Percutaneous insertion of craniocaudal expandable implants for vertebral compression fracture

5.1 Death was reported in 2 patients in an observational study of 103 patients treated by a vertebral craniocaudal expandable implant. The first death occurred 52 days after the procedure and was caused by acute kidney failure; the other death occurred 204 days after the procedure and was caused by an acute respiratory syndrome. The authors stated that the deaths were neither implant- nor procedure-related.

5.2 Pneumonia was reported in 1 patient out of 36 in the vertebral craniocaudal expandable implant group and in 2 patients out of 39 in the vertebroplasty group in a retrospective comparative study of 75 patients, within 12-month follow-up (no further details provided).

5.3 Cement extravasation measured immediately after the procedure and assessed on X-ray by an independent laboratory was reported in 55% (98/177) of vertebra levels in patients treated by a vertebral craniocaudal expandable implant and in 58% (103/178) of levels in patients treated by balloon kyphoplasty in a randomised controlled trial (RCT) of 300 patients treated by an implant (n=153) or by balloon kyphoplasty (n=147). There was no statistically significant difference between the groups (‑3%, Bayesian credible interval [BCI] ‑13% to 8%). However, in a secondary analysis, cement extravasation was reported statistically significantly less frequently in the implant group than in the balloon kyphoplasty group (17% [30/177] of levels compared with 26% [46/178] of levels, difference -9%, BCI -17% to -0.33%). Cement leaks were reported statistically significantly less frequently in the implant group (3% [4/133] of vertebras) than in the balloon kyphoplasty group (10% [12/122] of vertebras; p≤0.05) in an RCT of 185 patients treated by a vertebral craniocaudal expandable implant (n=92) or by balloon kyphoplasty (n=93). Intracanal leaks were reported in none of the patients treated by the implant and in 2% (2/86) treated by balloon kyphoplasty. Cement leaks identified by CT scan were reported in 14% (11/77) of patients in a retrospective case series of 77 patients treated by a vertebral craniocaudal expandable implant. All patients had post-traumatic fractures. One patient had nerve root pain caused by the cement leaking along a secondary fracture line in the pedicle (see section 5.5).

5.4 Dural tear was reported in 1 patient in a case series of 57 patients. It occurred during the initial pedicle access with the Jamshidi needle. It was treated with Gelfoam and there were no residual or permanent sequelae.

5.5 Adjacent level fracture was reported in 21% (28/134) of the as-treated population in the implant group and in 22% (29/130) of the as-treated population in the balloon kyphoplasty group in the RCT of 300 patients treated by an implant (n=153) or by balloon kyphoplasty (n=147). There was no statistically significant difference between the groups (-1%, BCI -11% to 8%). In the same study, a fractured pedicle was reported in 1 patient in the implant group. It was associated with the use of the implant in the setting of sclerotic bone. This resulted in back pain at the time of discharge, which was treated with analgesics. New fractures were reported in 12% (3/26) of patients in the implant group and in 54% (14/26) of patients in the balloon kyphoplasty group in a retrospective matched-paired comparative study of 52 patients. The difference between the groups was statistically significant, p<0.0001. Adjacent fractures were reported in 8% (2/26) of patients in the implant group and in 35% (9/26) of patients in the balloon kyphoplasty group.

5.6 Pain after the procedure was reported in 1 patient in the implant group in the RCT of 300 patients treated by an implant (n=153) or by balloon kyphoplasty (n=147).

5.7 Skin infection that started in hospital was reported in 1 patient in the retrospective case series of 77 patients. The infection was probably caused by contamination from an oral infection and was treated with antibiotics. Urinary tract infection was reported in 17% (6/36) of patients in the vertebral craniocaudal expandable implant group and in 21% (8/39) of patients in the vertebroplasty group in the retrospective comparative study of 75 patients (no further details provided).

5.8 Haematoma was reported in 1 patient in a prospective case series of 32 patients treated by a vertebral craniocaudal expandable implant; revision was not needed.

5.9 Minor loss of height of the stabilised L2 vertebral body in an osteoporotic fracture was reported in 1 patient in the prospective case series of 32 patients. The Beck Index changed after the procedure from 1.0 to 0.96 and the Cobb angle changed from 11 degrees to 13 degrees. The visual analogue scale score remained unchanged.

5.10 Collapse of the treated vertebral body resulting in canal compromise, haematoma and neurological symptoms was reported in 1 patient 16 days after the procedure in the observational study of 103 patients; the condition of the patient had improved at 12‑month follow-up (no further details reported).

5.11 Dislocation of posterior wall secondary to surgery and leading to a sensory deficit was reported in 1 patient 4 days after the procedure in the observational study of 103 patients. The patient had been treated outside of the device instructions for use and was subsequently treated by decompression and posterior instrumentation.

5.12 Device migration was reported in 1 patient in the retrospective case series of 77 patients; this reflected a technical problem that occurred with an instrument prototype.

5.13 In addition to safety outcomes reported in the literature, specialist advisers are asked about anecdotal adverse events (events which they have heard about) and about theoretical adverse events (events which they think might possibly occur, even if they have never done so). For this procedure, specialist advisers listed the following anecdotal adverse events: incorrect placement of the implant, implant tilt in osteoporotic bone and endplate fracture so that vertebral body height was not restored. They considered that the following were theoretical adverse events: failure to deploy the implant correctly and implant-related problems such as failure to raise the endplates.