Introduction

Extremity injuries account for the majority of injuries sustained in combat [14] and are responsible for the majority of in-hospital costs and associated long-term disabilities [3, 13]. Patients who sustain high-energy limb-threatening trauma are often faced with the decision between amputation and limb salvage. In the largest study evaluating outcomes after limb salvage or amputation for high-energy lower extremity trauma in a civilian population, patients had equivalent patient-reported functional outcomes [2, 11]. However, regardless of whether a patient underwent amputation or limb salvage, patients remained severely disabled both at 2 and 7 years after injury [2, 12]. Additionally, a recent retrospective study evaluating patient-reported outcomes in a military combat-wounded population also found service members who underwent limb salvage had high levels of disability that were more significant than the service members who had undergone bilateral lower extremity amputations [5].

Before injury, service members typically run, jump, and perform energy-intensive, high-impact tasks necessary for combat readiness. Although not all of these tasks may be possible after injury, many service members view the ability to run as a surrogate for returning to normal life activities. Continued activity limitation, weakness, and pain are three primary factors that play a role in a patient’s decision to abandon the limb salvage pathway and undergo late amputation (defined as > 90 days from injury) at a rate of up to 15% [10, 21].

The Return to Run clinical pathway is a dynamic high-intensity rehabilitation program designed to return military patients undergoing limb salvage to high-level physical function with the additional goal of decreasing the likelihood patients will opt for late amputation [16, 17]. The program combines high-intensity physical therapy with the Intrepid Dynamic Exoskeletal Orthosis (IDEO). The IDEO is a customized, carbon fiber energy storage and return orthosis that returns the energy stored during the stance phase of gait to help power pushoff. It consists of a proximal patellar tendon-bearing ground reaction cuff connected by carbon fiber posterior struts to a distal supramalleolar ankle-foot orthosis (Fig. 1). The IDEO has been shown in a small cohort of patients undergoing limb salvage to result in better functional performance and to decrease the number of patients undergoing limb salvage who were considering amputation [18]. However, this study occurred at a single time point and data on validated, patient-reported outcomes were not collected.

Fig. 1
figure 1

The IDEO is shown.

Full size image

The purpose of this study was to determine if 8 weeks of participation in the Return to Run clinical pathway resulted in improvement in validated patient-reported functional outcomes and objective physical performance measures in patients with lower extremity functional deficits or pain that limits their activities. We also determined if the magnitude of recovery differed between patients who enrolled > 2 years after injury, late-entry patients, compared with those who enrolled < 2 years after injury, early-entry patients. Finally, we sought to determine whether participation in this initiative decreases patients’ willingness to consider late amputation.

Patients and Methods

We conducted a prospective, longitudinal, observational dual-cohort study. Testing the patients before and after the intervention allowed the patients to act as their own controls. This is a short-term followup report of the first 84 patients enrolled between January 30, 2012, and December 20, 2012. Patients were selected based on the following inclusion criteria: active-duty men and nonpregnant women 18 years or older who had a functional deficit of the lower extremity as well as muscle weakness, nerve injury, volumetric muscle loss, or significant pain precluding normal function and full rehabilitation participation; able to bear full weight on bilateral lower extremities; who had completed definitive surgical treatment; and were beginning their participation in the Return to Run clinical pathway. Although this work focused on lower extremity trauma, this study included three patients with injuries at the level of the spine that affected the lower extremity making them functionally similar to the remainder of the cohort. Additionally, a healthy contralateral limb was not required for inclusion.

One hundred eight patients were eligible for the study. A total of 24 patients (22%) were excluded: 21 patients were lost to followup, two withdrew consent, and one withdrew for medical reasons. There was a small subset of patients who did not complete the testing at Week 0 as a result of the physical inability to do so and a small subset at the beginning of the study that did not complete testing at the end of Week 4; however, all patients included for the final analysis had full data for Week 8. The final study group consisted of 84 patients with a broad range of injury diagnoses including lower extremity fractures, nerve and other soft tissue injuries, and posttraumatic or degenerative osteoarthritis (Table 1). The majority of the patients had symptoms from multiple injury categories including mechanical pain, neuropathic pain, weakness, stiffness, and muscle loss with an average of 3.1 injury categories per patient (range, 1–5 categories) (Table 2). Patients began participation in the Return to Run an average of 3 years after injury. Thirty-one patients started the Return to Run clinical pathway > 2 years after their injury (late-entry cohort) and 53 patients started within 2 years after injury (early-entry cohort).

