AGREEing on clinical practice guidelines for idiopathic steroid-sensitive nephrotic syndrome in children

Background Nephrotic syndrome is the most common kidney disease in children worldwide. Our aim was to critically appraise the quality of recent Clinical Practice Guidelines (CPGs) for idiopathic steroid-sensitive nephrotic syndrome (SSNS) in children in addition to summarize and compare their recommendations. Methods Systematic review of CPGs. We identified clinical questions and eligibility criteria and searched and screened for CPGs using bibliographic and CPG databases. Each included CPG was assessed by four independent appraisers using the Appraisal of Guidelines for REsearch & Evaluation II (AGREE-II) instrument. We summarized the recommendations in a comparison practical table. Results Our search retrieved 282 citations, of which three CPGs were eligible and appraised: Kidney Disease: Improving Global Outcomes (KDIGO) 2012, Japan Society for Pediatric Nephrology (JSPN) 2014, and American Academy of Pediatrics (AAP) 2009. Among these, the overall assessment of two evidence-based CPGs scored > 70% (KDIGO and JSPN), which was consistent with their higher scores in the six domains of the AGREE II Instrument. In domain 3 (rigor of development), KDIGO, JSPN, and AAP scored 84%, 74%, and 41%, respectively. In domain 5 (applicability), they scored 22%, 16%, and 19%, respectively, and in domain 6 (editorial independence), they scored 94%, 65%, and 88%, respectively. Conclusions The methodological quality of the KDIGO CPG was superior, followed by JSPN and AAP CPGs with the relevant recommendations for use in practice. Systematic review registration The protocol was registered in the Center for Open Science (OSF) DOI: 10.17605/OSF.IO/6QTMD and in the International prospective register of systematic reviews PROSPERO 2020 CRD42020197511. Supplementary Information The online version contains supplementary material available at 10.1186/s13643-021-01666-w.


