Successful treatment of acute spleno-porto-mesenteric vein thrombosis after ChAdOx1 nCoV-19 vaccine. A case report

Abstract

Several cases of deep venous thrombosis in people who had recently received Vaxzevria (previously known as COVID-19 Vaccine AstraZeneca) have recently been reported, mainly presenting as cerebral vein/cerebral venous sinus thrombosis. This syndrome has been termed “vaccine-induced immune thrombotic thrombocytopenia (VITT)”. Acute spleno-porto-mesenteric vein thrombosis is an uncommon but serious condition with potential sequelae, such as small-bowel gangrene and end-stage liver failure. We describe a case of concomitant thrombosis of portal, superior mesenteric and splenic veins in a young female patient with no other risk factors who received Vaxzevria (previously ChAdOx1 nCoV-19 vaccine, AstraZeneca) 17 days before. The diagnostic workup and the successful endovascular treatment and systemic anticoagulation management is reported.

1. Background

Vaccination is considered as the critical weapon in the battle against COVID-19. In March 2021, several cases of deep venous thrombosis in people who had recently received Vaxzevria (previously known as COVID-19 Vaccine AstraZeneca) have been reported, mainly presenting as cerebral vein/cerebral venous sinus thrombosis. This syndrome has been termed vaccine-induced immune thrombotic thrombocytopenia (VITT). We present a case of a 36-year-old woman who developed a spleno-porto-mesenteric vein thrombosis few days after vaccination with Vaxzevria.

2. Case presentation

A previously healthy, Caucasian Italian 36-year-old female patient presented to the Emergency Departement with a 10-day history of upper abdominal pain. 17 days before she received her first dose of Vaxzevria COVID-19 Vaccine. A few days after the vaccination she suffered from a short course of fever, asthenia and diffuse ostheoarticular pain. She was not taking oral contraceptives, nor did she report a personal or familiar history of venous thromboembolism or thrombophilia.

On admission she was normotensive, vital parameters and physical examination were normal. Bloods showed a mild thrombocytopenia (133 * 10³/μL, normal values 150–350), while the remaining tests were normal. The nasopharingeal swab ruled out SARS-COV-2 infection. Chest X-ray was negative, her electrocardiogram showed sinus rhythm 98 b/min. Emergency US and a subsequent abdominal CT with contrast showed complete thrombosis of spleno-mesenteric-portal axis (Fig. 1 ), with abdominal and pelvic free fluid. A continuous infusion of unfractioned heparin was started and the patient was admitted to the ICU for close monitoring of her disease course and adverse events.

Fig. 1.

Contrast-enhanced MDCT scan of the abdomen.

Coronal MIP reconstruction that demonstrates splenic, portal and superior mesenteric veins massive thrombosis (arrowheads).

Considering the thrombotic burden, on the following day the patient underwent trans-hepatic portal vein venography via right internal jugular vein access that confirmed the complete thrombosis of splenic, superior mesenteric and portal vein (Fig. 2 A). Thrombus aspiration with a Penumbra catheter was performed (Fig. 2B). A porto-systemic shunt was performed between the right suprahepatic and the right portal vein. Venous pressure measures are reported in Table 1 . Ultrasound-accelerated thrombolysis was performed with two EkoSonic ultrasound-enhanced infusion catheters inserted through the right internal jugular vein. A 1 mg/h rtPA infusion was started and continued for 24 h. The results of coagulation tests are reported in Table 2 .

Fig. 2.

Venography of the splanchnic veins during the first angiography.

Panel A: Venography of the splenic vein that confirms portal, splenic and superior mesenteric veins massive thrombosis. Panel B: Venography of the portal and superior mesenteric vein during thrombus aspiration manoeuvres with partially restoration of blood flow into these vessels.

Table 1.

Values of venous pressures (in mmHg) in the different sites involved in the thrombosis, during the first and the second angiography (performed one week after the first).

Site of measurement	1st angiography	2nd angiography
Right atrium	15	18
Portal vein	n.a.	21
Portal vein (TIPS occluded)	n.a.	31.5
Splenic vein	26	22
Superior mesenteric vein	24	22
TIPS	17	20.5

TIPS: Transjugular Intrahepatic Porto-systemic Shunt; n.a. not available.

Table 2.

Results of the coagulation tests performed.

