Skip to content
Home  /  FRACS Orthopaedic Surgery  /  Study notes  /  Scaphoid Non-union - Bone Graft Selection: Autograft, Allograft, Synthetic Substitutes, and BMP-2

Scaphoid Non-union - Bone Graft Selection: Autograft, Allograft, Synthetic Substitutes, and BMP-2

FRACS Orthopaedic Surgery LO FRACSORTHO_RECON_1 2,568 words
Free preview. This study note covers learning objective FRACSORTHO_RECON_1 from the FRACS Orthopaedic Surgery curriculum. Inside PRIMEX you get AI-graded SAQ practice on this topic, voice viva with the AI examiner, MCQs across the full syllabus, and a curriculum tracker that ticks off every learning objective.

Overview

Scaphoid non-union management requires graft selection matched to the biological environment at the non-union site: the degree of avascular necrosis (AVN), the presence of humpback deformity, defect size, and proximal pole vascularity. This note covers the comparative properties of autograft, allograft, and synthetic bone substitutes, and the indications and risks of rhBMP-2. The Russe classification, Herbert screw mechanics, and Matti-Russe operative anatomy are assumed as background knowledge.


Scaphoid Blood Supply, Implications for Graft Selection

Approximately 70-80% of the scaphoid's blood supply enters via the dorsal ridge through a branch of the radial artery at the waist, supplying the proximal 70-80% of the bone in a retrograde fashion. The distal pole receives an independent volar supply via the scaphoid tubercle. Consequences:


Bone Graft Mechanisms

All grafts act through one or more of three mechanisms:

Comparative Properties by Graft Type

Property Cancellous Autograft Corticocancellous Autograft Cortical Autograft Cancellous Allograft Cortical Allograft DBM Calcium Phosphate Ceramics BMP-2/ACS
Osteogenesis ++ + +
Osteoinduction + ± ± + ± +++ +++
Osteoconduction ++ ++ + ++ ± ++ +++ + (carrier-dependent)
Mechanical strength + ++ ++ + ++ ++
Vascularity
Donor site morbidity Yes Yes Yes No No No No No
Disease transmission No No No Low Low Low No No

Vascularised autograft adds vascularity (++) and maintains osteogenesis (++) and mechanical strength (++), see dedicated section below.

(DBM = demineralised bone matrix; ACS = absorbable collagen sponge)


Autograft

Harvest Sites

Site Graft Type Volume Key Advantage Key Morbidity
Iliac crest (anterior) Cancellous ± corticocancellous Large Highest MSC concentration; optimal osteogenic source Chronic donor pain ~15-30%; lateral femoral cutaneous nerve injury; haematoma
Distal radius (volar) Cancellous Moderate Same operative field; no additional incision Limited volume; distal radius fracture if over-harvested
Olecranon Corticocancellous Small-moderate Low morbidity; convenient Limited volume
Proximal tibia Cancellous Moderate Low morbidity Separate tourniquet; rare fracture

The iliac crest is the optimal source for MSCs and embedded endogenous BMPs. It remains the reference standard for scaphoid non-union grafting.

Cancellous vs. Corticocancellous Autograft

Systematic review data (all non-vascularised grafts) demonstrate:

Parameter Cancellous Only Corticocancellous
Time to union ~11 weeks ~16 weeks
Carpal alignment correction Inferior Superior
Best indication Stable waist non-union; proximal pole non-union without major deformity Humpback deformity requiring structural correction (Fisk-Fernandez wedge)
Structural support Requires screw fixation Provides some intrinsic column support

In the Matti-Russe inlay technique, a volar cortical strut graft is used either alone or in combination with K-wires. In the Fisk-Fernandez technique, a volar distal radius or iliac crest corticocancellous wedge graft is customised to correct scaphoid malalignment and carpal alignment simultaneously.

Even in the setting of proximal pole AVN, corticocancellous iliac crest autograft with rigid screw fixation has achieved union rates approaching 100% in small series, challenging the older view that non-vascularised grafts invariably fail with AVN. The critical prerequisite is adequate proximal fixation providing both stability and compression.

