- Email: [email protected]
Thoracoabdominal aneurysm resection after previous infrarenal abdominal aortic aneurysmectomy
Thoracoabdominal Aneurysm Resection After Previous Infrarenal Abdominal Aortic Aneurysmectomy Andrew D. Fox,
MD,
Henry D. Berkowitz,
Thoracoahdominal aneurysms in 51 patients were repaired over 5 years ending in February 1991. Fourteen (27%, 14 of 5 1) patients had a prior infrarenal aneurysm resection (PRA) ; their data are analyzed separately. The average age of patients who had undergone PRA was 67 years (range: 56 to 86 years). The mean aneurysm diameter was 8.6 cm (range: 5 to 12 cm), and the mean time interval between initial aneurysm surgery and suhsequent resection of the thoracoahdominal aneurysm was 8.5 years (range: 2 to 17 years). Three patients in the PRA group were operated on emergently, two because of clinical evidence of rupture; the other patients underwent elective repair. Early mortality (30 days) in the PRA group was significantly related to age (72 years or older versus younger than 72 years: 75% versus lo%, p = 0.04), proximal extent of aneurysm (above diaphragm versus below diaphragm: 50% versus 0%, p = 0.05)‘) ruptured aneurysm (ruptured versus nonruptured: 100% versus 16%, p = 0.06)) and a cardiac history of myocardial infarction (57% versus O%, p = 0.03), congestive heart failure (66% versus O%, p = 0.01) , or arrhythmia (80% versus O%, p = 0.005). s imilar results were seen with the entire group of patients with thoracoahdominal aneurysms except that the proximal extent of the aneurysm was not related to mortality. These results demonstrate that thoracoahdominal aneurysm resection after prior infrarenal aneurysmectomy is not associated with increased mortality or morbidity.
From the Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania. Requests for reprints should be addressed to Henry D. Berkowitz, MD, Department of Surgery, Hospital of the University of Pennsylvania, 34th and Spruce Streets, Philadelphia, Pennsylvania 19104. Presented at the 19th Annual Meeting of the Society for Clinical Vascular Surgery, Kauai, Hawaii, April 3-6, 1991.
142
THE AMERICAN
JOURNAL
OF SURGERY
MD, Philadelphia, Pennsylvania
R
ecurrent aneurysm proximal to a previously implanted infrarenal aortic graft is an infrequent late complication of prior abdominal aortic surgery [I]. Although the natural history of these proximal or thoracoabdominal aneurysms has not been clearly defined, the poor prognosis of nonsurgically treated primary aneurysma1 disease of the descending thoracic and high intraabdominal aorta has been well established. Estimated mortality for patients with nonsurgically treated thoracoabdominal aneurysms has been shown to be 76% at 2 years [2]. This review presents our 5-year experience with 14 patients with thoracoabdominal aneurysm resection who had a previous infrarenal aneurysmectomy (PAR). This group represents 27% (14 of 51) of all the patients with thoracoabdominal aneurysms operated on during the period February 1986 to February 1991. Although this condition can tax the skill, judgment, and ingenuity of the most experienced vascular team, an aggressive approach to this problem is warranted. This review presents our experience with this interesting group of patients and emphasizes important lessons learned from their management. PATIENTS AND METHODS From February 1, 1986, to February 1, 1991,51 patients with a mean age of 70.5 years underwent thoracoabdominal aneurysm resection. Fourteen patients (13 men, 1 woman) had undergone PAR. All patients in the PAR group smoked, and the mean age was 67 years (range: 56 to 86 years). Patient clinical profiles are summarized in Table I. Hypertension, congestive heart failure, myocardial infarction, diabetes mellitus, cerebrovascular disease, and renal insufficiency were prevalent. One patient was undergoing long-term hemodialysis when his aneurysm became symptomatic. The average aneurysm diameter was 8.6 cm (range: 5 to 12 cm), and the mean time interval