Introduction

Postoperative pain after cardiothoracic surgery can be severe and attributed to a constellation of factors including rib retraction, sternal fracture, prolonged immobility, intercostal nerve stimulation and pleural irritation from chest tubes. Per current recommendations by the enhanced recovery after surgery society (ERAS), a consistent multimodal analgesic approach is essential to patient comfort, early mobility, avoidance of opioids and reduced likelihood of pulmonary complications following thoracic and cardiac surgery.1 These guidelines recommend regional anaesthesia, scheduled administration of acetaminophen and non-steroidal anti-inflammatory drugs (NSAIDs), ketamine and dexamethasone. Gabapentinoids also feature heavily in common multimodal anaesthetic pathways despite data suggesting limited effects following cardiothoracic procedures.2 Additionally, these guidelines may not provide sufficient guidance for patients with underlying liver or kidney disease, making it important to explore alternative options such as muscle relaxants.

Muscle relaxants as a class are commonly used to treat spasticity and musculoskeletal disorders, with well-documented effectiveness in chronic spastic musculoskeletal pain, multiple sclerosis and spinal cord injuries.3 4 Baclofen, tizanidine and dantrolene are approved to treat spasticity, though baclofen withdrawal can be life-threateningly severe.5 Baclofen and tizanidine are centrally acting and block GABAB receptors found in the spinal cord or brainstem,6 7 while dantrolene directly inhibits muscle contraction by reducing the release of calcium from the skeletal muscle sarcoplasmic reticulum.8 More pertinent to postoperative pain, agents such as carisoprodol, chlorzoxazone, cyclobenzaprine, metaxalone, methocarbamol and orphenadrine are approved for the treatment of musculoskeletal disorders.5 While used in the treatment of acute injuries, we hypothesise that there may be a role for use following cardiothoracic surgery.

This review aims to summarise the current literature on the effectiveness of non-paralytic muscle relaxants in postoperative pain management for cardiac and thoracic surgery patients. The aim is to provide and evidence basis for alternative analgesics beyond those offered by current guidelines.

Objectives

The objective of our study is to systematically review the literature for the use of non-paralytic skeletal muscle relaxants as adjuncts for non-opioid pain control following cardiac or thoracic surgery. This will include all human adult studies available in the literature for postoperative pain control.

Methods and designPopulation

The systematic review will focus on studies that include patients aged >18 years who have undergone cardiac or thoracic surgery.

Interventions

The interventions to be evaluated include any non-paralysing muscle relaxant as part of a pain control regimen for cardiac or thoracic surgery. This includes centrally acting muscle relaxants such as cyclobenzaprine and methocarbamol as well as antispastics and antispasmodics such as baclofen. This specifically excludes studies evaluating the role of paralytics in pain control.

Comparisons

We will compare adult patients undergoing cardiac and thoracic surgery and receiving standard multimodal analgesia (acetaminophen, gabapentin and others) with those receiving non-depolarising skeletal muscle relaxants.

Outcome

The primary outcomes required of included studies will be one of the following: qualitative and quantitative assessment for pain control, either through patient satisfaction Visual Analogue Scale or numeric rating scheme, days of pain control, time to pain control or measures of effect on functional status. Additional outcome measures may include sedation or other side effects and comparison in opioid usage. Secondary outcomes will also include effect on length of stay, incidence of falls, anticholinergic side effects and withdrawal episodes on removal of muscle relaxant in addition to any other reported adverse events.

Time

The outlined outcomes will be reported from the postoperative phase through to hospital discharge.

Study design

The systematic review and meta-analysis will include all therapeutic studies investigating the use of a non-paralysing muscle relaxant for pain control following cardiac or thoracic surgery within a week of surgery and following outcomes up until hospital discharge. This will include systematic reviews, meta-analyses, randomised control trials, adaptive clinical trials, prospective cohort and observational studies, non-randomised clinical trials, retrospective cohorts, case–control studies, case reports and case series. Studies that compare one intervention to usual practices recommended by ERAS as the control will be included. No minimum number of included subjects will be required. The review will exclude studies without a control group using either placebo or current standard care. Excluded study types include cross-sectional studies, editorials, preclinical or animal models and studies in the outpatient setting. Finally, grey literature shall be omitted from consideration. This review has been registered on the International Prospective Register for Systematic Reviews (PROSPERO) number CRD42023397917.

Search strategy

A three-step process will be used to identify eligible studies, including an initial search, title and abstract screening and full-text manuscript review. A professional systematic review librarian will provide guidance in developing the search criteria with the authors to include all relevant studies pertaining to adult, human studies non-paralysing muscle relaxants for pain control after thoracic and cardiac surgery. The databases that will be searched are PubMed, Web of Science, Cochrane Central and EMBASE from inception. No language restrictions will be applied. Figure 1 shows an example search algorithm for PubMed. Initial deduplication will be performed using EndNote (Clarivate Analytics, Philadelphia, Pennsylvania, USA).

