| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 1997;63:1533-1544
© 1997 The Society of Thoracic Surgeons
Departments of Cardiothoracic Surgery and Biomathematics, Mount Sinai School of Medicine, New York, New York
Abstract
Background. The decision whether or not to recommend resection of moderately large descending thoracic and thoracoabdominal aneurysms requires weighing the relatively high mortality and significant risk of paraplegia associated with operation against the likelihood that the aneurysm will rupture spontaneously, with an almost invariably fatal outcome. To better define the risk of aneurysm rupture, we undertook a prospective study of patients who had not had operation on their moderately large descending thoracic and thoracoabdominal aneurysms.
Methods. Patients were enrolled at the time of their second computed tomographic scans: three-dimensional computer-generated reconstructions allowed determination of several dimensional parameters for each study, including diameters and cross-sectional areas at the site of maximal dilatation in the descending aorta and in the abdomen as well as total thoracoabdominal surface area. Comparisons of serial studies permitted calculation of yearly rates of change in these dimensions.
Results. Of 114 patients, 8 died of causes unrelated to the aneurysm, 26 died of rupture, 20 met previously determined criteria for operation, and 60 survived without operation or rupture. Multivariate regression analysis identified maximal diameter in the descending and in the abdominal aorta as independent risk factors for rupture, as well as older age, the presence of even uncharacteristic pain, and a history of chronic obstructive pulmonary disease. A piecewise exponential model enabled construction of an equation allowing calculation of rate of rupture in patients in whom the values of the risk factors are known, and also of the probability of rupture in a given individual over a specified time interval.
Conclusions. Because using this equation-based on easily determined risk factors (age, pain, chronic obstructive pulmonary disease, maximal thoracic and maximal abdominal aortic diameter)-allows the risk of aneurysm rupture within a given interval to be estimated fairly accurately for each individual patient, it is our current practice to recommend operation when the calculated risk of rupture within 1 year exceeds the anticipated mortality of elective operation, rather than relying on general operative guidelines based almost exclusively on aneurysm size.
There is a reasonable consensus regarding the need to resect large, rapidly expanding, and symptomatic descending thoracic or thoracoabdominal aneurysms electively in patients with no major operative contraindications, but the asymptomatic high-risk patient with an aneurysm of moderate size still presents a dilemma for most surgeons. Although mortality for elective resection of thoracic aortic aneurysms is reasonably low, neurologic morbidity is a major concern, deterring a liberal approach toward early surgical treatment. On the other hand, the natural history of thoracic aortic aneurysms indicates that rupture occurs in a high proportion of untreated patients, almost invariably resulting in death. Making a choice between the risk of a devastating complication of operation and the risk of almost certain death if the thoracic aortic aneurysm ruptures would be considerably less difficult if it were possible to determine more precisely which aneurysms are most likely to rupture: one could then operate electively on those patients before rupture occurs. This study is part of an ongoing attempt to better define the factors predisposing to rupture of descending thoracic and thoracoabdominal aneurysms, including various characteristics of the aneurysm and its pattern of expansion.
Our approach has involved the study of aneurysms and their behavior using various measurements derived from three-dimensional reconstructions of serial computed tomographic (CT) scans in patients who have not been operated on. In an earlier analysis [1], we examined possible risk factors leading to rapid aneurysm expansion, assuming that rapid expansion was a prelude to rupture: this assumption was necessary to derive any meaningful conclusions from a relatively small number of patients, few of whom had rupture while under observation. Our major finding was that in multivariate analysis only aneurysm size predicted growth rate, and that overall survival of patients with large aneurysms (>5 cm) was poorer than survival in patients with smaller aneurysms [1]. Ascertainment of risk factors for rupture and quantitative assessment of their relative importance was not possible in our earlier study, making the decision whether or not to recommend operation a rather crude balancing act, as depicted in Figure 1
.
|
Patients and Methods
Patient Population
Since 1988 at Mount Sinai Hospital, patients with chronic thoracic aortic aneurysms have been followed up clinically using a protocol that involves serial evaluation of their aneurysms using computer-generated three-dimensional reconstructions of CT scans. The aorta is considered ectatic or aneurysmal if it has reached a maximal diameter of 3.5 cm or more, or if it measures twice the diameter of the adjacent normal aorta. One hundred fourteen patients with descending thoracic or thoracoabdominal aortic aneurysms who did not meet standard criteria for immediate surgical intervention and had at least two CT studies separated by a minimum interval of 3 months were entered into this prospective natural history study at the time of their second CT scan. Patients with chronic dissection were excluded. During this interval, 233 patients underwent resection of descending thoracic or thoracoabdominal aneurysms at our institution; thus approximately 1 patient in 3 was not selected for operation after the first evaluation.
During the study, 8 patients died of causes judged unrelated to their aneurysms: 2 with malignancies, 2 with documented myocardial infarcts, 1 with pneumonia, 1 after a pulmonary embolus, and 2 of unknown causes; none will be considered further. Sixty patients remained alive without rupture or operation.
Twenty-six patients died after unexpected rupture of their aneurysms verified by autopsy records, or had aortic rupture documented at the time of emergency operation. All patients in the study who experienced rupture died, including 3 who underwent emergency aneurysm resection; the outcome in all 10 patients who underwent emergency resection for ruptured descending thoracic aortic aneurysms during this interval at Mt Sinai Medical Center showed a 50% survival.
