A new definition of wrist sprain necessary after findings in a prospective MRI study

Full text

A new definition of wrist sprain necessary after findings in a prospective
MRI study
T.H. Bergh
a,b,
*, T. Lindau
b,c
, S.V. Bernardshaw
a
, M. Behzadi
d
, L.A. Soldal
a
, K. Steen
a
, C. Brudvik
a,b
a
Bergen Accident and Emergency Department, Bergen, Norway
b
Department of Surgical Sciences, University of Bergen, Bergen, Norway
c
The Pulvertaft Hand Centre, Kings Treatment Centre, Royal Derby Hospital, Derby, UK
d
Department of Radiology, Stavanger University Hospital, Stavanger, Norway
Pain and swelling in the wrist after a trauma is a common
presentation in clinical practice. The routine diagnostic work-up for
acute wrist trauma consists of a physical examination, usually with
radiography.This diagnosticstrategy probably identifies themajority
of fractures and dislocations, but provides very little information
about injuries to soft tissues, such as tendons, inter-carpal ligaments
and the triangularfibrocartilage complex (TFCC).
1
Another diagnostic
problem in the injured wrist is the relatively high rate of occult carpal
bone injuries not identified with traditional X-ray.
2
There are, to the best of our knowledge, no prospective studies
examining the distribution of injuries of a wrist sprain using acute
magnetic resonance imaging (MRI). Most articles describe the
findings of injury by the use of different radiological investigations
when fractures are suspected.
3,4
Wrist injuries with negative X-rays are usually diagnosed as
acute wrist sprains. Wrist sprain is defined by the International
Wrist Investigator Workshop (IWIW) as a partial ligament tear,
5
often without identifying exactly which ligament is injured.
Allegedly, there is sometimes a change in the dimensions of the
affected ligaments, with or without loss of its structural integrity.
5
The treatment is often PRICE (Protection, Rest, Ice, Compression,
Elevation).
6
The prognosis is usually good,
6
but some patients
suffer from prolonged pain and reduced wrist function. This is
probably due to missed diagnoses of more serious pathology.
7
The
aim of this study was to investigate acute wrist sprains with MRI to
detect which pathoanatomical structures are injured.
8–12
Patients and methods
This prospective study was conducted at Bergen Accident
and Emergency Department (A&E), Norway from 5 November 2009
to 4 November 2010. Bergen A&E is an outpatient clinic
Injury, Int. J. Care Injured 43 (2012) 1732–1742
ARTICLE INFO
Article history:
Accepted 25 June 2012
Keywords:
Wrist
Sprains
Magnetic resonance imaging
Fractures
Bruises
Triangular fibrocartilage complex
Ligaments
Soft-tissue injuries
ABSTRACT
Introduction:
Wrist injuries with negative X-rays are diagnosed as acute wrist sprains. The prognosis is
usually good, but some patients suffer from long -lasting pain and reduced wrist function, probably due to
missed diagnosis followed by inappropriate treatment. The aim of this study was to investigate acute
wrist sprains with MRI to detect the pathoanatomy of the injury.
Patients and methods: This prospective magnetic resonance imaging (MRI) study included patients
between 18 and 49 years, who attended the Accident and Emergency Department (A&E) Bergen, Norway,
after sustaining an acute wrist trauma within the previous week. Initial X-rays of the wrist were normal.
MRI was done within a median of 1 day (range 0–31 days) after the trauma, 80% within 4 days. The study
period lasted from 5 November 2009 to 4 November 2010.
Results: A total of 155 acute MRIs were done, out of which 30 were completely normal. Patients with
positive MRI had a median of two (range 0–8) pathological findings. We found 54 fractures and 56 bone
bruises, mostly located to the radius followed by the scaphoid, the triquetrum, the capitate and the
lunate. There were 73 soft-tissue injuries, which included 15 injuries to the triangular fibrocartilage
complex (TFCC) and five scapho-lunate (SL) ligament lesions.
Conclusions: Wrist sprain is an inaccurate diagnosis. In four out of five patients with normal X-rays, MRI
identified pathological findings and a large variety of injuries in different structures. We suggest that
wrist sprain should be defined as ‘‘occult partial or complete soft tissue (ligament, tendon, muscle) or
bony injury in relation to a trauma with negative X-ray’’. The MRI findings led to a more differentiated
treatment in more than a third of the patients. We recommend that MRI should be considered as a part of
an early investigation, especially when the wrist pain does not settle within the first couple of weeks.
ß2012 Elsevier Ltd. All rights reserved.
* Corresponding author at: Bergen Accident and Emergency Department, Vestre
Strømkaien 19, 5008 Bergen, Norway. Tel.: +47 55568700; fax: +47 55327934.
E-mail address: torbjorn.bergh@kir.uib.no (T.H. Bergh).
Contents lists available at SciVerse ScienceDirect
Injury
journal homepage: www.elsevier.com/locate/injury
0020–1383/$ – see front matter ß2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.injury.2012.06.028

treating almost all of the minor injuries in Bergen, the second-
largest city in Norway. Annually, 100 000 patients attend our A&E,
out of which 40 000 are treated at the surgical division at the A&E
due to different types of injuries. We included patients aged 18–49
years who attended the A&E within a week after sustaining an
acute wrist trauma requiring X-rays. Patients with clear radio-
graphic fractures or dislocations were excluded. Other exclusion
criteria were patients with rheumatoid arthritis and previous wrist
fractures. Furthermore, patients with contraindications for MRI
such as pregnancy, metal implant and claustrophobia were also
excluded.
