Intraoperative fracture during aseptic revision total knee arthroplasty Adam A. Sassoon MD, Cody C. Wyles BS, German A. Norambuena Morales MD, Matthew T. Houdek MD, Robert T. Trousdale MD PII: DOI: Reference:
S0883-5403(14)00324-6 doi: 10.1016/j.arth.2014.05.009 YARTH 53979
To appear in:
Journal of Arthroplasty
Received date: Revised date: Accepted date:
7 February 2014 11 May 2014 16 May 2014
Please cite this article as: Sassoon Adam A., Wyles Cody C., Norambuena Morales German A., Houdek Matthew T., Trousdale Robert T., Intraoperative fracture during aseptic revision total knee arthroplasty, Journal of Arthroplasty (2014), doi: 10.1016/j.arth.2014.05.009
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1 INTRAOPERATIVE FRACTURE DURING ASEPTIC REVISION TOTAL KNEE
RI P
T
ARTHROPLASTY
Adam A. Sassoon, MD
SC
Cody C. Wyles, BS
NU
German A. Norambuena Morales, MD Matthew T. Houdek, MD
ED
MA
Robert T. Trousdale, MD
55905
AC
CE
PT
Mayo Clinic, Department of Orthopedic Surgery, 200 First Street SW, Rochester, MN
Corresponding Author:
Robert T. Trousdale MD, Mayo Clinic, 200 First Street SW, Rochester, MN 55905 Office: 507-284-3663; Fax: 507-284-8935; Email:
[email protected] ACCEPTED MANUSCRIPT 2 Abstract Bone encountered during revision knee arthroplasty is compromised and
T
predisposed to fracture. This study reports the rate, location, timing, treatment, and
RI P
outcome of intraoperative fractures occurring during revision knee arthroplasty. Between 1997 and 2011, 2836 aseptic revisions were performed and retrospectively reviewed.
SC
Ninety-seven fractures were identified in 89 patients (3%). Fifty occurred in femora, 42
NU
in tibiae, and 5 in patellae. Forty-six occurred during exposure, 21 during bony preparation, 17 during trialing, and 13 during final component placement. Treatment
MA
included fixation (n=43), observation (n=21), component build-up (n=17), bone grafting (n=6), and a combination (n=3). Ninety-four percent of fractures united. Radiographic
ED
lucencies were noted around 19 components; however, only 3 were found to be clinically
PT
loose. Fifteen patients required a re-revision (17%), of which infection was the leading
CE
cause (n=5).
AC
Key Words: revision; total knee arthroplasty; aseptic; fracture
ACCEPTED MANUSCRIPT 3 Introduction The incidence, location, operative timing, and treatment of intraoperative
T
fractures during primary total knee arthroplasty (TKA) [1, 2] and two-stage revision TKA
RI P
[3] for infection have been reported. To our knowledge, similar focused details of this complication have not been reported for aseptic revision TKA. The reported incidence of
SC
intraoperative fractures occurring during two stage revision TKA performed in the setting
NU
of infection [3] was nearly a six-fold increase over that reported in primary TKA [1] from the same institution. We hypothesized that patients undergoing aseptic revision TKA
MA
would also have a similarly increased incidence of this complication when compared to primary TKA, especially when revision was performed for osteolysis, periprosthetic
ED
fracture, instability, or arthrofibrosis, as these indications may impose challenges with
necessitating removal.
PT
respect to exposure and available bone stock, while still having a well-fixed component
CE
This study aimed to report the incidence, location, and operative timing of
AC
intraoperative fractures occurring during aseptic revision TKA in a focused form. Additionally we aim to present a variety of management strategies for this complication. Lastly, we investigated the impact of this complication on overall survivorship and aimed to determine if an association between this complication and the need for re-revision exists.
Patients and Methods Following IRB approval, a retrospective review of surgical cases performed between 1997 and 2010 was completed using our institution’s total joint registry, which
ACCEPTED MANUSCRIPT 4 identified 2836 revision TKA’s placed for aseptic indications; 1528 were implanted in females and 1308 in males. Ninety-seven fractures occurred in 89 patients within this
T
cohort (3%). The indication for revision in the patients who sustained an intraoperative
RI P
fracture were osteolysis/aseptic loosening in 46 knees, instability in 22, arthrofibrosis in 12, extensor mechanism dysfunction in 6, and periprosthetic fractures in 3 (Figure 1).
