priate clinical setting, s The diagnosis is made by ar- teriography, and the treatment is surgical, with em- bolectomy, vascular reconstruction, or resection of the infarcted segment as necessary, 5 depending on the clinical circumstances. Mesenteric artery thrombosis does not have a well defined clinical picture. It may often be characterized by malaise and abdominal pain of more than 24 hours duration, which becomes steadily and progressively more severe.3, 4 This is followed by nausea, vomiting, and bloody diarrhea. Patients generally have signs of artherosclerosis at other locations.3, 4 I have not seen any other case of embolism to the superior mesenteric artery causing massive bowel gangrene from the ligament of Treitz to the ascending colon. Infarction of the entire small bowel implies that thrombosis had occurred. The patient with non-occlusive mesenteric infarc- tion is classically in shock, has severe congestive heart failure and ~s taking digitalis, or is septic. 4 Digitalis may potentiate this disease because it may cause splanchnic vasoconstriction. 4 Non-occlusive infarction is also associated with decreased cardiac output and the atherosclerotic narrowing of the mesenteric ves- sels. Clinically, it is difficult, if not impossible, to dis- tinguish between venous thrombosis and non- occlusive mesenteric infarction. An angiogram is necessary to differentiate the two. The surgical success on the former is dismal. The latter can be treated with an infusion of a vas- odilator,4, ~ such as papaverine or prostaglandin (Fig- ures 1 and 2). Once the diagnosis is established, most surgeons will explore the abdomen to determine if it is necessary to resect bowel that has not recovered from ischemia. Mesenteric infarction due to aortic dissec- tion is unusual, and is associated with a very poor prognosis. 1 Venous thrombosis may be associated with
to laparotomy. Unfortunately, by the time peritoneal signs have developed, the mortality and morbidity rate are extremely high. I would like to mention a few points about intes- tinal angina, since the clinical presentation and prog- nosis are quite different from superior mesenteric ar- tery occlusion. Intestinal angina2, 6 is characterized by abdominal pain developing about 30 minutes after a meal. Since meals are followed by pain, the patient eats very small amounts of food. At rest, about one fourth of the cardiac output goes to the splanchnic cir- culation. But after meals about a third more blood flow is directed to the gut, so that the lesion is silent until the patient eats. Intestinal angina is most com- mon in patients aged 60 or 70 who often have a past history of stroke or myocardial infarction. On physical examination, the patient appears thin and mal- nourished. There may or may not be an abdominal bruit. Angiography is necessary for the diagnosis, and treatment consists of vascular reconstruction. 1,2 CONCLUSION In summary, one must maintain a high level of suspicion in order to make the diagnosis of mesenteric infarction. Cases that are due to embolism are more likely to be salvaged, while those that are thrombotic in nature may be unsalvageable regardless of treat- ment. The highest survival rates are to be expected with early angiography and early surgery. For patients with intestinal angina, I think the Question: How much time do you have from the onset of embolization until ~he development of gan- grene? ': Dr. Berguer: That depends on many factors, such as the location of the occlusion, the degree of athero- sclerosis, and the extent of collateral circulation. Once you occlude the blood supply to a segment of bowel, mucosal destruction is apparent within about 30 min- utes. 4 Dead bowel results in shifts of fluid into the peritoneal cavity, electrolyte and acid-base distur- bances, and peritonitis. The more time that elapses from infarction to surgery, the worse the prognosis. Patients may have a successful outcome as late as 12 to 24 hours after the embolic event, but gangrene may develop in as short as six hours. Some patients may even survive without treatment. We know this is so because we occasionally see patients with bowel stric- tures, that probably developed a s a result of small em- bolizations. 4 Clinically, these may have been repre- sented by episodes of nonspecific abdominal pain. As far as the patient's initial evaluation, I think that on occasion, crucial time may be lost because the emergency physician is preoccupied with obtaining an array of laboratory studies, for example, serum en- zymes, repeat white blood cell count, electrolytes and amylase, and by the time the surgeon is consulted, it is too late to do anything. A high index of suspicion for the diagnosis, early surgical consultation, and early angiography are necessary to increase the survival rate in bowel infarction. 7 Dr. Krome: In the initial evaluation, don't forget to look for a source of emboli such as rheumatic heart disease, atrial fibrillation, or myocardial infarction. What is the minimum length of small bowel that is necessary for survival? Dr. Berguer: There are many variables, but gen- erally, the prognosis is dismal if there is less than three feet of bowel. Intravenous hyperalimentation can certainly increase the survival rate, and can be continued indefinitely. But it doesn't make much sense to continue hyperalimentation when there is no hope whatsoever of increasing the length or the reab- sorptive capacity of the bowel. Bowel transplants have been attempted, but have been unsuccessful. Question: Is thumb printing a valuable sign on an abdominal film? Dr. Berguer: Thumb printing s is a sharply de- fined, filling defect of the bowel wall which looks like a mucosal or intramural tumor. It is usually multiple, variable in size, and asymmetrical in location, and is thought to be due to submucosal hemorrhage. Thumb printing is a rather late sign in bowel infarction. Other radiologic signs associated with bowel infarc- tion s are small and large bowel distention to the level of the splenic flexure, evidence of true mechanical obstruction, generalized or local ileus, and, rarely, gas in the portal system2 Question: In certain settings, might it be possible to proceed directly with laparotomy and omit angiog- raphy? Dr. Berguer: Certainly it is best to proceed with angiography in order to differentiate between emboli- zation, thrombosis, and non-occlusive infarction. The angiogram should be performed with a catheter tech- nique, not with the translumbar approach. Both an- teroposterior and lateral views should be obtained. If the surgeon does not have e~/sy access to technically good angiography, it may be better to proceed directlyconditions such as cirrhosis, sepsis, neoplasm or blood dyscrasia. 