In this scope review we identified research and literature on DTI as a result of transtibial prosthetic use. We addressed three key areas: Firstly, the aetiology of DTI in prosthetic users is only partially understood. Whilst direct deformation seems to be an important factor, the contribution of ischemia, ischemia reperfusion, and an impaired lymphatic drainage has yet to be evaluated. Secondly, we identified several risk factors. Intrinsic determinants of DTI damage are tissue morphology and its mechanical properties, as well as sensory impairment of the residuum. Extrinsic determinants are socket design and prosthetic components. Finally, we found that methodological approaches in both aetiological and risk factor research focussed mainly on biomechanics. The most common research design combined static or dynamic loading with medical imaging and computational analysis. However, the diversity of input data, modelling assumptions, and outcome measures, together with the variability in geometrical patient characteristics and prosthetic components make a quantitative comparison of results and their clinical translation difficult.
Damage pathways
Within the last two decades, major research efforts led to advancements in our understanding of DTI. The pathways of tissue damage from external loading were investigated on different organisational levels, from cell cultures over engineered muscle constructs to in vivo animal studies [7, 46]. The results indicate that the relationship between the degree of deformation and time is essential: Whilst high strain can cause immediate direct deformation damage, moderate strain may injure the tissue by occluding the vascular and lymphatic system over a longer period.
Loading biomechanics and direct deformation
The biomechanical profile at the stump-socket interface of lower-limb amputees differs significantly from the frequently studied immobile individuals [15, 16, 18, 47–51]. Prolonged loading is only seen in situations of static weight bearing like standing or sitting with a donned prosthesis. In contrast, with over 8000 steps taken daily by transtibial amputees [52], the residuum often experiences dynamic, cyclic loading. During gait, complex internal stresses and strains develop: weight bearing causes pressure at the socket-limb interface [53]; the vertical pistoning movement between the residuum and the socket invokes shear forces [54]; and gait action and bone movement lead to the development of additional forces and moments [55, 56]. The combination of these forces and moments results in internal compressive, tensile, and shear stress [57]. However, the individual contribution of each stress component to the development of DTI is difficult to quantify.
So far, DTI research focussed mainly on compressive loading. Researchers have defined threshold levels for direct deformation damage, based on the relationship between pressure or compressive strain, and time [42, 43]. Subsequently, they compared these thresholds to computed values of internal compressive stress and strain in prosthetic patients to predict the likelihood of DTI development [30, 31, 33]. However, the results of these studies must be interpreted with caution.
Firstly, the threshold levels were deducted from experiments on animal tissue and bioartificial muscle constructs, and not intended to be translated directly to human studies [42, 43]. Secondly, the experimental design of threshold studies was oriented towards measuring direct deformation damage, whilst the microvascular and lymphatic system were mostly neglected. And lastly, the application of a static, compressive load differs from residuum biomechanics. Researchers showed that a cyclic force application may reduce the static threshold levels [40, 51, 58, 59], implying that the predicted risk of DTI in transtibial amputees has been underestimated. The addition of a shear component might also induce a greater damage potential by reducing the capillary closure pressure, which in turn increases the risk for ischemia [60, 61].
To minimise these limitations, new research approaches are needed. The effects of shear and distortion on tissues and cells should be examined with mechanical, microvascular, and lymphatic damage in mind. Additionally, researchers repeatedly stressed the need for a dynamic threshold level that combines compression, tension, and shear [29, 30, 62]. For this reason, our research group is working on investigating tissue response to a cyclic loading protocol under a variety of loading conditions that represent transtibial prosthetic use. This will allow for a better interpretation of existing data on internal loading conditions and contribute to the knowledge of DTI development.
Microvasculature and the lymphatic system
Our review revealed a gap in the aetiological research. Whilst the connection between deformation damage and prosthetic use has been established experimentally, the role of microvasculature and the lymphatic system was only discussed in general review papers.
