Interventional procedure overview of radiofrequency ablation as an adjunct to balloon kyphoplasty or percutaneous vertebroplasty for palliation of painful spinal metastases

Safety summary

RFA-related adverse events

Procedure-related adverse events occurred in 3% (16/583) of patients (range 0% to 11%) in the systematic review of 9 studies. All 9 studies reported complications. Increased pain and numbness (n=6) and post-procedure radicular symptoms and pain (n=5) were the most frequently described procedure-related adverse events. Procedure-related adverse events were not reported in 4 of the studies (Rosian 2018).

Immediate onset of lower extremity paralysis, diminished sensation, and bowel and bladder dysfunction after the RFA procedure was reported in 1 patient. Radiology reports indicated suspicion for thermal injury to the ventral nerve roots at L1. Authors state that this might have been caused by inaccurate placement of the RFA probe or more extensive zones of thermal ablation. The patient had prolonged inpatient rehabilitation, and sensation in the lower extremities was noted at 1-year follow up, but severe neurogenic bladder and bowel dysfunction continued. Electromyography study showed severe membrane instability with no active motor units in L1, L2, L3, L4 and L5, consistent with a conus medullaris or cauda equina injury (Huntoon 2020).

Four adverse events were reported in the case series of 100 patients, of which 2 resulted in hospitalisation for pneumonia and respiratory failure (Levy 2020).

In the retrospective cohort study of 166 patients (266 tumours), the total complication rate related to RFA was 3% (8/266). Major complication (bilateral lower extremity weakness, difficulty in urination, lack of erection, attributed to spinal cord thermal injury during the procedure) was reported in 1 patient. This was treated with high-dose intravenous steroids that only gave slight improvement. Minor complications were reported in 7 patients. These included periprocedural transient radicular pain in 4 patients (resolved after transforaminal steroid injections), delayed secondary lumbar vertebral fracture where vertebral augmentation was not done in 1 patient (treated with analgesics) and asymptomatic spinal cord oedema in 2 patients seen on imaging (resolved without treatment; Tomasian 2021).

Pain and numbness

Increased pain and numbness were reported in 1% (6/583) of patients in the systematic review of 9 studies with 583 patients (Rosian 2018).

Contralateral lower limb pain and numbness during the procedure was reported in 16% (4/25) of patients in a case series (Yang 2017) included in the systematic review of 8 studies. These symptoms spontaneously resolved with temperature decrease after RFA (grade 1 complication) in 2 patients, and the other 2 patients needed steroids. One of them experienced heaviness in the legs 1 day after RFA without any subsequent consequence at 1-week follow up (grade 2 to 3a complications; Cazzato 2018).

Post-procedure radicular symptoms and pain

Post-procedure radicular symptoms and pain were reported in 1% (5/583) of patients in the systematic review of 9 studies with 583 patients (Rosian 2018).

Post-procedure radicular pain needing selective nerve block (grade 3a complication) was reported in 1 patient in a case series included in the systematic review of 8 studies (Cazzato 2018).

Postoperative radicular pain was reported in 4 patients in a study (Wallace 2015) included in the systematic review of 8 studies on combined RFA and vertebral stabilisation techniques for patients with palliative treatment of vertebral metastases (Greif 2019).

Neural damage

Transient neural damage related to the high temperature rise during RFA treatment was reported in 1 patient (Nakatsuka 2009) in a systematic review of 8 studies. This resolved 2 days after the procedure with intravenous administration of steroids (Cazzato 2018).

Vertebroplasty-related adverse events

Cement extravasation was reported in 10% of patients in the systematic review of 15 studies. Only 1 patient reported moderate pain and needed surgical removal (Murali 2021).

The rate of vertebroplasty-related adverse events ranged from 4% to 73% in the systematic review of 9 studies with 583 patients. Cement extravasation after vertebroplasty was the most frequently reported adverse event, occurring in 15% (67/437) of patients. It is not clear if these events resulted in any clinically relevant episodes (Rosian 2018).

Bone cement leakage outside the vertebral body occurred in 71% (26/37) of patients at 31 levels in a case series (Gregory 2009) included in the systematic review of 8 studies (Cazzato 2018).

Extravasation after RFA followed by kyphoplasty was reported in 19 patients and anterior leaks were reported in 2 patients in a case series (Lane 2011) included in the systematic review of 8 studies on combined RFA and vertebral stabilisation techniques for patients with palliative treatment of vertebral metastases. Cement leakage (in 7 and 8 patients) was reported in 2 studies (Munk 2009, Madaelil 2016) included in the same systematic review (Greif 2019).

The retrospective comparative case series of 87 patients with spinal metastatic tumours (125 vertebral bodies) compared RFA combined with bone cement (35 patients with 47 vertebral segments) with bone cement alone (52 patients with 78 vertebral segments). It reported that the permeability of bone cement in the RFA combined with bone cement group was significantly lower than those in the bone cement alone group (6% compared with 21%, p<0.05; Lv 2020).

Haematoma

Haematoma at the site of treatment was reported in 6 patients from 3 studies (Burgard 2014, Toyota 2005, Hoffman 2008) included in the systematic review of 8 studies on combined RFA and vertebral stabilisation techniques for patients with palliative treatment of vertebral metastases (Greif 2019).

Mortality

24% (109/462) of patients died at a median follow up of 1 year across 10 studies in the systematic review of 15 studies. One death was related to postoperative sepsis and 2 studies reported that 13 deaths were unrelated to the RFA procedure. The cause of death was not mentioned in other studies (Murali 2021).

Deaths (5 and 10) were reported in 2 studies in the systematic review of 9 studies (Rosian 2018). There were 30 deaths (attributed to the underlying malignancy and not related to the study procedure) reported in the case series of 100 patients (Levy 2020). There were 27 deaths (related to other causes without spinal cord compression) reported in the retrospective cohort study of 166 patients (Tomasian 2021).