Yield of Repeated Screening for Abdominal Aortic Aneurysm After a 4-Year Interval FREE

ULTRASOUND SCREENING in older men to detect abdominal aortic aneurysm (AAA) has been a topic of considerable discussion in the medical literature. The only published randomized trial of ultrasound screening included 15 775 subjects and reported a substantial but statistically nonsignificant 41% reduction in AAA-related deaths in men.¹ Similar trials are under way in Denmark² and Western Australia.³ Most published cost-effectiveness analyses have judged ultrasound screening for AAA in men older than 60 years to be beneficial and associated with costs per year of life saved that are acceptable by current standards.⁴ The Vascular Surgery Society of the United Kingdom has recommended widespread screening of older men,⁵ whereas the Canadian and US Preventive Services Task Forces considered the evidence insufficient to make a recommendation,^6- 7 but plan to revisit the topic when more trial data are available (Michael Pignone, MD, written communication, 1998).

A number of public AAA screening programs have been established, particularly in England. The optimal design of an AAA screening program remains undetermined, however, including whether screening should be performed once in a lifetime, as proposed by some authors,^5,8 or repeated at intervals, as suggested by others.^9- 10 Which strategy is preferable depends largely on the yield of a second screening, and little information is available on the rate at which new AAA develop. The purpose of our study was to determine the yield of a second ultrasound screening for AAA in a population typical of those targeted by screening programs.

A subgroup of subjects who underwent initial screening for AAA with ultrasound as part of Veterans Affairs (VA) Cooperative Study No. 379, the Aneurysm Detection and Management Study, was identified for invitation to a second screening after an interval of 4 years. Details of the initial screening program have been reported previously.¹¹ In brief, all active patients aged 50 through 79 years at participating VA medical centers were invited to have an ultrasound examination to detect AAA, preceded by a questionnaire addressing demographic information and possible risk factors for AAA. Abdominal aortic aneurysm was defined as infrarenal aortic diameter of 3.0 cm or larger.

Thirteen VA medical centers participated in the second screening study, and these centers performed initial screening for 16 643 subjects from December 1, 1992, through June 30, 1993. Of these, 15 098 had infrarenal and suprarenal aortic diameter of 3.0 cm or less on the initial ultrasound examination and no previous history of AAA. From this latter group, 5151 were selected for a second screening using computer-generated random numbers. These subjects were invited to undergo ultrasound examinations from December 1, 1996, through June 30, 1997. The second ultrasound was thus scheduled to fall within 6 months of the 4-year anniversary of the original ultrasound and as close to that date as possible.

Local VA records and national VA databases (the Beneficiary Identification and Records Locator Subsystem and the Patient Treatment and Outpatient Record files) were searched to identify deaths and AAA diagnoses made during the study interval in the randomly selected subgroup. Cause of death was obtained from death certificate information through the National Death Index, which is reported to have a sensitivity of 97% in veterans.¹²

Interim diagnoses of AAA were considered separately from AAA detected at rescreening regardless of whether the subject attended the rescreening ultrasound examination. This was because subjects cannot undergo screening for a condition that is already diagnosed, and the detection of an interim AAA may have influenced the subject's decision to attend the rescreening. We considered the appropriate denominator for all subjects with interim diagnoses of AAA to be the entire randomly selected subset of 5151 subjects rather than the smaller group that actually underwent rescreening.

Analyses to determine associations between the items on the questionnaire and the presence of AAA were performed using univariable and multivariable logistic regression, details of which have been described previously.¹¹

Of the 5151 subjects selected for a second screening, 598 (11.6%) had died, 20 (0.4%) had an interim diagnosis of AAA, 2622 (50.9%) underwent a second screening, and the remaining 1911 (37.1%) were alive but did not undergo a second screening. Characteristics of these groups are shown in Table 1. Subjects who were alive but did not undergo a second screening were slightly younger, less likely to be male or white, more likely to be current smokers, and more likely to have peripheral vascular disease than those who attended the second screening.

Subjects who died during the 4-year interval were older, smoked more, and had a higher rate of reported diseases than those who attended the second screening. Of the 598 deceased subjects, matches to National Death Index death records were obtained for 558 (93.3%), and none of these listed aortic aneurysm as a cause of death.

