Patient Reported Outcomes for Maisonneuve Fractures using PROMIS Physical Function Scores

Abstract

Background

Limited research exits surrounding Maisonneuve fractures due to rarity. Additionally, even less data has been published evaluating the patient reported outcomes of surgically treated Maisonneuve fractures using the Patient-Reported Outcomes Measurement Information System (PROMIS) Physical Function (PF) and Pain Intensity (PI) score. The aim of this study was to put forth a cohort of patient reported outcomes using PROMIS PF and PROMIS PI

Methods

Maisonneuve fracture patients with minimum 12-month follow-up treated at a level-1 trauma center from 2006 to 2020 completed PROMIS PF and PROMIS PI tests. Medical records were reviewed to gather patient characteristics, mechanism of energy, operative variables, and complications.

Results

The final cohort consisted of 28 patients with a mean follow up of 59.6 months with a mean time between injury and primary operation of 7.9 days. The mean PROMIS PF and PROMIS PI scores of the cohort were 56.3 (SD 10.8 [95% CI 15.9, 171.0] and 40.9 (SD 7.5 [95% CI 36,44]) respectively. The mean Foot and Ankle Ability Measure (FAAM) Activities of Daily Living (ADL) and FAAM Sports were 90.4 (15.4 [95% CI 84, 109] and 75.2 (SD 28.9 [95% CI 62, 91] respectively. Thirteen patients received syndesmotic screws only, nine received plate + syndesmotic screw, three received suture-button devices only, and three patients received syndesmotic screws + suture-button device. When the methods of syndesmotic fixation were divided into three groups, (1- syndesmotic screws only, 2-plate + syndesmotic screw, 3- suture-button device only and syndesmotic screw + suture-button device), the PROMIS PF scores for each treatment group was 60.2 (SD 8.5 [95% CI 54, 66]], 46.5 (SD 9.7 [95% CI 41, 53]), 62.8 (SD 9.7 [95% CI 54, 77]), respectively. The treatment group of plate + syndesmotic screw has significantly worse PROMIS PF compared to the syndesmotic screw only group.

Conclusion

This study is the largest cohort utilizing PROMIS PF and PI to evaluate surgically treated Maisonneuve fractures. The resultant PROMIS PF an PI scores further demonstrate the favorable outcomes for surgical treatment of Maisonneuve fractures.

Level of Evidence: III

Introduction

Maisonneuve fractures are typically the result of high energy mechanisms which cause a proximal fibula fracture with concomitant injuries to the tibiofibular syndesmosis and the medial malleolus. The insult to the medial malleolus often results in damage to the deltoid ligament or an avulsion fracture of the medial malleolus. Posterior structures of the tibiotalar joint are also often associated with this type of injury.¹

The Maisonneuve fracture has generally been described using the AO (44C.1-3), Lauge-Hansen, and/ or Weber classification systems.^2,3,4 Treatment for these injuries typically involves surgical fixation via open reduction with internal fixation. However, debate remains as to what type of syndesmotic fixation is most efficacious.⁵ Additionally, some studies have been published of patients being treated conservatively without surgical management when clinicians decide certain clinical criteria are met.^5-8

While the fracture pattern involved in the Maisonneuve fracture is well known, the actual occurrence is relatively uncommon with rate of only 5 percent of ankle fractures.³ The rarity of the injury causes difficulty in accurately assessing outcomes of different types of treatment. Furthermore, publications of long-term outcomes in general have been limited with most outcomes being reported via case reports and other small-scale studies describing functional outcomes after surgical treatment.^3,5,10

To date, very little data exists on patient reported outcomes for Maisonneuve fractures using Patient-Reported Outcomes Measurement Information System (PROMIS) as a mode of evaluating outcomes.^11,12 Thus, the main objective of this study was to contribute to the literature and provide a larger cohort of patient reported outcomes using PROMIS Physical Function (PF) and Pain Intensity (PI) for surgically treated Maisonneuve fractures. Additionally, we hypothesized that patients treated with plate-assisted syndesmotic fixation would demonstrate worse physical function and activity scores compared to those treated with syndesmotic screws alone or suture-button devices.