Twenty patients underwent elective aneurysm resection. Indications for operative treatment included (1) the presence of pain or other symptoms suggestive of rapid expansion or impending rupture; (2) an absolute aortic diameter of more than 7 cm; (3) an increased rate of growth, defined as an increase in diameter of 1 cm per year or more; and (4) marked irregularity of aneurysm contour suggestive of localized wall weakness. Sixteen of these 20 surgical patients had three or more preoperative CT studies, and their data are included up to the time of their penultimate scans; 4 patients who had only two CT scans before operation are excluded from all but the data on patients who underwent operation, because whether or not their aneurysms would have ruptured after their last CT scan could not be known. The data from the interval preceding the last preoperative scan in all patients who underwent operation were tabulated and analyzed separately.
Data from 102 patients were included in the longitudinal analysis: 90 with descending thoracic and 12 with thoracoabdominal aortic aneurysms. Patients were monitored by means of clinical examinations and CT scans of the aorta, which were scheduled at 6-month intervals. For the purpose of the study, follow-up was begun at the time of each patient's second CT scan, and was terminated either at the time of rupture, the date of the last CT scan preceding an elective operation, or the last date at which the patient was confirmed to be alive without rupture or elective operation, which was July 1, 1996, or later. The median number of CT scans was four (range, from 2 to 13), and the median length of follow-up was 28 months.
A medical history geared toward maximizing information about factors likely to contribute to aneurysm rupture was elicited from each patient. A history of possibly relevant symptoms was obtained: if pain was severe, of recent onset, and not amenable to an alternative explanation, it was usually attributed to the aneurysm and the patient was referred for operation, but often if the pain was more chronic and could be explained as angina or osteoarthritis or attributed to the presence of an ulcer or hiatus hernia, it did not result in a decision to operate. A careful smoking history included whether the patient had ever smoked, whether the patient was smoking currently, and approximately how many packs per year were consumed, but for the purpose of this study no attempt was made to quantify tobacco use: any patient with a history of smoking was considered a smoker. Documentation of the presence and duration of hypertension and of diabetes was recorded. Evidence of a history of chronic obstructive pulmonary disease (COPD) was carefully elicited, and forced expiratory volume in 1 second (FEV1) was measured in patients during the latter part of the study, and reported as percent FEV1 predicted.
Dimensional Variables
A computer program was developed that allows three-dimensional reconstruction of the aorta from serial sections of CT scans, as has been reported previously [2]. Data are obtained by tracing the outline of the aneurysm from each CT slice using a translucent digitizing tablet and a digitizing puck. As seen from the example in Figure 2, the computer program permits (1) a three-dimensional reconstruction of the aorta as a series of stacked cylinders that can be viewed from any angle desired (usually a left anterior oblique projection); (2) portrayal of the four different aortic segments in different colors (ascending in blue, arch in black, descending in red, abdominal in turquoise); (3) determination of the diameter of each aortic slice; (4) an indication of the largest diameter in each segment; (5) the volume of each segment, calculated by multiplying the area of each slice by its height (usually 1 cm) and summing the numbers of slices per segment; (6) the total volume of the aorta; and (7) the tortuosity index, a ratio of the measured length of the descending aorta to its minimal theoretical height. A linear graph of the diameter and cross-sectional area of each aortic slice is also plotted, allowing easy comparison of serial studies. The descending thoracic aortic segment, as seen in red, is demarcated by the underside of the arch cranially and the crura of the diaphragm caudally.
|
.
|
, the circumference of the aorta is modeled in the computer as a set of equidistant points that define a polygon. The diameter is calculated by measuring the length of a line drawn from each point to the point opposite and passing through or near as possible to the center of mass of the polygon; this is done for all the points in the polygon. The program then reports the smallest diameter of the elliptic cross-section as the slice diameter. When all the slices have been processed, the highest value among the diameters of the individual slices is reported as the maximal diameter. The closest approximation to this value without having access to this computer technology would be obtained by measuring the lesser diameter of the ellipse formed by the aneurysm in the region of maximal dilatation of the aorta on a conventional CT scan.
|
Furthermore, with the hypothesis that impending rupture may be heralded by an accelerated rate of expansion, several yearly rates of aneurysm growth were also calculated: among these, the growth in maximal descending aortic diameter per year, in maximal abdominal diameter, and in maximal cross-sectional area per slice in the descending aorta are reported in the current study.
Statistical Analysis
Comparisons of demographic and dimensional data between patients with and without aneurysm rupture and between patients who did or did not undergo operation were undertaken using 
2 and Wilcoxon tests of significance, as appropriate.
For the prospective analysis, the interval of follow-up was partitioned at the times of successive CT scans to incorporate changes in age and in dimensional features of the aneurysm. Factors related to risk of rupture were studied using a piecewise exponential model, with the assumption that for a given set of factors, risk of rupture is constant and is not influenced by the time since diagnosis or by the length of time the patient has been under surveillance.
The assumption of constant risk for a given set of factors was tested by introducing a Weibull model, and also by partitioning the time after each scan into 4-month intervals: no significant time effects were found. Increasing numbers of CT scans also did not significantly influence risk of rupture. These observations were thought to justify use of the piecewise exponential model.
