Overview

This evidence report is the product of a systematic literature review including synthesis and analysis of data from all relevant sources that met predefined criteria. The protocol for this review was prospectively designed to define:study objectives, search strategy, patient populations of interest, study selection criteria and methods for determining study eligibility, outcomes of interest, and data elements to be abstracted and methods for abstraction. The protocol and report were developed with considerable guidance from expert advisors external to the Blue Cross and Blue Shield Association Technology Evaluation Center (TEC).

To maximize the accuracy of study selection and data abstraction, two independent reviewers were involved in each step in this protocol. Disagreements were infrequent and were resolved by consensus of the reviewers. The protocol provided that disagreement that could not be resolved by the reviewers would be referred for discussion with the external advisors.

This is a review of published evidence, including both full reports and abstracts from scientific meetings. Abstracts were included in the evidence review to maximize the comprehensiveness of the evidence presented. The decision to include abstracts was made in consultation with advisory experts who suggested that the number of full reports addressing the key questions was not large and that the overall usefulness of the report would be enhanced by including analysis and synthesis of data reported in scientific abstracts. Full reports of study methodology and findings that have gone through a peer review process are viewed with greater confidence than are abstracts presented at scientific meetings. Inherent limitations of abstracts relate primarily to the limited detail of information provided in the allowable space as well as the lack of a rigorous peer review process. Thus, results of abstracts are included to provide information of potential interest to the reader, but the evidence from abstracts would be considered preliminary and of lesser quality as compared with evidence drawn from a full report.

During the course of the literature review, the EPC team followed the status of three international studies that evaluated the effects of increasing Hct above 36 percent in patients with CRF (Jacobs, 1999). These studies include: (1) a multicenter RCT conducted in Canada by Foley, Parfrey, Morgan, et al. (1998) comparing normalization of Hct (40.5 percent target) with partial correction of anemia (30 percent target) in 146 hemodialysis patients with asymptomatic cardiomyopathy; (2) a multicenter study conducted by the Spanish Cooperative Renal Patients Quality of Life Study Group comparing outcomes before and after raising the Hct above 36 percent with epoetin in a group of 156 stable hemodialysis patients who did not have severe comorbidity (Moreno, Sanz-Guajardo, Lopez-Gomez, et al., 2000); and (3) a multicenter, RCT conducted in Scandinavia comparing normalized Hct (target of 43.5 to 48 percent in males, 40.5 to 45 percent in females) with partial correction of anemia (target 27 to 36 percent) in 416 patients with CRF (293 hemodialysis, 51 continuous ambulatory peritoneal dialysis, and 72 predialysis) ("Scandinavian Multicentre Trial," see description in Jacobs, 1999).

Quality of life and safety outcomes were assessed in all three studies. Echocardiographic changes were also assessed in the Canadian study and the Scandinavian study. Although both the Canadian and Scandinavian studies have been completed, the final results have not been published as full reports. Results of the Canadian study have been published in two abstracts (Foley, Parfrey, Morgan, et al. 1998; Wells, Coyle, Lee, et al. 1998). Results from the Spanish study have been published as a full report (Moreno, Sanz-Guajardo, Lopez-Gomez, et al., 2000) and have been included in this review. Results from the Scandinavian study have not yet been published. Supplemental review of these unpublished data may be performed in the future as full reports of these studies become publicly available.

Technical Advisors

The Technical Advisory Group (TAG) served as the primary advisory panel and was consulted extensively through all phases of this project. The Renal Disease TAG included three members appointed by the TEC: Allen Nissenson, MD, and Joseph Eschbach, MD, nephrologists; and Earl Steinberg, MD, MPP, an internist with expertise in outcomes research. Dr. Eschbach chaired and Dr. Nissenson was a member of the Anemia Work Group of the NFK-DOQI^TM. Dr. Steinberg was the NFK-DOQI Scientific Director. In addition, Alan Lichtin, MD, a hematologist appointed by the American Society of Hematology to serve on the TAG for our companion systematic review on the use of epoetin in oncology, was included in some of the Renal Disease TAG discussions.

An early work product for this report, consisting of a preliminary analysis of evidence, was reviewed by the Blue Cross and Blue Shield Association TEC Medical Advisory Panel. This interdisciplinary panel comprises experts in technology assessment methods and clinical research and also includes managed care physicians from Blue Cross and Blue Shield and Kaiser Permanente health plans.

