Fear of reinjury is associated with failure to return to preinjury level of activity after operative repair of ankle fractures

Abstract

Objectives:

Prior literature indicates fear of reinjury (FORI) as a limitation in recovery from various orthopaedic conditions. However, no studies evaluate whether fear affects return to activity after ankle fractures. This study, therefore, investigates whether FORI is associated with failure to return to preinjury activity level regardless of fracture subtype. This study also sought to determine the proportion of patients experiencing fear after ankle fractures.

Design:

Retrospective cohort study.

Setting:

Urban, quaternary care academic medical center.

Patients/Participants:

All patients who underwent ankle fracture surgery at a single institution between 2016 and 2021 were stratified by fracture subtype and final follow-up reduction quality. Multiple injuries, pilon variants, open fractures, nonanatomically reduced fractures at final follow-up, and patients with inadequate follow-up were excluded. One hundred eighty-two eligible patients completed postoperative questionnaires regarding ability to return to activity and whether patients experienced FORI.

Intervention:

The exposure was FORI.

Main Outcome Measures:

Primary outcome was whether FORI was associated with failure to return to activity.

Results:

Seventy percent (128/182) reported return to preinjury level of activity. All 36 patients who experienced FORI failed to return to preinjury activity level. Fear was significantly associated with failure to return to activity, as 100% (36/36) of patients with fear failed to return compared with 12.3% (18/146) of patients without FORI (P < 0.001). FORI was twice as common in moderate-high preinjury activity levels compared with sedentary–light levels.

Conclusion:

FORI may limit patient ability to return to activity after ankle fracture surgery. This effect may be greater for those with higher activity levels.

1. Introduction

1.1. Background

Ankle fractures are common injuries with an incidence of 187 per 100,000 annually.^1,2 Ankle fractures frequently necessitate surgical intervention to restore ankle integrity, reduce pain, and decrease risk of developing post-traumatic osteoarthritis.^3–5 Operative management of ankle fractures produces favorable functional outcomes.^6–11 Current literature supports that most patients return to baseline activity after operative repair of rotational, unstable ankle fractures.^6,7,12–23 Most of these studies evaluate return to sporting activities and report 33.3% to 100% of patients returning to activity, although the low (33.3%) and high (100%) ends of this range are reported in studies with small sample sizes (31 and 13 patients, respectively).^6,7,12–23

Prior studies have evaluated postoperative return to preinjury activity level based on fracture subtype,^{6,13,14,16–18,20} syndesmotic involvement,^{6,13,18,22,23} age,^{6,7,13–15,18,20} gender,^6,7,13,18 medical comorbidities,^24,25 and American Society of Anesthesiology score.^7,13 However, no studies have evaluated whether fear of reinjury (FORI) influences patient ability to return to activity after operative repair of unstable ankle fractures.

The impact of FORI on return to activity has been well-documented in other conditions including anterior cruciate ligament (ACL) ruptures, Achilles tendinopathies, Achilles tendon ruptures, meniscus tears, chronic ankle instability, and shoulder instability.^26–52 Fear has most notably has been documented in ACL reconstruction, with 12% to 25% of patients reporting inability to return to activity because of FORI.^26,34,38 As for patients with foot and ankle injuries, a meta-analysis conducted by Vivekanantha et al⁵² comparing 705 patients suffering from Achilles tendinopathy found that fear resulted in poor functional outcomes. Kaalund et al³⁷ also found that 15% (6/39) of badminton players who suffered Achilles tendon ruptures did not return to preinjury activity levels out of FORI. Another study evaluating 80 patients with chronic lateral ankle instability demonstrated that 64% (51/80) of these patients undergoing an augmented Broström procedure to restore ankle stability suffered from FORI postoperatively with significantly worse patient-reported functional outcomes as measured by Patient-Reported Outcomes Measurement Information System (PROMIS) scores compared with those without FORI.³⁵

1.2. Objectives

It, therefore, remains unclear to what extent FORI influences return to preinjury level of activity for the general population after ankle fracture surgery. This study investigates whether fear is associated with return to preinjury activity level, regardless of fracture subtype. Our hypothesis was that FORI would be associated with failure to return to preinjury level of activity regardless of fracture subtype among anatomically reduced fractures.

