Session 2: Hand and Wrist
Onsdag den 23. oktober
09:00 - 10:30
Lokale: Vingsal 1
Chairmen: Torben Bæk-Hansen og Maiken Stilling
11. Ligament Reconstruction Interposition Arthroplasty combined with Suspension Arthroplasty for CMC-1 Osteoarthritis results in Better Mobility and Less Pain at 6 months
Rasmus Wejnold Jørgensen , Kiran Anderson, Claus Hjorth Jensen
Håndkirurgisk Klinik, Ortopædkirurgisk afd. , Gentofte Hospital; Håndkirurgisk klinik, Ortopædkirurgisk afd., Gentofte University Hospital, Copenhagen, Denmark; Håndkirurgisk klinik, Gentofte Hospital
Background: Recovery time after surgery for thumb CMC-1 J
osteoarthritis usually approximates three to six
months regardless of operative technique.
Purpose / Aim of Study: The purpose of this study was to examine if adding
a suspension arthroplasty (Mini Tight Rope, Arthrex)
to Ligament Reconstruction Tendon Interposition of
the CMC-1 joint would add stability to the thumb and
thereby give a faster recovery and less pain at 6
months postoperatively.
Materials and Methods: 12 consecutive patients underwent thumb CMC-1 J
LRTI arthroplasty a.m. Burton Pellegrini. In addition,
they underwent a suspension arthroplasty using two
strands of #2 FiberWire suture which are fixed with
two oblong stainless-steel buttons for cortical
fixation between first and second metacarpal bone
(Mini Tight Rope, Arthrex). Patients were then given
a soft cast and mobilization was begun at 3-5 days
after surgery. The patients are followed using Quick-
DASH questionnaire, VAS pain, satisfaction, range
of motion and pinch strength preoperatively and at
3, 6, 12 and 24 months postoperatively. Each
patient was matched with three controls from our
own database of 250 patients receiving only LRTI
arthroplasty matched on gender, age and
preoperative Q-DASH.
Findings / Results: Preoperative Q-DASH for cases has a median of 44
(range 13-70) and for controls a median of 45
(range 13-75). 6-month postoperative Q-DASH for
cases has a median of 6.8 (range 0-27) and controls
a median of 20,5 (range 0-70). Preliminary results
show a significant difference in Q-DASH for cases
and controls at 6 months postoperatively.
Regarding question 9 of the Q-DASH (1=no pain
and 5=extreme pain) cases has a preoperative
median of 4 (range 2-4) and for controls a median of
3 (range 3-5). 6-month postoperative results show a
median of 1 (range 1-3) for cases and a median of 2
(range 1-5) for controls. This is not statistically
significant.
Conclusions: Treating thumb CMC-1 osteoarthritis with a
combination of LRTI and suture suspension
arthroplasty between first and second metacarpal
bone show significantly lower Q-DASH at 6-months
follow up compared to LRTI alone.
However, the pain level at 6-months follow up is not
significantly different for case and controls.
12. Higher preoperative pain catastrophizing increases the risk of low patient reported satisfaction after carpal tunnel release
Sebastian Breddam Mosegaard, Maiken Stilling, Torben Bæk Hansen
University Clinic for Hand, Hip and Knee Surgery, Department of Orthopaedics, Regional Hospital West Jutland; University Clinic for Hand, Hip and Knee Surgery, Department of Orthopaedics, Regional Hospital West Jutland; University Clinic for Hand, Hip and Knee Surgery, Department of Orthopaedics, Regional Hospital West Jutland
Background: Carpal tunnel syndrome is a common upper
limb nerve compression disease where
operative treatment is used when
conservative treatment fails. However,
several factors may influence the results of
the operation including preoperative pain
catastrophizing.
Purpose / Aim of Study: Based on demographic characteristics,
PROMs and with special reference to pain
catastrophizing the purpose of this study
was to identify risk factors for low patient-
reported satisfaction following surgical
treatment of idiopathic CTS with CTR. The
main hypothesis of this study was that:
Higher preoperative pain catastrophizing
increases the risk of low postoperative
patient reported satisfaction.