Under this model, the rate of rupture, 
, is given by (X1, X2, ..., Xk) = exp(
+ ß1X1 + ...ßkXk), where (X1, X2, ..., Xk) are the values of k factors that independently influence the rupture rate, (, ß1, ..., ßk) are constants, and (expß1, expß2, ..., expßk) are the relative rates for the corresponding factors. The probability of rupture within t days is then [1 - exp{-( + ß1X1 + ...ßkXk)t}]. Significant risk factors associated with rupture rate were identified by multivariate regression analysis using the piecewise exponential model, and estimates obtained of the corresponding coefficients. The significant factors and their corresponding coefficients were then used to obtain estimates of rupture rates for patients with specific values of the risk factors. These, in turn, allowed estimation of the probability of rupture of the aneurysm in individual patients within a given period of time. All calculations were implemented with SAS programs on a VAX computer [3].
Results
Characteristics of the Entire Study Population
A summary of the demographic data and information on the presence of risk factors in the 102 patients included in the longitudinal natural history study are depicted in Table 1. There were 59 men and 43 women, with a median age of 72 years. Half the patients were smokers, and a history of hypertension was found in 2 of every 3 patients. Chronic obstructive pulmonary disease was identified from historical data in 15 patients; in the 44 patients screened for pulmonary function, the median FEV1 was 68% of predicted values. Only 2 patients were identified as having Marfan's disease, and neither of these aneurysms ruptured or required operation.
|
.
|
Comparison of those patients who eventually had rupture, whether at operation or at autopsy, versus the patients in whom rupture did not occur is shown in Tables 3 and 4. Table 3 shows that among the patients who had rupture, there is a significantly higher proportion of women, a significantly higher median age, and a higher incidence of pain and of COPD. All patients who had rupture died, including 3 who underwent operation after rupture.
|
|
The overall rate of rupture among patients initially managed nonoperatively is 23%, but actually the rupture rate for all patients treated for descending thoracic and thoracoabdominal aneurysms at our institution, most of whom underwent operation, is considerably lower (only 11%).
Multivariate Analysis
Multivariate risk factor analysis demonstrated that age, pain, COPD, and maximal descending aortic and abdominal aortic diameters were independent risk factors for aneurysm rupture (Table 5). Demographic factors that were analyzed but failed to add any additional prognostic information regarding rupture include sex and a history of hypertension, smoking, or diabetes.
|
All dimensional parameters were also evaluated to see whether, if substituted for descending aortic diameter in the model, they would equal-if not improve-prognostication of rupture if the abdominal diameter and all the nondimensional risk factors were kept unchanged. It was found that descending aortic cross-sectional area was equal in significance in the multivariate analysis if substituted for maximal descending aortic diameter. If descending aortic volume is substituted for maximal descending thoracic aortic diameter in the multivariate equation, the significance decreases; this is also true if thoracoabdominal surface area is substituted for maximal diameter of the descending aorta, as shown in Table 6.
|
, after a CT scan is given below:
 | (1) |
where pain and COPD = 1 if present and 0 if absent or not reported, and age refers to the time of the most recent scan
 | (2) |
To make estimation of probability of rupture easier, we have depicted, in Figure 5, the calculated probability of rupture at three different ages according to maximal descending aortic diameter, with different combinations of other major risk factors such as the presence of COPD and pain. In the graphs suitable for use in estimating rupture for patients with an abdominal diameter less than 5 cm, the average abdominal diameter, 3.8 cm, in patients with small abdominal diameters (<5 cm) in the entire patient group was used in the equation. For patients with large abdominal aortic diameters, the average value in the study patients with abdominal diameters greater than 5 cm, which was 5.8 cm, was used to generate the graphs.
|
. In Appendix 1 we have also included a formula that allows calculation of 95% confidence limits for the probability of rupture.
The graphs in Figure 5 reinforce the importance of all the risk factors cited: age, a history of COPD and of pain, and the influence of aneurysm size on risk of rupture.
Patients Who Underwent Operation
We thought it would be of interest to examine the 20 patients originally included in the prospective study who underwent elective repair of their descending or thoracoabdominal aortic aneurysms for the indications previously outlined. All 20 patients survived operation, but paraplegia developed in 1 patient.
The specific indications for operation in these patients are shown in Table 7; some patients had more than one of the necessary criteria. The prevalence of pain, large aneurysm diameter, and rapid rate of growth in these patients may have resulted in an underestimation of these factors in the multivariate prognostic equation because data from the last CT scan in these patients were not included in the general analysis.
|
. These data show no significant differences apart from a preponderance of men among patients who were operated on, which we are at a loss to explain.
|
; the data for patients who were operated on refers to their last preoperative study. Significantly greater values for all measurements in the descending aorta are seen in the patients selected for operation, as well as a significantly higher thoracoabdominal surface area and a higher tortuosity index. All the selected dimensional measures of expansion are also significantly greater in the patients who underwent operation. These results are not surprising in view of the fact that dimensional criteria and rapid growth rates were used as indications for operation.
|
If the 3 patients with a very low risk of rupture according to our formula who nevertheless underwent operation are more closely examined, some of the potential pitfalls of too exclusive a reliance on this prognostic formula are revealed. One patient was operated on because a serious cough developed, an unusual symptom not accounted for in the formula, but a sign of recurrent laryngeal nerve irritation suggestive of rapid distal arch and proximal descending thoracic aortic expansion, and therefore a reasonable indication for operation. Another patient reported recent onset of severe pain, again undoubtedly a valid indication for elective operation. Pain by itself may result in a low calculated risk of rupture from our formula because aneurysm-related pain is an indication for operation, and therefore many patients presenting with pain were operated on after only one CT scan, and never became part of the nonoperative follow-up group, leading to an underestimation of the importance of pain in the prognostic formula. The third patient with a low likelihood of rupture based on the formula was classified as having a thoracoabdominal aneurysm but he had chiefly a large, rapidly expanding abdominal aneurysm, with the thoracic component only modestly enlarged and stable; it is not too surprising that this formula, which was designed to predict rupture in descending thoracic and thoracoabdominal aneurysms, may not be accurate when evaluating a lesion that is almost exclusively an abdominal aneurysm.
