- FACET (Foraminotomy ACDF Cost-Effectiveness Trial)
- Class of Evidence
- Multicenter investigator-blinded noninferiority randomized clinical trial.
- Randomization
- Randomly assigned in a 1:1 ratio (n256)
- Posterior cervical foraminotomy (posterior surgery): n132
- Anterior cervical discectomy with fusion (anterior surgery): n133
- Number of Patients
- At the 1-year follow-up, 228 patients were included in the analysis (110 in the posterior group and 118 in the anterior group), as 15 patients were lost to follow-up.
- Length of Follow-up (1 years)
- Clinical follow-up to 2 years is ongoing.
- Number of Centres Involved with the Study
- 9 hospitals in the Netherlands.
- Outcomes
- Primary Outcomes:
- Proportion of success using Odom criteria (a 4-point rating scale, with "excellent" or "good" considered a successful score).
- Decrease in arm pain using a visual analogue scale (VAS) score from 0 to 100.
- Results at 1 year:
- This difference of –0.11 percentage points (1-sided 95% CI, −0.01) indicated noninferiority of posterior surgery compared to anterior surgery regarding the success rate
- The proportion of success was
- 0.88 (86 of 98) in the posterior surgery group
- 0.76 (81 of 106) in the anterior surgery group.
- The mean between-group difference in arm pain at 1 year was −2.8 (1-sided 95% CI, −9.4), which also indicated noninferiority of posterior surgery.
- However, the mean between-group difference for the change score in arm pain was 3.4 (1-sided 95% CI, 11.8), crossing the 10% noninferiority margin, meaning there was insufficient evidence to establish noninferiority for the change in arm pain.
- Despite this, the authors consider such small between-group differences to be clinically similar.
- Secondary Outcomes: These included neck pain (VAS score), disability (Neck Disability Index), quality of life (EuroQol 5-Dimensions 5-Level questionnaire), work status (Work Ability Index Single-item), treatment satisfaction, reoperations, and complications.
- Results at 1 year: All secondary outcomes showed small between-group differences, with 2-sided 95% CIs clustered around 0, indicating comparable outcomes between the posterior and anterior surgery groups.
- Adverse events:
- Surgery-associated adverse events occurred in 26 individuals (22%) in the posterior group and 22 (18%) in the anterior group.
- Serious surgery-associated adverse events occurred in 7 patients (6%) in both groups.
- Reoperations occurred in 6 patients (5%) in the posterior group and 4 (3%) in the anterior group.
- The types of surgery-associated adverse events varied between groups, with dysphagia and hoarseness more frequent in anterior surgery, and wound infections more common in posterior surgery.
- Critique of the Study
- Unmet Sample Size:
- The predefined sample size of 308 patients was not reached; only 265 were randomized. Although an interim analysis suggested it was safe to conclude early, a lower inclusion rate than anticipated occurred, partly due to COVID-19-related cancellation of nonemergent health care.
- Power for Arm Pain Outcome:
- The study was powered primarily on the proportion of success (Odom score) and not on the mean decrease in arm pain, which may explain why the noninferiority was confirmed for arm pain at 1 year but not for the change in arm pain.
- Empirical Noninferiority Margin:
- The 10% noninferiority margin was chosen empirically, as an evidence-based margin for this research question does not exist.
- Inability to Blind:
- A significant limitation was the inability to blind surgeons and patients to the intervention due to the nature of the surgical procedures.
- Potential Selection Bias:
- Despite clear inclusion criteria, possible selection bias occurred in the assessment of eligible patients, as some surgeons had strong preferences for one of the interventions (mostly anterior surgery), which may have negatively affected participant enrollment.
Made with Bullet