The use of non-vascularised osteochondral autograft for reconstruction of articular surfaces in the hand and wrist

Abstract

Bone loss involving articular surface is a challenging problem faced by the orthopaedic surgeon. In the hand and wrist, there are articular defects that are amenable to autograft reconstruction when primary fixation is not possible. In this article, the surgical techniques and clinical outcomes of articular reconstructions in the hand and wrist using non-vascularised osteochondral autografts are reviewed.

Full text

1448 THE JOURNAL OF BONE AND JOINT SURGERY
INSTRUCTIONAL REVIEW: UPPER LIMB
The use of non-vascularised osteochondral
autograft for reconstruction of articular
surfaces in the hand and wrist
C. Y. Ng,
A. C. Watts
From Wrightington
Hospital, Wigan,
United Kingdom
C. Y. Ng, MBChB(Hons),
FRCSEd(Tr & Orth), DipSEM,
Upper Limb Fellow
A. C. Watts, BSc, MBBS,
FRCS(Tr & Orth), Consultant
Elbow, Hand & Upper Limb
Surgeon
Upper Limb Unit, Wrightington
Hospital, Hall Lane, Appley
Bridge, Wigan WN6 9EP, UK.
Correspondence should be sent
to Mr C. Y. Ng; e-mail:
chyeng@gmail.com
©2012 British Editorial Society
of Bone and Joint Surgery
doi:10.1302/0301-620X.94B11.
30082 $2.00
J Bone Joint Surg Br
2012;11:1448–54.
Bone loss involving articular surface is a challenging problem faced by the orthopaedic
surgeon. In the hand and wrist, there are articular defects that are amenable to autograft
reconstruction when primary fixation is not possible. In this article, the surgical techniques
and clinical outcomes of articular reconstructions in the hand and wrist using non-
vascularised osteochondral autografts are reviewed.
Combined loss of bone and articular cartilage
from joint surfaces presents reconstructive
challenges. The disruption results in joint
incongruity that may lead to reduction in
movement and function. The reconstructive
options may be divided into prosthetic or bio-
logical replacements with either autologous or
allogeneic grafts. Using autografts avoids the
potential risks of disease transmission and lim-
ited availability associated with the use of allo-
grafts.1 An ideal autograft would render
minimal donor site morbidity, afford long-
lasting anatomical reconstruction, and provide
mechanical stability to the reconstructed joint.
In the hand and wrist, there are defects that are
amenable to autograft reconstruction.
Articular defects may be caused by trauma,
nonunion, avascular necrosis (AVN) and osteo-
chondritis dissecans. Consideration has to be
given to local factors, such as the size and loca-
tion of defect, associated soft-tissue injury, previ-
ous intervention, presence of infection or
complex regional pain syndrome; patient factors
including the age, general health, expectations,
functional demand and motivation; and the sur-
gical factors of technical skill and experience.
Reconstruction using vascularised bone grafts
has already been comprehensively reviewed.2
In this article, we review the surgical tech-
niques and clinical outcomes of articular
reconstructions in the hand and the wrist using
non-vascularised osteochondral autografts.
The biology of osteochondral graft
The structural osteochondral graft consists of
both bone and cartilage. The rationale is to recre-
ate the lost articular surface using native hyaline
cartilage with a similar contour that is supported
on subchondral bone. The success of reconstruc-
tion relies on the immediate mechanical stability
and the long-term biological fate of the graft.3
Fixation is generally achieved with screws or
wires. Non-vascularised bone grafts undergo a
process of revascularisation and remodelling as
new bone grows into the graft by creeping substi-
tution.4,5 Experimental works in dogs suggest
that the thickness of the graft should be ≤ 5 mm
for reliable integration.6,7 Temporary weakening
of the subchondral bone may occur as the
necrotic graft is resorbed.4,8 Animal studies have
shown that systemic treatment before
harvesting4 or direct local treatment8 of bone
graft with bisphosphonates may reduce resorp-
tion of the grafted bone. Bisphosphonates adhere
to the bone mineral of calcium, which is internal-
ised by the osteoclasts during resorption, and
interfere with the cell metabolism, resulting in
apoptosis.9 The mechanical support afforded by
the graft is thus maintained for a longer period
during which in-growth and remodelling of new
bone can occur. This may help to reduce the risk
of failure and collapse of load-bearing osteo-
chondral graft.4 However, bisphosphonate treat-
ment before osteochondral bone grafting is not
yet established in clinical practice.
