Bilateral tibial fractures associated with powered exoskeleton use in complete spinal cord injury – a case report & literature review

Abstract

Introduction

Powered robotic exoskeleton (PRE) physiotherapy programmes are a relatively novel frontier which allow patients with reduced mobility to engage in supported walking. Research is ongoing regarding their utility, risks, and benefits. This article describes the case of two fractures occurring in one patient using a PRE.

Case

We report the case of a 54 year old man who sustained bilateral tibial fractures while using a PRE, on a background of T10 AIS A SCI. The initial session was discontinued due to acute severe bilateral knee swelling after approximately 15 min. The patient attended their local hospital the following day, where radiographs demonstrated bilateral proximal tibial fractures. The patient was treated with manipulation under anaesthetic and long-leg casting for five weeks, at which point he was stepped down to hinged knee braces which were weaned gradually while he remained non-weight bearing for 12 weeks. The patient was investigated with DEXA scan and was diagnosed with osteoporosis. He was liaised with rheumatology services and bone protection was initiated. Fracture healing was achieved and weight-bearing precautions were discontinued, however this period of immobilisation led to significant spasticity. The patient was discharged from orthopaedic services, with ongoing rehabilitation and physiotherapy follow-up.

Conclusion

PRE assisted physiotherapy programmes are a promising concept in terms of rehabilitation and independence, however they are not without risk and it is important that both providers and patients are aware of this. Furthermore, SCI patients are at increased risk for osteoporosis and should be monitored and considered for bone protection.

Introduction

Technological advancements have driven progress in medicine for well over a century, continuously expanding our repertoire of investigations and treatments for disease states. The concept of powered robotic exoskeletons (PRE) was first explored by General Electric and the United States’ military in 1965, and exoskeleton projects have continued to attract attention and funding in the decades since for their potential applications to industry, military purposes, and medicine. From recent bibliometric analyses, the most interest in the field is focused on rehabilitation applications of these devices – and specifically with regard to the lower limbs and assisted walking [1].

Spinal cord injury (SCI) is associated with a characteristic pattern of changes in body composition – individuals experience a progressive decline in lean muscle mass and bone mineral density (BMD), with increased rates of obesity and associated metabolic complications [2–4]. There is ongoing research into potential strategies to attenuate or reverse these effects [4–9]. PREs are a particular area of interest for their potential to restore locomotor function to non-ambulatory SCI patients, and the subsequent psychological and quality-of-life benefits associated with this [10–13].

The use of PREs for rehabilitation in SCI has been associated with some adverse events however – most commonly relating to skin lesions, blood pressure changes, dizziness and syncope, and fractures [11–16]. To date, five fractures occurring in three patients using PREs have been reported in the literature, with only one previously described proximal tibia fracture [14–16].

Case presentation

Patient information

We report the case of a 54 year-old man with a background of T10 AIS A traumatic SCI since 2016, who presented to our emergency department with bilateral knee swelling. He reported that he had engaged with a physiotherapy programme in a national university which involved the use of the Ekso GT (Ekso Bionics Holdings Inc., San Rafael, California, USA) PRE, and during the first session he experienced some dizziness during an initial sit-to-stand transition. The exoskeleton stopped responding, and it was noted that both of the patient’s knees had become swollen.

The session was abandoned and the patient returned home. He reports experiencing a headache and noticed further swelling of his knees, without any skin changes or bruising.

The following day he presented to his primary care doctor, who advised him to attend the emergency department.

Plain film radiographs demonstrated bilateral proximal tibia fractures, shown in Fig. 1, and he was referred to the orthopaedic service for further management. On examination of the lower limbs, both demonstrated swelling from the level of the knee extending distally. Dorsalis pedis and posterior tibial pulses were easily palpable bilaterally, and SCI precluded sensorimotor testing of the lower limbs.

Fig. 1.

Antero-posterior (AP) and lateral radiographs of both knees from initial presentation; left side on top row.

He underwent manipulation under anaesthetic and application of above-knee circumferential casts as seen in Fig. 2. The patient was followed up in the outpatient clinic to monitor his progress, and at five weeks post-presentation cast-immobilisation was discontinued and hinged knee braces were applied for a further seven weeks. Radiographs from his follow-up appointments are shown in Figs. 3 and 4.

Fig. 2.

AP and lateral radiographs of both knees from intraoperative screening; left side on top row.

Fig. 3.

AP and lateral radiographs of both knees from outpatient department review at 5 weeks; left side on top row.

Fig. 4.

AP and lateral radiographs of both knees from outpatient review at 12 weeks; left side on top row.

The patient has a background of AIS A SCI at the level of T10 from a trauma in 2016, and a pre-existing diagnosis of osteopaenia. He has no other medical history of note. Prior to this presentation his only regular medication was for bowel care.

He is a non-smoker and his body mass index is 19.7. Prior to his SCI he was an active cyclist, and following his injury he has developed an interest in table-tennis.

Follow-up & outcomes

The patient achieved a good outcome with regard to fracture healing, and was discharged from orthopaedic follow-up at 20 weeks. Due to the mechanism of injury and his background SCI a DEXA scan was performed; this demonstrated a z-score of −3.2 in the proximal femur and the patient was referred to rheumatology for bone protection treatment.

