Potential Predictive Factors for Clinical Outcomes Following Progressive Pain Reduction Associated With Conservative Treatment for Recently Diagnosed Osteoporotic Vertebral Fractures: A Case Control Study

The purpose of the present study was to elucidate a potential predictive factor for poor clinical outcomes in the conservative treatment of osteoporotic vertebral fractures. Materials and Methods This is a study of the vertebrae of 41 patients selected from 63 patients with recent osteoporotic fractures admitted to our hospital from April 2013 to March 2015. Patients with dementia were excluded. We analyzed visual analog scale (VAS: 0 ‒ 100 mm) scores and dynamic radiographic images over time for up to 12 weeks after admission. Our statistical analyses were designed to identify factors that were signicantly associated with pain prolongation. Results Logistic regression analysis using a stepwise method revealed that the independent factor signicantly associated with failure for 50% pain reduction at the nal follow-up visit was VAS reduction rate at week 3 (OR = 9.0, 95% CI 0.5–17.6, p = 0.04). Patients who responded well to treatment reported a reduction in pain during the rst 3 weeks of treatment versus nonresponders who did not. It may be possible to predict the prognosis of osteoporotic vertebral fractures early during the rst 3 weeks after hospital admission.


Introduction
In principle, osteoporotic vertebral fractures (OVF) can be treated conservatively and the clinical outcomes of treatment with rest and orthoses have been well reported [1,2] .However, there is no standardized treatment regimen and the outcome of OVF may be affected by multiple factors, such as age, sex, underlying diseases, and the type of fracture, all of which are reportedly predictive of prognosis [3]. We prioritize the process of pain reduction when assessing patients' conditions; it is among the most important objectives of treatment considering that when pain is prolonged, an underlying cause generally exists. Most patients who develop issues, such as delayed palsy, complain of severe, long-term pain. Thus, a focus on the process of pain reduction may inform clinical prognosis predictions. A rst step to verify this hypothesis is to identify how pain changes in patients who respond well to conservative treatment. The aim of the present study was to elucidate whether the an early reduction in pain can be one of the predictive factors for clinical outcomes associated with the conservative treatment for OVF.

Page 4/12
The study subjects were 41 patients (5 men and 36 women) selected from 63 patients with recently diagnosed OVFs who were admitted to our hospital from April 2013 to March 2015 (Fig. 1). Patients with dementia and those for whom clinical and imaging assessments for 12 weeks after admission were not available were excluded. Pain was assessed using a visual analog scale (VAS; 0-100 mm) score at admission and after 3, 6, and 12 weeks. The degree of vertebral instability was con rmed using dynamic radiographs of patients in standing and supine positions that are in accordance with the method of Abe et al [4]. Gross instability was de ned as a difference of ≥5 degrees in the vertebral wedge angle on a lateral radiograph with vertebral sway or the presence of a vacuum cleft. Kyphotic angle was de ned as an angle formed by the cranial end plate of the vertebral body just above the fractured vertebra and the caudal end plate of the vertebra just below the fractured vertebral body.
A favorable clinical outcome was de ned as a 50% or greater reduction in pain at the nal 12-month visit.
Failure was de ned as cases with no reduction in pain, or when surgical intervention was required. We performed statistical analyses using a 50% or greater reduction in pain at the nal visit as an objective variable and course of pain reduction (VAS at admission and VAS reduction rate at week 3, as well as the following imaging factors: vertebral instability at admission and vertebral instability, kyphotic angle, and wedge angle-all at week 3) as explanatory variables. Factors identi ed with a signi cance of p < 0.1 in the initial univariate analysis were used in further analyses. Possible confounding among explanatory factors during univariate analysis was assessed. Logistic regression analysis using a stepwise method was performed to identify possible independent risk factors which had a signi cant impact on clinical outcomes. Data are expressed as mean ± standard deviation (SD). The odds ratio (OR) for the incidence of failure and 95% con dence intervals (CI) were calculated as an approximation of the relative risk estimates. All statistical analyses were conducted using JMP 15 (SAS Institute Inc., Cary, NC, USA).

Results
The demographic characteristics of patients are shown in Table 1. The mean follow-up was 7.6 months. The mean VAS score improved over time from 67 mm at admission to 31 mm at week 3, 24 mm at week 6, and 20 mm at week 12. The mean vertebral instability also improved from 4.39 degrees at week 3, 2.86 degrees at week 6, and 2.21 degrees at week 12. Four of the 41 patients required surgery (percutaneous vertebroplasty in three cases and a posterior spinal fusion in one case) owing to severe pain which was classi ed as a reduction in pain of less than 50% since admission. In addition, there were four cases with no improvement in pain (Fig. 2). Using a univariate analysis, the VAS reduction rate at week 3, vertebral instability at admission, and vertebral instability at week 3 were identi ed as potencial signi cant factors (p < 0.1) for predicting failure and were used in subsequent analyses. Logistic regression analysis revealed that the independent factor signi cantly associated with failure to achieve ≥ 50% reduction in pain by the nal follow-up visit was a VAS reduction rate at week 3 (OR = 9.0, 95% CI 0.5-17.6, p = 0.04) ( Table 2). The receiver-operator characteristic analysis revealed that the cutoff value for the VAS reduction rate at week 3 was 27% or grater (Fig. 3), indicating that patients achieving a 27% reduction in pain at week 3 had a 50% or grater reduction in pain by the nal follow-up visit.

