Ann Thorac Surg 2007;83:S851-S855
© 2007 The Society of Thoracic Surgeons
Supplement
Long-Term Follow-Up After Thoracoabdominal Aortic Aneurysm Repair
Marc A.A.M. Schepens, MD, PhDa,*,
Johannes C. Kelder, MDb,
Wim J. Morshuis, MD, PhDa,
Robin H. Heijmen, MD, PhDa,
Eric P. van Dongen, MD, PhDc,
Huub T.M. ter Beek, MDc
a Department of Cardiothoracic Surgery, St. Antonius Hospital, the Netherlands
b Department of Cardiology Research and Statistical Analysis, St. Antonius Hospital, the Netherlands
c Department of Anesthesiology and Intensive Care, St. Antonius Hospital, the Netherlands
* Address correspondence to Dr Schepens, St. Antonius Hospital, Department Cardiothoracic Surgery, Koekoekslaan 1, 3435 CM Nieuwegein, the Netherlands (Email: m.schepens{at}antonius.net).
Presented at Aortic Surgery Symposium X, New York, NY, April 27–28, 2006.
 |
Abstract
|
|---|
BACKGROUND: Early mortality and morbidity after thoracoabdominal aortic aneurysm (TAAA) repair has been analyzed extensively; however, very few studies have examined the risk factors for late death.
METHODS: We analyzed 500 consecutive TAAA repairs performed at St. Antonius Hospital between 1981 and March 30, 2006. Survival and freedom from aortic reoperation were calculated using the Kaplan-Meier method, and the effects of preoperative, intraoperative, and postoperative risk factors were evaluated using Cox proportional hazard analysis. Survival was compared with a Dutch population matched for age, sex, and date of operation.
RESULTS: Patient survival with 95% confidence intervals (CI) was 83% (80% to 86%), 63% (58% to 67%), 34% (29% to 40%), 16% (9% to 20%), and 6% (2% to 11%) after 1, 5, 10, 15, and 20 years, respectively, compared with 100%, 99%, 85%, 36%, and 15% for the matched Dutch population. Hazard analysis showed an early phase of high hazard falling to low levels 9 months postoperatively and a late phase in which the hazard of death gradually increased. Incremental risk factors for late death were depressed left ventricular function (p < 0.001), increased age (p < 0.001), urgency (p = 0.007), postoperative dialysis (p < 0.001), and postoperative neurologic deficit (p = 0.016). Freedom from reoperation on the aorta with 95% CI was 98% (97% to 99%), 92% (89% to 94%), 86% (82% to 90%), 83% (78% to 87%), and 83% (78% to 87%) after 1, 5, 10, 15, and 20 years, respectively.
CONCLUSIONS: Survival remains suboptimal, especially in the early years after TAAA repair, compared with a matched population. Avoidance of postoperative problems such as dialysis and neurologic deficits and performing elective surgery on relative young patients with unimpaired ventricular function will increase the likelihood of late survival.
|
Introduction
|
|---|
Risk factors for and results of operative mortality and morbidity after thoracoabdominal aortic aneurysm (TAAA) repair have been investigated extensively [1–6]. Most previous studies have focused on risk factors for short-term outcomes, however, and little is known about long-term results. Crawford and colleagues [1] showed that age, rupture, renal dysfunction, type of aneurysm, and aortic dissection are important factors influencing late death. We were able previously to demonstrate the important impact of age, rupture, and postoperative dialysis on late survival [7, 8]. The goal of the present study was to examine the effects of variables on late survival in a large cohort of patients with TAAA repairs. This may contribute to better preoperative risk stratification and counseling of patients.
|
Patients and Methods
|
|---|
Between 1981 and March 2006, 500 consecutive TAAA repairs were evaluated. The local Ethics Committee approved this retrospective study and waived the need for individual patient consent because only completely anonymous data collection from medical records was used. There were 182 women (36.5%) and 318 men (63.5%), and their mean age was 65.6 years (range, 21.3 to 89.7 years). All aneurysms limited to the thoracic aorta (ascending, arch or descending) were excluded, which left 110 type I (22.0%), 234 type II (46.8%), 103 type III (20.6%), and 53 type IV (10.6%) aneurysms.
Surgical Procedure
Our operative technique has been described in detail [4, 7–9]. In summary, we initially used simple cross-clamping (from 1981 to 1987), and from then on, left heart bypass combined with staged clamping, together with evoked potential monitoring (somatosensory potentials since 1984 and motor potentials since 1994), moderate hypothermia, and liberal reimplantation of intercostal and lumbar vessels (especially between the level of T8 and L2, during the last years additionally guided by motor evoked potentials).
