Scaphoid Fractures Requiring Bone Graft: Does Graft Source Matter?

© 2024 HMP Global. All Rights Reserved.

Any views and opinions expressed are those of the author(s) and/or participants and do not necessarily reflect the views, policy, or position of ePlasty or HMP Global, their employees, and affiliates.

The scaphoid is the most commonly fractured carpal bone, accounting for 60% of all carpal fractures.¹ Of these, 66% to 88% occur in males and are most common in the second and third decades of life.^2-5 The estimated annual incidence in the United States is approximately 1.47 fractures per 100,000 person years.²

Bone graft is most often used in the repair of scaphoid nonunions. The reported incidence of nonunion is wide-ranging, with rates of 14% to 92% in displaced fractures and 0% to 23% in nondisplaced fractures.^6-7 When left untreated, patients can develop pain, stiffness, and weakness, with rates of radio-carpal osteoarthritis cited as high as 100%.^8-9 Bone grafting may also be used for the correction of deformity or for the restoration of normal carpal anatomy.¹⁰ Failure to correct malalignment can similarly result in pain, stiffness, and arthritis, even in cases of union.¹¹

Bone graft can be vascularized or nonvascularized, with union rates in the literature of 92% and 88%, respectively.¹² Autograft is traditionally harvested from the iliac crest. According to Herbert, this donor is superior due to osteogenic and mechanical properties.¹³ However, there is well-established morbidity associated with these grafts.¹⁴ As such, distal radius grafts have become more popular under the assumption that the union rate should be similar with lower donor site morbidity and risks.

The purpose of this retrospective study is to compare the use of iliac crest grafts against distal radius grafts when used in the fixation of scaphoid waist fractures. We hypothesize that rates of union and complication rates will not vary by graft source.

Institutional Review Board approval was granted for this study. Patients undergoing repair of a scaphoid fracture with bone graft at our institution between 2010 and 2020 were identified via Current Procedural Terminology code. Bone graft was used in patients with nonunion, humpback deformity, or for correction of scaphoid alignment, with bleeding confirmed at the bony surfaces prior to bone graft insertion. Patients were included in the study if they had fracture of the middle third of the scaphoid, fixation with either a headless compression screw (HCS) or nitinol compression staple, and fixation using bone graft from either the distal radius or iliac crest. Patients were excluded if they had prior surgical intervention to the affected scaphoid, pathologic fracture, less than 4 weeks between injury and surgery, less than 12 weeks of follow-up, or if autograft was supplemented with allograft. A full retrospective chart review was completed for all qualifying patients. A total of 39 patients were included in the study.

The primary outcome measure was union rate as determined by postoperative X-ray or computed tomography (CT) scan. Secondary outcomes included complication rate, rate of unplanned secondary surgery, time to union, immobilization time, and postoperative scapholunate angle. Complications included new or persistent avascular necrosis (AVN), malalignment, and issues related to hardware. Postoperatively, all patients were casted for 2 months, followed by wrist splinting until healing was observed. Range of motion and activity level were determined by clinical examination and radiographic findings.

Outcomes were stratified by bone graft source. Fisher exact tests were used to compare categorical variables. For comparison of continuous variables, Student t tests were used for parametric variables, and Mann-Whitney U tests were used for nonparametric variables. Missing data were excluded from statistical analyses. Statistical significance was set at P < .05.

Of the 39 patients included in the study, 32 (82%) were male. The median age was 22 years. Twenty-two (56%) injuries occurred on the right side. All patients presented with pain as a primary complaint. The most common mechanism of injury was a fall (67%). Fifteen (40%) presented with avascular necrosis prior to surgery. Twenty-one (54%) patients had fixation with HCS, and 18 (46%) had fixation with a nitinol compression staple. The median follow-up time was 5.6 months (Table 1).

Twenty-nine (74%) fractures were repaired with distal radius graft, and 10 (26%) were repaired with iliac crest graft. Both distal radius and iliac crest groups were primarily male (79% vs 90%, P = .65), and the median age was similar (19 vs 24 years, P = .57). A majority of patients in the distal radius group were fixed with HCS (62%), while the majority in the iliac crest group were fixed with staple (70%); this difference was not significant (P = .14). The most common injury mechanism was a fall in both groups (76% and 40%, respectively). The mean preoperative scapholunate angle was 57.8 degrees in the distal radius group and 66.2 degrees in the iliac crest group (P = .12).

A total of three (8%) repaired fractures failed to achieve radiographic union. One of the 29 (3%) distal radius grafts failed to achieve union, and 2 of the 10 (20%) iliac crest grafts failed to achieve union. There was no statistically significant difference in union rate between the 2 groups (P = .16) (Table 2).

