Introduction

Low back pain (LBP) is a common global cause of morbidity (1). It is a leading cause of hospital admission and healthcare spending (1,2). It is estimated that up to 60-90% of the general population will develop low back pain in their lifetime (1, 2). Because of this magnitude of the disease, LBP is an important cause of workplace absenteeism and a leading consumer of healthcare funding (1, 2). It is estimated to cost over

149 million lost workdays and 100 to 200 billion dollars per year in the United States (3). As a result, early symptom relief and resumption of daily activity is always a priority. Management of non-specific acute low back pain (NSALBP) entails pharmacological and non- pharmacological interventions for pain control and back rehabilitation with the aim of rapid return to normal activity (1) Lumbosacral corset (also known as support or brace) as a form of physical therapy for acute low back has been used with varying results (1-3). Potential benefits of lumbosacral corsets include stabilization and reduction of spine motion, reduction of mechanical loading; and provision of the miscellaneous effects of massage, heat and as a placebo (1,4,5). They are also thought to protect the spine by avoiding extreme trunk motion and offering tactile biofeedback (5). There is a paucity of data on the use of lumbosacral corset for acute low back pain. Existing literature on the use of a lumbosacral corset is deficient in high-quality randomized trials assessing the importance of this modality of treatment on acute low back pain with regard to pain relief and disability improvement (6). No studies exist locally to shed light on the same. The decision to use or not use a corset is made on a physician by physician basis; sometimes dependent on the patient's request or the physiotherapist input.

​A study assessing the impact of additional lumbosacral corset to other modes of therapy should, besides adding to the body of knowledge; form evidence-based reference locally and guide the choice of intervention for patients with NSALBP. To the authors’ knowledge, this is the first study in English literature addressing the use of lumbosacral corset in patients with NSALBP.

 

Methods

This study aimed to assess the outcome (change in back-specific disability) of patients with non-specific acute low back pain treated with a lumbosacral corset in addition to a standard analgesic protocol compared to those on a standard analgesic protocol alone over 3- week duration at the Aga Khan University Hospital, Nairobi, Kenya. The secondary objectives were to measure the change in pain score, quantify days off duty due to back pain, describe duration to maximal change in pain and disability and to assess the patients’ compliance to use of a lumbosacral corset.

​Eighty-two patients were randomly assigned to either corset or no corset arms by picking opaque envelopes. The primary outcome was the change in back-specific disability as measured on the Oswestry Disability Index (ODI). Secondary outcomes included pain as assessed by Visual Analogue Score (VAS), number of days off duty and number of days off analgesia. The patients were followed up weekly for three weeks. At each point, their ODI score was calculated and recorded, as were other secondary outcomes. The research protocol has been submitted separately for publication elsewhere.

​

Results

A total of 90 patients were assessed for eligibility, out of which 82 were recruited into the study. Three subjects (2 intervention, 1 control) were lost to follow up leaving 79 patients for the final analysis of the primary outcome (38 Intervention, 41 control). Four patients required admission for one day (3 intervention and 1 control). Of these four, one patient (control arm) was admitted at the end of the first week due to persistent pain and given parenteral analgesics, while 1 patient (intervention arm) required an MRI scan due to worsening pain. She was found to have nerve root compression from a prolapsed intervertebral disc and she underwent microdiscectomy.

​Baseline characteristics of the control and intervention groups were comparable. Sex and occupation were also similar between groups (p=0.363 and 0.643 respectively) (Table 1).

​

​

                                                                                                     Click To View Table 1

 

 

The results show that there was a significant effect of time on the ODI, F= (1.38, 106.56) = 207.89, p=.000. These results suggest that with the standard treatment protocol, with or without the lumbosacral corset, ODI improved over time. Between-subject effects showed a significant difference between the groups, (F (1,1=4.23, p=0.043) suggesting that patients in the intervention group had a greater improvement in ODI over the study period. (Table 2, Figure 1)

 

 

​

                                                                                                     Click To View Table 2

​

 

                                                                                                     Click To View Figure 1.

