Interventions to Modify Psychological Processes in Patients Undergoing Spine Surgery: A Systematic Review

Abstract

Objectives

Among patients undergoing elective spine surgery, psychological processes such as kinesiophobia and pain catastrophizing are associated with postoperative disability and poor quality of life. These represent risk factors which could be modified to improve surgical outcomes. We reviewed perioperative interventions to modify psychological processes and their effects on psychological and surgery-related outcomes.

Methods

We searched MEDLINE, EMBASE, and Cochrane databases for studies examining any interventions for modifying psychological processes in adult patients undergoing spine surgery. Two reviewers screened studies for eligibility, extracted data in duplicate, and performed risk of bias assessments. Outcomes included pain, disability, quality of life, kinesiophobia, self-efficacy, and pain catastrophizing.

Results

368 titles and abstracts were retrieved, of which 27 studies underwent full-text screening. We included 12 studies which reported on 1263 patients. Eight were randomized controlled trials. Interventions included preoperative and postoperative cognitive behavioural therapy, cognitive-behavioural-based physical therapy, a web-based interactive platform, an information booklet, and music therapy. The psychological and surgery-related outcomes of interventions were variable, with some studies reporting significant benefits and others reporting no differences between groups. The greatest potential benefits were found in studies of post-operative cognitive-behavioural-based physical therapy. Risk of bias among studies was high due primarily to lack of blinding and limited standardization of interventions.

Conclusions

Several interventions to potentially modify psychological processes in patients undergoing spine surgery have been reported. Post-operative cognitive-behavioural-based physical therapy might be associated with improved outcomes, but confidence is limited by inconsistency, risk of bias, and limited long-term follow-up.

Background

For many patients living with debilitating spine conditions, surgery can offer substantial improvements in pain, function, and quality of life. However, some patients also have comorbid psychological processes such as anxiety, depression, pain catastrophizing, and kinesiophobia (an excessive, irrational, and debilitating fear of physical movement and activity ¹ ), which have been associated with worse outcomes and less satisfaction^2-10 even among patients who experience clinical improvement. ¹¹ As such, these psychological processes may represent modifiable risk factors that could potentially be altered to improve patient care.

Interventions to modify psychological processes include cognitive behavioral therapy, specialized physical therapy, music therapy, biofeedback, mindfulness, and meditation, ¹² but the potential effects of these interventions for patients undergoing spine surgery are poorly understood. Although there is evidence to support various psychological interventions as adjuncts in the treatment of chronic spinal pain,^13,14 studies of their implementation and effectiveness in the context of surgery have shown mixed results. Two recent meta-analyses^15,16 suggested no benefit for preoperative “prehabilitation” programs on functional or psychological outcomes in spine surgery, and one meta-analysis of perioperative cognitive behavioural therapy (CBT) ¹⁷ suggested a significant effect on postoperative quality of life, psychological outcomes, disability, and pain despite substantial heterogeneity between studies.

In order to inform patient care and guide further research in this field, we performed a systematic review to answer the following questions:

1. Among patients undergoing elective spine surgery, what interventions are used to modify preoperative psychological processes?

2. What are the psychological and spine-related outcomes of these interventions?

Methods

Selection of Studies

We performed a systematic review according to the methodology outlined in the Cochrane Handbook for Systematic Reviews. MEDLINE, EMBASE and Cochrane Databases were searched using MESH and EMTREE headings and subheadings in various combinations, supplemented with free text (Appendix 1). All databases were searched from inception to March 18, 2024. Two reviewers (AA and MG) performed the initial title and abstract screen of retrieved studies, full manuscript review, and data abstraction from the final list of included studies. In cases of disagreement, a third author (NE) screened the studies with discrepancies resolved by consensus.