Costo-osteochondral graft comprises a rib
graft with a segment of costal cartilage and
the adjoining bone. This was first used to
reconstruct the temporomandibular joint10
and has since proven to be a versatile donor in
small joint11,12 and carpal bone reconstruc-
tions.13-17 An experiment using transgenic
mice has shown that costal cartilage shares
phenotypic similarities with articular carti-
lage.18 In clinical studies biopsies of the
grafted cartilage have shown evidence of via-
ble chondrocytes.13,19,20 When used in recon-
struction of the temporomandibular joint in
children, the graft also appeared to grow in
the longer term.21

THE USE OF NON-VASCULARISED OSTEOCHONDRAL AUTOGRAFT FOR RECONSTRUCTION OF ARTICULAR SURFACES 1449
VOL. 94-B, No. 11, NOVEMBER 2012
Harvesting techniques
Osteochondral graft. Osteochondral grafts may be har-
vested from the injured hand thereby limiting surgery to a
single limb. Available donor sites include the distal or prox-
imal ends of the second or third carpometacarpal (CMC)
joint,22,23 radial styloid,22,24 the trapezoid,22 dorsal
hamate25 and ulnar base of the fifth metacarpal.26
The hemi-hamate autograft has received particular atten-
tion and become popular as a graft to reconstruct the volar
lip of the base of middle phalanx.3,25,27-29 Suitability of the
graft and donor site stability after harvest was confirmed by
a cadaver study.29 The graft is harvested from the dorsum
of the ipsilateral hand.25 The bases of the fourth and fifth
metacarpals are localised using fluoroscopy before a trans-
verse incision is made proximal to the CMC joint, directly
over the hamate. The extensor tendons are retracted radi-
ally to allow capsulotomy with the capsule then reflected to
expose the hamate (Fig. 1). The distal articular ridge of
hamate will become the proximal articular ridge of middle
phalanx and is used as the reference point for planning the
graft (Fig. 2). The graft should be harvested, preferably
using an osteotome to avoid the risk of thermal necrosis
associated with a saw, slightly larger than required to allow
for sizing error and subsequent shaping. The graft is provi-
sionally fixed to the prepared bed with a Kirschner (K)-wire
before definitive fixation with screws (Figs 3 to 5).
The ulnar base of the fifth metacarpal is harvested
through a longitudinal dorsal incision. The extensor carpi
ulnaris tendon attachment is carefully elevated sub-
periosteally, in continuity with the diaphyseal periosteum
of the metacarpal for later reattachment.26 Typically a
5 mm- to 6 mm-wide graft is harvested. The shape of the
ipsilateral base of the metacarpal resembles the radial con-
dyle of the proximal phalanx. For the ulnar condyle recon-
struction, the graft is reversed.26 Distant osteochondral
grafts from the metatarsophalangeal (MTP) or proximal
interphalangeal joints (PIPJ) of the foot have been used to
reconstruct the corresponding joints of the hand.30-34
Boulas et al31 described dorsal excision of the metatarsal
head or base of the proximal phalanx or both of the ipsilat-
eral second, third or fourth MTP joints (depending on size
match) to fill the defect in metacarpophalangeal (MCP)
joints of index fingers. At the donor site, interposition of
Fig. 1
Clinical photograph showing the dis-
tal hamate donor site with bases of the
fourth and fifth metacarpals exposed
(Reprinted with permission from
Williams et al. Treatment of unstable
dorsal proximal interphalangeal frac-
ture/dislocations using a hemi-
hamate autograft. J Hand Surg Am
2003;28:856–865).
Fig. 2
Line drawing showing the prepared base of middle
phalanx (dimensions of the defect shown) and the cor-
responding donor site on the hamate (note the central
ridge and dimensions of the required graft) (Reprinted
with permission from Williams et al. Treatment of
unstable dorsal proximal interphalangeal fracture/dis-
locations using a hemi-hamate autograft. J Hand Surg
Am 2003;28:856–865).
Fig. 3
Intra-operative photograph showing the proximal interphalangeal joint
of the middle finger exposed through a ‘shot gun’ volar approach. The
volar plate was detached distally. The articular defect involved the volar
half of the middle phalanx.

