Introduction  

Diagnosis and treatment of vascular malformations (VASM) of the maxillofacial region (MFR) is still an urgent problem. According to many authors, the detection rate of VASM localized in the MFR reaches 60-80%, while patients with venous malformations (VM) account for about 40% of the total number of patients with VASM localized in the head and neck region [1, 2].

Ultrasound examination (US), being non-invasive, safe, easily accessible, and without contraindications, the method is currently one of the standard methods for the diagnosis of vascular pathology [3]. US allows visualizing the presence of volume formations, including VM, in the soft tissues of the MFR. Portable ultrasound devices take a special place among modern ultrasound diagnostic devices. They allow performing a diagnostic procedure directly in the operating room.

Here are the following indications for the US of the soft tissue in patients with suspected malformation:

• VASM verification.
• determination of the structure of the malformation (venous or arteriovenous).
• determination of the depth of the malformation location.
• navigation in the treatment of VM by sclerosing with the foam form of ethoxysclerol.
• specifying the size of the VM when planning a surgical operation.
• assessment of the condition of the surrounding soft tissues. US can be used to improve the safety and effectiveness of sclerosis of VM surface [4].

Despite great achievements in modern maxillofacial surgery, the treatment of VM of MFR cannot be considered completely resolved yet, since none of the methods of treating VM is universal, and therefore the problem remains relevant. Considering the anatomical features of MFR, the proximity of large vessels, sensory and motor nerves, vital organs (ENT, visual organs, brain), this area is one of the most difficult for vascular, maxillofacial, and plastic surgeons [5].

The most common methods of treatment of VM of MFR include surgical removal and sclerosis with 70% ethanol. Recently, such treatment methods as laser obliteration and microwave therapy have become widespread.

Such a wide range of methods of treatment of VM of MFR is due to the peculiarity of their clinical manifestations, localization, and size.

Surgical treatment is advisable to use if VM can be removed entirely within healthy tissues without significant cosmetic damage and functional impairment, in cases when the use of other methods of treatment is not so effective [6].

The use of 70% ethanol for sclerosing VM is based on its properties of causing cell death by causing cell membrane lysis, protein denaturation followed by vascular occlusion. Injection of 70% ethanol is accompanied by severe pain, the possibility of coagulation necrosis of the surrounding organs and tissues with the development of persistent paresis of the branches of the trigeminal and facial nerves. Also, when sclerosing medium and large malformations, 70% ethanol causes a toxic effect on the entire body [7].

The choice of an optimal treatment method of VM of MFR is especially important when they are localized in aesthetically significant areas. An optimal treatment method is determined only on an individual basis, considering clinical manifestations, localization, size, depth, and the number of anatomical areas affected by VM.

Materials and Methods

We analyzed the archival records of the Department of Maxillofacial and Plastic Surgery of the Moscow State University of Medicine and Dentistry for the period from 1997 to 2016. This analysis showed that out of 742 patients who were on inpatient treatment with various forms of VASM of the face, neck, organs, and tissues of the oral cavity, 450 (60.6%) were diagnosed with VM. According to the statistics of the polyclinic division of the Department of Maxillofacial and Plastic Surgery of the Moscow State Medical University for the period from 2010 to 2017, out of 276 patients with VASM of MFR, 188 (68.1%) were diagnosed with VM. All these numbers indicate that the largest number of patients with vascular pathology who were treated in inpatient and outpatient settings are patients with VM. These factors make the development and implementation of modern, effective, and minimally invasive treatment methods urgent.

The diagnosis of VM of MFR, especially those located superficially, is based on the typical clinical manifestations and anamnesis of the disease [8] and does not present great difficulties. The clinical method of the examination remains the main method of diagnosing VM [9, 10]. Often examination, palpation, and the presence of a positive symptom of "filling" are enough to make a correct diagnosis. Difficulties may occur in the diagnosis of deep-seated VM due to the absence of external clinical manifestations. The etiology of this disease can be different. According to the analysis of the archival records of the Department of Maxillofacial and Plastic Surgery of the Moscow State Medical University, for the period from 1997 to 2016, most commonly VM was diagnosed from birth (54.4%), 20.9% of patients associate the occurrence of VM with sustained injuries, 23.1% cannot associate the occurrence of VM with any causes (Figure 1).

From 2015 to 2018, the Department of Maxillofacial and Plastic Surgery of I.M.Sechenov First Moscow State Medical University examined and treated 60 patients aged 18 to 80 years (Figure 2), including 21 (35%) men and 39 (65%) women. In 44 patients, VM occupied one anatomical area, in 16 patients - 2 areas or more.

Predominant localization of VM was upper lip and cheek area in 42 (70%) patients (Figure 3).

The size and shape of VM of MFR were diverse; most of the small, medium-sized, and relatively regular rounded shape (Table 1).

Our proposed method of treatment of patients with VM of MFR consisted of sclerosing it with 3% ethoxysclerol in the foam form, which had been made in an operating room according to the method of L. Tessari (2000) [11]. This technique is also used in the clinical practice of phlebologists.

Ethoxysclerol is a detergent that damages the cytoplasmic membrane of target endothelial cells lining the inner wall of blood vessels. As a result, platelets are collected at the site of injury and attached to the venous wall. Then the vessel is clogged with a dense network of platelets, cellular debris, and fibrin. At a later stage, the obliterated vein is replaced by fibrous-adipose tissue.

