Clinical Kidney Journal, 2020, vol. 13, no. 4, 513–521 doi: 10.1093/ckj/sfz187 Advance Access Publication Date: 25 January 2020 CKJ Review CKJ R EV IEW Evidence in chronic kidney disease–mineral and bone disorder guidelines: is it time to treat or time to wait? Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 Jordi Bover1, Pablo Uren ~ a-Torres2, Silvia Mateu1, Iara DaSilva1, Silvia Gràcia1, Maya Sánchez-Baya1, Carolt Arana1, Leonor Fayos1, Lluis Guirado1 and Mario Cozzolino 3 1 Fundació Puigvert, Department of Nephrology, IIB Sant Pau, RedinRen, Barcelona, Catalonia, Spain, 2 Department of Dialysis, AURA Nord Saint Ouen, Saint Ouen and Department of Renal Physiology, Necker Hospital, University of Paris Descartes, Paris, France and 3Renal Unit, ASST Santi Paolo e Carlo, Department of Health Sciences, University of Milan, Milan, Italy Correspondence to: Jordi Bover; E-mail: jbover@fundacio-puigvert.es ABSTRACT Chronic kidney disease–mineral and bone disorder (CKD–MBD) is one of the many important complications associated with CKD and may at least partially explain the extremely high morbidity and mortality among CKD patients. The 2009 Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guideline document was based on the best information available at that time and was designed not only to provide information but also to assist in decision-making. In addition to the international KDIGO Work Group, which included worldwide experts, an independent Evidence Review Team was assembled to ensure rigorous review and grading of the existing evidence. Based on the evidence from new clinical trials, an updated Clinical Practice Guideline was published in 2017. In this review, we focus on the conceptual and practical evolution of clinical guidelines (from eMinence-based medicine to eVidence-based medicine and ‘living’ guidelines), highlight some of the current important CKD–MBD-related changes, and underline the poor or extremely poor level of evidence present in those guidelines (as well as in other areas of nephrology). Finally, we emphasize the importance of individualization of treatments and shared decision-making (based on important ethical considerations and the ‘best available evidence’), which may prove useful in the face of the uncertainty over the decision whether ‘to treat’ or ‘to wait’. Keywords: CKD, CKD–MBD, EBM, evidence-based medicine, KDIGO INTRODUCTION entitled ‘KDIGO Clinical Practice Guideline for the Diagnosis, A new definition and classification of chronic kidney disease– Evaluation, Prevention and Treatment of CKD–MBD’ [2]. These mineral and bone disorder (CKD–MBD) was proposed in 2005 [1] steps constituted recognition that CKD–MBD is one of the many and later adopted in 2009 in a guideline publication from the important complications associated with CKD. As is already Kidney Disease: Improving Global Outcomes (KDIGO) initiative well known, CKD–MBD represents a ‘systemic’ condition Received: 24.9.2019; Editorial decision: 2.12.2019 C The Author(s) 2020. Published by Oxford University Press on behalf of ERA-EDTA. V This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/ licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com 513 514 | J. Bover et al. Uncontrolled Rigorous invesgaon experience of health intervenons CLAIRVOYANCE M E INENCE- V E IDENCE- based medicine based medicine TRUST IN THE NUMBERS TRUST IN THE EXPERTS WAIT? THERAPEUTIC NIHILISM? TREAT? OVERTREATMENT? Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 PRIMUM NON-NOCERE FIGURE 1: Schematic representation of eMinence-based medicine vs eVidence-based medicine. manifested by either one or a combination of laboratory abnor- Controversies Conference determined that there was sufficient malities [calcium, phosphate, parathyroid hormone (PTH) or new evidence to support updating some of the previous CKD– vitamin D metabolism], bone abnormalities (turnover, minerali- MBD recommendations [10, 11]. This systematic update process zation, volume, linear growth or strength) and vascular and/or finally resulted in 21 updated recommendations/suggestions, other soft tissue calcifications [1–3]. By linking kidney, bone dis- and several societies have already commented on this update, ease (evolution of the ‘old’ concept of ‘renal osteodystrophy’, drawing attention to remaining critical issues and points now restricted to histomorphometric analysis) [1] and cardio- relevant to correct interpretation [11–14] (J. V. Torregrosa et al., vascular disease, the CKD–MBD complex represents an attempt submitted for publication in the Spanish Society of Nephrology to explain, at least in part, the extremely high morbidity and journal “Nefrologia”). In this review, we will examine the con- mortality of CKD patients [4, 5]. ceptual/practical evolution of clinical guidelines, highlight some The 2009 KDIGO Clinical Practice Guideline document was of the important CKD–MBD-related changes and underline based on the best information available at that time and was some ethical considerations that may prove of importance in designed not only to provide information but also to assist in the face of uncertainty. decision-making [2]. It was not intended to define a standard of care, and it was clearly stated that ‘it should not be construed as EVIDENCE-BASED MEDICINE VERSUS one, nor should it be interpreted as prescribing an exclusive course of management’ [2]. The KDIGO guidelines were created as EMINENCE-BASED MEDICINE a ‘global’ initiative, and it was recognized that variations in prac- Since the time of Hippocrates, as Djulbegovic and Guyatt tice would inevitably and appropriately occur as clinicians take recently remarked in The Lancet [15], medicine has struggled to into account the needs of individual patients, available resources balance the uncontrolled experience of ‘healers’ with observa- and limitations unique to an institution or type of practice [2]. tions obtained by rigorous investigation of claims regarding the Not only did