Mauchlys’ test indicated sphericity did not hold, (χ2 = 64.03, p=.000). Correction with Greenhouse- Geisser correction (ε=0.461) revealed a significant effect of time on the Visual Analog Score for pain, (F= (1.98, 152.39) = 381.21, p=.000), suggesting an improvement in the VAS with time among subjects in both arms. There was however no significant difference between the control and intervention arms (F (1, 78) =0.189, p=0.665). (Figure 2).

​

                                                                                                     Click To View figure 2

​

There was no statistical difference in days off duty from work due to back pain between the control and intervention groups (F1 = .000, p=.986). These data are summarized in Table 2.

​Days off analgesia were calculated for the first one week. Data was complete for 22/79 patients. No statistical difference was detected between the arms (F1 = .837, p=.371) (Table 2). Change in disability (ODI) and pain scores were similar between the intervention and control groups as summarized in Table 3. Mean duration of time off corset was 0.76 days (SD 0.863) (n=37). This was not significantly correlated with sex, F (1, 36) = 0.269, p= .607 nor age, F (1, 36) = 3.76, p=.061.

​

                                                                                                     Click To View Table 3

​

Discussion

 

This study established a positive correlation between change in both ODI and VAS and time, such that both the back specific disability and the pain score reduced over time in both arms. This was statistically significant. Our findings are in agreement with other authors who have observed that 90% of patients with NSALBP will heal within six weeks (7). In fact Calmels et al argue that in some cases 50% of patients with NSALBP will recover in one week and 90% within two weeks (1). In a randomized clinical trial involving patients with all classes of LBP, Valle-Jones et al evaluated pain control, analgesic requirements and resumption of duty among patients treated either with analgesics alone or those with additional lumbosacral corset over 21 days and demonstrated a similar trend (8). This shows that in general, Non specific LBP resolves quickly over time.

​For the primary outcome, this study found that the use of the lumbosacral corset in addition to analgesic protocol was associated with a statistically significant improvement in back specific disability as assesed via the ODI. Other researchers have demonstrated better outcome using various tools on different subsets of patients with LBP (1, 8). In a multicentre randomized trial, Calmels and colleagues found that the use of elastic lumbar belts among patients with sub acute LBP significantly reduced the pain score, analgesic requirement and back disability over a 90 day duration(1). The duration of follow-up for Valle-Jones et al trial was 21 days similar to ours. They too demonstrated earlier resumption of work in the lumbosacral corset arm but they did not use any back specific disability scoring tool (8). Their sample included patients with acute, sub acute and chronic LBP. Ours was the first study to investigate this on patients with NSALBP only. Different possible mechanisms by which lumbosacral corsets act have been postulated (5). It is not possible to describe how the corset in this study acted to improve the disability outcomes, and whether one or all mechanisms could be applicable. As stated earlier, it would be hypothesized that the effect of corset may have been by reduction of extreme trunk motion via tactile biofeedback (5, 9). It is also possible that the presence of a lumbosacral corset gave the patients the courage to initiate back activity earlier compared to the control arm patients who might have feared to worsen their pain and hence limited their back activity. It would be interesting to find out if there is a specific domain on the ODI tool that persistently changed more in the intervention arm to sway the results of the cumulative ODI score.

