- Email: [email protected]
Bone graft harvesting from the distal radius, olecranon, and iliac crest: A quantitative analysis
Bone Graft Harvesting From the Distal Radius, Olecranon, and Iliac Crest: A Quantitative Analysis Roderick J. Bruno, MD, Mark S. Cohen, MD, Aivars Berzins, MD, Dale R. Sumner, PhD, Chicago, IL The purpose of this study was to quantify the volume and density of cancellous bone available from 3 commonly used bone graft sites in upper extremity surgery: the distal radius, olecranon process, and anterior iliac crest. Sixteen cadavers (age range, 59 –98 years) with no prior history of bone harvest or metabolic conditions affecting bone were used. Cancellous graft was obtained using standardized techniques designed to simulate the clinical setting. Packed cancellous bone volume was determined as the volume occupied by the harvested bone after compression packing with a uniform load. Defect volume was determined by measuring the volume of the site from which bone was harvested. The distal radius and olecranon provided similar volumes of packed cancellous bone (2.7 and 2.8 cc, respectively). The anterior iliac crest provided approximately twice this amount (5.3 cc). The packed cancellous bone volume/defect volume ratio was not different between the 3 sites studied. Male gender was associated with a significantly greater amount of packed cancellous bone volume for all sites. We believe the olecranon to be an alternative to the distal radius as a source of bone graft for upper extremity procedures. We suggest using the anterior iliac crest when a large volume of cancellous bone is required. (J Hand Surg 2001;26A:135–141. Copyright © 2001 by the American Society for Surgery of the Hand.) Key words: Bone graft, iliac crest, radius, olecranon.
Cancellous bone graft is often used to fill gaps and facilitate bone union in fracture treatment and arthrodesis. Numerous sites have been advocated for obtaining graft for use in the upper extremity. There is little consensus, however, regarding the best location for harvesting bone for a given need. Furthermore, information is lacking on the quantity and
quality of cancellous bone available from different sites. The purpose of this study was to determine the amount of bone graft available from the distal radius, olecranon process, and anterior iliac crest using standardized techniques. By measuring the harvested defect, bone density was also determined. This information will aid in determining the optimum location for bone harvest in upper extremity procedures.
From the Departments of Orthopaedic Surgery and Anatomy, RushPresbyterian-St Luke’s Medical Center, Chicago, IL. Received for publication December 3, 1999; accepted in revised form August 13, 2000. 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. Reprint requests: Mark S. Cohen, MD, 1725 W Harrison St, Suite 1063, Chicago, IL 60612. Copyright © 2001 by the American Society for Surgery of the Hand 0363-5023/01/26A01-0021$35.00/0 doi: 10.1053/jhsu.2001.20971
Materials and Methods Sixteen fixed cadaveric specimens were studied. Distal radius, olecranon, and iliac crest cancellous bone grafts were obtained bilaterally. No specimen used in the study had previous distal radius or olecranon fracture or previous iliac crest bone graft harvesting. Cadavers with a history of metabolic bone disease or carcinoma were excluded. There were 6 white men, 8 white women, 1 African-American The Journal of Hand Surgery 135
136 Bruno et al / Bone Graft Harvesting
man, and 1 African-American woman. The mean age at the time of death was 81 ⫾ 11 years (⫾SD; range, 59 –98 years). The technique of bone graft harvest was standardized to simulate the clinical setting.1,2 For the distal radius, bone graft was obtained by creating an opening in the dorsal cortex with a 1-cm dowel (Kaltec Precision Bone Graft System, Adelaide, South Australia). The distal extent of the defect was placed 1 cm proximal to the end of the radius in line with Lister’s tubercle (Fig. 1). For the olecranon, a similar 1-cm opening was made in the cortex beginning 1 cm distal to the tip of the olecranon (Fig. 2). Iliac crest cancellous bone graft was obtained through a 1-cm rectangular opening in the superior cortex of the crest. The anterior extent of the defect was placed 1 cm posterior to the anterior superior iliac spine (Fig. 3). For all regions studied, cancellous bone was harvested using straight and curved curettes (Codman 23-1046/56, Randolph, MA), taking care not to violate the cortical defect created. All grafts were obtained by the same 2 individuals working together. Care was taken to remove all available bone from the individual harvest sites without a time limit. The cortical entry fragment was stripped of all cancellous bone and not included in the graft measurements. Bone was collected in an airtight container for subsequent analysis. The volume of the defect created by the removal of cancellous bone was measured by injecting each site with a silicone polymer (General Electric, Waterford, NY). This material was chosen based on its ability to adequately fill the created defect and subsequently harden for analysis. The filler cured in a uniform fashion and was solid within 48 hours. It was then harvested by removing enough cortical bone from each site to facilitate extraction (Fig. 4). The volume of each donor site was initially determined by both a volume displacement method and by weighing the samples and calculating the volume based on the known density of the polymer. No significant differences were observed between these methods; thus, the volume calculation by weight was used due to its ease. Quantitative analysis of the harvested bone was performed by determining the “packed bone volume,” a measure of the “finger-packed” volume.3 To accurately and consistently reproduce the force used to make this determination, a servohydraulic material testing machine was used to compress the harvested graft material (Instron 1321, Canton, MA). A com-
Figure 1. (A) Distal radius harvest was performed through a dorsal cortical window placed 1 cm proximal to the distal extent of the radius in line with Lister’s tubercle. (B) The window was created with a 1-cm dowel. (C) Cancellous graft was obtained using straight and curved curettes. (D) The defect created was filled with a silicone polymer.
