Peter C. Pairolero

Graft-related complications after abdominal aortic aneurysm repair: Reassurance from a 36-year population-based experience

PURPOSEnGraft-related complications must be factored into the long-term morbidity and mortality rates of abdominal aortic aneurysm (AAA) repair. However, the true incidence may be underestimated because some patients do not return to the original surgical center when a problem arises.nnnMETHODSnTo minimize referral bias and loss to follow-up, we studied all patients who underwent AAA repair between 1957 and 1990 in a geographically defined community where all AAA operations were performed and followed by a single surgical practice. All patients who remained alive were asked to have their aortic grafts imaged.nnnRESULTSnAmong 307 patients who underwent AAA repair, 29 patients (9.4%) had a graft-related complication. At a mean follow-up of 5.8 years (range, < 30 days to 36 years), the most common complication was anastomotic pseudoaneurysm (3.0%), followed by graft thrombosis (2.0%), graft-enteric erosion/fistula (1.6%), graft infection (1.3%), anastomotic hemorrhage (1.3%), colon ischemia (0.7%), and atheroembolism (0.3%). Complications were recognized within 30 days after surgery in eight patients (2.6%) and at late follow-up in 21 patients (6.8%). These complications were observed at a median follow-up of 6.1 years for anastomotic pseudoaneurysm, 4.3 years for graft-enteric erosion, and 0.15 years for graft infection. Kaplan-Meier 5- and 10-year survival free estimates were 98% and 96% for anastomotic pseudoaneurysm, 98% and 95% for combined graft-enteric erosion/infection, and 98% and 97% for graft thrombosis.nnnCONCLUSIONSnThis 36-year population-based study confirms that the vast majority of patients who undergo standard surgical repair of an abdominal aortic aneurysm remain free of any significant graft-related complication during their remaining lifetime.

Surgical treatment of hepatic and pulmonary metastases from colon cancer

BACKGROUNDnSurgical resection of isolated hepatic or pulmonary metastases secondary to colorectal cancer has been shown to yield acceptable long-term survival. However, results are inconclusive for surgical resection of both hepatic and pulmonary metastases.nnnMETHODSnWe reviewed the records of all patients who underwent surgical resection of both hepatic and pulmonary metastases from colorectal cancer between 1980 and 1998.nnnRESULTSnA total of 58 patients underwent resection of both hepatic and pulmonary metastases secondary to colorectal cancer. All patients had local control of their primary cancer before metastasectomy. There were no operative deaths. Morbidity occurred in 12% of patients. Follow-up was complete in all patients, with a median duration of 62 months (range, 6 to 201 months). The 5- and 10-year survivals were 30% and 16%, respectively. A premetastasectomy carcinoembryonic antigen level greater than 5 ng/mL increased the risk of early death (p = 0.029). Neither the number of pulmonary lesions nor the time interval between the primary surgery and the metastasectomy had a significant impact on survival (p = 0.67). At 5 years, 55% of patients were free of disease. Four patients had lymph node involvement at the time of pulmonary resection and all 4 patients died within 22 months of their pulmonary metastasectomy.nnnCONCLUSIONSnResection of both hepatic and pulmonary metastases secondary to colorectal cancer in highly selected patients is safe and results in long-term survival. Thoracic lymph node involvement and elevated carcinoembryonic antigen levels before pulmonary metastasectomy are associated with reduced survival.

Ischemic injury to the spinal cord or lumbosacral plexus after aorto-iliac reconstruction☆

