Optical Coherence Tomography Angiography and Attenuation Imaging for Label-Free Observation of Functional Changes in the Intestine after Sympathectomy: A Pilot Study

Abstract

We present in this study optical coherence tomography angiography (OCTA) and OCT attenuation imaging (OCTAI) for in vivo non-destructive visualization of intramural blood and lymphatic vessels of the intestine wall. Rabbit small intestine in the norm and after thoracolumbar sympathectomy served as the object of the intraoperative study. Compared to OCTA real-time imaging, OCTAI takes several minutes and can be termed as “nearly real time”. OCTAI signal processing was modified to take into account the signal-to-noise ratio and the final thickness of the intestine wall. The results showed that, after sympathectomy, changes in functioning of intramural blood and lymphatic vessels were observed with a high statistical significance. The occurrence of trauma-induced constriction of the blood and lymphatic vessels led to an especially pronounced decrease in the length of small-caliber (<30 µm) blood vessels (p < 10⁻⁵), as well as in the volumetric density of lymphatic vessels (on average by ~50%) compared to their initial state. Remarkably, OCTA/OCTAI modalities provide the unique ability for “nearly-instant detection” of changes in functional status of the tissues, long before they become visible on histology. The proposed approach can be used in further experiments to clarify the mechanisms of changes in intestinal blood and lymph flows in response to trauma of the nervous system. Furthermore, potentially it can be used intraoperatively in patients requiring express diagnosis of the state of intramural blood and lymph circulation.

1. Introduction

OCT technology is no longer limited to the structural visualization of objects [1,2]. A variety of methods for numerical processing of received OCT signals provide angiography (OCTA) [3,4] and lymphangiography [5,6], as well as neurography [7,8] and elastography [9,10,11]. These new modalities have greatly expanded the scope of medical application OCT, since its extensions provide monitoring of the functional state of tissues and systems at a precise moment of time, which gives a significant advantage over traditional “gold standard” histological examination.

Monitoring of intestinal blood circulation together with lymph flow is of key importance for diagnostics of a number of gastrointestinal diseases associated with insufficient tissue oxygenation and impaired water–salt balance, immune response, and transport of various substances, among many others, since the role of the circulatory and lymphatic systems in the proper functioning of the digestive system is enormous [12,13]. One of the clinically important tasks is to study changes in blood and lymph circulation in response to sympathetic denervation. Sympathetic denervation can be the result of a surgical operation (sympathectomy), which is performed for a curative purpose, but also often occurs as a result of various types of injury/trauma/disease [14,15]. Violation of the enteral innervation can cause ischemic phenomena and even may become the morphological basis for gastrointestinal dysfunction [16]. The spread of ischemia to the submucosal and muscle layers leads to damage to the metasympathetic innervation of the intestine; disturbances in homeostasis are aggravated in the distant post-traumatic period, and a vicious pathogenetic circle is closed [17]. Studies of blood perfusion in the intestinal wall demonstrate that sympathetic sympathectomy leads to spasm of the intestinal vessels [17]. This effect is due to the secondary discirculatory damage to the parasympathetic centers of innervation of the intestinal tract [18,19,20,21].

The system of intramural blood and lymph circulation of the intestine has been quite well studied in terms of anatomical and morphological aspects [22,23]. Until recently, dynamic functional disorders of the intestinal microcirculation have been poorly understood due to the limitations of methods suitable for this task. However, with an introduction of novel imaging techniques, research work in this area has intensified [24,25,26].

Intestinal lymphatic-vessel network is another system which has been poorly investigated in vivo [27]. Although the importance of the lymphatic system in metabolic disorders in the damaged intestine is beyond dispute, very few complex studies of the lymphatic microvessels of the denervated intestine in vivo have been undertaken due to the complexity of its objective visualization [13,28]. Real-time visualization of lymphatic vascular networks would certainly facilitate the analysis of the state of the lymphatic system and would allow a quantitative comparison of lymphatic vessels in pathological conditions with the norm.

This study aimed to reveal changes in the blood and lymphatic circulation following sympathetic denervation of the gut, since we believe that new knowledge about changes in the circulatory and lymphatic systems will be helpful in solving the most common problems of chronic inflammation and malabsorption in patients with damaged intestinal innervation [29].