Background
Nephrotic syndrome is considered the most common kidney disease in children worldwide. It is defined by a clinical characteristic of hypoalbuminemia < 25 g/L, edema and nephrotic range proteinuria > 40 mg/m 2 /h, or protein/creatinine ratio > 200 mg/mmol in a spot urine sample [1,2]. There are many classifications of nephrotic syndrome: one of the classifications is based on the clinical response to steroids. Most children with nephrotic syndrome respond to steroids within 4 weeks of proper steroid therapy (i.e., steroid-sensitive nephrotic syndrome [SSNS]); however, these children behave differently afterward [3].
The Kidney Disease: Improving Global Outcomes (KDIGO) stated the important scientific definitions of nephrotic syndrome: Patients with steroid-dependent nephrotic syndrome (SDNS) are defined as patients who have relapse while weaning the steroid dose or within 14 days of steroid discontinuation. Frequently relapsing nephrotic syndrome (FRNS) is defined as two or more relapses in 6 months after the initial response or four and more relapses in any 1-year period [2].
The global incidence rate of nephrotic syndrome of childhood is variable among countries and ranges from 1.15 to 16.9 per 100,000 children annually [4,5]. Children with nephrotic syndrome require prolonged use of immunosuppressive agents, with multiple adverse effects, including infections and other side effects. A study conducted in a tertiary care center in Saudi Arabia by Alfakeeh et al. showed that the cumulative yearly dose of steroids is an important independent risk of infection [6].
In our practice, we noted center-to-center differences in managing patients diagnosed with SSNS, SDNS, and FRNS. The main differences we observed were in the duration of steroid therapy, steroid weaning, selection of second-line immunosuppressive agent and its targeted levels, and other practice parameters [7][8][9].
We would like to adapt a unified national evidencebased clinical practice guideline (CPG) for the management of these patients. Our aim from developing this CPG is to unify the practice between centers and improve patients' outcomes and experience.
CPGs are statements that include recommendations intended to optimize patient care that are informed by a systematic review (SR) of evidence and an assessment of the benefits and harms of alternative care options [10]. To date, there are no national CPGs to provide evidencebased guidance for healthcare professionals during the provision of clinical care for children with idiopathic SSNS in Saudi Arabia. In 2019, a decision was made to launch a project for adaptation of a national evidence-based CPG for the management of children with SSNS by the Department of Pediatrics, College of Medicine, King Saud University (KSU) in collaboration with the Saudi Society of Nephrology and Transplantation, as the governing body of nephrology including pediatric nephrology practice in Saudi Arabia, to provide guidance and recommendations to pediatricians, nephrologists, pharmacists, nurses, pathologists, children with SSNS, and all related stakeholders in Saudi Arabia who care for these children. The project is guided by the "KSU-Modified-ADAPTE" as a formal CPG adaptation methodology consisting of three phases: setup, adaptation, and finalization [11][12][13].
The Appraisal of Guidelines for REsearch & Evaluation (AGREE II) instrument is the gold standard for the quality appraisal of CPGs. AGREE II is a validated CPG appraisal tool cited in > 1013 articles and endorsed by several healthcare organizations [14][15][16]. AGREE II identifies components that should be addressed by CPGs to improve their quality and trustworthiness and obtain positive patient outcomes [11,[14][15][16].
Since the SR of CPGs, using AGREE II, is a key step in the CPG adaptation process, we have dedicated this study to report the results of this SR and critically appraise recently published CPGs for childhood SSNS using AGREE II [11,17,18].
We utilized the PIPOH Model [i.e., Population (P), Intervention (I), Professionals (P), Outcomes (O), and Healthcare setting or context (H)] [11,12,17] where the Population (P) included children aged 2-12 years with non-congenital, idiopathic SSNS, including new-onset nephrotic syndrome, SDNS, or FRNS without any comorbidities. Intervention(s) (I) included all pharmacological therapeutic agents. Professionals (P) or target users of CPGs included mainly pediatric nephrologists, general pediatricians, and pharmacists and nurses with relevant nephrology experience. Outcomes (O) included prevention of disease relapse, appropriateness of prescription (i.e., duration of steroid courses in newly diagnosed SSNS and drug of choice of the second agent in SDNS or FRNS). Healthcare settings or context (H) included CPGs to be used in secondary and tertiary healthcare settings. The four main health questions were prioritized for this review. Additionally, we have utilized the PICAR statement where P: Population, clinical indication(s), and condition(s), I: Intervention(s), C: Comparator(s), Comparison(s), and (key) Content, A: Attributes of eligible CPGs, and R: Recommendation characteristics [17].

Methods
The protocol for this study was registered in PROSPERO (International Prospective Register of Systematic Reviews) (Protocol ID: CRD42020197511) and in the Center for Open Science (OSF) (DOI 10.17605/OSF.IO/6QTMD). Our CPG working group included expert pediatric nephrologists, general pediatricians, a pediatric intensivist, a clinical pharmacist, a renal pathologist, and a specialized nurse guided by two pediatricians with expertise in CPG methodologies. Two external international experts in nephrology were invited as international collaborators to contribute to this CPG project.

Data sources and search strategy
We systematically searched MEDLINE and EMBASE databases for relevant guidelines using the Ovid platform and hand-searched EBSCO DynaMed Plus (USA), ECRI Guidelines Trust, Guidelines International Network, International Guideline Library, National Institute for Health and Care Excellence (UK), Scottish Intercollegiate Guidelines Network (UK), and Australian National Health and Medical Research Council (Australia). Moreover, we searched databases of national and international societies specializing in fields related to our health topic of SSNS, including the Japanese Society of Pediatric Nephrology (JSPN), KDIGO, International Society of Nephrology, American Society of Pediatric Nephrology, National Kidney Foundation, American Academy of Pediatrics (AAP), and Scottish Paediatric Renal and Urology Network. The search terms used included combinations of subject headings and keywords with various synonyms for idiopathic SSNS, nephrotic syndrome, nephrology, pediatrics, pediatric medicine, child health, treatment, management, pharmacology, practice guidelines, CPGs, healthcare quality, patient safety, evidence-based medicine, AGREE II instrument, quality assessment, critical appraisal, and evidence-based pediatrics (see search strategy in additional file 1). The search was limited to published or updated CPGs between January 1, 2009, and December 31, 2019. We have decided on the last 10 years as the cutoff for dates of publication because typically CPGs are updated every 2-5 years [19]. The search was conducted by two CPG methodologists (RA and YA). We utilized the PIPOH model in addition to the PICAR statement (additional file 2) to support the CPG eligibility identification process [11,12,17]. Two reviewers (MA and AA) independently screened titles and abstracts of retrieved CPGs and articles meeting the inclusion criteria. The screening and full-text review were checked by three different reviewers (MH, AA, and AA). Disagreements were resolved by focus group discussions after retrieving and reviewing the full-text articles or full CPG documents.

