- Email: [email protected]
A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures
Accepted Manuscript A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures Xing-wen Zhao, Jian-xiong Ma, Xin-long Ma, Xuan Jiang, Yin Wang, Fei Li, Bin Lu PII:
S1743-9191(17)30046-8
DOI:
10.1016/j.ijsu.2017.01.044
Reference:
IJSU 3465
To appear in:
International Journal of Surgery
Received Date: 13 September 2016 Revised Date:
6 January 2017
Accepted Date: 11 January 2017
Please cite this article as: Zhao X-w, Ma J-x, Ma X-l, Jiang X, Wang Y, Li F, Lu B, A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures, International Journal of Surgery (2017), doi: 10.1016/j.ijsu.2017.01.044. 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
A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures Xing-wen Zhao a, b,1. Jian-xiong Ma a,1, *. Xin-long Ma a, b, c,1. Xuan Jiang a, c. Yin
RI PT
Wang a. Fei Li a, b. Bin Lu a.
a. The Orthopaedic Institute, Tianjin Hospital, Tianjin 300050, People’s Republic of
SC
China.
b. Tianjin Medical University, Tianjin 30070, People’s Republic of China.
M AN U
c. Department of Orthopaedics, Tianjin Hospital, Tianjin 300211, People’s Republic of China.
*Corresponding author. The Orthopaedic Institute, Tianjin Hospital, Tianjin 300050,
TE D
People’s Republic of China.
E-mail addresses: [email protected](X.-w. Zhao), [email protected] (J.-x. Ma). [email protected](X-l. Ma)
AC C
EP
1 These authors contributed equally to this work.
January 3, 2017
ACCEPTED MANUSCRIPT 1 2
A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures
3 4
ABSTRACT
5
Both external fixation(ExFx) and open reduction and internal fixation(ORIF) were used to treat
7
complex tibial plateau fractures, but it was not sure which one was better. So we did this
8
meta-analysis to evaluate the outcomes of ExFx and ORIF in managing complex tibial plateau
9
fractures.
RI PT
6
SC
10
Methods
12
Articles published before August 5, 2016 were selected from PubMed, Cochrane library, and some
13
other electronic database. Relevant journals were also searched manually with no language limited.
14
Two independent reviewers searched and assessed the literatures. A fixed effect model was initially
15
used for meta-analyses with RevMan 5.3.
16
M AN U
11
Results
18
When compared with ORIF, cases undergoing ExFx were more likely to return to the preinjury state
19
at the early stage, but no difference in the later period of follow-up. However, ExFx group had higher
20
infection rate (OR 1.98, 95 % CI 1.08-3.63, P=0.03), higher venous thromboembolism rate (OR 1.56,
21
95 % CI 0.49-4.96, P=0.45), higher re-operation rate (OR 0.87, 95 % CI 0.47-1.62, P=0.66) and
22
lower compartment syndrome rate (OR 0.61, 95 % CI 0.12-3.22, P=0.56), lower TKA rate (OR 0.51,
23
95 % CI 0.20-1.34, P=0.17). There were no statistically significant differences in the rate of deep
24
infection, venous thromboembolism, compartment syndrome and VTE between the two groups.
EP
TE D
17
AC C
25 26
Conclusion
27
Although external fixation may offer some advantages, both were acceptable strategies in managing
28
complex tibial plateau fractures. According to our analysis results, we strongly recommend that
29
selection of definitive fixators should base on the fracture patterns, soft-tissue condition as well as
30
the injury stages in clinical practice. More important, further multicentered, randomized controlled
31
studies should be implemented to get a more reliable and clear result.
32 33
Introduction
34
Tibial plateau fractures accounted for 1-2% of all the fractures, approximately 8% of them occured in 1
ACCEPTED MANUSCRIPT elderly[1] and over a half of cases were male[2]. They constitute high-energy injuries with associated
36
insult on the soft tissue envelope[3]. Due to the complex anatomy of the tibial plateau, intra-articular
37
lesions, severe soft-tissue damage, osseous compromise of the proximal tibia and high risk of
38
complications, most scholars regarded the Schatzke type V and VI[4] or the AO/OTA type C (C1,
39
C2, C3) [5] as the complex tibial plateau fractures, which remained a challenge to the surgeons even
40
the most experienced [6].
RI PT
35
41
All kinds of treatment, from conservative treatment to surgical management, were aimed at
43
anatomic reduction of the articular surface, restore of tibial length and alignment and prevent
44
secondary displacement of the fracture fragments [1]. Open reduction and internal fixation (ORIF)
45
with plates and screws through a extensile anterior approach was the first choice to achieve this
46
goal[7], which can direct reduction of fracture and offer an optimal visualization[8]. However, as
47
complex tibial plateau fractures associated with severe soft tissue damage, ORIF often led to higher
48
rate of complications over the past two decades [9]. Despite the evolution of treatment strategies and
49
quality of fixation implants, a poor outcomes were reported continuously [10].
M AN U
SC
42
50
Adequate fixation and early motion were important for a good prognosis and satisfied postoperative
52
rehabilitation, so fire-wire external fixation, like Taylor spatial frame, Ilizarov circular frame,
53
Monticelli-Spinelli circular fixator were good alternative interventions, which allowed for early and
54
adequate initial weight bearing without limitations related to skin condition, was considered as an
55
ideal method to these cases, who cannot use internal fixation due to trauma of the soft tissue envelop,
56
deficiency of bone stock, and bony comminution [11, 12].
EP
57
TE D
51
We performed this meta-analysis to discuss whether the external fixation (ExFx) provided better
59
radiological and clinical outcomes and fewer post-operative complications than open reduction and
60
internal fixation (ORIF) for managing complex tibial plateau fractures.
61
AC C
58
62
Materials and Methods
63
Literature search strategy
64
Electronic searches were performed by using PubMed, Cochrane library, Cochrane Central Register
65
of Controlled Trials (CCTR), China national knowledge internet database, Wan Fang database
66
without restriction for publication date and languages in August 5th, 2016. The following medical
67
subject headings (MeSH) and terms were used to achieve broad and specific searches: “internal
68
fixation”, “external fixation”, “complex tibial plateau fracture”, “Schataker 6”, “Schataker VI”, 2
ACCEPTED MANUSCRIPT 69
“Schataker 5”, “Schataker V” with the Boolean operators AND or OR. Additional records were
70
identified through google search engine or other available databases according to the Cochrane
71
Collaboration guidelines.
72
Inclusion and exclusion criteria
74
A study was included in the analysis as following criteria:
75
(1) Studies on complex tibial plateau fractures and conducted on human subjects (RCTs or
76
quasi-RCTs).
RI PT
73
(2) Studies directly compared the effectiveness of ExFx and ORIF.
78
(3) Reporting the data of outcomes (radiological, clinical and others) and complications (OA,
80 81
infections, DVT and others).
(4) The follows were excluded: reviews, isolated case reports, pathological fractures, biomechanics analyses and others.
M AN U
79
SC
77
82
Quality Assessment
84
The risks of bias were assessed independently by two of the authors, with Randomized control trial
85
were assessed by using the Cochrane Collaboration Risk of Bias Tool[13]. The contents were seven
86
parts---random sequence generation (selection bias), allocation concealment (selection bias),
87
blinding of participants and personnel (performance bias), blinding of outcome assessment data
88
(detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias) and
89
other bias. Each items were recorded as “high risk”, “low risk”, “unclear risk”. For non-RCT, the
90
Methodological Index for Non-Randomized studies (MINORs) scale was applied for quality
91
assessment[14], which scored from 0 to 24. Any disagreement was discussed and resolved with a
92
third independent author.
EP
TE D
83
AC C
93 94
Study selection and data extraction
95
Datum were independently extracted from articles, tables and figures by two investigators, including
96
the first authors, study design, the publication date, sample size, follow-up duration, interventions,
97
outcomes as well as adverse event. Any disagreement was discussed and resolved with a third
98
independent author.
99 100
Data Analysis and Statistical Methods
101
Statistical analyses were conducted with Review Manager Version 5.3 (Cochrane Collaboration,
102
Software Update, Oxford, UK). Statistical heterogeneity was assessed using the I2 value and 3
ACCEPTED MANUSCRIPT 103
chi-square test. When I2 >50%, P<0.1 was considered to be significant heterogeneity, random-effect
104
model was applied for meta-analysis. Otherwise, fixed-effect model was performed. If possible,
105
sensibility analysis was conducted to search the origins of heterogeneity. Dichotomous outcomes
106
were expressed as odds ratios(OR) with 95% confidence intervals (CIs). While continuous outcomes,
107
mean differences (MDs) and 95% confidence intervals (CIs) were calculated.
RI PT
108
Results
110
Search result
111
A total of 672 studies were identified with an initial decision, this yielded 243 titles for initial
112
screening after removal of 429 duplicates. Following initial screening and application of the
113
inclusion/exclusion criteria, there were one RCT and eight retrospectives, a total of 11 articles were
114
carried out on all[15-25]. The characteristics were described as Table 1. Two of them came from a
115
same study, conducted by the Canadian orthopaedic trauma society (COTS)[23, 24], which was also
116
the only RCT. Two came from a same institution, located in Boston[18, 22]. The search process was
117
performed as Figure 1.
M AN U
SC
109
118
Risk of assessment
120
The details about the included studies were summarized in Table 1. Cochrane Handbook for
121
Systematic Review of Interventions was consulted to assess the quality of RCTs. The COST [23, 24],
122
the only RCT, was assessed to be at low risk of bias in almost all terms (Table 2), except for blinding
123
of assessor and reporting bias. Due to there was no blinding of the evaluators and the protocol was
124
not published before recruitment commenced. The risk of bias was assessed for the eight
125
retrospective studies by MINORs [14] and showed as Table 3. Except for two of them were lost to
126
follow-up >5% [21, 25], the others were assessed as high quality.
EP
TE D
119
AC C
127 128
Study characteristics
129
There were 515 individuals with 519 fractures in our pooling of studies (four people suffer from
130
double knee fractures), 239 (46%) fractures in ExFx group and the others (54%) in ORIF. Mckee et
131
al. [23, 24], which was the only RCT, accounted for 83(16%) and used Hospital for Special Surgery
132
knee score (HSS score)[26] to measure the primary outcome, used McMaster Universities
133
Osteoarthritis Index (WOMAC)[27], the quality of radiographic reduction, the presence of
134
degenerative osteoarthritis and the scores on the Short-Form 36(SF-36) [28] to measure the
135
secondary outcome.