the international KDIGO Work Group include effects of health interventions. Dr Guyatt first coined the term worldwide experts but in addition, an independent Evidence ‘eVidence-based medicine’ (EBM) in 1991 [16], contending Review Team was assembled to ensure a rigorous review and that although there is a role for all empirical observations, ran- appraisal of the existing evidence. Briefly, the process included domized controlled clinical trials (RCTs), systematic reviews refining questions, developing the literature search strategy, and meta-analyses provide more trustworthy evidence than do extracting data and critically appraising the literature, summa- unsystematic, uncontrolled observations, case reports, biologi- rizing the evidence, revising the recommendation statements, cal experiments and the experiences of individual clinicians or and grading evidence quality and the strength of recommenda- experts (‘eMinence-based medicine’ or, even worst, ‘eLoquence- tions using the Grading of Recommendations Assessment, based medicine’, ‘eLegance-based medicine’ or ‘vehemence- Development and Evaluation (GRADE) approach [2, 6]. based medicine’ [17, 18]) (Figure 1). The practice of medicine Therefore, each recommendation was accompanied by the should be based on systematic reviews that summarize the strength of the recommendation and an evidence grade. ‘best “available” evidence’, even though uncontrolled clinical Guideline statements that provided general advice or guidance experience and incomplete or fragmented physiological reason- (and thus were not based on systematic review) were desig- ing have maintained a dominant position as drivers of usual nated as ‘not graded’. The guideline development process con- clinical practice. As a matter of fact, past inadequate research cluded with an external public review to ensure widespread led to fatal bone marrow transplantation for women with breast input from patients, experts, industry and national organiza- cancer, prophylactic antiarrhythmic therapy in patients with tions. After the initial publication [2], several national societies myocardial infarction and the prescription of hormone replace- and/or organizations followed-up with commentaries, interpre- ment therapy in millions of healthy women on the basis of a tations, updates and local adaptations [7–9]. Based on the evi- hypothetical cardiovascular risk reduction [15, 19–21]. dence from new clinical trials, an updated Clinical Practice Moreover, evidence should be evaluated in totality rather than Guideline was published in 2017 after the Madrid 2013 KDIGO focusing on a selection of evidence that favours a particular Evidence in CKD–MBD guidelines | 515 claim. ‘Anchoring’ is a fact, i.e. once we adopt a belief regarding results from different systematic reviews and meta-analyses a concept, when confronted with new data we tend, on average, may also differ substantially [23–28]. not to move as far from our accustomed position as we should EBM integrates the best published scientific evidence (what [22]. If the new data are positive, then the ‘believers’ accept the the literature says) with clinical expertise (clinical judgment or data, whereas ‘non-believers’ find some reason to discredit what the clinician knows) and patient values, preferences or them (e.g. because data derive from different races, countries or expectations (what the patient wants) (Figure 2). An extension continents or because of variation and/or validity of different of EBM methodology that is more relevant today is evidence- endpoints) [22]. Similarly, we tend to remember and refer to based clinical practice, which takes into account the healthcare studies that confirm our views and ignore those that do not setting and circumstances in which we practice. On the other (and this is a problem when heterogeneity is common) [22]. In hand, the GRADE system, first published in 2004 [31], provides a any case, we have to remember that well-conducted systematic much more refined hierarchy of evidence, and its use in the reviews can summarize not only RCTs but also cohort studies, KDIGO Clinical Practice Guidelines for CKD–MBD represented a case–control studies and even case reports [15]. Nevertheless, very important step forward as compared with previous mineral metabolism guidelines [32, 33]. GRADE also addresses the Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 process of moving from evidence to recommendations (Grade 1 ¼ strong evidence ¼ right for all or almost all patients; Grade 2 ¼ weak evidence ¼ right for most but not all patients, requiring Individual Clinical presentation of evidence that facilitates shared decision-mak- Expertise (Judgment) ing) (Table 1). Note that the need to consider patients’ values What the clinician knows and preferences is acknowledged [15]. Consequently, although not all evidence-based practices are alike [34, 35], GRADE has been adopted by >100 organizations, including the Cochrane Collaboration, National Institutes of Health, World Health Organization (WHO) and, obviously, KDIGO. Evidence-based Practice/Medicine External Patient Values, KDIGO CKD–MBD GUIDELINES Best (Published) Preferences & KDIGO CKD–MBD guidelines (from both 2009 and 2017) repre- Research (Scientific) Expectations Evidence sent the most important academic work on the subject to date What the patients wants What the literature says (and a major effort in terms of precise use of grammar!), but given their worldwide potential implications for health authorities, and also their political, economic and even legal FIGURE 2: Schematic representation of evidence-based practice/medicine consequences, it was rather disappointing to see a lack of strong (adapted from references [29, 30]). clinical evidence in almost all areas. This definitely highlights Table 1. Nomenclature and description for rating guideline recommendations (GRADE) [11, 30] Implications GRADEa Patients Clinicians Policy Level 1 Most people in your situation