Prusa AM, Wolff KS, Sahal M, Polterauer P, Lammer J, Kretschmer G, Huk I, Teufelsbauer H. Abdominal Aortic Aneurysms and Concomitant Diseases Requiring Surgical InterventionSimultaneous Operation vs Staged Treatment Using Endoluminal Stent Grafting. Arch Surg. 2005;140(7):686-691. doi:10.1001/archsurg.140.7.686
Copyright 2005 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2005
To investigate whether staged or synchronous treatment of infrarenal abdominal aortic aneurysms (AAAs) and concomitant diseases (CDs) requiring surgical repair plays a clinical role. We considered endovascular aneurysm repair (EVAR) in particular.
Review of a prospectively gathered database.
Tertiary care university teaching hospital.
We reviewed a total of 946 patients receiving elective AAA exclusion from 1980 through 2002. We divided the period into 2 observation intervals: 1980-1994, when only open graft replacement was available (n = 331), and 1995-2002, with 615 patients (326 who had open graft replacement and 289 who had EVAR). With regard to the physical status, expressed by the score from the American Society of Anesthesiologists (Park Ridge, Ill), we recorded in-hospital mortality rates and checked possible differences.
Main Outcome Measures
Indications for therapy and mortality rates before and after the availability of EVAR.
During the first interval, 14 simultaneous operations were carried out. During the second period, 19 patients received simultaneous operations while 49 underwent staged treatment using EVAR. The overall mortality rate was 3.7%. Irrespective of the American Society of Anesthesiologists classification, the mortality rate for patients who had EVAR was 0% in comparison with 13.6% for patients in American Society of Anesthesiologists class 3 or 4 after open graft replacement (P<.03).
The coincidence of a patient having both an AAA and a CD is rare but should not be neglected. Staged treatment of AAAs using EVAR followed by surgical therapy for CDs can be an effective causal therapy with an acceptable mortality rate provided that suitable aneurysm anatomy exists.
Concomitant occurrence of significant pathologic abnormalities in the abdominal cavity, in the retroperitoneal space, or in the abdominal wall represents a therapeutic dilemma in the presence of infrarenal abdominal aortic aneurysms (AAAs).1- 6 With increasing life expectancy, the prevalence of AAAs rises7 in parallel with the occurrence of significant nonvascular pathologic abnormalities. Because at one time only open graft replacement of the dilated aortic segment was available,8 surgeons discussed whether to perform simultaneous or staged procedures and, if they chose subsequent operations, which abnormality they should treat first.2,8,9
There is some evidence that open graft replacement and simultaneous treatment of intraperitoneal lesions and especially retroperitoneal lesions can be performed in a safe way. Ochsner et al10 demonstrated that aortic graft replacement and operations for concomitant diseases (CDs) requiring surgical repair, especially malignancies of the colon and rectum, might be performed simultaneously without an increase in either graft infection or in-hospital mortality rate. Conversely, other authors report an increased complication rate after simultaneous treatment,11- 13 reflecting a therapeutic dilemma.
The aim of this study was to investigate whether excluding the aneurysm with stent-grafting techniques can circumvent the obvious dilemma, thereby offering superior results.
From January 1980 through December 2002, 946 consecutive patients were electively treated for infrarenal AAAs. We excluded patients with thoracoabdominal or juxtarenal aortic aneurysms, as well as those requiring cross-clamping above the renal arteries. Three hundred thirty-one patients underwent classic open graft replacement of the aneurysmatic aortic segment from 1980 through 1994. Six hundred fifteen patients were treated from January 1995 through December 2002: 326 patients received open graft replacement, and 289 patients received endovascular aneurysm repair (EVAR) as a minimally invasive procedure for AAA exclusion.
We reviewed patient records to identify patients with AAAs and CDs that both required surgical intervention, but we excluded patients who required subsequent peripheral vascular surgical procedures. Patient characteristics and preoperative risk factors of the respective patients were analyzed. Diseases were regarded as concomitant when they were detected in preoperative AAA evaluation or when an AAA needing intervention was detected during the preoperative evaluation of a nonvascular disease. The interval between 2 staged interventions was up to 1 year.