136
4
ACCEPTED MANUSCRIPT The other eight retrospective studies[15-17, 19-21, 25] accounted for 436 (84%)of the published
138
cases in our meta-analysis (82% of ExFx and 85% of ORIF). What they used in the ExFx group
139
included Taylor Spatial Frame (TSF), Ilizarov circular frame, Hoffman II with limited internal
140
fixation, Synthes AO fixator or Ilizarov circular frame and so on, while used medial and lateral
141
non-locking buttress plates locking plate and the Synthes Less Invasive Stabilization System (LISS)
142
in ORIF group. They used WOMAC [15, 17, 25], Rasmussen’s system[16, 17], Kellgren-Lawrence
143
score[16], IOWA knee score[19], numeric rating scale (NRS) [17], Knee Injury and Osteoarthritis
144
Outcome Score(KOOS) [20], Honkonen and Jarvinen (HJ) Criteria[25] as well as range of motion
145
(ROM) [21] to measure the outcome of surgeries. Three studies reported the soft tissue damage cases
146
without distinguishing[20, 24, 25], 23 cases had associated with cruciate ligament avulsion, 20 had
147
meniscal ligaments avulsion, 12 had Collateral ligament avulsion, 6 had neurologic injury and 1 had
148
patellar tendon avulsion. Except for Chan et al. [16] reported one (4%) in ORIF and 7(20%) ExFx
149
were alcohol dependency, there was no significant difference between the two groups in baseline
150
information.
M AN U
SC
RI PT
137
151 152
Outcomes of Meta-Analysis
153
The results of meta-analysis were showed as Table 4.
154
Hospital stay time
156
The COST [23, 24] and Conserva et al. [17] were the only two studies reported the mean time of
157
hospital stay. Interestingly, the results of the two studies were almost the same: the ExFx group were
158
Significant shorter than ORIF group (9.9 to 23.4 days in COST, P = 0.024; 7.8 to 14.2 days in
159
Conserva et al., P = 0.002).
EP
TE D
155
160
Blood loss and surgery time
162
Only the COST [19,20] give the volume of blood loss in surgery and surgery time, it seemed that
163
ExFx group took less time in surgery (170min to 183min, P = 0.229), and less blood loss (213ml to
164
544ml, P = 0.006).
165
AC C
161
166
Radiographic assessment
167
Chan et al.[16] and Conserva et al. [17]used Rasmussen’s system, which was based on joint
168
depression, condylar widening, and varus/valgus angulation. But these studies reported no
169
significant differences (28.81 to 31.73 in Krupp et al., u = 0.512; 25 to 24.9 in Conserva et al., P =
170
0.47). Jansen et al. [20]used the Kellgren-Lawrence score, but they only given the average score 5
ACCEPTED MANUSCRIPT 171
(66.2) without distinguishing between the two groups.
172
Two studies [17, 21] compared the time to radiographic union: Krupp et al. reported 7.4 months
174
(range 3-15) in ExFx group and 5.9 months (range 3-14) in ORIF group; Conserva et al. reported
175
15.9 months (range 7.5-32) in ExFx group and 17.2 months (range 9.1-45) in ORIF group. Krupp et
176
al. reported 13 (5%) in ExFx and 4(1%) in ORIF were malunion, while 11 in ExFx (5%) and 7 in
177
ORIF (3%) were union after 6 months based on radiography evidence.
RI PT
173
178
There were 67 fractures (13% of total fractures) assessed of osteoarthritis (OA) [16, 17, 19, 21, 23,
180
24] based on radiography. The COTS and Chan et al. both relied on subjective assessment by a
181
single unblinded assessor to do the assessment and used radiographs taken after the same follow-up
182
period (24 months postoperative). The analysis results were showed as Figure 2. (OR 1.52, 95% CI
183
0.86-2.66, P = 0.15).
M AN U
SC
179
184
Functional outcomes
186
Six studies reported the functional outcomes[15-17, 20, 21, 23-25]. The COST reported the ExFx
187
group seemed higher HSS than ORIF group (72 to 61, 95% CI 2.03-19.97, P = 0.064) in 6 months
188
after-operative, but there was no difference at 12 months (72 to 67, 95% CI -2.59-12.59, P = 0.406)
189
and 24 months (75 to 68, 95% CI -1.45-15.45, P = 0.307). Also, it seemed that the ROM of flexion
190
and the extension in ExFx group was better than ORIF group (flexion: 123° to 113°, P=0.114;
191
extension: 3° to 4°, P=0.449). This result was almost the same as Krupp et al. [21]. Jansen et al.
192
[20]reported outcomes by Lysholm score and KOOS as the average of the two groups without
193
distinguishing. Three studies’ results of WOMAC showed no significant difference between
194
the two groups [15, 17, 25]. And the SF-36 in two studies[23-25] also showed no significant
195
difference. In terms of pain value, Ahearn et al. [15]used VAS and Conserva et al. [17]used
196
NRS, which was no significant difference in their reports.
EP
AC C
197
TE D
185
198
Re-operation rate after surgery
199
As Figure 3A. showed, four studies reported the re-operation rate after surgery [15, 17, 21, 23, 24],
200
which included knee amputation, osteotomy, pin-track debridement[23, 24], screw pull-out[17],
201
however, no study took the planned frame removal into account during analyses. There were 28
202
cases (12%) in ExFx group and 30 cases (11%) in ORIF group, but the results were not statistically
203
significant (OR 0.87, 95% CI 0.47–1.62, P=0.66). The COST trial reported 27 frames were removed
204
and their data indicated that ORIF group had higher the re-operation rate than ExFx group. 6
ACCEPTED MANUSCRIPT 205
Knee arthroplasty after surgery
207
Five studies (403 fractures,78% of total) reported on subsequent knee arthroplasty (TKA) [16, 17, 19,
208
21, 23, 24]. Figure 3B. showed that the cases of TKA was 7(3%) in the ExFx and 12(4.2%) in the
209
ORIF groups (OR 0.51, 95% CI 0.20–1.34, P=0.17). Guryel et al. [19] didn’t give us a clear time of
210
follow-up, so we cannot know the detail of subsequent knee arthroplasty. Also, due to the range of
211
follow-up of other four studies (6-72 months) and the subsequence TKA occurred in different stages,
212
which made it difficult to evaluate how many cases required total knee arthroplasty exactly.
RI PT
206
213
Complications
215
Infections
216
All the eight retrospective studies reported the infection cases in the follow-up period (436
217
fractures)[15-22, 25]. As Figure 4. showed, ExFx group had higher infective rate than ORIF group
218
both in total infective rate (17% versus 8%, OR 1.98, 95% CI 1.08–3.63, P= 0.03, Fig.4A),
219
superficial infective rate (13% versus 5%, OR 2.72, 95% CI 1.31–5.65, P=0.008, Fig.4B) and deep
220
infective rate (5% versus 4%, OR 1.08, 95% CI 0.48–2.46, P=0.85, Fig.4C).
M AN U
SC
214
221
Compartment syndrome
223
Figure 5A. showed the compartment syndrome in two studies [16, 20], 4.5% in ExFx group and 8.9%
224
in ORIF group (OR 0.61, 95% CI 0.12–3.22, P=0.56).
TE D
222
225
Venous thromboembolism (VTE)
227
Three studies [15, 16, 19] described the rate of venous thromboembolism (VTE), with no statistically
228
significant differences between the two groups (OR 1.56, 95% CI 0.49–4.96, P= 0.45, Fig.5B).
EP
226
AC C
229 230
Joint stiffness
231
Krupp et a. [21]and Conserva et al.[17] reported eight cases of joint stiffness (7 in ExFx group and
232
one in ORIF group), with statistically significant differences between two groups (OR 5.14, 95% CI
233
0.85–30.91, P= 0.07, Fig.5C).
234 235
Discussion
236
Notoriously, complex tibial plateau fractures were a challenging fractures associated with severe bone
237
and soft tissue injury, which led to high complication rates and poor clinical outcomes [6]. Since the
238
publication of the first study presented in our meta-analysis in 1992[18, 22], the management of it has 7
ACCEPTED MANUSCRIPT 239
witnessed new developments. In these years, different methods have been proposed and utilized in
240
clinic, including limited open reduction and stabilization system, contoured locking periarticular plates,
241
hybrid external fixation and so on. And several literatures reported the good efficacy of these methods
242
[29-31].
243
External fixation , which preserve soft tissues and reduce the rate of deep infection, has led many
245
surgeons to apply it as a minimally invasive surgery to treat complex tibial plateau fractures[4, 31].
246
Catagni et al.[32] reported the use of circular external fixator for Schatzker V and VI tibial plateau
247
fractures and received good results. Katsenis et al.[33] got the similar results when used small wire
248
transfixion frames. Open reduction and internal fixation was successful in restoring articular
249
congruity and further compromise the soft tissue envelope. However, some case series have
250
highlighted the dangers of complications following ORIF. Young et al. [9]reported the 88% deep
251
infections rates with dual plating, the similar higher deep infections rate of ORIF was also reported
252
by Jensen et al. [34] and Mckee et al. [24]. When comparing the two surgery methods, the main
253
foundings in these studies were divided. Mckee et al. [24]and Hall et al. [35]reported external fixator
254
trend towards better results, but Krupp et al.[21] and Ahearn et al. [15] found no significant
255
differences between the two methods. So we tried to get a uniform conclusion on whether external
256
fixation can provide better outcomes with fewer complications when compared to the open reduction
257
and internal fixation for complex tibial plateau fractures.
SC
M AN U
TE D
258
RI PT
244
One potential bias was that only one RCT (16% fractures) and eight retrospective studies (84%
260
fractures) have directly compared the ExFx and ORIF in treating complex tibial plateau fractures.
261
However, they were high quality research in which the main focus was on clinical outcomes, we also
262
could get some valuable and reliable information. Firstly, we found almost half of the cases were
263
high energy trauma fracture, like motor vehicle accident or fall from a height[15, 17, 25] and
264
numbers of patients associated with ligaments injuries[20, 24, 25]. Secondly, the severity of
265
intra-articular displacement and soft tissue compromise has a significant impact on the choice of
266
operation method and the timing of surgery[25, 36].
AC C
267
EP
259
268
Initially, we hoped to test the functional outcomes of the two methods, nevertheless, various
269
evaluation system made it difficult to performance this compare. Except for the COST, which
270
reported a higher HSS at the early stage and better ROM in ExFx group [23, 24, 35], the others
271
showed no significant difference in terms of SF-36, ROM, and WOMAC. Maybe in the further
272
studies, we can take a uniform evaluation system to make the compared more reliable. 8
ACCEPTED MANUSCRIPT 273
Theoretically, more minimal invasive surgery, like Taylor Spatial Frame or Ilizarov circular
275
frame, would be more helpful for the healing. Indeed, Conserva et al.[17] and the COST trail
276
[21, 23, 24]reported that individual who undergoing ExFx seemed had less blood loss, shorter
277
surgery time, faster return to work or pre-injury activities. Krupp et al. [21] reported the average time
278
to union was shorter in ORIF group (7.4 months in ExFx and 5.9 months ORIF ), however, it was
279
conflict with Conserva et al.[17]. We thought it maybe result from the type of plates they used and
280
the reduction skill of surgeons.
RI PT
274
281
As the analysis results showed above, we could know that the ExFx group seemed had higher
283
osteoarthritis(OA) rate, lower subsequence total knee arthroplasty(TKA) rate and lower re-operations
284
[16, 17, 19, 21, 23, 24], however, the results were no statistical difference and the small sample size
285
and the short follow-up duration made us cannot sure the rate of OA and subsequence TKA exactly,
286
which were a long-term complication of intra-articular fractures that can occur in any periods after
287
trauma. And no studies reported the cases who planned to remove the implants. Although, the data
288
showed ExFx had higher cost effectiveness, we should notice that higher advert event rates and the
289
further cost to deal with them in using external fixation.