Test	Unit of measure	Normal values	Result
Prothrombin time	sec		13
INR	INR	0.8–1.2	1.08
Activated partial thromboplastin time	ratio	0.8–1.2	0.96
Fibrinogen	mg/dl	150–450	501
Protein C	%	72–142	78
Protein S	%	65–128	44
Antithrombin III	%	70–120	68
IgG/IgM Anti-cardiolipin Antibodies	U/ml	<20	<2.6
IgG/IgM Anti-b2 glycoprotein Antibodies	U/ml	<20	7
Lupus anticoagulant		Negative	Negative
Paroxysmal nocturnal hemoglobinuria clone		Negative	Negative
Factor V Leiden mutation		Absent	Absent
Factor II mutation		Absent	Absent
Anti-heparin PF4 antibodies	OD	0.4	2.503

INR: International normalized ratio. PF4: platelet factor 4. OD: Optical density.

Anti-PF4/heparin immunoglobulins G were measured by an enzyme-linked immunosorbent assay (ELISA, PF4 Enhanced test, Immucor, Norcross, Georgia, US); the cut-off for normal values is <0.4 Optical Density.

Serum and urine protein electrophoresis excluded a myeloproliferative disorder. Given the findings of anti-heparin PF4 antibodies, the continuous infusion of heparin was discontinued and a continuous infusion of argatroban 2 μg/kg/min, with a target aPTT of 45″, was started. A diagnosis of an autoimmune HIT was suggested, and a 5-day course of 0.4 g/kg intravenous Immunoglobulin was started.

During the second day of ICU stay, the patient developed melena and the haemoglobin levels dropped to 5.6 g/dl; her BP dropped to 90/60 mmHg and heart rate rose to 130 b/min. Three units of packed red blood cells were transfused, with normalisation of blood pressure and heart rate. Urgent esophagogastroduodenoscopy was performed and several vascular lesions at greater curvature and gastric fundus were found, which were treated with local epinephrine injection with no further evidence of bleeding.

The patient stayed in the ICU for 6 further days. Her clinical conditions remained stable. A mesenteric angiography was repeated on the 7th day. The superior mesenteric, portal and splenic veins were patent, and a partial reperfusion right intrahepatic portal vein branches (Fig. 3 ). Venous pressures are reported in Table 1. The patient was then eventually discharged from the ICU and transferred to a gastroenterology unit in good conditions. Long-term anticoagulation with apixaban was started. At 5 weeks of follow up no relapse had occurred.

Fig. 3.

Venography of the splanchnic veins during the second angiography.

Venography of the superior mesenteric (Panel A) and splenic-portal veins (Panel B) 7 days after the first endovascular procedure demonstrates good recanalization of the whole venous system (arrowheads) and good patency of the TIPS (arrow).

3. Discussion

We report the case of a young, healthy, female patient in whom total spleno-porto-mesenteric thrombosis developed few days after Vaxzevria (previously COVID-19 Vaccine AstraZeneca), associated with thrombocytopenia and antiPF4 antibodies with no prior history of heparin exposure.

As of late March 2021, >20 million people in Europe and the UK had received the Vaxzevria vaccine. Cases of thrombosis and thrombocytopenia, mainly presenting as cerebral vein/cerebral venous sinus thrombosis, have been reported in persons who had recently received the vaccine, mostly within 14 days after vaccination. In some of these cases, mainly in women under 55, a combination of thrombosis and thrombocytopenia and bleeding, was described and attributed to an autoimmune form of Heparin-induced thrombocytopenia (HIT), which was termed vaccine-induced immune thrombotic thrombocytopenia (VITT) [1].

During the review process of this report, several other papers [[2], [3], [4], [5], [6], [7]] described cases of a new syndrome characterized by thrombocytopenia and deep thrombosis, which developed 5 to 24 days after administration of ChAdOx1 nCoV-19, a recombinant chimpanzee adenoviral-vector COVID-19 vaccine. Most of the patients were young (<50 years), healthy women, and a remarkably high percentage of them had thromboses at unusual sites, i.e. cerebral venous sinus or splanchnic thrombosis. All patients presented with concomitant thrombocytopenia; notably, none of them had received any form of heparin before onset of symptoms, and they all tested positive for the PF4-heparin antibodies. In total, the case-series describe the course of 43 patients; of these, 8 (18.6%) had splanchnic vein thrombosis. The European Medicines Agency estimated the incidence of this condition to be about 1 in 100,000 vaccinated people, with cerebral vein thrombosis being about three times more common than splanchnic vein thrombosis [8].

HIT is a prothrombotic condition caused by antibodies targeting platelet factor 4, which form PF4/heparin/IgG immune complexes and further activating platelets in a heparin-dependent fashion. Autoimmune HIT (aHIT) refers to a specific form of HIT in which of antiPF4 antibodies cause platelet activation despite not having previously received heparin [9]. We found antibodies against PF4, which induce massive platelet activation via the Fc receptor in analogy to heparin-induced thrombocytopenia (HIT mimicry), possibly as part of the inflammatory reaction and immune stimulation subsequent to the vaccine administration [10]; this lead to a treatment with a 5-day course of intravenous Immunoglobulin. Given the autoimmune process, steroids were also considered steroids for treating VITT. However, we didn't want to prevent the immune response to vaccine and platelet response to intravenous Immunoglobulin was complete. For the same reason plasma exchange, albeit potentially indicated in case of treatment failure, was not considered.