Here we used a variant of OCTA based on a speckle-variance combined with high-pass filtering to isolate blood vessels, the principles of which were described in [30,31]. From our extensive experience, this approach operating in contact mode works fairly well and stably for real-time visualization of intestinal blood vessel networks in laboratory animals (rats, rabbits) [32,33] and was used without any changes. As for lymphatic vessels, their imaging has been developed based on the estimation of the depth-resolved attenuation coefficient, which was initially applied to examination of human hairless skin [34]. Its application to the intestine of experimental animals required certain adaptation of the algorithm taking into account the final thickness of the intestine walls.

In fact, there are known various approaches to evaluate lymphatic vessels in vivo [13,35,36,37,38,39]. These techniques either have poor spatial resolution or suffer from contrast agent injection procedures with potential chemical or radiation toxicity. For many applications it is important to avoid contrast agent injection because these agents can distort the lymph microenvironment, so that evaluation of lymphatic vessels in these conditions can lead to misinterpretation of the results. In this context, optical coherence lymphography is a promising contrast-agent free high resolution lymphangiography technique. OCT lymphangiography as an OCT modality was proposed over a decade ago [39]. Lymphatic vessels are clearly visible on B-scans as areas with a low scattering coefficient [39]. The approach based on isolation of low-scattering areas in OCT scans attracted many researchers in the field of experimental oncology (see, e.g., [7] and literature cited therein), as well as in studies of pathogenesis of benign conditions in the clinic [5,34].

In this study, for enabling OCT lymphangiography, we implemented OCT-based attenuation imaging, OCTAI—the method that detects low-attenuation regions inside the tissue [5]. Specifically, this work reports the results of adaptation of OCTAI in combination with OCTA for characterization of the states of intestine walls before and after experimental modeling of spinal trauma with sympathetic sympathectomy of the gut. For these experimental conditions we demonstrate the applicability of OCTA and OCTAI to investigate changes in intramural blood flow and lymph vessel structure.

2. Materials and Methods

2.1. Animals and Sympathectomy Model of Small Intestine

Six male rabbits (two in the control group and four in the target group) weighing 1.1–1.3 kg were included in the described pilot study. The animals were kept in accordance with the rules adopted by the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes (Strasbourg, 1986). Prior to the study, all animals were vaccinated against rabies, dewormed, and quarantined for 30 days. Before the start of the study, during the adaptation period, the animals were monitored for the manifestation of abnormalities in the state of health. Animals with skin diseases and other pathologies, as well as with an unbalanced type of higher nervous activity, were not used in the experiments. Clinically healthy animals were kept under equal vivarium conditions (ambient temperature, food, light) in individual cages. The study was approved by the Ethics Committee of the Privolzhsky Research Medical University (Protocol #17 from 10 November 2019).

Surgical interventions were carried out under general anesthesia with a mixture of solutions of 3.5% tiletamine hydrochloride, zolazepam, and 2% xylazine hydrochloride (in a volume proportional to the body weight), which were injected intraperitoneally to laboratory animals. After the introduction of the animal into anesthesia, a mid-median laparotomy was performed and a loop of the small intestine up to 20 cm in length was brought out through the wound (Figure 1a).

An acute trauma of the sympathetic nervous system was modeled by resection of the sympathetic ganglia of the spinal cord due to the thoracic-abdominal approach at the Th6-L2 level (Figure 1b) [40]. With such an injury, a violation of innervation occurs both in the small and large intestines. We focused on the study of lymph and blood microcirculation in the small intestine, because it absorbs all the main nutrients, such that changes in the microvasculature of the small intestine are considered to be a trigger mechanism in the development of enteric insufficiency in patients in the acute period of trauma of the sympathetic nervous system [8].

2.2. Histology

After completion of OCT imaging of the intestinal wall the scanned area was marked with histological ink and samples were taken for histological evaluation. Samples of the intestinal wall without pathology were obtained from similar sections of the intestine of animals in the control group. The specimens were fixed in 10% formalin for 48 h and the histological sections were prepared through the marked areas, so that their planes corresponded to the depth-wise OCT images. Hematoxylin and eosin (H&E) staining was used. A histopathologist, to whom the intestinal condition was unknown, evaluated the histological samples for the presence of edema and inflammation as well as normal or damaged blood and lymphatic vessels. Digital photographs were obtained in transmitted light with a Leica DM2500 DFC (Leica Microsystems, Wetzlar, Germany) microscope, equipped with a digital camera. The results of the histopathology were compared with the corresponding OCTA and OCTAI-based findings.

