Fax: (011) 86-21-8187-0291
Health-related quality of life in patients with spinal metastases treated with or without spinal surgery
A prospective, longitudinal study
Article first published online: 13 MAY 2010
Copyright © 2010 American Cancer Society
Volume 116, Issue 16, pages 3875–3882, 15 August 2010
How to Cite
Wu, J., Zheng, W., Xiao, J.-R., Sun, X., Liu, W.-Z. and Guo, Q. (2010), Health-related quality of life in patients with spinal metastases treated with or without spinal surgery. Cancer, 116: 3875–3882. doi: 10.1002/cncr.25126
- Issue published online: 4 AUG 2010
- Article first published online: 13 MAY 2010
- Manuscript Accepted: 27 OCT 2009
- Manuscript Revised: 26 OCT 2009
- Manuscript Received: 2 SEP 2009
- quality of life;
- spinal metastasis;
- Functional Assessment of Cancer Therapy;
- health-related quality of life
The objective of this study was to determine whether surgery for patients with spinal metastases could improve the quality of remaining life and prolong survival.
In total, 96 patients who had spinal metastases were recruited from Changzheng Hospital at the Second Military Medical University in Shanghai, China, over the period from July 2007 to June 2009. The patients received treatments with or without spinal surgery (surgery group, n = 46 patients; nonsurgery group, n = 50 patients), and all patients received adjuvant therapies according to their original cancer and individual conditions. Patients' quality of life (QOL) was assessed at 5 time points—at the initial diagnosis (baseline) and at 1 month, 3 months, 6 months, and 9 months of follow-up—using the Functional Assessment of Cancer Therapy-General QOL questionnaire. Information on survival also was collected.
Sixty-seven of 96 patients completed all 5 follow-up assessments, including 33 patients in the surgery group and 34 patients in the nonsurgery group. The other 29 patients died within 9 months after the initial diagnosis. At the end of the study (June 2009), 22 patients (47.8%) remained alive in the surgery group, and with 16 patients (32%) remained alive in the nonsurgery group. The surgery group had significantly higher total QOL scores, physical well being scores, emotional well being scores, and functional well being scores than the nonsurgery group over the 9-month assessment period. There was no statistically significant difference in survival between the 2 groups (P = .056).
The current results indicated that surgical treatment greatly improved and maintained the QOL of patients who had spinal metastases over the 9-month assessment period and that surgery is an effective treatment for spinal metastases. Cancer 2010. © 2010 American Cancer Society.
There has been growing recognition that patient-reported outcome measures, and particularly measures of health-related quality of life (HR QOL), can convey important information to aid in making decisions about cancer treatment.1 In recent years, successful quality of life (QOL) evaluations have been performed on patients with cancers of the prostate,2 breast,3 lung,4 and liver5 after treatment.
Spine metastases are common among patients with advanced-stage cancer, which can lead to pain, pathologic fracture, hypercalcemia of malignancy, and spinal cord compression.6, 7 Spinal cord compression is a severe complication of cancer that occurs in 5% to 14% of patients who have metastatic cancer and can cause intractable pain, loss of mobility, and incontinence.8, 9 Because most spinal metastases are detected at later stages, when they are generally incurable, the impact of treatment on QOL should be an important consideration in the choice of treatment.
However, QOL data on patients with spinal metastases are limited. To our knowledge, there has been no previous report on QOL outcomes after surgery in patients with spinal metastases. The objective of the current prospective, longitudinal study was to identify and compare the QOL of patients with spinal metastases over a certain time after spinal surgery.
MATERIALS AND METHODS
Ninety-six patients who presented between July 2007 and September 2008 at Changzheng Hospital, the Second Military Medical University in Shanghai, China were eligible for this observational study. Inclusion criteria for all participants were age ≥18 years, histologically confirmed spinal metastasis of epithelial cancer origin with at least 1 neurologic sign or spinal symptom (mainly focused on pain), spinal cord compression restricted to a only single area (but could include several contiguous vertebral segments), and an estimated survival >3 months. Exclusion criteria were primary spinal tumor, previous surgery for a spinal tumor, total paraplegia for >72 hours before study entry, and health too poor to undergo surgery.
All eligible patients were separated into 2 groups, depending on whether or not they would undergo surgery (surgery group, n = 46 patients; nonsurgery group, n = 50 patients). From a clinical perspective, surgery could be suitable for all patients who were eligible for this study. However, it was up to the patient and his or her family members to decide whether to undergo surgery. Written informed consent was obtained from all patients.
This was a prospective, longitudinal study with 2 treatment groups: a surgery group and a nonsurgery group. At the time of diagnosis, each eligible patient completed a baseline QOL assessment using the Functional Assessment of Cancer Therapy-General (FACT-G) QOL questionnaire.
