Autogenous bone graft has been found to be more effective than allograft bone for augmenting fusion in spinal arthrodesis3,11. The posterior iliac crest is a readily available source of bone graft for posterior arthrodesis of the spine. However, the donor site is frequently a source of morbidity, including infection, hematoma, pain, numbness, and hypersensitivity or irritability of local tissue6,9,10. Some complications, such as hematoma and infection, need additional intervention and come to the attention of the surgeon; some complications, however, such as persistent pain, neuroma, skin irritability, and numbness caused by injury to the cluneal nerves, may be overlooked16.
We compared donor site morbidity associated with the procurement of bone graft from the iliac crest through an incision parallel to the superior cluneal nerves with that associated with procurement through an incision parallel to the posterior iliac crest.
Materials and Methods
Bone graft was procured from the iliac crest of sixty consecutive patients through a modified incision1 parallel to the superior cluneal nerves and perpendicular to the posterior iliac crest (the study group) after we noted the morbidity associated with the use of a standard oblique incision parallel to the posterior iliac crest. We compared donor site morbidity in this group of sixty patients with that in fifty-nine consecutive patients who had had a standard incision before we elected to modify the incision (the so-called control group). The intent to study the two groups of patients was presented to the Institutional Review Board, which had no objection to the change in incision. The primary operation consisted of spinal reconstruction in 106 patients (fifty-three in each group), spinal arthrodesis to treat traumatic injury in ten patients (five in each group), and bone-grafting to treat non-union of a long bone in three patients (two in the study group and one in the control group).
The standard incision, which was approximately ten centimeters long, was made parallel to and directly over the posterior crest and extended laterally from the posterior superior iliac spine, and the modified incision was made perpendicular to the posterior iliac crest (Fig. 1). The dissection was extended onto the iliac wing, and the fascia was elevated to expose approximately ten centimeters of the iliac crest without an attempt to visualize the cluneal nerves. The method used to elevate the fascia and to procure the bone was similar in the two groups. Care was taken to avoid broaching of the inner table of the ilium, the sacro-iliac joint, and the greater sciatic notch. The wound was closed in anatomical layers over a large Gelfoam pledget (Upjohn, Kalamazoo, Michigan) without a drain.
Six patients in the control group and three in the study group were lost to follow-up. An interviewer, who was blinded with respect to the type of incision that had been used, reviewed the records of the remaining 110 patients and interviewed the patients over the telephone within one year after the operation. The patients were asked to describe the symptoms at one and six months after the operation. Demographic analysis included the initial diagnosis, symptoms (sciatica or pain in the buttock), and body-mass index8,17 (a reflection of height and weight). An analog scale of 0 to 10 points was used to rank the severity of the pain at the primary operative and donor sites. The use of analgesics was recorded with use of a 4-point analog scale. One point indicated no use of drugs; 2 points, use of non-steroidal drugs only; 3 points, occasional use of narcotics; and 4 points, regular use of narcotics. The use of analgesics or walking aids; a limp, tenderness, or skin irritability over the donor site during activities of daily living; numbness over the donor site and cephalad portion of the buttock; deep pain at the donor site; satisfaction of the patient with the result of the primary operation; postoperative wound complications; and a pending Workers' Compensation claim were also recorded.
Statistical analysis included two-tailed t tests of analog ranked and continuous variables and chi-square analysis of non-continuous variables. A p value of less than 0.05 was considered significant.
Analysis of confounding variables and intergroup homogeneity showed that the rate of preoperative sciatica (forty-five patients [85 per cent] in the control group and fifty-one [89 per cent] in the study group) and postoperative sciatica (twenty-one and twenty-eight patients [40 and 49 per cent], respectively), the prevalence of pain in the buttock (thirty-seven and forty-three patients [70 and 75 per cent]), the body-mass index (evenly distributed), the mean analog score for pain at the primary operative site at one month (6.1 and 5.7 points) and at six months (2.0 and 1.8 points) after the operation, and the mean analog score for the level of satisfaction with the result of the primary procedure (2.0 and 2.2 points) were similar in the two groups.