Methods

A single-institution retrospective cohort study was performed. After Institutional Review Board (IRB) approval was obtained, a query was made to identify skeletally-mature patients with operative ankle fractures from January 2006 to February 2020. Radiographs for each patient were then reviewed to identify patients with Maisonneuve fractures. Exclusion criteria included any imprisoned patients, non-English speaking patients, and patients who died during the follow-up period. Forty-one Maisonneuve fractures were treated with definitive fixation and one fracture was treated non-operatively at our institution from January 2006 to December 2019. Of those, two patients were deceased, and one patient was imprisoned. Of the remaining eligible patients, twenty-eight (28/41) with at least one year follow-up postoperatively were contacted and agreed to complete the outcome scores (Figure 1).

Figure 1:

Illustration of how the final cohort size was obtained after patients were excluded from the study and/or lost to follow up.

From the electronic medical record, demographic data were gathered including age, gender, and mechanism of injury. End stage reconstruction was defined as either tibiotalar arthrodesis or total ankle arthroplasty. An IRB consent information packet was mailed to patients. Upon consent via mail, each patient was subsequently contacted by telephone or emailed a survey link to complete their PROMs. Patients were asked a series of questions that make up the PROMIS physical function (PF) and pain interference (PI) outcome measures. Each question relates to the patient’s current level of function and/or pain and is answered in a Likert-type scale from “Without any difficulty” to “Unable to do.” Scores range from 0 – 100 with higher scores indicating more of the domain, i.e higher function or more pain interference. Additionally, patients completed the Foot and Ankle Ability Measure (FAAM) score for both activities of daily living (ADL) and Sports. This measurement system asks questions related to the patient’s current level of function and/or pain and the Likert-scale is from “not at all [difficult]” to “unable to do” or N/A if the patient felt the question was not suitable for their current injury. For the FAAM, raw scores range from 0 – 84 on the ADL scale and 0 to 32 on the Sports scale. These raw scores are then transformed to percent scores (0 – 100%) with higher scores indicating higher function. Patients were also asked if they had undergone any subsequent surgeries on the ankle after permanent fixation or if they had ever had a surgical site infection on the affected ankle.

Statistical Analysis

Patient demographics and fixation characteristics were analyzed with descriptive statistics. A multivariate analysis of variance with a Fisher’s Least Significant Difference post-hoc test examined the differences in PROMIS PF, PROMIS PI, FAAM ADL, and FAAM SPORT across fracture fixation methods. Confidence intervals of 95 percent were also calculated for the mean of each outcome measure. Age, gender, and patients CCI score were controlled for in the model. The model met the assumptions for homogeneity, normality, and variance. No analysis was possible of medial malleolar fixation as the data failed to satisfy the assumptions for a multivariate model. Hypothesis testing was based on an alpha level <0.05. All statistics were completed with SPSS 28 (IBM, New York, United States).

Results

Our final cohort consisted of 28 patients with a mean follow up of 59.6 (range, 12.4-113) [95% CI, 15.9, 171.0] months and a mean time between injury and primary operation of 7.9 (range, 0-55) days [95% CI 40.5-119.9]. The average patient age was 41.5 (range, 18-70) [95% CI 44 -55] (Table 1). The mean and median, PROMIS PF scores of the cohort were 56.3 (SD 10.8 [95% CI 52.1, 60.5]) and 55.0, respectively. The PROMIS PI scores had a mean and median of 40.9 (SD 7.5 [95% CI 38.-, 43.8]) and 36.3, respectively. The mean and median scores for the FAAM ADL were 90.4 (SD 15.3 [95% CI 84.5, 96.3]) and 97, respectively. The mean and median of the FAAM SPORTS was 75.2 (SD 28.9 [95% CI 64.0, 86.4]) and 83, respectively (Table 2).

