Incidence and significance of nonaneurysmal-related computed tomography scan findings in patients undergoing endovascular aortic aneurysm repair
Article Outline
Objective
This study examined the frequency and nature of incidental findings seen on computed tomography (CT) scans during preoperative and postoperative follow-up in patients undergoing endovascular aortic aneurysm repair (EVAR).
Methods
Between January 1, 2000, and March 1, 2006, 176 consecutive patients who underwent EVAR at our institution were retrospectively reviewed. Patients were included in the study if all preoperative and postoperative surveillance CT scans were performed at our institution. Eighty-two patients, 26 women (32%) and 56 men (68%), met this criterion. Their mean age was 76 years (range, 51-103 years). Official CT scan reports were reviewed. Findings were considered primary incidental if they were noted on preoperative CT scans and secondary incidental if they appeared on surveillance CT scans but not on the preoperative study. Primary and secondary incidental findings were considered either benign (eg, gallstones, diverticulosis) or clinically significant if they warranted further workup (eg, suspicious masses or changes suggestive of malignancy, internal or diaphragmatic hernias, and diverticulitis). The median follow-up was 29 months (range, 3-60 months). Each incidental finding was counted only once, on the first scan in which it appeared.
Results
Of the 82 patients, 73 (89%) had at least one primary incidental finding, and 14 (19%) of these were clinically significant. Secondary incidental findings, many of which were clinically significant, continued to appear throughout the follow-up period. The most common clinically significant primary incidental finding was the presence of a lung mass (n = 4). The most common clinically significant secondary incidental findings were lung mass (n = 6), liver mass (n = 6), and pancreas mass (n = 3). There was a significant difference in the proportion of men to women in the group with clinically significant incidental findings vs the group without clinically significant incidental findings (P = .03959). Differences between the groups with respect to age or aneurysm size were not significant.
Conclusion
CT scans yielded surprisingly large numbers of both primary and secondary incidental findings, many of which were clinically significant. Primary incidental findings were more common than secondary incidental findings; however, clinically significant findings were found at a consistent rate throughout the study period.
Patients undergoing endovascular abdominal aortic aneurysm (AAA) repair (EVAR) require lifelong surveillance to detect graft patency, size of the aneurysm sac, endoleak, and graft migration. There are no uniformly accepted guidelines for surveillance after EVAR. In general, most patients undergo a combination of imaging and office visits at postoperative intervals of 1, 6, and 12 months, followed by 6- and 12-month intervals in the second postoperative year and beyond.1
Although computed tomography (CT) scanning continues to be the gold standard in monitoring patients after EVAR, other modalities are being used to lessen patients' lifelong exposure to radiation. Duplex ultrasound (DUS) scanning is comparable to CT scan in the surveillance of EVAR and can reduce the need for repeated CT scans.2, 3, 4 Wireless pressure transducers have been successfully used to correlate decreases in sac pressures with concomitant decreases in sac size and successful sac exclusion after EVAR.5, 6 In addition, technetium-99m (99mTc)-labeled red blood cell scans have been used to detect endoleaks.7 Although these modalities can be less expensive, less invasive, and require little or no exposure to radiation, they focus specifically on the endograft and its site of placement. A CT scan provides a more complete image and has the advantage of identifying coexistent intra-abdominal and extra-abdominal findings.
Unexpected CT scan findings are common in vascular patients and of major concern: One-fourth of these patients with potentially serious lesions did not undergo further evaluation.8 The purpose of this study was to examine the frequency and nature of incidental findings seen on CT scan during preoperative and postoperative follow-up in patients undergoing EVAR at our institution.
Methods
We retrospectively reviewed 176 consecutive patients who underwent EVAR at our institution between January 1, 2000, and March 1, 2006. Patient demographics and findings not related to the EVAR were recorded. Patients were included in the study if all preoperative and postoperative surveillance CT scans were performed at our institution. CT scanning was done using either a 16-slice helical scanner (Phillips Medical Systems, Bothell, Wash) or a 64-slice helical scanner (General Electric Medical Systems, LLC, Waukesha, Wisc), which have equal resolution of body images. All results were read by an attending radiologist and interpreted as official results. Official reports on each patient were reviewed. An attending radiologist verified all new findings by comparing them with previous imaging studies.
Findings not related to EVAR were considered primary incidental if they were noted on patients' preoperative CT scans and secondary incidental if they appeared on surveillance CT scans but were not present on the preoperative studies. Primary and secondary incidental findings were considered either benign, such as gallstones or diverticulosis, or clinically significant if they warranted further workup, such as suspicious masses or changes suggestive of malignancy, internal or diaphragmatic hernias, and diverticulitis. The median follow-up was 29 months (range, 3-60 months). Each incidental finding was counted only once, on the first scan in which it appeared. Surveillance CT scans were grouped by 6-month, and 1-, 2-, 4-, and 5-year follow-up. Within each interval, the number of patients, secondary incidental findings, and clinically significant findings were recorded.