2.3. OCT Experimental Setup and Data Acquisition

In this study, a common path spectral-domain OCT device with cross-polarization detection and multimodalities capability was used as described in, e.g., [31,41]. The central wavelength of this device is 1310 nm with spectral width of 100 nm, resulting in axial resolution of 10 μm, lateral resolution of 15 μm, and scanning depth of 2 mm in air. The device has a 20,000 A-scan/s acquisition rate and performs 2D lateral scanning over an area of 2.4 × 2.4 mm² to obtain the 3D distribution of backscattered light with polarization parallel and orthogonal to the polarization of the probing beam [31]. Polarization sensitivity is not discussed in this paper. Scanning is performed in contact mode and takes 26 s to acquire a 3D data set. The presence of contact between the OCT imaging arm and the living tissue allowed us to efficiently immobilize the tissues in a minimally invasive and minimally traumatic way without additional devices like immobilizing chambers, the utilization of which was absolutely unacceptable in our study. The contact mode made it possible to minimize artifacts caused by living tissue motions (especially peristalsis-related streak artifacts), besides, the contact mode enables deeper visualization of tissues.

Intramural blood and lymph flow was studied by OCT from the side of the serous membrane of the small intestine (Figure 1d) before and after modeling an acute trauma of the sympathetic nervous system. The region of interest (ROI), from which the OCT data were recorded, was marked with surgical thread in order to re-examine it in 3 h after thoracolumbar sympathectomy which corresponds to the most acute period of injury. Each rabbit had 4 ROI examined before and after sympathetic denervation. During the acquisition of OCT data, the intestinal loop was released with care through the laparotomy wound, and then tucked back into the abdominal cavity.

Blood vessel visualization was enabled in real time, and then post-processing was performed for detailed estimates of the microvasculature metrics similar to earlier works [42,43] (see Section 2.4. Data Processing for OCT Angiography (OCTA) for details). To construct and quantify the lymphatic vessel network post-processing was used (see Section 2.5. Data Processing of OCTAI).

2.4. Data Processing for OCT Angiography (OCTA)

The OCTA data processing is based on our custom-made algorithm enabling real-time microvasculature visualization, which was described in detail in [31]. Briefly, we perform computationally efficient convolution-based high-pass filtering of temporal speckle variations caused by the blood flow within the regions of the scan with a sufficiently high pre-selected signal-to-noise (SNR) ratio. The scanning is in the continuous regime. The compared B-scans are sufficiently slowly shifted with a significant overlap of neighboring scans (i.e., the shift of neighboring B-scans is several times smaller than the beam diameter). Namely, the beam FWHM diameter is 15 µm. The scanning speed in the slow lateral direction is 4.6 µm/B-scan so that each subsequent B-scan is shifted by 4.6 µm relative to the previous one. Roughly we can estimate that for beams of 15 µm diameter we have 4 overlapped B-scans in the sliding mode. At the same time each subsequent B-scans overlap at 70% of the beams FWHM size and the time gap between them is 26/512 = ~0.05 s. Due to the fact that OCTA processing uses computationally efficient convolution-based high-pass filtering, our method efficiently visualizes the blood microcirculation inside the tissue in real time, enabling generation of 2D en-face OCTA images directly during the tissue scanning.

The calculation of vessels parameters is described in our previous work [42]. Briefly, each 3D microvascular dataset was converted to maximum intensity projection images representing the whole tissue depth (~1 mm) onto the 2D-plane. These images were binarized and skeletonized. Vessel thickness was calculated as 2x the distance between the vessel’s binary image border and its skeleton; vessels with overlapped binary borders and skeletons were assigned a thickness of 1-pixel [42]. The lateral resolution of the current OCT imager corresponds to vessel diameters less than ~15 μm. The quantitative assessment of OCTA images consisted in calculating total length of all visualized blood vessels per unit square (“L”, mm/mm²) and similarly the normalized lengths of blood vessels in the following groups: with small diameter (<30 µm corresponding to capillaries and post-/pre-capillaries), medium ones (30–65 µm, arterioles and venules), and large diameter (>65 µm, arterioles/venules and arteries/veins) [23]. Such a division by diameters was chosen to see which components of the microvasculature are perturbed to a higher extent.

2.5. Data Processing of OCTAI

In contrast to blood vessels, the lymphatic vessels look like optically transparent inclusions inside the tissue (Figure 2a) [44] that cannot be detected by OCTA due to very low SNR ratio in such vessels (close to the background noise level).

