Home > Methods - Chronic Kidney Disease Stages...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Fink HA, Ishani A, Taylor BC, et al. Chronic Kidney Disease Stages 1–3: Screening, Monitoring, and Treatment [Internet]. Rockville (MD): Agency for Healthcare Research and Quality (US); 2012 Jan. (Comparative Effectiveness Reviews, No. 37.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Methods

Topic Refinement

The initial nominator of this topic, first titled “Management of Mild Renal Impairment,” proposed questions related to clinical typology, frequency of monitoring, calculation of creatinine clearance, management, and secondary prevention of mild renal impairment. Subsequently, a second nominator proposed questions related to screening for and treatment of screen-detected CKD. It was determined to be feasible to combine the two sets of questions. The scope of the combined questions explicitly excluded management of patients with more advanced kidney disease.

Key Questions were drafted with input from representatives of the nominating organizations. These Key Questions and project scope were submitted for AHRQ approval and then posted on the Effective Health Care web site for public comment.

Comparative Effectiveness Review

Public comments were reviewed with AHRQ and the nominators, and incorporated as appropriate in a draft protocol. The draft protocol was circulated to a Technical Expert Panel (TEP) composed of researchers, clinicians, and representatives from professional organizations and federal and state agencies including the American College of Physicians, United States Preventive Services Task Force, National Kidney Foundation, American Association for Clinical Chemistry, Centers for Disease Control and Prevention, American Academy of Family Practice, and KDIGO. Based on TEP feedback, including on the relevance and scope of the review, the protocol was revised and a final protocol, including the revised Key Questions and proposed project methods, was approved by the Agency for Healthcare Research and Quality (AHRQ) and posted on the Effective Health Care website.

Based on feedback received during protocol development, the terminology used in this project was changed to be consistent with the currently accepted terminology for referring to impairments in kidney function and kidney damage as established by the National Kidney Foundation's KDOQI2 and later modified by the KDIGO.33 In addition, its title was changed for the protocol to “Screening for and Management of Chronic Kidney Disease Stages 1–3.” Finally, based on public and peer reviewer feedback to the draft report, the final report title was changed to more accurately and transparently reflect its content and organization: “Chronic Kidney Disease Stages 1–3: Screening, Monitoring, and Treatment.”

Systematic Review

Search Strategy

We developed separate search strategies for the screening, monitoring, and treatment Key Questions. Search strings were developed and tested to identify randomized controlled trials (RCTs) or controlled clinical trials (CCTs). We included studies that enrolled an adult population (18 years of age and older), were published since 1985, and were written in the English language. Evidence suggests that for systematic reviews of conventional medicine, as were evaluated in the present review, restriction to include only English language trials should not bias estimates of the effectiveness of the interventions.34 Only full articles were included. We searched MEDLINE® and the Cochrane Database of Systematic Reviews. Details of the major search strategies are provided in Appendix A.

To identify systematic reviews related to the three topic areas, we completed a search of MEDLINE® and the Cochrane Database of Systematic Reviews using the same search strategies as above with the addition of publication type terms to identify systematic reviews. We manually searched the reference lists of the identified systematic reviews to identify any RCTs or CCTs not detected in our electronic literature search. We also manually searched reference lists of the primary reports that were eligible for inclusion in the review. Per project protocol, because we did not find evidence from RCTs or CCTs to directly address whether screening or monitoring impact clinical outcomes or harms, we conducted a nonsystematic search for observational studies to identify indirect evidence regarding the benefits and harms of screening for and monitoring of CKD. All citations then were imported into EndNote X and Excel for abstract review and database management.

A broad search of the grey literature was completed by the AHRQ Scientific Resource Center librarian. Grey literature, which by definition is literature that is not systematically stored or indexed,35 included abstracts presented at conferences, unpublished trial data, government documents, and pharmaceutical company scientific information packets on medications evaluated in this topic.

We conducted the initial searches in March and April of 2010. All searches were updated in January 2011.

Inclusion/Exclusion Criteria

We developed criteria for inclusion and exclusion of studies based on patient populations, interventions, outcome measures, and types of evidence relevant to the Key Questions. Within the sections for each pair of Key Questions immediately below, inclusion criteria are detailed in the ‘Patients’ sections and exclusion criteria are detailed in the ‘Study Selection’ sections. We retrieved full-text articles of potentially relevant abstracts and conducted a second review for inclusion by reapplying the inclusion criteria. If no abstract was available electronically, the full text of the article was obtained for review.

Key Questions 1 and 2

Patients

We restricted the review to studies that enrolled adults who were without known CKD, were with or without recognized risk factors for CKD, and who were systematically screened for CKD. Because much of our search period preceded the development and wide implementation of the current CKD staging system, studies whose definitions of CKD at least closely approximated the current KDOQI and KDIGO definitions for CKD stages 1–3 were considered eligible.