Most patients should receive the The recommendation can be eval- ‘We recommend’ would want the recommended recommended course of action uated as a candidate for devel- course of action, and only a oping a policy or a performance small proportion would not measure Level 2 The majority of people in your sit- Different choices will be appropri- ‘We suggest’ uation would want the recom- ate for different patients. Each The recommendation is likely to mended course of action, but patient needs helps to arrive at a require substantial debate and many would not management decision consis- involvement of stakeholders be- tent with her or his values and fore policy can be preferences determined GRADE Quality of evidence Meaning A High We are confident that the true effect lies close to the estimate of the effect B Moderate The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different C Low The true effect may be substantially different from the estimate of the effect D Very low The estimate of the effect is very uncertain, and often will be far from the truth a The additional category ‘not graded’ is typically used to provide guidance based on common sense or when the topic does not allow adequate application of evidence. The ungraded recommendations are generally written as simple declarative statements, but are not meant to be interpreted as being stronger recommendations than level 1 or 2 recommendations. Within each recommendation the strength of recommendation is indicated as Levels 1,2, or ‘not graded’, and the quality of the supporting evidence is shown as A,B,C or D. 516 | J. Bover et al. A B therapy (written in alphabetical order); (v) parathyroidectomy RCT for those who fail to respond to medical treatment; and (vi) re- Systematic Cohort study Reviews / Meta-Analysis/ striction of the dose of calcium-based phosphate binders in NW Meta-Analysis adult patients receiving phosphate-lowering treatment (impor- Case control Randomized Decreasing bias study Decreasing bias Controlled Trials tantly, the evidence grade for this statement has increased from Increasing quality of Case series Increasing quality of evidence 2C to 2B since the 2009 guidelines). It should be underlined that evidence Non-randomized Controlled Trials the 2009 KDIGO guidelines recommended that the dose of Case reports Observational Studies with calcium-based phosphate binders be restricted in the presence Comparison groups Animal research of persistent or recurrent hypercalcaemia (1B) and/or adynamic Case Series & Case Reports bone disease (2C) and/or persistently low serum PTH levels (2C), In-vitro research as well as in the presence of arterial calcification (2C). Actually, Expert experience Expert Opinion the last remark represented a step forward since the earlier 2003 National Kidney Foundation Kidney Disease Outcomes FIGURE 3: Traditional (A) and new (B) evidence-based medicine hierarchies of QUality Initiative (NKF KDOQI) guidelines stated that non-cal- cium-containing phosphate binders were preferred in dialysis Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 evidence (adapted from references [15, 39–41]). Note that systematic reviews are ‘chopped-off’ from the pyramid since a meta-analysis of well-conducted RCTs patients with ‘SEVERE’ vascular and/or other soft tissue calcifi- at low risk of bias cannot be equated with a meta-analysis of observational stud- cations [33]. The evidence grade for use of a dialysate calcium ies at higher risk of bias [41]. Some recent reports suggest that lines separating concentration between 1.25 and 1.50 mmol/L (2.5 and 3.0 mEq/L) the study designs should be wavy instead of straight [41]. NW, Network. (Guideline 4.1.4) was also upgraded (to 2C from 2D) [2, 11]. The new guidelines included a very important change of par- adigm in respect of BMD testing. In 2009, it had been suggested the need for rigorous RCTs in this field (as well as others) and that, in patients with CKD Stages 3–5D, BMD testing should for most of us offers a lesson in the need for humility. Humans ‘NOT’ be performed routinely because BMD does ‘NOT’ predict are ‘informavores’ [36], and in an era in which we have moved fracture risk as it does in the general population, and BMD does from ignorance to infoxication [37, 38], an increasingly sophisti- NOT predict the type of renal osteodystrophy (2B). However, cated hierarchy of evidence and systematic summaries of the new evidence from several RCTs now suggests that in such best evidence to guide care are of the utmost importance patients, BMD testing SHOULD BE performed to assess fracture (Figure 3). However, in the case of CKD–MBD (and most areas of risk ‘if’ results will impact clinical decisions (Guideline 3.2.1; evi- nephrology), the level of evidence is poor or extremely poor; dence again graded as 2B despite the very significant change). nevertheless, we are required to act [42–44]. This guideline is related to Guideline 4.3.2, which suggests treat- Not many things really changed in 2017 [11–13, 45]. The new ment as for the general population in patients with CKD Stages guidelines are mostly graded as suggestions (Level 2) or are ‘not 3A and 3B with PTH in the ‘normal’ range and osteoporosis and/ graded’ at all. Moreover, the quality of supporting evidence is or high risk of fracture as identified by WHO criteria (2B). mainly low (Grade C). Thus, among all current [‘old’ 2009 (no up- Consequently, there is currently an important debate over the date needed) and ‘new’ (updated 2017)] recommendations for extent to which these changes represent an important diagnos- ‘adults’, only one (old) is graded as Level 1 A evidence (Guideline tic and therapeutic challenge for nephrologists (in the absence 4.3.1), namely the ‘recommendation’ that patients with CKD of clear-cut treatment evidence) [46, 47], additionally consider- Stages 1 and 2 with osteoporosis and/or high risk of fracture, as ing that the inability to perform a bone biopsy may not justify identified