Appendix 1: methods used for guideline development

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT AIMS

T

HE OVERALL AIM of the project was to develop evidence-based recommendations for blood pressure management in individuals with CKD. The Work Group sought to develop an “evidence base” for the recommendations that is derived from a systematic review of the available scientific literature on the evaluation of blood pressure in individuals with CKD, measurement of blood pressure, treatment with lifestyle modifications, and treatment with pharmacological therapy, including selection of antihypertensive agents and treatment targets. Topics considered are listed in Table 165. As a result, this report consists of a set of clinical practice guidelines and the summary tables presenting the evidence on which the guidelines are based. Definitions Table 166 gives a list of definitions used in the guidelines. In addition, the Work Group developed the following operational definitions for these guidelines: Antihypertensive therapy includes lifestyle modifications and pharmacological therapy that lower blood pressure, in patients with or without hypertension. Lifestyle modifications include changes in diet, exercise and habits that may lower the risks of progression of CKD and CVD. These guidelines focus specifically on lifestyle modifications that lower blood pressure. Lifestyle modifications are discussed in more detail in Guideline 6. Pharmacological therapy includes selection of antihypertensive agents and blood pressure goals. General principles of pharmacological therapy and target blood pressure for CVD risk reduction are discussed in Guideline 7. Antihypertensive agents are defined as agents that are usually prescribed to lower blood pressure. Other agents may also lower blood pressure as a side-effect. Many antihypertensive agents have effects in addition to lowering systemic blood pressure and are used for indications other than hypertension. “Preferred agents” are classes of antihypertensive agents that have beneficial effects on progres-

sion of CKD or reducing CVD risk, in addition to their antihypertensive effects, such as reducing proteinuria, slowing GFR decline, and inhibiting other pathogenetic mechanisms of kidney disease progression and CVD. In certain types of CKD, specific classes of antihypertensive agents, notably those that inhibit the RAS, are preferred agents for slowing progression of CKD. Thus, the Work Group developed guidelines that recommend the use of specific classes of antihypertensive agents in certain types of CKD, even if hypertension is not present. Target Population Based on the results of clinical and epidemiological studies, the Work Group defined the target population for these guidelines as adults or children with CKD Stages 1-4. CKD is defined as kidney damage (abnormalities on biopsy, urinalysis, abnormal imaging studies) and/or GFR ⬍60 mL/min/1.73 m2 and includes the following causes of CKD: diabetic kidney disease, glomerular diseases, nephrosclerosis, tubulointerstitial diseases, cystic diseases, diseases in the kidney transplant, and renal artery disease. Stages 1-4 correspond to a GFR ⱖ15 mL/min/1.73 m2. Individuals on dialysis or those with a GFR ⬍15 mL/min/1.73 m2 were excluded based on the different pathophysiology and treatment modalities applicable to them. Kidney transplant recipients with CKD Stages 1-4 were included. Patients with CKD Stage 5 (kidney failure) were excluded for several following reasons. First, kidney disease progression may not be as important in patients who have already reached the stage of kidney failure. Second, the relationship between CVD risk and level of blood pressure is complex in kidney failure. Third, blood pressure in hemodialysis patients is affected by intermittent fluid gains and fluid removal. The Work Group has included recommendations for both adults and children. Guideline 13 for children was written with careful consideration to past recommendations for the treatment of hyper-

© 2004 by the National Kidney Foundation, Inc. 0272-6386/04/4305-0104$30.00/0 doi:10.1053/j.ajkd.2004.03.008

American Journal of Kidney Diseases, Vol 43, No 5, Suppl 1 (May), 2004: pp S231-S248

S231

S232

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

tension in children as noted in JNC 615 and from the National High Blood Pressure Education Program Work Group for children and adolescents.10 Figure 62 shows the evolution of National Kidney Foundation guidelines for the management of hypertension in patients with CKD. Intended Users These guidelines are intended for use by physicians, nurse practitioners, registered nurses, registered dietitians, masters prepared social workers, pharmacists, physician assistants, and other professionals caring for patients with CKD. The information contained in these guidelines can and should be conveyed to patients and their families in an understandable manner by their physician and/or other health-care professionals.

The development of educational support materials designed specifically for patients and their families should be part of the implementation of these guidelines. Anticipated Updates All guidelines should be updated whenever new, pertinent information becomes available. To anticipate when these guidelines may need to be updated, the Work Group discussed ongoing, controlled trials in the general population and in patients with CKD, as those results may be pertinent to some recommendations (Table 167).6 Given the potential for these and other studies to provide information relevant to the assessment and treatment of hypertension in patients with CKD, it was concluded that these guidelines

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

Fig 62. The evolution of National Kidney Foundation Guidelines on Hypertension and Antihypertensive Agents in CKD.

should be updated in about 3 to 4 years from the time of publication, and sooner if new, pertinent information becomes available before then. The Work Group will monitor the progress of these trials and recommend updating these guidelines as indicated. Overview of Process

K/DOQI Principles and Process The development of these guidelines followed four basic principles set forth by K/DOQI (Table 168). The guidelines were developed using an evidence-based approach similar to that endorsed by the Agency for Healthcare Research and Quality. Development of the guideline and evidence report required many concurrent steps (Table 169). The Work Group reviewed all pertinent, published evidence, and critically appraised the quality of studies and the overall strength of evidence supporting each recommendation.

S233

Creation of Work Group and Evidence Review Team The Work Group was convened by the NKF Kidney Disease Outcomes Quality Initiative (K/ DOQI) in response to recommendations of the NKF Task Force on CVD in Chronic Renal Disease.3 The Co-Chairs of the K/DOQI Advisory Board selected the Work Group Chair and Director of the Evidence Review Team, who then assembled groups responsible for the development of the guidelines and the evidence report, respectively. The Work Group and Evidence Review Team first met in August 2000. Both groups collaborated closely throughout the project. The Work Group consisted of “domain experts,” including individuals with expertise in adult and pediatric nephrology, cardiology, epidemiology, nutrition, social work, and family medicine. In addition, the Work Group had liaison members from the National Institute of Diabetes, Digestive and Kidney Diseases, from the National Heart Lung and Blood Institute and from the Renal Physicians Association. The Work Group also maintained communication with members of the JNC 75 and ADA 20036 and with investigators of the Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial (ALLHAT)107 for which results were announced during the time the Work Groups held meetings. The first task of the Work Group chair and the members was to define the overall goals and topics, including specifying the target condition, target population, and target audience. The domain experts then assisted the evidence review team in further refining topics, developing the literature search strategy, and drafting data extraction forms. The Work Group members were the principal reviewers of the literature. They used the evidence compiled in summary tables as a basis for the guidelines and took the lead in writing the guidelines and rationale statements. The Evidence Review Team consisted of nephrologists and methodologists from Tufts-New England Medical Center and elsewhere with expertise in systematic review of the medical literature. They were responsible for seeing to project timelines and compiling of the evidence report. Specific tasks included refining the questions to be addressed, developing literature search strate-

