ORIGINAL ARTICLE
Year : 2022 | Volume
: 8 | Issue : 2 | Page : 103--109
Early experience managing complex deformities using Autostrut™ robotic-controlled hexapod external fixators
Jason Shih Hoellwarth, S Robert Rozbruch, Taylor J Reif, Adam Daniel Geffner, Austin T Fragomen
Limb Lengthening and Complex Reconstruction Service, Hospital for Special Surgery, New York, NY, USA
Correspondence Address:
Adam Daniel Geffner
519 East 72nd Street, New York, NY 10021
USA
Abstract
Context: Hexapod circular external fixators allow bone manipulation in all planes to correct complex deformities. However, the patient must perform the strut adjustments consistently and correctly, often multiple times daily for weeks or months, to achieve intended corrections. This presents a potential source of variability, error, and anxiety to the patient. A computer-programmed, robotic automated motorized strut adjustment technology (Maxframe Autostrut™ Multi-Axial Correction System, Orthospin Ltd., Yoqneam, Israel) has been developed which automatically adjusts the struts without patient or clinician involvement. Aims: The aims of this study were as follows: first, to determine whether the motors performed the programmed initial and residual schedules and, second, to identify technology-specific problems and their management. Settings and Design: This was a retrospective observational study of a consecutive series of the first 16 patients who had the motorized hexapod frame applied. Subjects and Methods: A chart review was performed to record demographic information, indications and goals for hexapod frame care, whether the care goals were achieved, and whether unexpected and/or adverse events occurred (such as technical difficulties and medical complications) and the management of those issues. Statistical Analysis Used: Not applicable. Results: All patients achieved the index and residual adjustments as programmed. Conclusions: The Autostrut™ system appears reliable and safe. It executes programmed index and residual adjustments as well as strut change scenarios as directed. The system recognizes unexpected mechanical or programming issues and ensures patient safety by halting progress and alerting the patient. Future versions of the technology may benefit from added features such as remote reprogramming or current strut position monitoring.
How to cite this article:
Hoellwarth JS, Rozbruch S R, Reif TJ, Geffner AD, Fragomen AT. Early experience managing complex deformities using Autostrut™ robotic-controlled hexapod external fixators.J Limb Lengthen Reconstr 2022;8:103-109
|
How to cite this URL:
Hoellwarth JS, Rozbruch S R, Reif TJ, Geffner AD, Fragomen AT. Early experience managing complex deformities using Autostrut™ robotic-controlled hexapod external fixators. J Limb Lengthen Reconstr [serial online] 2022 [cited 2023 Mar 25 ];8:103-109
Available from: https://www.jlimblengthrecon.org/text.asp?2022/8/2/103/366303 |
Full Text
Introduction
Deformity correction using external circular frame fixation allows control of deformities in all planes: sagittal, coronal, axial, and length. The two fundamental styles of circular external fixation are the traditional Ilizarov system based on threaded rods and hinges, and the six-strut hexapod system first introduced as the Taylor Spatial Frame (Smith and Nephew, London, England).[1],[2] Both systems require manual adjustment of the threaded rod nuts or struts to achieve the gradual correction that is desired. Those adjustments are performed by the patient or the patient's caretaker, which, although quite often successful, impart some anxiety in the patient and/or caretaker related to performing the adjustments.[3]
A recent innovation is a computer-programmed, robotic automated motorized strut adjustment technology "Autostrut™" (Maxframe Autostrut™ Multi-Axial Correction System, Orthospin Ltd., Yoqneam, Israel) [Figure 1]. This system is fundamentally a standard hexapod external fixator with struts that are controlled by electronic motors, which are coordinated by a small cell phone-sized computer mounted to the patient's frame. The adjustment schedule, which is traditionally performed by the patient by manually adjusting the struts, is instead programmed into the computer and executed automatically at the instructed timing. Up to 20 adjustments per strut can occur daily, allowing very gradual length changes to occur. By replacing patients' manual adjustments with automated computer-controlled adjustments, it is expected that patient safety can be improved by reducing patient error. It is also expected that the smaller, more frequent adjustments should reduce patient pain associated with adjustments and also promote improved tissue histogenesis. Clinical use of the system has been described by one other surgical group to provide lengthening and alignment correction for ten children. They reported no adverse events.[5],[6]{Figure 1}
No prior literature describes the use of the Autostrut™ for correction of joint deformities or device-related issues and their management. This study was performed to address those knowledge gaps. Our full consecutive series of patients is described, representing the correction of bone and also joint deformities. Additionally described are device-specific issues and how they were managed.
Subjects and Methods
Following institutional ethics approval, a retrospective chart review was performed of the 16 patients who had an Autostrut™ applied. The review included demographics, indications for surgery, whether the intended correction was achieved, any general complications that occurred (such as superficial pin site infection), and any specific Autostrut™ issues that occurred.