Eligibility

We included randomized controlled trials, prospective and retrospective cohort studies, and prospective case series that were available in English. We excluded case studies, review articles, and conference abstracts. Patient populations were limited to adults over 18 years of age undergoing spine surgery for an indication at any spinal level. Studies were eligible if psychological processes were measured pre- and postoperatively using a standardized scale or questionnaire and a preoperative or postoperative intervention to modify psychological processes was evaluated. Data abstracted from full-text review included type of surgery; type, duration, and protocol of intervention implemented; administrator of intervention and their training level; time points for outcome evaluation; and outcomes evaluated.

Outcomes

Outcome data abstracted included all psychological and/or surgery-related outcomes, which ultimately included: pain catastrophizing, kinesiophobia, fear avoidance beliefs, self-efficacy, perceived control over pain and ability to decrease pain, anxiety and depression, global rating of mental health; disability, axial and extremity pain, quality of life, physical activity, tests of physical performance and range of motion, and return to work and activities of daily living.

Risk of Bias Assessments

Included studies were evaluated for Risk of Bias using the RTI Item Bank on Risk of Bias and Precision of Observational Studies ¹⁸ or the Revised Cochrane Risk-of-Bias Tool for Randomized Trials ¹⁹ as applicable.

Synthesis

Data abstracted from each study were compiled and interpreted descriptively. A meta-analysis was not planned or performed due to the anticipated and observed substantial heterogeneity of study design and outcomes among the included studies.

Results

Study Selection

The study selection process is summarized in Figure 1. We identified 368 potentially relevant studies, of which 27 were selected for full-text review after screening of titles and abstracts. Screening of reference lists from these articles as well as prior systematic reviews identified 3 more potentially relevant studies. In total, 12 studies were included. Study designs, timing of interventions, patient populations, and interventions are summarized in Figure 2.

Figure 1.

Study selection.

Figure 2.

Duration of interventions (green) and follow-up (violet).

Study Characteristics

The 12 included studies reported on 1263 patients. Three studies^20-22 investigated preoperative interventions and nine^23-31 postoperative interventions. All 3 studies of preoperative interventions and five studies of postoperative interventions^25,28-31 (67%) were conducted as randomized controlled trials. Other study designs included prospective cohort studies,^24,27 a prospective case series, ²³ and a mixed methods study. ²⁶ Study populations included patients undergoing lumbar fusion (seven studies, N = 778), lumbar decompression with or without fusion (one study, N = 86), lumbar discectomy (one study, N = 129), anterior cervical decompression and fusion (two studies, N = 171), posterior cervical decompression (one study, N = 39), and anterior, posterior, or combined spinal fusion not otherwise specified (one study, N = 60).

Nine studies (75%) reported on cognitive behavioural therapy (CBT)-based interventions^{20,21,23,25,27-31} including cognitive behavioural-based physical therapy (CBPT, seven studies^{21,23,25,27-29,31}). Two (17%) reported on educational platforms including a printed information booklet ²⁴ and a web-based platform. ²² One (8%) studied music therapy. ²⁶ The duration of follow-up varied: all twelve studies reported outcomes at short-term follow-up (<3 months), five^{20,21,23,25,31} at intermediate follow-up (≥3 months, <12 months) and three^20,29,31 at long-term follow-up (≥12 months). Only one study reported outcomes at each of short-, intermediate-, and long-term follow-up. ³¹ The duration and intensity of interventions varied from self-directed educational materials^22,24 to brief sessions on postoperative days 0-4^26,30 to a CBPT course lasting up to 20 weeks postoperatively. ²⁵ Four studies (33%) had therapy courses lasting 4-6 weeks.^23,27-29 The timing of interventions and follow-up in included studies are summarized in Table 1.

Table 1.

Study characteristics.

Effects of Interventions

Results in the included studies are summarized in Table 2. Eleven of twelve studies (92%) reported at least one statistically significant result in favour of the psychological intervention studied. Eight studies^{20,23-25,27-29,31} (67%) reported positive results for one or more psychological outcome and seven (58%) for surgery-related outcomes including disability or pain.^{21,23,26,28-31} Four studies^23,28,29,31 (33%) reported positive results in both psychological and surgery-related outcomes associated with psychological intervention.