An additional panel of External Reviewers provided input to the design of the prospective protocol and reviewed the draft report. This panel included clinical and research specialists involved in the treatment of patients with CRF who were either selected by the Technology Evaluation Center (TEC) or were designated by one of the national stakeholder organizations contacted by the TEC. In addition, a patient representative was designated by the American Association of Kidney Patients. The External Reviewers also included technical staff of Ortho-Biotech, Inc., which markets epoetin alfa for the treatment of predialysis patients with progressive renal failure, and technical staff of Amgen, Inc., which markets epoetin alfa for the treatment of patients undergoing dialysis for ESRD. A medical officer from the Health Care Financing Administration, which requested the systematic review of evidence on the normalization of Hct in ESRD, was also a reviewer for this report.

Appendix A lists the members and institutional affiliations of the TAG, External Reviewers, and the Blue Cross and Blue Shield Association TEC Medical Advisory Panel.

Search Strategy for the Identification of Articles

A comprehensive literature search was performed to identify publications of relevant clinical studies. The search process began with the MEDLINE and EMBASE databases. Two separate electronic search strategies were performed for this systematic review. The first search strategy pertains to Part I (Adult Patients with Chronic Renal Failure) and Part II (Pediatric Patients with Chronic Renal Failure). This search strategy listed "kidney failure" as a required search term. The second search strategy, which pertains to Part III (Subpopulations of Interest with or without Chronic Renal Failure), did not require "kidney failure" as a search term.

Search Strategy for Results and Conclusions, Parts I and II

These online sources were searched for all articles published since 1985 that included at least one of the following text words (tw) or Medical Subject Heading (MeSH^®) terms in their titles, their abstracts, or their keyword lists:

• erythropoietin (MeSH^®)
• epoetin alfa (MeSH^®)
• erythropoietin (tw)
• epoetin (tw)
• Epogen (tw)
• Procrit (tw)
• epo (tw)
• Anemia/drug therapy (MeSH^®; including all subheadings)
• Anemia/therapy (MeSH^®; including all subheadings)
• Anemia/diet therapy (MeSH^®; including all subheadings)

The search results were then limited to include only those articles that were indexed under the Medical Subject Headings^® "kidney failure" (including all subheadings) and that addressed studies on human subjects. The MEDLINE and EMBASE databases were last searched in December 1998. To supplement the above strategy, issues of Current Contents on Diskette were searched through October 30, 1999, to identify recently published articles that had not yet been indexed by the online databases.

An additional source of bibliographic information and reprints of clinical studies was provided by Ortho-Biotech, Inc., and Amgen, Inc. Finally, all relevant review articles, editorials, and letters published in 1995 or later were retrieved. Reference lists from these articles were searched for relevant studies not identified by the above methods.

Search Strategy for Results and Conclusions, Part III

An electronic search was conducted to identify articles pertaining to patient subgroups of interest defined by clinical characteristics that have been postulated to warrant maintaining Hct above 36 percent. Relevant clinical characteristics included: congestive heart failure, coronary artery disease, arterial occlusive disease including peripheral vascular disease, cerebrovascular disease, chronic obstructive pulmonary disease, living at high altitude, and adolescent age.

Studies considered to be relevant for this section of the systematic review were not required to include only patients with CRF. Our rationale for including studies in patients without CRF was that the clinical outcomes resulting from different levels of Hct observed in patients with one of these clinical characteristics might be generalizable to CRF patients with the same clinical characteristic. Thus, for example, if an Hct level above 36 percent was shown to improve health outcomes in a general population of patients with peripheral vascular disease, then the EPC team felt it might be reasonable to generalize that finding to CRF patients with peripheral vascular disease.

The following online search strategy was employed for this portion of the evidence review using the following tws or MeSHeading terms in their titles, abstracts, or keyword lists:

• erythropoietin (MeSH^®)
• epoetin alfa (MeSH^®)
• erythropoietin (tw)
• Epoetin (tw)
• Epogen (tw)
• Procrit (tw)
• epo (tw)
• Anemia/drug therapy (MeSH^®; including all subheadings)
• Anemia/therapy (MeSH^®; including all subheadings)
• Anemia/diet therapy (MeSH^® including all subheadings)

AND at least one of the following items:

• heart failure (MeSH^®>)
• congestive disease (MeSH^®)
• coronary disease (MeSH^®)
• arterial occlusive disease (MeSH^®)
• cerebrovascular disease (MeSH^®)
• lung diseases, obstructive (MeSH^®)
• altitude (MeSH^®)
• adolescent (MeSH^®)

The search results were then limited to include only those articles that addressed studies on human subjects.