Of the 20 interim AAAs, one was measured to be 4.1 cm only 5 months after a diameter of 1.6 cm was recorded at screening, presumably representing an AAA missed at screening. This AAA enlarged to 5.5 cm and was repaired near the end of the 4-year study interval. None of the remaining 19 interim AAA strongly suggested false-negative findings on the initial screen. One AAA was 4.8 cm (first diagnosed at 4.1 cm 2 years after an initial screening measurement of 2.4 cm); 5 AAAs were 3.5 to 3.9 cm; and 13 AAAs were 3.0 to 3.4 cm when last measured during the 4-year period. Apart from the 2 AAAs larger than 4.0 cm, all other interim AAAs increased by less than 1.5 cm in diameter from the initial screening measurement, and 4 increased by less than 0.5 cm. Subjects with interim AAA were older and had smoked longer than the subjects undergoing rescreening who did not have AAA (Table 1).

Of the 2622 subjects who underwent a second screening, 99% did so within 6 months of the 4-year anniversary of their original ultrasound. The mean infrarenal aortic diameter measurement of the entire second screening group decreased slightly between the initial and second ultrasound measurements (−0.07 cm; 95% confidence interval [CI], −0.47-0.33 cm), presumably the result of random measurement variation. Of the subjects undergoing rescreening, 58 (2.2%; 95% CI, 1.7%-2.8%) had new AAAs, of which 3 were 4.0 to 4.9 cm, 10 were 3.5 to 3.9 cm, and 45 were 3.0 to 3.4 cm. In 3 of these latter 45, the infrarenal aortic diameter had enlarged by less than 0.5 cm since the initial ultrasound. Two subjects had an increase of 2.0 cm or greater in infrarenal aortic diameter; one, of 2.6 cm (from 2.3 to 4.9 cm); and the other, of 2.4 cm (from 1.2 to 3.6 cm). Four additional subjects had an increase of 1.5 cm or greater.

Subjects in the second screening group in whom new AAA developed had a mean infrarenal aortic diameter on the initial ultrasound of 2.3 cm, compared with a mean of 2.0 cm in subjects without new AAA (P<.001), and 29% (17 subjects) had an infrarenal aortic diameter of 2.5 cm or greater on the initial ultrasound, compared with 7% of the subjects without a new AAA (P<.001).

We compared characteristics of the 58 subjects in whom new AAAs were detected with those of the subjects who underwent rescreening and did not have new AAAs (Table 1). In a univariable logistic model for predicting new AAAs using the characteristics shown in Table 1, the following variables were significant: ever smoked (odds ratio [OR], 2.20; 95% CI, 1.04-4.66), number of years smoked (OR, 1.26 per 10 years; 95% CI, 1.08-1.47), current smoker at the time of the initial screening (OR, 3.31; 95% CI, 1.92-5.72), and coronary artery disease (OR, 1.73; 95% CI, 1.03-2.91). When the regression model was reanalyzed with coronary artery disease, cerebral vascular disease, and claudication combined into a single variable of "any atherosclerosis," this variable was also significant (OR, 1.93; 95% CI, 1.14-3.28). In a forward stepwise multivariable logistic model using the significant factors from the univariable model, 2 variables remained significant: current smoker (OR, 3.09; 95% CI, 1.74-5.50) and coronary artery disease (OR, 1.81; 95% CI, 1.07-3.07). In the model using any atherosclerosis, this variable remained significant (OR, 1.97; 95% CI, 1.16-3.35).

Assuming that the likelihood of interim diagnosis was similar in subjects who did and did not undergo a second screening, then the 4-year incidence of AAA 3.0 cm or greater can be estimated by adding the interim diagnosis rate (0.4%) to the rescreening diagnosis rate (2.2%) to obtain 2.6%. The true incidence could be higher if undetected AAAs developed in some subjects, who then died of some other cause during the 4-year interval.

Our results provide information on the yield of a second ultrasound after 4 years to detect AAA in a high-risk population typical of those targeted for screening. Age, male sex, and a history of smoking are the principal risk factors for AAA,¹¹ and also characterize our study population. The prevalence of AAA of 3.0 cm or greater in the original screening of this population was 4.6%.¹¹ In our study, the 4-year incidence of AAA of 3.0 cm or greater was estimated to be 2.6%. This incidence is more than half the original prevalence, suggesting a mean disease duration of about 8 years, which is consistent with our knowledge of the natural history of AAA. The rescreening diagnosis rate of 2.2% also implies that a screening interval of 8 years or more may provide yields comparable to initial (prevalence) screening. Smoking at the time of the original screening ultrasound and atherosclerotic disease were independent predictors of new AAA at the second screening.