Materials and Methods: A total of 417 hands from 417 patients
(64.5% females) with a mean age of 58.0
years was included in this 1-year
prospective follow-up study. Data on DASH
score, EQ-5D, distal motor latency and pain
catastrophizing was collected preoperatively
and data on DASH score, EQ-5D and
patient satisfaction was collected 12 months
postoperatively. Wilcoxon matched-pairs
signed-rank test was used to test for
improvement in DASH and EQ-5D. Risk
factors for low postoperative patient
reported satisfaction was examined using
stepwise multiple logistic regression
analysis.
Findings / Results: We found a general improvement in patients
DASH score (12.29 [95% CI: 10.65 –
13.90], p<0.001) and EQ-5D (0.14 [95% CI:
0.13 – 0.16], p<0.001) from preoperative to
12 months postoperative. In the fully
adjusted multiple regression analysis there
was only statistically significant effect of
preoperative pain catastrophizing. A 1-unit
increase on the preoperative pain
catastrophizing scale lead to an increased
risk of OR=1.05 (p=0.022) for low 12-month
postoperative patient satisfaction. There
was no statistically significant predictive
effect of preoperative EQ-5D (p=0.869),
DASH (p=0.076), distal motor latency
(p=0.067), age (p=0.505), or gender
(p=0.222) on 12-month postoperative
patient satisfaction.
Conclusions: Patients improved both DASH and EQ-5D
scores from preoperative to 12 months
postoperative. Higher preoperative score on
the Pain Catastrophizing Scale seems to
have a negative effect on postoperative
patient reported satisfaction 12 months after
carpal tunnel release.
13. TFCC foveal re-attachment by modified ulnar tunnel technique - significant clinical improvement at one-year follow-up
Sabine Simonsen, Søren Ohrt-Nissen, Robert Gvozdenovic
Orthopaedic Surgery, Herlev-Gentofte Hospital; Orthopaedic Surgery, Hvidovre Hospital; Orthopaedic Surgery, Gentofte Hospital
Background: Traumatic disruption of the Triangular
fibrocartilage complex (TFCC) is a
common cause of ulnar sided wrist
pain and is often associated with
decreased grip strength and impaired
function .TFCC injuries are present in
up to 80% of displaced distal radius
fractures, but can also occur after a
minor wrist trauma. TFCC injuries that
do not respond to conservative
treatment should be offered diagnostic
arthroscopic surgery and re-
attachment if indicated.
Purpose / Aim of Study: The aim of this study was to evaluate
the clinical and patient-reported
outcome one year after TFCC re-
attachment by modified ulnar tunnel
technique.
Materials and Methods: Between April 2013 and June 2018, 32
patients underwent surgery with the
modified Iwasaki ulnar tunnel
technique for foveal re-attachment of
TFCC injury. All patients had ulnar-
sided wrist pain and mild instability of
the distal radioulnar joint (DRUJ).
Diagnosis was finally established
intraoperatively by the hook-test. All
patients were immobilized for 2 weeks
in a sugar tong splint and then with a
removable orthosis with limited rotation
for further 4 weeks. Full weight-bearing
activities were allowed at 3 months.
Prospective evaluation included
assessment of pain (VAS score), grip
strength, range of motion and q-DASH
score.
Findings / Results: The study included 32 patients with 31
available for one-year follow-up. 21
patients had preoperative MRI of which
13 (62 %) showed no signs of TFCC
injury. Median preoperative activity
VAS score was 60 [50-70], which
decreased to 14 [6-20] (p=0.001).
Median q-DASH score was 41 [36-43],
which improved to 11 [4-25] (p ¡Ü
0.001). Grip strength also improved
significantly (p ¡Ü 0.001), while
improvement in rotation of the forearm
was observed without significance. All
patients achieved full stability, except
one who needed re-fixation due to
early discontinuation of the splint. One
patient underwent synovectomy
without improvement of symptoms.
One patient was referred to SL
ligament reconstruction.
Conclusions: Modified ulnar tunnel method improves
stability and symptoms in patients with
TFCC injury. Preoperative MRI is not
sensitive for TFCC injury and should
primarily be considered for differential
diagnostic purposes.
14. Total Joint Arthroplasty of the Trapeziometacarpal Joint in WALANT.
Lotte Priess Larsen, Torben Bæk Hansen
Orthopedic , Regional Hospital West Jutland; Orthopedic, Regional Hospital West Jutland
Background: In recent years the concept of WALANT
(Wide Awake Local Anaesthetic No
Tourniquet) surgery has gained in popularity
and is increasingly reported internationally
as useful in hand surgery. WALANT
provides benefits for both the patient and
the surgeon, but concerns are present
regarding the use in more complex surgical
procedures. Numerous case reports of
WALANT in hand surgery are available but
studies describing total joint arthroplasty
(TJA) of the trapeziometacarpal joint are yet
to be published.
