Secondary Logo

Journal Logo

Original Article

Posterior Lumbar Interbody Fusion Using Dense Hydroxyapatite Blocks and Autogenous Iliac Bone

Clinical and Radiographic Examinations

Asazuma, Takashi MD; Masuoka, Kazunori MD; Motosuneya, Takao MD; Tsuji, Takashi MD; Yasuoka, Hiroki MD; Fujikawa, Kyosuke MD

Author Information
Journal of Spinal Disorders & Techniques: February 2005 - Volume 18 - Issue - p S41-S47
doi: 10.1097/01.bsd.0000112043.70321.1a
  • Free


Posterior lumbar interbody fusion (PLIF) was described by Jaslow1 and was established and popularized by Cloward.2,3 With the worldwide spread of the pedicle screw systems, PLIF became one of the standard surgical techniques for spondylolisthesis, herniated disc, spinal canal stenosis, and failed back syndrome, as one can achieve both dural sac and nerve root decompression as well as a simultaneous stabilization of the lumbar spine. In a meta-analysis of various kinds of lumbar spine fusion techniques, Turner et al4 found that PLIF provided the highest fusion rate and the most satisfactory clinical results. However, some problems such as collapse,5 retropulsion of the grafted bone,6-8 and pseudoarthrosis in 8-44% of the cases3,8-10 have been reported when autogenous or cadaveric bone was used. Furthermore, bone graft donor site problems were also reported.11,12 To resolve such problems, apatite and wollastonite-containing glass ceramics,13 titanium cages,14,15 and carbon fiber cages16-20 intervertebral spacers were developed and applied for clinical use.

To avoid the collapse of the grafted bone and to reduce donor site problems, we have used dense hydroxyapatite (HA) block as a substitute for autogenous bone graft for PLIF. The purpose of this study is to analyze the clinical and radiographic results of PLIF using dense HA blocks and clarify the efficacy of this surgical procedure.


HA is categorized as a bioactive ceramic. The molecular formula is Ca10(PO4)6(OH)2. The molecular weight of dense HA is 1000.63 and its calcination temperature is 1150°C. It has a density of 2.83-3.14 g/cm3 and a bending strength of >58.5 MPa. In this series, the size of the HA block was 8 mm in width, 20 mm in depth, and 10, 12, or 15 mm in height. There are two kinds of shapes. One is a block with incline (taper shape), and the other is a rectangular parallelepiped (Fig. 1). The blocks were manufactured by Sumitomo Osaka Cement Co., Ltd. (Japan).

Dense HA blocks manufactured by Sumitomo Osaka Cement Co., Ltd. (Japan). There are three sizes of blocks with or without an incline.


The patients were placed on the operating table to permit a standard posterior midline approach to expose the spine. The paravertebral muscles were separated to the laminae and transverse processes. After the pedicle screws were inserted, neural canal decompression was performed and space for the PLIF procedure was made in both sides using a shaver, a ring curette, and a file. Care was taken to preserve the cortical endplate, although it was not always possible, especially in cases of severe osteoporosis. Two iliac bone blocks harvested from one posterior iliac crest with one-side cortex and one HA block were grafted together into the interbody space such as in a sandwich under a proper degree of distraction. Cancellous bone chips, locally harvested, were also grafted onto the anterior and lateral aspect of the HA block according to Suzuki’s method (Fig. 2).21 One-side cortex and one HA block were placed as anteriorly as possible. After finishing the PLIF procedures, spinal implants were applied, and segmental compression was added to obtain lordotic alignment and stabilization. After 2 weeks of bed rest, patients were allowed to ambulate with a hard corset. We have found that the period of bed rest can be shortened to 5 days and have used this shortened time in our recent cases. The corset is used for about 3-4 months.

The schema of the axial view of the Suzuki method. Two iliac bone blocks and one HA block are grafted together with cancellous bone chips as in a sandwich.


