- Citado por SciELO
versión impresa ISSN 1130-0108
Rev. esp. enferm. dig. vol.96 no.6 jun. 2004
Superior mesenteric venous thrombosis: a retrospective study of thirteen cases
S. Muñoz, P. Cubo, J. González-Castillo, J. A. Nuevo, E.J. García-Lamberechts and A. Sanz
Service of Internal Medicine III. Hospital Clínico San Carlos. Madrid. Spain
Objective: to analyze the epidemiology, associated risk factors, clinical presentation, diagnostic methods, treatment, and evolution of patients diagnosed with superior mesenteric venous thrombosis (SMVT) at an university hospital in Madrid.
Experimental design: retrospective and descriptive study. We review the medical records of patients with this diagnosis in our hospital from January 1998 to December 2002. Data were processed by using the SPSS vs. 11 software.
Patients: all thirteen subjects diagnosed with SMVT in that period were included.
Results: associated risk factors included tumoral conditions (5 patients), acute abdominal pathology (2), polyglobulia (1), prothrombin gene mutation (1), and anticardiolipin antibodies (1). No predisposing factor was found in 3 patients. Clinical presentation for all patients was abdominal pain, with nausea and vomiting being the second symptom in frequency (7). The diagnosis was reached by abdominal CT (9), arteriography (2), ultrasounds (1), and histology after intestinal resection (1). Treatment with only anticoagulation was initiated in 4 patients, whereas anticoagulation and surgery were performed in 5 cases. In 4 subjects no specific treatment was prescribed and only palliative measures were established due to a baseline end-stage condition. Five patients died, and four of them had a neoplasic condition as associated risk factor. Mortality in our series was 38.5%.
Conclusions: SMVT is a very rare disease that is often associated with neoplasic pathology, which influences its high mortality. Due to non specific symptoms, imaging is essential for the diagnosis and the detection of associated risk factors. In our series, computed tomography imaging was the most profitable test.
Key words: Mesenteric venous thrombosis. Mesenteric ischemia. Intestinal infarction.
Muñoz S, Cubo P, González-Castillo J, Nuevo JA, García-Lamberechts EJ, Sanz A. Superior mesenteric venous thrombosis: a retrospective study of thirteen cases. Rev Esp Enferm Dig 2003; 96: 385-394.
Correspondencia: Sonia Muñoz Albarrán. servicio Medicina Interna III. Hospital Clínico San Carlos. C/ Dr. Martín Lagos, s/n. 28040 Madrid. Telf.: 91 330 38 02.
Superior mesenteric venous thrombosis (SMVT) is a well known cause of intestinal ischemia or infarction. It is uncommon, but represents 5-15% of mesenteric ischemic events (1,2), with mortality rates between 15 and 40% according to different series. It is influenced by its difficult and delayed diagnosis (3,4). Originally, it was described by Elliot, in 1895 (5), but it was then characterized as an independent clinical entity by Warren and Eberhard, in 1935 (6).
Some risk factors predispose to SMVT: congenital and acquired prothrombotic states (hematological diseases, neoplasms, etc.), abdominal infectious and inflammatory diseases, abdominal postoperative states, hepatic cirrhosis and portal hypertension, and abdominal trauma (1,3,7-12).
Clinically, there is an acute form of presentation in which symptoms appear suddenly (minutes, hours) and is associated with a high risk of intestinal ischemia. In the subacute form of presentation, abdominal pain is present for days or weeks; intestinal ischemia is less frequent than in the acute forms, and bleeding from esophageal varices is rare. The limit between acute and subacute presentations is confusing; sometimes difficult to determine. Disproportionate abdominal pain in relation to physical examination findings, nausea, vomiting, anorexia and diarrhea are the most frequent symptoms. Hematemesis, hematochezia and melenas are less frequent. At last, there is a chronic form of presentation (months, years) whose symptoms are splenic or portal thrombosis-related, with bleeding from esophageal varices (1).
Diagnosis is difficult and delayed, in most cases because of nonspecific symptoms and the absence of known predisposing risk factors. Sometimes a diagnosis is reached after laparotomy or autopsy.
It is important that an early diagnosis be reached, and that treatment be initiated as soon as possible to avoid complications and to reduce mortality. A high level of suspicion, a guided anamnesis for risk factors, an appropriated physical examination, and current imaging techniques (ultrasounds, computed tomography -CT-, arteriography) permit early diagnosis.