Purpose / Aim of Study: With this study we wanted to do a
descriptive study of possible adverse effects
and patient acceptance of WALANT in TJA
of the trapeziometacarpal joint and to
compare the duration of the operation in
WALANT with TJA performed in a bloodless
field under general or regional anaesthesia.
Materials and Methods: In a matched cohort study we compared
duration of the operation in 27 patients at
mean age 58 (range 48-69) receiving TJA in
WALANT with 27 patients at mean age 59
(range 43-72) receiving TJA in general or
regional anaesthesia with the use of
tourniquet. All patients were operated by the
same surgeon in the same institution and
with the same operative team. In the
WALANT group we also registered
conversion to general anaesthesia,
intraoperative supplemental local
anaesthesia, use of tourniquet, adverse
effects to the anaesthesia and willingness to
repeat (WTR).
Findings / Results: There were no conversions to general
anaesthesia, 2 patients required additional
local anaesthesia intraopertively, 1 patient
required the use of tourniquet for 2 minutes,
1 case of adrenalin rush was registered.
WTR in the WALANT group was 100%.
There was no significant difference in
operating time in the two groups.
Conclusions: The use of WALANT in thumb TJA surgery
is well tolerated by patients and provides a
safe and reliable method of anaesthesia.
WALANT provides similar working
conditions as experienced with the use of
tourniquet during general/regional
anaesthesia and is not associated with
prolonged operating time.
15. Lower recurrence rate of Dupuytrens Contracture following collagenase clostridium histolyticum treatment compared with percutaneous needle fasciotomy.
Stig Jørring, Claus Hjorth Jensen
Department of orthopaedic surgery, Herlev Gentofte Hospital; Department of orthopaedic surgery, Herlev Gentofte Hospital
Background: Few studies report on recurrence rates
following treatment with collagenase
histolyticum injections (CI) of
Dupuytrens contracture in comparison to
other treatment modalities.
Purpose / Aim of Study: To evaluate CI against another
minimally invasive treatment modality a
RCT comparing CI and percutaneous needle
fasciotomy (PNF) of the MCP joints in
Dupuytrens Contracture (DC) was undertaken.
Materials and Methods: 77 patients suffering from DC with
isolated extension deficit > 30 degrees
in the MP joint were randomized in a
RCT. The patients were allocated to
either CI injection in accordance with
manufacturers guidelines or PNF. Only
one finger was treated in each patient,
and only one injection was given to the
patients in the CI group. A correction
of 20 degrees in extension deficit at 1
month was considered an acceptable
result. Patients were followed 1-3
years. Recurrence of > 30 degrees or
new treatment of the finger were
considered failure.
Findings / Results: 5 patients withdrew from the study prior
to treatment leaving 34 patients in the
CI group and 36 in the PNF group. At 1
month 32 patients in the CI group (94%)
and 35 patients in the PNF group (97 %)
were considered having an acceptable
result and were consequently planned for
subsequent follow up at 1, 2, 3 and 5
years. Analysis of durability in
accordance with Kaplan Meyers Statistics
showed a significantly lower recurrence
rate of DD in favour of CI treatment 1
to 3 years after treatment. Log Rank P
0.005.
Conclusions: Considering recurrence CI yields
superior results compared to PNF at
follow up at 1-3 years in the treatment
of isolated DC in the
metacarpophalangeal joints.
