The effect of alendronate sodium on spinal fusion: a rabbit model

Bisphosphonates affect bone remodeling and increase bone mass through the inhibition of osteoclasts. Their effect on osteoblasts, and the balance between osteoblastic and osteoclastic activity on bone turnover and healing, is not completely understood. Specifically, the effect of bisphosphonates on spinal fusion has yet to be determined. With the increasing use of bisphosphonates in the elderly population, this effect needs to be delineated.

To evaluate the effect of alendronate sodium after an intertransverse process spinal fusion in a rabbit model.

Randomized double-blinded in vivo study of the effect of alendronate sodium in a spinal fusion model.

Fifty New Zealand white rabbits underwent a posterolateral L5-L6 intertransverse process arthrodesis with autogenous iliac crest bone graft. The rabbits were then randomly divided into two groups. Group I received 3 cc of saline placebo per oral gavage, and Group II received 200 micrograms (approximately 0.05 mg/kg/day) of alendronate sodium dissolved in 3 cc of saline per day for 8 weeks. Upon completion, the rabbits were sacrificed and the lumbar spines harvested, radiographed and graded for motion across the fusion site with manual palpation. Two independent pathologists then prepared and sectioned each left and right fusion mass. Three random x10 fields were examined and graded for both the cephalad and caudad ends of each section (516 fields). Fusion quality was graded using an established histological scoring scale (score 0 to 7 based on fibrous and bone content of the fusion mass).

Two rabbits died on the day of operation, and 48 rabbits survived the operation. Five additional rabbits died within the first 2 postoperative weeks. Thus, 43 rabbits (21 in Group I, 22 in Group II) completed the 8-week course of treatment. Grading each side separately, 26 of 42 fusion masses (62%) in Group I and 24 of 44 fusion masses (55%) in Group II had radiographic evidence of fusion (p=.76). With gross palpation, 11 of 21 motion segments (52%) in Group I versus 13 of 22 motion segments (59%) in Group II were determined to have a solid fusion (p=.76). Histologically, Group I had a higher median score (6.0; range, 0 to 7 vs. 1.0; range, 0 to 7; p<.0001) and a higher fusion rate (76% vs. 45%; p=.004) than Group II.

Alendronate sodium appears to inhibit or delay bone fusion in a rabbit model. Presumably, this occurs as a result of uncoupling the balanced osteoclastic and osteoblastic activity inherent to bone healing. These findings suggest that a discontinuance of alendronate sodium postoperatively during the acute fusion period may be warranted.