Inclusion and exclusion criteria
Teams of two reviewers, independently and in duplicate, screened titles and abstracts and potentially eligible full-text reports to determine eligibility. Disagreements were resolved through a review by RA. The eligibility criteria were as follows: (1) evidence-based with a clear record of their development methods; (2) English or Arabic language; (3) original source CPGs (de novo development); (4) national or international scope; and (5) published by an organization or group authorship and accessible from a CPG database or peer-reviewed journal. Only the most current version of each source CPG was appraised.
The exclusion criteria were CPGs that were published earlier than 2009, not in the English or Arabic language, adapted from other CPGs, presented as consensus or expert-based statements, or had a single author.

AGREE II instrument workshop
The two CPG methodologists (YA and RA) conducted a capacity building workshop for the review team through hands-on sessions in the concepts of evidence-based medicine and evidence-based CPG standards using the AGREE II instrument tool in 2 days. During the workshop, participants refined the research questions of interest to adapt a CPG to local practice (see the abovementioned health questions). Afterward, each reviewer scored his/her assigned CPGs. All four reviewers critically appraised each CPG. All appraisers reviewed the full CPG documents, including any updates with any relevant supplementary information or links to online web pages related to the CPG methods or CPG implementation tools. For each item, AGREE appraisers were instructed to record the justifications for their scores in the "Comment" section [20].

Assessment of childhood SSNS CPGs using AGREE II
The AGREE II instrument (www.agreetrust.org) consisted of 23 items organized into six domains: scope and purpose, stakeholder involvement, rigor of development, clarity of presentation, applicability, and editorial independence [14,15]. Each item was scored on a 7-point Likert scale. The AGREE II evaluation was guided by utilizing its online version: "My AGREE PLUS," which supports having a CPG appraisal group for each CPG that compiles and calculates the items' ratings into domain ratings and comments [14,15]. The four AGREE II appraisers for each CPG comprised a multidisciplinary group with expertise in pediatric nephrology (consultant physicians and head nurse) and pediatric clinical pharmacology (one clinical pharmacist), in addition to a general pediatrician with expertise in CPG methodologies.
Wide discrepancies between the assessors' scores of items or questions (i.e., whenever there was a difference between these scores of > 3) were resolved by discussion with the appraisal group. The online My AGREE PLUS automatically calculated the standardized AGREE domain scores or ratings (%). We agreed upon a cutoff point of 70% for each AGREE standardized domain score or rating. After the appraisal, more weight was emphasized on the scores of domains 3 and 5 to facilitate the filtration and final evaluation of the reporting quality of included CPGs. Similar cutoff values were reported [21][22][23]. In addition to the classification of the six AGREE II domains, the evidence base of the included CPGs, their references' sections, was screened for SRs or meta-analyses, specifically Cochrane reviews.

Analysis plan
For each AGREE II domain, we calculated standardized scores ranging from 0 to 100% using the methods suggested by the AGREE II instrument. The key recommendations of the eligible CPGs were summarized in a comparative tabular format. The quality of CPGs was classified based upon the rating of domain 3 (rigour of development) where a high-quality CPG will receive a standardized domain rating of more than or equal to 70%, moderate quality CPG (40-69%), and low quality (less than 40%).