Table 2.

Study characteristics, results, and risk of bias assessments for 12 included studies.

Reference	Study Type	Population, Sample Size	Psychological Assessment	Intervention and Protocol Description	Outcomes and Results	Risk of Bias Assessment
Pre-operative interventions
Rolving et al, 2015	Randomized controlled trial	• DDD or spondylolisthesis undergoing lumbar fusion • N = 90	• Fear avoidance beliefs questionnaire • Catastrophizing subscale of coping strategies questionnaire	Cognitive behavioural therapy • 6 3-hour group sessions (4 preoperative and 2 postoperative) • Managing team involved a psychologist, OT, PT, social work, spine surgeon, and previously operated patient • All staff attended a 2-day training program to develop basic cognitive behavioural skills	Oswestry disability index • Trend to better results, but no difference vs control during first year Fear avoidance beliefs questionnaire • Improved vs control at 6 months (P = 0.01) • No difference vs control at 1 year (P= 0.52) Catastrophizing subscale of coping strategies questionnaire • Improved vs control at 6 months (P = 0.04) • No difference vs control at 1 year (P = 0.51) Return to work • No difference during first year Low back pain rating scale • No difference during first year	High
Lotzke et al, 2019	Randomized controlled trial	• Disc herniation, foraminal stenosis, spondylolisthesis undergoing lumbar fusion • N = 118	• Pain catastrophizing scale • Tampa scale for kinesiophobia • Self-efficacy for exercise scale • Hospital anxiety and depression scale	Modified cognitive behavioural fear-avoidance model • 4 1-hour preoperative sessions (beginning 8-12 weeks preoperatively) and 1 30-minute session by telephone 2 weeks after surgery • Administered by a physical therapist with >10 years of experience with patients with low back pain and a graduate diploma in cognitive and behavioral psychotherapy	Oswestry disability index • Both groups reached minimal important change (MIC) at 8 weeks postoperatively • No difference vs control up to 6 months Back and leg pain VAS • Both groups reached MIC at 8 weeks postoperatively • No difference vs control up to 6 months Pain catastrophizing scale • Both groups reached MIC at 8 weeks postoperatively • No difference vs control up to 6 months Tampa scale for kinesiophobia • No difference vs control up to 6 months Self-efficacy for exercise scale • No difference vs control up to 6 months Hospital anxiety and depression scale • Both groups reached MIC at 8 weeks postoperatively • No difference vs control up to 6 months EQ-5D index and EQ-5D VAS • Significant group and time interaction effect (P = 0.01) • Both groups reached MIC at 8 weeks postoperatively Patient-specific functioning scale • Intervention group reached MIC at 3 months vs 6 months for control; trend to favor intervention at each follow-up • No difference vs control at 6 months Physical activity • No difference vs control up to 6 months, though trend to favor intervention group with respect to intensity, steps per day and one leg stand test	High
Strom et al, 2019	Randomized controlled trial	• DDD or spondylolisthesis undergoing lumbar fusion • N = 114	• Hospital anxiety and depression scale	A web-based spine platform featuring interaction and information by animation • Included a 15-minute baseline introduction followed by online access anytime • Tablets provided if needed • Included animated information regarding the course of treatment and postsurgical rehabilitation, access to an internet support group, and a diary for tracking pain, analgesia, and physical activity	Hospital anxiety and depression scale • No difference vs control up to 3 months Oswestry disability index • No difference vs control up to 3 months Quality of life (EQ-5D-5L) • No difference vs control up to 3 months Low back pain rating scale • No difference vs control up to 3 months	High
Post-operative interventions
Abbott et al, 2010	Randomized controlled trial	• Spinal stenosis, spondylosis, spondylolisthesis, or DDD undergoing lumbar fusion with or without decompression • N = 107	• Mental health subscale of medical outcome study short form 36 (SF-36) • Self-efficacy scale • Tampa scale for kinesiophobia • Catastrophizing, control over pain, and ability to decrease pain subscales of coping strategy questionnaire (CSQ-CAT, CSQ-COP, CSQ-ADP)	Psychomotor therapy program • 12-week program with combination of home and outpatient therapy sessions • Physical therapy combined with cognitive-behavioural strategies including patient education, motivational discussion, self-monitoring • Control group underwent 12-week home exercise program without cognitive-behavioural strategies • Administered by 3 physiotherapists experienced in cognitive-behavioural methods	Oswestry disability index • Improved at 3 months with