Mechanisms of injury were divided into high- and low-energy
trauma, defining low-energy trauma as equivalent to that
generated by a fall from standing position or lower. Falls from
higher positions or other injury mechanisms during sports were
defined as high-energy trauma.
Both the clinical examination and the interpretation of the X-
rays were done by the doctors on duty at the A&E as a part of their
daily clinical practice. All doctors were instructed in appropriate
wrist examination.
We used a standardised X-ray protocol with extended wrist
views including the distal part of the radius and the proximal parts
of the metacarpals. A series of four images were used (Fig. 1 and
legends) (Philips Diagnost with micro focus (0.3 mm), Agfa CR and
digitiser, Agfa Picture Archiving and Communication System
(PACS)).
The MRI protocol included coronal T1 spin echo (SE)
(fractures), coronal short tau inversion-recovery (STIR) (bone
bruise), coronal T2* (T2 gradient echo) (ligament- and TFCC
ruptures) and axial proton density fat-saturated (PD FS) (field
of view (FOV): 11 cm). Slice thickness: overall 3 mm but 1 mm
on T2. MRI scans were performed in a 1.5-T whole-body scanner
withawristcoil.FiveoftheMRIsweredoneatadifferent
institute. However, our radiologist accepted the quality of
the pictures and found them acceptable to be included in our
study.
MRI was done with a median of 1 day (0–31 days) after the
trauma. All MRIs were interpreted by an experienced musculo-
skeletal radiologist (MB). As an internal quality control, we used
three other experienced radiologists to interpret a random
selection of 10% of the MRI scans. Their results were in accordance
with the reported findings by MB.
The following definitions were used in our study for pathologi-
cal findings:
Fracture. A complete occult fracture extends through the entire
cross-section of the bone. An incomplete fracture does not
extend through the full transverse width of a bone, only
[(Fig._1)TD$FIG]
Fig. 1. X-ray wrist in four projections. The first was a standard posterior–anterior (PA) view with the wrist in ulnar deviation. The second was a lateral view of the carpus and
distal radius/ulna. The third and fourth views were oblique: one with the hand in 458supination the other with 458pronation (Pat id 145).
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1733

partly across the bone. In general, the findings of bone injury
may be subtle on PD, T2-weighted and T2*-weighted sequences
and are more conspicuous in T1-weighted, fat saturated T2/PD
or STIR sequences. Fractures appear as linear foci of decreased
signal intensity on T1-weighted images. High-signal intensity
oedema is seen on STIR sequences, classically surrounding a low
signal-intensity fracture line.
Bone bruise. Trabecular injuries that result from impaction
forces are known as bone bruise, bone contusions or micro-
trabecular fractures. Pathological studies of these injuries
reveal trabecular fractures and oedemas and haemorrhage
in the adjacent bone-marrow fat. MRI findings become
evident within hours of the injury. Bone bruise is most
conspicuousonSTIRorFST2orFSPDimages,whereit
appears as focal areas of increased signal intensity, usually
with poorly defined margins, presumably secondary to
the haemorrhage and oedema related to the trabecular
fractures.
Partial tendon rupture/tendinitis. The term ‘tendinitis’ is
reserved for tendon injuries that involve larger-scale acute
injuries accompanied by inflammation.
Tenosynovitis/paratendinitis/peritendinitis/tenovaginitis.
These terms all refer to injuries of the outer layers of tendons
with fluid of unknown cause.
Ganglion cysts. Ganglion cysts are small pouches from the joint
capsule if they are found around a joint, or tendon lining tissue
if they are found near a tendon. In general, ganglion cysts
consist of an outer fibrous coat and an inner lining, and contain
a clear, colourless, gelatinous fluid. Ganglion cysts can be seen
after injuries but are also seen without any known cause.
Ganglion cysts are most conspicuous on STIR, FS PD and T2*
sequences and appear as focal areas of increased signal
intensity.
Synovitis or bleeding. Findings of fluid of unknown nature in
the joint.
Oedema. Findings of fluid in various soft tissue places as
opposed to the oedema found in bone (defined as bone bruise).
Treatment
The following treatment protocol was agreed upon with
Consultants in Hand, Plastic and Orthopedic Surgery at Haukeland
University Hospital, Bergen, Norway ahead of the study, as a
response to possible pathological MRI findings in patients with
injured wrists:
Fractures. Immobilising in a cast for 3–6 weeks depending on
the identified fracture and its location.
Bone bruises. Immobilising in a supportive bandage for a few
days until 2 weeks or in a cast for 2 weeks if pain was
pronounced.
TFCC ligament ruptures. Immobilising in plaster above the
elbow with the wrist in neutral rotation for 4 weeks followed by
a low cast for 2 weeks; referral to a hand surgeon for
consideration of operative intervention.
TFCC ligament partial rupture. Immobilising in a cast for 2
weeks; if persistent pain or clinical instability of the distal
radioulnar (DRU) joint, referral to a hand surgeon for
consideration of operative intervention.
Intercarpal ligament lesions without dissociation. Immobi-
lising in a cast for 2 weeks; if still pain after 2 weeks, referral to a
hand surgeon.