2. Methods

2.1. Participants

This investigation was an Institutional Review Board (IRB)-approved retrospective cohort study of all patients who underwent operative repair of unstable ankle fractures at a single institution between January 1, 2016 through December 31, 2021. Two fellowship-trained orthopaedic surgeons independently reviewed all radiographs. Patients were stratified by Weber classification fracture subtype, fixation technique, and reduction quality at final follow-up. Patients with multiple injuries at time of presentation, open injuries, pilon variants, nonanatomically reduced fractures at final follow-up, and those with insufficient (defined as less than 1 year) follow-up were excluded.

2.2. Study design and Setting

This study was conducted at an urban, quaternary care academic medical center. Three hundred ninety-one eligible patients were sent questionnaires via REDCap.⁵³ Patients who did not respond were contacted by subsequent phone calls and emails for a total of 5 contact attempts before ceasing contact to respect patient privacy. The protocol for patients who did not respond included a follow-up email, followed by one phone call, and then 2 final emails, unless patients declined consent. Postoperative surveys queried patients' activity level (sedentary, light, moderate, and high) before and after surgery. “Sedentary” activity was defined as little to no activity; “light” as little daily activity including walking; “moderate” as moderate daily activity, on feet most of the day, moderate intensity recreational athletics, shelf stacking, street salespeople, etc.; and “high” as hard daily activity, working long days with intense labor, high intensity athletics, etc. Patients self-reported their activity levels based on these descriptions. The questionnaire asked whether patients experienced any current at time of survey completion limitations in returning to preinjury level of activity because of FORI, with a binary yes/no response option. The survey asked patients to self-report whether they were able to achieve their preinjury activity levels.

Fractures were characterized by the Danis–Weber, or Weber, classification system, which reliably describes fracture patterns based on the level of the primary fibular fracture about the distal tibiofibular syndesmosis.^54–56

2.3. Outcomes

The study's primary outcome was whether FORI was associated with ability to return to preinjury level of activity. Fear and preinjury activity level were measured based on the responses to the survey questions detailed in the previous section. Additional outcomes included the proportion of patients who reported FORI, and whether patient characteristics (including, but not limited to, fracture subtype, sex, body mass index [BMI]) were associated with return to preinjury activity level and FORI. All patients analyzed had minimum 1-year radiographic follow-up, as any patient with less than 1-year radiographic follow-up was excluded from the study onset. Radiographic follow-up was defined as time from surgery to last ankle radiographs (AP, lateral, and mortise) obtained. Clinical follow-up was defined as time from surgery to survey completion or last contact with our department.

2.4. Description of surgery

All ankle fractures were operatively fixed according to the AO-guidelines with fluoroscopic guidance. When clinically indicated, syndesmotic reduction was achieved with Syndesmosis TightRope XP (TightRope, Arthrex, Naples, FL), InternalBrace (Arthrex), or syndesmotic screws, and the deltoid ligament complex was repaired with DX FiberTak all-suture anchor (Arthrex) according to previously described techniques.^57,58 Before incision, patients received preoperative antibiotic prophylaxis. Anatomic reduction and adequate fixation were confirmed intraoperatively before wound closure. Postoperatively, the ankle was placed in a short leg plaster splint in neutral position. Patients were instructed to remain non–weight bearing with use of assistive devices until postoperative evaluation in clinic. During immobilization, all patients were treated with either aspirin or low-molecular weight heparin for thrombosis prophylaxis. After transitioning out of the splint, patients began weight bearing under physical therapy guidance and supervision.

2.5. Statistical analysis

Hypothesis-driven testing analyzed with R Version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria) was used. With a sample size of 182, this study achieved 80% power to detect an effect size of 0.20 (Cohen w) with an alpha level of 0.05. All comparisons were made using Pearson chi-square and Fisher exact test.

3. Results

Patient characteristics are displayed in Table 1. About 182/391 patients originally contacted consented to participate and were included for final analysis. This response rate of 47% exceeds the acceptable response rate reported in the literature for studies involving orthopaedic trauma including ankle fractures.⁵⁹ About 146/182 reported no FORI and were evaluated as a subgroup analysis.

Table 1.

Patient characteristics.