S234

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

S235

gies, establishing inclusion and exclusion criteria, running searches, screening of abstracts and articles, verification of data extraction, compilation of evidence and summary tables, consultation on all aspects of methodology and process, coordination and tracking of project tasks. Throughout the project the Evidence Review Team led discussions on the methodology of systematic review, literature searches, data extraction, assessment of quality of articles, summary reporting, formulation of guidelines statements and rationale text, and grading of the strength of guideline recommendations and rationale statements. Development of Topics The goals of the Work Group spanned the following topics: (1) evaluation of individuals with CKD or hypertension, (2) measurement of blood pressure including ambulatory blood pressure monitoring (ABPM), (3) evaluation of renal artery disease (RAD), (4) treatment of hypertension with lifestyle modifications, and (5) use of antihypertensive agents, including selection of antihypertensive agents and blood pressure targets, to slow progression of kidney disease and reduce CVD risk in CKD. Refinement of Topics and Development of Materials The Work Group and Evidence Review Team developed (1) draft guideline statements; (2) draft rationale statements that projected the pertinent evidence; (3) mock summary tables providing a shell for the expected evidence; and (4) data extraction forms prompting the reviewers for relevant data elements to be retrieved from the primary articles. The development process included creation of initial mock guidelines and tables by the Work Group Chair and Evidence Review Team, followed by iterative refinement by the Work Group members. The refinement process began prior to literature retrieval and continued through all stages of the project until the drafting of the final report. The refinement occurred by e-mail, telephone, and in-person communication regularly with local experts and with all experts during in-person meetings of the Evidence Review Team and Work Group members. Throughout the process of topic refinement,

S236

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

the type of study design that would be appropriate to address the topics of interest was carefully considered. Data extraction forms were designed to capture information on various aspects of the primary articles. Data extraction included objective information such as study setting and patient demographics, eligibility criteria, causes of kidney disease, numbers of subjects, study design, study funding source, definitions of measures, interventions in controlled trials and outcomes (mortality, CVD, kidney, and other outcomes as well as adverse events). Work Group members were also asked to provide their assessments of applicability of the study population (see below), methodological quality of the study (based on criteria appropriate for each study design, see below), appropriateness of the selected mea-

sures, and assessment of biases and other comments. The Evidence Review Team conducted training sessions for the Work Group members to learn systematic data extraction from primary articles. Evidence Review Team reviewed and verified the data forms completed by Work Group members. Feedback was provided to individual members and discrepancies were reconciled by e-mail and teleconferences. Review of Existing Guidelines on CKD There were several prior guideline recommendations for blood pressure management in CKD, as well as other documents that became available during the course of guideline development and these were reviewed by the Work Group. The recommendations issued in 1997 by the Sixth

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

Joint National Commission (JNC 6) were considered a reference source.15 Thus, the search for other guidelines with recommendations on blood pressure management in CKD was restricted to guidelines published since 1997. Searches were conducted through the National Guideline Clearing House on the following keywords: hypertension, kidney disease, diabetes, and cardiovascular disease, spanning the time from 1997 until March 2003. Approximately 150 guideline citations were found. All guidelines that could be downloaded free of charge or retrieved from published medical journals were searched for recommendations for blood pressure management and use of antihypertensive agents in CKD. Through this search, we found 11 guidelines with pertinent sections and supplemented this with other known recommendations (Table 170). The relevant recommendations of these other guidelines are reviewed in this guideline document in Guidelines 8, 9, 10, and 13. Review of Existing Guidelines on CVD in the General Population Because we found few studies that examined the effect of pharmacological antihypertensive therapy on CVD outcomes in CKD, the Work Group decided to extrapolate the results from studies conducted in the general population and in other, high-risk populations to those with CKD. Some would argue that no guideline statements should be made in the absence of evidence on clinical outcomes in the target population. However, the limited number of studies on antihypertensive therapy in CKD with CVD outcomes, compared to the large number of studies in the general population, required development of criteria for extrapolating evidence from the general population to the target population. Table 171 compares evidence derived from studies in the general population to evidence derived from studies in patients with CKD. Based on this comparison, the Work Group adopted the criteria developed by the NKF Task Force on Cardiovascular Disease in Chronic Renal Disease for extrapolating evidence collected in trials studying cardiovascular outcomes, mortality, or total mortality from the general population to patients with CKD. 1. The mechanisms and expression of CVD in CKD should be similar to those observed in the

S237

general population. Specifically, the features of CVD, the relationship of the risk factor (hypertension) to CVD outcomes, the mechanism of risk factor reduction (blood pressure lowering), and the responsiveness of the risk factor to therapies (lifestyle modifications and pharmacological therapy) should be similar in patients with CKD and in the general population. 2. Therapies in patients with CKD should be as safe, or nearly so, as in the general population. In particular, there should not be additional adverse effects of a specific therapy that limits its usefulness in patients with CKD, either because of altered pharmacokinetics, drug interactions, or increased risk of toxicity to the kidney. 3. The duration of therapy required to improve CVD outcomes in the general population should not exceed the life expectancy of patients with CKD. In other words, it should not already be too late to intervene in this generally elderly, sick, and frail population. Determining whether patients with CKD can survive for long enough to gain the benefit of therapy for CVD is a difficult question. Numerous studies show a dramatically shortened life expectancy for patients with CKD, especially patients with kidney failure. For example, the USRDS395 has estimated that the average life expectancy of 60- to 64-yearold patients treated by dialysis ranges from 3.6 to 5.1 years, depending on gender and race. On the other hand, the most common cause of death in kidney failure is CVD, and numerous studies of CVD in the general population have shown a benefit of interventions within 2 to 5 years, with greater and earlier benefits in patients at highest risk. Thus, it is likely that patients with kidney failure could benefit from more effective treatment of CVD. Because of their longer life expectancy, patients with earlier stages of CKD might be most likely to benefit. However, a systematic review of all the pharmacological blood pressure trials with CVD outcomes in the general population was felt to be beyond the scope or the expertise of the Work Group. Therefore, it was decided to review existing guidelines for recommendations on blood pressure targets and specific antihypertensive agents for prevention of CVD. The same search that was conducted through the National Guideline Clearinghouse to identify existing guide-

S238

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

lines with recommendations on managing blood pressure in CKD (see above) was also screened for sections containing recommendations on risk reduction for CVD using antihypertensive agents

(Table 172). This yielded 13 guidelines which were abstracted and compiled into tables and reviewed by Work Group experts. The relevant recommendations are reviewed in Guideline 7.

