*Article* **Incoherent Optical Fluctuation Flowmetry: A New Method for the Assessment of Foot Perfusion in Patients with Diabetes-Related Lower-Extremity Complications**

**Polina Glazkova 1,\*, Alexey Glazkov 1, Dmitry Kulikov 2,3, Sergei Zagarov 1, Yulia Kovaleva 1, Alina Babenko 4, Yulia Kononova 4, Elena Kitaeva 4, Timur Britvin 1, Natalia Mazur 1, Roman Larkov <sup>1</sup> and Dmitry Rogatkin <sup>1</sup>**


**Abstract:** (1) Background: To date, there are no studies evaluating the ability of the incoherent optical fluctuation flowmetry (IOFF) method to assess foot tissue perfusion. The aim of this study was to evaluate the correlation between perfusion values measured by IOFF and TcPO2 in patients with diabetes-related lower-extremity complications. (2) Methods: This was an observational, crosssectional, two-center study. Diabetic patients with peripheral artery disease and/or diabetic foot ulcers were studied (n = 27, examinations were carried out on 54 legs). Perfusion in the foot tissues was assessed using TcPO2 (reference standard for this study) and the IOFF method. (3) Results: High correlation coefficients of all perfusion parameters measured by IOFF with TcPO2 (Rs 0.7 to 0.76) were shown. The study demonstrated that the IOFF method allows, with a sensitivity of 85.7% and a specificity of 90.0%, the identification of patients with a critical decrease in TcPO2 < 20 mmHg. (4) Conclusions: The high correlation of IOFF parameters with TcPO2 and the moderately high sensitivity and specificity in detecting patients with severe ischemia of foot tissues shows the promise of the method for assessing a tissue perfusion in patients with diabetes-related lower-extremity complications.

**Keywords:** microcirculation; peripheral artery disease; diabetes mellitus; critical limb ischemia; laser doppler flowmetry; incoherent optical fluctuation flowmetry (IOFF)

#### **1. Introduction**

Diabetic foot syndrome (DFS) is a complication of diabetes mellitus and is described as a group of symptoms including neuropathy, reduced blood supply and infection leading to tissue breakdown, and morbidity that may be followed by amputation. DFS is one of the most serious complications of diabetes mellitus (DM) [1,2]. Peripheral arterial disease (PAD) often has a more aggressive course in diabetic patients and, in turn, can lead to DFS [3,4]. In the population of people living with DM, PAD is characterized by multilevel atherosclerotic lesions as well as greater involvement of the arteries below the knee [5]

Reliable assessment of foot tissue perfusion in patients with limb ischemia is essential for predicting limb outcomes and choosing the treatment algorithm [6]. Measurements of transcutaneous oxygen tension (TcPO2) are widely applied for indirect assessment of foot tissue perfusion in patients with limb ischemia [7]. In a number of studies, TcPO2 measurement has shown predictive value in the management of patients with limb ischemia and diabetic foot ulcers [6,8,9]. This method was included in the Wound, Ischemia, and Infection of the Foot (WIfI) classification system, which is recommended for assessing the risk of limb amputation and potential benefit from successful revascularization in patients with lower extremity atherosclerotic occlusive disease [10]. Thus, the results of TcPO2 assessment can directly influence the management of patients. The significance of this

**Citation:** Glazkova, P.; Glazkov, A.; Kulikov, D.; Zagarov, S.; Kovaleva, Y.; Babenko, A.; Kononova, Y.; Kitaeva, E.; Britvin, T.; Mazur, N.; et al. Incoherent Optical Fluctuation Flowmetry: A New Method for the Assessment of Foot Perfusion in Patients with Diabetes-Related Lower-Extremity Complications. *Diagnostics* **2022**, *12*, 2922. https://doi.org/10.3390/ diagnostics12122922

Academic Editor: Viktor Dremin

Received: 21 October 2022 Accepted: 18 November 2022 Published: 23 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

method (TcPO2) increases in patients with DM, as the ankle–brachial index (ABI) may be falsely elevated due to the medial artery calcification [9,11].

However, in addition to a number of clinical limitations (high variability, a dependence on ambient temperature, the presence of edema, a small area of the analyzed tissue, etc.), TcPO2 measurement is also limited by the cost of equipment, consumables, and duration of the measurement [6,12,13]. Alternative methods of assessing foot tissue perfusion have either not found clinical application (laser Doppler flowmetry, laser Doppler imaging) or also have a number of limitations (skin perfusion pressure, 2D perfusion imaging, fluorescence angiography, etc.) [6,9,14].

Thus, the development of new tools for the assessment of tissue perfusion in patients with atherosclerotic peripheral artery disease and DFS is an extremely important task.

An incoherent optical fluctuation flowmetry (IOFF) method has been developed in the Laboratory of Medical and Physical Research, Moscow Regional Research and Clinical Institute ("MONIKI") [15]. This is the first study to assess the clinical perspectives of the method in healthcare.

This study aimed to evaluate the correlation between perfusion values measured by IOFF and TcPO2 values in patients with diabetes-related lower-extremity complications.

#### **2. Materials and Methods**

*2.1. Study Design, Patients, and Data Sources*

This was an observational, cross-sectional, two-center study. The study was conducted at 2 centers:

A: Moscow Regional Research and Clinical Institute ("MONIKI").

B: Federal State Budgetary Institution "V.A. Almazov National Medical Research Center" of the Ministry of Health of the Russian Federation.

Patients with DM and PAD of lower extremities and/or diabetic foot ulcers were included. PAD was diagnosed by detecting hemodynamically significant stenoses/occlusions of the main arteries of lower extremities. Hemodynamically significant stenoses were defined as the presence of stenosis of at least 50% of the diameter reduction by ultrasound duplex scanning [16–18].

Exclusion criteria: pregnancy; diagnosed systemic autoimmune diseases; severe heart rhythm disorders (atrial fibrillation, frequent extrasystoles); presence of acute respiratory viral infections, fever of any genesis; exacerbation of concomitant chronic diseases; blood diseases—thrombocytopenia, anemia (hemoglobin less than 90 g/L); skin diseases that prevent the study; stage 5 chronic kidney disease (glomerular filtration rate < 15 mL/min/1.73 m2 according to MDRD or CKD-EPI); taking hormone replacement therapy, oral contraceptives; regular use of steroids, nonsteroidal anti-inflammatory drugs (therapy with antiaggregants was not exclusion criterion).

Macro- and microhemodynamics in the vessels of the lower extremities were assessed in all patients. To verify the diagnosis and assess the macrohemodynamic status in the extremities, all patients underwent ultrasound duplex scanning and ABI measurements.

Perfusion in the foot tissues was assessed both using TcPO2 (as the reference standard for this study) and the IOFF method.

TcPO2 was measured by using the TCM4 (Radiometer, Copenhagen, Denmark). The skin was first wiped with an alcohol solution. The sticky fixation ring was fixed at the dorsum of the foot between the first and second metatarsal heads just proximal to the first and second toes. After a few droplets of an electrolyte solution enhancing contact between the electrode and the skin, the electrode was fixed in the fixation ring (Figure 1a). Registration of TcPO2 was carried out when they stabilized after 15–20 min of local heating. A probe temperature of 44 ◦C was selected. The two feet were measured in sequence.

#### *2.2. Perfusion Measurement Using the IOFF Method*

The foot tissue perfusion measurement using the IOFF method was performed using prototypes of a new diagnostic device developed by the joint-stock company "Elatma Instrument-Making Enterprise" (Ryazan, Russia).

**Figure 1.** (**a**) TcpO2 measurement procedure; (**b**) procedure for measuring perfusion using the IOFF method.

The IOFF method is based on the analysis of low-frequency fluctuations (0–10 Hz) of optical signals backscattered by tissues, initially emitted by the incoherent source—a lightemitting diode (LED). The prototype device uses three LED emission sources operating in the wavelength range of 560–580 nm and one silicon photodiode in the optical sensor. The perfusion value calculated during signal processing is similar to that in laser Doppler flowmetry (LDF) [15,19]. The method allows the assessment of perfusion measured in perfusion units (PU).

The sensor was placed on the dorsal surface of the foot at the first intermetatarsal space (the same location as for the TcPO2 electrode) (Figure 1b). The measurement was carried out sequentially, first on the left leg, then on the right one. A total of 54 measurements were taken in 27 patients.

Perfusion was recorded during a local heating test. The time duration of the test was 6 min. Using heating plates integrated into the perfusion sensor, a thermoneutral temperature of 32 ± 0.5 ◦C was maintained for the first 60 s, and the baseline perfusion level (BP) was assessed. The heating plates were then heated at a rate of 1.5 ◦C/s to a temperature of 42 ± 0.5 ◦C. This plate temperature was maintained until the end of the measurement (for 5 min). An example of the perfusion curve obtained during the heating test is shown in Figure 2.

The following parameters were assessed:

Baseline perfusion (BP) was calculated as the median baseline perfusion level for the first 60 s of measurement.

Local thermal hyperemia 1–5 min (LTH 1–5 min) was calculated as the median perfusion for each minute of heating.

Before the IOFF and TcPO2 measurements, all patients were at rest for 10 min in the supine position, relaxed, head and heels supported, in a room with a comfortable temperature. All patients were asked to refrain from smoking within 3 h prior to the examinations. Thus, the IOFF perfusion measurement and the TcPO2 measurement were performed at the same location and under the same conditions (patient positioning and preparation). Because the IOFF perfusion measurement and the TcPO2 measurement are accompanied by a heat test, measurements were taken on different days.