Table 1 Clinical characteristics
Full size table
Table 2 Symptom-related injury categories
Full size table

The Return to Run clinical pathway has previously been described [15–17]. Briefly, the Return to Run program is a sports medicine-style rehabilitation program that incorporates a strength training program, plyometrics, and agility training. Once the IDEO is received, the program focuses on running, cutting, jumping, and military-specific training while using the energy-storing orthosis. For this study, patients initially completed a 4-week physical therapy program focused on strength training, functional movement, and core muscle strengthening to improve their force-generating ability. For the majority of patients, this portion of therapy was home-based. The patients were then fit with the IDEO and underwent an additional 4 weeks of physical therapy focused on proper use of the device to maximize energy storage and return during both low- and high-impact activities.

Patient demographics and injury data were collected at the patient’s initial visit for enrollment into the study (Week 0). Physical performance evaluations and validated patient-reported outcomes questionnaires, as described subsequently, were completed at Week 0, the end of Week 4 before receiving the IDEO, and at Week 8 after completing the physical therapy training in the IDEO. A patient questionnaire was also completed at Week 0 and Week 8 to determine the prevalence of patients who were considering amputation at the start of the program and the number who still were considering amputation at the end of the program.

The patient-reported outcomes included the Short Musculoskeletal Function Assessment (SMFA), the Veterans Rand 12 item health Survey (VR-12), and the visual analog pain scale (VAS). The SMFA is designed to detect differences in patient function and how bothersome the patient perceives their injury to be [6]. The VR-12 is a 12-question health survey based on the SF-12 that assesses physical functioning, role limitations resulting from physical problems, bodily pain, general health perceptions, vitality, social functioning, role limitations resulting from emotional problems, and mental health [9]. The VR-12 is used to compute two scores: a physical and mental component score. The VAS was used to assess the patient’s average level of pain while wearing the IDEO.

Patients enrolled in this study were all active-duty service members and at preinjury baseline were highly functioning, physically fit individuals. This study assessed four physical performance measures: the four square step test, the timed stair ascent, self-selected walking velocity [22], and the 20-m shuttle run at each time point. All of the measures except the 20-m shuttle run have been previously validated in young, healthy active-duty service members [22]. At the beginning of this study, 30 healthy subjects underwent testing for the 20-m shuttle run. They completed three trials of the measure on 2 separate days. Patients completed three trials of each measure. Each patient was timed by one of six research assistants (SG, KA, DL, MF, DM, AG) who had been trained in the testing method. The four square step test is a test of balance and agility [4]; the patient must change directions quickly while maintaining balance. A 1-inch high obstacle in the shape of a Maltese cross is placed on the floor. The patient was instructed to begin in the back, left quadrant and then to move (1) forward; (2) sideways right; (3) backward; then (4) sideways left and then to move in the reverse direction back to the original square. They were to move as fast as possible with one foot at a time but each foot had to touch the floor in each square. The timed stair ascent is a test of power. The patient was instructed to ascend 12 stairs as fast as possible while touching each stair with alternating feet and without using the handrail. Time started on the command “go” and ended when both feet reached the 12th step. Self-selected walking velocity is a test of general physical mobility and overall disability [8]. Patients were instructed to walk 25 m at a comfortable pace and the middle 15 m was timed to calculate the patient’s speed. The 20-m shuttle run assesses power, deceleration, and ability to change pace. Patients were instructed to move to a cone 10 m away and then return to the start point as fast as possible. They were allowed to rest for 5 minutes among each of the three trials.