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

Literature Search of Primary Articles The Work Group and Evidence Review Team decided in advance that a systematic process would be followed to identify primary studies on the topics of interest. Only fully published articles with original data were included. Review articles, editorials, letters or abstracts were generally excluded. The only exception were abstracts of the ALLHAT trial reporting kidney disease progression and CVD outcomes in a large subgroup of individuals with CKD.292,293 The Work Group members and Evidence Review Team selected textbooks and review articles based on personal knowledge. Studies for the literature review were identified primarily through Medline searches of the English language literature. The Medline literature searches were conducted between July 2001 and July 2002 to identify clinical studies published from 1966 through the search dates. Separate search strategies were developed for each topic. Development of the search strategies was an iterative process that included input from all members of the Work Group. The text words or medical subject headings (MeSH) included kidney or kidney diseases or kidney function tests or hypertension or renal, diabetic nephropathy, renal artery, and ambulatory blood pressure monitoring. The searches were limited to human studies. Studies that focused on hemodialysis or

S239

peritoneal dialysis, pregnancy, neonates, malignant hypertension, acute renal failure, or pharmacokinetics were excluded. The Evidence Review Team screened citations identified by the Medline search. Potentially relevant articles were identified from abstracts and titles, based on study population, relevance to specific guideline topics, and the study design. In general, studies with fewer than 10 subjects per treatment arm were excluded. However, for pediatric topics, due to the small number of articles, only case reports were excluded. Relevant articles known to domain experts and reviewers supplemented these searches. After retrieval, each paper was screened according to the established criteria to verify its relevance and appropriateness. Work Group members assigned to the specific topic reviewed the articles and made the final decision for inclusion or exclusion of articles. However, they had to provide the reason for rejection. Data extraction was performed on all included articles and their data compiled into evidence tables. Additional relevant studies published since July 2002 were added by experts. Table 173 lists the details of the literature search and review for each topic. Overall, 11,688 abstracts were screened by the Evidence Review Team, 899 articles were retrieved and reviewed, and data were extracted from 177 articles. Forty-

S240

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

seven articles were added by the Work Group. Finally, results from 76 articles were systematically included in the summary tables. The Medline search strategies are shown in Appendix 2.

Study Selection Evaluation and blood pressure measurement. Literature review did not yield any controlled trials of adequate quality or study size in individuals with CKD on the topics of evalua-

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

tion, blood pressure measurement (with the exception of studies on ABPM), and lifestyle modifications as treatment for blood pressure. For these topics, we used existing guidelines. ABPM in CKD. The Work Group felt that a detailed discussion on technical aspects of ABPM and its use in the management of blood pressure in CKD was important. However, since the evidence was not strong enough to recommend routine use of ABPM in CKD at the present time, a detailed review of this topic was included as a technical appendix. The issues the Work Group had identified to address regarding the use of ABPM in CKD were (1) correlation of office blood pressure with ambulatory blood pressure (ABP); (2) prevalence of white coat hypertension (WCH); (3) prevalence of abnormal blood pressure patterns; (4) association between ABP and kidney or CVD outcomes; (5) risk relationship of ABPM with subsequent clinical outcomes; and (6) impact of ABPM on blood pressure control or antihypertensive therapy. The search strategy for ABPM is shown in Appendix 2. During the course of the guideline development process, an evidence report on “Utility of Blood Pressure Monitoring Outside of the Clinic Setting” commissioned by the Agency of Health-

S241

care Research and Quality (AHRQ) became available.142 This report had systematically synthesized the available evidence on ABPM addressing the questions our Work Group had drafted and using very similar screening criteria. In addition to monitoring of blood pressure by ABPM, the AHRQ report had also examined the utility of self-measured blood pressure. The inclusion criteria of the AHRQ report were not restricted to, but included, studies conducted in the target population of adults or children with CKD. Thus, the Work Group felt that duplication of effort was not warranted and formal data abstraction of the articles identified by the search or synthesis of results into summary tables was not pursued. Rather, it was decided to incorporate the findings of the AHRQ report into this guideline and to cite the previously retrieved studies in a narrative format. Renal Artery Disease (RAD). For the topic of evaluation of individuals with CKD for RAD, we examined the literature for meta-analyses comparing the accuracy of diagnostic tests for RAD. This yielded one relevant study, which was reviewed according to the criteria in Tables 174 and 175. Another study that contains a clinical

S242

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

prediction model was added by the Work Group to estimate the probability of RAD. Nonpharmacological therapies. We searched for nonpharmacological therapy trials such as lifestyle modifications with diet, exercise, and weight reduction that were conducted to lower blood pressure and thereby reduce the risk for kidney or heart outcomes. We did not find any studies with a randomized or nonrandomized controlled design that had at least 10 CKD subjects per treatment arm and reported outcomes of interest. Pharmacological therapy with selected anti-

hypertensive agents or blood pressure treatment targets. For pharmacological therapy, we examined RCTs using antihypertensive agents or blood pressure modifying therapies. We reviewed studies providing comparisons on the two interventions of interest: use of selected antihypertensive agents or achievement of predefined blood pressure targets (Table 176). Outcomes of interest were kidney outcomes, cardiovascular outcomes, and total mortality and adverse effects. For evaluation of antihypertensive agents, we reviewed RCTs in patients with CKD with or

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

without elevated blood pressure for secondary prevention. For example, we included studies of patients with the earliest stage of CKD due to diabetes (microalbuminuria), regardless of blood pressure at baseline (including normotensive patients). On the other hand, we excluded studies of patients with diabetes without kidney disease, even if hypertensive, thereby excluding primary prevention studies in diabetes. For evaluation of the level of blood pressure, we reviewed RCTs comparing the effect of different predefined blood pressure ranges that were targeted with a combination of antihypertensive agents. Since the number of RCTs using predefined blood pressure targets was small, we also screened prospective cohort studies that reported risks of outcomes stratified by blood pressure levels. We were interested in data that allowed us to look at rates of events stratified by different blood pressure levels to discover nonlinear relationships between blood pressure level and risk for outcomes and to investigate threshold relationships in order to help define optimal blood pressure levels. However, since these results were reported in many different ways, they were difficult to compare across studies and did not allow us to answer the question of a threshold relationship with confidence. Thus, observational studies were not included. Instead, the recommendations for blood pressure targets were derived from the review of existing guidelines that built on the knowledge of a positive correlation between blood pressure level and CVD risk and the evidence that reduction of blood pressure reduced CVD risk. In addition, effect modifiers such as demographic or genetic factors were searched for as a routine part of the data abstraction of each treatment study. The questions of preferred agents and optimal blood pressure targets were posed for all types of kidney disease. Given the paucity of studies in kidney transplant recipients, we lowered the criteria. For kidney disease in the kidney transplant and outcome proteinuria, we accepted n ⱖ 10 per arm (instead of n ⱖ 50/arm) and follow-up ⱖ1 months (instead of ⱖ3 months) (Table 176). For RAD, screening yielded a limited number of studies that were not standardized very well, with overall short follow-up. The Work Group deemed this body of evidence to be insufficient

S243

to recommend selection of pharmacological agents or blood pressure targets. In addition, the Work Group decided that giving recommendations on how to treat individuals with RAD with other treatment modalities was beyond the scope of this project. This decision was based on the recognition that interventional therapy of renal artery disease is a field in flux because of the advent and testing of rapidly evolving novel interventional devices and techniques. In children, the searches yielded only few studies with small numbers and short follow-up. These studies are referred to in narrative form as appropriate rather than summarized in summary table format. Adverse events of preferred agents. The objective of this topic was to characterize the safety and tolerance of ACE inhibitors or ARBs in individuals with CKD. The rationale was that these agents would be recommended as preferred agents for a large number of individuals who have certain types of CKD. We wanted to address the concern that these agents can cause acute loss of kidney function or hyperkalemia, especially in individuals with decreased GFR. Thus, we collected data on the frequency and severity of adverse events from ACE inhibitors or ARBs, specifically, “early decrease of GFR” and hyperkalemia. We used RCTs, non-RCTs, prospective or retrospective cohort studies, and case-control studies in all types of CKD including RAD. Format for Evidence Tables Two types of tables were prepared using data extracted from accepted articles. Evidence tables contain data derived from the data extraction forms that covered features of the study design, patient demographics, disease characteristics, interventions, definitions of the outcomes and their results. These detailed evidence tables, with information spanning across several pages for each study, were made available to Work Group members for the purpose of reviewing the data and writing of guidelines. The evidence tables are not published along with the guidelines. Using the information in the evidence tables, the Evidence Review Team also prepared a set of summary tables that succinctly describe the characteristics for each study in six areas: study size; applicability (type of study subjects); baseline