Descriptive statistics were calculated for continuous data. No statistical comparisons were performed.
Surgical and technique notes
The Autostrut™ is an add-on accessory designed to be used with the Maxframe™ Multi-Axial Correction System (DePuy Synthes, West Chester, PA, USA). This system is fundamentally similar to a hexapod ring fixator system, and the surgical techniques are identical. The main relevant difference with using the Autostrut™ system is the application and use of the automated motorized struts. Rings are mounted to the patient as desired based on the intended deformity correction, the same as would be done if manually adjusted struts were to be used. We usually perform all necessary steps of the surgery (such as soft tissue releases) before performing the osteotomy which we perform as the last step; so right before performing the osteotomy, temporary manual struts are placed and length is recorded. The sterile struts are then manually set to these lengths (this can be done on a sterile back table or on the surgical field) and they would replace the temporary manual struts after the osteotomy is performed. Three strut sizes are available (short, medium, and long) ranging in length from 97 to 309 mm. As shown in [Figure 1], these struts have a built-on holster for the motor which will drive the struts based on the computer program created by the clinician. These struts are secured the same way as typical struts using shoulder bolts. The sterile portion of the surgery should be completed, including dressings, before placing the motors. Once sterile procedures are completed, the motors can be placed into the struts in a nonsterile manner. This can be done immediately after surgery or at any time before the intended start of the deformity correction schedule (such as in the recovery room, in the inpatient ward, or at follow-up office visit). The computer (white box) is permanently connected by wires to each of the six motors which are numbered for identification (in the same circular order as standard hexapod systems). The motors key into the holster and are secured with clips.
The correction schedule is created online using a browser-based interface, as is common for hexapod systems. Once the schedule is created, the program must be transferred from the online system to the patient's frame's Autostrut™ computer using a proprietary Universal Serial Bus cable; this can be performed by a product representative or a clinician using a laptop computer and the company's software. The system will perform several short checks to ensure hardware is working properly and the strut positions are as expected. Following this, no further interaction with the system is necessary unless there is a malfunction or if additional residual schedules are necessary.
Strut changes can be performed in a typical manner as any hexapod system would, without disrupting the automated routines and without interfacing with the computer. The process can be performed by the surgeon, other clinicians, or a product representative. The strut is a unique design with a motor housing built-on (the motor housing is not transferred from the prior strut to the new strut). First, the frame is stabilized with an additional strut, followed by removing the motor attachment clip and removing the motor from the strut, followed by removal of the strut and placement of the next sized strut which is manually set to the same length as the strut being replaced, followed by placing the motor back into the housing and securing with the attachment clip, followed by removal of the supplemental stabilization strut. The computer will perform the next adjustment at the scheduled time. When adjustments are completed, the computer and motors can be removed and the strut positions secured with clips in an office setting. This is a frame-only modification with no invasive component, so no analgesia is necessary. At this point, the frame becomes a standard hexapod frame (but static, as struts cannot be adjusted unless the computer and motors are replaced) and the frame can eventually be removed the same as any hexapod system, based on surgeon preference. At any time, if the surgeon desires to convert from the automated process to a manual process, this can be done by stopping the adjustments (using the STOP button on the computer), and replacing the automated struts with the system's manual struts. This can be performed in the office like a typical strut exchange, with care taken to ensure frame stability during the entire exchange.