	N (%)
Age
18‒25	19 (10.4)
26‒50	103 (56.6)
51‒75	58 (31.9)
76‒100	2 (1.1)
Sex
Female	123 (67.6)
Male	59 (32.4)
BMI
18.5–24.9	50 (28.2)
25–29.9	53 (29.9)
30 or greater	74 (41.8)
Diabetes
No	175 (96.2)
Yes (type I or II)	7 (3.8)
Posterior malleolus fracture
No	126 (69.2)
Yes	56 (30.8)
Syndesmosis rupture
No	92 (50.5)
Yes	90 (49.5)
Weber classification
Weber B	140 (76.9)
Weber C	42 (23.1)
Fracture subtype
Bimalleolar	46 (25.2)
Fibula only	80 (44.0)
Lateral malleolar and posterior malleolar	28 (15.4)
Trimalleolar	28 (15.4)

About 128/182 (70%) reported returning to preinjury activity levels. Average radiographic follow-up for the entire cohort (n = 182) was 15.3 ± 1.1 months. Average clinical follow-up for the cohort was 35.7 ± 7.2 months. There was no significant difference in radiographic or clinical follow-up between those with fear and those without FORI (P = 0.76 and P = 0.85, respectively).

FORI was significantly associated with failure to return to preinjury level of activity, as 100% (36/36) of patients with fear failed to return to activity compared with 12.3% (18/146) of patients without FORI who failed to return to activity (P < 0.001, Table 2). Across all groups, the highest incidence of return to preinjury level of activity was observed among patients who reported sedentary–light activity (81.8%, 54/66, Table 2). The lowest proportion of patients who returned to preinjury level of activity was those who reported high activity (60.0%, 24/40, Table 2).

Table 2.

Ability to return to preinjury level of activity.

	Overall	Return to preinjury level of activity		P
		No	Yes
Total	182	54 (29.7)	128 (70.3)
Preinjury level of activity tier				0.03
High	40 (22)	16 (40)	24 (60)
Moderate	76 (41.8)	26 (34.2)	50 (65.8)
Sedentary–light	66 (36.3)	12 (18.2)	54 (81.8)
Fear of reinjury				<0.001
No	146 (80.2)	18 (12.3)	128 (87.7)
Yes	36 (19.8)	36 (100)	0 (0)
Weber classification				0.45
Weber B	140 (76.9)	44 (31.4)	96 (68.6)
Weber C	42 (23.1)	10 (23.8)	32 (76.2)
Posterior malleolus fracture				0.15
No	126 (69.2)	40 (31.7)	86 (68.3)
Yes	56 (30.8)	14 (25)	42 (75)
Syndesmosis rupture				1
No	92 (50.5)	27 (29.3)	65 (70.7)
Yes	90 (49.5)	27 (30)	63 (70)
BMI				0.26
18.5–24.9	50 (28.2)	12 (24)	38 (76)
25–29.9	53 (29.9)	14 (26.4)	39 (73.6)
30 or greater	74 (41.8)	27 (36.5)	47 (63.5)
Age				0.84
18–25	19 (10.4)	7 (36.8)	12 (63.2)
26–50	103 (56.6)	30 (29.1)	73 (70.9)
51–75	58 (31.9)	17 (29.3)	41 (70.7)
76–100	2 (1.1)	0 (0)	2 (100)
Sex				<0.001
Female	123 (67.6)	47 (38.2)	76 (61.8)
Male	59 (32.4)	7 (11.9)	52 (88.1)
Diabetes				0.62
No	175 (96.2)	52 (29.7)	123 (70.3)
Yes	7 (3.8)	2 (28.6)	5 (71.4)

In the overall cohort (n = 182), preinjury level of activity was significantly associated with ability to return to activity (P = 0.03). However, among patients without FORI (n = 146), although, there was no significant association between preinjury level of activity and ability to return to preinjury level of activity (P = 0.10, Table 3).

Table 3.

Return to preinjury activity level with patients who reported no fear of reinjury.