Once a diagnosis has been reached, anticoagulation therapy with systemic heparin is the basis of treatment. Surgery with subsequent anticoagulation is restricted only for cases with suggestive evidence of intestinal infarction.
In the present study we analyzed the epidemiological, clinical and diagnostic-therapeutic data of patients with a diagnosis of SMVT in our hospital. Until now, publications of case series with this pathology are rare. Results of different studies are similar in some points, but differ in others. We attempt to establish the current situation in our hospital and to compare our results with those of previous studies, when possible.
PATIENTS, MATERIAL AND METHODS
This study was performed in the "Hospital Clínico San Carlos", Madrid. This is a third-level hospital with a mean of 45,000 inpatients per year. Approximately 11,600 of them belong in Internal Medicine, Gastrointestinal and General Surgery departments. We included patients diagnosed with SMVT from January 1998 to December 2002 in these departments. Thirteen patients were obtained (n = 13), seven males (53.8%) and six females (46.2%). Mean age was 68±12.53 years, and ages varied from 52 to 83 years. We analyzed epidemiological data (age and sex), associated risk factors (coagulation abnormalities, immunological factors, presence of tumors, hematological diseases, abdominal infections or inflammatory conditions, postoperative states, presence of cirrhosis and portal hypertension), symptoms of clinical presentation, form of presentation (acute, subacute, chronic), laboratory results upon arrival in the emergency department (leukocytes, hemoglobin, hematocrit and venous gasometry), results of imaging techniques (plain abdominal radiography, ultrasonography, abdominal CT, arteriography), treatment of patients (anticoagulation, surgery or both) and evolution (outcome or decease).
The information was obtained from the anamnesis and supplementary exams on medical records of patients with SMVT.
When evaluating associated risk factors, we considered those that were present before patient admission to the hospital, and those that were diagnosed in the hospital or during patient follow-up at the outpatient clinic. Acute forms were considered those cases with a standing time shorter than 48 hours; chronic forms were those who had symptoms for more than one month. The remaining patients were considered to have subacute presentations.
We conducted a descriptive retrospective study. Due to the limited number of patients, a statistical signification study was not performed. Data were analyzed using the SPSS vs. 11 software.
Five patients (38.5%) presented tumoral pathology, and three of them (60%) had associated portal hypertension. Tumors had a pancreatic origin in three cases (23.1%), one was originated from the thyroid gland (7.7%) and, in the last case, multiple hepatic lesions were detected (7.7%).
Two patients (15.4%) developed SMVT following intraabdominal acute conditions; one case was due to diverticulitis, the other to the presence of a small bowel volvulus.
One patient had been diagnosed with polyglobulia (7.7%). A prothrombin gene mutation was found during a coagulation study in another patient (7.7%), and anticardiolipin antibodies were positive in a third one (7.7%).
Three patients had no associated risk factors (23%). One of them (7.7%) was admitted to hospital again because of recurrent pulmonary embolism and deep venous thrombosis. Only one patient, this with positive anticardiolipin antibodies, had a history of deep vein thrombosis in another localization and recurrent pulmonary embolism (7.7%).
In our series, three cases of SMVT were chronic (23%), six were acute (46.2%), and four were subacute forms (30.8%). All cases developed abdominal pain (100%), while nausea and vomiting were the second most frequent symptoms (53.8%). Regarding laboratory results, in patients with chronic forms, we found no alterations in hemoglobin or hematocrit levels, and neither leukocytosis nor neutrophilia. On the contrary, patients with acute or subacute forms had leukocytosis and neutrophilia. A baseline arterial gasometry was performed in six patients (46.2%), but only in one case (16.7%) an evident metabolic acidosis was found; an intestinal infarction was demonstrated later. Gasometry was normal in two patients (33.3%); one of them was treated surgically, and an intestinal infarction was detected. In the other patients, HCO3- was in its lower normal range, with normal pH.
Urgent plain abdominal radiographs were taken in eleven patients (84.6%). Seven of them had pathologic findings (63.6%). Most frequent findings included small bowel dilatation (71.4%), air-fluid levels (57.1%), colon dilatation (28.6%), and colonic stop (28.6%). Abdominal radiographs were normal in the remaining four cases (36.4%).