The much more rapid rate of growth of aneurysms in the small group of patients who underwent operation compared with the larger group who were not operated on suggests that, because these patients were removed from the pool of those at risk of rupture because rapid growth rate was one of the factors prompting referral for operation, patients with rapid aneurysm growth are also probably underrepresented in the group from which the prognostic formula was derived, leading to an underestimation of rate of growth as a risk factor for aneurysm rupture. But if aneurysm size is a powerful predictor of growth rate, as shown in our previous study [1], it is also conceivable that even a pure natural history study, obviously impossible on ethical grounds, might not show that growth rate improves prognostication if used in addition to size alone.
The importance of aneurysm diameter, as has already been pointed out for pain, may also have been underestimated in the prognostic formula because of the removal from consideration of patients selected for operation because of large aneurysm size after only one CT study. It should be noted, however, that our size criterion for immediate operation (a maximal diameter > 7 cm) was quite conservative, so that the study population is not limited to very small aneurysms, but includes a number of patients with aneurysms from 5 to 7 cm in maximal diameter.
Comment
In the current study of the natural history of descending thoracic and thoracoabdominal aortic aneurysms, the indications for operative intervention were quite conservative. These criteria were based on the recognition that operative mortality and morbidity, although steadily improving, were still quite substantial and that there were few guidelines to suggest which patients could justifiably be exposed to these high surgical risks, apart from those with very large aneurysms or striking symptoms. The conservative operative indications led to inclusion of a considerable number of patients with relatively large diameter lesions and other attributes predisposing to rupture, and has therefore enabled us to identify and quantify risk factors for rupture of descending thoracic and thoracoabdominal aneurysms. The possible relevance of some of these factors had been suggested by earlier studies of abdominal aortic aneurysms. By permitting more accurate estimation of risk of rupture, the observations emerging from this study should enable us in the future to reduce the spontaneous rupture rate among patients with thoracic and thoracoabdominal aortic aneurysms by more aggressively recommending operation to those most at risk.
Our data once again affirm that the diameter of the aorta at the site where there is maximal dilatation is a major risk factor for aneurysm rupture. That bigger aneurysms are more likely to rupture should be no surprise: the importance of aneurysm size has been well documented in a number of studies both of abdominal [4, 5] and of thoracic aneurysms [2, 6–8], and is an observation that is easily explicable in terms of what is known about the physical forces involved in the physiology of aneurysm expansion and rupture. It is important to emphasize, however, that the diameter that we used to determine aneurysm size is not a casual measurement of a possibly elliptically distorted section of an aneurysmal aorta on a CT scan, but is an estimate of the true diameter of the aorta in the area of maximal dilatation, obtained as explained from a computer-generated three-dimensional aortic reconstruction.
The importance of age as a risk factor for rupture is somewhat surprising, but consistent with data from a recent study of ruptured thoracic aortic aneurysms in Sweden [7]. The prevailing wisdom in surgical circles has generally favored elective aneurysm resection in younger individuals, with the idea that younger patients are generally better operative candidates with a longer anticipated lifespan if the threat of aneurysm rupture can be removed, and with better potential for recovery if serious neurologic complications occur. But these considerations must now be balanced by the recognition that a younger patient has a smaller risk of rupture than his or her more elderly counterpart.
The role of pain as an indication for elective aneurysm repair must also come under renewed scrutiny. Pain is very frequent in this group of patients without operation despite the fact that patients with pain thought to be related to the aneurysm are immediately referred for operation and thus should be rare among the group involved in serial follow-up. Thus, the pain that emerged as a risk factor in the prognostic equation was predominantly pain that experienced surgeons had either dismissed as unrelated to the aneurysm, variously attributing it to ostearthritis, angina, ulcer or hiatus hernia, or had not considered acute or severe enough to take seriously as an operative indication. That the presence of such pain adds to the risk of aneurysm rupture suggests that our ability to discriminate between pain generated by the aneurysm and other kinds of pain may be suboptimal. If one adds to the demonstrated role of uncharacteristic pain as a risk factor for rupture in patients undergoing serial follow-up the likelihood that characteristic, striking aneurysmal pain was scarce among the study patients, pain becomes even more significant as a risk factor for impending rupture. Based on these observations, it seems worthwhile to make a real effort to evaluate the source of pain in patients with thoracic aortic aneurysms, and to recognize that the presence of otherwise unexplained pain increases the probability of rupture in these patients.
