Implant survival after total elbow arthroplasty: a
retrospective study of 324 procedures performed
from 1980 to 2008
Hans Christian Plaschke, MDa,*, Theis M. Thillemann, MD, PhDb,
Stig Brorson, MD, PhDa, Bo S. Olsen, MD, PhDa
aShoulder and Elbow Clinic, Department of Orthopedic Surgery, Copenhagen University Hospital, Herlev, Denmark
bDepartment of Orthopedic Surgery, Horsens Regional Hospital and Aarhus University Hospital, Horsens, Denmark
Background: Total elbow arthroplasty (TEA) is an established treatment for late-stage arthritis of the
elbow. Indications have expanded to osteoarthritis and nonunion in distal humeral fractures. Information
on implant survival and risk factors for revision is still sparse. The aim of this study was to evaluate implant
survival and risk factors for revision of TEAs inserted in patients in the eastern part of Denmark in the
period from 1980 until 2008.
Material and methods: The Danish National Patient Register provided personal identification numbers for
patients who underwent TEA procedures from 1980 until 2008. On the basis of a review of medical reports
and linkage to the National Patient Register, we calculated revision rates and evaluated potential risk fac-
tors for revision, including, age, sex, period, indication for TEA, and implant design.
Results: We evaluated 324 primary TEA procedures in 234 patients at a mean follow-up of 8.7 years
(range, 0-27 years). The overall 5-year survival was 90% (95% confidence interval [CI], 88%-94%),
and 10-year survival was 81% (95% CI, 76%-86%). TEAs performed with the unlinked design had a rela-
tive risk of revision of 1.9 (95% CI, 1.1-3.2) compared with the linked design. Fracture sequelae was asso-
ciated with a relative risk of revision of 1.9 (95% CI, 1.05-3.45).
Conclusions: We found acceptable implant survival rates after 5 and 10 years, with a higher revision rate
for the unlinked design and primary TEA due to fracture sequelae. Patient-related outcome measures
should be included in future studies for further elaboration of the outcomes after TEA.
Level of evidence: Level III, Retrospective Cohort Design, Treatment Study.
? 2014 Journal of Shoulder and Elbow Surgery Board of Trustees.
Keywords: Total elbow arthroplasty; elbow; arthroplasty; prosthesis; elbow release; rheumatoid arthritis;
elbow fracture; joint replacement
The elbow is a complex joint with a critical role in upper
extremity function. Mobility and stability of the elbow joint
are important for daily, leisure, and professional activities.
Even minor trauma or transient disease involvement
can result in limited and painful elbow motion. The
biomechanical function and the complex articular anatomy
The Scientific Ethics Committee for the Hovedstaden Region approved this
*Reprint requests: Hans Christian Plaschke, MD, Copenhagen Uni-
versity Hospital, Department of Orthopedic Surgery, Aavej 34, Hareskov,
DK-3500 Vaerloese, Denmark.
E-mail address: firstname.lastname@example.org (H.C. Plaschke).
J Shoulder Elbow Surg (2014) 23, 829-836
1058-2746/$ - see front matter ? 2014 Journal of Shoulder and Elbow Surgery Board of Trustees.
of the elbow are important to understand when performing
reconstructive surgery to this joint.
Since the first commercial total elbow arthroplasty
(TEA) was introduced by Dee4
implant designs have been introduced.2,21,23The TEA im-
plants are grouped by linked and unlinked design. The
linked design consists of a humeral and an ulnar component
that are physically connected. In the early linked design,
the connection between the components was a fixed hinge
that did not allow any varus-valgus motion between the
humeral and ulnar components.4,5This design was associ-
ated with high failure rates secondary to transmission of
high stresses toward the implant cement–bone interface and
mechanical failure.16,21,24In contrast to the earlier design,
today’s linked design has a sloppy hinge that allows varus-
valgus motion between the components.5,9,16
The unlinked design consists of a humeral and an ulnar
component that are not mechanically linked. This design
relies on matching shapes of the bearing surfaces, adequate
in 1972, different
bone stock, and soft tissue support for stability.6,8,11,13,19
The unlinked design, with the possibility of subluxation
and dislocation of the joint, is less favorable when elbow
destruction is more severe.8,24
The unlinked and linked design with sloppy hinge have
both been used for decades, but whether one design is
better than the other is uncertain.7,22The linked and the
unlinked designs are represented by different implant
brands provided by different product companies.21As with
the implant design, whether any implant brand is superior
to others is uncertain.7,10,18,22
Surgical techniques have improved, and indications for
TEA have expanded from severe rheumatoid arthritis
(RA) to include osteoarthritis, post-traumatic arthritis, and
post-traumatic conditions, such as nonunion after distal
humeral fracture, and more recently, as the initial treat-
ment for comminuted distal humeral fractures in the
elderly.3,14,15,17,21This places higher demands on implants
and may lead to higher failure rates.5,7,14
National Patient Register; TEA, total elbow arthroplasty.