SC
Sixty patients were female and 29 were male with an average age of 69 years (range, 45-
NU
91 years) and an average clinical followup of 63 months (range, 3-189 months). In cases where multiple fractures occurred in the same patient (n=8) a different anatomic location
MA
was recorded for each fracture, as was the operative timing. Bone defects at the time of reimplantation were also noted and graded according
ED
to the Anderson Orthopaedic Research Institute (AORI) classification [4, 5]. This
PT
information was taken directly from the operative report when available, and estimated via radiographic review of pre-revision radiographs by the senior author otherwise. Bone
CE
stock was compromised in all cases, the extent of which varied greatly as demonstrated
AC
by distribution of AORI scores recorded (Table 1). Bone loss was combated with a variety of component configurations; femoral reconstruction required stems in 80 cases, augments in 51, and metaphyseal cones or sleeves in 18. Tibial reconstruction required stems in 77, metaphyseal cones or sleeves in 22, and augments in 13. Implant constraint varied as well; 37 cases required constrained condylar equivalent components, 27 utilized posterior stabilized components, 23 required hinged components, and 2 utilized cruciate retaining components. Additionally, 3 patients required extensor mechanism augmentation as part of their revision procedure.
ACCEPTED MANUSCRIPT 5 The fracture location, timing of fracture occurrence and fracture treatment were recorded. A consensus of authors assessed radiographic evidence of fracture healing
T
(bridging of 3 cortices in a diaphyseal fracture, and disappearance of fracture lines in a
RI P
metaphyseal fracture) and component stability at the most recent followup. The need and reason for a subsequent re-revision were also determined. The average follow-up was 63
SC
months (range: 3-189 months). All patients were followed for at least 1 year with the
NU
exception of three patients, who were followed for 3 months and subsequently lost to
MA
followup.
Results
ED
Ninety-seven fractures occurred in 89 patients imparting an incidence of 3%. The
PT
most common location for the fracture was the femur, accounting for 48 fractures (49%), occurring in 44 patients. The specific locations of fracture occurrence in the femur are
CE
outlined in Figure 2. There were 35 simple fractures isolated to the tibia (36% of
AC
fractures), occurring in 34 patients; the specific locations of which are outlined in Figure 3. There were an additional 5 complex fractures of the tibia (5% of fractures), occurring in 4 patients; 2 of these involved the lateral cortex primarily but extended into the posterior cortex, 1 primarily involved the medial plateau but extended into the anterior cortex, 1 involved the anterior and medial cortices, and another involved the medial and posterior cortices. Two patients had fractures involving both the femur and tibia (5% of fractures). The first had a complex fracture of the femur involving both the medial condyle and anterior cortex with an accompanying fracture of the medial tibial plateau. The second had a fracture of the anterior femoral cortex in addition to a fracture of the
ACCEPTED MANUSCRIPT 6 medial tibial plateau. There were an additional 5 fractures of the patella (5% of fractures), occurring in 5 patients.
T
The most common timing for fracture occurrence was during exposure,
RI P
accounting for 46 events (47%). Another 21 fractures occurred during bony preparation (22%), 17 during trialing (18%), 12 during placement of final components (12%), and 1
SC
during tibial insert placement (1%) (Figure 4). Following identification of the fracture,
NU
management strategies were varied and tailored to fracture severity and location (Table 2). Management can be roughly categorized into fixation (n=43; 44%), observation
MA
(n=21; 22%), component build-up (n=16; 16%)), bone grafting (n=6; 6%), a combination of fixation and bone grafting (n= 2; 2%), and a combination of fixation and component
ED
build-up (n=1; 1%). Weight bearing limitations were imposed as an adjunctive treatment
PT
in 22 patients as well (25% of patients). At final followup 91 fractures were united (94%) (Figure 5), while 6 were non-
CE
united (6%) (Figure 6). Stable components were found in 86 patients (97%). Component
AC
stability was determined by an absence of radiographic findings suggestive of loosening (circumferential lucent lines, component migration, etc) in conjunction with an absence of pain. Radiographic lucencies were observed in 18 patients (20%). Seven were isolated to the tibial component, 2 were isolated to the femoral component, and 2 were isolated to the patellar component. Seven patients had lucencies around both their femoral and tibial components. Only 3 patients with lucencies had loose components, as the majority of these lesions were focal. Re-revision was required in 15 patients (17%) at a mean of 30 months (2-102). The reason for re-revision was infection in 5, loosening in 3 (2 tibial components, 1
ACCEPTED MANUSCRIPT 7 femoral component), mechanical hinge failure in 2, extensor mechanism failure in 2, periprosthetic fracture in 1, strut graft non-union in 1, and painful patellar crepitus in 1.