between initial surgery and resection for thoracoabdominal aneurysm was 8.5 years (range: 2 to 17 years). Two of the patients presented with clinical evidence of rupture and a third was explored emergently; the other patients underwent elective repair. In eight patients, the proximal extent of the aneurysm was the descending thoracic aorta (type III) and six aneurysms originated in the supraceliac aorta (type IV). There were no type I or type II aneurysms. All patients with elective operations underwent detailed cardiopulmonary assessment prior to surgery, including pulmonary function testing, dipyridamole-thalliurn scans, and cardiac catheterization when indicated. Patients were monitored intraoperatively with intra-arte-
VOLUME 162 AUGUST 1991
THORACOABM)MINALANEURYSMRESECTIONAFTERlNFRARENALANEURYSMECTOMY
TABLE 11
TABLE1
Univariate Analysis of Cllnical Variables Related to Early Death for Previous Aneurysm Resections (n = 14)
Patient Characteristics in Previous Aneurysm Resections n FM
Variable Age (range) Hypertension Smoke Arteriosclerotic heart disease Myocardial infarction Diabetes mellitus Cerebrovascular accident Creatinine > 2.0 mg/lOO mL
67 14 13 10 7 4 4 2
(56-66) (100) (93) (78) (60) (21) (21) (15)
rial pressures, pulmonary artery catheters, and thermodilution cardiac output determinations. Spinal fluid drainage was routinely performed, and cerebrospinal fluid pressure was maintained at less than 8 cm of water. Surgery consisted of graft inclusion with direct branch vessel attachment to openings in the graft as described by Crawford [3,4]. Associated visceral artery occlusion was treated by either endarterectomy or less frequently by dacron bypass grafting. Hypertension after proximal aortic clamping was controlled by nitroprusside and/or nitroglycerin. No attempt was made to perfuse the legs with shunts or cardiopulmonary bypass. Aortic clamp times varied depending upon the extent of the disease in the aorta and the difficulty in obtaining a satisfactory suture line. Distal control was obtained by inserting a Foley balloon catheter distally. This allowed control without the need for tediously dissecting out the old graft. The old graft suture line was excised, and the distal anastomosis was made to the proximal old graft. The graft was never removed. Autotransfusion of washed red cells with a cell saver was routine, as was use of a rapid infuser pump. Autotransfused blood was supplemented by homologous red cells, fresh frozen plasma, platelets, and cryoprecipitate. Room temperature was kept at 80°F to minimize heat loss from the patient. RESULTS The overall operative mortality in the PAR group was 28% (4 of 14). Paraplegia occurred in 14% (2 of 14) and dialysis was required in 2 patients (14%) with ruptured aneurysms. A third patient was already undergoing diilysis when he underwent an emergent operation. Early (30day) mortality was significantly related to several variables (Table II). The incidence of early deaths was expressed as a percentage for each of the preoperative and intraoperative variables, and the statistical significance of each was determined by univariate analysis. This analysis identified age, rupture, extent of aneurysm, and the presence of three cardiac variables (myocardial infarction, congestive heart failure, and arrhythmia) as the most significant predictors of death. No increase in early death risk was noted to be significantly related to gender, preoperative creatinine level, chronic obstructive pulmonary disease, diabetes mellitus, blood loss, or ischemic clamp times. Statistical analysis of the entire group of THE AMERICAN
Age (yrs) <72 272 Extent of aneurysm Ill IV Rupture Nonrupture Symptomatic* Asymptomatic History of Myocardial infarction No myocardiil infarction Chronic heart failure No chronic heart failure Arrhythmia No arrhythmia
No. of Cases
30-Day Mortality (%)
10 4
IO 75
0.04
8 6 2 12 8 6
50 0 100 16 38 20
0.05
7 7 6 8 5 9
57 0 66 0 80 0
0.03
P Value
0.06 0.08
0.01 0.005
‘Includes rupture.