Patients in the surgery group underwent surgical treatment within 3 days after diagnosis. There were no restrictions imposed on the surgeons in terms of resection methods or fixation devices. The objectives of the surgery were to excise and remove the metastatic spinal tumors completely, provide immediate decompression of the spinal cord, control pain, and maintain the stability of the vertebral column. Patients subsequently received different postoperative treatments, such as combined therapies, including chemotherapy, radiotherapy, hormone therapy, and bisphosphonate therapy, depending on their original cancer and general medical status. Patients were assessed again at 4 time points: 1 month, 3 months, 6 months, and 9 months during the postoperative period. Patients in the nonsurgery group received combined therapies immediately after diagnosis but without undergoing spinal surgery. QOL assessments also were performed in the nonsurgery group at 1 month, 3 months, 6 months, and 9 months time points after the diagnosis.
Baseline interviews were performed in the hospital. Follow-up assessments were performed by mail. All QOL data were collected and checked by the research assistant, who would call the patients for responses on missing items to minimize missing data. For those patients who survived for <9 months after diagnosis, QOL assessments could be performed only up to the last follow-up visit.
Instrument for QOL Assessment
The FACT-G questionnaire is 1 of the most widely used instruments for QOL assessment of patients with cancer and has verified as applicable to patients from diverse cultural backgrounds.10-13 The FACT-G (version 4.0) questionnaire is a self-report instrument that measures multidimensional QOL using a total of 27 items in the following 4 specific life domains: 1) physical well being (GP) (7 items), 2) social/family well being (GS) (7 items), 3) emotional well being (GE) (6 items), 4) and functional well being (GF) (7 items). Each item is answered with regard to the past 7 days. All FACT-G items are rated on a 5-point scale ranging from 0 (not at all) to 4 (very much). The maximum combined score for GP, GS, and GF is 28; the maximum score for GE is 24, and the total score can range from 0 to 108.14 Higher scores indicate better QOL. In the current study, the Chinese language version of the FACT-G questionnaires was administered at baseline and at 1 month, 3 months, 6 months, and 9 months follow-up. In addition, the FACT-G questionnaire includes the item, “I have pain” (GP; item 4). The data from this item was used to assess pain control in this study.
Chi-square tests and t tests were used to compare the baseline characteristics of patients in the surgery and nonsurgery groups. Two types of comparisons were made. First, repeated-measures analyses of variance were used to identify changes in QOL over time within the same treatment group and between the 2 treatment groups. Second, paired t tests were used to compare the QOL results from the 2 treatment groups and their respective scores at baseline, 1 month, 3 months, 6 months, and 9 months.
Kaplan-Meier curves were generated, and the log-rank test was used to determine whether there was any statistically significant difference in terms of overall survival between the 2 groups. A significance level of P = .05 was used in all analyses. All statistical analyses were performed using the SAS 9.1.3 software package (SAS Institute, Cary, NC).
In total, 67 patients (69.8%) completed all 5 assessments, including 33 patients (71.7%) in the surgery group and 34 patients (68%) in the nonsurgery group. At the end of this study (June 2009), 22 patients (47.8%) remained alive in the surgery group compared with 16 patients (32%) in the nonsurgery group (Fig. 1).
The overall median follow-up was 10.5 months (range, 1-23 months) in the surgery group and 11 months (range, 1-20 months) in the nonsurgery group. According to World Health Organization criteria, perioperative death was defined as death within 30 days of surgery.9 In the current study, 3 patients died within 1 month after undergoing surgery (at 16 days, 17 days, and 28 days), and the surgical perioperative mortality rate was 6.5%. Two patients died of acute liver failure, and the patient other died of severe pulmonary infection.
After surgery, 69.8% of patients in the surgery group received radiotherapy, 37.2% received chemotherapy, and 14% received hormone therapy, and the corresponding percentages for the nonsurgery group were 90%, 32%, and 14%, respectively. One month after treatments, 53.5% of patients in the surgery group had pain control compared with only (30%) of patients in the nonsurgery group.