M AN U
SC
282
290
Infections, or other advert events, in theory, was superior in open reduction and internal fixation.
292
However, our results were against to this concept. As compared to ORIF group, the rate of infection
293
(both superficial and deep infection), venous thromboembolism (VTE) and joint stiffness were
294
higher in the ExFx group. Fortunately, no evidences showed there were serious infections occurred
295
as well as the compartment syndrome, and no studies reported the Pulmonary embolism (PE).
EP
TE D
291
296
Only one RCT, small sample size, various assessment tools and short follow-up duration made it
298
difficult to achieve our goal which was to recommend the better surgical method in treating complex
299
tibial plateau fractures. Fortunately, the pooled studies and accumulated datum were relatively high
300
quality, we also got some valuable conclusion.
301
AC C
297
302
Conclusion
303
Based on our analysis results, external fixation may offer some advantages in terms of soft tissue
304
healing. However, with the use of minimally invasive internal fixation locking plates, the superiority
305
of external fixation was no longer obvious. The currently existing evidences didn’t give us a clear
306
recommendation on whether ExFx was better than ORIF in managing complex tibial plateau 9
ACCEPTED MANUSCRIPT fractures, and longer duration studies on comparing ORIF and ExFx were required. Although
308
orthopedists should be mindful the complications and the subsequence procedures after surgery, both
309
ExFx and ORIF were acceptable strategies in managing complex tibial plateau fractures. In clinical
310
practice, we should select proper fixator based on the fracture patterns, soft-tissue condition as well
311
as the injury stages. Meanwhile, further multicentered, randomized controlled studies should be
312
implemented to get a more reliable and clear result.
RI PT
307
313 314 315
SC
316 317 318
M AN U
319 320 321 322 323 324
TE D
325 326 327
EP
328 329 330
AC C
331 332 333 334 335 336 337
Table 1. Characteristics of including studies
338 339
Table 2. Quality of the randomized controlled trials (the Cochrane Collaboration Risk of Bias
340
Tool) 10
ACCEPTED MANUSCRIPT 341 342
Table 3. Quality assessment for non-randomized trials (MINORs)
343 344
Table 4. Results of meta-analysis
345
Figure 1. Flowchart of the study selection process
RI PT
346 347 348
Figure 2. Forest plot showing the meta-analysis of the osteoarthritis(OA)
349
Figure 3. Forest plot showing the meta-analysis of re-operation (A) and total knee
351
arthroplasty(B)
SC
350
352
Figure 4. Forest plot showing the meta-analysis of the rate of post-operative infections(A), the
354
rate of superficial infections(B), the rate of deep infections(C)
M AN U
353
355 356 357 358
TE D
359 360 361
EP
362 363 364
AC C
365 366 367 368 369 370 371
References
372 373
1.
AD. WJaW. Rockwood and Green's fractures in adults: Fractures of the proximal tbial and fibula Bucholz
374
RW and Heckman JD Ed 5th ed Philadelphia: Lip-pincott Williams and Wilkins2001;2:1799-839. 11
ACCEPTED MANUSCRIPT 375
2.
Albuquerque RP, Hara R, Prado J, Schiavo L, Giordano V, do Amaral NP. Epidemiological study on tibial
376
plateau fractures at a level I trauma center. Acta Ortopedica Brasileira2013;21(2–:109-15.
377
3.
378
International Wound Journal2014;11 Suppl 1(s1–:3 5.
379
4.
380
Clinical orthopaedics and related research1979 Jan-Feb(138–:94-104.
381
5.
382
Springer Berlin Heidelberg; 1990.
383
6.
384
treatment of tibial plateau fractures: Five- to 27-year follow-up results (Journal of Orthopaedic Trauma (2007–
385
21, (5-10––. 2007.
386
7.
387
complex
388
trauma1992;6(1–:78-86.
389
8.
390
comparison of treatment methods. Journal of Trauma1981;21(5–:376-81.
391
9.
392
review1994;23(23–:149-54.
393
10. Nikolaou VS, Tan HB, Haidukewych G, Kanakaris N, Giannoudis PV. Proximal tibial fractures: early
394
experience using polyaxial locking-plate technology. Injury Extra2011;35(8–:1215-21.
395
11. Dendrinos GK, Kontos S, Katsenis D, Dalas A. Dendrinos GK, Kontos S, Katsenis D, et al. Treatment of
396
high-energy tibial plateau fractures by the Ilizarov circular fixator. Bone & Joint Journal1996;78.
397
12. Subasi M, Kapukaya A, Arslan H, Ozkul E, Cebesoy O. Outcome of open comminuted tibial plateau
398
fractures treated using an external fixator. Journal of Orthopaedic Science2007;12(4–:347-53.
399
13. Higgins J, Green SE. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The
400
Cochrane Collaboration (Eds–. 2011;2011.
401
14. Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y, Chipponi J. Methodological index for non-randomized
402
studies ( MINORS –: development and validation of a new instrument. Anz Journal of Surgery2003;73(9–:712-6.
403
15. Ahearn N, Oppy A, Halliday R, Rowett-Harris J, Morris SA, Chesser TJ et al. The outcome following
404
fixation of bicondylar tibial plateau fractures. The bone & joint journal2014 Jul;96-b(7–:956-62.
405
16. Chan C, Keating J. Comparison of outcomes of operatively treated bicondylar tibial plateau fractures by
406
external fixation and internal fixation. Malaysian orthopaedic journal2012 Mar;6(1–:7-12.
407
17. Conserva V, Vicenti G, Allegretti G, Filipponi M, Monno A, Picca G et al. Retrospective review of tibial
408
plateau fractures treated by two methods without staging. Injury2015 Oct;46(10–:1951-6.
Brem MH, Bail HJ, Biber R. Value of incisional negative pressure wound therapy inorthopaedic surgery.
RI PT
Schatzker J, McBroom R, Bruce D. The tibial plateau fracture. The Toronto experience 1968--1975.
Müller MME, Koch MP, Nazarian MS. The Comprehensive Classification of Fractures of Long Bones:
SC
Rademakers MV, Kerkhoffs GMMJ, Sierevelt IN, Raaymakers ELFB, Marti RK. Erratum: Operative
Benirschke SK, Agnew SG, Mayo KA, Santoro VM, Henley MB. Immediate internal fixation of open, plateau
fractures:
treatment
by
a
standard
protocol.
M AN U
tibial
Journal
of
orthopaedic
Waddell JP, Johnston DW, Neidre A. Fractures of the tibial plateau: a review of ninety-five patients and
AC C
EP
TE D
Young MJ, Barrack RL. Complications of internal fixation of tibial plateau fractures. Orthopaedic
12
ACCEPTED MANUSCRIPT 18. Covall DJ, Fowble CD, Foster TE, Whitelaw GP. Bicondylar tibial plateau fractures: principles of treatment.
410
Contemporary orthopaedics1994 Feb;28(2–:115-22.
411
19. Guryel E. MS, Sharma R., Newman K., Elliott D., Khaleel A. THE CHERTSEY CLASSIFICATION OF
412
TIBIAL PLATEAU FRACTURES AND A COMPARISON OF THE OUTCOME OF TREATMENT WITH
413
INTERNAL FIXATION OR AN ILIZAROV FINE WIRE CIRCULAR FRAME
414
Orthopaedic Proceedings2010.
415
20. Jansen H, Frey SP, Doht S, Fehske K, Meffert RH. Medium-term results after complex intra-articular
416
fractures of the tibial plateau. Journal of Orthopaedic Science2013 2013/07/01;18(4–:569-77.
417
21. Krupp RJ, Malkani AL, Roberts CS, Seligson D, Crawford CH, 3rd, Smith L. Treatment of bicondylar tibia
418
plateau fractures using locked plating versus external fixation. Orthopedics2009 Aug;32(8–.
419
22. Mallik AR CD, Whitelaw GP. Internal versus external fixation of bicondylar tibial plateau fractures. Orthop
420
Rev1992.
421
23. S. Pirani MM. RANDOMIZED TRIAL OF BICONDYLAR TIBIAL PLATEAU FRACTURES TREATED WITH
422
AO OR RING FIXATOR METHODS: EARLY REINTERVENTION RATES FOR COMPLICATIONS.
423
Orthopaedic Proceedings2008.
424
24. SP MMP. Open reduction and internal fixation compared with circular fixator application for bicondylar
425
tibial plateau fractures. Results of a multicenter, prospective, randomized clinical trial. The Journal of bone and
426
joint surgery American volume2006 Dec;88(12–:2613-23.
427
25. Tul B Pun VPK, Pradeep M Poonnoose, Anil T Oommen, and Ravi J Korula. Outcome of Schatzker type V
428
and VI tibial plateau fractures. 2014.
429
26. Caplan N, Kader DF. A Comparison of Four Models of Total Knee-Replacement Prostheses. Journal of
430
Bone & Joint Surgery American Volume1976;58(6–:754-65.
431
27. Guermazi M, Poiraudeau S. Translation, adaptation and validation of the Western Ontario and McMaster
432
Universities Osteoarthritis Index (WOMAC– for an Arab population: the Sfax modified WOMAC. Osteoarthritis &
433
Cartilage2004;12(6–:459-68.
434
28. Jenkinson C, Coulter A, Wright L. Short Form 36 (SF-36– health service questionnaire. Normative data for
435
adults of working age. Bmj Clinical Research1993;306(6890–:1437-40.
436
29. Orthopaedics JCHXPLXWDYDo. Open Reduction and Internal Fixation of High-energy Tibial Plateau
437
Fractures. Orthopedic Clinics of North America2003;33(1–:177-98.
438
30. Jeray KJ, Lochow SC. Staged Open Treatment of High-Energy Tibial Plateau Fractures. Techniques in
439
Knee Surgery2005;4(4–:214-25.
440
31. Babis GC, Evangelopoulos DS, Kontovazenitis P, Nikolopoulos K, Soucacos PN. High energy tibial
441
plateau
442
Research2011;6(1–:1-7.
AC C
EP
TE D
M AN U
SC
RI PT
409
fractures
treated
with
hybrid
external
13
fixation.
Journal
of
Orthopaedic
Surgery
&
ACCEPTED MANUSCRIPT 443
32. Catagni MA, Ottaviani G, Maggioni M. Treatment strategies for complex fractures of the tibial plateau with
444
external circular fixation and limited internal fixation. Journal of Trauma2007;63(5–:1043-53.
445
33. Katsenis D, Athanasiou V, Megas P, Tyllianakis M, Lambiris E. Minimal internal fixation augmented by
446
small
447
trauma2005;19(4–:241-8.
448
34. Jensen DB, Rude C, Duus B, Bjerg-Nielsen A. Tibial plateau fractures. A comparison of conservative and
449
surgical treatment. Bone & Joint Journal1990;72(1–:49-52.
450
35. Hall JAB, M. J. McKee, M. D. Open reduction and internal fixation compared with circular fixator
451
application for bicondylar tibial plateau fractures. Surgical technique. The Journal of bone and joint surgery
452
American volume2009 Mar 1;91 Suppl 2 Pt 1:74-88.