Between January 1, 1980, and June 30, 1989, 9 patients (6 males and 3 females) developed ischemic injury to the spinal cord or lumbosacral plexus following 3,320 operations on the abdominal aorta (0.3%). The incidence of this complication was 0.1% (2 of 1,901) after elective and 1.4% (3 of 210) after emergency abdominal aortic aneurysm repair, and 0.3% (4 of 1,209) after repair for occlusive disease. Three of the latter had prior clinical evidence of distal embolization. Eight grafts were bifurcated (aorto-iliac:four, aorto-femoral: three, aorto-ilio-femoral:one). One patient underwent extra-anatomic revascularization. Only two patients had supraceliac aortic cross-clamping and one patient underwent exclusion of both internal iliac arteries. Four patients had hypotension. Early mortality was 22% (two of nine). Severe perioperative complications, mostly due to associated visceral and somatic ischemia and sepsis, were present in seven of the nine patients. The extent and type of the neurologic injury correlated with long-term outcome. Patients with ischemic injury of the lumbosacral roots or plexus had better recovery. Attention to the pelvic circulation and the collateral blood supply is important. Use of gentle technique to prevent embolization, avoidance of hypotension and prolonged supraceliac cross-clamping, revascularization of at least one internal iliac artery, and the use of heparin may decrease but not eliminate paraplegia. Once this unexpected complication occurs, careful neurologic evaluation should be done to localize the lesion and aid prognosis.

Influence of complete revascularization on chronic mesenteric ischemia

Complete revascularization for chronic intestinal ischemia is controversial. Fifty-eight patients (119 arteries) underwent mesenteric revascularization between 1981 and 1988. There were 46 women and 12 men (mean age: 63 years). Sixty percent of patients had three-vessel disease. Twenty-one patients underwent concomitant aortic reconstruction. Operative mortality was 10%. Four of the six deaths occurred in patients undergoing aortic surgery. Late graft failure occurred in five patients (10%). Five-year survival for patients with three-vessel involvement who underwent three-vessel repair was 73%, compared with 57% for two-vessel repair and 0% for one-vessel repair (p = NS). Similarly, graft patency in patients with three-vessel disease was highest in those patients who had complete revascularization (90%, 54%, and 0%, respectively) (p = NS). We conclude that increased graft patency and survival in patients with three-vessel disease was most frequent with complete revascularization. Diseased inferior mesenteric arteries should be repaired if feasible. Concomitant aortic operations should be avoided if possible.

Inflammatory abdominal aortic aneurysms: A case-control study

PURPOSEnThis study was designed to identify significant differences in the clinical and radiologic characteristics and outcome between patients with inflammatory and noninflammatory abdominal aortic aneurysms (AAAs).nnnMETHODSnWe reviewed 29 consecutive patients who underwent repair of an inflammatory AAA between 1985 and 1994. This group was matched in a case-control fashion by date of surgery and by the performing surgeon to a group of 58 patients who underwent repair of noninflammatory AAAs.nnnRESULTSnThe two groups had comparable characteristics of age, gender, and cardiovascular risk factors. Patients with inflammatory AAAs were significantly more symptomatic than those with noninflammatory AAAs (93% vs 9%, p < 0.001), were more likely to have a family history of aneurysms (17% vs 1.5%, p = 0.007), and tended to be current smokers (45% vs 24%, p = 0.049). The most significant laboratory difference was an elevated sedimentation rate in patients with inflammatory AAAs (mean, 53 mm/hr vs 12 mm/hr, p < 0.00001). Inflammatory AAAs also were significantly larger than noninflammatory AAAs at presentation (6.8 cm vs 5.9 cm, p < 0.05). Although operative mortality was low in both groups, patients with an inflammatory AAA tended to have higher morbidity, including sepsis (p < 0.01) and renal failure (p = 0.04). Five-year survival rates, however, were similar for the two groups (79% for inflammatory and 83% for noninflammatory AAAs). On follow-up computed tomographic scans, the retroperitoneal inflammatory process resolved completely in 53% of the patients, but 47% of patients had persistent inflammation that involved the ureters in 32% and resulted in long-term solitary or bilateral renal atrophy in 47%.nnnCONCLUSIONSnThis case-control study provides preliminary evidence that inflammatory AAAs may have a relatively strong familial connection and that current smoking may play an important role in the inflammatory response. The study also documents that persistent retroperitoneal inflammation may be more prevalent than has been previously reported, and stresses the need for an improved understanding of the pathogenesis and long-term management of inflammatory AAAs.