		No	Yes	P
Total	146	18 (12.3)	128 (87.7)
Preinjury level of activity tier				0.1
High	31 (21.2)	7 (22.6)	24 (77.4)
Moderate	57 (39)	7 (12.3)	50 (87.7)
Sedentary–light	58 (39.7)	4 (6.9)	54 (93.1)
Weber class				0.25
Weber B	112 (76.7)	16 (14.3)	96 (85.7)
Weber C	34 (23.3)	2 (5.9)	32 (94.1)
Posterior malleolus fracture				0.6
No	97 (66.4)	11 (11.3)	86 (88.7)
Yes	49 (33.6)	7 (14.3)	42 (85.7)
Syndesmosis rupture				0.46
No	76 (52.1)	11 (14.5)	65 (85.5)
Yes	70 (47.9)	7 (10)	63 (90)
BMI				0.6
18.5–24.9	43 (30.3)	5 (11.6)	38 (88.4)
25–29.9	43 (30.3)	4 (9.3)	39 (90.7)
30 or greater	56 (39.4)	9 (16.1)	47 (83.9)
Age				0.96
18–25	13 (8.9)	1 (7.7)	12 (92.3)
26–50	83 (56.8)	10 (12)	73 (88)
51–75	48 (32.9)	7 (14.6)	41 (85.4)
76–100	2 (1.4)	0 (0)	2 (100)
Sex				0.03
Female	92 (63)	16 (17.4)	76 (82.6)
Male	54 (37)	2 (3.7)	52 (96.3)
Diabetes				0.87
No	141 (96.6)	18 (12.8)	123 (87.2)
Yes	5 (3.4)	0 (0)	5 (100)

FORI was twice as common in patients with moderate preinjury level of activity (25.0%, 19/76) and high preinjury level of activity (22.5%, 9/40) compared with patients with sedentary–light preinjury level of activity (12.1%, 8/66, Fig. 1).

Figure 1.

Percentage of patients who report fear of reinjury by activity level tier.

Weber classification subtype was not associated with return to preinjury level of activity (P = 0.45, Table 2). Number of malleoli involved (uni-, bi-, tri-) was not associated with return to preinjury level of activity (P = 0.09, P = 0.72, P = 0.69, respectively). Syndesmotic injury and posterior malleolus involvement were not associated with ability to return to preinjury activity level (P = 1.0 and P = 0.15, respectively, Table 2).

Sex was significantly associated with return to preinjury level of activity, with a lower proportion of females returning than males who returned. Approximately 61.8% (76/123) of females reported return to activity compared with 88.1% (52/59) of males (P < 0.001, Table 2). A greater proportion of females also reported FORI compared with males: 25.2% (31/123) of females reported fear compared with 8.5% (5/59) of males (P = 0.01).

Diabetes, age, and BMI were not associated with return to preinjury activity levels (Table 2). Figure 2 displays proportion of patients with each preinjury activity level by patient characteristics. A greater proportion of BMIs greater than 30 had sedentary–light activity versus high activity, although this value did not reach statistical significance.

Figure 2.

Preinjury level of activity tier by patient characteristics.

4. Discussion

FORI may limit patient return to baseline activity after ankle fracture surgery. Fear was significantly associated with failure to return to preinjury activity level. Approximately 66.7% (36/54) of patients who failed to return to baseline activity expressed activity limitations because of FORI. This large proportion of patients who do not achieve preinjury activity levels out of fear constitutes 19.8% (36/182) of patients in this study's total cohort. No other study has documented FORI as a limitation after operative repair of unstable ankle fractures. Consequently, efforts need to be directed toward addressing psychological aspects of recovery to improve outcomes in these patients recovering from ankle fracture surgery.

In our study, only 60% (24/40) of high activity patients returned compared with 82% (54/66) of sedentary–light activity patients. Colvin et al¹³ described that at 1 year postoperatively, 88% of recreational athletes returned to preinjury sporting level compared with 11% of competitive athletes. Although this study by Colvin et al focused on return to sport rather than activities more representative of the general population, it underscores that perhaps those with a more competitive, high intensity activity demand may suffer from greater limitations in returning to their preinjury levels of activity—similar to the findings of the current investigation revealing a 60% return rate for patients with a high intensity activity level compared with the 82% return rate of patients reporting a sedentary–light activity level.