Abdominal ultrasounds were performed in only seven cases (53.8%), yielding clearly pathological results in six of them (85.7%), and inconclusive findings in another patient (14.3%). In two cases, ultrasounds demonstrated a tumor (28.6%) (pancreatic tumor, multiple focal hepatic lesions), and SMVT associated with portal thrombosis in one case (14.3%), and isolated portal thrombosis in another (14.3%). Other relevant findings by the ultrasound study included small bowel and colonic wall enlargement, and the presence of fluid within the abdominal cavity.
An abdominal CT scan was performed in eleven patients (84.6%). In nine of them, this study diagnosed SMVT (81.8%), with associated portal thrombosis in four patients (44.4%) and femoral vein thrombosis in two cases (22.2%). In the other two patients (18.2%), CT demonstrated colonic dilatation in one patient, and enlargement and dilatation of the small bowel in another patient, as well as free liquid within the peritoneal cavity. These are the most frequent findings. In two cases, CT showed related factors associated with SMVT that had not been diagnosed by ultrasounds (acute diverticulitis and a pancreatic neoplasm).
An arteriography was performed in three patients only (23.1%). In one of them, the patient with acute diverticulitis, a colonic angiodisplasia was diagnosed (33.3%). In the two remaining cases (66.6%), arteriography demonstrated SMVT, which had not been diagnosed using the previous methods. In one patient, SMVT was associated with portal vein thrombosis and, in the latter case, with a cava vein thrombosis and vascularized pancreatic tumor.
In summary, among our thirteen patients with SMVT, one was diagnosed by ultrasounds (7.7%); abdominal CT established the diagnosis in nine of them (69.2%), and arteriography in two cases (15.4%). Only in one case the diagnosis was obtained by histology after intestinal resection (7.7%). The diagnostic yield of each of the previously mentioned diagnostic methods is displayed in figure 1.
In our series, six patients (46.2%) presented an acute evolution, and four of them developed intestinal ischemia (66.6%). In four cases symptoms were subacute, and three of them suffered from intestinal ischemia (75%). Amongst the seven patients with ischemic signs (53.8%), surgical treatment was perfomed in only five cases (71.4%); the remaining cases were not believed to be candidates for surgical management as they were considered palliative patients.
When treatment was reviewed, we found that four patients received anticoagulation therapy with sodium heparin as the only therapeutic measure (30.8%); three of them were discharged after total recovery from our hospital (75%), while one of them died (25%). In the latter, a superficial intestinal infarction was demonstrated in the autopsy. Five patients received both anticoagulation and surgical treatment (38.5%); all of them had a good clinical evolution and were discharged following recovery. In the last four patients no treatment was administered (30.7%); three of them presented as chronic SMVT and one as an acute SMVT, but were considered palliative patients (three cases with pancreatic neoplasm and one case with multiple hepatic lesions) (Fig. 2).
Overall, five patients died (38.5%) and eight were discharged after recovery (61%) with oral anticoagulation or therapeutic doses of low molecular weight heparin (LMWH) as treatment.
SMVT is a difficult-to-diagnose uncommon condition with a high mortality. Occasionally it is associated with thrombosis in other areas (portal, femoral veins,…). Clinical suspicion is essential for the diagnosis, which must be afterwards confirmed through image testing (radiography, ultrasonography, CT and arteriography in some cases). Early treatment is essential to reduce the high mortality of this condition. In our hospital we found only thirteen cases in the last five years, with a mortality rate of 38.5%. This is similar to the mortality rate observed in series published by other authors (3,4).
Risk factors associated with SMVT include: a) pro-thrombotic states: antithrombin III deficiency, protein C or protein S deficiency, presence of factor V Leiden or mutation in prothrombin gene, antiphospholipid antibodies, hyperhomocysteinemia, oral contraceptive use, pregnancy or neoplasm; b) hematological disorders: polycythemia vera, essential thrombocytosis, paroxysmal nocturnal hemoglobinuria; c) inflammatory diseases: pancreatitis, peritonitis and intraabdominal sepsis, inflammatory bowel disease or diverticulitis; d) postsurgical states: abdominal operations, splenectomy, sclerotherapy for esophageal varices; e) cirrhosis and portal hypertension; and f) other: blunt abdominal trauma (1). When none of these factors is detected, SMVT is referred to as idiopathic.