Aneurysms of larger than 4.0 cm were uncommon at the second screening, as expected based on published mean AAA enlargement rates of 0.20 to 0.33 cm/y.^13- 14 Some smaller AAAs detected at rescreening in our study could represent only measurement variation. Ultrasound measurement variation has been reported to be less than 0.4 cm in 95% of cases.^15- 16 Only 3 of the 58 AAAs diagnosed at the second screening in our study had an increase in diameter of less than 0.5 cm since the initial ultrasound.

We are aware of only 1 previous study of a second screening for AAA, in which 2 (1.1%) of 189 men (95% CI, 0%-2.6%) aged 70 through 71 years had an AAA of 3.0 cm or larger 5 years after aortic diameter was measured to be normal (defined as <2.6 cm).¹⁷ The higher yield observed in our study is within the CIs of the yield of that study and also reflects our inclusion of subjects with initial aortic diameters of 2.6 to 2.9 cm. Of the 58 subjects with AAA diagnosed in our rescreening group, 16 had initial aortic diameters of 2.6 to 2.9 cm. The 4-year incidence of AAA in our study in the 2495 subjects in the rescreening group with initial diameters of less than 2.6 cm was 1.7% (95% CI, 1.4%-1.9%).

Our findings indicate that a second screening is of little practical value after 4 years, mainly because the AAAs detected are small. However, our data also suggest that a second screening after longer intervals (ie, >8 years) may provide total yields similar to those seen in initial screening, although possibly with fewer large AAAs. Rescreening only in subjects with infrarenal aortic diameter of 2.5 cm or greater on the initial ultrasound would have missed more than two thirds of the new AAAs in our series. Further study of the yield of screening after longer intervals is needed to determine an optimal rescreening strategy.

Accepted for publication June 29, 1999.

This study was supported by the Cooperative Studies Program of the Department of Veterans Affairs Office of Research and Development, Washington, DC.

The Aneurysm Detection and Management Veterans Affairs Cooperative Study Investigators also include the following: Gary W. Barone, MD, Department of Surgery, Veterans Affairs Medical Center, Little Rock, Ark; Gregory L. Moneta, MD, Department of Surgery, MD Veterans Affairs Medical Center, Portland, Ore; Raymond G. Makhoul, MD, Department of Surgery, Veterans Affairs Medical Center, Richmond, Va; Gerald B. Zelenock, MD, Department of Surgery, Veterans Affairs Medical Center, Ann Arbor, Mich; Steven Sparks, MD, Department of Surgery, Veterans Affairs Medical Center, San Diego, Calif; Alex Furst, MD, Department of Surgery, Veterans Affairs Medical Center, Miami, Fla; Joseph H. Rapp, MD, Department of Surgery, Veterans Affairs Medical Center, San Francisco, Calif; Robert A. Cambria, MD, Department of Surgery, Veterans Affairs Medical Center, Milwaukee, Wis; Louis M. Messina, MD, Department of Surgery, University of California–San Francisco; David J. Ballard, MD, PhD, Center for Clinical Evaluation Sciences, Emory University, Decatur, Ga; and Donovan B. Reinke, MD, and Howard J. Ansel, MD, Diagnostic Radiology Service, Veterans Affairs Medical Center, Minneapolis, Minn.

Frank A. Lederle, MD (cochairman), Veterans Affairs (VA) Medical Center, Minneapolis, Minn; Samuel E. Wilson, MD (cochairman), University of California–Irvine Medical Center, Orange, Calif; Michele A. Haas (project coordinator) and Jean M. Larson, VA Medical Center, Minneapolis; and Donovan B. Reinke, MD, and Howard J. Ansel, MD, Central Computed Tomographic Scan Laboratory, VA Medical Center, Minneapolis.

