Experimental and Clinical Study of Flap Monitoring with Analysis of Clinical Course of the Flap Using Infrared Thermal Camera

1. Introduction

Flap surgery is a technique in plastic and reconstructive surgery where any type of tissue is lifted from a donor site and moved to a recipient site with intact blood supply. This is done to fill a defect such as a wound as a consequence of injuries or surgery, when the remaining tissue is unable to support a graft, or to rebuild more complex anatomic structures such as the breast or jaw. It stands as a cornerstone of reconstructive procedures, playing a pivotal role in the restoration of form and function following trauma, oncologic resection, or congenital malformation [1]. The success rate of flap surgery relies on meticulous surgical technique and postoperative monitoring to assess tissue viability and mitigate complications [2].

Post-operative monitoring of free flaps is important in flap survival, immediate action increases flap salvage rate. Although various methods are available, room for improvement remains. Currently, free flap monitoring can be done through a variety of methods, including physical examination (skin color, turgor, temperature, and capillary refill), surface temperature recording, external doppler, implantable doppler, color duplex sonography, laser doppler flowmetry, infrared thermography, near-infrared spectroscopy, white light spectroscopy, and microdialysis. Despite the extensive exploration of monitoring methodologies, the absence of a definitive gold standard persists, which necessitates continued investigation into novel approaches [3,4].

Thermal cameras have proven their value in medicine and are nowadays readily available at a low cost. Their non-invasive nature and ability to provide immediate, quantitative data make infrared thermal imaging an attractive adjunct in flap monitoring [5]. Various clinical applications involving depth analysis of burns, detection of complications for diabetic foot ulcer and preoperative planning of perforators using infrared thermal imaging have been made in plastic surgery [6,7]. Furthermore, there have been several publications describing flap monitoring using infrared thermal imaging [5].

This study aims to present a new perspective on the clinical course of flaps in flap monitoring by taking advantage of the convenience of forward-looking infrared (FLIR) thermal cameras, focusing on the correlation between flap necrosis and temperature comparison of flaps from a different thermal analysis perspective than previous studies.

2. Materials and Methods

2.1. Rat Models

The in vivo experimental protocols and study were approved by the Institutional Animal Care and Use Committee (IACUC) of Soonchunhyang University (Approval number: SCH-0062). In total, 28 transgenic Spraque-Dawley (SD) rats (12 weeks) were involved in this study. On the day of implantation, all rats were trimmed over the backside under anesthesia with isoflurane (Terrel, USA), oxygen and N₂O. In this study, a reverse McFarlane skin flap of 3×9㎝ size was elevated using a blade to induce skin necrosis models to evaluate skin flaps. The flap was immediately closed by suturing with 4-0 nylon (AILEE Co., Korea) (Figure 1). The entire area was covered by gauze and then appropriately bandaged without disturbing the gauze over the wounds. To relieve their pain and prevent infection, tramadol and enrofloxacin were given intramuscularly. All rats were individually isolated in each cage at controlled temperatures (25℃) and sacrificed on the planned time points (Day 7). The experimental photographs were taken using a Cannon EOS R6 Mark Ⅱ (Tokyo, Japan) camera at a standard distance (40㎝) to assess the wounds and flaps on postoperative day (POD) 3 and POD 7. The thermal images were obtained by FLIR camera on POD 3 and POD 7. We calculated the ratio of necroitc area/total area and analyzed the difference between necrotic area and non-necrotic area using Fiji-ImageJ software(ImageJ, US).

2.2. Patients

The protocol of this study was approved by the Institutional Review Board approval (IRB number: 2023-12-071) Patients who underwent pedicled flap or free flap surgery on any locations of the body between December 2021 and September 2023 were included. Patients with distinct hypothermia or hyperthermia were excluded. Also, the buried flaps were excluded. All patients were advised to take absolute bed rest for a week and given intravenous antibiotics for preoperative and postoperative infection control. The clinical photographs were taken using an ILCE-7M4 (Sony®, Tokyo, Japan) camera at a standard distance (40㎝) to assess the wounds and flaps.