Tendon lesions and other soft-tissue injuries. Symptomatic
treatment with or without a supportive bandage and early
mobilisation.
The data was analysed by the use of Statistical Package for the
Social Sciences (SPSS) (version 18, PASW; IBM, Armonk, NY, USA).
The study was approved by the Norwegian Ethical Committee for
Medical Research.
Table 1
Registration form for all patients that went into the study.
1 Patient ID number ————
2 Date of birth; (dd. mm. yy) ————
3 MR taken 0. no 1. yes
4 MR pathology 0. no 1.1 finding 2.2 findings 3.3 f. 4.5.6.7.8. findings
5 Scaphoid 0. neg 1. fracture 2. bonebruise
6 Scaphoid localization 0. neg 1. prox 2. waist 3. distal 4. combination
7 Lunate 0. neg 1. fracture 2. bonebruise 3. impaction syndrome
8 Triquetrum 0. neg 1. fracture 2. bonebruise
9 Pisiforme 0. neg 1. fracture 2. bonebruise
10 Trapezium 0. neg 1. fracture 2. bonebruise
11 Trapezoid 0. neg 1. fracture 2. bonebruise
12 Capitate 0. neg 1. fracture 2. bonebruise
13 Hamate 0. neg 1. fracture 2. bonebruise
14 Metacarp 1 0. neg 1. fracture 2. bonebruise
15 Metacarp 2 0. neg 1. fracture 2. bonebruise
16 Metacarp 3 0. neg 1. fracture 2. bonebruise
17 Metacarp 4 0. neg 1. fracture 2. bonebruise
18 Metacarp 5 0. neg 1. fracture 2. bonebruise
19 Radius 0. neg 1. fracture 2. bonebruise
20 Ulna 0. neg 1. fracture 2. bonebruise 3. sublux 4. sublux + fracture
21 TFCC injury 0. no 1. yes
22 TFCC localization 0. neg 1. disc 2. ulnocarpallig 3. volar RUL 4. dorsal RUL 5. other
23 Intercarpal lig injury 0. neg 1. total rupture 2. partial rupture
24 Intercarp lig localization 0. neg 1. SL lig 2. LT lig 3. other
25 ECU tendon 0. neg 1. total rupture 2. partial rupture 3. post traumatic tenosynovitis
26 Other tendons 0. neg 1. total rupture 2. partial rupture 3. post traumatic tenosynovitis
27 Muscle contusions 0. neg 1. thenar 2. other
28 Ganglion uni-loculated 0. neg 1. volar 2. intercarpal 3. dorsal 4. many locations
29 Ganglion multi-loculated 0. neg 1. volart 2. intercarpal 3. dorsal 4. many locations
30 Synovitis 0. neg 1. DRUJ 2. intercarpal 3. radiocarpal 4. many locations
31 Edema 0. no 1. yes
32 Incidental finding 0. none 1. Enchondroma 2. Bone cyst 3. other
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1734

Results
A total of 155 patients were consecutively included during the
1-year study period. Median age of the patients was 28 years
(range: 18–49 years), 85 men and 70 women. The right wrists were
injured in 85 patients and the left in 70. The majority of the injuries
were caused by falls.
According to our definition, we found that 80% of the women
and 38% of the men acquired their wrist injury after a low-energy
trauma (p= 0.0005), but we did not find any significant difference
in the percentage of fractures between the patients with high- and
low-energy trauma or between men and women (p= ns).
Thirty of the 155 MRIs were normal. Thus, 125 patients, four out
of five, had a total of 303 pathological MRI findings according to our
definitions. All the different pathological MRI findings were
registered (Table 1). A median of 2 (range: 0–8) pathological
findings per patient were found, including 54 fractures, 56 bone
bruises, 15 Triangular TFCC injuries and five scapho-lunate (SL)
ligament lesions (Tables 2 and 3 and Fig. 2).
Fractures and bone bruises
Twenty-eight percent (44 out of 155) of the patients had a total
of 54 fractures (Table 2). Fractures within the radius and scaphoid
represent half of all fractures. More than 90% of the fractures of the
carpal bones were located to the scaphoid, the triquetrum or the
capitate (Fig. 3). The 13 scaphoid fractures were localised in the
following parts of the bone: proximal (n= 2), waist (n= 5), distal
(n= 3) and a combination of these (n= 3). The vast majority of
occult fractures found with MRI was undisplaced, such as the
scaphoid fractures. The trapezium fracture and the trapezoid
fracture were both avulsion fractures. Of the six triquetral
fractures, five were undisplaced and one was an avulsion fracture.
Three of the four capitate fractures were undisplaced and one was
a comminuted intra-articular fracture in the distal part of the
capitate and with a step in the articulate surface of 2 mm. All of the
14 radius fractures were undisplaced. Eight of the 14 fractures
were in the metaphysis and seven were intra-articular. One of the
ulnar fractures was an avulsion fracture at the tip of the ulnar
styloid and the other was a comminute, but undisplaced fracture in
the metaphyseal part of the ulna at the base of the ulnar styloid. All
13 metacarpal fractures were undisplaced, but six of them were
intra-articular.