​Regarding the secondary outcome of change in pain score, there was no statistically significant difference between the arms. In the same setting and using a similar methodology, Mohammedali et al did not demonstrate any difference in pain score between NSALBP patients treated with analgesics alone compared to those treated with additional ultrasound therapy (10). This means the lumbosacral corset did not contribute to pain reduction in our study population. This is in contradistinction to the ODI score. Pain severity is the first component of the 10-section ODI tool. In fact all the other 9 sections assessing various domains of back related disability are constructed assuming that pain is the limiting factor (11). Whether it is actually pain itself or the fear of anticipated pain that limits patients’ activity along these 9 domains is not easy to establish. The findings of our study may indicate that it is the fear of hurting one’s back that would limit their activity, and not the back pain itself. Thus once the pain was relieved by analgesia, the patients with a corset may have felt more confident to engage in activities that required the use of their back while those on analgesics alone were still cautious even with good pain control. It could as well be from various other mechanisms that have been postulated to explain the effects of lumbosacral corsets (5,9,11). Regarding days off duty, data was available for 68/79 patients analyzed (86%). The days {mean, (SD)} off duty were 1.93 (3.5) and 1.9 (3.05) for the intervention and control arms respectively. The difference was not statistically significant (F1 = .837, p=.371). This means that NSALBP kept patients off duty on average for two days. This concurs with our earlier finding of improvement over time in both back specific disability and pain score, indicating that patients recovered quickly enough to be able to resume duties. As highlighted by other authors, NSALBP generally runs a favorable prognosis with patients improving as early as one week (1,7,8). Our study was not however powered to detect differences in days off duty and thus we cannot draw strong conclusions in this regard. The maximal change in pain score and ODI occurred after a week of treatment. There was no statistically significant difference between the two arms. This means that irrespective of the arm, the treatment initiated brought about a clinical improvement. Mohammedali et al demonstrated the same finding in the same setting (10). The study was however not powered to assess these secondary outcomes so strong conclusions cannot be drawn.

​Compliance to the use of the corset was good with an average of 1 day off the corset. We notice that this information may have been subject to information bias from patients. The lumbosacral corset is fairly tolerated as it has no side effects apart from minor discomfort or skin irritation (8). As observed by other authors, patients actually approve of lumbosacral corsets, hence the good compliance (8).

​This study had several limitations. It depended on patients’ information about days off duty and compliance to use of the lumbosacral corset. This could have suffered from recall or information bias. We also did not stratify patients on other confounders of NSALBP such as type of mattress used and body mass index. Although randomization was supposed to cater for these, they remain as our main limitations. The study, however, draws strength in being the first to examine the effect of the lumbosacral corset among patients with NSALBP.

 

Conclusion

We have demonstrated that the use of lumbosacral corset in addition to analgesic protocol among patients with NSALBP results in a significant improvement in their back specific disability compared to use of analgesia alone. The precise mechanism by which this occurs should be the subject of future trials. Future trials should also control for various confounders and be done across multiple centers.

 

References

1. Calmels P, Queneau P, Hamonet C et al. Effectiveness of a Lumbar Belt in Subacute Low Back Pain. Spine. 2009; 34(3):215–20

2. Risks of Current Treatments. Curr Med Res Opin. 2010;26(1):179–90

3. Jellema P, Van Tulder MW, Van Poppel M et al. Lumbar Supports for Prevention and Treatment of Low Back Pain. Spine (Phila Pa 1976). 2001; 26(4):377–86.

4. Van Poppel MN, de Looze MP, Koes BW et al. Mechanisms of action of lumbar supports: a systematic review. Spine (Phila Pa 1976). 2000;25(16):2103–13.

5. Van Duijvenbode I, Jellema P, van Poppel M et al. Lumbar Supports for Prevention and Treatment of Low Back Pain. Cochrane Database Syst Rev. 2008; 2.

6. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976). 2000; 25(22):2940–52

7. Costa LCM, Maher CG, Hancock MJ et al. Prognosis of Acute and Persistent Low Back Pain. CMAJ 2012; 184(11): 613-24

8. Valle-Jones JC, Walsh H, O’hara J et al. Controlled Trial of a Back Support (‘Lumbotrain’) in Patients with Non-specific Low Back Pain. Curr Med Res Opin 1992;12(9):604–13.

9. Steele J, Bruce-Low S, Smith D. A Reappraisal of the Deconditioning Hypothesis in Low Back Pain: Review of Evidence From a Triumvirate of Research Methods on Specific Lumbar Extensor Deconditioning. Curr Med Res Opin. 2014;30(5):865-911.

10. Mohamedali S, Mutiso SK, Oroko P et al. Effect of Additional Ultrasound to Analgesics in Treatment of Acute Low Back pain: A Randomized Controlled Trial Conducted at The Aga Khan Univeristy Hospital, Nairobi (AKUH,N). SA Orthopaedic Journal. 2014; 13(3):64-9.

11. Fairbank JC, Pynsent PB. The Oswestry Disability Index. Spine (Phila Pa 1976). 2000; 25(22):2940–5