Vertebral fractures are a common consequence of osteoporosis. Annually, there are 400,000 osteoporotic vertebral fractures (OPVFs) occurring in the European Union 6 (United Kingdom, Italy, France, Germany, Spain, plus Sweden)1 and this number is expected to increase by 23% by the year 2030.1 Typically, OPVFs result in significant and persistent back pain, spinal deformity, and reduced mobility and quality of life.2-4 OPVFs also are a marker of poor bone health and predict future fractures, both vertebral and non-vertebral.5-6 Lindsay et al5 reported that a single vertebral fracture is associated with a five-fold increase in further vertebral fractures. OPVFs are also a burden to the community with the mean cost of care for the 12-month period following an OPVF ranging from €1928 to €14,474 (approximately 13,156 BRL to 98,763 BRL).1

Surgical and pharmacological interventions commonly used for the management of OPVFs have limited efficacy and potential risk of harm. A recent Cochrane review7 reported moderate to high certainty evidence that vertebral augmentation, a percutaneous injection of bone cement under image guidance into a fractured vertebra for patients with OPVFs (acute or subacute), showed no important change for pain, quality of life, or disability when compared to sham treatments. The authors reported uncertainty regarding risk of this procedure, including risk of future fracture. Another systematic review showed very low certainty evidence that opioids and non-steroidal anti-inflammatory medications provide acute pain relief for people with OPVFs when compared to alternative medicine.8 Concerningly, opioids showed no significant effects for pain relief when compared to placebo, and were associated with more adverse events such as gastrointestinal disorders.8

In comparison to vertebral augmentation and analgesia or anti-inflammatory medication, non-surgical and non-pharmacological management is non-invasive and might be expected to result in less harm. Further, these interventions (e.g. exercise therapy, bracing) may address problems associated with OPVFs including pain and limited activity. Previous systematic reviews have shown very low certainty evidence that exercise interventions for people with OPVFs7-9 may have small benefits for pain, quality of life and physical performance and few adverse events. Rzewuska et al's10 systematic review demonstrated low certainty evidence for the use of spinal orthoses for managing pain and disability in patients with OPVFs in the medium-term and very low certainty evidence for the effectiveness of a rigid orthosis compared to a soft spinal orthosis in reducing pain in the short-term.

Currently, there is insufficient evidence about the benefits and harms of non-surgical and non-pharmacological interventions to guide clinicians, which may lead to variable and suboptimal care for patients with OPVFs. Previous systematic reviews have not been comprehensive in their inclusion of a range of non-surgical and non-pharmacological interventions for OPVFs and little emphasis has been given on synthesising the evidence on harms as well as benefits.7-10 As literature searches in other reviews have not been updated since 2017 there is a need to consider new randomized controlled trials completed since.

This systematic review aims to evaluate the evidence from randomized controlled trials on the benefits and harms of non-surgical and non-pharmacological management of people with OPVFs compared with standard care (control); and, to compare the benefits and harms of non-surgical and non-pharmacological management of people with OPVFs compared with an alternative non-pharmacological, non-invasive intervention.

A systematic review and meta analyses were conducted following the PRISMA Statement.11 This systematic review was prospectively registered on the PROSPERO database (CRD42012002936).

Meta-analyses22-23,25-26,29-31 provided low certainty evidence that exercise interventions compared to no exercise were effective in reducing pain in people with OPVFs. Meta-analyses34,36 provided low certainty evidence that rigid bracing intervention compared with no bracing was effective in reducing pain in patients with OPVFs. No health-related quality of life or activity improvements were demonstrated for exercise interventions, bracing, electrotherapy, or multimodal interventions. Meta-analyses showed low to moderate certainty evidence of no differences in harms comparing exercise and no exercise groups. Our review includes six trials not included in previous reviews and systematically evaluated the effect of non-surgical and non-pharmacological management of OPVFs on harms as well as benefits.