effect sustained up to 2-3 years (P = 0.01) Quality of life (EQ-5D) • Trend towards improvement at 3 months but no significant difference Mental health (SF-36) • No difference vs control up to 2-3 years Self-efficacy scale • Improved at 3 months up to 2-3 years (P < 0.01) Back beliefs questionnaire • Improved at 3 months up to 2-3 years (P = 0.02) Tampa scale for kinesiophobia • Improved at 3 months up to 2-3 years (P < 0.01) Catastrophizing (CSQ-CAT) • Improved at 6 months (P = 0.02) and 2-3 years (P = 0.01) but not other time points or with repeated measures analysis of covariance Control over pain (CSQ-COP) • Improved at 3 months up to 1 year (P = 0.01) Ability to decrease pain (CSQ-ADP) • Improved at 3 months up to 1 year (P < 0.01) Back pain VAS • Improved at 3 months up to 6 months (P = 0.01)	High
Reichart et al, 2012	Randomized controlled trial	• Degenerative spinal disease with spinal stenosis and instability, undergoing posterior lumbar interbody fusion • N = 39	• Fear avoidance beliefs questionnaire	Short psychological intervention • Carried out by psychologist trained in behavioral therapy • 1 session 30 min preop; involved strategy for reducing fear and fear-avoidance beliefs by fostering motivation based on the concept of mental contrasting • 2nd session 30 min 2-4 days postop; meant to reduce fear-avoidance beliefs by problem solving and setting resolutions, using the concept of “implementation intentions”	German pain questionnaire • No difference vs control for mean pain intensity • Significantly lower highest pain intensity in intervention group (P = 0.031) at 6 weeks Fear-avoidance beliefs questionnaire • No difference vs control Hannover back function questionnaire (physical fitness) • Significantly improved fitness in intervention group (P = 0.05) at 6 weeks	Low
Monticone et al, 2014	Randomized controlled trial	• Spondylolisthesis and/or lumbar spinal stenosis and/or sciatica undergoing lumbar fusion with or without decompression • N = 130	• Tampa scale for kinesiophobia • Pain catastrophizing scale	Exercises plus cognitive behavioural therapy • Individual 60-minute inpatient cognitive behaviour therapy sessions twice a week for 4 weeks • 90-minute exercise sessions five times a week for 4 weeks • Program involved 2 physiatrists, a psychologist, and 4 physiotherapists	Oswestry disability index • Intervention group had a greater treatment effect up to 1 year, and a significant group × time interaction effect (P < 0.001) Tampa scale for kinesiophobia • Significant effect of time, group and interaction in favour of intervention group (P < 0.001) Pain catastrophizing scale • Significant effect of time, group and interaction in favour of intervention group (P < 0.001) Pain numerical rating scale • Significant effect of time, group and interaction in favour of intervention group (P < 0.001) Short-form health survey • Significant effect of time, group and interaction in favour of intervention group (P < 0.001)	High
Archer et al, 2016	Randomized controlled trial	• Lumbar degenerative condition undergoing a laminectomy with or without arthrodesis • Tampa scale for kinesiophobia score >38 (high fear of movement) • N = 86	• Tampa scale for kinesiophobia • 9-Item patient health questionnaire for depressive symptoms • Pain self-efficacy questionnaire	Cognitive behavioural-based physical therapy • Initial 60-minute session in-person, followed by weekly 30-minute sessions over the telephone for 6 weeks • Sessions led by a physical therapist • A clinical psychologist and physical therapist rated the sessions for treatment integrity	Brief pain inventory for back and leg pain • Improvement at 3 months in intervention group vs control (P = 0.007) Oswestry disability index • Improvement at 3 months in intervention group vs control (P < 0.001) SF-12 physical component score • Improvement at 3 months in intervention group vs control (P = 0.001) SF-12 mental component score • Improvements post-treatment and at 3 months in intervention group vs control (P < 0.001) Performance based physical tests: 5-Chair stand, timed up and go, 10-meter walk • No difference between intervention and control at 3 months Tampa scale for kinesiophobia • Improvement at 3 months in intervention group vs control (P = 0.009) Pain self-efficacy questionnaire • Improvement at 3 months in intervention group vs control (P < 0.001)	Low