Univariate comparison of variables associated with the presence of suspicious lesions was performed with the χ2 test. Significance was determined at the level of P < .05.
Results
Of the 176 patients reviewed between January 1, 2000, and March 1, 2006, 82 (46%) had a preoperative and at least one follow-up CT scan at our institution. They consisted of 26 women (32%) and 56 men (68%), and their mean age was 76 years (range, 51-103 years). The other 94 patients had scans done at outside institutions or were lost to follow-up.
Primary incidental findings were non-AAA findings noted on patients' preoperative CT scans. Of the 82 patients, 73 (89%) had at least one primary incidental finding, and 14 (19%) of these were clinically significant (Fig, Table I). Secondary incidental findings in the 82 patients who underwent surveillance ranged from 40% to 59% during the follow-up intervals. At the same intervals, the secondary incidental findings that were clinically significant ranged from 11% to 27%.
Fig.
Incidental secondary findings and clinically significant secondary findings after endovascular aortic aneurysm repair during follow-up with computed tomography scan at 6 months, and 1, 2, 3, 4, and 5 years.
Table I. Primary and secondary incidental findings with corresponding clinically significant findingsa
| Findings | Pre-op | 6 months | 1 year | 2 years | 3 years | 4 years | 5 years |
|---|---|---|---|---|---|---|---|
| Patients with CT scans | 82 | ||||||
 Primary incidental findings | 73(89) | ||||||
| Clinically significant | 14(19) | ||||||
| Patients with surveillance CT scans | 64 | 50 | 33 | 22 | 14 | 5 | |
| Secondary incidental findings | 32(50) | 23(46) | 19(59) | 9(41) | 7(50) | 2(40) | |
| Clinically significant | 7(11) | 6(12) | 9(27) | 3(14) | 2(14) | 1(20) |
aData are presented as numbers with percentages in parenthesis. |
Many of the 59 patients with benign primary findings had more than one finding, as was the case with the patients within each follow-up interval who had benign secondary findings (Table II). Clinically significant findings were noted as new results on the CT scans. A total of 42 clinically significant findings were documented in 35 patients (3 patients had 2 findings; 2 patients had 3 findings); of these, 13 were primary and 29 were secondary clinically significant findings (Table II). Most consisted of intra-abdominal findings, followed by thoracic, retroperitoneal, and soft tissue areas of the torso.
Table II. Total clinically insignificant and significant primary and secondary incidental findings
| Primary (n = 137) | No. | Secondary (n = 93) | No. |
|---|---|---|---|
| Benign findings | |||
| Gallstones | 13 | Gallstones | 12 |
| Adrenal nodule <2.5 cm | 14 | Adrenal nodule <2.5 cm | 9 |
| Pleural thickening | 1 | Renal cyst | 6 |
| Hepatic cyst | 15 | Inguinal hernia | 11 |
| Hepatic hemangioma | 2 | Prostatic hypertrophy | 12 |
| Renal cyst | 45 | Diverticulosis | 14 |
| Diverticulosis | 16 | Fibroid uterus | 1 |
| Fibroid uterus | 2 | Cardiomegaly | 2 |
| Prostatic hypertrophy | 9 | Hydronephrosis | 1 |
| Ovarian cyst | 1 | Renal calculi | 4 |
| Ventral hernia | 6 | Splenic granuloma | 2 |
| Inguinal hernia | 5 | Duodenal diverticulum | 1 |
| Retroaortic renal vein | 1 | Bladder diverticulum | 4 |
| Pancreatic cyst <1 cm | 1 | Splenomegaly | 3 |
| Pleural effusion | 1 | Umbilical hernia | 1 |
| Cardiomegaly | 5 | Hiatal hernia | 3 |
| Splenic granuloma | 2 | Spigelian hernia | 1 |
| Hiatal hernia | 4 | Femoral hernia | 1 |
| Lung granuloma | 1 | Pancreatic head calcification | 1 |
| Pelvic kidney | 1 | Celiac artery stenosis | 1 |
| Femoral artery aneurysm | 1 | Spinal stenosis | 1 |
| Bladder diverticulum | 2 | Bladder prolapse | 1 |
| Duodenal diverticulum | 1 | Renal artery stenosis | 1 |
| Atrophic kidney | 3 |
| Primary (n = 14) | Secondary (n = 28) | ||
|---|---|---|---|
| Clinically significant findings | |||
| Cecal mass | 1 | Lung mass | 6 |
| Renal mass | 1 | Lytic bone lesion(s) | 2 |
| Lung mass | 4 | Bladder wall thickening | 1 |
| Pancreas mass | 1 | Adrenal mass | 1 |
| Liver mass | 1 | Kidney mass | 1 |
| Diaphragmatic hernia | 1 | Diverticulitis | 2 |
| Celiac artery aneurysm | 1 | Pancreas mass | 3 |
| Distal ileal thickening | 1 | Colonic fistula | 1 |
| Lytic bone lesion(s) | 1 | Liver mass | 6 |
| Gastric GIST | 1 | Carcinomatosis | 1 |
| Breast nodule | 1 | Appendix mucocele | 1 |
| Internal hernia | 1 | ||
| Inguinal adenopathy | 1 | ||
| Rectal mass | 1 |
