1 Introduction

Wound complications, infection, and malunion can be devastating complications after surgical management of pilon fractures.^1–4 These complications can lead to poor functional outcomes and even threaten limb viability.^5–7 Historically, early definitive fixation of these fractures led to high rates of wound complications and infection.^1,8 Staged management with delayed open reduction internal fixation (ORIF) has significantly decreased the rates of infections and wound complications.^1,8 Pilon fractures are often associated with fibula fractures. Proper management and the timing thereof for concurrent fibula fractures in these injuries remains contested.^9,10

Some have argued that fibula fixation is not routinely necessary to prevent malunion or construct failure.^11–13 Others have argued that ipsilateral fibula fixation is important for optimizing outcomes by reducing the risk of malunion and post-traumatic arthritis.¹⁴ Some studies have suggested that concurrent fibula fixation may be associated with a higher risk of wound complications, although these studies had limited statistical power.^12,13

Prior studies investigating outcomes after ipsilateral fibula fixation in pilon fractures have been limited by small sample sizes.^12–14 The purpose of this study is to harness statistical power from a large national database to compare complications between pilon fractures managed with tibial fixation alone versus combined tibia and fibula fixation. Specifically, we will investigate the rates of wound complications, surgical site infection, and reoperation.

2 Methods

Data were obtained through the American College of Surgeons’ National Surgical Quality and Improvement Project (ACS NSQIP) database.^15,16 This database reports patient demographic data and outcome measures after various surgical procedures from over 700 participating institutions.¹⁵ Trained clinical reviewers collect and review the data.¹⁵ The ACS NSQIP database has been previously validated and utilized for numerous studies across the surgical subspecialties.^17–22

The ACS NSQIP database was queried to identify patients that underwent ORIF for tibial plafond fractures between January 1, 2015 to December 31, 2020. Data were extracted using current procedural terminology (CPT) codes: 27827 (open treatment of the tibial plafond with internal fixation of tibia only) and 27828 (open treatment of the tibial plafond with internal fixation of tibia and fibula). Cases with incomplete or missing data were not included in the final analysis. Open fractures were excluded from this analysis. Cases with concurrent open procedures for other injuries were excluded from this analysis, including open procedures on the ipsilateral extremity (i.e. proximal tibia fractures, metatarsal fractures, talus fractures, etc.).

Patient variables of interest included age, sex, body mass index (BMI), American Society of Anesthesiologist (ASA) classification, and medical comorbidities. The specific medical comorbidities of interest were diabetes (insulin versus non-insulin dependent), smoking, chronic obstructive pulmonary disease (COPD), ascites, congestive heart failure (CHF), hypertension, renal failure, dialysis use, chronic steroid use, and bleeding disorder. The ASA classification was reported as previously described in the literature.^23,24

Outcomes of interest included 30-day mortality, unplanned hospital readmission, reoperation, non-home discharge, and both surgical and medical complications. Medical complications included pneumonia, pulmonary embolism, unplanned intubation, ventilator dependence for greater than 48 h, renal insufficiency and failure, urinary tract infection (UTI), cerebrovascular accident (CVA), cardiac arrest, myocardial infarction, deep venous thromboembolism (DVT), systemic sepsis, and septic shock. Surgical complications included surgical site infection, dehiscence, and blood loss requiring transfusion.

Chi-squared analysis was used to compare categorical variables and the incidence of various complications between the two groups of interest. Fisher exact tests were used when appropriate – when cell counts were less than 20 or if any given cell had expected frequency of 5 or less. Multivariate logistic regression was used to identify any variables associated with the various outcomes of interest: 30-day mortality, unplanned readmission, reoperation, non-home discharge, and surgical and medical complications. Statistical significance was defined as p < 0.05. All statistical analyses were undertaken using IBM SPSS Version 24.0 (Armonk, NY: IBM Corp). Institutional Review Board approval was not required for this study, which utilized de-identified data from an available national surgical outcomes database.

3 Results

A total of 3340 patients were identified in the initial database query. Sixty-nine cases were excluded for having other concurrent open procedures performed during the index procedure. One-hundred and fifty-one cases were excluded for having open pilon fractures. A total of 3120 patients were included in the final cohort: 1530 patients underwent tibia fixation alone and 1590 patients underwent both tibia and fibula fixation.