The applied OCTAI processing based on attenuation coefficient estimation is described in our previous publication [34], however, to process data from the intestine having a final thickness in OCT scans we added the tissue region detection (Figure 2b). Briefly, the lymphatic vessels are the interstitial structures that do not cause signal absorption or scattering, so in the absence of other signal attenuation effects (such as roll-off) there is no signal loss on the lymph [34,44]. Therefore, the lymphatic vessel visualization obtained from the depth-resolved attenuation coefficient distributions is corrected for the noise [34]. Due to the fact that the intestinal wall is thinner than the image size and we see the entire thickness of the wall, we denote the pixels with signal attenuation as the tissue pixels (Figure 2b, orange color). Thus, the resultant 3D dataset is presented as a discrete map where one type of pixel indicates the relation to lymphatic vessels (Figure 2b, white color), the second type of pixel indicates the relation to tissue (Figure 2b, orange color), and the third is assigned to other pixels (Figure 2b, black color). Figure 2c presents the resultant en face image of average intensity projection (AIP) calculated only for lymphatic vessels.

2.6. Statistical Analysis

For OCTA images we calculate the 95% confidence intervals for the estimated mean values of the specific (per square millimeter) lengths of blood vessels before and after sympathectomy [45]. Mann–Whitney–Wilcoxon test (Matlab: mwwtest) was used to indicate the significance of the difference between the parameters calculated from the OCTA images.

3. Results

3.1. OCTA Study of the Healthy and Denervated Small Intestine

The parameters of OCTA images, which were obtained before intestine sympathectomy, were taken as normal. In OCTA images of the intact intestine, numerous blood vessels of different diameters, large-paired vessels (vessel with a large diameter—a vein, with a smaller one—an artery) were visualized (Figure 3a).

Typically, a network of microvessels is fairly evenly distributed over the image. The parameter of the length of all blood vessels per square millimeter (expressed in mm/mm²) of the OCTA image was found to be equal to 6.26 ± 0.4 mm/mm². After trauma of the sympathetic nervous system the specific length of all blood vessels decreased down to 4.99 ± 0.37 mm/mm². This difference is statistically significant (p < 10⁻⁴) (the data are presented in Figure 3c). As indicated in Figure 3b by white dashed lines, the decrease in the length of all blood vessels mostly occurred due to a substantial drop in the amount of capillaries and small vessels (<30 µm in diameter): before trauma it was 1.09 ± 0.079 mm/mm² and after trauma it became 0.63 ± 0.065 mm/mm² (p < 10⁻⁵) (Figure 3d). For comparison, the amount of large vessels (>65 µm) decreased from 2.98 ± 0.32 mm/mm² before trauma down to 1.96 ± 0.2 mm/mm² after trauma (p < 10⁻⁴) (Figure 3d).

In this context the absence of a similar decrease for the vessels with intermediate thickness (30–65 µm) and even a slight statistically insignificant increase shown in Figure 3d is very consistent with the above-made statements about the thinnest and thickest vessels. Indeed, the thicker vessels (>65 µm), in which the blood flow gets weaker after sympathectomy, become less visible in the angiograms, so that a part of such previously thicker vessels with reduced blood flow are interpreted as somewhat thinner vessels in the automated processing of the OCTA images and, consequently, are counted in the intermediate group (39–65 µm). For this reason there are no clear visible variations in the intermediate group.

The histological sections shown in Figure 3e,f do not allow for quantitative estimates of the parameters of blood microcirculation, however, qualitatively, histology confirms the above-made conclusions about the decrease in the blood flow intensity. Namely, in Figure 3f the signs of red blood cell stasis in the small-diameter vessel cross sections are clearly seen as indicated by dotted arrows, whereas there are no similar phenomena in the vessel cross sections before sympathectomy. The overall histological examination of the small intestine wall structure, post sympathectomy, does not show obvious changes (Figure 3f) in comparison to the norm (Figure 3e). Each layer is clearly differentiated; pronounced edema, sludge, and microthrombi in arterial and venous vessels were not found, the cellular composition corresponded to the norm, the height of the villi and crypts was comparable to the data of normal histology (Figure 3e). Only two out of 16 histological specimens showed mild diffuse edema.

