Richard J. Gusberg

Vascular complications associated with spontaneous aortic dissection

Three hundred twenty-five cases of spontaneous aortic dissection seen at two institutions between 1965 and 1986 were reviewed to assess the incidence, morbid sequelae, and specific management of aortic branch compromise. Noncardiac vascular complications occurred in 33% of the study group, and in these patients the overall mortality rate (51%) was significantly (p less than 0.001) higher than in patients without (29%) such complications. Although aortic rupture was the strongest correlate of mortality (90%), death specifically related to vascular occlusion was common when such occlusion occurred in the carotid, mesenteric, and renal circulation. There was a strong correlation between stroke and carotid occlusion (22/26 cases), yet specific carotid revascularization was only used during the chronic phase of the disease. Similarly, peripheral operation was ineffective in reducing the mortality rate in the setting of mesenteric (87%) and renal (50%) ischemia. Fifteen patients required either fenestration or graft replacement of the abdominal aorta for acute obstruction, rupture, or chronic aneurysm development. Thirty-eight patients (12%) demonstrated some degree of lower extremity ischemia, and one third of these required a direct approach on the abdominal aorta or iliofemoral segments to restore circulation. Selected patients with acute aortic dissection may require peripheral vascular operation in accordance with a treatment strategy that directs initial attention to the immediate life-threatening complications.

Long-term experience with descending aortic dissection: The complication-specific approach

We analyzed long-term results in 71 patients (45 men and 26 women) treated over 17 years for documented descending aortic dissection. Forty-nine patients were treated medically and 22, surgically. Actuarial survival was 65% at 1 year, 57% at 3 years, 50% at 5 years, and 28% at 10 years for the whole group. For the group treated medically, survival was 73%, 63%, 58%, and 25% at 1 year, 3 years, 5 years, and 10 years, respectively, and for the group treated surgically, 47%, 40%, and 28% at 1 year, 3 years, and 5 years, respectively. Ten (20.4%) of the 49 medically treated patients died early (5 of rupture), and 14 (28.6%) died late (8 of dissection). Five medically treated patients crossed over to surgical management for complications of dissection. Among the surgically treated patients, 6 underwent standard graft replacement of the proximal descending aorta, 8 underwent the fenestration procedure (with a standardized retroperitoneal abdominal approach), and 4 underwent the thromboexclusion operation. Specific analysis of fenestration in 14 patients (including some with persistent descending aortic dissection after replacement of the ascending aorta for dissection) found it to be safe and effective. Actuarial survival after fenestration was 77%, 77%, and 53% at 1 year, 3 years, and 5 years, respectively. Thromboexclusion was found effective, and postoperative studies confirmed thrombosis of the descending aorta with preservation of the lowest intercostal arteries. Fifteen of the 21 surviving medically treated patients agreed to return for follow-up imaging. Nine had thrombosis of the false lumen. An interesting radiographic finding was that 4 of the 15 restudied patients had a saccular aneurysm in the aorta at the level of the left subclavian artery. We recommend a complication-specific approach to the management of descending aortic dissection. Uncomplicated dissection is treated medically, whereas complicated dissection is treated surgically, with realized rupture treated by standard graft replacement, limb ischemia treated by fenestration, and enlargement or impending rupture treated by thromboexclusion.

Ischemic colitis: patterns and prognosis.

We identified 47 patients with nonocclusive ischemia of the large intestine over a seven-year period. The mean age at presentation was 56.2 years, with a 2∶2∶1 male predominance. Associated medical illnesses were diabetes (17 percent), renal failure (5 percent), and hematologie disorders (5 percent). Six patients developed ischemic colitis after aortic surgery. The mean delay in diagnosis was 1.8 days (range, three hours to 23 days). The right colon was involved in 21 patients (46 percent). Overall, 15 of 16 patients were successfully treated non-operatively with bowel rest and antibiotics; one patient who was managed nonoperatively died. Among the 31 requiring intestinal resection, enteric continuity was reestablished in 14. Second-look laparotomy in eight patients revealed further ischemia in two (20 percent). Mortality in the operative group was 29 percent (9 of 31). No patient has developed recurrent ischemia (mean follow-up, 5.3 years). Ischemic colitis often occurs without an obvious predisposing event, may involve all segments of the large intestine, and frequently requires surgery. While its course may be self-limited, elderly and diabetic patients, as well as those developing ischemia following aortic surgery or hypotension, continue to have a poor prognosis.

Aneurysm Rupture after EVAR: Can the Ultimate Failure be Predicted?