TABLE III Unlvariate Analysis of Clinical Variables Related to Early Death for all Thoracoabdominal Aneurysms (n = 51)
Variable Age (Y@ <72 272 Extent of aneurysm Ill IV Rupture Nonrupture Symptomatic* Asymptomatic History of Myocardial infarction No myocardial infarction Chronic heart failure No chronic heart failure Arrhythmia No arrhythmia
No.of Cases
30-Day Mortalii (%)
P Value
33 18
15 44
0.02
24 27 8 43 32 19
38 15 62 18 36 11
NS
28 23 21 30 19 32
48 0 63 0 58
0.008 0.05
0.002 0.001 0.003
1
*Includes
rupture. NS = not significant.
patients with thoracoabdominal aneurysms yielded similar results (Table III). Stepwise logistic regression identified age, rupture, and the presence of congestive heart failure as significant predictors of early death (Table IV). COMMENTS The vast majority of abdominal aortic aneurysm reconstructions produce satisfactory long-term results. However, some patients will require remedial operations years after the original reconstructions. Despite the extensive surgery needed to repair these new thoracoabdo-
JOURNAL
OF SURGERY
VOLUME
162
AUGUST
1991
143
FOX AND BERKOWITZ
TABLE
IV
Multivariate Analysis of Early (30-Day) Death (Stepwise Logistic Regression) for All Thoracoabdomlnal Aneurysms Variable Age Rupture Congestive Constant
heart
Coefficient
p Value
10.6 0.41 22.0 -42.0
co.05 <0.05 co.05
4. Crawford ES, Crawford JL. Diseases of the aorta including an atlas of angiographic pathology and surgical technique. Baltimore: Williams and Wilkins, 1984. 5. Hollier LH, Symmonds JB, Pairolero PC, Cherry KJ, Hallett JW, Gloviczki P. Thoracoabdominal aortic aneurysm repair. Arch Surg 1988; 123: 871-5. 6. Cambria RP, Brewster DC, Moncure AC, et al. Recent experience with thoracoabdominal aneurysm repair. Arch Surg 1989; 124: 620-4.
DISCUSSION Dr. Perry:
minal aneurysms, the mortality of this subgroup is no higher than that observed in primary thoracoabdominal repair. In both, age greater than or equal to 72 years is associated with a higher mortality (Tables II and III), and 35% of our patients were in this older group. The mortality rate in our patients 72 years of age and older is considerably higher than that reported by Crawford [2]. However, the mortality in the group younger than 72 years of 10% to 15% is only moderately higher than the overall 8% operative mortality reported by Crawford. One explanation for the increased mortality may be the large number of ruptured aneurysms in our series (8 of 5 1,16%), which is 4 times the rate reported by Crawford. Patient age and rupture were closely related. Sixty-three percent (5 of 8) of the ruptured thoracoabdominal aortic aneurysms were in patients 72 years or older, and all ruptured aneurysms in the PRA group were in patients 72 years old or older. Hollier and co-workers [5] also observed a significant positive correlation between age and emergency surgery on mortality. Similar to our experience, though, they found several clinical variables related to death, postoperative paraplegia and dialysis were not similarly associated to observed and measured clinical variables. Two of the patients in our study required dialysis postoperatively-both had ruptured aneurysms and both eventually died. No other variables were associated with need for postoperative dialysis in the PRA group. In the total group, a preoperative creatinine level greater than 2.0 mg/ 100 mL was predictive of the need for dialysis. We agree with several authors [2,5,6j that current indications for thoracoabdominal aneurysm repair are largely related to the natural history of this disease as well as to the surgical risk-benefit ratio. Our results suggest that patients with PRA should be treated similarly.
REFERENCES 1. Fulenwider JT, Smith RB, Johnson RW, Johnson RCJ, Salam AA, Perdue GD. Reoperative abdominal arterial surgery-a tenyear experience. Surgery 1983; 93: 20-8. 2. Crawford ES, Crawford JL, Sat? HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients. Vast Surg 1986; 3: 389-404. 3. Crawford ES. Thoracoabdominal and abdominal aortic aneurysms involving renal, superior mesenteric and celiac arteries. Ann Surg 1974; 179: 763-72.
