Anticoagulant-induced intramural intestinal hemorrhage
Brief Report
Anticoagulant-induced intramural intestinal hemorrhage
Chia-Ying Tseng MDa,b, Ju-Sing Fan MDb,c,e, Shu-Chuan Yang RNc, Hsien-Hao Huang MDb,c,e,?, Jen-Dar Chen MDd,
David Hung-Tsang Yen MD, PhDb,c,e, Chun-I Huang MDb,c
aEmergency Department, China Medical university hospital, Taiwan 40447
bEmergency Medicine, College of Medicine, National Yang-Ming University, Taiwan 11221 cDepartment of Emergency Medicine, Taipei-Veterans General Hospital, Taiwan 11217 dDepartment of Radiology, Taipei-Veterans General Hospital, Taiwan 11217
eInstitute of Emergency and Critical Care Medicine, College of Medicine, National Yang-Ming University, Taiwan 11221
Received 24 May 2009; revised 31 July 2009; accepted 3 August 2009
Abstract
Background: Long-term use of warfarin can provide benefits in the treatment of many diseases, but adverse bleeding events are unpreventable because of a narrow therapeutic range.
Objective: The aim of this retrospective chart review with data abstraction was to investigate the clinical presentations of intestinal intramural hemorrhage in emergency department (ED) patients.
Methods: We reviewed the cases of 17 patients with acute abdominal pain in our ED. Medical records including demographic data and results of Abdominal computed tomography were retrospectively reviewed and analyzed.
Results: The mean +- SD age of the reviewed patients was 77.7 +- 8.5 years (range, 60-93 years). The mean +- SD duration from onset of symptoms to ED visit was 2.5 +- 1.3 days (range, 1-5 days). All patients had abdominal pain, and 64.7% had nausea/vomiting. A total of 64.7% of patients had peritoneal signs. The jejunum was most commonly involved (88.2% of all cases). The maximal mean +- SD wall thickening of the bowel was 14.1 +- 4.4 mm (range, 7.4-26.7 mm), and the estimated mean +- SD length was 35.6 +- 24.4 cm (range, 9-105 cm). The mean +- SD prothrombin time and activated partial thromboplastin time were prolonged to 86.5 +- 26.9 and 116.2 +- 43.1 seconds, respectively. All patients received medical treatment and survived. At the last follow-up (mean, 27.4 months), none of the patients had recurrence of intestinal intramural hemorrhage or intestinal obstruction.
Conclusion: prolonged prothrombin time and drug history can indicate the possibility of intramural intestinal hemorrhage, and abdominal computed tomography may help to exclude surgical diseases and prevent unnecessary surgery.
(C) 2010
* Corresponding author. Department of Emergency Medicine, Taipei Veterans General Hospital, Taipei, Taiwan, R.O.C. Tel.: +886 2 8757628; fax: +886 2 8757842.
E-mail address: [email protected] (H.-H. Huang).
Introduction
Spontaneous intramural intestinal hematoma is a rare complication of Oral anticoagulant therapy, with a reported rate of 1 in 2500 [1]. Because of the increasing application rate in cases of pulmonary embolism, deep vein thrombosis,
0735-6757/$ - see front matter (C) 2010 doi:10.1016/j.ajem.2009.08.002
prosthetic valves, or persistent atrial fibrillation [2,3], emergency physicians should be aware of the life- threatening complications such as intracranial, pericardial, intraperitoneal, or gastrointestinal bleeding [4,5]. The clinical symptoms and signs of intramural intestinal hematoma usually mimic those of surgical acute abdomen [6]. The differential diagnosis of acute abdomen must take intramural intestinal hematoma into consideration, especial- ly in patients taking long-term oral anticoagulant therapy. Only a few reports in the literature have described the severity of intramural intestinal hemorrhage, including prolonged prothrombin time , length of involved segment, and the treatment when peritonitis appeared. The purpose of our retrospective chart review was to evaluate the clinical symptoms and the image findings in anticoag- ulant-induced intramural intestinal hemorrhage.