2.3. Infrared Thermal Imaging

Thermal images were captured using the FLIR C5 camera (Teledyne FLIR LLC, US). The FLIR C5 has two cameras a thermal imager (160×120 pixels) and 5-megapixel visual camera (640×480 pixel), and LED flashlight. Two images were obtained and merged by the Multi Spectral Dynamic Imaging (MSX) technology, resulting in one thermal image with resolution of 640×480 pixels. It has the capacity of detect temperature ranging of -20°C to 400°C and the thermal sensitivity is <70mK.

All flaps were monitored at POD 0, 1, 2, 3 & 7, and some flaps even up to 2 weeks with FLIR camera and clinical findings such as skin color and capillary refill time. The thermal imaging was performed at operating room with a controlled temperature (18-22°C) and humidity (50-55%). To reduce the temperature bias, thermal imaging was performed 10 minutes after the patient entered the operating room. The images were captured at a distance of 40㎝ from the flaps. After monitoring, the dressing methods for all flaps were equally controlled with the use of ointments and gauze. If the location of the flap was in an area where splinting was possible, the instability of the flap was reduced through the use of a splint.

2.4. Infrared Thermal Image Analysis

The authors also used the FLIR Thermal Studio software (Teledyne FLIR LLC, US) to analyze the thermal images which allows to identify the temperature of a specific point or area on a thermal image and calculate the temperature difference (Figure 2).

On the obtained thermal images, the authors marked the range of the total flap area and measured the average of temperature value of the flap area (ATF) at imaging software program. In addition, we measured the temperature of the perforator of the flap (PF), as well as the normal skin (NS) at the point 2cm proximal from the flap in anatomical position. The temperature difference (dT) between normal skin and the perforator of the flap was compared. In addition, the temperature difference of the perforator and the average temperature of flap (AFP) was obtained.

2.5. Statistical Analysis

Statistical significance was determined using the Mann-Whitney u test for independent groups. All statistical analyses were performed using SPSS (ver 27; IBM Corp., Armonk, NY, USA). A p-value of less than 0.05 was considered statistically significant.

3. Results

3.1. Rat Models

The degree of necrosis of the flaps was evaluated by flap color and skin turgor on POD 3 and POD 7. The necrotic part of the flap was dark in color and appeared to dry out hard. On POD 7, all rats were euthanized, and sutures were removed to evaluate the undersurface of the flap, and the difference in necrosis of the flap was more noticeable on the undersurface of the flap. The mean surface area of total flap was 27.12±0.90㎠, and mean ratio of total flap area over the necrotic area were 24.4% and 27.5% in POD 3 and POD7 respectively. On thermal imaging, the necrotic flap was clearly distinguished from the viable flap by color differences, with areas of high temperature being white and areas of low temperatures being red (Figure 3). The mean temperature difference in each area was 0.8℃ in POD 3 and 0.792℃ in POD 7 (Table 1).

3.2. Patients

3.2.1. Patient Demographics & Flap Characteristics

In total, 22 patients and 22 flaps were included in this study. There was a total of 16 male and 6 female patients, with a total average age of 57.45 years (range: 27 to 100). All flaps were either pedicled flaps (14 in total) or free flaps (8 in total). Various pedicled flaps were made depending on the location of the defect and the surrounding vascularity and an anterolateral thigh (ALT) free flap was used for free flap reconstruction in all cases except 2. Necrosis occurred in 2 cases of the free flap group due to arterial insufficiency, and 1 case of total necrosis occurred in the free flap group due to venous congestion. In the pedicled flap group, 1 case of partial necrosis occurred. Initially, arterial insufficiency occurred in 3 cases, but despite salvage procedures being performed in 2 of these cases, flap failure occurred, and a new flap was reapplied. One of these cases survived stably without necrosis, so it was not included in the necrosis group. The location of the flap site varied from the scalp to the lower extremity. There was no other complication such as hematoma or infection (Table 2, Table 3).