Neoplasms, oral contraceptive use, and hematological diseases, as well as pancreatitis, peritonitis and abdominal surgeries, are considered to be the most frequent factors associated with SMVT by some authors (1). According to other series, previous abdominal surgery (50% of patients) and abdominal infections (appendicitis, cholecystitis, and inflammatory bowel disease, among others) fol-lowed by prothrombotic states (28%) are the factors most frequently found involved in the etiology of SMVT (13). In other series prothrombotic states and hematological diseases are the most commonly found factors (3,4,9,11,14). In our series, the risk factor that was present in a major number of cases was tumors (38.5%), followed by acute intraabdominal conditions (15.4%), prothrombotic states (15.4%) and hematological disorders (7.7%).
The number of idiopathic cases varies in different studies. It ranges between 21 (13) and 37 (3) or 66.7% (4). These rates are not comparable as diagnostic methods used in patients diagnosed with SMVT are not specified. None associated risk factor was detected in 23% of our patients. The number of cases with unknown etiology is decreasing due to advanced imaging methods.
Most of the cases in our series followed an acute or sub-acute course. In these cases the related etiology was diverse. However, it is important to emphasize that all cases with a chronic evolution were associated with splenic-portal axis tumors, and 66.7% of them had portal hypertension as well. Abdominal pain, nausea and vomit-ing are constant symptoms, which are present in acute-subacute and chronic forms. Bleeding from esophageal varices is a less frequent symptom even in chronic forms.
The presence of leukocytosis with neutrophilia, the increasing levels of hemoglobin and hematocrit, and the presence of metabolic acidosis when an intestinal infarction is established are laboratory changes that contribute to the diagnosis (1,4). In our series there seems to exist some differences between acute-subacute and chronic forms. Leukocytosis and neutrophilia are present in all acute and subacute forms, while leukocytes remain within the normal range in chronic SMVT. We found no hemoconcentration in chronic forms, whereas it was present in only 50% of acute and subacute forms. In the only case with evident metabolic acidosis, an intestinal infarct was subsequently confirmed by gasometry. However, gasometry was normal in another patient in whom the presence of an intestinal infarct was later confirmed by laparotomy. In this way, data collected from abdominal exploration are of great value. The presence of peritoneal irritation signs should suggest the presence of an established intestinal infarct.
Once the suspected diagnosis is established by anamnesis, symptoms, physical examination and laboratory tests, imaging techniques must be performed. Plain abdominal radiographs are abnormal in 50-75% of cases, according to different authors (1). We obtained a similar percentage in our series (63.6%). When pathological, these techniques show indirect signs such as small bowel dilatation, air-fluid levels, wall edema, thumbprinting or partial obstruction of the small bowel (1,3,4,14). It may be normal in 25-50% of cases. Abdominal ultrasounds provide indirect information, like enlargement of the bowel wall, presence of free liquid within the abdominal cavity, or presence of a thrombus in the superior mesenteric vein. At the same time it informs about possible causes of SMVT (1,3,4). However, abdominal CT is by far the test of choice when SMVT is suspected, establishing the diagnosis in 90% of cases. However, it is less sensitive for the diagnosis of small-vessel thrombosis (1,3,4,7,16). In our series, it was diagnostic in a minor percentage (81.8%) of cases, though it was pathological in all patients in whom it was carried out. It shows a thrombus as a radiolucent area in the superior mesenteric vein, and contributes to detect possible causes of thrombosis. Other indirect information includes thickening of the bowel wall, pneumatosis intestinalis, gas in the portal vein, and presence of collateral circulation in cases with long standing time (1,7,15,17,18).
Nuclear magnetic resonance has a high sensitivity and specificity for the diagnosis of SMVT as well; however, it does not offer any advantage over CT, except in patients with intolerance to intravenous contrasts due to hypersensitivity, or renal failure (1,3,4). It was not performed in any of the patients in our series.
Mesenteric arteriography shows the thrombus and its associated vasospasm. A delay in the repletion of the superior mesenteric vein, prolonged opacification of the arterial arcades, or arterial reflux may all be observed. It is an invasive diagnostic method and for this reason it should be reserved for highly suspected cases when a diagnosis has not been reached by previous imaging me-thods or when small-vessel thrombosis is suspected (1). It helps differentiate between venous and arterial thrombosis, and provides also an access route for vasodilator administration (3). In our series, it was performed in only three cases; it confirmed the diagnosis in two patients and did not demonstrate the presence of thrombosis in the other one. Latter patient receiving intravenous sodium heparin during the last four days due to highly suspected SMVT according to abdominal CT.