1450 C. Y. NG, A. C. WATTS
THE JOURNAL OF BONE AND JOINT SURGERY
joint capsule into the residual space creates an excision
arthroplasty. Similarly through a dorsal approach, the
condyle of the proximal phalanx of the toe may be har-
vested together with a segment of collateral ligament.32 The
length of the graft and radio-ulnar width available are nor-
mally adequate but the dorso-volar height is always slightly
deficient.32 Sizing of the graft is thus critical, requiring care-
ful consideration of the potential bone loss that might result
from a saw cut (Fig. 6). No morbidity has been described in
relation to the donor toes with the donor joints being
managed successfully with arthrodesis.32 In contrast, com-
plaints of discomfort with weather change and shortening
of the donor toe have been noted with the MTP joint.31
Costo-osteochondral graft. The graft may be harvested
from any of the fourth to ninth ribs via a transverse sub-
mammary incision on the same side as the injured limb.11-15
It is advisable to start with a small incision in the mid-
clavicular line to identify the osteochondral junction first
before extending the incision to achieve adequate expo-
sure.14 The intercostal muscles are detached from the rib
using electrocautery, after which a curved rib rasp is care-
fully inserted deep to the rib between the perichondrium
and the parietal pleura. The posterior surface of the rib is
then cleared from laterally to medially. Care is taken to pre-
serve the perichondrium in order to avoid separation of the
graft components. It is critical to determine the appropriate
size of resection and at least 2.5 cm on either side of the
costo-chondral junction is harvested to ensure that the final
graft is of adequate size.14 Once harvested the graft can be
shaped with a blade and bone nibblers. The aim is to leave
only a thin segment of bone at the costo-chondral junction
to permit fusion to a prepared bone surface.
Even though there are understandable concerns with
potentially serious complications associated with harvesting
the graft, no major complication or death has been reported
in the clinical series.11-14 Nonetheless the scar may be
unsightly and patients should be counselled appropriately
before surgery.11-14 The defect at the donor ribs is replaced
with hard connective tissue and appears to cause no subse-
quent symptoms.19 Meticulous surgical technique and
appropriate training are crucial to performing a safe harvest.
Clinical application and outcomes
Hand. Fracture-dislocations of the PIPJ most commonly
occur in a dorsal direction with variable bone loss involving
the volar lip of the middle phalanx.35 When more than 50%
Fig. 4
Intra-operative photograph showing how the contoured hemi-hamate
graft was placed to fill the defect. A Kirschner wire was used to control
and temporarily fix the graft.
Fig. 5
Post-operative posteroanterior (a)
and lateral radiographs (b) showing
the hemi-hamate reconstruction of
the proximal interphalangeal joint.
Fig. 6
Line drawing showing the area used when
harvesting a partial toe joint osteochondral
graft with a segment of the collateral liga-
ment (Reprinted with permission from Gaul
JS Jr. Articular fractures of the proximal
interphalangeal joint with missing ele-
ments: repair with partial toe joint osteo-
chondral autografts. J Hand Surg Am
1999;24:78–85).

THE USE OF NON-VASCULARISED OSTEOCHONDRAL AUTOGRAFT FOR RECONSTRUCTION OF ARTICULAR SURFACES 1451
VOL. 94-B, No. 11, NOVEMBER 2012
of the volar surface is lost, or when more than 30° of flexion
is required to keep the joint reduced, the buttressing effect of
the volar lip needs to be recreated to restore joint stability.25
Ishida et al23 reported the outcome of ten cases of reconstruc-
tion of PIP or distal interphalangeal (DIP) joints using a graft
harvested from the second or third CMC joint. After a min-
imum follow-up of two years, the mean active range of
movement increased from 22° to 38° and the mean angular
deformity decreased from 33° to 4°.23
Williams et al25 used the hemi-hamate autograft to recon-
struct the base of the middle phalanx. In their initial clinical
series of 13 patients, the mean articular involvement was
60% (40% to 80%).27 After a mean follow-up of 17 months,
the authors reported a mean range of movement of 85° at the
PIPJ with 100% rate of union. However, there were two
cases of recurrent dorsal subluxation. Hemi-hamate recon-
struction can be used for both acute and chronic cases.
Calfee et al28 reviewed 22 patients with such injuries
(14 acute, eight chronic) at a mean follow-up of 4.5 years.
Active movements averaged 70° with a mean flexion con-
tracture of 19°. Grip strength was about 95% of the opposite
hand. However the mean pain scores were higher following
chronic reconstructions compared with acute cases.
Collapse of a non-vascularised graft may lead to degen-
erative change. Afendras et al3 reviewed eight patients after
a minimum of four years following hemi-hamate arthro-
plasty. All grafts had united radiologically but two patients
developed severe degenerative changes secondary to AVN
and collapse of the graft and two had mild changes at the
reconstructed joints but only one of them had significant
symptoms post-operatively.
