Surg Endosc (1992) 6:283-284
Surgical EndOSCOpy 9 springer-VerlagNew York Inc. 1992
Mesenteric venous thrombosis M o d e r n m a n a g e m e n t and endoscopic diagnosis Terence P. Wade l, William R. Jewell 2, and Charles H. Andrus l i Departments of Surgery, St. Louis University, 3635 Vista Avenue at Grand Boulevard, St. Louis, MI 63111-0250 and the 2 University of Kansas, 39th Street at Rainbow Blvd., Kansas City, KS 66103
Summary. Mesenteric venous thrombosis has an obscure etiology, a prolonged onset of symptoms, and controversial methods of diagnosis and treatment. Eleven collected patients with pathologic confirmation of this diagnosis are presented, with one patient demonstrating the endoscopic appearance of this disease in human small bowel. Other methods of diagnosis are also reviewed. Treatment of mesenteric venous thrombosis requires resection of necrotic bowel and, we believe, anticoagulation. The role of planned "second-look" laparotomy is reviewed, but it was not necessary in those patients in whom definitive resection could be combined with intraoperative anticoagulation.
Key words: Mesenteric venous thrombosis- Smallbowel endoscopy - Intestinal ischemia
localized or diffuse involvement of bowel and mesentery. Controversy exists both in the diagnosis and treatment of this disorder. Although associated with a history of thrombotic disorders [2, 3], prior splenectomy [9], oral estrogens, and portal stasis [3], mesenteric venous thrombosis most commonly presents without apparent preexisting cause [8]. Physical examination is remarkable for diffuse abdominal tenderness without localizing signs. Radiologic studies are not usually of value, and laboratory exams usually reveal only a leukocytosis. Most authors recommend anticoagulation and resection when the diagnosis is made, but the requirement for a later, planned second-look laparotomy to assure bowel viability is in debate [5,6, 10]. The 11 patients presented here illustrate the diagnostic problems and suggest therapeutic management of this disease.
Materials and methods Intestinal ischemia can be caused by obstruction to blood flow in the arterial or venous system. The presentation of arterial ischemia is usually sudden and painful, and systemic changes often follow quickly. Obstruction of mesenteric venous flow is thrombotic, and the onset and progression of symptoms are slower than seen with arterial occlusion. Mesenteric venous thrombosis is the etiology of 10-30% [6, 7, 10] of cases with acute intestinal ischemia, yet its insidious clinical presentation makes preoperative diagnosis a rarity. First described as a clinical entity in 1935 [12], mesenteric venous thrombosis was recently reviewed in a case report [ 11] and as a compilation of all reported cases [1]. The disease is characterized by a prolonged onset of symptoms, a long interval between symptomatic onset and diagnosis, and either
Offprint requests to: Terence P. Wade
All cases of acute intestinal ischemia which occurred at the University of Kansas Medical Center between 1974 and 1983 were reviewed. Nine cases of mesenteric venous thrombosis were pathologically documented, representing 10% of the patients with intestinal ischemia. Two additional cases have been added to the authors' experience since 1983. The l l cases were reviewed with regard to complaints at presentation; symptoms and findings on exam; patient demographics; preoperative laboratory and diagnostic tests; findings at operation or autopsy; and final outcome.
Results The 11 patients were age 16-84 with a mean of 46 years. Seven were male, nine were white, and two were black. All 11 patients complained of abdominal pain lasting 3-7 days prior to exploration. Other complaints included nausea or vomiting in nine patients; six had diarrhea. Three patients had a coexisting malignancy, five had prior clotting anomalies, and two were taking oral estrogens. On exam, all had diffuse abdominal tenderness without localization, and most had decreased bowel sounds.
284
Nine of the 11 had a leukocyte count in excess of 10,000 cells/mm3; 8 of these had a count o v e r 15,000 and in 5 it was over 20,000. Of the additional two patients, one had a normal leukocyte count and one had a leukopenia (700 WBC/mm 3) secondary to chemotherapy. All abdominal x-rays revealed nonspecific dilated small bowel with air fluid levels, and one patient had a nondiagnostic visceral angiogram. The most recent patient had a gastric and proximal small-bowel endoscopic examination performed with a pediatric colonoscope (PCF, Olympus Corp.) p e r os, which revealed a sharply demarcated area of mucosal ischemia and edema beginning in the first jejunal loop. This segment of ischemic small bowel was estimated at 60 cm past the pylorus, or 150 cm from the incisors. Due to the ability to "telescope" nonfixated segments of small bowel over the endoscope, the reach of oral endoscopy past the pylorus is variable and difficult to quantitate. Abdominal CT scan in this patient revealed this jejunal loop to be thick-walled and distended while the remainder of the bowel appeared normal. Four patients died of their disease: two were discovered at postmortem exam, and one died immediately after operation. The fourth patient had been previously anticoagulated with heparin for migratory thrombophlebitis associated with carcinoma of the lung. The heparin had been discontinued prior to laparotomy, and the patient died of a massive pulmonary embolism 48 h after operation. Anticoagulation was started on the day of surgery in 8 of the 11 patients, and in all 7 survivors. However, appropriate levels of anticoagulation (PT or PTT of 1.5 times normal) were diffiult to maintain. One patient survived multiple pulmonary emboli in the first 48 h after laparotomy despite heparin therapy, and two others had postoperative bleeding requiring a dosage modification. "Second-look" laparotomy (routine, planned reexploration to detect recurrent bowel ischemia without signs or symptoms) was performed in five patients, four of which were heparinized. All of these were negative. No patient who was anticoagulated suffered recurrentbowel ischemia. Four wound complications occurred in three patients. Two patients required drainage of subcutaneous infections (and one developed an incisional hernia that required repair) while another had a wound hematomaJ seroma.