by WHO criteria, should be managed as for the withholding antiresorptive therapy from patients at high risk of general population. Two guidelines (old) are graded as 1B: one fracture [11, 48]. This new guideline also begs many questions recommends that clinical laboratories should inform clinicians with respect to implementation. For example, what exactly are of the actual assay method in use and report any changes in ‘risk factors for osteoporosis’ in CKD patients when CKD itself, methods, sample source and handling specifications (Guideline among others, is clearly such a risk? What is the accuracy of the 3.1.6) and the other recommends measuring serum calcium and diagnosis of the underlying bone phenotype? What is the ‘nor- phosphate at least weekly, until stable, during the immediate mal’ range for serum PTH in CKD patients? [12]. post-kidney transplant period (Guideline 5.1). Three more revis- It is also worth mentioning that statements in the chapters ited guidelines (old) are graded as 1C and recommend: (i) moni- devoted to the treatment of CKD–MBD, targeted at lowering toring serum levels of calcium, phosphate, PTH and alkaline high serum phosphate levels and maintaining serum calcium, phosphatase activity, beginning in CKD Stage 3A; (ii) basing and the treatment of abnormal PTH levels, are mostly graded as therapeutic decisions on ‘trends’ rather than a single laboratory 2C or not graded, with the ‘obvious’ clear exceptions mentioned value; and (iii) avoiding the long-term use of aluminium-con- above. Optimal PTH goals in the non-dialysis CKD setting are taining phosphate binders and dialysate aluminium contami- not known; therefore, beyond the avoidance of hypercalcaemia nation (Guidelines 3.1.1, 3.1.4 and 4.1.7, respectively). An (Guideline 4.1.3; evidence upgraded to 2C from 2D), the lowering additional guideline (old) is graded 2A (3.3.2) and ‘suggests’ that of elevated phosphate levels ‘towards’ the normal range now patients with known vascular or valvular calcification should be in all CKD stages (2C) and the avoidance of ‘preventive’ considered at the highest vascular risk. Six others [two old and phosphate-lowering treatment, the guidelines only suggest (at last!) four new] are graded as 2B: Guidelines 3.2.1, 3.2.3, 4.1.6, that, in dialysis patients, intact serum PTH levels should be 4.2.4, 4.2.5 and 4.3.2. These relate to (i and ii) bone mineral den- maintained in the range of approximately two to nine times the sity (BMD) testing to assess fracture risk and suggested circum- upper normal limit for the assay (2C), and that marked and con- stances to treat CKD patients with CKD Stages 3A and 3B; (iii) sistent changes in PTH in either direction even within this range measurements of PTH and bone-specific alkaline phosphatase should prompt initiation of or change in therapy to avoid pro- to evaluate bone disease (predictors of bone turnover); (iv) the gression to values outside of this range (2C) (the so-called use of calcimimetics, calcitriol or vitamin D analogues (or com- extremes of risk). While the implementation of the guidelines is binations thereof) in dialysis patients requiring PTH-lowering not an easy task and misinterpretations or shortcuts are Evidence in CKD–MBD guidelines | 517 possible [leading to over- or under-diagnosis and over- or un- long-term prospective RCTs [54]. As nicely stated by Drs Chen der-treatments (‘therapeutic nihilism’)], it has been docu- and Bushinsky, it would be desirable for us, ‘as a nephrology mented that mean serum PTH levels have increased in most community’, to ‘focus our investigative efforts on testing hy- countries since the 2009 KDIGO publication [49]. potheses in RCTs with clinically important endpoints such as A further suggestion in the guidelines is that calcitriol and CKD progression, cardiovascular events, fractures or mortality, vitamin D analogues should NOT BE routinely used in adult which will enable stronger recommendations to be made in the patients with CKD Stages 3A–5 who are not on dialysis (2C). This next revision of the KDIGO CKD–MBD guidelines’ [45]. suggestion is based on the results of the Paricalcitol Capsule Benefits in Renal Failure-Induced Cardiac Morbidity (PRIMO) and Effect of paricalcitol on left ventricular mass and function ETHICAL CONSIDERATIONS in CKD (OPERA) studies and some meta-analyses that included Very significant gaps remain in our knowledge and there is a studies performed at a time when a certain degree of hypercal- danger that the consequences of a ‘misunderstood’ EBM will be caemia was considered useful, in accordance with the previ- ‘therapeutic nihilism’, especially under financial pressure. ously available drug armamentarium. The PRIMO and OPERA Absence of evidence is not evidence of absence (argumentum studies were negative regarding the primary endpoint (reduc- ad ignorantiam) [62]. While clinical experience and expert opin- Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 tion of left ventricular hypertrophy but NOT the control of sec- ion can, of course, be criticized and must be analysed, we ondary hyperparathyroidism). Thus, these patients were greatly believe that their significance cannot be completely dis- treated with relatively very high doses of paricalcitol, especially missed [63]. For instance, in the absence of clear evidence in taking into account the fact that most patients had only minor some areas (such as antithrombotic therapy in end-stage renal or mild elevations of PTH levels [50, 51]. However, somewhat disease), expert opinion is still very welcome so that nephrolo- surprisingly, in the new guidelines, it is considered reasonable gists are not left alone when facing difficult and sometimes to reserve the use of calcitriol and vitamin D analogues for risky clinical and therapeutic decisions for which supporting ev- patients with CKD Stages 4 and 5 who have ‘severe’ and pro- idence is lacking [44]. All these ethical/philosophical/practical gressive