Central venous sinus thrombosis in a young girl with ulcerative colitis

Abstract

Paediatric central venous sinus thrombosis (CVST) is an uncommon but important life-threatening complication of inflammatory bowel disease (IBD). As the incidence of IBD has increased in the last four decades, paediatricians need to be aware of this complication. There is currently no consensus on when children with IBD should receive prophylactic anticoagulation. We present the case of a young girl with ulcerative colitis who suffered an acute ischaemic event secondary to a CVST during an acute flare of her disease. We aim to bring awareness to CVST in IBD due to its high risk of morbidity and mortality.

Background

Paediatric central venous sinus thrombosis (CVST) is an uncommon event with an incidence of 0.67 case per 100 000 children, mainly occurring in the neonatal period.¹ While thrombosis is a well-recognised complication of inflammatory bowel disease (IBD) in adults, its prevalence in children is less well known. CVST is an uncommon but important life-threatening complication of IBD. The incidence of IBD has increased dramatically over the last four decades with approximately 5 million individuals affected in North America and Europe. As 20% of these are diagnosed in childhood,² CVST is a complication that paediatricians need to be aware of.

Case presentation

We present the case of a young girl who attended our paediatric assessment unit with a 2-week history of general malaise, lethargy and pallor. She had no abdominal symptoms nor any relevant medical, surgical or family history. Ancillary tests showed her haemoglobin (Hb) was 46 g/L with a hypochromic microcytic anaemia and ferritin <1 µg/L. She was diagnosed with ‘severe anaemia of unknown origin’ and received multiple blood transfusions. She was discharged with a Hb of 115 g/L and scheduled for an oesophagogastroduodenoscopy (OGD). During her follow-up, it was noted that her Hb had dropped to 83 g/L 8 weeks after discharge.

Unfortunately, the girl was not brought to several appointments and her follow-up was interrupted. Four months after her initial presentation, she re-presented to the paediatric assessment unit. She now had a 4-week history of weight loss, abdominal pain, faecal urgency and blood in semiformed stools. Her weight had dropped from 39.8 kg (30th centile) 4 months earlier to 35.2 kg (25th centile). She was still awaiting OGD; however, based on her clinical presentation, IBD was suspected and she was started on intravenous methylprednisolone and intravenous antibiotics (cefotaxime and metronidazole). Her faecal calprotectin was >3000 µg/g.

Eight days into her admission, she was transferred to a tertiary paediatric gastroenterology centre for further management. On day 5 of this admission, she had three episodes of left focal seizures with secondary generalisation which were terminated by lorazepam.

Investigations

A contrast CT brain was performed which showed two focal areas of venous ischaemia/infarction in the left and right superior parietal lobe with two discrete separate focal filling defects consistent with sinus thrombosis in the superior sagittal sinus.

A magnetic resonance venogram (MRV) brain was performed 4 days later, which showed (1) focal signal abnormality evident in the posteroparietal cortex and subcortical white matter at the site previously described on CT; and (2) evidence of non-occlusive thrombus in the superior sagittal sinus. An OGD and colonoscopy the following day showed a severe pancolitis with IBD-unspecified, favouring ulcerative colitis (UC). She subsequently had a histological diagnosis of UC.

Treatment

She was started on therapeutic low-molecular-weight heparin (LMWH), and thankfully, our patient showed no focal deficits following the event.

Outcome and follow-up

She continued anticoagulation therapy for 6 months and has been in remission from UC for the past 5 years with no further thrombotic episodes.

Discussion

Thromboembolic events (TEs) are a recognised complication of IBD. As the risk of TE is lower in the general paediatric population, it is often assumed that the risk of TE in children with IBD is also lower. This potentially places children with IBD at risk due to a lack of appropriate TE prophylaxis.³

