1.

Introduction

A reliable assessment of tissue viability is critical to effectively treating patients who sustain traumatic extremity injury. Despite their relatively inert appearance, bone, tendons, and ligaments are, physiologically speaking, highly active. Bones, specifically, receive $\sim 10\%$ of the entire cardiac output of blood volume.¹ Dysvascular tissues have limited potential to regenerate, heal, or fight infection due to insufficient delivery of inflammatory cells, growth factors, osteoprogenitor cells, endogenous immune cells, and antibiotics. It is clear that, in orthopaedic injury, nonviable or poorly perfused bone and soft tissue inhibits healing and increases the risk for infection. Because of this, assessing bone and soft tissue viability is critical to making effective treatment decisions, and the comparative translucency of these tissues to red and near-infrared light makes optical methods feasible.

2.

Current Paradigm and Challenges

To date, the eyes and hands of the surgeon remain the dominant “imaging modality” used to make decisions regarding the health of soft tissue or bone. Methods currently used to assess tissue and bone perfusion are subjective and entail non-quantitative clinical cues. These techniques include clinical judgement based on the color and swelling of skin, presence of skin wrinkles, color, and turgor of deep tissues, measurements of prior cadaveric studies (for example, regarding perfusion of the meniscus in the knee),^2,3 extent of soft tissue stripping off bone, and the “paprika sign” (defined as scattered pinpoint bleeding on bone).^4,5 The subjective or imprecise nature of these assessments leads to substantial variation in treatment and a lack of advancement in objectively driven treatment protocols.

Imaging blood perfusion to soft tissue and bone can be done using traditional imaging modalities, such as positron emission tomography/computed tomography (PET/CT)^6,7 and contrast-enhanced magnetic resonance imaging (MRI).⁸ These imaging modalities provide detailed and accurate information of the fractured or diseased bony and soft tissue anatomical structure, yet they have issues with limits to spatial resolution and artifacts from metallic implants. The logistical limitations are the cost recovery, the need to schedule them as a separate procedure, and the limits to using them for routine real-time assessment or repeated re-examination.^6,7 MR-based techniques are limited in orthopaedic patients due to metal artifact and signal dropout in association with metal implants. A limited number of studies have used PET/CT or MRI to measure blood flow to bone for surgical indications, including infection, osteomyelitis, and nonunion.^8–13 However, there is limited to no access to these imaging methods in real-time in the operating room. At the present time, fluoroscopy and plain radiographs are used regularly in the operating room to visualize bones and assess fracture alignment and/or hardware placement. In contrast to these radiologic modalities, optical imaging can obtain metabolic information that provides physiological abnormities of the tissue on microvascular perfusion and/or bone and tissue viability. The main limitation of optical imaging is the inability to achieve both high spatial resolution and high tissue penetrance: any added value, then, is either in high-resolution imaging at low penetrance or more macroscopic assessment of tissue function at a more moderate to medium penetrance. Thus, optical imaging can be a unique complementary imaging modality to the conventional imaging modalities.

3.

Potential Optical Imaging and Sensing Needs

4.

Optical Tools Utilized to Date

When optical imaging is used for maximum benefit, light of different wavelengths or different interaction mechanisms can be used to capture biophysical changes in the tissue, occurring in the vascular as well as extravascular and cellular matrix compartments.^81–83 The key to most of the applications, as with all imaging systems, is to have a good biophysical understanding of the compartmentalization of the endogenous or exogenous chromophores that contribute to each image so that the images can be interpreted appropriately. When compared with conventional medical imaging modalities, such as MRI, PET, or x-ray CTCT, optical imaging detects changes in light reflectance, absorption, and scattering of the tissue and thus offers advantages of nonionizing, low-cost, and portable imaging for routine, low-cost, longitudinal monitoring of response to therapy or during human surgery.^81–83 The logistics of use of optical imaging, being real time and at the point of care, make it more comparable to C-arm x-ray or ultrasound. Therefore, any determination of the value of optical imaging in tissue injury care needs to be done within the context of what information can be assessed at the point of surgery decision making.

The challenge of applying optical tools to bone and soft tissue assessment has been the high scattering nature of both tissues, limiting the spatial depth dependence of the information. Clearly surface camera-based imaging tools interrogate a range of depths in the millimeter to centimeter range,⁴⁵ with rapidly decreasing sensitivity and spatial resolution with small increases in depth. So, the two major attractions to optical tools come in the categories of one of three rough regimes:

These regimes are defined by the type of tool used and dictate what information can be obtained during imaging. The matching of the capabilities of each tool with the needs in each application are essential to evaluate.

Table 1 presents an overview of the tools that have been tried in soft and hard tissue trauma and disease studies. The engineering aspect of the contrast mechanism is used to categorize the tools in column 1. However, in terms of utility and function for trauma surgery, it is likely much more functional to consider the depth of interaction, column 3, because this is the factor that will determine what biomedical information the tool can supply. For example, optical coherence tomography (OCT) has been widely used^91,92 in bone imaging but with penetrance on the order of $<1\mathrm{mm}$, thereby allowing its use only for largely research studies and unlikely as a major surgical tool. Similarly, ultraviolet (UV) fluorescence imaging provides information bone necrosis, and the addition of antibiotics has been shown to enhance the ability of this fluorescence to provide diagnostic value about osteonecrosis. However, the penetrance of both UV and antibiotic fluorescence to just tens to hundreds of microns ($\mu \mathrm{m}$) does limit their use to situations in which the imaging is directly on the bone surface in question.

