Introduction

Open fractures are usually sustained from high-energy trauma causing significant soft tissue damage, thus increasing the risk of associated complications such as deep infection and non-union. Despite the adoption of classification systems such as the Gustilo–Anderson classification and the continued development of management principles for open fractures of varying severity, complication rates remain high [1]. A systemic review conducted by Papakostidis et al. [2] on complications of open tibial fractures, the most common (44%) long-bone to sustain open fractures, showed deep infection rates of up to 36% and non-union rates of up to 64%. Similarly, a systemic review conducted by Giannoudis et al. [3] showed infection rates of up to 13.3% for open femur fractures.

In addition, up to 17% of open femur fractures required re-operation while tibial fractures that are open have been shown to be an independent predictor of re-operation, with a relative risk of 5.21 compared to closed tibial fractures [3, 4]. Open tibial fractures in a level 1 trauma centre with a mean injury severity score of 15 was also associated with a significant length of hospital stay of 21 days [5]. High complication rates coupled with high re-operation rates and lengthy hospital stays contribute to significant morbidity and economic costs incurred by open fractures. Clinical pathways have been implemented successfully in the treatment of other conditions such as hip fractures and could potentially be implemented in the management of open fractures.

In view of the high morbidity and economic costs in managing open fractures and their complications, in addition to the purported improved outcomes clinical pathways can bring, our tertiary trauma centre implemented an Open Extremity Fracture Clinical Pathway in January 2015. The clinical pathway was implemented electronically where a documentation template was made available on the electronic health record system. A brief outline of the key components of the pathway is shown in Fig. 1 (“Appendix”). The same template was used throughout the entire length of stay of a patient which facilitated communication within the treatment team especially during inter-shift handovers. Also, a secondary goal of the clinical pathway was to keep the expected length of stay of patients with open extremity fractures to within 10 days.

This study reviews and compares the clinical outcomes before and after the implementation of an open extremity clinical pathway, focusing on open tibial and femur fractures.

Methodology

Approval from the local research ethics committee was obtained prior to the start of the study. A retrospective review of medical records was conducted. All patients with open tibial and femur fractures admitted in the 2-year period before implementation of the pathway (January 2013–December 2014) and the subsequent 2-year period after pathway implementation (January 2015–December 2016) were identified using the institution’s case mix database based on diagnosis codes. Patients with follow-up data for at least 1 year were included. Patients with a delay in presentation for more than 48 h, who received a primary or delayed amputation, or who were lost to follow up were excluded from our study.

Data of selected patients were extracted from the institution’s electronic medical records and operating theatre records and was abstracted for patient demographics (age and sex), fracture characteristics (location and Gustilo–Anderson classification type) and fixation methods. The primary outcome measures were complications of superficial wound infection (involving skin and subcutaneous tissues only), implant infection, delayed/non-union and flap failure, occurring in the immediate post-operative period and 1 year follow-up period. Secondary outcomes including length of stay, time from emergency department (ED) entrance to first wound debridement, time from ED to flap coverage (in Gustilo–Anderson type IIIb and above cases) and total number of operations required (including operations for complications) were calculated.

Standard descriptive statistics were used to describe groups. Chi square test was used to evaluate categorical variables (sex, fracture location, Gustilo–Anderson classification type, fixation method and primary outcomes) and the non-parametric Mann–Whitney U test was used to compare continuous variables (age and secondary outcomes) between the pre-pathway and post-pathway groups. Significance was set at P < 0.05 for all tests.

Results

A total of 43 pre-pathway and 46 post-pathway patients were included in our study. Patient demographics and description of fractures of both groups are shown in Table 1. There was no significant difference found with regards to demographic distribution (age and sex) and fracture description (location of fracture and Gustilo–Anderson classification type) between the two groups. Notably, there was a statistically significant decrease in the proportion of patients who underwent external fixation after implementation of the pathway from 47 to 26% (P = 0.045).

Table 1 Patient demographics and fracture types of pre-pathway and post-pathway groups
Full size table

A total of 15 pre-pathway and 14 post-pathway complications were noted (Table 2). Three pre-pathway and 1 post-pathway patients had more than one complication. Similarly, there was no significant difference found in the number of complications after the implementation of the pathway.