We compared the 35 patients with clinically significant findings with the 47 patients without clinically significant findings (Table III). The mean age of patients with clinically significant findings was 76.06 years (range, 55-103 years) compared with 75.45 years (range, 51-91 years) for those without clinically significant findings, which was not significant (P = .1685). A significant difference (P = .03959) was found in the proportion of men to women when the 35 patients with clinically significant findings were compared with the 47 patients without clinically significant findings, respectively, 22 (62.86%) men and 13 (37.14%) women vs 34 (72.34%) men and 13 women (27.66%). The odds ratio of 1.55 indicates a slightly higher chance of finding a clinically significant result in women. The aneurysm size at the time of repair of patients with clinically significant findings was mean 5.415 cm (range, 3-9.5 cm) compared with 5.425 cm (range, 3.2-9 cm) for those without clinically significant findings, which was not statistically significant (P = .8409).
Table III. Characteristics of patients with and without clinically significant computed tomography scan findings
| Variable | Clinically significant findings | P | |
|---|---|---|---|
| Patients with | Patients without | ||
| Patients, No. | 35 | 47 | |
| Age, mean (range) years | 76.06(55-103) | 75.45(51-91) | .1685 |
| Men vs women, % | 62.86 vs 37.14 | 72.34 vs 27.66 | .03959a |
| Aneurysm size at repair, mean (range) cm | 5.415(3-9.5) | 5.425(3.2-9) | .8409 |
aOdds ratio, 1.55. |
Discussion
To accurately size endografts, patients with AAAs must receive a contrast-enhanced CT scan and be followed up with imaging after EVAR to detect graft patency, size of the aneurysm sac, endoleak, and graft migration. The technology of CT scanning continues to improve, including the evolution of dynamic CT scanning. Recent evidence suggests static CT imaging may not adequately identify patients with large aortic pulsatility, potentially resulting in endograft undersizing, stent graft migration, intermittent type I endoleaks, and poor patient outcomes. The current standard regimen of 10% to 15% oversizing based on static CT may be inadequate for some patients.9 Although some evidence suggests magnetic resonance imaging may be more sensitive in detecting endoleaks, CT scanning remains the imaging modality most used to monitor patients after EVAR.10
Recent evidence suggests DUS scanning approaches the accuracy of CT scanning in detecting endoleaks.2, 3, 4 Wireless pressure transducers have been successfully used to correlate decreases in sac pressures with concomitant decreases in sac size and successful sac exclusion after EVAR.5, 6 In addition, 99mTc-labeled red blood cell scans have been used to detect endoleaks, although they are less sensitive than CT scanning.7 These modalities have the obvious advantages of less exposure to radiation and the avoidance of iodinated contrast; however, they do not allow for detection of other findings that may be present in patients with AAAs. For these reasons, a CT scan may be a more complete form of surveillance as evidenced by the high number of incidental findings in these patients.
Our results showed 89% of patients had a nonaneurysmal-related finding at their initial CT scan, and 19% of these were clinically significant. Between 6 months and 5 years of follow-up, nonaneurysmal findings on patients' surveillance CT scans ranged between 40% and 50%, depending on the interval of follow-up (Table I). During this same follow-up period, 11% to 27% of these findings were clinically significant. Why the nonaneurysmal findings at the initial CT scan are so much higher than the range in these patients' surveillance CT scans is unclear. Perhaps because most of these were the first CT scans in this particular patient population is one possibility.
The incidence of nonaneurysmal findings during the surveillance period is in keeping with a recently published report by McDougal et al8 of a 50% incidence of extravascular pathology in vascular patients undergoing CT angiography. That study did, however, include patients undergoing CT angiography for vascular disease other than AAA, who comprised only 43% of their cohort, and it is unclear if these findings were compared with a patient's prior CT scan.
Likewise, our discovery of 19% clinically significant primary and 11% to 27% secondary findings was higher the 5% reported by McDougal et al.8 This difference may also be due to the lack of a uniform definition for clinically significant findings and that these findings can be subjective. Their clinically significant findings consisted mostly of solid-organ lesions or masses. Although we also considered these clinically significant, we also considered findings such as internal hernia, carcinomatosis, diverticulitis, diaphragmatic hernia, and findings suggestive of metastasis disease, such as lytic bone lesions, to be significant. One factor might be that our patients had AAAs that required repair and their cohort comprised patients with vascular disease; however, further study is warranted until this can be justified.
