Risk factors for implant removal after spinal surgical site infection.

PURPOSE: Few studies have investigated the risk factors for implant removal after treatment for spinal surgical site infection (SSI). Therefore, there is no firmly established consensus for the management of implants. We aimed to investigate the incidence and risk factors for implant removal after SSI managed with instrumentation, and to examine potential strategies for avoiding implant removal. METHODS: Following a survey of seven spine centers, we retrospectively reviewed the records of 55 patients who developed SSI and were treated with reoperation, out of 3967 patients who had spinal instrumentation between 2003 and 2012. We examined implant survival rate and applied logistic regression analysis to assess the potential risk factors for implant removal. RESULTS: The overall rate of implant retention was 60% (33/55). A higher implant retention rate was observed for posterior cervical surgery than for posterior-thoracic/lumbar surgery (100 vs. 49%, P < 0.001). On univariate analysis, significant risk factors for implant removal included greater blood loss, delay of reoperation, and delay of intervention with effective antibiotics. Multivariate analysis revealed that a delay in administering effective antibiotics was an independent and significant risk factor for implant removal in posterior-thoracic/lumbar surgery (odds ratio 1.17; 95% confidence interval 1.02-1.35, P = 0.028). CONCLUSIONS: Patients with SSI who underwent posterior cervical surgery are likely to retain the implants. Immediate administration of effective antibiotics improves implant survival in SSI treatment. Our findings can be applied to identify SSI patients at higher risk for implant removal.

A multi-center study of reoperations within 30 days of spine surgery.

PURPOSE: To investigate the incidence and causes of reoperations within 30 days of spine surgery. METHODS: Patients who underwent spine surgery between 2002 and 2012, at one of 9 institutions, were enrolled. The causes of all reoperations, within 30 days of the index surgery, were reviewed. The incidence of reoperations within 30 days were calculated and compared according to the preoperative diagnosis, use of instrumentation, surgical level, and approach. Moreover, the distribution of the causes for reoperations within the 30-day period was also assessed. RESULTS: The overall incidence of reoperations, within 30 days of spine surgery, was 1.6% (175/10,680). Patients who underwent instrumentation procedures demonstrated a higher incidence of reoperations than patients who underwent procedures without instrumentation (P < 0.001). Moreover, patients diagnosed with preoperative trauma had a higher incidence of reoperation than those with other diagnoses (P < 0.001). Surgical site infection (SSI), postoperative epidural hematoma, pedicle screw malposition, and inadequate decompression were the four main causes of reoperation. Motor paralysis, due to epidural hematoma, was the predominant cause of reoperations during the hyper-acute phase (0-1 days, postoperatively), whereas SSI was the dominant cause during the sub-acute phase (11-30 days, postoperatively). CONCLUSIONS: This large, multi-center study indicated the causes and incidence of reoperations, within 30 days of the initial spinal surgery. Based on these data, spinal surgeons can provide patients with information that allows them to better understand the surgical risk and expected post-surgical management.

[Home oxygen therapy for terminal lung cancer cases].

Home oxygen therapy (HOT) was introduced as part of the terminal care of 52 lung cancer patients attending our hospital from 1997 to 2000, all of whom subsequently died from their illnesses. We administered a questionnaire about HOT to their bereaved family members. The results confirmed that many of those interviewed thought that HOT improved their quality of life, but probably made the patients more aware of the progression of their disease. As the incidence of lung cancer has been increasing, lung cancer patients have begun to form an increasing proportion of patients receiving HOT in recent years. In addition, the number of patients with a PaO2 of > or = 60 torr at rest at the start of HOT has been increasing. It may be necessary when dealing with terminal lung cancer patients, in whom PaO2 can decrease rapidly over a relatively short period of time, to explore indication criteria for HOT different from those conventionally employed.