Methods
We retrospectively reviewed the emergency medical charts in the Taipei Veterans General Hospital’s emergency department (ED), which has a capacity of around 80,000 visits per year. This retrospective chart review was approved by our institutional review board. Cases of intestinal intramural hemorrhage were first identified from a computer search for hospital discharge codes (International Classifi- cation of Diseases) of 569.8 and 578. We reviewed the discharged diagnosis and included patients with intestinal intramural hemorrhage and increased international normal- ized ration (INR) due to warfarin. Between 2003 and 2008,
17 patients were reviewed as nontraumatic, spontaneous intramural hematoma, caused by over-anticoagulation due to warfarin sodium. All patients had acute abdominal pain at the ED, and the final diagnosis was confirmed by an abdominal
computed tomography (CT) scan. The demographic data, initial clinical presentations, predisposing diseases, results of laboratory tests, and findings of the CT scans were evaluated. The one-sample Kolmogorov-Smirnov test was used for analysis of quantitative data.
Results
In the reviewed 17 patients, the mean +- SD age was 77.7 +-
8.5 years (range, 60-93 years). Cerebral vascular accidents and atrial fibrillation were the most common etiologies for patients with oral warfarin sodium. Intestinal hemorrhage and mesenteric artery occlusion disease were the most common initial impressions before performing abdominal CT (Table 1).
The mean +- SD duration from onset of symptoms to the ED visit was 2.5 +- 1.3 days (range, 1-5 days) (Table 2). All (100%) of the patients had abdominal pain and 11 (64.7%) had nausea/vomiting. Eleven (64.7%) patients had peritoneal signs such as abdominal rebounding pain or muscle guarding. All patients had Upper gastrointestinal bleeding proven by positive gastric or stool occult blood test results.
Table 2 Summary of initial clinical presentations, laboratory data, and radiological examinations in all study patients (N = 17)
Duration of symptoms before ED visit (d) Symptoms and signs
Abdominal pain Nausea/vomiting Melena stool Physical examination Local tenderness Diffuse tenderness
Hypoactive bowel sound Rebounding pain Muscle guarding
Laboratory data (reference range) WBC (4500-11 000/uL)
Hb (male: 14-18 g/dL; female: 12-16 g/dL)
Platelet (x1000 /uL)
Creatinine (0.5-1.5 mg/dL)
LDH (85-213 U/L)
CRP (0-0.05 mg/dL)
PT (8-12 s)
aPTT (23.9-35.5 s)
KUB X-ray of Kidney, Ureter and Bladder: ileus
Abdominal CT Jejunum Ileum
Maximal wall thickening (mm) Estimated length (cm)
2.5 +- 1.3
17 (100)
11 (64.7)
7 (41.2)
11 (64.7)
6 (35.3)
7 (41.2)
11 (64.7)
6 (35.3)
11 870.6 +- 4033.3
11.4 +- 2.7
247.1 +- 85.8
1.6 +- 0.6
252.9 +- 67.3
5.7 +- 2.9
86.5 +- 26.9
116.2 +- 43.1
14 (82.4)
15 (88.2)
2 (11.8)
14.1 +- 4.4
35.6 +- 24.4
Values are presented as mean +- SD or no. (%), where applicable. CRP indicates C-reactive protein; WBC, white blood cell; Hb, hemoglobin; LDH, lactate dehydrogenase.
Table 1 Summary of demographic data, underlying disease, and initial impressions before abdominal CT in all study patients (N = 17)
Age (y) Male:female Underlying disease
Cerebral vascular accident Atrial fibrillation
Prosthetic cardiac valve replacement Deep vein thrombosis
Pulmonary embolism Transient ischemic attack
Initial impression before performing abdominal CT Intestinal hemorrhage
Mesenteric artery occlusion disease Hemoperitoneum
Intestinal obstruction Ruptured appendicitis
77.7 +- 8.5
13:4
11 (64.7)
9 (52.9)
6 (35.3)
2 (11.8)
1 (5.9)
1 (5.9)
6 (35.3)
4 (23.5)
3 (17.6)
2 (11.8)
2 (11.8)
Values are presented as mean +- SD or no. (%), where applicable.