paction force of 40 N was chosen based on preliminary trials of finger packing. This force was then used for graft compression at a rate of 250 mm/min. Paired t-tests for the 2-volume measurements (packed bone volume and defect volume) indicated that there were no significant side-to-side differences. Therefore, the values for the 2 sides were averaged to obtain one set of volume measurements per site per cadaver. The ratio of the packed bone volume to the defect volume was also calculated,
The Journal of Hand Surgery / Vol. 26A No. 1 January 2001 137
t-tests, as appropriate, to determine which specific differences were significant. The analyses were performed with a statistical package (SPSS for Windows, release 8.0.0; SPSS, Inc, Chicago, IL).
Results The absolute amount of bone (ie, packed bone volume) was nearly 2-fold greater at the iliac crest than the distal radius or olecranon (Table 1). This difference was statistically significant (Table 2). Specifically, both the distal radius and the olecranon yielded less packed bone volume than the iliac crest. The variability of packed bone volume obtained was
Figure 2. (A) Olecranon harvest was performed though a cortical window 1 cm proximal to the tip. (B) A 1-cm dowel was used to create the cortical defect. (C) Curved and straight curettes were used to remove cancellous graft. (D) The defect was filled with a silicone polymer for determination of the defect volume.
providing a comparison of relative bone densities from the 3 harvest sites studied. The data were analyzed by repeated-measures ANOVA with anatomic site as the within-subjects factor, gender as the between-subjects factor, and age as a covariate. Mauchly’s test of sphericity was applied to the within-subjects effect, necessitating use of a multivariate test for site and the site-by-age and site-by-gender interactions (specifically, Pillai’s trace). Analyses showing a significant main effect were further analyzed with paired t-tests or Student’s
Figure 3. (A) A rectangular cortical window was made in the anterior iliac crest beginning 1 cm posterior to the anterior superior iliac spine. (B) The cortical window was 1 cm long. Pure cancellous graft was harvested with curved and straight curettes to simulate the clinical setting. (C) The defect created was filled with the silicone polymer between the intact tables.
138 Bruno et al / Bone Graft Harvesting
Figure 4. Cured silicone spacers harvested from the (left) iliac crest, (center) olecranon, and (right) distal radius donor sites. Note that the shape of each filler is analogous to the medullary cavity of the respective bones. These were used to calculate defect volume following harvest.
relatively constant, as the coefficient of variation (mean ⫾ SD ⫻ 100%) was 33%, 25%, and 32% for the distal radius, olecranon, and iliac crest, respectively. Gender was found to be a significant betweensubjects effect for the packed bone volume, with greater amounts harvested from the men than from
the women (Table 2). In general, approximately 50% more packed bone volume was obtained from the men than from the women (3.3 ⫾ 0.7 vs 2.2 ⫾ 0.7 cc for the distal radius, 3.4 ⫾ 0.7 vs 2.4 ⫾ 0.4 cc for the olecranon, and 6.2 ⫾ 1.0 vs 4.6 ⫾ 1.8 cc for the iliac crest in males and females, respectively.) The differences among the sites for the defect
Table 1. Volumes of Packed Bone, the Defect, and the Ratio of Packed Bone to the Defect Volume Packed bone (cc) Defect (cc) Packed bone/defect ratio
Distal Radius
Olecranon
Iliac Crest
2.7 ⫾ 0.9 (1.45–4.05) 5.0 ⫾ 1.2 (3.24–8.12) 0.5 ⫾ 0.1 (0.33–0.81)
2.8 ⫾ 0.7 (1.90–4.35) 4.3 ⫾ 0.5 (3.09–5.29) 0.7 ⫾ 0.2 (0.39–0.87)
5.3 ⫾ 1.7 (2.95–9.00) 10.8 ⫾ 5.3 (5.52–22.99) 0.5 ⫾ 0.2 (0.28–0.93)
Data are presented as mean values ⫾ SD (range).