**Figure 2.** Example of a perfusion curve obtained by IOFF during the local heating test.

#### *2.3. Assessment of Macrohemodynamic Parameters*

The ABI of each foot was calculated by dividing the higher pressure in the posterior tibial or dorsalis pedis arteries by the higher systolic blood pressure in the right or left arm. To record blood pressure, a Doppler probe was placed over the pulsing artery at a 45◦ to 60◦ angle to the surface of the skin [20].

The level and severity of arterial stenosis were assessed by the duplex ultrasound scanning of the lower limb arteries using the Philips Affinity 50 (Philips Ultrasound, Tampa, Florida, USA) and the Vivid 7 Dimension (GE Healthcare, Chicago, Illinois, USA). The ultrasound protocol involved assessing the presence of hemodynamically significant stenoses (at least 50% of the diameter reduction) in 6 main arteries of each lower limb (common femoral artery, deep femoral artery, superficial femoral artery, popliteal artery, anterior tibial artery, posterior tibial artery).

#### *2.4. Statistical Analysis*

Statistical analysis of the data was carried out using the RStudio 2021.09.0 Build 351 program using the R version 4.1.1 language. Medians and quartiles (Me (LQ; UQ)) were calculated for quantitative variables. Absolute (n) and relative (%) frequencies were used for qualitative variables. Spearman's correlation coefficient was used to assess the correlation between quantitative variables. Thresholds for quantitative variables were estimated using ROC analysis (pROC 1.18.0 package). The required sample size was calculated using the "power.roc.test()" function. The level of the type I error (α) was set equal to 0.05; null hypotheses were rejected at *p* < 0.05.

The required sample size was calculated using the power.roc.test() function from the pROC package. The power of the study was set at 90%. The expected area under the ROC curve for identifying limbs with low TcPO2 (<20 mmHg) was 0.8. The ratio of the number of limbs without TcPO2 reduction to the number of limbs with TcPO2 reduction was set at 2 to 1 (κ = 2). Thus, at least 39 observations had to be recruited into the study in order to achieve a 90% power level under these conditions. Since the different limbs of the patients were analyzed as independent cases, at least 20 patients had to be included in the study.

#### **3. Results**

#### *3.1. Study Population and Baseline Characteristics*

A total of 27 patients were included in the study. The study was carried out on 54 feet. The characteristics of the group are listed in Table 1.


**Table 1.** Characteristics of the studied patients.

ABI—ankle-brachial index; eGFR using CKD-EPI—Estimated Glomerular Filtration Rate according to Chronic Kidney Disease Epidemiology Collaboration; HbA1—glycated hemoglobin; LQ—lower quartile; Me—median; UQ—upper quartile; MNSI—The Michigan Neuropathy Screening Instrument. \* ABI was not detected on 11 limbs due to the absence of a pulse in the arteries of the foot or due to severe pain syndrome.

#### *3.2. Comparison of IOFF and TcPO2 Measurement Results*

It was revealed that all perfusion parameters analyzed by the IOFF method correlated significantly with the TcPO2 measurement with high correlation coefficients (Table 2).

**Table 2.** Correlation between the parameters of the IOFF signal and value of the TcPO2 assessed on 54 measurements (27 patients). The table shows the correlation coefficients.


BP—baseline perfusion; LTH 1–5 min—local thermal hyperemia for each minute of heating; *p*—statistical significance.

In order to assess this phenomenon in detail, all measurements were divided into three subgroups: subgroup 1—TcPO2 < 20 mmHg (n = 14); subgroup 2—TcPO2 20–39 mmHg (n = 17); subgroup 3—TcPO2 ≥ 40 mmHg (n = 23). These TcPO2 limits have been used because, according to a number of studies, ulcer healing and limb prognosis are generally poor if TcPO2 is <20 mmHg and are generally good if >40 mmHg [7].

The perfusion parameters estimated by the IOFF method were analyzed in each of the subgroups separately and then compared (Figure 3).

**Figure 3.** Median and interquartile ranges of perfusion measured by IOFF at different time points of the local heating test in legs with different TcPO2 levels (\*—*p* < 0.05, \*\*—*p* < 0.01, \*\*\*\*—*p* < 0.0001).

Baseline perfusion and local thermal hyperemia measured by the IOFF method differed significantly in all three subgroups. Limbs with critically low transcutaneous oxygen tension (TcPO2 < 20 mmHg) had a significantly lower baseline perfusion than those in the subgroups 2 and 3. In the first subgroup, the increase in perfusion in response to heating was also significantly less pronounced than in the limbs with higher TcPO2 values. In this subgroup, there was low variability in all analyzed perfusion parameters. In cases where the TcPO2 value was greater than 40 mmHg, the levels of BP and LTH were significantly higher than in the subgroups 1 and 2. The hyperemic response peaked at 3 and 4 min of heating.

Receiver operating characteristic (ROC) curve analyses were performed to assess the ability to identify patients with a critical decrease in TcPO2 (<20 mmHg) based on the IOFF perfusion measurements (Table 3).


**Table 3.** Area under the ROC curve showing the diagnostic potential of IOFF and ABI for identification legs with a critical decrease in TcPO2 (<20 mmHg).

ABI—ankle-brachial index; AUC—area under the ROC curve; LCL—lower 95% confidence limit; UCL—upper 95% confidence limit; LTH 1–5 min—local thermal hyperemia for each minute of heating; BP—baseline perfusion.

All analyzed perfusion parameters from the IOFF measurement had a high diagnostic potential in identifying patients with critically low TcPO2. The area under the ROC curve for all IOFF parameters, including BP, was higher than that for ABI.

The area under the curve (AUC) of the LTH, 2 min was 0.943 (95% CI 0.874–1) and had an optimal cutoff value for the identification of limbs with critically low TcPO2 (0.989 PU), with a sensitivity of 85.7% and a specificity of 90% according to the ROC analysis (Figure 4).

**Figure 4.** ROC curve showing the sensitivity and specificity of LTH, 2 min for identification of legs with a critical decrease in TcPO2 (<20 mmHg).

#### **4. Discussion**

This study was the first to investigate the informative value of the IOFF technique in assessing a foot tissue perfusion in patients with diabetes-related lower-extremity complications. High correlation coefficients of all perfusion parameters measured by IOFF with TcPO2 (Rs 0.7 to 0.76) were demonstrated. The study showed that the IOFF method allows, with a moderately high sensitivity of 85.7% and a specificity of 90.0%, the identification of patients with a critical decrease in TcPO2 < 20 mmHg. Such a decrease in TcPO2 is known to indicate a poor prognosis for ulcer healing and limb preservation, so identifying patients in this group is of a great clinical importance [7,21].

Current assessment standards in lower extremity artery disease focus on macrovascular function with less emphasis on foot tissue perfusion measurements. However, measurement of foot perfusion is extremely important because macro- and microvascular disorders are not always congruent [22]. A variety of noninvasive diagnostic technologies have been proposed as promising methods for assessing foot tissue perfusion (LDF, LDI, Laser speckle contrast imaging, 2D perfusion imaging, fluorescence angiography, near-infrared spectroscopy, cone-beam computed tomography, etc.) [6,22–26]. However, most of these methods have not yet become widespread in clinical practice [6,25].

The ideal method for the assessment of a foot tissue perfusion should be inexpensive, readily available, and reproducible, and its results should help the clinician to predict wound healing and provide information that influences patient management [6]. The development of such new methods is complicated by the lack of a widely applicable, highly sensitive method that could be called the "gold standard" for studying perfusion and microcirculation in tissues. In this study, TcPO2 measurement was used as a reference method. TcPO2 method has been shown to be highly informative in predicting the ulcer healing and foot amputation, but the method is limited in application due to the cost of equipment, consumables, and duration of the measurements [6,12,13]. Thus, the development of new, widely available methods of assessing foot tissue perfusion can significantly improve the quality of management of patients with diabetes-related lower-extremity complications.

A number of studies have evaluated the applicability of LDF and laser Doppler imaging as a noninvasive optical method for assessing a tissue perfusion [6,27–30]. However, these tools have not found widespread clinical application due to the high frequency of

artefacts, sensitivity to ambient temperature, poor reproducibility, and high operator dependence. In addition, the disadvantage of the LDF method is the very small volume of probed tissue (1 mm3 or smaller) [9,31]. Previously, LDF has been shown to be less predictive of ulcer healing and forefoot amputation than the TcPO2 testing [28].

The new IOFF method makes it possible to analyze the perfusion index similar to that calculated by the LDF method. The use of LEDs makes it possible to significantly reduce the cost of the sensors and device. LEDs also eliminate the need for lasers and fiber optics. This reduces the impact of wire positioning on instrument performance. The penetration depth at IOFF is 2–3 mm, which is slightly greater than in LDF. In addition, because of the probe design, the signal is analyzed from a larger area of skin (~25 mm2). Due to this, the signal backscattered from the tissue is collected from a larger tissue volume than in LDF and includes deeper vascular plexuses and larger vessels [14,15]. This also reduces the impact of local vascular network heterogeneity on measurement variability.