To estimate operative risk, we investigated several parameters preoperatively. We considered patients to have renal dysfunction if their serum creatinine levels exceeded 1.5 mg/dL (133 μmol/L). Pulmonary dysfunction was assumed when the results of lung function tests were below 65% of the expected values or moderate or severe dyspnea on exertion was found. Hepatic dysfunction was indicated by serum γ-glutamyl transpeptidase levels higher than 30 U/L. We chose the respective cutoff levels according to accepted international limits14 as well as standard thresholds of our Department of Laboratory Medicine. We defined cardiac risk factor as a history of heart disease, including myocardial infarction, stable angina pectoris, balloon dilatation of coronary arteries, bypass grafting, evidence of reduced ventricular performance, and presence of pulmonary hypertension. Other risk factors were diabetes, hypertension, and evidence of severe cerebrovascular disease. In patients with clear indications for revascularization of coronary or carotid arteries, the respective interventional or operative repair was done prior to AAA exclusion. Patients who had untreatable unstable angina pectoris, had New York Heart Association class IV function, needed oxygen support at rest, or had end-stage malignant disease were not treated for their AAAs. We used American Society of Anesthesiologists (ASA) scoring15 to estimate the overall status of health of the individual patient. Because of the small sample size, the data for patients in ASA class 1 or 2 (“low-risk” patients) as well as patients in an ASA class higher than 3 (patients with increased risk) were summarized and analyzed as a respective group.
Regarding survival data, most recent information is periodically obtained from the Austrian Central Bureau of Statistics (Statistik Austria) in Vienna. Austrian law requires that all deaths and causes be reported to this institution. At least once a year, all these data are transferred to the mainframe computer of the Medical University of Vienna. The last upgrade of data before our study was performed on January 1, 2003.
Open graft replacements used a median transperitoneal approach. Exclusion of AAAs was performed by implantation of commercially available bifurcated or tubular grafts. Endoluminal stent grafting was performed via a transfemoral route. Commercially available modular stent grafts were implanted through arteriotomy. With bifurcated stent grafts, the extension graft for the second limb was inserted by transcutaneous puncture or arteriotomy of the contralateral common femoral artery. Interdisciplinary teams of vascular surgeons and interventional radiologists deployed all prostheses. Concomitant diseases were treated by standard surgical procedures.
For basic description, age was expressed as median, interquartile range, and absolute range. We checked significant differences with Wilcoxon tests. All other data were displayed as frequencies and percentages. We performed the Fisher exact test or Mantel-Haenszel test, wherever appropriate, for bivariate analysis of categorical data. Mortality rates were calculated according to the method described by Kaplan-Meier. We analyzed differences in mortality rates with Fisher exact tests. All statistical analyses were carried out with version 6.09E of SAS (SAS Institute, Inc, Cary, NC) on an IBM 4090 mainframe.
During the 23-year observation period, 82 patients were identified as suffering from both AAAs and CDs (8.67% of the total clinical series). We divided the period into an interval before (1980-1994) and one after the availability of EVAR (1995-2002). We found 35 surgical repairs of CDs in 33 patients (2 patients with 2 surgical interventions) receiving open graft replacement, representing simultaneous operations. In 49 patients, 66 staged treatments were performed (EVAR and surgical therapy for CDs). Table 1 demonstrates that the percentage of patients with AAAs and CDs requiring surgical intervention increased 3-fold from 4.23% to 11.06% whereas the number of patients receiving the alternative simultaneous operations decreased by a quarter from 4.23% to 3.09%.
Table 2 lists the numbers and types of simultaneous operations of CDs in the respective intervals. Renal or adrenal gland surgery included nephrectomy because of renal cell carcinoma in 1 patient each in both observation periods, as well as bilateral adrenalectomy due to central Cushing disease. Simultaneous hernia repair took place for 6 inguinal hernias. We excluded other types of hernia (eg, umbilical, incisional, or spigelian). Small- and large-bowel surgery included resections of small-bowel segments because of carcinoid tumors (n = 2) and 2 Meckel diverticula (1 patient having both conditions) and a sigmoid resection for ulcerative diverticulitis. Oversewing a duodenal fistula and an enterotomy due to intestinal obstruction were each performed once. The gastric surgery procedure in the first observation period was a Billroth I resection due to a bleeding ulcer, and in the second interval, partial gastrectomy was performed for a gastric neoplasia. All splenectomies (n = 5) took place in the second period. Also in the second interval, 1 patient underwent extensive adhesiolysis after abdominal irradiation.