Sometimes diagnosis is obtained after intestinal resection. Only in one of our patients the diagnosis of SMVT was reached histologically after intestinal resection for infarction; it showed small-vessel thrombosis.
In our series, CT and arteriography were the most valuable tests for establishing an SMVT diagnosis. Due to the invasive character of arteriography, CT should be the selected procedure for all patients with suspected SMVT, because of its high sensitivity and smaller number of complications.
The treatment of acute or subacute SMVT consists of anticoagulation with or without surgery. Once a diagnosis has been established, treatment with intravenous heparin has to be initiated as soon as possible, maintaining APTT over two times its normal value. Even in patients with intestinal bleeding, treatment has to be initiated if the hemorrhagic risk is lower than that of an intestinal infarction (1). It has been demonstrated that treatment with heparin improves survival and decreases the risk of recurrence (1,19,20). It also prevents thrombus propagation into the portomesenteric system (4,19,21).
Anticoagulation with warfarin should only be initiated when there is no evidence of ischemic progression, and should be given for six to twelve months if there are no associated risk factors or when these have already disap-peared. Otherwise, anticoagulation must be given for life (1).
Supporting measures like fluids, bowel rest and nasogastric suction play a fundamental role in treatment. However, the usefulness of intravenous antibiotics is not clearly established when intestinal infarction is absent (1).
Thrombolytic therapy may be another treatment option for patients whose diagnosis is well established, who are clinically stable, and have no evidence of intestinal infarction and no contraindication for this kind of therapy. So, this would be an effective treatment in a limited number of patients (22-24). A thrombectomy has been performed in some cases (25). Neither thrombolysis nor thrombectomy were performed in any of our patients.
Only patients with peritoneal signs on physical examination require an urgent surgical exploration. Surgery is the only effective treatment in cases with intestinal infarction. As soon as an intraoperative diagnosis of SMVT has been established, heparin perfusion must be initiated (1) because it decreases both mortality and recurrences (3). The aim of surgery is to preserve as much bowel as possible, only extirpating non-viable tissue. The aim of "second look" laparotomy, 24 hours after the first surgical procedure, is to avoid unnecessary resection of viable intestine. This is worthwhile in patients with extended intestinal ischemia in which venous irrigation is still present, though (26,27).
In chronic presentations, treatment is symptomatic and is aimed at controlling esophageal varices bleeding or recurrent bleeding with pharmacological agents like propranolol, endoscopic therapy or portosystemic shunts (1).
In our series, SMVT presented a high rate of intestinal infarction. Excluding chronic forms, in which intestinal infarction is exceptional, 53.8% of patients developed ischemia with intestinal infarction. In contrast with other authors (1), patients in our series with subacute SMVT developed more intestinal infarctions (75%) than acute forms (66.6%).
Patients treated only with anticoagulation had a mortality rate of 25% (one patient in each four), and intestinal infarction was demonstrated by autopsy. Patients treated with anticoagulation and surgery had a good outcome and all of them were discharged as cured (five of five). "Second-look" surgery was not performed in any case.
In our series mortality appeared to be related in most cases with the underlying condition causing SMVT, a tumor most often. Patients without a tumor had a good clinical outcome with a suitable treatment, even when intestinal infarction was present.
In conclusion, SMVT is an infrequent pathology with a high mortality rate. It should be considered in the differential diagnosis of patients with abdominal symptoms, particularly when risk factors are present. In many cases it is associated with a neoplastic condition or thrombosis in other areas (portal, femoral veins,...). Nonspecific symptoms (abdominal pain, nausea, vomiting) delay the diagnosis. A high suspicion rate, early diagnosis with the help of modern techniques (ultrasounds, abdominal CT, arteriography), and early treatment with heparin or surgery when needed all contribute to improved survival and the prevention of recurrence.