Distal Radius Autograft

Practical for same-field harvest; volume is limited and it is best suited to stable waist non-unions without major structural defect or AVN. Used in both the Fisk-Fernandez wedge and as a cancellous supplement in the Russe-type approach.


Vascularised Bone Grafts (Autograft Subtype)

Vascularised grafts bring an independent arterial supply to the non-union site. They function as both a structural and vascular augment in the setting of mechanical malalignment and as a regenerative augment in proximal pole non-unions with compromised healing capacity. Their use is controversial; most indications centre on waist non-unions with suspected AVN or proximal pole non-unions with established AVN.

Pedicled Options

Graft Pedicle Approach Indication
1,2 Intercompartmental supraretinacular artery (1,2 ICSRA), Zaidemberg Branch of radial artery Dorsal Proximal pole AVN; waist non-union with AVN
Volar distal radius Volar carpal artery Volar Waist non-union with AVN
Pronator quadratus, Braun/Kawai-Yamamoto Anterior interosseous artery Volar Proximal pole; waist with AVN
4th/5th extensor compartment artery Branch of anterior interosseous Dorsal Waist/proximal pole AVN
Distal ulna Ulnar artery branch Ulnar/dorsal Alternative when radial options unavailable

Free Vascularised Graft: Medial Femoral Condyle

The medial femoral condyle (MFC) free flap, based on the superomedial genicular artery, provides corticocancellous bone with reliable periosteal vascularity. Recent series suggest potentially superior union rates compared to pedicled distal radial grafts for AVN cases. Indications:

Limitations: microsurgical expertise required; second operative field; longer operative time.

Outcomes: Vascularised vs. Non-Vascularised Grafts

Meta-analytic data (Merrell, Wolfe, and Slade, 36 eligible reports from 1121 articles):

Clinical Scenario Non-Vascularised Graft + Screw Vascularised Graft K-wire + Wedge Graft
Non-union without AVN ~94% union Not routinely required ~74%
Non-union with AVN ~47% union ~88% union ,

Union rates for pedicled distal radial grafts range from 27% to 100% across published series, with poor results at the lower end concentrated in AVN cases. The MFC free flap has shown superior rates in AVN compared to pedicled radial options in more recent series.


Allograft

Types

Type Properties Role in Scaphoid Non-union
Cancellous allograft Osteoconductive; minimal osteoinduction if processed; no osteogenesis Rare primary use; possible extender
Cortical allograft Structural; slow remodelling; poor osteoinduction Structural scaffold adjunct only
Demineralised bone matrix (DBM) Osteoinductive (BMP exposure) + osteoconductive; available as gel/putty/strips Adjunct; not structural; not a primary graft substitute
Whole scaphoid allograft Structural replacement; high non-union risk Salvage only for fragmented proximal pole where reconstruction not feasible

Why Allograft is Rarely the Primary Choice

  1. The small bone volumes required are readily met by autograft harvest.
  2. Allograft lacks viable osteoprogenitor cells, most critical when the biological environment is compromised.
  3. Processing (freeze-drying, irradiation) degrades growth factors, reducing osteoinductivity.
  4. No published level I evidence supports allograft equivalence to autograft in scaphoid non-union.

DBM may supplement autograft as an osteoinductive adjunct or serve as a BMP-2 carrier but should not substitute for autograft where regenerative capacity is the primary requirement.

Regulatory and Consent Considerations

In Australia, allograft is regulated by the TGA and must be sourced from accredited tissue banks. Residual risks include viral transmission (HIV estimated <1 per million with current screening) and theoretical prion transmission. These must be disclosed in consent. Use in elective non-union surgery where autograft is readily available requires explicit justification.


