Patrick J. O'Hara

Surgical management of infected abdominal aortic grafts: Review of a 25-year experience*

Eighty-four patients with infected abdominal aortic grafts managed from 1961 through February 1985 were reviewed. Thirty-three patients had associated aortoenteric fistula formation. Twenty-eight infections (33%) and 13 aortoenteric fistulas (39%) originated at The Cleveland Clinic, yielding an incidence of aortic graft infection of 0.77% (28 of 3652 grafts) and aortoenteric fistula formation of 0.36% (13 of 3652 grafts) at this center. Staphylococcus organisms alone or in combination with other organisms were isolated from 34% of the series. Management consisted of graft removal and extra-anatomic bypass in 54 patients (64%), graft removal alone in 14 (17%) patients, partial graft removal and extra-anatomic bypass in seven (8%) patients, and miscellaneous operations in nine (11%) patients. Twenty-three patients (27%) required major amputations, nine of which were bilateral. Life-table analysis yielded 30-day and 1-year survival rates of 72% and 42%, respectively. Thirty-day survival of the aortoenteric fistula subset (49%) was less than that (86%) of the nonaortoenteric fistula subset (p = 0.003). One-year survival of patients treated since 1980 (54%) was superior to that of patients treated before 1980 (31%, p = 0.035). No difference in operative or 1-year survival was demonstrated between the group treated with extra-anatomic bypass and subsequent graft removal and another in which both procedures were performed simultaneously, although the staged group experienced substantially fewer (p = 0.04) amputations (7%) than the combined group (41%).

Thoracoabdominal aneurysm repair: a representative experience.

Between May 1966 and June 1991, 129 patients underwent surgical repair of thoracoabdominal aneurysms, with an overall 30-day mortality rate of 35%. In 75 operations (58%) performed electively, 11 deaths (15%) occurred, and in 54 cases (42%) of either symptomatic or ruptured aneurysms 34 deaths (63%; p less than 0.001) occurred. No one survived among six patients with preoperative hypotension (less than 90 mm Hg) or cardiac arrest. In 16 patients (12%) the etiology of aneurysms was a result of chronic aortic dissection, and the mortality rate in this subgroup was 44%. In the remaining 113 patients (88%) where the etiology was atherosclerosis, 38 deaths occurred (34%; p = 0.433). Spinal cord ischemia occurred in 25 cases (21%) among 116 patients who survived operation. Partial ischemia occurred in six cases (25%), and complete paraplegia occurred in the remainder. Complete and partial paraplegia occurred in 16 of 42 cases (38%) when all of the thoracic aorta was replaced (Crawford groups I, II) and in 9 of 74 cases (12%) when only the abdominal or lower thoracic aorta was replaced (Crawford groups III, IV; p = 0.016). Other complications included myocardial infarction (14 cases, 11%), respiratory failure (46 cases, 36%), and renal failure (33 cases, 27%). The major prospect for improved early survival of patients with thoracoabdominal aneurysms seems to be early detection and elective repair before the occurrence of symptoms.

Early outcome assessment for 2228 consecutive carotid endarterectomy procedures: The Cleveland Clinic experience from 1989 to 1995

PURPOSE Several randomized trials now have established guidelines regarding patient selection for carotid endarterectomy (CEA) that have been widely accepted but have little relevance unless they are considered in the context of perioperative risk. The purpose of this study was to demonstrate the feasibility of early outcome assessment using a computerized database. METHODS Since 1989 demographic information and in-hospital results for all surgical procedures performed by the members of our department have been entered into a prospective registry. For the purpose of this report, we have analyzed the stroke and mortality rates for 2228 consecutive CEAs (2046 patients), including 1924 that were performed as isolated operations and 304 that were combined with simultaneous coronary artery bypass grafting (CABG). This series incidentally contains a total of 153 reoperations for recurrent carotid stenosis. RESULTS The respective stroke and mortality rates were 0.5% and 1.8% for all isolated CEAs, 4.3% and 5.3% for all CEA-CABG procedures, and 4.6% and 2.0% for carotid reoperations. According to a multivariable statistical model, the composite stroke and mortality rate for isolated CEA was significantly influenced by female gender (p = 0.050), by the urgency of intervention (p = 0.026), and by carotid reoperations (p = 0.024). Gender (p = 0.030) and urgency (p = 0.040) also were associated with differences in the stroke rate alone; furthermore, the incidence of perioperative stroke was higher in conjunction with synthetic patching (odds ratio, 2.6; 95% confidence interval, 1.2 to 5.3) and was marginally higher with primary arteriotomy closure (odds ratio, 2.7; 95% confidence interval, 0.8 to 9.5) compared with vein patch angioplasty (1.3%). The method used to repair the arteriotomy was the only independent factor that qualified for the multivariable composite stroke and mortality models that were applied to the combined CEA-CABG procedures, but too few patients in this cohort had synthetic patches or primary closure to validate the perceived superiority of vein patching. CONCLUSIONS Prospective outcome assessment is essential to reconcile the indications for CEA with its actual results, and it may lead incidentally to important observations concerning patient care.