S244

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

information (kidney function, level of proteinuria, and blood pressure in comparator group one); information on therapy in each treatment arm (antihypertensive agents or blood pressure targets tested as well as the blood pressure at the end of the study); results of the primary outcome (the main or composite primary outcome and the magnitude of the effect); and methodological quality. For outcomes, only the primary outcome is represented. Composite outcomes that were primarily kidney outcomes (eg, doubling of creatinine and ESRD) but also contained allcause mortality were coded as kidney out-

comes. Similarly, composite outcomes combining outcomes related to kidney function and proteinuria were summarized under kidney outcomes. Thus, the outcome for proteinuria is only coded in the proteinuria column for studies where proteinuria was the sole primary outcome. Within each of the summary tables, studies were ordered first by methodological quality (highest to lowest), then by applicability (most to least), and then by study size (largest to smallest). In summary tables reporting side effects, studies are ordered first by type of disease (diabetic kidney diseases first), then by study design (randomized

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

controlled trials first), then by study size (largest first). Summary tables are presented in the guideline section that corresponds to the respective topic. An example of a summary table is shown in Table 177. Grading of Individual Studies

Study Size The number of patients (N, sample size) is used as a measure of the weight of the evidence. In general, large studies provide more precise estimates of effects or associations. In addition, results from large studies are more likely to be applied to a broader population; however, large size alone does not guarantee a high degree of applicability. A study that enrolls a large number of selected patients may be less generalizable than several smaller studies that include a broad spectrum of patient populations. Applicability Applicability (also known as generalizability or external validity) addresses the issue of whether the study sample is sufficiently broad so that the results can be applied to the population of interest at large. The study sample is defined by the inclusion and exclusion criteria. The target population was de-

S245

fined to include patients with CKD and those at increased risk of CKD, except where noted. A designation for applicability was assigned to each article, according to a three-level scale. In making this assessment, sociodemographic characteristics were considered, as were the stated causes of CKD and prior treatments. If the study is not considered broadly generalizable, reasons for the limited applicability are reported. Table 178 describes our approach to assessing applicability. Results The type of results available in each study is determined by the study design, the purpose, and the question(s) being asked. The Work Group decided that the outcomes of interest were primary outcomes documenting effects on kidney disease progression, all-cause mortality, or cardiac outcomes in RCTs. Given the large spectrum of different outcomes, results were collapsed into four categories: direct health outcomes related to kidney disease progression (kidney failure, changes in GFR, changes in serum creatinine), surrogate outcome category related to kidney disease progression (proteinuria), direct health outcomes related to CVD (clinical events), and surrogate outcomes related to CVD (LVH or carotid artery intima thickness). Table 179 shows our approach in reporting comparisons of blood pressure targets or antihypertensive

S246

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

agents. For studies reporting adverse effects of ACE inhibitors and ARBs, the incidence of adverse effects (ie, drug reduction or discontinuation for hyperkalemia, for early decrease in GFR or for any adverse effect) were reported as percentages. Methodological Quality Methodological quality (or internal validity) refers to the design, conduct, and reporting of the clinical study. Because studies with a variety of types of design were evaluated, a generic threelevel classification of study quality was devised (Table 180). Summarizing Review Articles and Selected Original Articles Work Group members had wide latitude in summarizing review articles. They selected original articles for topics that were determined, a priori, not to require a systematic review of the literature. The use of published or derived tables and figures was encouraged to simplify the presentation. Guideline Format This document contains 13 guidelines. The format for each guideline is outlined in Table 181. Each guideline contains one or more specific, numbered “recommendations” or “guideline statements.” Each guideline contains background information, which is generally sufficient to interpret the guideline. The rationale for each guideline contains definitions, if appropriate, and a section on the strength of evidence. The strength of evidence includes a series of specific “rationale statements,” each supported by evidence, and “summary tables” (if appropriate) compiling and evaluating original

reports of studies. The guideline concludes with a discussion of limitations of the evidence review and a brief discussion of clinical applications, implementation issues, and research recommendations regarding the topic. Rating the Strength of Guidelines and Rationale Statements The overall strength of each guideline statement was rated by assigning either “A,” “B,” or “C” (Table 182). An “A” rating indicates “it is strongly recommended that clinicians routinely follow the guideline for eligible patients. There is strong evidence that the practice improves health outcomes, and benefits substantially outweigh harms.” The “B” rating indicates “it is recommended that clinicians routinely follow the guideline for eligible patients. There is moderate evidence that the practice improves health outcomes.” A “C” rating indicates “it is recommended that clinicians consider following the guideline for eligible patients. This recommendation is based on either weak evidence, or on the opinions of the Work Group and reviewers, that the practice might improve health outcomes.” The strength of evidence was graded using a rating system that takes into account: (1) methodological quality of the studies; (2) whether or not the study was carried out in the target population, ie, patients with CKD, or in other populations; and (3) whether the studies examined health outcomes directly, or examined surrogate measures for those outcomes, eg, reducing proteinuria rather than slowing progression of CKD (Table 183). These three separate study characteristics were combined in rating the strength of a body of evidence provided by the composite of the pertinent studies.

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

S247

S248

APPENDIX 1: METHODS USED FOR GUIDELINE DEVELOPMENT

In addition, the Work Group adopted a convention for using existing expert guidelines issued for populations other than the target population. Grades assigned by the guideline-issuing bodies for the strength of evidence were adopted. When the guideline or the evidence were not graded, this Work Group assumed that the guideline would be based on at least moderately strong evidence. The extrapolation of ungraded guideline recommendations from the general populations to the target population was considered to support grade B recommendations. Limitations of Approach While the literature searches were intended to be comprehensive, they were not exhaus-

tive. Medline was the only database searched, and searches were limited to English language publications. Hand searches of journals were not performed, and review articles and textbook chapters were not systematically searched. Important studies known to the domain experts and reviewers that were missed by the literature search were included in the review. Due to the wide variety of methods of analysis, units of measure, definitions of CKD, and methods of reporting in the original studies, it was often very difficult to standardize the findings for this report.

APPENDIX 2: MEDLINE SEARCH STRATEGIES SEARCH 1: CONTROLLED TRIALS IN CKD WITH HYPERTENSION

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

15. 16. 17.

