Abstract
We present a rare case of a 47-year-old woman admitted to our maxillofacial surgery department for a large cervicofacial venous malformation. The patient underwent fluoroscopy-guided intralesional foam sclerotherapy with sodium tetradecyl sulfate and air under general anaesthesia. On awakening, after 48 h of endotracheal intubation, she displayed dysarthria and dysmetria. Her brain CT scan showed no haemorrhagic lesions. A chest CT scan ruled out a potential pulmonary embolism. Suspicion for a paradoxical embolism was high and echocardiography confirmed a patent foramen ovale, which acted as a passageway for the embolus. Transcranial ultrasound showed mild right-to-left heart shunting. The dysarthria and dysmetria disappeared gradually over 48 h, thus confirming a reversible ischaemic neurological deficit. A brain MRI performed 1 week later showed no ischaemic or haemorrhagic lesions. The patient recovered completely. She was advised cardiosurgical follow-up and discharged.
Background
Surgery has a long-standing role in venous malformation (VM) treatment. Nevertheless, surgical treatment of large VMs affecting highly functional areas such as the head and neck can lead to unacceptable results from an aesthetic as well as functional point of view.
When addressing such lesions, surgeons can count on alternative therapies, such as percutaneous intralesional injection of sclerosant substances, which allow for good to excellent functional and morphological results, coupled with a relatively low complication rate. In large VMs the procedure is better performed under fluoroscopic guidance (roadmap procedure).
Embolic brain ischaemia, specifically, has never been reported after intralesional sclerotherapy for VM, despite being described—though exceptionally—after intralesional sclerotherapy for varicose veins.
Embolic brain ischaemia following intralesional sclerotherapy is always due to paradoxical embolism (PDE), where a thrombus or an air bubble reaches the brain through a patent foramen ovale (PFO), which can be readily identified via transthoracic echocardiography during the preoperative assessment. Our experience suggests complete routine multistep work up of patients with VM candidate to intralesional sclerotherapy with an echocardiography, a quick, cost-effective and informing procedure.
Case presentation
A 47-year-old woman was admitted to our maxillofacial surgery department for a cervicofacial VM. The VM had been growing slowly since her birth and she underwent non-specified radiotherapy on the malformation 45 years before, without significant improvements.
The patient was otherwise healthy and not currently under any medication.
At admission, the VM extended from the right cervical region subcutaneously to the oral floor cranially, to the carotid space medially, to the posterior neck triangle posteriorly and to the manubrium sterni caudally (figures 1 and 2). Furthermore, the mass dislocated the hypopharynx, the cervical oesophagus, the larynx and the trachea. The mass could be observed as a bluish pyriform sinus bulge during an endoscopical laryngeal examination.
Figure 1.
Preoperative photograph of the patient. The venous malformation can be seen as a bluish/purple swelling extending from the chin down to the manubrium sterni.
Figure 2.
Sagittal image from the preoperative MRI. The image showing a contrast-enhanced mass extending from the floor of the mouth down to the sternum.
The VM had no connection to the intracranial vessels, as confirmed by an MR angiography of the head. Therefore, as a first line treatment, the patient underwent a fluoroscopy-guided intralesional sclerotherapy with sodium tetradecyl sulfate (STS) under general anaesthesia. Only 15 mL of 3% STS was used as foam. The STS was mixed with angiographic contrast medium and air 3:1:5, until reaching 45 mL foam volume. The aim was to obtain a partial reduction of the VM and to perform further periodic intralesional injections until substantial regression of the mass, keeping well below the 50 mL volume, which is considered a per administration limit in the current literature.
After the procedure, which was apparently uneventful, the patient was transferred to the intensive care unit, monitored and ventilated via an endotracheal tube to secure the upper airways after the procedure. This period of assisted breathing was considered prudent owing to the amplitude of the VM that was already displacing the trachea.
On awakening the patient and after removing the endotracheal tube, we noticed that she was dysarthric and dysmetric; this was confirmed by a formal examination by the neurologist on duty.
Investigations
After the neurologist confirmed dysmetria and dysarthria, the patient underwent a head CT scan in order to rule out potential haemorrhagic lesions: the CT reported no signs of haemorrhage or other findings.
After the patient showed progressive improvement of symptoms over the course of 48 h, we had a high suspicion of a reversible ischaemic neurological deficit, which could be due to a pulmonary embolism or a PDE. A chest CT scan with contrast medium allowed us to rule out a potential pulmonary embolism and showed signs of pneumonia with mild pleural effusion. However, echocardiography confirmed a PFO, which could have acted as a passageway for the embolus. A transcranial ultrasound analysis after administration of 100 mL of saline+air mixture confirmed a clear right-to-left heart shunting. A brain MRI performed 1 week after showed no ischaemic or haemorrhagic lesions and no signs of other focal injuries.
Differential diagnosis
After a general anaesthesia procedure and prolonged endotracheal ventilation, it is often difficult to tell neurological anomalies arising from benign and self-limiting cognitive slowing due to hypnotics and sedatives. Nevertheless, carefully performed complete neurological examinations allow identifying precise signs, such as dysarthria and dysmetria.