The various types of staged treatments of CDs with the advent of EVAR are presented in Table 3. Cardiac surgery (n = 11) included coronary artery bypass grafting, or transmyocardial laser revascularization, aortic valve repair, and mitral valve repair and reconstruction. Fifteen patients underwent surgical repair of inguinal hernias, which used laparoscopic transabdominal preperitoneal inguinal herniorrhaphy for 3 cases and open tension-free procedures for the others. For all 3 patients undergoing cholecystectomy, a laparoscopic approach was performed. Resection of small- and large-bowel segments was needed in 6 cases (colorectal cancer, n = 4; sigma diverticulitis, n = 2). Renal surgery represents partial and total nephrectomy due to renal cell carcinoma while 1 patient suffered from adrenal adenoma. All other resections and extirpations were deemed necessary because of neoplasia (bladder, gastric, liver, lung, prostate, uterus). Procedures summarized as “others” included kidney transplantation, total hip replacement, surgery because of recurrent pancreatitis, extracorporeal shock wave lithotripsy, cataract, benign prostatic hypertrophy, and bladder-associated stricture of urethra.
Table 4 outlines the distribution of various risk factors as found preoperatively and indicates the uneven distribution of these factors. We found significant higher percentages of comorbidities and reduced physical status in the group of patients treated by EVAR.
Overall mortality rate was 3.7%. Table 5 demonstrates that for patients with good or reduced physical status who had EVAR, mortality rates were 0% compared with 13.6% for patients in ASA class 3 or 4 after open graft replacement (P<.03). The table clearly indicates the value of EVAR allowing sequential treatment, eliminating the need for a simultaneous procedure with increased operative burden.
After open graft replacement was introduced as the standard of care in the management of AAAs, what to do with CDs became a difficult question. Up to now, the 2 strategies have involved either simultaneous procedures or staged procedures. Both approaches offer some advantages and disadvantages. Tsuji et al3 concluded that the simultaneous surgical treatment of an AAA and a CD is safe and beneficial for the patient, but Bergan in the invited commentary to the article recommended doing only 1 procedure per patient per day. These conflicting statements indicate that a dilemma does indeed exist.
Following staged AAA management, performing the operation on the CD first, reports described an increased risk of AAA rupture in the postoperative period following unrelated laparotomy.16 A clear reason for this risk remains unknown. The simultaneous treatment of AAA and CD means there is no subsequent surgery, but the combined approach augments the procedure and the operative risk. Heparin administration and difficult hemostasis are specific problems. Following simultaneous procedures, the most feared but relatively rare complication is graft infection, observed in a prevalence of 1.6% for aortoiliac grafts and 1.3% to 3.1% for aortofemoral grafts, whereas mortality was observed to be between 30% and 60%.17 Nonvascular procedures carried out concomitantly are possible causes, particularly when hollow organs with potentially infected contents were opened. To reliably compare the mortality rates of the combined vs staged treatments, both reports should date from the same period or at least the same decade. Because AAAs and CDs very rarely coincide, meeting this standard is not feasible. Therefore, what is recommended is performing the vascular procedure by implanting an allograft first, closing the retroperitoneal space as meticulously and tightly as possible, and then performing the nonvascular procedure. Several authors report that simultaneous surgery was carried out without adverse events, especially without allograft infection or an increase in mortality. Other authors reporting on the topic of aortic graft infection have described anecdotal cases in which performing synchronous procedures was considered the source of the catastrophic infective event.18,19 As a consequence, these authors strictly avoided performing simultaneous procedures.
Benign, often inflammatory, and malignant lesions may be found together with dilating vascular changes in the aortoiliac region. Table 6 summarizes the frequency of usually benign biliary tract problems, as published in the literature. Patients with AAAs are usually 60 years of age or older, and biliary calculi are commonly seen in the same, advanced age group. The incidence varies between 4.7% and 19.0%: this is obviously the most commonly seen CD. According to the literature, concomitant cholecystectomy was carried out at a rate of 2.4% to 12.5% and with a reported mortality of synchronous operations of 0% to 8.3%. Three biliary tract complications—possibly avoided with a simultaneous cholecystectomy—were observed in a pooled series of 2000 aortic graft replacements collected from the literature.3,4,11- 13,20- 22 In the presence of an active biliary problem (ie, acute cholecystitis, biliary colic, pancreatitis), it is accepted to perform a stepwise procedure. A gangrenous cholecystitis is able to mask an AAA in rupture or vice versa. We adhered to a very restrictive treatment policy in carrying out simultaneous cholecystectomy (1.09%).