18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38.

exp kidney diseases/ exp Kidney Glomerulus/ exp Kidney Function Tests/ 1 or 2 or 3 exp hypertension/ exp hypertension, renal/ hypertens$.af. high blood pressure.af. (eleva$ adj6 blood pressure).tw. 5 or 6 or 7 or 8 or 9 4 and 10 limit 11 to human limit 12 to english language limit 13 to (addresses or bibliography or biography or comment or dictionary or directory or editorial or festschrift or interview or lectures or legal cases or letter or news or periodical index) case report/ 13 not (14 or 15) limit 16 to (guideline or meta analysis or practice guideline or review or review literature or review of reported cases or review, academic or review, multicase or review, tutorial) 16 not 17 follow-up studies/ follow-up.tw. exp Case-Control Studies/ case-control.tw. exp Longitudinal Studies/ longitudinal.tw. exp Cohort Studies/ cohort.tw. (random$ or rct).tw. exp Randomized Controlled Trials/ exp random allocation/ exp Double-Blind Method/ exp Single-Blind Method/ randomized controlled trial.pt. clinical trial.pt. (clin$ adj trial$).tw. ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$ or mask$)).tw. exp PLACEBOS/ placebo$.tw. exp Research Design/

39. 40. 41. 42.

43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69.

70. 71. 72. 73.

74.

exp Evaluation Studies/ exp Prospective Studies/ exp Comparative Study/ 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 18 and 42 (random$ or rct).tw. exp Randomized Controlled Trials/ exp random allocation/ exp Double-Blind Method/ exp Single-Blind Method/ randomized controlled trial.pt. clinical trial.pt. (clin$ adj trial$).tw. ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$ or mask$)).tw. exp PLACEBOS/ placebo$.tw. exp Comparative Study/ exp controlled trials/ 44 or 45 or 46 or 47 or 48 or 49 or 50 or 51 or 52 or 53 or 54 or 55 or 56 18 and 57 43 not 58 18 and (dh or dt or rt or su or th).fs. 60 and 57 18 and dt.fs. 62 and 57 61 not 63 43 not 58 43 and (dh or dt or rt or su or th).fs. limit 18 to all child 具0 to 18 years典 18 not 67 limit 68 to (adult 具19 to 44 years典 or middle age 具45 to 64 years典 or “aged 具65 and over典” or “aged 具80 and over典”) 18 not 69 limit 18 to “aged 具65 and over典” 18 not 71 limit 72 to (all infant or all child 具0 to 18 years典 or newborn infant or infant 具1 to 23 months典 or preschool child 具2 to 5 years典 or child 具6 to 12 years典 or adolescence 具13 to 18 years典 or adult 具19 to 44 years典 or middle age 具45 to 64 years典) 18 not 73

American Journal of Kidney Diseases, Vol 43, No 5, Suppl 1 (May), 2004: pp S249-S251

S249

S250

APPENDIX 2: MEDLINE SEARCH STRATEGIES

SEARCH 2: CONTROLLED TRIALS IN CKD OR PROTEINURIA WITH HYPERTENSION OR WITH ANTIHYPERTENSIVE AGENTS

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17. 18.

19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37.

exp Proteinuria/ exp kidney diseases/ exp Kidney Glomerulus/ exp Kidney Function Tests/ 2 or 3 or 4 exp hypertension/ exp hypertension, renal/ hypertens$.af. high blood pressure.af. (eleva$ adj6 blood pressure).tw. 6 or 7 or 8 or 9 or 10 5 and 11 limit 12 to human limit 13 to english language limit 14 to (addresses or bibliography or biography or comment or dictionary or directory or editorial or festschrift or interview or lectures or legal cases or letter or news or periodical index) case report/ 14 not (15 or 16) limit 17 to (guideline or meta analysis or practice guideline or review or review literature or review of reported cases or review, academic or review, multicase or review, tutorial) 17 not 18 follow-up studies/ follow-up.tw. exp Case-Control Studies/ case-control.tw. exp Longitudinal Studies/ longitudinal.tw. exp Cohort Studies/ cohort.tw. (random$ or rct).tw. exp Randomized Controlled Trials/ exp random allocation/ exp Double-Blind Method/ exp Single-Blind Method/ randomized controlled trial.pt. clinical trial.pt. (clin$ adj trial$).tw. ((singl$ or doubl$ or trebl$ or tripl$) adj (blind$ or mask$)).tw. exp PLACEBOS/

38. 39. 40. 41. 42. 43.

44. 45. 46. 47. 48. 49.

50. 51. 52. 53.

54. 55. 56. 57. 58. 59. 60.

61. 62.

63. 64.

placebo$.tw. exp Research Design/ exp Evaluation Studies/ exp Prospective Studies/ exp Comparative Study/ 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29 or 30 or 31 or 32 or 33 or 34 or 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 exp Antihypertensive Agents/ 1 and 44 45 not 19 limit 46 to human limit 47 to english language limit 48 to (addresses or bibliography or biography or comment or dictionary or directory or editorial or festschrift or interview or lectures or legal cases or letter or news or periodical index) 48 not 49 Case Report/ 50 not 51 limit 52 to (guideline or meta analysis or review or review literature or review of reported cases or review, academic or review, multicase or review, tutorial) 52 not 53 5 and 44 limit 55 to human limit 56 to english language 57 not 19 43 and 58 limit 59 to (addresses or bibliography or biography or comment or dictionary or directory or editorial or festschrift or interview or lectures or legal cases or letter or news or periodical index) 59 not (60 or 51) limit 61 to (guideline or meta analysis or practice guideline or review or review literature or review of reported cases or review, academic or review, multicase or review, tutorial) 61 not 62 54 or 63

SEARCH 3: AMBULATORY BLOOD PRESSURE MONITORING

1. exp Blood Pressure Monitoring, Ambulatory/ 2. exp Kidney Failure, Chronic/

APPENDIX 2: MEDLINE SEARCH STRATEGIES

3. 4. 5. 6. 7. 8. 9. 10.

exp Kidney Diseases/ exp Proteinuria/ (circadian rhythm adj6 blood pres$).tw. (chronobiolog$ adj6 blood pres$).tw. 1 or 5 or 6 2 or 3 or 4 limit 7 to human limit 9 to English language

S251

11. limit 10 to (addresses or bibliography or biography or comment or dictionary or directory or editorial or festschrift or interview or lectures or legal cases or letter or news or periodical index) 12. 10 not 11 13. 12 and 8 14. 12 not 13

APPENDIX 3: TECHNICAL REPORT ON AMBULATORY BLOOD PRESSURE MONITORING IN CKD BACKGROUND

JNC 7 summarizes the use of ambulatory blood pressure monitoring (ABPM) as follows5,5a: ● ABPM provides information about blood pressure during daily activities and sleep. ● ABPM is warranted for evaluation of “white coat” hypertension (WCH) in the absence of target organ injury. It is also helpful to assess patients with apparent drug resistance, hypotensive symptoms with antihypertensive medications, episodic hypertension, and autonomic dysfunction. ● The ambulatory blood pressure values are usually lower than clinic readings. Awake, individuals with hypertension have an average blood pressure of more than 135/85 mm Hg and during sleep, more than 120/75 mm Hg. ● The level of blood pressure measurement using ABPM correlates better with targetorgan injury than office measurements. ● ABPM also provides a measure of the percentage of blood pressure readings that are elevated, of the overall blood pressure load, and of the extent of blood pressure reduction during sleep. ● In most individuals, blood pressure decreases by 10% to 20% during the night; those in whom such reductions are not present are at increased risk for cardiovascular events. It was the opinion of the Work Group that special considerations regarding blood pressure patterns in CKD justified an evaluation of the usefulness of ABPM in CKD. This evaluation suggested to the Work Group that ABPM should be used more often as a diagnostic tool and guide to antihypertensive therapy in CKD, and that controlled trials were needed to determine more precisely the circumstances in which it should be routinely used in CKD. All members of the Work Group concluded that a more thorough review of the subject was warranted. Suggestions in this Appendix reflect the opinions of several members of the Work Group. Scope of Work Key definitions for this ABPM report can be found in Guideline 3. In addition to a considerS252