3.2.2. Infrared Thermal Imaging Analysis

The temperature difference between normal skin and the perforator (dT of NS-PF) in the non-necrosis group showed a gradual decrease over time, while that in necrosis group increased.

In POD 0, the temperature difference between normal skin and perforator (dT of NS-PF) was shown to be lower in the necrosis group (1.55) than in the non-necrosis group (1.817). However, from POD 1 to 7, the Necrosis group showed higher dT of NS-PF values than the non-necrosis group. Among them, there was a statistically significant difference between the non-necrosis group and the necrosis group in POD 3 and POD 7. When the necrosis group was subdivided, the difference between the total necrosis group and the non-necrosis group became higher.

The difference between the perforator temperature and the average temperature of the flap (dT of PF-AFP) showed positive values from POD 0 to POD 7 in the non-necrosis group, while the necrosis group showed negative values from POD 1 to POD 7. On POD 1, the dT of PF-AFP was significantly different between the non-necrosis group (0.35) and the necrosis group (-0.175) with a p-value of 0.002. This trend continued with significant differences on POD 2 (p = 0.003), POD 3 (p < 0.001), and POD 7 (p < 0.001)(Table 4).

3.2.3. Case of Non-Necrosis Group (Case #7 & Case #13)

Case #7 is a 100-year-old female patient with no past medical history aside from squamous cell carcinoma of the left upper eyelid. She underwent reconstruction through ALT free flap after resection of the full-thickness skin including the tumor on the left upper eyelid. Case #13, a healthy 23-year-old female patient, was reconstructed through superficial circumflex iliac artery perforator (SCIP) free flap after distal finger necrosis caused by the use of vasopressors due to septic shock. Both cases showed stable progress without necrosis, and the temperature difference between normal skin and perforator (dT of NS-PF) showed a decreasing trend over time (Figure 4). On POD 0, the dT of NS-PF for Case #7 and Case #13 were 0.9 and 1.4, respectively. However, by POD 7, the dT of NS-PF had decreased to 0.4 and 0.5 for Case #7 and Case #13, respectively, indicating a reduction in temperature difference over time. This decrease in dT of NS-PF reflects the stabilization and successful integration of the flap without vascular compromise. The thermal images of both cases on POD 7 showed a similar color to the surrounding area except for some margins, indicating no signs of necrosis. Both flaps survived stably without any necrosis and the patients were discharged after POD 14.

3.2.4. Case of Venous Congestion in Necrosis Group (Case #10 & Case #21)

Case #10 is a 51-year-old male patient who has been taking medication for diabetes for 15 years and underwent ALT free flap reconstruction for a defect in his left foot that occurred after orthopedic surgery. From POD 2, the color of the flap gradually turned purplish and showed a congestive appearance. Conservative salvage using heparin gauze was attempted, but the effect was not significant. Subsequently, venous congestion progressed, and the dT of NS-PF value continued to increase. By POD 7, the flap became darker in color and eventually led to total necrosis. The dT of NS-PF was 1.2 on POD 0 and increased to 3.5 by POD 7, indicating significant venous congestion and the resulting necrosis. Case #21 is a 49-year-old female patient with no medical history or medication history who suffered a right distal tibia-fibula fracture after a traffic accident. A peroneal artery perforator-based fasciocutaneous flap was performed to reconstruct a defect exposing the metal plate after reduction with plate fixation for an ankle fracture. From POD 1, the flap began to become hyperemic from the distal portion and progressed with a congestive appearance. Salvage was performed using leech therapy. Until POD 7, a similar appearance was maintained, but thereafter, the central portion showed improvement in congestion, and only partial necrosis of the distal portion remained on POD 14. The dT of NS-PF initially showed an increasing trend, being 1.0 on POD 0 and 2.5 on POD 7, but later decreased to 0 by POD 14, indicating improvement in congestion and partial recovery. (Figure 5).