16. Clinical Dorsal Wrist Ganglion: Clinical implications after imaging and pathological assessment
Jamila Eriksen, Niels Henrik Søe, Dimitar Ivanov Radev, Xiong Xie, Eva Balslev, Merete Juhl Kønig, Lisbeth Vesterløkke, Helle Raagaard Larsen, Nana Vermehren, Guðlaug Rósa Sigurðardóttir, Britt Ebstrup,
Department of Handsurgery, Nordsjællands Hospital; Department of Handsurgery, Herlev and Gentofte University Hospital; Department of Radiology, Bispebjerg Hospital; Department of MRI, Herlev and Gentofte University Hospital; Department of Pathology, Herlev and Gentofte University Hospital; Department of ultrasound, Herlev and Gentofte University Hospital; Department of Handsurgery, Herlev and Gentofte University Hospital; Department of Handsurgery, Herlev and Gentofte University Hospital; , Aleris-Hamlet Hospital; Department of Anæestesiology, Rigshospitalet; Anæstesiology and Day Care Surgery, Herlev and Gentofte University Hospital; ,
Background: The clinical dorsal wrist ganglion accounts for 60-
70% of the hand and wrist ganglia. Little is known of
the ætiology or pathogenesis of this unsightly and
often painful soft tissue mass and recurrence after
lege artis surgical incision is seen in up to 30% of
cases.
Purpose / Aim of Study: The purpose of the study is to gather as much
knowledge about the dorsal wrist ganglion as
possible through patient history, imaging and
pathological examination to see whether this can
improve diagnosis and treatment and hinder
recurrence.
Materials and Methods: 42 patients with clinical dorsal ganglion have over
two years been included into the study at two
Danish handsurgery departments. The affected
wrists, which have not formerly been treated, have
been examined by X-ray, MRI imaging and
ultrasound imaging. After excision of the dorsal wrist
ganglion by the same surgeon, the tissue has been
stained with CD68, vimentin and D2-40 and
examined microscopically. Patients are followed at 3
months, 6 months, 1 year and 2 years and in case
of recurrence. The study is ongoing.
Findings / Results: We discovered that approximately 50% of the
clinical dorsal ganglia were ganglion cysts and 50%
synovial cysts. Preliminary results show greater
recurrence rates in patients with synovial cysts. X-
ray, MRI and ultrasound imaging cannot precisely
give the exact origin nor predict the type of tissue
the mass is composed of. Neither is there a clear
connection to gender, former trauma or occupation.
Conclusions: This study shows that although the clinical dorsal
ganglion has the same appearance, location and
similar characteristics on X-ray, MRI and ultrasound
imaging, the tissue excised is clearly one of two
distinct tissue types, namely ganglion and synovial
cysts. The aforementioned imaging techniques are
therefore not clinically diagnostic and may not be
justified in the diagnosis and treatment of dorsal
wrist ganglia. The discovery of differences in
recurrence rate in the two tissue types is also new
information, which can lead to improved patient
information and begin to question the choice of
treatment method whether it be open or arthroscopic
excision.
17. Scapholunate ligament reconstruction. One-year follow-up using the SLAM procedure in 43 patients.
Lars Soelberg Vadstrup
Ortopædkirurgisk afdeling, Herlev Gentofte Hospital
Background: The objective of this study is to evaluate
scapholunate ligament reconstruction using a
modified ScaphoLunate Axis Method (SLAM) in
patients with a dynamic or reducible static
scapholunate ligament tear. Many different methods
have been suggested in the past, some with
promising results and some with less promising
results. The minimal invasive SLAM procedure is
reliable in treating SL ruptures and the results are
promising.
Purpose / Aim of Study: To evaluate scapholunate ligament reconstruction
using a modified ScaphoLunate Axis Method
(SLAM) in patients with a dynamic or reducible static
scapholunate ligament tear.
Materials and Methods: Between June 2014 and June 2019, 43
consecutive patients have been operated using
the ScaphoLunate Axis Method (Arthrex) for
reconstruction of a scapholunate ligament tear.
The SL lesion was identified by wrist
arthroscopy, and the scaphoid reducibility was
identified prior to minimal open ligament
reconstruction in a single session. Follow-up
was 12 months. Evaluation pre-operatively and
at 3, 6 and 12 month after surgery included
assessment of range of motion (ROM), grip
strength and Disabilities of the Arm, Shoulder
and Hand (quick-DASH) Score.
Findings / Results: There were three complications during surgery the
operative procedures or the recovery / hand therapy.
All but one patient improved in the patient reported
outcome measure at 12 months as q-DASH values
improved significantly (p<0.05). Mean q-DASH value
pre-operatively was 39 (range 9-80), and 21 (range
0-72) post-operatively. ROM and grip strength was
unchanged at 12 month compared to pre-operative
measures. Mean grip strength was 39 Kilogram-
Forces (KgF) pre-operatively, and 43 KgF post-
operatively (86 % strength of the contralateral side).
