This document is a doctoral thesis that examines total wrist arthroplasty (TWA), patient-rated outcome measures related to TWA, and periprosthetic osteolysis following TWA. The thesis includes studies using a multicenter international registry of TWA patients, radiographical and histopathological evaluations, and validation of translated patient-rated outcome measures. The overall aim is to evaluate outcomes of TWA, assess patient-rated outcomes, and investigate periprosthetic osteolysis.

1. Total
Wrist
Arthroplasty:
outcomes,
patient
rated
outcome
measures
and
periprosthetic
osteolysis.
Michel
E.
H.
Boeckstyns
Doctoral
Thesis
University
of
Copenhagen
Gentofte
Hospital,
Clinic
for
Hand
Surgery

2. 2
Denne afhandling er af Det Sundhedsvidenskabelige Fakultet ved Københavns Universitet antaget til
offentligt at forsvares for den medicinske doktorgrad.
København, den 21/5 2015
Professor Ulla Wewer
dekan
Forsvaret finder sted fredag den 12. juni 2015, kl. 14 i Hannover Auditoriet, Panum, Blegdamsvej 2B,
2200 København N
Officielle opponenter:
Professor Leiv M. Hove, Bergen Universitet, Norge
Professor Torben Bæk Hansen, Aarhus Universitet.
Copyright © 2015 Michel E. H. Boeckstyns
All rights reserved
Printed in Denmark
Eget forlag
ISBN 978-87-998283-0-2
Address for correspondence:
mibo@dadlnet.dk

3. 3
Tak
til…
…
først
og
fremmest
min
gode
ven
og
kollega
Søren
Merser,
som
har
ydet
mig
en
utroligt
uselvisk
og
kvalificeret
hjælp
og
rådgivning,
ikke
mindst
i
det
statistiske
…
Iben,
som
har
udvist
stor
tålmodighed
…
Stig
Sonne-­‐Holm
for
hans
velvillige
og
kvalificerede
kommentarer
…
Gentofte
Hospital
for
at
give
mig
frihed
til
at
lave
arbejdet
…
Guillaume
Herzberg
for
det
værdifulde
samarbejde
…
de
øvrige
medarbejdere
i
mine
projekter,
uden
hvem
de
ikke
ville
være
blevet
fuldført:
Allan
Ibsen
Sørensen,
Alex
Herup,
Anders
Toxværd,
Karsten
Krøner,
Lars
Soelberg
Vadstrup,
Laurent
Obert,
Manjula
Bansal,
Peter
Axelsson,
Philippe
Liverneaux.
I
express
my
gratitude
to
Søren
Merser
for
his
invaluable
help
…
my
wife
Iben
for
being
so
patient
…
Stig
Sonne-­‐Holm
for
his
helpfulness
…
Gentofte
Hospital
for
giving
me
the
possibility
to
do
this
work
…
Guillaume
Herzberg
for
his
invaluable
contributions
…
all
my
other
co-­‐workers
in
this
project:
Allan
Ibsen
Sørensen,
Alex
Herup,
Anders
Toxværd,
Karsten
Krøner,
Lars
Soelberg
Vadstrup,
Laurent
Obert,
Manjula
Bansal,
Peter
Axelsson,
Philippe
Liverneaux.

4. 4
Contents
List
of
terms
and
definitions
5
Abbreviations
11
List
of
papers
13
Introduction
and
Background
14
1.
Historical
background
and
current
issues
concerning
Total
Wrist
Arthroplasty
(TWA).
14
2.
Patient-­‐rated
Outcome
measures
(PROMs)
and
TWA
16
Aims
of
the
Thesis
17
Initiatives
17
Methods
and
Methodological
Considerations
19
1.
The
PRISMA
statements
19
2.
The
multicenter
international
Re-­‐motion
registry
22
3.
The
cumulated
implant
survival
25
4.
Radiographical
measurements
26
5.
Histopathological
evaluations
27
6.
Translation
of
PROMs
28
7.
Validation
of
PROMs
28
Results
30
General
Discussion
and
Comparison
with
Other
Research
49
Review
of
the
literature
49
The
multicentre
international
Re-­‐motion
registry.
52
Periprosthetic
osteolysis
53
Validation
of
PROMs
61
Future
perspectives
63
Summary
and
Conclusions
of
the
Thesis
65
Dansk
resume
og
konklusioner
67
References
70
Paper
I
76
Paper
II
85
Paper
III
89
Paper
IV
95
Paper
V
100
Paper
VI
106
Paper
VII
115
Paper
VIII
119

5. 5
List
of
terms
and
definitions
Anchor
based
method:
method
that
uses
some
external
anchor,
such
as
patient
judgments
of
change,
which
is
then
used
to
compute
a
minimal
clinically
important
difference
(MCID).
Cohen’s
kappa:
Cohen's
kappa
coefficient
is
a
statistical
measure
of
inter-­‐rater
agreement
for
qualitative
(categorical)
items.
Condition-­‐specific
questionnaire:
addresses
specific
conditions,
like
carpal
tunnel
syndrome,
Dupuytren’s
disease,
distal
radius
fractures
etc.
Construct:
A
well-­‐defined
and
precisely
demarcated
subject
of
measurement
Construct
validity:
The
degree
to
which
the
scores
of
a
scale
are
consistent
with
a
priori
hypotheses
concerning
the
construct
to
be
measured.
Content
validity:
Content
validity
is
the
extent
to
which
an
outcome
measure
instrument
appears
to
measure
all
facets
of
what
it
was
intended
to
measure.
Criterion
validity:
Criterion
validation
assesses
how
a
person
who
scores
at
a
certain
level
on
a
scale
does
on
a
gold
standard
or
some
other
validated
criterion
measure.
Cronbach's
alpha:
a
coefficient
of
internal
consistency
that
indicates
the
intercorrelation
among
test
items.
Scales
are
considered
to
be
internally
consistent
if
Cronbach’s
alpha
is
between
0.7
and
0.9
1.
Cronbach’s
alpha
in
excess
of
0.9
suggests
possible
redundancy
in
the
questionnaire.
DASH:
Disabilities
of
Arm,
Shoulder
and
Hand
(Handicaps
i
Arm,
Skulder
og
Hånd).
Generic
patient
rated
outcome
measure
of
the
upper-­‐extremity
2.
Domain:
a
sub-­‐score
within
a
questionnaire
meant
to
cover
a
specific
condition
of
interest,

6. 6
e.g.
motion,
pain,
strength
etc.
Floor
and
ceiling
effect:
the
floor
and
ceiling
effects
show
the
proportion
of
individuals
who
achieve
the
highest
or
lowest
possible
numeric
value
of
a
score
and
are
considered
present
when
more
than
15%
of
the
individuals
achieve
these
values.
3
4
A
ceiling
or
floor
effect
indicates
that
the
measurement
instrument
cannot
be
used
for
the
entire
continuum
of
patients
seen.
General
outcome
assessment
(GOA):
assessment
of
outcome
made
by
an
observer
–
e.g.
a
surgeon
or
hand
therapist
–
on
the
basis
of
physical
tests,
like
measuring
motion,
stability
etc.
=
evaluation
from
the
clinician’s
perspective.
Generic
questionnaire:
questionnaire
addressing
more
general
health
concerns.
E.g.:
DASH
addresses
general
upper
limb
function.
Intraclass
correlation
(ICC):
The
ICC
is
used
to
assess
the
consistency,
or
conformity,
of
measurements
made
by
multiple
observers
measuring
the
same
quantity.
While
it
is
viewed
as
a
type
of
correlation,
unlike
most
other
correlation
measures
it
operates
on
data
structured
as
groups,
rather
than
data
structured
as
paired
observations
1.
ICC1:
Each
target
is
rated
by
a
different
judge
and
the
judges
are
selected
at
random.
It
is
sensitive
to
differences
in
means
between
raters
and
is
a
measure
of
absolute
agreement.
ICC2:
A
random
sample
of
k
judges
rate
each
target.
The
measure
is
one
of
absolute
agreement
in
the
ratings.
ICC3:
A
fixed
set
of
k
judges
rate
each
target.
There
is
no
generalization
to
a
larger
population
of
judges.
ICC2
and
ICC3
remove
mean
differences
between
judges,
but
are
sensitive
to
interactions
of
raters
by
judges.
The
difference
between
ICC2
and
ICC3
is
whether
raters
are
seen
as
fixed
or
random
effects