Aggressive thoracotomy for pulmonary metastatic osteogenic sarcoma in children and young adolescents

During a 6-yr period, 57 children and young adults less than 21 yr of age underwent 111 thoracotomies for pulmonary metastatic osteogenic sarcoma. Follow-up after the initial thoracotomy averaged 36.8 no and ranged from 1 to 78 mo. Twenty-eight patients (49%) underwent more than one thoracotomy--the number of multiple thoracotomies averaged 2.9 and ranged from two to eight. Twenty-six of the 57 patients are currently alive, and 25 of these are free of disease. Median survival was 36 mo. Actuarial survival curves demonstrated a 5-yr survival of approximately 40%. Seventy-one percent of patients who had a tumor-free interval of greater than 1 yr are currently alive, compared with 39% of patients who had a tumor-free interval of less than 1 yr. Patients with disease confined to one lobe at the initial thoracotomy had a better survival (64%) than patients with diffuse or bilateral disease (41%). An aggressive surgical approach toward osteogenic sarcoma with pulmonary metastasis thus appears to be justified.

Pathologic comparison of video-assisted thoracic surgical lung biopsy with traditional open lung biopsy

Video-assisted thoracic surgical lung biopsy is an alternative to traditional open lung biopsy for diagnosis in patients with pleuropulmonary diseases. Between January 7, 1991, and August 3, 1993, 71 consecutive patients had video-assisted thoracic surgical lung biopsy and 42 patients had traditional open lung biopsy. A specific histologic diagnosis that correlated with the clinical findings was sought in each case and the yield was compared between the two groups. Procedure-related artifactual changes were also evaluated; the extent of traumatic hemorrhage and neutrophil margination as a result of tissue manipulation was significantly greater for patients in the video-assisted thoracic surgical lung biopsy group than for those in the open lung biopsy group, but the changes were generally minor and did not affect diagnostic yield. Complications developed in 11 (15%) of 71 patients in the video-assisted thoracic surgical lung biopsy group including 5 patients with prolonged air leakage (more than 10 days); 2 with pneumonia; and 1 each with bleeding, late pneumothorax necessitating readmission, mucus plug necessitating bronchoscopy, and a hypoxic episode necessitating mechanical ventilation. On the other hand, 7 (17%) of 42 patients in the open lung biopsy group had complications including 4 patients with prolonged air leakage (more than 10 days) and 3 with pneumonia. There were 6 (8%) operative deaths in patients who had video-assisted thoracic surgical lung biopsy and 7 (17%) in the open lung biopsy group; all had preoperative respiratory failure. We conclude that video-assisted thoracic surgical lung biopsy is an acceptable alternative to open lung biopsy for diagnosis of pulmonary infiltrates or indeterminate nodules.

Reconstruction of the superior vena cava: Benefits of postoperative surveillance and secondary endovascular interventions ☆ ☆☆ ★ ★★

PURPOSEnSuperior vena cava (SVC) reconstructions are rarely performed; therefore the need for surveillance and the results of secondary interventions are unknown.nnnMETHODSnDuring a 14-year period 19 patients (11 male, 8 female; mean age 41.9 years, range 8 to 69 years) underwent SVC reconstruction for symptomatic nonmalignant disease. Causes included mediastinal fibrosis (n = 12), indwelling foreign bodies (n = 4), idiopathic thrombosis (n = 2), and antithrombin III deficiency (n = 1). Spiral saphenous vein graft (n = 14), polytetrafluoroethylene (n = 4), or human allograft (n = 1) was implanted.nnnRESULTSnNo early death or pulmonary embolism occurred. Four early graft stenoses or thromboses (spiral saphenous vein graft, n = 2, polytetrafluoroethylene, n = 2) required thrombectomy, with success in three. During a mean follow-up of 49.5 months (range, 4.7 to 137 months), 95 imaging studies were performed (average, five per patient; range, one to 10 studies). Venography detected mild or moderate graft stenosis in seven patients; two progressed to severe stenosis. Two additional grafts developed early into severe stenosis. Four of 19 grafts occluded during follow-up (two polytetrafluoroethylene, two spiral saphenous vein graft). Computed tomography failed to identify stenosis in two grafts, magnetic resonance imaging failed to confirm one stenosis and one graft occlusion, and duplex scanning was inconclusive on graft patency in 10 patients. Angioplasty was performed in all four patients with severe stenosis, with simultaneous placement of Wallstents in two. One of the Wallstents occluded at 9 months. Repeat percutaneous transluminal angioplasty was necessary in two patients, with placement of Palmaz stents in one. Only one graft occlusion and one severe graft stenosis occurred beyond 1 year. The primary, primary-assisted, and secondary patency rates were 61%, 78%, and 83% at 1 year and 53%, 70%, and 74% at 5 years, respectively.nnnCONCLUSIONnLong-term secondary patency rates justify SVC grafting for benign disease. Postoperative surveillance with contrast venography is indicated in the first year to detect graft problems. Endovascular techniques may salvage and improve the patency of SVC grafts.