144
THE AMERICAN
JOURNAL OF SURGERY
Is the visceral ischemia that you mentioned kidney or bowel? Dr. Fox: When we speak of visceral, we are referring to the time required to perfuse the superior mesenteric, celiac, and right renal arteries. Dr. Perry: So this includes renal ischemia? Dr. Fox: Yes. Dr Bergen: How
often should screening be done following infrarenal procedures to detect suprarenal ectasia, and what are the parameters when one prescribes reoperation? Dr. Fox: From the literature, it appears that only 2% to 3% of infrarenal aneurysm resections are followed by suprarenal aneurysm. Therefore, screening would not be cost-effective. Careful clinical follow-up with attention to pulsatile masses or symptoms in the abdomen would indicate the performance of a computed tomographic (CT) scan. If ectasia of the suprarenal aorta is noted above the renals at the time of infrarenal aneurysm resection, we would consider extending the procedure above the kidneys. Dr. Hollier: In our experience, patients with hypertension are at a much higher risk and have a 20% incidenceof aneurysm or pseudoaneurysm. Have you found this? Also, what is the incidence of isolated thoracic aneurysm in your series? We have found that doing a thoracoabdominal aneurysm above an infrarenal graft is much easier and the mortality is lower, i.e., 4% to 8%. We found, in these cases, that sewing a graft above to a graft below is accompanied by a greater incidence of embolism. Would you comment on these findings? Dr. Fox: We have no experience with isolated thoracic aneurysms above a previous infrarenal aneurysm repair, since these are usually referred to cardiac surgeons. We also agree that sewing the thoracic graft into the old aortic graft does simplify the procedure if distal control is maintained by a balloon catheter. Dr. Corson: Do you know whether the patients had suprarenal extension of their aneurysms at the original operation? How do you avoid missing a suprarenal aneurysm in your initial work-up? How do you manage a patient who develops suprarenal extension? Dr. Fox: In 6 of the 11 patients, records indicated that the aorta above the aneurysm was normal at the original operation. In answer to the second question, we use every modality available to work-up the patient preoperatively, including arteriography, CT scanning, and magnetic resonance imaging.
VOLUME 162 AUGUST 1991
MD,
Henry D. Berkowitz,
Thoracoahdominal aneurysms in 51 patients were repaired over 5 years ending in February 1991. Fourteen (27%, 14 of 5 1) patients had a prior infrarenal aneurysm resection (PRA) ; their data are analyzed separately. The average age of patients who had undergone PRA was 67 years (range: 56 to 86 years). The mean aneurysm diameter was 8.6 cm (range: 5 to 12 cm), and the mean time interval between initial aneurysm surgery and suhsequent resection of the thoracoahdominal aneurysm was 8.5 years (range: 2 to 17 years). Three patients in the PRA group were operated on emergently, two because of clinical evidence of rupture; the other patients underwent elective repair. Early mortality (30 days) in the PRA group was significantly related to age (72 years or older versus younger than 72 years: 75% versus lo%, p = 0.04), proximal extent of aneurysm (above diaphragm versus below diaphragm: 50% versus 0%, p = 0.05)‘) ruptured aneurysm (ruptured versus nonruptured: 100% versus 16%, p = 0.06)) and a cardiac history of myocardial infarction (57% versus O%, p = 0.03), congestive heart failure (66% versus O%, p = 0.01) , or arrhythmia (80% versus O%, p = 0.005). s imilar results were seen with the entire group of patients with thoracoahdominal aneurysms except that the proximal extent of the aneurysm was not related to mortality. These results demonstrate that thoracoahdominal aneurysm resection after prior infrarenal aneurysmectomy is not associated with increased mortality or morbidity.
From the Department of Surgery, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania. Requests for reprints should be addressed to Henry D. Berkowitz, MD, Department of Surgery, Hospital of the University of Pennsylvania, 34th and Spruce Streets, Philadelphia, Pennsylvania 19104. Presented at the 19th Annual Meeting of the Society for Clinical Vascular Surgery, Kauai, Hawaii, April 3-6, 1991.