Twelve patients had a urine test, and 7 (58.3%) had hematuria. The number of values that exceeded the upper limit of PT and Activated partial thromboplastin time values were 9 (52.9%) and 4 (23.5%) respectively. While PT greater than 100 seconds and aPTT greater than 156 seconds were calculated as 100 seconds and 156 seconds respective- ly, the mean and aPTT were prolonged to 86.5 +- 26.9 and
116.2 +- 43.1 seconds, respectively.
All patients were diagnosed with intramural intestinal hemorrhage after abdominal CT. In the CT findings, the jejunum was most commonly involved. The maximal mean +- SD wall thickening of the bowel was 14.1 +-
4.4 mm (range, 7.4-26.7 mm) and the estimated mean +- SD length was 35.6 +- 24.4 cm (range, 9-105 cm). The mean +- SD dosage of warfarin sodium was 3.8 +- 2.4 mg/d. Fresh- frozen plasma (FFP) and vitamin K1 were intravenously given at 2.6 +- 1.1 U (range, 0-4 U) and 10.0 +- 3.5 mg (range, 0-20 mg), respectively. After 13.4 +- 7.9 hours, the mean +- SD values of INR, PT, and aPTT had been corrected to 1.4 +- 0.3, 16.1 +- 4.8 seconds, and 34.5 +- 5.8 seconds, respectively.
The mean +- SD hospital days were 12.7 +- 11.6 days (range, 0-43 days). All patients received medical treatment and survived. At the last follow-up (mean, 27.4 months), none of the patients had recurrence of intestinal tract hematoma, abdominal pain, or intestinal tract obstruction.
Discussion
About 1.1% to 8.1% of patients receiving long-term Warfarin therapy per year report major bleeding [7], and the incidence of spontaneous intestinal intramural hemorrhage has been reported as 1 in 2500 [1]. Risk factors include old age, serious illness (cerebral, cardiac, kidney, or liver disease), cerebrovascular or peripheral vascular disease, and an unstable anticoagulant effect [7].
The patients typically visited the ED because they had abdominal pain [8]. Nausea/vomiting (64.7%) and peritoneal signs (64.7%) indicated the severity of abdominal pain and the possibility of surgical intervention. After analysis of the initial impressions before performing abdominal CT, 47.1% (8/17) of the patients were suspected to have surgical abdomen (4 mesenteric artery occlusion diseases, 2 intestinal obstruction, and 2 ruptured appendicitis).
Drug history is a key factor, and the detection of prolonged PT and INR can help emergency physicians take Bleeding complications from warfarin overdose into consid- eration [9]. A total of 52.9% (9/17) of patients were suspected to have nonsurgical abdomen, and the suspected major problem of acute abdomen was bleeding (6 bowel hemorrhage and 3 hemoperitoneum) before performing abdominal CT.
In the CT findings, the jejunum was most commonly involved [10], and the rate was 88.2% in our study. The mean
thickness and length of intramural intestinal hemorrhage were about 15.6 mm (range, 9.1-2.67 mm) and 35.6 cm (range, 9-105 cm), respectively. The rate of hemoperitoneum was about 94.1%. These results are compatible with those of previous studies [6,10]. A total of 47.1% (8/17) of patients had more than 30 cm (range, 34-105 cm) of intramural intestinal hemorrhage. A higher rate of hemoperitoneum and longer length of involved intestinal hemorrhage could explain why the rate of peritonitis increased to 64.7%.
The management of warfarin overdose without bleeding, 1 to 2.5 mg of oral vitamin K, reduces the INR from 5.0-9.0 to 2.0-5.0 within 24 to 48 hours [11,12]. In patients with serious bleeding and increased INR, vitamin K (10 mg) by slow intravenous infusion supplemented with FFP, repeated every 12 hours for persistent INR elevation, is recommended [13]. Compared with a previously published study [8], our review had longer prolonged PT and aPTT (86.5 +- 26.9 and
116.2 +- 43.1 seconds, respectively, in our study vs 35.4 +-
32.1 and 55.8 +- 25.4 seconds, respectively, in the previous study). Although the coagulation conditions were so severe, intravenous 2.8 U FFP and 10 mg vitamin K1 successfully corrected the PT from 86.5 to 16.1 seconds in the mean follow-up period of 14.7 hours.