Although it may appear that activity demand is associated with likelihood of return, that is not the case when FORI is taken into account. When patients with fear were excluded, 88% of patients returned to preinjury activity levels (compared with 70% overall) regardless of fracture subtype, baseline activity, or sex. These findings demonstrate that FORI may be a substantial barrier to return to preinjury activity level. FORI was twice as common in patients with moderate (25%) and high activity (22.5%) compared with patients with sedentary–light activity (12.1%), which indicates that FORI may serve as a larger barrier for patients with moderate to high preinjury activity levels than for patients with sedentary–light activity levels.

This study did not evaluate strategies to tackle FORI, as it was previously unknown that fear played such a significant role in returning to activity. Discussing fear with patients who are not progressing may be as important as the objective measurements of range of motion and strength. Now that this study identified fear as a barrier, surgeons may consider working closely with primary care physicians, physical therapists, and sports psychologists, especially for those who are high risk and are failing to progress appropriately because of fear. Therefore, in addition to focusing on physical aspects of recovery, a multidisciplinary team may also be able to discuss and address psychological aspects of recovery for patients recovering from ankle fracture surgery.

Patients should also be screened for comorbid anxiety and depression, as evidence suggests that patients with psychiatric illnesses suffer worse outcomes after ankle fracture surgery.^50,60 Future research should focus on understanding this psychological component and deriving interventions that help mitigate FORI as a barrier to functional outcomes. Moreover, Saliba et al⁶¹ argued that the Ankle Ligament Reconstruction Return to Sports after Injury (ALR-RSI) score proves a valid metric for assessing psychological readiness for return to play after ankle fracture surgery with a sensitivity of 81% and a specificity of 75%. Tools such as the ALR-RSI score should perhaps be used more frequently to assess fear in hopes of identifying patients who may suffer worse outcomes because of psychological limitations in their recoveries. Future interventions from this study team will aim to screen for patients with baseline anxiety and depression with the Generalized Anxiety Disorder-7 (GAD7) and Patient Health Questionnaire 2/9 (PHQ2/9) screening tools, work closely with physical therapists and primary care physicians to evaluate for and address fear, administer the ALR-RSI to identify patients suffering from fear but not readily apparent, and offer the services of sports psychologists to patients who may be at high risk of poorer outcomes because of FORI.

FORI has been well documented in other orthopaedic conditions, particularly ACL rupture.^{26,32,34,38,43} Regarding, ankle injuries specifically, prior studies identified that patients with Achilles tendinopathy, Achilles rupture, and ankle instability who suffer from fear have worse outcomes.^{30,33,35,37,52} Vivekanantha et al⁵² found that fear resulted in poor functional outcomes as measured by Victorian Institute of Sports Assessment Achilles and Foot and Ankle Outcome Scores Quality of Life (FAOS-QoL) scores in patients with Achilles tendinopathy. Kaalund et al³⁷ found that 15% (6/39) of badminton players who suffered Achilles tendon ruptures did not return to baseline activity out of FORI. Another study evaluating 80 patients who underwent a Broström for chronic lateral ankle instability determined that FORI was associated with clinically and statistically significant deficits in PROMIS physical function compared with those without FORI.³⁵ However, these studies did not evaluate strategies to combat fear. Prior literature does, nevertheless, suggest some helpful strategies, particularly in the form of a multidisciplinary team working closely with physical therapists. Kvist et al underscores the importance of reassuring patients regularly and tailoring rehabilitation exercises to build and promote confidence as patients regain function.³⁹ Although not in an ankle population, Culvenor et al³³ reported that psychological specific interventions such as guided imagery, relaxation, coping modeling, and visual imagery combined with an intensive rehabilitation program may reduce fear and anxiety after ACL reconstruction. Patients with fear who have this psychological component addressed early may also progress more quickly in the recovery process, as suggested by Chmielewski et al.³⁰ Therefore, in addition to focusing on physical aspects of recovery, rehabilitation protocols may also be able to address psychological aspects of recovery, especially in connection with activity progression, for patients recovering from ankle fracture surgery.⁴⁷

Seventy percent of patients reported achieving preinjury activity levels, which is consistent prior literature.^{6,7,13,18,20,22} Fracture subtype was not associated with return to preinjury level of activity, which corroborates prior literature.^6,13,18,20 Weber C fractures typically involve disruption of the distal tibiofibular syndesmosis, a critical component of ankle stability.^62,63 Thus, Weber C fractures may be considered higher level of injury fracture patterns given the associated soft tissue trauma compared with Weber B fractures. However, similar to prior literature, our findings demonstrate that anatomic reduction, regardless of fracture type, results in most patients achieving preinjury levels of activity.⁶ It is, therefore, possible that psychological limitations, rather than higher energy, more severe fracture patterns, contribute to patients' activity limitations postoperatively.