Table 1.

Demographics and Injury Classifications of the Patients with Maisonneuve Fractures Who Were Managed Surgically

Variable	(N=28)
Sex
F	7 (25%)
M	22 (75%)
Age
Mean (SD)	41.5 (15.6)
Range	(18, 70.0)
Mechanism
Ground level fall	8 (31.0%)
Fall from height	2 (6.06%)
MVC/MVA/MCC	2 (6.9%)
Other	10 (30.30%)
Open Fracture	0 (0%)

Table 2.

Patient Reported Outcomes of the Patients with Maisonneuve Fractures Managed Surgically Using PROMIS PF, PROMIS PI, FAAM ADL, and FAAM Sports

Variable	(N=28)
PROMIS-PF
Mean	56.3 (SD 10.8)
Median	55.0
PROMIS-PI
Mean	40.9 (SD 7.5)
Median	36.3
FAAM ADL
Mean	90.4 (SD 15.3)
Median	97
FAAM SPORTS
Mean	75.2 (SD 28.93)
Median	83
Hardware Removal	13 (46.4%)
Infections	2 (7.1%)
Ankle Fusion	1 (3.6%)

Methods of syndesmotic fixation included syndesmotic screws+/-plate and/or suture-button device. Thirteen patients received syndesmotic screws (SS) only, three received suture-button devices (SBD) only, and nine patients received SS with a plate. Additionally, three patients received SSs and a SBD (Table 3). When the methods of syndesmotic fixation were divided into three groups, (Fixation: 1- SS only, 2-plate+SS, 3- SBD only, and SS+SBD), the PROMIS PF scores for each treatment group was 60.2 (SD 8.5 [95% CI 54.1, 66.3]), 46.5 (SD 9.7 [95% CI 39.6, 53.5]), 62.8 (SD 9.7 [95% CI 55.9, 69.7]) respectively. Comparing the PROMIS PF, FAAM ADL, and FAAM Sport scores between the different treatment groups, syndesmotic screw with plate performed significantly worse than SS or suture button device (Table 4).

Table 3.

Subsets of Surgical Management for the Patients with Maisonneuve Fractures

Variable	(N=28)
Syndesmotic Fixation-
Screw only	13 (46.4%)
Suture-button device only	3 (10.7%)
Plate + Syndesmotic Screw	9 (32.1%)
Screw + Suture-button device	3 (10.7%)
Medial Malleolus Management-
Screw	10 (35.7%)
Deltoid repair	5 (17.8%)
Plate	1 (3.6%)
None	12 (42.8%)

Table 4.

Statical Analysis Using MANOVA Stats Controlling for Age, Gender, and CCI Score as Covariates for PRO of Different Syndesmotic Constructs

PRO Score	SS		Plate + SS		Suture-button device SS +SB
PROMIS PF	60.2	(SD 8.5) 95% CI (54.1, 66.3)	46.5 *(SS)	SD 9.7 95% CI (39.6, 53.5)	62.8	SD 9.7 95% CI (55.9, 69.7)
PROMIS PI	40.4	SD 6.6 95% CI (35.7, 45.1)	43.0	SD 9.8 95% CI (36.0, 50.0)	39.0	SD 6.1 95% CI (34.6, 43.4)
FAAM ADL	96.2	SD 5.4 95% CI 92.3, 100.1)	79.5	SD 23.4 95% CI (62.8, 96.2)	93.2	SD 11.3 95% CI (85.1, 101.3)
FAAM Sport	84.7	SD 18.0 95% CI (75.4, 94.0)	54.3	SD 38.6 95% CI (26.7, 81.9)	89.2	SD 14.2 95% CI (79.0, 99.4)

* P=0.005; $ p =>0.001.