Stump pressure, the contralateral carotid artery, and electroencephalographic changes

Abstract Electroencephalographic (EEG) monitoring and measurement of stump pressure are the most widely employed methods of assessing the risk of cerebral ischemia during carotid endarterectomy. The status of the contralateral carotid artery has also been thought to influence the need for placing a shunt. The relationship of EEG monitoring, stump pressure, and the contralateral carotid artery has not been completely delineated. We retrospectively reviewed these three variables in 113 patients undergoing 124 carotid endarterectomies. The contralateral artery was classified as occluded, stenotic (greater than 50% decrease in diameter), or nonstenotic. There was a 48% incidence of EEG changes with contralateral occlusion, 18% with stenosis, and 21% with nonstenotic arteries (p = 0.014). There was a 73% incidence of EEG changes when the stump pressure was less than 25 mm Hg, 32% when the stump pressure was 25 to 50 mm Hg, and 2% when the stump pressure was greater than 50 mm Hg (p All patients with contralateral occlusion and a stump pressure less than 25 mm Hg had EEG changes. No patient with a stump pressure greater than 50 mm Hg and a patent contralateral artery had EEG changes. Although the incidence of EEG changes in the majority of patients was not accurately predicted by the stump pressure and the status of the contralateral carotid artery, stump pressure less than or equal to 50 mm Hg was sensitive, identifying 97% of patients with EEG changes.

Advanced carotid disease in patients requiring aortic reconstruction

Perioperative stroke is a devastating complication of abdominal aortic operations. Patients requiring aortic reconstruction with advanced carotid occlusive disease pose a particularly challenging management problem regarding timing of operations. All patients (n = 121) undergoing both carotid artery endarterectomy (CEA) and abdominal aortic reconstruction (AAR) within 1 year of each other between 1979 and 1989 were reviewed. The sequence of operation was analyzed to determine its effect on early and late outcome. CEA was the first operation in 99 patients (group I); AAR was performed first in 22 patients (group II). Age, gender, number, types of risk factors, and associated medical problems were similar in both groups. Indications for CEA were: transient ischemic attacks (TIAs), recent ipsilateral stroke, or high-grade asymptomatic carotid artery stenosis exceeding 80%. Indications for aortic operation included: abdominal aortic aneurysm, aortoiliac occlusive disease, and combined aortic and renovascular disease. There were five perioperative strokes, two in group I (2%) and three in group II (14%) (p < 0.04). All strokes occurred after AAR. There were five perioperative deaths (4%), four in group I (4%) and one in group II (5%). Overall survival was significantly greater in group I compared to group II (p < 0.04); 5-year survival was 77% and 51%, respectively. Multivariate analysis demonstrated age, hypertension, and diabetes to adversely affect survival; CEA as the first procedure, however, had a protective effect. Importantly, eight strokes occurred in group I in late follow-up, but only one was ipsilateral to the CEA. We conclude that CEA in selected patients who require AAR is safe, and, when performed prior to abdominal aortic repair, reduces perioperative stroke and may improve long-term survival.