Cavadas et al26 performed an anatomical study and
reported their clinical experience of using the ulnar base of
the little finger to reconstruct the condyle of proximal pha-
lanx. A total of 15 patients, who underwent 16 unicondylar
reconstructions, were reviewed after a mean follow-up of
4.5 years. No significant pain was reported and the mean
arc of movement was 48°. Notably, variable degrees of
graft resorption were identified but these did not appear to
correlate with pain or loss of function.
Wu et al22 reported their experience of reconstructing
finger joints in 16 patients using osteochondral grafts taken
from the bases of the second or third metacarpals, the
radial styloid or the trapezoid. The mean age was
36.5 years and they were followed up from 12 to
62 months. This was a heterogeneous series including ten
complex open injuries. At final review, ten patients had no
pain in the reconstructed joints, five had mild discomfort
and one had severe pain ultimately requiring fusion of the
PIP joint. The mean range of movement was 55.8% of that
of the contralateral normal joint. Donor site morbidity was
observed in only one patient who required fusion of the sec-
ond CMC joint to prevent instability. The authors particu-
larly favoured the radial styloid and the base of second
metacarpal as donor sites partly because of the possibility
of harvesting a part of the radial collateral ligament or the
extensor carpi radialis longus (ECRL) tendon for collateral
ligament reconstruction of the finger joint.22
Hasegawa and Yamano12 transplanted a costal cartilage
graft from the seventh or eighth rib to reconstruct articular
defects involving the proximal phalanx at the PIP joint. In
their first two cases, only costal cartilage was used and the
grafts did not survive, resulting in bony ankylosis. In their
subsequent five cases, costo-osteochondral grafts were har-
vested and all of them united. A mean range of movement of
64° was achieved. From this small series, one may infer the
crucial supportive role played by the adjoining bone when
using a rib graft. Sato et al11 expanded the application of
costo-osteochondral graft and used it to reconstruct MCP,
PIP, DIP and thumb IP joints with post-traumatic articular
cartilage loss or AVN.19 The articular defect ranged from
50% to 100%. A total of 30 joints in 29 patients, including
three joints with complete bony ankylosis, were recon-
structed using the autografts.11 All grafts united within a
mean of 58 days and all gained some improved movement.
However, two patients complained of discomfort when mov-
ing the joints. In seven patients (eight joints), a biopsy of the
grafted cartilage was performed at the time of screw
removal.19 Histology of the specimens revealed scattered
chondrocytes within a matrix, similar to that of normal hya-
line cartilage.19 Zappaterra et al36 reported their results of
the use of rib graft in reconstructing PIP and MCP joints in
six patients aged between 15 and 46 years. After a mean
follow-up of 16 months, the PIP joints achieved a mean arc
of movement of 33° and the MCP joints 37°. The mean
Buck-Gramko score37 was 11/15; the mean Strickland
score38 58%; and the mean quick Disabilities of the Arm
Shoulder and Hand (DASH)39 score 17.
Partial osteochondral grafts from the foot have been used
with mixed results. Boulas et al31 reviewed five patients
who underwent MCP joint reconstruction using an MTP
autograft. After a mean follow-up of 33 months, all
achieved full active extension with 74° of flexion and grip
strength equal to 86% of the contralateral hand. One
patient complained of crepitus. All grafts united but one
developed a partial AVN. Gaul32 reported on five patients
who underwent reconstruction of the PIPJ using a partial
toe joint autograft. The mean age was 36 years with a mean
follow-up of seven years. Two cases of graft resorption
occurred, one was converted to arthroplasty while the other
underwent repeat grafting. The arc of movement varied
from 10° to 80°.
