Influence of patient characteristics and treatment options on outcome of patients with prosthetic aortic graft infection.

This study was undertaken to determine the influence of patient characteristics and treatment options on survival and limb loss after treatment of prosthetic aortic graft infection. Fifty-three patients treated for prosthetic aortic graft infection were reviewed. Twenty-three presented with groin infection, 12 with sepsis, 10 with aortoenteric fistula, 4 with limb ischemia, and 4 with pseudoaneurysm. Treatment included staged extraanatomic bypass (EAB) plus graft excision in 23 patients, simultaneous EAB and graft excision in 18, in situ graft replacement in 5, and local therapy only in 7. Axillofemoral bypass was done for revascularization in 53 limbs and axillopopliteal bypass in 16 limbs. The results of this study showed that morbidity and mortality of prosthetic aortic graft infection is influenced by the presentation and type of treatment of the infected graft. Staged axillofemoral bypass (when possible) plus graft excision appears to be associated with acceptable outcome (survival with limb salvage in 74%).

Attenuation of skeletal muscle ischemia/reperfusion injury by inhibition of tumor necrosis factor.

PURPOSE: Tumor necrosis factor alpha (TNF-alpha) has been shown to play a role in pulmonary injury after lower-extremity ischemia/reperfusion (I/R). However, its role in direct skeletal muscle injury is poorly understood. The hypothesis that endogenous TNF production contributes to skeletal muscle injury after hindlimb I/R in rats was tested. METHODS: Juvenile male Sprague-Dawley rats underwent 4 hours of bilateral hindlimb ischemia and 4 hours of reperfusion (IR) or sham operation (SHAM). A subset was treated with a soluble TNF receptor I construct (STNFRI, 10 mg/kg) 1 hour before ischemia (PRE) or at reperfusion (POST). Direct skeletal muscle injury (SMII) and muscle endothelial capillary permeability (MPI) were quantified by means of Tc99 pyrophosphate and I125 albumin uptake. Pulmonary neutrophil infiltration and hepatocellular injury were assessed by means of myeloperoxidase content (MPO) and aspartate aminotransferase (AST) concentrations, respectively. Serum TNF bioactivity was measured with the WEHI bioassay. RESULTS: Hindlimb I/R (IR vs SHAM) resulted in a significant (P <.05) increase in the SMII (0.52 +/- 0.06 vs 0.07 +/- 0.01) and MPI (0.35 +/-.04 vs 0.06 +/- 0.01). Pretreatment with STNFRI (PRE vs IR) significantly ameliorated both SMII (0.30 +/- 0.05 vs 0.52 +/- 0.06) and MPI (0.23 +/- 0.02 vs 0.35 +/- 0.04), whereas treatment at reperfusion (POST vs IR) had no effect. Hindlimb I/R (IR vs SHAM) resulted in both significant pulmonary neutrophil infiltration (MPO 16.4 +/- 1.06 U/g vs 11.3 +/- 1.4 U/g) and hepatocellular injury (AST 286 +/- 45 U/mL vs 108 +/- 30 U/mL), but neither was inhibited by pretreatment with STNFRI before ischemia. Detectable levels of TNF were measured during ischemia in a significantly higher percentage of the IR group compared with SHAM (9 of 12 vs 3 of 12), and the maximal TNF values were also significantly greater (51.1 +/- 12.6 pg/mL vs 5.5 +/- 2.9 pg/mL). No TNF was detected in any treatment group during reperfusion nor after administration of the STNFRI. CONCLUSION: Acute hindlimb IR initiates a systemic TNF response during the ischemic period that is partly responsible for the associated skeletal muscle injury.

Impact of a clinical pathway for elective infrarenal aortic reconstructions.