Regarding ulcer disease, it is observed that after the Bouhoutsos et al19 article in 1974 (7 ulcers out of 51 AAAs), neither Bickerstaff et al11 nor Tennant and Baird21 reported on that problem. This trend reversal in the treatment of ulcer disease is ascribed to the availability of histamine2 receptor antagonists. Our experience showed the same tendency.
Hernias of the abdominal wall are observed at a rate of 1.9% to 5.9%.11,21 This comorbidity should be treated simultaneously because it is located in the access route to the AAA. Valid evidence to support appendectomy or resection of the Meckel diverticulum simultaneously with open graft replacement is not available in the literature.
Malignant lesions usually shorten the life expectancy of patients. Elective aortic graft replacement is prophylactic and directed toward prolonging the patient’s life by preventing aneurysm rupture and death from hemorrhage. The operative risk of AAA treatment must be weighed against the patient’s life expectancy. As summarized in Table 7, colorectal tumors are the most commonly seen concomitant malignancies, although generally occurring in a low incidence rate.
Table 8 outlines the incidence of tumors in the urogenital tract. The recommendation to treat tumors or nodules located in the retroperitoneal space synchronously is reasonable. Staging the procedure may cause complications because of extensive scar formation. Because the treatment is a mere ablative procedure, infective problems have not been reported. On the contrary, AAA and urinary bladder malignancies may cause intraoperative difficulties, especially owing to lesions of the ureter. Difficult lymphadenectomies or reconstructions of the urinary conduit may be technically challenging to perform. Therefore, performing the aortic graft replacement synchronously with the lymphadenectomy is recommended, followed by cystectomy at a later stage.
For simultaneous procedures, the probability of survival is smaller than that for a solitary lesion. Only 25% of the patients survive for a median of 18 months. One has to bear in mind that all of the observed clinical series are small and inhomogeneous, that data collection was done in retrospect, and that various malignancies are pooled.
Usually, the presence of an AAA and a neoplastic CD, both of which need surgical treatment, moves the patient into a “high-risk” group. With the advent of EVAR, an innovative approach allows the treatment of the aortoiliac aneurysm first in a minimally invasive way, followed by a standard procedure to treat the CD. As we demonstrated, EVAR offers acceptable mortality rates, particularly in high-risk patients.26 A risk-adjusted use of EVAR significantly reduces the operative mortality. For patients with advanced age and reduced physical status as expressed by the ASA classification, EVAR offers a reduction of the surgical burden, therefore making an active intervention feasible in patients for whom operative repair otherwise would have been denied. In our patients, no death occurred in low-risk patients, and in ASA class 3 or 4 patients, we found a significant difference in mortality rates between the 2 treatment concepts.
A different problem is how to manage a CD that is discovered intraoperatively during a planned open graft replacement. With a staged policy when CD repair is performed first, reports indicate an increased risk of AAA rupture in the postoperative period following unrelated laparotomy. This effect can be circumvented by surgically treating the CD instead of performing open graft replacement and by excluding the AAA by EVAR a few days later during the same hospitalization period.
In the treatment of obliterative arterial disease and CDs, it is possible to use different endoluminal or extra-anatomic procedures, thereby avoiding orthotopic reconstructions. In the case of concomitant occurrence of AAAs and CDs, it is reasonable to recommend endoluminal aneurysm repair first, followed by standard management of the CD. Because around 60% of patients with AAAs can be managed with the current use of endovascular prosthesis,27 the obvious dilemma might not be solved in all but a subgroup of patients, in whom the anatomical preconditions of the AAAs are suitable for EVAR.
Correspondence: Harald Teufelsbauer, MD, Department of Vascular Surgery, Medical University of Vienna, Waehringer Guertel 18-20, 1090 Vienna, Austria (firstname.lastname@example.org).
Accepted for Publication: November 26, 2004.
Author Contribution: Drs Prusa and Wolff contributed equally to this work.