Comparison between open reduction with internal fixation to circular external fixation for tibial plateau fractures: A systematic review and meta-analysis

Peer-reviewed published studies on tibial plateau fractures treated with either open reduction with internal fixation (ORIF) or circular external fixation were reviewed to compare functional, radiological outcomes, postoperative complications, and reoperation rates between the two methods. A systematic search of various databases including Medline, Cochrane Controlled Register of Trials (CENTRAL), ScienceDirect, and Google Scholar from inception until June 2019 was performed. 17 studies with 1168 participants were included in the review. Most of the studies (76%) were retrospective in nature and had low or unclear bias risks. Incidence of total infection (Odds ratio [OR], 2.58; 95% CI, 1.33–5.02) and malunions (OR, 2.56; 95% CI, 1.12–5.84) were higher and length of hospital stay was shorter in patients treated with circular external fixator (Mean difference [MD], -6.1; 95% CI, -11.1–-1.19). There were no differences in the incidence of secondary osteoarthritis (OR, 1.49; 95% CI, 0.92–2.42), range of motion (MD, 2.28; 95% CI, -11.27–15.82) non-union (OR, 1.44; 95% CI, 0.14–14.27) and reoperation rates (OR, 1.84; 95% CI, 0.90–3.78) between the two groups. Results from this investigation suggest that circular fixation may offer some advantages over ORIF such as a shortened length of hospital stay and early return to preinjury activities. Definitive clinical recommendations cannot be made as it also presents higher rates of postoperative complications than ORIF.


Introduction
The incidence of tibial plateau fractures has increased over the past decade [1]. These fractures have a bimodal distribution affecting younger children (due to high energy traumas) and elderly individuals (due to increased bone fragility) [2]. Timely and adequate management of these injuries is important to avoid complications like knee instability or stiffness, secondary osteoarthritis (OA), compartment syndrome, and soft tissue damage [3]. Tibial plateau fractures can be managed either by conservative methods or by operative treatment. Anatomical a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 reduction and restoration of alignment is important to prevent secondary displacement of the fracture fragments and is most commonly achieved by open reduction and internal fixation (ORIF) with plates and/or screws [4]. ORIF of tibial plateau fractures, however, is associated with significant complications like skin necrosis, early-onset osteoarthritis, and infection, especially in complex fractures and cases with significant soft tissue injury [5]. To overcome these drawbacks, minimally invasive methods have been developed which stabilize bone fragments by means of a circular external framework. External fixation usually requires a small and limited number of incisions thereby avoiding the detrimental effects of soft tissue dissection and preventing devascularisation of the osseous fragments [2,6,7].
Several studies have compared outcomes with the two treatment modalities for the management of tibial plateau fractures, but with conflicting results. A few systematic reviews and meta-analyses have also attempted to summarize existing evidence on the superiority of one treatment over the other [8][9][10]. Boutefnouchet et al [8] in their systematic review of 5 studies and Metcalfe et al [9] in their meta-analysis of 7 studies demonstrated similar results with either fixation technique for tibial plateau fractures. Later in 2017, in a meta-analysis of 11 studies, Zhao et al [10] also reported similar results but pointed out the deficiency of highquality studies in the literature comparing the two fixation methods. Since then many new studies have been published and there is a need for a comprehensive and updated meta-analysis to provide quality evidence on this controversial topic. Therefore, the purpose of this study was to perform an updated literature search and meta-analysis comparing outcomes of ORIF with circular external fixation for the management of tibial plateau fractures.

Inclusion criteria and search strategy
A systematic search to identify studies that directly compared the effectiveness ORIF to circular external fixation for the management of tibial plateau fractures was performed. An electronic search was conducted from the following databases: Medline, ScienceDirect, Cochrane Bone, Joint and Muscle Trauma Group Specialised Register, Cochrane Central Register of Controlled Trials (CENTRAL), clinical trial registries like ClinicalTrials.gov, and WHO International Clinical Trials Registry Platform (ICTRP). The search performed used a combination of medical subject heading (MeSH) and free-text terms including "tibia", "tibial Fractures" "fractures", "open fracture reduction", "internal fixation", "circular fixator", "surgical fixation", "postoperative complications" and "wounds and injuries" for all English language publications from databases inception to June 2019. The search strategy of the MEDLINE database is presented as S1 File. The references list of all the primary trials obtained through the electronic search was also included and relevant articles were included in the analysis. Authors were contacted by email for missing data.
Inclusion criteria were all randomized or quasi-randomized controlled trials (RCT), and prospective or retrospective cohort studies comparing ORIF and circular external fixation for tibial plateau fracture management. Outcomes of the study were to include either radiographic outcome (secondary osteoarthritis [OA]), functional outcomes (range of motion, length of hospital stay) or postoperative complications (both superficial and deep infections), malunion, non-union, stiffness, thromboembolism and reoperation rate.

