1402

JOURNAL OF PALLIATIVE MEDICINEVolume 10, Number 6, 2007© Mary Ann Liebert, Inc.DOI: 10.1089/jpm.2007.9841

Prognostication in Patients Receiving Dialysis #191

MATTHEW HUDSON, STEVEN WEISBORD, M.D., and ROBERT M. ARNOLD, M.D.

END-STAGE RENAL DISEASE (ESRD) is a highlyprevalent and rapidly increasing condition. While

dialysis prolongs life in patients with ESRD, life ex-pectancy remains only a third to a sixth as long as sim-ilar patients not on dialysis. The overall 1- and 5-yearmortality rates are 25% and 60%, respectively. Ap-proximately 20% of ESRD patient deaths occur aftera decision to stop dialysis, highlighting the importanceof discussions of prognosis and goals of care with thischronically ill population. This Fast Fact reviews thecurrent data regarding prognostication in patients re-ceiving chronic hemodialysis and peritoneal dialysis.Note: renal transplantation reduces mortality and thefollowing data do not consider patients with function-ing kidney transplants.

PROGNOSTIC FACTORS

Several patient-specific factors influence prognosis:

• Age: for 1-year increments beginning at age 18,there is a 3% to 4% increase in annual mortalitycompared to the general population. One- and 2-yearmortality rates go from 10% and 12% at 25–29 yearsof age, to 25% and 42% at 65–69 years, to 39% and61% at 80–84 years of age.

• Functional status: the relative risk of dying within 3years of starting dialysis is 1.44 for those withKarnofsky Performance Status scores of � 70 com-pared to those with a score � 70 (see Fast Fact #13).

• Albumin: serum albumin level, both at baseline andduring the course of dialysis treatment, is a consis-tent and strong predictor of death. For example, the1 and 2 year survival of patients with an albuminof � 3.5 g/dL is 86% and 76% respectively, com-pared to 50% and 17% if one’s albumin is less than3.5.

PROGNOSTIC TOOLS

It has long been recognized that patient comorbid-ity is strongly correlated with prognosis in ESRD. Anage-modified Charlson Comorbidity Index (CCI),which stratifies patients based on medical comorbidi-ties and age, has been successfully used to predict mor-tality in dialysis-dependent patients8:

MODIFIED CHARLSON COMORBIDITY INDEX

Total score is the sum of the comorbidity points.

For example, a 66-year old male on dialysis with ahistory of CHF, COPD, and diabetes with retinopathywould have a CCI score of 9 and a nearly 50% chanceof dying within a year. Using this, a provider could

*Fast Facts are edited by Drew Rosielle, M.D., Palliative Care Center, Medical College of Wisconsin. For comments/questionswrite to: drosiell@mcw.edu. The complete set of Fast Facts is available at www.eperc.mcw.edu

Fast Facts and Concepts*

Comorbidity Points

1 point each for coronary artery disease, congestive heart failure,peripheral vascular disease, cerebrovascular disease, dementia,chronic pulmonary disease, connective tissue disorder, peptic ulcerdisease, mild liver disease, diabetes

1 point for every decade over 40 (e.g., a 65-year-old would receive 3 points).

2 points each for hemiplegia, moderate-to-severe renal disease(including being on dialysis), diabetes with end-organ damage,cancer (including leukemia or lymphoma)

3 points for moderate-to-severe liver disease

6 points each for metastatic solid tumor or AIDS

Modified CCI Low score Moderate High Very High (�8)Score Totals (�3) (4–5) (6–7)

Annual 0.03 0.13 0.27 0.49mortality rate

FAST FACTS AND CONCEPTS 1403

discuss with the patient his prognosis and use this tofacilitate further discussion regarding planning for thefuture, including end-of-life decisions. The Index ofCoexistent Disease (ICED), a general illness severityindex, has also shown predictive power in ESRD. Thescale’s complexity and length however (it entails ask-ing over 100 questions) limit its clinical usefulness.

SUMMARY

The age-modified CCI, in conjunction with otherprognostic factors such as serum albumin and func-tional status, can be used to help facilitate discussionswith dialysis-dependent patients and their families re-garding goals of care and end-of-life planning.

REFERENCES

1. United States Renal Data System. Incidence and preva-lence. Annual data report, 2006. Minneapolis, MN: US-RDS Coordinating Center; 2006. �www.usrds.org/2006/pdf/02_incid_prev_06.pdf� (Last accessed November 5,2007).

2. Cohen LM, Moss AH, Weisbord SD, Germain MJ: Renalpalliative care. J Palliat Med 2006;9:977–992.

3. Renal Physicians Association and American Society ofNephrology: Shared Decision-Making in the AppropriateInitiation of and Withdrawal from Dialysis, Clinical Prac-tice Guideline No 2. Washington, DC: Renal PhysiciansAssociation, 2000.

4. Ifudu O, Paul HR, Homel P, Friedman EA: Predictive valueof functional status for mortality in patients on maintenancehemodialysis. Am J Nephrol 1998;18:109–116.

5. Owen WF, Lew NL , Yiu Y, Lowry EG, Lazarus JM: Theurea reduction ratio and serum albumin concentration aspredictors of mortality in patients undergoing hemodialy-sis. N Engl J Med 1993;329:1001–1006.

6. Owen WF, Price D: African-Americans on maintenancedialysis. Adv Ren Replace Ther. 1997;4:3–12.

7. Lowrie EG, Lew NL: Death risk in hemodialysis patients.Am J Kidney Dis 1990;15:458–482.

8. Beddhu S, Bruns FJ, Saul M, Seddon P, Zeidel ML: A sim-ple comorbidity scale predicts clinical outcomes and costsin dialysis patients. Am J Med 2000;108:609–613.

