Abstract
Aim
Surgical site infection (SSI) is associated with substantial morbidity, mortality and economic burden. Management of spinal SSI is becoming more challenging especially in instrumented cases, but is not well recognized as high risk procedure. The objective of this study was to determine the impact of procedure type comparing SSI risk with arthroplasties among all orthopaedic procedures.
Method
Using prospectively collected data of consecutive samples in multi-center orthopedic SSI surveillance, we explored the differences in SSI rates within 30 days after surgery by procedure types. Patients who underwent surgery of single site between November 2013 and May 2016 were enrolled. SSI was our primary outcome. Urinary tract infection (UTI), and respiratory tract infection (RTI) were also evaluated. The definition of SSI was based on the CDC definition with slight modifications. All patients were followed for 30 days postoperatively. Multivariate logistic regression analyses were done, and variables were carefully selected for adjustments.
Results
In total 8,907 single site surgeries were analyzed. There were four major procedure types, fracture repair 31%, arthroplasty 30%, spinal surgery without instrumentation 14.7% and spinal instrumentation surgery 13%. Patient backgrounds were male 41.4%, diabetes 13.5%, rheumatoid arthritis 3.8 %, present smoker 13.4%, mean BMI 23+4, and operative time 144+92 minutes. Cefazolin was administered in more than 98% of all cases, and were administered appropriately before surgery. SSI occurred in 102 cases (1.2%), and the SSI rates were 2.5% in spinal instrumentation surgery and 0.6% in arthroplasty. After adjustment with several clinically relevant variables such as age, sex, diabetes and ASA, spinal instrumentation surgery was the only procedure which remained significant with adjusted odds ratio (aOR) of 3.3 (1.8–6.2, P<0.01) compared with arthroplasties. The risk remained stable after adding further clinically relevant variables (aOR of 2.2 to 3.3). The risk was not significant for spinal surgery without instrumentation (aOR, 1.8; 0.9–3.5, P=0.10). Moreover, the risk of spinal instrumentation surgery was highest for UTI (aOR, 4.7; 2.9–7.6), P<0.01) and RTI (aOR, 3.7; 1.6–8.9), P<0.01) among all procedures.
Conclusions
From our study, spinal instrumentation surgery was the only procedure to be significant after multivariate analysis, and the risk for SSI remained 2.2 to 3.3 fold higher compared with arthroplasties. The risk was also highest for several other major healthcare-associated infections. Considering the disastrous consequences, more interests and improvements in total perioperative care are needed for this procedure.
