Article Text
Abstract
Background and purpose There has been no prospective evaluation of vertebroplasty using a validated instrument. We describe the pain and functional status of 72 patients before and after vertebroplasty, as prospectively evaluated by the Vertebral Compression Fracture Pain and Functional Disability Questionnaire.
Methods Of 161 consecutive patients, 72 consented to participate in the study and self-completed the questionnaire prior to undergoing vertebroplasty. Differences in pain and distress before and after vertebroplasty, and between the first and second follow-up intervals, were evaluated. Mean scores for each of 24 activities of daily living (ADLs) were plotted at the baseline and first and second follow-up intervals.
Results The mean (SD) patient age was 74 (10) years; 80% were female. On the 0 (no pain) to 10 (pain as bad as it could be) visual analog pain scale, patients reported significantly more pain, on average, before undergoing percutaneous vertebroplasty (PV) than at the first follow-up interval (mean 5.8 vs 3.5, p<0.001). The reduction in reported pain following vertebroplasty persisted at the second follow-up on both the visual analog and adjectival pain scales. Among the 24 ADLs, between 25% and 69% of patients reported a mean improvement of at least 1 level on the 5-point ADL scale, and between 14% and 55% reported a mean improvement of at least two levels. The majority of the improvement in reported functional status following vertebroplasty was sustained at the second follow-up interval.
Conclusion PV resulted in substantial, lasting reduction in pain and improvement in ability to perform ADLs.
- Spine
- Interventions and basic science
Statistics from Altmetric.com
Vertebral compression fractures (VCFs) are the most common complication of osteoporosis. These fractures frequently result in debilitating pain and impaired functional status and quality of life. Percutaneous vertebroplasty (PV) is a treatment modality originally developed in France in the late 1980s,1 and first performed in the USA in the mid-1990s.2 3 The procedure involves percutaneous injection of bone cement (polymethylmethacrylate(PMMA)) into collapsed vertebrae, and has been used for treatment of patients with osteoporosis who have prolonged pain following vertebral fracture, as well as those with aggressive hemangiomas and osteolytic neoplasms.
Several recent literature reviews on the efficacy of PV have concluded that the procedure, when used in the setting of osteoporotic compression fractures, results in substantial and immediate pain relief, improved functional status, and minimal short-term complications.4 5 6 7 However, despite this overall endorsement of the procedure, to our knowledge, all published reports to date2 8 9 10 11 12 13 14 15 16 17 18 19 suffer from at least one of three primary methodological limitations: (i) small sample size; (ii) retrospective assessment of patient status (pain and functional ability); and (iii) assessment of patient status using either non-validated instruments (eg, self-developed) or validated instruments non-specific to VCFs.
Recently, members of our group reported high levels of internal consistency reliability, and moderate-to-high criterion and construct validity, for the Vertebral Compression Fracture Pain and Functional Disability Questionnaire,20 an instrument developed specifically to assess pain and functional status among patients with vertebral fractures. We now report the pain and functional status of the 72 patients we evaluated in the development of the Vertebral Compression Fracture Pain and Functional Disability Questionnaire.
Methods
Patient population
Following an institutional review board-approved protocol, a series of 161 consecutive patients treated with PV were invited to enroll in the study which consisted of completion of the Vertebral Compression Fracture Pain and Functional Disability Questionnaire. In addition to the ability to read English, the only additional inclusion criterion was patient suitability and treatment for vertebroplasty.2 This included radiographic confirmation of one or more new or progressive VCFs, diagnosis of moderate or severe back pain and/or functional disability as determined from patient history and clinical examination, ability to tolerate lying prone for 1–2 h, and no symptoms secondary to spinal stenosis, retropulsed fragments, or disc herniation. Of 161 consecutive patients, 72 consented to participate in the study and self-completed the Vertebral Compression Fracture Pain and Functional Disability Questionnaire prior to undergoing PV. Post-procedure questionnaires were then mailed to the participant at 1 and 6 weeks, respectively, along with a self-addressed stamped return envelope. Instructions were printed at the top of each questionnaire that simply stated each questionnaire would take approximately 1.5 h to complete, and to complete all questions as they appeared on the form. Seventy-one (99%) and 68 (94%) of the 72 study patients completed the questionnaire at first and second follow-up intervals that occurred at a median time of 20 days and 49 days, respectively, after undergoing vertebroplasty. In addition, all study participants completed the Oswestry Low Back Pain Questionnaire21 that was used to help validate the Vertebral Compression Fracture Pain and Functional Disability Questionnaire, and is not described in the present report.
