Vascularized versus Nonvascularized Bone Grafts: What Is the Evidence?

There is a general perception in practice that a vascular supply should be used when large pieces of bone graft are used, particularly those greater than 6 cm in length for long-bone and large-joint reconstructions. However, the scientific source of this recommendation is not clear.

We wished to perform a systematic review to (1) investigate the origin of evidence for this 6-cm rule, and (2) to identify whether there is strong evidence to support the importance of vascularization for longer grafts and/or the lack of vascularization for shorter grafts.

Two systematic reviews were performed using SCOPUS and Medline, one for each research question. For the first research purpose, a review of studies from 1975 to 1983 matching article title ("bone" and "graft") revealed 725 articles, none of which compared graft length. To address the second purpose, a review of articles before 2014 that matched "bone graft" AND ("vascularised" OR "vascularized") AND ("non-vascularised" OR "non-vascularized") revealed 633 articles, four met prespecified inclusion criteria and were evaluated qualitatively. MINORS ratings ranged from 16 to 18 of 24, and National Health and Medical Research Council [NHMRC] Evidence Hierarchy ratings ranged from III-2 (comparative studies without concurrent controls) to III-3 (comparative studies with concurrent controls).

No evidence was found that clarified grafts longer than 6 cm should be vascularized. The first reference to the 6-cm rule cites articles that do not provide strong evidence for the rule. Of the four articles found in the second systematic review, none examined osseous union of vascularized and nonvascularized grafts with respect to length. One study (III-3, MINORS 18 of 24) of fibular grafts to various limb defects found that vascularization made no difference to union rate or time to union. Vascularized grafts were more likely to require surgical revision for wound breakdown, nonunion, graft fracture, or mechanical problems (hazard ratio [HR], 5.97, p = 0.008) and grafts smaller than 10 cm had fewer complications requiring revision (HR, 0.88; p = 0.03). Three studies (III-2 to III-3, MINORS 16 to 18 of 24) that examined fibular grafts to the femoral head found that vascularized grafts had superior Harris hip and pain scores. Two of the three articles showed that vascularization was associated with superior radiologic measures of collapse progression.

No compelling evidence was found to illuminate the origin of the 6-cm rule for vascularized bone grafts, or that such a rule is based on published research. The evidence we found for grafts to long-bone defects suggested that vascularization might increase the risk of complications that require a surgical revision without increasing union rates or time to union. For large joints, vascularization may result in better functional scores and pain scores, while the evidence that they improve radiologic measures of progression is mixed. There were no studies of long-bone or large-joint reconstructions that examined the role of length with respect to osseous union. We suggest that future studies should present data for graft lengths quantitatively and with individual data points rather than categories of length ranges.

Level III, therapeutic study.