453
36. Kokkalis ZT, Iliopoulos ID, Pantazis C, Panagiotopoulos E. What's new in the management of complex
454
tibial plateau fractures? Injury-international Journal of the Care of the Injured2016;47(6–:1162-69.
frames
for
high-energy
tibial
plateau
fractures.
Journal
AC C
EP
TE D
M AN U
455
14
of
orthopaedic
RI PT
transfixion
SC
wire
ACCEPTED MANUSCRIPT Dear editor: The study was supported by the National Natural Science Foundation of China (NO. 81572154), Tianjin Health Bureau Science and Technology Foundation (NO. 15KG123), and Key project of Tianjin science and technology (NO.13ZCZDSY01700). We were grateful to all the participating patients in these trails and all the staffs in these studies. Thanks very much for your attention to our paper. Sincerely yours,
RI PT
Xing-wen Zhao Corresponding author: Jian-xiong MA E-mail [email protected]
AC C
EP
TE D
M AN U
SC
January 6, 2017
1
ACCEPTED MANUSCRIPT
Table 2. Quality of the randomized controlled trials (the Cochrane Collaboration Risk of Bias Tool)
COTS
Random sequence
Allocation
Blinding of
Blinding of
Incomplete
Selective reporting
generation
concealment
participants and
outcome
outcome data
(reporting bias)
（selection bias） ）
（selection bias） ）
personnel
assessment
(attrition bias)
(performance bias)
(detection bias)
Low risk
High risk
Low risk
Low risk
Mckee et al. (2006)
AC C
EP
TE D
M AN U
SC
Pirani et al. (2008)
RI PT
Studies
1
Low risk
Unclear risk
Others
Low risk
ACCEPTED MANUSCRIPT
Table 3. Quality assessment for non-randomized trials (MINORs) Quality assessment for Non-RCT
Boston (
Krupp et
Guryel et
Chan et
Jansen et
Pun TB et
Ahearn et
Conserva et
Covall et al.
al. (2009)
al. (2010)
al. (2012)
al. (2013)
al. (2014)
al. (2014)
al. (2015)
Mallik et al. （1992） ）) 2
2
2
2
2
2
2
Inclusion of consecutive patients
2
2
2
2
2
2
2
2
Prospective data collection
0
0
0
0
0
0
0
0
Endpoints appropriate to the aim of the study
2
2
2
2
2
2
2
2
Unbiased assessment of the study endpoint
2
2
2
2
2
2
2
2
Follow-up period appropriately to the aims of
2
2
2
2
2
2
2
2
Less than 5 % loss to follow-up
2
1
2
2
2
2
1
2
Prospective calculation of the sample size
0
0
0
0
0
0
0
0
An adequate control group
2
2
2
2
2
2
2
2
Contemporary groups
2
2
2
2
2
2
2
2
Baseline equivalence of groups
1
2
2
2
2
1
2
2
Adequate statistical analyses
2
2
2
2
2
2
2
2
Total score
19
19
20
20
20
20
17
20
AC C
EP
TE D
study
SC
2
M AN U
A clearly stated aim
RI PT
(1994) and
1
ACCEPTED MANUSCRIPT
Studies
Boston
Country
U.S.
Design
Non-RCT
Mean Ages
Cases
Fracture
Interventions (knees and methods)
(years)
(knees)
types(knees)
45 years
17(17)
NM
Covall et
Monticelli-Spinellib
Bilateral buttress,
Complications,
semi-tubular
re-operations
ORIF: 33 months
Medial and lateral
Primary: HSS
6 12 24 months
cannulated screws (7)
EF: 46.2
82(83)
V-18, VI-65
Percutaneous lag
SC
RCT
2
C1-20,
screw, and Ilizarov
non-locking
Secondary: WOMAC,
buttress plates ±
reoperations,
bone grafting (40)
complications, SF-36,
years
al. (2006)
ORIF: 43.3
C2-39,
circular frame (43)
Pirani et
years
C3-24
M AN U
Canada
1
Mckee et
al. (2008) U.S.
Non-RCT
EF: 48.8
1
58(58)
V-37
years
al. (2009)
VI-21
ORIF: 46.64
U.K.
Non-RCT
NM
NM (124)
al. (2010)
3
radiological results
Hoffman II Hybrid
Dual plating (8),
Clinical outcomes:
EF:16.4 months
(16),
Lateral locking
union rate, time to
ORIF: 10.2
circular frames (14)
plate + medial
union, malunion,
months
screws or buttress
ROM, complications,
plate (20)
re-operations
NM (79)
IOWA knee score,
TE D
years Guryel et
EF:10 months
plates, or
al. (1992)
Krupp et
(months)
circular fixator (10)
Mallik et COTS
Follow-up
ORIF
al. (1994)
4
Outcome Measures
ExFx
RI PT
Table 1. Characteristics of including studies
Vagus-62,
Ilizarov circular
varus-14,
frame (45)
NM
ROM, complications
U.K.
Non-RCT
EF:52.03 years
(2012)
58(59)
AC C
Chan et al.
EP
axial-48
ORIF:45.04
2
C1-18,
Ilizarov circular
Buttress plate
Complications,
C2-2,
frame (23), Hoffman
(21), LISS (4)
re-operations
C3-39
II with limited
years Jansen et al. (2013)
Germany
Non-RCT
46 years
22(23)
3,6,12,24 months
internal fixation (13) 2
C1-7,
Ilizarov circular
LISS± additional
KOOS, IKDC,
C2-7,
frame (2)
plates± artificial
Lysholm Score,
bone substitute
Complications, ROM
C3-9
(21) 1
67 months
Pun TB et
Indian
Non-RCT
43.85 years
ACCEPTED MANUSCRIPT V-11, Circular frame + VI-10 medial percutaneous screws (12)
1
21(21)
al. (2014)
dual plate (9)
HJS (clinical,
30 months
functional, radiological outcomes) WOMAC, infection rates
U.K.
Non-RCT
44 years
55(55)
C3-30,
Others were
al. (2014)
Taylor Spatial Frame
lateral locking
Primary: WOMAC,
(21)
plate ± medial
SF-36
plate fixation (34)
Secondary:
RI PT
Ahearn et
2
Conserva
Italy
Non-RCT
54.1years
79(79)
1IV-23,
et al.
V-29,
(2015)
VI-27
M AN U
SC
unknown
Percutaneous lag
LCP±bone graft
screw+ hybrid
substitute (38)
external fixator (41)
EF:31 months ORIF:40.5 months
satisfaction scale, VAS, complications, reoperations, radiological outcomes Rasmussen,
EF:39.4 months
WOMAC, NRS,
ORIF:35.1months
complications, reoperations
TE D
Abbreviations: KOOS, Knee Injury and Osteoarthritis Outcome Score. IKDC, International Knee Documentation Committee score. LCP, locking compression plate. NRS, Numeric Rating Scale. HSS, Hospital for special surgery knee score. WOMAC, Western Ontario and McMaster Universities Arthritis Index. ROM, Range Of Motion. HJS, Honkonen and Jarvinen score. SF-36, Short Form 36. VAS, Visual Analogue Score. LISS, Less Invasive Stabilization System. COST, Canadian Orthopaedic Trauma Society. U.S., United State. U.K., United Kingdom. ExFx, External Fixation. ORIF, Open Reduction and Internal Fixation. RCT, Randomized Control Trial. Non-RCT, Non-Randomized Control Trial. NM, Not Mentioned.
AC C
EP
Note: 1, Schatzker classification; 2, Orthopaedic Trauma Association classification. 3, The Chertsey classification of tibial plateau fractures. 4, Multicenter randomized controlled trial.
2
ACCEPTED MANUSCRIPT
Outcome
Studies
Overall effect Effect estimate
95%CI
RI PT
Table 4. Results of meta-analysis Heterogeneity 2
P value
I (%)
P value
0.55
0
0.15
0.31
17
0.66
5
1.52
[0.86,2.66]
Re-operation
4
0.87
[0.47,1.62]
Subsequence TKA
5
0.51
[0.20,1.34]
0.99
0
0.17
Infection rate
8
1.98
[1.08,3.63]
0.26
22
0.03
Superficial infection
8
2.72
[1.31,5.65]
0.44
0
0.008
Deep infection rate
8
1.08
[0.48,2.46]
0.11
43
0.85
Compartment syndrome
2
0.61
[0.12,3.22]
0.35
0
0.56
VTE
3
1.56
[0.49,4.96]
0.13
50
0.45
Joint stiffness
2
5.14
[0.85,30.91]
0.78
0
0.07
M AN U
AC C
EP
TE D
rate
SC
Osteoarthritis (OA)
1
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 1. Flowchart of the study selection process
1
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 2. Forest plot showing the meta-analysis of the osteoarthritis(OA)
1
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
Figure 3. Forest plot showing the meta-analysis of re-operation (A) and total knee arthroplasty(B)
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 4. Forest plot showing the meta-analysis of the rate of post-operative infections(A), the rate of superficial infections(B), the rate of deep infections(C)
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 5 Forest plot showing the meta-analysis of the rate of compartment syndrome(A), the rate of venous thromboembolism(B), and the rate of knee joint stiffness(C).
1
ACCEPTED MANUSCRIPT We aimed to evaluate the effectiveness of external fixation(ExFx) and open reduction and internal fixation(ORIF) in treating complex tibial plateau fractures. According to our results, ExFx had some advantages when compared with ORIF, but there were no
RI PT
statistically significant differences. So we strongly recommend that selection of definitive fixators and time of intervention should base on the fracture patterns,
soft-tissue condition as well as the injury stages in clinical practice. What’s more,
SC
due to the limits of this review, we recommended further researches were needed to
AC C
EP
TE D
M AN U
achieve high quality and credible results.
1
ACCEPTED MANUSCRIPT
International Journal of Surgery Author Disclosure Form
Please state any conflicts of interest
SC
The authors declared that they have no conflict of interest.
RI PT
The following additional information is required for submission. Please note that failure to respond to these questions/statements will mean your submission will be returned. If you have nothing to declare in any of these categories then this should be stated.
M AN U
Please state any sources of funding for your research None of the authors have any financial funding.
TE D
Please state whether Ethical Approval was given, by whom and the relevant Judgement’s reference number
EP
No.
AC C
Research Registration Unique Identifying Number (UIN) Please enter the name of the registry and the unique identifying number of the study. You can register your research at http://www.researchregistry.com to obtain your UIN if you have not already registered your study. This is mandatory for human studies only.
reviewregistry111
1
ACCEPTED MANUSCRIPT
SC
RI PT
Author contribution Please specify the contribution of each author to the paper, e.g. study design, data collections, data analysis, writing. Others, who have contributed in other ways should be listed as contributors. Xing-wen Zhao and Xin-long Ma did the design. Xing-wen Zhao, Jian-xiong Ma, Xuan Jiang, Yin Wang, Fei Li, and Bin Lu did the data collection. Xing-wen Zhao and Jian-xiong Ma did data analysis and writing.
M AN U
Guarantor The Guarantor is the one or more people who accept full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
EP
TE D
Jian-xiong MA
Thanks very much for your attention to our paper.
