Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial
BMJ 2018; 361 doi: https://doi.org/10.1136/bmj.k1551 (Published 09 May 2018) Cite this as: BMJ 2018;361:k1551
All rapid responses
We are interested in the conclusions that are drawn from this trial, which included outpatients with a pain duration of up to 9 months, and we appreciate the authors' efforts to conduct this well-constructed randomised trial (1). However, we have several questions, some of which are related to the distinct possibility of selection bias.
First, the authors did not sufficiently define the nature of pain in this study. We frequently observe two different types of pain in the clinical setting, namely movement-induced pain and pain at rest. In the early stages, patients tend to remain bed-bound because of a stinging motion-induced pain; after a few weeks, the pain typically becomes a dull pain at rest that disturbs long-term posture maintenance. We assume that the former type of pain (motion induced) is directly attributable to fractures, whereas the latter (pain at rest) is a multifactorial pain consisting of burden on the back muscles or psychological elements. Notably, the benefits of cement augmentation, which provides internal fixation of unhealed vertebral fractures, might be more pronounced in motion-induced pain. As the median number of days with pain in the vertebroplasty group is relatively long (about 6 weeks) and the subjects are outpatients who do not exhibit major problems with body movement, it is possible that the subjects in this trial included few or no patients whose main pain component is motion induced. We believe that the perceived discord between the results of VAPOUR (2) and the present trial reflects this difference in patient populations.
Second, we disagree with the suggestion that periosteal infiltration alone in the early phase provides sufficient pain relief. Theoretically, it is unlikely that the effects of a single administration of local anaesthesia will persist for 12 months. Hence, the pain reduction in the sham procedure group was more likely related to the effects of placebo, natural healing, and regression to the mean than to the effect of local anaesthesia. Moreover, to confirm the effect of local anaesthesia, we believe that a randomised controlled trial with placebo comparison is essential.
Finally, we have a serious concern that 156 subjects, almost the same number as those who underwent randomisation, declined to participate. Many subjects would have chosen to undergo open-label vertebroplasty and some might have been unable to endure 10 days from diagnosis to the procedure. Is it not possible that the group who declined to participate is ‘a place for vertebroplasty’？ Furthermore, it should be noted that subjects with neurological deficits were excluded from enrolment. The natural history of a complicated vertebral fracture is short; it often causes neurological symptoms within a few weeks (3, 4). Given that the median time from pain onset to the surgical procedure in this trial was approximately 6 weeks and none of the patients developed neurological symptoms during the study period, it is possible that fragile fractures with high likelihood of progression were scarcely included in the trial. We believe that the group with impending neurological symptoms is ‘a place for vertebroplasty’ as well.
1. Firanescu CE, de Vries J, Lodder P, Venmans A, Schoemaker MC, Smeet AJ, et al. Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial. Bmj. 2018;361:k1551.
2. Clark W, Bird P, Gonski P, Diamond TH, Smerdely P, McNeil HP, et al. Safety and efficacy of vertebroplasty for acute painful osteoporotic fractures (VAPOUR): a multicentre, randomised, double-blind, placebo-controlled trial. Lancet. 2016;388(10052):1408-16.
3. Heggeness MH. Spine fracture with neurological deficit in osteoporosis. Osteoporos Int. 1993;3(4):215-21.
4. Ito Y, Hasegawa Y, Toda K, Nakahara S. Pathogenesis and diagnosis of delayed vertebral collapse resulting from osteoporotic spinal fracture. Spine J. 2002;2(2):101-6.
Competing interests: No competing interests
I probably have performed more Kyphoplasty/vertebroplasties than anyone in Florida over the past 15 years. There are 3 procedures that yield me 90+ success rates - kyphoplasty for acute (within 3 weeks) compression fracture, microdiscectomy for acute HNP with radiculopathy, and ACDF/replacement for acute cervical hnp with radiculopathy. Unfortunately in England acute means within 6 months. If I were to see a patient with an “acute” compression fracture “an average of 43 days post fracture”, only patients with significant pain would receive the procedure because most would have nearly healed with conservative management by the time they were seen. My happiest patient is my patient treated within 1st week of injury.