Investigators at VA Medical CentersAnn Arbor, Mich: Gerald B. Zelenock, MD, Charles Shanley, MD, Clare Sekerak, RN, Mary Lingg, RN, and Latonya Trohallis; Cleveland, Ohio: Allen W. Averbook, MD, Steven J. Busuttil, MD, Linda Graham, MD, Charles L. Mesh, MD, Krista Kaelin, RN, Theresa O'Malley, RN, Evelynn Wingard, RN, MS, John Lindesmith, RN, and Julie Francosky, RT; Miami, Fla: Alex Furst, MD, Marwan Tabbara, MD, Anselmo A. Nunez, MD, Ghislaine Paperwalla, RN, Kay Nunez, RN, Christine Estep, RN, and James Taylor, RT; Denver, Colo: William C. Krupski, MD, Pamela Strecker, RN, and Linda Schoening, CCVT; Hines, Ill: Fred N. Littooy, MD, John Maggio, PhD, and Wendy Cote, RVT; Little Rock, Ark: Gary W. Barone, MD, Mohammed Moursi, MD, Bernard W. Thompson, MD, Brenda Kackley, RN, Becky Chesser, BSN, CRT, Rob Little, RT, and Terry Wood, BSRT; Long Beach, Calif: Ian Gordon, MD, Cheryl Kohl, RN, Gary Scully, RT, Rebecca Complin, RVT, and Manju Akkinepali; Madison, Wis: Charles W. Acher, MD, Judy Archibald, RN, Jennifer Fitzsimons, BS, and Traci Jo Havlik, RTR; Minneapolis: Edmund P. Chute, MD, William D. Payne, MD, Michael L. Schwartz, MD, Catherine Proebstle, RN, Ronald Hedblad, RT, and Michael Richardson, RT; Pittsburgh, Pa: Michel S. Makaroun, MD, Satish Muluk, MD, Cynthia Slivka, RN, Nina Carnegie, RN, Tricia DeBoo, RN, Janice Freidel, DMS, RT, and Jean Fuhs, RT; Portland, Ore: Gregory L. Moneta, MD, Richard A. Yeager, MD, James M. Edwards, MD, Roland Jemerson, RN, Troy Williams, RVT, and Robert McCartney, RVT; Richmond, Va: Raymond G. Makhoul, MD, Ann Grimsdale, RN, Laura Bartnicki, RN, and Linda Tyree Joyner, RT; San Francisco, Calif: Louis M. Messina, MD, Joseph H. Rapp, MD, Shelley Dwyer, RN, Donna Collins, RN, Steven Vann Smith, CVT, CCPT, and Harry Daughenbaugh, RDCS; San Diego, Calif: Steven Sparks, MD, Robert J. Hye, MD, Edward J. Plecha, MD, Yehuda G. Wolf, MD, Gerry Cali, RN, Terri Scala, RN, and Roger Hull, RCVT; Tampa, Fla: Dennis Bandyk, MD, Dolores Bou-Eid, RN, and Maureen Jackson Maynard, RT; and Milwaukee, Wis: Robert A. Cambria, MD, Julie Freischlag, MD, Christa Kallio, RN, Susan Framberg, RN, Sandra Towne, RN, and Cleveland Weller, RT.

Planning Committee Jack L. Cronenwett, MD (Hanover, NH); David B. Matchar, MD (Durham, NC); and Christopher R.B. Meritt, MD (New Orleans, La).

Executive Committee David J. Ballard, MD, PhD (Decatur, Ga), and C. William Cole, MD (Ottawa, Ontario).

Data Monitoring Board G. Patrick Clagett, MD (cochairman), University of Texas Southwestern Medical Center, Dallas; John D. Corson, MB, ChB (cochairman), The University of Iowa, Iowa City; William C. Cushman, MD, VA Medical Center, Memphis, Tenn; C. Seth Landefeld, MD, VA Medical Center, San Francisco; Theodore G. Karrison, PhD, University of Chicago, Chicago, Ill; and John P. Matts, PhD, University of Minnesota, Minneapolis.

Cooperative Studies Program Coordinating CenterWest Haven, Conn: Gary R. Johnson, MS (study biostatistician); Angela Hudson, MPH, Velma Williams-Estes, Richard Vinisko, Rae Bartozzi, Ray Kilstrom, MBA, and Kathy Riester (statistical assistants); Kathy Newvine and Robert Goodwin, MS (computer programmers); Peter Peduzzi, PhD (director), Dorothea Collins, ScD; Peggy Antonelli (administrative officer); Mary Smith (forms design); Lillie Franklin, Pattie Collins, Stella Marcinauskis, and Bonita Hunter.

Cooperative Studies Program Administration at VA HeadquartersWashington, DC: John R. Feussner, MD, MPH (chief), and Ping C. Huang, PhD.

Reprints: Frank A. Lederle, MD, Minneapolis Veterans Affairs Medical Center, Department of Medicine (III-0), One Veterans Drive, Minneapolis, MN 55417.