Monticone et al, 2017	Prospective population-based cohort study	• Lumbar fusion with or without a decompression for chronic low back pain, spondylolisthesis, and/or lumbar stenosis • N = 180	• Tampa scale for kinesiophobia	Exercises plus cognitive behavioural therapy • Individual 60-minute inpatient cognitive behavioural sessions twice a week for 4 weeks • 90-minute exercise sessions five times a week for 4 weeks • Program involved 2 physiatrists, a psychologist, and 4 physiotherapists Note – identical protocol but different study population as Monticone et al, 2014	Tampa scale for kinesiophobia • Improvements observed after treatment at 4 weeks (mean score 34.1+/−7.0 pre-treatment; 21.6+/−7.7 post-treatment) • Total score was responsive to treatment (ROC AUC 0.999) • 2 subscales (harm and activity avoidance) were equally responsive to treatment Note - Aim of study was to evaluate responsiveness and MCID for Tampa Scale for Kinesiophobia after lumbar fusion	Some concerns
Claus et al, 2017	Prospective, controlled, time-series study	• Patients undergoing first-time lumbar discectomy • N = 129	• Fear-avoidance beliefs questionnaire • Back beliefs questionnaire (higher score suggest better coping abilities)	Evidence-based information booklet titled “your back operation” given to patients • Promotes a proactive approach to patient self-management • Key messages: Restrictions may delay recovery and return to work, “let pain be your guide” is a helpful approach, early rehabilitation improves function in activities of daily living, and early return to work may produce faster recovery and better outcomes	Quebec back-pain disability scale • No difference at 2 months between groups Return to work and daily living activity • No difference at 2 months between groups Leg and back pain VAS • No difference at 2 months between groups Fear-avoidance beliefs questionnaire • Improved physical activity score in intervention group at 2 months (P = 0.008) Back beliefs questionnaire • Improvement after reading booklet in intervention group (P = 0.0006) • No difference at 2 months between groups	Some concerns
Mondanaro et al, 2017	Mixed-methods study	• Anterior, posterior, or combined spinal fusion • N = 60	• Hospital anxiety and depression scale	Music therapy • 1 30-minute music therapy session within 72 hours after surgery • Included patient-preferred live music, singing, rhythmic drumming, breathwork and/or visualization supporting tension release and relaxation • Psychoeducation about mind-body awareness explained by the therapist • Included 2 licensed, board-certified music therapists and masters-level music therapy interns	Pain VAS • Improvement in intervention group after therapy (P = 0.01) Hospital anxiety and depression scale • No difference between groups after therapy Tampa scale for kinesiophobia • No difference between groups after therapy	High
Wibault et al, 2017	Randomized controlled trial	• Patients with cervical radiculopathy scheduled for anterior or posterior cervical surgery • N = 202	• Catastrophizing subscale of the coping strategy questionnaire (CSQ-CAT) • Self-efficacy scale	Neck-specific exercises with a behavioural approach • Weekly sessions with a physiotherapist for 6 weeks, then twice weekly sessions for a maximum of 20 weeks • Sessions combined exercise therapy for neck, trunk and scapula muscles with a cognitive behavioural approach to pain and stress management	Neck muscle endurance and cervical range of motion • No difference between groups at 6 months Self-efficacy scale • No difference between groups at 6 months Pain catastrophizing (CSQ-CAT) • Patients with >50% attendance at treatment sessions had larger improvements at 6 months vs control (P = 0.04) Perceived control over pain • No difference between groups at 6 months	High
Coronado et al, 2021	Prospective case series	• Patients undergoing anterior cervical discectomy and fusion for cervical radiculopathy, myelopathy or myeloradiculopathy • N = 8	• Tampa scale for kinesiophobia • Pain catastrophizing scale • Pain self-efficacy questionnaire	Early telephone-supported home exercise program with cognitive-behavioural approach • Provided activity instruction for neck/shoulder/core strengthening and range of motion, daily walking, deep breathing and distraction/relaxation techniques • Patients were mailed a manual with instructions for each activity and exercise, and instructed to perform them daily for 6 weeks • Supported by weekly phone calls with a physical therapist	Neck disability index • 75% demonstrated a clinically relevant reduction at 6 weeks; 88% at 6 months Numeric rating scale for neck and arm pain intensity • 88% demonstrated a clinically relevant reduction in arm pain at 6 weeks and 6 months • 88% demonstrated a clinically relevant reduction in neck pain at 6 weeks; 75% at 6 months Tampa scale for kinesiophobia • 75% demonstrated a clinically relevant reduction at 6 weeks; 50% at 6 months Pain catastrophizing scale • 88% demonstrated a clinically relevant reduction at 6 weeks; 100% at 6 months Pain self-efficacy questionnaire • 75% demonstrated a clinically relevant reduction at 6 weeks and at 6 months Average steps per day • 29% recorded an increase in average steps per day by 6 weeks; 43% by 6 months	High