3.2. OCT Visualization of Lymphatic Vesssels in the Healthy Small Intestine and Its Response to Sympathectomy

We calculated the difference in OCTAI for each ROI before and after sympathectomy taking 15 out of 16 ROI from four rabbits for this analysis (four locations for each rabbit, but one ROI was excluded due to vast edema developed after suppression of peripheral nerves, which introduced a significant artifact into the result). The results for the volumetric density of low-attenuation (visually “empty”) structures estimated using OCTAI are represented in

Table 1. OCTAI volumetric density (VD) values before and after sympathectomy.

	Cases of a Significant Decrease in the VD Values										Cases of a Slight Decrease/Increase in the VD Values
ROI	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
VD/10−2 before sympathectomy	4.14	4.31	2.75	3.46	5.00	1.93	5.44	4.37	5.61	2.80	0.73	0.57	1.78	0.68	1.65
VD/10−2 after sympathectomy	0.49	1.28	0.88	1.14	2.30	0.97	2.79	3.33	5.02	2.58	0.71	0.58	1.83	0.86	2.17
Change, %	−88.2	−70.3	−68.0	−67.1	−54.0	−49.7	−48.7	−23.8	−10.5	−7.9	−2.74	1.75	2.81	26.5	31.5

.

These data demonstrate that in 10 out of 15 pairs the volumetric density exhibits pronounced decrease after sympathectomy. In these pairs (#1–#10) the median decrease in the volumetric density is 49.74% (mean decrease is 44.62% and standard deviation is 23.5%). The other five cases (#11–#15) demonstrate either slight decrease or increase: cases #11–#13 showed a very small decrease (#11) or small increase (#12 and #13) in volumetric density. In two cases (#14 and #15) the observed increase in values is possibly associated with tissue edema, which was confirmed by histology. Notably, the initial values of the OCTAI signal in all those cases were rather low, which increases the errors of the processing algorithm and makes it difficult to adequately track changes.

The typical changes of lymphatic vessels after sympathectomy are represented in Figure 4.

Structural OCT images—both B-scans (Figure 4(a1,b1)) and en face (Figure 4(a4,b4))—demonstrate reduction of low-attenuation structures (lymphatic vessels) after sympathectomy (Figure 4(b1,b4)) compared to the initial state (Figure 4(a1,a4)). Correctness of operation of the proposed OCTAI algorithm is confirmed by the comparison of the visualization of lymphatic vessels on structural images (Figure 4(a1,b1,a4,b4)) with the corresponding processed OCTAI images (Figure 4(a2,b2,a3,b3)), where the lymphatic vessels are segmented.

For the control group, the histological images of the intestinal wall demonstrate an empty cross section of lymphatic vessels (Figure 4c) while after sympathectomy the lumens of lymphatic vessels are filled with protein precipitates (Figure 4d). This fact qualitatively indicates a change in the composition of the lymph and the occurrence of a post-traumatic decrease in lymph flow. In this regard, it can be expected that the lumens of the lymphatic vessels may be narrowed after sympathectomy. However, we could not confirm this by reliable quantitative analysis because of the insufficient number of samples in the reported pilot study.

4. Discussion

Emerging modalities of OCT, such as OCTA and OCTAI, which enable in vivo label-free non-destructive visualization of blood and lymphatic vessels, opened a broad range of various biomedical applications due to their ability to provide previously unavailable information about functional status of tissues. Although the application of OCT for both blood microcirculation and lymphatic vessel visualization has been used previously, the particular realizations of these modalities are characterized by several important features that, to the best of our knowledge, are not typical of other groups and studies.

First, the chosen contact mode of the OCT examination deserves special discussion. There is a widespread opinion that non-contact mode is always preferable. In this regard, we have to recall that the contrast-agent-free OCTA is based on discrimination of own motions of blood particles. However, natural motions of living tissues produce strong motion artefacts that often completely corrupt OCTA data. For this reason apparently “non-contact” OCT/OCTA methods described in the literature are operable, but only with the use of various tissue-immobilizing devices (like special chambers with transparent windows, etc.) In our study utilization of such immobilizing chambers and other devices on the intestine was impossible because such devices can be very traumatic. At the same time, the peristaltic motions of the bowel are rather pronounced and without immobilization it would be absolutely impossible to obtain artifact-free images to analyze the blood and lymph vessels. In this situation the utilization of contact-mode OCTA allowed us to solve the problem of motion artifacts and enabled rather efficient immobilization, while in comparison with conventionally used immobilization devices our contact-mode OCTA is minimally invasive and minimally traumatic, even allowing application on patients. In addition, deeper visualization of tissues was an extra advantage of the contact-mode OCT.