Discussion
This experience represents a comparatively recent collection of patients with mesenteric venous thrombosis and suggests that both the diagnosis and management of this disease may be changing. Preoperative CT scanning may show changes suggestive of this diagnosis, including bowel-wall thickening [1] and thrombosis within larger mesenteric veins [4]. This paper reports an endoscopic demonstration of mesenteric venous ischemia in the human. The muco-
sal edema and necrosis observed were very similar to that previously demonstrated in a canine experimental model [8]. Photographs of the canine model in this paper were very similar to those from the patient in the present report. The mucosal edema and thickening of the intestinal circular folds are the key to the endoscopic diagnosis of mesenteric venous thrombosis; arterial ischemia is characterized by a pale, nondistended mucosa as seen intraluminally. The sharp segmental mucosal change seen endoscopically in this patient was also seen later at operation on the serosa of the bowel segment. The demonstration of laparotomy of thickwalled small bowel and/or mesenteric veins distended with clot should also suggest this diagnosis. Small intestinal endoscopy could make an important contribution to diagnosis in the future if this work can be duplicated in further patients with mesenteric ischemia. Routine second-look procedures are not supported by this series. We see two explanations for this departure from prior reports: (1) anticoagulation was used frequently and usually was begun intraoperatively at the time of diagnosis; and (2) those patients who were not reexplored had a well-demarcated ischemic segment(s) of small bowel, and anastomoses were performed in clearly viable bowel. When all involved bowel and mesentery can be safely resected and heparin is begun during exploration, we feel routine, planned reexploration to evaluate the viability of the bowel is not necessary. However, reexploration still has a role in those patients with diffuse invovlement of long segments of small bowel in whom initial definitive resection is impossible. References 1. Abdu R, Zakhout B, Dallis D (1987) Mesenteric venous thrombosis. Surgery 101:383-389 2. Arko RF, Jewell CT, Dainer P, MacLeod WAJ (1979) Primary mesenteric vein thrombosis associated with antithrombin Ill deficiency. JAMA 242(21): 2324-2325 3. Drott C, Henning R, Hedner U (1978) Massive thrombosis of the superior mesenteric, splenic and portal veins. Acta Chir Scand 144:57-60 4. Harward TRS, Green D, Bergan JJ, Rizzo RJ, Yao JST (1989) Mesenteric venous thrombosis. J Vasc Surg 9(2): 328-333 5. Khodadadi J, Rozencwajg J, Nacash N, Schmidt B, Feuchtwanger MM (1980) Mesenteric vein thrombosis. The importance of a second-look operation. Arch Surg 115:315-317 6. Levy PJ, Krauz MM, Manny J (1990) Acute mesenteric ischemia: Improved results - - a retrospective analysis of ninety-two patients. Surgery 107(4): 372-380 7. Levy PJ, Krauz MM, Manny J (1990) The role of second-look procedure in improving survival time for patients with mesenteric venous thrombosis. Surg Gynecol Obstet 170(4): 287-291 8. MacFadyen BV, Gliga L, AI-Kaisi NK, Vaneska GJ, Ponskey J (1988) Endoscopic and histologic correlates of intestinal ischemia in a canine model. Am Surg 54:69-72 9. Mitty WF, Purisima CO (1977) Agnogenic venous mesenteric thrombosis. J Am Geriatr Soc 25(11): 514-520 10. Nagasue N, Inokuchi K, Kobayashi M, Saku M (1977) Mesenteric venous thrombosis occurring late after splenectomy. Br J Surg 64:781-783 11. Sachs S, Morton JH, Schwartz S (1982) Acute mesenteric ischemia. Surgery 92(4): 646-653 12. Scully RE, ed (1990) Case records of the Massachusetts General Hospital (Case 9-1991). N Engl J Med 324(9): 613-623
Surgical EndOSCOpy 9 springer-VerlagNew York Inc. 1992
Mesenteric venous thrombosis M o d e r n m a n a g e m e n t and endoscopic diagnosis Terence P. Wade l, William R. Jewell 2, and Charles H. Andrus l i Departments of Surgery, St. Louis University, 3635 Vista Avenue at Grand Boulevard, St. Louis, MI 63111-0250 and the 2 University of Kansas, 39th Street at Rainbow Blvd., Kansas City, KS 66103
Summary. Mesenteric venous thrombosis has an obscure etiology, a prolonged onset of symptoms, and controversial methods of diagnosis and treatment. Eleven collected patients with pathologic confirmation of this diagnosis are presented, with one patient demonstrating the endoscopic appearance of this disease in human small bowel. Other methods of diagnosis are also reviewed. Treatment of mesenteric venous thrombosis requires resection of necrotic bowel and, we believe, anticoagulation. The role of planned "second-look" laparotomy is reviewed, but it was not necessary in those patients in whom definitive resection could be combined with intraoperative anticoagulation.
