Cerebral venous thrombosis is a rare complication of inflammatory bowel disease. There have been at least 30 cases reported in the world literature (1-23). Fewer than 10 of these cases involved children (3,8,17). Strokes appear to be more common in patients with ulcerative colitis than in those with Crohn's disease (22). The outcome in the colitis group is extremely poor; in a review of 15 cases of ulcerative colitis complicated by cerebral venous thrombosis, more than 80% of the patients had serious permanent neurologic sequelae or died (22).
The mechanisms of thrombogenesis in inflammatory bowel disease are still incompletely understood, although thrombocytosis; elevated factors V, VIII, and fibrinogen; and decreased antithrombin III have been well documented (22,24-26). Increased production of tumor necrosis factor α (27) and disruption of intestinal glycosaminoglycans (28) could also have important procoagulant effects. Transient protein S deficiency has been reported in one adult patient with active ulcerative colitis, who had a cerebral venous thrombosis (14); and there is one report of a patient with ulcerative colitis and transient combined protein C and S deficiency having a cerebral arterial thrombosis (29).
We report the unfortunate case of a 14-year-old girl diagnosed with a severe active proctocolitis, who died of multiple cerebral venous thrombosis within 24 hours of colonoscopy and the initiation of intravenous steroid treatment.
A 14-year-old girl was referred in March 1994 by her primary care physician with a history of lethargy, joint pains, loose stool, and documented iron-deficiency anaemia. Ten months previously, she had seen her physician complaining of frontal headaches and lassitude. At that time, a full blood count revealed a haemoglobin of 6 g/dl. Her ferritin was 4 ng/ml. She was admitted to her local hospital in May 1993 where she underwent a 3-unit blood transfusion and was subsequently discharged with a provisional diagnosis of dietary anaemia. She began taking iron supplements. In the next 6 months she was seen on three occasions at her local hospital. On each clinic visit she was noted to be well with no specific symptoms. However her haemoglobin fell from 10.8 to 10.3 g/dl during this period. In November 1993 she began taking folic acid in addition to the iron supplements. In March 1994 she revisited her physician and complained of a recurrence of tiredness. She also reported loose stools three to four times daily with a semiformed consistency and no blood loss. Other symptoms included episodic, vague, lower abdominal pains, unrelated to eating or defecation, and 3 months of symmetric arthralgias affecting her large joints, particularly her wrists and ankles. Apart from 200 mg ferrous sulfate twice daily and folic acid 5 mg once daily, she was taking no other medications. There was no family history of gastrointestinal disease. Although her appetite was poor, her weight had remained stable in recent months. Her menses had started 2 years before and had remained normal with a 28-day cycle throughout this period.
On examination she was pale and gaunt with a weight of 46.2 kg (tenth percentile) and a height of 157 cm (twenty-fifty percentile). She had a low-grade fever of 37.7 °C. Apart from a sinus tachycardia of 110, results of cardiovascular and respiratory examinations were normal. Her abdomen was soft, nontender, with no palpable organomegaly or masses. There were no focal neurologic abnormalities.
Initial laboratory investigations revealed a microcytic, hypochromic anaemia with a haemoglobin of 8 g/dl. Her white count was elevated at 13 × 109/l, with an 85% granulocytosis. The platelet count was normal at 289 × 109/l. Renal and liver biochemistries were normal. Her B12 level was normal, but the folate level was increased at 20 ng/ml. The level of C-reactive protein was elevated at 23.6 mg/l. The erythrocyte sedimentation rate (ESR) was only 13 mm/hr. The international normalized ratio (INR) and activated partial thromboplastin time (APPT) ratios were both normal. Three stool cultures were positive for cysts of Blastocystis hominis. No other pathogens, including Cryptosporidium were identified. Results of stool electron microscopy were negative. Midstream urine (MSU) and blood cultures were sterile. The 1-hour blood xylose absorption test was normal.