AC C
Sincerely yours, Xing-wen Zhao
Corresponding author: Jian-xiong MA E-mail：[email protected] January 6, 2017
2
S1743-9191(17)30046-8
DOI:
10.1016/j.ijsu.2017.01.044
Reference:
IJSU 3465
To appear in:
International Journal of Surgery
Received Date: 13 September 2016 Revised Date:
6 January 2017
Accepted Date: 11 January 2017
Please cite this article as: Zhao X-w, Ma J-x, Ma X-l, Jiang X, Wang Y, Li F, Lu B, A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures, International Journal of Surgery (2017), doi: 10.1016/j.ijsu.2017.01.044. 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
A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures Xing-wen Zhao a, b,1. Jian-xiong Ma a,1, *. Xin-long Ma a, b, c,1. Xuan Jiang a, c. Yin
RI PT
Wang a. Fei Li a, b. Bin Lu a.
a. The Orthopaedic Institute, Tianjin Hospital, Tianjin 300050, People’s Republic of
SC
China.
b. Tianjin Medical University, Tianjin 30070, People’s Republic of China.
M AN U
c. Department of Orthopaedics, Tianjin Hospital, Tianjin 300211, People’s Republic of China.
*Corresponding author. The Orthopaedic Institute, Tianjin Hospital, Tianjin 300050,
TE D
People’s Republic of China.
E-mail addresses: [email protected](X.-w. Zhao), [email protected] (J.-x. Ma). [email protected](X-l. Ma)
AC C
EP
1 These authors contributed equally to this work.
January 3, 2017
ACCEPTED MANUSCRIPT 1 2
A meta-analysis of external fixation versus open reduction and internal fixation for complex tibial plateau fractures
3 4
ABSTRACT
5
Both external fixation(ExFx) and open reduction and internal fixation(ORIF) were used to treat
7
complex tibial plateau fractures, but it was not sure which one was better. So we did this
8
meta-analysis to evaluate the outcomes of ExFx and ORIF in managing complex tibial plateau
9
fractures.
RI PT
6
SC
10
Methods
12
Articles published before August 5, 2016 were selected from PubMed, Cochrane library, and some
13
other electronic database. Relevant journals were also searched manually with no language limited.
14
Two independent reviewers searched and assessed the literatures. A fixed effect model was initially
15
used for meta-analyses with RevMan 5.3.
16
M AN U
11
Results
18
When compared with ORIF, cases undergoing ExFx were more likely to return to the preinjury state
19
at the early stage, but no difference in the later period of follow-up. However, ExFx group had higher
20
infection rate (OR 1.98, 95 % CI 1.08-3.63, P=0.03), higher venous thromboembolism rate (OR 1.56,
21
95 % CI 0.49-4.96, P=0.45), higher re-operation rate (OR 0.87, 95 % CI 0.47-1.62, P=0.66) and
22
lower compartment syndrome rate (OR 0.61, 95 % CI 0.12-3.22, P=0.56), lower TKA rate (OR 0.51,
23
95 % CI 0.20-1.34, P=0.17). There were no statistically significant differences in the rate of deep
24
infection, venous thromboembolism, compartment syndrome and VTE between the two groups.
EP
TE D
17
AC C
25 26
Conclusion
27
Although external fixation may offer some advantages, both were acceptable strategies in managing
28
complex tibial plateau fractures. According to our analysis results, we strongly recommend that
29
selection of definitive fixators should base on the fracture patterns, soft-tissue condition as well as
30
the injury stages in clinical practice. More important, further multicentered, randomized controlled
31
studies should be implemented to get a more reliable and clear result.
32 33
Introduction
34
Tibial plateau fractures accounted for 1-2% of all the fractures, approximately 8% of them occured in 1
ACCEPTED MANUSCRIPT elderly[1] and over a half of cases were male[2]. They constitute high-energy injuries with associated
36
insult on the soft tissue envelope[3]. Due to the complex anatomy of the tibial plateau, intra-articular
37
lesions, severe soft-tissue damage, osseous compromise of the proximal tibia and high risk of
38
complications, most scholars regarded the Schatzke type V and VI[4] or the AO/OTA type C (C1,
39
C2, C3) [5] as the complex tibial plateau fractures, which remained a challenge to the surgeons even
40
the most experienced [6].
RI PT
35
41
All kinds of treatment, from conservative treatment to surgical management, were aimed at
43
anatomic reduction of the articular surface, restore of tibial length and alignment and prevent
44
secondary displacement of the fracture fragments [1]. Open reduction and internal fixation (ORIF)
45
with plates and screws through a extensile anterior approach was the first choice to achieve this
46
goal[7], which can direct reduction of fracture and offer an optimal visualization[8]. However, as
47
complex tibial plateau fractures associated with severe soft tissue damage, ORIF often led to higher
48
rate of complications over the past two decades [9]. Despite the evolution of treatment strategies and
49
quality of fixation implants, a poor outcomes were reported continuously [10].
M AN U
SC
42
50
Adequate fixation and early motion were important for a good prognosis and satisfied postoperative
52
rehabilitation, so fire-wire external fixation, like Taylor spatial frame, Ilizarov circular frame,
53
Monticelli-Spinelli circular fixator were good alternative interventions, which allowed for early and
54
adequate initial weight bearing without limitations related to skin condition, was considered as an
55
ideal method to these cases, who cannot use internal fixation due to trauma of the soft tissue envelop,
56
deficiency of bone stock, and bony comminution [11, 12].
EP
57
TE D
51
We performed this meta-analysis to discuss whether the external fixation (ExFx) provided better
59
radiological and clinical outcomes and fewer post-operative complications than open reduction and
60
internal fixation (ORIF) for managing complex tibial plateau fractures.
61
AC C
58
62
Materials and Methods
63
Literature search strategy
64
Electronic searches were performed by using PubMed, Cochrane library, Cochrane Central Register
65
of Controlled Trials (CCTR), China national knowledge internet database, Wan Fang database
66
without restriction for publication date and languages in August 5th, 2016. The following medical
67
subject headings (MeSH) and terms were used to achieve broad and specific searches: “internal
68
fixation”, “external fixation”, “complex tibial plateau fracture”, “Schataker 6”, “Schataker VI”, 2
ACCEPTED MANUSCRIPT 69
“Schataker 5”, “Schataker V” with the Boolean operators AND or OR. Additional records were
70
identified through google search engine or other available databases according to the Cochrane
71
Collaboration guidelines.
72
Inclusion and exclusion criteria
74
A study was included in the analysis as following criteria:
75
(1) Studies on complex tibial plateau fractures and conducted on human subjects (RCTs or
76
quasi-RCTs).
RI PT
73
(2) Studies directly compared the effectiveness of ExFx and ORIF.
78
(3) Reporting the data of outcomes (radiological, clinical and others) and complications (OA,
80 81
infections, DVT and others).
(4) The follows were excluded: reviews, isolated case reports, pathological fractures, biomechanics analyses and others.
M AN U
79
SC
77
82
Quality Assessment
84
The risks of bias were assessed independently by two of the authors, with Randomized control trial
85
were assessed by using the Cochrane Collaboration Risk of Bias Tool[13]. The contents were seven
86
parts---random sequence generation (selection bias), allocation concealment (selection bias),
87
blinding of participants and personnel (performance bias), blinding of outcome assessment data
88
(detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias) and
89
other bias. Each items were recorded as “high risk”, “low risk”, “unclear risk”. For non-RCT, the
90
Methodological Index for Non-Randomized studies (MINORs) scale was applied for quality
91
assessment[14], which scored from 0 to 24. Any disagreement was discussed and resolved with a
92
third independent author.
EP
TE D
83
AC C
93 94
Study selection and data extraction
95
Datum were independently extracted from articles, tables and figures by two investigators, including
96
the first authors, study design, the publication date, sample size, follow-up duration, interventions,
97
outcomes as well as adverse event. Any disagreement was discussed and resolved with a third
98
independent author.
99 100
Data Analysis and Statistical Methods
101
Statistical analyses were conducted with Review Manager Version 5.3 (Cochrane Collaboration,
102
Software Update, Oxford, UK). Statistical heterogeneity was assessed using the I2 value and 3
ACCEPTED MANUSCRIPT 103
chi-square test. When I2 >50%, P<0.1 was considered to be significant heterogeneity, random-effect
104
model was applied for meta-analysis. Otherwise, fixed-effect model was performed. If possible,
105
sensibility analysis was conducted to search the origins of heterogeneity. Dichotomous outcomes
106
were expressed as odds ratios(OR) with 95% confidence intervals (CIs). While continuous outcomes,
107
mean differences (MDs) and 95% confidence intervals (CIs) were calculated.
RI PT
108
Results
110
Search result
111
A total of 672 studies were identified with an initial decision, this yielded 243 titles for initial
112
screening after removal of 429 duplicates. Following initial screening and application of the
113
inclusion/exclusion criteria, there were one RCT and eight retrospectives, a total of 11 articles were
114
carried out on all[15-25]. The characteristics were described as Table 1. Two of them came from a
115
same study, conducted by the Canadian orthopaedic trauma society (COTS)[23, 24], which was also
116
the only RCT. Two came from a same institution, located in Boston[18, 22]. The search process was
117
performed as Figure 1.
M AN U
SC
109
118
Risk of assessment
120
The details about the included studies were summarized in Table 1. Cochrane Handbook for
121
Systematic Review of Interventions was consulted to assess the quality of RCTs. The COST [23, 24],
122
the only RCT, was assessed to be at low risk of bias in almost all terms (Table 2), except for blinding
123
of assessor and reporting bias. Due to there was no blinding of the evaluators and the protocol was
124
not published before recruitment commenced. The risk of bias was assessed for the eight
125
retrospective studies by MINORs [14] and showed as Table 3. Except for two of them were lost to
126
follow-up >5% [21, 25], the others were assessed as high quality.
EP
TE D
119
AC C
127 128
Study characteristics
129
There were 515 individuals with 519 fractures in our pooling of studies (four people suffer from
130
double knee fractures), 239 (46%) fractures in ExFx group and the others (54%) in ORIF. Mckee et
131
al. [23, 24], which was the only RCT, accounted for 83(16%) and used Hospital for Special Surgery
132
knee score (HSS score)[26] to measure the primary outcome, used McMaster Universities
133
Osteoarthritis Index (WOMAC)[27], the quality of radiographic reduction, the presence of
134
degenerative osteoarthritis and the scores on the Short-Form 36(SF-36) [28] to measure the
135
secondary outcome.
136
4
ACCEPTED MANUSCRIPT The other eight retrospective studies[15-17, 19-21, 25] accounted for 436 (84%)of the published
138
cases in our meta-analysis (82% of ExFx and 85% of ORIF). What they used in the ExFx group
139
included Taylor Spatial Frame (TSF), Ilizarov circular frame, Hoffman II with limited internal
140
fixation, Synthes AO fixator or Ilizarov circular frame and so on, while used medial and lateral
141
non-locking buttress plates locking plate and the Synthes Less Invasive Stabilization System (LISS)
142
in ORIF group. They used WOMAC [15, 17, 25], Rasmussen’s system[16, 17], Kellgren-Lawrence
143
score[16], IOWA knee score[19], numeric rating scale (NRS) [17], Knee Injury and Osteoarthritis
144
Outcome Score(KOOS) [20], Honkonen and Jarvinen (HJ) Criteria[25] as well as range of motion
145
(ROM) [21] to measure the outcome of surgeries. Three studies reported the soft tissue damage cases
146
without distinguishing[20, 24, 25], 23 cases had associated with cruciate ligament avulsion, 20 had
147
meniscal ligaments avulsion, 12 had Collateral ligament avulsion, 6 had neurologic injury and 1 had
148
patellar tendon avulsion. Except for Chan et al. [16] reported one (4%) in ORIF and 7(20%) ExFx
149
were alcohol dependency, there was no significant difference between the two groups in baseline
150
information.