There were no statistically significant differences in demographic or clinical characteristics between the 2 groups at baseline (Table 1). The Frankel score was used to assess neurologic impairment. In this scoring system, scores from E to A suggests increasing impairment of neurologic function.15 At the beginning of the study, 4 patients (8.7%) in the surgery group and 3 patients (6%) in the nonsurgery group had no perceivable neurologic impairment (Frankel score, E), 34 patients (73.9%) in the surgery group and 37 patients (74%) in the nonsurgery group had some neurologic impairment and were still able to mobilize (Frankel score, D), and 8 patients (17.4%) in the surgery group and 10 patients (20%) in the nonsurgery group presented with paraparesis or paraplegia (Frankel score, C-A). All patients who had little or no neurologic impairment (Frankel score, D and E) all experienced intractable pain. In total, 40 patients (87%) in the surgery group and 41 patients (82%) in the nonsurgery group presented with pain ranging from “a little bit” to “very much.”
|Characteristic||Surgery Group (n=46)||Nonsurgery Group (n=50)||Pa|
|Mean±SD||54.9 ± 8.9||57.4 ± 9.3|
|Frankel score at entry||.525|
|No. of spinal metastases||.648|
|Pain at entry||.504|
|Total QOL score at baselineb||.282|
|Mean±SD||45.3 ± 4.4||44.5 ± 4.0|
Descriptive statistics are listed in Table 2 and were based on total FACT-G scores and all 4 specific life domains scores at the baseline assessment and at the subsequent assessments. At baseline, the QOL scores for both groups were similar: The mean total score was 45.3 for the surgery group and 44.5 for the nonsurgery group (P = .282). In summary, patients in the surgery group had significantly higher QOL scores over time compared with the nonsurgery group, especially at the 3-month follow-up assessment (mean score, 59.3 vs 40.1, respectively; P < .0001).
|FACT-G Domain||QOL Scores: Mean±SD||Pa|
|Baseline Assessment||1-Month Assessment||3-Month Assessment||6-Month Assessment||9-Month Assessment|
|Surgery (n=46)||Nonsurgery (n=50)||Surgery (n=43)||Nonsurgery (n=50)||Surgery (n=41)||Nonsurgery (n=48)||Surgery (n=40)||Nonsurgery (n=43)||Surgery (n=33)||Nonsurgery (n=34)|
|Physical well being||13.7 ± 2.2||12.8 ± 2.1||16.1 ± 2.0b||11.4 ± 2.0c||18.0 ± 2.1b||11.5 ± 2.3c||16.0 ± 1.7b||10.8 ± 1.5c||14.7 ± 1.7b||10.0 ± 1.7c||.000|
|Social/family well being||14.8 ± 2.2||14.4 ± 2.1||15.3 ± 1.8||14.4 ± 1.9||15.4 ± 1.9||13.6 ± 2.1c||15.7 ± 1.9b||14.2 ± 1.5||15.2 ± 2.0||13.8 ± 2.0||.132|
|Emotional well being||8.9 ± 2.1||8.4 ± 1.8||14.0 ± 1.9b||8.8 ± 1.5||14.2 ± 2.0b||7.1 ± 1.7c||13.5 ± 1.8b||8.7 ± 1.7||11.5 ± 1.9b||7.7 ± 1.9||.000|
|Functional well being||7.9 ± 2.4||8.8 ± 2.3||2.9 ± 2.5b||8.5 ± 1.8||11.7 ± 1.8b||7.9 ± 2.0c||12.0 ± 2.1b||7.1 ± 2.0c||14.0 ± 1.9b||7.5 ± 2.5||.000|
|Total well being score||45.3 ± 4.4||44.5 ± 4.0||58.3 ± 4.7b||43.1 ± 3.8c||59.3 ± 4.2b||40.1 ± 4.0c||57.2 ± 3.3b||40.8 ± 4.0c||55.4 ± 3.9b||39.0 ± 4.0c||.000|
Figures were generated to illustrate visually and clearly the changing trend in QOL scores. Figure 2 illustrates how the pattern of total QOL after surgery in the surgery group differed from that of the nonsurgery group. Total QOL scores for the surgery group increased gradually, reached the maximum at 3 months, then were maintained at a fairly high level; whereas total QOL scores for the nonsurgery group decreased gradually. The amplitude of change in the surgery group was greater than that in the nonsurgery group. Figure 3 depicts GP scores from the FACT-G for both groups. The changes in GP scores differed vastly between the 2 groups. A sharp increase in the GP score was noted for the surgery group, especially at 3 months, whereas a sharp decrease was observed in the nonsurgery group, in which the lowest GP score was recorded at 9 months. Similar patterns of change were observed for GS scores between the 2 groups, with no significant changes over time for both groups (Fig. 4). Differences in the GS score between the 2 groups were observed but had no statistical significance (P = .132). In the surgery group, GE scores rose and then fell but still were above baseline scores by Month 9. In the nonsurgery group, there were no significant changes in GE scores, except in Month 3, when the scores were lowest (Fig. 5). In the surgery group, the amplitude of GF scores was large, and all scores were above the baseline. In the nonsurgery group, the amplitude of GF scores was small and decreased gradually (Fig. 6).
There were significant decreases in QOL scores from baseline in almost all domains at 1 month, 3 months, 6 months, and 9 months of follow-up in the nonsurgery group. Conversely, there were significant QOL increases in the surgery group (Table 2).
