Abstract
Amputations are an operation that can strike fear into the heart of a trauma and orthopaedic surgeon with a sense of doom and gloom on occasion for both the surgeon and the patient. At one extreme there may be the emergency life-saving setting, and on the other the more controlled elective setting often following multiple attempted limb-salvage procedures. Either way, there can be a feeling of failure at the end of the treatment road. This need not be the case. Patients coming out of a protracted treatment pathway sometimes wonder why they did not have the limb removed sooner and are usually grateful for the sense of freedom that the operation has given them. This review aims to provide a brief historical perspective through to an update on current techniques with a dedicated discussion around osseointegration. There are suggested techniques, that will provide at the very least a refresher or indeed an alternative thought process to approach both the operation and the patient. Amputations are often the beginning of a new life, but not necessarily the end of the old one.
Introduction
‘Amputation’ is derived from the latin ‘amputare’ meaning to prune or cut back; however, the art of surgical amputation is believed to date back more than 31,000 years – following the relatively recent discovery of skeletal remains in Liang Tebocave, East Kalimantan, Indonesian Borneo. It is reported that the individual is likely to have survived a surgical amputation of the lower limb, during childhood, and is believed to have lived for a further 6–9 years. Estimates had previously dated the first surgical amputation around 7000 years ago following the discovery of the skeletal remains of a European Neolithic farmer in France.
As quoted by Apley and Solomon, the indications for amputation can broadly be divided into the 3Ds – “Dead, dangerous or damn nuisance”. These three terms represent a “dead” – non-viable limb through disease or ischaemia; a limb that is “dangerous” to the patient – from infection, crush injury, burns or tumour; or considered a “damn nuisance” – non-functional secondary to congenital deformity or a result of significant trauma.
Within the orthopaedic world, the most commonly encountered amputations are those in the lower limb, frequently as a result of ischaemia and diabetes, closely followed by trauma. Less frequently, the orthopaedic surgeon will be required to perform amputation for some congenital deformities or tumours – both primary bone or soft tissue malignancies and occasionally in the setting of metastatic disease. Whilst the rising incidence of peripheral vascular disease is resulting in more amputations for the vascular surgeon, there is a similar trend within the orthopaedic community following the rise in diabetes with an increasing number of patients requiring multiple surgeries often ending in amputation.
There is an ongoing debate surrounding amputation versus limb salvage within the setting of trauma, tumour and congenital deformity cases. Whilst advancements in trauma management, microsurgical skills and fracture fixation techniques have allowed surgeons to push the boundaries for limb salvage, it has not come without its complications. Research has suggested that salvaging a limb that is insensate with compromised functional outcome is often not in the best interest of the patient as these patients will frequently endure repeated surgeries in an attempt to keep their limb; only after many procedures to end up with an amputation. Quite often, these patients have had a lengthy journey with multiple hospital admissions, have failed to rehabilitate and many will have developed chronic pain leading to less favourable outcomes after amputation. It is believed that trauma patients with significant limb-threatening injuries who go straight to amputation have a quicker recovery, develop fewer chronic pain issues and generally rehabilitate faster.
Similarly, there have been many developments in the world of tumour surgery with advancements in endoprosthetics and biological reconstructions meaning that patients presenting with more complex tumours can be considered for limb salvage options whereas 30–40 years ago, the only surgical option would have been amputation. However, again, whilst every patient’s wish is to try to salvage a limb whenever possible, this should never be at the expense of the oncological resection. Equally, the surgeon must also consider the anticipated function of the limb after reconstruction and any potential adjuvant treatments that may also influence both the success of the procedure and the final functional outcome for the patient. If a patient is facing resection of a joint, with significant soft tissue excision and endoprosthetic reconstruction and requires post-operative radiotherapy – not only is the functional outcome of the limb compromised, it is also at high risk of infection and failure. Thus, when a tumour is either encasing the neurovascular structures or the resection and reconstruction is significant enough to compromise function, then the patient should at the very least be considered for a discussion around amputation.
