PATIENTS WITH SPINAL METASTASIS SUBMITTED TO NEUROLOGICAL DECOMPRESSION AND STABILIZATION

Objective: To analyze retrospectively the surgical outcomes of a group of patients with bone metastases and multiple myeloma in the spine, which underwent neurological decompression and arthrodesis using pedicle screws, by isolated posterior approach, to check whether the operated patients present clinical improvement regarding the pain and neurological deficit compared to the preoperative period. Methods: This is a retrospective, cross-sectional study of case series, that analyzed data from medical records of patients with bone metastases in the spine who underwent surgical treatment between January 2007 and February 2011. Results: Of the 42 patients in the sample, according to the Kolmogorov-Smirnov test (p=0.000) there was improvement in pain with respect to the preoperative, with 33 patients (78.6%) reporting improvement and only 9 (21.4%) maintaining the initial pain complaints. Of the 25 patients available for evaluation of neurological improvement (≠ Frankel E) 9 patients (36%) had some kind of improvement postoperatively, and no operated patient presented neurological worsening, indicating statistical significance according to the Kolmogorov-Smirnov test (p = 0.000). It was also observed statistical correlation (p=0.042) between Frankel functional score postoperatively and pain relief, using the chi-square test. Conclusions: Surgical treatment for patients with spinal metastasis through arthrodesis with pedicle instrumentation and decompression may have significant clinical benefits, especially as regards the improvement in pain symptoms and improved neurological function.


InTRODUCTIOn
Metastatic tumors are the most common malignant bone lesions, occurring 40 times more often than all other primary malignant bone tumors combined.¹The main primary sites include the breasts, lungs, prostate, kidneys, gastrointestinal system, and the thyroid gland.Multiple myeloma is a plasma cell dyscrasia, in which bone lesions are common in functioning bone marrow sites, such as the vertebral bodies.It is the most common malignant neoplastic skeletal lesion.¹Recent advances in chemotherapy, radiotherapy, and other oncological treatments have significantly improved the survival of many patients with these lesions.With improved survival rates, previously silent spine metastases are becoming clinically apparent and significantly impairing the quality of life of these patients.Metastatic disease affects the spine in 50% to 80% of patients, most frequently affecting the vertebral bodies of the lumbar spine, followed by the thoracic, cervical, and sacral regions.¹Approximately 10% of cancer patients are diagnosed with symptomatic vertebral metastasis during the course of the disease and up to 50% of these patients require treatment of spine metastases because of neurological deficit or pain.²It is estimated, from studies performed on cadavers, that the frequency of vertebral lesions in patients who die from malignant neoplasia ranges from 30% to 90%.³ Currently, there are various ways of addressing spine metastases, taking into account the degree of spinal involvement, the presence of neurological deficit, immunocompetence and the clinical conditions of the patient, the number of metastases to other organs, the number of metastases to the appendicular skeleton, and the primary tumor type.There are several studies that have attempted using prognostic scores to determine the best type of therapeutic approach, including wide excision of the lesion, marginal or intralesional excision, palliative surgery, and non-surgical treatment. 4he main objectives of surgical treatment are pain control, maintenance of spinal stability, and the preservation of neurological function.Therefore, surgery is reserved for patients with neurological compromise, radioresistant tumors, vertebral instability, and non-treatable mechanical pain. 1,2he objective of this study is to retrospectively analyze the surgical outcomes of a group of patients diagnosed with bone metastases and multiple myeloma of the spine, who underwent isolated posterior approach medullary decompression surgery associated with fusion with pedicle screw instrumentation.This is a surgical procedure of intermediate complexity with rapid recovery, a low expected rate of complications, mainly serious complications in the short term, but with a long-term risk of recurrence and local complications.We will evaluate whether the patients who underwent surgery experienced pain relief and neurological changes as compared to the preoperative period and we will also evaluate short-and long-term surgical complications and survival time.