ation of reviews, an evaluation of primary articles was performed. For the primary literature review, articles were examined on individuals with essential hypertension as well as CKD, regardless of blood pressure, which (1) reported the correlation of office blood pressure with ABPM; (2) defined the prevalence of WCH; (3) described the prevalence of abnormal patterns of ABPM (nondipping or reverse dipping) in individuals with CKD; and (4) evaluated the relationship between ABPM and outcomes in CKD. Searches were also performed to identify studies examining how antihypertensive therapy affects ABPM, and its correlation with outcomes. Screening criteria for articles in ABPM are given in Appendix 2. Articles were reviewed by one member of the Work Group (RJP); data were not abstracted into evidence tables, and strength of evidence is not graded explicitly. ABPM in Essential Hypertension Recent authoritative reviews suggest advantages of ABPM in essential hypertension. ABPM has been widely used for the past decade but its use has been limited in the United States because of the expense of the equipment, the time involved in patient training and data analysis, and—until recently—the lack of consistent reimbursement. In Europe and parts of Asia, ABPM has become the standard for the measurement of blood pressure. (1) Monitoring of blood pressure in an ambulatory setting provides multiple, objective blood pressure measurements in the patient’s own environment over a full circadian period. A study comparing the diagnosis of hypertension made by casual blood pressure (CBP) with ABPM is shown in Fig 63. When there is agreement between the two methods, the diagnosis of “true” normotension (area 3) or “true” hypertension (area 2) results. If CBP is elevated but ABP normal, this is defined as WCH (area 1). These patients have a level of end organ involvement that is not different from normotensive patients. When CBP is normal but ABP is elevated, this is termed “masked hypertension” (area 4). Patients with masked hypertension have end-organ dam-

American Journal of Kidney Diseases, Vol 43, No 5, Suppl 1 (May), 2004: pp S252-S261

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

S253

Fig 63. Comparison of the diagnosis of hypertension using casual blood pressure and ambulatory blood pressure monitoring. Figure compares the diagnosis of hypertension made by CBP and ABPM in 573 patients. Area 1 represents WCH; Area 2 represents true hypertensives; Area 3 represents true normotensives; Area 4 represents masked hypertension.598a

age similar to that in persistently hypertensive patients. All of these combinations may be observed in patients with CKD, but there are no studies specifically addressing the prevalence of each combination in CKD patients. In adults, diagnostic ABP thresholds have been carefully defined in multicenter studies. The definitions take into account distributions of ABP in normal and hypertensive populations, validation of diagnostic thresholds in terms of LVH, and in terms of morbidity and mortality (Table 184).175 A meta-analysis of 18 studies demonstrated that CBP levels were greater than 24-hour, wake, or sleep ABP.142 (2) ABPM facilitates the diagnosis of WCH. ABP has been most useful in making the diagnosis of WCH.170 In a meta-analysis, the prevalence of WCH was found to be approximately 20% in the population of adults diagnosed as hypertensive by CBP. Too few studies were available for the analysis to determine the reproducibility of this classification. The question of whether WCH is an intermediate state between normotension and hypertension, or a prehypertensive state, remains unanswered. The ABPM Task Force168

determined that event-based studies in hypertensive patients have convincingly demonstrated that the risk of CVD is less in patients with properly defined WCH than those with higher ABP levels, even after controlling for concomitant risk factors. Based on prognosis, WCH can be defined as a CBP that is persistently ⱖ140/90 mm Hg with an average daytime ABP of below 135/85 mm Hg.168 WCH was associated with reduced risk of CVD compared to sustained hypertension, although the authors concluded that no prospective study adequately compared risk in patients with WCH versus normotensive patients. Under standardized conditions, the selfmeasurement of blood pressure can identify WCH, although there are insufficient data to compare the prognostic accuracy of SMBP with CBP and ABPM.142 The importance of the white coat effect is controversial. While WCH is defined by a normal ABP, the white coat effect takes into consideration the decrease in blood pressure seen with ABP compared to CBP. For example, a patient with a CBP of 180/120 mm Hg may have an ABP of 150/90 mm Hg. The patient is still hyperten-

S254

sive by ABP. Most, but not all, experts feel that the white coat effect (in this patient 30/30 mm Hg) is not of prognostic significance. (3) Level of blood pressure and abnormal patterns by ABPM show a high correlation with end-organ damage. ABP has been shown to have an excellent correlation to hypertensive endorgan damage, such as cardiovascular mortality and events, abnormal left ventricular mass and function, cerebrovascular events, glomerular and tubular proteinuria, and is capable of identifying abnormal circadian blood pressure rhythms such as nocturnal hypertension. ABP is also important in demonstrating the rapid rise in early morning blood pressure that is associated with a high risk for CV events. Many of these studies are summarized in Table 185.599 An example of some of the patterns commonly seen with ABPM are seen in Fig 64. There is a growing body of evidence showing that nondipping is associated with a worse prognosis, irrespective of whether night-time dipping is studied as a continuous or a class variable.174

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

The meta-analysis confirmed that a nondipping or reverse dipping pattern was associated with an increased risk of adverse events. A comparison of prediction of risk by ABP versus CBP in the meta-analysis was not deemed possible due to the uncertain quality of the CBPs, not ABPs. While repeated ABPM has excellent reproducibility for group data, this may not be the case for the individual patient.149 The international conference on ABPM in 2001 agreed that, while the incidence of cardiovascular events has some correlation with CBP, ABPM significantly refined the prediction. In the Syst-Eur Trial, 24-hour SBP measurements were found to be a significant predictor of cardiovascular risk.600 Some, but not all, studies have reported an independent and positive relationship between measures of variability (standard deviation of blood pressure divided by the mean blood pressure level) of daytime and night-time blood pressure and cardiovascular outcome. Finally, it has been suggested that ABP patterns can be predictive of pre-eclampsia early in pregnancy