Wrist. Nonunion and AVN of the proximal pole of the scaph-
oid remain challenging to treat. Sandow13 described a
replacement arthroplasty of the proximal scaphoid using a
costo-osteochondral autograft harvested from the fifth or
sixth rib. The graft was sculpted with a blade to reproduce the
shape of the proximal scaphoid pole and fixed with two lon-
gitudinal K-wires with the tips bent like a hockey-stick and
buried. He reviewed 22 patients after a median follow-up of
24 months and reported statistically significant improvement
in wrist function score, grip strength and arc of movement. In

1452 C. Y. NG, A. C. WATTS
THE JOURNAL OF BONE AND JOINT SURGERY
one patient a small haemothorax developed that did not
require specific treatment and four patients subsequently
underwent re-operation due to poor movement or radial sty-
loid impingement. The author took the opportunity to biopsy
the graft, which demonstrated variable survival of the chon-
drocytes with changes suggesting adaptive metaplasia at the
articulating surface of the chondral portion. Interestingly
there was a firm fibrous connection between the scaphoid
graft and the lunate in all these cases. Tropet et al17 reported
their results in 18 patients after a mean follow-up of
4.1 years. In 15 patients excellent or good results were
achieved two were fair and one was poor due to subluxation
Table I. Summary of case series describing osteochondral autograft reconstructions in the hand and wrist
Author/s Reconstruction*Sample
size
Mean age (yrs)
(range)
Mean follow-up
(mths) (range) Outcomes†
Afendras et al3Volar lip of middle
phalanx using hemi-
hamate
8 49 (25 to 66) 60 (48 to 69) Mean ROM 67°. Mean grip strength 91% of uninjured
side. Two wrists with Kellgren and Lawrence grade
IV and two with grade II arthritis, but only one was
symptomatic
Boulas et al31 MCP joint using MTP
of 2/3/4th toes
5 41 (23 to 60) 33 (17 to 50) Full active MCP extension, mean flexion of 74°. Mean
grip strength 86% of uninjured side. All united. One
case of partial AVN
Calfee et al28 Volar lip of middle
phalanx using hemi-
hamate
22 34 (14 to 63) 54 (12 to 84) Mean active ROM 70°. Mean flexion contracture 19°.
Mean pain score 1.4/10. Mean DASH 5. Mean grip
strength 95% of opposite hand
Cavadas et al26 Condyle of proximal
phalanx using ulnar
base of the little fin-
ger metacarpal
15 38 (24 to 52) 59 (12 to 90) Mean active ROM 48°. No significant pain. No pain
or instability at the donor area
Gaul32 Condyle of proximal
phalanx using proxi-
mal phalanx of toe
joint
5 36 (17 to 59) 84 (24 to 144) Active ROM: 80° in two digits, 45° in one and 10° in
one. Two instances of resorption: one converted to
arthroplasty and one re-grafted
Hasegawa and
Yamano12
Condyle of proximal
phalanx using costo-
osteochondral graft
5 32 (12 to 52) 10 to 114 Mean active ROM 64°. Two cases of ankylosis when
only cartilage was grafted
Huard et al15 Lunate using costo-
osteochondral graft
4 40 (32 to 51) 27 (6 to 36) Resolution of pain. Mean flexion-extension arc 108°.
Mean grip strength 83% of opposite side. Mean
DASH 6
Ishida et al23 DIP/PIP joints using
2nd/3rd CMC joint
10 13 (6 to 34) 64 (25 to 113) Mean active ROM increased from 22° to 38°. Mean
angular deformity decreased from 33° to 4°. Mean
grip strength 84% of opposite side
Lo and Chang24 Dorsal base of proxi-
mal phalanx using
radial styloid
2 36 (29 to 43) 34 (36 to 32) Pain-free. Both had flexion arc of 5° to 50° at the MCP
joints. Stable reduction with incorporation of graft
Sandow13 Proximal pole of
scaphoid using
costo-osteochondral
graft
22 31 (20 to 44) 26 (12 to 72) All reported increased ROM, improved grip strength
and less pain. Mean modified Green and O’Brien
Wrist Function Score improved from 53 to 80. 100%
union rate. One small haemothorax. Four
re-operations
Sato et al11 DIP/PIP/MCP and
thumb IP joints using
costo-osteochondral
grafts
16 37 (18 to 68) 28 (18 to 57) Mean increase in ROM 45°. Mean time to union of 58
days. Chondrocytes appeared viable in the grafts
Sato et al 19 DIP/PIP/MCP and
thumb IP joints using
costo-osteochondral
grafts
29 34 (14 to 68) 22 (4 to 57) Mean increase in ROM: 59° (MCP); 49° (PIP); 39°
(DIP); 34° (thumb IP). 100% union rate. Two had
discomfort on movement
Veitch et al40 Proximal pole of
scaphoid using
costo-osteochondral
graft
14 26 (21 to 53) 64 (27 to 103) 13 excellent/good, one fair. Mean grip strength
improved from 33 kg to 42 kg. 100% union rate
Williams et al27 Volar lip of middle
phalanx using hemi-
hamate
13 29 (15 to 50) 16 (6 to 43) Mean ROM 85°. Mean grip strength 80% of uninjured