A week after her initial physician referral, she was admitted to the paediatric hospital where she underwent a 4-unit blood transfusion. A distal duodenal biopsy was taken using a Crosby capsule. The histopathologist commented that the villi were of normal length but that there were focal collections of polymorphs within the lamina propria associated with mild crypt hyperplasia. It was concluded that she had a mild, nonspecific, acute duodenitis. It also became apparent at this time that the patient was passing small amounts of blood rectally. Rectal examination confirmed this, and the presence of a granular mucosa was also noted. She was referred for colonoscopy. On the day before the procedure she was given two sachets of Picolax (Nordic Pharmaceuticals Ltd., UK). Apart from a low-grade fever of 37.5 °C, the patient was reported well. Three other abnormal parameters were noted on the day of her colonoscopy: Her white count had increased from 13 to 54.2 × 109/l, her platelet count from 289 to 971 × 109/l, and her plasma sodium was recorded at 153 mM/l. Apart from the raised sodium level, there was no other objective evidence of haemoconcentration, in that both the haematocrit and urea were measured at 0.273 and 1.1 mM/l, respectively, on the day of her colonoscopic examination. Blood and urine cultures were negative. She underwent the colonoscopy under intravenous sedation with 5 mg midazolam and 25 mg pethidine. The endoscopist documented a florid colitis with confluent erythema and no ulceration, extending from the rectum to the midtransverse colon, which was the limit of this examination. She was started on 100 mg intravenous hydrocortisone four times daily, Predsol (Evans Medical Ltd., UK) enemas twice daily and a low-residue diet. At 11:30 PM that night she appeared well. She cheerfully conversed with the night nurse and walked to the toilet without difficulty. The following morning at 6:30 AM, she was found cold and blue with fixed, dilated pupils. All attempts at resuscitation were unsuccessful.
Early haemorrhagic infarction was noted throughout the cerebrum, cerebellum, and brain stem. Many of the small vessels adjacent to the infarcts contained fibrin thrombi and the walls were infiltrated by neutrophils. However, there was no necrosis of the vessel walls. Consequently the neutrophilic infiltrates were thought to be secondary to a thrombotic process rather than to a primary cerebral vasculitis. No significant abnormalities were found in the cardiovascular, respiratory, or renal systems. The stomach and small bowel were normal. She had a pancolitis with diffuse mucosal inflammation. There was no evidence of transmural inflammation or microscopic granulomas; consequently, the microscopic findings were thought to be more in keeping with ulcerative colitis, than with Crohn's disease. The liver revealed marked fatty change with patchy portal chronic inflammation, probably related to the preexisting colitis. Furthermore, neutrophils were noted in the sinusoids, suggesting antemortem sepsis. These findings from the coroner's postmortem evaluation were confirmed by an independent pathologist.
The cause of death was determined to be multifocal cerebral infarction associated with widespread venous thrombi, diffuse panproctocolitis, fatty liver with portal inflammation, and underlying systemic sepsis.
The association of thrombosis with inflammatory bowel disease is well known. Data from a review of autopsies showed 39% of patients who died with a diagnosis of inflammatory bowel disease had evidence of thromboembolic complications (30). In results of a clinical study the detected incidence was much lower at 1.2% (31). Neurologic thrombotic episodes are even less common, perhaps accounting for less than 10% of all thromboembolic complications. In a recent review, 42 case reports of strokes in association with inflammatory bowel disease were accrued from the world literature (22). The incidence was equal between sexes. The mean age of the patients was 31. The mean duration from diagnosis of inflammatory bowel disease to cerebrovascular complications was 30 months. Thirty of these cases were associated with ulcerative colitis and only 12 with Crohn's disease. Fourteen out of the 42 patients had a proven cerebral venous thrombosis. It was also documented that 60% of these inflammatory bowel disease patients were taking systemic steroids at the time of their stroke, 10% occurred after surgery, and 1 patient had ongoing sepsis. The mechanism for the increased thrombotic tendency is unclear. There appears to be a link between thrombocytosis, increased levels of factors V and VIII, and decreased antithrombin III (22,24-26). The relative importance of each of these factors is completely unknown, as is the significance of the procoagulant effects of increased production of tumor necrosis factor α (27) and intestinal glycosaminoglycan disruption (28). There is a single case report of an ulcerative colitis patient with protein S deficiency who sustained a cerebral venous thrombosis (14). There is another report of a colitis patient with combined protein C and S deficiencies who had a cerebral arterial thrombosis (29). Corticosteroids have also been implicated as thrombotic agents (32), because they can inhibit fibrinolysis that could potentially contribute to a hypercoagulable state (33). Conversely, their antiinflammatory action will have antithrombotic effects. This latter effect may be clinically more important: It is known that the majority of patients are having active exacerbations of their inflammatory bowel disease at the time at which they sustain a thrombotic complication.