The Journal of Hand Surgery / Vol. 26A No. 1 January 2001 139
Table 2. Significance Levels From the Analyses of Variance Variable
Site
Gender
Age
Site by Age
Site by Gender
Packed bone Defect Packed bone/defect ratio
.031* .689 .203
.004† .060 .162
.458 .763 .575
.135 .968 .341
.459 .165 .230
P values less than .05 were considered to be significant and are shown in bold. *Radius ⬍ iliac crest (p ⬍ .001), olecranon ⬍ iliac crest (p ⬍ .001). †Female ⬍ male for radius (p ⫽ .006), olecranon (p ⫽ .004), and iliac crest (p ⫽ .052).
volume and packed bone/defect volume ratio were not significantly significant (p ⬎ .05). In addition, except as noted for the packed bone volume, gender was not a significant factor. Finally, age was not a significant covariate in any of the analyses, nor were there significant statistical interactions between anatomic site and age or between anatomic site and gender.
Discussion Autologous cancellous bone graft is used in upper extremity surgery for acute fracture care, nonunions, arthrodesis, and filling of osseous tumor defects. The most common site for bone graft harvest is the anterior iliac crest. This provides an abundant amount of cancellous bone but does require preparation of a distant anatomic site with associated donor site morbidity4,5 Furthermore, general anesthesia is typically required for iliac crest harvest and ambulation can be problematic when surgery is performed on an outpatient basis. In an effort to decrease the morbidity of iliac crest harvest, percutaneous techniques with biopsy trochars have been introduced.6 The iliac crest is often covered by a thick soft tissue envelope, however, and percutaneous techniques also can be complicated by bleeding and hematoma formation.7 For upper extremity procedures, various investigators advocate obtaining bone graft from the olecranon or distal radius.8 –14 This allows the surgical procedure to be limited to the involved extremity without additional anesthetic requirements. The primary disadvantage of local upper extremity graft is the limited amount of bone available. There also are isolated reports of pathologic fracture following olecranon and distal radius harvest, although this is quite rare.8,15 No quantitative data have previously existed regarding the actual amount of bone that can be obtained from the different sites. Several studies have attempted to evaluate healing using upper extremity sites in lieu of the iliac crest.
Hull et al16 compared iliac crest graft with the distal radius for scaphoid nonunion, reporting a 73% union rate for iliac crest graft versus a 47% union rate for the distal radius graft. These investigators surmised that the increased union rate seen with the iliac crest was a function of the osteogenic potential of the bone. All the iliac grafts were corticocancellous, however, and were placed palmarly. Eleven of the 17 radius grafts involved only a plug of bone inserted through a drill hole in the scaphoid and only 1 of these achieved union. Schnitzler et al17 compared bone structure between the distal radius and the iliac crest in vitro and concluded that the distal radius was structurally inferior and had a lower bone turnover compared with the iliac crest. These investigators pointed out, however, that radius autograft may very well be suitable for “compacting into a small bone defect.” Furthermore, the biologic differences were felt to be of minimal clinical significance when the bone was used as graft material.18 For hand and wrist procedures, numerous reports suggest that the distal radius is a useful option.11–14 Mirly et al8 reported an 82% healing rate when anterior distal radius graft was used for cysts, arthrodesis, and scaphoid nonunions. One complication, a fracture through the donor site at approximately 4 years after surgery, occurred in 131 cases. The fracture was nondisplaced and healed uneventfully. McGrath and Watson9 reported on local bone graft for 124 procedures over a 10-year period. Bone was obtained from the distal radius in 78 cases, the proximal ulna in 19 cases, and the small hand bones in 27 cases. At a minimum follow-up period of 22 months, they had 1 nonunion and 1 patient with chronic donor site pain. These investigators concluded that upper extremity regional bone grafts provide a reliable and adequate amount of bone with minimal complications for upper extremity procedures. They report that “several” cubic centimeters of cancellous bone
140 Bruno et al / Bone Graft Harvesting
can be obtained from the distal radius and that up to 4.5 cubic centimeters of bone is available from the proximal ulna. The harvested bone volumes, however, were not further quantified. We evaluated bone graft harvested from the distal radius and olecranon relative to that from the anterior iliac crest. We found the average volume of packed cancellous bone obtainable from the distal radius to be similar to the olecranon and approximately half of that available from the anterior iliac crest (Table 1). We also found approximately twice as much bone available in males than females in the population studied. It must be noted, however, that a considerable range exists for each site, most likely based on the size and gender of the individual. Furthermore, the actual volume of bone available in vivo is most likely somewhere in between our measured packed and defect volumes, especially in younger individuals in the absence of osteopenia. Perhaps the major limitation of this study is the age of the cadaveric specimens. One can only extrapolate our findings to young adults if the relative amounts of cancellous bone found at the distal radius, olecranon, and iliac crest are reasonably constant from skeletal maturity through aging. Although there are no data on cancellous bone volume for the olecranon, the existing data for the iliac crest and distal radius indicate that the age-related rate of cancellous bone loss is roughly equivalent at these 2 sites.19 –25 Thus, we believe it is safe to assume that in the patient population most often in need of upper extremity bone grafting, the iliac crest would supply approximately twice the volume of cancellous bone as the distal radius or proximal ulna. Our findings indicate that the olecranon is an alternative to the distal radius as a source of bone graft in upper extremity procedures and that it is a viable option when the distal radius is not available, eg, for distal radius fracture. When a large volume of cancellous bone is required, the anterior iliac crest provides approximately twice the amount of cancellous autograft and would remain the preferred site. This information should be useful when planning cancellous bone harvest for a particular need in the upper extremity.