We did not use visceral perfusion except in 2 patients. Partial or total extracorporeal circulation, with or without deep hypothermic circulatory arrest, was used only in cases of aortic arch involvement, or if the TAAA was so big that safe entry into the chest was impossible, or when single lung ventilation was not tolerated. We used extracorporeal circulation in 34 patients (6.8%) and deep hypothermic circulatory arrest in 4.
Follow-Up
Follow-up information on all patients was gathered by the principal author (M.A.S.) and updated between November 2005 and March 30, 2006. This information was acquired from our outpatient clinic, from the referring cardiologist, vascular surgeon or other treating specialist, from the family doctor, or by direct telephone contact with the patient. Follow-up was 100% complete (500/500), despite the international referrals. Total follow-up was 2520 patient-years (mean, 5.0 ± 4.4 years). The longest follow-up was 21 years.
Cardiovascular reinterventions imply coronary artery bypass grafting, percutaneous balloon coronary angioplasty, mitral valve repair or replacement, aortic valve replacement, cardiac transplantation, bypass operations for peripheral occlusive disease, carotid endarterectomy, carotid dilatation and stenting, amputation of the lower limb, or correction of aneurysms of the abdominal aorta, the popliteal, femoral, iliac, or subclavian artery.
Statistical Analyses
Preoperative, intraoperative, and postoperative variables (Appendix) thought to be clinically important were analyzed to detect their influence on late mortality. Standard univariate methods (Yates corrected 
2 or the Student t test) were used. The end points for this study were all-cause mortality from the time of TAAA repair, and late reoperation on the aorta or cardiovascular system. Cumulative survival curves were constructed using the Kaplan-Meier product-limit method [10] accompanied by 95% confidence intervals (CI).
Survival was compared with the expected survival of a Dutch population matched by calendar year, age, and sex (Statistics Netherlands, www.cbs.nl). The Cox proportional hazard model was used to predict the independent determinants of long-term survival [11]. All computations were made with the aid of SAS 8.2 (SAS Inc, Cary, NC).
|
Results
|
|---|
During follow-up there were 216 (43.3%) late deaths (Table 1). Risk-unadjusted survival with 95% CI was 83% (80% to 86%), 63% (58% to 67%), 34% (29% to 40%), 16% (9% to 20%), and 6% (2% to 11%) at 1, 5, 10, 15, and 20 years after TAAA repair, respectively (Fig 1), and 100%, 100%, 85%, 36%, and 15%, respectively, for the age-matched and sex-matched population.
View larger version (18K):
[in this window]
[in a new window]
|
Fig 1. Overall survival after thoracoabdominal aortic aneurysm surgery. The solid line represents the survival of the study population (with the 95% upper and lower confidence limits as dotted lines). The dash-dot line represents the life-table estimate survival of a Dutch population matched for age, sex, and time of operation. Numbers in parentheses are patients at risk.
|
|
The hazard of death was highest immediately after operation and fell to a very low level after 9 months (Fig 2). Then, it began to rise steadily and was similar to that predicted for the Dutch population matched for age, sex, and time of operation. The hazard function consisted of a brief early phase and a gradually rising mid-term and late phase. Independent risk factors for death included left ventricular ejection fraction, age, postoperative dialysis, and postoperative neurologic deficit (Table 2). The time period (year of the operation) was not a significant factor in the analysis for late death, but it reached statistical significance in the analysis for hospital death: mortality was much higher in the patient group operated on before 1996 (p = 0.0014), with a relative risk of 6.6 (2.1 to 21.4).
View larger version (16K):
[in this window]
[in a new window]
|
Fig 2. Hazard function. The solid line indicates the hazard of dying of the study population (with the 95% upper and lower confidence limits as dotted lines). The dash-dot line represents the hazard of dying for a Dutch population matched for age, sex, and time of operation. (TAAA = thoracoabdominal aortic aneurysm.)
|
|
During follow-up, 55 reoperations on the aorta were necessary in 50 patients. The causes are summarized in Table 3. Apart from the 5 patients who were successfully operated on for an acute type A dissection, 5 additional patients sustained an acute type A dissection during follow-up. These patients all died because the diagnosis was missed and they did not undergo operation (positive postmortem findings).
As can be seen from Table 3, 2 patients received an endoprosthesis at 5 and 9 years after the original intervention for a false aneurysm at the reimplanted intercostal artery patch. Another patient with a nongrowing false aneurysm at the reimplanted intercostal artery patch has been monitored for 3 years. During follow-up, 2 patients died from a ruptured false aneurysm: one at the level of the renal arteries at 10 years and another at the intercostal arteries at 9 years. The 9 patients who needed an exclusion for an abdominal aortic aneurysm received open surgical repair.
