Reoperation rates in the surgical treatment of spinal metastases.

BACKGROUND CONTEXT: The surgical treatment in spinal metastases has been shown to improve function and neurologic outcome. Unplanned hospital readmissions can be costly and cause unnecessary harm.

PURPOSE: Our aim was to first analyze the reoperation rate and indications for this revision surgery in spinal metastases from an academic tertiary spinal institute and, second, to make comparisons on outcome (neurology and survival) against patients who underwent single surgery only.

STUDY DESIGN/

SETTING

This was an ambispective review of all patients treated surgically over an 8-year period considering their neurologic and survival outcome data. Statistical analysis was performed using IBM SPSS 20. Because all scale values did not follow the normal distribution and significant outlier values existed, all descriptive statistics and comparisons were made using median values and the median test. Crosstabs and Pearson correlation were used to calculate differences between percentages and ordinal/nominal values. For two population proportions, the z test was used to calculate differences. The log-rank Mantel-Cox analysis was used to compare survival.

PATIENT SAMPLE: During the 8 years' study period, there were 384 patients who underwent urgent surgery for spinal metastasis. Of these, 289 patients were included who had sufficient information available. There were 31 reoperations performed (10.7%; mean age, 60 years; 13 male, 18 female). Exclusion criteria included patients treated solely by radiotherapy, patients who had undergone surgery for spinal metastasis before the study period, and those who had other causes for neurologic dysfunction such as stroke.

OUTCOME MEASURES: The outcomes considered in this study were revised Tokuhashi score, preoperative/postoperative Frankel scores, and survival.

METHODS

We performed an ambispective review of all patients treated surgically from our comprehensive database during the study period (October 2004 to October 2012). We reviewed all patient records on the database, including patient demographics and reoperation rates.

RESULTS

Reoperations were performed in the same admission in the majority of patients (n=20), whereas 11 patients had their second procedure in subsequent hospitalization. The reasons for their revision surgery were as follows: surgical site infection (SSI; 13 of 31 [42%]), failure of instrumentation (9 of 31 [29%]), local recurrence (5 of 31 [16%]), hematoma evacuation (2 of 31 [6%]), and others (2 of 31 [6%]).When comparing the "single surgery" and "revision surgery" groups, we found that the median preoperative and postoperative Frankel scores were similar at Grade 4 (range, 1-5) for both groups (preoperative, p=.92; postoperative, p=.87). However, 20 patients (8%) from the single surgery group and 7 (23%) from the revision group had a worse postoperative score, and this was significantly different (p=.01). No significant difference was found (p=.66) in the revised Tokuhashi score. The median number of survival days was similar (p=.719)-single surgery group: 250 days (range, 5-2,597 days) and revision group: 215 days (range, 9-1,352 days).

CONCLUSION

There was a modest reoperation rate (10.7%) in our patients treated surgically for spinal metastases over an 8-year period. Most of these were for SSI (42%), failure of instrumentation (26%), and local recurrence (16%). Patients with metastatic disease could benefit from revision surgery with comparable median survival rates but relatively poorer neurologic outcomes. This study may help to assist with informed decision making for this vulnerable patient group.