Regarding medial malleolus management, fifteen of the twenty-eight patients had either avulsion fractures or medial clear space widening. Of the twenty-eight patients, ten patients received a medial malleolus screw, thirteen received no operative management of the deltoid ligament (DL) and five had DL repair via suture or suture anchor (Table 2). The mean PROMIS PF score of the nontreatment group was 53.4 (SD 9.4 [95% CI 47.7, 59.1]) and the deltoid repair group was 61.6 (SD 13.2 [95% CI 45.2, 78.0]) (p-value of 0.34). Hardware removal was completed in 42.8% (12/28) patients. Of those who underwent hardware removal mean PROMIS PF and PROMIS PI was 56.8 (SD 8.3 [95% CI 51.5, 62.1]) and 41.7 (SD 7.8 [95% CI 36.8, 46.6]), respectively. For those who never underwent hardware removal PROMIS PF, PROMIS PI, was 57.3 (SD 12.1 [95% CI 50.8, 63.8]) and 39.7 (SD 7.4 [95% CI 35.8, 43.6), respectively.

Discussion

Due to the relative rarity of the Maisonneuve fracture, limited research exists about the topic. The present study is the largest cohort of patient reported outcomes utilizing PROMIS for surgically managed Maisonneuve fractures. The injury entails a proximal fibular fracture along with the failure of the deltoid ligament or medial malleolus and diastasis of the distal tibiofibular syndesmosis.¹³ The high energies required for syndesmotic injury in addition to the proximal fibular fracture results in a severely unstable ankle fracture.^14,15 As a result, nearly all Maisonneuve fractures are treated surgically, and post-surgical outcomes have generally been reported as good.^1,11,12,16

PROMIS Physical Function scores have been validated as reliable and responsive tools for assessing outcomes in patients with lower extremity trauma, including ankle fractures.¹⁷ The findings of this study are largely consistent with those reported by Sanchez et al., which also emphasized the impact of demographic factors, comorbidities, and radiographic findings on PROMIS scores.¹² The present study further supports the generally favorable outcomes of surgical management of Maisonneuve fractures. We found the mean PROMIS PF, PROMIS PI to be 56.3(SD 10.8) and 40.9 (SD 7.5), respectively. Franovic et al. established a population average for PROMIS PF of 52.9 (SD 7.6) and PROMIS PI of 43.6 (SD 7.6) for people >40 years old.¹⁸ Our results further demonstrate the favorable outcomes of surgical management of Maisonneuve fractures, as most patients within the current cohort returned to a level of function similar to that described for the general population.

FAAM has also been validated as a reliable measure of foot and ankle outcomes.^6,19,20 Lambers et al. published FAAM outcomes for a cohort of forty-four patients who underwent surgical fixation of Maisonneuve fractures using syndesmotic screws only.²¹ Within Lambers et al.’s study, the mean FAAM score was 94, with no distinction being made between the FAAM ADL and FAAM SPORT. These results are similar to the FAAM ADL results presented in our present study of 90.4 (SD 15.3). Furthermore, Fox RS et al. published an abstract in 2020 with a cohort of patients with Maisonneuve fractures who were surgically treated with suture button devices.²² The results of their study showed a mean FAAM ADL of 91.1 (vs 90.4 [SD 15.3] for our study) and a mean FAAM-Sports of 81.7 (vs FAAM sports of 75.2 [SD 28.9] for our study). These results appear to be consistent with the FAAM ADL scores within our current study of Maisonneuve fractures.

In evaluating syndesmotic treatment for Maisonneuve fractures, we found management with plate + SS to have significantly worse PROMIS PF compared to treatment with SS only, which may be a reflection of more severe injuries necessitating larger constructs. There was no significant difference between the SS only group and the SS+SBD group. While syndesmotic injuries are commonly treated with SS+plate, at the time of the literature review, we found no studies that directly compared the outcomes of syndesmotic injury treated surgically by SS vs SS+plate.²³ The consensus on the best treatment remains unclear and requires further investigation.²⁴ Nonetheless, comparable findings have been demonstrated in other studies evaluating the effectiveness between SS+/-plate and SBD in stabilization of syndesmotic injuries.²⁵ Most notable, a randomized control trial between the two methods of syndesmotic fixation in the setting of pronation-external rotation ankle fractures was completed by Lehtola et al.²⁶ The results showed the mean Olerud-Molander Ankle Outcome Score (OMAS) to be 88 in the syndesmotic screw group and 78 in the suture button group (difference between means 7.1, 95% CI: -7.0-21.1, P = 0.32).