Congenital deformity patients can also have a similar albeit more controlled journey to the multiply operated trauma patient. Ultimately, many of these children will often spend their childhood years in and out of hospital, investing time and energy into their limb with lengthening procedures, or corrective osteotomies. Whilst these may improve the appearance of the limb and address a limb length discrepancy, it does not always follow that these patients will have good function. Some will be left with pain, stiffness and contractures whilst others will complain of instability and after many years of surgeries, some will choose to have the limb amputated, often reporting that their function with the aid of a good prosthetic is much improved.
What is important in all of these patient groups and scenarios is that amputation should not be considered the end of the road for them but rather the beginning of a new journey and an opportunity to regain independence. Some of the most significant changes we have seen in the world of amputation in recent years surround the psychosocial aspect and the concept of prehabilitation. Over recent decades, society has become far more accepting of people who have suffered limb loss and there are many charities and support groups that are accessible to patients of all ages and backgrounds. There has been a significant shift in acceptance within society and a drive to support amputees to empower individuals to live their lives to the fullest. Following on from various conflicts across the world, our military personnel have defied the odds and through the power of support and initiatives such as the Invictus games, they have demonstrated the impossible, taking on physical and mental challenges that were perceived as being beyond their reach. Their determination and physical fitness together with advancements in prosthetics has provided new lifelines not only within the military setting but more broadly across the wider public. It is the change in perception and the drive to optimize function that has led to the development of prehabilitation. Certainly within the elective or semi-elective setting, the preoperative engagement with an amputation therapy team should not be underestimated. Patients who are able to optimize their fitness, arrange the logistics of their day to day lives and prepare psychologically for an amputation, report significant benefits compared with those who either cannot or do not wish to engage pre-operatively. Patients often report feeling better prepared for their journey, both from a physical and mental perspective and will often value the opportunity to share this with their families prior to the surgery.
Technical aspects
Preparation and good surgical technique are fundamentally important in achieving the best outcomes for patients. When preparing a patient for an ablative procedure, it is imperative that all avenues have been explored with the patient and that they feel supported in their decision to proceed with amputation. Unlike many other surgeries, an amputation is an irreversible procedure which has lifelong consequences and thus adequate preparation is essential.
The indication for amputation must be clear and the patient should be optimized prior to surgery. Appropriate imaging should be performed and reviewed to allow planning of transection level and soft tissue coverage. Whenever possible, the surgeon should aim to preserve as much length as possible and should also consider disarticulations where appropriate. Whilst the exact level of amputation is often dictated by the disease/pathology, when planning the transection, it is worth bearing in mind that there are optimal distances to aid with prosthetic fitting. For example, in the lower limb, when performing a below-knee amputation – it is preferrable to maintain the proximal 15 cm of tibia if possible, which will result in an adequate stump length to accommodate a prosthesis. Despite a relatively long lever arm, a below-knee amputation typically incurs a 25% increase in energy expenditure compared with the normal gait of a person with two limbs. Similarly, when planning an above-knee amputation, it is preferrable to preserve as much length as possible and a transection point approximately 12 cm above the joint line again results in an adequate stump for a prosthesis. However, the energy expenditure in an above-knee amputation is increased by 65% and in a bilateral above-knee amputation setting the figure is 200%.
When consenting a patient for an amputation, one should discuss risks of pain, infection, bleeding, scarring, anaesthetic risks including deep vein thrombosis (DVT) and pulmonary embolism (PE), neurovascular injury and wound complications. In addition to the general surgical risks, patients should be informed of the risks of phantom pain, phantom sensation, flap failure/demarcation, neuroma formation, bony spurs, skin and soft tissue complications including skin irritation with the prosthetic interface, bursa formation and atrophy leading to bony prominence and problems with prosthetic fitting.