The assessment of perfusion by IOFF and TcPO2 are similar. Both procedures involve the local heating test, and the measurement results are affected by a tissue blood supply. However, perfusion assessment using the IOFF method is significantly less time consuming than measuring TcPO2 (6 min vs. 20 min). In addition, the IOFF method does not require expensive consumables, and the use of the light-emitting diodes makes the technology easy to implement and potentially widely available. Expected cost of IOFF devices would be about USD 2000. This low estimated cost is an important potential advantage of the IOFF method over other methods of perfusion assessment (such as LDF, TcPO2, 2D perfusion imaging, cone-beam computed tomography, etc.)

Thus, the new method IOFF may be promising as an informative, available and convenient way to assess foot tissue perfusion in patients with diabetes-related lowerextremity complications.

Additionally in this study, the IOFF method was compared with the results of measuring ABI. The area under the ROC curve for all IOFF parameters was higher than that for ABI for identification legs with a critical ischemia. However, it is known that the usefulness of the ABI is limited in people with DM because of medial arterial calcification [11]. Thus, in this population, the toe–brachial index (TBI) may be more informative [32]. However, this index is not applicable for patients with an amputated big toe. Though TBI was not measured in our study, it will be of great interest to compare the results of TBI and IOFF assessment in further studies.

This first study of the IOFF technique foreshadows further necessary longitudinal research, which should focus on endpoint analysis and the derivation of specific perfusion thresholds for the probability of wound healing. For a more accurate assessment of the applicability of the method IOFF in practical healthcare, further study of reproducibility, testing of the methodology on large sample sizes, and evaluation of the predictive ability of the method is required. It is also of great interest to explore the association between flow and perfusion [33] and to assess the diagnostic informativeness of the method as a screening tool for the detection of PAD in both diabetic and nondiabetic patients.

#### **5. Limitations**

In this study, we used strict inclusion/exclusion criteria. This reduced possible measurement bias (e.g., associated with heart rhythm disorders or anemia) but may affect the reproducibility of the result in "real-world practice".

It is known that the duplex ultrasound scanning is an operator-dependent procedure. The study was conducted at two independent centers, so the assessment of the duplex ultrasound scanning was performed by several experts, which could affect the accuracy.

#### **6. Conclusions**

The results of the pilot study demonstrated a high correlation between the perfusion parameters assessed by IOFF and TcPO2. A sensitivity of 85.7% and specificity of 90.0% in identifying patients with critically decreased TcPO2 < 20 mmHg suggests that the IOFF

technique may be promising as an informative, rapid, and noninvasive method of assessing tissue perfusion.

**Author Contributions:** Conceptualization, P.G., A.G., D.K. and D.R.; methodology, P.G., A.G., D.K., Y.K. (Yulia Kovaleva), A.B., Y.K. (Yulia Kononova) and E.K.; software, A.G.; validation, A.G., R.L., T.B. and N.M.; formal analysis, A.G.; investigation, P.G., A.G., S.Z., Y.K. (Yulia Kovaleva), A.B., Y.K. (Yulia Kononova), E.K. and N.M.; resources, D.K., S.Z., A.B., T.B., R.L. and D.R.; data curation, P.G., A.G., S.Z., Y.K. (Yulia Kovaleva), A.B., Y.K. (Yulia Kononova) and N.M.; writing—original draft preparation, P.G.; writing—review and editing, A.G., D.K., S.Z., Y.K. (Yulia Kovaleva), A.B., Y.K. (Yulia Kononova), E.K., T.B., N.M., R.L. and D.R.; visualization, A.G.; supervision, D.K., A.B. and D.R.; project administration, D.K., A.B. and D.R.; funding acquisition, D.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** The recruitment and examination of patients was funded by JSC "Elatma Instrument-Making Enterprise" (Ryazan, Russia). Data analysis and evaluation was carried out by researchers without funding from JSC "Elatma Instrument-Making Enterprise" within the framework of the research project "New Approaches to the Comprehensive Assessment of Peripheral Hemodynamic Parameters in the Management of Patients with Diseases of Various Etiologies", funded by the State Budget of the Moscow Region. The funding source had no role in this manuscript.

**Institutional Review Board Statement:** The study was conducted in accordance with the Declaration of Helsinki. The trial protocol was approved by Research Ethics Committees at participating institutions: Moscow Regional Research and Clinical Institute independent ethics committee (Protocol No. 13 dated 7 November 2019) and Almazov National Medical Research Centre ethics committee (No. 27112019, meeting No.11-19 dated 11 November 2019). The photograph of the subjects' legs was accompanied by the written permission of the subjects.

**Informed Consent Statement:** Written informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data that support the findings of this study are available from the corresponding author (Glazkova P.) upon reasonable request.

**Conflicts of Interest:** The authors declared no potential conflict of interest with respect to the research, authorship, and/or publication of this article.

#### **References**


## *Article* **Biochemical and Mechanical Analysis of Occlusal and Proximal Carious Lesions**

**Sahar Al-Shareefi 1, Ali Addie <sup>2</sup> and Lamis Al-Taee 1,\***


**Abstract:** A precise evaluation of caries excavation endpoint is essential in clinical and laboratory investigations. Caries invasion differentiates dentin into structurally altered layers. This study assessed these changes using Raman spectroscopy and Vickers microhardness. Ten permanent molars with occlusal and proximal carious lesions were assessed and compared at 130 points utilizing four Raman spectroscopic peaks: phosphate v1 at 960 cm<sup>−</sup>1, amide I (1650 cm−1), amide III (1235 cm−1) and the C-H bond of the pyrrolidine ring (1450 cm−1). The phosphate-to-amide I peak ratio and collagen integrity peak ratio (amide III: C-H bond) of carious zones were calculated and compared in both lesions. The former ratio was correlated to 130 Vickers microhardness indentations through lesions. The caries-infected dentin (CID) exhibited low phosphate peak, but higher amide I, III and C-H bond peaks than other zones in both lesions. The peaks in amide regions (I and III) varied in occlusal versus proximal lesions. A high correlation was found between mineral: matrix peak ratio and equivalent microhardness number within carious lesions, while the collagen integrity peak ratio was applied in proximal lesions only. Raman spectroscopy detected changes in the mineral and matrix contents within different carious zones and regions.

**Keywords:** caries-infected dentin; caries-affected dentin; collagen integrity; Raman microscopy; Vickers microhardness

#### **1. Introduction**

Dental caries is the most common multifaceted disease with a significant societal impact that spreads worldwide [1]. The occlusal and proximal tooth surfaces are the most susceptible sites for demineralization from the bacterial acidic by-products [2]. Although the predominance of occlusal caries was higher [3], in which the bacterial accumulations receive the best protection in the deepest part of the groove-fossa system, carious progression in smooth surfaces is also popular due to the difficulty of early detection using the standard visual-tactile evaluation [4]. However, the radiographs and fiber-optic transillumination can be beneficial, but they cannot always lead to a definitive diagnosis [5]. Carious dentin is classified clinically and histologically into two main zones, superficial infected (CID) and deeper affected layer (CAD), in which the delineation between them is critical in clinical and laboratory investigations. The use of minimally invasive caries management reduces unnecessary tissue removal and the risk of pulp exposure, whilst maximizing the reparative potential of the dentin-pulp complex [6]. This includes complete excavation of the superficial, bacterially contaminated and denatured dentin while preserving the remaining harder affected dentin that can be sealed with therapeutic restorations [7]. Clinically, the discrimination between the carious zones based on tissue hardness, moisture, color, fluorescence properties and dye stainability. Additionally, carious tissue removal methods can be used to determine the endpoint, such as self-limiting burs and chemomechanical removal agents. These methods are implemented in in vitro investigations but lack sufficient clinical validation [8]. The laboratory studies assessed the mechanical properties of the carious

**Citation:** Al-Shareefi, S.; Addie, A.; Al-Taee, L. Biochemical and Mechanical Analysis of Occlusal and Proximal Carious Lesions. *Diagnostics* **2022**, *12*, 2944. https://doi.org/ 10.3390/diagnostics12122944

Academic Editors: Daniel Fried and Viktor Dremin

Received: 28 September 2022 Accepted: 19 November 2022 Published: 25 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

dentin through relative changes in tissue hardness via Knoop or Vickers hardness test as a gold standard to delineate the excavation margins. However, this test is considered as an invasive, low-resolution test that damages tissues irreversibly preventing further tissue analysis [9]. Furthermore, hardness clarifies the mechanical integrity without any correlation to the biochemistry of the carious lesion. In contrast, the use non-invasive analytical technology for in vitro studies such as Raman spectroscopy will help in assessing the biochemical tissue changes during the caries process. It can be sufficiently sensitive to the differences in the mineral and organic compositions of healthy and carious dentin tissues through their specific molecular vibrational energy signatures [10]. Previous studies utilized FTIR and Raman spectroscopy to identify the spectral characteristics of mineral-tomatrix peak ratio, which represents the phosphate and amide I distributions through the carious dentin lesion [11–14]. The phosphate peak intensity is related to the mineral content, while the amide I peak intensity represents the organic component in dentin [15–18]. Thus, this correlation is regarded as a biochemical measure for dentin demineralization, in which the corresponding hardness values will help to identify the lesion characteristics in each zone. The evaluation of the collagen in carious dentin has a significant importance, since it affects the bond stability when bonded to adhesive resin restorations. The integrity of the collagen triple helix in type I collagen membrane obtained from bovine Achilles deep tendon was previously measured using attenuated total reflection Fourier transform infrared (ATR-FTIR) [19]. In this method, a peak ratio of amide III: C-H bond of the pyrrolidine ring (1235 cm<sup>−</sup>1:1450 cm−1) was used as an indicator for the collagen integrity if it is close to 1 [20]. However, this ratio was not tested on a natural dentin carious lesion using Raman microscopy, to validate using this ratio as an indicator for the collagen integrity in vitro investigations. The biochemical characterization of proximal carious lesions is scarce in the literature. The susceptibility of different tooth surfaces for caries development and progression is markedly differed in occlusal versus smooth surfaces [21], added to differences in the inflammatory pulp response in proximal lesions as compared to occlusal lesions [22]. This might necessitate an assessment of the structural changes in proximal carious lesions in comparison to occlusal in the same tooth model. Accordingly, this study evaluated and compared the changes in the mineral content and collagen matrix in different carious zones of occlusal vs. proximal carious lesions using Raman microscopy. The null hypotheses: (i) there were statistically no significant changes in the mineral and matrix contents of demineralized carious lesions (CID and CAD) from sound dentin in each lesion and between lesions, (ii) the collagen integrity ratio (Amide III: CH- bond of the pyrrolidine ring) cannot be applied for a natural carious dentin lesion, and (iii) there was no statistical correlation between Raman peak ratio of phosphate ν1:amide I ratio and the equivalent VHN in occlusal and proximal carious dentin lesions.