side. 100% union. Mean pain score 1.3/10. Two
recurrent dorsal subluxations
Wu et al22 DIP/PIP/MCP joints
using base of 2nd/
3rd metacarpal,
radial styloid, trape-
zoid
16 36.5 (17 to 63) 16.8 (12 to 62) Mean ROM 56% of the opposite joint. 15 had no or
mild discomfort. Three had mild narrowing of the
joint space and two had slight joint subluxation. One
had severe pain and went on to have fusion of PIP
joint. No reported donor site morbidity
Zappaterra et al36 PIP/MCP joints using
costo-osteochondral
grafts
6 29 (15 to 46) 16 (9 to 25) Mean ROM 33° (PIP); 37° (MCP). Mean Buck-Gramko
score 11/15. Mean Strickland score 58%. Mean Quick-
DASH 17
* MCP, metacarpophalangeal; MTP, metatarsophalangeal; DIP, distal interphalangeal; PIP, proximal interphalangeal; CMC, carpometacarpal; IP, inter-
phalangeal
† ROM, range of movement; AVN, avascular necrosis; DASH, Disabilities of the Arm Shoulder and Hand

THE USE OF NON-VASCULARISED OSTEOCHONDRAL AUTOGRAFT FOR RECONSTRUCTION OF ARTICULAR SURFACES 1453
VOL. 94-B, No. 11, NOVEMBER 2012
of the graft. Veitch, Blake and David40 prospectively reviewed
14 patients who underwent the procedure. After a mean
follow-up of 64 months, the modified wrist function score of
Green and O’Brien41 improved from a mean of 54 to 79. Grip
strength increased from a mean of 33 kg to 44 kg. All but one
achieved a good or excellent outcome. Carpal alignment
appeared to be maintained, or even improved in some that
were abnormal pre-operatively. This reconstruction appears
to be a viable treatment option in those patients who have a
deficient proximal pole that are not suitable for conventional
bone grafting or those who have had previous failed opera-
tions.13,40 However, it is contraindicated in those who have
developed significant degenerative carpal changes.13 Wrists
with grade 1 scaphoid nonunion advanced collapse (SNAC)
change may still be suitable for the reconstruction given the
potential of improving carpal alignment and thereby off-
loading the styloid-scaphoid articulation. Early to medium-
term results would suggest that the mechanical integrity of
the radial column of the carpus is restored by this procedure.
The clinical use of costochondral graft continues to be
expanded. Huard et al15 treated four patients with
advanced Kienbock’s disease with excision of the lunate
and replacement with costochondral autograft harvested
from the ninth rib. The age of the patients ranged from
32 to 51 years. They were prospectively reviewed and after
a mean follow-up of 27 months, they reported resolution of
wrist pain, a mean DASH score of 6, arc of movement of
108° and grip strength 83% of the contralateral wrist. The
early results appear encouraging but the long-term out-
come remains unknown. Costo-osteochondral graft has
also been used successfully to reconstruct the proximal pole
of the capitate. In a series of four patients, who were fol-
lowed up for 12 to 60 months post-reconstruction, signifi-
cant pain relief and improved wrist function scores were
reported.14 Tropet et al16,17 described the application of
costochondral grafting in the treatment of trapeziometa-
carpal arthritis. After a mean follow-up of 5.6 years,
100 patients, who underwent partial trapeziectomy and
autologous costochondral grafting, were reported to have
achieved better results than those of trapeziectomy with
tendon interposition or ligamentous reconstruction.17 The
authors attributed their favourable results to better stability
of the height of the thumb ray following costochondral
grafting. The graft has also been used to reconstruct the
articular surface of the distal radius.17
Conclusion
Articular reconstruction using autograft is technically
demanding. In complex trauma to the hand, the tissue bank
concept is particularly valuable where a non-salvageable
digit may serve as the source of donor tissues for the dam-
aged hand.42 Careful patient selection, accurate characteri-
sation of the articular defect, meticulous harvesting of the
graft and subsequent shaping to fit the defect, and a well-
planned post-operative rehabilitation are vital for a good
outcome.
Evidence-based decision making is hampered by the poor
levels of evidence in the literature due to the heterogeneous
case mix of conditions requiring articular reconstruction and
the paucity of comparative studies. Much of the literature
comprises case series from specialist centres with early to
medium term results only (Table I). Studies of the long-term
function of the reconstructed joints would be valuable.
