fistulae for periods of 6 to 12 months. Blood flow approximated 200 mL/min, and dialysate flow was 300 mL/min; the dialysate contained potassium and phosphate. The procedure was remotely monitored.
Results The children had unrestricted diets and fluid allowance and did not require phosphate binders. Persistent relative hypotension developed in 2 of 4 children. Weekly Kt/V urea values were consistently >10; other biochemical measures varied. Quality of life and school attendance improved in 3 of 4 children. The workload and reported emotional burden of NHD was substantial. No significant complications occurred. Dialysate losses of calcium, phosphate and carnitine required supplementation. The annual cost per patient was $64,000 Canadian, which represented a 27% savings compared with thrice weekly incenter hemodialysis. Conclusions NHD is feasible in selected children, allows free dietary and fluid intake, and improves patient wellbeing. The burden on the family is substantial, and NHD requires support of a dedicated multidisciplinary team. (J Pediatr 2005; 147:383-7) emodialysis (HD) was first introduced as a maintenance therapy for children with chronic renal failure in 1968.1 Approximately a decade later, continuous ambulatory peritoneal dialysis was introduced.2 Since then, peritoneal dialysis (PD) has become the most widely used dialysis modality for North American children.3 The use of hemodialysis as a home nocturnal treatment (NHD) was developed for adults in 1994,4,5 and the use of this treatment has expanded in the adult renal failure population, in whom it reportedly improves dialysis delivery and patient wellbeing, at reduced cost.6-12 We describe our experience with NHD in 4 children at The Hospital for Sick Children in Toronto, including clinical, psychosocial, and economic outcomes.
H
PATIENTS Patient 1 Patient 1 was a 13-year-old male with end-stage renal disease caused by focal segmental glomerulosclerosis who started PD at age 4 years, and subsequently underwent bilateral nephrectomy and renal transplantation at age 8. At age 11, because of recurrent disease in the graft and chronic allograft nephropathy, he underwent graft nephrectomy and restarted PD. He had recurrent peritonitis and 2 episodes of pancreatitis, which necessitated a switch to HD. On HD, he remained malnourished, had multiple central venous access clots and infections, was markedly hypertensive and repeatedly fluid overloaded, and was unable to comply with dietary and medication prescriptions. In June 2002, he started NHD.
Patient 2 Focal segmental glomerulosclerosis developed in this 15.5-year-old female patient at age 5 years, and she started maintenance PD 3 years later. In 1996, she received a cadaveric AV CVL HD
Arteriovenous Central venous lines Hemodialysis
NHD PD
Nocturnal hemodialysis Peritoneal dialysis
From the Division of Nephrology, Department of Paediatrics, The Hospital for Sick Children and University of Toronto, Toronto, Ontario, Canada. Supported by a grant from the CHANGE foundation. Submitted for publication Nov 10, 2004; last revision received Apr 7, 2005; accepted Apr 14, 2005. Reprint requests: Denis F. Geary, MB, MRCP(UK), FRCP(C), Chief, Division of Nephrology, The Hospital for Sick Children, 555 University Ave, Toronto, ON M5G 1X8. E-mail: denis. [email protected]. 0022-3476/$ - see front matter Copyright ª 2005 Elsevier Inc. All rights reserved. 10.1016/j.jpeds.2005.04.034
383
renal transplant, which was complicated by recurrence of her original disease, development of seizures and facial palsy, chronic allograft nephropathy, and malnutrition. In 1998, she restarted cycling PD and had a gastrostomy tube inserted for supplemental feeds. She was referred to our hospital in 2003 because of increasing morbidity associated with ultrafiltration failure, despite approximately 20 hours daily of cycling PD. Because of her distance from the hospital, the family elected to start NHD.
This 16-year-old boy developed Wegener’s granulomatosis with crescentic glomerulonephritis at age 15 years. He progressed to end-stage renal disease and started PD. The family requested a switch to NHD after 6 months because of peritonitis and severe hypertension.
Patient 4 Hemolytic uremic syndrome developed in this 12.5year-old girl with end-stage renal disease and severe hypertension at age 3 years, requiring bilateral nephrectomies. In 2001, she underwent a living-donor kidney transplant, but chronic allograft nephropathy and end-stage renal disease developed. Because of geographic location and fears of increasing obesity if she was treated with PD, the family elected to start NHD.