Regarding the vascular system, the effects of a cyclic loading protocol with ischemic and reperfusion phases altering at high frequencies are unknown. The literature on similar loading scenarios is ambiguous: Some experimental results indicate an accumulation of cellular and tissue damage in comparison to continuous loading. However, the studies were either based on animal skin [58, 59, 63] or used prolonged time intervals [50, 51]; these were far below frequencies of about 1 Hz commonly seen during prosthetic gait [31, 64]. Other studies report on the adaptive capabilities of soft tissue [65–68], which indicates an increased tissue tolerance towards ischemic and reperfusion damage [69].
The structure and function of the lymphatic system and its response to loading are barely understood. In general, compression in combination with shear stress may affect lymph formation and propulsion [70]. The resulting dysfunction of lymphatic drainage can lead to an accumulation of toxic waste products, which damages cells and tissues [71, 72]. In the last decade, this theory was supported by a small number of studies. Researchers confirmed a reduction of lymphatic clearance during compressive loading [73], as well as toxic effects of accumulated bio-wastes on cell viability [74]. Reperfusion seems to add additional stress on the lymphatic system [75]. These factors might be aggravated in transtibial amputees. The initial amputation process causes traumatic injury to the lymphatic system [39, 76], followed by a reduced muscle tone post-amputation, which might reduce the propulsive capabilities of lymphatic vessels and results in oedema [77].
For a better understanding of the role of the vascular and lymphatic system in DTI development, the simulation of high-frequency cyclic loading protocols with perfused tissue will be necessary. Beside histological analysis of muscle damage in animal models, arterial spin labelling or dynamic contrast MRI could relate the damage to muscle perfusion and oxygenation [78]. For clinically oriented studies, near-infrared (NIR) spectroscopy is potentially capable of evaluating the oxidative function of transtibial residua [79, 80]. The lymphatic system could be explored in a similar fashion by visualising lymph flow with NIR fluorescent imaging [73]. Together with information from direct deformation studies, the results of microvascular and lymphatic experiments can feed into the prevention, diagnosis, and treatment of DTI.
Population-specific risk factors
Identifying population-specific characteristics that may have a negative influence on the soft tissue in amputees is equally as important as estimating internal loading states. We found a variety of risk factors that link transtibial prosthetic use to DTI development, as detailed in the following paragraphs.
Influence of tissue characteristics
Our review identified a range of tissue-related determinants of DTI risk. Firstly, tissue morphology and geometry change as a result of the surgical process. By revisiting established amputation procedures and considering alternatives, for example the use of bone bridges [81] instead of bevelled bone edges, or different flap techniques that change the location of surgical scars [82, 83], some of these risk factors might be avoidable.
On the other hand, naturally occurring risk factors are hard to eliminate. The residuum commonly undergoes changes of intramuscular fat infiltration and muscular atrophy post-amputation [66, 84], which were both shown to enhance internal tissue deformation in SCI patients [13]. Additionally, amputees often experience complications after surgery like osteophytes or angular deformities [24, 25], which also have adverse effects on tissue loading [36, 39]. Overall, a multitude of amputation-related tissue characteristics aggravates the challenging loading conditions in transtibial amputees.
The role of co-morbidities
The main cause of transtibial amputation in the high-income countries is disvascularity. In the UK and the US, most patients suffer from diabetes and other peripheral vascular diseases [85–87]. This affects the perfusion of their residua, which facilitates ischemic damage and exacerbates inflammation. In diabetic patients, neuropathy is another common complication. The review papers we identified stressed that as the body’s protective pain mechanism is suspended, affected individuals may not distribute potentially damaging loads anymore [26, 40].
To validate this link between co-morbidities and DTI development in transtibial amputees, epidemiological studies would be beneficial. These studies also offer a chance to uncover possible correlations of other factors, such as time since amputation or age. Additionally, they could give an insight into the extent of the prevalence of DTI in transtibial amputees in general. The epidemiology of DTI in lower-income countries should also be investigated, as their amputation population profiles, surgical procedures, and prosthetic componentry differ significantly from high-income countries [87–89].
Prosthesis-related risk factors
Prosthetic sockets are in direct contact with the residuum. Hence, they are the component most frequently associated with DTI development. Researchers highlighted the influence of the socket shape on internal pressure distribution and soft tissue deformation [9, 40, 41]. However, studies that quantify differences in internal loading conditions between the two main socket designs PTB and TSB are missing. To close the gap, biomechanical and physiological studies could be conducted. This information would help to define good prosthetic fit and improve clinical practice.