The authors prefer a local reconstructive option if avail-
able, in order to limit surgery to the injured limb. Specifi-
cally, hemi-hamate autograft is our graft of choice to
reconstruct the volar lip of middle phalanx. For bone loss
involving the convex surface of PIP or MCP joints, partial
toe osteochondral grafts provide the closest anatomical
match. Costo-osteochondral grafting demands skilful shap-
ing and is associated with the risk of unsightly scar. None-
theless the rib graft has an important role in reconstruction
of proximal scaphoid.
No benefits in any form have been received or will be received from a commer-
cial party related directly or indirectly to the subject of this article.
References
1. Sutherland AG, Raafat A, Yates P, Hutchison JD. Infection associated with the use of
allograft bone from the north east Scotland Bone Bank. J Hosp Infect 1997;35:215–222.
2. Malizos KN, Dailiana ZH, Innocenti M, et al. Vascularized bone grafts for upper limb
reconstruction: defects at the distal radius, wrist, and hand. J Hand Surg Am
2010;35:1710–1718.
3. Afendras G, Abramo A, Mrkonjic A, et al. Hemi-hamate osteochondral transplantation
in proximal interphalangeal dorsal fracture dislocations: a minimum 4 year follow-up in
eight patients. J Hand Surg Eur Vol 2010;35:627–631.
4. Tägil M, Aspenberg P, Astrand J. Systemic zoledronate precoating of a bone graft
reduces bone resorption during remodeling. Acta Orthop 2006;77:23–26.
5. Chacha PB. Vascularised pedicular bone grafts. Int Orthop 1984;8:117–138.
6. Pap K, Krompecher S. Arthroplasty of the knee: experimental and clinical experiences.
J Bone Joint Surg [Am] 1961;43-A:523–537.
7. Depalma AF, Tsaltas TT, Mauler GG. Viability of osteochondral grafts as determined by
uptake of S35. J Bone Joint Surg [Am] 1963;45-A:1565–1578.
8. Tägil M, Astrand J, Westman L, Aspenberg P. Alendronate prevents collapse in
mechanically loaded osteochondral grafts: a bone chamber study in rats. Acta Orthop
Scand 2004;75:756–761.
9. Rogers MJ. New insights into the molecular mechanisms of action of bisphosphonates.
Curr Pharm Des 2003;9:2643–2658.
10. Gilles HD. Plastic surgery of the face. London: Oxford Univ ersity Press, 1920:13–15,177–
182.
11. Sato K, Nakamura T, Nakamichi N, et al. Finger joint reconstruction with costal oste-
ochondral graft. Tech Hand Up Extrem Surg 2008;12:150–155.
12. Hasegawa T, Yamano Y. Arthroplasty of the proximal interphalangeal joint using costal
cartilage grafts. J Hand Surg Br 1992;17:583–585.
13. Sandow MJ. Proximal scaphoid costo-osteochondral replacement arthroplasty. J Hand
Surg Br 1998;23:201–208.
14. Sandow MJ. Costo-osteochondral grafts in the wrist. Tech Hand Up Extrem Surg
2001;5:165–172.
15. Huard S, Rochet S, Lepage D, Garbuio P, Obert L. New treatment of advanced Kien-
böck disease: replacement of lunate with costochondral autograft. Chir Main 2011;30-
3:211–217 (in French).
16. Tropet Y, Ridoux PE, Garbuio P, Vichard P. Treatment of rhizomelic arthrosis of the
thumb by partial resection of the trapezium and rib cartilage autograft: preliminary results.
Chirurgie 1994;120:563–567 (in French).
17. Tropet Y, Lepage D, Gallinet D, et al. Articular reconstructions by a costochondral graft-
ing (or osteochondral costal grafting). Bull Acad Natl Med 2006;190:1439–1456.
18. Kitaoka E, Satomura K, Hayashi E, et al. Establishment and characterization of chon-
drocyte cell lines from the costal cartilage of SV40 large T antigen transgenic mice. J Cell
Biochem 2001;81:571–582.
19. Sato K, Sasaki T, Nakamura T, Toyama Y, Ikegami H. Clinical outcome and his-
tologic findings of costal osteochondral grafts for cartilage defects in finger joints. J
Hand Surg Am 2008;33:511–515.

1454 C. Y. NG, A. C. WATTS
THE JOURNAL OF BONE AND JOINT SURGERY
20. Iwai S, Sato K, Nakamura T, et al. Costo-osteochondral graft for post-traumatic
osteonecrosis of the radial head in an adolescent boy. J Bone Joint Surg [Br]
2011;93-B:111–114.