Next, attenuation distribution visualization is known in analysis of OCT scans (e.g., [46,47]). Our research group contributed to the field by introducing a robust-to-noise method for depth-resolved attenuation coefficient calculation [48]. However, for lymphatic vessel visualization, to the best of our knowledge the estimation of the local attenuation coefficient has been recently applied only in our previous work [34]. Conventionally, the lymphatic vessels on OCT scans are identified as local regions of low-level signal, whereas the correction for depth-dependent signal decay is used as an auxiliary procedure for correction of the threshold value, below which the signal is considered “low-level” [5]. The closest to our utilization of the low-attenuation filter (but yet unrelated to lymphatic system visualization) is for detection of the edema region, where the rate of the signal decay is reduced [47]. The zones of increased transparence in lymphatic vessels correspond to both low signal zone and low attenuation zone, so that in principle both may be used for detection of lymphatic vessels. The advantage of the attenuation-based method [34] used here is that in regions of low signal-to-noise ratio the detection of the signal-level decrease becomes problematic, whereas the estimate of the decrease in the attenuation is feasible by accounting for noises and enables better tolerance to the noise influence.

Concerning the object of the study, we present here the application of these OCT modalities for visualization of intramural blood and lymphatic vessels of the intestine wall in laboratory animal experiments. Namely, rabbit small intestine in the norm and after thoracolumbar sympathectomy (which simulates either consequences of spinal-cord trauma or therapeutic sympathectomy) was chosen as the object of the study.

The choice of the model is because of the critical importance of the function of the intestinal lymphatic system in patients with spinal injury [49]. It is known that insufficiency of the intestinal lymphatic system is associated with the risk of fat malabsorption [50], impaired metabolism of triglycerides, cholesterol, and chronic inflammation of the intestine [51]. The results of current research indicate that therapy aimed at improving lymphatic function is a potentially effective way to prevent these complications in patients with spinal injury [51,52]. However, the development of this area of lymphotropic therapy requires accurate data on the time of development, manifestations, and in vivo methods of controlling/monitoring the state of the lymphatic system in the intestine.

The obtained results indicate that, in the acute period of small intestine sympathectomy (after 3 h), statistically significant reduction in the blood circulation (perfusion) occurs, which is especially pronounced for small-diameter vessels. Previously, it was found that open abdomen and even laparoscopic abdominal surgery are accompanied by decreased mesenteric arterial blood flow, which may be a spastic reaction of the microvasculature [53]. It is possible that the mechanism of spasm of the intramural vessels of the small intestine after sympathetic denervation is based on the pathological reaction of the intramural nerve plexuses.

The OCTAI results indicated a pronounced decrease in the volumetric density after sympathectomy in 10 of the 15 evaluated cases presented in Table 1. The reasons for somewhat different or unclear reaction in the other five cases were discussed in detail in Section 3: Results, directly after Table 1. In each of those cases it was argued that the data do not affect the overall conclusion about the reduction in the OCTAI signal which was very clear in all 10 representative cases.

Additionally, to avoid over-interpretation, we should indicate several other limitations of OCTAI. First, the approach does not distinguish transparent liquid in lymphatic vessels from other transparent structures. It should be noted that transparent inclusions may also correspond to edema [54] or nerve fascicles [7,8], so that both may contribute to the evaluated volumetric density parameter. If edema develops it should increase the volumetric density of the OCTAI signal. On the contrary, if nerve fascicles degrade, they should become less transparent and correspondingly should decrease this parameter.

Edema may also develop in various forms. One form is an extensive area of interstitial transparent liquid that may look like a weakly absorbing ovaloid inclusion. The development of such a large inclusion in the visualized region certainly increases pronouncedly the estimated volumetric density of the OCTAI signal, which has no relation to variations in the state of lymphatic vessels, so that such cases should be excluded from the analysis. Another type of edema may have a form of thin interstitial layers containing weakly absorbing/scattering liquid. Such thin layers have not yet been resolved in the OCT images, so that the tissue may look fairly homogeneous, but with somewhat reduced attenuation. Therefore, edema of this type may moderately increase the volumetric density of the OCTAI signal, which evidently occurred in cases #14 and #15, for which edema was not obvious in the OCT scans but was confirmed histologically.