It is interesting that this young girl initially went to her physician in May 1993 with frontal headaches. It is known that migraine is the second most common predisposing cause of cerebral infarction in the young (34). Alternatively, she may have had an episode of cerebral thrombosis at that time. Her sudden death a year later was presumably secondary to brain stem infarction, perhaps precipitated by her active colitis, sepsis, dehydration after receiving picolax, or the administration of intravenous steroids. There is no doubt that she had a panproctocolitis. The histologic findings are compatible with an active ulcerative colitis. During postmortem examination, no macro- or microscopic abnormalities were found in her small bowel. The significance of the Crosby-capsule biopsy findings of a mild, nonspecific duodenitis are unclear. In addition, she may have had underlying sepsis; but she had no specific symptoms indicating a focus of infection, except her diarrhoea. Her urine and blood cultures were sterile. Her chest radiograph was normal. The only organism isolated was B. hominis, on three separate occasions from her stool cultures. The pathogenicity of this protozoan remains controversial. Some authors consider it a pathogen associated with diarrhoeal illnesses or exacerbations of colitis (35-37). The evidence for this is circumstantial. That B. hominis can be cultured from the stools of patients with diarrhoea or colitis does not prove a causal association. A recent study found no statistical difference in the prevalence of B. hominis between asymptomatic and symptomatic patients (38). A further study on 12 patients with exacerbations of inflammatory bowel disease, who also had positive stool cultures for B. hominis, found that the majority of patients improved with steroids and medical therapy directed at the underlying illness (39). We believed that B. hominis was unlikely to be a significant pathogen in inflammatory bowel disease.
This case offers a salutary lesson to all gastroenterologists. It is difficult to offer definitive recommendations when the underlying mechanisms of systemic thrombosis in inflammatory bowel disease are still incompletely understood. However a case can be made for considering treatment with heparin in patients with active colitis associated with marked thrombocytosis. The beneficial effects of heparin in the treatment of refractory colitis are increasingly recognised (40,41). Furthermore, heparin has also been shown to be useful in the treatment of some of the systemic complications of ulcerative colitis (pyoderma gangrenosum), in which increased neutrophil endothelial adherence can be inhibited by a variety of heparin's antithrombotic and antiinflammatory effects, including antagonism of neutrophil elastase, leukocyte heparanase, and binding lactoferrin. A comprehensive review of these and other effects can be found in Dwarakanath et al. (42)
This is the first case report of a young patient with an active ulcerative colitis and B. hominis infection, complicated by fatal multiple cerebral venous thromboses.
1. Yerby MS, Bailey GM. Superior sagittal sinus thrombosis ten years after surgery for ulcerative colitis. Stroke
2. Schneiderman JH, Sharpe JA, Sutton DM. Cerebral and retinal complications of inflammatory bowel disease. Ann Neurol
3. Mayeux R, Fahn S. Strokes and ulcerative colitis. Neurology
4. Sigsbee B, Rottneberg DA. Sagittal sinus thrombosis as a complication of regional enteritis. Ann Neurol
5. Ryan FP, Timperley WR, Preston FE, Holdsworth CD. Cerebral involvement with disseminated intravascular coagulation in intestinal disease. J Clin Pathol
6. Rousseau B, Belanger C, Lacoursiere Y, et al. Colite ulceruse et thrombose veinuse cerebrale. Union Medicale du Canada
7. Harrison MJ, Truelove SC. Cerebral venous thrombosis as a complication of ulcerative colitis. American Journal of Digestive Diseases
8. Markowitz RL, Ment LR, Gryboski JD. Cerebral thromboembolic disease in paediatric and adult inflammatory bowel disease: Case report and review of the literature. J Pediatr Gastroenterol Nutr
9. Kiff RS, Denton GW. Fatal cerebral venous thrombosis complicating ulcerative colitis. Am J Gastroenterol
10. Cherqui B, Grippon P, Lubetski C, et al. Une complication rare des colites inflammatoires. La thrombophlebite cerebrale. Presse Med
11. Usui Y, Matsukawa M, Kobayashi S, et al. A case of ulcerative colitis causing superior sagittal sinus thrombosis. Rinsho Hoshasen