Autostrut™ generation notes
The Autostrut™ is currently in its second generation. Generation 2 has replaced Generation 1 which is no longer supplied. The first seven patients in our series had Generation 1 and the rest had Generation 2 [identified in [Table 1]]. The major improvements in Generation 2 are as follows. Generation 2 has a single control unit instead of three, which reduces weight and bulk. The metal casing on Generation 1 struts was eliminated to improve radiographic visibility. Twenty daily adjustments can be made (instead of four). The entire system is Ingress Protocol 68 (IP68) certified, impervious to dust, and resists water when submerged up to 1.5 m up to 30 min.{Table 1}
Results
The patient summaries are presented in [Table 1]. The average age was 47.7 ± 18.5 years. There were eight males and eight females. The primary deformity was the tibia for 13 patients; the other three were one knee, one ankle, and one foot. The average weight was 80.7 ± 23.1 kg. The average duration of motors being used was 60 ± 29 days, the average duration of frame use was 169 ± 84 days, and the average duration of follow-up time was 9.2 ± 4.0 months. The final patient still has her frame on. The outcome of patient 2 is visualized in [Figure 2]. The outcome of patient 6 is visualized in [Figure 3] and [Figure 4].{Figure 2}{Figure 3}{Figure 4}
[Table 2] summarizes the deformities corrected, the residuals, and the strut change events. The most common deformity corrected was in the coronal plane, followed by length. Six patients had a joint involved in the deformity correction. There were ten residual programs and ten distinct times of strut changes. The computer was able to negotiate these residuals and strut changes without issue.{Table 2}
Adverse events will be presented as "general frame-related" and "Autostrut™ specific events." The general frame events will be presented first. Eleven patients had minor wire or pin site infection treated by a 10-day course of oral antibiotics. One patient (Patient 7) had persistent fifth metatarsal drainage from a wire site even after frame removal, which was successfully resolved with irrigation and debridement. Three patients (Patients 8, 13, and 16) had persistent poor-appearing regenerate, all who presented with tibial nonunion. Patient 8 had iliac crest autograft and bone morphogenetic protein-2 (BMP-2) applied to the regenerate at seven months, which facilitated frame removal 4 months later. Patient 13 had ipsilateral femur autograft and bone marrow aspirate concentrate (BMAC) applied to the regenerate site at 7 months, which facilitated frame removal 4 months later. Patient 16 had regenerate drilling with BMAC and BMP-2 applied to the regenerate site at 7 months (less than a month before this study was performed) and remains in her frame.
Three Autostrut™-specific unexpected events occurred. In Patient 2, the struts were applied in reverse orientation (inner threaded rod proximal and outer numbered tube distal) in order to improve visualization of the osteotomy behind the struts; this is not an issue in nonmotorized hexapod systems. This resulted in the device immediately alerting the patient that the strut position was not as expected and a halt to the program. The patient returned for strut reorientation and the program completed without issue. In Patient 5, the computer halted progress for an unspecified error; she returned to the office for a device interrogation. No error could be diagnosed, so the system was reset; the program then resumed and completed normally. The patient also subsequently missed a necessary strut change appointment due to contracting COVID-19 and requiring an extended quarantine; the computer identified that the end of the strut had been reached and stopped the program until the new strut was applied. Patient 10's computer halted adjustments and displayed alert light signals 2 months into her program. Upon inspection, one of the struts had become loose; this was tightened in the office and the program completed without further issue. In no situation was the patient's outcome compromised due to using an automated system.
Discussion
This study has two paramount points. First, the confirmation that the Autostrut™ system is suitable for correcting a variety of complex deformities, no differently in surgical technique and preoperative planning than standard hexapod systems. Second, that when device-specific issues occur, the patient safety is ensured by the system halting and alerting the patient to seek attention; all such issues were quickly resolved and resulted in achievement of the intended corrections. There were no compromises to patient care or outcomes associated with the new technology. In fact, the system potentially averted patient harm by virtue of its self-monitoring technology.
There have been two prior reports regarding Autostrut™. The fundamental Autostrut™ technology and its successful application in a 10-year-old patient with a multiplanar tibial deformity were highlighted by Geffner et al.[4] The other study, of ten patients under 21 years old, confirmed that the adjustments are performed correctly, within the limits of inter-rater reliability of reading strut numbers.[5] No unexpected events during program execution were reported in either article. Our study expands beyond those studies in two important regards: first, by exploring more diverse and complex deformity corrections and, second, by identifying unexpected system-specific events and their ramifications.
In our current series, the utility of the frame's use for procedures beyond lengthening was explored. The frame's ability to achieve correction just the same as would be expected by a hexapod frame was confirmed in difficult situations such as adjusting through a joint (six patients), and performing bone transport (two patients). Ten residual adjustment programs were programmed and executed without issue, and ten struts were changed without issue. The motorized struts allow the surgeon to plan and execute correction just as would be done with any hexapod frame using manual struts.
Three unexpected system-specific events occurred. All were rapidly addressed without adverse outcomes or patient inconvenience, other than an additional visit to the office. The occurrence of these unexpected events is important to understand, because indeed any treatment system can have device-specific issues, and how they are managed ultimately determines the safety of the device. The situation of reversed strut orientation for Patient 2 would no longer be a problem. The system software has been updated to allow the clinician to indicate that struts have been reversed, so this orientation is now supported. It must be noted that all of the struts must be in the same orientation (all normal, or all reversed); the system still will not accept mixed orientations. For Patient 5, the reason for the computer halting and indicating an error could not be exactly identified despite evaluation of the computer log history. However, no harm resulted to the patient and upon a single visit to the office produced expected performance for the remainder of the program. We can only speculate that perhaps there may have been some error in the code that was sent to the computer the first time. The situations of the computer automatically stopping adjustments when the end of strut was reached (also Patient 5) or when a strut became loose (Patient 10) in fact are safety features which were successfully invoked by the computer. These demonstrate that the computer is not merely "blindly" performing adjustments regardless of the situation but, in fact, is potentially providing value by performing safety checks. In the case of Patient 10, the safety check may have actually prevented erroneous positioning from being imparted had a manual system been in place and the patient not detected the strut becoming loose.