Cronenwett and associates [5, 9] were the first to recognize the importance of a history of COPD in increasing the risk of rupture in abdominal aortic aneurysms. It seems likely that whatever connective tissue pathology predisposes to aneurysm formation may also play a role in the pathogenesis of COPD. Given the large proportion of smokers with both diseases, it is tempting to implicate the destructive effects of smoking as significant in their etiology. But whether or not we can explain it, there is no question empirically that the presence of COPD is a powerful risk factor both for accelerated expansion of thoracic aneurysms, as recently documented by Cambria and colleagues [10], and for rupture of thoracic aortic aneurysms in our series. Our data thus far do not support the idea that measurement of FEV1, when compared with simply obtaining a history of COPD, increases our sensitivity in detecting patients likely to have rupture, but only about half our patients had FEV1 measured. We should also note that we have not found that mild to moderate COPD seriously affects the mortality or morbidity of elective aneurysm resection, so we feel strongly that patients with COPD should be offered earlier elective operation than their counterparts without chronic lung disease.
In contrast to data regarding rupture in abdominal aneurysms [11], and the findings of our earlier study, which did not consider COPD, smoking did not emerge as a risk factor for rupture in the current study. The introduction of COPD as a variable may have obscured the relevance of smoking, and the inclusion of all patients who had ever smoked as smokers may have diluted the impact of heavy and recent tobacco use. We therefore still urge patients with aneurysms to stop smoking. A rationale for our approach can be found in the recent study by MacSweeney and co-workers [12], who found that abdominal aneurysms in patients who stopped smoking enlarged significantly less rapidly (0.09 versus 0.16 cm per year; p = 0.04) than in those patients who continued to smoke, and that the growth rate of aneurysms correlated with blood levels of nicotine metabolites.
We continue to believe that aspects of aneurysms in addition to their maximal diameter are important in determining risk of rupture, even though we have had some difficulty until now in showing definitively that this is so. The importance of the abdominal diameter in addition to the maximal descending aortic diameter in the calculation of risk of rupture in the current study corroborates our notion that the length of aorta over which dilatation extends affects the risk of rupture. In addition, the highly significant differences found between many of our dimensional measurements in patients with ruptured and unruptured aneurysms reinforce the value of these data, and the even greater number of dimensional measurements with differences between patients who did or did not undergo operation suggest that the value of some of these measurements may have been underestimated in the prognostic equation. Although we appreciate the additional input that the three-dimensional computer reconstructions provide and the ease with which they allow comparison of serial studies, we recognize that much of the essential information can also be obtained from careful manual scrutiny of CT studies.
We attribute the inability to demonstrate a role for growth rate as a risk factor for rupture in multivariate analysis to its dependence on initial aneurysm diameter, a relationship previously documented by us [1] and by others [4, 6], but that has also been questioned [10]. We also think that the removal of patients with rapid growth rates from the pool of patients who continue to have serial studies without operation may have resulted in an underestimation of the importance of rapid expansion as a warning of impending rupture in our prognostic formula. The failure to show that rapid aneurysm expansion is an important risk factor for rupture may also be a consequence of the underrepresentation of patients with large aneurysm diameter in the population under study; large aneurysm diameter, which is often associated with rapid aneurysm growth, is an indication for early elective operation, and therefore its presence results in exclusion from the group of patients without operation undergoing serial studies in whom growth rates can be measured.
The failure of hypertension to emerge as a significant risk factor for rupture is also somewhat unexpected. Because all patients with thoracic aortic aneurysms are treated for hypertension if it is present, however, the lack of impact of a history of hypertension on risk of rupture may reflect the successful treatment of hypertension in most patients. Although we attempt to monitor blood pressure as part of the clinical evaluation during follow-up visits, some patients are seen locally and only their films are sent for detailed study, so we have had too few patients with reliable serial blood pressure measurements to investigate further the impact of treating hypertension on risk of rupture.
In summary, we have generated an equation from the data in this study of patients who were not operated for their chronic descending thoracic and thoracoabdominal aneurysms that allows us to weigh more precisely the risk of rupture in a given patient. The more detailed consideration of various risk factors associated with rupture must be balanced against a similarly nuanced consideration of the risk of operation, including not only the risk of death but also of paraplegia. In our most recent series of 95 patients with descending thoracic and thoracoabdominal aneurysms, which included 18% emergency cases, hospital mortality was 10.5% [13]. The overall rate of paraplegia was 2%, but it was as high as 10% in very extensive aneurysms, those requiring sacrifice of more than ten intersegmental arteries. In contrast to the gradually improving results of elective operation, the surgical mortality once rupture has occurred remains dismal: in a recent study from Sweden [7], overall mortality of ruptured thoracic aortic aneurysms was 97% even though 41% of patients reached the hospital alive.
In light of these surgical statistics, the 23% rate of rupture experienced by the patients in this series seems high, and suggests that the operative indications used in this study may be somewhat too conservative for future use. Rather than devising a new set of general guidelines to help determine when an elective operation is indicated, it seems more appropriate to use the equation we have developed, which incorporates information about various important risk factors for rupture, to calculate the risk of rupture over a 1-year interval in each patient, and then to recommend an elective operation only if the calculated risk of rupture exceeds the anticipated operative mortality of elective aneurysm resection for that patient. By carefully balancing in this way the risk of an elective operation for descending thoracic and thoracoabdominal aneurysms against the risk of rupture, as depicted in Figure 6, we may now be in a position to advise patients more rationally, and to recommend operation more confidently to those patients who are likely otherwise to succumb to rupture of their aneurysms.