#### **2. Materials and Methods**

#### *2.1. Sample Preparation for Raman and Hardness Measurements*

Ten permanent, freshly extracted human 1st molars were collected using an ethics protocol approved by the health research committee (Ref No. 285521, 31 March 2021). The teeth showed two active carious dentin lesions; occlusal and proximal at the contact area and above the cementoenamel junction. The lesions showed score 4 following the international caries detection and assessment system (ICDAS) [23], in which the lesion extended halfway through the dentin without pulp exposure, then were stored in distilled water in a cold cabinet (+4 ◦C). Samples were hemi-sectioned longitudinally (Isomet 1000, Buehler, Lake Bluff, IL, USA) using a water-cooled diamond blade (330-CA/RS-70300, Struers, Detroit Rd. Westlake, LLC, Cleveland, OH, USA), and then embedded in epoxy resin molds. The surfaces were polished under running water using a polishing machine (Laryee Technology, China), and silicon waterproof papers in a sequential pattern (P1200 for 10 s, P2500 for 10 s, and P4000 for 4 min) [13] to gain flat and smooth surfaces for accurate measurements. A small steel round rotary bur was used to create a reference dot at the enamel–dentin junction occlusally and proximally. These dots are visible under Raman

spectroscopy and Vickers microhardness with a 500 μm distance between the examined points, as shown in Figure 1.

**Figure 1.** The prepared lesion hemi-section was firstly analyzed using confocal Raman microscopy followed by Vickers microhardness indenter across the lesion. The assessment started from the enamel–dentin junction through the lesions towards the pulp in a straight path occlusally (a) and an oblique path proximally (b). A control measurement was taken in an area of clinically sound dentin (c) in the same sample. Six measurement areas were taken in each lesion with a 500 μm distance between the examined points.

#### *2.2. Raman Spectroscopy*

A high-resolution confocal Raman microscope (Senterra, Bruker Optics, Ettlingen, Germany) operating in line scan mode was used to scan the carious lesions and sound dentin. A total of 130-point scans were made over twenty carious lesions (*n* = 13 per tooth). The distance between the scanned points was 500 μm, and it was controlled using a programmable sample stage with a 1μm resolution. Spectra acquisition was performed using a 785 nm near-infrared diode laser and a 400 line/mm diffraction grating. An Olympus 20X/0.40 NA objective lens was used to focus the laser on the sample surface with a spot size of about 5 μm. The sectioned tooth was mounted on the sample stage using plastic molding putty, and the Raman spectra for carious lesions and sound dentin were measured over the range of 200–3600 cm−<sup>1</sup> with 100 mW of laser power on each point. The integration time for each spectrum was typically 30 s, with three accumulations. Spectrum acquisition was conducted using the image stitching technique to collect a large area of the mounted tooth (Figure 1). Baseline correction was performed by Raman processing software (OPUS, Bruker Optics, Germany). After acquisition and spectra processing, four Raman spectroscopic peaks were identified. The phosphate peak intensity v1 vibration at 960 cm−1, amide I peak intensity at 1650 cm−1, amide III at 1235 cm−<sup>1</sup> and C-H bond of the pyrrolidine ring at 1450 cm<sup>−</sup>1. Peak height intensities were calculated and averaged in both lesions (occlusal and proximal) and the control point (sound dentin). The inorganic to organic ratios of dentin components were assessed from the band intensities of phosphate V1 at (960 cm−1) to amide I (1650 cm−1). While the collagen integrity at each zone was assessed by calculating the absorbance ratio of 1235 cm−1:1450 cm−<sup>1</sup> (*n* = 2 per zone, then averaged).

#### *2.3. Vickers Microhardness*

The microhardness of demineralized dentin zones (caries-infected, caries-affected) and sound dentin was measured using the Vickers microhardness tester (TH714, Obsnap Instruments Sdn Bhd, Selangor, Malaysia). A square-based pyramid diamond-shaped indenter

was used, with a load of 300 gf for 15 s [24]. A total of 130 indentations (*n* = 13 per lesion) were made in the same straight path line that previously assessed by Raman microscopy. These indentations started from the enamel–dentin junction occlusally and proximally toward the pulp, with 500 μm interval between the examined points (Figure 1). Then, the Vickers hardness number was recorded automatically using the manufacturer's software.

#### *2.4. Statistical Analysis and Spectral Correlation Processing with Hardness Measurements*

The statistical analysis was performed using SPSS software version 25 (IBM, Chicago, IL, USA). Shapiro–Wilk test was used to evaluate the normality of data distribution. The data were statistically analyzed using one-way ANOVA followed by Tukey post hoc multiple comparisons (*p* > 0.05) regarding the intensities of four Raman peaks (A.U.) and Vickers microhardness number (VHN) in each zone per lesion. The independent t-test (Minitab 14, Minitab LLC, Chicago, IL, USA) was performed for pairwise comparisons between groups (α = 0.05) to assess the differences in these data between the occlusal and proximal lesions at each zone. Additionally, Pearson's correlation coefficient test was used to explore if there is a correlation between Raman mineral: matrix peak ratio and their equivalent microhardness values (VHN) at each point per lesion. The peak ratios were determined after baseline correction in single spectra by dividing the intensity of phosphate to the amide I (960 cm<sup>−</sup>1/1650 cm−1).

#### **3. Results**

#### *3.1. Biochemical Analysis of Sound and Demineralized Zones in Occlusal and Proximal Lesions*

The relative Raman band intensities (Mean ± SD) with statistical correlations in cariesinfected, caries-affected and sound dentin zones of the occlusal and proximal carious lesions are shown in Table 1. All dentin layers (sound and demineralized) showed the evidence of the four characteristic peaks; the phosphate peak intensity (symmetric P-O stretching mode, v1-PO) at 960 cm−<sup>1</sup> which represents the inorganic part of dentin, the amide I peak intensity at 1650 cm<sup>−</sup>1, amide III at 1235 cm−<sup>1</sup> and the C-H bond of the pyrrolidine ring at 1450 cm−<sup>1</sup> which refer to the organic part of sound and demineralized dentin. One-way ANOVA showed a statistically significant difference between sound and demineralized (lesion) dentin regarding all Raman peaks in both lesions (*p* = 0.000). Fur-ther analysis using Tukey post hoc multiple comparisons revealed that the phosphate band at 960 cm−<sup>1</sup> in both lesions was statistically the lowest in caries-infected dentin zone (CID, *p* = 0.000), but it was increased with increasing the mineral content in the caries-affected (CAD), and then sound dentin with non-significant difference between occlusal and proximal lesions in each zone (independent *t*-test, *p* > 0.05), Table 1.

In both carious lesions, the intensities of amide I, III and C-H bond of pyrrolidine ring peaks (1650, 1235 and 1450 cm−1, respectively) were statistically higher in CID, with no significant difference between CAD and sound dentin (*p* > 0.05), except the C-H bond of pyrrolidine ring peak in the proximal carious lesions which was higher in CAD than in sound dentin (*p* = 0.023), Table 1, Figures 2 and 3.

By comparing the peak intensity between the occlusal and proximal carious lesions, it was found that there were statistically no significant differences in the intensity of the phosphate peak (960 cm−1) between both lesions (*p* > 0.05) at each zone. For the Amide I (1650 cm−1), the peak intensity in the caries-infected zone was higher in the occlusal lesion as compared to the proximal (*p* = 0.014), while there was statistically no significant difference in CAD and sound dentin zones between lesions (*p* > 0.05). The intensity of amide III peak (1235 cm<sup>−</sup>1) was significantly higher in the occlusal carious lesion in all zones (CID, CAD and sound dentin) in comparison to proximal carious lesion (*p* < 0.05). However, the intensity of C-H bond of the pyrrolidine ring peak at 1450 cm−<sup>1</sup> was comparable between the occlusal and proximal lesions at each zone (*p* > 0.05).


**Table 1.** Raman band intensities A.U. (Mean ± SD) of caries-infected, caries-affected and sound dentin in occlusal and proximal carious lesions.

(\*) significant difference between sound and demineralized dentin (caries-infected or caries-affected), one-way ANOVA test and Tukey post hoc tests (alpha level of 0.05). (**ˆ**) significant difference between occlusal and proximal carious lesion (Independent *t*-test, *p* < 0.05). Similar letters in rows indicate no significant differences between lesions (*p* > 0.05).

