Cholesterol microcrystals and neutrophil extracellular traps detection during open aortic surgery

Abstract

Objective

Improved aortic surgery outcomes are linked to a broader comprehension of the pathogenesis of thromboembolic complications. This study aims to evaluate the involvement of cholesterol microcrystals and neutrophil extracellular traps (NETs) in postoperative thrombotic complications following open aortic surgery.

Methods

Aortic blood smears were examined precisely to identify the presence of cholesterol microcrystals (CMs) using polarized light microscopy, Coherent Anti-Stokes Raman spectroscopy (CARS), and fluorescence microscopy to detect NETs. The data obtained, including CMs quantity, perimeter, and NETs quantity, were evaluated as possible predictors of the postoperative complication rate.

Results

Fifty-five patients (85%) had an uneventful postoperative period, while 10 patients (15%) experienced early postoperative complications, there was a statistically significant positive correlation between the average perimeter of the CMs and the number of NETs in the blood smears in patients who experienced a complicated postoperative period (rho = 0.67; p = .03).

Conclusion

In some cases, complications in the early postoperative period after aortae surgery may be caused by CMs embolism (CE) of the distal vascular bed, accompanied by NETs-mediated thrombosis. The protocol for assessing arterial blood allows for the identification and evaluation of CMs and NETs characteristics as predictors of perioperative thromboembolic complications.

Introduction

Abdominal aortic aneurysm and aortoiliac occlusive disease remain the most relevant challenges in the modern vascular surgery. Despite significant achievements of modern aortae reconstructive interventions and complex postoperative treatment, perioperative thrombotic complications still reach 10%, even during elective surgery.^1,2 It is known that the processes of thrombosis and inflammation are of great importance in the development and progression of atherosclerosis. ³ Forward-looking hypothesis confers leading role to neutrophil extracellular traps (NETs) in the thrombotic complication evolution. NETs are DNA strands with manifested pro-coagulant activity, secreted by neutrophils as a result of programmed cell death, the so-called NETosis process that also induced by the structural components of atherosclerotic plaques, in particular, cholesterol microcrystals (CMs).^4,5 In case of an unstable atherosclerotic plaque, presenting the cap rupture, CMs exit into the bloodstream can lead to acute ischemic damage due to arterial embolization.^6–8 Therefore, open aortic repair is associated with an increased risk of thromboembolic complications, especially in presence of unstable atherosclerotic plaques and parietal thrombotic masses at the level of aortic clamping and proximally.^6,9 The role of NETs and CMs in development of acute ischemic events in unstable course of atherosclerosis has been investigated in several studies in the coronary and cerebral arteries.^6–9 To date, there are only a few reports on the impact of NETs and CMs on the results of open aortic surgery. ¹⁰

The purpose of this study was to assess the relationship between the presence of CMs and NETs in the arterial blood of patients with abdominal aortic aneurysm and aortoiliac occlusive disease with different patterns of the postoperative period.

Material and methods

Design and study population

This retrospective study analysed 65 patients who underwent surgery for abdominal aortic aneurysm and aortoiliac occlusive disease at our University between 2021 and 2023. The patients’ ages ranged from 44 to 75 years (63.3 ± 0.9), with a male-to-female ratio of 8:1. A median laparotomy approach was used for all patients. Due to significant differences in the pathogenesis of abdominal aortic aneurysm and aortoiliac occlusive disease, surgical treatment, and its results, all patients were divided into two groups: group 1 (n = 35) included patients with abdominal aortic aneurysm, and group 2 (n = 30) included patients with aortoiliac occlusive disease. Institutional review board requirements were waived for this retrospective study.

Arterial blood collection and sample preparation

Following vascular anastomoses and restoration of blood flow, 1 ml of arterial blood was obtained via a sterile syringe by puncturing the main vascular prosthesis branch. A dry, clean glass slide was used to prepare a smear with 0.01 ml of arterial blood. A cover glass was placed on top, covering one part of the arterial blood smear and leaving the rest uncovered. After preparing the smear, the specimen was stored in an airtight container at room temperature until microscopy was performed.