M AN U
SC
RI PT
137
151 152
Outcomes of Meta-Analysis
153
The results of meta-analysis were showed as Table 4.
154
Hospital stay time
156
The COST [23, 24] and Conserva et al. [17] were the only two studies reported the mean time of
157
hospital stay. Interestingly, the results of the two studies were almost the same: the ExFx group were
158
Significant shorter than ORIF group (9.9 to 23.4 days in COST, P = 0.024; 7.8 to 14.2 days in
159
Conserva et al., P = 0.002).
EP
TE D
155
160
Blood loss and surgery time
162
Only the COST [19,20] give the volume of blood loss in surgery and surgery time, it seemed that
163
ExFx group took less time in surgery (170min to 183min, P = 0.229), and less blood loss (213ml to
164
544ml, P = 0.006).
165
AC C
161
166
Radiographic assessment
167
Chan et al.[16] and Conserva et al. [17]used Rasmussen’s system, which was based on joint
168
depression, condylar widening, and varus/valgus angulation. But these studies reported no
169
significant differences (28.81 to 31.73 in Krupp et al., u = 0.512; 25 to 24.9 in Conserva et al., P =
170
0.47). Jansen et al. [20]used the Kellgren-Lawrence score, but they only given the average score 5
ACCEPTED MANUSCRIPT 171
(66.2) without distinguishing between the two groups.
172
Two studies [17, 21] compared the time to radiographic union: Krupp et al. reported 7.4 months
174
(range 3-15) in ExFx group and 5.9 months (range 3-14) in ORIF group; Conserva et al. reported
175
15.9 months (range 7.5-32) in ExFx group and 17.2 months (range 9.1-45) in ORIF group. Krupp et
176
al. reported 13 (5%) in ExFx and 4(1%) in ORIF were malunion, while 11 in ExFx (5%) and 7 in
177
ORIF (3%) were union after 6 months based on radiography evidence.
RI PT
173
178
There were 67 fractures (13% of total fractures) assessed of osteoarthritis (OA) [16, 17, 19, 21, 23,
180
24] based on radiography. The COTS and Chan et al. both relied on subjective assessment by a
181
single unblinded assessor to do the assessment and used radiographs taken after the same follow-up
182
period (24 months postoperative). The analysis results were showed as Figure 2. (OR 1.52, 95% CI
183
0.86-2.66, P = 0.15).
M AN U
SC
179
184
Functional outcomes
186
Six studies reported the functional outcomes[15-17, 20, 21, 23-25]. The COST reported the ExFx
187
group seemed higher HSS than ORIF group (72 to 61, 95% CI 2.03-19.97, P = 0.064) in 6 months
188
after-operative, but there was no difference at 12 months (72 to 67, 95% CI -2.59-12.59, P = 0.406)
189
and 24 months (75 to 68, 95% CI -1.45-15.45, P = 0.307). Also, it seemed that the ROM of flexion
190
and the extension in ExFx group was better than ORIF group (flexion: 123° to 113°, P=0.114;
191
extension: 3° to 4°, P=0.449). This result was almost the same as Krupp et al. [21]. Jansen et al.
192
[20]reported outcomes by Lysholm score and KOOS as the average of the two groups without
193
distinguishing. Three studies’ results of WOMAC showed no significant difference between
194
the two groups [15, 17, 25]. And the SF-36 in two studies[23-25] also showed no significant
195
difference. In terms of pain value, Ahearn et al. [15]used VAS and Conserva et al. [17]used
196
NRS, which was no significant difference in their reports.
EP
AC C
197
TE D
185
198
Re-operation rate after surgery
199
As Figure 3A. showed, four studies reported the re-operation rate after surgery [15, 17, 21, 23, 24],
200
which included knee amputation, osteotomy, pin-track debridement[23, 24], screw pull-out[17],
201
however, no study took the planned frame removal into account during analyses. There were 28
202
cases (12%) in ExFx group and 30 cases (11%) in ORIF group, but the results were not statistically
203
significant (OR 0.87, 95% CI 0.47–1.62, P=0.66). The COST trial reported 27 frames were removed
204
and their data indicated that ORIF group had higher the re-operation rate than ExFx group. 6
ACCEPTED MANUSCRIPT 205
Knee arthroplasty after surgery
207
Five studies (403 fractures,78% of total) reported on subsequent knee arthroplasty (TKA) [16, 17, 19,
208
21, 23, 24]. Figure 3B. showed that the cases of TKA was 7(3%) in the ExFx and 12(4.2%) in the
209
ORIF groups (OR 0.51, 95% CI 0.20–1.34, P=0.17). Guryel et al. [19] didn’t give us a clear time of
210
follow-up, so we cannot know the detail of subsequent knee arthroplasty. Also, due to the range of
211
follow-up of other four studies (6-72 months) and the subsequence TKA occurred in different stages,
212
which made it difficult to evaluate how many cases required total knee arthroplasty exactly.
RI PT
206
213
Complications
215
Infections
216
All the eight retrospective studies reported the infection cases in the follow-up period (436
217
fractures)[15-22, 25]. As Figure 4. showed, ExFx group had higher infective rate than ORIF group
218
both in total infective rate (17% versus 8%, OR 1.98, 95% CI 1.08–3.63, P= 0.03, Fig.4A),
219
superficial infective rate (13% versus 5%, OR 2.72, 95% CI 1.31–5.65, P=0.008, Fig.4B) and deep
220
infective rate (5% versus 4%, OR 1.08, 95% CI 0.48–2.46, P=0.85, Fig.4C).
M AN U
SC
214
221
Compartment syndrome
223
Figure 5A. showed the compartment syndrome in two studies [16, 20], 4.5% in ExFx group and 8.9%
224
in ORIF group (OR 0.61, 95% CI 0.12–3.22, P=0.56).
TE D
222
225
Venous thromboembolism (VTE)
227
Three studies [15, 16, 19] described the rate of venous thromboembolism (VTE), with no statistically
228
significant differences between the two groups (OR 1.56, 95% CI 0.49–4.96, P= 0.45, Fig.5B).
EP
226
AC C
229 230
Joint stiffness
231
Krupp et a. [21]and Conserva et al.[17] reported eight cases of joint stiffness (7 in ExFx group and
232
one in ORIF group), with statistically significant differences between two groups (OR 5.14, 95% CI
233
0.85–30.91, P= 0.07, Fig.5C).
234 235
Discussion
236
Notoriously, complex tibial plateau fractures were a challenging fractures associated with severe bone
237
and soft tissue injury, which led to high complication rates and poor clinical outcomes [6]. Since the
238
publication of the first study presented in our meta-analysis in 1992[18, 22], the management of it has 7
ACCEPTED MANUSCRIPT 239
witnessed new developments. In these years, different methods have been proposed and utilized in
240
clinic, including limited open reduction and stabilization system, contoured locking periarticular plates,
241
hybrid external fixation and so on. And several literatures reported the good efficacy of these methods
242
[29-31].
243
External fixation , which preserve soft tissues and reduce the rate of deep infection, has led many
245
surgeons to apply it as a minimally invasive surgery to treat complex tibial plateau fractures[4, 31].
246
Catagni et al.[32] reported the use of circular external fixator for Schatzker V and VI tibial plateau
247
fractures and received good results. Katsenis et al.[33] got the similar results when used small wire
248
transfixion frames. Open reduction and internal fixation was successful in restoring articular
249
congruity and further compromise the soft tissue envelope. However, some case series have
250
highlighted the dangers of complications following ORIF. Young et al. [9]reported the 88% deep
251
infections rates with dual plating, the similar higher deep infections rate of ORIF was also reported
252
by Jensen et al. [34] and Mckee et al. [24]. When comparing the two surgery methods, the main
253
foundings in these studies were divided. Mckee et al. [24]and Hall et al. [35]reported external fixator
254
trend towards better results, but Krupp et al.[21] and Ahearn et al. [15] found no significant
255
differences between the two methods. So we tried to get a uniform conclusion on whether external
256
fixation can provide better outcomes with fewer complications when compared to the open reduction
257
and internal fixation for complex tibial plateau fractures.
SC
M AN U
TE D
258
RI PT
244
One potential bias was that only one RCT (16% fractures) and eight retrospective studies (84%
260
fractures) have directly compared the ExFx and ORIF in treating complex tibial plateau fractures.
261
However, they were high quality research in which the main focus was on clinical outcomes, we also
262
could get some valuable and reliable information. Firstly, we found almost half of the cases were
263
high energy trauma fracture, like motor vehicle accident or fall from a height[15, 17, 25] and
264
numbers of patients associated with ligaments injuries[20, 24, 25]. Secondly, the severity of
265
intra-articular displacement and soft tissue compromise has a significant impact on the choice of
266
operation method and the timing of surgery[25, 36].
AC C
267
EP
259
268
Initially, we hoped to test the functional outcomes of the two methods, nevertheless, various
269
evaluation system made it difficult to performance this compare. Except for the COST, which
270
reported a higher HSS at the early stage and better ROM in ExFx group [23, 24, 35], the others
271
showed no significant difference in terms of SF-36, ROM, and WOMAC. Maybe in the further
272
studies, we can take a uniform evaluation system to make the compared more reliable. 8
ACCEPTED MANUSCRIPT 273
Theoretically, more minimal invasive surgery, like Taylor Spatial Frame or Ilizarov circular
275
frame, would be more helpful for the healing. Indeed, Conserva et al.[17] and the COST trail
276
[21, 23, 24]reported that individual who undergoing ExFx seemed had less blood loss, shorter
277
surgery time, faster return to work or pre-injury activities. Krupp et al. [21] reported the average time
278
to union was shorter in ORIF group (7.4 months in ExFx and 5.9 months ORIF ), however, it was
279
conflict with Conserva et al.[17]. We thought it maybe result from the type of plates they used and
280
the reduction skill of surgeons.
RI PT
274
281
As the analysis results showed above, we could know that the ExFx group seemed had higher
283
osteoarthritis(OA) rate, lower subsequence total knee arthroplasty(TKA) rate and lower re-operations
284
[16, 17, 19, 21, 23, 24], however, the results were no statistical difference and the small sample size
285
and the short follow-up duration made us cannot sure the rate of OA and subsequence TKA exactly,
286
which were a long-term complication of intra-articular fractures that can occur in any periods after
287
trauma. And no studies reported the cases who planned to remove the implants. Although, the data
288
showed ExFx had higher cost effectiveness, we should notice that higher advert event rates and the
289
further cost to deal with them in using external fixation.
M AN U
SC
282
290
Infections, or other advert events, in theory, was superior in open reduction and internal fixation.
