Original article
Revascularization of the superior mesenteric artery after acute thromboembolic occlusion M. BjoÈrck, S. Acosta*, F. Lindberg², T. TroeÈng* and D. Bergqvist Departments of Surgery, Uppsala University Hospital, Uppsala, *Blekinge County Hospital, Karlskrona and ²SkellefteaÊ District Hospital, SkellefteaÊ, Sweden Correspondence to: Dr M. BjoÈrck, Department of Surgery, University Hospital, SE-751 85 Uppsala, Sweden (e-mail:
[email protected])
Background: The outcome and prognostic factors after revascularization of acute thromboembolic
occlusion of the superior mesenteric artery (SMA) are poorly documented. Methods: Sixty patients with acute thromboembolic occlusion of the SMA had revascularization procedures at 21 hospitals from 1987 to 1998. They were registered prospectively in the Swedish Vascular Registry. Patient ®les were analysed retrospectively. Results: The median age of the patients was 76 years; 73 per cent suffered from cardiac disease and 23 per cent had previous vascular surgery. Onset of symptoms was classi®ed as sudden (30 per cent), acute (33 per cent) or insidious (37 per cent). The occlusions were thought to be either embolic (67 per cent) or thrombotic (33 per cent). The diagnosis was suspected on ®rst examination in 32 per cent of patients, a group whose median time to operation was shorter (P = 0´01). Fifty-eight patients had an exploratory laparotomy and subsequent revascularization, and two were treated with thrombolysis alone. Secondlook laparotomy was performed in 41, and third look in eight patients; 19 required an additional bowel resection. The overall mortality rates were 43, 52, 60 and 67 per cent at 30 days, discharge, 1 and 5 years, respectively. No patient was dependent on intravenous nutrition after 1 year. Previous vascular surgery resulted in a higher institutional mortality rate (79 per cent; P = 0´02). Patients who had a sudden onset of symptoms outside hospital had a better outcome (mortality rate 27 per cent; P = 0´02). Conclusion: Many non-diagnostic radiological examinations were performed and a routine second-look is warranted. The results suggest that attempts at revascularization procedures for acute mesenteric ischaemia may improve the outcome. Paper accepted 21 March 2002
Introduction
Acute thromboembolic occlusion of the superior mesenteric artery (SMA) is a condition with a dismal prognosis. In a report on 53 patients who had surgery between 1969 and 1984 the mortality rate was 85 per cent1. Among 57 patients who had surgery between 1987 and 1993 the mortality rate was similar, 83 per cent2. Endean et al.3 reported a somewhat lower mortality rate of 60 per cent among 43 patients treated between 1993 and 2000, 14 of whom had undergone revascularization surgery; this suggested that contemporary treatment with more frequent use of revascularization might improve the prognosis. In a review in 1995, Bradbury et al.4 stated that `data on the long-term results of emergency mesenteric revascularization are sparse and dif®cult to access'. In the Swedish Vascular Registry (Swedvasc), indications for surgery, preoperative risk factors, details regarding the ã 2002 Blackwell Science Ltd
British Journal of Surgery 2002, 89, 923±927
operative procedure, and outcome at 1 month and 1 year are reported prospectively5. Previously data from the registry have been used to study the incidence of, and risk factors for, bowel ischaemia after aortoiliac surgery6,7 by combining the prospective data from the registry with retrospective data from the patient records. The aim of this study was to review the clinical course of patients with acute thromboembolic occlusion of the SMA, and to try to identify risk factors associated with survival in the subgroup of patients who undergo revascularization. Patients and methods
Records from Swedvasc between 1987 and 1998 were examined; only patients undergoing a vascular surgical reconstruction were registered. Among 7634 patients who had surgery for acute thromboembolism, 84 patients (1 per cent) from 26 hospitals had reconstruction of the SMA. The British Journal of Surgery 2002, 89, 923±927
923
924
Revascularization after acute occlusion of the superior mesenteric artery · M. BjoÈrck, S. Acosta, F. Lindberg, T. TroeÈng and D. Bergqvist
case notes of these patients were requested for review. Five hospitals, in which 14 patients were treated, decided not to participate. Ten of the remaining 70 patients were excluded: ®les could not be found for four, in three no revascularization procedure had been performed and one patient underwent reconstruction of the SMA after resection of a tumour. Two patients had surgery for aortic occlusive disease within 30 days before thromboembolic occlusion of the SMA. They were included in a cohort of patients with intestinal ischaemia complicating aortoiliac surgery, and had been studied previously6,7. The remaining 60 patients from 21 hospitals were included in the study. Statistical analysis Statistical analysis of continuous variables was performed with Levene's test for equality of variances, followed by Student's t test. Discrete variables were analysed with the Pearson c2 test. Postoperative survival was analysed according to the Kaplan±Meier method. P < 0´05 was considered signi®cant. Results