Twenty-two percent (33 out of 155) of the patients had a total of
56 bone bruises. A bone bruise was found in all of the eight carpal
bones in addition to the radius, the ulna and the metacarpal bones
(Table 2 and Fig. 4). The bone bruise of the scaphoid, the capitate
and the lunate represented nearly 60% of the carpal bone bruises.
The lunate bone had no fractures, but six bone bruises, represent-
ing 15% of the carpal bone bruises.
Soft-tissue injuries
The distribution of the pathological soft-tissue findings is
shown in Table 3.
Ligaments
We found 15 TFCC injuries, four in the volar radioulnar
ligament, three in the membranous disc and eight with more than
one localisation (Fig. 5). There were no dislocations in the DRU
joint due to TFCC tears. In three cases of tears to the volar
radioulnar ligament, we noticed a subtle dorsal subluxation, which
may be caused by an offset image as it did not cause any long-term
problems.
The five inter-carpal ligament lesions were all partial SL
ligament tears. In four of these, MRI showed a ganglion with
connection to the SL joint (Fig. 6). All SL-ligament lesions were
partial ligament tears: four of them without signs of widened
distance between the scaphoid and the lunate on MRI. One case
had a small widening less than 3 mm.
Tendons
A total of 13 patients had partial tendon rupture or
tenosynovitis according to our definitions. Four patients had
Table 2
Distribution of MRI findings in the bone in 155 patients with normal X-rays after a
wrist sprain.
Pathological findings
Fracture Bone bruise
Proximal bones
Radius 14 7
Ulna 2 2
Carpal bones
Scaphoid 13 9
Lunate 0 6
Triquetrum 6 3
Pisiforme 0 3
Trapezium 1 4
Trapezoid 1 3
Capitate 4 8
Hamate 0 3
Distal bones
Metacarpals 13 8
Total 54 56
Table 3
Distribution of MRI findings in the soft tissues, synovitis and oedema in 155 patients
with normal X-rays after a wrist sprain.
Identified structure Localisation Pathological findings
TFCC Membranous disc 3
Palmar RUL 4
Other 8
Inter-carpal ligament tear SL-partial tear 5
Tendon Partial rupture 7
Tenosynovitis 6
Muscular contusions 8
Ganglion Uni-locular 23
Multi-locular 8
Synovitis 69
Oedema 38
Abbreviations: TFCC, triangular fibro-cartilage complex; RUL, radio-ulnar ligament;
SL, scapho-lunate ligament.
[(Fig._2)TD$FIG]
Fig. 2. Almost all part of the wrist had some sort of pathology when MRI was used in
patients with sprained wrist and negative X-ray.
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1735

partial rupture in the extensor carpi ulnaris (ECU) and one of
them also had a tenosynovitis in ECU (Fig. 7). Five patients
had tenosynovitis in both extensor carpi radialis brevis (ECRB)
and extensor carpi radialis longus (ECRL), and one of these
patients had an additional tenosynovitis in the extensor
digitorum communis. Three patients had a partial rupture in
the extensor pollicis longus (EPL) and one of them had an
additional partial rupture in the extensor pollicis brevis (EPB).
One patient had tenosynovitis in the flexor carpi radialis (FCR).
Probably, these findings most likely represent tracking of
blood along the tendons caused by the sprain, rather than
the paratendinous fluid typically seen with inflammatory
tenosynovitis.
Muscles
Seven patients had contusions of the thenar muscle and one
patient had contusion in the distal part of the musculus extensor
digitorum (Fig. 8).
Ganglions
There were 23 uni-loculated ganglions in the wrist, 13 located
volarly and 10 dorsally (Fig. 9). Fifteen of the uni-loculated
ganglions had relation to the SL joint. In three of these wrists, MRI
showed injury of the SL ligament. Two of the uni-locular ganglions
were located near the ulnar styloid, in which MRI further indicated
injury to the TFCC.
We found eight multi-loculated ganglions. Seven of them were
located volarly and one dorsally.
Synovitis
Synovitis, according to our definition, was identified in 69
wrists and was located in the following areas: 36 inter-carpal, 10 in
the DRU joint and 23 in more than one localisation (Fig. 8). These
findings probably represent bleeding in the joints rather than
synovitis fluid secondary to an inflammatory condition.
Oedema
We found soft-tissue oedema in 38 wrists (Fig. 10).
Treatment
In this study, we pre-empted all potential findings and agreed
with the hand surgeons at Haukeland University Hospital in
Bergen regarding treatment. As the pathology was only visible on
MRI, and not on X-rays, we followed a Norwegian tradition of
conservative management in most of these injuries.
Nearly all patients were treated conservatively. Seven patients
with TFCC injuries or SL-ligament injuries were referred to a
specialist and evaluated for further treatment. Two of these seven
patients underwent arthroscopy and suture of the torn TFCC.
Discussion
This prospective 1-year study, from a high-volume A&E centre
in Norway, has revealed that four out of five patients with acute
[(Fig._3)TD$FIG]
Fig. 3. Occult fracture in the scaphoid and bone bruise of the capitate shown with MRI (Pat id 68).
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1736

wrist trauma and negative X-ray had pathological findings on
MRI, including occult fractures, bone bruise, SL ruptures and
TFCC injuries. Up until now, a wrist sprain has been defined as a
partial ligament tear with a normal X-ray.