Below-knee amputation
In the below-knee amputation there are two techniques that have been popularized in terms of soft tissue flap coverage. There is the long posterior flap vs the fish-mouth flap. As with the transection level, the choice of flap will often be dictated by the pathology/disease leading to the amputation and generally speaking, the long posterior flap has a more reliable blood supply and thus is more commonly used, particularly in the ischaemic patient.
Flaps should be developed, ideally in continuity with the underlying fascia to provide a strong envelope for closure. Once through the fascia, the muscle is divided and the neurovascular bundles identified and ligated. A transfixation suture is often used for the artery and double ties for the vein. The nerves are sharply transected with a knife under slight tension to allow retraction.
The transection point on the tibia is identified and a saw is used to complete the cut at the appropriate level whilst protecting the surrounding soft tissues. The anterior aspect of the tibia is cut at an angle and bevelled to a smooth edge and the fibula is transected with a saw 1.5–2 cm proximal to the tibial cut on a slight oblique angle so as not to leave a prominent edge. The posterior fascia is repaired to the periosteum overlying the anterior tibia to allow full coverage of the cut end of the tibia. The superficial fascia is then repaired to approximate the flaps and the subcutaneous and dermal layers closed sequentially. If a tourniquet has been used, this should be deflated before closure to ensure adequate haemostasis. A layered compressive bandage or negative pressure wound therapy dressing is then typically applied.
Above-knee amputation
The most commonly used technique for above-knee amputation utilises the fish mouth incision or the skew flap whereby the fish mouth is rotated off the midline of the femur possibly to incorporate a previous scar or to navigate a soft tissue lesion whilst trying to maintain length in the transected bone.
One can measure and mark out the flap by taking a piece of surgical tape and marking the circumference of the leg at the point of transection. This length can then be halved and held over the anterior thigh at the level of transection to mark the apices of the two flaps. The tape can then be halved again (quarter of original length) and held vertically from the transection point along the midline anterior and midline posterior points to mark the distal extent of the skin flaps. Two semi-circles can then be drawn from the distal anterior marking to the apices and from the distal posterior marking to the apices. The same principle can be used for the skew flap taking into consideration the degree of rotation required.
As for the below-knee amputation, the dissection should be continued in one plane to the fascia and ideally should be kept in continuity with the subcutaneous layer. Dissection is continued through the muscle compartments taking care to identify the vascular bundle medially – whereby a transfixation suture and double ties can be used to ligate the vessels. The sciatic nerve should again be identified and sharply transected under slight tension to allow retraction. It is worth noting that the sciatic nerve has an associated artery and this should be ligated separately to avoid bleeding following transection of the nerve. A saw is used to transect the femur at the appropriate planned level and the edges are bevelled to a smooth finish. A 2 mm drill should be used to make perforations in the distal aspect of the cut end of the femur to allow myodesis of the adductor muscle. This will prevent the residual limb from drifting into an abducted position and will allow better positioning within the prosthetic. The deep fascia of the anterior and posterior compartment muscles can then be sutured together in the form of a myoplasty with a small amount of tension. The superficial fascia can then be closed over a drain and the subcutaneous tissues and dermis closed as sequential layers. It is worth noting that in the ischaemic patient every effort should be made to avoid any undue tension in the closure so as not to compromise the vascularity of the flap and impede healing. If a tourniquet has been used, this should be deflated before closure to ensure adequate haemostasis. A layered compressive bandage is applied (not in the ischaemic limb).