The median age was 76 (range 35±90) years in the 35 women and 25 men studied. Preoperative risk factors included previous reconstructive vascular surgery in 14 patients, diabetes in ten, hypertension in 25, cerebrovascular disease in 14, pulmonary disease in six, renal insuf®ciency in four and smoking in nine patients. Forty-four patients suffered cardiac disease (Table 1), but only two patients were taking anticoagulant medication. Both had interruption of anticoagulation treatment under protection of low molecular weight heparin after mesenteric embolization occurred. Eight of the patients reported previous symptoms compatible with intestinal ischaemia: four described intestinal
Table 1 Characterization of cardiac disease in 60 patients with acute thromboembolic occlusion of the superior mesenteric artery No. of patients Atrial ®brillation Valvular disease Cardiomyopathy Ischaemic heart disease Any kind of cardiac disease
34 2 2 15 44
(57) (3) (3) (25) (73)
Values in parentheses are percentages. Some patients had multiple pathology
British Journal of Surgery 2002, 89, 923±927
www.bjs.co.uk
angina, two had signi®cant weight loss and three had chronic diarrhoea. Symptoms and signs Seventeen patients (30 per cent) suffered a sudden onset of severe abdominal pain (within minutes), 19 (33 per cent) developed acute pain (within 1 h), 21 (37 per cent) had an insidious onset of pain and in three the onset was unknown. The onset was no more sudden among patients thought to have embolic disease (P = 0´3). The cause of acute occlusion was considered to be embolic in 40 patients and thrombotic in 20. The sources of emboli were atrial ®brillation (30 patients), acute myocardial infarction (®ve), aortic thrombus (one), cardiomyopathy (one) or unknown (three). Diarrhoea was reported by 21 patients and was bloodstained in 11; 33 patients had vomiting. On presentation, 13 had had abdominal pain for less than 1 h, 22 for less than 2 h, 42 for less than 12 h and 15 for more than 12 h. The signs on physical examination at presentation were recorded in 51 patients. In 27 (53 per cent) there was pain out of proportion to the signs, i.e. no sign of peritonitis in spite of intense pain. Fourteen patients (27 per cent) had signs of peritonitis. Other physical signs were distended (nine) or tympanic (four) abdomen, hyperperistalsis (two) and bloody stools on rectal examination (three). Two patients had simultaneous embolization to leg. The diagnosis of acute SMA occlusion was suspected at the time of the ®rst examination in 19 patients (32 per cent) because of atrial ®brillation (11), recent stroke (six) or acute embolism to the leg (two), or the history of sudden onset (eight).
Diagnosis Patients were recorded prospectively into the Swedvasc database; analyses of laboratory results were retrospective. The median white blood cell count on admission was 16 200 (range 5000±81 000) per mm3; only two patients had a normal count. Plain abdominal radiography was performed in 35 patients and barium follow-through in ®ve. Paralytic ileus was seen in eight patients, intestinal obstruction in four and in 23 patients the radiological picture was inconclusive. Five patients were examined with duplex ultrasonography. One examination was impossible due to excessive bowel gas, in three patients arterial ¯ow was veri®ed in the proximal segment of the SMA and in one patient no ¯ow was detected. Plain abdominal ultrasonography was performed in 16 patients. ã 2002 Blackwell Science Ltd
M. BjoÈrck, S. Acosta, F. Lindberg, T. TroeÈng and D. Bergqvist · Revascularization after acute occlusion of the superior mesenteric artery 925
Four computed tomography (CT) examinations without contrast were not diagnostic. Among six CT examinations with contrast enhancement two were diagnostic. In the false-negative examinations contrast was observed in the proximal SMA. The diagnosis was veri®ed by angiography in 11 and at laparotomy in 49 patients. One patient had angiography after laparotomy, followed by revascularization. Among the 12 patients examined by angiography ®ve had a patent proximal segment of the SMA, with either a visible embolus or a distal occlusion. These lesions were interpreted as embolic. Seven patients had thrombotic occlusions at the origin of the SMA. After investigation, a median of 16 (range 1´5±408) h after presentation, 58 patients underwent exploratory laparotomy. At the time of the operation 38 (66 per cent) were suspected to have mesenteric ischaemia. Among the 20 patients in whom the diagnosis was not suspected the preoperative diagnosis was peritonitis (nine), perforated appendicitis (®ve), intestinal obstruction (three), caecal volvulus, diverticulitis and gastrointestinal bleeding (one each). The median time between presentation and operation or thrombolysis was signi®cantly shorter among patients in whom the diagnosis was suspected at ®rst examination (6 h) than among those in whom it was not (24 h) (P = 0´01).