|
This section shows the derivation of the 95% confidence limits for P(t), where P(t) = 1 - exp(-t) is the probability of rupture within t days of a CT scan, based on k independent factors, X1, X2, ..., Xk, with = exp[ + ß1X1 + ...ßkXk]. Let X = (1,X1,X2, ..., Xk), and b be the corresponding (k + 1) vector of estimated coefficients. The confidence limits for P(t) are:
 |
To calculate the confidence interval, is estimated by exp(X` b), and V by X` CovX, where Cov is the estimated covariance matrix of b. The results are illustrated in Appendix Table 1 for a patient aged 65 years with pain but no COPD, abdominal diameter of 5 cm, and increasing sizes of descending diameter.
|
Acknowledgments
We thank Christine Christopher, RN, BS, Elaine Kaplan, RN, BS, Peter E. Seissler, BES, and Donna Smith-Jordon, BS, for invaluable help in obtaining accurate follow-up data. We are also grateful to the Department of Surgery, Oulu University Hospital, University of Oulu, Finland's Academy of Science, and Finland's Society of Angiology for their support of Dr Juvonen.
Footnotes
Presented at the Thirty-third Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Feb 3–5, 1997.
Address reprint requests to Dr Griepp, Department of Cardiothoracic Surgery, Mount Sinai Medical Center, One Gustave L. Levy Place, New York, NY 10029.
References
This article has been cited by other articles:
|
|
 |
|
  S. Trimarchi, F. H. W. Jonker, S. Hutchison, E. M. Isselbacher, L. A. Pape, H. J. Patel, J. B. Froehlich, B. E. Muhs, V. Rampoldi, V. Grassi, et al. Descending aortic diameter of 5.5 cm or greater is not an accurate predictor of acute type B aortic dissection J. Thorac. Cardiovasc. Surg., September 1, 2011; 142(3): e101 - e107. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Schepens, P. A. Verrelst, W. Ranschaert, E. Graulus, and D. J. Meester Indications for thoracoabdominal aortic surgery J. Thorac. Cardiovasc. Surg., December 1, 2010; 140(6_suppl): S121 - S124. [Full Text] [PDF]
|
|
|
|
 |
|
 American College of Cardiology Foundation, American Heart Association Task Force on Practice, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interve, Society of Interventional Radiology, Society of Thoracic Surgeons, Society for Vascular Medicine, et al. 2010 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM Guidelines for the Diagnosis and Management of Patients With Thoracic Aortic Disease J. Am. Coll. Cardiol., April 6, 2010; 55(14): e27 - e129. [Full Text] [PDF]
|
|
|
|
 |
|
 WRITING GROUP MEMBERS, L. F. Hiratzka, G. L. Bakris, J. A. Beckman, R. M. Bersin, V. F. Carr, D. E. Casey Jr, K. A. Eagle, L. K. Hermann, E. M. Isselbacher, et al. 2010 ACCF/AHA/AATS/ACR/ASA/SCA/SCAI/SIR/STS/SVM Guidelines for the Diagnosis and Management of Patients With Thoracic Aortic Disease: A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, American Association for Thoracic Surgery, American College of Radiology, American Stroke Association, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of Thoracic Surgeons, and Society for Vascular Medicine Circulation, April 6, 2010; 121(13): e266 - e369. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Pochettino, W. T. Brinkman, P. Moeller, W. Y. Szeto, W. Moser, K. Cornelius, F. W. Bowen, Y. J. Woo, and J. E. Bavaria Antegrade Thoracic Stent Grafting During Repair of Acute DeBakey I Dissection Prevents Development of Thoracoabdominal Aortic Aneurysms Ann. Thorac. Surg., August 1, 2009; 88(2): 482 - 490. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Iliopoulos, E. P. Kritharis, A. T. Giagini, S. A. Papadodima, and D. P. Sokolis Ascending thoracic aortic aneurysms are associated with compositional remodeling and vessel stiffening but not weakening in age-matched subjects. J. Thorac. Cardiovasc. Surg., January 1, 2009; 137(1): 101 - 109. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Nienaber, I. Akin, R. Erbel, and A. Haverich CHAPTER 31 Diseases of the Aorta and Trauma to the Aorta and the Heart ESC Textbook of Cardiovascular Medicine, January 1, 2009; 2(1): med-9780199566990-chapter - med-9780199566990-chapter. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-P. Chang, J.-C. Liu, Y.-M. Liou, S.-S. Chang, and J.-Y. Chen The Role of False Lumen Size in Prediction of In-Hospital Complications After Acute Type B Aortic Dissection J. Am. Coll. Cardiol., September 30, 2008; 52(14): 1170 - 1176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Patel and G. M. Deeb Ascending and Arch Aorta: Pathology, Natural History, and Treatment Circulation, July 8, 2008; 118(2): 188 - 195. [Full Text] [PDF]
|
|
|
|
|
|
M. F. Conrad and R. P. Cambria Contemporary Management of Descending Thoracic and Thoracoabdominal Aortic Aneurysms: Endovascular Versus Open Circulation, February 12, 2008; 117(6): 841 - 852. [Full Text] [PDF]
|
|
|
|
|
|