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

S255

Fig 64. Standardized common patterns of ABPM. Standardized common patterns of ABPM (ECF Medicatl Ltd., Blackrock, Co. Dublin, Ireland, www.ecfmedical.com). Common to all plots: vertical axes show blood pressures; horizontal axes shot 24-hour clock times; horizontal bands indicate normal values for 24-hour SBP and DBP; shaded vertical areas indicate night-time. (A) Normal ABPM pattern. This ABPM suggests normal 24-hour SBP and DBP (128/78 mm Hg daytime, 110/62 night-time), (B) White-coat hypertension. This ABPM suggest WCH (175/95 mm Hg) with otherwise normal 24-hour SBP and DBP (133/71 mm Hg daytime, 119/59 mm Hg night-time). (C) White-coat effect. This ABPM suggests mild daytime systolic hypertension (149 mm Hg), borderline daytime diastolic hypertension (87 mm Hg), borderline night-time systolic hypertension (121 mm Hg), and normal night-time DBPs (67 mm Hg) with white-coat effect (187/104 mm Hg). (D) Systolic and diastolic hypertension. This ABPM suggests mild daytime systolic and diastolic hypertension (147/93 mm Hg), but normal night-time SBP and DBP (111/66 mm Hg). (E) Isolated systolic hypertension. This ABPM suggests severe 24-hour isolated systolic hypertension (176/68 mm Hg daytime, 169/70 mm Hg night-time). (F) Hypertensive dipper. This ABPM suggests severe daytime systolic hypertension (181 mm Hg), moderate daytime diastolic hypertension (117 mm Hg) and normal night-time SBP and DBP (111/68 mm Hg). (G) Hypertensive nondipper. This ABPM suggest severe 24-hour systolic and diastolic hypertensive (210/134 mm Hg daytime, 205/130 mm Hg night-time).

and that intervention with aspirin may prove beneficial in prevention.601 There are insufficient studies to assess the correlation of SMBP to end-organ damage.142 (4) There is insufficient evidence to conclude that adjusting medications using ABP is superior to using CBP measurements. The meta-analysis concluded that ABP has not been studied adequately to assess its role in antihypertensive therapy. Nonetheless, ABP’s predictive power

for end-organ damage must be considered in drug trials. Further, ABP can give information unavailable through CBP, such as the trough to peak ratio over a period of time, rather than a single measurement. The missed-dose analysis can be helpful in demonstrating the duration of action of medication. Further, the ABPM may eliminate the need for placebos in drug trials and require fewer patients enrolled in order to prove efficacy.602

S256

Fig 65. Blunting of the diurnal blood pressure rhythm of predialysis patients is more severe at lower GFR (higher serum creatinine concentration). To convert to mg/dL, divide by 88.4. Reproduced with permission.604

ABPM in CKD CKD is associated with alterations in the circadian patterns of blood pressure; the most common of these rhythm abnormalities is sleep-associated hypertension and a nondipping blood pressure pattern. Patients with CKD have a high prevalence of a nondipping pattern (67%).176-178 These patterns are reproducible, with 67% of patients with CKD maintaining their baseline circadian rhythm and 82% having a predictable pattern on repeated monitoring. These patterns were even more reproducible in transplant and dialysis patients.174 The mechanisms for this pattern have not been clearly elucidated.603 It was also demonstrated that the prevalence of nondipping patterns is higher at lower GFR. Fifty-three percent of CKD patients with serum creatinine of ⬍4.5 mg/dL (⬍400 ␮mol/L) had nondipping pattern progressing to 75% with creatinine of ⬎6.8 mg/dL (⬎600 ␮mol/L) (Fig 65).604 The nondipping pattern observed in patients with type 1 and 2 diabetes605 has also been related to the presence of microalbuminuria.606 CKD patients, particularly kidney transplant patients, may have a reversed blood pressure pattern of higher sleep than wake BPs. Abnormal ambulatory patterns have been shown to be related to varying degrees of abnormal protein excretion. Abnormal or nondipping pattern by ABPM is strongly correlated with microalbuminuria, macroalbuminuria, and proteinuria in a manner superior to CBP. This has been shown in studies of type 1 diabetes,179,607 type 2 diabetes,605,608 type 1 and type 2 diabetes combined,609-611 and essential hypertension.611-613 Some of the cross-sectional studies were per-

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

formed in individuals labeled as having “essential hypertension” or “diabetes,” but not specifically CKD. However, many individuals with microalbuminuria or macroalbuminuria would now be considered to have CKD based on the presence of this marker of kidney damage. Thus, these studies are pertinent to the target population for these guidelines. In the few studies examining possible mechanisms, autonomic dysfunction has been implicated.607,611 Furthermore, it has been shown that ABPM can identify individuals at risk for development of CKD. In type 1 diabetes, nocturnal blood pressure elevations determined by ABPM predated the development of microalbuminuria (Fig 66).179 Identifying individuals at risk for development of microalbuminuria provides an important window of opportunity for primary prevention. Abnormal ambulatory patterns have been shown to be related to CV damage and events. There are few studies in patients with CKD examining the relationship of ABP to morbid or mortal events. Three hundred and twenty-five patients with NIDDM were studied with ABPM with the circadian rhythm assessed by cosinor analysis.183 After an 8-year follow-up, 201 patients had a normal circadian rhythms and 87 had a reversed one. Twenty subjects in the group with normal rhythm (9.9%) had nonfatal (cerebrovascular, cardiovascular, peripheral vascular or retinal events) or fatal events versus 56 patients (64.4%) in the reversed group. Circadian pattern of blood pressure and age had the greatest effect on adverse events in Cox proportional hazards model (Fig 67). The meta-analysis of 25 studies examining the relationship of ABP levels with target-organ damage was positively associated with ABP. In 10 studies, at least one dimension of ABP predicted subsequent clinical events prospectively. The correlation of end-organ damage to cardiovascular end points is similar in patients with CKD to that reported in essential hypertensive patients. The same correlation between ABP and measures of left ventricular mass is seen, regardless of the cause of CKD: ADPKD,614 CKD,184,615 glomerulonephritis,616 hypertension with albuminuria,186,612 IgA nephropathy,617 NIDDM,183 and in kidney transplant recipients178,618 (Fig 68). The recurring theme in these articles is the finding of a positive correlation of left ventricular

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

S257

Fig 66. Kaplan-Meier curves showing the probability of developing CKD (microalbuminuria) according to the pattern of daytime and nighttime systolic pressure in diabetes. The probability of microalbuminuria differed significantly between the two groups (P ⴝ 0.01 by the log-rank test; chisquare ⴝ 6.217 with 1 df). The risk of microalbuminuria was 70% lower in the subjects with a normal nocturnal pattern than in those with an abnormal nocturnal pattern.179

mass with ABP parameters; most commonly, a nondipping pattern or sleep-related hypertension. In each of the studies there were either weak or, more commonly, no correlations between CBP and these outcomes. The combination of the 24-hour systolic ABP with a measure of dipping, the night to day ratio, was a strong predictor of cardiovascular end points in the Systolic Hypertension in Europe Trial (Fig 69). Abnormal ambulatory patterns have been related to more rapid progression of kidney disease. There are five studies that examine the association of ABP with kidney disease progression. Out of 126 patients with IgA nephropathy,