Table 1

Listing of optical imaging tools utilized in soft and hard tissue trauma and disease studies, categorized by the type of contrast, depth regime, and a brief strengths and weaknesses list.

In terms of molecular information about bone, perhaps no other measure has the potential to provide high feature information than Raman spectroscopy.^99–105 The measures of bone regrowth have been well studied, and measurements through even centimeter thick tissues have been demonstrated with spatially offset Raman or Raman near-infrared (NIR) tomography.^106,109 However, practical issues in the very low signal strength and the specificity of the signal have limited the routine application of this technique to research studies to date. More exotic methods such as terahertz imaging have been tried, although their specificity to biological information beyond water content of the tissue is unclear and they are constrained to fairly superficial tissue thicknesses ($\mu \mathrm{m}$ to mm).

At the other end of the penetrance scale (mm to cm), near-infrared spectroscopy (NIRS) can be used through several centimeters of tissue, albeit with very limited spatial resolution, for deep tissue sensing of absorption and scattering features. Time-resolved NIRS, which measures photon time-of-flight allowing for selection of late-arriving photons to increase depth sensitivity, may offer improvements in resolution and quantification and has been used recently in joint imaging.⁹⁶ So, while NIRS has its limitations, it is likely a more useful tool for macroscopic tissue assessment and is largely limited to the measurement of blood, oxygen saturation, water, and lipid content of the tissue or scattering features related to the tissue matrix composition. Thus, it can be used in conjunction with ICG as a dynamic contrast-enhanced NIRS.¹³⁰ Future work might examine the value of NIRS for osteonecrosis assessment or flow. The assessment of blood flow can be done microscopically with OCT imaging. However, macroscopic bone blood flow has been one of the challenges. Recently, the flow assessment of imaging the fluorescence from ICG has been shown to provide predictive value for flow and even potentially differentiate between endosteal and periosteal flow.^27,44,45,49

Perhaps the most promising tools are ones in which there is a hybrid approach to imaging information, so the tool is not being evaluated as a single diagnostic, but rather is complementary to existing clinical tools or as a combination of more than one tool. The most obvious of these is the use of ICG fluorescence imaging in which it has been added onto color laparoscopy imaging, and the two can be codisplayed on a video monitor for the surgeon to use.¹³¹ This technique has high penetrance, with several commercial systems approved for use.¹³² The creation of new molecular-specific optical contrast agents is evolving rapidly with many clinical trials ongoing. These should be eventual contrast agents for fluorescence-guided surgery in trauma, even if many of the initial studies are devoted to diseases such as cancer or cardiovascular disease.

Photoacoustic imaging exists as a hybrid optical-ultrasound tool, which can in some cases provide both optical contrast as well as ultrasound images, although like ultrasound it has its strengths in soft tissue imaging and can have large artifacts from the interfaces between hard and soft tissues or voids within either tissue. Still, the photoacoustic mode of imaging vascularity and even flow can be significant and provide exquisite detail on vascular patterns and function when deployed appropriately.^127,128,133 The clinical utility of these systems is evolving rapidly, and evaluating their use in trauma surgery should be a high priority.

Figure 1 shows an overview of the needs in trauma surgery, based upon the depth of penetrance of the imaging tool and the biological timeline of the disease. This figure provides a basic template from which to think about the tools and their value to trauma or disease diagnostics. While fine structured fractures might be seen with high-resolution tools, this is the realm of x-ray imaging or microscopy with optics such as OCT.^91,92 Vascular function such as flow or perfusion to the tissue can be assessed microscopically or macroscopically, based upon the scope of the problem, but most clinical evaluations would require macroscopic sensing such as with NIRS, fluorescence, or photoacoustics. Biological functions such as regrowth or necrosis have two very different sampling scales, either microscopic methods such as UV and antibiotic fluorescence or macroscopic tools such as NIRS and contrast agent fluorescence.^27,44,45,49 Again, for most clinical issues, the sampling of macroscopic signals is likely needed, and the combination with existing clinical tools such as ultrasound or x-ray would be beneficial.

Fig. 1

The matching process of the capabilities of each optical tool with the needs in each application.

5.

Conclusions

Research is needed to optimize optical imaging techniques for tissue trauma surgical applications, and the developments described in Sec. 4 are each important to test. Based on the safety profile and previous implementation success to other surgical domains, the barriers to the translation of these techniques to orthopaedic surgery are relatively low. We believe that using optical imaging to provide surgeons with real-time objective data regarding bone and tissue perfusion will lead to more effective patient-specific management of common orthopaedic conditions with lower morbidity and will result in decreased variability associated with how these conditions are managed. Information provided should go beyond the basic soft and hard tissue structures available with x-ray imaging and incorporate functional flow and perfusion or tissue metabolism features that are likely to have a higher specific correlation to the outcome of the procedure. Optical tools have the best opportunity to impact surgery because of their inherent point-of-care use, their relatively low capital costs, and their compatibility with intraprocedure measurement.