Sudden and non-traumatic onset of focal neurological deficits in patients with no predisposing factors whatsoever could be due to brain infarction or haemorrhagic events, often related to unknown intracranial aneurysms. Head CT scans can readily and swiftly identify haemorrhages, while ischaemic events can be identified by MRI with contrast only and might be due to PDE.
Furthermore, it has to be considered that either of two types of embolism might follow foam sclerotherapy: a paradoxical thrombotic embolism or a paradoxical air embolism. The former ensues a deep vein thrombosis following sclerotherapy. It is rare and symptoms occur after a few days or a week in most cases, with MRI showing vessel occlusion. The latter is due to air embolism by air bubbles and may arise from using large foam volumes. Its features are immediate onset, no clots, quick and in most cases complete recovery, and no permanent vessel occlusions in imaging. In this specific case, paradoxical air embolism appears as the soundest diagnosis.
If not for the fluoroscopic guidance, a misinjection of the foam into a cervical artery should have also been taken into account.
Treatment
After the injective procedure the patient was treated with intravenous dexamethasone 12 mg twice daily, with progressive tapering during the following 10 days to prevent and treat reactive cervical and upper airways oedema. The patient was also treated with subcutaneous calciparin 4000 IU three times daily during the next month starting on identification of neurological signs, after having properly ruled out any possible haemorrhagic event. Since the thoracic CT scan described signs of an initial lobar pneumonia, the patient was also treated with oral levofloxacin 500 mg four times a day for 14 days. Gastric protection with proton pump inhibitors (omeprazole 20 mg four times a day) was also prescribed for 1 month, in order to prevent corticosteroid gastric side effects.
Outcome and follow-up
The patient recovered completely and was discharged after 1 week with indications for cardiosurgical follow-up.
One month later the patient underwent uneventful closure of the PFO in order to carry on with the intralesional sclerotherapy schedule for her VM, which is currently ongoing.
The mass showed significant reduction, mostly in the lower portion, right from the first sclerotherapy session (figures 3 and 4).
Figure 3.
Postoperative photograph of the patient. Right from the first sclerotherapy procedure, the patient had a significant aesthetic gain. The mass is considerably less visible, especially in the neck region.
Figure 4.
Sagittal image from the postoperative MRI. A considerable reduction of the neck mass can be seen, mirroring the more than satisfying aesthetic results.
Discussion
Surgery has a definite role in the treatment VMs, dating back several decades.1
Carefully planned and conducted procedures allow for excellent functional and aesthetic results, given that the smaller the VM, the better the results. Nevertheless, when dealing with large VMs involving the head and neck areas, surgical treatment at times presents abysmal morphological and functional outcomes.2 Some specific procedures can aid in treating such large unresectable VMs.3
In order to avoid such wide resections, alternative procedures have been developed, such as percutaneous intralesional injection of sclerosant substances (eg, ethanol, bleomycin and STS). Sclerosants induce apoptosis and necrosis of the endothelial lining, which is followed by a cascade of luminal thrombosis and fibrosis, which ultimately leads to shrinkage and, sometimes, complete regression of the lesions, oftentimes after several sessions.4–7
STS is an ionic surfactant with a soapy consistency that is able to induce intimal necrosis and vascular closure and forms relatively stable foam when mixed with gases. STS is a less toxic alternative to ethanol and holds a lower rate of skin necrosis, nerve injuries and systemic complications.5 7–10
Minor known complications of VM sclerotherapy using STS include local swelling, pain, erythema and other skin complications that can span from blistering, hyperpigmentation and ulceration to necrosis. According to the literature, minor complications can affect up to 14% of patients.9 10 Major complications include nerve injuries, muscular contracture, compartment syndrome, haemolysis, renal toxicity with haemoglobinuria and thrombophlebitis; luckily in an acceptably low rate of patients (2% of cases).9 10
Cerebral embolism is usually not reported among possible major complications of intralesional sclerotherapy for VM, despite being recognised as a potential (although extremely rare) complication of intralesional sclerotherapy of varicose veins in patient with a PFO.11–14
Aside from paradoxical embolism, some authors reported that the release of high vasospastic substance concentrations (endothelin-1) after venous sclerotherapy may have a definite role in visual and neurological symptoms onset.15 16 In vitro models of sclerotherapy suggest a role for antiendothelin agents in preventing such neurological sequelae of sclerotherapy.17
The prevalence of PFO, which is virtually the sole source of PDE, is approximately 26% in the general population and can be readily identified by a simple transthoracic ecocardiography.18 Transthoracic echocardiography, the examination of choice for PDE diagnosis and prevention and cryptogenic cerebral vascular events management is an inexpensive outpatient examination that, despite its operator-dependency, provides insightful data on presence, position and width of PFOs, with sensibility and specificity nearing 100%.14 18
Given the high health cost of vascular cerebral events coupled with the young age of most patients with VM, and considering the lack of widely accepted guidelines for diagnosis and treatment of patients with VM, we strongly recommend including transthoracic echocardiography in the routine work up of patients with VM candidate to intralesional sclerotherapy. Whenever a patient is diagnosed with a PFO, the defect should be addressed surgically before undertaking any intralesional procedure.