hyperparathyroidism. Although this statement is ‘not dilemmas actually extend to most areas within general ne- graded’, vitamin D is a classic ‘preventive’ and therapeutic ma- phrology (Table 2), causing nephrologists to struggle to choose noeuvre that we have been employing (even enforcing!) for between passivity (adopting a ‘wait and see’ approach and de- many years [32, 33]. This new guideline calls attention to the ciding to act only when Level 1 evidence becomes available) and suggestion that a ‘passive’ attitude (with non-native forms of vi- an exceedingly proactive attitude based on long-lasting ‘beliefs’ tamin D) should be adopted until ‘something’ becomes suffi- that sometimes are not confirmed [66, 67]. For many reasons, ciently ‘severe’ to warrant the initiation of treatment. Moreover, including the frequent loss of patients (e.g. due to renal trans- the perception of ‘severity’ remains at the discretion of the prac- plantation) and the complex nature of uraemia and dialysis, titioner despite active surveillance. Of note, it is still unknown RCTs in nephrology are and probably will remain extremely what the optimal serum PTH level is in these patients. A differ- scarce [60, 61]. A variety of additional problems have been de- ent reasonable approach is suggested by some (J. V. Torregrosa scribed with respect to the guidelines and/or EBM [15], including et al., submitted for publication in the Spanish Society of delays in updating new relevant evidence when it becomes Nephrology journal “Nefrologia”) to avoid an induced and prob- available (Table 3). ‘Living’ systematic reviews and ‘living’ ably inadequate normalization of serum PTH levels in these guideline recommendations [68, 69] may help solving this last patients without the need for passive ‘waiting’ until something problem. However, unawareness or infoxication, difficulty in the becomes sufficiently ‘severe’ to consider treatment. Actually, achievement of goals, limited adherence, disbelief in current some degree of secondary hyperparathyroidism may be benefi- recommendations, sometimes induced by continuously chang- cial in CKD patients (PTH is a phosphaturic hormone and is nec- ing recommendations or significant differences among different essary for a normal bone formation rate) due to the presence of societies that are more or less conservative, misinterpretation PTH hyporesponsiveness (PTH resistance) in CKD [52, 53]. by non-experts and ‘shortcutting’ have been described as addi- Whether these 2017 KDIGO changes will result in a greater tional difficulties in guidelines [37, 44, 54, 55, 70–73]. In any case, number of patients reaching later PTH extremes of risk or real the methodical analysis of one’s own decision-making is man- improvement in patient-level outcomes remains to be seen. datory. While ‘precision medicine’ in nephrology is still far The rationale for all these KDIGO-suggested interventions is away [74] and big data will transform our clinical decision-mak- based, with very few exceptions, on our evolving understanding ing [75] (some consider that the role of the physician will be en- of the complex CKD–MBD pathophysiology, the assumed biolog- hanced, not diminished, as evidence and data grow) [76], the ical plausibility for a treatment effect, epidemiological studies main ethical principles (Table 4) [77] should be seriously and other considerations [45]. We are used to achieving a taken into account. In this context, we believe that both desired target value for an ion or hormone mainly on the basis ‘individualization’ of treatments (applying the principles of of retrospective studies; however, it is becoming clear not only non-maleficence and beneficence) and, especially, ‘shared’ that the achievement of these artificial goals is difficult [54, 55] decision-making (respecting the patient’s autonomy and social (there are many interactions among biomarkers and among justice) should help in resolving difficult dilemmas between treatments), but also that the evidence base for them is of lim- ‘treating’ or ‘waiting’ in any particular situation. ited value—since, as mentioned before, most are graded as 2C. We should all ‘go to the balcony’ and recognize that, from This is so even if in some studies (mainly retrospective, post hoc that wider perspective, guidelines assist in the provision of rec- or using predefined but nominally significant ‘secondary’ out- ommendations at the individual personalized (shared!) patient- comes) it seems that the mortality rates are reduced by the level, where straightforward ethical principles may be of valu- achievement of certain targets or the use of certain drugs able help (Table 4). As a matter of fact, the most significant [56–61]. Therefore, the benefit of achieving the difficult KDOQI change in the current update is a move towards a more articu- or the more recent KDIGO targets [54, 55], in terms of improve- lated pragmatic and personalized approach to management for ments in hard outcomes or survival, remains to be confirmed in each patient [12, 13]. The attached guideline expert research 518 | J. Bover et al. Table 2. Summary of evidence grades in the 2012 KDIGO glomerulonephritis and 2009 transplant recipient guideline statements compared with the 2017 CKD–MBD guidelines [64, 65] Statements Statements Statements Evidence grades (glomerular), n (%) (transplant), n (%) (CKD–MBD), n (%) 1A 6 (3.1) 3 (1.2) 1 (2.2) 1B 22 (11.5) 15 (6.2) 2 (4.3) 1C 17 (8.9) 18 (7.5) 3 (6.5) 1D 0 15 (6.2) 0 Total GRADE 1 (recommendations) 45 (23.5) 51 (21.1) 6 (13) 2A 0 1 (0.4) 1 (2.2) 2B 10 (5.2) 11 (4.6) 6 (13) 2C 51 (26.6) 59 (24.5) 17 (37) 2D 60 (31.3) 76 (31.5) 4 (8.7) Total GRADE 2 (suggestions) 121 (63.1) 147 (61) 28 (60.9) Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 Not graded 26 (13.5) 43 (17.8) 12 (26.1) Total number of statements 192 (100) 241 (100) 46 (100) Note that the number of statements in CKD–MBD guidelines is (obviously) lower as is the percentage of Grade 1 evidence