1 Introduction

Goals for the treatment of tibial plateau fractures include restoration of articular congruity, joint stability, alignment, and range of motion while limiting complications. The influence of tibial articular incongruity on symptoms and limitations varies, and its importance relative to alignment, stability, and motion are unclear. Open reduction internal fixation (ORIF) of tibial plateau fractures involves incisions that allow for reduction maneuvers and placement of instrumentation. Fracture reduction is assessed with the use of fluoroscopy and direct visualization. Arthroscopic-assisted fixation (AAF) of tibial plateau fractures is typically performed with percutaneous reduction and fixation methods, and evaluation of fracture reduction relies on fluoroscopic and arthroscopic images.

The post-operative articular reduction has been evaluated for different techniques described for achieving and maintaining reduction in tibial plateau fractures. In one prospective cohort study, patients with tibial plateau fractures treated with ORIF underwent postoperative computed tomography (CT) which identified an articular step off or gap greater than 2 millimeters (mm) in 21 of 65 patients with fractures (32%).¹ Treatment was further divided into reduction based on fluoroscopy alone or the addition of a submeniscal arthrotomy, and rates of an articular 2 mm gap or step off were reported to be 41% and 16% respectively.¹ A smaller retrospective study found that patients treated with AAF had a mean articular step off of 1 mm (range 0–3 mm) based on post-operative radiographs.² Due to the lack of post-operative CT evaluation of patients treated with AAF as well as no randomized studies that include both techniques, the difference in the radiographic outcome of articular congruity between techniques is currently unknown.

AAF of tibial plateau fractures has the theoretical potential to improve articular visualization and reduction while minimizing comorbidity with preservation of the blood supply to soft tissues and bone. A number of retrospective studies have compared AAF to ORIF for the treatment of tibial plateau fractures. The studies found AAF was associated with fewer complications, better knee range of motion, and better Rasmussen functional scores.^3–9 The conclusions of these studies are limited due to the lack of randomization and inclusion of only single-center trials. Although there may be benefit to AAF, there remains a need for a prospective, randomized controlled trial that is multi-center in nature to confirm these findings.

The purpose of this study was to measure the influence of arthroscopy on the interobserver reliability in classification, treatment choice, determination of intra-articular pathology, and quality of fracture reduction for fractures of the tibial plateau.

2 Methods

Our institutional review board approved the use of deidentified radiographs and arthroscopic images of tibial plateau fractures. Members of the Science Of Variation Group (SOVG) and Orthopaedic Trauma Association (OTA) were invited to participate in this online study. Participation is voluntary and participants are aware that completing the survey indicates informed consent. After initial invitations were sent, three reminder emails followed over the course of one month in case they did not respond.

All participants were asked demographic questions regarding gender, years in independent practice, supervising trainees in the operating room, completion of an orthopaedic fellowship training, specialization, geographic practice location, setting of practice, and utilization of arthroscopy in practice and in treatment of a tibial plateau fracture. Surgeons were randomized at a 1:1 ratio to review eight cases of patients with either 1) anteroposterior and lateral knee radiographs alone or 2) radiographs and arthroscopic images. We used SurveyMonkey© (San Mateo, CA, USA)¹⁰ to develop the survey and allocate randomization. For each patient scenario, surgeons were asked: (1) How would you classify this injury based on the Schatzker classification?¹¹ (2) Would you recommend surgery? (3) Do you suspect any intra-articular pathology (meniscal or cruciate ligament injuries) that would benefit from surgery? And (4) based on the intraoperative images provided, do you find this reduction acceptable? Pre-reduction imaging was provided for questions 1, 2, and 3. Imaging after reduction maneuvers were applied was provided for question 4 (Fig. 1). The patients presented in the survey had tibial plateau fractures that were treated with arthroscopic-assisted fixation (AAF) at an outpatient surgery center from 2016 to 2018. All eight patients were active individuals that were younger than 50 years old.