292
However, our results were against to this concept. As compared to ORIF group, the rate of infection
293
(both superficial and deep infection), venous thromboembolism (VTE) and joint stiffness were
294
higher in the ExFx group. Fortunately, no evidences showed there were serious infections occurred
295
as well as the compartment syndrome, and no studies reported the Pulmonary embolism (PE).
EP
TE D
291
296
Only one RCT, small sample size, various assessment tools and short follow-up duration made it
298
difficult to achieve our goal which was to recommend the better surgical method in treating complex
299
tibial plateau fractures. Fortunately, the pooled studies and accumulated datum were relatively high
300
quality, we also got some valuable conclusion.
301
AC C
297
302
Conclusion
303
Based on our analysis results, external fixation may offer some advantages in terms of soft tissue
304
healing. However, with the use of minimally invasive internal fixation locking plates, the superiority
305
of external fixation was no longer obvious. The currently existing evidences didn’t give us a clear
306
recommendation on whether ExFx was better than ORIF in managing complex tibial plateau 9
ACCEPTED MANUSCRIPT fractures, and longer duration studies on comparing ORIF and ExFx were required. Although
308
orthopedists should be mindful the complications and the subsequence procedures after surgery, both
309
ExFx and ORIF were acceptable strategies in managing complex tibial plateau fractures. In clinical
310
practice, we should select proper fixator based on the fracture patterns, soft-tissue condition as well
311
as the injury stages. Meanwhile, further multicentered, randomized controlled studies should be
312
implemented to get a more reliable and clear result.
RI PT
307
313 314 315
SC
316 317 318
M AN U
319 320 321 322 323 324
TE D
325 326 327
EP
328 329 330
AC C
331 332 333 334 335 336 337
Table 1. Characteristics of including studies
338 339
Table 2. Quality of the randomized controlled trials (the Cochrane Collaboration Risk of Bias
340
Tool) 10
ACCEPTED MANUSCRIPT 341 342
Table 3. Quality assessment for non-randomized trials (MINORs)
343 344
Table 4. Results of meta-analysis
345
Figure 1. Flowchart of the study selection process
RI PT
346 347 348
Figure 2. Forest plot showing the meta-analysis of the osteoarthritis(OA)
349
Figure 3. Forest plot showing the meta-analysis of re-operation (A) and total knee
351
arthroplasty(B)
SC
350
352
Figure 4. Forest plot showing the meta-analysis of the rate of post-operative infections(A), the
354
rate of superficial infections(B), the rate of deep infections(C)
M AN U
353
355 356 357 358
TE D
359 360 361
EP
362 363 364
AC C
365 366 367 368 369 370 371
References
372 373
1.
AD. WJaW. Rockwood and Green's fractures in adults: Fractures of the proximal tbial and fibula Bucholz
374
RW and Heckman JD Ed 5th ed Philadelphia: Lip-pincott Williams and Wilkins2001;2:1799-839. 11
ACCEPTED MANUSCRIPT 375
2.
Albuquerque RP, Hara R, Prado J, Schiavo L, Giordano V, do Amaral NP. Epidemiological study on tibial
376
plateau fractures at a level I trauma center. Acta Ortopedica Brasileira2013;21(2–:109-15.
377
3.
378
International Wound Journal2014;11 Suppl 1(s1–:3 5.
379
4.
380
Clinical orthopaedics and related research1979 Jan-Feb(138–:94-104.
381
5.
382
Springer Berlin Heidelberg; 1990.
383
6.
384
treatment of tibial plateau fractures: Five- to 27-year follow-up results (Journal of Orthopaedic Trauma (2007–
385
21, (5-10––. 2007.
386
7.
387
complex
388
trauma1992;6(1–:78-86.
389
8.
390
comparison of treatment methods. Journal of Trauma1981;21(5–:376-81.
391
9.
392
review1994;23(23–:149-54.
393
10. Nikolaou VS, Tan HB, Haidukewych G, Kanakaris N, Giannoudis PV. Proximal tibial fractures: early
394
experience using polyaxial locking-plate technology. Injury Extra2011;35(8–:1215-21.
395
11. Dendrinos GK, Kontos S, Katsenis D, Dalas A. Dendrinos GK, Kontos S, Katsenis D, et al. Treatment of
396
high-energy tibial plateau fractures by the Ilizarov circular fixator. Bone & Joint Journal1996;78.
397
12. Subasi M, Kapukaya A, Arslan H, Ozkul E, Cebesoy O. Outcome of open comminuted tibial plateau
398
fractures treated using an external fixator. Journal of Orthopaedic Science2007;12(4–:347-53.
399
13. Higgins J, Green SE. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. The
400
Cochrane Collaboration (Eds–. 2011;2011.
401
14. Slim K, Nini E, Forestier D, Kwiatkowski F, Panis Y, Chipponi J. Methodological index for non-randomized
402
studies ( MINORS –: development and validation of a new instrument. Anz Journal of Surgery2003;73(9–:712-6.
403
15. Ahearn N, Oppy A, Halliday R, Rowett-Harris J, Morris SA, Chesser TJ et al. The outcome following
404
fixation of bicondylar tibial plateau fractures. The bone & joint journal2014 Jul;96-b(7–:956-62.
405
16. Chan C, Keating J. Comparison of outcomes of operatively treated bicondylar tibial plateau fractures by
406
external fixation and internal fixation. Malaysian orthopaedic journal2012 Mar;6(1–:7-12.
407
17. Conserva V, Vicenti G, Allegretti G, Filipponi M, Monno A, Picca G et al. Retrospective review of tibial
408
plateau fractures treated by two methods without staging. Injury2015 Oct;46(10–:1951-6.
Brem MH, Bail HJ, Biber R. Value of incisional negative pressure wound therapy inorthopaedic surgery.
RI PT
Schatzker J, McBroom R, Bruce D. The tibial plateau fracture. The Toronto experience 1968--1975.
Müller MME, Koch MP, Nazarian MS. The Comprehensive Classification of Fractures of Long Bones:
SC
Rademakers MV, Kerkhoffs GMMJ, Sierevelt IN, Raaymakers ELFB, Marti RK. Erratum: Operative
Benirschke SK, Agnew SG, Mayo KA, Santoro VM, Henley MB. Immediate internal fixation of open, plateau
fractures:
treatment
by
a
standard
protocol.
M AN U
tibial
Journal
of
orthopaedic
Waddell JP, Johnston DW, Neidre A. Fractures of the tibial plateau: a review of ninety-five patients and
AC C
EP
TE D
Young MJ, Barrack RL. Complications of internal fixation of tibial plateau fractures. Orthopaedic
12
ACCEPTED MANUSCRIPT 18. Covall DJ, Fowble CD, Foster TE, Whitelaw GP. Bicondylar tibial plateau fractures: principles of treatment.
410
Contemporary orthopaedics1994 Feb;28(2–:115-22.
411
19. Guryel E. MS, Sharma R., Newman K., Elliott D., Khaleel A. THE CHERTSEY CLASSIFICATION OF
412
TIBIAL PLATEAU FRACTURES AND A COMPARISON OF THE OUTCOME OF TREATMENT WITH
413
INTERNAL FIXATION OR AN ILIZAROV FINE WIRE CIRCULAR FRAME
414
Orthopaedic Proceedings2010.
415
20. Jansen H, Frey SP, Doht S, Fehske K, Meffert RH. Medium-term results after complex intra-articular
416
fractures of the tibial plateau. Journal of Orthopaedic Science2013 2013/07/01;18(4–:569-77.
417
21. Krupp RJ, Malkani AL, Roberts CS, Seligson D, Crawford CH, 3rd, Smith L. Treatment of bicondylar tibia
418
plateau fractures using locked plating versus external fixation. Orthopedics2009 Aug;32(8–.
419
22. Mallik AR CD, Whitelaw GP. Internal versus external fixation of bicondylar tibial plateau fractures. Orthop
420
Rev1992.
421
23. S. Pirani MM. RANDOMIZED TRIAL OF BICONDYLAR TIBIAL PLATEAU FRACTURES TREATED WITH
422
AO OR RING FIXATOR METHODS: EARLY REINTERVENTION RATES FOR COMPLICATIONS.
423
Orthopaedic Proceedings2008.
424
24. SP MMP. Open reduction and internal fixation compared with circular fixator application for bicondylar
425
tibial plateau fractures. Results of a multicenter, prospective, randomized clinical trial. The Journal of bone and
426
joint surgery American volume2006 Dec;88(12–:2613-23.
427
25. Tul B Pun VPK, Pradeep M Poonnoose, Anil T Oommen, and Ravi J Korula. Outcome of Schatzker type V
428
and VI tibial plateau fractures. 2014.
429
26. Caplan N, Kader DF. A Comparison of Four Models of Total Knee-Replacement Prostheses. Journal of
430
Bone & Joint Surgery American Volume1976;58(6–:754-65.
431
27. Guermazi M, Poiraudeau S. Translation, adaptation and validation of the Western Ontario and McMaster
432
Universities Osteoarthritis Index (WOMAC– for an Arab population: the Sfax modified WOMAC. Osteoarthritis &
433
Cartilage2004;12(6–:459-68.
434
28. Jenkinson C, Coulter A, Wright L. Short Form 36 (SF-36– health service questionnaire. Normative data for
435
adults of working age. Bmj Clinical Research1993;306(6890–:1437-40.
436
29. Orthopaedics JCHXPLXWDYDo. Open Reduction and Internal Fixation of High-energy Tibial Plateau
437
Fractures. Orthopedic Clinics of North America2003;33(1–:177-98.
438
30. Jeray KJ, Lochow SC. Staged Open Treatment of High-Energy Tibial Plateau Fractures. Techniques in
439
Knee Surgery2005;4(4–:214-25.
440
31. Babis GC, Evangelopoulos DS, Kontovazenitis P, Nikolopoulos K, Soucacos PN. High energy tibial
441
plateau
442
Research2011;6(1–:1-7.
AC C
EP
TE D
M AN U
SC
RI PT
409
fractures
treated
with
hybrid
external
13
fixation.
Journal
of
Orthopaedic
Surgery
&
ACCEPTED MANUSCRIPT 443
32. Catagni MA, Ottaviani G, Maggioni M. Treatment strategies for complex fractures of the tibial plateau with
444
external circular fixation and limited internal fixation. Journal of Trauma2007;63(5–:1043-53.
445
33. Katsenis D, Athanasiou V, Megas P, Tyllianakis M, Lambiris E. Minimal internal fixation augmented by
446
small
447
trauma2005;19(4–:241-8.
448
34. Jensen DB, Rude C, Duus B, Bjerg-Nielsen A. Tibial plateau fractures. A comparison of conservative and
449
surgical treatment. Bone & Joint Journal1990;72(1–:49-52.
450
35. Hall JAB, M. J. McKee, M. D. Open reduction and internal fixation compared with circular fixator
451
application for bicondylar tibial plateau fractures. Surgical technique. The Journal of bone and joint surgery
452
American volume2009 Mar 1;91 Suppl 2 Pt 1:74-88.
453
36. Kokkalis ZT, Iliopoulos ID, Pantazis C, Panagiotopoulos E. What's new in the management of complex
454
tibial plateau fractures? Injury-international Journal of the Care of the Injured2016;47(6–:1162-69.
frames
for
high-energy
tibial
plateau
fractures.