**Figure 2.** Representative Raman spectra of the caries-infected (CID), caries-affected (CAD) and sound dentin in the occlusal carious lesion. The spectra were normalized based on the peak intensities of v1-PO at 960 cm−1, which is the strongest signal among all Raman spectra, it is the highest in sound dentin. Amide I at 1650 cm<sup>−</sup>1, Amide III at 1235 cm−<sup>1</sup> and C-H bond of the pyrrolidine ring at 1450 cm−<sup>1</sup> are the highest in CID. The inserted figure shows the magnified, same spectra in the range from 1000 to 1700 cm<sup>−</sup>1.

**Figure 3.** Representative Raman spectra of the caries-infected (CID), caries-affected (CAD) and sound dentin in the proximal carious lesion. The spectra were normalized based on the peak intensities of v1-PO at 960 cm−1, which is the strongest signal among all Raman spectra, it is the highest in sound dentin. Amide I at 1650 cm<sup>−</sup>1, Amide III at 1235 cm−<sup>1</sup> and C-H bond of the pyrrolidine ring at 1450 cm−<sup>1</sup> are higher in CID and CAD than sound dentin. The inserted figure shows the magnified, same spectra in the range from 1000 to 1700 cm<sup>−</sup>1.

To analyze the integrity of the collagen triple helix, a mean absorbance ratio of amide III bands at 1235 cm−<sup>1</sup> and the pyrrolidine ring at 1450 cm−<sup>1</sup> was calculated. In the proximal carious lesion, the mean ratio in the caries-infected dentin was 0.7 ± 0.2, while the mean ratio in CAD was 0.9 ± 0.2 that was close to sound dentin (1.0 ± 0.3). This demonstrates a lack of collagen denaturation in the CAD zone. However, this ratio was not applied in the occlusal carious lesion that showed a very high absorbance band of the amide III bands at 1235 cm−<sup>1</sup> (amide III) than that of the pyrrolidine ring at 1450 cm<sup>−</sup>1, Figure 2.

#### *3.2. Vickers Microhardness*

The change in the mineral content of the carious lesions was also measured via a mi-crohardness test. One-way ANOVA revealed a statistically significant difference among different dentin layers (sound vs. demineralized, *p* = 0.000) with no significant difference found between occlusal and proximal lesions at each zone (independent *t*-test, *p* > 0.05). Further analysis by Tukey post hoc multiple comparisons test showed that there was a marked reduction in VHN in CID at occlusal and proximal lesions (20.2, 22.6, respectively), which was significantly higher in CAD (34.4, 32.5, respectively, *p* = 0.000), followed by sound dentin (54.1, 53.2, respectively, *p* = 0.000), as shown in Figure 4.

**Figure 4.** Means of Vickers microhardness number (VHN) of the sound and demineralized dentin (caries-infected and caries-affected) in occlusal and proximal lesions. (\*) Statistically significant difference of demineralized dentin from sound (*p* < 0.001). Similar litters mean statistically nonsignificant differences (*p* > 0.05) between different dentin zones in occlusal vs. proximal lesions.

#### *3.3. Raman Spectral Correlation with Vickers Microhardness (VHN)*

The peak ratios of phosphate ν1: amide I of the selected points in each carious lesion (occlusal and proximal) were calculated and plotted against Vickers hardness numbers (VHN) as a scatter diagram, as shown in Figure 5. The coefficient of determination was calculated across the assessed points in the occlusal and proximal lesions, in which the R2 = 0.90 and 0.94, respectively, (*p* = 0.000). The statistically significant high correlation between peak ratio and VHN in both lesions indicates that they showed a logarithmic regression relationship which enable the calculation of the tissue hardness when the peak ratio was measured.

**Figure 5.** A scatter plot and a regression line (R) demonstrating the presence of a logarithmic regression relationship (*p* < 0.000) between the Raman peak ratios (phosphate v1: amide I) and Vickers microhardness (VHN) in the sound and demineralized dentin (CID and CAD) in the occlusal lesion (**A**) and proximal lesion (**B**). The peak ratio is lower in CID with reduced mineral contents which gradually increased towards sound dentin (more phosphate v1).

#### **4. Discussion**

The improved understanding of the caries process and biology of the dentin-pulp defense and the regenerative responses encouraged the application of minimally invasive caries removal rather than the traditional surgical excavation approach. This approach relies on accurate caries diagnosis, then identifying the excavation endpoint to exclude the irreversibly caries-infected dentin while preserving the remineralizable caries-affected dentin to enhance the long-term survival of the dentin-pulp complexes. Caries invasion leads to the differentiation of dentin into zones with altered composition, collagen integrity and mineral identity. However, the understanding of these changes from the fundamental perspective of molecular structure is limited. Accordingly, this study provided a map of the biochemical changes through the different carious dentin zones in two lesion models (occlusal and proximal) utilizing Raman spectroscopy to extract the molecular information of each zone regarding the hydroxyapatite's structural changes and collagen denaturation as the dentin transition from the superficial caries-infected zone (CID) into sound dentin. The integrity of collagen's triple helical structure was also evaluated based on spectra collected from demineralized dentin (carious lesions) of the selected teeth. The results support the argument that there are statistically significant changes (*p* < 0.000) in the biochemical components across the carious lesions from the superficial layers towards sound dentin.

From a biochemical perspective, the mineral content was detected via the phosphate peak (PO4 <sup>−</sup><sup>3</sup> *ν*1) at 960 cm−1, which was the strongest signal among all Raman spectra. This peak is referred to the degree of demineralization in natural carious enamel and dentin [25,26]. In accordance with Almahdy et al. (2012) and El-Sharkawy (2019) [18,26], the mineral content was dropped significantly (*p* = 0.000) in CID, but higher in CAD towards sound within the same sample with no significant differences in the intensities between the occlusal and proximal carious lesions (*p* > 0.05) at each zone. The presence of amorphous Ca/P provides a local ion-rich environment, which is favorable for in situ generation of prenucleation clusters, succeeding further dentin remineralization [27]. Accordingly, the first stated hypothesis was partially rejected as the mineral content represented by the phosphate peak intensity differed among different dentin zones (sound and demineralized) but was comparable between occlusal and proximal lesions.

Amide I peak (1650 cm−1) is the most prominent organic component of dentin, predominantly type I collagen [28]. This band is assigned to Y8a tyrosine side chain of solution-phase collagen representing the secondary structure of proteins [29]. It was used in many studies to detect changes in the molecular structure of collagen [16,30]. The amide I content was higher in CID in both lesions, and decreased in CAD and sound dentin, as shown in Figures 2 and 3. It was also higher in occlusal than proximal carious lesion. This might be attributed to the higher proteins in the infected layer than sound dentin [31]. The intensity of amide I is correlated to the non-reducible cross-links in collagen [32], which means that there is an apparent change in the molecular structure of collagen in the superficial CID layer in both lesions. This alteration in collagen is correlated to the presence of esters in the carious tissues [17] derived from the bacterial lipid components. This will promote esterification of the carboxylic side-chains of aspartate and glutamate residues catalyzed by the acidic environment (lactic acid) [33]. The closer proximity in the intensity of this band between CAD and sound dentin in both lesions might indicate the importance of preserving this layer since the organic matrix in dentin regulates the growth and maturation of apatite crystals and thus the mineralization process.

The organic component of carious dentin has a major role in the progression of carious development [16,34]. It represents the integral component of the mineralized tissue, but after demineralization, they become exposed and altered structurally [17]. The demineralization process in carious dentin can be assessed by the difference in the mineral/protein band ratio of sound and carious dentin. The band ratio of amide I at ∼1650 cm<sup>−</sup>1, which is the most prominent organic moiety in dentin, and the phosphate ion at ∼960 cm−<sup>1</sup> was used in spectroscopic studies [13,14,17,35] to analyze the distribution of mineral content in the dental tissues. Using this band ratio, the Raman peaks graphs showed the heterogeneous nature of caries dentin in terms of mineral distribution with lower values for carious-infected dentin compared to CAD and sound dentin. This band ratio increased by increasing the mineral content in the caries-affected zone followed by sound dentin in both lesions, as shown in Figure 5.

The integrity of the collagen triple helix was evaluated by assessing the ratio of the absorbance bands of amide III at ∼1235 cm−<sup>1</sup> and CH2 scissoring at ∼1450 cm−<sup>1</sup> (stereochemistry of the pyrrolidine rings). The former peak is sensitive to the presence of secondary structure of collagen, while the latter is independent of the ordered structure of collagen [20]. The ratio indicates collagen with an integral triple helical structure. If the value is close to 1, but if the ratio <0.8, it means that there is a breakdown of the triple helix in this zone, such as gelatin [19,20]. The mean ratio of amide III: CH2 (1235 cm<sup>−</sup>1:1450 cm−1) was 0.7 in CID in the proximal lesion, 0.9 in CAD that was close to sound dentin (1.0) which indicates the integrity of collagen fibers that is essential for the remineralization processes. This supported that the infected layer with comprised denatured collagen, loses the potential for remineralization and must be removed. Conversely, the affected layer, that is partially demineralized and remineralizable with collagen fibrils retaining their natural structure around intact dentinal tubules, is to be preserved to maximize the reparative potential and reduce the risk of pulp exposure [36]. However, this ratio was not achieved in the occlusal carious lesion model. This is due to the higher absorbance of amide III than CH2-CH- bands in the CID zone, as shown in Figure 2, with statistically no significant difference between CAD and sound dentin (*p* > 0.05). This is referred to as the increased aliphatic side-groups of various amino acid residues in the infected tissues in these lesions which promote the demineralization process [37]. The difference between the occlusal and proximal lesions indicate that this ratio cannot be applied for all carious studies due to the variation in collagen denaturation in various tissues and regions. This is due to the variety in depth, degree of organic matrix destruction and loss of mineral content which affect tissue response to different therapeutic materials in an attempt to repair and remineralize the damaged tissues. Accordingly, the second stated hypothesis was partially accepted.