Data extraction and outcomes
Two independent investigators performed the literature search. Articles were screened by their titles and abstracts for possible inclusion in the review. Full-text articles of relevant studies were obtained for final scrutiny based on inclusion criteria. Disagreements between investigators were resolved by consultation with another investigator. The Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines were followed for reporting this review [11]. The work has been reported in line with AMSTAR (Assessing the methodological quality of systematic reviews) Guidelines.
Two reviewers extracted the following details from the included studies:

Risk of bias
Two independent investigators assessed the risk of bias for included RCTs using the Cochrane risk of bias tool [12], focusing on the following domains: random sequence generation, allocation concealment, blinding of the participants and outcome assessment, incomplete outcome data, selective reporting of outcome and other sources of bias. For non-randomized studies, the risk of a bias assessment tool for non-randomized studies was used [13] based on the following domains: the selection of participants, confounding variables, intervention measurements, blinding of outcome assessment, incomplete outcome data, and selective outcome reporting.

Statistical analysis
The software "Review Manager" (RevMan, version 5.3; Nordic Cochrane Centre [Cochrane Collaboration], Copenhagen, Denmark; 2014) was used for the meta-analysis. Continuous variables were summarized using Mean Difference with 95% confidence intervals (CI). Categorical variables were summarized using Mantel-Haenszel Odds Ratios (OR) with 95% CI. Predicting heterogeneity in the included studies, due to different study types and other methodological variations, we preferred a random-effects model to calculate the pooled effect size for all analyses [14]. Heterogeneity was classified according to I 2 as mild (I 2 <25%) moderate (I 2 between 25 and 75%) or substantial (I 2 >75%) [15]. Study-specific and pooled estimates were graphically depicted through forest plots. Publication biases for postoperative infection rates were assessed as other outcomes did not have the required number of studies to assess the publication bias (minimum of 10 studies) and this was graphically represented by a funnel plot.

Results
Search flow chart of the study is presented in Fig 1. A total of 17 studies with 1168 participants were included in the meta-analysis. Table 1 lists the characteristics of the studies analyzed. Two studies were RCTs, 2 were prospective, and 13 studies were retrospective. The mean age of study participants ranged from 12.4 to 55.7 years in the ORIF cohort, and that in the circular external fixation cohort ranged from 13.6 to 50.4 years. Of the 1168 participants, 550 patients were treated with the circular external fixator and 618 with ORIF. The sample sizes of studies in the ORIF cohort varied  Table 2.   (2) Malunion (5) Infection (22) Consolidation delay (10) Malunion (16) Total (5) Mobilization under anesthesia (1) Arthroscopic arthrolysis (1) Cannulated screw removal (2) Surgical debridement (3) Total (22) Autograft (5) Mobilization under anesthesia (3) Arthroscopic arthrolysis (3) Surgical debridement (8) Fasciotomy (1) Berven 2018 [18] Non-union (9) Knee stiffness (9) Knee instability (3) Infection (7) Heterotopic ossification (9) Osteoarthritis (34) DVT (1) Malunion (15) Non-union (3) Knee stiffness (7) Knee instability (5) Infection (31) Heterotopic ossification (1) Osteoarthritis (42) Peroneus paresis (3) DVT (2) Malunion (16) NR Boston (Malik) 1994 [19] Infection (3) Infection (5) (7) Infection (9) DVT (4) Compartment syndrome (2) Osteoarthritis (9) Fasciotomy (3) Wound debridement (2) Fasciotomy (2) Wound debridement (2) Chertsey (Nawaz) 2013 [22] Infection (2) DVT (2) -NR Assessments of risk of bias for RCTs and non-randomized studies is presented in Table 3. The percent of cases lost to follow-up in the included studies ranged from 0% to 12.5%. Both RCTs [24,31] included in the review had low risks in almost all the domains except for high risk in the blinding of participants and outcome assessments, and they had an unclear risk of selective reporting of outcomes given that the published protocols for the studies could not be identified. Among the non-randomized studies, all the studies had a low risk of intervention measurements and unclear risks of selective reporting of outcomes (protocols not published) or blinding of outcome assessment (not mentioned in the studies). Almost all the studies except Jansen 2013 [27] had low risks with respect to the selection of participants. Four out of 15 had high risks of bias concerning incomplete outcome data due to loss to follow up of the study participants.

