The present review protocol is being reported in accordance with the reporting guidelines, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocol (PRISMA-P) statement (see PRISMA-P checklist in Additional file 2) . This review protocol was registered within the International Prospective Register of Systematic Reviews (PROSPERO) (registration number CRD42019135289) .
The authors identified and defined the essential components of enhanced recovery within the area of spinal surgery. The authors performed this process by reviewing the current externally validated enhanced recovery protocols as recommended by the ERAS Society. We compared and contrasted the current published guidelines and reviews for gastrointestinal surgery, thoracic surgery, orthopedic surgery, and gynecological surgery [3,4,5,6,7,8, 12]. We identified and applied the relevant components to the field of spinal surgery. In addition, we compared and contrasted the technical elements present in the published prior reviews on enhanced recovery in spinal surgery [8, 11, 12]. A brief summary of this process has been provided in Additional file 1.
Our review protocol falls within the framework and definition of a complex systematic review due to the underlying intervention complexity and intertwined causal pathways. The planned review targets multiple groups of participants and will be applicable at multiple organizational levels and will require multifaceted adoption. It will need to be applied in a dynamic multidimensional environment. The planned review will therefore be performed according to the methodological standards for complex reviews [21,22,23,24,25,26,27,28].
Studies will be selected according to the following criteria below (and a summary in Table 2).
Types of studies
We will include systematic reviews and meta-analysis, randomized controlled trials, non-randomized controlled studies, and observational studies (e.g., cohort studies, case-control studies, cross-sectional studies, and case series).
We will include studies published in the English language. Due to the need for a contemporaneous review, studies published prior to 1990 will be excluded. Studies will not be excluded based on the publication status.
Types of participants
We will include all adult and pediatric population undergoing spinal surgical procedures (Table 2). We will include patients undergoing spinal surgical procedure on any anatomical site including cervical (anterior or posterior cervical decompression and fusion), thoracic (e.g., thoracic decompression and fusion), lumbar (e.g., lumbar decompression and fusion, lumbar laminectomy, lumbar micro-discectomy), sacral (e.g., scoliosis correction), or any combination of these.
Types of interventions
The interventions of interest have been classified in 5 perioperative pillars: preadmission period, preoperative period, intraoperative period, postoperative period, and audit and compliance processes. The interventions have been defined as the pre-determined components of the enhanced recovery pathway as applicable to spinal surgery (Table 1). The interventions may have been studied alone or in any one combination.
Types of outcome measures
One of the hallmarks of enhanced recovery reviews is the heterogeneity of outcomes due to the depth and breadth of the patient care received. Our primary outcomes will be all-cause mortality, morbidity outcomes (e.g., pulmonary, cardiac, renal, surgical complications), patient-reported outcomes and experiences (e.g., pain, quality of care experience), and health services outcomes (e.g., length of stay and costs) (see Table 3 and the “Outcomes and prioritization” section below).
Information sources and search strategy
We will search the following electronic databases (from 1990 onwards): MEDLINE via Ovid SP; EMBASE via Ovid SP; and Cochrane Library (Cochrane Database of Systematic Reviews and CENTRAL). In addition, the authors will search the grey literature through the following specific search engines: Google Scholar, OpenGrey, and GreyNet [29,30,31]. For the search strategy, we will combine keyword and subject headings in combination with validated filters in each of the pre-determined electronic databases . The draft search strategy specific to MEDLINE is included in Additional file 3.
Data selection and screening process
All articles identified from the literature search will be screened by two reviewers independently using an electronic screening form (Covidence web platform: http://www.COVIDENCE.org ). First, titles and abstracts of articles returned from initial searches will be screened based on the eligibility criteria outlined above. Second, full texts will be examined in detail and screened for eligibility. Third, references of all considered articles will be hand-searched to identify any relevant report missed in the search strategy. Any disagreements will be resolved by discussion to meet a consensus, if necessary.
A PRISMA flow diagram showing details of studies included and excluded at each stage of the study selection process will be provided.
Study information will be stored and managed using Endnote X9 throughout the review process.
A data extraction form will be designed and used to extract equivalent information from each study. Information of interest will include the following:
Study characteristics: study design, year of publication, journal, year (or period) of study conduct, sample size, setting, and other fields to capture data relevant to the assessment of study methodological quality (see the “Risk of bias assessment” subsection).
Participant characteristics: population sampled, age (e.g., mean with standard deviation, range) and gender (e.g., percentage of female participants), type of spinal surgery.
Intervention (individual components of the pathway) and comparator characteristics and definitions.
Outcome results: definitions and assessment tools.
Data extraction forms will be piloted initially on a small number of included studies. Subsequently, each of the included studies will be abstracted by two team members, independently, and potential conflicts will be resolved through discussion. Authors of primary publications will be contacted for data clarifications or missing outcome data, as necessary.