The caries-infected dentin showed higher accentuated amide I and III peaks in occlusal lesions in comparison to proximal carious lesions, which might be due to the higher aliphatic content. This is also might be attributed to the fact that the dentin subjacent to proximal enamel caries forms mineral crystals obliterating the main transport pathways (dentinal tubules), creating sclerotic dentin close to the enamel–dentin junction even before the demineralization reaches the enamel–dentin junction [38]. However, a histological study [39] reported a higher frequency of deep dentin demineralization (>50%) in proximal lesions than occlusal lesions under microradiography with contrast (MRC). This might suggest the presence of facilitated transport pathways for dentinal fluid in dentin underlying the proximal enamel lesion, since the tubules in sclerotic dentin are partially obliterated [40]. This induces a facilitated transport of dentinal fluid into the pores of carious enamel mixed with enamel fluid, and then with bacterial biofilm fluid, and thus promoting the carious progression in proximal lesions.

The microhardness values remain the gold standard for characterizing sound and demineralized dentin tissues in most in vitro studies. It is correlated to the clinical mechanical excavation procedures, and also can predict the behavior of dentin/restoration interfaces, as the regional differences in tissue hardness can alter the distribution of stresses along the interface and thus affecting the preferential location of failures. Additionally, it represents a direct measure of the hydroxyapatite (HAp) crystals present in dental hard tissues. This study showed a gradual increase in tissue hardness from the superficial carious layer towards the deeper area of the carious lesions towards sound dentin (*p* = 0.000). The caries-infected dentin showed the lowest VHN values in both lesions <25, Figure 4, which is attributed to the greater dissolution of HAp compared to the deeper CAD layer (<40) that has the potential to be repaired when sealed with self-adhesive restorations. However, the carious dentin lesion in vivo progresses in a continuous wave rather than distinct zones with gradual transitions in histology and bacteriology from enamel to pulp, it is perhaps rather simplistic to consider the lesion as three distinct zones as described above, but clinically, this analysis has merit because it allows for more reliable, efficient and reproducible operative treatment [41].

Raman can potentially evaluate the carious dentin non-destructively, as an alternative to the invasive, low-resolution hardness tests. This study found statistically significant correlations (R<sup>2</sup> = 90, 94%; *p* = 0.000) between Raman peak ratio (phosphate v1: amide I) and Vickers microhardness number (VHN) in both lesions as shown in Figure 5. This Raman ratio was selected since the mineral content is considered as an indicator of the inherent physical properties of hard tissues. This correlation supports the use of this noninvasive high-resolution technique for chemical characterization in in vitro hard tissue studies rather than the physical tissue microhardness. This has important implications for the tissue pathogenesis and the minimally invasive treatment modalities. Consequently, the third stated hypothesis was rejected.

#### **5. Conclusions**

Based on the overall results of this study, the use of Raman spectroscopy and subsequent spectra analyses are highly useful for probing the molecular structure of carious dentin in various zones and regions. Raman detected structural changes in the inorganic and organic components of demineralized dentin in comparison to sound dentin in various regions. In which the mineral distribution was the lowest in the caries-infected dentin zone with no differences between occlusal and proximal lesions, whilst the collagen crosslink and triple helical structure were altered differently in carious zones and lesions. The presence of a statistical correlation between the Raman phosphate to amide I peak ratio with VHN supports the use of this non-invasive high-resolution technique for chemical characterization in in vitro studies. This will help further understanding of carious progression and assessing the remaining dentin tissues after different caries removal techniques following the minimally invasive approaches.

**Author Contributions:** S.A.-S.: methodology, investigation, data curation, writing—original draft preparation. A.A.: methodology, resources, analysis. L.A.-T.: conceptualization, methodology, writing—review and editing, supervision. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Teeth were collected using an ethics protocol approved by health research committee of Baghdad College of Dentistry (Ref No. 285521, 31 March 2021).

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** Not applicable.

**Acknowledgments:** This work was supported by Baghdad College of Dentistry, University of Baghdad, Ministry of Higher Education and Scientific Research/Iraq.

**Conflicts of Interest:** The authors declare that they have no known competing financial interest or personal relationships that could have appeared to influence the study reported in this paper.

#### **References**


## *Article* **Diagnostic Accuracy of Line-Field Confocal Optical Coherence Tomography for the Diagnosis of Skin Carcinomas**

**Elisa Cinotti 1,2,\*, Tullio Brunetti 1, Alessandra Cartocci 1, Linda Tognetti 1, Mariano Suppa 2,3, Josep Malvehy 4,5, Javiera Perez-Anker 4,5, Susanna Puig 4,5, Jean Luc Perrot <sup>6</sup> and Pietro Rubegni <sup>1</sup>**


**Abstract:** Line-field confocal optical coherence tomography (LC-OCT) is a new, noninvasive imaging technique for the diagnosis of skin cancers. A total of 243 benign (54%) and malignant (46%) skin lesions were consecutively enrolled from 27 August 2020, to 6 October 2021 at the Dermatology Department of the University Hospital of Siena, Italy. Dermoscopic- and LC-OCT-based diagnoses were given by an expert dermatologist and compared with the ground truth. Considering all types of malignant skin tumours (79 basal cell carcinomas (BCCs), 22 squamous cell carcinomas, and 10 melanomas), a statistically significant increase (*p* = 0.013) in specificity was observed from dermoscopy (0.73, CI 0.64–0.81) to LC-OCT (0.87, CI 0.79–0.93) while sensitivity was the same with the two imaging techniques (0.95 CI 0.89–0.98 for dermoscopy and 0.95 CI 0.90–0.99 for LC-OCT). The increase in specificity was mainly driven by the ability of LC-OCT to differentiate BCCs from other diagnoses. In conclusion, our real-life study showed that LC-OCT can play an important role in helping the noninvasive diagnosis of malignant skin neoplasms and especially of BCCs. LC-OCT could be positioned after the dermoscopic examination, to spare useless biopsy of benign lesions without decreasing sensitivity.

**Keywords:** optical coherence tomography; tumor; basal cell carcinoma; imaging; squamous cell carcinoma

#### **1. Introduction**

Line-field confocal optical coherence tomography (LC-OCT) is a new, noninvasive skin imaging technique that combines the advantages of optical coherence tomography (OCT) and reflectance confocal microscopy (RCM) in terms of spatial resolution, penetration, and image orientation, overcoming their respective limits [1–5]. LC-OCT has a higher resolution than OCT (~1 μm) [6–8] and higher penetration [9] depth than RCM (~500 μm), and it creates both vertical and horizontal images in real time [10–13].

Recently, LC-OCT [14] has been gaining attention because it has been shown that this device can help the clinical diagnosis of different neoplastic [15–18], inflammatory [19–23], and infectious [24,25] skin diseases. In particular, it has proven to be very effective in identifying basal cell carcinoma (BCC) [26], even managing to differentiate its histological subtypes [27] and to follow-up noninvasive treatment [28]. Furthermore, this noninvasive diagnostic technique can be used to help the differentiation of actinic keratosis (AK) from squamous cell carcinoma (SCC) [29–31] and to monitor the field of cancerization-directed treatments [32].

**Citation:** Cinotti, E.; Brunetti, T.; Cartocci, A.; Tognetti, L.; Suppa, M.; Malvehy, J.; Perez-Anker, J.; Puig, S.; Perrot, J.L.; Rubegni, P. Diagnostic Accuracy of Line-Field Confocal Optical Coherence Tomography for the Diagnosis of Skin Carcinomas. *Diagnostics* **2023**, *13*, 361. https:// doi.org/10.3390/diagnostics 13030361

Academic Editor: Viktor Dremin

Received: 23 December 2022 Revised: 9 January 2023 Accepted: 12 January 2023 Published: 18 January 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Many descriptive studies have shown the relevance of this imaging device for the diagnosis of cutaneous tumours, and our study aimed to evaluate the sensitivity and specificity of LC-OCT compared to dermoscopy for the diagnosis of skin tumours in a real-life setting in a third-level dermatology department.

#### **2. Materials and Methods**

#### *2.1. Study Design*

Prospective observational, monocentric study.

#### *2.2. Setting*

Patients were enrolled from the 27 August 2020 to the 6 October 2021 at the Dermatology Department of the University Hospital of Siena, Italy, from the melanoma prevention outpatient ambulatory. The study was conducted according to the criteria set by the Declaration of Helsinki. All data were deidentified before use.

#### *2.3. Participants*

We enrolled consecutive patients with cutaneous lesions of clinical and/or dermoscopic uncertain diagnosis of possible malignant skin tumours that needed to be removed or followed up according to a skin imaging expert dermatologist (E.C.) and that had LC-OCT examination.