Conservative nonoperative therapy is the optimal treat- ment [8,10,14], and surgery is indicated only if generalized peritonitis or intestinal obstruction develops [8]. Peritonitis is a crucial sign for clinical physicians to determine whether surgical intervention is suitable. As previous studies have shown, peritonitis occurs initially in 7.7% to 14.3% of patients, and about 15.3% to 28.5% undergo surgical exploration because of peritoneal signs or obstructing mass (hematoma) [8,10]. In our review, up to 64.7% of patients had peritoneal signs, and none of them underwent surgical exploration. In such conditions, abdominal CT plays an important role in making decisions for emergency physicians or surgeons considering surgery. If the possibility of complete intestinal obstruction, hollow organ perforation, or Mesenteric ischemia bowel disease can be excluded by abdominal CT, conservative treatment is the best choice for these patients. All patients in this review successfully received medical treatment and were discharged without recurrence of intestinal tract hematoma, abdominal pain, or intestinal tract obstruction in 27.4 months of follow-up.
Conclusion
Nontraumatic spontaneous intramural intestinal hemor- rhage is a complication of anticoagulant therapy that presents as an acute abdomen in the ED. Prompt and early recognition is crucial to prevent unnecessary explor- atory surgery. Prolonged PT and drug history can indicate the possibility of intramural intestinal hemorrhage, and abdominal CT helps to exclude surgical diseases and prevent unnecessary surgery.
References
- Bettler S, Montani S, Bachmann F. Incidence of intramural digestive system hematoma in anticoagulation. epidemiologic study and clinical aspects of 59 cases observed in Switzerland (1970-1975). Schweiz Med Wochenschr 1983;113(17):630-6.
- Hirsh J. Oral anticoagulant drugs. N Engl J Med 1991;324(26): 1865-75.
- Shah P, Kraklow W, Lamb G. Unusual complication of Coumadin toxicity. Wis Med J 1994;93(5):212-4.
- Singer DE, Albers GW, Dalen JE, et al. antithrombotic therapy in atrial fibrillation: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126(3): 429-56.
- Celik A, Ozkan N, Ersoy OF, et al. clinical challenges and images in GI. Intramural intestinal hematoma. Gastroenterology 2008;134(2): 387, 647.
- Macari M, Chandarana H, Balthazar E, et al. intestinal ischemia versus intramural hemorrhage: CT evaluation. AJR Am J Roentgenol 2003; 180(1):177-84.
- Gallus AS, Baker RI, Chong BH, et al. Consensus guidelines for warfarin therapy. Recommendations from the Australasian Society of Thrombosis and Haemostasis. Med J Aust 2000;172(12):600-5.
- Polat C, Dervisoglu A, Guven H, et al. Anticoagulant-induced intramural intestinal hematoma. Am J Emerg Med 2003;21(3):208-11.
- Chaiteerakij R, Treeprasertsuk S, Mahachai V, et al. Anticoagulant- induced intramural intestinal hematoma: report of three cases and literature review. J Med Assoc Thai 2008;91(8):1285-90.
- Abbas MA, Collins JM, Olden KW. Spontaneous intramural small- bowel hematoma: imaging findings and outcome. AJR Am J Roentgenol 2002;179(6):1389-94.
- Weibert RT, Le DT, Kayser SR, et al. Correction of excessive anticoagulation with low-dose oral vitamin K1. Ann Intern Med 1997; 125:959-62.
- Crowther M, Donovan D, Harrison L, et al. Low-dose oral vitamin K reliably reverses over-anticoagulation due to warfarin. Thromb Haemost 1998;79:1116-8.
- Ansell J, Hirsh J, Hylek E, et al. Pharmacology and management of the vitamin K antagonists. Chest 2008;133(6):160S-98S.
- Eiland M, Han SY, Hicks GM. Intramural hemorrhage of the Small intestine. JAMA 1978;239(2):139-42.