A questionnaire pertaining to amputation was included at 8 weeks. The initial question asked, “Have you ever considered amputation of your injured leg?” It then asked the reasons why (pain, weakness, inability to run/jump). The next question was “Are you now favoring keeping your injured leg?” The answers to these questions were used to determine who was considering amputation and the number of patients favoring limb salvage after 8 weeks of the Return to Run clinical pathway.

A one-way analysis of variance (ANOVA) with repeated measures was used to examine the change in each of the physical performance measures and patient-reported outcomes during the course of the study (Week 0, Week 4, and Week 8) in both the full 84-patient cohort and the 31-patient late entry group. A two-way ANOVA with repeated measures and the Tukey-Kramer adjustment for multiple comparisons were used to determine if there were differences in the rate and magnitude of improvement in the scores in each of the physical performance measures and questionnaire scores between the early-entry and late-entry groups of patients. If the overall magnitude of change or rate of change was found to be statistically significant, the individual week p values were listed at the bottom of the table. The effect size was determined using the difference of the means. SAS 9.2 (SAS Institute Inc, Cary, NC, USA) was used for all statistical analysis. Statistical significance was set at p < 0.05.

Results

Physical Performance, Pain, and Patient-reported Outcomes

Improvements were seen in each of the four physical performance measures from Week 4 to Week 8 and in the four square step test, self-selected walking velocity, and timed stair ascent from Week 0 to Week 8 (effect size range, 24%–166%). No improvements were found for Week 0 to Week 4 (Fig. 2A–D). Despite instructions to walk at a comfortable pace during the self-selected walking velocity test, 17 patients completed the test in a manner inconsistent with the instruction (eg, walking very rapidly). Previously published data from young, healthy active-duty individuals suggest a mean self-selected walking velocity of 1.5 m/s. To prevent overestimation of treatment effect, the 17 patients who walked at a pace greater than the highest velocity in the healthy population (> 1.93 m/s) were excluded from the self-selected walking velocity test analysis [22].

Fig. 2A–D
figure 2

The graphs show the change in the (A) time for the four square step test (effect size 4.5 seconds, 41%); (B) time for the timed stair ascent (effect size 3.2 seconds, 40%); (C) speed for the self-selected walking velocity (effect size 0.3 m/s, 24%); and (D) speed for the 20-m shuttle run (effect size 1.6 m/s, 165%).

Full size image

Patients with lower extremity disability improved in the VAS pain scores and SMFA scores except in the function of the arm and hand subcategory (effect sizes range, 23%–35% improvement, all p < 0.0001). Physical health VR-12 scores improved from Week 0 to Week 8 and Week 4 to Week 8 (p < 0.0001). The mental health VR-12 score demonstrated improvement from Week 0 to Week 8 only (p < 0.013) (Table 3).

Table 3 Patient-reported outcome measures for the combined cohort (n = 84)
Full size table

Influence of Early versus Late Enrollment in the Program

The late-entry patients improved in all physical performance measures and all patient-reported outcomes from Weeks 0 and 8 (effect size ranges, 16%–45% and 27%–38% improvement, respectively, p value range 0.002 to < 0.0001) and Weeks 4 and 8 except the arm and hand function subcategory of the SMFA and the mental health VR-12 score (Table 4). There were similar rates of improvement between the early-entry and late-entry groups in all physical performance measures (p > 0.097). For all physical performance measures, except the timed stair ascent, the magnitudes of improvement were similar (p values listed Table 5). In the timed stair ascent, the early-entry group started out slower but the groups were similar at Week 8 (Table 5).

Table 4 Change over time for the greater than 2 years after injury cohort (n = 31)
Full size table
Table 5 Magnitude of change comparison of physical performance measures in early entry (n = 53) versus late entry (n = 31) patients
Full size table

There were similar magnitudes and rates of improvement (p > 0.3) for all patient-reported outcomes except VAS between the early-entry and late-entry groups as well (magnitudes; Table 6). The late-entry group had a higher VAS score at Week 0 (4.9 versus 3.3, p = 0.001) that remained higher at Week 8 (3.3 versus 2.3, p = 0.04). However, the rate of improvement of change was similar between the late-entry and early-entry groups (p = 0.3) (Table 6).