Polarized light microscopy

Optical light microscopy was performed using a polarizing microscope, the Leica DM4500P (Leica Microsystems, Germany), at various modes and magnifications. The covered part of the arterial blood smear was evaluated in five areas, each with dimensions of 200 × 200 µm, located in the corners and central part of the cover glass. The quantity of microcrystals with morphological signs and optical properties typical for CMs was then counted. The study evaluated the double refraction property of CMs that were diamond-shaped, plate-shaped, or needle-shaped. The assessment was conducted by observing the interference color (blue or yellow) and direct extinction of the crystals when the stage was rotated at a right-angle position of the microscope analyzer and the polarizer. The study found that these signs are highly specific for CMs. Additionally, the sizes of the CMs were also assessed.

For the purpose of clarifying quantitative analysis and qualitative confirmation of the presence of CMs, as well as the detection of NET, CARS and fluorescence microscopy of 20 arterial blood smears was performed: 10 with a complicated course of the postoperative period, another 10 with an uncomplicated course (five smears from each group selected randomly).

Coherent anti-Stokes Raman spectroscopy (CARS-microscopy)

The Confotec CARS spectrometer (SOL Instruments Ltd, Belarus) was used to record high spatial resolution and high scanning speed microimages of CARS. The CARS signal was generated using a Nd³⁺:YVO₄ laser operating in passive synchronization mode with diode pumping (1064 nm, 7 ps, 85 MHz, 5 W, linewidth ∼ 5–7 cm^–1), as well as an optical parametric oscillator (OPO) with synchronous pumping from the second harmonic of radiation at 1064 nm. The CARS process utilized a tunable OPO beam (690–990 nm, 6 ps, 150–350 mW) that was collinearly overlapped with a portion of the 1064 nm beam, which served as the pump wave (λ_p) and Stokes wave (λ_S), respectively. This resulted in the generation of an anti-Stokes signal at λ_aS = λ_p/(2−λ_p/λ_S), which was detectable by the system within the Raman shift range of 990–3580 cm^–1.

CARS microscopy employed a high numerical aperture water immersion objective (NA = 1.2, UPLANAPO-60×, Olympus, Japan) to tightly focus the beams. The anti-Stokes signal was collected at a wavelength of 660.6 nm (Raman frequency −2870 cm^–1) in the reverse backward direction (E-CARS) using the same objective. The signal was separated from the excitation beams using two bandpass filters (FF01-720/SP-25, SemRock, LLC, USA) and directed to the entrance slit of a 520-mm focal length grating monochromator-spectrograph with a 600 grooves per millimeter grating. The H6780-01 photomultiplier tube (Hamamatsu, Japan) detected CARS signals. A computer-controlled XY galvanometric scanner (VM1000, ScanLab, Germany) rapidly scanned the sample in the lateral focal plane of the lens. CARS images of 250 × 250 pixels were obtained by raster scanning the surface area of a 225 × 225 μm sample at room temperature. The signal integration time was 3 μs/pixel. Fifteen frames were scanned for each sample. The perimeter of the areas with positive pixels of CMs was calculated for each frame using the ImageJ program version 1.53q (USA). The total perimeter value for all frames was then determined and the average of the 15 values was calculated.

Fluorescence microscopy

The open area of the smear (without a coverslip) was stained with DAPI. Using a NIKON Eclipse Ts2R-FL fluorescence microscope, an area of approximately 1 cm² was scanned in each sample, resulting in around 1000 frames per patient. The ImageJ program was used to calculate the number of NETs in each frame, and the total number of NETs was determined for each patient by summing across all frames.

Statistical analysis

Statistical data analysis was conducted using MedCalc Ver. 20.010 (Belgium). To test the distribution type, the Shapiro-Wilk test was applied. Quantitative values are presented as mean and standard deviation (M ± SD) for normal distribution and as median and interquartile range for abnormal distribution. Student's t-test for independent samples was used to compare data from normally distributed populations. The Mann‒Whitney test was used to compare quantitative data from populations with abnormal distributions. Fisher's exact test (two-sided) was used to compare two or more relative indicators that characterize the frequency of a particular trait. The Spearman correlation coefficient (rho) was used to assess the relationship between characteristics. Differences were considered statistically significant at p < .05. The study investigated the requirements of the Declaration of Helsinki of the World Medical Association (2013).

Results

Postoperative course

In each group, two subgroups were distinguished: with and without complications in the early postoperative period. Fifty-five (85%) patients had uneventful postoperative period, 10 (15%) patients (five cases in each group) had early postoperative complications. No deaths were observed. The list of complications is shown in Table 1.