All 3 nonunions required revision surgery. Two additional patients required secondary surgeries: 1 treated with a distal radius autograft underwent an extensor tenolysis, and 1 treated with iliac crest graft underwent surgery for removal of a buried pin (Table 2).

Six (15%) patients had postoperative complications. One patient with an iliac crest graft developed both AVN and malalignment postoperatively, which was treated with revision surgery. Three additional patients, all of whom received a distal radius graft, developed postoperative AVN. All 3 patients had poor follow-up, and no specific intervention was identified during chart review. Two patients developed issues related to hardware: 1 in the distal radius group and 1 in the iliac crest group (Table 2). The iliac crest patient required hardware removal and revision fixation. The distal radius patient had hardware removal.

The median time to union was 2.8 months in the iliac crest group and 3.2 months in the distal radius group; this difference was not significant (P = .62). The postoperative scapholunate angle was larger in the iliac crest group compared with the distal radius group, but the difference was not significant (mean 61.0 vs 52.5 degrees, median 64.0 vs 53.6 degrees, P = .05; Table 2). Note that for the scapholunate angle, mean values were provided as a reference to preoperative values, but median values were compared statistically due to the variable’s nonparametric nature.

Bone graft is used for various reasons in the repair of scaphoid fractures. The goal of our study was to assess whether there was any difference in outcome with graft harvested from the distal radius compared with the iliac crest in scaphoid waist fractures. In agreement with our hypothesis, our data show similar rates of union and similar secondary outcomes between both graft sources.

Most of the existing literature investigating the difference between distal radius and iliac crest autografts has been exclusive to nonunions. Overall rates of postoperative union in these studies have ranged from 67% to 91% for distal radius grafts and 66% to 100% for iliac crest grafts.^15-18 None of these studies showed a significant difference in union rate based on graft source. Our data similarly showed no difference by graft source. While our 80% union rate for iliac crest grafts is consistent with these studies, our 97% union rate for distal radius grafts is higher than the aforementioned series.

This study adds to the current body of literature in several ways. To our knowledge, this is the first series to include the use of bone graft in the nonunion setting utilizing staples and HCS in the treatment of scaphoid waist fractures. One study by Tambe et al included fixation with HCS and K-wires, whereas our study was done with HCS or staples.¹⁵ Braga-Silva et al compared vascularized distal radius graft against nonvascularized iliac crest grafts; however, the difference in vascularization of the graft adds a confounding factor to the comparison.¹⁶ Studies by Goyal et al and Christodoulou et al included proximal pole fractures, which are less likely to unite, whereas our study was limited to waist fractures and therefore had fewer confounding variables.^17-18 One large systematic review with 1602 patients having fixation of a scaphoid nonunion cited union rates of 89% for distal radius grafts and 87% for iliac crest grafts.¹² However, this study was a collection of 48 publications with a fair amount of heterogeneity, making a true comparison between the 2 grafts challenging.

In our review, there was insufficient data regarding donor site complications to make any meaningful conclusions. However, common complications from both sites have been documented in the literature. Complications from distal radius graft harvest include soft tissue contracture, radiocarpal ligament disruption, adhesions, De Quervain tenosynovitis, and graft site fractures.¹⁹ One study showed an overall complication rate of 4% over 4.5 years; the most common complications were the need for a repeat autograft with iliac crest (2.3%) and the development of De Quervain tenosynovitis (1.3%).²⁰ In contrast, complications from iliac crest autograft are more common and more severe. Major complications include numbness, hernia, pelvic fracture, infection, significant hematoma, and seroma. Minor complications include wound drainage, temporary sensory loss, superficial infection, delayed healing, minor hematoma, and minor wound problems. The rates of these complications have been cited as high as 29%.¹⁴ Furthermore, iliac crest bone grafting typically requires general anesthesia, whereas distal radius graft harvest can be done using regional anesthesia without any additional anesthesia.

There are several limitations to this study. While we found a difference in union rates between distal radius and iliac crest bone grafting, there was no statistical difference. This is likely attributable to the small sample size leading our study to be underpowered. Second, the study was retrospective in nature and lacked randomization. Third, there was insufficient information regarding donor-site complications; as such, we had to rely on historical data from the literature, which were then applied to our study. Lastly, patient charts had inconsistent/incomplete information on functional outcomes such as range of motion, grip strength, Disabilities of Arm, Shoulder and Hand (DASH) score, visual analog scale (VAS) pain score, and time to return to work. Without this data, we were unable to draw conclusions about functional outcomes between the 2 graft sources.

Based on our results, surgeons should consider using distal radius grafts over iliac crest grafts. In agreement with existing literature, we have shown no difference with respect to radiographic outcomes. Given the higher morbidity associated with the iliac crest donor site, it is reasonable to instead consider the use of distal radius autograft. To strengthen this claim, further research is needed in the form of a prospective randomized-controlled trial with a larger sample size and special attention paid to donor site complications.