OBJECTIVES To provide insight into the causes and timing of AAA rupture after EVAR. DESIGN Original data regarding AAA ruptures following EVAR were collected from MEDLINE and EMBASE databases. Data were extracted systematically and patient and procedural characteristics were analyzed. RESULTS 270 patients with AAA ruptures after EVAR were identified. Causes of rupture included endoleaks (in 160: type IA 57, type IB 31, type II 23, type III 26, type IV 0, endotension 9, unspecified 14), graft migration 41, graft disconnection 11 and infection 6. Most of the described AAA ruptures occurred within 2-3 years after EVAR. Mean initial AAA diameter was relatively large (65 mm). No abnormalities were present in 41 patients during follow-up before rupture. Structural graft failure was described in 96 and a fatal course in 119 patients. CONCLUSIONS Focus of surveillance on the first 2-3 years after EVAR may possibly reduce the AAA rupture rate, especially in patients with increased risk of early rupture (relatively large initial AAA diameter or presence of endoleak or graft migration). Better stent-graft durability and longevity is required to further reduce the AAA rupture risk after EVAR. Complete prevention will however remain challenging since AAA rupture may occur even if no predisposing abnormalities are present.

Initial experience with venous stents in exertional axillary-subclavian vein thrombosis ☆ ☆☆ ★

PURPOSE Exertional thrombosis of the axillary and subclavian veins, also known as Paget-Schrötter syndrome, has been increasingly recognized in recent years as a cause of long-term morbidity. Recent aggressive approaches to treating Paget-Schrötter syndrome have suggested the association of early failure with residual subclavian vein stenosis. As a result, the use of endoluminal stents has been proposed as an aid to venous percutaneous transluminal angioplasty for this disorder. METHODS This report outlines the therapy of 11 consecutive patients with Paget-Schrötter syndrome who were treated at our institution between October, 1992, and December, 1995. Stents were placed when percutaneous transluminal angioplasty was unsuccessful at achieving an adequate residual lumen. RESULTS Stents were placed after initial thrombolysis in six patients and in late follow-up in two patients. Of the six patients who had stents placed at initial thrombolysis, first-rib resection was eventually performed in four. In two patients first-rib resection was not performed, and stent fracture occurred in both. Late patency was achieved in the stents of six of the eight patients. CONCLUSIONS Trials to evaluate stents as an adjunct to conventional therapy seem warranted. The use of stents alone without first-rib resection, however, appears to be associated with stent fracture.

Simplified captopril renography in diagnosis and treatment of renal artery stenosis.

To improve the diagnosis and forecast the response to surgery or renal angioplasty in patients with hypertension and renal artery stenosis, we employed a simplified captopril renography protocol in conjunction with renal arteriography in 94 clinically selected patients. Fifty hypertensive patients (group 1) with a high clinical likelihood of renovascular hypertension were evaluated using a simplified captopril renography protocol and renal angiography on the arterial side. Criteria for normal captopril renal scintigrams were established based on this original cohort and validated in an additional 44 clinically comparable patients (group 2). Renal revascularization or nephrectomy was performed in 39 patients, and success of the procedure was determined in the 34 patients for whom 3-month follow-up was available. In the 94 patients, 44 (47%) had renal artery stenosis. Simplified captopril renography was 91% sensitive and 94% specific in identifying or excluding renal artery stenosis in the combined group, with no difference in the diagnostic utility between groups 1 and 2, or in those with renal insufficiency (n=38) or those with bilateral disease (n=17). Scintigraphic abnormalities induced by captopril were strongly associated with cure or improvement in blood pressure control following revascularization or nephrectomy (15 of 18), while the lack of captoprilinduced changes was associated with failure of such intervention (13 of 16) (/?=0.0004). We conclude that simplified captopril renography is highly sensitive and specific in the diagnosis of renal artery stenosis in a clinically selected high-risk population and that the test accurately predicts the success or failure of therapeutic intervention. {Hypertension 1991; 18:)

Concentrations of Glucagon and the Insulin: Glucagon Ratio in the Portal and Peripheral Circulation

Summary In postoperative patients studied in the postabsorptive state by mean of an umbilical vein catheter, concentrations of glucagon and insulin in portal blood exceeded those in peripheral venous blood. However, the portal—peripheral gradient for insulin (1.8-4.0) was consistently greater than that observed for glucagon (1.1-1.5). Consequently, the insulin:glucagon (I/G) molar ratio in the portal vein was 100% higher than in peripheral blood. A direct linear correlation however, was observed between portal and peripheral I/G molar ratios. The data indicate that peripheral measurements underestimate the absolute ratio of insulin to glucagon in portal blood, but nevertheless provide an index of relative hormonal availability in the hepatic circulation.

Gastrointestinal hemorrhage from pseudoaneurysms in pancreatic pseudocysts

Gastrointestinal hemorrhage secondary to hemosuccus pancreaticus is a rare condition that poses a significant diagnostic and therapeutic challenge. It is reported to occur most commonly in the setting of acute or chronic pancreatitis with rupture of pseudoaneurysms of the spleen or hepatic artery into the pancreatic duct. In this report three such cases have been reported. Abdominal ultrasound and CT scanning can noninvasively define pancreatic pseudocysts with a high degree of accuracy. Real-time ultrasonography may document a pulsatile pseudoaneurysm. Radionuclide arterial scanning, by demonstrating pooling of blood in the area of a pseudocyst, can point to the source of bleeding in patients with pancreatitis and gastrointestinal hemorrhage. Selective celiac angiography, however, is the only diagnostic test that can definitively outline a pseudoaneurysm and demonstrate its rupture into a pseudocyst or into the pancreatic duct. Pancreatic resection including excision of the pseudoaneurysm and pseudocyst (when present) is the treatment of choice. In cases where resection is not possible, ligation of the artery proximal and distal to the pseudoaneurysm and drainage of the pseudocyst into the gastrointestinal tract is an acceptable alternative procedure. Although intraarterial catheter embolization of the bleeding vessel can be a lifesaving procedure in these very sick patients, subsequent resection of the lesion is warranted as the definitive treatment.