Through-knee amputation
While not commonly performed, a through-knee amputation or disarticulation is preferred when possible in the paediatric cohort as it avoids the issue of bony spurs and provides an end bearing stump for crawling. However, it poses significant challenges for prosthetic fitting and is considered less favourable cosmetically. This is due to the entire length of the femur being preserved, thus when in a sitting position wearing a prosthetic – the prosthetic knee is proud giving the illusion that the prosthetic leg is longer than the contralateral leg which can cause some distress to patients. The technical aspects of the procedure are similar to the above-knee amputation with a fish mouth incision located distal to the knee joint. The tibia is disarticulated and the patella tendon advanced to attach to the anterior cruciate ligament. Debate persists around preserving the patella due to concerns that this can then become a point of pain where the patella articulates with the femur. The same technique can be used, preserving the patella tendon but excising or enucleating the patella itself thereby mitigating this issue. Alternative techniques include excision of the femoral condyles thus removing the distal femoral flare proposedly improving the fit of a prosthesis. A further technique that has been used is the Gritti–Stokes technique whereby the distal femur is excised and the patella attached to the cut surface as an ‘endcap’ to form a smooth contour and aim to prevent bony spurs. Closure is as per above-knee amputation.
Foot amputations
There are a number of amputation levels that can be considered in the foot. However, these often pose more of a challenge in terms of prosthetics.
Lesser toe amputation: dependent on the level of amputation – if a terminalization is to be performed, then a long plantar flap can be used and if the amputation is more proximal towards the metatarsophalangeal joint then a racquet incision is typically used. Full thickness flaps are created and the flexor and extensor tendons divided. The neurovascular bundles are identified and ligated – however, care should be taken to ensure that in the setting of a long plantar flap the soft tissue flap remains vascularized. The bone is transected or disarticulated and the soft tissues closed. When performing a second toe amputation, where possible, it is preferrable to leave as much of the proximal phalanx as possible in an attempt to minimize valgus drift of the hallux.
Ray amputations are classified either as border ray amputations (first or fifth ray) or central ray amputations. A racquet incision can be used for both border or central ray amputation which is extended to the level of the tarsometatarsal joint. It is considered technically easier to transect the metatarsal rather than disarticulate the joint and the cut should be made obliquely to aid soft tissue coverage. During the approach, full thickness flaps should be created to avoid devascularization and subperiosteal dissection continued. When disarticulating the first metatarsophalangeal joint care should be taken to preserve the penetrating branch of the dorsalis pedis artery.
Transmetatarsal amputations: while not commonly performed, this technique again utilizes a long plantar flap and the dorsal incision is sited at the level of the intended transection. The medial aspect of the flap is kept long due to increased thickness of the foot and to aid with closure. Extensor and flexor tendons are divided under tension to allow retraction and the neurovascular bundles identified and ligated. The metatarsals are then transected at the pre-determined level and edges smoothed. If a tourniquet is used, this should be deflated before closure to ensure heamostasis and a single layer closure is achieved with non-absorbable interrupted sutures.
Midfoot and hindfoot amputations: whilst similar in principle to the transmetatarsal amputation, the midfoot amputation involves a more proximal disarticulation – either at the level of the tarsometatarsal joints – Lisfranc amputation or at the midtarsal joint – Chopart amputation. Neither amputation is commonly used and both tend to drift into equinovarus with time.
The hindfoot amputation is often referred to as a Syme’s amputation. It is considered a favourable option as it allows an end-bearing stump on a short leg which allows a patient to mobilize short distances without a prosthetic. However, it poses significant challenges for prosthetic fitting and requires an expert prosthetist and is considered inferior in terms of cosmesis. The skin incision is made by creating a transverse limb from the tip of the lateral malleolus across the ankle joint ending approximately 2 cm below the medial malleolus and a vertical limb under the heel which meets the medial and lateral aspects of the transverse limb. Dissection is performed as a full thickness flap and continued down to bone taking care to preserve the posterior tibial artery. The foot is put into an equinus position to allow excision of the talus and calcaneus. The distal tibia is transected approximately 0.5 cm above the joint line to create a surface parallel with the ground and the edges smoothed. Remaining tendons are divided and the neurovascular bundles ligated, taking care to ensure that the posterior skin flap remains vascularized. Drill holes are created in the anterior tibia and the heel pad is mobilized and secured over the cut surface of the tibia by suturing through the drill holes. Care should be taken to ensure that the flap is not under excess tension to avoid devascularization. Closure is with interrupted non-absorbable sutures.