Freedom from reoperation on the aorta with 95% CI was 98% (97% to 99%), 92% (89% to 94%), 86% (82% to 90%), 83% (78% to 87%), and 83% (78% to 87%) after 1, 5, 10, 15 and 20 years, respectively (Fig 3). In the Marfan patients, these numbers were 83% (65% to 100%), 77% (56% to 97%), and 38% (0% to 77%) after 1, 5, and 10 years (p = 0.0002). Freedom from all cardiovascular reinterventions with 95% CI (including all operations on the heart, the aorta and the peripheral vascular tree) was 97% (96% to 99%), 87% (84% to 90%), 79% (75% to 83%), 77% (72% to 82%), and 75% (70% to 81%) after 1, 5, 10, 15, and 20 years, respectively (Fig 4).
View larger version (11K):
[in this window]
[in a new window]
|
Fig 3. Freedom from reoperation on the aorta. The solid line represents the freedom from reoperation on the aorta (with the 95% upper and lower confidence limits as dotted lines). Numbers in parentheses are patients at risk.
|
|
View larger version (12K):
[in this window]
[in a new window]
|
Fig 4. Freedom from cardiovascular reoperation. The solid line represents the freedom from cardiovascular reoperation (with the 95% upper and lower confidence limits as dotted lines). Numbers in parentheses are patients at risk.
|
|
|
Comment
|
|---|
Surprisingly, the long-term survival of TAAA repair has not been described extensively, in contrast with most other degenerative vascular diseases; however, morbidity and early results have been thoroughly analyzed. It was the late Stanley Crawford who introduced the surgical technique of simple cross-clamping and was also the first to describe the late survival of 60% ± 3% at 5 years and 32% ± 6 % at 10 years [1] in a series of 605 TAAA repairs. These numbers were confirmed in a larger analysis by Svensson and colleagues [2] in 1993.
After multivariate analysis, Crawford and colleagues [1] found that age, rupture, renal dysfunction, and aortic dissection all had an important influence on long-term survival. Coselli and colleagues [12] reported a late survival of 74% at 5 years. The risk-unadjusted survival of our patients of 34% at 10 years is very similar to the Crawford series, but our series extends an additional decade, with a maximal survival of 21 years. The relative short mean follow-up of 5 years was because many more cases were operated on during the last years.
The type of aneurysm has no influence on late survival in our series. When survival between the treated patients and population matched for age, sex, ethnicity, and time of operation-matched is compared, it is obvious that the hospital mortality initially accounts for the significant difference. Surprisingly, after only 5 years, both curves show a parallel decline, with a similar ongoing downsloping trend indicating that the impact of the repair is much greater than expected. In general, survival seems to be suboptimal, especially for the first 5 years, but it is equivalent to that in other degenerative vascular diseases, such as after abdominal aortic aneurysm repair. Moreover, when compared with the natural history of this disease—a survival 2 years after admission of only 24% [13]—the surgical approach is more than justified.
The risk of dying after the intervention is highest during the first months, but then levels off and reaches a comparable hazard as in an age-matched and sex-matched Dutch population, something that might seem to be evident. This illustrates that surgery for TAAA has a formidable impact on the patient, despite the multimodality approach using left heart bypass, spinal fluid drainage, reimplantation of patent intercostal or lumbar arteries, or both, and other adjuncts. Avoiding postoperative renal and neurologic problems, central as well as spinal, has major consequences for late survival.
Our results underscore that elective intervention has an important impact on survival and that emergent or urgent intervention should be avoided, if possible. Actually, at St. Antonius Hospital, procedures in patients with a symptomatic TAAA and signs of rupture are postponed if they hemodynamically stable until the most experienced team is available to perform the repair. It is obvious that we need further optimization of immediate and early postoperative care if we want to reduce further the impact of complications such as respiratory and renal failure, which are very often the beginning of multiorgan failure. This attempt must start with the best possible preoperative preparation, with optimization of cardiovascular and respiratory status.
Because postmortem examination is not obligatory in The Netherlands, the precise cause of late death remains unknown in a substantial number of patients; however, we have clear evidence that aorta-related death occurred in 29 patients (5.8%). This number might even be higher if one considers the likelihood that at least a few of the unknown deaths might be attributable to an aorta-related problem. It underscores the necessity of periodic evaluation by specialists who are focused not only on the heart or lungs but also on the entire cardiovascular system. The substantial number of aortic reoperations at the location of the original repair as well as at remote sites emphasizes the notion that aortic degenerative pathology is an ongoing disease and reinforces the necessity of yearly aortic surveillance.
