- Email: [email protected]
Contrast Enhanced Duplex Ultrasound Imaging of the Mesenteric Arteries
Contrast Enhanced Duplex Ultrasound Imaging of the Mesenteric Arteries John Blebea, MD,1 Nikolaos Volteas, MD,2 Marsha Neumyer, RVT,1 John Ingraham, BS,1 Katherine Dawson, RN,1 Shahin Assadnia, MD,1 Karla M. Anderson, MD,1 and Robert G. Atnip, MD,1 Hershey, Pennsylvania and Thessaloniki, Greece
Duplex ultrasound of the visceral arteries is a technically challenging procedure. We examined the clinical usefulness of per¯utren intravenous ultrasound contrast to improve the diagnostic accuracy of such studies. Seventeen patients were prospectively studied. A color duplex imaging study of the visceral vasculature was performed with and without the contrast agent. Vessels were imaged and peak systolic velocity and Doppler waveforms of the aorta, celiac artery, superior mesenteric artery, and the inferior mesenteric artery were examined. These results were independently compared to those of contrast angiography. From this analysis we concluded contrast-enhanced duplex imaging of the mesenteric arteries is safe but not routinely required when performed by an experienced sonographer. Ultrasound contrast may be helpful in dif®cult patients when the vessels are not initially successfully visualized.
INTRODUCTION Duplex ultrasound evaluation of the visceral arteries is a technically challenging procedure hampered by the lack of reproducible uniform criteria among different institutions for the diagnosis of hemodynamically signi®cant stenosis.1-3 Intravenous microbubble contrast agents enhance vascular re¯ective acoustic signals and have been found to improve ultrasound diagnostic accuracy in the examination of the heart, liver, and peripheral vas-
1 Division of Vascular Surgery, The Pennsylvania State University College of Medicine, Hershey, PA. 2 Department of Surgery, Thessaloniki School of Medicine, Thessaloniki, Greece.
Presented at the Twenty-sixth Annual Meeting of the Peripheral Vascular Surgery Society, Baltimore, MD, June 9, 2001 Correspondence to: J. Blebea, MD, Division of Vascular Surgery, Penn State College of Medicine, 500 University Drive, M.C. H053, Hershey, PA 17033-0850, USA, E-mail: [email protected]. Ann Vasc Surg 2002; 16: 77-83 DOI: 10.1007/s10016-001-0144-2 Ó Annals of Vascular Surgery Inc. Published online: 21 January 2002
culature.4,5 We prospectively examined the potential clinical usefulness of an ultrasound contrast agent in the duplex examination of the mesenteric vasculature.
PATIENTS AND METHODS During a 9-month period from March to November 2000, a study examining duplex imaging of the mesenteric arteries was performed at the Milton S. Hershey Medical Center of the Pennsylvania State University School of Medicine as part of a prospective multicenter open-label phase 2 clinical trial to assess the usefulness of contrast-enhanced imaging of renal artery stenosis sponsored by DuPont Pharmaceuticals Co. (Billerica, MA). The study was approved by the Institutional Review Board of the Penn State College of Medicine. Adult patients who had undergone abdominal contrast angiography for suspected renal or peripheral atherosclerotic disease within the prior 90 days were eligible for participation. Two patients (12%) had signs and symptoms of chronic mesenteric ischemia. After an 8-h fast, a complete abdominal color 77