Discussion
The management of tibial plateau fractures has seen technological advances [2,6,7]. A lack of systematic and high-quality research assessing the effectiveness of these newer techniques make it difficult to determine if the safety and efficacy surpass previous standards of care. The  purpose of this analysis was to compare the efficacies of ORIF to circular external fixation, in terms of functional and radiological outcomes, postoperative complications, and reoperation

PLOS ONE
rates among patients with tibia plateau fracture. A total of 17 studies consisting of 1168 participants met the inclusion criteria for this analysis. Out of these, only 2 studies were RCTs and 2 were prospective studies, while the remainder (76% of the studies) were retrospective studies. Most of the studies in this review had either unclear or low bias risks. Substantial heterogeneity among the reported outcomes in the studies was observed. Given the small number of studies under each of the outcomes (less than 10) a meta-regression to explore the source of heterogeneity could not be performed. Except for the length of hospital stay, all other outcomes (functional and radiological outcomes, and postoperative complication and reoperation rates) were

PLOS ONE
superior after ORIF than after circular external fixation. However, the only conclusive and significant evidence for the risk of postoperative complications such as all infections, superficial infections, and malunions were more frequent in patients undergoing external circular fixation. The pooling of all other variables amongst studies did not show any statistically significant differences in terms of functionality, radiological outcomes, and complications between the study cohorts.
Patients undergoing circular external fixations were more likely to develop postoperative infections and malunions than the patients undergoing ORIF. However, patients who undergo circular external fixations return to their preinjury activities sooner than patients who underwent ORIF surgery. Similar findings were reported in the review by Metcalfe et al (2015) [9] in which authors found conclusive evidence for postoperative infections (all infections and superficial infections) being more frequent in patients after external fixations than after ORIF. That review included only 7 studies and functional outcomes (range of motion, length of hospital stay) and postoperative complications (knee stiffness, malunions, and non-union) were not assessed. Results from this investigation provide evidence to aid surgeons to determine which surgical intervention is preferable for their patients with tibia plateau fractures.
The major strengths of this study include the comprehensive literature search and the broad search strategy to capture relevant publications. Besides, this review compares functional outcomes such as knee range of motion and postoperative complications (malunions, non-unions, and knee stiffnesses) between patients treated by ORIFs or circular external fixations.
There were limitations in this analysis; there was a high level of heterogeneity across the studies included in the review and it was not possible to identify the source of heterogeneity as each of the outcomes was present in only a few studies (less than 10) making meta-regressions not possible. As a result, it was necessary to apply a random-effects model to account for this heterogeneity. Also, this review included only 2 RCTs among the 17 studies included. Since most of these studies were retrospective in nature, it is not possible to infer causal associations between the interventions and the outcomes. More trials of adequate size need to be conducted to make stronger recommendations on treatment. It was not possible to determine which fixation method between ORIF or circular external fixation is superior for the management of tibia plateau fractures. However, most of the studies in this investigation noted that the external fixator method was inferior to ORIF due to postoperative complication rates. Uncertainties regarding the functional and radiological outcomes persist. Adequately powered RCTs or prospective studies are needed to strengthen the evidence for recommendations on how to best manage patients with tibia plateau fractures. Conclusions Circular fixator method may offer some advantages over ORIF such as shortened length of hospital stays and early returns to preinjury activities, however, had a higher rate of postoperative complications than the ORIF cohort. Inconclusive evidence in terms of the functional and radiological outcomes was observed. Future studies with larger sample sizes are needed to identify the best management strategy for tibia plateau fractures.