21. Munro IR, Phillips JH, Griffin G. Growth after construction of the temporoman-
dibular joint in children with hemifacial microsomia. Cleft Palate J 1989;26:303–
311.
22. Wu WC, Fok MW, Fung KY, Tam KH. Autologous osteochondral graft for trau-
matic defects of finger joints. J Hand Surg Eur Vol 2012;37:251–257.
23. Ishida O, Ikuta Y, Kuroki H. Ipsilateral osteochondral grafting for finger joint
repair. J Hand Surg Am 1994;19:372–377.
24. Lo CY, Chang YP. Osteochondral grafting of the metacarpophalangeal joint in
rheumatoid arthritis. J Hand Surg Br 2003;28:94–97.
25. Williams RM, Hastings H, Kiefhaber TR. PIP fracture/dislocation treatment
technique: use of a hemi-hamate resurfacing arthroplasty. Tech Hand Up Extrem
Surg 2002;6:185–192.
26. Cavadas PC, Landin L, Thione A. Reconstruction of the condyles of the proximal
phalanx with osteochondral grafts from the ulnar base of the little finger metacar-
pal. J Hand Surg Am 2010;35:1275–1281.
27. Williams RM, Kiefhaber TR, Sommerkamp TG, Stern PJ. Treatment of unsta-
ble dorsal proximal interphalangeal fracture/dislocations using a hemi-hamate
autograft. J Hand Surg Am 2003;28:856–865.
28. Calfee RP, Kiefhaber TR, Sommerkamp TG, Stern PJ. Hemi-hamate arthro-
plasty provides functional reconstruction of acute and chronic proximal inter-
phalangeal fracture-dislocations. J Hand Surg Am 2009;34:1232–1241.
29. Capo JT, Hastings H, Choung E, et al. Hemicondylar hamate replacement arthro-
plasty for proximal interphalangeal joint fracture dislocations: an assessment of
graft suitability. J Hand Surg Am 2008;33:733–739.
30. Menon J. Reconstruction of the metacarpophalangeal joint with autogenous met-
atarsal. J Hand Surg Am 1983;8:443–446.
31. Boulas HJ, Herren A, Büchler U. Osteochondral metatarsophalangeal autografts
for traumatic articular metacarpophalangeal defects: a preliminary report. J Hand
Surg Am 1993;18:1086–1092.
32. Gaul JS. Articular fractures of the proximal interphalangeal joint with missing ele-
ments: repair with partial toe joint osteochondral autografts. J Hand Surg Am
1999;24:78–85.
33. Graham WC, Riordan DC. Reconstruction of a metacarpophalangeal joint with a
metatarsal transplant. J Bone Joint Surg [Am] 1948;30-A:848–853.
34. Boulas HJ. Autograft replacement of small joint defects in the hand. Clin Orthop
Relat Res 1996;327:63–71.
35. Ng CY, Oliver CW. Fractures of the proximal interphalangeal joints of the fingers. J
Bone Joint Surg [Br] 2009;91-B:705–712.
36. Zappaterra T, Obert L, Pauchot J, et al. Post-traumatic reconstruction of digital
joints by costal cartilage grafting: a preliminary prospective study. Chir Main
2010;29:294–300.
37. Buck-Gramcko D, Dietrich FE, Gogge S. Evaluation criteria in follow-up studies of
flexor tendon therapy. Handchirurgie 1976;8-2:65–69 (in German).
38. Strickland JW, Glogovac SV. Digital function following flexor tendon repair in
Zone II: a comparison of immobilization and controlled passive motion techniques. J
Hand Surg Am 1980;5:537–543.
39. Gummesson C, Ward MM, Atroshi I. The shortened disabilities of the arm, shoul-
der and hand questionnaire (QuickDASH): validity and reliability based on responses
within the full-length DASH. BMC Musculoskelet Disord 2006;7:1–7.
40. Veitch S, Blake SM, David H. Proximal scaphoid rib graft arthroplasty. J Bone Joint
Surg [Br] 2007;89-B:196–201.
41. Green DP, O’Brien ET. Open reduction of carpal dislocations: indications and oper-
ative techniques. J Hand Surg Am 1978;3:250–265.
42. Chase RA. The damaged index digit: a source of components to restore the crippled
hand. J Bone Joint Surg [Am] 1968;50-A:1152–1160.