References 1. Chapman MW. Bone grafting. In: Chapman MW, ed. Operative orthopaedics. 2nd ed. Philadelphia: JB Lippincott, 1993:139 –149.
2. Crenshaw AH. Surgical techniques and approaches. In: Canale ST, ed. Campbell’s operative orthopaedics. 9th ed. St Louis: Mosby, 1998:40 – 47. 3. Lindberg EJ, Katchis SD, Smith RW. Quantitative analysis of cancellous bone graft available from the greater trochanter. Foot Ankle Int 1996;17:473– 476. 4. Beirne JC, Barry HJ, Brady FA, Morris VB. Donor site morbidity of the anterior iliac crest following cancellous bone harvest. Int J Oral Maxillofacial Surg 1996;25:268 – 271. 5. Younger EM, Chapman MW. Morbidity at bone graft donor sites. J Orthop Trauma 1989;3:192–195. 6. Gartsman GM, Lane JM. A simple method of obtaining bone graft by bone biopsy trochar. J Hand Surg 1981;6: 627– 628. 7. Billmire DA, Rotatori S. Use of the CORB needle biopsy for the harvesting of iliac crest bone graft. Plast Reconstr Surg 1994;93:416 – 418. 8. Mirly HL, Manske PR, Szerinski JM. Distal anterior radius bone graft in surgery of the hand and wrist. J Hand Surg 1995;20A:623– 627. 9. McGrath MH, Watson HK. Late results with local bone graft donor sites in hand surgery. J Hand Surg 1981;6:234 – 237. 10. Voche P, Merle M. Removal of cancellous bone graft through the base of the excised radial styloid process. J Hand Surg 1992;17B:499 –500. 11. Eckenrode JF, Louis DS, Greene TL. Scaphoid-trapeziumtrapezoid fusion in the treatment of chronic scapholunate instability. J Hand Surg 1986;11A:497–502. 12. Frykman EB, Ekenstam FA, Wadin K. Triscaphoid arthrodesis and its complications. J Hand Surg 1988;13A:844 – 849. 13. Kleinman WB, Steichen JB, Strickland JW. Management of chronic rotary subluxation of the scaphoid by scaphotrapezio-trapezoid arthrodesis. J Hand Surg 1982;7:125– 136. 14. Watson HK, Hempton RF. Limited wrist arthrodeses. I. The triscaphoid joint. J Hand Surg 1980;5:320 –327. 15. Walker LG, Meals RA. Pathologic fractures of the proximal ulna through a bone graft donor site. J Hand Surg 1990;15A:781–784. 16. Hull WJ, House JH, Gustillo RJ, Kleven L, Thompson W. The surgical approach and source of bone graft for symptomatic nonunion of the scaphoid. Clin Orthop 1976;115: 241–247. 17. Schnitzler CM, Biddulph SL, Mesquita JM, Gear KA. Bone structure and turnover in the distal radius and iliac crest. J Bone Miner Res 1996;11:1761–1768. 18. Biddulph SL. Bone donor site: iliac crest or distal radius. J Hand Surg 1999;24B:645– 646. 19. Boonen S, Cheng XG, Nijs J, et al. Factors associated with cortical and trabecular bone loss as quantified by peripheral computed tomography at the ultradistal radius in aging women. Calcif Tissue Int 1997;60:164 –170. 20. Han ZH, Palnitkar S, Rao DS, Nelson D, Parfitt AM. Effect of ethnicity and age or menopause on the structure and geometry of iliac bone. J Bone Miner Res 1996;11:1967– 1975. 21. Martin JC, Reid DM. Radial bone mineral density and
The Journal of Hand Surgery / Vol. 26A No. 1 January 2001 141 estimated rates of change in normal Scottish women: assessment by peripheral quantitative computed tomography. Calcif Tissue Int 1999;64:126 –132. 22. Mellish RWE, Garrahan NJ, Compston JE. Age-related changes in trabecular width and spacing in human iliac crest biopsies. Bone Miner 1989;6:331–338. 23. Meunier P, Courpron P, Edouard C, Bernard J, Bringuier J, Vignon G. Physiological senile involution and pathological
rarefaction of bone: quantitative and comparative histological data. Clin Endocr Metab 1973;2:239 –256. 24. Ouverton TR, Basu TK. Longitudinal changes in radial bone density in older men. Eur J Clin Nutr 1999;53:211–215. 25. Schnitzler CM, Pettifor JM, Mesquita JM, Bird MDT, Schnaid E, Smyth AE. Histomorphometry of iliac crest bone in 346 normal black and white South African adults. Bone Miner 1990;10:183–199.