1. Kumar S, Sarr MG, Kamath PS. Mesenteric venous thrombosis. N Engl J Med 2001; 345: 1683-8. [ Links ]
2. Grendell JH, Ockner RK. Mesenteric venous thrombosis. Gastroenterology 1982; 82: 358-72. [ Links ]
3. Choudhary AM, Grayer D, Nelson A, Roberts I. Mesenteric venous thrombosis. A diagnosis not to be missed! J Clin Gatroenterol 2000; 31: 179-82. [ Links ]
4. Divino CM. In: Park S, Angel LP, Ellozy S, Spiegel R, Kim U. A retrospective study of diagnosis and manegement of mesenteric vein thrombosis. Am J Surgery 2001; 181: 20-3. [ Links ]
5. Elliot JW. The operative relief of gangrene of intestine due to occlusion of the mesenteric vessels. Ann Surg 1895; 21: 9-23. [ Links ]
6. Warren S, Eberhard TP. Mesenteric venous thrombosis. Surg Gynecol Obstet 1935; 61: 102-21. [ Links ]
7. Harward TRS, Green D, Bergan JJ, et al. Mesenteric venous thrombosis. J Vasc Surg 1989; 9: 328-33. [ Links ]
8. Boley SJ, Kaleya RN, Brandt LJ. Mesenteric venous thrombosis. Surg Clin North Am 1992; 72: 183-201. [ Links ]
9. Valla K. Venous Thrombosis in gastroenterology. J Mal Vasc 1991; 16: 243-8. [ Links ]
10. Nair HT, Muscroft TJ. Idiopathic mesenteric venous thrombosis. Br J Clin Pract 1990; 44: 651-2. [ Links ]
11. Alving BM. The hipercoagulable states. Hosp Pract (Off Ed) 1993; 28: 109-21. [ Links ]
12. Slutzker DM, Pigula F. Small bowel infarction and death from primary mesenteric venous thrombosis. Am J Emerg Med 1987; 5: 126-9. [ Links ]
13. Warchauer DM, Lee JKT, Mauro MA, White GC II. Superior mesenteric vein thrombosis with radiologically occult cause: a retrospective study of 43 cases. AJR 2001; 177: 837-41. [ Links ]
14. Grendell JH, Ockner RK. Mesenteric venous thrombosis. Gastroenterology 1982; 82: 358-72. [ Links ]
15. Boley SJ, Kaleya RN, Brandt LJ. Mesenteric venous thrombosis. Surg Clin North Am 1992; 72: 183-201. [ Links ]
16. Vogelzang RL, Gore RM, Anschuetz SL, Blei AT. Thrombosis of the splacnic veins: CT diagnosis. AJR Am J Roentgenol 1988; 150: 93-6. [ Links ]
17. Rosen A, Korobkin M, Silverman PM, et al. Mesenteric venous thrombosis: CT identification. AJR Am J Roentgenol 1984; 143: 83-6. [ Links ]
18. Matos C, Gansbeck V, Zalcman M, et al. Mesenteric venous thrombosis: early CT and US diagnosis and conservative management. Gastrointest Radiol 1986; 11: 322-5. [ Links ]
19. Abdu RA, Zakhour BJ, Dallis DJ. Mesenteric venous thrombosis-1911 to 1984. Surgery 1987; 101: 383-8. [ Links ]
20. Matthews JE, White TP. Primary mesenteric venous occlusive dis-ease. AM J Surgery 1972; 122: 579-83. [ Links ]
21. Chen M, Brown MC, Willson RA, et al. Mesenteric vein thrombosis. Four cases and review of the literature. Dig Dis 1996; 14: 382-9. [ Links ]
22. Poplausky MR, Kaufman JA, Geller SC, Waltman AC. Mesenteric venous thrombosis treated with urokinase via the superior mesenteric artery. Gastroenterology 1996; 110: 1633-5. [ Links ]
23. Yankes JR, Uglieta JP, Grant J, Braun SD. Percutaneus transhepatic recanalization and thrombolysis of the superior mesenteric vein. AJR Am J Roentgenol 1991; 151: 289-90. [ Links ]
24. Rivitz SM, Geller SC, Hahn C, Waltman AC. Treatment of acute mesenteric venous thrombosis with transjugular intramesenteric urokinase infusion. J Vasc Interv Radiol 1995; 6: 219-28. [ Links ]
25. Bergentz S, Ericsson B, Hedner U, et al. Thrombosis in the superior mesenteric and portal veins: report a case treated with thrombectomy. Surgery 1974; 76: 286-90. [ Links ]
26. Kodadadi J, Rozencwajg J, Nacasch N, Smidt B, Feuchtwanger MM. Mesenteric vein thrombosis: the importance of a second-look operation. Arch Surg 1980; 115: 315-7. [ Links ]
27. Levy PJ, Krausz MM, Manny J. The role of second-look procedure in improving survival time for patients with mesenteric venous thrombosis. Surg Gynecol Obstet 1990; 170: 287-91. [ Links ]