In addition to this search process, reference lists from relevant reviews, editorials, and letters published since 1995 were searched for studies not identified by the electronic search methods. Furthermore, all of the external advisors involved in the project were asked to inform the project team of any additional studies potentially relevant to the key questions that had not yet been identified. This search process identified over 3,106 titles and abstracts.

Determining Study Eligibility

Study eligibility was assessed in five general areas: (1) types of participants, (2) type of study, (3) Hct range being studied, (4) sample size, and (5) type of outcome reported.

Stages of Study Selection

Three stages were used in systematically selecting articles to be included in the systematic review.

Stage 3: Evaluation of Potential Data Articles and Identifying Studies Eligible for Analysis of Evidence

Articles identified as potentially containing data were evaluated by two independent reviewers according to a prespecified set of study selection criteria. A list of predefined reasons for exclusion are described in Table 2. Reviewers were not required to list more than one reason for exclusion in the database even if more than one reason was applicable. Disagreements were reviewed by a third reviewer. In most cases, a consensus opinion could be reached through discussion. When a consensus decision was not reached among the three reviewers, a decision was made in consultation with the TAG members. A total of 25 studies, 15 full reports and 10 abstracts, were identified as eligible for inclusion in the review of evidence.

Table

Table 2. Reasons for study exclusion.

Data Abstraction

The data abstraction forms utilized for this part of the systematic review can be found in Appendix B.

Two reviewers independently abstracted data from each eligible study and recorded the data directly into a word-processing tabular format. Information was gathered describing the patient population (including demographics and type of dialysis), study design, anemia treatment information, and outcome information. These tables were visually compared for consistency by a single reviewer, and simple numerical discrepancies were resolved by referring to the original article. Inconsistencies not related to a simple error were discussed and consensus was reached.

Evaluation of Study Quality

The primary assessment of study quality that we used in this systematic review is based on study design with a three-tier hierarchy of quality. Accordingly, each section of the Results is subdivided into three categories (listed in order of strongest to weakest design): RCT, nonrandomized controlled clinical trial, or cross-sectional analysis. Discussion of results within each category of study design addresses both results described in full reports and results described in abstracts. Results from full reports are viewed with greater confidence than are those from abstracts presented at scientific meetings.

Although it was the intent of this systematic review to prospectively apply additional measures to assess study quality, the overall paucity of available evidence and considerable variation in study designs utilized by various investigators grossly limited the feasibility and utility of constructing a formal quality evaluation schema. In consultation with the TAG, the decision was made not to apply any formalized quantitative quality scoring system. However, during analysis of the evidence, five issues pertaining to study quality and relevance were identified.

For the purpose of this report, the first two dimensions serve to define better studies (uses double-masked design, or compares Hct >36 percent with 33 to <36percent), and the second two dimensions serve to define worse studies (loss to followup or missing data for >10 percent, or potential bias in control group comparison). The EPC team recognizes that the comparison of Hct range is not a measure of study quality; however, this issue pertains to the relevance of the study to the primary focus of this systematic review. Thus, the team included this dimension among the quality assessment criteria for simplicity. Finally, the extent to which cross-sectional analyses identified and controlled for confounding influences in the data analysis was described.

These five dimensions of study quality and relevance are described and tabulated for each study included in the systematic review.

Footnotes

4

After puberty, the normal range of blood hemoglobin is 16 ± 2.0 g/dL in males and 14 ± 2.0 g/dL in females (Perkins, 1999). Employing the commonly used conversion factor of multiplying by 3 to convert hemoglobin to Hct, normal Hct ranges are approximately 48±6 percent for males and 42±6 percent for females.

5

The normal range of blood hemoglobin level for children between the ages of 3 months and 10 years is 12.2 ±2.3 g/dL (Perkins, 1999). Employing the commonly used conversion factor of multiplying by 3 to convert hemoglobin to Hct, normal Hct range is approximately 36.6 ± 6.9 percent.