METHODS The technical considerations, including home renovations, dialysis machines and water purification units, the possible need for water softeners, and the remote monitoring process, were previously reported for our program and for adults.13,14 Vascular access was provided initially with central venous lines (CVL) in all 4 patients and, subsequently, arteriovenous (AV) fistulae in 2 patients. During NHD, central venous lines are held in place with a locking device described by Pierratos.7 Access through AV fistulas has been with both single-needle and 2-needle techniques; we used a ‘‘buttonhole’’ technique by repeated insertion of the access needles at the same location to form a needle-track,15 which facilitated access by the care provider at home. The parent supervisor, patient, or both was taught this technique for access of a fistulae by the teaching nurse. Dialysate flows of 200 to 300 mL/min were used, with blood flows of 150 to 200 mL/min. Dialysis was provided for approximately 8 hours nightly, 6 to 7 times per week, usually while the patient was sleeping. The dialysate contained potassium at a concentration of 2 to 3 mEq/L. Calcium content of the dialysate was adjusted between 3.0 and 3.5 mEq/L, because chronic negative calcium balance may develop with large ultrafiltration volumes. Phosphate was added to the acid dialysate concentrate as a sodium phosphate (Fleet) enema, which provided a concentration approximating 3 mg/dL. Geary et al
The NHD program was a collaborative effort between the nurses, technologist, social workers, dieticians, and medical staff. The nursing commitment was substantial and demanded expertise in HD and family education. Collaboration with Fresenius Medical Care greatly facilitated the process because of their staff expertise and provision of teaching materials.
Cost Analysis
Patient 3
384
Staff Requirements
Estimation of program costs, expressed in Canadian (Cd) dollars, included: home renovations, including plumbing, electrical, cost of water testing, and installation of phone lines, were measured as actual costs and amortized over 1 year. Costs for the 6-week training period were itemized prospectively and amortized over 1 year. Equipment, including dialysis machines, water softeners, reverse osmosis machines, computer hardware, and home centrifuge, were itemized individually and amortized over 5 years. Dialysis disposables/surgical supplies were individually itemized and included as actual costs. Staff costs were prospectively itemized and are presented as the mean for the 4 patients; nursing costs for ongoing patient care were calculated with a nurse-to-patient ratio of 1:8, with training costs extra. Program development costs (including initial supplies, teaching materials, staff time) are provided separately. No costs are included for physicians or medication.
Psychosocial and Clinical Evaluations Psychosocial evaluations were performed by a social worker (G.P.) on the basis of comments provided by the children, their parents, or both, an analysis of the number of clinic visits for 6 months preceding and subsequently during each 6 months while undergoing NHD, and a record of school attendance. Quality of life was estimated using the PedsQL version 4.0 generic core scale.16 This 23-item self-report questionnaire includes 8 items to determine physical functioning; psychosocial functioning is evaluated by 5 questions each related to emotional, social, and school-functioning. Responses to all questions are scored to provide a total of 0 to 100, with higher scores indicating a better health-related quality of life. A large number of clinical and biochemical tests were monitored to evaluate efficacy of dialysis and clinical wellbeing, and results are reported in tabular form. Single pool Kt/V urea was estimated with the Daugirdas logarithmic formula.17 Clinical and biochemical outcomes were collected at baseline and 3, 6, and 12 months. Results are presented in descriptive form, because only 4 patients are included. Comparative statistical analysis is not provided. The NHD program was initiated for clinical care reasons. However, because data were collected prospectively and there was intent to publish our experience, Research Ethics Board approval was obtained before patients started the program. Also, because these patients are a small group that might be identifiable, written consent was obtained to include the data in the case reports. The Journal of Pediatrics September 2005
RESULTS Patient Selection The 4 families who met eligibility criteria agreed to participate in the program and completed the training requirements. No families have been denied access to this program. Clinical and biochemical outcomes are reported at 3 and 6 months for all patients; 12-month data are available for patients 1 and 2. Patient 1 received NHD for 1 year before switching to a hybrid form of dialysis of NHD Sunday to Wednesday nights inclusive, with an in-center HD on Friday for respite purposes. There have been no dropouts from the program, and no patient deaths.