Synthetic Bone Substitutes

Classification

Class Examples Mechanism Role in Scaphoid Non-union
Calcium sulphate , Osteoconduction; rapid resorption (weeks) Void filler only; no structural role
Calcium phosphate (HA, β-TCP, biphasic) Chronos (β-TCP), Vitoss Osteoconduction; variable resorption Adjunct void filler; not primary graft
Bioactive glass , Osteoconduction; bonds to host bone Experimental; no scaphoid-specific evidence
Demineralised bone matrix Grafton, Osteoset Osteoinduction + osteoconduction Adjunct only; not structural
Bone marrow aspirate (BMA) , Osteogenesis (MSC-dependent) Adjunct to scaffold; not standalone
Composite grafts DBM + calcium phosphate Combined osteoconduction + osteoinduction Potential adjunct when autograft limited
BMP-2 on carrier INFUSE (rhBMP-2/ACS) Osteoinduction +++ Off-label; see section below

Synthetic substitutes lack osteogenesis and do not replicate iliac crest autograft. Their role is as extenders when autograft volume is insufficient, as carriers for biological agents, or as void fillers with adequate fixation.


BMP-2: Indications and Risks

Biology

Recombinant human BMP-2 (rhBMP-2) is a member of the TGF-β superfamily and the most potent osteoinductive agent in clinical use. It binds BMP receptors on MSCs, activating the SMAD 1/5/8 signalling cascade and driving osteoblastic differentiation:

$$\text{MSC} \xrightarrow{\text{BMP-2} \to \text{SMAD 1/5/8}} \text{Osteoprogenitor} \to \text{Osteoblast} \to \text{New bone}$$

The commercial product (INFUSE) delivers rhBMP-2 at 1.5 mg/mL on an absorbable collagen sponge (ACS) carrier.

Regulatory Status and Approved Indications

rhBMP-2 (INFUSE) TGA-listed indications in Australia:

Use in scaphoid non-union is off-label. Informed consent must specifically address the off-label nature; RACS/AOA guidance on off-label use of therapeutic goods applies.

Theoretical Indications in Scaphoid Non-union

Off-label consideration may be justified when:

Optimal use: combine with rigid headless compression screw fixation and supplementary cancellous autograft (osteogenic cells alongside osteoinductive stimulus); do not use as sole graft substitute.

Risks and Adverse Effects

Risk Category Specific Adverse Effect Mechanism / Comment
Inflammatory Early osteolysis; soft tissue swelling Supraphysiological dosing triggers local inflammatory cascade; especially problematic in constrained spaces (wrist)
Ectopic ossification Heterotopic bone formation in radiocarpal/midcarpal joint Off-target BMP signalling; can cause permanent stiffness
Wound complications Seroma; wound dehiscence ACS acts as fluid reservoir; excessive swelling in tight dorsal/volar wrist incisions
Tendon irritation Flexor (volar approach) or extensor (dorsal approach) adhesions Proximity of tendons to graft site in the wrist
Carpal tunnel Elevated carpal tunnel pressure Inflammatory oedema from ACS/BMP reaction
Oncological Theoretical cancer risk Post-market surveillance (AMPLIFY trial) raised concern; subsequent analyses suggest confounding; current consensus is no proven carcinogenic risk at approved doses
Immune / allergic Anti-BMP-2 or anti-bovine collagen antibodies May limit efficacy on re-exposure; rare anaphylaxis
Resorption paradox Initial net bone resorption before formation Dose-dependent; small-bone applications risk net bone loss in early phase
Bone quality Woven, poorly mineralised bone Rapidly formed bone may be architecturally inferior to lamellar autograft-derived bone
Cost Significant cost premium Not PBS-subsidised for this indication; economic justification requires careful case selection

BMP-2 in the Constrained Wrist Environment

The wrist capsule is immediately adjacent to the graft site on both volar and dorsal approaches. Inflammatory swelling from the BMP-2/ACS reaction can cause flexor or extensor tendon irritation and adhesions, carpal tunnel pressure elevation, and ectopic bone in the radiocarpal or midcarpal joint. Cautious, low-dose supplementary use is strongly preferred over BMP-2 as a sole graft substitute.