Acute thrombosis of the splanchnic venous circulation is a rare condition in the general population, accounting for about 5% of all abdominal ischaemic events [11]. Nevertheless, it is of vital importance to the clinician, given it clinical non-specific and highly variable presentation [12]. Clinical expression varies from asymptomatic cases to abdominal cramps, nausea, anorexia, vomiting, diarrhoea, and/or melaena [13]; potentially life-threatening severe manifestations include mesenteric ischaemia and variceal bleeding and progressive liver failure [14]. Improved imaging techniques have led to an increased recognition of this condition [15], allowing clinicians to treat the thrombosis conservatively, in an effort to avoid bowel resection and its associated complications [16].

Our case presented significant management challenges because the three major veins comprising the portal circulation were all acutely occluded. Moreover, since this was categorized as an unprovoked deep vein thrombosis event in a young patient, thrombophilia testing was performed as suggested [17]. While we were waiting for the blood test results, anti-coagulation therapy was started with an unfractioned heparin continuous infusion. Blood tests excluded other forms of thrombophilia, such as protein C or protein S deficiency, antithrombin III deficiency, disseminated intravascular coagulation, antiphospholipid syndrome or the presence of lupus anticoagulant. The finding of antiPF4 antibodies suggested a possible diagnosis of aHIT or the newly described VITT [1].

Intravenous immunoglobulin and aggressive anticoagulation with argatroban was started as suggested [9], with the aim of interrupting HIT antibody-induced platelet activation, leading to rapid platelet count recovery. Although the role of heparin in the pathogenesis of VITT is uncertain, to date there is no evidence that heparin dislodges PF4 bound to endothelium and other cellular surfaces also in VITT, thus increasing anti-PF4/heparin antibodies. Thus, according to ISTH recommendations [18] non-heparin anticoagulants are recommended. We chose argatroban for its short half-life in a patient with high bleeding risk.

As far as the specific treatment for the spleno-porto-mesenteric vein thrombosis is concerned, there is not a universally accepted protocol for management and treatment [13]. Systemic anticoagulation has shown a six-month recanalization rate of about 50%, as well as a failure rate of 10% [19]. Local infusion of thrombolytic therapy, given via a transjugular route, was shown effective to provide recanalization [12]. Adjunctive endovascular techniques have recently been developed to reduce thrombolytic drug exposure, and improve efficacy compared with standard thrombolysis [20,21], which include balloon angioplasty, thrombectomy devices, ultrasound–accelerated thrombolysis, aspiration thrombectomy and TIPS creation. Ultrasound–accelerated thrombolysis involves the simultaneous endovascular delivery of low-intensity ultrasounds and thrombolytic agent. In in vitro studies, this technique proved effective in accelerating clot lysis by increasing clot permeability and penetration of the thrombolytic agent, exposing additional plasminogen receptor sites to the thrombolytic agent [22]. In a multicentre, retrospective report, this technique was shown to be associated to a reduced drug infusion time, with a greater incidence of complete lysis and a reduction in bleeding rates [23]. In the case reported, effective venous recanalization was achieved; however, a major haemorrhagic complication occurred. Gastric bleeding, despite potentially associated with splenic vein thrombosis [24], is most likely a major complication of the systemic anticoagulation and the regional thrombolysis. In our centre, the EkoSonic ultrasound-enhanced thrombolysis system is routinely used in cases of intermediate-high risk pulmonary embolism. The decision to use the device in this case of splanchnic thrombosis, as well as the prolonged duration of the rtPA infusion, was empiric and based on our previous experience, as well as on the significant thrombotic burden. Given the only recent description of VITT, no data are available for the risk of recurrence. Hence, long-term duration of anticoagulant therapy is still an open issue.

4. Conclusion

While the vaccination campaign to progress, further studies will be required to allow for a better understanding of the real incidence and the exact pathogenesis, as well as the optimal treatment of this condition. In the meantime, we suggest that to (1) consider spleno-porto-mesenteric thrombosis in the differential diagnosis of epigastric abdominal pain, (2) perform a complete thrombotic work-up to elucidate abnormalities that could be contributing to a pro-thrombotic state and (3) initiate aggressive endovascular and systemic measures in order to avoid decompensation and a significant adverse outcome.

Declarations of interest

None.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.