According to the results of the current study, reoperation rates were relatively high with 46.6% (13/28) undergoing reoperation, when compared to Andrew NS et al. which had a reoperation rate of 8.3% (2/24).¹¹ The most common reason for reoperation within our study was for symptomatic relief via hardware removal of syndesmotic screws. When PROMIS PF scores within our cohort were further subdivided into patients who underwent syndesmotic screw removal and those who did not the outcomes were nearly identical (56.8 [SD8.3] vs 57.3 [SD 12.1]). A 2021 systemic review and a 2021 meta analysis found the current literature to show no difference in function outcomes, complications rates, or pain scores for syndesmotic screw removal and concluded there is no evidence for the basis of routine removal of syndesmotic screws.^27,28 While our current study had no complications with the removal of the syndesmotic screws, infection is an obvious known complication.²⁷ Thus, as in our current study, we recommend that syndesmotic screw removal be assessed on a case-by-case basis with a patient focused conversation regarding potential benefits and risks of the procedure.

Medial malleolus ligamentous disruption is common in Maisonneuve fractures. However, the fracture pattern has been described without medial malleolus disruption.^8,29 With the main stabilizer of the ankle being the deltoid ligament, a Maisonneuve fracture with an intact deltoid ligament could be consider for nonoperative management.³⁰ Within our present study’s cohort, fourteen of the twenty-eight patients had either avulsion fractures or medial clear space widening. Of the fourteen patients, twelve received no operative management of the deltoid ligament (DL) and five had DL repair via suture or suture anchor. The mean PROMIS PF score of the non treatment group was 53.4 (SD 9.4 [95% CI 47.7, 59.1]) and the deltoid repair group was 61.6 (SD 13.2 [95% CI 45.2, 78.0]), which is within one standard deviation mean score for the general population.¹⁸ However, Guo et al. published a meta-analysis of 388 participants who experienced Weber type B or C fractures and found that the American Orthopaedic Foot and Ankle Society (AOFAS) ankle-hindfoot scores were significantly better in patients who received deltoid ligament (DL) repairs.⁹ Obvious limitations exist in comparing our present study to that of Guo et al.’s, as our study is specific to Maisonneuve fractures. None the less, our findings further support Guo et al.’s conclusion that more high-quality and prospective studies with long follow-up durations are needed to further evaluate the superiority of DL repair in ankle fractures.

Our study has limitations. While the cohort size is relatively large in comparison to the other patient reported outcome studies, the actual cohort size is small. As a result, statistical analysis within the cohort is limited. Because of the retrospective design with a longer catchment period, some patients were lost to follow up. None the less, the follow up percentage for the study was nearly 70%, which is in accordance with other retrospective orthopedic trauma studies.³¹ The mean post operative follow up for our study was nearly 5 years, with some responses occurring nearly a decade after surgery; as such, the data may not accurately reflect early patient reported outcomes. Furthermore, the study was conducted at a single institution and our results may not translate well to other centers.

In conclusion, our current study puts forth the largest cohort utilizing PROMIS PF and PI to evaluate surgically treated Maisonneuve fractures. The resultant PROMIS PF and PI scores further demonstrate the favorable outcomes for surgical treatment of Maisonneuve fractures. Additionally, treatment of the syndesmotic injury with plate+SS showed significantly worse PROMIS PF scores compared to the SS only group. These results may help counsel patients about prognosis. However, much research still is needed to further evaluate these rare injuries.