Bilateral fixed flexion contractures of approximately 20° were noted on removal of casts at five weeks, with painless passive range of motion of 20–90°. He was liaised with local physiotherapy and the national rehabilitation service.

The rehabilitation medicine team initiated a treatment plan of regular baclofen and stretching for his contractures and an interval admission for intensive physiotherapy following fracture healing.

At most recent contact – more than two years post presentation – the patient reports that he is doing well with regard to his knees, but has residual increased tone at his hip joints which he manages with prone lying.

Discussion

SCI is an extremely challenging condition for both patients and clinicians, with multifactorial and multifaceted effects on patients and their secondary health outcomes. The loss of mobility associated with SCI gives rise to undesirable changes in body composition leading to increased risk of fractures and metabolic complications, and directly puts patients at risk of complications such as decubitus ulcers, spastic contractures, and deep vein thrombosis [2–4]. In the case we have described, osteoporosis due to the patient’s SCI prevented him from engaging with the PRE rehabilitation programme to any significant degree.

PREs are a promising new frontier for restoring locomotion and rehabilitation in SCI.

In the setting of incomplete SCI the primary outcome measures of interest are quite clear – patients can be assessed in terms of muscle power or mobility-based outcome measures such as a time or distance based walk-test, and stride metrics such as length, cadence, velocity; either on single measurement, or before and after the training programme [11–13, 17, 18].

For patients with complete SCI the outcome measures are more opaque – research to date suggests PRE training programmes can improve spasticity, bowel functionality, and confer a greater degree of autonomy based on patient reported outcome measures [11–13]. Studies have demonstrated improvements in mobility-based outcome measures as described above, however these measures are only relevant in the context of using the PRE during training sessions given that this patient population are obligate non-ambulators and the cost of these units is generally prohibitive to private ownership [11–13].

It has been postulated that supported weightbearing in the context of PRE training may attenuate bone loss and improve BMD – however these effects are not borne out in the literature and there is no strong evidence to support this effect [2, 4, 19].

At present it is not feasible to provide private individuals with PREs for extended personal use – however, one study from van Dijsseldonk et al. assessed usability in the home and community setting and demonstrated strong patient satisfaction with the PRE for exercise and social purposes, although some limitations were recognised in terms of activities of daily life [20].

The use of PREs is not without risk, as described above the most common adverse events reported to date relate to skin lesions, blood pressure lability and dizziness, syncope, and fracture [11–16]. Table 1 details previously reported fractures associated with PRE use [14–16].

Table 1.

Previously reported fractures associated with PRE programmes.

Author	Year of publication	Article type	Exoskeleton manufacturer	Total patients (n)	Affected patients (n)	Fractures (n)	Fracture site
Van Herpen	2019	Case series	ReWalka	2	2	2	Tibia
Bass	2020	Case report	Eksob	1	1	2	Calcaneus
Benson	2016	Feasibility study	ReWalka	10	1	1	Talus

^aReWalk Robotics GmbH, Leipziger Platz 15, 10117 Berlin, Germany.

^bEkso Bionics Holdings Inc., San Rafael, California, USA.

Given that this population is by definition high-risk for these sequelae, it seems prudent to implement screening programmes prior to enrolling SCI patients in a rehabilitation programme and optimise their comorbidities where possible – some studies have proposed criteria for such patient selection measures [21]. It has also been suggested that a graduated training algorithm, designed to reduce risk of fracture, may be helpful [22].

Lastly, new research has been targeted at developing prototype sensors and instrumented simulation to measure forces generated by PREs and recognise potential hazards such as malalignment [23, 24]. Further to this, sensors are also in development to measure shear forces to avoid skin breakdown due to strap placement [25].

There is conflicting evidence with regard to the feasibility of implementing PRE training programmes [12, 16, 26, 27]. This is attributed in part to the medical comorbidities of the patients themselves and the interplay these may have with adverse events, and additionally due to attrition from the training programme not meeting candidates’ expectations [16]. It is difficult to make definite statements on the topic of PREs due to the relative novelty of the technology and heterogeneity in the literature with regard to model of exoskeleton, training programme protocols, and patient populations.

Conclusion

Our patient achieved a good clinical outcome with regard to bone healing, and – with the exception of increased tone at his hip joints – he has experienced no further sequelae related to his fractures to date.

There is ongoing research into the utility, applications, and safety of PREs in the context of SCI – we suggest that patients undergo screening and optimisation of their existing health status with particular attention to bone health, and advise that patients and healthcare professionals remain vigilant for adverse events when engaging with novel technologies.

Author contributions

JM was the main author, and was responsible for data collection and liaison with the patient, review of the literature, and writing the manuscript. LN contributed to review of the literature, data collection, and contributed to writing the manuscript. DOS and MC were involved in initial literature search, and contributed to review and feedback on the manuscript. AD and CGM are senior authors who provided feedback and final approval of the manuscript.

Funding

This project did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

Information generated and analysed during the writing of this paper can be found within the text itself and referenced articles.

Competing interests

The authors declare no competing interests.

Ethical approval

Ethical approval was not required for this project given that it was a case report concerning one patient, whose informed consent was obtained for production and publication of the manuscript.

Informed consent

The authors confirm that the patient is aware that data concerning his case would be submitted for publication and proceed with his informed consent.