5
However, we found
that only 13% of our patients had partial ligament tears defined
as TFCC injuries or other ligament tears. Consequently, the
clinical and radiological investigations fail to adequately diag-
nose the complete pathology in wrist sprains. Alternatively, the
diagnosis and definition of a sprained wrist are incomplete and
inaccurate according to the present definition
5
and ought to be
changed.
It has previously been demonstrated that MRI is a sensitive
method of prompt assessment of both bone and ligament injuries
in the wrist in young adults when the initial X-ray is negative.
3
This
corresponds well with our experience. We also find, like others,
MRI to be a good diagnostic tool to identify injuries associated with
wrist sprain.
4
Proper MRI modalities enable simultaneous diagno-
sis of the soft tissue-, ligament- and bone injuries.
13
An experienced musculoskeletal radiologist did the MRI
analyses in our study. Previous studies have shown that correct
interpretation of the MRI images strongly depends on the
experience of the radiologist. Ample experience is needed to
distinguish between swelling and oedema, micro-fractures and
incomplete/complete fractures or complete fractures that are
undisplaced.
13
Inclusion criteria for this study were patients with a
wrist injury and a negative X-ray.
[(Fig._4)TD$FIG]
Fig. 4. MRI showing bone bruise of the MC 2–4, the hamate, the capitate, the trapezium and the radius as well as an occult fracture in the scaphoid (Pat id 145).
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1737

Our study is accompanied with some limitations. At times, the
activity in the A&E was extremely busy to the point where we
probably have lost patients who potentially could have been
included in this study. We do not believe that this would have
changed the outcome of our study.
Furthermore, all patients were clinically examined by
the doctors on call before X-ray investigation. The doctors
had various degrees of clinical and radiological experience. The
X-rays were interpreted by the same doctor together with the
radiographer on duty in accordance with current practice.
We chose not to revise the X-ray interpretations done by our
doctors on duty to keep the study design as close to ordinary
practice as possible, as this will better reflect reality. There is
therefore a small risk that more pathology could have been
detected if the X-rays would have been revised by experienced
doctors.
Five of the MRIs were done at a different institute with a
different MRI machine. In spite of that fact, our radiologist accepted
the quality of the images and found them acceptable to be included
in our study. We considered excluding these cases, but we believe
that it will not make a difference to the final outcome to keep them
in our material as presented.
[(Fig._5)TD$FIG]
Fig. 5. MRI showing a TFCC injury, volar RU ligament, close to the fovea of the ulnar head as well as synovitis in the DRU joint (Pat id. 121).
[(Fig._6)TD$FIG]
Fig. 6. MRI showing a scapho-lunate (SL) injury (Pat id 175).
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1738

[(Fig._7)TD$FIG]
Fig. 7. MRI showing a partial rupture of the extensor carpi ulnaris (ECU) tendon (Pat id 161).
[(Fig._8)TD$FIG]
Fig. 8. MRI showing a contusion to the thenar muscle and synovitis in carpal joints (Pat id 163). Synovitis may in fact represent bleeding in the joint rather than fluid second to
an inflammatory condition.
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1739

There are no specific differentiating features on MRI examina-
tions separating a traumatically induced tear on the TFCC from one
caused by degeneration.
9
However, the age of the patient, the site
of the tear, associated lesions and the absence of secondary
changes in the lunate and/or ulnar head helped us in this regard.
We still agree that we could not prove for sure that all TFCC tears
were caused by the trauma. Thus, determining the clinical
relevance of these lesions and their correlation with the patient’s
symptoms were difficult.
The main problem with this study was to decide how the
pathological findings should be managed. Many of our findings
have an unknown clinical relevance. We therefore implemented a
treatment protocol ahead of the study. In the absence of consensus
of treatment in the literature, we agreed to a reasonable
symptomatic treatment for each diagnosed pathological finding.
We integrated local experience from the University hospital
with the best current evidence found in the limited literature.
Long-term outcome studies are needed to find the best clinical
practice for these previously unknown injuries. We found MRI
pathology in 80% of the patients. This is far more than what was
found in a retrospective cohort study of patients with a clinically
suspected scaphoid fracture with normal X-ray.
14
They found
abnormal MRI findings in 41% (266 out of 651) of the wrists, in spite
of a normal X-ray.
14
In another study from the same centre, they
found 44% positive MRI findings out of 378 patients with clinical
suspicion of underlying injury in the wrist and negative X-ray.
4
This
high proportionof abnormalities discoveredin our study might both
be due to the good technical quality of the MRI modalities and the
number of sequences used, and also the interpretations of an
experienced radiologist. Some of the difference may also be due to
over-reporting pathology as seen in MRI investigations of both
adults and children with asymptomatic wrists.
15,16
In accordance with other trauma-related injuries described in
the literature, we found that women acquire their wrist injuries
after low-energy trauma and men after high-energy trauma.
Despite the use of the accepted definition of high- or low-energy
trauma, falling injuries in sports does not necessarily involve more
energy than during walking on slippery surfaces. Our findings
possibly reflect that even in a modern society, men tend to be
physically more active and careless in sports than women.
Bone
The distribution of radiographically diagnosed fractures in
wrist injuries has shown a majority of scaphoid fractures (58–
61%),
17,18
followed by triquetrum fractures (29–32%). By the use of
MRI in radiographically normal wrists, we also found the scaphoid
to be the most frequently affected bone with 52% of all occult
carpal fractures. This is in accordance with earlier studies with
normal X-rays and suspicion of underlying wrist injuries, which
have shown approximately 25% occult fractures on MRI. They also
found that scaphoid fractures made up most of the occult fractures
in the carpus.