Technical and Vascular Access Evaluations TECHNICAL. Few technical problems were encountered. Remote monitoring was used for all procedures, but no phone calls to patient homes were required because of a failure to respond to an alarm in timely manner. One family was called when an air detect alarm was noted by the monitor. VASCULAR ACCESS. Episodes of CVL dysfunction were treated with interdialytic instillation of intraluminal tissue plasminogen activator. One CVL infection (Staphyloccoccus aureas) was documented in patient 1 when he fell in a lake at camp, and 1 tunnel infection was observed in patient 2 (diphtheroid organism). The frequency of these episodes was not considered different than in our general HD population. The use of the ‘‘buttonhole’’ technique to access a fistula repeatedly through the same site was associated with the development of an aneurysm at the site, but to date this has not interfered with its continued use. No line disconnections were reported. STAFF REQUIREMENTS. Patients and their parents were trained for 6 weeks: 3 in-center 6-hour sessions per week for 5 weeks, followed by 1 week (3-5 sessions) of nights at the hospital, where the parents are required to perform all duties; the nurse is also available overnight. In addition, parents were required to write 7 tests, with a passing grade considered to be >80% for each examination. Ongoing monitoring of safety and efficacy includes remote monitoring, weekly blood tests, and monthly clinic visits. COST EVALUATION. The costs for development of the NHD program were Cd $4200 (approximately $3400 in US dollars). Patient/family training cost averaged Cd $9000 for the 6 weeks, as aforementioned. The costs of the training are amortized over 1 year and included in the overall annual cost estimate of Cd $64000.
Psychosocial and Clinical Evaluations Variability in many of these outcomes is thought to reflect the diversity, complexity, and multiplicity of the problems that were present before initiation of NHD in these patients. Consistent findings among the 4 patients included normal plasma calcium and phosphate values, increased singleHome Nocturnal Hemodialysis In Children
pool Kt/V urea measurements and high b2-microglobulin levels (Table I; available online at www.jpeds.com). Bone density remained poor in patients 2 and 4, and deteriorated in patients 1 and 3, which suggests the need to avoid sustained negative calcium and phosphate balance. Both free and total plasma carnitine values were below reference range in all patients, until supplemented. The clinical significance of this finding is uncertain, although it might represent a deficiency caused by solute clearance. Many inconsistencies were also observed in outcome measures. The rise in PTH levels in patient 1 between 6 and 12 months could be caused by frequent interruption to his NHD schedule and line function problems during this time (Table I). This patient had a fistula created and transferred to hybrid dialysis after 12 months for respite. Increased CRP levels were also noted in 3 of the 4 patients; the significance of this is unknown. Information on medications for blood pressure, anemia (erythropoietic therapy), phosphate binders and carnitine supplements are presented in Table II (available online at www.jpeds.com). Although all the children had unrestricted diets and fluid intakes, weight gain was variable. Requests for accurate documentation of dietary intake were not regularly adhered to, so the effect of frequent dialysis on nutritional state remains unclear. However, for patient 2, who was completely G-tube dependant at the start of NHD, it was documented that her oral intake increased to 25% to 50% of her dietary requirements. Also, when weight loss has occurred, it is unclear whether this is in part caused by increased exercise/activity. Normalized protein catabolic rate was not measured. Blood pressure control was also variable. Patients 2 and 3, with native kidneys in situ, required anti-hypertensive medication, whereas relative hypotension developed and persisted in patients 1 and 4. These latter 2 anephric children received Midodrine before dialysis to prevent exacerbation of their hypotension. Quality of life (Table I; online at www.jpeds.com) physical and psychological scores improved in 3 of 4 patients. Patient 1, who had a decreased quality of life score between 6 and 12 months, required respite relief and started hybrid dialysis at 12 months. School attendance, as a reflection of psychosocial behavior, improved substantially in 3 children, and teacher reports included improved interaction with peers and improved performance overall. Patient 3 had persistent and significant numbers of days missed from school related to complications of his underlying Wegener’s granulomatosis and the associated immunosuppressant treatment. Specific comments from the children and parents included: ‘‘Because this program has changed my life completely, by making the quality of my life ten times better.’’; ‘‘I feel like a normal kid’’; ‘‘She can’t remember the last time she felt so good’’; ‘‘I used to push her around the block in a wheelchair, and now she is canvassing doorto-door to raise money for her school.’’ Negative comments about NHD reflected the intensity of the workload and parental anxieties related to the 385
complexity of care provided at home. The mother of patient 1 was psychologically and emotionally worn out, until respite was provided when a hybrid form of dialysis was instituted. Others described disruption for other family members and difficulty establishing a new routine within the home.