Table 6 Magnitude of change comparison of patient-reported outcomes in early entry (n = 53) versus late entry (n = 31)
Full size table

Does Participation Decrease the Number of Patients Considering Late Amputation?

At enrollment into the program, 50 patients had considered an amputation. The most common reasons were an inability to run/jump (44 [88%]) followed by mechanical pain (43 [86%]), weakness (34 [68%]), and neuropathic pain (27 [54%]). At the conclusion of the 8 weeks in the study, 41 (82%) of those 50 patients now favored limb salvage.

Discussion

Extremity injuries account for the majority of injuries sustained in combat [14] and are responsible for the majority of in-hospital costs and associated long-term disabilities [3, 13]. Previous studies of the Return to Run clinical pathway have demonstrated an improvement in physical performance and higher return to duty and deployment rates than described before initiation of this program [1, 18, 19]. To date, however, there has not been a prospective study to evaluate the effect of the Return to Run clinical pathway and use of the IDEO on patients’ physical performance over time and on validated patient-reported outcomes. The purpose of this study was to determine if 8 weeks of participation in the Return to Run clinical pathway resulted in improvements in patient-reported outcomes and validated physical performance measures. This study also sought to determine if the magnitude of recovery differed between patients who enrolled > 2 years after injury and those who enrolled < 2 years after injury and if the Return to Run clinical pathway affected patients’ willingness to consider a late amputation.

There were several limitations to this study. We did not have a control group that completed only physical therapy without the use of the IDEO for 8 weeks. A previous study at this institution found the IDEO improved patients’ physical performance on specific tasks more than two other commercially available ankle-foot orthoses and no brace [18]. Given the apparent clinical success of the Return to Run clinical pathway with the IDEO in returning patients to duty [1], the limited improvements with rehabilitation alone, as demonstrated by the lack of significant improvement in the first 4 weeks of this study when the IDEO was not worn, and we believe this issue had already been addressed by our prior study [18], we believe it would have been inappropriate to randomize the patients to rehabilitation with and without the IDEO.

This study was conducted solely in active-duty service members, the majority of whom sustained high-energy injuries in a combat environment. These are generally highly motivated patients who are physically fit before injury; it is unknown how civilian patients would respond to this program. However, the physical therapy portion of the Return to Run clinical pathway, as modified for this study, involved 4 weeks of home-based physical therapy to increase extremity and core strength and functional capacity before receiving the IDEO; it is not unlike other physical therapy programs used in the civilian sector. The IDEO combined with physical therapy may be useful in a subset of patients who are motivated and have exhausted other measures to improve functional capabilities and decrease distal lower extremity pain. Also, the severity of the injuries may differ between military and civilian patients, particularly blast injuries. The followup for this cohort was only 8 weeks. In light of growing concerns regarding the cost of medical care, a program capable of delivering positive results over a short period of time is advantageous. However, long-term followup will be necessary to determine whether these improvements can be maintained. Finally, of the 108 patients initially enrolled, 24 (22%) did not participate in the 8-week followup. A unique limitation of our population is the majority of service members in this study traveled to our institution from remote facilities for short periods of time and sometimes they may have been required to leave the training earlier than anticipated. This may have precluded them from completing full followup.

Finally, patients in this study had a wide complement of injuries and symptoms associated with those diagnoses resulting in a heterogeneous patient population (Tables 1, 2). Based on this, there were two challenges in this study with subgroup analysis. The first is the majority of the patients had multiple symptoms in their limb (average of 3.1 per patient) contributing to their lower extremity functional limitations, making division into subgroups challenging. Additionally, this cohort consisted of a relatively small number of patients for the number of diagnoses and would result in relatively small subgroups for analysis. A subgroup analysis or multivariable regression analysis would be appropriate in the future with a larger number of patients.