A prospective study of vein patch angioplasty during carotid endarterectomy. Three-year results for 801 patients and 917 operations.

From 1983 through 1985, 801 consecutive patients (mean age: 66 years) underwent 917 primary carotid endarterectomies at the Cleveland Clinic. Conventional arteriotomy closure was performed during 483 operations, while patch angioplasty using a distal segment of saphenous vein was employed in 434. Preoperative risk factors, surgical management, and antiplatelet therapy were equivalent in the vein patch (VP) and nonpatch (NP) groups. Early results were evaluated by intravenous angiography (DSA) in 715 patients (89%), and 332 reconstructions (36%) have been reassessed by objective imaging during a mean follow-up interval of 21 months. Ischemic strokes occurred after 18 (1.9%) of the 917 procedures (0.7% VP, 3.1% NP; p = 0.0084), and symptomatic (N = 9) or unsuspected (N = 8) thrombosis of the internal carotid artery was confirmed by neck exploration or routine DSA after 1.9% of all operations (0.5% VP, 3.1% NP; p = 0.0027). Only ten patients (1.2%) have required reoperations for severe recurrent lesions, but the cumulative 3-year incidence of new defects (greater than or equal to 30% stenosis) documented by objective studies in the VP and NP groups was 9% and 31%, respectively (p = 0.0066). These results strongly suggest that VP angioplasty enhances the safety and durability of carotid endarterectomy.

The management of severe aortoiliac occlusive disease: Endovascular therapy rivals open reconstruction

OBJECTIVE Aortobifemoral bypass (ABF) grafting has been the traditional treatment for extensive aortoiliac occlusive disease (AIOD). This retrospective study compared the outcomes and durability of recanalization, percutaneous transluminal angioplasty, and stenting (R/PTAS) vs ABF for severe AIOD. METHODS Between 1998 and 2004, 86 patients (161 limbs) underwent ABF (n = 75) or iliofemoral bypass (n = 11), and 83 patients (127 limbs) underwent R/PTAS. All patients had severe symptomatic AIOD (claudication, 53%; rest pain, 28%; tissue loss, 12%; acute limb ischemia, 7%). The analyses excluded patients treated for aneurysms, extra-anatomic procedures, and endovascular treatment of iliac stenoses. Original angiographic imaging, medical records, and noninvasive testing were reviewed. Kaplan-Meier estimates for patency and survival were calculated and univariate analyses performed. Mortality was verified by the Social Security database. RESULTS The ABF patients were younger than the R/PTAS patients (60 vs 65 years; P = .003) and had higher rates of hyperlipidemia (P = .009) and smoking (P < .001). All other clinical variables, including cardiac status, diabetes, symptoms at presentation, TransAtlantic Inter-Society Consensus stratification, and presence of poor outflow were similar between the two groups. Patients underwent ABF with general anesthesia (96%), often with concomitant treatment of femoral or infrainguinal disease (61% endarterectomy, profundaplasty, or distal bypass). Technical success was universal, with marked improvement in ankle-brachial indices (0.48 to 0.84, P < .001). Patients underwent R/PTAS with local anesthesia/sedation (78%), with a 96% technical success rate and similar hemodynamic improvement (0.36 to 0.82, P < .001). At the time of R/PTAS, 21% of patients underwent femoral endarterectomy/profundaplasty or bypass (n = 5) for concomitant infrainguinal disease. Limb-based primary patency at 3 years was significantly higher for ABF than for R/PTAS (93% vs 74%, P = .002). Secondary patency rates (97% vs 95%), limb salvage (98% vs. 98%), and long-term survival (80% vs 80%) were similar. Diabetes mellitus and the requirement of distal bypass were associated with decreased patency (P < .001). Critical limb ischemia at presentation (tissue loss, hazard ratio [HR], 8.1; P < .001), poor outflow (HR, 2; P = .023), and renal failure (HR, 2.5; P = .02) were associated with decreased survival. CONCLUSION R/PTAS is a suitable, less invasive alternative to ABF for the treatment of severe AIOD. Repair of the concomitant femoral occlusive disease is often needed regardless of open or endovascular treatment. Infrainguinal disease negatively affects the durability of the procedure and patient survival.