1 This patient was a heavy drinker who developed abdominal signs and symptoms that the initial exam- iner thought were compatible with pancreatitis. Cer- tainly, we know better than to exclude the diagnosis of pancreatitis on the basis of a normal amylase. The ab- dominal films were consistent with this diagnosis also. I'm afraid that during the initial period, I think I would have treated him just as you did, as a case of acute alcoholic pancreatitis. Ronald L. Krome, MD: In my experience, it is un- usual for a patient to present with acute pancreatitis and delirium tremens (DTs) at the same time, since the former occurs with acute alcohol ingestion, while the latter is an alcohol withdrawal syndrome, develop- ing about 48 hours after the cessation of drinking. Dr. Berguer: Even so, I don't think that DTs and acute pancreatitis are an impossible combination. Soon thereafter, the patient developed the signs and symptoms so characteristic of mesenteric infarc- tion, that is, bloody diarrhea and diffuse abdominal pain. His WBC rose from 15,000/cu mm to 24,000/cu mm over a few hours. UnfortunateIy, by the time these signs and symptoms developed he was unsal- vageable. One very important correlation that was overlooked, and which could have led to the diagnosis earlier, was the patient's prior history of embolization combined with his foot and abdominal pain. A logical corollary might be that he was throwing clots again. In addition, as I read over the patient's hospitali- zation chart, I note that another examiner was able to obtain a more detailed history that indicated that the patient had actually had intermittent abdominal pain for two days prior to admission. Although it would have been difficult and time-consuming to obtain a thorough history in this patient, it might have helped in making the diagnosis sooner. Another maneuver that should have been done was testing the stool for occult blood. Once the patient was shown to have oc- cult blood in the stool, an angiogram might have been performed. Dr. White: This patient's prothrombin time was prolonged 23 seconds despite the fact that he had not been taking anticoagulants. Couldn't rectal bleeding have been attributed to the prolonged prothrombin time? Dr. Berguer: At this point, our patient had ab- dominal pain and tenderness, diarrhea, and a marked leukocytosis. Blood in the stool points to bowel infarc- tion and ~n angiogram is indicated. Dr. White: Would the elevated prothrombin time lead you away from the diagnosis of mesenteric artery thrombosis? Dr. Berguer: No, absolutely not, since the initiat- ing factor in mesenteric artery thrombosis is most likely a platelet aggregate, with secondary thrombus formation leading to vascular occlusion. We occasion- ally see recurrent vascular occlusion in patients re- ceiving chronic anticoagulant therapy, despite the fact that prothrombin time is at a therapeutic level. Dr. Krome: Is the elevated prothrombin time a contraindication to angiography? Dr. Berguer: I don't believe so. In a circumstance such as this, I would correct the coagulopathy with fresh frozen plasma, and perform the angiogram. The most likely complication is a hematoma at the arterial puncture site, which can be controlled.r fluid levels visible on the upright film. Surgical consultation was obtained. The abdominal pain became progressively more severe, and the patient began having loose burgundy stools. His blood pressure was now 114/90, his pulse rate was 180, and his temperature was 38.8 C (102 F). In the operating room, surgeons found diffuse necrosis of the small bowel extending from the ligament of Treitz to the ascending colon, secondary to mesenteric artery thrombosis. There was thought to be no likeli- hood of survival if the bowel was resected, so the ab- domen was closed. The patient died 24 hours later. Ramon Berguer, MD. Unfortunately, the final diagnosis was probably not suspected because the pa- tient was relatively young, and because his history of alcohol abuse, in combination with the early physicial findings, led the physicians to consider acute pan- creatitis as the diagnosis. Mesenteric infarction can be caused by arterial embolism, venous or arterial thrombosis, non- occlusive infarction, aortic dissection, or arteritis.1, 3 Thrombosis usually occurs at the origin of the superior mesenteric artery and embolic occlusion usually occurs at a vessel bifurcation. 1 Occlusion of a mesenteric vessel results in submucosal and mucosal edema and hemor- rhage into a bowel lumen. 4 Classically, acute embolic infarction is characterized by the development of sudden, acute periumbilical pain, out of proportion with the physical findings.I, 4 The pain is very severe, rather like that of a perforated peptic ulcer. This is followed by nausea, vomiting, and diarrhea. At first diarrhea may not be bloody, but in a few hours, occult blood will be detectable in the stool, and finally, the stool will be grossly bloody. A history of rheumatic heart disease, myocardial infarction, or previous embolization would support a diagnosis of mesenteric artery embolism in the appro-