Cancellous bone graft is often used to fill gaps and facilitate bone union in fracture treatment and arthrodesis. Numerous sites have been advocated for obtaining graft for use in the upper extremity. There is little consensus, however, regarding the best location for harvesting bone for a given need. Furthermore, information is lacking on the quantity and
quality of cancellous bone available from different sites. The purpose of this study was to determine the amount of bone graft available from the distal radius, olecranon process, and anterior iliac crest using standardized techniques. By measuring the harvested defect, bone density was also determined. This information will aid in determining the optimum location for bone harvest in upper extremity procedures.
From the Departments of Orthopaedic Surgery and Anatomy, RushPresbyterian-St Luke’s Medical Center, Chicago, IL. Received for publication December 3, 1999; accepted in revised form August 13, 2000. 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. Reprint requests: Mark S. Cohen, MD, 1725 W Harrison St, Suite 1063, Chicago, IL 60612. Copyright © 2001 by the American Society for Surgery of the Hand 0363-5023/01/26A01-0021$35.00/0 doi: 10.1053/jhsu.2001.20971
Materials and Methods Sixteen fixed cadaveric specimens were studied. Distal radius, olecranon, and iliac crest cancellous bone grafts were obtained bilaterally. No specimen used in the study had previous distal radius or olecranon fracture or previous iliac crest bone graft harvesting. Cadavers with a history of metabolic bone disease or carcinoma were excluded. There were 6 white men, 8 white women, 1 African-American The Journal of Hand Surgery 135
136 Bruno et al / Bone Graft Harvesting
man, and 1 African-American woman. The mean age at the time of death was 81 ⫾ 11 years (⫾SD; range, 59 –98 years). The technique of bone graft harvest was standardized to simulate the clinical setting.1,2 For the distal radius, bone graft was obtained by creating an opening in the dorsal cortex with a 1-cm dowel (Kaltec Precision Bone Graft System, Adelaide, South Australia). The distal extent of the defect was placed 1 cm proximal to the end of the radius in line with Lister’s tubercle (Fig. 1). For the olecranon, a similar 1-cm opening was made in the cortex beginning 1 cm distal to the tip of the olecranon (Fig. 2). Iliac crest cancellous bone graft was obtained through a 1-cm rectangular opening in the superior cortex of the crest. The anterior extent of the defect was placed 1 cm posterior to the anterior superior iliac spine (Fig. 3). For all regions studied, cancellous bone was harvested using straight and curved curettes (Codman 23-1046/56, Randolph, MA), taking care not to violate the cortical defect created. All grafts were obtained by the same 2 individuals working together. Care was taken to remove all available bone from the individual harvest sites without a time limit. The cortical entry fragment was stripped of all cancellous bone and not included in the graft measurements. Bone was collected in an airtight container for subsequent analysis. The volume of the defect created by the removal of cancellous bone was measured by injecting each site with a silicone polymer (General Electric, Waterford, NY). This material was chosen based on its ability to adequately fill the created defect and subsequently harden for analysis. The filler cured in a uniform fashion and was solid within 48 hours. It was then harvested by removing enough cortical bone from each site to facilitate extraction (Fig. 4). The volume of each donor site was initially determined by both a volume displacement method and by weighing the samples and calculating the volume based on the known density of the polymer. No significant differences were observed between these methods; thus, the volume calculation by weight was used due to its ease. Quantitative analysis of the harvested bone was performed by determining the “packed bone volume,” a measure of the “finger-packed” volume.3 To accurately and consistently reproduce the force used to make this determination, a servohydraulic material testing machine was used to compress the harvested graft material (Instron 1321, Canton, MA). A com-
Figure 1. (A) Distal radius harvest was performed through a dorsal cortical window placed 1 cm proximal to the distal extent of the radius in line with Lister’s tubercle. (B) The window was created with a 1-cm dowel. (C) Cancellous graft was obtained using straight and curved curettes. (D) The defect created was filled with a silicone polymer.
paction force of 40 N was chosen based on preliminary trials of finger packing. This force was then used for graft compression at a rate of 250 mm/min. Paired t-tests for the 2-volume measurements (packed bone volume and defect volume) indicated that there were no significant side-to-side differences. Therefore, the values for the 2 sides were averaged to obtain one set of volume measurements per site per cadaver. The ratio of the packed bone volume to the defect volume was also calculated,
The Journal of Hand Surgery / Vol. 26A No. 1 January 2001 137
t-tests, as appropriate, to determine which specific differences were significant. The analyses were performed with a statistical package (SPSS for Windows, release 8.0.0; SPSS, Inc, Chicago, IL).