Syndesmotic disruption and posterior malleolus involvement were not associated with return to preinjury level of activity. Mixed data exist on whether syndesmosis disruption^{13,14,18,24,64,65} and posterior malleolus involvement^66–69 affect outcomes, especially regarding return to activity.

Several studies have demonstrated that females experience worse outcomes compared with males after ankle fractures although there is no anatomical or biomechanical basis for why females would have poorer outcomes.^4,6,7,18,50 This study found that 25.2% of females failed to return to preinjury level of activity because of FORI compared with 8.5% of males. As a result, this study revealing fear as a potential limitation in recovery and a statistically significantly greater proportion of females experiencing fear compared with males may provide insight into why females might have worse outcomes compared with males after operative repair of ankle fractures. Future efforts should focus on understanding the etiology of sex differences in outcomes to address this health care disparity.

BMI, age, and diabetes were not significantly associated with ability to return to preinjury levels of activity. Mixed data exist in the literature regarding whether age and diabetes affect outcomes.^6,7,13,64 Obesity, in contrast, has been well documented as an increased risk factor for worse outcomes after ankle fractures.^25,64,70 BMI likely was not associated with return to preinjury activity level in this study because a larger proportion of these patients had sedentary to moderate activity levels.

4.1. Limitations and strengths

This study has some limitations. Most notably, patients self-reported ability to achieve preinjury activity level in a binary yes/no survey, which may have introduced recall bias. However, the survey assessed current at time of survey completion FORI. A validated, reliable metric used to assess fear needs to be developed, as the binary yes/no survey evaluating fear may have limited assessment of FORI. The retrospective design may introduce additional bias. Future studies should include prospective methods to obtain more accurate representations of preinjury activity levels. The activity level scale used may have been interpreted differently by patients, which may have led to some measurement bias. The imperfect response rate (182/391, 47%) may introduce nonresponse bias. In particular, it is likely that the survey responses overrepresented females, which could influence the results, as a higher proportion of females experienced fear than males. However, this survey response rate exceeds the acceptable response rate of 30% reported in the literature for orthopaedic trauma reported by Nguyen et al.⁵⁹ In addition, psychosocial factors that may affect the presence of FORI were not collected and analyzed. It is possible that those with lower socioeconomic status may suffer from greater degrees of fear than those with higher socioeconomic status because of the severe consequences that come from missing work for patients who cannot adequately modulate their lifestyle in the postoperative period to optimize recovery out of financial necessity. In addition, it is possible that those with comorbid depression and anxiety may also suffer from higher rates of FORI.⁵⁰ We also did not have any data on catastrophizing behavior, which may have a strong relationship with FORI.^71,72 There are several scoring systems for catastrophizing behavior, so future studies should analyze these scores in relation to fear and clinical outcomes.^71,72 Future research should overall include evaluations of psychosocial factors that influence FORI in patients. Future directions should also evaluate any potential associations between the development of complex regional pain syndrome type symptoms and FORI.

Despite these limitations, this investigation has many strengths. Most notably, this study is the first to describe FORI serving as a potential limitation in patients recovering from ankle fracture surgery and may provide insight into why females have worse outcomes than males after ankle fractures. Although limitations are present in this investigation, the study yields important insight into an area of ankle fracture surgery—the psychological components of recovery—that is currently completely unaddressed and needs to start being evaluated to improve outcomes for these patients. This study is also the largest and first retrospective study evaluating the influence of preinjury level of activity on rate of return to any prior activity level—not just sporting activity—for the general population.

5. Conclusion

FORI may serve as a barrier in postoperative recovery from ankle fracture surgery, as it is significantly associated with failure to return to preinjury level of activity. Fear is an aspect of ankle fracture surgery recovery currently not being addressed, as it was previously unknown that FORI played a role in limiting patient recovery from operative repair of ankle fractures. Future studies should address psychological barriers that affect functional outcomes.

Data Availability Statement

Study data is available upon request; please contact the senior or corresponding author.