OBJECTIVE: To determine the impact of a clinical pathway for elective infrarenal aortic reconstruction on outcome, resource utilization, and cost in a university medical center. SUMMARY BACKGROUND DATA: Clinical pathways have been reported to control costs, reduce resource utilization, and maintain or improve the quality of patient care, although their use during elective aortic reconstructions remains unresolved. METHODS: A clinical pathway was developed for elective infrarenal aortic reconstructions by a multidisciplinary group comprised of representatives from each involved service. The prepathway practice and costs were analyzed and an efficient, cost-effective practice with specific outcome measures was defined. The impact of the pathway was determined by retrospective comparison of outcome, resource utilization, and cost (total and direct variable) between the pathway patients (PATH, n = 45) and a prepathway control group (PRE, n = 20). RESULTS: There were no significant differences in the patient demographics, comorbid conditions, operative indications, or type of reconstruction between the groups. There were no operative deaths and the overall complication rate (PRE, 35% vs. PATH, 34%) was similar. The pathway resulted in significant decreases in the total length of stay and preoperative length of stay and a trend toward a significant decrease (p = 0.08) in the intensive care length of stay for the admission during which the operation was performed. The pathway also resulted in significant decreases in both direct variable and total hospital costs for this admission, as well as a significant decrease in the overall direct variable and total hospital costs for the operative admission and the preoperative evaluation (< or =30 days before operative admission). Despite these reductions, the discharge disposition, 30-day readmissions, and number of postoperative clinic visits within 90 days of discharge were not different. CONCLUSIONS: Implementation of a clinical pathway for elective infrarenal aortic reconstructions dramatically decreased resource utilization and hospital costs without affecting the quality of patient care and did not appear to shift the costs to another setting.

Improving the cost-effectiveness of carotid endarterectomy.

PURPOSE: Carotid endarterectomy (CEA) has been shown to significantly reduce the risk of stroke caused by carotid artery stenosis. Limiting the costs of CEA without increasing the risks will improve the cost-effectiveness of this procedure. METHODS: Results were prospectively collected from 63 consecutive CEAs performed in 60 patients who were entered into a clinical pathway for CEA that included avoidance of cerebral arteriography, preferential use of regional anesthesia, selective use of the intensive care unit (ICU), and early hospital discharge. The mortality rate, complications, hospital costs, and net income in these patients were then compared with results from 45 CEAs performed in 42 consecutive patients immediately before beginning the CEA pathway. Age, comorbid risk factors, incidence of symptoms, and degree of carotid artery stenosis were similar in both patient groups. RESULTS: The rates of mortality and complications associated with CEA were low (mortality rate, 0%; stroke, 0.9%; transient ischemic attack, 2.8%) and did not vary between the two groups. Implementation of the CEA pathway resulted in significant (p < 0.001) reductions in the use of arteriography (74% to 13%), general anesthesia (100% to 24%), ICU use (98% to 30%), and mean hospital length of stay (5.8 days to 2.0 days). These changes resulted in a 41% reduction in mean total hospital cost ($9652 to $5699) and a 124% increase in mean net hospital income ($1804 to $4039) per CEA (p < 0.01). For the 39 patients (62%) who achieved all elements of the CEA pathway, the mean hospital length of stay was 1.3 days, the mean hospital cost was $4175, and the mean hospital income was $4327. CONCLUSIONS: Costs associated with CEA can be reduced substantially without increased risk. This makes CEA an extremely cost-effective treatment of carotid disease against which new therapeutic approaches must be measured.

Intraoperative prebypass arteriography for infrageniculate revascularization.

BACKGROUND: Selection of the optimal distal target for infrageniculate arterial revascularization remains difficult in patients with multilevel occlusive disease due to poor visualization of the distal vasculature by preoperative arteriography. Prebypass, intraoperative arteriography (IOA) with direct injection of contrast into the infrageniculate arteries may improve distal arterial visualization and bypass target selection. METHODS: One hundred fourteen extremities in 104 consecutive patients requiring infrageniculate bypass were prospectively studied. All patients underwent preoperative contrast arteriography (CA) of the aortoiliac and lower extremity arteries using iodinated and/or CO2 contrast and digital subtraction techniques. IOAs were obtained at operation to confirm the adequacy of the distal runoff from the predicted bypass target and to identify potentially superior targets. The preoperative plan formulated from the CA was compared to the actual procedure performed based on the IOA. The CA and IOA were also independently reviewed postoperatively by two blinded vascular surgeons to determine the number of patent vessel segments visualized and the number of segments with <50% stenosis. RESULTS: Revascularization was done in 105 of 114 extremities (92%), whereas amputation was required as the initial procedure in 9 patients (8%). The IOA altered the operative plan based on the CA in 27 of 144 patients (24%). Changes in the planned bypass procedures included selection of a more distal anastomotic site in 13 of 102 patients (13%), selection of a more proximal anastomotic site in 4 of 102 (4%), selection of a different artery for the distal anastomosis in 3 of 102 (3%), and amputation rather than bypass in 2 of 102 patients (2%) with no suitable distal bypass target on the IOA. The IOA also resulted in bypass rather than planned amputation in 5 of 12 patients (42%) deemed unreconstructable on the preoperative CA. A mean of 13 minutes was required for IOA and an adequate study was obtained on the first attempt in 98 of 144 patients (86%). On postoperative review, more patent vessel segments but fewer segments with <50% stenosis were identified on the IOA compared to the CA. CONCLUSIONS: Prebypass intraoperative arteriography facilitates selection of the optimal distal bypass target during infrageniculate revascularization and can result in initial limb salvage in select patients deemed unreconstructable by preoperative contrast arteriography.