Table 1 summarizes the patient demographics and medical comorbidities of the population sample. The combined tibia and fibula fixation group had more female patients (55.3 % vs. 39.9 %, p < 0.001), fewer patients aged 18–39 years (33.0 % vs. 39.3 %, p < 0.001), and more patients aged 50–59 years (21.1 % vs. 17.2 %, p = 0.006) and 60+ years (27.1 % vs. 22.0 %, p < 0.001). The combined group also had more underweight patients (7.5 % vs. 5.6 %, p = 0.029), more patients with hypertension (28.5 % vs. 24.5 %, p = 0.012), fewer patients with ASA class 1 (15.7 % vs. 18.8 %, p = 0.025), and more patients with ASA class 3 (25.9 % vs. 22.0 %, p = 0.011).

Table 2 compares the 30-day outcome measures between the two groups. The combined group had a higher rate of reoperation (2.3 % vs. 1.1 %, p = 0.013) and non-home discharge (14.8 % vs. 11.2 %). However, the two groups had similar rates of mortality, hospital readmission, medical complications, and surgical complications (p > 0.05). Specifically, the two groups had similar rates of surgical site infection (2.4 % vs. 2.2 %, p = 0.663) and wound dehiscence (0.6 % vs. 0.5 %, p = 0.696).

Table 3 summarizes the multivariate logistic regression assessing the association between several variables of interest and 30-day outcome measures. Notably, combined tibia and fibula fixation was independently associated with reoperation (1.939 [1.081–3.477], p = 0.026). Older age, diabetes, and bleeding disorders were also associated with reoperation (p < 0.05). Older age, diabetes, COPD, and ASA class of 3 or 4 were independently associated with unplanned hospital readmission (p < 0.05). Male gender, older age, diabetes, COPD, CHF, hypertension, chronic steroid use, bleeding disorder, and ASA class of 3 or 4 were independently associated with a non-home discharge (p < 0.05). Bleeding disorders and ASA class of 3 and 4 were independent predictors of overall surgical complications. Older age, COPD, and ASA class 4 were independent predictors of overall medical complications.

4 Discussion

This study provides the largest sample to date comparing short-term complications after pilon ORIF with and without fibula fixation. Patients in the combined tibia and fibula fixation group had a slightly higher rate of 30-day reoperation and non-home discharge. However, they had similar rates of medical and surgical complications, including surgical site infection and wound dehiscence. Additionally, patients in the combined fixation group were also older and with a higher comorbidity burden. After controlling for confounding variables via multivariate logistic regression, combined tibia and fibula fixation was only independently associated with a higher risk of unplanned reoperation. Other factors, including age, medical comorbidities, and ASA classification, were more predictive of short-term complication measures.

Rüedi and Allgöwer classically described operative management of tibial plafond fractures that were previously considered non-amenable to surgical fixation.²⁵ One of the four key principles described was establishing the correct length of any associated fibula fracture.²⁵ However, historically high infection rates after immediate ORIF of pilon fractures were described in several cohorts, often attributed to compromised soft tissues from high energy injuries.^1,8 Sirkin et al. then described a staged protocol comprised of ankle-spanning external fixation of the tibia and immediate ORIF of the fibula when fractured, followed by definitive ORIF after soft tissue rest.¹ For closed injuries, the authors had a 100 % wound healing rate with only 17 % of patients having partial-thickness skin necrosis, which were managed with local wound care and oral antibiotics.¹

Some prior studies have investigated short-term complications after ORIF of pilon fractures using national datasets. However, to our knowledge, none have assessed the influence of supplemental fibula fixation on short-term outcomes. Panton et al. demonstrated that patients with metabolic syndrome had a higher risk of unplanned hospital readmission, infectious complications, and major adverse events after pilon ORIF.²⁶ Masrouha et al. described that patients undergoing pilon ORIF had significantly higher rates of wound complications and hospital length-of-stay relative to patients with rotational ankle injuries.²⁷ They described a wound complication rate of 1.43 % in a group of 1122 cases of pilon ORIF.²⁷ These studies did not compare cases with or without supplemental fixation, nor investigate the influence of supplemental fibula fixation on short-term complication rates.