Close

Fig. 1 Anteroposterior and lateral knee radiographs demonstrating a depressed lateral plateau fracture in a patient with a previous anterior collateral ligament (ACL) reconstruction. a) Imaging provided to surgeons in the control group for questions 1, 2, and 3.b)Arthroscopic images of the same patient demonstrating the fracture displacement, lateral meniscus, and ACL graft. Imaging included with Fig. 1a to surgeons randomized to the intervention group for questions 1, 2, and 3.c)Anteroposterior and lateral knee radiographs after a reduction maneuver. Imaging provided to the control group for question 4.d)Arthroscopic imaging demonstrating fracture line after reduction maneuver. Imaging included with Fig. 1c to intervention group for question 4.

Export to PPT

3 Results

4 Discussion

Prior studies have demonstrated that there may be benefit to AAF in the treatment of tibial plateau fractures, but no previous studies have evaluated the reliability of interpretation of arthroscopic images amongst orthopaedic surgeons treating tibial plateau fractures. The purpose of this study was to measure the influence of arthroscopic images on the interobserver reliability in classification, treatment choice, and evaluation of intra-articular pathology and fracture reduction for tibial plateau fractures.

A number of limitations of the study need to be taken into consideration. We cannot calculate a response rate because we do not know if all email addresses are active or how many of the invited participants treat tibial plateau fractures. However, because we randomize the response rate it is less important and only affects the generalizability of the findings with respect to the comparison. In addition, participants may have been influenced to recommend surgery based on the presence of surgical implants on radiographs in the survey. Although only injury radiographs accompanied questions related to surgical treatment, the surgeon could have returned to previous questions and changed answers if later radiographs demonstrated the presence of surgical implants. In addition, the relatively low rate of utilization of arthroscopy amongst surgeons surveyed may contribute to detection bias. This effect should have been similar between both groups and should not have affected our comparison of the two groups.

Our finding that arthroscopic images did not improve the reliability of fracture classification suggests that disagreement about classification may relate more to heuristics (mental short cuts) than images. Several previous studies have reported mean kappa values for inter-observer reliability of classification of tibial plateau fractures using radiographs alone, and values ranged from 0.38 to 0.47.^13–15 Our overall lower reliability (mean kappa 0.17) compared to prior studies is common with SOVG studies, perhaps because each surgeon is less prepared for the specific study than one involving just a few raters, perhaps because of the more controlled study methods, and perhaps subjects go through the images more rapidly and with less care when doing an internet survey. It is impossible to know the relative influence of these conjectures.

The observation that recommendation for surgery was not associated with arthroscopy, but was associated with surgeons supervising trainees in the operating room, suggests that personal factors such as practice setting are important influences on variation in treatment. The use of surgical procedures varies across geographical regions.¹⁶ Attitudes and beliefs about the surgeon's indications for surgery as well as surgeon personality influence surgical variation.^17–19

The observation that the agreement on adequate fracture reduction was not improved by arthroscopic images, suggests that radiographs alone provide adequate assessment of the articular surface. The fact that agreement was only fair for both groups indicates that any benefit to arthroscopy can only be realized by developing new surgeon heuristics, perhaps using specific training methods.

From a clinical perspective, the use of arthroscopy would potentially assist with intraoperative decision-making by providing direct visualization of all aspects of the tibial plateau and articular fracture alignment. However, the lack of improvement in interobserver agreement with the addition of arthroscopic images in our study may relate to the challenge of interpretation of arthroscopic images. The interpretation of arthroscopic images involving a fracture, similar to the interpretation of radiographs, is likely an important aspect of clinical experience that develops and improves over time. A first step to determining the influence of arthroscopy on the surgical treatment and outcomes of tibial plateau fractures is reliable assessment of fracture characteristics. Previously reported benefits of AAF, including lower rates of complications and better functional outcomes,^3–9 may be related to the surgeon expertise with the technique, evaluator bias, less soft tissue dissection with the use of associated percutaneous fixation, or bias towards using arthroscopic assistance with relatively simple fractures. Future studies will be necessary to determine if AAF of tibial plateau fractures is generalizable to surgeons of different training backgrounds. Future randomized controlled trials should be designed to include multiple centers with surgeons of varying training background and number of years in practice.