(A) Flowchart shows exclusions and inclusions. (B) Distribution of included and excluded procedures over the 3 decades. NRP,
830H.C. Plaschke et al.
Literature on TEA is increasing but is still sparse con-
cerning the indication for surgery and on survival and
revision rates for different designs. Recent studies from the
Finnish, Norwegian, and Scottish registers show no differ-
ence in survival or revision rates by implant.10However,
the indication for TEA seems to influence the survival, with
primary TEA inserted due to fracture sequelae of the distal
humerus having the highest risk for revision.7,10,22
the period 1980 to 2008, to compare the revision rate of the
different TEA designs used, and to analyze the difference in
survival by choice of design, implant, and indications for
Material and methods
This study was based on data provided by the Danish National
Patient Register (NPR) for patients who underwent TEA pro-
cedures in eastern part of Denmark between 1980 and 2008.
The NPR holds data on all discharges from public and private
hospitals in Denmark since 1977, including dates of all admissions
and discharges, up to 20 diagnoses for every discharge, and the
surgical procedures performed.1The diagnoses are classified ac-
cording to the Danish version of the International Classification of
Diseases (ICD). The physician who discharges the patient assigns
all discharge diagnoses. With use of the NPR, it is possible to
construct the complete hospitalization history for each patient. The
NPR was used to identify all primary and revision TEAs performed
in Sealand (eastern part of Denmark) from 1980 to 2008.
The Civil Registration System
A unique personal identification number is given to all Danish
citizens at birth or according to their date of birth. The Civil
Registration System (CRS) records information on changes in
vital status of all Danish citizens, including changes of address,
date of emigration, and (since 1968) the date of death. Accurate
linkage between the public Danish registers is possible at the level
of individual patients by using the personal identification number.
Thus, we used the personal identification number to merge data
from the NRP and the CRS.
Retrieval of data was based on the ICD-8 criteria from 1980 to the
end of 1993 and the ICD-10 criteria from 1994 to 2008. The
criteria included all types of primary and revision prosthesis for
the elbow; caput radii prosthesis, isolated ulnar component or
humeral component, hemiarthroplasty, and TEAs, as well as
choice of cemented or uncemented technique. Given these criteria
(ICD-9: 70010, 70011, 70012, 70013, 70014, 70015, 70016,
70019, 70110, 70111, 70112, 70113, 70114, 70115, 70116, and
70117; and ICD10: KNCB0, KNCB1, KNCB20, KNCB30,
KNCB40, KNCB59, KNCB69, KNCB99, KNCC0, KNCC1,
KNCC2, KNCC3, KNCC4, KNCC59, and KNCC99), 1664 per-
sonal identification number were provided covering all of
Denmark. Of these, 526 patients underwent operations in the
eastern part of Denmark at 12 different hospitals. To verify data,
we studied the patients’ medical reports at the different hospitals.
Medical report data
From the medical reports we obtained demographic data, age, and
sex, indication for surgery, brand of implant used, complications,
revision surgery, and reason for revision. Any revision performed
outside the eastern part of Denmark was identified in the data
provided by the NPR. By going through the medical reports, 324
procedures were verified as being primary TEA procedures per-
formed on 234 patients in the period 1981 until 2008 (Fig. 1). In
68 cases of the 324 primary TEAs, revision TEA procedures were
gery over time (A) by implant design and (B) by indications for
Developments in total elbow arthroplasty (TEA) sur-
Implant survival after total elbow arthroplasty831
performed. An isolated survival analysis of these revision TEAs
was also performed. None of the procedures performed in 1980
could be confirmed.