The two groups were also similar with respect to age, gender distribution, and underlying disease. The only potential confounding variable between the two groups was a pending Workers' Compensation claim: thirty-nine patients in the control group and twenty-eight patients in the study group had a claim pending (p < 0.05). Patient demographics, presenting symptoms, and pain at the primary operative and donor sites were similar in terms of severity and frequency regardless of whether or not there was a claim pending. However, analysis of six pooled symptoms six months after the operation showed a significant difference in frequency between the two groups. At one month postoperatively, the compensation claimants had a mean of four symptoms compared with a mean of three for the other patients (p = 0.2); at six months, the claimants had a mean of three symptoms compared with a mean of two for the other patients (p = 0.002). The level of satisfaction with the result of the primary operation was the same for the claimants as for the other patients.
Donor Site Morbidity
The donor site was a frequent source of symptoms in all patients at one and six months after the operation; however, the frequency and severity of symptoms were consistently greater in the control group. The mean analog score for pain at the donor site was 7 points in the control group and 6 points in the study group (p = 0.001) at one month and 3 and 2 points, respectively, at six months (p = 0.001). The score for use of analgesics at one month was 4 points in the control group and 3 points in the study group. These differences were significant (p = 0.03). Forty-two patients (79 per cent) in the control group used narcotics regularly to alleviate pain one month after the operation, compared with thirty-one patients (54 per cent) in the study group. At six months, analgesic use was similar in the two groups.
The prevalence of symptoms was greater in the control group than in the study group. At one month, thirty-nine patients (74 per cent) in the control group had numbness, compared with twenty-five (44 per cent) in the study group (p = 0.001); at six months, thirty-one patients (58 per cent) in the control group had numbness, compared with fourteen (25 per cent) in the study group (p = 0.0002). Thirty-six patients (68 per cent) in the control group and twenty-four patients (42 per cent) in the study group (p = 0.005) had tenderness over the incision at one month, and twenty-seven (51 per cent) in the former group and eleven (19 per cent) in the latter (p = 0.0003) had tenderness at six months. Twenty-four patients (45 per cent) in the control group had irritation over the donor site during activities of daily living at six months, compared with sixteen patients (28 per cent) in the study group (p = 0.046). Pooling of these symptoms with a limp, the use of walking aids, and deep pain at the donor site resulted in a mean of four symptoms in the control group and a mean of three in the study group (p = 0.02) at one month and of three and two, respectively, at six months (p = 0.001).
Forty patients (75 per cent) in the control group had deep pain at the donor site at one month, compared with forty-six patients (81 per cent) in the study group (p = 0.29). At six months, thirty-two patients (60 per cent) in the control group had deep pain at the donor site, compared with thirty-one patients (54 per cent) in the study group (p = 0.49). The increased prevalence of morbidity in the control group did not result in an appreciable difference in the level of satisfaction with the operative treatment for the primary disease: forty-four patients (83 per cent) in the control group and fifty patients (88 per cent) in the study group were satisfied with the result (p = 0.13).
Complications in the control group included superficial infection (two patients), wound dehiscence (one patient), hematoma (one patient), and hypertrophic scar (one patient). Complications in the study group included superficial infection (one patient), hematoma (two patients), and hypertrophic scar (one patient).
Bone for use as a graft can be obtained from the patient or from a donor. The transmission of disease through allograft bone continues to be a possibility despite standardized bone-banking and tissue-testing protocols2. The use of autogenous bone alleviates this risk, but it introduces a potential source of morbidity at the donor site. Efforts to decrease this morbidity include the use of percutaneous methods when a limited volume of bone graft is needed15,18; however, an open approach is needed to obtain larger amounts of bone.