142
THE AMERICAN
JOURNAL
OF SURGERY
MD, Philadelphia, Pennsylvania
R
ecurrent aneurysm proximal to a previously implanted infrarenal aortic graft is an infrequent late complication of prior abdominal aortic surgery [I]. Although the natural history of these proximal or thoracoabdominal aneurysms has not been clearly defined, the poor prognosis of nonsurgically treated primary aneurysma1 disease of the descending thoracic and high intraabdominal aorta has been well established. Estimated mortality for patients with nonsurgically treated thoracoabdominal aneurysms has been shown to be 76% at 2 years [2]. This review presents our 5-year experience with 14 patients with thoracoabdominal aneurysm resection who had a previous infrarenal aneurysmectomy (PAR). This group represents 27% (14 of 51) of all the patients with thoracoabdominal aneurysms operated on during the period February 1986 to February 1991. Although this condition can tax the skill, judgment, and ingenuity of the most experienced vascular team, an aggressive approach to this problem is warranted. This review presents our experience with this interesting group of patients and emphasizes important lessons learned from their management. PATIENTS AND METHODS From February 1, 1986, to February 1, 1991,51 patients with a mean age of 70.5 years underwent thoracoabdominal aneurysm resection. Fourteen patients (13 men, 1 woman) had undergone PAR. All patients in the PAR group smoked, and the mean age was 67 years (range: 56 to 86 years). Patient clinical profiles are summarized in Table I. Hypertension, congestive heart failure, myocardial infarction, diabetes mellitus, cerebrovascular disease, and renal insufficiency were prevalent. One patient was undergoing long-term hemodialysis when his aneurysm became symptomatic. The average aneurysm diameter was 8.6 cm (range: 5 to 12 cm), and the mean time interval between initial surgery and resection for thoracoabdominal aneurysm was 8.5 years (range: 2 to 17 years). Two of the patients presented with clinical evidence of rupture and a third was explored emergently; the other patients underwent elective repair. In eight patients, the proximal extent of the aneurysm was the descending thoracic aorta (type III) and six aneurysms originated in the supraceliac aorta (type IV). There were no type I or type II aneurysms. All patients with elective operations underwent detailed cardiopulmonary assessment prior to surgery, including pulmonary function testing, dipyridamole-thalliurn scans, and cardiac catheterization when indicated. Patients were monitored intraoperatively with intra-arte-
VOLUME 162 AUGUST 1991
THORACOABM)MINALANEURYSMRESECTIONAFTERlNFRARENALANEURYSMECTOMY
TABLE 11
TABLE1
Univariate Analysis of Cllnical Variables Related to Early Death for Previous Aneurysm Resections (n = 14)
Patient Characteristics in Previous Aneurysm Resections n FM
Variable Age (range) Hypertension Smoke Arteriosclerotic heart disease Myocardial infarction Diabetes mellitus Cerebrovascular accident Creatinine > 2.0 mg/lOO mL
67 14 13 10 7 4 4 2
(56-66) (100) (93) (78) (60) (21) (21) (15)
rial pressures, pulmonary artery catheters, and thermodilution cardiac output determinations. Spinal fluid drainage was routinely performed, and cerebrospinal fluid pressure was maintained at less than 8 cm of water. Surgery consisted of graft inclusion with direct branch vessel attachment to openings in the graft as described by Crawford [3,4]. Associated visceral artery occlusion was treated by either endarterectomy or less frequently by dacron bypass grafting. Hypertension after proximal aortic clamping was controlled by nitroprusside and/or nitroglycerin. No attempt was made to perfuse the legs with shunts or cardiopulmonary bypass. Aortic clamp times varied depending upon the extent of the disease in the aorta