Correspondence
Abdominal Aortic Aneurysm
N Engl J Med 1993; 329:1275-1277October 21, 1993
- Article
To the Editor:
Ernst (April 22 issue)1 provided a comprehensive review of the complex issue of the effectiveness of elective aortic reconstruction for abdominal aortic aneurysms. His presentation, however, was scientifically selective in favor of elective surgery. No physician is in a position to inform patients that they “will require” an operation (particularly an elective one). It is important that physicians make an unbiased presentation of the benefits and risks of this procedure (Table 1Table 1
Examples of Referral Bias in the Literature Pertaining to the Risks and Benefits of Elective Surgery for Abdominal Aortic Aneurysm.).Ernst refers to population-based studies suggesting annual increases in the diameters of aneurysms that were approximately half those reported in referral-center case series (0.2 cm vs. 0.4 cm). He did not indicate which of these estimates may be more valid in the context of clinical decision-making. With respect to the risk of rupture, several population-based studies published since 19893,4 support the findings regarding the natural history of aneurysms reported by the Rochester Epidemiology Project5 and the Oxford Regional Health Authority6.
In reviewing the results of studies of aortic reconstruction, Ernst included only those with at least 200 patients undergoing repair of nonruptured aneurysms -- an arbitrary criterion. This excludes all the population-based studies in the United States, including one from the Rochester Epidemiology Project reporting a 30-day mortality of 4.6 percent among 131 Olmsted County residents undergoing elective aneurysm surgery5. In a clinical setting with the same operating rooms, anesthesiologists, and surgeons, unselected community patients had a nearly twofold greater risk of death than patients in the two Mayo Clinic referral-center case series reported by Ernst in Table 1 of his article. The Olmsted County results of surgery for nonruptured abdominal aortic aneurysm are similar to the results from the prospective Canadian national registry data cited in Table 1 by Ernst.
The review article does not present a balanced picture of the subjective methods that have been used to evaluate indications for aneurysm surgery. Although Ernst acknowledged a report by three vascular surgeons recommending repair of aneurysms at least 4 cm in diameter, or twice the diameter of the normal infrarenal aorta, planning groups in Canada, the United Kingdom, and the United States concluded that a randomized trial was necessary to resolve scientific uncertainty regarding the relative benefits and risks of elective resection of aneurysms measuring 4.0 to 5.4 cm in diameter6. Similarly, a multispecialty panel2 has been less enthusiastic about the effectiveness of elective aneurysm surgery than the three vascular surgeons referred to by Ernst.
6 ReferencesDavid J. Ballard, M.D., Ph.D.
Thomas Jefferson Health Policy Institute, Charlottesville, VA 229011
Ernst CB . Abdominal aortic aneurysm. N Engl J Med 1993;328:1167-1172
Full Text | Web of Science | Medline2
Ballard DJ, Etchason JA, Hilborne LH, et al. Abdominal aortic aneurysm surgery: a literature review and ratings of appropriateness and necessity. Santa Monica, Calif.: RAND, 1992.
3
Johansson G ,Nydahl S ,Olofsson P ,Swedenborg J . Survival in patients with abdominal aortic aneurysms: comparison between operative and nonoperative management. Eur J Vasc Surg 1990;4:497-502
CrossRef | Medline4
Glimaker H ,Holmberg L ,Elvin A , et al. Natural history of patients with abdominal aortic aneurysm. Eur J Vasc Surg 1991;5:125-130
CrossRef | Medline5
Roger VL ,Ballard DJ ,Hallett JW Jr ,Osmundson PJ ,Puetz PA ,Gersh BJ . Influence of coronary artery disease on morbidity and mortality after abdominal aortic aneurysmectomy: a population-based study, 1971-1987. J Am Coll Cardiol 1989;14:1245-1252
CrossRef | Web of Science | Medline6
Report of a meeting of physicians and scientists, University College London Medical School: abdominal aortic aneurysm
CrossRef | Web of Science | Medline
To the Editor:
Ernst states that “a recent population-based study reporting a very low rupture rate for small aneurysms has been questioned.” Although crossovers to elective surgery will contaminate even the three ongoing randomized trials of aneurysm surgery, none of the Rochester residents who underwent repair of an aneurysm less than 5 cm in diameter between 1971 and 1987 had symptoms, and those who underwent elective aneurysm repair did not have a more rapid rate of change in the size of their aneurysms than untreated patients. The observation that five patients with aneurysms diagnosed before 1984 had ruptured aneurysms at the initial ultrasound examination is not surprising, given that many other studies have reported that some patients initially present with rupture. To estimate the risk of rupture in a valid way for the 176 patients who presented with an unruptured aneurysm, it is necessary to exclude the five prevalent cases of rupture from the incident-cohort analysis1.