MATERIAL AnD METhODS
We retrospectively evaluated all 42 patients with bone metastases and multiple myeloma of the spine who underwent surgery during the period between January 2007 and February 2011.The study was approved by the Institutional Review Board.
The variables evaluated in the study were: sex, age, primary tumor type, pre-and postoperative frankel ability scores, short-and long-term surgical complications (whether present or not and description of the type), postoperative pain assessment (improved or not improved), survival time, satisfaction with the procedure.
We evaluated medical records, conducted routine outpatient consultations, and made telephone calls to each living patient to get answers to the following two questions: "Were you satisfied with the surgical procedure performed?"and "Would you undergo the same surgery again to treat the same problem?".
The patients underwent posterior approach vertebral fusion with pedicular instrumentation of one or two levels above and below the affected vertebra, associated with localized decompressive laminectomy with little or no involvement of the lesion of the vertebral body, but performing, when necessary, partial posterolateral approach resections of lesions of significant mass effect.All the patients with radiosensitive tumors underwent local adjuvant radiotherapy 20 to 30 days following surgery.
A significance level of 5% and a confidence level of 95% were used for analysis of the variables collected and the statistical tests.SPSS version 17.0 was the software used.
Descriptions of frequency and averages were used to evaluate the demographic and epidemiological data.
The Kolmogorov-Smirnov test was used to test the null hypothesis that there was no postoperative pain relief or of the Frankel functional ability score.
The chi-squared test was used to analyze the correlation between the variables of pain relief and Frankel functional ability score, short-and long-term surgical complications and type of tumor, and complication rate and level of satisfaction with the procedure.
Fisher's exact test was used to evaluate the relationship between the postoperative Frankel scale and post-surgical complications.
The ANOVA was used to evaluate the correlation between the variable of time of survival and type of primary tumor.

RESULTS
Of the patients evaluated, 17 (40.5%)were male and 25 (59.5%) were female.At the time of surgery, more than half of the patients (54.8%) were between 31 and 60 years of age.A total of 17 (40.5%)were 61 years of age or older.Only 2 patients (4.8%) were younger than 30 years of age.The youngest patient at the time of surgery was 11 years old, the oldest was 81, and the average age was 55.33 years.Excluding the outlier factor (the 11-year-old patient), the average age was 56.4 years.(Table 1) The most common primary tumors among the patients studied were of the breast, occurring in 12 (28.6%), of the prostate in 7 (16.7%),and multiple myeloma in 6 (14.3%).Of the 42 patients, four (9.5%) had had no prior diagnosis.
The most frequently occurring preoperative Frankel functional ability score was E, in 17 patients (40.5%), but most of the patients who underwent surgical treatment had some type of change (59.5%).None of the patients had a Frankel score of A.
For the analysis of improvement using the Frankel functional scale, the patients with a preoperative Frankel score of E were excluded since they had no motor or sensory deficits and the goal of surgery in these cases was not to improve the neurological profile, but rather to improve pain and spinal stabilization.Thus, 24 patients were available for an evaluation of neurological improvement.Of these patients, 36% (9 patients) showed some type of postoperative improvement.None of the patients who underwent surgery worsened from a neurological perspective.According to the Kolmogorov-Smirnov test, the null hypothesis that there was no improvement in the functional ability score should be rejected, i.e., it can be considered that there was some type of improvement (p=0.000).
According to the Kolmogorov-Smirnov test, of the 42 patients from the sample that were analyzed, the pain profile showed an improvement over the preoperative values (p=0.000), with 33 patients (78.6%) reporting improvement and only 9 of them (21.4%)maintaining their initial complaints of pain.
We observed a statistically significant correlation (p=0.042) between the postoperative Frankel functional score and pain relief using the chi-squared test.It can be seen in the table that the worst results in relation to pain occurred in patients whose neurological profiles did not improve.
Surgical complications were separated into short-and longterms categories, with 11 patients (26.2%) presenting some sort of short-term surgical complication.(Figure 1) There were a total of 17 short-term complications among these patients, some of whom experienced more than one complication.The most frequently observed complication was deep infection, occurring five times (29.4% of the complications in 11.9% of the patients).Another complication was perioperative bleeding (patients who needed   blood transfusions for hemodynamic stabilization in the peri-and immediate postoperative periods), occurring four times (23.5% of the complications in 9.5% of the patients), and superficial infection three times (17.6% of the complications in 7.1% of the patients).One of the patients who was bleeding (2.3% of the sample) died during the early postoperative period as a result of sustained shock.
There was no correlation between the rate of complications and the type of primary tumor, either in the short term (p=0.126) or in the long term (p=0.958).
In relation to long-term surgical complications, 8 (19.0%) of the 42 patients had some type of complication.(Figure 2) The most frequent among the eight patients affected was residual pain, occurring in five of them, followed by deep infection, which occurred four times, and the loosening of the osteosynthesis material, three times.(Table 2) Some patients experienced more than one complication.According to the Fisher's exact test, no relationship exists between the postoperative Frankel functional scale score and postsurgical complications (p=1.000).
Half of the patients had died and half were still living at the cutoff date, 07/31/2012.Of the 21 patients who had died, six had a survival time of up to six months (28.6%), one patient had a survival time between 6 and 12 months (4.8%), and 14 survived longer than 12 months (66.7%).(Table 3) The shortest patient survival time was 2 days and the longest was 48 months.The average survival time among the deceased was 20.76 months.The average survival time of the patients who were still living on 07/31/2012 was 35.86 months.According to the ANOVA results, the variables of survival time and type of primary tumor are independent (p=0.954).
The satisfaction with the surgical procedure questionnaire was answered by 21 patients (living patients), with 19 (90.5%) who were satisfied and who would undergo the surgery again.There was no correlation between the occurrence of complications and the degree of satisfaction with the procedure as analyzed using the chi-squared test.