and the difficulty in obtaining a satisfactory suture line. Distal control was obtained by inserting a Foley balloon catheter distally. This allowed control without the need for tediously dissecting out the old graft. The old graft suture line was excised, and the distal anastomosis was made to the proximal old graft. The graft was never removed. Autotransfusion of washed red cells with a cell saver was routine, as was use of a rapid infuser pump. Autotransfused blood was supplemented by homologous red cells, fresh frozen plasma, platelets, and cryoprecipitate. Room temperature was kept at 80°F to minimize heat loss from the patient. RESULTS The overall operative mortality in the PAR group was 28% (4 of 14). Paraplegia occurred in 14% (2 of 14) and dialysis was required in 2 patients (14%) with ruptured aneurysms. A third patient was already undergoing diilysis when he underwent an emergent operation. Early (30day) mortality was significantly related to several variables (Table II). The incidence of early deaths was expressed as a percentage for each of the preoperative and intraoperative variables, and the statistical significance of each was determined by univariate analysis. This analysis identified age, rupture, extent of aneurysm, and the presence of three cardiac variables (myocardial infarction, congestive heart failure, and arrhythmia) as the most significant predictors of death. No increase in early death risk was noted to be significantly related to gender, preoperative creatinine level, chronic obstructive pulmonary disease, diabetes mellitus, blood loss, or ischemic clamp times. Statistical analysis of the entire group of THE AMERICAN
Age (yrs) <72 272 Extent of aneurysm Ill IV Rupture Nonrupture Symptomatic* Asymptomatic History of Myocardial infarction No myocardiil infarction Chronic heart failure No chronic heart failure Arrhythmia No arrhythmia
No. of Cases
30-Day Mortality (%)
10 4
IO 75
0.04
8 6 2 12 8 6
50 0 100 16 38 20
0.05
7 7 6 8 5 9
57 0 66 0 80 0
0.03
P Value
0.06 0.08
0.01 0.005
‘Includes rupture.
TABLE III Unlvariate Analysis of Clinical Variables Related to Early Death for all Thoracoabdominal Aneurysms (n = 51)
Variable Age (Y@ <72 272 Extent of aneurysm Ill IV Rupture Nonrupture Symptomatic* Asymptomatic History of Myocardial infarction No myocardial infarction Chronic heart failure No chronic heart failure Arrhythmia No arrhythmia
No.of Cases
30-Day Mortalii (%)
P Value
33 18
15 44
0.02
24 27 8 43 32 19
38 15 62 18 36 11
NS
28 23 21 30 19 32
48 0 63 0 58
0.008 0.05
0.002 0.001 0.003
1
*Includes
rupture. NS = not significant.
patients with thoracoabdominal aneurysms yielded similar results (Table III). Stepwise logistic regression identified age, rupture, and the presence of congestive heart failure as significant predictors of early death (Table IV). COMMENTS The vast majority of abdominal aortic aneurysm reconstructions produce satisfactory long-term results. However, some patients will require remedial operations years after the original reconstructions. Despite the extensive surgery needed to repair these new thoracoabdo-
JOURNAL
OF SURGERY
VOLUME
162
AUGUST
1991
143
FOX AND BERKOWITZ
TABLE
IV
Multivariate Analysis of Early (30-Day) Death (Stepwise Logistic Regression) for All Thoracoabdomlnal Aneurysms Variable Age Rupture Congestive Constant
heart
Coefficient
p Value
10.6 0.41 22.0 -42.0
co.05 <0.05 co.05
4. Crawford ES, Crawford JL. Diseases of the aorta including an atlas of angiographic pathology and surgical technique. Baltimore: Williams and Wilkins, 1984. 5. Hollier LH, Symmonds JB, Pairolero PC, Cherry KJ, Hallett JW, Gloviczki P. Thoracoabdominal aortic aneurysm repair. Arch Surg 1988; 123: 871-5. 6. Cambria RP, Brewster DC, Moncure AC, et al. Recent experience with thoracoabdominal aneurysm repair. Arch Surg 1989; 124: 620-4.