Hip disarticulation/hindquarter amputation
A hip disarticulation involves removal of the entire limb through the hip joint leaving the pelvis intact, whereas a hindquarter amputation again removes the entire limb but with resection of part of the pelvis. These procedures have a significant impact on patient lifestyle and all patients being considered for surgery should be seen by the pre-amputation therapy team. Very few patients progress to a prosthesis and the majority will be confined to a wheelchair and some will require high levels of care, occasionally preventing them from returning to their home. Sitting balance is typically preserved after a hip disarticulation but lost following a hindquarter amputation. Most indications for such extensive surgeries include large pelvic tumours encasing neurovascular bundles or significant trauma whereby limb salvage is not feasible. These procedures can be performed with either curative or palliative intent however, they are highly morbid with high complication rates and generally poor functional outcome.
Postoperative instructions
Patients should be moved to the appropriate recovery level for their needs and pain management should be initiated immediately. Debate persists around the effectiveness of postoperative ketamine infusion for the first 48 hours with little convincing evidence to prove its superiority over more traditional opioid based analgesics. Patients should be encouraged to start bed exercises from day 1. Drains, when used are typically removed at 48–72 hours and the initial compression stockings can usually be fitted from day 5 onwards. Where applicable, patients can be fitted with sockets once the wound has healed sufficiently.
Complications in amputation surgery
Wound/skin issues
Depending on the indication for amputation, there are certain scenarios that increase the likelihood of wound breakdown or complications. In the arteriopath, or trauma patient there may be concern regarding the viability of the flap and risk of flap necrosis. Whilst in the planning and during surgery every effort is taken to avoid this, when there is sufficient concern, some advocate deliberately leaving the flap long to allow it to demarcate. Patients can also develop skin irritation at the stump-prosthesis interface which can usually be resolved by making modifications to the prosthetic.
Bony spurs
The development of bony spurs at the distal end of the remaining bone is more common in children than in adults – hence the recommendation where possible to consider a disarticulation rather than transosseous amputation in children. Many attempts have been tried over the years to overcome this issue by essentially filing the intramedullary space with graft – for example when performing a below-knee amputation – a bone bridge can be used from the fibula to the tibia (Ertl technique) or less commonly, if appropriate a toe can be salvaged from the amputated foot and can be transplanted into the distal tibia as an osteochondral graft creating an ‘endcap’ (similar to the Gritti–Stokes technique with the patella in the through-knee amputation).
Phantom pain/sensation and neuromas
It has been estimated that up to 80% of patients experience phantom sensations in the first 6 months following amputation. It is important to realize that not all of these patients will be describing a painful stimulus, hence the expression of phantom limb sensation rather than pain. Some will complain of experiencing temperature changes in the amputated limb or will describe wanting to reposition the limb in space. They can experience heaviness in the limb and some will describe telescoping of the limb whereby they feel that the limb is getting shorter as time goes by.
However, there is a proportion of patients who do experience phantom limb pain (PLP) which they tend to characterize as either a sharp or burning painful stimulus. It tends to be that the more distal the pain experience in the missing limb, the more exaggerated the pain and it is believed that this may be due to a larger area of representation of the foot/hand in the cortical homunculus.
What is fundamentally important is that these patients should be managed within a specialized multidisciplinary amputation team made up of physiotherapists, occupational therapists, psychologists, specialist pain management consultants, surgeons and prosthetists. Patients will often require combination therapy to try to manage PLP with a multimodal approach encompassing drug therapy, mirror therapy, virtual reality and cognitive behavioural therapy.
Neuroma formation following amputation is common however it is not always symptomatic and thus not all neuromas require intervention. In a symptomatic patient non-operative management strategies should be attempted in the first instance such as modification of the prosthetic so as not to cause unnecessary pressure over the affected nerve end. Ultrasound-guided injections can be used as both diagnostic and therapeutic interventions and one must always consider that a patient may have a concurrent problem that is causing pain – such as osteoarthritis in the adjacent joint – hence a thorough clinical examination is essential.