Kurz et al. and Banwart et al. reported several complications associated with the procurement of bone graft from the iliac crest. The major early complications were deep wound infection and neurovascular injury, the minor early complications were hematoma and severe pain, the major late complications were hernia and pelvic instability, and the minor late complications were delayed wound-healing and chronic pain. Younger and Chapman, in a retrospective study of fifty-four patients in whom the incision for the primary procedure was also used to procure the graft from the posterior iliac crest, reported eleven (20 per cent) major and fifteen (28 per cent) minor complications.
A review of the anatomy of the iliac crest is essential to decrease the associated morbidity. The superior cluneal nerves arise from the first, second, and third lumbar dorsal rami and provide sensation to the region of the posterior iliac crest and the cephalad portion of the buttock12,13. They pass deep to the erector spinae muscles and then through both the deeper dorsolumbar fascia and the more superficial layer of the latissimus dorsi. The nerves cross the iliac crest approximately eight centimeters lateral to the midline12 and seven centimeters cephalad to the posterior superior iliac crest19 (Fig. 2). They pass through the thoracolumbar fascia proximal to the crest and lie in the subcutaneous tissue over the ilium, and they are at risk of being divided or of sustaining a traction injury during procurement of a bone graft13,16. In order to avoid this complication, several alternative incisions have been suggested, including a midline incision followed by a subcutaneous or subfascial approach onto the ilium7,14 as well as a separate lateral incision over the iliac crest in line with the iliac wing staying close to the posterior superior iliac spine in order to avoid injury to the superior cluneal nerves5,7,13.
The use of a separate incision parallel to the posterior iliac crest and limitation of the dissection to an area within eight centimeters of the posterior superior iliac spine have been reported to prevent injury to the superior cluneal nerves5,13. Despite our attempts to adhere to this recommendation in the control group, thirty-one patients (58 per cent) had areas of numbness over the incision and the cephalad portion of the buttock. It is uncertain whether these symptoms were secondary to a traction injury or to division of the cluneal nerves at the time of the operation.
Fernyhough et al. reported chronic pain at the donor site one year postoperatively in forty-four (29 per cent) of 151 patients, suggesting that donor site morbidity may not be a temporary condition. Those authors found that the separate incision over the iliac crest was associated with a greater frequency of altered sensation (43 per cent; ten of twenty-three) compared with the midline subcutaneous approach (19 per cent; four of twenty-one). The two groups of patients were dissimilar, and a larger number of patients, who were managed for a traumatic injury to the spine, had a midline incision.
Hutchinson and Dall reported a 15 per cent rate of complications after 200 midline subcutaneous approaches to the posterior iliac crest compared with a rate of 3 per cent after 200 midline fascial-splitting approaches. The fascial-splitting approach makes it possible to dissect the soft tissues within the relatively avascular planes of the dorsolumbar fascia and more superficial fascia of the latissimus dorsi, possibly accounting for the infrequent formation of hematomas.
Kreibich et al. reported moderate or severe postoperative pain after 25 per cent (ten) of forty procedures performed through a percutaneous approach and after 61 per cent (twenty) of thirty-three procedures performed with an open technique.
While it has been suggested that an incision perpendicular to the posterior iliac crest may decrease morbidity at the donor site4,5, we are not aware of any study to support this contention. The modified incision (Fig. 1) used in the present study minimizes injury to the cluneal nerves by dissection of the subcutaneous tissues in line with the course of the nerves over the iliac crest, as described (but not evaluated) by Craig and Asher.
We assessed the impact of donor site morbidity on the patient by analyzing the activities of daily living and any limitation of function. In the present study, a single structured interview, with the interviewer blinded as to the incision used, focused on subjective donor site morbidity one and six months after the operation. The use of analog scales made it possible to quantify pain at both the primary operative site and the donor site. Donor site morbidity and pain at the primary operative site were not significantly different between patients who had a pending compensation claim and the remaining patients at one month but were significantly different at six months. It is uncertain whether this reflects issues of potential secondary gain. Deep pain at the donor site associated with the actual procurement of bone graft from the ilium was similar in the control and study groups. Therefore, we believe that the difference in donor site morbidity between those groups was attributable to the use of the modified incision.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
Investigation performed at McMaster University, Hamilton
1. Banwart, J. C.; Asher, M. A.; and Hassanein, R. S.: Iliac crest bone graft harvest donor site morbidity. A statistical evaluation. Spine
, 20: 1055-1060, 1995.