T
Re-revision for infection occurred at a mean of 31 months (range 1-102 months), with 2
RI P
occurring within a year of surgery. All of these underwent a 2-stage re-implantation.
SC
Discussion
NU
This study represents the first report we are aware of to document in a focused format the incidence, location, timing, and treatment of intraoperative fractures occurring
MA
during revision TKA when performed for aseptic indications. Bone loss occurring from component loosening, osteolysis, periprosthetic fracture, or removal of well-fixed
ED
components creates an increased risk for intraoperative fracture in the setting of revision
PT
TKA. The reported incidence of intraoperative fracture occurring during primary TKA is 0.39% [1]. This series demonstrated a rate of 3%, a near 10-fold increase when
CE
compared to primary TKA, reflective of the bone stock compromise often encountered in
AC
the revision setting.
The two most common sites for fracture in our series were the lateral femoral condyle and the anterior femoral cortex. The medial tibial plateau followed these closely behind. The most common portion of the case during which fractures occurred was exposure, followed by bony preparation. A variety of treatment strategies were observed, but fixation of the fracture was most commonly employed. Weight bearing limitation was also common as an adjunctive treatment measure. Fracture union occurred in an overwhelming majority of cases (94%) at final followup and stable components were even more common (97%). Despite the
ACCEPTED MANUSCRIPT 8 encouraging results related to these clinical parameters, the re-revision requirement was still 15%, the leading cause for which was infection. This is greater failure rate than that
T
observed in the general aseptic revision TKA population at our institution during a
RI P
similar time period [6]. This may be due in part to the added time required to address the fracture intraoperatively, predisposing the patient to a higher risk for infection. The
SC
aseptic failure rate in our cohort was 10%, which was also higher than which was
NU
observed from our institution in the general aseptic revision TKA population [6]. This may not be a result of the fracture itself, but rather a situation where the fracture serves as
MA
a prognostic indicator signifying decreased mechanical integrity of bone supporting the revision prosthesis.
ED
When compared to existing literature on intraoperative fractures encountered
PT
during a revision TKA, the results of this study mirror those reported by Sassoon et al reported in the setting of two-stage revisions for periprosthetic infections [3]. In both
CE
studies the most common location for fractures was the femur. Fracture union occurred
AC
in the overwhelming majority of patients in both studies (94% in the present and 89% in the previous). Additionally, both series noted a higher rate of re-revision in patients who sustained an intraoperative fracture when compared to historic cohorts of otherwise similar clinical circumstance. Furthermore, in both studies the leading reason for rerevision was infection. This study suffers from several limitations including those inherent to its retrospective design. The reported incidence of 3% likely represents an underestimate as discovery of intraoperative fractures within our total joints registry relies on the surgeon to dictate its occurrence within the operative report. Thus minor fractures not detected by
ACCEPTED MANUSCRIPT 9 the treating surgeon, or those requiring no alteration of the operative plan may have been omitted. The patient cohort is also heterogeneous with regards to treating surgeon,
T
operative technique, prosthesis, and post-operative rehabilitation, which may introduce
RI P
substantial variability into the clinical results and survivorship. Cementless technique and prosthetic designs with larger box cuts in smaller femurs may predispose to
SC
intraoperative fractures and these factors were not controlled for in our analysis.
NU
In conclusion, the rate of intraoperative fractures occurring during aseptic revision TKA is roughly 3%, with fractures most commonly occurring during exposure and
MA
involving the femur. Recognition of the fracture and adequate treatment resulted in union in 90% of cases and stable implants in 97% of cases. Despite these results a lower
ED
survivorship at 5 years was observed in this patient cohort, and although strict causation
PT
between intraoperative fractures and decreased survivorship was not proven, this association should be appreciated and may warrant closer follow-up between mid- and
AC
CE
long-term intervals.
ACCEPTED MANUSCRIPT 10 References: 1.
SC
NU
MA
ED
6.
PT
5.
CE
4.
AC
3.
RI P
T
2.