Several near-term modifications may further improve the utility of this automated system. One, the ability to mount struts in preferred orientation (each individually, not requiring all to be in normal or reversed orientation) may allow more flexibility if space becomes difficult. Next, remote monitoring and programming capability can help: A clinician could check on the position and provide a residual program remotely instead of requiring a technician to locally plug into the machine to adjust the program. In addition, the system could provide reminders to the patient to return to the office for necessary maintenance such as strut changes (not simply stop upon the end of a strut being reached).
Beyond the safety provided by the computer performing system checks, several potential benefits may be achievable because of the automated system. Dynamization (physiologic axial compression and associated relaxation) performed late in distraction, or reverse dynamization (dynamization performed at the beginning of callus formation)[7] can improve bone healing for fractures and improve the bone healing index for lengthening.[8] Dynamization with manual systems would not be practically achieved if patients had to constantly adjust their struts a dozen or more times daily. Autostrut™ can currently perform up to 20 adjustments per day, which can achieve dynamization without any patient effort. The dynamization technique has not yet been investigated for Autostrut™. In addition, smaller and more frequent adjustments are better tolerated by the regenerate than larger and less frequent adjustments,[9] which would be expected to reduce patient pain and improve bone regenerate condition. Again, because the motors can make up to 20 adjustments per day, much more minute and gradual adjustments can be performed than with typical manual systems. This study was not designed or adequately powered to determine these potential benefits. Finally, while the issue of erroneous adjustment has not been previously directly investigated, automated adjustment should eliminate patient-related errors regarding strut adjustment. The reliability of the automated system has so far proven essentially perfect, provided the computer does not identify a problem.
The limitations of this study center on the relatively small sample size of 16 patients. Certainly, a greater volume of experience can allow the recognition of more device-specific issues. Further, all deformities were for the lower extremities; there may be unexpected user experience issues related to upper extremity use. However, our study is the largest in the brief history of the technology. The variety of deformity correction is also quite diverse, and includes all geometric axes. Further, several device-specific unexpected events occurred and were described. We believe there is substantial merit to the technology and feel confident in the safety of this system to use in many future cases where hexapod frame correction is indicated.
Conclusions
The Autostrut™ system appears reliable and safe. It executes programmed index and residual programs as well as strut change scenarios as directed. The system recognizes unexpected mechanical or programming issues and ensures patient safety by halting progress and alerting the patient. Future versions of the technology may benefit from added features such as remote reprogramming or current strut position monitoring. Further study of this technology appears merited, particularly regarding potential benefits such as dynamization and more frequent more gradual adjustments.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
| 1 | Manner HM, Huebl M, Radler C, Ganger R, Petje G, Grill F. Accuracy of complex lower-limb deformity correction with external fixation: A comparison of the Taylor Spatial Frame with the Ilizarov ring fixator. J Child Orthop 2007;1:55-61. |
| 2 | Fragomen AT, Rozbruch SR. The mechanics of external fixation. HSS J 2007;3:13-29. |
| 3 | Yilihamu Y, Keremu A, Abulaiti A, Maimaiti X, Ren P, Yusufu A. Outcomes of post-traumatic tibial osteomyelitis treated with an Orthofix LRS versus an Ilizarov external fixator. Injury 2017;48:1636-43. |
| 4 | Geffner AD, Reif TJ, Fragomen AT, Rozbruch SR. Use of orthospin technology in the correction of complex limb deformities. Ann Robot Automation 2021;5:037-40. |
| 5 | Gigi R, Mor J, Lidor I, Ovadia D, Segev E. Auto Strut: A novel smart robotic system for external fixation device for bone deformity correction, a preliminary experience. J Child Orthop 2021;15:130-6. |
| 6 | Eitan S, Yakov M, Lidor I, Dror O, Roy G. auto strut: A novel smart robotic system for external fixation device for bone deformity correction, a preliminary experience. J Child Orthop 2021;15:130-6. |
| 7 | Glatt V, Samchukov M, Cherkashin A, Iobst C. Reverse dynamization accelerates bone-healing in a large-animal osteotomy model. J Bone Joint Surg Am 2021;103:257-63. |
| 8 | Ferreira N, Tanwar Y, Burger M. The effect of fixation dynamization on fracture healing: A systematic review. J Limb Lengthening Reconstr 2020;6:7. |
| 9 | Hasler CC, Krieg AH. Current concepts of leg lengthening. J Child Orthop 2012;6:89-104. |
|