L. G. Svensson, N. T. Kouchoukos, D. C. Miller, J. E. Bavaria, J. S. Coselli, M. A. Curi, H. Eggebrecht, J. A. Elefteriades, R. Erbel, T. G. Gleason, et al. Expert Consensus Document on the Treatment of Descending Thoracic Aortic Disease Using Endovascular Stent-Grafts Ann. Thorac. Surg., January 1, 2008; 85(1_Supplement): S1 - S41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Kahn, M. E. Stone, and D. M. Moskowitz Anesthetic Consideration for Descending Thoracic Aortic Aneurysm Repair Seminars in Cardiothoracic and Vascular Anesthesia, September 1, 2007; 11(3): 205 - 223. [Abstract] [PDF]
|
|
|
|
|
|
A. Zierer, R. K. Voeller, K. E. Hill, N. T. Kouchoukos, R. J. Damiano Jr, and M. R. Moon Aortic Enlargement and Late Reoperation After Repair of Acute Type A Aortic Dissection Ann. Thorac. Surg., August 1, 2007; 84(2): 479 - 487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. G. Gleason and L. C. Benjamin Conventional Open Repair of Descending Thoracic Aortic Aneurysms Perspectives in Vascular Surgery and Endovascular Therapy, June 1, 2007; 19(2): 110 - 121. [Abstract] [PDF]
|
|
|
|
|
|
J. C. Halstead, M. Meier, C. Etz, D. Spielvogel, C. Bodian, M. Wurm, R. Shahani, and R. B. Griepp The fate of the distal aorta after repair of acute type A aortic dissection J. Thorac. Cardiovasc. Surg., January 1, 2007; 133(1): 127 - 135. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Zierer, S. J. Melby, J. G. Lubahn, G. A. Sicard, R. J. Damiano Jr, and M. R. Moon Elective Surgery for Thoracic Aortic Aneurysms: Late Functional Status and Quality of Life Ann. Thorac. Surg., August 1, 2006; 82(2): 573 - 578. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Fattori, C. A. Nienaber, H. Rousseau, J.-P. Beregi, R. Heijmen, M. Grabenwoger, P. Piquet, L. Lovato, C. Dabbech, S. Kische, et al. Results of endovascular repair of the thoracic aorta with the Talent Thoracic stent graft: The Talent Thoracic Retrospective Registry. J. Thorac. Cardiovasc. Surg., August 1, 2006; 132(2): 332 - 339. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Sueyoshi, I. Sakamoto, M. Uetani, and Y. Matsuoka CT Analysis of the Growth Rate of Aortic Diameter Affected by Acute Type B Intramural Hematoma Am. J. Roentgenol., June 1, 2006; 186(6_Supplement_2): S414 - S420. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. R. Davies, A. Gallo, M. A. Coady, G. Tellides, D. M. Botta, B. Burke, M. P. Coe, G. S. Kopf, and J. A. Elefteriades Novel Measurement of Relative Aortic Size Predicts Rupture of Thoracic Aortic Aneurysms Ann. Thorac. Surg., January 1, 2006; 81(1): 169 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Panos, A. Kalangos, P. Christofilopoulos, and G. Khatchatourian Combined Surgical and Endovascular Treatment of Aortic Type A Dissection Ann. Thorac. Surg., September 1, 2005; 80(3): 1087 - 1090. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. F. Immer, E. Krahenbuhl, U. Hagen, M. Stalder, P. A. Berdat, F. S. Eckstein, J. Schmidli, and T. P. Carrel Large Area of the False Lumen Favors Secondary Dilatation of the Aorta After Acute Type A Aortic Dissection Circulation, August 30, 2005; 112(9_suppl): I-249 - I-252. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. F. Immer, U. Hagen, P. A. Berdat, F. S. Eckstein, and T. P. Carrel Risk factors for secondary dilatation of the aorta after acute type A aortic dissection Eur J Cardiothorac Surg, April 1, 2005; 27(4): 654 - 657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kirali, H. Ardal, V. Erentug, D. Mansuroglu, N. U Bozbuga, and C. Yakut Surgical Outcome of Subtypes of Aortic Arch Dissection Asian Cardiovasc Thorac Ann, December 1, 2004; 12(4): 300 - 305. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Kaschina, M. Stoll, M. Sommerfeld, U. M. Steckelings, R. Kreutz, and T. Unger Genetic kininogen deficiency contributes to aortic aneurysm formation but not to atherosclerosis Physiol Genomics, September 16, 2004; 19(1): 41 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. Sueyoshi, I. Sakamoto, K. Hayashi, T. Yamaguchi, and T. Imada Growth Rate of Aortic Diameter in Patients With Type B Aortic Dissection During the Chronic Phase Circulation, September 14, 2004; 110(11_suppl_1): II-256 - II-261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Demers, D. C. Miller, R. S. Mitchell, S. T. Kee, D. Sze, M. K. Razavi, and M. D. Dake Midterm results of endovascular repair of descending thoracic aortic aneurysms with first-generation stent grafts J. Thorac. Cardiovasc. Surg., March 1, 2004; 127(3): 664 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ueda, H. Shimizu, K. Hashizume, K. Koizumi, M. Mori, H. Shin, and R. Yozu Mortality and morbidity after total arch replacement using a branched arch graft with selective antegrade cerebral perfusion Ann. Thorac. Surg., December 1, 2003; 76(6): 1951 - 1956. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. P. Ehrlich, M. Schillinger, M. Grabenwoger, A. Kocher, E. M. Tschernko, P. Simon, A. Bohdjalian, and E. Wolner Predictors of Adverse Outcome and Transient Neurological Dysfuntion Following Surgical Treatment of Acute Type A Dissections Circulation, September 9, 2003; 108(2011): II-318 - II-323. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Okamoto, H. Xu, N. T. Kouchoukos, M. R. Moon, and T. M. Sundt III The influence of mechanical properties on wall stress and distensibility of the dilated ascending aorta J. Thorac. Cardiovasc. Surg., September 1, 2003; 126(3): 842 - 850. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Coselli and P. L. Moreno Descending and Thoracoabdominal Aneurysm Card. Surg. Adult, January 1, 2003; 2(2003): 1169 - 1190. [Full Text]