Alden, K.J., et al., Intraoperative fracture during primary total knee arthroplasty. Clin Orthop Relat Res, 2010. 468(1): p. 90-5. Pinaroli, A., et al., Intraoperative fractures and ligament tears during total knee arthroplasty. A 1795 posterostabilized TKA continuous series. Orthop Traumatol Surg Res, 2009. 95(3): p. 183-9. Sassoon, A.A., N.J. Nelms, and R.T. Trousdale, Intraoperative Fracture During Staged Total Knee Reimplantation in the Treatment of Periprosthetic Infection. J Arthroplasty, 2014. Engh, G.A. and D.J. Ammeen, Classification and preoperative radiographic evaluation: knee. Orthop Clin North Am, 1998. 29(2): p. 205-17. Engh, G.A. and D.J. Ammeen, Bone loss with revision total knee arthroplasty: defect classification and alternatives for reconstruction. Instr Course Lect, 1999. 48: p. 167-75. Mabry, T.M., et al., Revision total knee arthroplasty with modular cemented stems: long-term follow-up. J Arthroplasty, 2007. 22(6 Suppl 2): p. 100-5.
ACCEPTED MANUSCRIPT 12 Figure Legends Figure 1:
Graphic representation of the indications for revision TKA, as encountered in
Graphic representation of the anatomic femoral fracture location by
RI P
Figure 2:
T
this series.
percentages.
Graphic representation of the timing of fracture occurrence, as encountered in
NU
Figure 4:
SC
Figure 3: Graphic representation of the anatomic tibial fracture location by percentages.
this series.
AP (a) and lateral (b) radiographs of a TKA revised for aseptic loosening of
MA
Figure 5:
the tibial component. The patient sustained a tibial shaft fracture
ED
intraoperatively (c-d), which was treated by increasing the length of the
PT
revision tibial stem to bypass the fracture (e-f). The patient subsequently healed his fracture (g-i) and had an excellent clinical result.
CE
Anteroposterior (AP) view of a TKA undergoing revision for instability and osteolysis secondary to polyethylene wear (a). During the revision the
AC
Figure 6:
patient sustained a supracondylar fracture treated with strut grafting and cables (b-c), which unfortunately failed to unite at 24 months (d-e) and was the source of significant pain and functional limitation. The knee underwent re-revision to a distal femoral replacement (f-g), resulting in improvements in pain and function.
ACCEPTED MANUSCRIPT
PT
ED
MA
NU
SC
RI P
T
13
AC
CE
Figure 1
ACCEPTED MANUSCRIPT
PT
ED
MA
NU
SC
RI P
T
14
AC
CE
Figure 2
ACCEPTED MANUSCRIPT
PT
ED
MA
NU
SC
RI P
T
15
AC
CE
Figure 3
ACCEPTED MANUSCRIPT
PT
ED
MA
NU
SC
RI P
T
16
AC
CE
Figure 4
ACCEPTED MANUSCRIPT
Figure 5A
AC
CE
PT
ED
MA
NU
SC
RI P
T
17
ACCEPTED MANUSCRIPT
Figure 5B
AC
CE
PT
ED
MA
NU
SC
RI P
T
18
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
19
Figure 5C
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
20
Figure 5D
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
21
Figure 5E
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
22
Figure 5F
ACCEPTED MANUSCRIPT
ED
MA
NU
SC
RI P
T
23
AC
CE
PT
Figure 5G
ACCEPTED MANUSCRIPT
ED
MA
NU
SC
RI P
T
24
AC
CE
PT
Figure 5H
ACCEPTED MANUSCRIPT
ED
MA
NU
SC
RI P
T
25
AC
CE
PT
Figure 5I
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
26
Figure 6A
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
27
Figure 6B
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
28
Figure 6C
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
29
Figure 6D
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
30
Figure 6E
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
31
Figure 6F
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
MA
NU
SC
RI P
T
32
Figure 6G
ACCEPTED MANUSCRIPT 33
Femora
Tibiae
1
32
35
2
28
35
3
29
19
AC
CE
PT
ED
MA
NU
SC
RI P
AORI Score
T
Table 1: AORI scores encountered at the time of revision TKA
ACCEPTED MANUSCRIPT 34 Table 2: Treatment of Intraoperative Fractures Treatment
Treatment Details
Adjunctive Weight
T
Category
RI P
Bearing
SC
Limitations
Screws/Pins - 31 Wires - 10
NU
Fixation - 43
Suture - 2
MA
Observation - 21
6 3 0 9
Increased stem length/augments - 15
1
up - 16
Increased constraint - 1
0
Strut - 1
1
PT
Bone Grafting - 6
ED
Component build
AC
Combination - 3
CE
Cancellous Chips - 5
0
Fixation + Strut Bone Grafting - 2
2
Fixation + Component build up - 1
0