Comparative Effectiveness Between Primary and Revision Foraminotomy for the Treatment of Lumbar Foraminal Stenosis

Background: Foraminotomy has demonstrated clinical benefit in patients with lumbar foraminal stenosis (LFS), as evidenced by several small retrospective investigations. However, there is a subset of patients who have recurrent symptoms following the operation and therefore require revision surgery. Yet, despite this phenomenon, the relative efficacy of revision foraminotomy (RF) is not well elucidated due to limited literature on the quality of life (QOL) outcomes and cost associated with primary foraminotomy (PF) and RF.

Purpose: To compare the effectiveness of PF and RF in terms of QOL outcomes and relative costs.

Study design/setting: This is a retrospective cohort study conducted at a single tertiary-care institution. The patient sample consisted of patients undergoing foraminotomy for the treatment of LFS between 2008 and 2016. The primary outcome measure was improvement in postoperative QOL, as measured by EuroQol 5-Dimensions (EQ-5D), and secondary outcome measures included Pain Disability Questionnaire (PDQ) and Patient Health Questionnaire-9 (PHQ-9) perioperative cost as well as minimum clinically important difference (MCID).

Methods: A retrospective chart review was conducted to identify individuals who underwent PF or RF for LFS and to collect clinical, operative, and demographic data. QOL scores (EQ-5D, PDQ, and PHQ-9) were collected between 2008 and 2016, and perioperative financial data were extracted via the institution's cost utilization engine. Paired t tests were used to assess changes within treatment groups, and Fisher exact tests were used for intercohort comparisons.

Results: Five hundred seventy-nine procedures were eligible: 476 (82%) PF and 103 (18%) RF. A significantly higher proportion of males underwent RF than PF (71% versus 59%, P = .03), and PF was done on a significantly higher number of vertebral levels (2.2 versus 2.0, P = .04). There were no other significant differences in demographics. Preoperatively, mean PDQ-Functional scores (50 versus 54, P = .04) demonstrated significantly poorer QOL in the RF cohort. Postoperatively, EQ-5D index showed significant improvement in both the PF (0.547→0.648, P < .0001) and the RF (0.507→0.648, P < .0001) cohorts. Similarly, total PHQ-9 improved significantly in the PF cohort (7.84→5.91, P < .001) and in the RF cohort (8.55→5.53, P = .02), as did total PDQ (PF: 77→63, P < .0001; RF: 85→70, P = .04). QOL scores were also compared between groups preoperatively and postoperatively, and the only significant difference between PF and RF was observed in the preoperative PDQ-Functional score (49.7 versus 54.3, P = .04). The proportion of patients achieving MCID was not significantly associated with cohort. Finally, perioperative cost did not differ significantly between cohorts (PF: $13,383 versus RF: $13,595, P = .82).

Conclusions: Both PF and RF produced significant improvement in nearly all measures in patients with LFS. There was no significant difference in cost between PF and RF, but both PF and RF showed postoperative QOL improvements as compared with preoperative scores, indicating that RF remains a reasonable treatment option for patients with recurrent symptoms of LFS.