7. 7
Internal
consistency:
The
degree
of
the
interrelatedness
among
the
items.
The
degree
to
which
the
sum-­‐score
actually
reflects
the
trait
to
be
measured,
related
to
the
degree
of
unidimensionality
(see
also
Cronbach’s
alpha).
Item:
A
single
question
within
a
domain
or
questionnaire.
Items
with
dichotomous
response
options:
items
responded
by
whether
the
item
is
endorsed
or
not
(e.g.
yes/no
or
agree/disagree).
Items
with
polytomous
response
structure:
items
with
several
response
options.
The
structure
can
be
categorical,
ordinal,
intervallic.
Likert
scale:
A
rating
scale
in
which
raters
express
their
opinion
on
a
given
subject
by
marking
a
box
within
a
continuum
of
disagree-­‐agree
statements.
Minimal
clinically
important
difference
(MCID):
The
smallest
difference
that
patients
perceive
as
beneficial
(or
detrimental).
5.
There
are
several
methods
to
estimate
so-­‐called
minimal
clinically
important
differences
(MCIDs).
One
is
the
‘‘one
SEM’’
method,
which
defines
the
MCID
by
the
baseline
standard
deviation
multiplied
by
the
square
root
of
1
minus
the
reliability
coefficient
of
the
scale.
Another
is
the
anchor
based
method
(see
that
word).
Patient Rated Wrist Evaluation: the
Patient
Rated
Wrist
Evaluation
questionnaire
(PRWE)
6
was
originally
designed
as
a
specific
instrument
for
the
assessment
of
distal
radius
fractures
and
wrist
injuries.
Patient-­‐related
/
patient
reported
/
patient
rated
outcome
measures
(PROM):
evaluation
of
outcome
made
by
the
patient
without
interference
by
the
clinician
or
others
=
evaluation
from
the
patient’s
perspective.
A
questionnaire
used
in
a
clinical
trial
or
a
clinical
setting,
where
the
responses
are
collected
directly
from
the
patient.
Pearson’s
correlation
coefficient
(Pearson’s
r):
is
a
measure
of
the
linear
correlation
(dependence)
between
two
variables
X
and
Y,
giving
a
value
between
+1
and
−1
inclusive,
where
1
is
total
positive
correlation,
0
is
no
correlation,
and
−1
is
total
negative
correlation.
A

8. 8
correlation
value
between
0.8
and
1.0
or
-­‐0.8
and
-­‐1.0
is
considered
a
very
strong
relationship,
between
0.6
and
0.8
a
strong
relationship,
between
0.4
and
0.6
a
moderate
relationship,
between
0.2
and
0.4
a
weak
relationship
and
between
0
.0
and
0.2
a
very
weak
or
absent
relationship.
Periprosthetic
osteolysis
(PPO):
a
biological
process
of
bone
resorption
adjacent
to
prosthetic
joint
implants,
seen
as
radiolucent
lines
or
-­‐areas
on
radiographs.
QuickDASH:
Shortened
version
of
the
DASH-­‐questionnaire,
comprising
11
of
the
30
items
in
the
full
DASH7
Reproducibility:
The
extent
to
which
scores
for
patients,
whose
clinical
status
has
not
changed,
are
the
same
for
repeated
measurement
under
several
conditions:
over
time
(test-­‐
retest),
by
different
persons
on
the
same
occasion
(inter-­‐rater)
or
by
the
same
persons
on
different
occasions
(intra-­‐rater).
The
Pearson’s
and
Spearman’s
correlation
coefficients,
the
Intraclass
correlation
coefficient
and
the
Kappa
coefficient
are
commonly
used
statistic
to
evaluate
reliability.
Response
rate:
the
proportion
of
respondents
in
relation
to
all
patients
who
received
the
questionnaire.
Responsiveness:
the
extent
to
which
an
outcome
measure
instrument
is
able
to
detect
and
assess
how
a
patient
responds
to
treatment
or
a
meaningful
or
important
change
in
a
clinical
state
(the
ability
to
demonstrate
the
impact
of
treatment).
Responsiveness
can
be
assessed
by
the
standardized
effect
size
(SE)
or
the
standardized
response
mean
(SRM).
8
Sensitivity
of
a
measurement
instrument:
The
ability
of
an
instrument
to
measure
change
in
a
state
irrespective
of
whether
it
is
relevant
or
meaningful
to
the
decision
maker
9.

9. 9
Spearman’s
correlation
coefficient
(Spearman’s
rho):
is
a
nonparametric
measure
of
statistical
dependence
between
two
variables.
It
assesses
how
well
the
relationship
between
two
variables
can
be
described
using
a
monotonic
function.
If
there
are
no
repeated
data
values,
a
perfect
Spearman
correlation
of
+1
or
−1
occurs
when
each
of
the
variables
is
a
perfect
monotone
function
of
the
other.
Spearman's
coefficient
is
appropriate
for
both
continuous
and
discrete
variables,
including
ordinal
variables.
A
correlation
value
between
0.8
and
1.0
or
-­‐0.8
and
-­‐1.0
is
considered
a
very
strong
relationship,
between
0.6
and
0.8
a
strong
relationship,
between
0.4
and
0.6
a
moderate
relationship,
between
0.2
and
0.4
a
weak
relationship
and
between
0
.0
and
0.2
a
very
weak
or
absent
relationship.
Standard error of measurement (SEM): the
standard
deviation
of
repeated
test
Standardized
effect
size
(SE):
is
used
for
assessing
responsiveness
and
equal
to
the
mean
score
difference
(follow-­‐up
minus
baseline)
divided
by
the
score’s
standard
deviation
at
baseline.
An
ES
>0.80
is
considered
as
large,
0.50–0.79
as
moderate,
0.20–0.49
as
small,
and
0.00–0.19
as
very
small.
Standardized
response
mean
(SRM):
The
standardized
response
mean
is
one
of
several
available
and
widely
used
effect
size
indices,
used
to
gauge
the
responsiveness
of
scales
to
clinical
change.
The
SRM
is
computed
by
dividing
the
mean
score
change
(i.e.,
follow-­‐up
minus
baseline)
by
the
standard
deviation
of
the
change.
Cohen
has
advocated
thresholds
for
the
interpretation
of
effect
size
indices:
‘trivial’
(ES
<0.20),
‘small’
(ES
0.20-­‐0.50),
‘moderate’
(ES
0.50-­‐0.80),
or
‘large’
(ES
>0.80).
According
to
Middel
&
van
Sonderen
however,
this
may
lead
to
over-­‐
or
underestimation
of
the
magnitude
of
intervention-­‐related
change
over
time10.
Surrogate
measures:
measures
that
are
used
in
place
of
the
clinically
most
relevant
measures.
PROM’s
are
preferred
compared
to
surrogate
measures,
such
as
biomarkers
.

10. 10
Thurstone
scale:
is
made
up
of
statements
about
a
particular
issue,
and
each
statement
has
a
numerical
value
indicating
how
favourable
or
unfavourable
it
is
judged
to
be.
It
is
an
attempt
to
approximate
an
interval
scale.
E.g.
in
the
DASH-­‐questionnaire:
No
difficulty
(1),
Mild
difficulty
(2),
Moderate
difficulty
(3),
Severe
difficulty
(4),
Unable
(5).
Unidimensionality:
in
a
unidimensional
construct,
the
variable
is
identified
and
mapped
in
a
single
real
number
line.
A
questionnaire
that
measures
a
single
construct
is
described
as
unidimensional.
Items
(questions)
in
a
unidimensional
questionnaire
can
be
added
to
provide
a
single
scale
score.
Validity:
The
degree
to
which
a
PROM
measures
the
construct(s)
it
purports
to
measure.
Visual
Analogue
Scale
(VAS):
A
visual
analogue
scale
is
a
psychometric
response
scale,
which
can
be
used
in
questionnaires.
It
is
a
measurement
instrument
for
subjective
characteristics
or
attitudes
that
cannot
be
directly
measured.
When
responding
to
a
VAS
item,
respondents
specify
their
level
of
agreement
to
a
statement
by
indicating
a
position
along
a
continuous
line
between
two
end-­‐points.
This
continuous
(or
"analogue")
aspect
of
the
scale
differentiates
it
from
discrete
scales
such
as
the
Likert
scale.
There
is
evidence
showing
that
visual
analogue
scales
have
superior
metrical
characteristics
than
discrete
scales,
thus
a
wider
range
of
statistical
methods
can
be
applied
to
the
measurements.

11. 11
Abbreviations
CTS:
Carpal
Tunnel
Syndrome
DASH: Disabilities
of
Arm,
Shoulder
and
Hand
(Handicaps
i
Arm,
Skulder
og
Hånd):
Generic
patient
rated
outcome
measure
of
the
upper-­‐extremity
ICC:
Intraclass
Correlation
IWH:
Institute for Work & Health.
MCID:
Minimal
Clinically
Important
Difference
OA: Osteoarthritis
PPO: Periprosthetic Osteolysis
PRWE: Patient Rated Wrist Evaluation
PRISMA: Preferred Reporting Items for Systematic reviews and Meta-Analyses
PRO(M): Patient Related Outcome/ Patient Reported Outcome/ Patient Rated Outcome (Measure)
PT: Posttraumatic
PWA: Partial Wrist Arthroplasty
RA: Rheumatoid Arthritis
RSA:
Radiostereometric
Analysis
SEM: Standard Error of Measurement
SES:
Standardized
Effect
Size
SRM:
Standardized
Response
Mean
SLAC:
Scapholunate
Advanced
Collapse
SNAC:
Scaphoid
Nonunion
Advanced
Collapse
TAA: Total Ankle Arthroplasty
TEA: Total Elbow Arthroplasty
THA: Total Hip Arthroplasty

12. 12
TKA: Total Knee Arthroplasty
TSA: Total Shoulder Arthroplasty
TWA: Total Wrist Arthroplasty
TWF: Total Wrist Fusion
VAS: Visual Analogue Scale