Exogenously administered interleukin-10 decreases pulmonary neutrophil infiltration in a tumor necrosis factor-dependent murine model of acute visceral ischemia.

INTRODUCTION: Visceral ischemia and reperfusion associated with thoracoabdominal aortic aneurysm (TAAA) repair results in lung injury, which appears to be mediated in part by proinflammatory cytokines. The purpose of this study was to determine the effect of exogenous administration of the antiinflammatory cytokine, recombinant human IL-10 (rhIL-10), on proinflammatory cytokine production (IL-6 and TNF alpha) and pulmonary neutrophil infiltration after acute visceral ischemia-reperfusion. METHODS: Two hours before 25 minutes of supraceliac aortic occlusion, 80 C57BL/6 mice (20 to 22 g) received an intraperitoneal injection of rhIL-10 (0.2 microgram [n = 20], 2 micrograms [n = 20], 5 micrograms [n = 25], or 20 micrograms [n = 15]), and 16 mice received murine anti-IL-10 IgM 200 micrograms. Twenty-five additional mice underwent visceral ischemia-reperfusion without treatment (controls), and 16 mice underwent laparotomy without aortic occlusion (sham). RESULTS: Pretreatment with exogenous rhIL-10 resulted in significant reductions in lung neutrophil infiltration with 0.2 microgram, 2 micrograms, and 5 micrograms per mouse of rhIL-10 compared with lung neutrophil levels in control mice that underwent acute visceral ischemia-reperfusion alone (p < 0.05). In addition, serum TNF alpha was detected in 50% of control mice and in 75% of mice that received murine anti-IL-10, but in none of the mice that received rhIL-10 (2 micrograms per mouse) or the mice that underwent sham operative procedures (p < 0.05 by chi 2 analysis). CONCLUSION: Exogenous IL-10 limits pulmonary neutrophil recruitment and the appearance of TNF alpha in this model of visceral ischemia-reperfusion injury. Thus the use of exogenous IL-10 may offer a novel therapeutic approach to decrease the complications that are associated with TAAA repair.

Intraoperative autologous transfusion during elective infrarenal aortic reconstruction.

Intraoperative autologous transfusion devices have been purported to reduce allogenic transfusions and their associated complications. However, the value of their routine use during elective cardiovascular operations remains undefined. This study was designed to examine the efficacy of the Haemonetics Cell Saver (CS) during elective aortic reconstructions and identify predictors of clinically significant (> or = 500 cc) and cost-efficient (> or = 1250 cc) salvage volumes. The medical records of all patients undergoing elective infrarenal aortic reconstructions between January 1991 and June 1995 were retrospectively reviewed to determine blood loss, CS return, predictors of clinically significant/ cost-efficient CS returns, blood products transfused, and estimated cost per unit CS return. The CS was used for 138 (82.1%) of all reconstructions during the study period. Estimated blood loss (2127 +/- 1467 vs 1415 +/- 1047) and CS return (927 +/- 790 vs 515 +/- 408) were significantly greater in patients with aneurysms (AAA, N = 63) compared to those with aortoiliac occlusive disease (AIOD, N = 75). CS returns > or = 500 cc were common (79.4% AAA, 52.0% AIOD) and predictors of > or = 500 cc CS returns were large aneurysms (6.79 +/- 1.84 vs 5.72 +/- 0.71 cm) and male sex (82.0 vs 46.2%) in AAA patients and lower preoperative platelet counts (262 +/- 93 vs 311 +/- 113 K/mm3), concomitant renal revascularizations (20.5 vs 0%), and prolonged operative time (7.9 +/- 2.4 vs 6.9 +/- 2.1 hr) in AIOD patients. In contrast, CS returns > or = 1250 cc were relatively uncommon (28.6% AAA, 5.3% AIOD), and predictors of these CS returns were found only for AAA patients and included any concomitant vascular procedures (38.8 vs 15.6%) and the need for suprarenal aortic clamping (27.8 vs 6.7%). Despite the use of the CS, 73.8% of all patients required allogenic packed red blood cells with a mean of 3.0 +/- 3.1 units transfused in the perioperative period; no difference was seen between AAA and AIOD patients. The calculated cost for a unit of CS return was +128.77 for the AAA patients and +231.91 for the AIOD patients. Not using the CS and substituting the return with allogenic packed red blood cells would have saved +252.80 and +352.84 for the AAA and AIOD patients, respectively. Routine use of the CS during elective infrarenal aortic reconstructions is not cost efficient and should be abandoned. Use of the device should be reserved only for complex reconstruction.