Decision Framework: Graft Selection

Clinical Scenario Recommended Graft Strategy
Stable waist non-union, no AVN, no deformity Cancellous iliac crest or distal radius autograft + headless compression screw
Waist non-union + humpback deformity, no AVN Corticocancellous iliac crest autograft (Fisk-Fernandez wedge) + headless compression screw
Proximal pole non-union, viable proximal fragment Corticocancellous iliac crest autograft + antegrade headless screw (dorsal approach)
Proximal pole non-union with AVN Vascularised bone graft (pedicled 1,2 ICSRA or MFC free flap) ± cancellous autograft augment
Recalcitrant non-union, failed ≥2 procedures MFC free flap preferred; BMP-2 off-label adjunct with autograft as last resort
Unable to tolerate autograft harvest DBM + BMA as adjunct to synthetic scaffold; counsel regarding inferior biology

Outcomes Summary

Fixation / Graft Method Union Rate
Non-vascularised graft + headless compression screw ~94%
Vascularised graft in AVN ~88%
Non-vascularised graft in AVN ~47%
K-wire + wedge graft ~74%
Corticocancellous iliac crest + screw (proximal pole, including AVN cases) Up to ~100% (small series)

Time to union: cancellous autograft ~11 weeks; corticocancellous autograft ~16 weeks.

CT confirmation is the gold standard for assessing union; plain radiographic assessment is unreliable.


Complications of Bone Grafting in Scaphoid Non-union

Complication Notes
Persistent non-union ~6-20% depending on AVN status and fixation adequacy
Donor site pain (iliac crest) 15-30% at 1 year; chronic pain ~10%
Lateral femoral cutaneous nerve injury ~5-10% with anterior iliac crest harvest
Ectopic bone (BMP-2) Variable; higher with excessive dosing or in constrained spaces
Hardware prominence / screw back-out Secondary removal required in ~5%
SNAC wrist progression Inevitable if symptomatic non-union untreated; staged salvage required
Carpal tunnel syndrome Volar approach with significant swelling

Outcome Measures

Measure Application
Patient-Rated Wrist Evaluation (PRWE) Most responsive PROM for wrist-specific pain and function
DASH / QuickDASH Upper limb disability; widely validated
Grip strength (% of contralateral) >85% considered good outcome
Wrist range of motion (% of contralateral) Flexion-extension arc; radial-ulnar deviation
CT-confirmed union Gold standard imaging endpoint
Time to union Surrogate endpoint; cancellous ~11 weeks vs. corticocancellous ~16 weeks

Paediatric Considerations

Scaphoid non-union is uncommon in the skeletally immature. When grafting is required:


Sources

PRIMEX

Practice this topic in the app

Attempt a graded SAQ on this exact LO, run a voice viva with the AI examiner, or work through MCQs that map to FRACSORTHO_RECON_1. Your free trial covers all 21 exams.

Start 7-day free trial

7-day free trial · Cancel anytime

Quick recall flashcards

A small sample of the deck for this topic. Tap a question to reveal the answer. The full deck and spaced-repetition scheduler live inside PRIMEX.

Describe the Russe classification of scaphoid non-union and its three types.
  • Type I (Stable fibrous non-union): no displacement, minimal resorption, adequate bone stock, good prognosis with grafting
  • Type II (Unstable non-union with displacement): variable resorption, progressive carpal malalignment possible, requires structural correction
  • Type III (Osteonecrosis of proximal pole): avascular change in proximal fragment, worst prognosis, requires vascularised graft consideration
What is the most commonly used internal fixation device for scaphoid non-union after bone grafting?
  • The Herbert differential-pitch headless compression screw
  • Its variable thread pitch generates compression across the non-union site without protruding hardware
  • Cannulated versions (Herbert-Whipple) allow fluoroscopically guided placement over a guide wire
What is the Matti-Russe technique for scaphoid non-union?
  • Volar approach between flexor carpi radialis and radial artery
  • Excavation of a cavity (inlay) bridging both proximal and distal fragments
  • Packing with cancellous bone graft to stimulate union
  • No intrinsic fixation in the original description; Kirschner wires or screws may be added
What union rate is achieved with bone grafting and screw fixation for scaphoid non-union without osteonecrosis?
  • Approximately 94% union rate
  • Compared with approximately 74% using Kirschner wire fixation with wedge grafting
  • Screw fixation provides superior compression and stability
Start free trial