Effects of smoking on pedicle screw-based dynamic stabilization: radiological and clinical evaluations of screw loosening in 306 patients

OBJECTIVE Cigarette smoking has been known to increase the risk of pseudarthrosis in spinal fusion. However, there is a paucity of data on the effects of smoking in dynamic stabilization following lumbar spine surgery. This study aimed to investigate the clinical outcomes and the incidence of screw loosening among patients who smoked. METHODS Consecutive patients who had lumbar spondylosis, recurrent disc herniations, or low-grade spondylolisthesis that was treated with 1- or 2-level surgical decompression and pedicle screw-based Dynesys dynamic stabilization (DDS) were retrospectively reviewed. Patients who did not complete the minimum 2 years of radiological and clinical evaluations were excluded. All screw loosening was determined by both radiographs and CT scans. Patient-reported outcomes, including visual analog scale (VAS) scores of back and leg pain, Japanese Orthopaedic Association (JOA) scores, and Oswestry Disability Index (ODI), were analyzed. Patients were grouped by smoking versus nonsmoking, and loosening versus intact screws, respectively. All radiological and clinical outcomes were compared between the groups. RESULTS A total of 306 patients (140 women), with a mean age of 60.2 +/- 12.5 years, were analyzed during an average follow-up of 44 months. There were 34 smokers (9 women) and 272 nonsmokers (131 women, 48.2% more than the 26.5% of smokers, p = 0.017). Postoperatively, all the clinical outcomes improved (e.g., VAS back and leg pain, JOA scores, and ODI, all p < 0.001). The overall rate of screw loosening was 23.2% (71 patients), and patients who had loosened screws were older (61.7 +/- 9.6 years vs 59.8 +/- 13.2 years, p = 0.003) and had higher rates of diabetes mellitus (33.8% vs 21.7%, p = 0.038) than those who had intact DDS screws. Although the patients who smoked had similar clinical improvement (even better VAS scores in their legs, p = 0.038) and a nonsignificantly lower rate of screw loosening (17.7% and 23.9%, p = 0.416), the chances of secondary surgery for adjacent segment disease (ASD) were higher than for the nonsmokers (11.8% vs 1.5%, p < 0.001). CONCLUSIONS Smoking had no adverse effects on the improvements of clinical outcomes in the pedicle screw-based DDS surgery. For smokers, the rate of screw loosening trended lower (without significance), but the chances of secondary surgery for ASD were higher than for the nonsmoking patients. However, the optimal surgical strategy to stabilize the lumbar spine of smoking patients requires future investigation.