Journal
AC C
EP
TE D
M AN U
455
14
of
orthopaedic
RI PT
transfixion
SC
wire
ACCEPTED MANUSCRIPT Dear editor: The study was supported by the National Natural Science Foundation of China (NO. 81572154), Tianjin Health Bureau Science and Technology Foundation (NO. 15KG123), and Key project of Tianjin science and technology (NO.13ZCZDSY01700). We were grateful to all the participating patients in these trails and all the staffs in these studies. Thanks very much for your attention to our paper. Sincerely yours,
RI PT
Xing-wen Zhao Corresponding author: Jian-xiong MA E-mail [email protected]
AC C
EP
TE D
M AN U
SC
January 6, 2017
1
ACCEPTED MANUSCRIPT
Table 2. Quality of the randomized controlled trials (the Cochrane Collaboration Risk of Bias Tool)
COTS
Random sequence
Allocation
Blinding of
Blinding of
Incomplete
Selective reporting
generation
concealment
participants and
outcome
outcome data
(reporting bias)
（selection bias） ）
（selection bias） ）
personnel
assessment
(attrition bias)
(performance bias)
(detection bias)
Low risk
High risk
Low risk
Low risk
Mckee et al. (2006)
AC C
EP
TE D
M AN U
SC
Pirani et al. (2008)
RI PT
Studies
1
Low risk
Unclear risk
Others
Low risk
ACCEPTED MANUSCRIPT
Table 3. Quality assessment for non-randomized trials (MINORs) Quality assessment for Non-RCT
Boston (
Krupp et
Guryel et
Chan et
Jansen et
Pun TB et
Ahearn et
Conserva et
Covall et al.
al. (2009)
al. (2010)
al. (2012)
al. (2013)
al. (2014)
al. (2014)
al. (2015)
Mallik et al. （1992） ）) 2
2
2
2
2
2
2
Inclusion of consecutive patients
2
2
2
2
2
2
2
2
Prospective data collection
0
0
0
0
0
0
0
0
Endpoints appropriate to the aim of the study
2
2
2
2
2
2
2
2
Unbiased assessment of the study endpoint
2
2
2
2
2
2
2
2
Follow-up period appropriately to the aims of
2
2
2
2
2
2
2
2
Less than 5 % loss to follow-up
2
1
2
2
2
2
1
2
Prospective calculation of the sample size
0
0
0
0
0
0
0
0
An adequate control group
2
2
2
2
2
2
2
2
Contemporary groups
2
2
2
2
2
2
2
2
Baseline equivalence of groups
1
2
2
2
2
1
2
2
Adequate statistical analyses
2
2
2
2
2
2
2
2
Total score
19
19
20
20
20
20
17
20
AC C
EP
TE D
study
SC
2
M AN U
A clearly stated aim
RI PT
(1994) and
1
ACCEPTED MANUSCRIPT
Studies
Boston
Country
U.S.
Design
Non-RCT
Mean Ages
Cases
Fracture
Interventions (knees and methods)
(years)
(knees)
types(knees)
45 years
17(17)
NM
Covall et
Monticelli-Spinellib
Bilateral buttress,
Complications,
semi-tubular
re-operations
ORIF: 33 months
Medial and lateral
Primary: HSS
6 12 24 months
cannulated screws (7)
EF: 46.2
82(83)
V-18, VI-65
Percutaneous lag
SC
RCT
2
C1-20,
screw, and Ilizarov
non-locking
Secondary: WOMAC,
buttress plates ±
reoperations,
bone grafting (40)
complications, SF-36,
years
al. (2006)
ORIF: 43.3
C2-39,
circular frame (43)
Pirani et
years
C3-24
M AN U
Canada
1
Mckee et
al. (2008) U.S.
Non-RCT
EF: 48.8
1
58(58)
V-37
years
al. (2009)
VI-21
ORIF: 46.64
U.K.
Non-RCT
NM
NM (124)
al. (2010)
3
radiological results
Hoffman II Hybrid
Dual plating (8),
Clinical outcomes:
EF:16.4 months
(16),
Lateral locking
union rate, time to
ORIF: 10.2
circular frames (14)
plate + medial
union, malunion,
months
screws or buttress
ROM, complications,
plate (20)
re-operations
NM (79)
IOWA knee score,
TE D
years Guryel et
EF:10 months
plates, or
al. (1992)
Krupp et
(months)
circular fixator (10)
Mallik et COTS
Follow-up
ORIF
al. (1994)
4
Outcome Measures
ExFx
RI PT
Table 1. Characteristics of including studies
Vagus-62,
Ilizarov circular
varus-14,
frame (45)
NM
ROM, complications
U.K.
Non-RCT
EF:52.03 years
(2012)
58(59)
AC C
Chan et al.
EP
axial-48
ORIF:45.04
2
C1-18,
Ilizarov circular
Buttress plate
Complications,
C2-2,
frame (23), Hoffman
(21), LISS (4)
re-operations
C3-39
II with limited
years Jansen et al. (2013)
Germany
Non-RCT
46 years
22(23)
3,6,12,24 months
internal fixation (13) 2
C1-7,
Ilizarov circular
LISS± additional
KOOS, IKDC,
C2-7,
frame (2)
plates± artificial
Lysholm Score,
bone substitute
Complications, ROM
C3-9
(21) 1
67 months
Pun TB et
Indian
Non-RCT
43.85 years
ACCEPTED MANUSCRIPT V-11, Circular frame + VI-10 medial percutaneous screws (12)
1
21(21)
al. (2014)
dual plate (9)
HJS (clinical,
30 months
functional, radiological outcomes) WOMAC, infection rates
U.K.
Non-RCT
44 years
55(55)
C3-30,
Others were
al. (2014)
Taylor Spatial Frame
lateral locking
Primary: WOMAC,
(21)
plate ± medial
SF-36
plate fixation (34)
Secondary:
RI PT
Ahearn et
2
Conserva
Italy
Non-RCT
54.1years
79(79)
1IV-23,
et al.
V-29,
(2015)
VI-27
M AN U
SC
unknown
Percutaneous lag
LCP±bone graft
screw+ hybrid
substitute (38)
external fixator (41)
EF:31 months ORIF:40.5 months
satisfaction scale, VAS, complications, reoperations, radiological outcomes Rasmussen,
EF:39.4 months
WOMAC, NRS,
ORIF:35.1months
complications, reoperations
TE D
Abbreviations: KOOS, Knee Injury and Osteoarthritis Outcome Score. IKDC, International Knee Documentation Committee score. LCP, locking compression plate. NRS, Numeric Rating Scale. HSS, Hospital for special surgery knee score. WOMAC, Western Ontario and McMaster Universities Arthritis Index. ROM, Range Of Motion. HJS, Honkonen and Jarvinen score. SF-36, Short Form 36. VAS, Visual Analogue Score. LISS, Less Invasive Stabilization System. COST, Canadian Orthopaedic Trauma Society. U.S., United State. U.K., United Kingdom. ExFx, External Fixation. ORIF, Open Reduction and Internal Fixation. RCT, Randomized Control Trial. Non-RCT, Non-Randomized Control Trial. NM, Not Mentioned.
AC C
EP
Note: 1, Schatzker classification; 2, Orthopaedic Trauma Association classification. 3, The Chertsey classification of tibial plateau fractures. 4, Multicenter randomized controlled trial.
2
ACCEPTED MANUSCRIPT
Outcome
Studies
Overall effect Effect estimate
95%CI
RI PT
Table 4. Results of meta-analysis Heterogeneity 2
P value
I (%)
P value
0.55
0
0.15
0.31
17
0.66
5
1.52
[0.86,2.66]
Re-operation
4
0.87
[0.47,1.62]
Subsequence TKA
5
0.51
[0.20,1.34]
0.99
0
0.17
Infection rate
8
1.98
[1.08,3.63]
0.26
22
0.03
Superficial infection
8
2.72
[1.31,5.65]
0.44
0
0.008
Deep infection rate
8
1.08
[0.48,2.46]
0.11
43
0.85
Compartment syndrome
2
0.61
[0.12,3.22]
0.35
0
0.56
VTE
3
1.56
[0.49,4.96]
0.13
50
0.45
Joint stiffness
2
5.14
[0.85,30.91]
0.78
0
0.07
M AN U
AC C
EP
TE D
rate
SC
Osteoarthritis (OA)
1
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
Figure 1. Flowchart of the study selection process
1
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
Figure 2. Forest plot showing the meta-analysis of the osteoarthritis(OA)
1
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
Figure 3. Forest plot showing the meta-analysis of re-operation (A) and total knee arthroplasty(B)
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 4. Forest plot showing the meta-analysis of the rate of post-operative infections(A), the rate of superficial infections(B), the rate of deep infections(C)
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
Figure 5 Forest plot showing the meta-analysis of the rate of compartment syndrome(A), the rate of venous thromboembolism(B), and the rate of knee joint stiffness(C).
1
ACCEPTED MANUSCRIPT We aimed to evaluate the effectiveness of external fixation(ExFx) and open reduction and internal fixation(ORIF) in treating complex tibial plateau fractures. According to our results, ExFx had some advantages when compared with ORIF, but there were no
RI PT
statistically significant differences. So we strongly recommend that selection of definitive fixators and time of intervention should base on the fracture patterns,
soft-tissue condition as well as the injury stages in clinical practice. What’s more,
SC
due to the limits of this review, we recommended further researches were needed to
AC C
EP
TE D
M AN U
achieve high quality and credible results.
1
ACCEPTED MANUSCRIPT
International Journal of Surgery Author Disclosure Form
Please state any conflicts of interest
SC
The authors declared that they have no conflict of interest.
RI PT
The following additional information is required for submission. Please note that failure to respond to these questions/statements will mean your submission will be returned. If you have nothing to declare in any of these categories then this should be stated.
M AN U
Please state any sources of funding for your research None of the authors have any financial funding.
TE D
Please state whether Ethical Approval was given, by whom and the relevant Judgement’s reference number
EP
No.
AC C
Research Registration Unique Identifying Number (UIN) Please enter the name of the registry and the unique identifying number of the study. You can register your research at http://www.researchregistry.com to obtain your UIN if you have not already registered your study. This is mandatory for human studies only.
reviewregistry111
1
ACCEPTED MANUSCRIPT
SC
RI PT
Author contribution Please specify the contribution of each author to the paper, e.g. study design, data collections, data analysis, writing. Others, who have contributed in other ways should be listed as contributors. Xing-wen Zhao and Xin-long Ma did the design. Xing-wen Zhao, Jian-xiong Ma, Xuan Jiang, Yin Wang, Fei Li, and Bin Lu did the data collection. Xing-wen Zhao and Jian-xiong Ma did data analysis and writing.
M AN U
Guarantor The Guarantor is the one or more people who accept full responsibility for the work and/or the conduct of the study, had access to the data, and controlled the decision to publish.
EP
TE D
Jian-xiong MA
Thanks very much for your attention to our paper.
AC C
Sincerely yours, Xing-wen Zhao
Corresponding author: Jian-xiong MA E-mail：[email protected] January 6, 2017
2