Conclusions: Short-term results of the SLAM procedure for
patients with a dynamic or reducible static
scapholunate ligament tear provided satisfactory
results with a few observed complications. The
presented technique using a tendon autograft (PL or
part of the FCR) placed along the axis of rotation of
the SL joint, fixated in both the scaphoid and the
lunate, minimize loss of the obtained SL reduction
and reconstructs the scapho-lunate ligament
complex.
18. Measurements on sagittal CT scans of the scaphoid bone. What are the normal values?
Cæcilie West, Robert Gvozdenovic, Dimitar Radev
Handsection, Department of Orthopeadic Surgery, Herlev Gentofte Hospital; Handsection, Department of Orthopeadic Surgery, Herlev Gentofte Hospital; Department of Radiology, Bispebjerg Frederiksberg Hospital
Background: Scaphoid waist fractures are often undisplaced
but can be complicated by a humpback deformity
with or without union. It is believed that not only
nonunion, but also malunion can cause impaired
outcome. However, there are no consensus on
reliable measurements of scaphoid malunion to
predict functional and subjective outcome. The
lateral intrascaphoid angle (LISA) has previously
shown to be a good predictor of outcome but is
suspected to have poor inter- and intra-rater
reliability. Height-length ratio (HLR) seems to have
a high intra- and interrater reliability but a poor
predictor of outcome and the dorsal cortical angle
(DCA) seems to be less prone to observer bias
than LISA but a poorer predictor of outcome. To
our knowledge, the normal values of HLR, LISA
and DCA, have never been investigated on a
larger population of normal scaphoids.
Purpose / Aim of Study: To determine the normal values of the most commonly
used measurements (HLR, LISA, DCA) on a sagittal
CT scans of the scaphoid.
Materials and Methods: We included CT scans of normal wrists and
scans with pathology not related to the scaphoid.
Exclusion criteria were age younger than 18, any
type of scaphoid pathology, signs of carpal
instability, osteoarthritis and wrist-near fractures
among others. All CT scans were reformatted
along the long axis of the scaphoid and images of
0,5-2mm thickness were obtained. 3 observers
(1 musculoskeletal radiologist, 1 hand surgeon
and 1 orthopedic resident) applied the
measurements. Descriptive values were
calculated using measurement from all three
observers. Interrater reliability was estimated
using intraclass correlation coefficient.
Findings / Results: We investigated CT scans from 62 patients (53 %
men), mean age 39 years. The mean HL-ratio was
0,58 (range 0,49-0,70), mean LISA 27 degrees (4-
58) and mean DCA was 128 degrees (106-151).
Interrater reliability was good (0,83) for HL-ratio,
moderate (0,74) for DCA and poor (0,03) for LISA.
Conclusions: This study provides an impression of the normal
values of the scaphoid and may aid in decision
making when defining malunion. We suggest that
LISA cannot be utilized as a measure of deformity.
Further studies on the correlation between HL-ratio,
DCA and clinical outcome are needed.
19. Primary treatment of trigger finger: Higher recurrence upon corticosteroid-injection
Frederik Flensted Andersen, Rasmus Wejnold Jørgensen, Claus Hjorth Jensen, Jens-Christian Vedel, Henrik Daugaard
Hand Clinic, Department of Orthopedics, Herlev-Gentofte University Hospital ; Department of Orthopedics, Herlev-Gentofte University Hospital ; Hand Clinic, Department of Orthopedics, Herlev-Gentofte University Hospital ; Department of Orthopedics, Herlev-Gentofte University Hospital ; Hand Clinic, Department of Orthopedics, Herlev-Gentofte University Hospital
Background: Treatment options for trigger digits are
typically either surgery or corticosteroid
injections. Though previous studies have
shown big differences in success- and
recurrence rates when comparing the
treatments, follow-ups have been short and
cohorts small
Purpose / Aim of Study: The purpose of the study was to compare
corticosteroid injection with surgical
treatment of first time and reoccurring
trigger finger
Materials and Methods: In a retrospective chart review, records of
815 patients referred with trigger finger
during the period 2014-2015 were identified.
