Pathway for Enhanced Recovery After Spinal Surgery

A Systematic Review of Evidence for Use of Individual Components

Ana Licina; Andrew Silvers; Harry Laughlin; Jeremy Russell; Crispin Wan


BMC Anesthesiol. 2021;21(74) 

In This Article

Implications of This Study and Future Directions

We have identified, delineated and presented the evidence base for first comprehensive multimodal program for Enhanced Recovery in Spinal Surgery (ERSS). A continuous issue when discussing enhanced recovery protocols is that of contention as to which components have the highest clinical utility, accompanied by somewhat arbitrary decisions on incorporating different elements into the program. We identified a high level of evidence for administration of pre-emptive analgesia, peri-operative blood conservation (antifibrinolytic use), surgical site preparation and antibiotic prophylaxis. Although evidence base for cessation of smoking in surgery of the spine is low, there is translational high level evidence from other surgical specialties. In contrast with prior ERSS reviews we identified moderate evidence base for utilization of minimally invasive surgery and use of multimodal analgesia.[12] Although early mobilization and dietary libertization are considered critical in enhanced recovery, we identified a moderate level of evidence for institution of these interventions. Some clinical units may choose to use certain aspects of this proposed perioperative program as suited best to their unique location and practice pattern.

Evidence base is low in certain research areas. Most of the studies assessed were conducted outside the context of enhanced recovery program. This may have a negative bias effect, where the effect of an individual component may be higher than estimated when used within the ERSS pathway. Their combination with other components in a particular pathway is thought to have a synergistic effect. In addition to including studies focusing on individual ERSS elements, we evaluated studies focusing on bundles of care. Through the additive incremental value of each component, this may have a positive bias towards patient care outcomes. Full compliance with ERAS protocols has been identified to be an issue in prior studies. Compliance with ERAS pathways has been deemed to be a 5-year survival measure.[265] Overall compliance with ERAS protocols has been shown to be associated with better patient reported outcome measures.[266]

Methodologically heterogenous studies including systematic reviews and meta-analysis, randomized controlled trials, non-randomized controlled studies, and observational studies were eligible for this review. Hence, methodological heterogeneity of included studies rendered any quantitative effect estimates unreliable. In line with our planned protocol, we did not conduct a meta-analysis of studies in this patient population. Inherent clinical heterogeneity was present in this complex systematic review of a multi-pathway intervention. Clinical heterogeneity arose from variability in the participants, types and timing of outcome measures. Participants varied in their nature due to the type of baseline disease (e.g. ERSS for surgery on anatomically abnormal spine versus ERSS in cancer patient population). Participants also varied in inherent comorbidities e.g. young patients undergoing scoliosis surgery versus elderly with co-morbidities. These multilevel participant heterogenous characteristics were combined with varied baseline analgesic consumption. Types and timing of outcome measures were compliant with our pre-determined outcome groups. As anticipated, outcome measures were broadly different to be suitable for meta-analytic process. Forest plots were obtained for some pre-determined outcomes however the statistical heterogeneity together with baseline clinical differences made these measures inappropriate for interpretation. In line with our protocol, we performed a pre-planned thematic synthesis.

We have undertaken a number of steps to minimize the underlying meta-biases in this systematic review of complex intervention. We disseminated this protocol through open literature in order to give transparency to our research structure. We assessed the risk of bias in all individual studies. Furthermore, we graded the risk of bias across outcomes.[267] As we have identified 22 components of this pathway, some selection bias due to not identifying all eligible studies was possible. Publication bias across studies, where only data published through positive findings are disseminated poses a risk in any systematic review. Detection bias may have arisen due to problems with classification of exposure or outcomes.