#### *2.4. Imaging Examination*

Dermoscopy was performed both with a hand-held 10× dermoscope (DermLite DL4, DermLite, San Juan Capistrano, USA) and a 20× videodermoscope (Vivacam, Mavig, Munich, Germany). LC-OCT (DeepLive, Damae, France) was performed in horizontal and vertical scans and 3D mode, and a video was acquired on the operator's judgment. Following manufacturer recommendations, lesions in the patients' periocular region were excluded from LC-OCT examination.

Dermoscopic and LC-OCT diagnoses were given by an expert in skin imaging (E.C.) during the imaging examination of the lesions and were registered on the LC-OCT device. Concerning LC-OCT, BCC was diagnosed by the presence of tumour lobules [27], SCC by the presence of atypical keratinocytes in the entire epidermis [31,32], and melanoma by the presence of atypical bright cells that were sparse inside the epidermis and/or inside melanocytic nests [12–15]. The lesions suggesting malignant skin tumours at dermoscopic and/or LC-OCT examination were biopsied or surgically removed for histological diagnosis. The others were followed up for at least one year.

#### *2.5. Statistical Analysis*

The sensitivity and specificity of each technique (LC-OCT and dermoscopy) for the diagnosis of BCC, SCC/Bowen disease (i.e., SCC or Bowen disease), AK/SCC/Bowen disease (i.e., AK or SCC or Bowen disease) group and malignant tumour were calculated with their exact 95% confidence interval (CI) by using the histopathological diagnosis obtained from an incisional or excisional biopsy as the gold standard; a distinct analysis was also performed, including the lesions that had a final diagnosis based on a follow up of at least one year without a histological examination.

The sensitivity and specificity of LC-OCT and dermoscopy for the diagnosis of the different skin tumours were compared by using the proportion test. *p* values < 0.05 were considered statistically significant. All statistical analyses were conducted by using R (version 4.0.3., R foundation for statistical computing).

#### **3. Results**

We included 196 patients (81 women, 115 men; mean age of 64.45 years, range 0– 96 years) with 243 lesions; 226 lesions were histopathologically confirmed (Table 1) and 17 lesions had a final diagnosis after a follow-up of at least one year. Sensitivity and specificity of dermoscopy and LC-OCT for BCC, SCC/Bowen disease group, AK/SCC/Bowen disease group, and malignant tumour considering histopathology as the gold standard are reported in Table 2.


**Table 1.** Confusion matrix: dermoscopy vs. histology and LC OCT vs. histology.

**Table 2.** Sensitivity and specificity of dermoscopy and LC-OCT considering only histologically confirmed cases.


TP, true positive; P, positive; TN, true negative; N, negative.

#### *3.1. Diagnostic Performances of Dermoscopy and LC-OCT Considering Only Cases with Histological Diagnoses*

3.1.1. Dermoscopy and LC-OCT Diagnostic Performances for BCC

Considering the 79 histopathologically confirmed BCCs, LC-OCT showed higher specificity (0.95, CI 0.90–0.98; *p* = 0.015) for BCC diagnosis than dermoscopy (0.86, CI 0.80– 0.91, *p* = 0.015), and no statistically significant difference in sensitivity (0.97 CI 0.91–1.00 for LC-OCT and 0.96 CI 0.89–0.99 for dermoscopy; Table 2). Dermoscopy had 20 false positive (FP) cases that histologically corresponded to two nevi, one AK, two Bowen diseases in situ, three inflammatory lesions, a scar, and 11 benign nonmelanocytic lesions (including a solar lentigo, a seborrheic keratosis (SK), a lichenoid keratosis, a xanthogranuloma, a sebaceoma, a trichilemmoma, a trichoblastoma, and a neurofibroma). LC-OCT had seven FP cases that histologically corresponded to one nevus, one inflammatory lesion, and five benign nonmelanocytic lesions. LC-OCT enabled us to correctly diagnose 13 of the 20 dermoscopic FP BCCs as four inflammatory lesions, two cases of normal skin, two Bowen diseases, one AK, one nevus, one neurofibroma, one scar, and one xanthogranuloma. Three dermoscopic false negative (FN) cases were diagnosed at dermoscopy as AK, SCC, and SK and histologically corresponded to two infiltrative BCCs and a superficial microinvasive BCC. Two of these FN cases were also FN at LC-OCT and were diagnosed as AK and SCC, while they corresponded to two infiltrative BCCs at histopathology.

3.1.2. Dermoscopy and LC-OCT Diagnostic Performances for the Diagnosis of SCC/Bowen Disease

LC-OCT showed a slightly higher sensitivity (0.86, CI 0.65–0.97) for SCC/Bowen disease diagnosis than dermoscopy (0.77, CI 0.55–0.92), which did not reach a statistically significant difference. Concerning specificity, no statistically significant difference was found (0.98, CI 0.94–0.99 for LC-OCT and 0.96,0.92–0.98 for dermoscopy; Table 2). Among the 22 histopathologically confirmed SCC/Bowen diseases, there were nine FP at dermoscopy, five FP at LC-OCT, five FN on dermoscopy, and three FN on LC-OCT (two of them were Bowen diseases in situ diagnosed as AK). The nine FP at dermoscopy were histologically diagnosed as one BCC, three AKs, two inflammatory lesions, one dermatofibroma, one granulomatous lesion, and one microcystic adnexal carcinoma; the five FP at LC-OCT were histologically diagnosed as one BCC, one AK, one inflammatory lesion, one dermatofibroma, and one granulomatous lesion.

The five FN at dermoscopy were diagnosed as one AK, two BCCs, one SK, and one granuloma, and corresponded to four Bowen diseases in situ and one microinvasive keratoacanthoma at histopathology; the three FN at LC-OCT were diagnosed as two AKs and one SK and corresponded to two Bowen diseases in situ and a microinvasive keratoacanthoma at histopathology.

3.1.3. Dermoscopy and LC-OCT Diagnostic Performances for the Diagnosis of AK/SCC/Bowen Disease

LC-OCT showed higher sensitivity and specificity than dermoscopy for AK/SCC/ Bowen disease diagnosis (sensitivity of 0.87 (CI 0.72–0.96) for dermoscopy and 0.95 (CI 0.82–0.99) for LC-OCT, specificity of 0.96 (CI 0.92–0.98) for dermoscopy and 0.97 (CI 0.93– 0.99) for LC-OCT (Table 2)). However, the difference in sensitivity and specificity did not reach statistical significance.

3.1.4. Dermoscopy and LC-OCT Diagnostic Performances for Malignant Tumour

Considering only the cases with a histologic diagnosis of malignancy, we observed a significant increase of specificity from 0.73 (CI 0.64–0.81) with dermoscopy to 0.87 (CI 0.79–0.93) with LC-OCT (*p* = 0.013) for a malignant tumour, whereas the sensitivity was similar with the two imaging techniques (0.95 CI 0.89–0.98 for dermoscopy and 0.95 CI 0.90–0.99 for LC-OCT). The group of malignant tumours included both skin carcinomas and melanomas.

3.1.5. Diagnostic Performances of Dermoscopy and LC-OCT Considering Both Histological and Follow-Up Diagnoses

The sensitivity and specificity of dermoscopy and LC-OCT for BCC and malignant tumour considering as comparison the diagnoses derived from histopathology and followup at least one year are reported in Table 3.

**Table 3.** Sensitivity and specificity of dermoscopy and LC-OCT considering histology and followup diagnoses.


3.1.6. Dermoscopy and LC-OCT Diagnostic Performances for BCC (Including 13 Cases without a Histological Diagnosis)

Considering both the cases with histology and follow-up of at least one year, LC-OCT enabled us to correctly diagnose 26 over 33 dermoscopic FP cases: seven inflammatory lesions, three cases of normal skin, three scars, two Bowenoid SCCs in situ, two AKs, three nevi, one neurofibroma, one xanthogranuloma, one seborrhoeic keratosis, one rosacea, one sebaceous hyperplasia, and one scaly crust with papillomatosis. Among these cases, LC-OCT allowed us to save 13 excisions. Considering these 13 FP lesions at dermoscopy for which BCC was excluded after LC-OCT and for which surgical excision was not done (assuming that the follow-up >1 year of these patients could confirm the absence of BCC), the specificity for BCC diagnosis increased from 0.79 (CI 0.72–0.85) for dermoscopy to 0.96 (CI 0.91–0.98) for LC-OCT (*p* < 0.001). Sensitivity was similar for LC-OCT (0.97, CI 0.91) and dermoscopy (0.96, CI 0.89–0.99, Table 3).

#### 3.1.7. Dermoscopy and LC-OCT Diagnostic Performances for Malignant Tumours (Including 17 Cases without a Histological Diagnosis)

Considering the 17 FP lesions at dermoscopy for which malignancy was excluded after LC-OCT and surgical excision was not done (assuming that the follow-up of these patients could confirm the absence of malignant tumour), the specificity for malignancy increased respectively from 0.64 (CI 0.55–0.72) of dermoscopy to 0.89 (CI 0.82–0.93, *p* < 0.001) for LC-OCT. Sensitivity was similar: 0.95 (CI 0.90–0.99) for LC-OCT and 0.95 (CI 0.89–0.98) for dermoscopy (Table 3).