Learning points.
Intralesional sclerotherapy is often the treatment of choice for massive venous malformation of the head and neck region.
Fluoroscopy-guided intralesional sclerotherapy represents the most effective and least risky technique for ample venous malformations.
Transthoracic echocardiography should be part of the routine work up of candidates to intralesional sclerotherapy for venous malformations.
Always take into account paradoxical embolism whenever a patient develops central neurological signs after procedures involving systemic vessels.
A patent foramen ovale should be addressed surgically before treating a patient with intralesional sclerotherapy.
Footnotes
Contributors: MM diagnosed the venous malformation. FA, DR and GC performed the procedure.
Competing interests: None.
Patient consent: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1.Colletti G, Valassina D, Bertossi D et al. Contemporary management of vascular malformations. J Oral Maxillofac Surg 2014;72:510–28. [DOI] [PubMed] [Google Scholar]
- 2.Belov S, Loose DA. Surgical treatment of congenital vascular defects. Int Angiol 1990;9:175–82. [PubMed] [Google Scholar]
- 3.Colletti G, Colombo V, Mattassi R et al. Strangling technique to treat large cervicofacial venous malformations: a preliminary report. Head Neck 2014;36:E94–8. [DOI] [PubMed] [Google Scholar]
- 4.Legiehn GM, Heran MK. A step-by-step practical approach to imaging diagnosis and interventional radiologic therapy in vascular malformations. Semin Intervent Radiol 2010;27:209–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hein KD, Mulliken JB, Kozakewich HP et al. Venous malformations of skeletal muscle. Plast Reconstr Surg 2002;110:1625–35. [DOI] [PubMed] [Google Scholar]
- 6.Tan KT, Kirby J, Rajan DK et al. Percutaneous sodium tetradecyl sulfate sclerotherapy for peripheral venous vascular malformations: a single-center experience. J Vasc Interv Radiol 2007;18:343–51. [DOI] [PubMed] [Google Scholar]
- 7.De Lorimier AA. Sclerotherapy for venous malformations. J Pediatr Surg 1995;30:188–94. [DOI] [PubMed] [Google Scholar]
- 8.Siniluoto TM, Svendsen PA, Wikholm GM et al. Percutaneous sclerotherapy of venous malformations of the head and neck using sodium tetradecyl sulphate (sotradecol). Scand J Plast Reconstr Surg Hand Surg 1997;31:145–50. [DOI] [PubMed] [Google Scholar]
- 9.Adatia S, Nambiar V, Kapadia R et al. Acute ischemic stroke caused by paradoxical air embolism following injection sclerotherapy for varicose veins. Neurol India 2013;61:431–3. [DOI] [PubMed] [Google Scholar]
- 10.Ma RW, Pilotelle A, Paraskevas P et al. Three cases of stroke following peripheral venous interventions. Phlebology 2011;26:280–4. [DOI] [PubMed] [Google Scholar]
- 11.Hahn M, Schulz T, Jünger M et al. Late stroke after foam sclerotherapy. Vasa 2010;39:108–10. [DOI] [PubMed] [Google Scholar]
- 12.Hanisch F, Müller T, Krivokuca M et al. Stroke following variceal sclerotherapy. Eur J Med Res 2004;9:282–4. [PubMed] [Google Scholar]
- 13.Ceulen RP, Sommer A, Vernooy K. Microembolism during foam sclerotherapy of varicose veins. N Engl J Med 2008;358:1525–6. [DOI] [PubMed] [Google Scholar]
- 14.Nayor M, Maron BA. Contemporary approach to paradoxical embolism. Circulation 2014;129:1892–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Frullini A, Felice F, Burchielli S et al. High production of endothelin after foam sclerotherapy: a new pathogenetic hypothesis for neurological and visual disturbances after sclerotherapy. Phlebology 2011;26:203–8. [DOI] [PubMed] [Google Scholar]
- 16.Frullini A, Barsotti MC, Santoni T et al. Significant endothelin release in patients treated with foam sclerotherapy. Dermatol Surg 2012;38:741–7. [DOI] [PubMed] [Google Scholar]
- 17.Frullini A, Da Pozzo E, Felice F et al. Prevention of excessive endothelin-1 release in sclerotherapy: in vitro and in vivo studies. Dermatol Surg 2014;40:769–75. [DOI] [PubMed] [Google Scholar]
- 18.Mirzada N, Ladenvall P, Hansson PO et al. Multidisciplinary management of patent foramen ovale (PFO) and cryptogenic stroke/TIA. J Multidiscip Healthc 2013;6:357–63. [DOI] [PMC free article] [PubMed] [Google Scholar]