13. 13
List
of
papers
I. Boeckstyns
MEH.
Wrist
arthroplasty
−
a
systematic
review.
Dan
Med
J
2014;61(5):A4834
11.
II. Herzberg
G,
Boeckstyns
M,
Sorensen
AI,
Axelsson
P,
Kroener
K,
Liverneaux
P,
et
al.
"Remotion"
total
wrist
arthroplasty:
preliminary
results
of
a
prospective
international
multicenter
study
of
215
cases.
J
Wrist
Surg.
2012
Aug;1(1):
17-­‐22.
12
III. Boeckstyns
ME,
Herzberg
G,
Sorensen
AI,
Axelsson
P,
Kroner
K,
Liverneaux
PA,
et
al.
Can
total
wrist
arthroplasty
be
an
option
in
the
treatment
of
the
severely
destroyed
posttraumatic
wrist?
J
Wrist
Surg.
2013
Nov;2(4):
324-­‐9.
13
IV. Boeckstyns
ME,
Herzberg
G,
Merser
S.
Favorable
results
after
total
wrist
arthroplasty:
65
wrists
in
60
patients
followed
for
5-­‐9
years.
Acta
orthopaedica.
2013
Aug;84(4):
415-­‐9.
14
V. Boeckstyns
MEH,
Herzberg
G:
Periprosthetic osteolysis after total wrist
arthroplasty. J Wrist Surg 2014;3:101–106 15
.
VI. Boeckstyns
MEH,
Toxværd
A,
Bansal
M,
Vadstrup
LS.
Wear
particles
and
osteolysis
in
patients
with
total
wrist
arthroplasty.
J
Hand
Surg
Am
2014;
39(12):2396-­‐2404
16
.
VII. Herup
A,
Merser
S,
Boeckstyns
M.
[Validation
of
questionnaire
for
conditions
of
the
upper
extremity].
Ugeskr
laeger
2010;172(48):
3333-­‐6.
17
VIII. Boeckstyns
MEH,
Merser
S:
Psychometric Properties of two Questionnaires in the
Context of Total Wrist Arthroplasty.
Dan
Med
J
2014;
61
(11):
A4939
18.

14. 14
Introduction
and
Background
1.
Historical
background
and
current
issues
concerning
Total
Wrist
Arthroplasty
(TWA).
Themistocles
Gluck
(1853-­‐1942)
is
said
to
have
performed
the
first
total
wrist
arthroplasty
(TWA)
19.
“A
19-­‐year-­‐old
male
patient,
named
Franz,
had
a
21-­‐month
history
of
tuberculosis
of
his
right
wrist,
presumably
due
to
a
trauma.
He
showed
progressive
loss
of
function
and
atrophy
of
the
hand.
On
9
June
1890
an
operation
was
performed.
A
dorsoradial
incision
in
the
manner
of
von
Langenbeck
was
used;
resection
of
the
joint
including
the
base
of
the
metacarpals,
the
two
carpal
rows
and
the
distal
part
of
radius
and
ulna
was
performed.
After
cleaning
the
wound
thoroughly
and
extirpation
of
the
capsule
a
device
made
of
ivory
was
placed,
a
ball
and
socket
articulation
with
forks
at
both
ends,
designed
so
that
one
fork
fitted
the
ulna
and
radius
and
the
other
in
the
medullary
canals
of
the
metacarpals.
Stable
fixation
was
achieved,
the
wound
was
closed
and
recovery
was
uneventful.
Today
the
device
is
fully
incorporated,
the
hand
is
not
shortened
and
no
pain
is
present.”
20.
At
a
follow-­‐up
of
more
than
one
year,
the
implant
was
still
in
place
with
a
good
range
of
motion,
but
a
chronic
fistula
was
present
due
to
the
nature
of
the
original
disease
process.
The
idea
of
wrist
arthroplasty
using
artificial
materials
was
then
abandoned
until
John
Niebauer
and
Alfred
Swanson
during
the
1960s
independently
introduced
the
concept
of
a
silicone
interpositional
spacer
for
joint
replacement
that
could
offer
immediate
stability
and
a
foundation
on
which
the
reparative
fibrous
tissue
could
grow
without
inhibiting
later
motion.
Swanson
started
using
these
silicone
implants
for
the
radiocarpal
joint
in
1967
and
reported
his
experience
in
1982
and
1984
21.
Lundkvist
&
Barfred
have
reported
on
a
Danish
experience
22.
The
results
have
been
generally
favourable
in
low
demand
rheumatoid
patients

15. 15
at
short
term
but
the
silicone
spacers
are
no
longer
in
use
for
wrist
replacement
due
to
problems
with
breakage,
subsidence
and
silicone
synovitis
23.
The
2nd
generation
of
implants,
introduced
in
the
1970s,
were
multicomponent
24
25-­‐28.
There
is
no
consensus
on
the
definition
of
second
generation.
In
this
thesis,
it
is
defined
as
an
implant
consisting
of
a
radial
component
and
a
carpal
component
that
is
fixated
in
one
or
more
of
the
metacarpal
bones.
Some
of
these
systems
have
been
developed
after
the
introduction
of
the
3rd
generation
27-­‐29.
The
3rd
generation
is
characterized
by
minimal
bone
resection
and
avoids
fixation
in
the
metacarpal
bones,
with
the
exception
of
an
optional
and
restricted
fixation
in
the
second
metacarpal.
They
attempt
to
mimic
the
natural
anatomy
and
biomechanics
of
the
wrist
and
are
largely
unconstrained
30-­‐32.
In
recent
years,
pyrocarbon
was
introduced
as
a
single
component
interposition
arthroplasty
33
or
hemiarthroplasty
34.
I
define
these
as
“4th
generation”.
Many
2nd
generation
implants
turned
out
to
have
deceiving
long
term
results
and
most
are
no
longer
available.
The
published
series
are
generally
rather
small
and
with
a
short
follow-­‐up.
The
most
well
documented
2nd
generation
implant,
the
Biax,
was
withdrawn
from
the
market
for
commercial
reasons.
Thus,
the
longevity
of
TWA
needs
to
be
investigated.
Reports
are
not
uniform.
Some
report
implant
survival
rates
at
8
years
at
a
level
of
80-­‐100%
25,35-­‐37,
whilst
others
report
markedly
lower
survival
38
39.
Problems
have
mainly
been
located
at
the
carpal
side.
Periprosthetic
osteolysis
has
also
been
a
problem.
It
may
be
associated
with
definite
implant
loosening
or
not
25
but
its
natural
history
and
clinical
consequences
have
not
been
well
described.
The
question
as
to
which
extent
and
on
what
indications
TWA
is
superior
to
total
wrist
fusion
(TWF)
also
needs
to
be
answered
definitely.
Although
many
patients
with
bilateral
procedures
-­‐TWA
on
one
side
and
TWF
on
the
other
–
would
have
preferred
arthroplasty
on

16. 16
both
sides,
this
is
not
always
the
case
36.
2.
Patient-­‐rated
Outcome
measures
(PROMs)
and
TWA
Patient-­‐rated
outcome
assessments
are
increasingly
emphasized
in
orthopaedic
surgery,
as
it
has
in
other
medical
specialties.
The
field
has
progressed
from
outcomes
defined
by
joint
motion
and
bony
union
to
standardized
assessments
of
function
and
disability
completed
by
the
patients.
It
is
essential
to
realize
that
the
choice
of
available
health
status
instruments
is
related
to
the
methodological
debate
on
the
psychometric
properties
of
instruments.
Generic
measures
allow
investigators
to
compare
health
status
across
different
diseases
and
interventions.
Condition-­‐specific
measures
focus
on
the
disease
being
studied,
allowing
greater
sensitivity
to
intervention
related
change
compared
to
generic
measures.
The
DASH
was
designed
as
a
measure
of
disability:
physical
function
(in
terms
of
disability)
and
symptoms
related
to
the
upper-­‐limb.
As
a
generic
upper-­‐limb
measure,
it
assesses
the
impact
of
disorders
on
the
whole
person
rather
than
on
a
specific
limb,
i.e.
the
whole
person’s
ability
to
function,
even
if
the
person
is
compensating
with
the
other
arm
or
using
devices.
The
DASH
and
the
shortened
version,
the
QuickDASH,
are
probably
the
most
widely
used
patient
rated
outcome
measure
instruments
(PROMs)
in
hand
surgery.
With
the
increased
international
focus
on-­‐
and
usage
of
PROMs
it
has
become
increasingly
important
that
they
are
properly
translated/culturally
adapted
and
well
validated
in
the
context
in
which
they
are
intended
to
be
used
40.