Results The absolute amount of bone (ie, packed bone volume) was nearly 2-fold greater at the iliac crest than the distal radius or olecranon (Table 1). This difference was statistically significant (Table 2). Specifically, both the distal radius and the olecranon yielded less packed bone volume than the iliac crest. The variability of packed bone volume obtained was
Figure 2. (A) Olecranon harvest was performed though a cortical window 1 cm proximal to the tip. (B) A 1-cm dowel was used to create the cortical defect. (C) Curved and straight curettes were used to remove cancellous graft. (D) The defect was filled with a silicone polymer for determination of the defect volume.
providing a comparison of relative bone densities from the 3 harvest sites studied. The data were analyzed by repeated-measures ANOVA with anatomic site as the within-subjects factor, gender as the between-subjects factor, and age as a covariate. Mauchly’s test of sphericity was applied to the within-subjects effect, necessitating use of a multivariate test for site and the site-by-age and site-by-gender interactions (specifically, Pillai’s trace). Analyses showing a significant main effect were further analyzed with paired t-tests or Student’s
Figure 3. (A) A rectangular cortical window was made in the anterior iliac crest beginning 1 cm posterior to the anterior superior iliac spine. (B) The cortical window was 1 cm long. Pure cancellous graft was harvested with curved and straight curettes to simulate the clinical setting. (C) The defect created was filled with the silicone polymer between the intact tables.
138 Bruno et al / Bone Graft Harvesting
Figure 4. Cured silicone spacers harvested from the (left) iliac crest, (center) olecranon, and (right) distal radius donor sites. Note that the shape of each filler is analogous to the medullary cavity of the respective bones. These were used to calculate defect volume following harvest.
relatively constant, as the coefficient of variation (mean ⫾ SD ⫻ 100%) was 33%, 25%, and 32% for the distal radius, olecranon, and iliac crest, respectively. Gender was found to be a significant betweensubjects effect for the packed bone volume, with greater amounts harvested from the men than from
the women (Table 2). In general, approximately 50% more packed bone volume was obtained from the men than from the women (3.3 ⫾ 0.7 vs 2.2 ⫾ 0.7 cc for the distal radius, 3.4 ⫾ 0.7 vs 2.4 ⫾ 0.4 cc for the olecranon, and 6.2 ⫾ 1.0 vs 4.6 ⫾ 1.8 cc for the iliac crest in males and females, respectively.) The differences among the sites for the defect
Table 1. Volumes of Packed Bone, the Defect, and the Ratio of Packed Bone to the Defect Volume Packed bone (cc) Defect (cc) Packed bone/defect ratio
Distal Radius
Olecranon
Iliac Crest
2.7 ⫾ 0.9 (1.45–4.05) 5.0 ⫾ 1.2 (3.24–8.12) 0.5 ⫾ 0.1 (0.33–0.81)
2.8 ⫾ 0.7 (1.90–4.35) 4.3 ⫾ 0.5 (3.09–5.29) 0.7 ⫾ 0.2 (0.39–0.87)
5.3 ⫾ 1.7 (2.95–9.00) 10.8 ⫾ 5.3 (5.52–22.99) 0.5 ⫾ 0.2 (0.28–0.93)
Data are presented as mean values ⫾ SD (range).
The Journal of Hand Surgery / Vol. 26A No. 1 January 2001 139
Table 2. Significance Levels From the Analyses of Variance Variable
Site
Gender
Age
Site by Age
Site by Gender
Packed bone Defect Packed bone/defect ratio
.031* .689 .203
.004† .060 .162
.458 .763 .575
.135 .968 .341
.459 .165 .230
P values less than .05 were considered to be significant and are shown in bold. *Radius ⬍ iliac crest (p ⬍ .001), olecranon ⬍ iliac crest (p ⬍ .001). †Female ⬍ male for radius (p ⫽ .006), olecranon (p ⫽ .004), and iliac crest (p ⫽ .052).
volume and packed bone/defect volume ratio were not significantly significant (p ⬎ .05). In addition, except as noted for the packed bone volume, gender was not a significant factor. Finally, age was not a significant covariate in any of the analyses, nor were there significant statistical interactions between anatomic site and age or between anatomic site and gender.