|
|
|
|
|
|
T. Fleck, C. Hamilton, M. P. Ehrlich, D. Hutschala, H. Koinig, E. Wolner, and M. Grabenwoger Thoracoabdominal Aortic Aneurysm Repair: Reducing Adverse Outcome with Left Heart Bypass, Selective Visceral Perfusion and Renal Protection Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2002; 6(4): 287 - 291. [Abstract] [PDF]
|
|
|
|
|
|
J. S. Coselli, L. D. Conklin, and S. A. LeMaire Thoracoabdominal aortic aneurysm repair: review and update of current strategies Ann. Thorac. Surg., November 1, 2002; 74(5): S1881 - S1884. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. N. Girardi, K. H. Krieger, N. K. Altorki, C. A. Mack, L. Y. Lee, and O. W. Isom Ruptured descending and thoracoabdominal aortic aneurysms Ann. Thorac. Surg., October 1, 2002; 74(4): 1066 - 1070. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Fleck, D. Hutschala, M. Czerny, M. P. Ehrlich, M.-T. Kasimir, M. Cejna, E. Wolner, and M. Grabenwoger Combined Surgical and Endovascular Treatment of Acute Aortic Dissection Type A: Preliminary Results Ann. Thorac. Surg., September 1, 2002; 74(3): 761 - 765. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. L. Moura, G. Melissano, F. Maisano, E. M. Marone, E. Civilini, O. Alfieri, and R. Chiesa Combined Endovascular Treatment of a Descending Thoracic Aortic Aneurysm and Off-Pump Myocardial Revascularization: A Case Report Vascular and Endovascular Surgery, July 1, 2002; 36(4): 305 - 309. [Abstract] [PDF]
|
|
|
|
|
|
R. R. Davies, L. J. Goldstein, M. A. Coady, S. L. Tittle, J. A. Rizzo, G. S. Kopf, and J. A. Elefteriades Yearly rupture or dissection rates for thoracic aortic aneurysms: simple prediction based on size Ann. Thorac. Surg., January 1, 2002; 73(1): 17 - 28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. LeMaire, C. C. Miller III, L. D. Conklin, Z. C. Schmittling, C. Koksoy, and J. S. Coselli A new predictive model for adverse outcomes after elective thoracoabdominal aortic aneurysm repair Ann. Thorac. Surg., April 1, 2001; 71(4): 1233 - 1238. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Treasure SURGERY: Cardiovascular surgery for Marfan syndrome Heart, December 1, 2000; 84(6): 674 - 678. [Full Text]
|
|
|
|
|
|
R S Bonser, D Pagano, M E Lewis, S J Rooney, P Guest, P Davies, and I Shimada Clinical and patho-anatomical factors affecting expansion of thoracic aortic aneurysms Heart, September 1, 2000; 84(3): 277 - 283. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. S. Coselli, S. A. LeMaire, C. C. Miller III, Z. C. Schmittling, C. Koksoy, J. Pagan, and P. E. Curling Mortality and paraplegia after thoracoabdominal aortic aneurysm repair: a risk factor analysis Ann. Thorac. Surg., February 1, 2000; 69(2): 409 - 414. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. P. Cambria Stent Graft Repair of Thoracic Aortic Pathology Perspectives in Vascular Surgery and Endovascular Therapy, January 1, 2000; 13(2): 1 - 13. [Abstract] [PDF]
|
|
|
|
 |
|
 Y. Kitami, Y. Takata, T. Fukuoka, J. Arimitsu, K. Hiwada, and T. Okura Giant Unruptured Aneurysm of the Thoracic Aorta: A Case Report Angiology, October 1, 1999; 50(10): 865 - 869. [Abstract] [PDF]
|
|
|
|
|
|
I. Shimada, S. J. Rooney, P. A. Farneti, P. Riley, P. Guest, P. Davies, and R. S. Bonser Reproducibility of thoracic aortic diameter measurement using computed tomographic scans Eur J Cardiothorac Surg, July 1, 1999; 16(1): 59 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Shimada, S. J. Rooney, D. Pagano, P. A. Farneti, P. Davies, P. J. Guest, and R. S. Bonser Prediction of thoracic aortic aneurysm expansion: validation of formulae describing growth Ann. Thorac. Surg., June 1, 1999; 67(6): 1968 - 1970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Juvonen, M. A. Ergin, J. D. Galla, S. L. Lansman, J. N. McCullough, K. Nguyen, C. A. Bodian, M. P. Ehrlich, D. Spielvogel, J. J. Klein, et al. RISK FACTORS FOR RUPTURE OF CHRONIC TYPE B DISSECTIONS J. Thorac. Cardiovasc. Surg., April 1, 1999; 117(4): 776 - 786. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Katsumata, A. Shinfeld, and S. Westaby Operation for chronic traumatic aortic aneurysm: when and how? Ann. Thorac. Surg., September 1, 1998; 66(3): 774 - 778. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