DISCUSSION This report describes the successful implementation of NHD for children. Although introduced approximately 10 years ago for adults, the only previous pediatric experience with NHD, published in abstract form from Sweden,12 described 4 children with ages similar to our own patients. They used CVL as blood access, whereas we have also successfully used AV fistulas. Overall, their brief report suggests that outcomes are comparable to our own, and that NHD provides an improvement in patient well-being compared with other dialysis modalities.12 Consequently, we recommend that this modality be considered for all teenagers, and some younger school-age children starting dialysis. The success of NHD is perhaps best exemplified by the continuation of the procedure in all patients. No patients have died, although the severity of uremia and its complications when starting NHD was extreme for patients 1 and 2. The Hemo Study for hemodialysis patients18 and the ADEMEX Study of peritoneal dialysis patients19 suggested that increasing the clearance of urea as measured with Kt/V urea, beyond certain limits, does not produce further improvement in patient outcomes. However, we do not believe that the results of these studies are applicable to NHD. Rather, the Hemo Study results suggest that for intermittent therapies, there are limitations to what can be achieved; the Hemo Study can be interpreted to suggest that more frequent dialysis, allowing for less interdialytic accumulation of toxins, may be the only remaining way to improve outcomes for patients undergoing HD. Data in small numbers of patients receiving NHD suggest that long-term survival is improved when compared with conventional dialysis modalities.8 As an alternative to NHD, the introduction of short daily dialysis at home may be considered. Short daily dialysis has the advantage that it can be provided during the daylight hours when the family is awake, and therefore remote monitoring will not be required. However, the amount of dialysis provided using a short daily regimen is less than that achieved by overnight dialysis, and because the set-up and tear-down time and supplies needed are the same for each modality, short daily dialysis is less cost-effective. The introduction of NHD to a patient’s home is a huge undertaking. However, the improvement in clinical status, school attendance and performance, which we have noted, and the small but consistent improvement in quality of life reported by our patients suggest that the outcomes justify the psychosocial burden that accompanies NHD. Significant concerns were expressed about the introduction of nocturnal cycling PD to patientsÕ homes.20 The same concerns can be expressed about the introduction of NHD. However, because 386
Geary et al
most children with end-stage renal disease are now treated with overnight cycling PD,3 it is reasonable to consider that with further experience and familiarity, NHD will become increasingly accepted if the improvements in patient well being that we report are substantiated by others. Remote monitoring of NHD was provided for all our patients. This provides a safety net and a psychological security blanket for patients and caregivers undertaking NHD. We recommend that some form of monitoring be in place for other patients starting this procedure, despite the fact that home monitoring is no longer recommended for all adult patients.8,21 However, it is also important to inform families that remote monitoring may be insufficient if a major dialysisrelated complication occurs. Our overall costs were estimated at Cd $63,670 per patient annually, which compares favorably with our in-center HD costs, both currently (Cd $88,000) and a decade ago (Cd $76,000).22 This represents a savings of 27% for each patient receiving NHD, compared with thrice-weekly in-center HD, and exceeds the savings reported in adult NHD.9,10 Although the cost of supplies is double that of conventional HD, savings still result because of the reduction in staff use. The savings are greater in children than adults because the nurse-to-patient ratio in the pediatric HD unit exceeds that of an adult unit. When the patients remain on NHD >1 year, our cost estimates would be lower because we amortized the costs of patient training, and home renovations over 1 year. If the equipment (eg, reverse osmosis and dialysis machines) is used for longer than 5 years, further savings would be realized. Finally, our estimate of appropriate nurse-to-patient ratio of 1:8 was arbitrary. Adult centers report a nurse-to-patient ratio of 1:20 for their home HD programs, but we feel that ratio will not be achievable for children because of the increased psychosocial component of pediatric care. The possibility that morbidity might result from excess dialysis must be considered for children receiving NHD. Phosphate and calcium losses in the