3.2.5. Case in Which a New Flap Was Applied after Salvage Procedure for Arterial Insufficiency (Case #12 & Case #14)

Case #12 is a 48-year-old male patient who underwent a thoracodorsal artery perforator (TDAP) free flap for a defect that occurred after orthopedic surgery due to a fracture of the left distal tibia. On POD 1, doppler sound of the anastomosed vessel could not be heard, and it was judged as arterial occlusion. Despite the salvage procedure, the flap failed, and an ALT free flap was re-applied. The new flap remained stable without necrosis on POD 7. Initially, the dT of NS-PF was 2.0 on POD 0 but increased to 2.8 on POD 1, reflecting the arterial insufficiency. After the application of the new flap, the dT of NS-PF decreased to 0.7 on POD 7, indicating the success of the new flap. Case #14 is a 75-year-old male patient who underwent burr-hole drainage several times due to subdural hemorrhage and was hospitalized due to surgical site infection. ALT free flap was performed for a scalp defect where a plate fixed to the skull after the site of the craniotomy was exposed. On POD 1, the flap was pale in the center and mottled around the margins of the flap, suggesting arterial occlusion. Despite the salvage procedure, the flap failed, and a vastus lateralis muscle free flap was performed. However, the new flap also showed signs of arterial occlusion and progressed to total necrosis. Initially, the dT of NS-PF was 1.8 on POD 0 and increased to 3.0 on POD 1, similar to Case #12. However, after the new flap was applied, the dT of NS-PF further increased to 3.5 on POD 7, indicating continued vascular compromise and eventual total necrosis. The contrasting clinical courses of the two cases were clearly distinguished by the decrease and increase in the dT of NS-PF values. In Case #12, the successful salvage with a new flap was reflected by a significant decrease in dT of NS-PF, while in Case #14, the failure of the new flap was indicated by the continued increase in dT of NS-PF. (Figure 6).

3.2.6. The Difference between the Temperature of Perforator Area and the Average Temperature of Total Flap (dT of PF-AFP).

On POD 0, both the necrosis and non-necrotic groups showed positive PF-AFP values, with 0.539 in the non-necrosis group and 0.175 in the necrosis group, showing no significant difference (p = 0.118). However, from POD 1 to POD 7, the necrosis group showed positive values while the non-necrosis group showed negative values. This difference was statistically significant (Table 4).

4. Discussion

Skin grafts and flap surgeries are two surgical techniques that are commonly utilized by plastic surgeons when a defect cannot be closed by primary or secondary intention. In plastic surgery, the choice between skin grafts and flap surgeries depends on the type and condition of the wound. These methods have distinct indications and purposes based on their technical differences and the specific needs of the reconstruction. Skin grafts involve transferring a thin layer of skin from a donor site to a recipient site, which is ideal for superficial defects with good vascular beds. However, they may not be suitable for areas with poor vascularization or deeper tissue loss. In contrast, flap surgeries involve transferring a full thickness of tissue, including skin, muscle, and blood vessels, from one part of the body to another, maintaining its own blood supply. This makes flap surgeries essential for reconstructing defects where blood supply is compromised or where complex three-dimensional tissue reconstruction is needed.

Within flap surgeries, there are two main types: pedicled flaps and free flaps. They represent distinct methods in reconstructive surgery, each with specific applications and characteristics [1]. Pedicled flaps retain their original blood supply through a connected pedicle, making them suitable for shorter distances between the donor and recipient sites and generally reducing the risk of vascular complications. They are favored for their reliability and less demanding microsurgical requirements. Free flaps, on the other hand, involve completely detaching the tissue, including its blood vessels, and microsurgically reconnecting it to blood vessels at the recipient site. This allows for tissue transfer over longer distances and offers greater versatility in tissue types, making them preferable for more complex reconstructions [8,9].

Flap surgery remains the most prevalent and practical surgical approach within the realm of plastic surgery. The demand for flap surgery is on the rise due to several factors, including the increasing incidence of malignant tumors and the growing need for immediate breast reconstruction following mastectomy [10,11]. Extensive research has been conducted on flap surgery, revealing its evolving complexity and the advancements in surgical techniques that continue to enhance its efficacy and outcomes.