Discussion Autologous cancellous bone graft is used in upper extremity surgery for acute fracture care, nonunions, arthrodesis, and filling of osseous tumor defects. The most common site for bone graft harvest is the anterior iliac crest. This provides an abundant amount of cancellous bone but does require preparation of a distant anatomic site with associated donor site morbidity4,5 Furthermore, general anesthesia is typically required for iliac crest harvest and ambulation can be problematic when surgery is performed on an outpatient basis. In an effort to decrease the morbidity of iliac crest harvest, percutaneous techniques with biopsy trochars have been introduced.6 The iliac crest is often covered by a thick soft tissue envelope, however, and percutaneous techniques also can be complicated by bleeding and hematoma formation.7 For upper extremity procedures, various investigators advocate obtaining bone graft from the olecranon or distal radius.8 –14 This allows the surgical procedure to be limited to the involved extremity without additional anesthetic requirements. The primary disadvantage of local upper extremity graft is the limited amount of bone available. There also are isolated reports of pathologic fracture following olecranon and distal radius harvest, although this is quite rare.8,15 No quantitative data have previously existed regarding the actual amount of bone that can be obtained from the different sites. Several studies have attempted to evaluate healing using upper extremity sites in lieu of the iliac crest.
Hull et al16 compared iliac crest graft with the distal radius for scaphoid nonunion, reporting a 73% union rate for iliac crest graft versus a 47% union rate for the distal radius graft. These investigators surmised that the increased union rate seen with the iliac crest was a function of the osteogenic potential of the bone. All the iliac grafts were corticocancellous, however, and were placed palmarly. Eleven of the 17 radius grafts involved only a plug of bone inserted through a drill hole in the scaphoid and only 1 of these achieved union. Schnitzler et al17 compared bone structure between the distal radius and the iliac crest in vitro and concluded that the distal radius was structurally inferior and had a lower bone turnover compared with the iliac crest. These investigators pointed out, however, that radius autograft may very well be suitable for “compacting into a small bone defect.” Furthermore, the biologic differences were felt to be of minimal clinical significance when the bone was used as graft material.18 For hand and wrist procedures, numerous reports suggest that the distal radius is a useful option.11–14 Mirly et al8 reported an 82% healing rate when anterior distal radius graft was used for cysts, arthrodesis, and scaphoid nonunions. One complication, a fracture through the donor site at approximately 4 years after surgery, occurred in 131 cases. The fracture was nondisplaced and healed uneventfully. McGrath and Watson9 reported on local bone graft for 124 procedures over a 10-year period. Bone was obtained from the distal radius in 78 cases, the proximal ulna in 19 cases, and the small hand bones in 27 cases. At a minimum follow-up period of 22 months, they had 1 nonunion and 1 patient with chronic donor site pain. These investigators concluded that upper extremity regional bone grafts provide a reliable and adequate amount of bone with minimal complications for upper extremity procedures. They report that “several” cubic centimeters of cancellous bone
140 Bruno et al / Bone Graft Harvesting
can be obtained from the distal radius and that up to 4.5 cubic centimeters of bone is available from the proximal ulna. The harvested bone volumes, however, were not further quantified. We evaluated bone graft harvested from the distal radius and olecranon relative to that from the anterior iliac crest. We found the average volume of packed cancellous bone obtainable from the distal radius to be similar to the olecranon and approximately half of that available from the anterior iliac crest (Table 1). We also found approximately twice as much bone available in males than females in the population studied. It must be noted, however, that a considerable range exists for each site, most likely based on the size and gender of the individual. Furthermore, the actual volume of bone available in vivo is most likely somewhere in between our measured packed and defect volumes, especially in younger individuals in the absence of osteopenia. Perhaps the major limitation of this study is the age of the cadaveric specimens. One can only extrapolate our findings to young adults if the relative amounts of cancellous bone found at the distal radius, olecranon, and iliac crest are reasonably constant from skeletal maturity through aging. Although there are no data on cancellous bone volume for the olecranon, the existing data for the iliac crest and distal radius indicate that the age-related rate of cancellous bone loss is roughly equivalent at these 2 sites.19 –25 Thus, we believe it is safe to assume that in the patient population most often in need of upper extremity bone grafting, the iliac crest would supply approximately twice the volume of cancellous bone as the distal radius or proximal ulna. Our findings indicate that the olecranon is an alternative to the distal radius as a source of bone graft in upper extremity procedures and that it is a viable option when the distal radius is not available, eg, for distal radius fracture. When a large volume of cancellous bone is required, the anterior iliac crest provides approximately twice the amount of cancellous autograft and would remain the preferred site. This information should be useful when planning cancellous bone harvest for a particular need in the upper extremity.