Table 2 Complication rates before and after implementation of pathway
Full size table

Differences in secondary outcomes between pre-pathway and post-pathway groups are demonstrated in Table 3. There was a significant reduction in length of hospital stay (P = 0.00932), with a 37.5% decrease in median duration of stay from 11.2 to 7 days after implementation of the pathway. No significant differences were found in terms of time to first wound debridement, total number of operations required and time to flap coverage between pre-pathway and post-pathway groups.

Table 3 Secondary outcomes of all cases before and after implementation of pathway
Full size table

We performed a subgroup analysis of the Gustilo–Anderson type III cases (Table 4). A total of 10 pre-pathway and 11 post-pathway complications were observed, with no significant difference in the total number of complications. With regards to secondary outcomes, there were significant reductions in length of hospital stay and total number of operations required after the implementation of the pathway. Median length of hospital stay decreased by 46.7% from 15 to 8 days and total number of operations required decreased by 50% from a median of four operations to two operations (P = 0.0001 for both). However, there was a significant increase in time to first wound debridement, from a median time of 5.20–5.92 h, an increase of 13.8% (P = 0.0367).

Table 4 Subgroup analysis of Gustilo–Anderson type III cases before and after implementation of pathway
Full size table

Discussion

Clinical pathways (CPWs), also variously called ‘integrated care pathways’ or ‘critical pathways’ are evidence-based workflow processes targeting specific clinical conditions within a given time period. Clinical pathways were first implemented in healthcare from the 1980s and have since become more prevalent due to multiple factors including the increased emphasis on standardized care, greater multidisciplinary input in patient care and the increasing need of institutions to project financial outcomes [6]. Clinical pathways assist in improving healthcare delivery by detailing sequential steps in the care process, improving coordination within the multi-disciplinary team, assisting in proper documentation, and aiding evaluation of variances and outcomes [7].

A 2010 Cochrane review on the outcomes of clinical pathways concluded that CPWs were associated with improved clinical outcomes such as reduced in-hospital complications without negatively affecting duration of patient stay and hospital costs [8]. While clinical pathways traditionally target conditions more chronic in nature with a longer continuum of care during which interventions can be made, there exist potential value in implementing clinical pathways to conditions that are more acute in nature such as open fractures. This has yet been fully explored in literature.

Open fractures are associated with high rates of complications, resulting in significant morbidity and healthcare costs. Established concepts in the management of open fractures include accurate fracture classification, early prophylactic antibiotic administration and early wound debridement [9]. Numerous publications comparing outcomes of various fracture fixation methods have also been published. Specifically for open tibial fractures, intramedullary nailing is currently the preferred method of fracture stabilization compared to plate fixation or external fixation due to lower secondary surgery and infection rates [10]. More recent trends also include the use of orthobiologics and adjunctive wound care techniques such as negative pressure wound therapy.

Beyond developments in the surgical treatment of open fractures, clinical pathways offer a promising approach to target the high complication rates and cost of open fractures. Although there is limited literature on clinical pathways implemented for open fractures, a meta-analysis of clinical pathways for hip fractures showed that the implementation of clinical pathways reduced rates of complications that commonly affect hip fracture patients [11]. Findings from our study show that there is potential for clinical pathways to improve outcomes of open fractures and reduce associated costs, which is especially pertinent in the current landscape of burgeoning healthcare costs and increasingly complex patient care involving multiple members of the multidisciplinary team.

In our study, we found a significant reduction in length of stay after the implementation of an open fracture clinical pathway. Reduction in length of hospital stay has often been noted after the implementation of clinical pathways in literature [8]. Furthermore, it has been noted that clinical pathways involving invasive interventions, such as surgical treatment of open fractures, could potentially result in a greater magnitude of length of stay reduction when compared to non-invasive interventions adopted in chronic medical diseases [12]. The premise for this is that invasive interventions are less varied and can more easily be standardized in a clinical pathway compared to non-invasive interventions. Hence invasive interventions would tend to yield a more significant economic benefit in terms of length of stay reduction when implemented in a clinical pathway. The reduction in length of patient stay is a reassuring finding that indicates that the clinical pathway adopted has been effective in improving outcomes in patients with open fractures. Better surgical planning with protocols and improved handover within the multidisciplinary team could explain this, potentially also contributing to cost savings for patients.