Competing interests: No competing interests
Dear Editor,
Firanescu et al1 did a wonderful job with the VERTOS IV trial, and have concluded no difference in outcome with the use of vertebroplasty for osteoporotic vertebral fractures. There are some queries regarding the methodology.
1. The patients recruited had pain for 6 to 9 weeks, even some had pain more than that, making this selection procedure more into sub acute category rather than acute.
2. More importantly, the patient’s demographics show that these patients were on drugs before this intervention (vertebroplasty/sham). The number of patients on stronger opioids were higher in vertebroplasty group as compared to sham group [42 (47) vs. 25 (29)], but the baseline VAS, Mean (SD) QUALEFFO score, Mean (SD) RMDQ score were same if not low in vertebroplasty group as compared to sham group. This doesn’t match as people requiring stronger opioids should have more pain and disability.
3. Baseline characteristics of the patients was measure after stopping of drugs or not. If yes, then for how many days?
4. The use of local anesthetic in sham procedure makes things more unclear as it can have multiple effect
a. Wet needling for erector spinae, potential cause of pain because of changed biomechanics.
b. May go to the facets near the level of infiltration.
c. May go lateral like an erector spinae plane block between transverse process and erector spinae.
5. What was the volume of local anesthetic used as it will have implications as described above.
6. How many patients stopped taking strong opioids in vertebroplasty/sham group of patients? What was the analgesic use outcome of the patients post procedure?
7. Why was a change of only 1.5 in VAS score considered significant? As their baseline VAS score is around 7 so they should incorporate at least 3 points so the category of the patients change from severe to moderate or mild.
Conflict of interest: None
Bibliography:
1. Firanescu Cristina E, de Vries Jolanda, Lodder Paul, Venmans Alexander, Schoemaker Marinus C, Smeet Albert J et al. Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial BMJ 2018; 361 :k1551
Competing interests: No competing interests
We are concerned by conclusions drawn in the VERTOS IV publication and accompanying editorial.
Randomised clinical trials are designed to assess the efficacy of treatments employed in clinical practice. The conclusions reached are very dependent on the populations recruited and their selection criteria. Patients presenting with acute vertebral fractures may encompass a wide spectrum varying from those with mild to severe osteoporosis, single or multiple comorbidities and those with a pain severity requiring simple analgesia to major opiates. Hospital inpatients are often much worse affected than those who are ambulatory outpatients. It would be an error of judgement to suggest that these cohorts are similar and should be managed equally.
VERTOS4 protocol (1) did not recruit patients with uncontrolled pain referred for vertebroplasty by their attending physicians, but from assessing spinal radiographs performed in 4 radiology centres. Patients with radiographic fractures were given a questionnaire and if back pain was less than 9 weeks duration and at least 5 out of 10 on VAS scale, were invited to participate. Only 34% were using strong opiates and hospitalised inpatients were ineligible. Their osteoporosis could be considered as mild with mean BMD T-scores 0f -2.4 and less than 50% receiving anti-osteoporotic therapies. Patients agreeing to participate were then referred for MRI and physician consult to assess eligibility, causing a mean 14-day delay. At time of vertebroplasty, pain duration was less than 11 weeks, with a median (and mean) of 6.1 weeks. Patient enrolment occurred from April 2011 to April 2014 with final data collection completed three years ago.
Of four blinded, randomised trials of vertebroplasty only the VAPOUR trial (2) has shown clinically important benefits from vertebroplasty. The authors of VERTOS4 acknowledge this but fail to explore the differences between the trials. VAPOUR targeted a different subgroup of patients, who were referred by attending physicians on clinical grounds, with uncontrolled, severe pain, defined as NRS ≥7/10 despite maximal medical therapy, including opiates. 57% were hospital inpatients and 43% outpatients and all had severe osteoporosis. Vertebroplasty was offered without further delay, to control the acute pain syndrome, facilitate rapid rehabilitation and prevent the downhill spiral so often seen in the elderly with catastrophic spinal fractures. Most patients (80%) underwent vertebroplasty within 3 weeks of fracture onset (mean fracture duration 2.6 weeks compared to 6.1 weeks in VERTOS4).