DISCUSSIOn
Choosing a treatment for patients with spine metastases is usually a challenge for the spine surgeon because there are several factors that must be taken into account before thinking about a surgical approach, such as life expectancy, the risks inherent in the surgical procedure, the possibility of some functional improvement or lack thereof, and the stage of malignancy of the primary tumor.][4] The main objective of this study was to analyze the surgical outcomes of patients with spine metastasis, with an emphasis on pain relief and improved neurological symptoms, using a treatment technique that is less complex than large tumor resections.Several studies have reported improved quality of life following surgical treatment of patients with spine metastases, with more than 80% of the patients satisfied or very satisfied. 5,6In this study, we observed pain relief of 78.6% in the total patient sample and, among those still living, we confirmed that 90.5% were satisfied with their decision to operate and would undergo the procedure again to treat the same problem, in agreement with the literature.
Most of the patients with metastatic spine lesions were assessed as Frankel E (40.5%) on the Frankel functional scale.Excluding these patients, there was a functional improvement in nine (36%) of the 25 patients with neurological deficit, showing that patients who underwent surgical decompression and stabilization had significant improvement in relation to the Frankel functional scale.In this study, we did not observe worsening of neurological symptoms in any patient and we observed some level of improvement in a third of them, in line with the literature. 2,4,5,7An interesting fact observed in the study was the correlation between pain relief and improved neurological symptoms, with pain relief in all patients who experienced neurological improvement and the absence of pain relief in the patients who experienced no change in the neurological lesion.
][8][9][10][11] Breast, prostate, and multiple myeloma were the most prevalent primary tumors in this study, totaling 59.6%.No statistical significance was observed (p=0.954) between the primary tumor type and patient survival.We believe that this lack of correlation is due to the small sample and the greater proportion of patients with slow-growing tumors.
2][13][14] The highest complication rates and the most serious complications apply mainly to larger operations, including vertebrectomies and large resections, which are associated with increased morbidity as compared to simpler procedures. 4,5,6,7,11,12,14In our study, we observed a rate of short-term surgical complications of 26.2%, most of them minor, with only one early death.The most often encountered complication in the series was infection of the surgical site in 47% of the more complex cases, which is in agreement with that observed in the literature. 5,7,15his study proposes that treatment of intermediate complexity be applied uniformly to all patients, regardless of tumor type, with a high rate of satisfaction (90.5%), good pain relief (78.6%), and some neurological recovery in about a third of the patients.This fact leads to questions about the need to perform more complex procedures, given that the reduced postoperative rehabilitation time is critical in patients with limited survival.
Hosono et al. 6 retrospectively reviewed a large series of patients with spine metastases and concluded that "a large prospective study of consecutive patients is essential to track possible prognostic factors in patients with spine metastases".This type of study can help to make more appropriate decisions about the scope of surgery to be recommended.

COnCLUSIOn
With this study, we conclude that surgical treatment for patients with spine metastases, using fusion with pedicular instrumentation and decompression, can bring significant clinical benefits, especially as regards pain relief and improved neurological symptoms.
All the authors declare that there are no potential conflicts of interest regarding this article.

Table 1 .
Description of the patients.

Table 3 .
Survival rate.Each author made significant individual contributions to the development of the manuscript.JAN and BPCF were the principle contributors to the writing of the article.They conducted the bibliographical research, the review of the manuscript, and contributed to the intellectual concept of the study.JAN, BPCF, RDAM, and CES performed the surgeries, monitored the patients, gathered the clinical data, and evaluated the data from the statistical analysis.