Fig 67. Survival curves of diabetic subjects with normal (N) and reversed (R) circadian blood pressure rhythms. The unadjusted relative risk for diabetic subjects with a reversed circadian blood pressure rhythm was 20.6-fold higher than that of subjects with a normal rhythm (P < 0.001; Cox-Mantel’s test).183

those who had ambulatory hypertension developed higher plasma creatinine values whereas the normotensive patients, with or without antihypertensive therapy, had stable serum creatinine. Nondippers made up 93% of the hypertensive patients. The normotensive patients who were nondippers also had higher creatinine values than dippers.180 In a 3-year study, 48 patients with CKD were studied: 28 dippers and 20 nondippers.181 Nondippers had a faster decline in GFR (4.4 mL/min/y) versus dippers (3.2 mL/ min/y) and a higher mean protein excretion. In a similar study of diabetic patients, nondippers had a 7.9 mL/min/y decline compared to 2.9 for dippers182 (Fig 70). Finally, Nakano et al619 examined 257 patients with NIDDM. Patients had a normal circadian rhythm (193) or a reversed pattern (63). More of the reversed patients (23.5%) progressed to hemodialysis compared to those with normal rhythm (3%).619 Jacobi et al620 demonstrated the direct relationship between ABP and serum creatinine 6 months posttransplant (Fig 71). Due to these abnormal patterns, the correlation between CBP and ABP is poor, as is the relationship of CBP to end-organ damage. Three studies of diabetic patients showed a significant association of ABP to urinary albumin excretion where CBP failed to do so. Once again, SBP and nocturnal elevations had the closest correlation to this measure of end organ damage. In nondiabetic CKD patients, five studies have

S258

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

Fig 68. Relationship between end-systolic left ventricular diameter and percent fall in blood pressure (sleep to awake) in hemodialysis patients and kidney transplant recipients. The less the fall in blood pressure during sleep, the more dilated the LV chamber. Reproduced with permission.618

demonstrated the same finding. McGregor et al178 demonstrated a nondipping pattern in kidney transplant recipients which was associated with LVH, whereas CBP did not. Tucker et al184 showed that casual SBP and DBP had a weak correlation (0.25 to 0.22) to LVH in CKD patients, while systolic ABP had a correlation of

Fig 69. Night to day blood pressure ratio and 24hour SBP at entry as predictors of the 2-year incidence of cardiovascular end-points in 393 patients randomized to the placebo group of the Systolic Hypertension in Europe Trial. Using Cox regression analysis, the event rate was standardized to female sex, mean age (69.6 years), a lack of previous cardiovascular complications, nonsmoking status, and residence in Western Europe. Incidence is given as the probability of an event per 100 patients. Reproduced with permission.600

0.52 and diastolic ABP 0.42. Nocturnal BPs had an even better correlation. Patients with CKD can have WCH; the diagnosis of WCH could affect the requirement for antihypertensive therapy. The prevalence of WCH has not been determined reliably in CKD, especially in relation to CKD stages. In essential hypertension, WCH is more likely to occur in patients with lesser stages of hypertension severity. Thus, the prevalence of WCH might be lower in patients with CKD. Nonetheless, the diagnosis may have importance in determining the need for antihypertensive therapy. There are three studies specifically examining the issue of WCH in various types of CKD, which mirror results observed in patients with essential hypertension. The prevalence of WCH among patients is definition-dependent. The more severe the hypertension, the lower the prevalence of WCH608 (Fig 72). The prevalence of end-organ damage is low in CKD patients with WCH.170 In the HARVEST study of 942 patients with Stage 1 hypertension, patients with WCH had significantly lower left ventricular mass and microalbuminuria compared to established hypertensive patients. There was no difference in end-organ damage between WCH and normotensive controls.185 The diagnosis of nocturnal hypertension in CKD could affect the requirement for antihypertensive therapy. Concerns in CKD patients who have altered patterns of blood pressure and a nondipping blood pressure pattern include (1) CBP may not correctly represent blood pressure

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

S259

Fig 70. Diurnal blood pressure rhythm in diabetic kidney disease. Comparison of two groups of patients with diabetic kidney disease matched for daytime blood pressure, cholesterol, proteinuria and diabetic control— one cohort (left) with normal diurnal blood pressure, and an average creatinine clearance decline of 2.9 mL/min compared to the other cohort (right) with abnormal diurnal blood pressure, and a rate of creatinine clearance decline of 7.9 mL/ min. Reproduced with permission.182

burden and risk; and (2) confining the definition of WCH to the daytime blood pressure comparisons by ABPM, a common practice, or SMPB may include patients with sleep-related hypertension in the WCH category. This issue will require further study. There is insufficient evidence to determine whether ABPM is a better monitoring tool than CBP for assessing the effectiveness of antihypertensive therapy to reduce kidney disease progression or CVD risk in CKD. Several studies in CKD have shown advantages of ABPM

on surrogate outcomes. In treating hypertensive CKD patients, LVH was found to be present only when the patients were noted to have persistently abnormal ABP, but not CBP. When treatment decisions were made by CBP, 13% of patients had LVH compared to none of the patients controlled by ABP criteria.186 A comparison of lisinopril with nisoldipine showed a decrease in microalbuminuria with no change between albuminuria groups by CBP; however, there were differences between groups for night-time SBP by ABP.623 In type 1 and type 2 diabetic patients,

Fig 71. Correlation of creatinine values in patients 6 months after kidney transplantation with 24-hour SBP (left) and DBP (right). Reproduced with permission.620

S260

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

Fig 72. Rates of major cardiovascular events in a normotensive group (A), two groups with WCH defined using restrictive (B) or liberal (C) criteria, and a group with ambulatory hypertension (D). Reproduced with permission.621,622

both pretreatment and posttreatment changes in microalbuminuria were associated with nocturnal blood pressure changes, but not CBP.609 In the Heart Outcomes Prevention Evaluation (HOPE) trial, the ACE inhibitor ramapril was clearly shown to reduce cardiovascular events in patients with vascular disease.105 Blood pressure measurement protocols were less than adequately defined. Further, the small changes noted in CBP (3/2 mm Hg) in light of the major improvement in events were interpreted to mean that this beneficial effect was due to the character of the

Fig 73. Comparison of cuff blood pressure, 24-hour ABPM, and night-time ABP in the HOPE Study and a HOPE Substudy.

drug itself, rather than the blood pressure lowering effects. However, a substudy using ABPM in the HOPE trial showed a clear and significant reduction in blood pressure, particularly at night, raising doubts about the prior conclusion (Fig 73).624,625 Most recently, the prognostic value of ABP was documented in the Office versus Ambulatory Blood Pressure Study, where 1,963 hypertensive patients were followed for a mean of 5 years.172 The investigators demonstrated that ABP was an independent and significant risk factor for CV events, even after adjustment for classic risk

APPENDIX 3: TECHNICAL REPORT ON ABPM IN CKD

factors, including office blood pressure. They also noted that CV outcomes are better predicted by ABP than CBP. Most notable was the finding that patients whose mean 24-hour ABP was ⱖ135 mm Hg when receiving treatment were nearly twice as likely to have CV events as patients with ⬍135 mm Hg, regardless of their CBP measurement. Clinicians may find benefit in recommending ABPM for patients with CKD. Although the Work Group did not recommend routine monitoring of ABP, it recognizes that health-care providers may find benefit in recommending this technique for individuals with CKD for the indications listed in JNC 7 and those listed in Table

S261

75. Indeed, its prognostic value has also been documented in patients with treated hypertension.172 ABPM results have a close correlation with end-organ damage, including progression of kidney disease and clinical CVD outcomes. It may be of particular benefit in patients with CKD who have a known alteration in the circadian blood pressure pattern that cannot be revealed using CBP measurements. The presence of a nondipping pattern or nocturnal hypertension in a patient with CKD places an already high-risk patient into an even higher-risk group. This knowledge may be helpful in suggesting an even more aggressive therapeutic regimen.