Although our findings suggest that most of our patients did not have important nonaneurysmal-related findings, 42 clinically significant findings were noted in 35 patients (Table II). Most of these findings were suggestive of malignancy; however, we did identify a celiac artery aneurysm, a diaphragmatic hernia, two patients with diverticulitis, a colonic fistula, and an internal hernia. A recent report of 500 men undergoing CT colonography found 315 (63.0%) had extracolonic findings, and 45 (9.0%) had clinically important extracolonic findings. Of the 596 extracolonic findings identified, 50 (8.4%) were thought to be clinically important.11 The average age of this male-only cohort was 62 years compared with our cohort of 68% men with an average age of 76 years. We had a mixed cohort with a higher average age, which may explain why our clinically significant findings were higher. In addition, the clinical marker we found associated with having a significant nonaneurysmal finding in our study was female sex (Table III). Aneurysm size and age did not significantly contribute to the likelihood that a patient would have a clinically significant incidental finding.
The first limitation in our study is that we initially reviewed 176 patients who underwent EVAR at our institution during a 6-year period, and 94 did not meet our study criteria. These patients had CT scans done at outside institutions, did not have follow-up, or did not survive, which limited the power of our study. Others analyzing follow-up CT imaging at major academic centers have had similar problems of losing a large number of patients to follow-up.8, 12 The CT scans in our study were read by more than one attending radiologist and are therefore subject to intraobserver variability. We were also unable to locate all of the medical records of those patients with clinically significant findings to determine if each one had an appropriate workup; however, these findings have influenced us to do so currently. Finally, our study is a retrospective analysis, and the true number of significant nonaneurysmal findings can only be elucidated by randomized prospective trials.
Conclusions
Our findings suggest CT scanning has the benefit of detecting nonaneurysmal-related findings in patients undergoing EVAR when compared with other methods such as DUS imaging, pressure sensor placement, angiography, and nuclear scanning. These data may support the more widespread use of CT scanning in patients undergoing EVAR. The identification of nonaneurysmal findings should not be the end point if it occurs. As vascular surgeons, we have the responsibility to convey these findings to patients' primary physicians to ensure they receive adequate care.
Author contributions
The authors wish to acknowledge the assistance of Alysia I. Privrat, MAPS, for her invaluable assistance in the preparation of this manuscript.
References
- . Mastery of vascular and endovascular surgery. Philadelphia, PA: Lippincott Williams & Wilkins; 2006;p. 163
- . Duplex ultrasound scanning versus computed tomographic angiography for postoperative evaluation of endovascular abdominal aortic aneurysm repair. J Vasc Surg. 2000;32:1142–1148
- Duplex ultrasound scanning is reliable in the detection of endoleak following endovascular aneurysm repair. Eur J Vasc Endovasc Surg. 2006;32:537–541
- . Fate of endoleaks detected by CT angiography and missed by color duplex ultrasound in endovascular grafts for abdominal aortic aneurysms. J Endovasc Ther. 2006;13:490–495
- Abdominal aortic aneurysm sac shrinkage after endovascular aneurysm repair: correlation with chronic sac pressure measurement. J Vasc Surg. 2006;43:2–7
- Initial results of wireless pressure sensing for endovascular aneurysm repair: the APEX Trial–Acute Pressure Measurement to Confirm Aneurysm Sac EXclusion. J Vasc Surg. 2007;45:236–242
- . Detection of endoleaks after endovascular aneurysm repair with use of technetium-99m sulfur colloid and (99m)Tc-labeled red blood cell scans. J Vasc Interv Radiol. 2006;17:1739–1743
- . Computed tomographic angiography has added value in patients with vascular disease. J Vasc Surg. 2006;44:998–1001
- Use of dynamic computed tomography to evaluate pre- and postoperative aortic changes in AAA patients undergoing endovascular aneurysm repair. J Endovasc Ther. 2007;14:44–49
- . Computed tomography versus magnetic resonance imaging of endoleaks after EVAR. Eur J Vasc Endovasc Surg. 2006;32:361–365
- Extracolonic abnormalities discovered incidentally at CT colonography in a male population. Radiology. 2005;236:519–526
- . Detection and follow-up of important extra-arterial lesions with helical CT angiography. Clin Radiol. 1999;54:294–300
Supported by a grant from the William J von Leibig Foundation.
Competition of interest: none.
PII: S0741-5214(08)00507-7
doi:10.1016/j.jvs.2008.03.057
© 2008 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