Identification of CPGs for SSNS in children
We retrieved a total of 282 records. After screening titles and abstracts, eight were included for full-text assessment, and only three were eligible for the review as illustrated in the PRISMA [24] flowchart ( Fig. 1) and the PRISMA checklist (additional file 3). These CPGs were developed by the AAP [25], JSPN [26][27][28], and KDIGO-Chapter 3 [29]. At the time of writing this manuscript, the 2020 KDIGO "CPG on Glomerular Diseases" update was under development as the public review has just closed in the official KDIGO website [30] Key characteristics of childhood SSNS CPGs Table 1 highlights the characteristics of all eligible CPGs. The CPG developer organizations were reference, specialized professional organizations in pediatrics or nephrology, including KDIGO, AAP, and JSPN. All organizations were from high-income countries.

Reporting the quality of Childhood SSNS CPGs
The AGREE II standardized domain ratings are summarized in Table 2.

Domain 1: scope and purpose
The AGREE II standardized score for domain 1 ranged from 65 to 100%. The scores of two CPGs were > 70% in domain 1 (KDIGO = 100% and AAP = 75%).

Domain 2: stakeholder involvement
The AGREE II standardized domain scores for domain 2 ranged from 60 to 86%. The score of a single CPG was > 70% in domain 2 (JSPN = 86%).

Domain 3: rigor of development
The AGREE II standardized scores for domain 3 ranged from 41 to 84%. The score of two CPGs were > 70% in domain 3 (KDIGO = 84% and JSPN = 74%). They both reported utilizing the Grading of Recommendations, Moreover, the KDIGO CPG reported its adherence to two sets of CPG standards, namely the Conference on Guideline Standardization Checklist for Reporting CPGs and the Institute of Medicine Standards for Systematic Reviews and Guidelines.

Domain 5: applicability
The AGREE II standardized scores for domain 5 ranged from 16 to 22%. None of the included CPGs scored > 70%.

Overall assessment
The AGREE II standardized domain scores for the first overall assessment ranged from 58 to 75%. Two CPGs scored > 70% (KDIGO and JSPN), which was consistent with the high scores in the six AGREE II domains.

Recommending the childhood SSNS CPGs for use in practice
The second (overall) assessment (i.e., recommendation for using the CPG in practice) revealed a consensus between the reviewers on recommending the use of two CPGs.  All included CPGs cited SRs in their reference list. The largest number of SR citations was observed in the JPNS CPG (n = 12), among them were six Cochrane SRs [26][27][28], followed by KDIGO-Chapter 3 (n = 5) including four Cochrane SRs [29], and lastly AAP (n = 4) with one Cochrane SR [25].

Discussion
Although several regional and national guidelines have been published [25][26][27][28][29][30], shared treatment guidelines are limited in Saudi Arabia, and consequently, physicians rely on the clinical expertise of each unit to select the best treatment option for pediatric patients with SSNS. To the best of our knowledge, this review is novel in that it systematically evaluates the quality of recently published CPGs of SSNS in children using the AGREE II instrument as a part of a national CPG adaptation initiative.
Three CPGs addressing the management of children with SSNS were assessed using the AGREE II instrument. This AGREE II assessment highlighted several areas of improvement in the methodological rigor of the included CPGs. Although the assessment of overall guideline quality and recommendation for use are standard components of AGREE II, it is possible that they are underreported: standardized domain scores for the first overall assessment ranged from 58 to 75%, with the KDIGO and JSPN scoring > 70%.
In this review, the scores of all three CPGs were > 70% in domain 4. However, it was previously suggested that domain 3 was the strongest indicator of guideline quality than other domains [31][32][33], with a high score signifying a low degree of bias and evidence-based guideline development [33]. Conversely, a low score suggests serious methodological flaws. This may be the case, for example, if the team in charge of developing the guideline includes experts with little experience in guideline development or if the systematic search strategy is inadequately described [31].
A summary for the key recommendations of the three included CPGs is presented in Tables 3 and 4  • Modified ISKDC regimen 1. 60 mg/m 2 /day or 2.0 mg/kg/day in three divided doses daily until confirmation of the resolution of proteinuria for at least 3 days but not exceeding 4 weeks (maximum: 60 mg/ day), followed by 2. 60 mg/m 2 or 2.0 mg/kg once in the morning on alternate days for 2 weeks (maximum: 60 mg on alternate days), followed by 3. 30 mg/m 2 or 1.0 mg/kg once in the morning on alternate days for 2 weeks (maximum: 30 mg on alternate days), followed by 4. 15 mg/m 2 or 0.5 mg/kg once in the morning on alternate days for 2 weeks (maximum: 15 mg on alternate days). Use immunosuppressive agents (e.g., cyclosporine, cyclophosphamide) in the treatment of frequently relapsing and steroid-dependent nephrotic syndrome (Grade C1) due to the development of various steroid-induced side effects.