The end point for survival was defined as revision involving
1 component or the entire implant (removal or exchange).
All patients were followed up from surgery to revision,
death, emigration, or January 1, 2013. Kaplan-Meier sur-
vival analyses were used to calculate implant survival rates
at 1, 5, 10, and 15 years. Survival data obtained in the
Kaplan-Meier analysis were compared by the log-rank test
(Mantel-Cox). The TEA was bilateral in 45 patients, and
each replacement procedure was considered a separate
The Cox multiple-regression model was used to calcu-
late relative risk (RR) estimates adjusting for confounding
factors of age, sex, period of procedure, and indication for
TEA. The factors studied with the Cox regression model
were sex, age at surgery, period where primary TEA was
performed, indications for TEA, implant design, and
implant brand. When calculating RR by brand of implant,
we only included implants that had been used in more than
40 procedures. Differences between groups were consid-
ered statistically significant if the P values were <.05 in a
two-tailed test. SPSS 20.0 software (IBM Corp, Armonk,
NY, USA) was used for statistical analysis.
Of the 324 TEA operations, 264 (81.5%) were performed
in women. At the time of primary TEA, the mean age was
62 years (range, 25-91 years), and 160 (49.4%) of the
TEAs were performed on the right elbow. Rheumatoid
arthritis (RA) was the most frequent indication for the
TEA operation, with 237 (73.1%) procedures performed.
Other indications for primary TEA were fracture sequelae
with 61 procedures, including 7 acute fractures (18.8%),
osteoarthritis (OA) in 18 (5.6%), and other indications,
including septic and psoriasis arthritis, in 8 (2.5%).
Development in distribution of indication for TEA surgery
and TEA design over time are shown in Figure 2. Mean
overall follow-up was 105 months for primary TEA and 89
months for revision TEA.
Seven different brands of TEA implants were used,
counting 152 linked and 172 unlinked designs: Souter
itellocondylar (CC) TEA (Johnson & Johnson Orthopaedics
Inc, Raynham, MA, USA), Pritchard ERS (DePuy, DePuy
Mitek, Raynham, MA, USA), Kudo type 3 (Biomet Inc,
Warsaw, IN, USA), Coonrad-Morrey (Zimmer Inc, Warsaw,
Distribution of implant brands divided into unlinked and linked designs
Variable No.PeriodMean age Gender Indication for surgery
Kudo type 3
F, female; M, male; OA, osteoarthritis; RA, rheumatoid arthritis.
)Includes septic arthritis and psoriasis arthritis.
Distribution of reason for revision divided into unlinked and linked designs
Infection Fracture of
832 H.C. Plaschke et al.
IN, USA), GSB III (Zimmer), and Discovery (Biomet;
Table I). From 1981 to 2000 the most commonly used
implant was the unlinked Souter Strathclyde. Since 2003,
the tendency has gone from the unlinked toward the linked
design, and since 2008, only 1 TEA implant, the linked
Coonrad-Morrey implant, has been used in eastern part of
Denmark. Since 2008, total elbow replacement procedures
have been centralized to a single unit in the eastern part of
During the period 1981 to 2008, 68 revisions and 2 cases of
permanent implant removal were performed. The most
common reason for revision was aseptic loosening in 39
(57.4%), followed by periprosthetic fracture in 11 (16.2%),
prosthesis dislocation in 8 (11.8%), infection in 5 (7.4%),
fracture of the prosthesis in 3 (4.4%), and loose hinge pin in
2 (2.9%). The difference in the reason for revision between
linked and unlinked design is summarized in Table II. The
66 revision TEAs used were Coonrad Morrey in 30
(45.5%), GSB III in 15 (22.7%), Souter Strathclyde in 14
(21.2%), the CC TEA in 3 (4.5%), Kudo in 3 (4.5%), and
Pritchard ERS in 1 (1.5%).