Between October 1995 and March 2001, 31 patients (16 males and 15 females) underwent PLIF using dense HA blocks and autogenous iliac bone grafts. Twenty-six patients (12 males, 14 females) who could be followed for a minimum of 2 years were examined in this study. Their ages ranged from 17 to 63 years (mean 40.3 years). The follow-up periods ranged from 2 to 6 years (mean 3 year 9 months). The spinal pathogeneses were lumbar disc herniation in 12 patients, degenerative spondylolisthesis in 7, spondylolytic spondylolisthesis in 4, failed back syndrome in 2, and lumbar canal stenosis in 1. The pedicle screw system was used in all the patients. The Crock-Yamagishi system (CY) was used in nine patients, Steffee VSP (VSP) in nine, ISOLA spinal system in three, NDMC system in three, and M8 in two. One-level fusion was performed in 19 patients, two-level fusion in 6, and three-level fusion in 1 (Table 1).

Table 1
Table 1:
Patients’ Profiles and Results

The surgical outcome was evaluated by the Japanese Orthopaedic Association’s Assessment of Treatment of Low Back Pain (JOA score) and the recovery rate (Table 2).22 The recovery rate was calculated using the following formula: (postoperative score − preoperative score)/(29 − preoperative score) × 100 (%). Intra- and postoperative complications were also analyzed. Radiographic evaluations were in five areas: bony union, clear zone in the upper or lower margin of HA block, cracking of HA block, sinking of HA block, and changes of lumbar sagittal alignment pre- and postoperatively. Clear zone (+) was defined as the presence of a readily discernible gap between the HA block and vertebral bodies that may or may not extend along the entire margin, and clear zone (−) was defined as the case in which no gap can be seen in lateral plain radiographs at follow-up time. Sinking of the HA block was deemed positive when >2 mm of the anterior or posterior margin of an HA block was found to have migrated into the adjacent vertebral body when comparing lateral radiographs taken immediately after surgery and at follow-up. The assessment of bony union depended on the continuous connection of bony trabeculae between grafted bone and upper/lower vertebral bodies in plain anteroposterior and lateral radiograph. Preoperative and postoperative sagittal alignment was determined by manual measurements using the method of Cobb as applied to the supra and subadjacent endplates according to Klemme et al.23 In the case of spondylolisthesis, the pre- and postoperative percentage slip and its correction rate were measured. Furthermore, we compared radiographic results of the cases in which rigid spinal implants such as Steffee VSP, ISOLA spinal system, and M8 were used with those that used semirigid spinal implants such as the CY system or the NDMC system (using the χ2 test).

Table 2
Table 2:
Clinical Results of Treatment for Low Back Pain (JOA Score)


Clinical Results

Preoperative JOA scores ranged from −1 to 22 (mean 10.7). Postoperative JOA scores ranged from 13 to 29 (mean 26.9). The overall recovery rate ranged from 42.9% to 100% (mean 88.3%). Intra- and postoperative complications included a transient L5 nerve palsy due to pedicle screw malinsertion and a postoperative deep infection. The L5 nerve palsy recovered spontaneously 6 months after surgery, and the postoperative deep infection subsided by inserting a continuous irrigation system without removing the spinal implant. Continuous donor site pain of the iliac bone occurred in one patient.

Radiographic Results

Bony union was confirmed in 25 patients (96.2%). Pseudoarthrosis was suspected at one of the segments in the patient with the three-level fusion. A clear zone was observed in 9 of 68 contact surfaces (13.2%). Sinking was observed in 8 of 34 segments (23.5%), and slight cracking of the HA block was observed in 6 segments (17.6%). Four segments (11.7%) cracked during surgery when they were impacted into the intervertebral space and an additional two segments (5.9%) were found to be cracked on follow-up x-rays. However, no collapse of the HA blocks occurred, nor did any patient require additional surgery based on the condition of the HA blocks. HA block dislocation was not observed. Preoperative sagittal lumbar alignment ranged from 17° kyphosis to 25° lordosis (mean 5.1° lordosis), and postoperative sagittal lumbar alignment ranged from 5° kyphosis to 21° lordosis (mean 6.4° lordosis) in L4-L5. Preoperative sagittal lumbar alignment ranged from 6° kyphosis to 30° lordosis (mean 13.8° lordosis), and postoperative sagittal lumbar alignment ranged from 2° kyphosis to 29° lordosis (mean 13.1° lordosis) in L5-S1. Preoperative average percentage slip ranged from 11.1% to 27.5% (20.6%) in 11 patients with degenerative spondylolisthesis and spondylolytic spondylolisthesis. Postoperative percentage slip ranged from 7.1% to 19% (mean 9.4%), and correction rate ranged from 0% to 75.6% (mean 50.1%).