The most important factor in flap surgery is flap vascularity, and most flap complications are associated with flap vascular compromise [12]. Vascular compromise is largely divided into two types: arterial occlusion and venous congestion. The pathophysiology of arterial occlusion causes insufficient oxygen delivery and a concomitant deficiency in the elimination of harmful metabolites from the tissues that are impacted. The outcomes include reactive oxygen species (ROS) buildup, an influx of inflammatory cells like neutrophils and macrophages, and a gradual release of cytokines, which creates a cycle of inflammation that eventually results in tissue necrosis [13]. The main variables that link the biochemical pathways between tissue necrosis and prolonged ischemia are reactive oxygen species (ROS), which cause microcirculatory damage that leads to permanent deterioration and harm. When arterial flow is persistent in venous congestion, the intravascular pressure rises and the microvasculature bleeds into the extra-vascular space. The vessels externally compress and collapse as a result of the elevated extravascular pressure. Edema that develops in the interstitial tissue prevents oxygen from diffusing through the tissue, causing more tissue damage [13,14,15]. The complications of flap surgery involve flap failure, leading to secondary flap surgery, partial or total loss of the flap. For the successful survival of the flap, flap monitoring is essential immediately after surgery, and, if necessary, procedures such as salvage are required.

In reconstructive microsurgery, perforator flaps have become a significant development due to their ability to preserve major vessels and muscle function while allowing less invasive reconstruction. A perforator flap is defined as a flap based on a vessel that perforates the envelope of the target tissue to be transferred. This envelope could be the superficial fascia for skin, the deep fascia for muscle, the periosteum for bone, or the perineurium for nerve. The evolution of perforator flaps has enabled surgeons to perform complex three-dimensional reconstructions with improved outcomes and reduced donor site morbidity. This new definition helps in precisely classifying all flaps, ensuring a better understanding and application in reconstructive procedures [16].

Conventionally, flap monitoring is often performed with subjective methods such as clinical observation of skin color, skin turgor, flap temperature, and capillary refill time. These methods rely on the surgeon’s experience and judgment to assess the viability of the flap postoperatively [17]. A normal flap typically exhibits several key criteria. It should feel warm to the touch, indicating good perfusion. The color should be a healthy, pink shade. The capillary refill time should be less than 2 seconds when pressed and released. The flap should have good skin turgor, with rapid return to its original position after being pinched. In some cases, the flap should retain normal sensation, although this can vary depending on the type and location of the flap [18]. When there is an issue with blood flow, specific signs can indicate whether the problem is arterial insufficiency or venous congestion. In the case of arterial insufficiency, the flap may appear pale or white due to insufficient arterial blood supply. It may feel cool to the touch and exhibit a prolonged capillary refill time, often more than 2 seconds. If not addressed, arterial insufficiency can lead to necrosis, or tissue death. On the other hand, venous congestion typically presents with cyanosis, a blue or dusky appearance due to poor venous outflow. The flap may become swollen and edematous. Despite the congestion, the flap may feel warm and have a delayed capillary refill time, similar to arterial insufficiency but with a bluish color [19]. In addition to these subjective methods, various objective methods utilizing medical devices are also used to monitor flaps. Numerous methods for monitoring flaps have been extensively studied, each presenting distinct advantages and disadvantages shown in Table 5, for example [20,21,22,23,24,25,26] (Table 5). However, despite the extensive exploration of monitoring methodologies, there is currently no universally established gold standard among these modalities. Each method has its limitations, such as the requirement for specialized equipment, potential invasiveness, and operator dependency.

Various studies related to infrared thermography for flaps have been conducted for a long time, and interest in this field has recently increased. Many studies, such as burn depth analysis and preoperative perforator mapping, have been conducted, and there are also many existing studies that apply infrared thermography to flap monitoring [27,28]. These studies show a diverse range of methods and results.