Other prosthetic componentry like prosthetic feet and liners have also shown to influence the internal loading conditions [32, 38]. Before this information can be utilised reliably in clinical practice, more studies with large sample sizes and low variability are needed. This applies not only to studies on prosthesis-related risk factors, but to all experimental DTI studies on human subjects. Thus, consistency in patient groups, socket types, alignment, and componentry are crucial.
Methodological Approaches
Tissue loading is undoubtedly a major factor in the development of DTI. Traditionally, the interface pressure between the socket and the residuum was quantified with sensor technologies [40, 90]. With the discovery of deep tissue layers as the potential origin of pressure ulcers [4, 11], the focus shifted towards assessing internal mechanics. New methodologies were needed as the forces applied at the interface do not reflect the internal loading states [91, 92].
Computational and analytical modelling
According to our review, modelling was the measure of choice. Medical images, interface pressure measurements, and data on mechanical properties of soft tissue informed numerical and analytical models. Researches then used the estimated stress and strain values of those models to determine general loading patterns during various scenarios and patient-specific DTI risks.
However, the interpretation of estimated stress and strain values beyond basic trends is limited. This comes down to the complexity of soft tissue mechanics. Skeletal muscle shows time-dependent behaviour under load, including hysteresis, stress relaxation, and creep [93, 94]. Due to its cellular arrangement in parallel fibrous structures, the direction of load application alters the mechanical properties, known as anisotropy [93]. Additionally, the residuum consists not only of skeletal muscle, but also adipose tissue, connective tissue, and skin, which all have different, patient-specific mechanical properties. Each layer might also be affected by inhomogeneities, like scar tissue or fat infiltration. Taken together, the intricate nature of soft tissues exceeds the modelling capabilities of existing computational approaches [41]. The research on biomechanics of the residual limb is no exception and usually draws on simplified concepts with assumptions of linear elasticity, isotropy, or homogeneity. Even though clinical translatability is a core issue in DTI research, the accuracy of the proposed patient-specific modelling approaches based on these simplifications is questionable.
Nevertheless, promising methodological developments may increase the accuracy of future research efforts (Table 8). An additional advancement is the integration of model post-processing. Muscle stiffening is a phenomenon observed after the onset of DTI, which redistributes subsequent loads towards adjacent sites and promotes damage progression in a positive feedback loop [44, 95]. Integrating this damage law into FE models for prolonged loading increases the accuracy of DTI prediction [44, 62]. Widening this approach to model pre-processing by incorporating the effects of the donning procedure, as Lacroix and Patino [96] did for transfemoral amputees, would improve modelling even further.
Alternatives to determine internal loading states
Modelling is not the only approach to determine internal loading states. Medical imaging offers ways to visualise soft tissue deformation directly. MRI tagging has been proposed and used to derive strain values of compressed muscle [99, 102]. Another option would be to measure physiological features that are indirectly related to the mechanical state of tissues. DT MRI for example captures changes in diffusivity due to compression [103], making it a potential measure for strain [104]. Given the complexity and limitations that the predictive nature of analytical and computational modelling inheres, it seems sensible to integrate technologies that have not yet been used in the context of pressure ulcer research.
Limitations
Whilst we believe that our systematic approach according to the PRISMA-ScR guidelines identified most of the relevant literature, there are various sources of potential bias. Firstly, the lack of registration of the review protocol prior to performing this scoping review may have introduced conduction and reporting bias. By restricting the search to English language and excluding grey literature, we might have missed studies and introduced selection bias. Further papers might have been overlooked due to the heterogeneities in wording and definitions of pressure injuries [8, 105, 106], despite the use of common DTI- and prosthesis-related expressions in our search strategy. Additionally, the review process was mainly conducted by one reviewer. We aimed to limit possible selection bias by consulting a second reviewer in case of uncertainty. Lastly, we focussed on transtibial prosthetic users, whereas transfemoral amputees might experience similar conditions and could therefore be included in future endeavours.