Elective surgical treatment of symptomatic chronic mesenteric occlusive disease: Early results and late outcomes

PURPOSE The purpose of this study was to determine the safety and efficacy of the elective surgical treatment of symptomatic chronic mesenteric occlusive disease (SCMOD) and to identify the factors that influence the results of this procedure. METHODS From 1977 to 1997, 85 patients (mean age, 62 years) underwent elective surgical treatment of SCMOD. The presenting symptoms were abdominal pain in 78 patients (92%) and weight loss in 74 patients (87%). The surgical procedures included retrograde bypass grafting in 34 patients (40%), antegrade bypass grafting in 24 patients (28%), transaortic endarterectomy in 19 patients (22%), local arterial endarterectomy with patch angioplasty in six patients (7%), thrombectomy alone in one patient (1%), and superior mesenteric artery reimplantation in one patient (1%). Thirty-five patients (41%) underwent concomitant aortic replacement. All the involved mesenteric vessels were revascularized in 21 patients (25%), whereas revascularization was incomplete for the remaining 64 patients (75%). Late information was available for all 85 patients at a mean interval of 4.8 years. RESULTS There were seven early (<35 days) postoperative deaths (8%). The cumulative 5-year survival rate was 64% (95% confidence interval [CI], 53% to 75%), and the 3-year symptom-free survival rate was 81% (95% CI, 72% to 90%). Serious complications occurred in 28 patients (33%). The results of univariate analysis identified advancing age at operation (P <.001), cardiac disease (P =.03), hypertension (P =.03), and additional occlusive disease (P =.05) as variables associated with mortality. Concomitant aortic replacement (P =.037), renal disease (P =.011), advancing age ( P =.035), and complete revascularization ( P =.032) were associated with postoperative morbidity including mortality. Late recurrent mesenteric occlusive disease was seen in 21 patients (16 symptomatic and five asymptomatic). Nine patients (43%) died, and 8 patients (38%) required subsequent surgical or endovascular procedures to treat their recurrent lesions. The 3-year survival rate from recurrent mesenteric occlusive disease was 76% (95% CI, 66% to 86%). CONCLUSION We conclude that the elective surgical treatment of SCMOD may be performed with reasonable early and late mortality rates and that most of the patients remain free from recurrent symptoms of mesenteric ischemia. Advancing age, cardiac disease, hypertension, and additional occlusive disease significantly influenced the overall mortality rates, and concomitant aortic replacement, renal disease, and complete revascularization were significantly associated with postoperative morbidity rates. Surveillance and appropriate correction of recurrent disease appear to be necessary for optimal long-term results.

Recurrent carotid stenosis. A five-year series of 65 reoperations.

From 1979 to 1983, 1726 carotid endarterectomies were performed at the Cleveland Clinic. During this period, 39 men (mean age, 60 years) and 22 women (mean age, 63 years) required 65 reoperations (3.8%) for correction of recurrent carotid stenosis occurring 3 to 194 months (mean, 42 months) after previous endarterectomy at this center (N = 43) or elsewhere (N = 22). Remedial procedures were necessary because of restenosis demonstrated by routine noninvasive testing in 32 asymptomatic lesions and because of neurologic symptoms in 33 others. The mean recurrence interval was 57 postoperative months for atherosclerosis (N = 37) in comparison to 21 months (p = 0.0007) for myointimal hyperplasia (N = 28), and was 48 months for men in comparison to 31 months for women (p = NS). Hypercholesterolemia appeared to be associated with late atherosclerotic recurrence (p = 0.05), but was not a feature of myointimal hyperplasia. Patch angioplasty (N = 59) or graft replacement (N = 3) was employed during 62 of the 65 reoperations, with a total of two operative deaths (3.1%), one nonfatal stroke (1.5%), and six transient cranial nerve injuries (9.2%). Three unrelated late deaths have occurred within a mean follow-up period of 23 months, but only three patients have experienced subsequent neurologic symptoms.