. A separate
• Initially: daily prednisone until the child has been in remission for at least 3 days • Then: alternate-day prednisone for at least 3 months. (2C) Long term steroid: prednisone to be given on alternate days in the lowest dose to maintain remission without major adverse effects. (2D) If not effective: daily prednisone at the lowest dose to be given to maintain remission without major adverse effects (2D)  To be given at an initial dose of 2.5-5 mg/kg/day in two divided doses, followed by dose adjustment according to monitoring of blood drug concentration • Not administered at the standard dose (4 mg/kg/day, maximum 150 mg/day) as it would be inadequately effective.
(Grade C1) To be administered at higher doses of 7-10 mg/kg/day once daily, with a peak blood mizoribine concentration (C2* 2 or C3* 3 ) ≥ 3.0 μg/mL, because of reported efficacy in preventing relapses  These key elements of the management of childhood SNSS included case definition, genetic testing, diet therapy, treatment of an initial episode of SSNS with steroids, treatment of relapsing SSNS with steroids, steroid therapy in FRNS and SDNS in children, treatment of FRNS and SDNS with steroid-sparing agents, renal biopsy, and vaccination in these children.
A set of strengths were noted in our work. First, we used a comprehensive search strategy to identify potentially relevant CPGs and performed quality assessment using the AGREE II tool by an expert specialized clinical team of pediatric nephrologists, general pediatricians, a clinical pharmacist, a renal pathologist, and a specialized nurse guided by two pediatricians with expertise in CPG methodologies, which adds a layer of strength to the AGREE II assessment. The results of this review can be used as a basis for CPG development or adaptation projects for the management of children with SSNS.
Furthermore, the results of our study propose the importance of including the AGREE II criteria in the capacity building of clinicians to guide their decisions in selecting CPGs for use in their daily practice.
Our study also has several limitations. First, some disadvantages of AGREE II have been addressed in the "AGREE-REX" (Recommendation EXcellence) tool, which addresses the clinical credibility of the CPG recommendations [31]. The selection of 70% as a cutoff point for standard domain ratings is another potential limitation as the original AGREE II does not mandate such a cutoff, but similar studies have also suggested so [22,23].
Other limitations, apart from those imposed by the AGREE II, include the following: (i) only English or Arabic language CPGs may have resulted in the exclusion of relevant CPGs intended for use in non-Englishspeaking and non-Arabic healthcare settings; (ii) this review mainly focused on CPGs for management of SSNS in children and did not evaluate other subcategories of nephrotic syndrome in children as it was out of the scope of this study.

Implications for practice: guidance for clinical guideline uptake
The adaptation of CPGs has been identified as a valid alternative to de novo development, which is a resource-extensive process [13]. Evidence-based practice initiatives in several countries in our region have opted to utilize CPG adaptation rather than de novo development [11,12]. Several CPG formal adaptation methodologies are presently available and could be further customized to local contexts [13]. Studies similar to our study could provide information on relevant CPG adaptation projects for the same health topics, especially for groups with little experience in using the AGREE II instrument.
This critical appraisal highlights the importance of quality assessment of CPGs by clinicians to ensure the transparency and strength of the CPG development process according to international CPG standards and support the best practice for children with SNSS. We recommend incorporating the AGREE II appraisal of CPGs in the capacity building of pediatricians and nephrologists.