Visceral ischemia-reperfusion injury promotes tumor necrosis factor (TNF) and interleukin-1 (IL-1) dependent organ injury in the mouse.

Acute visceral ischemia and subsequent reperfusion injury, which accompanies the surgical repair of a thoracoabdominal aorta aneurysm, is associated with high rates of morbidity and mortality. The purpose of the present study was to determine whether endogenous tumor necrosis factor-alpha (TNF-alpha) and interleukin-1 (IL-1) production contributes to organ dysfunction in animals subjected to visceral ischemia secondary to 30 min of supraceliac aortic occlusion. C57BL6/j mice were treated with either a TNF binding protein (TNF-bp-10 mg/kg) or an anti-IL-1 receptor type 1 antibody (150 micrograms) 2 h prior to 30 min of supraceliac aortic occlusion. An additional group of mice received 30 min of infrarenal aortic occlusion to determine the contribution of lower torso ischemia-reperfusion injury to the changes seen following supraceliac aortic occlusion. Visceral organ ischemia for 30 min produced by supraceliac aortic occlusion followed by 2 h of reperfusion produced measurable TNF-alpha in 38% of untreated mice, but TNF-alpha was undetectable in both sham-operated mice and following infrarenal aortic occlusion. After 2 h of reperfusion, lung myeloperoxidase levels were significantly elevated in the mice experiencing visceral ischemia-reperfusion compared with either a sham operation or infrarenal ischemia-reperfusion (11.6 +/- 1.3 U/g vs. 3.4 +/- .2 U/g and 3.7 +/- 1.0 U/g, respectively, p < .05). Pretreatment with TNF-bp and anti-IL-1 antibody decreased lung neutrophil recruitment (7.2 +/- 1.2 U/g and 4.6 +/- 1.1 U/g) and capillary membrane permeability changes in mice following visceral ischemia-reperfusion. The present study demonstrates that brief (30 min) clinically relevant visceral ischemia produces TNF-alpha and IL-1 dependent lung injury.

Prediction of cure of hypertension in atherosclerotic renal artery stenosis.

We analysed the outcome of 63 consecutive, adequate interventions for atherosclerotic renal artery stenosis and hypertension: 34 patients had percutaneous transluminal renal angioplasty, and 29 had surgical correction. Hypertension was cured in 21% of patients and improved in 47%, but 32% failed to respond. We analyzed clinical variables predictive of cure. Duration of hypertension, level of diastolic blood pressure, and sex were found to be predictive of cure. The highest probability of cure was found in men with a duration of hypertension of less than 10 years and an initial diastolic blood pressure of greater than 80 mm Hg. Use of these clinical variables in a tree-based model correctly classified 80% of cases, with a sensitivity of 92% and a specificity of 77%. We conclude that a tree-based clinical algorithm based on only three clinical criteria correctly predicted cure of hypertension in most patients with renal artery stenosis and may be useful in decision making. A prospective analysis will be required to evaluate the clinical validity of the algorithm.

Visceral ischemia and organ dysfunction after thoracoabdominal aortic aneurysm repair. A clinical and cost analysis.