Key words: Mesenteric venous thrombosis- Smallbowel endoscopy - Intestinal ischemia
localized or diffuse involvement of bowel and mesentery. Controversy exists both in the diagnosis and treatment of this disorder. Although associated with a history of thrombotic disorders [2, 3], prior splenectomy [9], oral estrogens, and portal stasis [3], mesenteric venous thrombosis most commonly presents without apparent preexisting cause [8]. Physical examination is remarkable for diffuse abdominal tenderness without localizing signs. Radiologic studies are not usually of value, and laboratory exams usually reveal only a leukocytosis. Most authors recommend anticoagulation and resection when the diagnosis is made, but the requirement for a later, planned second-look laparotomy to assure bowel viability is in debate [5,6, 10]. The 11 patients presented here illustrate the diagnostic problems and suggest therapeutic management of this disease.
Materials and methods Intestinal ischemia can be caused by obstruction to blood flow in the arterial or venous system. The presentation of arterial ischemia is usually sudden and painful, and systemic changes often follow quickly. Obstruction of mesenteric venous flow is thrombotic, and the onset and progression of symptoms are slower than seen with arterial occlusion. Mesenteric venous thrombosis is the etiology of 10-30% [6, 7, 10] of cases with acute intestinal ischemia, yet its insidious clinical presentation makes preoperative diagnosis a rarity. First described as a clinical entity in 1935 [12], mesenteric venous thrombosis was recently reviewed in a case report [ 11] and as a compilation of all reported cases [1]. The disease is characterized by a prolonged onset of symptoms, a long interval between symptomatic onset and diagnosis, and either
Offprint requests to: Terence P. Wade
All cases of acute intestinal ischemia which occurred at the University of Kansas Medical Center between 1974 and 1983 were reviewed. Nine cases of mesenteric venous thrombosis were pathologically documented, representing 10% of the patients with intestinal ischemia. Two additional cases have been added to the authors' experience since 1983. The l l cases were reviewed with regard to complaints at presentation; symptoms and findings on exam; patient demographics; preoperative laboratory and diagnostic tests; findings at operation or autopsy; and final outcome.
Results The 11 patients were age 16-84 with a mean of 46 years. Seven were male, nine were white, and two were black. All 11 patients complained of abdominal pain lasting 3-7 days prior to exploration. Other complaints included nausea or vomiting in nine patients; six had diarrhea. Three patients had a coexisting malignancy, five had prior clotting anomalies, and two were taking oral estrogens. On exam, all had diffuse abdominal tenderness without localization, and most had decreased bowel sounds.
284
Nine of the 11 had a leukocyte count in excess of 10,000 cells/mm3; 8 of these had a count o v e r 15,000 and in 5 it was over 20,000. Of the additional two patients, one had a normal leukocyte count and one had a leukopenia (700 WBC/mm 3) secondary to chemotherapy. All abdominal x-rays revealed nonspecific dilated small bowel with air fluid levels, and one patient had a nondiagnostic visceral angiogram. The most recent patient had a gastric and proximal small-bowel endoscopic examination performed with a pediatric colonoscope (PCF, Olympus Corp.) p e r os, which revealed a sharply demarcated area of mucosal ischemia and edema beginning in the first jejunal loop. This segment of ischemic small bowel was estimated at 60 cm past the pylorus, or 150 cm from the incisors. Due to the ability to "telescope" nonfixated segments of small bowel over the endoscope, the reach of oral endoscopy past the pylorus is variable and difficult to quantitate. Abdominal CT scan in this patient revealed this jejunal loop to be thick-walled and distended while the remainder of the bowel appeared normal. Four patients died of their disease: two were discovered at postmortem exam, and one died immediately after operation. The fourth patient had been previously anticoagulated with heparin for migratory thrombophlebitis associated with carcinoma of the lung. The heparin had been discontinued prior to laparotomy, and the patient died of a massive pulmonary embolism 48 h after operation. Anticoagulation was started on the day of surgery in 8 of the 11 patients, and in all 7 survivors. However, appropriate levels of anticoagulation (PT or PTT of 1.5 times normal) were diffiult to maintain. One patient survived multiple pulmonary emboli in the first 48 h after laparotomy despite heparin therapy, and two others had postoperative bleeding requiring a dosage modification. "Second-look" laparotomy (routine, planned reexploration to detect recurrent bowel ischemia without signs or symptoms) was performed in five patients, four of which were heparinized. All of these were negative. No patient who was anticoagulated suffered recurrentbowel ischemia. Four wound complications occurred in three patients. Two patients required drainage of subcutaneous infections (and one developed an incisional hernia that required repair) while another had a wound hematomaJ seroma.