The reduction of the allogenic transfusion requirement in aortic surgery with a hemoglobin-based solution******

OBJECTIVE Because of allogenic red blood cell (RBC) availability and infection problems, novel alternatives, including hemoglobin-based oxygen-carrying solutions (HBOC), are being explored to minimize the perioperative requirement of RBC transfusions. This study evaluated HBOC-201, a room-temperature stable, polymerized, bovine-HBOC, as a substitute for allogenic RBC transfusion in patients undergoing elective infrarenal aortic operations. METHODS In a single blind, multicenter trial, 72 patients were prospectively randomized two-to-one to HBOC (n = 48) or allogenic RBC (n = 24) at the time of the first transfusion decision, either during or after elective infrarenal aortic reconstruction. Patients randomized to the HBOC group received 60 g of HBOC for the initial transfusion and had the option to receive three more doses (30 g each) within 96 hours. In this group, any further blood requirement was met with allogenic RBCs. Patients randomized to the allogenic RBC group received only standard RBC transfusions. The efficacy analysis was a means of assessing the ability of HBOC to eliminate the requirement for any allogenic RBC transfusions from the time of randomization through 28 days. Safety was evaluated by means of standard clinical trial methods. RESULTS The two treatment groups were comparable for all baseline characteristics. Although all patients in the allogenic RBC group required at least one allogenic RBC transfusion, 13 of 48 patients (27%; 95% CI, 15% to 42%) in the HBOC group did not require any allogenic RBC transfusions. The only significant changes documented were a 15% increase in mean arterial pressure and a three-fold peak increase in serum urea nitrogen concentration after HBOC. The complications were similar in both groups, with no allergic reactions. There were two perioperative deaths (8%) in the allogenic RBC group and three perioperative deaths (6%) in the HBOC group (P = 1.0). CONCLUSION HBOC significantly eliminated the need for any allogenic RBC transfusion in 27% of patients undergoing infrarenal aortic reconstruction, but did not reduce the median allogenic RBC requirement. HBOC transfusion was well tolerated and did not influence morbidity or mortality rates.

Thrombolysis of peripheral arterial bypass grafts: Surgical thrombectomy compared with thrombolysis ☆ ☆☆: A preliminary report

Twenty-two patients were selected from a group of 33 patients who underwent recombinant human tissue-type plasminogen activator (rt-PA) thrombolysis for thrombosed infrainguinal bypass grafts of the lower extremity and were compared with 38 matched patients who had undergone surgical thrombectomy during the same period. The proportion of persons with diabetes mellitus, smokers, and types of bypass grafts was similar in both groups. More patients in the rt-PA-treated group had hypertension (p = 0.01). To evaluate the different lengths of follow-up, Kaplan-Meier survival analysis was used with a log-rank test to compare the proportion of persons with patent grafts in the two treatment groups. At 30 days, 86% of the rt-PA-treated grafts were still patent compared with 42% of the surgically treated grafts (p = 0.001). When risk factors on the Kaplan-Meier curves were compared, there was no statistical difference with regard to graft patency among the groups. According to simultaneous Cox regression analysis, no risk factor was significantly associated with graft patency. When amputation was evaluated between treatment groups simultaneously with other risk factors in a logistic regression analysis, smoking and age of the graft were marginally significant (p = 0.07), whereas all other factors were clearly not significant. In 91% of the rt-PA-treated patients, a secondary surgical procedure was required to maintain patency of the graft segment. Eighty-nine percent of the surgically treated patients required similar graft revisions. Two patients in the surgical group and one patient in the rt-PA-treated group had major complications.(ABSTRACT TRUNCATED AT 250 WORDS)

Surgical management of popliteal aneurysms. Trends in presentation, treatment, and results from 1952 to 1984.

Graft replacement was performed for 123 (77%) of 160 popliteal aneurysms evaluated at The Cleveland Clinic from 1952 to 1984, employing autogenous saphenous vein in 58 (36%), polytetrafluoroethylene in 19 (12%), and of historic interest, Dacron (12%) or arterial homograft (16%) in 46. The 10-year cumulative patency (CP) rate was 56% and the limb salvage (LS) rate was 83% following graft replacement, but late results were superior in patients who received vein bypass (CP, 94%; LS, 98%), in those who underwent revascularization before ischemic complications had occurred (CP, 92%; LS, 96%), and in those who recovered both pedal pulses (CP, 64%; LS, 96%). Long-term asymptomatic limbs were restored in 96%, 92%, and 89% of these subsets, respectively, compared with 65% of those receiving other graft materials (p = 0.00003), 59% of those with preoperative ischemic symptoms (p = 0.00001) and 68% of those regaining only an isolated popliteal pulse (p = 0.0326). These data indicate that popliteal aneurysms should be corrected by vein bypass to a patent tibioperoneal segment before spontaneous thrombosis or embolization eliminates the critical outflow bed.