12. Bewermeyer H, Dreesbach HA, Krakamp B. Zerebrale venen und sinusthrombose bei colitis ulcerosa. ROFO: Fortschritte auf dem Gebiete der Rontgenstrahlen und der Nuklearmedzin
13. Besson I, Mannant PR, Neau JP, Van der Marcq P, Beau P, Beauchant M. Thrombo-phlebite cerebrale au cours de la rectocolite hemorragique. Gastroenterol Clin Biol
14. Vaezi MF, Rustagi PK, Elson CO. Transient protein S deficiency associated with cerebral venous thrombosis in active ulcerative colitis. Am J Gastroenterol
15. Kermode AG, Ives FJ, Taylor B, Davis SJ, Carrol WM. Progressive dural venous thrombosis treated with local streptokinase infusion. J Neurol Neurosurg Psychiatry
16. Fukudome T, Shirabe S, Yoshimura T, Tsujihata M, Nagataki S. A case of cerebral venous thrombosis occurring in ulcerative colitis. Rinsho Shinkeigaku
17. Calderon A, Wong JW, Becker LE. Multiple cerebral venous thromboses in a child with ulcerative colitis. Clin Pediatr
18. Musio F, Older SA, Jenkins T, Gregorie EM. Case report: Cerebral venous thrombosis as a manifestation of active ulcerative colitis. Am J Med Sci
19. Cecchi L, Gianni R, Santoro E, Leopardi A, Taccetti G, et al. Trombosi del seno sagittale superiore e colite ulcerosa. Descrizione di un caso. Annali Italiani di Medicina Interna
20. Moriyama E, Shinohara C, Tokunaga K, Kamitani M, Norikane H. Cerebral sinus thrombosis in patients with ulcerative colitis. A case report. Neurol Med Chir
21. Garcia-Monco JC, Gomez-Belderrain M. Superior sagittal sinus thrombosis complicating Crohn's disease. Neurology
22. Johns DR. Cerebrovascular complications of inflammatory bowel disease. Am J Gastroenterol
23. Teggi A, Della Rocca C, Traditi F, Lanzalone CM, De Rosa F. Colite ulcerosa, trombosi dei seni cavernosi e del plesso prostatico e leuconevrassite acuta necrotizzante emorragica: Decrizione di un caso di associazione sindromica. Medicina Firenze
24. Conlan MG, Haire WD, Burnett DA. Prothrombotic abnormalities in inflammatory bowel disease. Dig Dis Sci
25. Dejong E, Porte RJ, Knot E, et al. Disturbed fibrinolysis in patients with inflammatory bowel disease. A study in blood plasma, colon mucosa and faeces. Gut
26. Talstad I, Rootwelt K, Gjone E. Thrombocytosis in ulcerative colitis and Crohn's disease. Scand J Gastroenterol
27. Murch SH, Braeger CP, Walker-Smith JA, MacDonald TT. Location of tumour necrosis factor α by immunohistochemistry in chronic inflammatory bowel disease. Gut
28. Murch SH. Sulphation of proteoglycans and intestinal function. J Gastroenterol Hepatol
29. Jorens PG, Hermans CR, Haber I, Kockx MM, Vermylen J, Parizel GA. Acquired protein C and S deficiency, inflammatory bowel disease and cerebral arterial thrombosis. Blut
30. Graef V, Baggenstoss AH, Sauer WG, et al. Venous thrombosis occurring in non specific ulcerative colitis. A necropsy study. Arch Intern Med
31. Talbot RW, Heppel J, Dozois RR, et al. Vascular complications of inflammatory bowel disease. Mayo Clin Proc
32. Paradis K, Bernstein ML, Adelson JW. Thrombosis as a complication of inflammatory bowel disease in children: A report of four cases. J Pediatr Gastroenterol Nutr
33. Gerrits WB, Flier OT, Meer J van der, et al. Corticosteroids and experimental intravascular coagulation. Scandinavian Journal of Haematology
34. Hilton Jones D, Warlow CP. The causes of stroke in the young. J Neurol
35. Vannatta JB, Adamson D, Mullican K. Blastocystis hominis
infection presenting as recurrent diarrhoea. Ann Intern Med
36. Sheehan DJ, Raucher BG, McKitrick JC. Association of Blastocystis hominis
with signs and symptoms of human disease. J Clin Microbiol
37. Russo AR, Stone SL, Taplin ME, Snapper HJ, Doern GV. Presumptive evidence for Blastocystis hominis
as a cause of colitis. Arch Intern Med
38. Udkow MP, Markell EK. Blastocystis hominis:
Prevalence in asymptomatic versus symptomatic hosts. J Infect Dis
39. Nagler J, Brown M, Soave R. Blastocystis hominis
in inflammatory bowel disease. J Clin Gastroenterol
40. Gaffney PR, O'Leary JJ, Doyle CT, et al. Response to heparin in patients with ulcerative colitis. Lancet
41. Gaffney PR, Doyle CT, Hogan J, Gaffney A, et al. Paradoxical response to heparin in 10 patients with ulcerative colitis. Am J Gastroenterol
42. Dwarakanath AD, Yu LG, Brookes C, Pryce D, Rhodes JM. Sticky neutrophils, pathergic arthritis, and response to heparin in pyoderma gangrenosum complicating ulcerative colitis. Gut