Characteristics of eligible trials of patient management following lumbar spinal fusion surgery
| Trial | Design | Participants and indication | Intervention and setting | Outcome measures | Main results | Comments |
| Abbott et al (2010)33
Sweden | RCT Two groups: A: Exercise therapy B: Psychomotor therapyRecruitment strategy: patients selected for lumbar spinal fusion by spinal surgeons at one University Hospital orthopaedic clinic. Recruitment over 2-year period 2005–2007. | Patients aged 18–65 years, presenting with back pain/sciatica >12 months without success for conservative management; a primary diagnosis of spinal stenosis, spondylosis, degenerative or isthmic spondylolisthesis or degenetaive disc disease; patient selected for lumbar fusion with or without decompression; competent in Swedish; with no previous lumbar fusion, rheumatoid arthritis, ankylosing spondylitis. Baseline: Pre-operative. A: n=54 Mean age (SD) 51 (11) 57% women B: n=53 Mean age (SD) 50 (10) 66% women |
Intervention: A: Exercise therapy. Inpatient respiratory and circulatory exercises, training in transfers, walking and other activities of daily living. Prior to discharge, 20 min instruction in home exercises including dynamic endurance exercises for back, abdominal and leg muscles, stretches, and cardiovascular exercises. Progression of intensity and quantity by patient's perceived level of pain. Home programme continued from 0 to 12 weeks. Activity restriction from contact sports, running, heavy lifting, outer-range lumbar movements for 6 months post-operation. B: Psychomotor therapy. Same inpatient physiotherapy and activity restriction as A. Prior to discharge, 20 min instruction in home exercises including therapeutic exercise for lumbo-pelvic stabilisation, based on cognitive–behavioural early intervention and motor re-learning principles. Home programme continued from 0–3 and then 9–12 weeks. Programme progressed during 90 min outpatient sessions at 3, 6 and 9 weeks post-operation. Setting: Outpatient physiotherapy clinic. Home exercise programme. |
Short term:
ODI V.2.0 (primary outcome) VAS back pain intensity latest week (0–100 mm)EQ-5D SF-36 mental health subscale (0–100) SES (8–64) BBQ (9–45) TSK (17–68) CSQ and subscales Long term: a/a Return to work Work status Compliance evaluated through self-reported diary. Assessments: Short term at 3 and 6 months post-operation. Long term at 12 months and 2–3 years post-operation. | Statistically significant improvement for group B compared with group A in: ODI, SES, BBQ and TSK at 3, 6 and 12 months, and 2–3 years; VAS at 3 and 6 months; EQ-5D at 12 months; CSQ-catastrophising subscale in group B at 6 months and 2–3 years; CSQ coping to control pain and ability to decrease pain subscales at 3, 6 and 12 months. More patients employed 2–3 years after surgery. And fewer patients had sickness leave duration >6 months after surgery. Authors did respond to request for data, but no additional data available. | Primary outcome measure specified No primary end point specified A priori power calculation conducted on ODI (α=0.05; power =90%, MCID =10) Loss to follow-up: Dropouts: At 3 months (6%): A: n=3 withdrew B: n=3 treatment protocol violation No dropouts at 6-month follow-up. At 12 months (7%): A: n=2 (n=1 more pain, n=1 re-operation) At 2–3 years (19%): A: n=4 (n=1 not stated, n=3 pain free) B: n=8 (n=2 deceased, n=3 pain free, n=1 more pain, n=2 re-operation) No exclusions Management of losses: missing values imputed taking into account level of pain, when dropout was not associated with pain. Co-interventions not explored. ITT analyses performed. |
| Christensen et al (2003)13
Sogaard et al (2008)8 Sogaard et al (2006)32 Denmark | RCT Three groups: A: VideoB: Back-cafe C: Training Recruitment strategy: Patients post-lumbar spondylodesis 3 months previously invited to participate. Recruitment period 1996–1999. Stratification for posterior lumbar access, posterior and anterior access combined.n=25 patients refused inclusion. Statistically significant differences between groups for climbing stairs without pain at baseline. | Patients post-lumbar spondylodesis 3 months previously, with severe chronic low back pain caused localised lumbar or lumbosacral segmental instability caused by isthmic spondylolisthesis I/II, primary degeneration (no previous surgery), secondary degeneration after decompression surgery, accelerating degeneration after decompression surgery; with no age <20 or >60 years, one or more comorbidity (eg, metabolic bone disease, osteoporosis), clinical indication of new lumbar nerve root compression, psychosocial instability. Baseline: 1 week after 3 months post-operation follow-up clinical/radiograph. 60 women and 30 men Mean age 45 (range 24–60) A: 29 Median age (range) 45 (24–59) 83% women B: 26 Median age (range) 47 (24–60) 50% women C: 26 Median age (range) 48 (26–59) 67% women Conflicting reporting of n in each group. |
Intervention: A: Video. Video demonstration rehabilitation exercises, followed by one oral instruction by a physiotherapist. Exercises designed to provide dynamic muscular training to enhance endurance back, abdominal and leg muscles, stretching, warm up and restrictions to activity (no contact sports, training on machines at a fitness centre, running/jogging). Video and written instructions provided. B: Back-cafe. As A with additional back-cafe meetings with a physiotherapist and other patients to exchange experiences, doubts, reassurances, tips and psychological support. Involved tea/coffee, 10 min small talk, duration 1.5 h, three over period of 8 weeks. C: Training. Warm up of 15 min, condition training to improve heart–lung function, dynamic muscular endurance training of the back, abdominals and legs, and stretching exercises. Lasted 1.5 h, twice weekly over 8 weeks. Progressed repetitions individually. Unclear if a group intervention. Setting: Outpatient physiotherapy sections of hospital. Home exercise programme. |
Short term:
LBPRS: back and leg pain LBPRS: function and psychological capacity Long term: LBPRS: back and leg pain LBPRS: function and psychological capacity RTW Work status Back-related healthcare Assessments: Short term at 6 months post-operation. Long term at 12 and 24 months post-operation. | Statistically significant improvement for: leg pain in groups A and B compared with group C at 24 months. Components of function for group B compared with groups A and C at 24 months. Sick leave for group B compared with groups A and C at 12 and 24 months. Working patients in groups B and C compared with group A at 12 months. Authors did not respond to request for data. | No primary outcome measure specified No primary end point specified A priori power calculation reported but lacks clarity (α =0.05; power =90%). Reported that n=29 initially included in each group to ensure validity. n=9 lost to follow-up to give n=81. Loss to follow-up: Losses of n=9 (11%) overall (n=3 dissatisfied with allocation to group, n=2 re-operation, n=1 operation for malignancy, n=1 additional treatment, n=1 language issues, n=1 moved). Dropouts at each follow-up not reported. Lack of clarity over n in each group ateach assessment. No exclusions No management of losses described Co-interventions not explored No ITT analyses reported Data analysis and reporting results unclear Inconsistency in data between trial and subsequent reports |
BBQ, Back Beliefs Questionnaire; CSQ, Coping Strategies Questionnaire; EQ-5D, European Quality of Life Questionnaire; ITT, intention-to-treat; LBPRS, Low Back Pain Rating Scale; MCID, minimum clinically important difference; ODI, Oswestry Disability Index; RCT, randomised control trial; RTW, return to work; SES, Self-efficacy Scale; SF-36, Short Form 36-item health questionnaire; TSK, Tampa Scale for Kinesiophobia; VAS, Visual Analogue Scale.