78 Blebea et al.
duplex ultrasound scan was performed by an experienced registered vascular technologist (M.N.), blinded to the results of the angiogram, using an ATL HDI 5000 (ATL Ultrasound, Bothell, WA) ultrasound scanner and a curved 2-5 MHz Doppler probe. The scanner's parameters in terms of transmit power, gain, repetition frequency, and dynamic range were individually optimized for each patient. The aorta, celiac artery (CA), and its splenic and hepatic branches, superior mesenteric artery (SMA), and inferior mesenteric arteries (IMA) were imaged in two-dimensional (2-D) grayscale. The peak systolic (PSV) and end diastolic (EDV) velocities were measured and Doppler spectral waveform characteristics evaluated. Care was taken to maintain the angle of insonation between 40° and 60° and obtain Doppler velocity measurements from the center of the vessel. Doppler criteria for hemodynamically signi®cant arterial stenosis were de®ned as a PSV of ³275 cm/sec for the SMA, and ³200 cm/sec for the CA and IMA.1,6 After the initial examination was completed, an identical contrast-enhanced study was performed by the same technologist with the same ultrasound scanner. A vial of per¯utren contrast agent (Definityä, DuPont Pharmaceuticals Co) was agitated for 45 sec at 4500 oscillations/min in a vial shaker. A volume of 1.3 mL was then injected into a 50-mL bag of 0.9% sterile saline solution. This was then infused intravenously at a constant rate of 2 mL/ min via an 18-guage needle in a forearm vein throughout the duration of the duplex examination. All studies were recorded on tape. The duplex results were independently recorded (N.V.) and compared with the results obtained with contrast angiography, by which vessel patency and degree of stenosis were measured. Clinical laboratory examinations, which included serum blood urea nitrogen (BUN) and creatinine, were performed at baseline before the duplex examination and at follow-up 48-72 hr later. Vital signs, consisting of respiratory rate (RR), heart rate (HR), and supine blood pressure, were measured within 15 min prior to the infusion of the contrast agent, during the infusion 30 min later, then at 1 and 3 hr. A follow-up visit was performed 2 to 3 days after the study and a history was compiled and clinical assessment made for possible adverse events. All values are reported as mean standard error of the mean. Statistical analysis was performed using Sigma Stat 2.0 statistical software package (Jandel, San Rafael, CA). Paired Student t-testing was used for data comparisons. A p-value of <0.05 was considered statistically signi®cant.
Annals of Vascular Surgery
Table I. Clinical risk factors Factor
Patients (%)
Hypertension Tobacco use Coronary artery disease Peripheral vascular disease Hyperlipidemia Diabetes mellitus
94 88 65 65 56 12
RESULTS There were a total of 17 patients in the study. There were 11 males and 6 females with a mean age of 62 3 years (range 47-79 years) and a body weight of 173 8 pounds. They had the expected risk factors for systemic atherosclerotic disease (Table 1). The rate of contrast infusion was 2 mL/min in all except one patient who had an infusion rate of 4 mL/min. The mean total volume of infusion in each patient was 24 2 mL of contrast solution. For the CA, near equivalent results were obtained with duplex ultrasound alone and with contrast. Both methods correctly identi®ed the two lesions of >70% stenosis but duplex alone could not visualize two vessels and falsely suggested a hemodynamically signi®cant stenosis in a vessel with only a 40% stenosis by contrast angiography. With the addition of contrast, all vessels were successfully identi®ed but two arteries with an angiographic 40% stenosis had a PSV of >300 cm/ sec, falsely suggesting a hemodynamically signi®cant stenoses. The accuracy of the methods for the CA was therefore 82% and 88%, respectively, with and without echocontrast. Examination of the SMA showed an accuracy of 71% for duplex alone, correctly identifying 1 of 2 occlusions, 2 of 3 stenoses, and 9 of 12 normal vessels. With contrast, both occlusions and all the stenosis were correctly identi®ed as well as two normal vessels (Fig. 1), for an accuracy rate of 94%. The diagnostic accuracy in the duplex examination of the IMA was signi®cantly decreased compared to that of the SMA (p < 0.001). In the IMA, duplex alone did not correctly diagnose two of seven occlusions, two of four normal vessels, and ®ve of six vessels with hemodynamically signi®cant stenoses, for an accuracy rate of 35%. The addition of contrast was associated with an accuracy rate of 38%, correctly identifying three of the normal vessels, four occlusions, and one of the three stenoses examined (Fig. 2). Combining all 68 vessels, including the aorta, the accuracy rate for duplex alone was 72% and 85% for those studies performed with the
Vol. 16, No. 1, 2002
Contrast imaging of mesenteric arteries 79
Fig. 1. A Abdominal color ¯ow duplex imaging in an obese patient fails to visualize either the aorta or mesenteric vessels, demonstrating only unidenti®ed branch arteries (arrows). B With the infusion of contrast, both the CA (straight arrow) and SMA (curved arrow) are easily seen at their origins. C Doppler spectra can now be obtained from the arteries.