statements, whereas ‘not graded’ evidence is more frequent. Regarding the percentage of Grade 2 evidence statements (similar percentage in all guidelines and more frequent than Grade 1), CKD–MBD guidelines include a higher percentage of 2A–2C statements versus 2D statements (which are more frequent in the other guidelines). Table 3. Summary of contributions (pros) and criticisms (cons) of Evidence Based Medicine (discussed in references [15, 68, 69]) Pros: contributions Cons: criticisms Focus on comparative research, over- or underdiagnosis Reductionism of the scientific method and over- or undertreatment Overly strict adherence to the evidence hierarchy pyramid [improved by the GRADE Measurements of the quality of care framework (see Figure 3 and Table 1)] Improving publishing standards Encouragement of ‘cookbook medicine’ (automatic decision-making—algorithms— Ensuring all trials are registered (investigators report discouraging deliberation) only 50% of trials, publication bias) Promotion of rule-based reasoning instead of intuitive and experimental thinking Avoiding waste in research production There is no high-quality evidence that its application has improved patient care Upcoming ‘living’ systematic reviews and ‘living’ (poor uptake of Evidence Based Medicine in clinical practice) guideline recommendations ‘Hijacking’ by commercial interests Publication bias (only ‘positive’ studies) Delay in updating new relevant evidence when it becomes available (‘evidence- practice’ or ‘know–do’ gap) Table 4. Main ethical principles (adapted from reference [77]) forward momentum will be maintained by following the best available evidence. Principle Definition Non-maleficence Duty to avoid causing harm and to minimize ACKNOWLEDGEMENTS harm to the patient J.B. belongs to the Spanish ‘Red Nacional RedinRen’ (RD06/ Respect for Duty to respect a patient’s right of self- 0016/0001 and RD12/0021/0033) and the ‘Red de Biobancos autonomy governance Nacional Espan ~ ola’ (RD09/0076/00064), as well as to the Beneficence Duty to maximize benefits and to enhance the patient’s well-being Catalan ‘Grupo Catalán de Investigación’ (AGAUR, Agència Justice Duty to treat patients fairly and equitably de Gestió d’Ajuts Universitaris i de Recerca) (2009 SGR-1116). Collaborations with the Fundación Renal In ~ igo Alvarez de Toledo (FRIAT) have also been arranged. We also thank Mr Ricardo Pellejero for his priceless bibliographic assistance recommendations are also extremely valuable in enabling the and Dr Sergi Sabaté, Dr Ana Vila, Dr Raul Alejandro Quiroga, detection of major flaws in current data. As Nicolaus Dr Verónica Coll and Dr Jackson Ochoa for their valuable Copernicus nicely put it ‘to know that we know what we know, contributions to this review. and to know that we do not know what we do not know, that is true knowledge’. However, while the history of medicine has CONFLICT OF INTEREST STATEMENT sometimes made fools of physicians [20–22], including those ex- treme advocates of EBM as very nicely reported in the system- J.B. declares advisory, lecture fees, and/or travel funding atic review of the parachute gravitational challenge [78], from Amgen, Abbvie, Sanofi-Genzyme, Shire, Vifor- comparison of the current status of our CKD population with Fresenius-Pharma, and Sanifit. P.U.T. declares advisory, lec- that in earlier decades clearly suggests that much progress has ture fees and/or travel funding from Abbvie, Amgen, been made in all our fields (before and after EBM) and that the Astellas, Medici, Sanofi, Vifor-Pharma, Fresenius Medical Evidence in CKD–MBD guidelines | 519 Care, and Hemotech. I.dS. declares lecture and/or travel (CKD-MBD). What are the real changes? G Ital Nefrol 2018; 35: funding fees from Vifor-Fresenius-Pharma and Alexion. pii M.C. declares advisory, lecture fees and/or travel funding 14. Isakova T, Nickolas TL, Denburg M et al. KDOQI US from Amgen, Abbvie, Sanofi-Genzyme, Shire, Vifor- Commentary on the 2017 KDIGO Clinical Practice Guideline Fresenius-Pharma, and Baxter. Update for the Diagnosis, Evaluation, Prevention, and Treatment of chronic kidney disease-mineral and bone dis- order (CKD-MBD). Am J Kidney Dis 2017; 70: 737–751 REFERENCES 15. Djulbegovic B, Guyatt GH. Progress in evidence-based medi- 1. Moe S, Drüeke T, Cunningham J et al. Definition, evaluation, cine: a quarter century on. Lancet 2017; 390: 415–423 and classification of renal osteodystrophy: a position state- 16. Guyatt GH. Evidence-based medicine. ACP J Club 1991; 114: ment from Kidney Disease: Improving Global Outcomes A–16 (KDIGO). Kidney Int 2006; 69: 1945–1953 17. Szajewska H. Clinical practice guidelines: based on emi- 2. Kidney Disease: Improving Global Outcomes (KDIGO) CKD- nence or evidence? Ann Nutr Metab 2014; 64: 325–331 MBD Work Group. KDIGO clinical practice guideline for the 18. Pincus T, Tugwell P. Shouldn’t standard rheumatology clini- Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 diagnosis, evaluation, prevention, and treatment of chronic cal care be evidence-based rather than eminence-based, elo- kidney disease-mineral and bone disorder (CKD-MBD). quence-based, or elegance-based? J Rheumatol 2007; 34: 1–4 19. Rettig RA, Jacobson PD, Farquhar CM et al. False Hope: Bone Kidney Int Suppl 2009; 113: S1–130 3. Cozzolino M, Uren ~ a-Torres P, Vervloet MG et al. Is chronic Marrow Transplantation for Breast Cancer. New York, NY: Oxford University Press, 2007 kidney disease-mineral bone disorder (CKD-MBD) really a 20. Moore TJ. Deadly Medicine. New York, NY: Simon & Shuster, syndrome? Nephrol Dial Transplant 2014; 29: 1815–1820 1995 4. Vervloet MG, Massy ZA, Brandenburg VM et al. Bone: a new 21. Rossouw JE, Anderson GL, Prentice RL et al. Risks and bene- endocrine organ at the heart of chronic kidney disease and fits of estrogen plus progestin in healthy postmenopausal mineral and bone disorders. Lancet Diabetes Endocrinol 2014; women: principal results From the Women’s Health 2: 427–436 Initiative randomized