References 1. Chapman MW. Bone grafting. In: Chapman MW, ed. Operative orthopaedics. 2nd ed. Philadelphia: JB Lippincott, 1993:139 –149.
2. Crenshaw AH. Surgical techniques and approaches. In: Canale ST, ed. Campbell’s operative orthopaedics. 9th ed. St Louis: Mosby, 1998:40 – 47. 3. Lindberg EJ, Katchis SD, Smith RW. Quantitative analysis of cancellous bone graft available from the greater trochanter. Foot Ankle Int 1996;17:473– 476. 4. Beirne JC, Barry HJ, Brady FA, Morris VB. Donor site morbidity of the anterior iliac crest following cancellous bone harvest. Int J Oral Maxillofacial Surg 1996;25:268 – 271. 5. Younger EM, Chapman MW. Morbidity at bone graft donor sites. J Orthop Trauma 1989;3:192–195. 6. Gartsman GM, Lane JM. A simple method of obtaining bone graft by bone biopsy trochar. J Hand Surg 1981;6: 627– 628. 7. Billmire DA, Rotatori S. Use of the CORB needle biopsy for the harvesting of iliac crest bone graft. Plast Reconstr Surg 1994;93:416 – 418. 8. Mirly HL, Manske PR, Szerinski JM. Distal anterior radius bone graft in surgery of the hand and wrist. J Hand Surg 1995;20A:623– 627. 9. McGrath MH, Watson HK. Late results with local bone graft donor sites in hand surgery. J Hand Surg 1981;6:234 – 237. 10. Voche P, Merle M. Removal of cancellous bone graft through the base of the excised radial styloid process. J Hand Surg 1992;17B:499 –500. 11. Eckenrode JF, Louis DS, Greene TL. Scaphoid-trapeziumtrapezoid fusion in the treatment of chronic scapholunate instability. J Hand Surg 1986;11A:497–502. 12. Frykman EB, Ekenstam FA, Wadin K. Triscaphoid arthrodesis and its complications. J Hand Surg 1988;13A:844 – 849. 13. Kleinman WB, Steichen JB, Strickland JW. Management of chronic rotary subluxation of the scaphoid by scaphotrapezio-trapezoid arthrodesis. J Hand Surg 1982;7:125– 136. 14. Watson HK, Hempton RF. Limited wrist arthrodeses. I. The triscaphoid joint. J Hand Surg 1980;5:320 –327. 15. Walker LG, Meals RA. Pathologic fractures of the proximal ulna through a bone graft donor site. J Hand Surg 1990;15A:781–784. 16. Hull WJ, House JH, Gustillo RJ, Kleven L, Thompson W. The surgical approach and source of bone graft for symptomatic nonunion of the scaphoid. Clin Orthop 1976;115: 241–247. 17. Schnitzler CM, Biddulph SL, Mesquita JM, Gear KA. Bone structure and turnover in the distal radius and iliac crest. J Bone Miner Res 1996;11:1761–1768. 18. Biddulph SL. Bone donor site: iliac crest or distal radius. J Hand Surg 1999;24B:645– 646. 19. Boonen S, Cheng XG, Nijs J, et al. Factors associated with cortical and trabecular bone loss as quantified by peripheral computed tomography at the ultradistal radius in aging women. Calcif Tissue Int 1997;60:164 –170. 20. Han ZH, Palnitkar S, Rao DS, Nelson D, Parfitt AM. Effect of ethnicity and age or menopause on the structure and geometry of iliac bone. J Bone Miner Res 1996;11:1967– 1975. 21. Martin JC, Reid DM. Radial bone mineral density and
The Journal of Hand Surgery / Vol. 26A No. 1 January 2001 141 estimated rates of change in normal Scottish women: assessment by peripheral quantitative computed tomography. Calcif Tissue Int 1999;64:126 –132. 22. Mellish RWE, Garrahan NJ, Compston JE. Age-related changes in trabecular width and spacing in human iliac crest biopsies. Bone Miner 1989;6:331–338. 23. Meunier P, Courpron P, Edouard C, Bernard J, Bringuier J, Vignon G. Physiological senile involution and pathological
rarefaction of bone: quantitative and comparative histological data. Clin Endocr Metab 1973;2:239 –256. 24. Ouverton TR, Basu TK. Longitudinal changes in radial bone density in older men. Eur J Clin Nutr 1999;53:211–215. 25. Schnitzler CM, Pettifor JM, Mesquita JM, Bird MDT, Schnaid E, Smyth AE. Histomorphometry of iliac crest bone in 346 normal black and white South African adults. Bone Miner 1990;10:183–199.