Failure of medical therapy to adequately control acute fracture pain indicated poor outcomes from placebo therapy in VAPOUR. Most placebo group patients still had moderate or severe pain at 6 months and 76% were still using analgesics, whereas the vertebroplasty group derived clinically important benefits at all time points. Early intervention makes vertebroplasty technically easier, as the fracture is soft and physically compliant, accepting a larger PMMA volume (7.5cc in VAPOUR vs 5.1cc in VERTOS4) with a reduced PMMA leak rate (34% in VAPOUR vs 91% in VERTOS4). It also facilitated vertebral height restoration. The very high leak rate in VERTOS4 suggests that the vertebral body had already consolidated.
Patients with severe pain (NRS ≥7/10) due to osteoporotic spinal fracture less than 6 weeks duration should be offered maximal medical therapy including opiate analgesia. If this fails to adequately control the acute pain syndrome, or if the patient cannot tolerate the opiates due to side effects, then they should be offered early vertebroplasty without further delay. This includes the most severely affected outpatients and hospital inpatients. This reinforces the notion that scientific evidence is an essential part of advancing medical knowledge, but it must be integrated with clinical experience to complete the process. Only then can we inform decision making in clinical care and deliver the best clinical outcomes available.
1. Firanescu C, Lohle PN, de Vries J, et al. A randomised sham controlled trial of vertebroplasty for painful acute osteoporotic vertebral fractures (VERTOS IV). Trials. 2011;12:93. doi:10.1186/1745-6215-12-93.
2. Clark, W, Bird, P, Gonski, P et al. Safety and efficacy of vertebroplasty for acute painful osteoporotic fracture (VAPOUR): a multicentre, randomised, double-blind, placebo-controlled trial. Lancet 2016;388:1408–16. doi.org/10.1016/S0140-6736(16)31341-1
Competing interests: We are authors of the VAPOUR trial.
Firanescu and colleagues examined the effect of percutaneous vertebroplasty versus sham procedure in patients with osteoporotic compression fractures after administration of local anesthetic at the level of vertebral body pedicle. (1) The authors appear, in having proceeded directly to real or sham treatment without reassessment of pain relief after local anesthetic administration, to have missed an important aspect of pathophysiology which calls the significance of this study’s results into question.
An investigation by Wilson and colleagues (2) was prompted by the great degree of relief seen in both true and sham vertebroplasty treatment in two well-publicized studies which employed prior local anesthetic administration. (3,4) Wilson wished to investigate the effect of local anesthetic administered in close anatomic proximity to the involved facet joint(s) on subsequent outcome, hypothesizing that disruption of posterior element anatomy in vertebral fractures might be a significant pain generator as opposed to, or in addition to, pain due to the vertebral body injury itself. This investigation, although admittedly non-controlled, yielded that of 61 patients with osteoporotic vertebral compression fractures and evidence of edema on magnetic resonance imaging (MRI), 21 had significant relief of their pain upon facet joint injection itself. Of a remaining 24 patients who underwent vertebroplasty, 23 achieved successful outcome with the procedure.
Although instillation of local anesthetic at the level of the pedicle is not identical to intra-articular facet injection, there exists concern that Firanescu may have failed to account for a recognized cause of pain in a significant number of these patients. This raises the distinct possibility that the authors’ results are confounded in the aspect of not recognizing the possible effect of two distinct disease processes in such a population.
REFERENCES:
1. Firanescu CE, de Vries J, Lodder P et al. Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial. BMJ 2018;360:k1551.
2. Wilson DJ, Owen S, Corkill RA. Facet joint injections as a means of reducing the need for vertebroplasty in insufficiency fractures of the spine. Eur Radiol 2011;21:1772-1778.
3. Buchbinder R, Osborne RH, Ebeling PR et al. A randomized trial of vertebroplasty for painful osteoporotic vertebral factures. N Engl J Med 2009;361:557-568.
4. Kallmes DR, Comstock BA, Heagerty PJ et al. A randomized trial of vertebroplasty of osteoporotic spinal fractures. N Engl J Med 2009;361:569-579.
