Preventing postlaminectomy adhesion. A new experimental model.

STUDY DESIGN: The efficacy of various interpositional membranes for prevention of extradural adhesion was investigated by a new animal model that quantified the biomechanical effect of scar formation. Twenty-one dogs were treated with autologous free fat graft, hyaluronic acid or no interpositional membrane after undergoing two-level laminotomy, nerve root exploration, and disk injury. An additional 11 dogs that did not undergo spine surgeries served as control animals. OBJECTIVES: Inter-animal variability in inherent propensity to form scar was first measured before any spine surgery. Twelve weeks after spine surgery, the lumbosacral spine of each dog was harvested en bloc for biomechanical testing of extradural adhesion ultimate load. Scar stiffness coefficient was also calculated. SUMMARY OF BACKGROUND DATA: Adhesion ultimate load was significantly less in the nonoperative control group when compared with the fat graft and no interpositional membrane group, but not when compared with the hyaluronic acid group. A beneficial effect of hyaluronic acid in lowering adhesion ultimate load was demonstrated, although a statistically significant difference from the fat graft and no interpositional membrane groups was not reached. No difference in scar stiffness coefficient was found between the four groups. METHODS: A new experimental model allowing objective biomechanical quantification of the effect of postoperative scar was described. Ultimate load of adhesions to both nerve roots and dura was measured. A biochemical assay that determined collagen content was also used to assess inter-animal propensity to form scar after a standardized surgical insult. Results were compared with other relevant studies. RESULTS: Findings suggest a beneficial effect of hyaluronic acid in decreasing the biomechanical strength of extradural adhesions following laminotomy, nerve root exploration, and disk injury when compared with use of fat graft or no interpositional membrane. These results support other recent investigations that study the use of hyaluronic acid treatment in a laminectomy model. The adverse consequence of intraoperative epidural bleeding was also demonstrated. CONCLUSIONS: The new experimental model described in the current study was reproducible and permitted objective quantification of the effect of postoperative adhesion rather than measuring its mere presence. A beneficial effect of hyaluronic acid treatment and a lack of such beneficial effect of free fat graft interpositional membrane was suggested. The importance of avoiding active epidural bleeding was also evident.

Comparison of hydroxyapatite granules to autogenous bone graft in fusion cages in a goat model.

BACKGROUND: With the increased use of fusion cages to achieve lumbar intervertebral fusion, the question arises as to the potential for bone ingrowth from the host bone through the entire cage. Is it even necessary to have an autogenous graft to achieve total bone incorporation? METHODS: Nine adult male goats had fusion cages implanted into three vertebral bodies. The design was Surgical Dynamics/Ray Fusion Cage, measuring 21 mm x 14 mm. In each animal, one fusion cage was filled with autogenous graft, one with hydroxyapatite, porous granules, and the other with nonporous granules. Amount of new bone formation was determined by backscatter electron microscopy at 3 months post implantation in all animals. RESULTS: The histologic section shows that there was total incorporation in all specimens at 3 months. There was slightly more new bone (43%) with the nonporous granules compared with the porous granules (35%). The amount of residual void space was about the same in all specimens, indicating that the amount of new bone formation was similar and not statistically different in cages filled with hydroxyapatite granules versus granules of autogenous bone. CONCLUSION: This study confirms that total incorporation by ingrowth of new bone can be expected in fusion cages. The amount of ingrowth is about the same for autogenous graft versus hydroxyapatite granules. Apparently, it is not necessary to use bone graft to achieve successful bone incorporation if an acceptable biocompatable lattice, such as hydroxyapatite granules, is used.

Is increased segmental motion early after lumbar discectomy related to poor clinical outcome 5 years later?

The purpose of the study was to compare segmental motion in the early postoperative phase after lumbar discectomy to the outcome 5 years postoperatively. The study population had radiologically verified symptomatic L4-L5 or L5-S1 lumbar disc herniation and was referred with an indication for lumbar discectomy. Radiostereometry was performed in the supine and standing positions. The L4-L5 and L5-S1 segments were analysed separately. L4-L5 segments adjacent to the operated L5-S1 segment constituted a reference segment for the operated L4-L5 and vice versa. Twenty-one patients were available for the follow-up at 5 years. Outcome was classified as functionally good or poor. Repeated or planned repeat surgery at the same level during follow-up was considered as poor outcome. The L4-L5 segments in the poor group showed different direction of sagittal rotation (anterior versus posterior) of L4 on L5 compared with the good group (p<0.01). On the L5-S1 segment, patients with poor outcome displayed an increased anterior translation of about 1 mm (p<0.01) compared with the reference segments. Our study suggests that increased inducible vertebral displacement in the early postoperative phase after discectomy is associated with a poor clinical outcome.

The effect of standard lumbar discectomy on segmental motion: 5-year follow-up using radiostereometry.

We measured the effects of lumbar discectomy on segmental motion over a period of 5 years. Twenty-four patients with lumbar disc herniation were treated by standard lumbar discectomy at the L4-L5 or L5-S1 level. Peroperatively, tantalum markers were inserted into L4, L5, and the sacrum. Radiostereometric analysis was performed at discharge from hospital and 5 years postoperatively. The treated level was compared with the corresponding untreated level. Thus, patients who had discectomy at the L4-L5 level served as controls for patients with L5-S1 lesions and vice versa. The relative rotation and translation in relation to the three cardinal axes were calculated. Inducible displacements over the two discs were calculated between the supine and standing positions. At the L4-L5 level, there were no differences in inducible displacements between the operated and control levels at discharge or 5 years postoperatively. At the L5-S1 level we found decreasing inducible movement in the sagittal plane over time for discectomy patients. The reason for decreasing mobility over time after discectomy at the L5-S1 but not at the L4-L5 level is unknown. Mechanical factors caused by the more vertical orientation of the L5-S1 disc in combination with degenerative changes could be one explanation.