17. 17
Aims
of
the
Thesis
The
aims
of
this
thesis
were:
1. To
review
the
literature
in
order
to
update
and
summarize
the
current
knowledge
on
total
wrist
arthroplasty
(TWA).
2. To
analyse
the
clinical
and
radiographical
results,
and
the
longevity
obtained
with
one
3rd
generation
TWA,
the
Re-­‐motion
prosthesis
(SBI
Inc.,
Morrisville,
PA,
USA,
previously
the
Avanta
TWA,
AVANTA
Orthopaedics
San
Diego,
CA)
.
3. To
obtain
knowledge
on
the
prevalence,
location,
possible
causes
and
clinical
implications
of
periprosthetic
osteolysis
(PPO).
4. To
assess
and
validate
patient
rated
outcome
measures
(PROMs),
commonly
used
in
the
context
of
TWA
with
special
focus
on
the
QuickDASH
–
questionnaire
and
the
PRWE.
Initiatives
The
thesis
is
the
result
of
working
with
PROMs
and
TWA
during
15
years.
It
has
included
a
systematic
search
of
the
literature
and
the
construction
of
a
multicentric
international
registry
for
the
Re-­‐motion
TWA
in
view
of
collecting
data
on
a
larger
sample
and
with
a
longer
follow-­‐up
period
than
currently
available.
The
Re-­‐motion
TWA
is
an
elliptic
ball
and
socket
design
consisting
of
radial
and
carpal
Cr-­‐Co
components
that
are
titanium-­‐coated,
and
an
intercalated
polyethylene
component
that
mainly
articulates
with
the
radial
component
but
also
permits
a
rotational
articulation
of
20
degrees
with
the
carpal
plate
(Figure
1).
The
carpal
plate
is
fixated
to
the
carpus
by
its
stem
and
2
screws,
of
which
only
the
most
radial
may
penetrate
the
metacarpal
for
a
very
short
distance
even
though
many
advocate
not
doing
so.

18. 18
Thus,
fixation
is
mainly
aimed
to
be
to
the
carpus
and
minimally
in
the
metacarpals.
The
fixation
is
typically
done
without
cement.
Figure
1.
The
Re-­‐motion
TWA
with
the
metallic
radial
and
carpal
components
and
the
intercalated
polyethylene
ball
(Courtesy
Acta
Orthopaedica
2013,
paper
IV).
Both
initiatives
have
revealed
that
periprosthetic
osteolysis
(PPO),
showing
as
periprosthetic
radiolucency
on
plain
X-­‐rays,
with
or
without
total
loosening
of
the
implant
components
is
a
frequent
occurrence.
Further
investigations
were
conducted
-­‐
together
with
Guillaume
Herzberg,
Lyon
-­‐
in
order
to
obtain
more
precise
information
on
the
nature,
location,
prevalence
and
possible
clinical
consequences
of
PPO.
Finally
a
study
was
done,
aiming
at
finding
relations
between
implant
wear
and
PPO.
Concurrently
the
psychometric
properties
of
widely
used
PROMs
were
investigated
upon:
the
shortened
version
of
the
Disability
of
Arm
Shoulder
and
Hand-­‐questionnaire
(QuickDASH)

19. 19
and
the
Patient
Rated
Wrist
Evaluation
(PRWE).
An
essential
part
of
this
process
was
the
cross-­‐cultural
adaptations
of
the
DASH
and
QuickDASH
to
Danish.
Methods
and
Methodological
Considerations
1.
The
PRISMA
statements
Systematic
reviews
are
essential
tools
for
summarizing
evidence
accurately
and
reliably
in
an
effort
to
assess
the
benefits
and
harms
of
health
care
interventions.
They
attempt
to
collate
all
empirical
evidence
that
fits
pre-­‐specified
eligibility
criteria
to
answer
specific
research
questions
and
may
be
used
to
summarize
evidence
other
than
that
provided
by
randomized
trials.
The
review
of
the
literature
in
this
thesis
was
conducted
according
to
the
PRISMA
–
guidelines.
The
overall
aim
of
PRISMA
–
consisting
of
a
27-­‐item
checklist
and
a
four-­‐phase
flow
diagram
-­‐
is
to
help
ensure
the
clarity
and
transparency
of
reporting
of
systematic
reviews
and
meta-­‐analyses
41.
A
search
was
made
using
a
protocolled
strategy
and
well-­‐
defined
criteria
in
PubMed,
in
the
Cochrane
Library
and
by
screening
reference
lists
(Fig.
2).
I
made
a
primary
search
through
PubMed
with
the
Mesh
terms
“Wrist
Arthroplasty”
and
“Wrist
Replacement”
but
restricted
the
search
to
the
1994-­‐2013-­‐
period,
considering
earlier
material
to
have
historical
value
only.
A
second
search
was
done
in
the
Cochrane
Library
and
a
continuous
supplementary
search
by
scanning
the
reference
lists
of
the
papers
first
included.
The
inclusion
criteria
were:
papers
with
primary
clinical
data
on
second,
third
and
fourth
generation
implants.
Excluded
were:
cadaveric
studies;
biomechanical
studies;
studies
not
accessible
in
journals,
books
or
online;
reviews
without
primary
data.
Double
publications

20. 20
and
articles
with
overlap
of
cases
were
relative
exclusion
criteria.
Articles
not
written
in
English,
Danish,
Swedish,
Norwegian,
French,
Dutch
or
German
were
evaluated
on
the
basis
of
an
English
abstract,
if
available.
Papers
with
less
than
ten
cases
were
considered
to
be
less
useful
and
are
therefore
only
mentioned
very
briefly.
Implant
longevity
was
primarily
evaluated
on
the
basis
of
papers
reporting
a
cumulated
implant
survival
of
at
least
five
years;
secondarily,
papers
with
a
follow-­‐up
of
a
minimum
of
two
years
in
each
case.
Function
was
evaluated
if
reported
by
well-­‐validated
and
relevant
outcome
measurement
tools
like
the
DASH/QuickDASH,
the
PRWE
or
the
MHQ.

21. 21
Figure
2:
Flowdiagram
of
the
search
strategy
in
the
systematic
review
of
the
literature
on
TWA
(1994-­‐2013)
1
Search
for
“wrist
arthroplasty”
and
“wrist
replacement”:
800
articles
(by
April
2013)
Additional
search
by
scanning
reference
lists:
13
articles
Number
of
articles
after
exclusions
through
Mesh-­‐
words:
248
Number
of
eligible
articles
after
duplicated
removed:
44
Number
of
articles
after
screening:
36
Number
of
eligible
articles:
56
16
articles
for
qualitative
evaluation
of
longevity
(articles
with
follow-­‐up
in
every
case
≥2
years
or
with
documented
implant
survival
at
≥5
years)
Articles
with
N
<
10
excluded
from
analysis
but
briefly
mentioned:
7
21
16
37
articles
for
qualitative
evaluation
of
clinical
outcome
Duplicates
/
overlap
excluded:
12
articles
Supplementary
search
by
December
2013:
7
articles

22. 22
2.
The
multicenter
international
Re-­‐motion
registry
The
initiative
to
create
this
registry
was
taken
in
2009
in
collaboration
with
Guillaume
Herzberg
and
with
the
technical
assistance
of
Søren
Merser.
The
launching
has
partly
been
supported
by
SBI
Inc.,
Morrisville,
PA,
USA
and
has
since
been
administrated
by
the
two
initiators,
independently
from
any
industrial
or
commercial
interests.
The
input
of
data
in
the
registry
is
made
online,
directly
by
the
participants
and
is
overviewed
and
supported
by
the
two
initiators.
Data
sampled
before
the
creation
of
the
registry
could
be
also
be
entered,
provided
they
were
collected
prospectively
and
according
to
the
guidelines
of
the
registry.
On
demand
by
any
participant,
statistical
calculations
and
the
generation
of
extensive
updated
reports
are
performed
automatically
in
real
time.
The
registry
is
accessible
at
https://statcom.dk/irwa.
Ideally,
the
participant
centres
should
perform
follow-­‐up
examinations
of
their
cases
annually
after
operation.
Quality
control
in
the
registry
The
registry
offers
the
possibility
to
the
administrators
to
perform
data
quality
control.
Primarily,
this
consists
of
surveillance
of
the
completeness
of
data.
Lacking
follow-­‐up
data
are
reported
automatically
to
the
administrators,
in
which
case
they
may
choose
to
contact
the
participants
and
urge
them
to
complete
the
data,
an
action
that
has
been
taken
with
success
before
important
communications
and
publications
in
peer-­‐reviewed
journals.
Also,
the
administrators
may
choose
to
exclude
data
from
participants
that
do
not
have
included
a
sufficient
number
of
cases
or
follow-­‐up
examinations.
The
registration
of
general
outcome
measures,
like
motion
measured
with
goniometer,
grip-­‐strength
measured
with
the
JAMAR
Hydrolic
Hand
Dynamometer
(Sammons
Preston
Rolyan,
Bolingbrook,
IL,
USA)
and
pain
on
a
Visual
analogue
scale
(VAS)
may
be
considered
as
biased,
due
to
a
potential
inter-­‐rater

23. 23
variation,
but
“outliers”
can
be
spotted
by
the
administrators
and
action
taken
for
correction.
The
quality
of
data
on
function
or
disability
is
assured
by
the
use
of
approved
and
validated
versions
of
the
QuickDASH.
The
radiographical
data
are
considered
as
weak
and
merely
estimates,
because
it
is
left
to
the
judgment
of
the
surgeons
who
contributed
to
the
register,
how
to interpret
the
radiographs.
The degree of uniformity of data across centres is shown in table 1, showing data for each of the
seven centres that were selected to provide data in view of 3 publications 12-14
, recorded
preoperatively and at latest follow-up (minimum 1 year), revision cases excluded.