It is of interest that false aneurysm at intercostal patches, occurring in 0.6 % in our series, can be treated noninvasively by insertion of a thoracic endoprosthesis, avoiding difficult reoperative surgery. In these circumstances, we advise extensive preoperative imaging using computed tomography scanning and angiography to delineate the presence of patent intercostal vessels. If these are present, we would not hesitate to overstent them because we think that the collateral circulation should be sufficient to avoid spinal cord problems.
The limitations of this study include its retrospective nature. Although the number of patients at risk at 20 years is very low, this single-center experience provides us with long-term survival after TAAA, without patients lost to follow-up. In addition, it is compared with a Dutch population matched for age, sex, and time of operation. The functional status of the patients at follow-up, however, was not evaluated.
|
Appendix
|
|---|
Preoperative, Intraoperative, and Postoperative Variables Used in the Analysis
|
References
|
|---|
- Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients J Vasc Surg 1986;3:389-404.[Medline]
- Svensson LG, Crawford ES, Hess KR, Coselli JS, Safi HJ. Experience with 1509 patients undergoing thoracoabdominal aortic operations J Vasc Surg 1993;17:357-370.[Medline]
- Safi HJ, Estrera AL, Miller CC, et al. Evolution of risk for neurologic deficit after descending and thoracoabdominal aortic repair Ann Thorac Surg 2005;80:2173-2179.[Abstract/Free Full Text]
- Schepens MA, Defauw JJ, Hamerlijnck RP, Vermeulen FE. Risk assessment of acute renal failure after thoracoabdominal aortic aneurysm surgery Ann Surg 1994;219:400-407.[Medline]
- Coselli JS, LeMaire SA, Conklin LD, Köksoy C, Schmittling ZC. Morbidity and mortality after extent II thoracoabdominal aortic aneurysm repair Ann Thorac Surg 2002;73:1107-1116.[Abstract/Free Full Text]
- Coselli JS, LeMaire SA, Köksoy C, Schmittling ZC, Curling PE. Cerebrospinal fluid drainage reduces paraplegia after thoracoabdominal aortic aneurysm repair: results of a randomized clinical trial J Vasc Surg 2002;35:631-639.[Medline]
- Schepens MA, Defauw JJ, Hamerlijnck RP, De Geest R, Vermeulen FE. Surgical treatment of thoracoabdominal aortic aneurysms by simple crossclamping: risk factors and late results J Thorac Cardiovasc Surg 1994;107:134-142.[Abstract/Free Full Text]
- Schepens MA, Dekker E, Hamerlijnck RP, Vermeulen FE. Survival and aortic events after graft replacement for thoracoabdominal aortic aneurysm Cardiovasc Surg 1996;4:713-719.[Medline]
- van Dongen E, Schepens MA, Morshuis WJ, et al. Thoracic and thoracoabdominal aortic aneurysm repair: use of evoked potential monitoring in 118 patients J Vasc Surg 2001;34:1035-1040.[Medline]
- Kaplan EL, Meier P. Nonparametric estimation for incomplete observations J Am Stat Assoc 1958;53:457-481.
- Cox DR. Regression models and life-tables J R Stat Soc B 1972;34:187-220.
- Coselli JS, LeMaire SA, Miller III CC, et al. Mortality and paraplegia after thoracoabdominal aortic aneurysm repair: a risk factor analysis Ann Thorac Surg 2000;69:409-414.[Abstract/Free Full Text]
- Crawford ES, DeNatale RW. Thoracoabdominal aortic aneurysm: observations regarding the natural course of the disease J Vasc Surg 1986;3:578-582.[Medline]
This article has been cited by other articles:
|
|
|
|
 
G. C. Hughes, J. J. Nienaber, E. L. Bush, M. A. Daneshmand, and R. L. McCann
Use of custom Dacron branch grafts for "hybrid" aortic debranching during endovascular repair of thoracic and thoracoabdominal aortic aneurysms
J. Thorac. Cardiovasc. Surg.,
July 1, 2008;
136(1):
21 - 28.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. C. Miller III, M. A. Villa, P. Achouh, A. L. Estrera, A. Azizzadeh, S. M. Coogan, E. E. Porat, and H. J. Safi
Intraoperative skeletal muscle ischemia contributes to risk of renal dysfunction following thoracoabdominal aortic repair
Eur. J. Cardiothorac. Surg.,
April 1, 2008;
33(4):
691 - 694.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