#### **4. Discussion**

Our study showed that LC-OCT can increase specificity for the noninvasive diagnosis of skin cancers compared to dermoscopy. Considering histopathology as a gold standard and analyzing only the cases with histological diagnosis (Table 2), we found an increase in specificity for the diagnosis of BCC from 0.86 (0.80–0.91 CI) for dermoscopy to 0.95 (0.90–0.98 CI) for LC-OCT (*p* = 0.015). The sensitivity was similar with the two methods (0.96 CI 0.89–0.99 for dermoscopy with three FN cases and 0.97 CI 0.91–1.00 for LC-OCT with two FN cases).

The same analysis including the cases that were diagnosed based on a follow-up of at least one year and that lacked histopathological examination obtained similar results. We found an increase in the specificity for the diagnosis of BCC from 0.79 (CI 0.72–0.85) for dermoscopy to 0.96 (CI 0.91–0.98) for LC-OCT (*p* < 0.001), whereas with regard to sensitivity we did not find any statistically significant difference between dermoscopy (0.96 CI 0.89–0.99) and LC-OCT (0.97 CI 0.91–1.00). Similar sensitivity results probably reflect the current use of LC-OCT as a secondary-level technique on skin lesions that are already identified as suspicious by dermoscopic examination. In most cases, LC-OCT easily confirms the dermoscopic diagnosis of a malignant tumour [27] and it is interesting to note that in clinical practice, LC-OCT is useful to increase the diagnostic confidence of the dermatologist and to confirm the need for surgical excision.

LC-OCT only missed two infiltrative BCCs, and retrospective examination of their images revealed hyperkeratosis thickness ranging from 200 to 300 μm with no visible dermis in one case and poor image in the other case

Concerning specificity, LC-OCT significantly reduced the cases of false positives (FP) BCCs in our series. 13 FP cases of BCC at dermoscopy were correctly diagnosed with LC-OCT (Table 3; Figures 1 and 2). These data are consistent with the latest studies on LC-OCT that highlight how this technique can easily recognize BCC imitators [20,33–35].

**Figure 1.** False positive case of basal cell carcinoma at dermoscopy. Clinical (**a**), dermoscopic (**b**), and LC-OCT images (**c**,**d**). Dermoscopy identified the lesion as a basal cell carcinoma, while LC-OCT as healthy skin. Histology confirmed the diagnosis of healthy skin. Dashed line (**b**) indicates the approximate area of the LC-OCT imaging. White scale bar in LC-OCT images: 100 μm.

**Figure 2.** False positive case of basal cell carcinoma at dermoscopy. Clinical (**a**), dermoscopic (**b**), and LC-OCT images (**c**,**d**). Dermoscopy identified the lesion as a basal cell carcinoma, while LC-OCT identified it as healthy skin. Histology confirmed the diagnosis of healthy skin. Dashed line (**b**) indicates the approximate area of the LC-OCT imaging. White scale bar in LC-OCT images: 100 μm.

There were only seven FP cases at LC-OCT, and three of them corresponded to benign skin tumours that can share histopathological similarities with BCC: a sebaceoma, a trichoblastoma, and a trichilemmoma. Sebaceoma (Figure 3) is characterized by multiple basal cell nests (Figure 3, asterisk) with a random mix of sebaceous cells in the upper and middle dermis with possible continuity with the basal layer of the epidermis [36]. Trichoblastoma (Figure 4) shows irregular nests of basal cells similar to BCC, with variable stromal thickening and pilar differentiation [36]. Trichilemmoma (Figure 5) is composed of one or more lobules (Figure 5, asterisk) in the dermis that extend in continuity with the epidermis (Figure 5, orange arrow) or the follicular epithelium, and there is a peripheral layer of columnar palisade cells [36]. Under LC-OCT, all these three tumours exhibited tumour islands with overlapping features of BCC. We should also consider that there could be biopsy sampling errors explaining some FP results. Interestingly, one LC-OCT FP case of our series, defined on the histological report of an incisional biopsy as solar lentigo, was later completely excised based on the retrospective revaluation of the LC-OCT images and had a final histological diagnosis of BCC.

**Figure 3.** Sebaceoma diagnosed as basal cell carcinoma at dermoscopy and line-field confocal optical coherence tomography (LC-OCT). Clinical (**a**), dermoscopic (**b**), and LC-OCT images (**c**,**d**). Dermoscopy shows a pinkish-yellowish background and linear vessels. LC-OCT reveals a large lobular structure (asterisk) with "feuilletage" and clefting, surrounded by hyperreflective stroma and connected to a hair follicle (red arrow). Dashed line (**b**) indicates the approximate area of the LC-OCT imaging. White scale bar in LC-OCT images: 100 μm.

In the literature, we could find only one prospective study on the diagnostic accuracy of LC-OCT for skin tumours, and it consists of similar real-life research on equivocal lesions. It showed that LC-OCT can significantly increase diagnostic confidence after dermoscopy and avoid potentially unnecessary biopsies [37]. However, it revealed a higher sensitivity (0.98) and a good, but lower, specificity (0.80) for LC-OCT compared to dermoscopy (sensitivity of 0.90 and specificity of 0.86).

The acquisition and interpretation of the LC-OCT images are operator-dependent and different results may be related to different investigator expertise and different lesion selection (equivocal aspect of the lesion at dermoscopy). Moreover, Gust et al. found that in 70% of the lesions, the LC-OCT diagnostic was provided with high confidence in comparison with dermoscopy which only provided high confidence in 48% of the lesions [37]. In this subgroup, the LC-OCT performance increased significantly, with a sensitivity of 100% and a specificity of 97% in agreement with our results. In the future, an effort should be made to define precise criteria for the LC-OCT diagnosis of skin tumours to have more reliable and comparable results. Moreover, artificial intelligence could help the identification of BCC tumour lobules and atypical cells [10]. Regarding the diagnosis of SCC/Bowen's disease/AK, we found an increase in both sensitivity and specificity for LC-OCT compared to dermoscopy without any statistically significant difference. Recently, many studies have shown that LC-OCT can identify several histological criteria of AK and SCC and this technique seems to be promising for the diagnosis of squamous cell tumours [16,29,31,32]. Our study could not prove a statistically significant benefit of LC-OCT possibly due to the relatively small sample size that has been analyzed.

**Figure 4.** Trichoblastoma diagnosed as basal cell carcinoma under dermoscopy and line-field confocal optical coherence tomography (LC-OCT). Clinical (**a**), dermoscopic (**b**), and LC-OCT images (**c**,**d**). Dermoscopy shows black, brown, and grey globules and dots on a pinkish and brownish background. LC-OCT reveals small well-delimited roundish lobules with "palisading" and some keratin cysts (white arrow). Dashed line (**b**) indicates the approximate area of the LC-OCT imaging. White scale bar in LC-OCT images: 100 μm.

Regarding the diagnosis of malignancy, considering histopathology as a gold standard and analyzing only cases with histological diagnosis, a significant increase in specificity (Table 2) was observed from 0.73 (IC 0.64–0.81) with dermoscopy to 0.87 (IC 0.79–0.93) with LC-OCT (*p* = 0.013). However, we did not detect any statistically significant difference in sensitivity between the two methods (0.95 IC 0.89–0.98 for dermoscopy and 0.95 IC 0.90–0.99 for LC-OCT), similar to the diagnosis of BCCs. The same analysis including the 17 FP cases of malignity at dermoscopy in which the diagnosis of malignant neoplasm was excluded with LC-OCT and surgery was not performed (assuming that the follow-up of these patients could confirm the absence of malignant tumour, (see Table 3)) obtained similar results. We detected an increase in specificity for malignant skin tumours from 0.64 (CI 0.55–0.72) for dermoscopy to 0.89 (CI 0.82–0.93) for LC-OCT (*p* << 0.001) while regarding the sensitivity we did not find any statistically significant difference between dermoscopy (0.96 CI 0.89–0.99) and LC-OCT (0.97 CI 0.91–1.00). These data on malignant tumours were mainly driven by BCCs and SCCs because melanomas were few. The increase in specificity with LC-OCT for the diagnosis of malignant skin tumours was mainly determined by the increase in specificity for the diagnosis of BCC. Although LC-OCT seems to play a possible role also for melanocytic tumours [12], to date there are few data on the diagnostic accuracy of LC-OCT for malignant skin tumours other than BCC and SCC [15].

**Figure 5.** Trichillemoma diagnosed as basal cell carcinoma under dermoscopy and line-field confocal optical coherence tomography (LC-OCT). Clinical (**a**), dermoscopic (**b**), and LC-OCT images (**c**,**d**). Dermoscopy shows a pinkish background with linear vessels and scales. LC-OCT reveals lobular structures (asterisk) with clefting, palisade, "feuilletage", and connection with the epidermis (orange arrow). Dashed line (**b**) indicates the approximate area of the LC-OCT imaging. White scale bar in LC-OCT images: 100 μm.

#### **5. Conclusions**


**Author Contributions:** Conceptualization, E.C. and J.L.P.; Methodology, E.C.; Software, A.C.; Validation, E.C.; Formal analysis, E.C. and A.C.; Investigation, E.C.; Resources, E.C.; Data curation, E.C., T.B. and A.C.; Writing—original draft, E.C. and T.B.; Writing—review & editing, E.C., L.T., M.S., J.M., J.P.-A., S.P., J.L.P. and P.R. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Ethical review and approval were waived for this study because this is on observational study and skin lesions were removed according to our clinical practice independently from this study.

**Informed Consent Statement:** Patient consent was waived due to the observational study type and data anonymization. Patient consent for image publication was obtained.

**Data Availability Statement:** Data are unavailable due to privacy restriction.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