One explanation for why exercise and bracing reduced pain could be both these interventions reduced loading on the spine. A common component of the programs included back extension strengthening or postural control, which reduces flexion and loading on the spine. Takahashi et al.44 found increased load during trunk flexion in healthy adults. Rohlmann et al.45 demonstrated high forces in spinal segments in activities of daily living that involved moving the center of mass anteriorly for example upper body flexion, lifting, or stair walking in patients with instrumented vertebral body replacements. Interestingly, walking was ranked as having the lowest resultant force (N). Pain in people with OPVFs is typically increased during loading activities such as bending and eased when lying down.46

While the results on outcomes for exercise and bracing appear promising for reducing pain, their clinical significance requires further consideration. For example, effect of exercise versus no exercise on pain, (MD = 1.01 units, 95% CI: 0.08, 1.93), is less than the minimum clinically important difference of 1.8 units.47 The pain outcome for bracing (MD = 2.61 units, 95%CI: 0.95, 4.27) exceeds the minimum clinically important difference; however, the lower band of the confidence interval is less than 1.8 units suggesting there remain some uncertainty about the clinical significance of the effect of rigid bracing on pain reduction for people with OPVFs. While there may be small benefits for pain, there was not a change in activity level or quality of life from exercise, bracing, or multimodal interventions. One possible explanation was that the small reduction in pain may not have been large enough to elicit a change in activity level or quality of life.

Even though the effects on pain may be relatively small and the clinical significance questioned, the effects of exercise and bracing compared to no exercise or brace respectively is consistent with management of other back conditions. A recent meta-analysis focused on exercise compared with no treatment/usual care in participants with persistent non-specific low back pain, demonstrated that exercise therapy on average reduced pain by 1.1 units on 0-10 visual analogue scale (95% CI: 0.74, 1.41),48 which is of similar magnitude to our results. Also, these results may be similar to the effects of commonly prescribed medications. One trial demonstrated significantly lower pain intensity in a group receiving tramadol (100 mg/day) compared to a control group allowed to receive a Chinese medicine for pain relief (0.32 g ‘as needed’), at both immediate- (SMD = -1.23, 95% CI: -2.42, -0.05) and short-term follow-ups (SMD = -1.58, 95% CI: -2.83, -0.33).10

The results of this review provide some support for prescribing exercise programs for people with OPVFs as they may have a small effect on pain and avert a small risk of harm. These results add to the uncertain evidence that exercise programs may also have a role in preventing fragility fractures associated with osteoporosis such as OPVFs.49 It might provide support for exercise as a first-line intervention for management of OPVFs before considering common treatments with questionable efficacy and known harms such as analgesia. However, this review has not considered cost effectiveness. Most programs involved supervision which may increase costs and adherence was as low as 58% completion of prescribed sessions. However, our preliminary data suggest the costs of exercise interventions were similar to standard care.41-42 A recent paper by Ferreira et al.,50 found that people with low back pain prefer brief low-cost home-based programs. Perhaps a simple low-cost, less resource-intense intervention for people with OPVFs that retains the important components (e.g. load reduction with relative extension and extensor muscle activation) and therefore may achieve the key benefits may be worthwhile. For example, a walking program might achieve the same benefits and also achieve better adherence. None of the exercise trials in this review included a walking program and this may be an area of further research.

This review followed the PRISMA reporting guidelines for systematic reviews and meta-analyses. All the trials included are randomized controlled trials increasing confidence in results, as findings are less subject to bias. The GRADE approach was applied to meta-analyses determining level of certainty in results.

A limitation of our search strategy was that we used a database filter to limit the search to randomized controlled trials rather than a validated filter such as those used in Cochrane reviews. This may have increased the risk of missing potentially relevant trials. However, the small number of additional trials identified through citation tracking suggests our search strategy was sufficiently sensitive. Only trials reported in English were included in our search, which is a limitation of this review. Computation of results for meta-analyses was immediately post intervention only; longer-term results have not been presented in this review. Related to this, each trial included employed variable length interventions, and this also could be considered a limitation. Finally, the fact that all trials were completed in countries classified as high or upper-middle income status means results may not be generalisable to countries of lower socio-economic status.

There is low certainty evidence that exercise interventions and rigid bracing can reduce pain in people with OPFVs without increasing risk of harm. No benefit for health-related quality of life and activity measures were demonstrated for exercise interventions, bracing, or multimodal interventions. Randomized trials testing the effectiveness of simple low-cost exercise interventions for example walking are needed.