The frequency of the appearance of a clear zone between the HA spacer and vertebral bodies was 15% (6/40) in surgeries that used a rigid system and 10.7% (3/28) in surgeries that used a semirigid system. There was no statistical difference between these two groups.


A 39-year-old man had persistent low back and right leg pain. We found the straight-leg-raising test to be positive at 80° on the right leg. Neither sensory disturbance nor muscle weakness could be detected. His preoperative JOA score was 12. Conventional radiographs showed an L4 spondylolytic spondylolisthesis slippage of 27.5%. Preoperative sagittal alignment was 3° kyphosis. Sagittal T2-weighted magnetic resonance images revealed slight dural sac compression and disc degeneration at L4-L5 (Fig. 3). PLIF was performed at L4-L5 using a dense HA block and autogenous iliac bone. The JOA score improved to 29 points (degree of recovery 100%) at 5 years postoperatively. Bony union was observed and the percentage slip improved to 7.5% in conventional radiographs. Postoperative sagittal lumbar alignment was 5° lordosis (Fig. 4).

A 39-year-old man. A, Lateral plain radiograph shows an L4 spondylolytic spondylolisthesis with a 27.5% slippage. B, Sagittal T2-weighted magnetic resonance images reveal compression of the dural sac and disc degeneration at L4-L5.
Postoperative anteroposterior (A) and lateral (B) radiographs. Bony union and improvement of percentage slip were observed, and neither cracking nor sinking of HA block was seen.


Hydroxyapatite was synthesized by Aoki et al24 and has been used clinically in the field of orthopaedic surgery.25,26 In in vivo animal models, HA showed excellent biologic compatibility, although significant rates of implant collapse were reported in cervical anterior fusion.27,28 The value of compressive strength and tensile strength was reported to be higher in dense HA than in cortical bone and porous HA.29 In an in vivo animal investigation, dense hydroxyapatite and apatite/wollastonite glass ceramic were found to bind strongly to the bone, both with a bonding strength that did not decrease even at 25 weeks after implantation compared with that found at 8 weeks after surgery.30 Pintar et al,31 in an in vitro animal study, reported that the fusion rate is same in dense HA as it is in autogenous bone and that the rate is better in the lumbar spine than in the cervical spine.

Based on a large number of experimental studies, many authors advocated the clinical usefulness of the porous HA block for anterior cervical fusion.32,33 In contrast, very few papers21 reported on the use of porous HA blocks for lumbar spine surgery in PLIF because of the relative mechanical weakness of porous HA. We also worked with porous HA blocks for the PLIF in 1990 but abandoned this material because of its relative fragility as it tended to crack at the time of insertion and as well as exhibit occasional late collapse. The availability of dense HA with its improved mechanical properties led us to re-examine the utility of HA. To our knowledge, there have been no reports on the use of dense HA blocks in PLIF in the medical literature published in English. Retropulsion of the grafted bone was reported for the use of autogenous or cadaveric bone as a graft.6-8 Recently, retropulsion of titanium cage implants after PLIF was also reported.34 We suggest that retroversion can be avoided if care is taken that the HA block is placed at least 10 mm anterior from the posterior edge of the vertebrae and that adequate compressive force is added between the HA block and vertebral bodies. We have incorporated these technical details in our surgical protocols and have experienced retropulsion of neither the HA bock nor the grafted bone.

Lumbar sagittal contour after PLIF was reported by Klemme et al.23 According to them, there was a mean lordotic gain of 5°/segment in patients undergoing fusion with vertically oriented mesh cages combined with posterior compression instrumentation. In our study, a mean loss of lordosis of 2.4° was found. There seemed to be a limitation of the sagittal alignment correction in the use of dense HA for the PLIF even if a block with an incline is used.