24. 24
Table 1.
Data recorded preoperatively and at latest follow-up after Re-motion TWA in 7 centres.
Centre (number of cases in brackets)
A (45) B (51) C (25) D (64) E (17) F (19) G (13)
Motion
Mean of Total
Extension-flexion in
degrees
(preop/postop)
75/61 72/65 47/48 64/65 77/72 69/73 75/70
Mean of total Ulnar-
radial flexion in
degrees
(preop/postop)
30/27 35/37 12/18 33/39 21/41 23/26 NA
Mean Grip strength
in kgF (preop/postop)
12/16 11/13 9/12 10/16 10/21 14/14 9/13
Median QuickDASH-
score (0-100)
(preop/postop)
47/30 61/45 67/42 55/25 50/20 92/72* 56/34
Median VAS-score
for Pain (0-100)
(preop/postop)
70/14 67/12 80/10 67/11 70/0 67/24 65/30
*: The QuickDASH was calculated in two cases only at this centre.

25. 25
3.
The
cumulated
implant
survival
We
consider
the
Kaplan-­‐Meier
method
for
the
analysis
of
implant
survival
as
a
powerful
tool
in
the
evaluation
of
the
durability
of
TWA
42.
The
method
makes
it
possible
to
analyse
data
from
patients
with
different
lengths
of
follow-­‐up,
taking
into
account
dropouts
for
any
reason.
A
disadvantage
is
the
fundamental
assumption
that
patients
who
are
lost
to
follow-­‐up
and
patients
who
have
died
have
the
same
failure
rate
as
those
who
comply
with
regular
follow-­‐
up
examinations,
which
is
not
necessarily
true.
The
most
widely
accepted
and
commonly
used
definition
of
failure
in
implant
survival
analysis
is
“revision”
(removal
of
implants).
This
endpoint
has
been
criticized
because
the
criteria
used
to
decide
the
need
for
removal
may
vary
between
patients
and
surgeons,
and
sometimes
it
is
argued
that
other
definitions
should
be
considered.
These
could
be
severe
pain
or
the
presence
of
radiolucency
or
subsidence
combined
with
moderate
or
severe
pain.
Still
other
definitions
can
be
considered,
but
it
will
remain
difficult
to
compare
survival
analyses
until
consensus
is
reached
about
which
other
outcome
measures
should
be
used
rather
than
revision.
In
the
studies
based
on
the
Re-­‐motion
registry,
the
decision
to
revise
implants
relies
on
the
judgment
of
several
surgeons
or
units
that
work
independently,
which
is
an
advantage
compared
with
studies
in
which
the
decision
is
made
by
a
single
surgeon
and
solely
dependent
on
this
person’s
views.
The
findings
must
be
interpreted
correctly:
The
survival
rate
at
the
“tail”
of
the
curve
is
less
reliable
because
of
the
relatively
small
number
of
patients
with
long
follow-­‐up,
and
it
must
be
expected
that
the
incidence
of
revision
will
increase
as
the
implants
inevitably
wear
out.

26. 26
4.
Radiographical
measurements
The
radiographical
data
in
the
Remotion
register
are
considered
as
weak
and
merely
estimates,
because
it
was
left
to
the
judgment
of
the
surgeons
who
contributed
to
the
register,
how
to
interpret
the
radiographs.
In
paper
V 15
, we (Guillaume Herzberg and myself) used the
measurement
software
provided
by
Sectra
(Sectra
AB,
Linköbing,
Sweden).
We
did
the
measurements
together
in
order
to
obtain
consensus
and
uniformity,
but
we
made
no
further
attempt
to
validate
the
measurements.
In
paper
VI
16
,
the
measurements
were
done
independently
with
the
Sectra
software
by
2
blinded
raters
(Lars
S.
Vadstrup
and
myself)
and
the
inter-­‐rater
reliability
assessed
on
a
total
of
820
measurements
of
the
width
of
radiolucent
zones.
It
can
be
argued
that
our
evaluation
of
the
osteolystic
area
may
not
reflect
the
volume
of
bone
resorption
correctly,
since
we
used
a
2-­‐dimensional
surrogate
for
a
3-­‐dimensional
space.
Nevertheless,
our
method
was
highly
reproducible
(Pearson’s
r
and
ICC
=
0.85)
and
corresponding
to
the
method
described
by
Cobb
et
al.
25
and
later
also
used
by
Takwale
et
al.
36,
although
these
do
not
operate
with
mean
values
of
2
zones.
We
considered
a
progressive
and
consistent
change
of
distances
indicating
the
position
of
the
implants
compared
to
the
bony
structures
form
the
first
postoperative
to
the
latest
radiograph,
of
≥
3mm
as
indicative
of
subsidence.
The
3
mm
threshold
was
considered
to
be
significant
because
any
error
due
to
obliquity
of
the
film
would
be
eliminated
and
because
it
has
been
used
in
other
studies
25
36.
Smaller
changes
that
were
consistent
and
progressive
on
the
serial
radiographs
could
theoretically
be
considered
as
possible
subsidence,
but
were
not
encountered.
Correspondingly,
change
of
angulation
of
≥
5
degrees
was
considered
as
indicative
of
tilting
of
the
implants.
The
inter-­‐rater
reliability
of
these
measurements
(4
x
78
measurements),
each

27. 27
performed
independently
by
the
two
blinded
investigators
was
also
very
high:
Pearson’s
r
and
ICC
were
between
0.87
and
0.98.
It
can
be
argued
that
other
methods
could
have
been
preferable
for
the
assessment
of
implant
loosening.
These
could
have
been
radiostereometric
analyses
(RSA)
or
CT-­‐based
methods.
CT-­‐based
methods
have
been
used
in
a
preclinical
cadaveric
trial
on
TWA,
but
have
not
been
validated
for
clinical
use
43.
RSA
has
been
applied
to
trapeziometacarpal
implants
44
but
to
date
not
to
TWA.
The
method
requires
standardization
and
validation,
which
is
beyond
the
scope
of
this
thesis.
Moreover,
traditional
marker-­‐based
RSA
requires
metal
markers
to
be
implanted
in
the
surrounding
bones
in
addition
to
the
polyethylene
component
of
the
implant
itself,
which
could
not
be
done
retrospectively.
Furthermore:
since
the
polyethylene
component
of
the
Re-­‐motion
is
mobile
compared
to
the
carpal
plate
and
hence
to
the
carpal
bones
(at
least
in
theory)
the
method
cannot
be
used
for
carpal
plate
assessment.
We
had
no
access
to
the
alternative
model-­‐based
RSA.
5.
Histopathological
evaluations
Histopathological
findings
may
be
subject
to
differences
of
interpretation
and
inter-­‐observer
variation
that
deserve
consideration
and
discussion.
For
this
reason,
in
paper
VI
16
two
blinded
pathologists
with
experience
in
examining
implant-­‐bone
interphases
reviewed
independently
the
biopsies
and,
after
a
general
common
discussion,
performed
an
independent
second
look
examination.
The
interobserver
agreement
between
the
two
investigators
concerning
different
findings
(“present”
or
“not
present”),
was
assessed
with
Cohen’s
kappa.
There
is
no
consensus
on
how
to
interpret
kappa
values
but
in
this
thesis,
I
adopted
values
<0.40
as
indicating
poor
agreement,
values
0.40-­‐0.75
fair
to
good
agreement

28. 28
and
>0.75
excellent
agreement
45.
Generally
the
inter-­‐observer
agreement
was
good
to
excellent
(Table
2)
.
Table
2:
The
reliability
of
histological
findings
(agreement
between
the
two
pathologists)
Findings
Kappa
Necrotic
tissue
0.68
Metal
particles
0.59
Polyethylene
debris
0.78
Preponderance
of
macrophages
compared
to
lymphocytes
1.00
Foreign
body
reaction
0.74
6.
Translation
of
PROMs
With
the
increased
international
focus
on-­‐
and
usage
of
PROMs
it
has
become
increasingly
important
that
they
are
properly
translated.
Translation
implies
a
cross-­‐cultural
adaptation
process.
Guillemin
et
al.
have
proposed
standardized
guidelines
for
this
process
46,
which
today
are
commonly
used.
The
process
of
cultural
adaptation
of
the
DASH
and
the
QuickDASH
included
forward
translation,
expert
panel
discussion,
back-­‐translation,
new
expert
panel
discussion,
pre-­‐testing
with
cognitive
interviewing
and
formulating
a
final
version
and
documentation
to
the
Institute
for
Work
&
Health
(IWH),
Toronto,
Canada.
The
cultural
adaptation
of
the
PRWE
has
been
performed
by
another
team
47.
7.
Validation
of
PROMs

29. 29
For
being
useful
PROMs
have
to
be
validated
in
the
specific
context,
in
which
they
are
intended
to
be
used.
This
can
be
done
by
classical
test
methods
or
by
item
response
methods
and
equivalents
45
,
chapter
12
40.
Sample
dependency
may
be
a
problem
connected
with
the
classical
test
theory
45
pp
299-­‐301,
which
emphasizes
the
need
of
validation
of
PROMs
in
the
specific
context
of
TWA.
At
this
stage
two
of
the
most
commonly
used
PROMs
in
wrist
surgery
–the
DASH/QuickDASH
and
the
PRWE
-­‐
were
validated
by
classical
test
methods
in
this
thesis.
The
validation
has
included
assessments
of
construct
validity,
reproducibility,
internal
consistency,
responsiveness
and
floor-­‐/ceiling
effect
17
18
.