Next, concerning the nervous fascicles that are also rather transparent, the literature does not describe in vivo morphological changes in the enteric nervous system that develop in the first 6 h after spinal injury. In the period from 6 to 24 h after spinal trauma, changes occur on the ultrastructural level [55], which cannot be fixed by the OCT method due to its insufficient resolution. Changes described at the cellular and tissue level (post-traumatic degradation and regeneration of nerves) are observed over periods measured in weeks and months [56,57].

In view of this, the clear decrease in the volumetric density observed in the most part of the examined cases (see Table 1) can be reasonably attributed to the contribution of lymphatic vessels, thus, the decrease in the volumetric density of the OCTAI signal does really indicate the decrease in the volume of lymphatic vessels after sympathetic denervation.

In this context it should be mentioned that the OCTAI approach does not provide any information about the lymph flow. This approach indicates only the changes in transparent structures via changes in the volumetric density of the OCTAI signal, so that we had to use additional information to distinguish between the contributions of possible edema, variations in the transparency of nerves, and lymphatic vessels themselves.

In the discussed observations we can hypothesize that the decrease in the volumetric density of the lymphatic vessels can be caused by the primary change in the blood flow intake because of spasm of blood vessels induced by thoracolumbar sympathectomy. The spasm of the precapillary sphincters [53], which block the blood flow to the subsequent capillary bed and thus limit the flow of fluid into the intercellular space, is of the greatest importance here. Because of this, the lymphatic capillaries have no liquid to absorb, so that they decrease in size, which then affects the larger lymphatic vessels as well. Both phenomena—the loss of the microcapillary link of the blood flow and the reduction in the volume of the detectable lymphatic cavities—were histologically confirmed in the present study. The above-described scenario is supported by the fact that histological examination of intestinal samples does not reveal tissue edema in cases where a notable decrease in the volumetric density of lymphatic vessels was recorded by OCTAI (Table 1, ROIs #1–#10) or when the changes were not pronounced (Table 1, ROIs #11–#13). Notably, histology registers only such changes that physically modify the structure of the tissue but cannot yet track its functional rapidly changing state. In contrast, OCTA and OCTAI modalities provide the unique ability to “nearly-instantly” detect changes in functional status of the tissues, which is impossible to detect and, moreover, characterize quantitatively using the “gold standard” histological examination.

We should mention that the formulation about “instant detection” of changes in the blood and lymph circulation should not be confused with real-time diagnosing. Nevertheless, we point out that even the post-processing analysis of OCT data requires only several minutes. Therefore, in comparison with conventional laborious and time-consuming histological examinations our in vivo obtained OCT results, even after a several-minute delay, can be considered as “nearly real-time”. In addition, we should emphasize that the used method of OCT angiography provides microcirculation visualization literally on-flight and, therefore, this is indeed a real-time visualization. This possibility is critically important for our OCT examinations of intestine that were performed intraoperatively at very limited available time intervals and could not be repeated afterwards. In this regard, unlike methods based on post-processing, the used on-flight OCTA visualization allowed us to avoid collection of data of insufficient quality (e.g., data corrupted by motion artefacts).

5. Conclusions

The described tests of the OCTAI and OCTA adapted to the specific features of the intestine wall were performed in an experimental model of sympathetic denervation of the gut. In an acute period after sympathectomy a reduction in the total length of intramural perfused vessels was detected. This is mostly due to the sharp drop in the small-caliber blood vessel length. We recall that this difference revealed by OCTA within a rather short post-trauma time interval is caused by the functional state of the blood vessels (development of stasis). The OCTAI signal also decreases in the majority of examined cases (10 of 15), which can be reliably attributed to the decrease in the lymphatic vessel density.

These results shed light on the possible mechanisms of external and auto-regulation of blood and lymphatic circulation in violation of the sympathetic innervation of the intestine and, we believe, are helpful in solving chronic inflammation and malabsorption problems appearing in patients with damaged intestinal innervation. The decrease in the functional activity of the lymphatic vessels of the intestine that we found indicates a high risk of impaired absorption and transport of lipids already in the first hours after injury. This should be taken into account when choosing a nutritional support program in the acute period of spinal injury, otherwise, it may not only be ineffective, but also become a trigger for chronic inflammation in the intestine. The beginning of lymphotropic therapy and control of the lipid profile during the first day after the injury, with a high probability, will help to avoid these complications. The OCT-based extensions enabling nearly real time evaluation of changes in the blood circulatory and lymphatic system state have proven to be very promising techniques for further study of the aforementioned challenging biomedical problems.