Table 1 and Table 2 in the article by Ernst do not make clear the length of the period during which patients are at risk for death after surgery; in fact, some of the “mortality” values pertain to deaths occurring 30 days after surgery, whereas others are confined to in-hospital deaths, and still others combine deaths occurring during hospitalization with those occurring 30 days postoperatively. Table 3 mixes studies that include operative deaths in the estimates of “late survival” with studies that omit operative deaths. The presentation of the data of Crawford et al.2 is particularly misleading. The 5-year and 10-year late-survival rates reported by Crawford et al. include perioperative deaths; the 1-year survival rate shown in Table 3 was not reported by Crawford et al., and the survival rate of 95 percent at 1 year ascribed to the Crawford case series by Ernst is not credible in view of the 4.8 percent rate of operative mortality in the Crawford series reported in Table 1.
Finally, the article does not describe the methods used to generate the data in what is referred to as “this study” and cites no peer-reviewed publication of the data from “this study.”
2 ReferencesMartin P. Nevitt, M.D., M.P.H.
Mayo Clinic, Rochester, MN 559051
Nevitt MP ,Ballard DJ ,Hallett JW Jr . Prognosis of abdominal aortic aneurysms: a population-based study. N Engl J Med 1989;321:1009-1014
Full Text | Web of Science | Medline2
Crawford ES ,Saleh SA ,Babb JW III ,Glaeser DH ,Vaccaro PS ,Silvers A . Infrarenal abdominal aortic aneurysm: factors influencing survival after operation performed over a 25-year period. Ann Surg 1981;193:699-709
CrossRef | Web of Science | Medline
To the Editor:
We would like to add cigarette smoking to Dr. Ernst's list of risk factors for abdominal aortic aneurysm. An association between cigarette smoking and abdominal aortic aneurysm has been shown in a number of studies1-4. Heavy smokers have as much as an eightfold increase in the incidence of abdominal aortic aneurysm1,4. Approximately one in four adults in the United States continues to smoke. We therefore believe it prudent to include smokers (55 years of age and older) among the persons for whom screening for abdominal aortic aneurysm is considered appropriate.
4 ReferencesAlwin F. Steinmann, M.D.
William Caramore, M.D.
Albany Medical College, Albany, NY 122081
Auerbach O ,Garfinkel L . Atherosclerosis and aneurysm of aorta in relation to smoking habits and age. Chest 1980;78:805-809
CrossRef | Web of Science | Medline2
Hammond EC ,Garfinkel L . Coronary heart disease, stroke, and aortic aneurysm: factors in the etiology. Arch Environ Health 1969;19:167-182
Medline3
Lederle FA ,Walker JM ,Reinke DB . Selective screening for abdominal aortic aneurysms with physical examination and ultrasound. Arch Intern Med 1988;148:1753-1756
CrossRef | Web of Science | Medline4
Krohn CD ,Kullmann G ,Kvernebo K ,Rosen L ,Kroese A . Ultrasonographic screening for abdominal aortic aneurysm. Eur J Surg 1992;158:527-530
Medline
To the Editor:
In his review Dr. Ernst cites a paper suggesting that octogenarians should perhaps be excluded from the list of patients eligible for repair of ruptured abdominal aortic aneurysm1. The life span of all 85-year-olds, including those who are very ill, is 5.7 years, and many octogenarians are in better physiologic condition than some 60-year-olds2. This is a common dilemma for physicians making decisions about aneurysm repair. Is there not a risk of undertreatment on the basis of age?