dialysate have been documented,8 and supplements of each were added to the dialysate in our patients to prevent hypocalcemia and hypophosphatemia. Calcium deficiency may be compounded by the fact that, unlike patients receiving conventional dialysis, patients receiving NHD do not take calcium-containing phosphate binders. Sustained negative calcium balance, although not documented by us, potentially may adversely affect bones. A deficiency of carnitine, requiring supplementation, was documented in our patients. The potential for carnitine deficiency with frequent dialysis has been addressed in the report of the Carnitine Consensus Conference,23 although this concern was not mentioned in previous reports of NHD in adults. Prolonged deficiency of carnitine may result in muscle weakness, cardiomyopathy, intradialytic hypotension, and anemia and should be considered in patients receiving NHD who manifest these symptoms. The authors wish to thank Dr Andreas Pierratos from the Humber River Regional Hospital in Toronto and Michaelene Ouwendyk from Fresenius Medical Care for their assistance in the development of our Home Nocturnal Hemodialysis program. The Journal of Pediatrics September 2005
REFERENCES 1. Fine RN, De Palma JR, Lieberman E, Donnell GN, Gordon A, Maxwell MH. Extended hemodialysis in children with chronic renal failure. J Pediatr 1968;73:706-13. 2. Irwin MA, Balfe JW, Hardy B. Continuous ambulatory peritoneal dialysis in pediatrics. J Am Assoc Nephrol Nurses Tech 1981;8:11-3. 3. Leonard MB, Donaldson LA, Ho M, Geary DF. A prospective cohort study of incident maintenance dialysis in children: A NAPRTC study. Kidney Int 2003;63:744-55. 4. Uldall R, Francouer R, Ouwendyk M, Wallace L, Langos V, Ecclestone A, et al. Simplified nocturnal home hemodialysis (SNHHD): a new approach to renal replacement therapy. J Am Soc Nephrol 1994;5:428. 5. Uldall R, Ouwendyk M, Francouer R, Wallace L, Sit W, Vas S, et al. Slow nocturnal home hemodialysis at the Wellesley Hospital. Adv Ren Replace Ther 1996;2:133-6. 6. Lockridge RS Jr, Spencer M, Craft V, Pipkin M, Campbell D, McPhatter L, et al. Nocturnal home hemodialysis in North America. Adv Ren Replace Ther 2001;8:250-6. 7. Pierratos A. Nocturnal home haemodialysis: an update on a five-year experience. Nephrol Dial Transplant 1999;14:2835-40. 8. Pierratos A. Daily nocturnal hemodialysis. Kidney Int 2004;65:1975-86. 9. McFarlane PA, Bayoumi AM, Pierratos A, Redelmeier DA. The quality of life and cost utility of home nocturnal and conventional in-center hemodialysis. Kidney Int 2003;64:1004-11. 10. Kroeker A, Clark WF, Heidenheim P, Kuenzig L, Rosemary L, Meyette M, et al. An operating cost comparison between conventional and home quotidian hemodialysis. Am J Kidney Dis 2003;42:S49-55. 11. Chan CT, Floras JS, Miller JA, Richardson RM, Pierratos A. Regression of left ventricular hypertrophy after conversion to nocturnal hemodialysis. Kidney Int 2002;61:2235-9. 12. Simonsen O. Slow nocturnal dialysis as a rescue treatment for children and young patients with end-stage renal failure. J Am Soc Nephrol 2000;11:327A.
Home Nocturnal Hemodialysis In Children
13. Geary DF, Piva E, Gajaria M, Tyrrell J, Picone G, Harvey EA. Development of a nocturnal home hemodialysis (NHHD) program for children. Semin Dial 2004;17:115-7. 14. Mehrabian S, Morgan D, Schlaeper C, Kortas C, Lindsay RM. Equipment and water treatment considerations for the provision of quotidian home hemodialysis. Am J Kidney Dis 2003;42:S66-70. 15. Twardowski Z. Constant site (buttonhole) method of needle insertion for hemodialysis. Dial Transplant 1995;24. 16. Varni JW, Seid M, Rode CA. The PedsQL: Measurement model for the pediatric quality of life inventory. Med Care 1999;37:126-39. 17. Daugirdas JT. Second generation logarithmic estimates of single pool variable volume Kt/V: an analysis of error. J Am Soc Nephrol 1993;4:1205-13. 18. Eknoyan G, Beck GJ, Cheung AK, Daugirdas JT, Greene T, Kusek JW, et al. Efect of dialysis dose and membrane flux in maintenance. New Engl J Med 2002;347:2010-9. 19. Paniagua R, Amato D, Vonesh E, Correa-Rotter R, Ramos A, Moran J, et al. Effects of increased peritoneal clearances on mortality rates in peritoneal dialysis: ADEMEX, a prospective, randomized, controlled trial. J Am Soc Nephrol 2002;13:1307-20. 20. Fine RN. Peritoneal dialysis update. J Pediatr 1982;100:1-7. 21. Heidenheim AP, Leitch R, Kortas C, Lindsay RM. Patient monitoring in the London Daily Nocturnal Hemodialysis Study. Am J Kidney Dis 2003; 42:S61-5. 22. Coyte PC, Young LG, Tipper BL, Mitchell VM, Stoffman PR, Willumsen J, et al. An economic evaluation of hospital-based hemodialysis and home-based peritoneal dialysis for pediatric patients. Am J Kidney Dis 1996;27:557-65. 23. Eknoyan G, Latos DL, Lindberg J. Practice recommendations for the usoe of L-Carnitine in dialysis-related carnitine disorder. Am J Kidney Dis 2003;41:868-76.
387