The difficulty of radiographic evaluation especially using plain radiographs for the bony union when carbon fiber cages or titanium mesh cages was used is reported by many authors.17,19,35,36 Plain radiographic evaluation of extracage fusion may be facilitated by placement of graft outside and around cages against the decorticated endplate.35 In our series, bony union is easy to recognize in that the existence of continuous connections of trabeculae between grafted bone and adjacent vertebral bodies can be readily seen in plain anteroposterior or lateral radiographs. It supports the fact that the fusion mass outside and surrounding the cage is easier to assess with plain radiography than the fusion within the cage.35

We found that no statistical difference in the rate of appearance of a clear zone between patients receiving a rigid system and those treated with a semirigid system. This fact indicates that either a rigid or a semirigid spinal implant can be applicable for PLIF using a dense HA block and autogenous iliac bone.

Recently, the use of titanium alloy cages have become widespread as intervertebral spacers for PLIF14,15 because of their ease of use and mechanical strength. Kasai et al reported that one-third of patients with titanium alloy spinal implants exhibited abnormal serum or hair metal concentrations and titanium or aluminum may travel to distant organs after solubilization of the metals from the spinal implants.37 The safety of the titanium alloy cages, which cannot be removed, even after obtaining bony union, has not been established as yet. From the viewpoint of biologic safety, HA is superior to artificial material as an intervertebral spacer, although further improvement in its mechanical properties would be necessary. To simplify the surgical procedures, to shorten the operation time, and to reduce the blood loss, we are developing a new method for PLIF using two blocks of HAB and locally harvested autograft without the iliac crest bone. Although we currently have experience with only five patients, the short-term clinical results appear quite promising.