Competing interests: No competing interests
Nothing has been more divisive in the treatment of osteoporotic vertebral compression fractures (OVCF) than the role of percutaneous vertebroplasty. The authors appear to firmly conclude that percutaneous vertebroplasty has no role as standard pain treatment for these patients [1]. Looking at the participants in the study, it does not reflect the patients I work with in secondary care who present as unplanned admissions to hospital. Their pain and disability following the fracture was so debilitating that a period of hospitalisation was required. The participants in this study were recruited from an outpatient setting with pain in the preceding 6-9 weeks. These patients' pain have an element of 'chronicity' despite the presence of oedema on their MRI. At this stage, back pain could be due to causes other than the fracture.
The non-acute nature of the study's participants, ie less than 6 weeks, was discussed in the paper but appeared lost behind the message that the authors were keen to focus on. The authors highlighted the VAPOUR study [2] which perhaps better reflects the acute hospital cohort I see. The acute hospital cohort with OVCF also represents an older multi-morbid group where effective pain relief to enable early mobilisation is crucial. Besides that, the authors also asked participants the day after treatment if they were able to guess which group they were randomised to as a proxy of how effective the sham procedure was. Although their numbers were higher than VAPOUR, 54% [2] and INVEST, 37% [3], both asked participants 14 days after the procedure and VERTOS IV asked the question the day after the procedure, 82% [1]. Patients with OVCF represent a heterogenous group from those discovered incidentally, symptoms that can be managed as an outpatient and those needing hospital admission. Symptomatic patients should not be grouped together. I strongly advocate research and treatment be individualised to this context and not be just a subjective patient reported score.
Reference
1. Firanescu C, de Vries J, Lodder P, Venmans A, et al. Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial. BMJ 2018; 360:k1551
2. Clark W, Bird P, Gonski P, et al. Safety and efficacy of vertebroplasty for acute painful osteoporotic fractures (VAPOUR): a multicentre, randomised, double-blind, placebo- controlled trial. Lancet 2016:388:1408-1416
3. Kallmes D, Comstock B, Heagerty P, et al. A randomized trial of vertebroplasty for osteoporotic spinal fractures. N Engl J Med 2009:361:569-579
Competing interests: I have received lecture fees from Eli Lilly
A few minor mistakes in the article OF Re: Vertebroplasty versus sham procedure for painful acute osteoporotic vertebral compression fractures (VERTOS IV): randomised sham controlled clinical trial
To the editor：In this article, the author intends to assess whether percutaneous vertebroplasty results in more pain relief than a sham procedure in patients with acute osteoporotic compression fractures of the vertebral body. After a Randomised, double blind, sham controlled clinical trial, the author concludes that Percutaneous vertebroplasty did not result in statistically significantly greater pain relief than a sham procedure during 12 months’ follow-up among patients with acute osteoporotic vertebral compression fractures. The article has undergone rigorous experimental design and supervision. The design of this paper is reasonable and the experimental results are of great significance. This is one of several high-quality RCT articles comparing the efficacy of vertebroplasty and placebo in recent years. In the future, this article will be quoted in large quantities by many meta-analysis and monograph articles, and its influence will be enormous. However, there is a few minor mistakes that can not be ignored in the article, that is, the expression of the duration of pain in the patients. It was clearly pointed out in the article that the inclusion criteria were T5-L5 focal back pain at the level of fracture for up to six weeks. In the results section, the author states that twenty two of the 90 participants (24%) in the vertebroplasty group and 12 of the 86 participants (14%) in the sham procedure group had pain for more than six weeks and less than nine weeks before randomisation. But in its table 1, it shows that the median time from onset of symptoms to treatment was 43 days (interquartile range 29-52 days) in the vertebroplasty group and 36 (24-51) days in the sham procedure group. The expression of pain duration is inconsistent. Therefore, the results of the experiment will be questioned by the reader.
As the conclusion of this article is of great significance, and the article has great influence in the future, we suggest that the author further modify the errors in this part and explain the contents of this part.
Competing interests: No competing interests