Survival of TEA
The overall survival rates for primary TEAwere 90% (95%
confidence interval [CI], 88%-94%) at 5 years and 81%
(95% CI, 75%-86%) at 10 years, and varied for the different
types of implants, as reported in Table III. The Kaplan-
Meier survival curves related to Table III are shown in
The cumulative survival rates for linked prostheses were
90% (95% CI, 84%-95%) at 5 years and 88% (95% CI,
83%-94%) at 10 years. These results were comparable with
the unlinked design at 5 years (90%; 95% CI, 87%- 96%)
and at 10 years (77%; 95% CI, 70%-84%). This result was
not significantly different from the linked designs but
indicated a tendency of better long-term survival for the
The implant-specific 10-year cumulative survival rate
for the 4 most commonly used implant brands were 88%
(95% CI, 81%-95%) for Coonrad-Morrey, 89% (95% CI,
80%-97%) for GSB III, 88% (95% CI, 78%-97%) for the
CC TEA, and 72% (95% CI, 62%-82%) for the Souter
The 5- and 10-year cumulative survival rates for revision
TEAs (Coonrad Morrey, n ¼ 30; GSB III, n ¼ 15; Souter
Strathclyde, n ¼ 14; CC TEA, n ¼ 3; Kudo, n ¼ 3;
Pritchard, n ¼ 1) were 96% (95% CI, 90%-100%) and 89%
(95% CI, 79%-100%), respectively.
We found that the unlinked design was associated with a
higher RR of revision of 1.88 (95% CI, 1.1-3.2; P ¼ .049)
compared with the linked design. The adjusted RR for
revision was calculated between the 4 most used implants:
Coonrad-Morrey (RR, 0.49; 95% CI, 0.2-1.22; P ¼ .12),
GSB III (RR, 0.42; 95% CI, 0.19-0.94; P ¼ .039), and the
CC TEA (RR, 0.50; 95% CI, 0.23-1.1; P ¼ .09), with
Souter Strathclyde set as the indicator because it has been
used the longest and with a high frequency (n ¼ 90;
Compared with RA, primary TEAs performed due to
fracture had an increased adjusted RR for revision of 1.9
(95% CI, 1.1-3.4; P ¼ .035). The adjusted RR for revision
was 0.88 (95% CI, 0.54-1.42; P ¼ .85) for patients older
than 60 years and was 1.75 (95% CI, 1.02-3.01; P ¼ .043)
for men compared with women.
In this retrospective study we found acceptable 5-, 10-,
and 15-year implant survival rates after primary TEA.
Patients receiving TEAs over time have increased, and
Cumulative survival rates after 1, 5, 10 and 15 years according to design and implant brand
324 97 (96-100) 31490 (88-94) 24581 (75-86)120 67 (59-72)46
6 88 (81-95)
CI, confidence interval; n, number; nR, numbers at risk; S, survival in %; TEA, total elbow arthroplasty.
Implant survival after total elbow arthroplasty833
since 2003, the choice of implant has been dominated by
the linked design. Revision rates and RR estimates were
higher for the unlinked design. The most common reason
for revision for both the linked and unlinked design was
aseptic loosening. TEA procedures performed due to
fracture and fracture sequelae were associated with 1.8-
times increased risk for revision compared with RA. The
unlinked Souter Strathclyde TEA was associated with
increased risk for revision compared with the linked GSB
III. Male patients were associated with a 1.8-times
increased risk of revision.
The overall 5-year and 10-year survival rates of 90%
and 81% are comparable to previous studies from the
Scottish Arthroplasty Project and the Norwegian and
Finnish Arthroplasty Registers, with 8% to 10% and 10%
to 18% revision rates, respectively.7,10,22The Scottish
Arthroplasty project lacks data related to implant types.
Fevang et al7included 30 linked TEAs compared with
475 unlinked, and there was no 10-year follow-up on
linked implants. In the study by Skytt€ a et al,22all of the
TEAs were unlinked and the indication was restricted
In this study, the 5-year and 10-year survival rates for the
linked design were 90% and 88% compared with 90% and
77% for the unlinked design. The survival results after
revision TEA are excellent, with 5-year and 10-year sur-
vival of 96% and 89%. It is uncertain though whether this
high percentage of survival is due to good functional TEAs
and a low level of pain or the unwillingness among sur-
geons to perform further revision surgery on already
We found the same association between a TEA proce-
dure performed due to fracture sequelae and an increased
risk for revision as Fevang et al.7The RR in their study was
5.8 and also significant, but they counted only 12 pro-
cedures of unknown design.7Our higher level of power
could explain the difference in the RR estimates. The
linked design was used in 67% of the TEA procedures
performed due to fracture sequelae in this study.