2. Buck, B. E.; Malinin, T. I.; and Brown, M. D.: Bone transplantation and human immunodeficiency virus. An estimate of risk of acquired immunodeficiency syndrome (AIDS). Clin. Orthop.
, 240: 129-136, 1989.
3. Burchardt, H.: The biology of bone graft repair. Clin. Orthop.
, 174: 28-42, 1983.
4. Cockin, J.: Autologous bone grafting—complications at the donor site. In Proceedings of the British Orthopaedic Society. J. Bone and Joint Surg.
, 53-B(1): 153, 1971.
5. Craig, C. C., and Asher, M. A.: Hemostasis in human iliac crest donor sites with microfibrillar collagen. Spine
, 2: 313-317, 1977.
6. de Boer, H. H.: The history of bone grafts. Clin. Orthop.
, 226: 292-298, 1988.
7. Fernyhough, J. C.; Schimandle, J. J.; Weigel, M. C.; Edwards, C. C.; and Levine, A. M.: Chronic donor site pain complicating bone graft harvesting from the posterior iliac crest for spinal fusion. Spine
, 17: 1474-1480, 1992.
8. Frankel, H. M.: Determination of body mass index [letter]. J. Am. Med. Assn.
, 255: 1292, 1986.
9. Friedlaender, G. E.: Current concepts review. Bone grafts. The basic science rationale for clinical applications. J. Bone and Joint Surg.
, 69-A: 786-790, June 1987.
10. Friedlaender, G. E.: Bone grafts. In Orthopaedic Knowledge Update 3; Home Study Syllabus, p. 172. Park Ridge, Illinois, The American Academy of Orthopaedic Surgeons, 1990.
11. Goldberg, V. M., and Stevenson, S.: Natural history of autografts and allografts. Clin. Orthop.
, 225: 7-16, 1987.
12. Grant's Atlas of Anatomy, edited by J. E. Anderson. Ed. 8, plates 5-25 to 5-29. Baltimore, Williams and Wilkins, 1983.
13. Hoppenfeld, S., and deBoer, P.: Surgical Exposures in Orthopaedics. The Anatomic Approach, pp. 295-299. Philadelphia, J. B. Lippincott, 1984.
14. Hutchinson, M. R., and Dall, B. E.: Midline fascial splitting approach to the iliac crest for bone graft. A new approach. Spine
, 19: 62-66, 1994.
15. Kreibich, D. N.; Scott, I. R.; Wells, J. M.; and Saleh, M.: Donor site morbidity at the iliac crest: comparison of percutaneous and open methods. J. Bone and Joint Surg.
, 76-B(5): 847-848, 1994.
16. Kurz, L. T.; Garfin, S. R.; and Booth, R. E. Jr.: Harvesting autogenous iliac bone grafts. A review of complications and techniques. Spine
, 14: 1324-1331, 1989.
17. National Institutes of Health Consensus Development Panel: Health implications of obesity. National Institutes of Health Consensus Development conference statement. Ann. Intern. Med.
, 103: 147-151, 1985.
18. O'Neill, D. B.: Tips of the trade: 22. Simplified iliac crest cancellous grafting. Orthop. Rev.
, 19: 383-384, 1990.
19. Xu, R.; Ebraheim, N. A.; Yeasting, R. A.; and Jackson, W. T.: Anatomic considerations for posterior iliac bone harvesting. Spine
, 21: 1017-1020, 1996.
20. Younger, E. M., and Chapman, M. W.: Morbidity at bone graft donor sites. J. Orthop. Trauma, 3: 192-195, 1989.