Study Selection

We sought RCTs or CCTs that assessed the direct impact of systematic screening for CKD stages 1–3 on clinical outcomes and harms. Examples of tests to screen for CKD that were considered eligible were direct measurements of GFR or creatinine clearance, estimation of GFR or creatinine clearance with creatinine-based formulae, serum creatinine, albuminuria, proteinuria, albumin/creatinine ratio, and cystatin C. The screening method must have been feasible within a primary care setting. Our exclusion criteria were as follows: nonadult population, study participants already diagnosed with CKD, not an RCT that assigned participants to systematic screening for CKD versus usual care or a comparator intervention, study followup duration less than 1 year, and sample size less than 1,000 randomized participants.

When no RCTs were identified that evaluated a CKD screening intervention and reported clinical outcomes and harms, indirect evidence was reviewed regarding its possible benefits and harms. This indirect evidence included observational studies on CKD prevalence, clinical recognition, accuracy and reliability of CKD screening tests, and RCTs of CKD treatments. Although these observational studies were not identified through a comprehensive literature search, whenever possible we evaluated data from large representative U.S. cohorts. Assessment of CKD treatment benefits and harms was based strictly on direct evidence from RCTs.

Comparators

Studies were to compare systematic screening for CKD stages 1–3 with no CKD screening, usual care, or an alternative CKD screening regimen. Any monitoring or treatment interventions that followed screening were allowed.

Outcomes

We restricted the review to studies that reported clinical outcomes or harms. Clinical outcomes included: all-cause mortality, cardiovascular mortality, MI (any, fatal, nonfatal), stroke (any, fatal, nonfatal), CHF (hospitalization, death), composite vascular outcomes, composite renal outcomes, ESRD (progression to kidney transplant or dialysis), quality of life, physical function, and activities of daily living. Intermediate outcomes included: progression to stage 4 or stage 5 kidney disease, composite renal outcomes, doubling of serum creatinine or halving of GFR, and conversion from microalbuminuria to macroalbuminuria. Harms included: any adverse events, serious adverse events, specific adverse events, and any renal adverse events.

Study Designs

We initially included only RCTs. As described above, when no relevant RCTs were identified, we expanded our search to include observational studies that could provide indirect evidence regarding these questions.

Key Questions 3 and 4

Patients

We restricted the review to studies that enrolled adults with CKD stages 1–3 who were systematically monitored for worsening of kidney function and/or damage. As above, studies whose definitions of CKD stages 1–3 at least closely approximated the current KDOQI and KDIGO definitions were considered eligible.

Study Selection

We sought RCTs or CCTs that assessed the direct impact of monitoring on clinical outcomes and harms. Examples of tests to monitor for worsening kidney function and/or damage that were considered eligible were direct measurements of GFR or creatinine clearance, estimation of GFR or creatinine clearance with creatinine-based formulae, serum creatinine, albuminuria, proteinuria, albumin/creatinine ratio, and cystatin C. The monitoring method must have been feasible within a primary care setting. Our exclusion criteria were as follows: nonadult population, population entirely or predominately not CKD stages 1–3, not an RCT that assigned participants to systematic monitoring for worsening of kidney function and/or damage versus usual care or a comparator intervention, and sample size of less than 50 randomized participants.

When no RCTs were identified that evaluated a CKD monitoring intervention and reported clinical outcomes or harms, indirect evidence was reviewed regarding its possible benefits and harms. This indirect evidence included observational studies on CKD progression, clinical recognition, accuracy and reliability of CKD monitoring tests, and RCTs of CKD treatments. Although these observational studies were not identified through a comprehensive literature search, whenever possible we evaluated data from large representative U.S. cohorts. Assessment of CKD treatment benefits and harms was based strictly on direct evidence from RCTs.

Comparators

Studies were to compare systematic monitoring of patients with CKD stages 1–3 for changes in kidney function and/or damage with usual care or an alternative CKD monitoring regimen. Any interventions that followed CKD monitoring were allowed.

Outcomes

We restricted the review to studies that reported clinical outcomes or harms. Clinical outcomes included: all-cause mortality, cardiovascular mortality, MI (any, fatal, nonfatal), stroke (any, fatal, nonfatal), CHF (hospitalization, death), composite vascular outcomes, composite renal outcomes, ESRD (progression to kidney transplant or dialysis), quality of life, physical function, and activities of daily living. Intermediate outcomes included: progression to stage 4 or stage 5 kidney disease, composite renal outcomes, doubling of serum creatinine or halving of GFR, and conversion from microalbuminuria to macroalbuminuria. Harms included: any adverse events, serious adverse events, specific adverse events, and any renal adverse events.

Study Designs

We initially included only RCTs. As described above, when no relevant RCTs were identified, we expanded our search to include observational studies that could provide indirect evidence regarding these questions.