2,4,14
Scaphoid pathology was also noted in 23% of all
bone bruises affecting the carpal bones (Table 2). This probably
reflects that we usually fall on the outstretched hand with the
[(Fig._9)TD$FIG]
Fig. 9. MRI showing a uni-loculated ganglion cyst volar of the scapho-lunate (SL)
articulation (Pat id 60).
[(Fig._10)TD$FIG]
Fig. 10. MRI showing oedema and tenosynovitis (Pat id 152). Tenosynovitis may in fact represent bleeding along the tendon rather than fluid second to an inflammatory
condition.
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1740

forearm in a pronated position. Interestingly, more than 90% of the
occult fractures of the carpal bones were located to the scaphoid,
the triquetrum or the capitate. This finding, combined with the fact
that fractures in the radius or the scaphoid represented half of all
MRI identified fractures, may imply that many wrist sprains, in
fact, are a part of what Mayfield described as the full-blown
perilunate dislocation.
19,20
By combining the radius, the scaphoid,
the capitate and the triquetrum which are the most affected bones
seen on MRI in our study, the greater arch mechanism for the
perilunate dislocation can be visualised, but in this simpler form
without the final perilunate dislocation. Thus, we believe that our
findings support the Mayfield theory of how the energy
sequentially passes through the wrist causing these injuries to
develop. In Mayfield’s original article, high-energy injuries cause a
full-blown greater arch perilunate dislocation, while we, in lower-
energy, radiographically invisible, but MRI-visible injuries, find
that the lesions follow the same arch.
Bone bruise
We found one or more bone bruises in 22% of the wrists. The
bone bruises of the scaphoid, capitate and lunate represented
nearly 60% of the carpal-bone bruises. An earlier study examined
125 X-ray negative wrists with MRI. They found bone bruises in 49
(39%) of the wrists, mainly in the scaphoid, lunate and triquetrum.
2
An interesting observation is that we did not find any fractures in
the lunate, but six bone bruises here (15% of all the carpal bone
bruises). These results are in accordance with a previous X-ray
study, where no lunate fracture was identified
18
. The lunate is a
crescent-shaped bone forming part of the wrist and is anatomically
located in the middle of the carpus. It does not function as a lever,
contrary to the scaphoid,
17
and might therefore be less exposed to
fracture-inducing energy. This is in line with our previous
suggestion of the Mayfield mechanism, where the lunate is
central, almost protected by the other carpal bones, where most of
the energy passes. Possibly, due to the impact, the lunate bone
bruise explains that there is a direct impact to the bone itself after
all, but because of the configuration of the bone, it maintains its
shape and does not fracture. This is further supported by the fact
that in a full-blown perilunate dislocation, being a lesser or greater
arch mechanism, the lunate is usually intact on X-rays. Perhaps an
MRI in these cases would prove that there is bone bruise as part of
the trauma mechanism.
A capitate fracture is a rare radiographic finding (0.5
17
–1.4%
18
).
We found it to be more frequent using MRI for diagnosis, with four
occult fractures (16% of all occult carpal bone fractures) and eight
bone bruises (21% of all carpal bone bruise). Avascular necrosis of
the capitate has been described as a rare condition with an
unknown aetiology,
21
but perhaps occult fracture or bone bruise is
an important factor in the pathogenesis of avascular necrosis of the
capitate. Consequently, if capitate fractures are identified and thus
immobilised, we might prevent the development of avascular
necrosis. Further studies are needed to confirm this theory.
The same theory is applicable to the pathogenesis of Kienbock’s
disease (avascular necrosis of the lunate), which has not been
identified yet. Mechanical factors such as acute trauma or
repetitive minor trauma have been suggested not to be the
primary cause.
22
We think that our findings of bone bruise in the
lunate may be of importance in understanding one possible cause
of Kienbock’s disease.
Soft tissue
The definition of soft-tissue injuries varies in different studies.
In our study, we included TFCC injuries, ligament tears, tendon
injuries, tenosynovitis, muscular contusions and ganglions as
soft-tissue injuries. According to this definition, we found 73 soft-
tissue injuries, 69 findings of synovitis and 38 cases of oedema. In
one study of patients with clinical suspicion of wrist fractures and
negative X-ray, MRI showed soft-tissue injuries in 20 of 36 wrists
with no fractures.
23
In a retrospective cohort of patients presenting
with symptoms and signs suggestive of scaphoid injury and
normal initial radiographs, the 651 MRIs identified 33% with
fracture or bone bruise, 8% with soft-tissue injuries and 59% were
normal. TFCC injuries made up 31% and SL-ligament injuries made
up 15% of the total soft-tissue injuries.
14
Ganglion
Ganglia are usually located dorsally in the wrist, roughly
representing 2/3 of all ganglia of the hand. In contrast to this
previous knowledge, we found 23 uni-loculated ganglions in the
wrist, 13 located volarly and 10 dorsally. This pattern different
from the normal presentation in the population might reflect that
these uni-loculated ganglia are caused by a rupture in the volar
capsule, forming these ganglia. The suggested traumatic cause of
some of these ganglia is further supported by the fact that 15 of the
uni-loculated ganglia had a connection to the SL joint, which in
three cases had an MRI showing partial SL-ligament tears. This is
further supported by the fact that three of our five SL-ligament
tears also had concomitant adjacent uni-loculated ganglia. Similar
findings of uni-loculated ganglia around the ulnar styloid were
found in association with TFCC injuries.