Hepatic and extrahepatic splanchnic glucose metabolism in the postabsorptive and glucose fed dog

In awake dogs we measured the glucose balance across the liver and extrahepatic splanchnic tissues in the postabsorptive state and during two hours of IV infusion of glucose or for three hours following ingestion of oral glucose and during four hours of sequential intraportal followed by oral glucose. The IV glucose infusion rate was adjusted to maintain a steady state glucose concentration of either euglycemic levels (insulin clamp, group 1, N = 4), 125 mg/100 mL above the postabsorptive glucose concentration (+125 mg glucose clamp, group 2, N = 3) or 200 mg/100 mL above basal glucose levels (+200 mg glucose clamp, group 3, N = 7). Oral glucose was given at a dose of either 1.5 g/kg (group 4, N = 7) or 2.5 g/kg (group 5, N = 12). In dogs that received IV glucose, basal gut glucose uptake (0.5 +/- 0.1 mg/min X kg) was stimulated by hyperglycemia (1.5 +/- 0.5 and 1.4 +/- 0.1 mg/min X kg for group 2 and 3, respectively, P less than 0.05). In these same animals basal hepatic glucose output (-2.7 +/- 0.3 mg/min X kg) was promptly suppressed and net hepatic glucose uptake occurred (2.8 +/- 0.2 and 2.4 +/- 0.5 mg/min X kg in group 2 and 3 respectively). Euglycemic hyperinsulinemia (group 1) suppressed postabsorptive hepatic glucose release but did not enhance glucose removal by either the liver or gut tissues. After oral glucose gut tissues released absorbed glucose into portal blood. Over three hours following the glucose meal 74% and 59% of the ingested glucose was absorbed in group 4 and 5, respectively. As with IV glucose, postabsorptive hepatic glucose production was suppressed and over the first two hours after feeding the liver took up glucose (3.4 +/- 1.0 and 3.1 +/- 0.7 mg/min X kg groups 4 and 5, respectively) at a rate similar to that seen with IV glucose. To further examine the effect of the route of glucose administration on liver glucose handling, hepatic glucose balance was measured serially over four hours in three dogs that received IV glucose into a mesenteric vein to produce portal hyperglycemia (+125 mg/dL portal glucose clamp N = 3). Oral glucose (2.5 mg/kg) was given at two hours, and the rate of the mesenteric glucose infusion adjusted to maintain portal glycemia constant. The hepatic glucose balance averaged 5.5 mg/min X kg over the 0 to 2 hour period and 4.2 +/- 1.0 mg/min X kg over the 2 to 4 hour time.(ABSTRACT TRUNCATED AT 400 WORDS)

Effect of renal insufficiency on outcome following infrarenal aortic surgery

BACKGROUND This study reviewed the effect of preoperative renal insufficiency on outcome following elective infrarenal aortic surgery. METHODS The charts of 210 consecutive patients undergoing aortic surgery (occlusive disease, 15%; aneurysmal disease, 78%; or combined disease, 7%) from 1990 to 1995 were categorized into three groups based on preoperative creatinine ([Cr] group 1 Cr < 1.5, n = 171; group 2 Cr 1.5 to 1.7, n = 22; and group 3 Cr > or = 1.8, n = 17) and calculated creatinine clearance ([CrCl] CrCl > 45 mL/min, n = 162 versus CrCl < 45 mL/min, n = 48). Patients with renal artery stenosis or those who required suprarenal cross clamping or emergency procedures were excluded. Differences in postoperative intensive care unit (ICU) stay, ventilator days, dialysis dependence, morbidity, and, mortality were compared. RESULTS Patients in groups 2 and 3 had an increased incidence of postoperative dialysis dependence (group 2 9%, group 3 8%) when compared with patients in group 1 (group 1: 0%, P < 0.05). Patients in the CrCl > 45 group had a lower mortality rate when compared with patients with a CrCl < 45 (CrCl > 45 0.6% versus CrCl < 45 8%, P <0.05) a lower incidence of dialysis (0% versus 7%, P <0.05), and a lower incidence of postoperative serum creatinine elevation from baseline (CrCl > 45 8% versus CrCl < 45 18%, P <0.05). There was no significant difference in morbidity, ICU stay, or ventilator days between the groups. Upon regression analysis, preoperative CrCl but not Cr was predictive of postoperative mortality (P <0.05). Serum Cr was more predictive than CrCl of impaired renal function postoperatively. CONCLUSIONS Preoperative CrCl is more accurate than Cr as a predictor of postoperative mortality. Patients with preoperative CrCl < 45 mL/minute who undergo elective aortic surgery have a significant increase in postoperative cardiac-related mortality and dialysis.