30. 30
Results
In
the
systematic
review
of
the
literature
on
TWA
(paper
I)
11
,
37
publications
describing
a
total
of
18
implants
were
selected
for
analysis
12,25,27,28,30,32,33,35-­‐39,48-­‐72
(Fig.
1).
Sixteen
of
the
publications
were
useful
for
the
evaluation
of
implant
longevity
12,25-­‐27,30,35-­‐39,51-­‐53,62,71-­‐73.
Despite
methodological
shortcomings
in
many
of
the
source
documents,
some
summary
estimate
was
possible.
It
seems
that
wrist
arthroplasty
(TWA
or
PWA)
has
a
good
potential
to
improve
function
through
pain
reduction
and
preservation
of
mobility.
The
risk
of
severe
complications
–
deep
infection
and
instability
problems
–
is
small
with
the
available
implants.
A
cumulated
implant
survival
of
0.9
to
1.0
at
five
years
is
reported
in
most
series
–
if
not
all
–
on
newer
second
generation
implants
(the
Biax
TWA)
and
third
generation
implants
(the
Re-­‐
motion
and
the
Universal
2)
,
but
it
declined
between
five
and
eight
years.
Periprosthetic
osteolysis
(PPO)/radiolucency
is
frequently
reported:
In
13
of
the
37
series,
no
useful
information
could
be
retrieved,
whereas
20
papers
did
report
osteolysis,
ten
of
these
mentioning
radiolucency
without
frank
loosening
of
the
implant
components.
In
a
consecutive
series
of
Biaxial
TWA
with
a
follow-­‐up
time
of
5-­‐9
years,
there
was
progressive
radiolucency
at
the
carpal
component
in
12
out
of
46
wrists,
seven
of
which
were
revised.
Subsidence
of
the
carpal
component
was
present
in
seven
cases
after
one
year
and
in
20
cases
at
final
follow-­‐up.
The
causes
and
consequences
of
PPO
are
not
clarified
25.
In
the
clinical
studies
of
this
thesis
(paper
II-­‐IV)
12-­‐14
,
based
on
the
multicentre
international
Re-­‐motion
registry,
the
cases
of
seven
centres,
contributing
with
at
least
15
inclusions
in
the
multicentre
international
Re-­‐motion
registry
and
adequate
follow-­‐up
examinations,
were
considered
12.
A
total
of
215
wrists
were
included
in
paper
I.
In
the
rheumatoid
arthritis
group
(RA;
129
wrists)
and
the
non
rheumatoid
arthritis
group
(non-­‐RA;
86
wrists),
there
were
5

31. 31
and
6%
complications
respectively,
requiring
implant
revision,
with
a
survival
rate
of
0.96
and
0.92
respectively
at
4
years
(Figure
3a
and
b)
and
of
92
%
at
8
years.
Within
the
whole
series,
only
one
dislocation
was
observed
in
one
non-­‐RA
wrist.
A
total
of
112
wrists
(75
RA
and
37
non-­‐RA)
had
more
than
2
years
of
follow-­‐up
(average
4
years,
range
2-­‐8
years).
In
the
RA
and
non-­‐RA
group,
the
mean
VAS-­‐score
for
pain
improved
by
48
and
54
points,
respectively,
and
the
mean
QuickDASH
score
improved
by
20
and
21
points,
respectively,
with
no
statistically
significant
differences
between
the
two
diagnostic
groups.
Average
postoperative
arc
of
wrist
flexion–extension
was
58
degrees
in
RA
wrists
(loss
of
1
degree)
compared
with
63
degrees
in
non-­‐RA
wrists
(loss
of
9
degrees)
with
no
statistically
significant
differences
(Table
3).
Grip
strength
improved
respectively
by
40
and
19%
in
RA
and
non-­‐RA
groups
(p
=
0.033).
Implant
loosening
seen
at
follow-­‐up
was
reported
in
4%
of
the
RA
wrists
and
3%
of
the
non-­‐RA
wrists
with
no
statistically
significant
differences
between
the
two
diagnostic
groups.
Radiolucency
without
migration
of
the
implant
components
was
reported
in
8
%
of
the
RA
cases
and
in
15
%
of
the
non-­‐RA
cases,
still
without
differences
between
the
groups.
Essentially,
this
study
suggested
that
the
Re-­‐motion
TWA
was
feasible
in
the
midterm
in
RA
as
well
as
selected
non-­‐RA
patients.

32. 32
Table
3:
Outcomes
of
TWA
in
112
wrists
with
at
least
2
years
of
follow-­‐up
Rheumatoid
Non-­‐rheumatoid
Statistical
significance
of
the
differences
between
the
diagnostic
groups
VAS
Pain
improvement
(100
point
scale)
48
points
54
points
ns
Quick
DASH
improvement
20
points
21
points
ns
Wrist
Extension
29°
(+2°)
36°
(-­‐4°)
ns
Wrist
Flexion
29°
(-­‐3°)
37°
(-­‐5°)
ns
Ulnar
Deviation
24°
(+7°)
28°
(+2°)
ns
Radial
Deviation
5°
(-­‐1°)
10°
(64°)
P=0,015
Grip
Strength
improvement
(%
of
pre-­‐operative
value)
40%
19%
P=0,033

33. 33
Figure
3a:
Cumulated
Implant
survival
curve
for
non-­‐rheumatoid
patients
(Courtesy
Thieme/J
Wrist
Surg,
Paper
II).
Figure
3b:
Cumulated
Implant
survival
curve
for
rheumatoid
patients
(Courtesy
Thieme/J
Wrist
Surg,
Paper
II).

34. 34
We
made
a
specific
analysis
of
the
cases
in
which
the
Re-­‐motion
TWA
was
used
as
a
salvage
procedure
for
severe
arthritis
due
to
posttraumatic
causes
(paper
III)
13.
Thirty-­‐five
cases
had
a
minimum
follow-­‐up
time
of
2
years.
Average
follow-­‐up
was
39
(24–96)
months.
Pain
had
improved
significantly
at
follow-­‐up,
mobility
remained
unchanged.
The
total
revision
rate
was
3.7%,
and
the
implant
survival
was
92%
at
4–8
years
(Figure
4).
The
clinical
relevance
of
this
paper
was
that
although
painful
posttraumatic
wrists
with
severe
joint
destruction
can
be
salvaged
by
TWF
and
sometimes
by
partial
wrist
fusion,
TWA
can
be
an
alternative
procedure
and
yields
results
that
are
comparable
to
those
obtained
in
rheumatoid
cases,
at
least
evaluated
at
short
to
mid-­‐term.

35. 35
Figure
4:
Cumulated
implant
survival
curve
for
posttraumatic
patients
(Courtesy
Thieme/J
Wrist
Surg,
Paper
III).

36. 36
In
a
specific
analysis
of
the
cases
operated
between
2003
and
2007
–
performed
in
order
to
have
a
series
with
a
minimum
of
5
years
follow-­‐up
in
each
case
-­‐
60
patients
had
been
operated
(5
bilaterally),
5
wrists
had
been
revised,
and
52
with
the
original
implant
in
situ
were
available
for
follow-­‐up
(paper
IV)
14.
The
pain
scores,
grip
strength,
QuickDASH
scores,
ulnar
flexion,
and
supination
for
the
whole
group
were
statistically
significantly
better
at
follow-­‐up
(table
4
and
figure
5).
There
were
no
statistically
significant
differences
between
the
rheumatoid
and
the
non-­‐rheumatoid
patients
except
for
motion,
which
was
better
in
the
non-­‐rheumatoid
group.
The
motion
obtained
depended
on
the
preoperative
motion
(in
average
and
in
the
individual
patients).
The
implant
survival
was
0.9
at
9
years,
both
in
rheumatoid
and
non-­‐rheumatoid
cases
(figure
6).
In
six
cases,
there
were
radiographic
signs
of
implant
loosening
(subsidence
or
tilting):
five
carpal
plates
and
one
radial
component
(five
rheumatoid,
one
idiopathic
osteoarthrosis
(OA)).
In
11
other
cases,
PPO
without
any
loosening
of
implant
components
was
reported:
three
carpal
alone,
seven
radial
alone,
and
one
radial
and
carpal
(eight
rheumatoid,
three
posttraumatic).