AUC = area under curve, DDD = degenerative disc disease, MIC = minimally important change, OT = occupational therapist, PT = physical therapist, QOL = quality of life, ROC = receiver operating characteristic, SF-12 = 12-Item Short Form Survey, SF-36 = 36-Item Short Form Survey, VAS = visual analogue scale.

Of the 9 studies of cognitive-behavioural-based interventions, seven^{20,23,25,27-29,31} (78%) reported a significant improvement in psychological outcomes and six^21,23,28-31 (67%) in surgery-related outcomes. All four studies which reported positive results in both domains used postoperative cognitive-behavioural physical therapy techniques. None of the preoperative interventions reported a significant benefit for postoperative pain or disability. The reported psychological ²⁰ and quality of life ²¹ effects from preoperative interventions were small and transient, suggesting a potential benefit in the immediate postoperative period with uncertain clinical significance.

Psychological Outcomes

The most commonly reported psychological processes were kinesiophobia, impaired self-efficacy, and pain catastrophizing. The number of studies reporting significant results for these processes are summarized in Figure 3. Seven^{20,23,24,27-29,31} of nine studies reporting on kinesiophobia used the Tampa Scale for Kinesiophobia or Fear Avoidance Beliefs Questionnaire and reported better postoperative scores in the intervention group, as did three^23,28,31 of five studies that used a self-efficacy scale (Pain Self-Efficacy Questionnaire, Self-Efficacy for Exercise Scale, Swedish Self-Efficacy Scale). Four^20,23,25,29 of six studies that measured pain catastrophizing (Pain Catastrophizing Scale or catastrophizing subscale of Coping Strategies Questionnaire (CSQ-CAT)) reported better scores with the intervention. Differences in kinesiophobia, self-efficacy, and catastrophizing were observed at early follow-up in six studies,^{23,24,27-29,31}from as early as 4 weeks with effects sustained at long-term follow-up in two of 3 studies.^29,31 Other studied psychological processes and outcomes included the Hospital Anxiety and Depression Scale and the SF-36 and SF-12 Mental Health scores.

Figure 3.

Significant results for psychological outcomes by study.

