Faculty Bibliography

Splenic vein aneurysms are rare and are usually caused by portal hypertension. Symptoms are unusual, but may include rupture or abdominal pain. Diagnosis can usually be made either by means of duplex ultrasonography or computed tomography scanning. Treatment varies from noninvasive follow-up to aneurysm excision. We report an expanding splenic vein aneurysm in a young woman with abdominal and back pain and no history of portal hypertension. She was treated with aneurysm excision and splenectomy.

Twenty-four patients who underwent surgery for pararenal aortic aneurysms between January 1992 and April 1997 are reviewed. Eighteen patients had primary atherosclerotic aneurysms, three patients had symptomatic infected aneurysms, two patients had an aneurysm proximal to a prior aortic repair, and one patient had a pseudoaneurysm of a proximal aortic graft anastomosis. Thirteen patients underwent elective operation, five had an urgent operation, and six patients underwent an emergency procedure. Five patients had the proximal aortic clamp placed between the renal arteries (Group I), three patients had it placed between the superior mesenteric and the renal arteries (Group II), and 16 patients had it placed in a supraceliac location (Group III). Aneurysm size, age, sex, preoperative blood chemistries (including hemoglobin, hematocrit, liver function studies, and coagulation studies) were similar in all groups. Two patients in Group III were on hemodialysis preoperatively. Preoperative renal function (blood urea nitrogen and creatinine) was the same in all groups. Visceral ischemic time was 43.4 +/- 9.37 min to the distal kidney in Group I, 26.6 +/- 7.63 min in Group II, and 24.5 +/- 6.22 min in Group III. Mean transfusion requirements were similar in all groups. Two patients in Group I required postoperative hemodialysis. No patient in either Group II or III developed renal insufficiency. Mortality was the same in each group but was related to the urgency of operation (elective 7.6%, urgent 40%, emergent 50%). Intrarenal clamping (Group I) was associated with more renal and gastrointestinal complications than either suprarenal or supraceliac clamping. Although suprarenal and supraceliac clamping had similar results, our preference is supraceliac clamping because it is technically easy to achieve and is associated with few end-organ complications.

The etiology of perioperative stroke in patients undergoing isolated coronary artery bypass grafting (CABG) is multifactorial. One significant cause is thought to be high-grade internal carotid artery stenosis. Between April 1992 and June 1995, 1686 patients undergoing isolated CABG underwent preoperative carotid duplex scanning. This represented 77% of patients who underwent CABG during that time period (2188 patients). Sixty-eight patients (4.0%) had 80%-99% stenosis of at least one carotid artery. Fifteen patients underwent CABG without carotid intervention (Group I) and 53 patients underwent either carotid endarterectomy prior to CABG or simultaneous with CABG (Group II). Age, sex, history of prior neurologic events, ejection fraction, number of distal bypasses performed, total pump time, and aortic cross clamp times were similar between the two groups. Three patients in Group I developed a permanent postoperative neurologic deficit (20%) and one patient developed a transient deficit. The defect was focal and ipsilateral to high-grade stenosis in three patients and global in one. No patient in Group II developed either a transient or permanent neurologic deficit. There was one death in Group I in the patient who developed a global neurologic deficit and one death in Group II 2 weeks after CABG in a patient who had undergone prophylactic preCABG-carotid endarterectomy. Statistical analysis (Fisher's exact test, 2-tail) demonstrated a significant decrease both in total neurologic events (p = 0.001) and permanent neurologic defects (p = 0.005) in those patient undergoing prophylactic CE (Group II). Patients with 80%-99% carotid stenosis undergoing CEA prior to or in conjunction with isolated CABG have a decreased incidence of neurologic events postoperatively.

Twenty-five patients who underwent resection of a thoraco-abdominal aneurysm between 1985-1993 were reviewed to study determinants of survival in patients undergoing the procedure in a community hospital. Twelve procedures were performed electively, six urgently and seven emergently. Type I aneurysm was present in one patient (n = 1), Type II n = 7; Type III n = 5 and Type IV = 12. Hypertension (n = 17), cardiac disease (n = 10) and renal insufficiency (n = 4) were most common risk factors. Aneurysms were repaired using inclusion method without special techniques for renal or spinal cord preservation. Eighteen patients survived and were discharged; four patients died 30 days and three patients died 30 days. Causes of death were multisystem failure (n = 3), acute myocardial infarction (n = 2) coagulopathy (n = 1) and bowel infarction (n = 1). Major complications included renal failure (n = 2) myocardial infarction (n = 3), bleeding (n = 3), paraplegia (n = 1). Statistical significance was determined using Fisher's exact test-2 tail. Risk factors for death and complication included emergency or urgent surgery (4 deaths-emergent, 2 deaths-urgent) and preoperative renal insufficiency (2 deaths; 1 dialysis) 52% of patients in a community hospital setting underwent emergent or urgent operation and this accounted for 87% of deaths and most morbidity. Mortality in elective procedures was 8%. Based on this data, we believe that thoracoabdominal aneurysm resection can be reasonably undertaken in a community-type hospital.

Spontaneous dissection of the infradiaphragmatic abdominal aorta is a rare form of aortic dissection. Its natural history and management are not well defined. We have recently treated five patients with spontaneous aortic dissection. Two patients had acute dissections and three had chronic dissections. Three patients developed saccular aneurysms and underwent resection. Two patients had stable dissections and were treated medically. All patients are alive and well at 6 months to 5 years. We think that spontaneous aortic dissection can be treated as a variant of type III dissection with initial medical management, reserving surgery for those patients in whom a saccular aneurysm or a complication of the dissection develops.

The effect of descending thoracic aortomyoplasty using conditioned latissimus dorsi muscle on cardiac output in five mongrel dogs with pharmacologically induced congestive heart failure was evaluated. A neurovascular left latissimus dorsi flap was lifted and through a left thoracotomy placed around the proximal descending thoracic aorta. The flap was conditioned for 4-6 weeks with a neurostimulator using the following parameters: amplitude 0.5 V, pulse width 210 microseconds and frequency 2 Hz. The neurostimulator was then removed and a cardiomyostimulator inserted and programmed to burst-stimulate the muscle during diastole. Baseline measurements of central venous pressure, heart rate, mean arterial blood pressure, pulmonary capillary wedge pressure, and cardiac output were obtained with the cardiomyostimulator off and on (study 1). Heart failure was induced with a combination of propranolol and verapamil, and measurements again taken with the stimulator off and on. The neurostimulator was reimplanted to continue stimulation of the latissimus dorsi muscle, and another set of measurements taken at 6 weeks with the cardiomyostimulator off and on (study 2). Counterpulsation in control conditions (before cardiac failure) in both studies demonstrated no significant increase in cardiac output. However, mean(s.d.) cardiac output was significantly (P < 0.1) increased by muscle stimulation in dogs with heart failure (study 1: from 2.39(1.10) to 3.14(1.41)l/min; study 2: from 1.89(0.64) to 2.38(0.57)l/min). There was no significant difference in the increase in cardiac output associated with muscle stimulation between studies 1 and 2. The results indicate that the model can increase cardiac output in heart failure and that this improvement is constant over a 4-6 week period, suggesting that muscle fatigue may not occur.