controlled trial. JAMA 2002; 288: 5. Covic A, Vervloet M, Massy ZA et al. Bone and mineral disor- 321–333 ders in chronic kidney disease: implications for cardiovascu- 22. Craig JC, Tong A, Strippoli GF. New approaches to trials in lar health and ageing in the general population. Lancet glomerulonephritis. Nephrol Dial Transplant 2017; 32: i1–i6 Diabetes Endocrinol 2018; 6: 319–331 23. Palmer SC, McGregor DO, Macaskill P et al. Meta-analysis: vi- 6. Kidney Disease Initiative Global Outcomes. Methods for tamin D compounds in chronic kidney disease. Ann Intern guideline development. Kidney Int Suppl (2011) 2012; 2: Med 2007; 147: 840–853 317–323 24. Palmer SC, McGregor DO, Craig JC et al. Vitamin D com- 7. Torregrosa JV, Bover J, Cannata Andı́a J et al. Spanish Society pounds for people with chronic kidney disease not requiring of Nephrology recommendations for controlling mineral dialysis. Cochrane Database Syst Rev 2009; 4: CD008175 and bone disorder in chronic kidney disease patients (S.E.N.- 25. Palmer SC, McGregor DO, Craig JC et al. Vitamin D com- M.B.D). Nefrologia 2011; 31 (Suppl 1): 3–32 pounds for people with chronic kidney disease requiring di- 8. Uhlig K, Berns JS, Kestenbaum B et al. KDOQI US commentary alysis. Cochrane Database Syst Rev 2009; 4: CD005633 on the 2009 KDIGO Clinical Practice Guideline for the 26. Sekercioglu N, Thabane L, Dı́az Martı́nez JP et al. Diagnosis, Evaluation, and Treatment of CKD-mineral and Comparative effectiveness of phosphate binders in patients bone disorder (CKD-MBD). Am J Kidney Dis 2010; 55: 773–799 with chronic kidney disease: a systematic review and net- 9. Goldsmith DJ, Covic A, Fouque D et al. Endorsement of the work meta-analysis. PLoS One 2016; 11: e0156891 Kidney Disease Improving Global Outcomes (KDIGO) 27. Habbous S, Przech S, Acedillo R et al. The efficacy and safety Chronic Kidney Disease-Mineral and Bone Disorder (CKD- of sevelamer and lanthanum versus calcium-containing MBD) Guidelines: a European Renal Best Practice (ERBP) com- and iron-based binders in treating hyperphosphatemia in mentary statement. Nephrol Dial Transplant 2010; 25: patients with chronic kidney disease: a systematic review 3823–3831 and meta-analysis. Nephrol Dial Transplant 2017; 32: 111–125 10. Ketteler M, Elder GJ, Evenepoel P et al. Revisiting KDIGO clini- 28. Jamal SA, Vandermeer B, Raggi P et al. Effect of calcium- cal practice guideline on chronic kidney disease-mineral based versus non-calcium-based phosphate binders on and bone disorder: a commentary from a Kidney Disease: mortality in patients with chronic kidney disease: an Improving Global Outcomes controversies conference. updated systematic review and meta-analysis. Lancet 2013; Kidney Int 2015; 87: 502–528 382: 1268–1277 11. KDIGO 2017 Clinical Practice Guideline Update for the 29. Schmillen H. What is Evidence-Based Practice and Evidence- Diagnosis, Evaluation, Prevention, and Treatment of Based Medicine. https://libguides.library.ohio.edu/evidence Chronic Kidney Disease–Mineral and Bone Disorder (CKD- (20 September 2019, date last accessed) MBD). Kidney Int Suppl (2011) 2017; 7: 1–59 30. Sacket DL, Rosenberg WMC, Gray JAM et al. Evidence based 12. Burton JO, Goldsmith DJ, Ruddock N et al. Renal association medicine: what it is and what it isn’t. BMJ 1996; 312: 71–72 commentary on the KDIGO (2017) clinical practice guideline 31. Atkins D, Best D, Briss PA et al.; GRADE Working Group. update for the diagnosis, evaluation, prevention, and treat- Grading quality of evidence and strength of recommenda- ment of CKD-MBD. BMC Nephrol 2018; 19: 240 tions. BMJ 2004; 328: 1490 13. Pasquali M, Bellasi A, Cianciolo G et al. A nome del GdS 32. Cannata-Andia J, Passlick-Deejen J, Ritz E. Management of Elementi Traccia e Metabolismo Minerale della Società the renal patient: experts’ recommendation and clinical Italiana di Nefrologia. Update 2017 of the KDIGO guidelines algorithms on renal osteodystrophy and cardiovascular risk on Chronic Kidney Disease-Mineral and Bone Disorder factors. Nephrol Dial Transplant 2000; 15: 39–57 520 | J. Bover et al. 33. National Kidney Foundation. K/DOQI clinical practice guide- 51. Wang AY, Fang F, Chan J et al. Effect of paricalcitol on left lines for bone metabolism and disease in chronic kidney dis- ventricular mass and function in CKD–the OPERA trial. J Am ease. Am J Kidney Dis 2003; 42 (Suppl 3): S1–201 Soc Nephrol 2014; 25: 175–186 34. Upshur RE. Are all evidence-based practices alike? Problems 52. Evenepoel P, Bover J, Uren ~ a Torres P et al. Parathyroid hor- in the ranking of evidence. CMAJ 2003; 169: 672–673 mone metabolism and signaling in health and chronic kid- 35. Schünemann HJ, Best D, Vist G et al.; GRADE Working Group. ney disease. Kidney Int 2016; 90: 1184–1190 Letters, numbers, symbols and words: how to communicate 53. Bover J, Uren~ a P, Evenepoel P et al. PTH receptors and skeletal grades of evidence and recommendations. CMAJ 2003; 169: resistance to PTH action. In: Uren ~ a P, Covic A (eds). 677–680 Parathyroid Glands in CKD. Springer (In press) 36. Pirolli P, Card S. Information foraging. Psychol Rev 1999; 106: 54. Cozzolino M. CKD-MBD KDIGO guidelines: how difficult is 643–675 reaching the ‘target’? Clin Kidney J 2018; 11: 70–72 37. D Agostino M, Mejı́a FM, Martı́ M et al. Infoxication in health. 