Key Questions 5 and 6

Patients

We restricted the review to studies that enrolled adults with CKD stages 1–3. Again, studies whose definitions of CKD stages 1–3 at least closely approximated the current KDOQI and KDIGO definitions were considered eligible.

Interventions

We included studies of both CKD specific and nonspecific treatments. Specifically, we attempted to identify studies of ACEI, ARB, calcium channel blockers (CCB), aldosterone antagonists, alpha blockers, beta blockers (BB), loop diuretics, thiazide and related diuretics, combination antihypertensive regimens, targeting thresholds of blood pressure control independent of specific antihypertensive agent(s), insulin, sulfonylureas, thiazolidinediones, biguanides (e.g., Metformin), targeting thresholds for glycemic control, HMG CoA-reductase inhibitors (i.e., statins), bile acid sequestrants, cholesterol absorption inhibitors (e.g., Ezetimibe), anorexiants, lipase inhibitors, low protein diets, and other diets.

Comparators

These studies compared active treatment of patients with CKD stages 1–3 with placebo, usual care/no treatment, or with other active treatments, including combination treatment and comparisons with the same active treatments using different dose levels or targeting different treatment thresholds.

Outcomes

We restricted the review to studies that reported clinical outcomes or harms. Clinical outcomes included: all-cause mortality, cardiovascular mortality, MI (any, fatal, nonfatal), stroke (any, fatal, nonfatal), CHF (hospitalization, death), composite vascular outcomes, composite renal outcomes, ESRD (progression to kidney transplant or dialysis), quality of life, physical function, and activities of daily living. Intermediate outcomes included: progression to stage 4 or stage 5 kidney disease, composite renal outcomes, doubling of serum creatinine or halving of GFR, and conversion from microalbuminuria to macroalbuminuria. Harms included: any adverse events, serious adverse events, specific adverse events, and any renal adverse events.

Study Designs

We only included RCTs.

Study Selection

Separate literature searches were completed for the three main topic areas: screening, monitoring, and treatment. Results of each literature search were imported to a spreadsheet for screening. Trained reviewers examined all titles and abstracts for eligibility based on the inclusion/exclusion criteria for the topic area of the search. Titles and abstracts with insufficient information to determine eligibility were pulled for full article text review. If the initial reviewer was uncertain about eligibility, one of the physician project leads reviewed the abstract (or article) and made a final decision about inclusion or exclusion. We selected a 10 percent sample (representing the work of all abstract reviewers) for repeat review. Based on discrepancies between the results of one initial reviewer and the second reviewer, all abstracts reviewed by that initial reviewers were reviewed a second time. Overall, we asked abstract reviewers to err on the side of inclusion rather than exclusion. Reasons for exclusion were tallied in the spreadsheet and entered in an EndNote file for reference list management. We also applied the inclusion/ exclusion criteria to studies identified in the hand search of reference lists and in the review of studies cited in relevant systematic reviews. Additional references suggested by members of our TEP and by the public during the comment period also were reviewed for eligibility. A list of excluded studies is included in Appendix B.

Data Extraction

For the treatment interventions, trained clinicians or research assistants extracted data onto a spreadsheet. After verifying study eligibility, we extracted the following data from each trial:

  • Study quality: Allocation concealment, intention-to-treat analysis, blinding, withdrawals, and dropouts adequately described;
  • Study characteristics: Location, number of sites, subject inclusion and exclusion criteria, source of study subjects, total number randomized, details of treatment and control group interventions;
  • Baseline participant data: age, weight, body mass index, gender, race/ethnicity, CKD stage, estimated or directly measured GFR, serum creatinine, urinary albumin or protein excretion rate, creatinine clearance, urine albumin or protein creatinine ratio, glycosylated hemoglobin or hemoglobin A1c (HbA1c), blood pressure, cholesterol, smoking status, and history of diabetes, hypertension, dyslipidemia, coronary artery disease, congestive heart failure (CHF), peripheral arterial disease, myocardial infarction (MI), stroke, and history of acute kidney injury;
  • Efficacy outcomes: Duration of followup, all-cause mortality, cardiovascular mortality, MI (any, fatal, nonfatal), stroke (any, fatal, nonfatal), CHF (hospitalization, death), composite vascular outcomes, ESRD (progression to kidney transplant or dialysis), progression to stage 4 or stage 5 kidney disease, composite renal outcomes, doubling of serum creatinine or halving of GFR, conversion from microalbuminuria to macroalbuminuria, whether continuous renal outcomes were reported, and whether quality of life, physical function or activities of daily living were reported; and
  • Withdrawals and adverse events: any withdrawals, withdrawals due to adverse events, any adverse events, serious adverse events, specific adverse events, and any renal adverse events.