9. Miskulin DC, Martin AA, Brown R, et al: Predicting 1-year Mortality in an outpatient hemodialysis population: Acomparison of comorbidity instruments. Nephrol DialTransplant 2004;19:413–420.

10. Moss AH: A new clinical practice guideline on initiationand withdrawal of dialysis that makes explicit the role ofpalliative medicine. J Palliat Med 2000;3:253–260.

11. Nicolucci A, Cubasso D, Labbrozzi D, et al: Effect of co-existent diseases on survival of patients undergoing dialy-sis. ASAIO J 1992;38:M291–M295.

Address reprint requests to:Robert M. Arnold, M.D.

Section of Palliative CareUPMC Montefiore-933W

200 Lothrop StreetPittsburgh, PA 15213

E-mail: rabob@pitt.edu

DOI: 10.1089/jpm.2007.9840

Opioid Use in Renal Failure #161

ROBERT M. ARNOLD, M.D., PEG VERRICO, R.Ph., and SARA N. DAVISON, M.D.

CHRONIC PAIN is common in chronic kidney diseaseimpacting 50% of hemodialysis patients, 82% of

whom experience moderate to severe pain.1,2 The ab-sorption, metabolism, and renal clearance of opioidsare complex in renal failure. However, with the ap-propriate selection and titration of opioids, patientswith renal failure can achieve analgesia with minimalrisk of adverse effects. This Fast Fact reviews rec-ommendations for opioid use in the setting of renalfailure and in patients receiving long-term dialysis.3–5

NOT RECOMMENDED FOR USE

Meperidine is not recommended in renal failure dueto accumulation of normeperidine, which may causeseizures.

Codeine has been reported to cause profound tox-icity which can be delayed and may occur after triv-ial doses. We recommend that codeine be avoided inpatients with a glomerular filtration rate (GFR) lessthan 30 mL/min.

Dextropropoxyphene is associated with central ner-vous system (CNS) and cardiac toxicity and is not rec-ommended for use in patients with renal failure.

Morphine is not recommended for chronic use inrenal insufficiency (GFR � 30 mL/min) due to therapid accumulation of active, nondialyzable metabo-lites that are neurotoxic. If morphine must be used,avoid long-acting preparations and monitor closely fortoxicity (see Fast Facts #57, 58).

USE WITH CAUTION

Oxycodone is metabolized in the liver with 19% ex-creted unchanged in the urine. There are reports of ac-cumulation of both the parent compound and metabo-lites in renal failure resulting in CNS toxicity andsedation.

Hydromorphone, as the parent drug, does not sub-stantially accumulate in hemodialysis patients. Con-versely, an active metabolite, hydromorphone-3-glu-curonide, quickly accumulates between dialysistreatments, but appears to be effectively removed dur-ing hemodialysis. With careful monitoring, hydro-morphone may be used safely in dialysis patients.However, it should be used with caution in patientswith a GFR � 30 mL/min who have yet to start dial-ysis or who have withdrawn from dialysis.

SAFEST IN RENAL INSUFFICIENCY

Fentanyl is considered relatively safe in renal fail-ure as it has no active metabolites. However, very lit-tle pharmacokinetic data exist regarding fentanyl inend stage renal disease. While some studies haveshown decreased clearance in renal failure, most stud-ies do not show drug accumulation. Fentanyl is not di-alyzable due to high protein binding and a high vol-ume of distribution.

Methadone is considered relatively safe in renal fail-ure. It has no active metabolites and limited plasmaaccumulation in renal failure due to enhanced elimi-nation in the feces. However, precautions regardingthe use of methadone exist (See Fast Facts # 75, 86);it does not appear to be removed by dialysis.

OPIOID DOSING

Given the lack of pharmacokinetic and pharmaco-dynamic data of opioids in renal failure, it is difficultto advocate for specific analgesic treatment algo-rithms. However, t he following guide has been pro-posed for the initial dosing of the safer opioids in re-nal failure.

• Cl CR � � 50 mL/min: Normal dosing• Cl CR � 10–50 mL/min: 75% of normal• Cl CR � � 10 mL/min: 50% of normal

The “normal opioid dose” for any given patient isthe dose that adequately relieves pain without unac-ceptable adverse effects (see Fast Fact #20). Whileopioids can be used in renal insufficiency, they requirecloser monitoring and constant reassessment to ensurethat accumulation of active metabolites does not resultin toxicity. This should not preclude the effective useof opioids in these patients.

REFERENCES

1. Chambers EJ, Germain M, Brown E (eds): Supportive Carefor the Renal Patient. New York: Oxford University Press,2004.

2. Davison SN: Pain in hemodialysis patients: Prevalence,cause, severity, and management. Am J Kidney Dis2003;42:1239–1247.

3. Murphy EJ: Acute pain management pharmacology for thepatient with concurrent renal or hepatic disease. AnaesthIntens Care 2005;33:311–322.

4. Dean M: Opioids in renal failure and dialysis patients. JPain Symptom Manage 2004;28:497–504.

5. Broadbent A, Khor K, Heaney A: Palliation and chronicrenal failure: Opioid and other palliative medications—Dosage guidelines. Prog Palliat Care 2003;11:183–190.

Address reprint requests to:Robert M. Arnold, M.D.

Section of Palliative CareUPMC Montefiore-933W

200 Lothrop StreetPittsburgh, PA 15213

E-mail: rabob@pitt.edu

FAST FACTS AND CONCEPTS1404