Surgery-Related Outcomes

Six^23,26,28-31 of ten studies reported on pain outcomes (Visual Analogue Scale, Numerical Rating Scale, Brief Pain Inventory, or German Pain Questionnaire) and found a significant difference favouring the intervention group, as did five^23,28-31 of eight studies which reported disability outcomes (Oswestry Disability Index, Neck Disability Index, Hannover Back Function Questionnaire). The numbers of studies reporting significant results for these traits are summarized in Figure 4. Pain outcomes were improved in the intervention group at short-term (<3 months) follow-up in six studies^23,26,28-31 with sustained effects seen at 6 months in two out of two studies^23,31 reporting pain outcomes at this timepoint and 12 months in one study ²⁹ out of two. For disability, an effect was seen at short-term follow-up in six studies,^21,23,28-31 up to 6 months in one study, ²³ and at long-term (≥12 months) follow-up in two studies.^29,31 One of 3 studies which reported on quality of life (EQ-5D/5L indices) found a significant improvement up to 8 weeks. ²¹

Figure 4.

Significant results for surgery-related outcomes by study.

Risk of Bias

Detailed bias assessments are reported in Appendix 2 and are summarized in Table 2. The risk of bias among studies was generally high due to lack of blinding of patients, intervention providers, and assessors, as well as limited adherence to and standardization of interventions.

Discussion

We performed a systematic review to identify interventions to modify psychological processes among patients undergoing spine surgery. We found that several interventions have been reported, but that post-operative cognitive behavioural therapy combined with exercise-based physical therapy might be most associated with improvements in kinesiophobia and pain catastrophizing, as well as postoperative pain and disability. Confidence in these findings is limited by inconsistency, heterogeneity, imprecision, risk of bias, and limited long-term follow-up.

Relation to Prior Literature

This study adds substantially to a small but growing body of literature about perioperative psychological interventions for patients undergoing spine surgery. In contrast to prior reviews on this topic, we report on both psychological and surgery-related outcomes, and we compared multiple pre- and postoperative intervention modalities. A recent systematic review by Scarone et al ³² studied the effect of perioperative psychological interventions on persisting spinal pain and secondary surgical outcomes including disability and quality of life in a more restricted population of patients undergoing fusion procedures. That review did not report psychological outcomes but similarly found that perioperative cognitive behavioural-based interventions may be associated with improved pain, disability, and quality of life postoperatively. A recent meta-analysis by Parrish et al ¹⁷ of RCTs on CBT-based interventions in lumbar spine surgery analyzed seven studies of 531 patients with generally more restrictive inclusion criteria excluding observational studies and non-CBT interventions. Using a sensitivity analysis, significant positive effects were identified for psychological outcomes (standard mean difference, SMD = 0.61), quality of life (SMD = 0.55), and to a lesser degree, pain and disability. On subgroup analysis of psychological outcomes, the largest effect size was observed for kinesiophobia (SMD = 0.67). This review quantitatively corroborates our findings in a more limited sample of studies amenable to meta-analysis. Similar heterogeneity was nonetheless observed between studies in this work.

Our findings suggest increased benefits from postoperative interventions rather than preoperative or “prehabilitation” strategies, which is supported by a pair of recent meta-analyses finding no significant effect of preoperative CBT on postoperative outcomes.^15,16 Some studies included in similar reviews, such as two RCTs analyzed in Parrish et al,^33,34 were excluded here because psychological processes were not measured preoperatively or post-operatively. Despite this finding, psychotherapeutic interventions have a strong body of evidence for treating specified comorbid diagnoses associated with poor surgical outcomes including substance use disorders, eating disorders, anxiety, and depression. It follows that psychological prehabilitation strategies could still have a clinical role in high-risk patients, ¹⁶ but further work in future studies including preoperative risk stratification will be necessary to define their potential benefit and optimal target population.