OBJECTIVE: Repair of thoracoabdominal aortic aneurysms (TAAAs) is associated with significant postoperative morbidity and mortality. Reperfusion of acutely ischemic abdominal viscera in animals leads to release of multiple factors that cause local and distant organ damage, and similar phenomena occurring in humans after TAAA repair could contribute to the high morbidity/mortality and cost associated with this procedure. METHODS: Twenty-nine patients undergoing elective TAAA repair were studied prospectively. Preoperative organ dysfunction and intraoperative risk factors (cross-clamp time, blood loss, operative time) were assessed and compared with postoperative organ dysfunction (defined as: pulmonary, positive pressure ventilation for > 7 days; renal, increase in serum creatinine > 2.0 mg/dL over baseline; hepatic, lactate dehydrogenase > 500 international units and total bilirubin > 3.0 mg/dL or serum transaminase level > 200 international units; hematopoietic, platelet count > 50 K or leukocyte count > 4.5 K, mortality, and costs. RESULTS: No relationship between preoperative organ dysfunction, blood loss, or operative time and postoperative organ dysfunction or mortality was seen; however, cross-clamp times > 40 minutes were associated with a significantly greater incidence of pulmonary (59%), renal (47%), hepatic (35%), and hematopoietic (47%) dysfunction. In addition, multiple-organ dysfunction (> 2 organ systems) was more common after > 40 minutes of visceral ischemia and led to significantly greater overall hospital ($88,465 + $76,155 vs. $41,782 + $31,244) and intensive care unit ($26,726 + $28,256 vs. $11,234 + $12,146) costs (p < 0.01, Mann-Whitney U test). Mortality associated with leukopenia was 67% compared with 4% without leukopenia (p < 0.01). CONCLUSION: Increasing durations of acute visceral ischemia led to significant multiple organ dysfunction after TAAA repair. Methods of limiting visceral ischemia or the systemic effects of visceral ischemia may decrease both the morbidity and mortality and the overall hospital cost associated with this procedure.

Increased soluble interleukin-1 type II receptor concentrations in postoperative patients and in patients with sepsis syndrome.

Plasma interleukin-1 (IL-1) activity is modulated in part through the simultaneous appearance of several inhibitors of IL-1 action, including interleukin-1 receptor antagonist (IL-1ra) and the soluble IL-1 type II receptor (IL-1RII). However, little is known concerning the plasma appearance of these inhibitors in patients following operative trauma or those with sepsis syndrome. In the present report, plasma IL-1beta, IL-1ra, and soluble IL-1RI and IL-1RII concentrations were evaluated in 118 patients with sepsis syndrome or after elective operative trauma. Plasma concentrations of IL-1ra increased significantly following elective operative repair of thoraco-abdominal and abdominal aortic aneurysms, and after bowel resection for inflammatory bowel disease, but did not increase after laparoscopic cholecystectomy. Plasma IL-1ra levels were also elevated in patients with sepsis syndrome. In contrast, soluble IL-1RII levels were only increased in patients after operative repair of thoraco-abdominal aortic aneurysms and in sepsis syndrome, whereas concentrations were unaffected by the other more modest surgical procedures. Plasma IL-1RI concentrations decreased in all postoperative patients in the first 24 hours after surgery. We conclude that both plasma IL-1ra and soluble IL-1RII concentrations often increase in sepsis and following some operative trauma. Less severe operative trauma increases the plasma concentration of only IL-1ra, whereas both IL-1ra and soluble IL-1RII are increased in patients with sepsis syndrome or following thoraco-abdominal aneurysm repair.

Operative mortality rates after elective infrarenal aortic reconstructions.