Discussion
In the present case series, we determined that a poor pain reduction rate (~ 27%) 3 weeks after the diagnosis of an OVF that caused pain was a factor signi cantly associated with inadequate pain relief at nal follow-up (14 months). Until now, few studies have addressed the process of pain recovery in patients with OVF treated with conservative therapy. Based on the experience of individual doctors in clinical practice or at hospitals managing these patients, wide variations in the initial treatments for OVF -such as drug administration, bracing, bedrest for a set duration, admission to a hospital, and so forthare among the possible reasons for a lack of reports addressing the course of pain reduction in the past. This wide variation makes it challenging to compare the actual time course of pain reduction between patients who received different therapeutic regimens. In contrast, in this case series we applied a conservative standardized treatment regimen that included 2 weeks of bedrest followed by rehabilitative treatment that required a rigid Jewett spinal brace to be worn; this enabled us to analyze the course of pain reduction over time. We consider this one of the strong points of our present study.
Toyoda and colleagues performed a stepwise cluster analysis using VAS scores and identi ed four different patterns of lumbar pain progression after OVF [5]. In that report, cluster 4 included the patients who initially exhibited moderate to strong pain without pain reduction over the clinical course. Another cluster included patients with initial moderate pain that was rapidly attenuated. In the present study, patients in the group who experienced an unfavorable level of pain reduction (< 50% at 3 weeks similarly exhibited an inadequate reduction in their pain level 12 weeks after hospital admission for OVF. Results at the nal follow-up visit largely overlap with the cluster 4 results in the report of Toyoda et al. Both reports showed that a failure to adequately relieve pain in the relatively early phase of treatment is signi cantly associated with inadequate pain relief at nal follow-up. In other words, the pain reduction rate at a relatively early phase of OVF could be one of the possible predictive factors for the prognosis of pain reduction. Further exploration with a larger prospective cohort in the near future is prerequisite to being able to elucidate the true usefulness of using pain mitigation early in the course of treatment as a predictive factor for pain prognosis in patients with OVF.
Previously, Toyone et al. reported that vertebral instability in the early phase after injury was signi cantly associated with pain reduction at nal follow-up [6]. In the present study, univariate analysis revealed that vertebral instability 3 weeks after diagnosis of OVF was signi cantly associated with nal outcomes; this aligns with the report of Toyone and colleagues. A possible reason for a discrepancy between the two studies is the cutoff of the dichotomous objective variable, ≥ 50% pain reduction. Therefore, it might be considered premature to conclude that vertebral instability is unassociated with nal outcome.
The present study has several strengths and limitations. We consider that the ease of evaluation is a strength of our study in addition to only evaluating pain intensity after an appropriate period of conservative therapy. Furthermore, the conservative standardized treatment regimen lends strength to our results. In contrast, the latter approach could hinder the generalizability of our results to patients receiving other types of conservative treatment. The limitations of this study included the small number of patients, a short follow-up period, and the fact that we did not consider other factors associated with the clinical course, such as fracture type and underlying disorders. The retrospective nature of the present study is another possible limitation given that we could not consider the patients who discontinued the study.
Currently, we apply conservative standardized therapy regimens to almost all patients with OVF as a way of screening. If the reduction in pain achieved via initial conservative treatment is inadequate, we consider other treatment options that include surgery. Our study results are encouraging and support our current treatment strategy.

Conclusion
Pain reduction rate 3 weeks after initiating conservative treatment for a newly diagnosed OVF is among the potential predictive factors for a prognosis of inadequate pain reduction at nal follow-up.
I con rm that I understand BMC Musculoskeletal Disorders is an open access journal that levies an article processing charge per articles accepted for publication. By submitting my article I agree to pay this charge in full if my article is accepted for publication.
No, I declare that the authors have no competing interests as de ned by BMC, or other interests that might be perceived to in uence the results and/or discussion reported in this paper.
The results/data/ gures in this manuscript have not been published elsewhere, nor are they under consideration (from you or one of your Contributing Authors) by another publisher.
I have read the BMC journal policies on author responsibilities and submit this manuscript in accordance with those policies.
All of the material is owned by the authors and/or no permissions are required.    Clinical outcomes at the 12-week assessment by fractured vertebral levels (n = 37). L, lumbar; T, thoracic; VAS, visual analog scale