The present data cover 172 unlinked and 152 linked
TEAs. Since 2006, only the linked design has been used
in TEA procedures. The tendency of increasing use of the
linked design was recently described by Giannicola et al8
in a review. Studies have indicated lower complication
rates and revisions in the linked design compared with
the unlinked design, but no studies have included survival
analysis.12,18The use of the linked design has increased
as has the increase in TEA procedures due to fracture and
fracture sequelae. The TEA procedures due to fracture
sequelae are associated with a higher revision rate, and
due to lack of sufficient bone stock and potential liga-
mentous insufficiency, literature advocates for the linked
design.12,18Patients with RA are known to generally have
lower demands than patients who undergo TEA pro-
cedures due to fractures and fracture sequelae. A lower
level of activity provides less strain on the prostheses,
thus decreasing the risk for revision. Younger and more
active patients have higher demands. Surgeons must keep
this in mind when considering TEA when the indication
is fracture sequelae and acute fracture in the distal
Because TEA procedures remain a challenge, the liter-
ature advocates centralization to ensure that surgeons with
adequate expertise perform the procedures.10,14,22,24Since
2008, the TEA procedures in the eastern part of Denmark
have been centralized to a single clinic.
The number of TEA procedures, the long-term follow-
up, inclusion of revision TEAs, as well as the equal
number of linked and unlinked implants provides strength
to this study. The weaknesses are the lack of radiographic
follow-up, identification of other complications that
could require revision, uncertainty of the numbers lost to
follow-up, the retrospective design of the study, and
the lack of functional and patient-related outcome
for total elbow arthroplasty (TEA) (A) by design and (B) by
Kaplan-Meier constructed overall survival curves
834H.C. Plaschke et al.
Survival analysis on TEA is sparse as are studies
comparing TEA designs. RA continues to be the most
common indication, but TEA procedures due to fracture
revision. The survival rates of revision TEAs are prom-
ising, which is of interest when performing primary TEA
surgery on younger patients, who are more likely than
older patients to live past the survival of their primary
TEA. The tendency goes toward the linked design, which
in this study is supported by the significantly higher RR
for revision for the unlinked design. A centralization of
the procedures should ensure adequate expertise and
volume to conduct further studies. Prospective studies
that include survival analysis, functional, and patient-
related outcomes would benefit to the studies on TEA
surgery, and further studies on revision TEAs should be
conducted. National registers and databases could
contribute to the knowledge about TEA as well.
The authors, their immediate families, and any research
foundations with which they are affiliated have not
received any financial payments or other benefits from
any commercial entity related to the subject of this
1. Andersen TF, Madsen M, Jørgensen J, Mellemkjoer L, Olsen JH. The
Danish National Hospital Register. A valuable source of data for
modern health sciences. Dan Med Bull 1999;46:263-8.
2. Bernardino S. Total elbow arthroplasty: history, current concepts, and
future. Clin Rheumatol 2010;29:1217-21. http://dx.doi.org/10.1007/
3. Cobb TK, Morrey B. Total elbow arthroplasty as primary treatment for
distal humeral fractures in elderly patients. J Bone Joint Surg Am
4. Dee R. Total replacement arthroplasty of the elbow for rheumatoid
arthritis. J Bone Joint Surg Br 1972;54:88-95.
5. Dee R. Reconstructive surgery following total elbow endoprosthesis.
Clin Orthop Relat Res 1982;170:196-203.
6. Espag MP, Back DL, Clark DI, Lunn PG. Early results of the Souter-
Strathclyde unlinked total elbow arthroplasty in patients with osteo-
arthritis. J Bone Joint Surg Br 2003;85:351-3. http://dx.doi.org/10.
7. Fevang B-TS, Lie SA, Havelin LI, Skredderstuen A, Furnes O. Results
after 562 total elbow replacements: a report from the Norwegian
Arthroplasty Register. J Shoulder Elbow Surg 2009;18:449-56. http://
Cox regression analyses of relative risk for revision adjusted for age, sex, period, and indication
VariableNo. Revisions Crude estimates (95% CI)RR (95% CI)P
Indication for primary TEA
1.88 (1.10-3.20) .049
CI, confidence interval; RR, relative risk; TEA, total elbow arthroplasty.