PURPOSE: This study was designed to test the hypothesis that cardiac complications (myocardial infarction, congestive heart failure, fatal arrhythmias) are no longer the leading cause of death after elective aortic reconstructions. METHODS: The medical records of all elective infrarenal aortic reconstructions performed between January 1982 and June 1994 were retrospectively reviewed. All perioperative deaths were analyzed to determine the cause of death and were compared with a subset of 266 survivors to identify any associated preoperative or intraoperative factors. RESULTS: Seven hundred twenty-two aortic reconstructions were performed for aneurysmal or occlusive disease, and there were 44 deaths (overall mortality rate of 6.1%). The mortality rate after aortic reconstruction alone was 4.9% and increased with the addition of renal (8.9%, p = 0.16) or lower extremity vascular procedures (15.8%, p = 0.01). Multisystem organ failure (MSOF) was the cause of death in 56.8%, of the patients (3.5% overall mortality rate) followed by cardiac events in 25% (1.5% overall mortality rate). Visceral organ dysfunction was the most common cause of MSOF leading to death in 14 patients (56.0%), and postoperative pneumonia was responsible for the fatal MSOF in nine patients (36.0%). Patient age, history of myocardial infarction/congestive heart failure, ejection fraction less than 50%, duration of operative time, and performance of additional procedures were associated with increased operative mortality rates by multivariate analysis. CONCLUSIONS: MSOF, predominantly from visceral organ dysfunction, was the leading cause of death after elective infrarenal aortic reconstruction. The risk of MSOF and operative death increases with the complexity of the procedure and the number of comorbid conditions.

Selection of patients for renal artery repair using captopril testing.

BACKGROUND: Prediction of improvement following surgical or radiologic intervention in patients thought to have renovascular hypertension (RVH) is often unreliable. Use of the angiotensin-converting enzyme inhibitor captopril in conjunction with measurement of peripheral renin levels or radioisotope renograms is thought to detect patients with functionally significant renal artery stenosis. However, it is unclear whether these tests can identify patients whose hypertension will significantly improve after renal artery repair. PATIENTS AND METHODS: The records of 52 consecutive hypertensive patients undergoing captopril studies followed by renal artery repair were reviewed. All patients had either renal artery stenosis > 75% or renal artery occlusion. Preprocedure evaluation included a captopril challenge test (measurement of peripheral renin levels after captopril ingestion) (n = 12) or a captopril renogram (determination of renal blood flow and glomerular filtration rate before and after captopril administration) (n = 40). Either renal artery bypass/nephrectomy (n = 41) or balloon angioplasty (n = 11) was done in all patients (18 bilateral/34 unilateral). No periprocedural deaths occurred. All surgically placed bypass grafts were shown to be patent by contrast or carbon dioxide arteriography before hospital discharge. RESULTS: Preprocedure captopril tests were positive (suggestive of RVH) in 39 patients (75%) and negative in 13 (25%). All patients with positive captopril tests had improvement in their RVH after intervention (17 cured, 22 improved) while 8 of 13 patients with negative captopril tests had no improvement in blood pressure control. Four of five false-negative tests were associated with a unilateral total renal artery occlusion, making detection of a postcaptopril effect impossible. If these 4 patients are excluded from analysis, preprocedure captopril testing was 98% accurate in predicting postprocedure outcome. CONCLUSIONS: Preprocedure captopril testing permits extremely accurate selection of patients with renal artery stenosis who will benefit from renal artery repair.

Limb-threatening ischemia due to multilevel arterial occlusive disease. Simultaneous or staged inflow/outflow revascularization.

SUMMARY BACKGROUND DATA: Limb-threatening ischemia due to severe multilevel arterial occlusive disease may require both inflow and outflow bypass to achieve limb salvage. Simultaneous inflow/outflow bypass has been advocated because the cumulative risks of separate staged inflow/outflow procedures can be avoided. However, the magnitude of complete revascularization is substantial; thus, the morbidity and mortality of simultaneous inflow/outflow bypass may be excessive. METHODS: The medical records of 450 patients undergoing lower extremity arterial reconstruction between 1988 and 1994 were retrospectively reviewed, allowing identification of 54 patients who had undergone simultaneous aortoiliac and infrainguinal bypasses. This group consisted of 38 men and 26 women (mean age: 64.7 years), with significant cardiac disease in 24, smoking history in 53, and diabetes mellitus in 15. Indications for surgery were limb-threatening ischemia in 48 (89%) and severe short-distance claudication in 6 (11%). Inflow disease was corrected by direct aortoiliac reconstruction in 28, whereas other extra-anatomic bypasses were constructed in 26. Outflow revascularization required infrainguinal bypass to the infragenicular arteries in 46 (below-knee popliteal: 21; tibial: 25), a concomitant profundaplasty in 26, and a composite bypass conduit in 14. RESULTS: Limb salvage was 97% at 30 days whereas morbidity/mortality were 61% and 19%, respectively. However, the majority of complications and deaths occurred in patients undergoing aortic inflow plus complex outflow procedures (profundaplasty and/or composite bypass conduits), in which the morbidity/mortality rates were 84.2% and 47.4%, respectively, compared with rates of 45.7% and 2.9% (p < 0.01) after all other inflow/outflow procedures. The increased difficulty of these complex procedures is reflected in the significantly greater blood loss and operative times (1853 mL and 10.0 hours) compared with similar values (1125 mL and 7.7 hours)(p < 0.01) for all other inflow/outflow procedures. CONCLUSION: Simultaneous inflow/outflow bypasses are effective and safe in patients with severe, multilevel arterial occlusive disease, except when a complex outflow procedure is needed in conjunction with direct aortoiliac reconstruction. In the latter setting, a staged procedure is recommended because it may be associated with less morbidity and mortality.