Articles identified as not meeting eligibility criteria during the extraction phase were tallied and documented on the study flow diagram. In preparing the tables and text, a second clinician or research assistant confirmed the accuracy of the extracted information by comparing the extracted information with the original article. A physician project lead verified all entries in tables included in the review and appendices.

Quality Assessment

Study quality for the individual RCTs was rated by using the following criteria based on the domains the Cochrane Collaboration recommends to assess the risk of bias of studies included in a systematic review:36 (1) adequate allocation concealment, based on the approach by Schulz and Grimes;37 (2) blinding methods (participant, investigator, and/or outcome assessor); (3) how incomplete data are addressed (did the study analyze the data based on the intention-to-treat principle, i.e., were all participants who were randomized included in the outcomes analyses); and (4) whether reasons for dropouts/attrition were reported. Studies were rated as good, fair, or poor quality. A rating of good generally indicated that the trial reported adequate allocation concealment, blinding, analysis by intent-to-treat, and reasons for dropouts/attrition were reported. Studies were generally rated poor if the method of allocation concealment was inadequate or not defined, blinding was not defined, analysis by intent-to-treat was not utilized, and reasons for dropouts/attrition were not reported and/or there was a high rate of attrition.

Rating the Body of Evidence

The overall strength of evidence for the randomized trials was evaluated using methods developed by AHRQ and the Effective Health Care Program.38 For each of several important clinical outcomes within each comparison evaluated, the strength of the evidence was evaluated based on four required domains: (1) risk of bias (do the studies for a given outcome or comparison have good internal validity); (2) consistency (the degree of similarity in the effect sizes, i.e., same direction of effect, of the included studies); (3) directness (reflecting a single, direct link between the intervention of interest and the outcome); and (4) precision (degree of certainty surrounding an effect estimate of a given outcome). The risk of bias, based on study design and conduct, is rated low, medium, or high. Consistency is rated consistent, inconsistent, or unknown/not applicable (e.g., a single study was evaluated). Directness can either be direct or indirect and precision is either precise or imprecise. A precise estimate is one that would yield a clinically meaningful conclusion. Based on these four domains, the overall evidence was rated as: (1) high, indicating high confidence that further research is very unlikely to change the confidence in the estimate of effect, meaning that the evidence reflects the true effect; (2) moderate, indicating moderate confidence that further research may change our confidence in the estimate of effect and may change the estimate; (3) low, indicating low confidence that further research is very likely to have an important impact on the confidence in the estimate of effect and is likely to change the estimate, meaning there is low confidence that the evidence reflects the true effect; and (4) insufficient, indicating that evidence either is unavailable or does not permit a conclusion. An overall rating of high strength of evidence would imply that the included studies were RCTs with a low risk of bias, with consistent, direct, and precise domains.

Applicability

Applicability of the results reported in this review is affected by the representativeness of the patient samples in the included studies to general populations and specific subpopulations of nonstudy patients with CKD stages 1–3, both those identified through screening and through other means. All treatment trials included patients with CKD stages 1–3, but because of the variability in CKD definitions used in identified studies, some trials also included some patients outside the bounds defined by CKD stages 1–3. This may limit the applicability of results reported here to patients who meet the currently accepted definition for CKD stages 1–3. Incomplete reporting of patient characteristics in many included trials also limits our ability to judge applicability of study results to specific CKD patient populations. The evidence tables in Appendix C identify reported details on the patient inclusion and exclusion criteria, as well as baseline patient characteristics.

Data Synthesis

Text; evidence, outcomes, and summary tables; and figures were organized by intervention. If clinical heterogeneity of patient populations, interventions, and outcomes was minimal, we pooled results. For many interventions, there were only one or two trials and reported outcomes did not overlap. Narratives provide details on study populations, interventions, clinical outcomes, and harms. Data were analyzed in Review Manager 5.0.39 Random effects models were used to generate pooled estimates of relative risks (RR) and 95 percent confidence intervals (CI). Statistical heterogeneity was summarized using the I2 statistic (50 percent indicates moderate heterogeneity and 75 percent or greater indicates high heterogeneity).40

Publication Bias

Grey literature was searched for relevant trials and other material to estimate the likelihood of publication bias. Sources of regulatory documents included Federal Drug Administration – Medical Reviews and Statistical Reviews, Health Canada – Drug Monographs, and Authorized Medicines for the European Union. Clinical trial registries accessed were ClinicalTrials.gov, Current Controlled Trials, Clinical Study Results, and World Health Organization's Clinical Trials. Conference papers and abstracts were identified from the CSA Conference Papers Index and Scopus.

Cover of Chronic Kidney Disease Stages 1–3
Chronic Kidney Disease Stages 1–3: Screening, Monitoring, and Treatment [Internet].
Comparative Effectiveness Reviews, No. 37.
Fink HA, Ishani A, Taylor BC, et al.

AHRQ (US Agency for Healthcare Research and Quality)

PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...