37. 37
Table
4:
Clinical
results
at
the
latest
follow-­‐up
(“Post”)
compared
to
preoperative
values
(“Pre”).
Mean
values
(SD),
but
median
(range)
for
QuickDASH
Rheumatoid
cases
Non-­‐rheumatoid
cases
P-­‐value1
All
cases
P-­‐value2
Pre
Post
Pre
Post
Pre
Post
Pain
(
0-­‐
100
on
VAS)
66
(20)
29
(26)
72
(12)
23
(38)
P=0.6
67
(17)
27
(29)
P<0.001
Grip
strength
(KgF)
9
(8)
14
(8)
16
(14)
19
(13)
P=0.3
10
(10)
15
(10)
P=0.03
QDASH
(0-­‐100)
61
(41-­‐
89)
41
(8-­‐84)
41
(14-­‐
79)
50
(0-­‐61)
P=0.5
58
(14-­‐
89)
42
(0-­‐84)
P<0.001
Motion
(degrees)
Supination
71
(22)
81
(13)
72
(35)
89
(4)
P=0.003
71
(25)
83
(12)
P=0.005
Pronation
71
(16)
80
(10)
82
(12)
85
(13)
P=0.3
79
(15)
81
(11)
P=0.5
Extension
27
(16)
28
(15)
43
(18)
43
(22)
P=0.06
30
(17)
31
(18)
P=0.8
Flexion
25
(21)
25
(16)
50
(19)
44
(23)
P=0.003
31
(23)
29
(19)
P=0.7
Radial
7
(11)
6
(8)
14
(8)
7
(5)
P=0.6
8
(11)
6
(8)
P=0.3
Ulnar
14
(8)
20
(14)
23
(14)
28
(16)
P=0.2
16
(11)
22
(14)
P=0.02
1
Significance
of
differences
between
the
rheumatoid
cases
and
the
non-­‐rheumatoid
cases
at
follow-­‐up.
2
Signifcance
of
differences
between
preoperative
values
and
values
at
follow-­‐up
for
the
total
sample.

38. 38
Figure
5.
QuickDASH-­‐score
before
operation
and
at
follow-­‐up.
The
dotted
line
represents
equivalency.
(Courtesy
Acta
Orthopaedica
2013,
Paper
IV)

39. 39
Figure
6:
Cumulated
implant
survival
curve
for
patients
operated
between
2003
and
2007.
(Courtesy
Acta
Orthopaedica
2013,
Paper
IV).

40. 40
In
terms
of
radiography,
a
major
weakness
of
the
above
mentioned
analyses
in
paper
II-­‐IV
was
that
there
were
not
given
precise
guidelines
for
the
evaluation
of
radiolucency
or
precise
criteria
for
implant
loosening:
this
was
left
to
the
judgment
of
the
participating
surgeons.
For
this
reason,
we
made
a
specific
analysis
of
the
prevalence,
location
and
natural
history
of
PPO
following
TWA
with
precise
measurements
on
the
radiographs
of
consecutive
patients
operated
in
2
wrist
centres
(paper
V)
15.
We
excluded
patients
with
less
than
2
years
follow-­‐
up
and
cases
that
had
been
revised
with
removal
of
implant
components.
Thus
we
analysed
44
consecutive
cases.
The
X-­‐ray
examinations
were
done
preoperatively,
at
6
months
after
operation
and
thereafter
annually.
We
defined
radiological
spots
for
the
measurement
of
radiolucency
on
digitalized
posteroanterior
radiographs
(figure
7)
and
measured
the
maximal
width
of
the
radiolucent
zones
at
these
spots.
We
found
significant
periprosthetic
radiolucency
(more
than
2
mm
in
width)
at
the
radial
component
side
in
16
of
the
cases
and
at
the
carpal
component
side
in
seven.
It
developed
gradually
around
the
prosthetic
components
near
the
joint
regardless
of
the
primary
diagnosis,
and
seemed
to
stabilize
in
most
patients
after
1-­‐3
years
(figure
8a
and
b).
In
a
small
percentage
of
the
patients,
the
periprosthetic
area
of
bone
resorption
was
markedly
larger.
In
general,
radiolucency
was
not
related
to
evident
loosening
of
the
implant
components
and
only
five
carpal
components
and
one
radial
had
subsided
or
tilted.

41. 41
Figure
7
Spots
for
the
measurement
of
the
width
of
radiolucency
on
serial
postero-­‐anterior
radiographs.
(Courtesy
Thieme/J
Wrist
Surg,
Paper
V).

42. 42
Figure
8a
(Courtesy
Thieme/J
Wrist
Surg,
Paper
V)
Width
of
radiolucent
zones
at
spot
4-­‐5
in
function
of
time.
Each
line
represents
a
single
case.
X-­‐axis:
length
of
follow-­‐up
in
years.
Y-­‐axis:
Width
of
radiolucency
in
mm.

43. 43
Figure
8b
(Courtesy
Thieme/J
Wrist
Surg,
Paper
V).
Width
of
radiolucent
zones
at
spot
9-­‐10
in
function
of
time.
Each
line
represents
a
single
case.
X-­‐axis:
length
of
follow-­‐up
in
years.
Y-­‐axis:
Width
of
radiolucency
in
mm.
The
arrows
indicates
a
maximal
width
of
radiolucency
under
the
carpal
plate
at
2
years
after
operation
in
this
particular
case
(6.2
mm),
and
at
4
years,
where
the
radiolucent
zone
was
reduced
to
almost
0
mm,
as
the
carpal
plate
sunk
into
the
carpus.
In
a
following
study
(paper
VI)
16,
we
aimed
at
determining
whether
the
amount
of
polyethylene
and
metallic
debris
in
the
interphase
tissue
between
prosthesis
and
bone
in
patients
with
TWA
correlated
to
the
degree
of
periprosthetic
osteolysis
(PPO).
We
also
measured
the
level
of
Cr-­‐
and
Co-­‐
ions
in
the
blood,
and
assessed
the
possible
role
of
infectious
or
rheumatoid
activity
in
the
development
of
PPO.
Biopsies
were
taken
from
the
implant-­‐bone
interphase
in
13
consecutive
patients
with
Re-­‐motion
TWA
and
with
at
least
3
-­‐
years
follow-­‐up,
and
histological
as
well
as
bacteriological
examinations
were
done.
Serial

44. 44
annual
radiographs
were
obtained
prospectively
for
the
evaluation
of
PPO.
Blood
samples
were
collected
for
white
blood
cell
count,
C-­‐reactive
protein,
and
metallic
ion
level.
A
radiolucent
zone
of
>
2mm
was
observed
juxta-­‐articular
to
the
radial
component
in
four
cases
and
at
the
carpal
component
in
three
cases.
Its
magnitude
tended
to
level
out
over
time.
Subsidence
of
the
implant
was
observed
in
three
cases
on
the
carpal
side
and
in
no
case
on
the
radial
side.
The
amount
of
polyethylene
and
metallic
debris
was
generally
small
and
did
not
correlate
with
the
width
of
the
radiolucent
zone
(Figure
9
and
10).
The
blood
levels
of
Cr-­‐
and
Co-­‐
ions
were
normal.
There
was
no
evidence
of
infectious
or
rheumatoid
activity.
Figure
9:
scatterplot
showing
the
width
of
radiolucency
in
zone
4-­‐5
vs.
the
amount
of
polyethylene
fragments
in
the
samples
taken
from
the
periprosthetic
tissue
between
the
radial
component
and
the
radius.

45. 45
Figure
10:
Polarized
light
microscopy,
x
200
magnification).
Polyethylene
fragments
engulfed
by
multinucleated
foreign
body
giant
cells,
semi-­‐quantitatively
estimated
as
“intermediate”
on
a
4-­‐grade
scale
from
“none”
to
“high”
for
the
amount
of
foreign
body
particles.
In
paper
VII,
we
evaluated
the
reproducibility
of
the
Danish
version
of
the
DASH
in
a
total
of
83
patients,
72
of
these
having
hand-­‐related
conditions
and
ten
having
shoulder
problems.
This
investigation
was
a
supplement
to
the
cultural
adaptation
of
the
DASH.
The
patients
included
were
all
in
a
stable
stage
of
their
disease,
which
typically
was
chronic,
and
they
were
invited
to
answer
the
questionnaire
in
connection
with
a
medical
examination
preceding
surgery
or
at
the
final
ambulatory
visit
after
treatment.
After
one
week,
a
second
questionnaire
was
sent
to
the
patients
and
the
correlation
between
the
two
scores
was
calculated.
Fifty-­‐four
of
the
patients
completed
both
questionnaires
at
an
interval
of
maximum
30
days.
.
The
DASH
scores
of
the
first
and
second
test
were
numerically
close
with
an
intra-­‐class
coefficient
of
0.85.
Cronbach's
alpha
was
0.96,
indicating
that
the
subtests
were

46. 46
internally
consistent.
Spearman's
correlation
coefficient
was
overall
0.90
and,
thus,
there
was
a
strong
correlation
between
the
first
and
the
second
response
(figure
11).
The
scores
were
evenly
spread
across
the
scale,
suggesting
a
good
discriminative
property.
Ten
percent
of
the
responses
were
insufficient
for
the
calculation
of
a
score.