Impact of angioscopy on infrainguinal graft patency.

To investigate the impact of angioscopy on infrainguinal graft patency, 50 consecutive cases with angioscopy as an adjuvant to infrainguinal arterial bypass performed during a 12-month period were reviewed (group I). For comparison, 42 similar cases of infrainguinal arterial reconstruction performed during the 12 months prior to introduction of routine intraoperative angioscopy were also reviewed (group II). Patients were followed up for 12 months and graft patency was determined at 1, 3, 6, and 12 months. An abnormality was identified in 13 (26%) group I patients (10, angioscopy alone; 1, arteriography alone; 2, both). Defects were anastomotic abnormalities (n = 7), vein sclerosis (n = 3), retained valve cusp (n = 2), and proximal artery stenosis (n = 1). A similar percentage, but different types of defects, were seen in group II; 11 patients (26%) had an abnormality (anastomotic abnormality [n = 3], vein sclerosis [n = 4], retained valve cusp [n = 1], and arterial outflow stenoses [n = 3]). All significant defects were surgically explored and corrected. Graft patency rates in group I and II at 1, 3, 6, and 12 months were 100% and 85% (P < 0.005), 94% and 80% (P < 0.05), 87% and 74% (P = non-significant [NS]), and 86.1% and 73.7% (P = NS), respectively. Intraoperative angioscopy detects anastomotic and vein graft defects not always seen on arteriography; the repair of these defects significantly improves early infrainguinal bypass graft patency rates.

Does routine stress-thallium cardiac scanning reduce postoperative cardiac complications?

OBJECTIVE: Prophylactic cardiac revascularization in patients with ischemic myocardium could reduce postoperative cardiac complications after aortic reconstruction. However, the effectiveness of this approach has not been documented. SUMMARY BACKGROUND DATA: Stress-thallium scanning can identify patients with ischemic myocardium. Morbidity and mortality after aortic reconstruction appears to be largely caused by co-existent coronary artery disease, and patients who have had recent cardiac revascularization have few postoperative cardiac complications. METHODS: Preoperative stress-thallium scanning was evaluated prospectively in 146 patients undergoing aortic reconstruction. Patients with positive studies underwent coronary arteriography and cardiac revascularization, when appropriate. Postoperative cardiac complications and long-term survival in these patients were compared with results from 172 similar patients undergoing aortic reconstruction without stress-thallium scanning. Results also were analyzed to determine predictors of postoperative cardiac events. RESULTS: Forty-one per cent of patients undergoing stress-thallium testing underwent coronary arteriography, and 11.6% had cardiac revascularization. In contrast, 14.7% of patients treated without stress-thallium testing had coronary arteriography, and 4.1% had revascularization (p < 0.01). Despite this, cardiac mortality, serious cardiac complications, and long-term cardiac mortality were similar in both groups. Only advanced age and intraoperative complications (but not a positive stress-thallium test) predicted postoperative cardiac events. CONCLUSIONS: Preoperative stress-thallium testing confirmed a high incidence of significant coronary artery disease in patients undergoing aortic reconstruction, but prophylactic cardiac intervention does not reduce operative or long-term mortality. Thus, the risk and expense of routine stress-thallium testing and subsequent cardiac revascularization cannot be justified.

Glutamine preserves gut glutathione levels during intestinal ischemia/reperfusion.