DISCUSSION Dr. Perry:
minal aneurysms, the mortality of this subgroup is no higher than that observed in primary thoracoabdominal repair. In both, age greater than or equal to 72 years is associated with a higher mortality (Tables II and III), and 35% of our patients were in this older group. The mortality rate in our patients 72 years of age and older is considerably higher than that reported by Crawford [2]. However, the mortality in the group younger than 72 years of 10% to 15% is only moderately higher than the overall 8% operative mortality reported by Crawford. One explanation for the increased mortality may be the large number of ruptured aneurysms in our series (8 of 5 1,16%), which is 4 times the rate reported by Crawford. Patient age and rupture were closely related. Sixty-three percent (5 of 8) of the ruptured thoracoabdominal aortic aneurysms were in patients 72 years or older, and all ruptured aneurysms in the PRA group were in patients 72 years old or older. Hollier and co-workers [5] also observed a significant positive correlation between age and emergency surgery on mortality. Similar to our experience, though, they found several clinical variables related to death, postoperative paraplegia and dialysis were not similarly associated to observed and measured clinical variables. Two of the patients in our study required dialysis postoperatively-both had ruptured aneurysms and both eventually died. No other variables were associated with need for postoperative dialysis in the PRA group. In the total group, a preoperative creatinine level greater than 2.0 mg/ 100 mL was predictive of the need for dialysis. We agree with several authors [2,5,6j that current indications for thoracoabdominal aneurysm repair are largely related to the natural history of this disease as well as to the surgical risk-benefit ratio. Our results suggest that patients with PRA should be treated similarly.
REFERENCES 1. Fulenwider JT, Smith RB, Johnson RW, Johnson RCJ, Salam AA, Perdue GD. Reoperative abdominal arterial surgery-a tenyear experience. Surgery 1983; 93: 20-8. 2. Crawford ES, Crawford JL, Sat? HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients. Vast Surg 1986; 3: 389-404. 3. Crawford ES. Thoracoabdominal and abdominal aortic aneurysms involving renal, superior mesenteric and celiac arteries. Ann Surg 1974; 179: 763-72.
144
THE AMERICAN
JOURNAL OF SURGERY
Is the visceral ischemia that you mentioned kidney or bowel? Dr. Fox: When we speak of visceral, we are referring to the time required to perfuse the superior mesenteric, celiac, and right renal arteries. Dr. Perry: So this includes renal ischemia? Dr. Fox: Yes. Dr Bergen: How
often should screening be done following infrarenal procedures to detect suprarenal ectasia, and what are the parameters when one prescribes reoperation? Dr. Fox: From the literature, it appears that only 2% to 3% of infrarenal aneurysm resections are followed by suprarenal aneurysm. Therefore, screening would not be cost-effective. Careful clinical follow-up with attention to pulsatile masses or symptoms in the abdomen would indicate the performance of a computed tomographic (CT) scan. If ectasia of the suprarenal aorta is noted above the renals at the time of infrarenal aneurysm resection, we would consider extending the procedure above the kidneys. Dr. Hollier: In our experience, patients with hypertension are at a much higher risk and have a 20% incidenceof aneurysm or pseudoaneurysm. Have you found this? Also, what is the incidence of isolated thoracic aneurysm in your series? We have found that doing a thoracoabdominal aneurysm above an infrarenal graft is much easier and the mortality is lower, i.e., 4% to 8%. We found, in these cases, that sewing a graft above to a graft below is accompanied by a greater incidence of embolism. Would you comment on these findings? Dr. Fox: We have no experience with isolated thoracic aneurysms above a previous infrarenal aneurysm repair, since these are usually referred to cardiac surgeons. We also agree that sewing the thoracic graft into the old aortic graft does simplify the procedure if distal control is maintained by a balloon catheter. Dr. Corson: Do you know whether the patients had suprarenal extension of their aneurysms at the original operation? How do you avoid missing a suprarenal aneurysm in your initial work-up? How do you manage a patient who develops suprarenal extension? Dr. Fox: In 6 of the 11 patients, records indicated that the aorta above the aneurysm was normal at the original operation. In answer to the second question, we use every modality available to work-up the patient preoperatively, including arteriography, CT scanning, and magnetic resonance imaging.
VOLUME 162 AUGUST 1991