1. Jaslow LA. Intercorporal bone graft in spinal fusion after disc removal. Surg Gynecol Obstet. 1946;82:215-218.
2. Cloward RB. The treatment of ruptured lumbar intervertebral discs by vertebral body fusion. J Neurosurg. 1953;10:154-168.
3. Cloward RB. Posterior lumbar interbody fusion update. Clin Orthop. 1985;193:16-19.
4. Turner JA, Ersek M, Herron L, et al. Patient outcomes after lumbar spinal fusions. JAMA. 1992;268:907-911.
5. Arai Y, Takahashi M, Kurosawa H, et al. Comparative study of iliac bone graft and carbon cage with local bone graft in posterior lumbar interbody fusion. J Orthop Surg. 2002;10:1-7.
6. Hutter CG. Posterior intervertebral body fusion- a 25 year study. Clin Orthop. 1983;179:86-96.
7. Collis JS. Total disc replacement: a modified posterior lumbar interbody fusion. Report of 750 cases. Clin Orthop. 1985;193:64-67.
8. Lin PM. Posterior lumbar interbody fusion technique: complications and pitfalls. Clin Orthop. 1985;193:90-102.
9. Ma GWC. Posterior interbody fusion with specialized instruments. Clin Orthop. 1985;193:57-63.
10. Brantigan JW. Pseudoarthrosis rate after allograft posterior lumbar interbody fusion with pedicle screw and plate fixation. Spine. 1994;19:1271-1280.
11. Summers BN, Eisenstein SM. Donor site pain from the ilium. J Bone Joint Surg Br. 1989;71:677-680.
12. Younger EM, Chapman MW. Morbidity at bone graft donor sites. J Orthop Trauma. 1989;3:192-195.
13. Shimizu K, Iwasaki R, Matsushita M, et al. Posterior lumbar interbody fusion using AW-GC vertebral spacer. Bioceramics. 1992;5:435-441.
14. Ray CD. Threaded titanium cage for lumbar interbody fusions. Spine. 1997;22:667-680.
15. Kevin RE, Bridwell KH, Ungacta FF, et al. Analysis of titanium mesh cage in adults with minimum two-year follow-up. Spine. 2000;25:2407-2415.
16. Brantigan JW, Steffee AD. A carbon fiber implant to aid interbody lumbar fusion. Spine. 1993;18:2106-2117.
17. Tullberg T, Brandt B, Rydberg J, et al. Fusion rate after posterior lumbar interbody fusion with carbon fiber implant: 1-year follow-up of 51 patients. Eur Spine J. 1996;5:178-182.
18. Agazzi S, Reverdin A, May D. Posterior lumbar interbody fusion with cages; an independent review of 71 cases. J Neurosurg (Spine 2). 1999;91:168-192.
19. Brantigan JW, Steffee AD, Lewis ML, et al. Lumbar interbody fusion using the Brantigan I/F cage for posterior lumbar interbody fusion and the variable pedicle screw placement system. Two-year results from a Food and Drug Administration investigational device exemption clinical trial. Spine. 2000;25:1437-1446.
20. Chitnavis B, Barbagallo G, Selway R, et al. Posterior lumbar interbody fusion for revision disc surgery: review of 50 cases in which carbon fiber cages were implanted. J Neurosurg (Spine 2). 2001;95:190-195.
21. Suzuki N, Iwamoto Y. Use of hydroxyapatite blocks in posterior interbody fusion. In: Yonenobu K, Ono K, Takemitsu K, eds. Lumbar Fusion and Stabilization. Tokyo: Springer-Verlag; 1993:371-378.
22. Izumida S, Inoue S. Assessment of treatment for low back pain. Japanese Orthopaedic Association. J Jpn Orthop Assoc. 1986;60:391-394.
23. Klemme WR, Owens BD, Dhawan A, et al. Lumbar sagittal contour after posterior interbody fusion. Spine. 2001;26:534-537.
24. Aoki H, Kato K. Application of apatite to biomaterials. Ceramics. 1975;10:469-478. Japanese.
25. Holms RE, Bucholz RW, Mooney V. Porous hydroxyapatite as a bone-graft substitute in metaphyseal defects. J Bone Joint Surg Am. 1986;68:904-911.
26. Bucholz RW, Carlton A, Holms RE. Hydroxyapatite and tricalcium phosphate bone graft substitutes. Orthop Clin North Am. 1987;18:323-334.
27. Cook SD, Dalton JE, Tan EH, et al. In vivo evaluation of anterior cervical fusions with hydroxylapatite graft material. Spine. 1994;19:1856-1866.
28. Zdeblick TA, Cooke ME, Kunz DN, et al. Anterior cervical discectomy and fusion using a porous hydroxyapatite bone graft substitute. Spine. 1994;19:2438-2457.
29. Jarcho M. Calcium phosphate ceramics as hard tissue prosthetics. Clin Orthop. 1981;157:259-278.
30. Yamamuro T, Nakamura T, Higashi S, et al. Artificial bone for use as a bone prosthesis. In: Atsumi K, Maekawa M, Ota K, eds. Progress in Artificial Organs-1983, Vol. 2. Cleveland: ISAO Press; 1984:810-814.
31. Pintar FA, Maiman DJ, Hollowell JP, et al. Fusion rate and biomechanical stiffness of hydroxylapatite versus autogenous bone grafts for anterior discectomy. Spine. 1994;19:2524-2528.
32. Koyama T, Handa J. Porous hydroxyapatite ceramics for use in neurosurgical practice. Surg Neurol. 1986;25:71-73.
33. Kim P, Wakai S, Matsuo S, et al. Bisegmental cervical interbody fusion using hydroxyapatite implants: surgical results and long-term observation in 70 cases. J Neurosurg. 1998;88:21-27.
34. Uzi EA, Dabby D, Tolessa E, et al. Early retropulsion of titanium-threaded cages after posterior lumbar interbody fusion. A report of two cases. Spine. 2001;26:1073-1075.
35. Eck KR, Bridwell KH, Ungacta FF, et al. Analysis of titanium mesh cages in adults with minimum two-year follow-up. Spine. 2000;25:2407-2415.
36. Cizek GR, Boyd LM. Imaging pitfalls of interbody spinal implants. Spine. 2000;25:2633-2636.
37. Kasai Y, Iida R, Uchida A. Metal concentrations in the serum and hair of patients with titanium alloy spinal implants. Spine. 2003;28:1320-1326.

hydroxyapatite; posterior lumbar interbody fusion; clear zone

© 2005 Lippincott Williams & Wilkins, Inc.