Ligament injuries
We found 15 injuries to the TFCC and five SL ligament lesions,
representing 13% of our patients. These findings are in accordance
with another study where 10% of patients with negative X-ray and
strong clinical suspicion of underlying injury had evidence of
ligamentous injury, also identified by MRI.
4
We suspect that we
still may fail to identify some of the SL injuries as MRI still is not
particularly sensitive to pick up these injuries.
Treatment
We found it useful to have a treatment protocol outlined ahead
of the study. In this way, the doctors were prepared once they
encountered previously unknown injuries. The doctors appre-
ciated this. Both doctors and patients expressed satisfaction with
an early and more precise diagnosis. The MRI findings led to a
change in treatment in more than one-third of the patients. This is
in agreement with other studies where wrist MRI had therapeutic
consequences for the patients in respectively 37 of 56 (66%)
wrists
23
and 45 of 98 (46%) wrists.
24
In contrast to another study,
1
we discharged most of our patients with a normal MRI. All of the
scaphoid fractures were occult, undisplaced fractures. They were
immobilised in a cast for 4–9 weeks depending on the location of
the fracture and the progress of healing as diagnosed with clinical
examination (localised pain) and, in some patients, with delayed X-
ray findings. All of the fractures healed with conservative
treatment. None of the patients returned with problems within
the following year. It is reassuring to find that our assumption that
occult undisplaced fractures do not need surgical intervention was
supported by our findings. We find no reason to consider
percutaneous screw fixation of these occult stable scaphoid
fractures. This is further supported by a recently published
meta-analysis of randomised controlled trials of acute, minimally
displaced and undisplaced scaphoid waist fractures. This analysis
did not support routine surgical treatment, and concluded that
aggressive conservative management should remain the mainstay
for scaphoid waist fractures.
25
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1741

The other occult fractures were treated with a plaster cast for 3–
5 weeks depending on the identified fracture and its location.
According to our pre-study guidelines, the decision of referring the
patient to the local hand specialist at the University hospital for an
operative treatment was made in the light of the clinical
examination of the patient combined with the MRI findings. Thus,
seven patients with TFCC injuries or SL-ligament injuries were
referred for further evaluation or treatment. Only two of these
seven patients had clinical symptoms which lead to arthroscopy
and suture of the peripheral TFCC injury.
Even though it is likely that repair of torn ligaments would
produce a better outcome than a late reconstruction, no study has
yet clearly demonstrated the benefits of early intervention, or even
if surgical repair at any stage is preferable to conservative
treatment.
7
We find that our conservative approach on these
difficult injuries seems to have been correct.
We changed the treatment for patients with occult fractures,
TFCC injuries and SL-ligament tears according to an agreed
protocol with the local hand unit at our University hospital. As
the consequences of these, previously unknown, injuries were not
possible to predict, we found it necessary to increase the
conservative management of these, so-called, sprained wrists.
We immobilised them in a supportive bandage or a cast for 2 weeks
if pain was pronounced. Fractures and ligament injuries diagnosed
by MRI alone, in the absence of X-ray findings, were treated in a
plaster cast for 2–6 (9) weeks. This treatment was adequate for all
patients with ‘occult’ scaphoid fractures. Only two of the patients
with TFCC-ligament tears needed operative treatment.
Conclusion
We found that four out of five patients with acute wrist trauma
and negative X-ray had pathological findings on MRI. Acute MRI
detects a large number of injuries not seen on X-rays, which is in
accordance with the view that a more complete imaging analysis of
wrist sprains is needed.
5
The current definition of a wrist sprain
being a partial ligament injury of the wrist, as defined by the
IWIW,
5
ought to be changed. We suggest that a wrist sprain should
be defined as an ‘‘occult partial or complete soft tissue (ligament,
tendon, muscle) or bony injury in relation to a trauma with
negative X-ray’’. Based on this study, we recommend early MRI of
acute wrist sprains with pain or functional symptoms that does not
settle within 2 weeks. This will provide a more specific diagnosis,
which hopefully will lead to more appropriate treatment.
Conflict of interest statement
Research grants have been received from The University of
Bergen and The Norwegian Research Council.
Acknowledgements
The authors acknowledge Frank van Betten, Arve Strandenes
and the remaining staff at Bergen A&E for their support. Leiv Hove,
Yngvar Krukhaug, Eivind Strandenes and Rakel Gudmundsdottir,
Colleagues at Bergen Hand Surgery Center are greatly acknowl-
edged. For internal radiographic cross-check we thank: Eivind
Strøm, Svein Halvorsen, Per M Kristoffersen at Department of
Radiology, Haukeland University Hospital, Bergen, Norway.
References
1. Nikken JJ, Oei EH, Ginai AZ, Krestin GP, Verhaar JA, van Vugt AB, et al. Acute wrist
trauma: value of a short dedicated extremity MR imaging examination in
prediction of need for treatment. Radiology 2005;234:116–24.