In rat models of this study, a model that induces flap necrosis is required. To understand this, it is essential to introduce the concepts of anterograde flaps and reverse flaps. Anterograde flaps have blood flowing in the natural direction, from the center of the body towards the periphery. This configuration typically results in smooth and stable hemodynamics, with minimal vascular resistance, ensuring adequate blood supply and higher survival rates of the flap. In contrast, reverse flaps involve blood flow in the opposite direction, from the distal part of the flap back towards the center. This reversal can increase vascular resistance and potentially lead to less stable blood flow, making the flap more prone to ischemia and necrosis [29,30,31,32]. In this study, a caudally based reverse McFarlane flap model was used in rats to intentionally induce flap necrosis for observational purposes. This model is advantageous for studying necrosis because the vascularity of the distal portion of the flap decreases over time, leading to necrosis [33,34]. Thermal imaging was employed to monitor these changes, with temperatures visualized in color gradients (white to blue, indicating decreasing temperatures). Areas undergoing necrosis displayed lower temperatures compared to viable tissue, particularly noticeable on the undersurface of the flap. On POD 3 and POD 7, the average temperature in the necrotic area did not significantly differ from the total flap area, likely due to the influence of core body temperature (Table 1). Since rats have minimal dorsal fat tissue, the skin flap is in direct contact with the body core, which affects the temperature distribution [35]. This core temperature effect minimized the temperature gradient between necrotic and non-necrotic areas, making the impact of external factors more prominent.

In the clinical cases, this study obtained the data by comparing the temperature between normal skin and the perforator of the flap and comparing the temperature of the perforator point of the flap with the average temperature of the entire flap area using infrared thermal imaging in the patients group.

In R. Kraemer’s study, they evaluated flap skin temperature and capillary microcirculation through postoperative monitoring using a digital infrared surface thermometer combined with laser-doppler and photospectrometry. They proved their hypothesis that the flap skin temperature decreases when there is arterial thrombosis or venous compromise [36]. The other previous studies demonstrated that if microvascular compromise occurs, the flap surface temperature decreases and temperature difference with adjacent skin increase [37,38]. In the same context as these, the non-necrosis group showed a decreasing trend in the temperature difference between normal skin and perforator (dT of NS-PF) over time, but cases with vascular compromise showed an upward trend in the temperature difference between normal skin and flap due to the flap’s temperature dropping further (Figure 7). Although this study did show a statistically significant difference in POD 3 and POD 7, there was no statistically significant difference between the non-necrosis group and the necrosis group until POD 2. However, because early rapid detection of vascular compromise and salvage of the flap is important, it should not be considered as absolute guidance but can be used as a supplementary indicator [39].

In Case #7 and #13, the dT of NS-PF value decreased as time passed compared to the initial stage, but in Case #10 and #21, which showed venous congestion, the temperature of the perforator decreased compared to normal skin as time passed, and the dT of NS-PF value increased. However, in Case #21, the temperature of the perforator increased again from POD 7 as congestion improved through leech therapy, and it ultimately progressed to partial necrosis rather than total necrosis. Therefore, when venous congestion occurs, it indicates that if the dT of NS-PF value does not increase steadily but decreases, the progression to total necrosis may not occur (Figure 5).

Case #12 and Case #14 are cases in which salvage procedures and new flap applications were performed on POD 1 due to arterial occlusion. In both cases, the dT of NS-PF value on POD 1 was higher than the average of the non-necrosis group. In addition, when comparing Case #14 with Case #10, where venous congestion is represented, it can be confirmed that the dT of NS-PF value is higher in Case #14 with arterial insufficiency than in Case #10 with venous congestion. However, after applying a new flap, a contrasting, downwards trend of dT of NS-PF value can be confirmed depending on the presence or absence of flap necrosis (Figure 6).

