This study evaluates the long-term survival of spinal implants after surgical site infection (SSI) and the risk factors associated with treatment failure.

A Kaplan-Meier survival analysis was carried out on 43 patients who had undergone a posterior spinal fusion with instrumentation between January 2006 and December 2008, and who consecutively developed an acute deep surgical site infection. All were appropriately treated by surgical debridement with a tailored antibiotic program based on culture results for a minimum of eight weeks.

A ‘terminal event’ or failure of treatment was defined as implant removal or death related to the SSI. The mean follow-up was 26 months (1.03 to 50.9). A total of ten patients (23.3%) had a terminal event. The rate of survival after the first debridement was 90.7% (95% confidence interval (CI) 82.95 to 98.24) at six months, 85.4% (95% CI 74.64 to 96.18) at one year, and 73.2% (95% CI 58.70 to 87.78) at two, three and four years. Four of nine patients required re-instrumentation after implant removal, and two of the four had a recurrent infection at the surgical site. There was one recurrence after implant removal without re-instrumentation.

Multivariate analysis revealed a significant risk of treatment failure in patients who developed sepsis (hazard ratio (HR) 12.5 (95% confidence interval (CI) 2.6 to 59.9); p < 0.001) or who had > three fused segments (HR 4.5 (95% CI 1.25 to 24.05); p = 0.03). Implant survival is seriously compromised even after properly treated surgical site infection, but progressively decreases over the first 24 months.

Cite this article: Bone Joint J 2013;95-B:1121–6.

Introduction: Whereas adults with Low Back Pain (LBP) who seek medical attention show a decrease on HRQoL, there is little information if patients are adolescents with LBP. The aim of our study is to assess the impact on HRQoL of adolescents referred to a hospital of geographical reference due to non-specific LBP (NS-LBP).

Methods: All consecutive adolescents with NS-LBP (patients) referred to the hospital between January06 and October07 completed a self-administered questionnaire including a generic quality of life (KIDSCREEN-52) and two LBP-specific (Roland-Morris, Hannover) instruments. Comparisons were performed among patients and two groups of schoolchildren (one of them with self reporting LBP and another one without) selected from a representative sample of 1470 schoolchildren from Barcelona and Friburg, paired by sex, age and country. Comparisons were made using t-tests and effect size (ES) estimation.

Results: Seventy-six patients (mean age 14.1y, 59.2% girls) completed the questionnaire and were compared with 304 controls (152 reporting LBP and 152 without LBP). Patients reported significantly higher frequency (p=0.014), duration (p=0.009) and intensity (p< 0.001) of pain than symptomatic schoolchildren. Perceived functional capacity (Roland Morris 5.5 vs 4.3, p=0.023, and Hanover 4.5 vs 3.5, p=0.032) was also worse, even tough the overall disability was not high in 65% of them. However, HRQoL (KIDSCREEN) was better in almost all dimensions in patients than in symptomatic schoolchildren.

Discussion: Overall adolescent LBP is associated to low disability, and scarce impact on QoL. Adolescents with LBP referred to the hospital have worse clinical and functional picture but better HRQoL than symptomatic peers from the general population using the Kidscreen instrument.