Outcome The 30-day and in-hospital mortality rates were 43 and 52 per cent respectively. The patients died a median of 6 (range 0±72) days after surgery. On routine follow-up the mortality rate had risen to 60 per cent at 1 year, but to only 67 per cent at 5 years. Many of the surviving patients were dependent on intravenous nutrition during the ®rst months. Among the 24 patients who were alive 1 year after surgery none was dependent on intravenous nutrition, but six required medication for diarrhoea. Prognostic factors In-hospital mortality was examined according to patientand procedure-related factors (Tables 2 and 3). A history of previous vascular surgery was associated with a higher inhospital mortality rate. Previous vascular surgical procedures were for embolism (four), chronic (four) or acute (two) ischaemia of the leg, aortic aneurysm (three) and thoracic aortic dissection. All procedures had been performed more than 1 year previously, except for two patients treated for embolism at the same admission.
Table 2 Patient-related prognostic factors after revascularization for acute mesenteric ischaemia
Treatment
No. of patients
In-hospital mortality (%) P²
All 40 patients with SMA embolism were treated by embolectomy; one had additional intraoperative thrombolysis. Among the 20 patients treated for SMA thrombosis, ten had thrombectomy. Adjunctive procedures included patch angioplasty (two), on-table angioplasty (two) and aortic thrombectomy (one). Eight patients were treated with mesenteric bypass from the aorta (seven) or the common iliac artery (one) to the SMA (®ve) or to both mesenteric arteries (three). Finally, two patients were treated by thrombolysis without laparotomy, one of whom also underwent angioplasty. In all patients revascularization was considered a prerequisite for survival. Ten of 40 patients with embolic disease and 14 of 20 with thrombosis also required bowel resection at the primary procedure. A second-look operation was performed in 41 patients and a third look in eight. Nineteen patients required additional bowel resection at relaparotomy. In all, 43 patients underwent bowel resection; 42 had a resection of the small intestine (median 145 (range 10±360) cm) and 20 had a colonic resection (median 30 (range 15±120) cm).
*Time from onset of pain until ®rst examination by physician. ²Pearson c2 test
ã 2002 Blackwell Science Ltd
www.bjs.co.uk
All patients 60 of 60 52 Female sex 35 of 60 57 Age > 70 years 48 of 60 50 Age > 80 years 19 of 60 63 Chronic bowel ischaemia 8 of 60 50 Smoking 9 of 51 33 Diabetes 10 of 60 40 Ischaemic heart disease 15 of 60 53 Previous vascular surgery 14 of 60 79 Hospitalized before onset 18 of 60 50 Embolic occlusion 40 of 60 50 Presenting symptoms and signs Sudden onset of pain 17 of 57 35 Sudden onset of pain at home 15 of 57 27 Acute or sudden onset of pain (< 1 h) 36 of 57 47 Insidious onset of pain (> 1 h) 21 of 57 62 Presentation within 1 h* 13 of 58 62 Presentation 1±12 h* 30 of 58 40 Presentation after 12 h* 15 of 58 67 Vomiting as presenting symptom 33 of 60 46 Diarrhoea as presenting symptom 21 of 60 43 Pain out of proportion 27 of 51 41 Peritonitis on admission 14 of 51 57
0´08
0´02
0´09 0´02
0´06
British Journal of Surgery 2002, 89, 923±927
926
Revascularization after acute occlusion of the superior mesenteric artery · M. BjoÈrck, S. Acosta, F. Lindberg, T. TroeÈng and D. Bergqvist
The median delay from the ®rst examination by a physician to laparotomy was shorter among the patients in whom the diagnosis was suspected, either initially or after diagnostic assessment (13 h), than among those in whom the diagnosis was not suspected (24 h) (P < 0´05), but this did not affect outcome. There was a non-signi®cant (P = 0´06) improvement in survival in patients who presented between 1 and 12 h after the onset of symptoms. All 13 patients who were evaluated within 1 h from onset of symptoms were in hospital at the time. Since there was a tendency to a lower mortality rate among those who had a sudden onset of symptoms, and suspecting that patients already in hospital might have an adverse outcome owing to multiple disease, these factors were combined. The 15 patients with a sudden onset of symptoms and who were not in hospital had a better outcome. Various factors related to management were not associated with outcome (Table 3), including timing of bowel resection (at primary operation or at repeat laparotomy), length of bowel resected, perioperative administration of mannitol, type of thromboprophylaxis and requirement for inotropic support. Discussion
In an 18-year retrospective review of 214 patients from Finland treated for acute intestinal ischaemia, the lowest mortality rate (51 per cent) was reported in the group of
Management-related prognostic factors after revascularization for acute mesenteric ischaemia
Table 3
All patients Suspicion at ®rst examination Time to operation (h)*