Discussion
This experience represents a comparatively recent collection of patients with mesenteric venous thrombosis and suggests that both the diagnosis and management of this disease may be changing. Preoperative CT scanning may show changes suggestive of this diagnosis, including bowel-wall thickening [1] and thrombosis within larger mesenteric veins [4]. This paper reports an endoscopic demonstration of mesenteric venous ischemia in the human. The muco-
sal edema and necrosis observed were very similar to that previously demonstrated in a canine experimental model [8]. Photographs of the canine model in this paper were very similar to those from the patient in the present report. The mucosal edema and thickening of the intestinal circular folds are the key to the endoscopic diagnosis of mesenteric venous thrombosis; arterial ischemia is characterized by a pale, nondistended mucosa as seen intraluminally. The sharp segmental mucosal change seen endoscopically in this patient was also seen later at operation on the serosa of the bowel segment. The demonstration of laparotomy of thickwalled small bowel and/or mesenteric veins distended with clot should also suggest this diagnosis. Small intestinal endoscopy could make an important contribution to diagnosis in the future if this work can be duplicated in further patients with mesenteric ischemia. Routine second-look procedures are not supported by this series. We see two explanations for this departure from prior reports: (1) anticoagulation was used frequently and usually was begun intraoperatively at the time of diagnosis; and (2) those patients who were not reexplored had a well-demarcated ischemic segment(s) of small bowel, and anastomoses were performed in clearly viable bowel. When all involved bowel and mesentery can be safely resected and heparin is begun during exploration, we feel routine, planned reexploration to evaluate the viability of the bowel is not necessary. However, reexploration still has a role in those patients with diffuse invovlement of long segments of small bowel in whom initial definitive resection is impossible. References 1. Abdu R, Zakhout B, Dallis D (1987) Mesenteric venous thrombosis. Surgery 101:383-389 2. Arko RF, Jewell CT, Dainer P, MacLeod WAJ (1979) Primary mesenteric vein thrombosis associated with antithrombin Ill deficiency. JAMA 242(21): 2324-2325 3. Drott C, Henning R, Hedner U (1978) Massive thrombosis of the superior mesenteric, splenic and portal veins. Acta Chir Scand 144:57-60 4. Harward TRS, Green D, Bergan JJ, Rizzo RJ, Yao JST (1989) Mesenteric venous thrombosis. J Vasc Surg 9(2): 328-333 5. Khodadadi J, Rozencwajg J, Nacash N, Schmidt B, Feuchtwanger MM (1980) Mesenteric vein thrombosis. The importance of a second-look operation. Arch Surg 115:315-317 6. Levy PJ, Krauz MM, Manny J (1990) Acute mesenteric ischemia: Improved results - - a retrospective analysis of ninety-two patients. Surgery 107(4): 372-380 7. Levy PJ, Krauz MM, Manny J (1990) The role of second-look procedure in improving survival time for patients with mesenteric venous thrombosis. Surg Gynecol Obstet 170(4): 287-291 8. MacFadyen BV, Gliga L, AI-Kaisi NK, Vaneska GJ, Ponskey J (1988) Endoscopic and histologic correlates of intestinal ischemia in a canine model. Am Surg 54:69-72 9. Mitty WF, Purisima CO (1977) Agnogenic venous mesenteric thrombosis. J Am Geriatr Soc 25(11): 514-520 10. Nagasue N, Inokuchi K, Kobayashi M, Saku M (1977) Mesenteric venous thrombosis occurring late after splenectomy. Br J Surg 64:781-783 11. Sachs S, Morton JH, Schwartz S (1982) Acute mesenteric ischemia. Surgery 92(4): 646-653 12. Scully RE, ed (1990) Case records of the Massachusetts General Hospital (Case 9-1991). N Engl J Med 324(9): 613-623