80 Blebea et al.
contrast agent (Table II). This was associated with a somewhat higher sensitivity for detecting stenosis or occlusion with the addition of contrast compared to that of duplex alone (81% vs. 55%, respectively) and similar speci®city (84% vs. 79%, respectively). Although duplex alone falsely indicated a hemodynamically signi®cant stenosis in 4 vessels and was falsely normal in another 3, 12 vessels (12/ 68; 18%) were not successfully visualized. This accounted for 63% (12/19) of misdiagnoses with duplex alone as compared to only 3 vessels (5%) not visualized when contrast was infused. Of the 12 arteries not seen by duplex alone, 9 (75%) were successfully visualized with the addition of contrast. The vessels not visualized by duplex ultrasound alone, but seen with contrast, included an occluded and a stenotic SMA, a stenotic IMA, along with a normal aorta and two normal inferior mesenteric and celiac arteries. The arteries with stenosis, both with and without contrast, had PSV signi®cantly higher than those without stenosis (Table III). Although there was a trend toward increasing PSV after the addition of contrast, this did not reach statistical signi®cance. In only one patient, with a 40% stenosis of the CA, did the addition of contrast result in an increase in the PSV, suggesting a hemodynamically signi®cant stenosis not ful®lling such velocity criteria without the presence of contrast infusion. There were no adverse physiologic effects seen with contrast infusion. Respiratory rate, heart rate, and blood pressure were all without change (Figs. 3 and 4). The BUN at baseline was 24 4 compared to 22 3 mg/dL after contrast. Similarly, creatinine was 1.2 0.2 before and 1.3 0.2 mg/dL after contrast. No serious adverse effects were seen with administration of contrast or related to the echocontrast during the study.
DISCUSSION Chronic mesenteric ischemia is an uncommon clinical condition usually seen in association with advanced systemic atherosclerotic occlusive disease. If untreated, this chronic condition can lead to acute mesenteric ischemia and bowel infarction, with a mortality rate as high as 70%.7 Because of the inaccuracy of clinical signs and symptoms in diagnosing this condition, duplex ultrasound was introduced to measure splanchnic blood ¯ow8 and the detection of stenosis of the mesenteric arteries in patients suspected of having chronic mesenteric ischemia.9 Unlike the experience with extracranial carotid arterial disease, however, it has been much more
Annals of Vascular Surgery
dif®cult to establish duplex ultrasound criteria for the diagnosis of hemodynamically signi®cant stenosis in the splanchnic circulation.Because of the depth of these intraabdominal vessels and the overlying bowel gas, visualization of the mesenteric arteries is technically challenging. In order to see them, a low-frequency transducer is needed, which results in deterioration of the B-mode image resolution. With this decreased visualization, inaccurate placement of the Doppler sample volume for velocity measurements can become a signi®cant source of error.1,8 Different institutions have also de®ned varying criteria for diagnosis. Harward et al. employed peak systolic frequencies3 while most other investigators have used velocity criteria. Within this group, Zwolak et al.2,9 found EDV more useful while Monetta et al. have utilized PSV.1 The latter group also de®ned a hemodynamically signi®cant lesion as one with a >70% diameter reduction on contrast angiograph, while others have used the very same velocity criteria for a 50% stenosis.6 Differences in threshold criteria used among diverse institutions may also be partially explained by different patient populations, prevalence of the disease, and intrinsic variation in velocity measurements among different brands of ultrasound scanner.2,10,11 In this study, we wished to examine the clinical usefulness of an ultrasound contrast agent in improving the diagnostic accuracy of duplex ultrasound imaging of the mesenteric arteries. Ultrasound contrast agents are stabilized gas microbubbles that enhance ultrasound signals through the difference between them and the surrounding blood in terms of compressibility and density. Although originally described in the context of room air rapidly agitated in a syringe of normal saline before injection, a variety of agents ranging from albumin to carbohydrates and ¯uorocarbons, have subsequently been developed to coat the microbubbles.12 These second-generation agents have been developed for increased density, compressibility, intravascular persistence, and acoustic backscatter. They re¯ect transmitted ultrasound strongly, at both fundamental and harmonic frequencies, which provides improved vascular imaging and tissue differentiation. In the present study, the per¯utren contrast agent was used.5,13 This is composed of three phospholipids and a per¯uoropropane gas which, when mixed together, form lipid encapsulated gas-®lled microbubbles. They are <5 lm in diameter, thus smaller than red blood cells, which allows them to pass easily through the pulmonary microcirculation. They are rapidly cleared from the systemic circu-
Vol. 16, No. 1, 2002
Contrast imaging of mesenteric arteries 81
Fig. 2. A The IMA is identi®ed without contrast but its origin is not seen and Doppler velocities of 109 cm/sec are measured beyond the area of greatest stenosis. B The infusion of ultrasound contrast better visualizes the focal ori®cial stenosis and allows measurement of PSV of 352 cm/sec, correctly diagnosing a hemodynamically signi®cant stenosis.
lation without metabolism through expired air, with none being detected within 5 min of termination of the infusion. In our study, we found a tendency toward increased visualization and identi®cation of the mesenteric vessels and improved diagnosis of hemodynamically signi®cant stenoses (Table II).