In Figure 7, the temperature difference between perforator and average of flap (dT of PF-AFP) in the non-necrosis group showed a positive value in all cases from POD 0 to POD 7, except for 1-2 points, while in the necrosis group, it almost always showed a negative value (Figure 8). A positive value of dT of PF-AFP can be interpreted as the temperature of the perforator being higher than the average temperature of the flap, with a negative value meaning the opposite. In the study by Whitaker IS et al., mapping of the deep inferior epigastric artery perforator was carried out using a dynamic infrared thermal camera by performing a cold challenge with a water pack and utilizing the hot spot appearing on the camera to measure the temperature at the perforator with the most abundant flow of heat through the vascular flow of the perforator as the flap re-warmed [40]. The study by Perng CK et al. explained that the surface temperature of the flap was determined by three mechanisms: heat from blood flow entering the flap, heat conduction from tissue located at the bottom of the flap, and heat loss to the air in contact with the flap surface [41]. When applied to this study, if the average temperature of the perforator and the entire area of the flap (which is limited to the inside of the flap) are compared, the other two heat sources are unified, and temperature comparison is possible as heat from blood flow. In addition, since all flaps entered theatre lower than room temperature and thermal imaging was performed 10 minutes later, this can be considered a form of cold challenge. Of course, although the temperature of the perforator site of the flap is not clearly distinguishable from the overall temperature of the flap in terms of color, it is possible to compare the temperature of the perforator and the average temperature of the flap using thermal imaging analysis software. If vascular compromise occurs due to reasons such as thrombosis, malfunction of the circulation inside the flap may occur, and the difference between the temperature of the perforator site and the average temperature of the entire flap may decrease. Therefore, if the dT of PF-AFP value is less than 0, it can be interpreted that there is microvascular compromise in the flap.

Many previous studies analyzed temperature in two methods in flap monitoring using infrared thermography. The first was to measure the absolute temperature gradient of the flap itself, and the other was to measure the gradient in temperature difference between the flap and adjacent normal skin. This study utilized the latter method but showed limitations in early flap monitoring due to variables in temperature measurement of normal skin. Since comparisons cannot be made between flaps on different patients or on different sites, there is a lot of bias towards environmental variables. However, comparing perforator temperature and average temperature within the flap itself can reduce other biases and allow a more accurate evaluation of vascular compromise.

There are several limitations of this study. The sample size, firstly, was too small to evaluate the efficacy of flap monitoring through thermal imaging - Due to the small number of necrosis cases and the inconsistent size or type of flap, it is assumed that the lack of sufficient sample size prevented the derivation of statistical significance. Secondly, we measured the temperature of normal skin at a specific point, but this is somewhat unreliable as it does not represent the temperature of normal skin throughout. For the study by S. Hummelink et al., the mean temperature value within a specific area of adjacent normal skin had been marked and used for comparison instead [37]. In addition, due to the nature of the FLIR camera, various biases are likely to occur due to the influence of the surrounding environment. MA Moran-Romero et al. argued that free flap monitoring through thermal imaging could produce ambiguous results [42]. However, we compensated for this by conducting temperature measurements in a surgical theatre where temperature and humidity were controlled and there was a unified dress code. Also, upon comparison of the results, we analyzed the gradient of temperature differences as opposed to absolute temperature values.

It is difficult to evaluate the microvascular flow of the flap with absolute value of results obtained through infrared thermal assessment using a FLIR camera. However, the gradient of the temperature difference between normal skin and the perforator over time can be a guide in interpreting the overall course of the flap. In addition, this study suggests a novel method for flap monitoring using infrared thermal imaging with the analysis of the temperature difference between the perforator and the average temperature of the flap, which has sufficient value as a supplementary indicator in judging vascular compromise. Above all, the FLIR camera is a contact-free modality that does not harm the flap, and its high convenience will be more frequently utilized in flap monitoring.

5. Conclusions

Numerous studies are investigating flap monitoring using infrared thermal imaging. FLIR cameras offer the benefits of being non-invasive and highly convenient. While further research is necessary before FLIR cameras can serve as definitive indicators for vascular compromise during flap monitoring, they can effectively function as auxiliary tools for assessing and predicting the overall clinical progression of flaps. This study introduces a novel method for flap monitoring via infrared thermography by analyzing the temperature differential between the perforator area and the average flap temperature, demonstrating significant potential for enhancing the flap monitoring process.