Figure 1

Example search algorithm for PubMed.

Study selection

Literature search results will be uploaded from EndNote and screened through Covidence (Melbourne, Victoria, Australia). Study titles and abstracts will be screened for relevance in duplicate, blindly and independently, by the three junior authors (SK, QW and MF) and adjudicated by the senior author (TWM). Eligible studies will then be assessed again for inclusion and for quality in secondary screening through review of full-text manuscripts before data abstraction. This process will be reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram. The PRISMA-Protocol checklist pertaining to this protocol is available as online supplemental file 1. Any conflicting remarks regarding studies will be adjudicated through discussion before inclusion in the final analysis.

Quality assessment

Each article will undergo initial title and abstract screening in parallel by two independent, blinded reviewers to minimise bias. Conflicts will be adjudicated with discussion and involvement of the senior author as necessary. All selected articles will be reviewed with the senior author during full-text review. Cochrane tools for assessment of study quality such as risk of bias in non-randomised studies of interventions (ROBINS-1) and version 2 of the Cochrane risk-of-bias tool for randomised trials (RoB 2.0) will be used to determine appropriateness for inclusion. Two independent authors, one being the senior author, will assess the risks of bias in studies considered for full-text review in order to determine feasibility of a meta-analysis.

Data extraction

Quantitative data will be extracted from studies meeting inclusion on full-text review by a professional biostatistician. Data extracted for meta-analysis will be specifically those pertinent to the primary and secondary outcomes specified in the systematic review. This will include demographics of patient populations, types of surgical procedures performed, medications administered and outcome parameters as well as any data that are available across all included studies. Data extraction will be independently cross-checked by the senior author and discrepancies resolved through discussion with the biostatistician. The raw data for this review including a dataset of articles screened will be published in a data repository.

Endpoint

Results of the systematic review will be grouped by pharmacological agent. The primary outcomes will be quality of pain control as measured by patient satisfaction scores, Visual Analogue Scale, numeric ratings, days of pain control reported, intensity of pain reported, timing to effective pain control, sedating side effects, seizures or other adverse effects, effect on functional mobility and effect on opioid requirements. We will also include secondary outcomes of effect on length of stay, incidence of falls or injury due to muscle relaxants, anticholinergic reactions, withdrawal events on cessation. Any follow-up duration will be accepted as there is considerable variability in the existing literature.

Certainty of cumulative evidence

We will assess the certainty of evidence for each outcome using the Grading of Recommendations Assessment, Development and Evaluation framework in reporting the endpoints discussed above. As there may not be sufficient data for a meta-analysis, the overall certainty will be communicated using the narrative statements recommended by the framework.

Patient and public involvement

No patients were involved with the planning of this protocol.

AnalysisDescriptive analysis

A narrative synthesis of the final studies included will be written summarising the different non-paralysing muscle relaxants identified. The impact of each of these agents on the primary and secondary outcomes will be described in addition to a formal meta-analysis of studies using each pharmacological agent. Included studies will be evaluated for statistical heterogeneity before formal analysis.

Statistical analysis

The primary focus of this review is to detect evidence for the impact of the pharmacological agents identified on pain control and opioid usage. ‘Pain control’ is quantified in different ways that are important to patients including intensity, days of control, improvement from prior pain and whether pain limits functional mobility. As the effects of non-paralysing muscle relaxants are studied differently, we expect some limitation in the ability to directly compare one agent to standard protocols or to one another. However, when available, pharmacological agents will be compared as equitably as possible using all available outcome parameters reported in the primary literature. Subgroup analysis within populations of similar operations and treated with the same pharmacological agents will be performed to identify populations most likely to benefit from a given agent. We will use the risk ratio with 95% CI to express the effect estimate for dichotomous outcome and will express the effect estimate as mean difference with 95% CI for continuous outcomes. Analysis of covariance will be used for comparing clusters treated with agents and procedures as the baseline characteristics of groups in different studies are likely to be different.

Data synthesis

A PRISMA flow diagram will be used to summarise study selection. Results will be presented in accordance with the PRISMA statement. Tabulated data showing qualitative and quantitative pain control for each pharmacological agent and surgical intervention will be presented. For secondary outcome variables, we will present synthesised data as available in separate tables but will otherwise provide a separate narrative summary of the data available for each agent. We will cluster studies of each pharmacologic agent for each surgical type (cardiac or thoracic) based on the quality of evidence available and outcome variables included.

Meta-analysis

We anticipate clinical and methodological heterogeneity of the included studies and will plan for a qualitative synthesis of data in narrative form and to the extent possible, a quantitative synthesis of the pharmacological agents. The results of this meta-analysis will then be compared with current ERAS practice guidelines to provide context and compare relative efficacy.

Ethics and dissemination

No ethical or safety considerations were considered based on the nature of this review. Dissemination of findings through a peer-reviewed publication on the conclusion of the meta-analysis.