2. Pierre-Jerome C, Moncayo V, Albastaki U, Terk MR. Multiple occult wrist bone
injuries and joint effusions: prevalence and distribution on MRI. Emergency
Radiology 2010;17:179–84.
3. Fotiadou A, Patel A, Morgan T, Karantanas AH. Wrist injuries in young adults:
the diagnostic impact of CT and MRI. European Journal of Radiology 2011;77:
235–9.
4. Sinha R, Smith FW. Large series vindicates early MRI for the diagnosis of wrist
pain following trauma. Injury Extra 2007;38:277–88.
5. Gilula LA, Mann FA, Dobyns JH, Yin Y. Wrist terminology as defined by the
International Wrist Investigators’ Workshop (IWIW). The Journal of Bone & Joint
Surgery 2002;84:1–66.
6. Haugstvedt J-R. Ha
˚nd- og ha
˚ndleddsskader. In: Bahr R, Mæhlum S, editors.
Idrettsskader. Bergen: Fagbokforlaget; 2010. p. 211–46.
7. Adolfsson L, Povlsen B. Arthroscopic findings in wrists with severe post-trau-
matic pain despite normal standard radiographs. Journal of Hand Surgery-British
and European Volume 2004;29:208–13.
8. Coggins CA. Imaging of ulnar-sided wrist pain. Clinics in Sports Medicine
2006;25:505–26. vii.
9. Zlatkin MB, Rosner J. MR imaging of ligaments and triangular fibrocartilage
complex of the wrist. Magnetic Resonance Imaging Clinics of North America
2004;12(301–31):vi–i.
10. Magee T. Comparison of 3-T MRI and arthroscopy of intrinsic wrist ligament
and TFCC tears. AJR American Journal of Roentgenology 2009;192:80–5.
11. Brydie A, Raby N. Early MRI in the management of clinical scaphoid fracture. The
British Journal of Radiology 2003;76:296–300.
12. Hobby JL, Tom BD, Bearcroft PW, Dixon AK. Magnetic resonance imaging of the
wrist: diagnostic performance statistics. Clinical Radiology 2001;56:50–7.
13. Rhemrev SJ, Ootes D, Beeres FJ, Meylaerts SA, Schipper IB. Current methods of
diagnosis and treatment of scaphoid fractures. International Journal of Emer-
gency Medicine 2011;4:4.
14. McCullough NP, Smith FW, Cooper JG. Early MRI in the management of the
clinical scaphoid fracture. European Journal of Emergency Medicine Official
Journal of the European Society for Emergency Medicine 2011;18:133–6.
15. Ejbjerg B, Narvestad E, Rostrup E, Szkudlarek M, Jacobsen S, Thomsen HS, et al.
Magnetic resonance imaging of wrist and finger joints in healthy subjects
occasionally shows changes resembling erosions and synovitis as seen in
rheumatoid arthritis. Arthritis and Rheumatism 2004;50:1097–106.
16. Muller LS, Avenarius D, Damasio B, Eldevik OP, Malattia C, Lambot-Juhan K, et al.
The paediatric wrist revisited: redefining MR findings in healthy children.
Annals of the Rheumatic Diseases 2011;70:605–10.
17. Hove LM. Fractures of the hand. Distribution and relative incidence. Scandina-
vian Journal of Plastic and Reconstructive Surgery and Hand Surgery Nordisk
plastikkirurgisk forening [and] Nordisk klubb for handkirurgi1993;27:317–9.
18. van Onselen EB, Karim RB, Hage JJ, Ritt MJ. Prevalence and distribution of hand
fractures. Journal of Hand Surgery-British and European Volume 2003;28:491–5.
19. Mayfield JK. Mechanism of carpal injuries. Clinical Orthopaedics and Related
Research 1980:45–54.
20. Mayfield JK, Johnson RP, Kilcoyne RK. Carpal dislocations: pathomechanics and
progressive perilunar instability. Journal of Hand Surgery-American Volume
1980;5:226–41.
21. Toker S, Ozer K. Avascular necrosis of the capitate. Orthopedics 2010;33:850.
22. Irisarri C. Aetiology of Kienbock’s Disease Handchirurgie Mikrochirurgie plastische
Chirurgie Organ der Deutschsprachigen Arbeitsgemeinschaft fur Handchirurgie
Organ der Deutschsprachigen Arbeitsgemeinschaft fur Mikrochirurgie der Periphe-
ren Nerven und Gefasse Organ der Vereinigung der Deutschen Plastischen Chirur-
gen2010;42:157–61.
23. Mack MG, Keim S, Balzer JO, Schwarz W, Hochmuth K, Windolf J, et al. Clinical
impact of MRI in acute wrist fractures. European Radiology 2003;13:612–7.
24. Hobby JL, Dixon AK, Bearcroft PW, Tom BD, Lomas DJ, Rushton N, et al. MR
imaging of the wrist: effect on clinical diagnosis and patient care. Radiology
2001;220:589–93.
25. Ibrahim T, Qureshi A, Sutton AJ, Dias JJ. Surgical versus nonsurgical treatment of
acute minimally displaced and undisplaced scaphoid waist fractures: pairwise
and network meta-analyses of randomized controlled trials. Journal of Hand
Surgery-American Volume 2011;36:1759–68. e1.
T.H. Bergh et al. / Injury, Int. J. Care Injured 43 (2012) 1732–1742
1742