There are multiple hypotheses for the pathogenesis of TE in IBD. It is thought to be largely due to the inflammatory component of IBD which produces a hypercoagulable state with coagulation cascade activation, platelet activation and fibrinolysis anomalies.⁴ Thrombosis in IBD involves both systemic TE and localised microthrombi in the vasculature of the intestine.⁵ Notably, there is evidence that microvascular thrombosis may be a part of the disease process in UC as treatment with enteral extended colon-release tablets of LMWH has shown to be beneficial.⁶ Nylund et al considered 96% of all paediatric inpatient visits in the USA at a stratified random sample across five time periods: 1997, 2000, 2003, 2006 and 2009. They found that of 7 448 292 paediatric discharges 61 076 identified with Crohn’s disease and 7318 identified with UC. Children with no IBD had an absolute risk of TE of 50.4/10 000, whereas those with Crohn’s disease and UC had elevated absolute risks of 119.8 (relative risk (RR) 2.37) and 101.7 (RR 1.99), respectively. The RR of CVST in children with UC was 15.5 compared with 5.4 in those with Crohn’s disease. This paper highlights the increased prevalence of TE in children with IBD. The absolute risk of CVST in UC was 15.5/10 000 hospitalisations.⁷

A case of CVST was also reported in a 5-year-old boy with UC.⁸ He was in the perioperative period following a colectomy which places patients at higher risk of TE.

CVST is one of the most common causes of paediatric and neonatal stroke. The incidence is estimated at 0.6/100 000 per year, with 30%–50% occurring in neonates.^{9 10} It has a high morbidity with neurological sequelae in up to 40% of survivors and mortality approaching 10%. The clinical presentation of CVST is often variable, and patients may present with a constellation of symptoms depending on the site and extent of thrombosis. The most common presentation includes signs and symptoms of raised intracranial pressure such as headache, reduced visual acuity, papilloedema, focal neurological deficits and seizures.¹¹

Neuroimaging, typically a CT, is the principal basis for the diagnosis. Brain MRI and MRV can diagnose CVST with a high sensitivity and specificity.¹² Digital subtraction angiography is the ‘gold standard’ for diagnostic purposes in adults; however, its use in children is limited due to its invasiveness.¹³

CVST treatment should start immediately after diagnosis. Antithrombotic treatment during the acute phase is similar to that used in adults. Surprisingly, the evidence is weak, with no large, high-quality randomised trials. The guidelines from the European Stroke Organisation¹⁴ recommend the use of LMWH as initial anticoagulation. The optimal duration of the anticoagulation is uncertain. In patients with one episode of CVST and transient risk factors (eg, dehydration, drugs, surgery, infection), expert opinion and guidelines recommend 3–6 months of anticoagulation. For patients with one CVST episode without known risk factors, they recommend anticoagulation for 6–12 months, and in those with >1 CVST or a prothrombotic condition, they recommend lifelong anticoagulation.¹⁵ There is some weak evidence for the use of prophylactic folic acid supplementation. This is thought to be related to hyperhomocystinaemia which is a risk factor for TE and can be caused by low folate levels.¹⁶

As children with IBD are at increased risk of potentially life-threatening TE, the question remains, to whom should we give prophylactic anticoagulation? Various studies and guidelines^{17 18} have recommended prophylactic LMWH in adults with acute severe colitis; however, this is not the usual practice in paediatrics where the advice tends to consider the patient’s other risk factors for TE. These include prolonged immobility, surgery, tobacco use, oral contraceptive use, obesity, central venous catheters, concomitant infection, family history and known prothrombotic disease.¹⁹ The guidelines proposed by the Canadian Gastroenterology Association recommend prophylaxis only in obese adolescents with severe acute colitis undergoing surgery and in hospitalised patients with a previous history of thrombosis.²⁰

In our case, the patient only had one risk factor for developing thrombosis—prolonged immobility due to her hospital stay—so would not have been placed on prophylactic anticoagulation under these guidelines.

Learning points.

• Central venous sinus thrombosis (CVST) is a rare but potentially life-threatening complication of inflammatory bowel disease (IBD) in children. Physicians should be aware of the symptoms and signs including headaches, lethargy, focal deficits, diplopia and seizures.

• There is currently no consensus on when children with IBD should receive prophylactic low-molecular-weight heparin (LMWH). Clinicians should be aware of the risk of thromboembolic events in children with IBD and think prophylaxis.

• There is also no consensus on the length of treatment with LMWH for children with CVST. It should be guided by the clinical picture in discussion with our haematology colleagues.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).