Table 1.

Types of early postoperative complications (N = 10).

Group	Gender, age	Number of arterial blood smear	Complication
1	М, 73	2	Acute lower extremity arterial embolism (popliteal artery)
	М, 64	17	Acute lower extremity arterial embolism (tibial arteries)
	М, 75	18	Paralytic ileus
	М, 74	19	Large Bowel Ischemia/Infarction
	М, 73	20	Thrombosis of the aorto-ilio-femoral bypass
2	М, 60	1	Thrombosis of the aorto-femoral bypass
	М, 57	8	Acute lower extremity arterial embolism (superficial femoral artery)
	F, 54	9	Thrombosis of the aorto-femoral bypass
	М, 72	10	Acute Ischemic Stroke
	М, 58	11	Acute lower extremity arterial embolism (tibial arteries)

CMs and NETs detection

Polarized light microscopy revealed CMs in 62 (94%) of 65 smears with the typical optical characteristics, mostly less than 10 μm (Figure 1(a) and (b)).

Figure 1.

(a) Cholesterol microcrystals and NETs detection. Polarized light microscopy of the arterial blood smears: cholesterol microcrystals with maximum size less than 50 μm (a) and more than 50 μm (b). CARS images of CMs with different sizes (c, d). Fluorescent images with NETs (e, f). The average CMs perimeter (g), the NETs number (h) in the arterial blood smears (N = 20).

CARS microscopy revealed CMs for all patients examined. An example is shown in Figure 1(c) and (d). Typical fluorescent images with NETs are shown in Figure 1(e) and (f). There was a strong correlation between the number of NETs and the corresponding graph data for those patients (Figure 1(g) and (h)).

The CMs quantity in arterial blood smears of both groups depending on the presence of complications are shown in Figure 2.

Figure 2.

Box plots shows the relation of CMs quantity in arterial blood smears and complications in post-operative period.

Although no statistically significant difference in the number of CMs was identified, there is a tendency for a higher median CMs number in the blood smears of group 2 (subjects who experienced complications in the early postoperative period). Arterial blood smears of four patients did not contain CMs; three of them had an uneventful postoperative period.

The detection rate of CMs, depending on their size and the presence of complications in both groups, is presented in Figure 3. CMs larger than 50 µm were detected only in three (5%) smears, two of them belonged to patients from group 2 who had complications in postoperative period.

Figure 3.

Detection rate of CMs depending on their size and the presence of complications in both groups.

Figure 4(a) and (b) presents the average perimeter of CMs and the number of NETs, as well as the course of the early postoperative period in both groups. The average CMs perimeter varied from 2.42 to 30.1 µm and NETs number per cm² varied from 13 to 215. Although no statistically significant difference in the analyzed parameters was identified, there is a tendency towards a higher level of the median NETs number in the arterial blood smears in group 2.

Figure 4.

Box plots shows relation of CMs perimeter (a) and the NETs number (b) in arterial blood smears (N = 20).

Correlation analysis

A correlation analysis was performed to assess the relationship between CMs perimeter (a) and the NETs number (Figure 5(a) and (b).

Figure 5.

Scatter plots showing CMs quantity (a) and CMs perimeter (b) regarding the NETs quantity (N = 20); CMs perimeter and the NETs quantity depending on the group (c, d) and the postoperative course (e, f) (N = 20).

A moderate positive correlation was found to be statistically significant between the average perimeter of the CMs and the number of NETs (rho = 0.46; p = .04). Subsequently, a subgroup analysis was conducted based on the group and the nature of the postoperative period, as illustrated in Figure 5(c)‒(f).

The presented diagrams illustrate a statistically significant positive correlation between the average perimeter of CMs and the NETs number in the blood smears in group 2 (rho = 0.7; p = .022) and with a complicated postoperative period (rho = 0, 67; p = .03).