55. Fernández E. Are the K/DOQI objectives for bone min- Health information overload on the Internet and the risk of eral alterations in stage 3-5 chronic kidney disease important information becoming invisible. Rev Panam Salud patients unreachable or inadequate? Nefrologia 2013; 33: 1–6 Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023 Publica 2018; 41: e115 38. Ortiz-Herbener F, Bover J, Gracia S et al. Nefrologı́a e Internet: 56. Block GA, Klassen PS, Lazarus JM et al. Mineral metabolism, del desconocimiento a la infoxicación. Dial Traspl 2004; 25: mortality, and morbidity in maintenance hemodialysis. J Am 203–214 Soc Nephrol 2004; 15: 2208–2218 39. Guyatt GH, Sackett DL, Sinclair JC et al. Users’ guides to the 57. Fernández-Martı́n JL, Martı́nez-Camblor P, Dionisi MP et al. medical literature. IX. A method for grading health care rec- Improvement of mineral and bone metabolism markers is ommendations. Evidence-Based Medicine Working Group. associated with better survival in haemodialysis patients: JAMA 1995; 274: 1800–1804 the COSMOS study. Nephrol Dial Transplant 2015; 30: 40. Golden SH, Bass EB. Validity of meta-analysis in diabetes: 1542–1551 meta-analysis is an indispensable tool in evidence synthe- 58. Danese MD, Halperin M, Lowe KA et al. Refining the defini- tion of clinically important mineral and bone disorder in he- sis. Diabetes Care 2013; 36: 3368–3373 modialysis patients. Nephrol Dial Transplant 2015; 30: 41. Murad MH, Asi N, Alsawas M et al. New evidence pyramid. 1336–1344 Evid Based Med 2016; 21: 125–127 ~ a-Torres P, Lloret MJ et al. Integral pharmacolog- 59. Teng M, Wolf M, Lowrie E et al. Survival of patients undergo- 42. Bover J, Uren ing hemodialysis with paricalcitol or calcitriol therapy. N ical management of bone mineral disorders in chronic kid- Engl J Med 2003; 349: 446–456 ney disease (part I): from treatment of phosphate imbalance 60. Suki WN, Zabaneh R, Cangiano JL et al. Effects of seve- to control of PTH and prevention of progression of cardio- lamer and calcium-based phosphate binders on mortal- vascular calcification. Expert Opin Pharmacother 2016; 17: ity in hemodialysis patients. Kidney Int 2007; 72: 1247–1258 1130–1137 43. Bover J, Uren~ a-Torres P, Lloret MJ et al. Integral pharmacolog- 61. Chertow GM, Block GA, Correa-Rotter R et al.; EVOLVE Trial ical management of bone mineral disorders in chronic kid- Investigators. Effect of cinacalcet on cardiovascular disease ney disease (part II): from treatment of phosphate in patients undergoing dialysis. N Engl J Med 2012; 367: imbalance to control of PTH and prevention of progression 2482–2494 of cardiovascular calcification. Expert Opin Pharmacother 62. Altman DG, Bland JM. Absence of evidence is not evidence of 2016; 17: 1363–1373 absence. BMJ 1995; 311: 485 44. Burlacu A, Genovesi S, Ortiz A et al. Pros and cons of antith- 63. Yee J. Eminence-based medicine: the King is dead. Adv rombotic therapy in end-stage kidney disease: a 2019 up- Chronic Kidney Dis 2012; 19: 1–2 date. Nephrol Dial Transplant 2019; 34: 923–933 64. KDIGO Clinical Practice Guidelines for Glomerulonephritis. 45. Chen W, Bushinsky DA. Chronic kidney disease: KDIGO Kidney Int Suppl 2012; 2: 139–274 CKD-MBD guideline update: evolution in the face of uncer- 65. KDIGO Clinical Practice Guideline for the Care of Kidney tainty. Nat Rev Nephrol 2017; 13: 600–602 Transplant Recipients. Am J Transplant 2009; 9: S1–S157 46. Wilson LM, Rebholz CM, Jirru E et al. Benefits and harms of 66. Lv J, Zhang H, Wong MG et al.; TESTING Study Group. Effect osteoporosis medications in patients with chronic kidney of oral methylprednisolone on clinical outcomes in patients disease: a systematic review and meta-analysis. Ann Intern with IgA nephropathy: the testing randomized clinical trial. Med 2017; 166: 649–658 JAMA 2017; 318: 432–442 47. Bover J, Uren ~ a-Torres P, Laiz Alonso AM et al. Osteoporosis, 67. Fried LF, Emanuele N, Zhang JH et al.; VA NEPHRON-D bone mineral density and CKD-MBD (II): therapeutic impli- Investigators. Combined angiotensin inhibition for the treat- cations. Nefrologia 2019; 39: 227–242 ment of diabetic nephropathy. N Engl J Med 2013; 369: 48. Moe SM, Nickolas TL. Fractures in patients with CKD: time 1892–1903 for action. Clin J Am Soc Nephrol 2016; 11: 1929–1931 68. Elliott JH, Turner T, Clavisi O et al. Living systematic reviews: 49. Tentori F, Wang M, Bieber BA et al. Recent changes in thera- an emerging opportunity to narrow the evidence-practice peutic approaches and association with outcomes among gap. PLoS Med 2014; 11: e1001603 patients with secondary hyperparathyroidism on chronic 69. Akl EA, Meerpohl JJ, Elliott J et al.; on behalf of the Living hemodialysis: the DOPPS study. Clin J Am Soc Nephrol 2015; Systematic Review Network. Living systematic reviews: 4. 10: 98–109 Living guideline recommendations. J Clin Epidemiol 2017; 91: 50. Thadhani R, Appelbaum E, Pritchett Y et al. Vitamin D ther- 47–53 apy and cardiac structure and function in patients with 70. Bover J, Górriz JL, Martı́n de Francisco AL et al.; Grupo de chronic kidney disease: the PRIMO randomized controlled Estudio OSERCE de la Sociedad Espan ~ ola de Nefrologı́a. trial. JAMA 2012; 307: 674–684 [Unawareness of the K/DOQI guidelines for bone and Evidence in CKD–MBD guidelines | 521 mineral metabolism in predialysis chronic kidney disease: 74. Hunter DJ, Longo DL. The precision of evidence needed to results of the OSERCE Spanish multicenter-study survey]. practice “precision medicine”. N Engl J Med 2019; 380: Nefrologia 2008; 28: 637–634 2472–2474 71. Roetker NS, Peng Y, Ashfaq A et al. Adherence to kidney dis- 75. Harari YN. 21 Lessons for the 21st Century. Spiegel & Grau, 2018 ease: improving global outcomes mineral and bone guide- 76. Williams RD. Of Eminence-Based and Evidence-Based Medicine. lines for monitoring biochemical parameters. Am J Nephrol https://www.aao.org/eyenet/article/of-eminence-based-and-evi 2019; 49: 225–232 dence-based-medicine (20 September 2019, date last accessed). 72. Skolnik N. Reexamining recommendations for treatment of 77. Jalaly J, Dineen K, Gronowski AM. A 30-year-old patient who hypercholesterolemia in older adults. JAMA 2019; 321: refuses to be drug tested. Clin Chem 2016; 62: 807–809 1249–1250 78. Smith GCS, Pell JP. Parachute use to prevent death and major 73. Dowell D, Haegerich T, Chou R. No shortcuts to safer opioid trauma related to gravitational challenge: systematic review prescribing. N Engl J Med 2019; 380: 2285–2287 of randomised controlled trials. BMJ 2003; 327: 1459–1461 Downloaded from https://academic.oup.com/ckj/article/13/4/513/5715959 by guest on 10 October 2023