Inclusion criteria was primary treatment of
the finger with corticosteroid injection or
surgery as first line treatment. 210 cases
were excluded, due to previous treatment of
the same finger or wrong diagnosis, as
patients were initially identified by the ICD-
10 code for trigger finger. Primary treatment
type, recurrence, secondary treatment type
and comorbidities were recorded. Minimum
follow-ups were 2 years.
Findings / Results: Of 605 primary trigger digits, 539 were
treated with corticosteroid-injection and 66
with surgery as first line treatment. 330
recurrences in the corticosteroid group as
compared to 8 in the surgery group were
observed, Chi square, (p<0,001). Increased
risk of recurrence was seen after treatment
of the third finger regardless the treatment
as compared to the other digits. Among
others, carpal tunnel syndrome was
associated with an increased risk of
recurrence among those treated with
injection.
As second-line treatment, 207 patients had
corticosteroid-injection, 106 were operated
and 25 had other treatment. After second
line treatment 106 trigger fingers recurred
again in the injection group as compared to
8 in the surgery group, Chi square,
(p<0,001).
Time to recurrence was significantly shorter
in the corticosteroid group, survival analysis
(p<0,001) with a Hazard ratio of 6,839
(CI95%: 3,39-13,79, p<0,05).
Conclusions: Significantly higher recurrence rate of
trigger finger was observed when
comparing corticosteroid-injection to
surgical treatment both primary and
secondary. Recurrence happened faster
upon surgery, but at any timepoint, the risk
of recurrence are higher among those
treated with corticosteroid-injection as
primary treatment.
20. Can pre-operative measurement of humpback deformity and the size of bony cysts predict the union rate and time to healing of Scaphoid Nonunions?
Benjamin Presman, Morten Bo Larsen , Dimitar Ivanov Radev, Stig Jørring, Claus Hjorth Jensen, Robert Gvozdenovic
Orthopaedic Surgery, Section of Hand Surgery, Gentofte Hospital, Copenhagen University, Hellerup, Denmark.; Orthopaedic Surgery, Section of Hand Surgery, Gentofte Hospital, Copenhagen University, Hellerup, Denmark.; Radiology , Gentofte Hospital, Copenhagen University, Hellerup, Denmark.; Orthopaedic Surgery, Section of Hand Surgery, Gentofte Hospital, Copenhagen University, Hellerup, Denmark.; Orthopaedic Surgery, Section of Hand Surgery, Gentofte Hospital, Copenhagen University, Hellerup, Denmark.; Orthopaedic Surgery, Section of Hand Surgery, Gentofte Hospital, Copenhagen University, Hellerup, Denmark.
Background: Non-union is reported to occur in
approximately 10% of all scaphoid
fractures. It is known, that the
occurrence of nonunion is affected by
factors such as fracture location,
displacement and the time from injury
to treatment. However, the impact of
humpback deformity and bony cysts on
union rate and time to healing after
surgery has not been studied.
Purpose / Aim of Study: The purpose of this study was to
examine, if the degree of humpback
deformity or the size of cysts along the
fracture line besides different treatment
methods are associated with the union
rate and healing time following surgery
of scaphoid nonunion.
Materials and Methods: 63 patients with scaphoid delayed or
nonunion and preoperative CT scans
were retrospectively analyzed. Four
orthopedic surgeons and one intra-
scaphoid angle, the height/length ratio,
the size of the cysts and displacement
of the fracture were radiologist
independently analyzed the CT scans.
The dorsal cortical angle, measured.
Healing was assessed by CT scan or
by conventional x-ray and clinical
examination.
Findings / Results: Open surgery was the treatment of
choice in 49 patients, 8 patients were
treated arthroscopically and 6 patients
with delayed union were operated with
percutaneous method. Overall union
rate was 86% (54/63) and was
achieved after 84 days (mean). The
failure rate and time to healing was not
associated with the degree of the
humpback deformity, size of the cysts
or displacement of the fracture.
Patients treated arthroscopically
achieved significantly faster healing
(42 days) as compared to patients
treated by open techniques (92 days)
(p¡Ü0.05). Time from injury to surgery
was significantly correlated with time to
union but not associated with the union
rate (p¡Ü0.05, p¡Ü0.4, respectively).
Conclusions: The degree of humpback deformity
and the size of cysts along the fracture
line have no predictive value for the
surgical result. Time to healing
following surgery is influenced by the
surgical technique and the time from
injury to the surgical treatment of
scaphoid nonunion.