Our assessment of the risk of bias among included studies differed from both Scarone et al and Parrish et al, with seven of eight RCTs graded as having a “high” risk of bias in the present work using the Cochrane RoB 2 tool ¹⁹ (Appendix 2). This discrepancy is attributable to the use of different bias assessment tools, and specifically the importance given to lack of blinding of patients, providers, and assessors, but we acknowledge that blinding is not possible for many psychological interventions. ³⁵

Outside of spine surgery, the observed effects of perioperative psychological interventions have been inconsistent. In elective orthopedic surgery, meta-analyses have suggested some improvements in postoperative pain, anxiety, quality of life, and disability, but results have varied between studies and even between similarly conducted reviews.^36-39 Both pre- and postoperative interventions have shown variable benefit for modifying psychological processes, but postoperative interventions may have a stronger impact on postoperative pain.^36,37 While cognitive behavioural-based methods have been the most widely studied perioperative psychological interventions, there is also evidence across surgical disciplines to support a variety of techniques less commonly cited in the spine literature, including relaxation, guided imagery, hypnosis-based interventions, and digital psychoeducation platforms.^40-42 Our review retrieved an insufficient sample of interventions besides cognitive behavioural-based methods to draw clear conclusions on other specific techniques.

Implications

The impact of psychological processes on postoperative outcomes in spine surgery is well-known,^2-11 but strategies to screen for and potentially intervene on patients at risk may be underutilized in clinical practice. The results of our review suggest that postoperative cognitive behavioural strategies combined with exercise-based physical therapy (CBPT) might be considered for patients undergoing elective spine surgery, but further research is warranted to confirm these findings and increase confidence when considering application to clinical care. While a survey of US spine surgeons ⁴³ demonstrated strong beliefs regarding the impact of psychological factors on pain relief, adherence to therapy, and return to work following surgery, approximately only one third of these surgeons employed presurgical psychological screening despite its validation as a predictor of postoperative outcomes. ⁴⁴ The extent of screening, tools used, and effects seen were variable among survey respondents.

The large and growing surgical and non-surgical patient volume at many spine surgery centres suggests that indiscriminate application of perioperative psychological interventions could be clinically and economically unfeasible, but routine psychological screening with validated questionnaires to identify candidates who could benefit from such interventions might be a reasonable approach. ⁴⁵ While not the primary focus of this review, we found that the Tampa Scale of Kinesiophobia ⁴⁶ was the most widely used instrument in the included studies. Perioperative kinesiophobia has been shown to be modifiable with CBT-based interventions in spine surgery ¹⁷ and has been previously validated as a predictor of postoperative outcomes.^7,8,47

Limitations

It is not possible to draw strong conclusions from this review due to the limitations and small volume of the included studies. There was substantial inconsistency of the effects seen in both psychological and clinical outcomes between studies with similar research protocols and patient populations. Risk of bias further limits the interpretation and utility of these results when considering the clinical and economic effectiveness of implementing specific intervention strategies. This study was also limited by the methodological heterogeneity of study design and intervention protocols which precluded a quantitative meta-analysis.

Conclusion

Several interventions to potentially modify psychological processes associated with poor outcomes in patients undergoing elective spine surgery have been reported. Postoperative cognitive behavioural therapy combined with exercise-based physical therapy may be associated with improved psychological and clinical outcomes, but interpretation of the literature is limited by inconsistency, imprecision, and risk of bias. Preoperative CBT has not been conclusively shown to significantly modify postoperative outcomes.

Supplemental Material

Supplemental Material - Interventions to Modify Psychological Processes in Patients Undergoing Spine Surgery: A Systematic Review

Supplemental Material for Interventions to Modify Psychological Processes in Patients Undergoing Spine Surgery: A Systematic Review by Matthew Skarsgard, Alysa Almojuela, Martin Gagliardi, Ganesh Swamy, Fred Nicholls, W Bradley Jacobs, Kenneth C Thomas, Alex Soroceanu, Denise Eckenswiller, Elias Soumbasis, Rob Tanguay, and Nathan Evaniew in Global Spine Journal.