2 ReferencesJeremy Walston, M.D.
Thomas Finucane, M.D.
Francis Scott Key Medical Center, Baltimore, MD 212241
Johansen K ,Kohler TR ,Nicholls SC ,Zierler RE ,Clowes AW ,Kazmers A . Ruptured abdominal aortic aneurysm: the Harborview experience. J Vasc Surg 1991;13:240-247
CrossRef | Web of Science | Medline2
Katz S ,Branch LG ,Branson MH ,Papsidero JA ,Beck JC ,Greer DS . Active life expectancy. N Engl J Med 1983;309:1218-1224
Full Text | Web of Science | Medline
Author/Editor Response
Dr. Ernst replies:
To the Editor: Ballard's comments reflect a certain misinterpretation of the review. The statement that it is inappropriate to inform patients that they “will require” an operation was taken out of context. No patient should be told that an operation is required. Such critical decisions are mutually made by the physician and the patient, and only after fully informed consent has been given. With respect to the size of an aneurysm and the likelihood of rupture, the studies to which Ballard refers were cited in the review. Although the expansion rates of 0.2 to 0.4 cm a year should be cause for concern, the absolute size of an aneurysm and the estimate of the risk entailed by either watchful waiting or operation are the critical factors in clinical decision making. Admittedly, the reports cited were chosen arbitrarily from those that provided the best available contemporary surgical results at experienced centers.
From his experience with the Rand study on abdominal-aortic-aneurysm surgery, Ballard is well aware of the vicissitudes of reviewing a variable literature to formulate treatment recommendations. Since the best results correlate with increased experience, only studies recording at least 200 elective aneurysm repairs were selected. Nevertheless, the combined 4 percent mortality rate after elective repair referred to in the review was not significantly different from the 4.6 percent mortality rate in the study that Ballard cites. His comments are appreciated since they support my plea for prospective, randomized trials to resolve the uncertainty regarding the management of small abdominal aortic aneurysms. To this end, the three ongoing studies to which he refers hold promise, and the results are eagerly awaited.
Nevitt's letter is in part an attempt to validate his report. His study, however flawed, was a conscientious attempt to define the risk of rupture of small aneurysms. However, the use of a “crossover” rate of approximately 25 percent and the exclusion of five additional patients make conclusions regarding prognosis suspect. Very few studies with such high crossover rates provide conclusive information for clinical decision making. Some of the 55 aneurysms that were less than 3.5 cm in diameter may not have met the definition of an aneurysm. The tables in my article were constructed from a variety of retrospective referral-based reports. They provide a realistic depiction of contemporary surgical results. Furthermore, the majority of the reports cited included “30-day” mortality or “operative mortality,” synonymous terms in the surgical lexicon. The one-year survival rate reported for the study by Crawford et al. refers to the 95 percent survival rate after elective operation and does not discredit the purpose of the table, which was to document late survival after aneurysm repair.
It should be obvious that the data designated as from “this study” are from the Henry Ford Hospital. The data base is the 35-year-old Vascular Surgery Registry, which is continually updated. Similar registry data have been cited in numerous peer-reviewed reports in the past, and no justification is required for their inclusion in a review article.
I agree with the views of Steinmann and Caramore and Walston and Finucane. If there are no compelling contraindications as a result of coexisting illnesses, elective aneurysm repair should not be denied on the basis of age alone, because emergency aneurysm repair in octogenarians is usually fatal.
Calvin B. Ernst, M.D.
University of Michigan Medical School, Ann Arbor, MI 48109- Citing Articles (2)
Citing Articles
1
(1997) Infrarenal Aortic Aneurysms. New England Journal of Medicine 336:24, 1756-1758
Full Text2
JAMES P. KAHAN, ROLLA EDWARD PARK, LUCIAN L. LEAPE, STEVEN J. BERNSTEIN, LEE H. HILBORNE, LORI PARKER, CAREN J. KAMBERG, DAVID J. BALLARD, ROBERT H. BROOK. (1996) Variations by Specialty in Physician Ratings of the Appropriateness and Necessity of Indications for Procedures. Medical Care 34:6, 512-523
CrossRef