47. 47
Figure
11:
correlation
between
DASH-­‐scores
in
a
test-­‐retest
trial
on
54
patients
with
a
stable
upper-­‐extremity
condition.
X-­‐axis
=
scores
at
first
test,
y-­‐axis
=
scores
at
retest
14-­‐30
days
later.
ICC
=
0.85,
Spearman’s
rho
=
0.90.
(figure
1
in
Herup
A,
Merser
S,
Boeckstyns
M.
[Validation
of
questionnaire
for
conditions
of
the
upper
extremity].
Ugeskr
laeger
2010;172(48):
3333-­‐6
(Paper
VII)).
As
the
DASH-­‐questionnaire
is
increasingly
replaced
by
the
QuickDASH,
we
decided
to
assess
the
psychometric
properties
of
the
Danish
QuickDASH
and
another
widely
used
PROM
for
wrist
conditions
–
the
PRWE
-­‐
in
patients
with
total
wrist
arthroplasty
(paper
VIII)
18.
In
a
prospective
cohort
of
102
cases,
we
evaluated
the
QuickDASH.
In
a
cross-­‐sectional
study
and
in
a
test-­‐retest
on
a
subgroup
of
the
patients
we
evaluated
both
the
QuickDASH
and
the
PRWE.
Internal
consistency
and
reproducibility
were
very
high
(Cronbach’s
alpha
0.96
/

48. 48
0.97;
Spearman’s
rho
0.90
/
0.91;
ICC
0.91
/
0.92)
and
there
were
no
floor-­‐
or
ceiling
effects.
The
responsiveness
of
the
QuickDASH
was
high
(SRM
1.06
and
ES
1.07).
The
construct
validity
of
both
scales
was
confirmed
by
three
a
priori
formulated
hypotheses:
a
moderate
negative
correlation
of
scores
with
grip-­‐strength,
a
moderate
positive
correlation
with
pain
and
a
very
weak
or
no
correlation
with
mobility.
Rheumatoid
patients
scored
significantly
higher
on
the
QuickDASH,
indicating
a
higher
degree
of
disability.
The
scores
of
both
questionnaires
were
very
closely
related
(figure
12,
Spearman’s
rho
=
0.90
)
.
Figure
12:
Scatter
plot
showing
the
QuickDASH-­‐
and
the
PRWE-­‐scores
in
patients
with
Total
Wrist
Arthroplasty.
The
curved
line
is
the
LOESS
(local
regression)
line,
the
thin
solid
line
is
the
regression
line
and
the
thick
solid
line
is
the
line
of
equivalency.
Dots
located
on
the
line
of
equivalency

49. 49
represent
patients
whose
QuickDASH
and
PRWE-­‐scores
are
equal.
Dots
below
this
line
indicate
a
QuickDASH-­‐score
that
is
lower
than
the
PRWE-­‐score
for
a
given
patient
and
vice-­‐
versa.
The
tendency
indicated
by
the
LOESS-­‐line
is
that
the
QuickDASH
scores
are
approximately
5
points
higher
than
the
PRWE-­‐scores
in
the
lower
end
of
the
scales,
while
they
are
approximately
10
points
lower
in
the
higher
end.
(figure
2
in
Boeckstyns
MEH,
Merser
S:
Psychometric
Properties
of
two
Questionnaires
in
the
Context
of
Total
Wrist
Arthroplasty.
Dan
Med
J
2014;
61
(11):
A4939
(Paper
VIII)).
General
Discussion
and
Comparison
with
Other
Research
Review
of
the
literature
According
to
the
systematic
review
of
the
literature
on
TWA
(paper
I),
no
more
than
17
publications
–
after
exclusion
of
one
paper
for
language
reasons,
one
paper
reporting
data
as
a
part
of
a
less
commonly
used
scoring
system
and
several
papers
because
of
an
important
overlap
of
the
materials
-­‐
were
prospective,
even
when
using
a
broad
definition:
data
collected
preoperatively
as
well
as
postoperatively.
Of
these
17
papers,
eight
used
a
validated
and
widely
used
outcome
measurement
system.
Weak
methodology
applied
mainly
to
studies
on
2nd
generation
implants.
In
at
least
16
of
the
37
papers
that
were
selected,
one
or
several
authors
were
involved
as
or
close
to
the
inventors,
developers
or
producers,
but
this
seemed
not
to
have
had
impact
on
the
reported
clinical
or
longevity
results.
The
majority
of
data
are
based
on
rheumatoid
cases,
although
other
diagnoses
are
increasingly
represented
in
recent
publications.
Previously,
the
general
opinion
has
been
that
better
longevity
must
be
expected
in
low-­‐
demand
patients,
typically
rheumatoid
patients.
It
is
not
possible,
throughout
the
different
series
in
the
literature
to
compare
outcomes
in
rheumatoid
and
non-­‐rheumatoid
patients,
except
that
the
series
of
Herzberg,
Boeckstyns
et
al.
(paper
I)
12
shows,
on
basis
of
prospective
data,
that
there
are
no
clinically
or
statistically
significant
differences
between
these

50. 50
diagnostic
groups.
This
is
consistent
with
a
growing
view
that
non-­‐rheumatoid
patients
may
do
better
because
of
a
better
bone
stock,
provided
that
their
level
of
activity
is
restricted
70.
In
general,
mean
values
for
motion
at
follow-­‐up
are
similar
for
most
implants
and
generally
within
the
functional
range
defined
by
Palmer
et
al.
74
although
somewhat
smaller
than
the
more
rigorous
range
defined
by
Ryu
&
Cooney
75.
An
exception
may
be
the
Maestro
that
showed
better
motion
in
the
single
series
with
this
implant
32.
On
the
other
hand,
there
is
less
consistency
concerning
the
change
in
motion
from
before
operation
to
follow-­‐up.
This
might
be
attributed
to
different
case
selections,
different
postoperative
protocols
or
factors
related
to
the
implant
itself
but
it
is
impossible
to
clarify
this
on
basis
of
the
published
data.
In
the
studies
based
on
the
Re-­‐motion
registry,
no
statistically
significant
change
in
motion
was
obtained
(see
below).
A
reasonable
appreciation
of
the
longevity
of
implants
was
possible
in
16
papers,
although
only
ten
provided
information
on
cumulated
implant
survival.
These
ten
papers
concerned
the
Biaxial,
Meuli,
Motec,
Re-­‐motion
and
Universal
prostheses.
Generally,
the
5-­‐year
implant
for
the
survival
rate
was
higher
than
90
%
but
declining
at
8
years
(0.81-­‐1.0
for
the
Biax
and
0.77
for
the
Meuli.).
One
exception
is
the
low
survival
of
the
Universal
1
reported
by
Ward
et
al.
38
(
0.75
at
5
years,
0.62
at
8
years).
This
series
contains
exclusively
rheumatoid
cases.
It
is
likely
that
the
lower
survival
rate
is
due
to
the
strategy
of
the
surgeon,
who
seemed
to
be
inclined
to
revise
implants
exhibiting
PPO
without
frank
loosening.
On
the
other
hand,
the
Universal
2
series
of
Ferreres
et
al.
report
a
100
%
survival
at
a
mean
follow-­‐up
of
5.5
years.
Another
notable
result
concerns
the
metal-­‐on-­‐metal
APH-­‐prosthesis:
36
of
37
implants
were
revised
during
a
follow-­‐up
time
of
2-­‐6.1
years.
.
Solitary
loosening
of
the
carpal
component
of
this
implant
was
predominant.
The
authors
believed
that
the
main
cause
of
loosening
was
bone
resorption
induced
by
titanium
debris
and
they
abandoned
the
use
of
this
implant
39.

51. 51
Krukhaug
et
al.
have
reported
on
the
survival
of
189
TWA
in
the
Norwegian
Arthroplasty
Register
76:
The
cumulated
survival
of
the
Biax
was
85
%
at
5
years
and
approximately
78
%
at
8
years.
The
survival
of
the
Gibbon/Motec
was
obviously
lower
than
published
by
Reigstad
et
al.
62,
which
possibly
can
be
attributed
to
underreporting
to
the
register
77.
Failed
TWA
can
successfully
be
revised
by
fusion
14,30,38,39,60,62,65,
by
total
or
partial
replacement
of
the
components
14,30,38,51,60,65
or
by
total
or
partial
removal
of
the
components
with
or
without
soft-­‐tissue
interposition,
typically
fascia
lata
25,38.
The
review
did
not
aim
to
make
a
comparison
between
TWA
and
TWF,
but
the
question
is
important
to
discuss.
Murphy
et
al
made
a
comparison
between
TWA
(Universal
1
in
24
rheumatoid
wrists)
and
TWF
(27
rheumatoid
wrists)
in
a
retrospective
design
78.
Treatment
groups
were
well
matched
by
patient
characteristics
and
radiographic
staging.
There
were
no
statistically
significant
differences
between
arthroplasty
and
arthrodesis
in
either
DASH
or
PRWE
scores.
Cavaliere
&
Chung
compared
TWA
with
TWF
in
a
systematic
review
of
the
literature
on
TWA
compared
with
TWF
for
rheumatoid
arthritis
79.
They
identified
18
total
wrist
arthroplasty
studies
representing
503
procedures
and
20
TWF
studies
representing
860
procedures
in
rheumatoid
patients.
They
concluded
that
the
outcomes
for
TWF
were
comparable
and
possibly
better
than
those
for
TWA.
One
major
limitation
in
that
study
was
that
the
methodology
in
the
source
publications
was
often
very
weak.
Furthermore,
the
TWA-­‐
implants
in
many
of
the
series
were
of
older
and
now
abandoned
designs,
as
well
as
the
techniques
for
TWF
varied
a
lot.
Nydick
et
al
compared
the
Maestro
TWA
(seven
wrists)
with
TWF
(15
wrists)
in
posttraumatic
arthritis
80.
The
PRWE
scores
were
significantly
better
in
the
arthroplasty
group,
but
there
were
no
differences
in
DASH
scores.
Besides
its
retrospective
design,
the
weakness
of
this
study
is
obviously
the
very
small
number
of
TWA
and
the
fact
that
all
cases
had
been
treated
at
the
same
clinic,
implying
that
there
had
been
a