Description of questionnaire
The Vertebral Compression Fracture Pain and Functional Disability Questionnaire (hereafter “the questionnaire”) is designed to collect information on patient pain and functional status before and after PV. Briefly, the first section of the questionnaire consists of 11 items that address: (i) the patient's prior and current magnitude of back pain and distress experienced; (ii) overall general health; (iii) use of prescription medications for non-fracture conditions; and (iv) expected relief of back pain to be achieved from treatment. The five “pain” and two “distress” items (due to the patient's back condition) are measured using either a visual analog scale (range 0–10) or an adjectival scale (range 0–8). The remaining four items are measured using either a 5-point Likert scale or dichotomous yes/no responses. In the second section of the questionnaire, respondents rate their ability to perform 24 individual activities of daily living (ADLs) on a 5-point scale ranging from 1 (absolutely unable to do because of pain) to 5 (able to do without pain). In addition, there is a response box to identify if the patient does not perform a particular ADL. The 24 ADLs, as reported on baseline and follow-up questionnaire administrations, have demonstrated high internal reliability consistency (Spearman rank coefficient: 0.87–0.98) and modest to high criterion and content validity with the pain and distress items on the questionnaire and dimensions of the Oswestry Low Back Pain Questionnaire.20 Further description of the content and psychometric properties of the questionnaire have been reported.20
Description of vertebroplasty procedure
The PV procedure was performed in accordance with standard procedures. Briefly, the patient was placed in the prone position on an angiography table under sterile conditions. Neuroleptic analgesics were administered. The treated vertebral body was localized under fluroscopic control and the skin overlying this area was prepped and draped. The skin over the center of the pedicle was anesthetized, followed by a deep injection of bupivacaine to and including the periosteum. A small skin incision was made and a disposable 11-gauge bone biopsy needle was positioned with its tip in the center of the pedicle as visualized on fluoroscopy and advanced until the stylet tip abutted the bone. Lateral fluoroscopy was used to advance the needle through the pedicle and into the anterior one-third of the vertebral body. The dead space of the needle was then filled with saline, and prepared polymethylmethacrylate injected into the vertebral body. Injection of the material was performed under lateral fluoroscopy, and was continued until at least two-thirds filling of the vertebral body was achieved, no more material could be pushed into the body, or extravasation into veins or the disk space was noted. Following the procedure, patients remained immobilized on strict bed rest for 1 h, and were subsequently discharged upon ability to ambulate and tolerate per orem intake.
Statistical analysis
Differences in patient self-reported pain and distress before and after PV, and between the first and second follow-up intervals, were evaluated by paired t tests. Mean scores for each of the 24 ADLs (based on the 1–5 scale) were plotted at the baseline and first and second follow-up intervals. In addition, percentages of patients who improved at least one or at least two levels on the ADL scale after undergoing PV were determined, as were the percentages of patients who worsened by one or two levels on the ADL scale between the first and second follow-up intervals. Finally, to gauge relative patient improvement in functional status, percentages of patients who reported being able to perform each ADL without pain or with only mild pain were plotted for the baseline and first follow-up questionnaires. All analyses were completed using the SAS system, version 8.1 (SAS Institute, Cary, North Carolina, USA).
Results
Demographic and clinical characteristics
The mean (SD) age of the 72 patients was 74 (10) years; 80% were female and 100% were Caucasian. Osteoporosis was the primary indication for PV in 89% of all patients with the remaining 11% attributed to malignant tumor, benign tumor, or spinal instrumentation preparation. Prior to PV, 13% of patients could not ambulate at all and 59% could not ambulate more than one block. In addition, 15% of patients used a wheelchair and 7% were bedridden. Thirty-nine percent of patients had a history of cancer, and 10% had a history of hip fracture. Finally, on a scale of 1 (none) to 10 (severe), physician assessment of pain and disability reported mean (SD) scores of 6.8 (1.9) and 6.1 (2.4), respectively.
Procedural characteristics and complications
Both local and neuroleptic anesthesia were used in the majority of all patients. Patients had a mean (SD) of 2.8 (2.2) vertebral fractures identified, of which, 1.7 (1.0) were treated with PV. Multiple vertebral fractures were present in 63% of all patients, and 44% of patients had more than one fracture treated with PV. Eighty-six percent of patients reported excellent tolerance of the procedure, and the remaining 14% reported good tolerance. There were no symptomatic complications. Technical complications included three patients with paravertebral vein filling, one patient with a small amount of PMMA in an epidural vein, and one patient with a tiny PMMA embolism to lung. Thus, the procedural complication rate (which was limited to these asymptomatic cement leaks) was 7%.
Reported pain and distress before and after vertebroplasty
On the 0 (no pain) to 10 (pain as bad as it could be) visual analog pain scale, patients reported significantly more pain, on average, before undergoing PV than at the first follow-up interval (mean 5.8 vs 3.5, p<0.001) (fig 1A). Similar results were reported on the 0 (none) to 8 (very intense) adjectival pain scale (mean 5.7 vs 3.8, p<0.001) (fig 1B). Importantly, the substantial reduction in reported pain following PV persisted at the second follow-up on both the visual analog and adjectival pain scales. Essentially identical results were observed for the measure of distress on both the visual analog and adjectival distress scale (fig 1C,D).
