The surgical site infection (SSI) rate in this study was 24 (11.7%), and most of these infections were superficial incisional SSIs (19, 79.2%). Among the several risk factors assessed for SSI, surgery was indicated for 19.4% of the patients with spondylotic disease and for 50% of the patients with bacterial spondylitis who developed SSI, whereas surgery was indicated for 9.4% of the patients with traumatic spine injury; this difference in the infection rate was found to be significant (p = 0.042). Similarly, infection rates were significantly greater for the lumbar spine (14, 22.2%) and thoraco-lumbar spine (4, 12.9%) surgeries (p = 0.009). Although the incidence of SSIs was greater in female patients, cigarette smokers, immunosuppressed patients, patients who underwent surgery via a posterior approach and patients who underwent surgery for >2 hours, this difference was not statistically significant. The multivariable regression model revealed that the spine segment affected was predictive of SSI (P = 0.001, odds ratio [OR]: 2.20, 95% confidence interval [CI]: 1.38, 3.47). Table 2 The median length of hospital stay (LOS) was found to be significantly longer (36.5 days) in patients with SSI than in those without SSI (23 days) (p = 0.008).
The prevalence of surgical site infections observed in this study, though relatively high, is the lowest observed in Nigeria, based on publications by other authors in Nigeria, who reported SSI rates between 11.8% and 36.1%.22-24 On a general note, based on the published literature on spine surgeries, SSI rates have been shown to range from 0.22% to 16.4% in some studies and from 1, 7 and 0.5% to 20% in others.4,5,8,9 This suggests that the infection rate in this study is similar to that reported by several other authors. This finding is relatively greater than the 3.1% overall pooled incidence of SSI in spine surgery reported in a Systematic Review and Meta-analysis by Zhou et al. 28
The number of publications related to the rate of SSI in spine surgery has increased in the past 15 years, indicating a growing interest in research related to this subject.29 Several theories may explain this growing interest. First, there is a persistently high rate of SSI after spine surgery, despite the availability and wide range of prophylactic antibiotics and several peri-operative aseptic techniques used by spine surgeons.8-10 Second, increased morbidity is caused by comorbid factors such as diabetes, antibiotic abuse, immune suppression and cigarette smoking. Third, the use of spinal instrumentation in spine surgeries has increased, and finally, diagnostic sensitivity has improved. This may explain why the topic “Risk Factors for spinal SSI” was the most published topic relating to spinal SSI in the past 15 years, as reported in a bibliometric analysis by Wang et al.29
This study reviewed several explanatory variables as possible risk factors for SSI, and we observed a significantly greater rate of SSI in patients who underwent surgery for spondylotic disease and acute bacterial spondylitis. Likewise, lumbar and thoraco-lumbar junction surgeries were associated with a significantly greater rate of SSI. Although all the SSIs occurred during surgery lasting more than two hours, this difference was not statistically significant. In the systematic review by Zhou et al., neuromuscular scoliosis was found to have the highest incidence of SSI, while idiopathic scoliosis had the lowest incidence; this finding was also previously reported by other authors.28,30-32 Although this study did not include any patients with scoliosis, other studies of non-deformation surgeries did not assess the indications for surgery. With respect to spine segment operation, the incidence of SSI in thoracic surgery was slightly greater than that in cervical and lumbar surgery according to Zhou et al., while other authors did not find spine segment operation to be a significant risk factor.28,33,34 This is contrary to the findings of this study, which revealed lumbar and thoracolumbar junction surgeries to be independent predictors of SSI.
Despite the finding in this study that SSIs occurred more often in female patients, cigarette smokers, immunosuppressed patients, patients who underwent posterior surgery and patients who underwent surgery for >2 hours, this difference was not statistically significant. Previous studies on these risk factors reported that posterior approach, immunosuppression and cigarette smoking are significant risk factors for SSI after spinal surgery. The duration of surgery as a risk factor for SSI has also been widely reported, with several authors attributing surgical timing ≥3 hours as a significant cut-off time that is associated with SSI, with Peng et al. reporting a fourfold-fold greater risk for SSI in patients who underwent spinal surgery lasting ≥3.28,35 However, other authors did not establish this relationship in their study.28,36,37 This finding is similar to our finding that all SSIs occurred in patients who underwent surgeries lasting >2 hours, but this difference was not statistically significant.
As one of the most common complications after spinal surgery, SSIs prolong the course of the disease and duration of hospital stay. This often results in increased medical bills, and the overall psychological consequences may affect the prognosis and outcome of care. This study revealed a significantly longer length of hospital stay in patients with SSI, although the cost of care for these patients was not evaluated. A systematic review by Patel et al. reported that patients who developed SSIs had a longer LOS (range, 7.1–19.3 d) than did those without SSIs (range, 4.0–9.3 d), which is also consistent with findings reported by other authors.8,38,39,40
The most common organism cultured in this study was Staphylococcus aureus, which is reportedly the most common causative microorganism of SSI after spine surgery. Patel et al. suggested that the pooled average contribution of S. aureus infections to spinal SSIs was 49.3%. However, in the meta-analysis by Zhou et al., slightly lower percentages of Staphylococcus aureus (37.9%) and Staphylococcus epidermidis (22.7%) were reported.
One of the major limitations of this study is its retrospective design, in which case some factors of potential significance could not be assessed properly due to missing or incomplete data on admission, which also resulted in a significant number of patients being excluded from the study. Second, the inclusion of nonrandomly selected participants may have resulted in difficulty eliminating unpredictable biases. This single-institution-based study had a limited dataset, and the lack of significance of some of the parameters of potential interest may reflect insufficient statistical power for generalization.