The length of stay reduction could also be attributed to the reduced proportion of patients treated with external fixation. With the implementation of the clinical pathway, trauma surgeons were engaged early upon admission and were guided by a protocolised timeline. As a result, surgeons were forced to be more discerning regarding their fracture fixation methods. Surgeons who previously opted for external fixation for patients who could have been treated with internal fixation now opted for internal fixation as the initial surgery.

The 1-day reduction in time to flap closure also reflects better coordination of surgeons and focus on expedient care, although this was not significant in our study possibly due to its small numbers.

As Gustilo–Anderson type III fractures are an independent risk factor for increased complication rates, a subgroup analysis was conducted to investigate if there were any differences in outcomes within Gustilo–Anderson type III fractures after pathway implementation [1, 13]. The reduction in length of hospital stay was also observed in the subgroup analysis. In addition, there was also a significant reduction in the number of operations required. These improvements were not at the expense of increased complication rates.

When compared to other studies involving open tibial fractures of similar severity, our study showed shorter length of patient stay and also reduced number of operations required. A systemic review by Wood et al. showed that studies of Gustilo–Anderson III tibial fractures with intermediate time to flap closure of 5–7 days, comparable to a median of 6.4 days in our study, had lengths of patient stay of 3.5–4.2 weeks [14,15,16]. A more recent study by Chua et al. involving 323 open tibial fractures also showed similar findings. Patients with Gustilo III tibial fractures had a mean length of stay of 33.8–47.9 days and required 4.32–7.84 operations on average [17]. In comparison, median length of stay in Gustilo–Anderson III fractures in our study was 8 days and number of operations required was 2. This further supports the use of a clinical pathway to improve patient outcomes.

However, there was a significant increase in time to first debridement of Gustilo–Anderson type III fractures after pathway implementation. We believe this to be a negligible increase in time to debridement, measured in terms of minutes. In addition, the post-pathway median time to debridement of 5.92 h is still well below the recommended 12–24 h cut-off suggested by guidelines such as that by the National Institute for Health and Care Excellence and the British Association of Plastic, Reconstructive and Aesthetic Surgeons [9, 18]. Recent reviews conducted by Schenker et al. and Crowley et al. also suggest that the historical “6 h rule” favoured by older guidelines has little support in available literature and should be re-evaluated [19, 20]. This may explain why the increase in time to debridement is not accompanied by increased complication rates or poorer secondary outcomes.

It should be noted that possible confounding factors to decreased length of stay of patients have been considered. Within the study period, there were no major administrative or organisational changes in the department or hospital that could have influenced the LOS of patients. Furthermore, as a major trauma centre in the country, negligible number of patients were referred to other hospitals and it is unlikely this would have changed within the study period.

Limitations of this study include our inability to accurately pinpoint the time of injury and the need to use the ED entrance time to calculate to time to first wound debridement and flap closure. However given the small geographical area of our country and an efficient ambulance response time, we do not foresee that the results of this study would be significantly affected by this limitation. Our study is also unable to provide information on which specific components of the pathway resulted in the improvements seen. Further studies could involve using clinical pathway audit tools to deconstruct the pathway and provide more information on how the pathway could be modified to enhance outcomes. Further studies are planned to evaluate the effect of the clinical pathway on open fractures affecting other sites such as the upper limbs, providing a larger sample size for a more robust analysis and more variables for subgroup analysis.

Conclusion

Our study has demonstrated that the implementation of an open extremity fracture clinical pathway significantly reduces the length of hospital stay and number of operations required in patients with open tibial and femur fractures, possibly by encouraging surgeons to be more discerning regarding their fracture fixation methods and decreasing the number of external fixation surgeries. These improvements are achieved without compromising complication rates at 1-year follow-up.