Importantly, it has been demonstrated that fibula fractures are independently associated with pilon fracture pattern and overall severity.^28,29 These findings are important to consider, because soft tissue complications may have otherwise occurred in these more severe injury patterns. Luk et al. published a case-control series demonstrating that the intact fibula group had a higher prevalence of AO B2 and B3 type fractures, while the fractured fibula group had more AO C3 type fractures.²⁸ Barei et al. found similar results in a series of pilon fractures that were ranked on a numerical scale of radiopgraphic severity by three orthopaedic traumatologists.²⁹ Fibula fractures were more commonly associated with AO C-type fractures.²⁹ However, there were no differences found in the radiographic severity of C-type fractures with or without fibular fractures.²⁹

The necessity and role of supplemental fibula fixation in pilon injuries remains somewhat unclear. Fixation of the fibula may improve the capability to obtain and maintain appropriate reduction in certain pilon injuries, but it may also increase rates of nonunion¹⁰ or soft tissue complications.¹¹ Previously, Lee et al. described a series of patients with pilon fractures and ipsilateral fibula fractures, and they found that supplemental fibula ORIF yielded a better clinical outcome, measured via the rating scale of Teeny and Wiss.¹⁴ They also reported a trend towards lower rates of post-traumatic arthrosis in the group with fibula fixation, although this was not statistically significant.¹⁴ However, Hong et al. recently found no difference in malunion, delayed union, nonunion, or implant failure in fractures managed with or without fibula fixation.¹³ Kurylo et al. found a higher rate of overall complications – mainly driven by removal of hardware – in patients with pilon injuries that underwent supplemental fibula fixation.¹² These complications were compared to two control groups: i) patients with pilon injuries without fibula fractures and ii) patients with pilon injuries with fibula fractures that were managed without supplemental fixation.¹²

One important consideration for supplemental fibula fixation is the relative risk of short-term complications, particularly surgical site infection, wound healing complications, and unplanned reoperation. However, many of the aforementioned studies had relatively small sample sizes and may have been underpowered to detect these differences. Our study provides a large sample size with significant statistical power to compare complications between pilon ORIF with and without fibula fixation using a validated data source. Overall, supplemental fibula fixation was associated with a slightly higher risk of 30-day reoperation, but it was not associated with a higher risk of surgical site infection or wound dehiscence.

This study does have limitations, which should be considered in the context of its conclusions and the current literature. First, the study is a retrospective study of a national surgical database, which is inherently subject to selection bias. The database itself does not provide injury-specific data that could be important metrics for analysis, such as AO fracture classification, extent of soft tissue injury, and the specific surgical approach or fixation strategy employed. Access to the ACS NSQIP database is limited to participating institutions, which tend to favor large health systems and academic institutions, so the population may not be entirely representative of the general population. Nonetheless, these data may help clinicians with decision-making and counseling their patients regarding the relative risk of specific complications after surgical management of pilon fractures.

5 Conclusion

In summary, supplemental fibula fixation during tibial plafond ORIF yielded a slightly higher incidence of unplanned reoperation and non-home discharge relative to tibia fixation alone. Supplemental fibula fixation was independently associated with a higher risk of unplanned reoperation. Fibula fixation was not associated with a higher rate of surgical site infection or wound dehiscence. Baseline patient demographics and comorbidity burden appear to be more important in predicting poor short-term outcomes rather than surgical fixation technique.

CRediT authorship contribution statement

Mark A. Plantz: Conceptualization, Data curation, Data Analysis, Investigation, Methodology, Manuscript Writing, Writing – review & editing, Visualization, Software. Jennings Dooley: Conceptualization, Data Analysis, Investigation, Methodology, Manuscript Writing, Writing – review & editing. Tyler Compton: Conceptualization, Data Analysis, Investigation, Methodology, Manuscript Writing, Writing – review & editing. Rachel Bergman: Conceptualization, Data Analysis, Investigation, Manuscript Writing, Writing – review & editing. Michael Peabody: Conceptualization, Data Analysis, Investigation, Manuscript Writing, Writing – review & editing. Jasmin Vargas: Conceptualization, Investigation, Writing – review & editing, Visualization. Erik B. Gerlach: Conceptualization, Investigation, Writing – review & editing, Visualization. Muhammad Mutawakkil: Conceptualization, Writing – review & editing, Management, Validation, Project administration. Milap Patel: Conceptualization, Writing – review & editing, Management, Validation, Project administration. Bennet Butler: Conceptualization, Writing – review & editing, Management, Validation, Project administration. Anish Kadakia: Conceptualization, Writing – review & editing, Management, Validation, Project administration.

Ethical statement

The data utilized herein were obtained from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database. This is an available database of deidentified data that is collected from participating institutions. Therefore, this study is waived from institutional board review.

Guardian/patient consent statement

The data utilized herein were obtained from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) database. This is an available database of deidentified data that is collected from participating institutions. Therefore, patient/guardian consent is not relevant for this database study.

Funding statement

No external or internal funding was utilized for purposes of this study.