The survival rates at 9 months and at 18 months were 67.4% and 58.9%, respectively in the surgery group, and 62% and 48.5%, respectively, in the nonsurgery group. There was no statistically significant difference between the 2 groups in terms of survival (P = .056). Nevertheless, Figure 7 indicates slightly increased cumulative survival in the surgery group after 9 months.
To reveal the impact of spinal surgery on the QOL of patients with spinal metastases, this prospective, longitudinal study was designed and implemented using a group of patients who did not undergo surgery as the nonsurgery control group. The results of study suggest that surgical treatment can improve patient QOL greatly but does not prolong survival.
Metastatic disease is generally incurable, and treatments usually are aimed at prolonging survival and the palliation of symptoms.16-18 QOL assessment is used to help evaluate the benefits and/or risks of any therapy. Some authors claim that the standard treatment for spinal metastases is radiotherapy and that surgery is used either as second-line treatment when radiotherapy has failed or merely for pain control.19-22 However, in recent years, with the development of surgical techniques, surgeries have been used increasingly for the treatment of spinal metastases. Can spinal surgery indeed improve the QOL of patients with spinal metastases? Although the importance of patient-reported outcome has been widely recognized, data from studies of the impact of surgery on QOL for patients with spinal metastases are far from sufficient. To our knowledge, the current study is actually the first attempt to address this issue.
The various domains in the FACT-G QOL questionnaire allow the evaluation of QOL in patients with cancer for several important aspects of life, and the total score is a good reflection of a patient's overall QOL.23 The individual QOL domain and the overall QOL outcomes for our patients who underwent spinal surgery were far better than those for patients in our nonsurgery group, except for the GS domain. A possible explanation for this exception may lie in the family ethics among traditional Chinese and in the inadequate social health insurance scheme. In China, patients with advanced cancer usually are attended by their families. Patients rely heavily on their families, relatives, and friends, regardless of their disease stage and no matter what kind of treatment they receive. This applied to both the surgery group and the nonsurgery group in the current study, and both groups had similar GS scores.
After surgery, scores on the GP, GE, and GF domains improved greatly, and these improvements were associated with the advantage of spinal surgery. For patients with spinal metastases, the main purpose of surgery was to relieve severe pain and regain or maintain the ability to walk. This may have contributed to the improvement in scores on some specific items from the 3 domains, such as “I have pain” and “I am forced to spend time in bed” in the GP domain, “I worry about dying” in the GE domain, and “I am sleeping well” in the GF domain.
The surgical perioperative mortality rate was 6.5% (3 patients), which confirmed the risk of spinal surgery. However, the mean total QOL score for the surgery group was greatly enhanced, reached the maximum at 3 months after surgery, and then remained at a fairly high level until 9 months; whereas the mean total QOL score for the nonsurgery group decreased gradually. These results suggest that surgical treatment benefits patients with spinal metastases and that chemotherapy, radiotherapy, hormone therapy, and bisphosphonate therapy should be considered as auxiliary treatments.
Another noteworthy observation is the slight decline in QOL scores at 9 months, even in the surgery group and especially in the GP and GE domains. This may be explained in part by the natural course of primary cancer and also confirms our thesis that the prevention and treatment of spinal metastases depend mainly on effective treatment of the original cancer.18
Some previous investigators concluded that surgical treatment can prolong survival in patients who undergo surgery,8 but our study did not produce the same result. There was no statistically significant difference between the 2 groups in survival (P = .056). However, a more moderate decline in survival was observed between 12 months and 22 months in the surgery group compared with that observed during the initial 12 months. Because of research schedule limitations, we did not observe all patients until death, and this is a limitation of the current study.
The patients with spinal metastases in this study were selected based on specific inclusion criteria. Therefore, the current results cannot be applied to all patients with spinal metastases and can be applied only to patients similar to those who were included in our study. In particular, from a clinical perspective, surgery was suitable for all patients in this study. However, it was up to the patient and his or her family members to decide whether to undergo surgery. To some extent, this increased the patient selection bias. We tried to balance the 2 groups at study entry. Data indicate that there was no statistically significant difference in demographic and clinical characteristics between the 2 groups at the baseline; thus, the 2 groups were considered comparable.
In conclusion, the current longitudinal, repeated, controlled evaluations of QOL outcomes in patients with spinal metastases demonstrated that their QOL improved and was maintained well during the 9 months after surgical treatment. This finding supports the use of surgery as an effective treatment for spinal metastases.
CONFLICT OF INTEREST DISCLOSURES
The authors had no financial disclosures.
- 18Bone metastases in lung cancer [article in Japanese]. Clin Calcium. 2008; 18: 455-459., .