Implant survival after total elbow arthroplasty835
8. Giannicola G, Sacchetti FM, Antonietti G, Piccioli A, Postacchini R, Download full-text
Cinotti G. Radial head, radiocapitellar and total elbow arthroplasties: a
review of recent literature. Injury 2013;19:1-9. http://dx.doi.org/10.
9. Gill DRJ, Morrey BF. The Coonrad-Morrey total elbow arthroplasty in
patients who have rheumatoid arthritis. A ten to fifteen-year follow-up
study. J Bone Joint Surg Am 1998;80:1327-35.
10. Jenkins PJ, Watts AC, Norwood T, Duckworth AD, Rymaszewski LA,
McEachan JE. Total elbow replacement: outcome of 1,146 arthro-
plasties from the Scottish Arthroplasty Project. Acta Orthop 2013;84:
11. Landor I, Vavrik P, Jahoda D, Guttler K, Sosna A. Total elbow
replacement with the Souter-Strathclyde prosthesis in rheumatoid
arthritis. Long-term follow-up. J Bone Joint Surg Br 2006;88:1460-3.
12. Little CP, Graham AJ, Carr AJ. Total elbow arthroplasty: a systematic
review of the literature in the English language until the end of 2003. J
Bone Joint Surg Br 2005;87:437-44. http://dx.doi.org/10.1302/0301-
13. Van der Lugt JC, Rozing PM. Systematic review of primary total
elbow prostheses used for the rheumatoid elbow. Clin Rheumatol
14. Mansat P, Bonnevialle N, Rongi? eres M, Mansat M, Bonnevialle P.
Experience with the Coonrad-Morrey total elbow arthroplasty: 78
consecutive total elbow arthroplasties reviewed with an average 5
years of follow-up. J Shoulder Elbow Surg 2013;22:1461-8. http://dx.
15. Mansat P, Nouaille Degorce H, Bonnevialle N, Demezon H, Fabre T.
Total elbow arthroplasty for acute distal humeral fractures in patients
over 65 years old–results of a multicenter study in 87 patients. Orthop
Traumatol Surg Res 2013;99:779-84. http://dx.doi.org/10.1016/j.otsr.
16. Morrey BF, Bryan RS. Complications of total elbow arthroplasty. Clin
Orthop Relat Res 1982;(170):204-12.
17. Morrey BF, Sanchez-Sotelo J. Approaches for elbow arthroplasty: how
to handle the triceps. J Shoulder Elbow Surg 2011;20(2 Suppl):S90-6.
18. Park SE, Kim JY, Cho SW, Rhee SK, Kwon SY. Complications and
revision rate compared by type of total elbow arthroplasty. J Shoulder
Elbow Surg 2013;22:1-7. http://dx.doi.org/10.1016/j.jse.2013.03.003
19. Potter D, Claydon P, Stanley D. Total elbow replacement using the
Kudo. J Bone Joint Surg Br 2003;85:354-7. http://dx.doi.org/10.1302/
20. Robertsson O, Ranstam J. No bias of ignored bilaterality when ana-
lysing the revision risk of knee prostheses: analysis of a population
based sample of 44,590 patients with 55,298 knee prostheses from the
national Swedish Knee Arthroplasty Register. BMC Musculoskelet
Disord 2003;4:1. http://dx.doi.org/10.1186/1471-2474-4-1
21. Sanchez-Sotelo J. Total elbow arthroplasty. Open Orthop 2011;5:115-
22. Skytt€ a ET, Eskelinen A, Paavolainen P, Ik€ avalko M, Remes V. Total
elbow arthroplasty in rheumatoid arthritis. Acta Orthop 2009;80:472-7.
23. Souter WA. Surgery of the rheumatoid elbow. Ann Rheum Dis 1990;
24. Voloshin I, Schippert DW, Kakar S, Kaye EK, Morrey BF. Complica-
tions oftotal elbowreplacement: a systematic review.J Shoulder Elbow
Surg 2011;20:158-68. http://dx.doi.org/10.1016/j.jse.2010.08.026
836 H.C. Plaschke et al.