The Lower Extremity Assessment Project study group found worsening patient-reported outcomes between 2 and 7 years after injury in patients who had sustained high-energy lower extremity trauma and underwent limb salvage, which contradicted prior thought that continued recovery would occur during this late phase [2, 12]. Multiple small studies have shown similarly poor outcomes with respect to patient-reported outcomes in patients undergoing limb salvage [7, 20]. In contrast, this prospective study showed improvement from Week 0 to Week 8 in patient-reported outcomes both in the full cohort of patients and the cohort that was > 2 years from the time of injury. Likewise, there was an improvement in validated physical performance measures. Other than a preliminary study performed here on patients undergoing limb salvage [18], we are unaware of any other studies prospectively evaluating validated physical performance measures in patients undergoing limb salvage to quantify their improvement. Comparison to the normative data in an active-duty military population [22] reveals that after 8 weeks in the Return to Run clinical pathway, patients were able to return to a near normal self-selected walking velocity (1.46 m/s compared with 1.5 m/s) and approach a normal time for the four square step test (6.5 seconds compared with 5.7 seconds). However, in the more physically demanding tasks, the timed stair ascent and 20-m shuttle run, the patients improved but were not as fast as their uninjured peers. Further study is needed to determine if performance continues to improve with time.

Ultimately, the patients gained enough improvement in the 8 weeks of the Return to Run clinical pathway that 41 of 50 patients (82%) who initially were considering amputation at Week 0 favored limb salvage after completing the 8 weeks of the program. A recent study in an active-duty military population found better patient-reported outcomes in amputees compared with patients undergoing limb salvage, even if the service member had undergone bilateral lower extremity amputation [5]. A key aspect of that study was that the injuries occurred between 2003 and 2007 before the development of the IDEO and the Return to Run clinical pathway. The lack of a standardized postinjury rehabilitation program and custom orthotic for the patients undergoing limb salvage likely contributed to their findings. The data in this current study suggest the IDEO may provide an advantage by combining the ability for energy storage and return similar to an energy-storing prosthetic. Furthermore, the proximal portion of the IDEO is modeled after a patella tendon-bearing prosthetic. This may allow enough off-loading of the lower leg to provide pain relief along with ground reaction propulsion.

Also with respect to amputation, we aimed to obtain data on how many patients enrolled in this intervention had pain or disability severe enough that they considered amputation. Likewise, we queried the effect this intervention had, at least in the short term, affected their feelings on amputation. We directly asked the patients in two dichotomous questions regarding their consideration of amputation surgery. Although the decision to pursue an amputation is a complex and multifactorial decision, we believe the patient’s perspective on this is one of the most important measures of this program. Also, the results were not shared directly with the treating clinicians to reduce the likelihood of participants biasing their answers to meet clinician preference. These questions were also intended to help guide our clinical recommendation of attempting 8 weeks of this intervention versus amputation. If patients in this study demonstrated improvements in pain and function but the majority still desired amputation after 8 weeks, our clinical recommendation of which patients should go through this intervention may need to include only those patients who do not initially consider amputation. However, as demonstrated in this population, 41 of 50 patients who initially considered amputation favored limb salvage at the end of 8 weeks. In the future, we hope to assess at long-term followup how many patients actually elect to have an amputation.

In conclusion, patients who completed 8 weeks of the Return to Run clinical pathway with the IDEO improved in terms of validated physical performance measures, patient-based outcomes, and decreased desire for amputation in a short timeframe. Importantly, improvements also occurred in the patients who began the Return to Run clinical pathway and enrolled in this study > 2 years after their injury, a patient population previously shown to become more disabled with time. The strengths of this study include the prospective assessment of patient-reported outcomes and physical performance measures at multiple time points before the Return to Run clinical pathway, before receiving the IDEO at the end of the intervention, allowing patient comparison over time. Additional study is necessary to determine if the documented benefits are sustained. Finally, efforts are underway to determine if the Return to Run clinical pathway with the IDEO can be successfully implemented at additional military centers and in patients > 2 years from injury while sustaining similar improvements in patient outcomes. The ability to translate this integrated orthotic and rehabilitation program into the civilian setting is unknown and warrants further investigation.