Traditionally, the surgical technique of handling the nerve during amputation is to sharply transect the nerve under slight tension to allow retraction of the nerve into the soft tissues. However, more recently, surgeons have been looking at alternative techniques to reimplant the nerve in a cuff of muscle (regenerative peripheral nerve interface, RPNI) or particularly if a myoelectrical prosthesis is to be considered to use targeted muscle reinnervation (TMR). In the most simplistic terms, these techniques essentially rehome the nerve providing an anatomical end point to the nerve or allowing it to adopt a new function rather than allowing it to be freely suspended within the soft tissues trying to regrow and find its transected counterpart. Results have thus far been variable, and the technique is continuing to evolve as we learn more about the neuromodulatory response to amputation.
Osseointegration
Background
An extremity amputation leads to major changes in a person’s function, body image, physiology and the overall quality of their life. It is estimated that less than 50% of amputees return to work and that the average time to return to work exceeds a year. Over 90% of bilateral above-knee amputees eventually end up confined to a wheelchair due to the difficulty of mobilizing with a dual socket.
The current standard of care is a traditional socket mounted prosthesis, technology which is now in the region of 600 years old. Sockets have evolved over time, but not much. The problems with a socket prosthesis that patients regularly face are ‘socket–skin interface’ related such as: friction, allergy due to reaction with the liner, blisters, infections, ulceration, chafing, ingrowing hair and other skin irritation issues. In addition, stump size can fluctuate, leading to issues with mobility and fit. Thirdly, there is ultimately a loss of proprioception, the ability of the person to feel the ground and walk over uneven terrain.
These socket-related problems are amplified in patients that do not have a well-padded stump, which can be caused by the extent of the original trauma such as bomb blasts from war, the location of a cancer or on occasion poor surgical technique. This can lead to there being a short residuum (stump), skin graft at the site, skin flaps that are over hanging, multiple scars, heterotopic bone ossification, atrophic fragile skin, inadequate soft tissue and hypersensitive neuromas.
OI is the implantation of a transcutaneous bone anchored metal nail which attaches to a prosthetic limb. This eliminates skin compression and prosthesis fit issues, providing improvement in mobility and quality of life versus a traditional socket prosthesis. Ultimately the socket is eliminated altogether.
The modern OI multidisciplinary team includes a physiotherapist, prosthetist, psychologist and a surgeon. Assessments by the individual specialists and a subsequent collaborative approach are needed for prehabilitation and to optimize patient outcomes.
Historical overview
The first documented attempts at a transcutaneous prosthesis were in animals. Vitallium screws were inserted into the intramedullary canal of 18 dogs. This was reported by Dr Elliot Cutler and Dr James Blodgett as early as 1942 at Harvard University. Transcutaneous prosthetic attempts in humans were first recorded in 1967. Dr Vert Mooney working at the rehabilitation centre in Rancho Los Amigos tried using a porous ceramic in a patient’s humerus.
Modern OI started in 1977, by Dr Per-Ingvar Branemark in Sweden. He described the phenomenon of bone growing directly onto and attaching to the surface of a material without an intermediate layer such as fibrous tissue. There are several different systems that exist, but the most common are the OPRA (Osseointegration Prosthesis for the Rehabilitation of Amputees), the ILP (Integral Leg Prosthesis), and the OPL (Osseointegrated Prosthetic Limb).
The ideal OI implant should provide the ability to rapidly fully weight bear, be mechanically durable and should resist infection, which the soft tissue implant interface aids with. The OPL offers this and is the system used by the author. Along with a wide range of sizes, it is performed in a single stage. The system was developed by Professor Munjed Al Muderis in Australia and is the most implanted OI system in the world. An X-ray demonstrating the implant in situ is shown ( Figure 1 ).