Outcomes and prioritization
We defined our primary outcomes in terms of the following groups:
Mortality from all causes.
Morbidity, including pulmonary, cardiac, and renal complication rates; surgical complication rates; and readmission rates.
Patient-reported experiences and outcomes (PREMs/PROMs), including pain-related outcomes (pain score rating, pain management satisfaction), readiness for surgery, quality of care patient scores, and quality of recovery outcomes.
Health service-related outcomes, including length of stay and reported economic/financial outcomes (e.g., costs of the length of stay).
Our chosen outcome measures although grouped in several primary headings are broad. This is due to the anticipated heterogeneity of the reporting for the pre-specified components of the pathway. Our approach to the outcome definition and synthesis is in line with other reported enhanced recovery specialty pathways. Outcomes analyzed in previous reviews consisted of a length of stay, readmission, complication rate, and patient satisfaction measures [3, 4, 7,8,9,10, 32].
Risk of bias in individual studies
Risk of bias in randomized controlled studies will be assessed using the Cochrane Risk of Bias tool . This tool considers several domains of bias: randomization, allocation concealment, blinding, accounting of patients and outcome events, and selective outcome reporting bias . Risk of bias appraisal in non-randomized and observational studies can introduce difficulty due to a lack of a generic robust bias assessment tool . We will use the ROBINS-I (risk of bias in non-randomized studies of interventions) to assess the risk of bias in non-randomized studies . ROBINS-I is the preferred tool to be used in Cochrane Reviews for non-randomized studies of interventions. The authors of the GRADE working group have recently issued recommendations on how the ROBINS-I tool should be used within the context of the GRADE guidelines . In order to assess the risk of bias in systematic reviews, we will use the revised AMSTAR-2 tool . We will use the risk of bias assessment in individual studies to inform our assessment of study limitations across the body of evidence.
We will perform qualitative (narrative) synthesis of included studies. We will perform this process through thematic synthesis for each proposed component of the complex intervention pathway . We plan to analyze, synthesize, and grade the quality of the body of evidence across outcomes with regard to individual pathway components . In the first stage of the synthesis, we will systematically describe the data obtained from each study and its findings. In the second stage, we will group the obtained data under the relevant outcome and individual pathway element. The endpoint of the GRADE evidence summary will consist of the Evidence Profile (EP) . This thematic narrative synthesis will not provide a single estimate measure for each outcome of each pathway component. Summary of findings table will not be constructed . Narrative analysis of each pathway component will present the summarized data used to inform the synthesis in the form of a forest plot . Forest plot will be used in order to summarize data for each relevant outcome of each pathway component. Finally, we will perform a narrative analysis of the obtained data pertaining to each pathway component. For each component of the pathway, we will assess clinical diversity across studies by examining the variability in the intervention used, the type of surgery and the participant group. We have chosen this method in order to answer the key question of “what happens when the complex intervention is implemented?”. This method of evidence synthesis is in line with other published enhanced recovery reviews [3, 5, 7, 15, 16, 41, 42]. Meta-analyses will not be conducted due to the anticipated inherent methodological heterogeneity of the studies with regard to design, populations, procedures, and outcomes . As such, it is unlikely that quantitative synthesis would be meaningful in this review of complex intervention .
Selective reporting of outcomes within studies will be reviewed by comparison with published protocols, review of study registration information on appropriate sources (e.g., trial registration websites, PROSPERO for systematic reviews) .
Confidence in cumulative evidence
We will use the GRADE Handbook to guide the process of rating the evidence. In the context of a systematic review, the ratings of the quality of evidence reflect the extent of our confidence that the estimates of the effect are correct . Quality of evidence will be classified according to the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) system into one of four categories: high, moderate, low, and very low . Grading of the quality of evidence according to this system will be performed for pertinent study outcomes classified by the individual elements of the pathway as outlined in Table 1. A recommended strategy for systematic review authors is to use this approach in order to rate the quality of evidence for each outcome across studies, i.e., for a body of evidence [38, 39]. Quality of the body of evidence will be assessed according to study limitations, imprecision, the inconsistency of results, indirectness of evidence, and publication bias . Evidence based on randomized controlled trials will be considered as high quality unless confidence in the evidence is decreased due to study limitations, the inconsistency of results, indirectness of evidence, imprecision, and reporting biases. Observational studies will be considered low quality; however, they may be graded higher if the treatment effect observed is very large or if there is evidence of a dose-response relationship [39, 40]. Quality of evidence can be seen as occurring in a continuum; therefore, any discrete categorization of continuous evidence base does involve a degree of arbitrariness. As we are undertaking a systematic review, rather than developing a guideline, we will not be making recommendations on the utility of the pathway components [27, 40].