The technique consists of combining two disposable plastic syringes with the capacity of 5 ml, connected through a three-way tap at an angle of 90°, which is used for infusion. One syringe was filled with 1 ml of 3% ethoxysclerol, and the other with 4 ml of air (in a sterile operating room setting). After that, the pistons of the syringes were moved alternately and rapidly (back and forth), mixing the contents from one syringe to another (mixing air with sclerosant) until 5 ml of a white foam emulsion was formed. Generally accepted criteria for preparing foam for injection into VM cavity is 20-25-fold mixing of it from one syringe to another syringe.

Sclerosant in the micro-foam form changes its physical properties, improving distribution in the venous vessels, reducing the rate of drainage from vessel lumen, increasing the exposure time of surfactant to endothelial cells of the vascular wall [12].

During the preoperative period, all patients underwent US to determine the prevalence and size of VM. In addition to that, we paid attention to large vessels located near the VM, especially arteries, since ingress of a sclerosant into the arterial system is an absolute contraindication and can lead to extensive soft tissue necrosis.

Patients underwent an ultrasound on IU-22 devices ("Philips", Netherlands) using a high-frequency linear sensor of 5-17 MHz and a high-frequency linear intraoral sensor of 7-15 MHz. In addition to that, the US was performed with the help of a portable ultrasound device S2 ("SonoScape", China) using a linear sensor of 1-15 MHz in V-modes and Doppler modes.

The clinical criteria for a complete filling of VM with foam was a transition from a bluish color to a pale pink color and its compaction.

The evaluation of treatment results was carried out during routine follow-up examinations with mandatory ultrasound and photographing of patients on the 7, 14 days, and in 1 month after treatment.

In the future, control examinations were performed once every 6 months or at the patient's request.

Statistics

We determined the mean, median, standard deviation, minimal and maximal parameters. After testing of normality of distribution in all groups and inside of them in dynamics, we proved or declined null hypothesis 1 (‘There is no difference between different size VMs for main characteristics’), null hypothesis 2 (‘There is no difference between different size VMs in the US-square’), null hypothesis 3 (‘There is no difference for different size VMs in the US-square inside the groups in dynamics’). The null hypothesizes were approved for p-value > .05 and declined for p-value < .05.

For analysis of VM of the maxillofacial region, we assessed different forms with their characteristics in points and compared them with the Kruskal-Wallis test due to the lack of distribution normality and inside the groups in dynamics, the hypothesis was proved or declined with the T-Wilcox test. For statistical analysis, the program RStudio (R version 3.6.3 (2020-02-29) was used.

Results and Discussion

We marked the main features of VM besides size: localization, number of the involved anatomic areas, the deepness of process, closeness of facial nerve, and possibility of complications after treatment.

For analysis of different forms of VM of maxillofacial surgery we coded these characteristics with points:

Localization: 1 point - infraorbital area; 2- tongue; 3- lower lip; 4- upper lip; 5- cheek.

The number of anatomic areas: 1 point- only one zone; 2 – 2 and more zones.

Deepness of process: 1 point- superficial; 2- deep.

Closeness with facial nerve: 1 point- far; 2- closely.

Complications (minor as light hyperaemia, short-term pain): 0 point- absence; 1 point- presence.

The analysis results are in Table 2.

In the treatment of 52 (86.7%) of our patients, one session of sclerosis with the foam form of 3% ethoxysclerol was sufficient to achieve a positive result. At the same time, 8 (13.3%) patients needed several surgical interventions, which was associated with the large size of the VM. Of these 3 (5%) patients were treated with 3% ethoxysclerol foam as the preparation for surgical excision, and 5 (8.3%) patients were treated with laser obliteration to reduce the size of VM and risk of bleeding.

In 6 (10%) patients the sclerosis was performed in the hospital setting under general anesthesia, in 54 (90%) - under local anesthesia Sol. Articaini 1:200 000 to 2 ml.

After receiving a positive aspiration test, 3% ethoxysclerol was injected into the foam form under continuous ultrasound control until the foam filled the VM cavities.

Based on the ultrasound data, it is possible to calculate the volume of the VM and the volume of injected sclerosant in the foam form.

The volume of VM was calculated by the formula:

where V is the volume, L is the length, W is the width, H is the anteroposterior size, 0.5 is the constant coefficient.

For example V=5×7×2×0,5=35 ml.

The optimal amount of 3% ethoxysclerol in the foam form was selected depending on localization, size, and depth of the lesion, as well as determination of the volume of the VM, which corresponded to the approximate volume of the VM. Based on the results of the ultrasound, a decrease in the formation by 25-35% was noted on the 7th day after controlled sclerosis with a 3% solution of ethoxysclerol in the foam form. The zone of reduced echogenicity was visualized around malformation, in the projection of which multiple hyperechogenic inclusions were determined due to the accumulation of air bubbles, that indicated the presence of a sclerosant in the lumen of the vein. There were signs of edema in surrounding soft tissues. Clinically, it was difficult to determine the change in the size of the VM due to collateral edema and infiltration of the soft tissues surrounding the VM. The skin surface may be slightly hyperemic, bruising may occur. Also, soft tissues above the surface of the VM may be slightly compacted and painful during palpation.

The dynamics of VM after treatment according to the US results are introduced in Table 3.