Discussion

In 2004, a milestone discovery was made that DNA not only serves as a genetic information store but also plays a role in innate immunity.^4,11 NETs are composed of modified released chromatin, granule proteins, nucleus, and cytoplasm. It has been subsequently shown that NETosis (NET formation) is not obligatory and can occur without cell death. It plays a significant role in the pathogenesis of many inflammatory or autoimmune diseases and can be activated by various stimuli, including bacteria, fungi, protozoa, viruses, and bacterial cell walls.^12–16

All compounds of the low-density lipoproteins, except for cholesterol, are utilized in foam cells. If cellular death occurs, cholesterol enters the intercellular space. Free cholesterol in the intracellular and extracellular space of atherosclerotic plaques spontaneously organizes into a crystalline form, causing inflammation. ¹⁷ Destroyed foam cells release lysosomal enzymes and CMs, contributing to the destruction of the plaque's fibrous cap, its rupture, and CE. ¹⁷ This leads to acute ischemic damage due to mechanical vessel obstruction and an inflammatory response from the intima, resulting in intravascular thrombus formation and leukocyte infiltration.^13,17,18

The study does not definitively exclude atheroembolism (cholesterol crystal embolism) as a cause of early postoperative complications. The risk of complications increases significantly due to mechanical intraoperative trauma during aorto-iliac reconstructions. Graft thrombosis is typically associated with technical errors in vascular anastomoses, as well as inadequate assessment of arterial outflow. However, in most presented cases, anastomosis reconstruction was not required after emergency thrombectomy (1/3, 33%). Open surgeries create conditions, including hemodynamic ones, that promote the development of athero- and thromboembolism at the level of aortic clamping and proximally. ⁷ These conditions arise from atheromatous unstable plaques. Circulating microemboli can cause damage to the arterioles and endothelium, leading to the activation of the coagulation cascade. This can result in microthrombosis, ischemia, and tissue necrosis. The pathogenesis of thromboembolic complications involves interrelated processes such as atherogenesis, inflammation, and thrombosis. ³ The presence of circulating microemboli is one of the triggers for netosis, which is a key element in intravascular thrombus formation. ¹⁵ According to another data, patients with long COVID and thrombovascular complications presented structural association of nets with amyloid fibrin microclots in the plasma. Thus, it is possible that CMs could potentially be structurally associated with nets. ¹⁹

Both groups included patients with ischemic stroke and sigmoid infarction, as it is not always possible to determine the true nature of ischemic complications. The available clinical and instrumental data did not allow us to exclude the possibility of thromboembolic causes of acute circulatory disorders in those organs. Additionally, one patient developed long-term dynamic bowel obstruction in the early postoperative period, which may also be caused by CE. Ischemic damage caused by CE can disrupt the intestinal mucosal barrier function, increase permeability of the enterocapillary barrier, and allow microorganisms to penetrate into the circulatory system. This is known as the ‘bacterial translocation syndrome’, which contributes to the development and progression of systemic inflammatory response syndrome and multiple organ failure, as seen in this case.

The most common sources of CE are atherosclerotic plaques and the aortic intima, which are encrusted with CMs. This is known as the ‘shaggy aorta’ syndrome, first described by Kazmier in 1989. ²⁰ Currently, this term refers to a diffuse, heterogeneous atherosclerotic lesion that involves more than 75% of the descending thoracic aorta, with a thickness of thrombotic deposits greater than 4 mm, detected by multislice-computed tomography. ⁶ The ‘shaggy aorta’ syndrome is a significant risk factor and an independent predictor of early postoperative embolic complications and mortality after both open and endovascular interventions.^7,8

It is assumed that the CE can only be reliably discovered through microscopic examination of the affected target organ. The specific histological sign of CE is the presence of slit-like spaces in the thrombotic masses in the arterioles lumen, which are formed due to the leaching of CMs during sample preparation. However, these microscopic changes are only found in half of the patients with clinical CE signs, and histology reveals only signs of an inflammatory tissue reaction. ²¹

Presented research has shown the presence of CMs and NETs in the arterial blood of patients with abdominal aortic aneurysm and aortoiliac occlusive disease, with a high probability of detection in vivo. A tendency towards a higher median number of CMs was observed in group 2 with a complicated postoperative period. In this subgroup, ‘large’ CMs were detected in the majority of cases, with a maximum size of over 50 microns. There was a tendency towards a higher median number of NETs in arterial blood smears of patients in group 2. A statistically significant positive correlation was identified between the average perimeter of CMs and the number of NETs in arterial blood smears. This may indicate an unstable course of atherosclerotic aortic lesions in patients with aortoiliac occlusive disease, leading to thromboembolic complications in the postoperative period.

Limitations

In the presented groups, a lack of patients who had complications in early postoperative period could have affected the statistical analysis.