Intestinal ischemia/reperfusion (I/R) causes formation of reactive oxygen intermediates (ROI) which lead to mucosal cell injury. Glutathione (GSH), an ROI scavenger, protects tissues from ROI-mediated cell injury. Since GSH biosynthesis is partially dependent on glutamine (Gln) levels, we tested the hypothesis that intravenous Gln infusion will assist in maintaining mucosal cell GSH levels and decrease membrane lipid peroxidation during intestinal I/R. The external jugular vein of male Sprague-Dawley rats was cannulated and infused with normal saline (NS) at 2 cc/hr. After 3 days, matched pairs of rats received either NS alone or NS+ 3% Gln for an additional 24 hr. Next, mucosal GSH levels were measured after a sham I/R in 6 rats and after either 30 or 60 min of ischemia/60 min of reperfusion in a group of 8 and 12 rats, respectively. Finally, conjugated diene (CD), a byproduct of membrane lipid peroxidation, was measured following 60 min of ischemia/60 min of reperfusion in a separate group of 12 rats. Control rats had the highest GSH levels and there was no difference between NS vs NS + 3% Gln rats (2.50 +/- 0.48 vs 2.50 +/- 0.43, P = NS). With 30 and 60 min of ischemia/60 min of reperfusion, GSH levels were significantly lower in NS-infused rats compared to those in NS + 3% Gln-infused rats (30 min: 1.54 +/- 0.14 vs 1.80 +/- 0.16, P < 0.05; 60 min: 1.27 +/- 0.15 vs 1.52 +/- 0.20, P < 0.04). In addition, CD levels were lower in NS + 3% Gln-infused rats compared to those in NS alone-infused rats (5.58 +/- 0.87 vs 7.94 +/- 0.55, P < 0.04). In conclusion, Gln supplementation partially maintains gut GSH levels during bowel I/R, which in turn lessens I/R-induced cell membrane lipid peroxidation.

Pneumonitis of unclear etiology in the surgical intensive care unit leading to the demise of an elderly, immunocompromised woman.

Multiple system organ failure, likely an expression of a dysregulated immune system, is a common cause of death in the surgical intensive care unit (ICU). While investigational modalities of therapy are on the horizon, current treatment primarily consists of supportive care. The case of an elderly woman who was admitted to the surgical ICU after a celiac axis-superior mesenteric artery bypass, is presented. Her course was complicated by Influenza-B pneumonitis, multiple system organ failure, and, ultimately, death. The course, pathophysiology, and therapeutic modalities involved in this syndrome are discussed. Additionally, the natural history of influenza infection is reviewed.

Multiple organ dysfunction after mesenteric artery revascularization.

PURPOSE: Mesenteric arterial reconstruction in patients with acute or chronic intestinal ischemia is associated with significant morbidity and mortality rates. Reperfusion of ischemic bowel in animals leads to release of multiple factors that cause organ dysfunction. Similar phenomena potentially occur in human beings after repair of symptomatic chronic mesenteric arterial disease. The purpose of this study was to define the incidence of multiple organ dysfunction seen after reperfusion of the chronically ischemic bowel. METHOD: Between January 1990 and January 1993, 18 patients underwent surgical revascularization for symptomatic chronic mesenteric arterial occlusive disease. Four patients required emergency surgery, whereas 14 procedures were done electively. Hepatic, renal, pulmonary, and coagulation function were evaluated in all patients. Bypass graft patency was confirmed by duplex scanning or repeat operation in all patients with deteriorating conditions. RESULTS: Immediately after surgery hepatic function deteriorated. The serum transaminases increased 90- to 100-fold whereas lactate dehydrogenase increased 25-fold. Simultaneously, platelet counts dropped below 40,000 mm3 in 11 patients. On postoperative day 2, 16 patients had acute pulmonary insufficiency demonstrated by an increase in the pulmonary shunt fraction to a mean of 32% +/- 3% (range 21% to 60%). Ten patients required reintubation, and lung failure lasted an average of 8.4 days (range 1 to 35 days). Hepatic and coagulation function changes were usually transient, returning to baseline within 7 to 10 days. Four of the patients who had multiple organ dysfunction died (two after elective surgery), and one patient died of myocardial infarction after emergency surgery. CONCLUSION: Revascularization of chronically ischemic bowel leads to significant multiple organ dysfunction, potentially as a result of intestinal ischemia and reperfusion injury.