Ability to perform ADLs before and after vertebroplasty
The ability to perform each of the 24 ADLs without pain or with only mild pain was substantially greater at both follow-up intervals compared with before undergoing vertebroplasty (fig 2, middle section). This improved functional status occurred irrespective of the level of difficulty prior to surgery. Substantial mean improvements were reported for all ADLs with the exception of the activity “doing gardening”. Among the 24 ADLs, between 25% and 69% of patients reported a mean improvement (between the baseline and first follow-up interval) of at least one level on the 5-point ADL scale, and between 14% and 55% reported a mean improvement of at least two levels (fig 2, right side). The reported worsening of ability to perform ADLs between the first and second follow-up intervals (ie, lack of sustained benefit) ranged from 0% to 36% of patients for at least one level, and from 0% to 14% of patients for at least two levels. Thus, the majority of the substantial improvement in reported functional status following vertebroplasty was sustained at the second follow-up interval.
Figure 3 graphically depicts the percentages of patients able to perform the individual ADLs without pain or with only mild pain at baseline and at the first follow-up interval. More than half of all patients were able to “wash dishes” and “use the toilet” without pain or with only mild pain before undergoing vertebroplasty. Thus, the magnitude of potential improvement in these ADLs following vertebroplasty was limited and was reported as modest. On the other hand, substantial improvements in the ability to perform ADLs without or with only mild pain were reported for nearly all of the 22 remaining ADLs in which at least 50% of patients reported at least moderate pain before undergoing vertebroplasty. The greatest relative improvements in patient ability to perform ADLs either pain free or with only mild pain were for the activities “social activities”, “doing laundry”, “getting in/out of bed”, “preparing meals”, “lifting light objects”, and “doing heavy housework”.
Discussion
Epidemiological estimates project that the incidence of osteoporotic insufficiency fractures will increase fourfold in the next 50 years, and the most common fracture associated with osteoporosis is the vertebral compression fracture.22 Annually, these fractures account for 161 000 physician office visits and more than 150 000 hospitalizations.23 24 The traditional treatment for VCFs is “conservative therapy”. Yet conservative therapy is not without risk. During bed rest virtually every organ system is adversely affected and these effects tend to be more pronounced in older patients who have less reserve than younger patients. For patients on bed rest, the incidence of deep vein thrombosis increases dramatically and fatal pulmonary embolism occurs in 5–10% of patients. Muscle strength declines 1–3% per day and bone density declines approximately 2% per week, a serious concern in patients already suffering from osteoporosis and unlikely ever to regain the lost bone mass. Adverse gastrointestinal, cardiac, urologic, and even neurologic effects have also been well documented.25 26
Given the adverse effects of bed rest and the high cost of hospitalization for VCFs (540 million annually), an alternative therapy would be welcome.27 A recent study suggests that PV facilitates rapid ambulation and early hospital discharge of patients admitted for refractory back pain from VCFs.28
This study demonstrates substantial improvement in pain and functional status following PV. As such, it supports the findings of the few other prospective reports in the literature.17 18 19 The strength of this study is that it uses a validated instrument specific to back pain in the elderly, and that the follow-up times are reasonably uniform. Weaknesses of the study include the absolute numbers of subjects (low compared with the Do study19) and the number of screened patients who agreed to be in the study (72 out of 161). However, the percentage of patients enrolled who responded to both the first and second questionnaires was high (99% and 94%, respectively).
The magnitude of effect in this study, while substantial, is not as large as most earlier studies. The patients in this study reported, on average, a lower initial pain score than in previous studies and this lower degree of initial pain likely explains the more modest effect observed.
The ADLs studied in this report are the types of activities most commonly performed by patients in this age group (eg, preparing meals, doing laundry, lifting light objects, social activities), and substantial improvements were observed at both the early and the later follow-up intervals. Although the time period to second follow-up was relatively short, the results are encouraging in that reported improvements in pain and functional status were generally sustained, at least in the short term. Thus, PV appears to be a promising treatment to substantially improve overall quality of life.
Finally, the greatest weakness of this study, as well as almost every other study of PV that precedes it, is the lack of an adequate control group. Two previous reports include medical control groups and report positive results, but it is well known that placebo effects occur after interventions in the spine.29 30 Thus, randomizing patients to PV or to sham surgery might provide a better control group, and such a trial is currently ongoing, the NIH-sponsored INVEST trial.31 Given the prevalence of this disease and the healthcare dollars at stake, level one evidence that PV is an effective procedure, would be a welcome addition to the medical literature.
References
Footnotes
Competing interests Dr Evans receives royalties from Cardinal Health Inc. This company and these products are not mentioned in this article, but can be used to perform vertebroplasty.
Ethics approval Ethics committee approval was obtained.