The diagnostic improvements with echocontrast were most evident in the celiac and superior mesenteric arteries, while examination of the IMA did not bene®t markedly by the infusion of contrast. The lack of statistical signi®cance most likely re¯ects the small number of patients in this study. Such results, however, are consistent with other
Table II. Comparison of accuracy of mesenteric artery imaging Angiogram
Duplex
Accuracy
Duplex + contrast
Accuracy
Normal (48) Stenosis ³50% (11) Occlusions (9) Accuracy (68)
38 5 6 49
79% 45% 67% 72%
44 7 6 57
90% 78% 67% 85%
82 Blebea et al.
Annals of Vascular Surgery
Table III. Doppler peak systolic velocities
Normal arteries Stenosis ³50% All arteries a b
Fig. 3. Respiratory rate in breaths per minute (dotted line) and heart rate expressed as beats per minute (solid line) at baseline just before injection (time 0) and following the infusion of contrast. There were no changes in either parameter.
Fig. 4. Contrast infusion induced no changes in either systolic or diastolic blood pressure.
series that have demonstrated the usefulness of echocontrast in the evaluation of cardiac and peripheral vascular circulation.5,13,14 In a multicenter placebo-controlled trial that included 140 patients with initially suboptimal non-contrast echocardiograms, Kitzman et al. found that with per¯utren, diagnostic studies were achieved in 48% of Patients.5 Using per¯enapent bubbles (EchoGen), Robbin et al. found that 93% of patients had vascular enhancement as compared to placebo, converting 15% of the nondiagnostic studies into complete studies and improving diagnostic con®dence in 54% of patients.4 Similarly, in the peripheral circulation, there was a marked increase in the intensity of pulsed-wave Doppler signals obtained in the femoral artery after a bolus injection of galactose-covered microbubbles (Levovist), with an attendant improvement in clinical diagnostic con®dence when used in initially suboptimal clinical exams.14
Duplex alone
Duplex + Contrast
110 9 190 2a 142 14
120 8 214 34b 157 15
p < 0.01 vs. normal. p < 0.001 vs. normal.
Even when performed by an experienced ultrasonographer, identifying the mesenteric arteries can be dif®cult (Fig. 1). Ten vessels (15%) were not successfully visualized in our study. This is not an unusual occurrence. In a series with 38 patients, Harward et al. noted that 18% of studies did not have a clear image of the mesenteric vessels, thereby precluding accurate positioning of the pulsed-wave Doppler sample volume for measurement of velocity.3 In 100 patients, Monetta et al. failed to visualize 7% of SMAs and 17% of celiac vessels.1 Artery identi®cation may be most problematic with diseased vessels. Bowersox et al. failed to visualize one-third (2 of 6) of CA or SMA occlusions and one-half (6/12) of the stenoses by duplex ultrasound.9 By contrast, Perko et al. found that in 10% (7/78) of occluded arteries, unenhanced duplex ultrasound falsely suggested ¯ow.6 The addition of echocontrast may be particularly applicable in this subset of patients with diseased vessels. In our experience, of the 10 nonvisualized vessels, 6 were successfully seen following the infusion of per¯utren ultrasound contrast. In addition, the overall rate of nonvisualization was only 7% when contrast was employed. Better visualization of the mesenteric vessels will allow more precise measurement of velocity and classi®cation of stenosis (Fig. 2, Table II). Accurate velocity measurements are extremely dependent on the angle of insonation of the Doppler beam in relation to the vessel axis, with a 2° change from 60° inducing a 7% error.8 Contrast may enable improved visualization of the site of stenosis and therefore a more accurate placement of the sample volume to allow determination of a hemodynamically signi®cant lesion. Like other investigators, we noted a 10% increase in PSV after per¯utren infusion (Table III). The increase in PSV does not appear to re¯ect real changes in ¯ow. Such changes in velocity measurements were documented in animal studies, even though measured blood ¯ow did not change.15 Similarly, in a ¯ow phantom under controlled conditions, there was a 17% increase in the Doppler shift velocity without changes in actual ¯ow volumes. Such increases in measured blood velocities are most likely artifactual and re-
Vol. 16, No. 1, 2002
lated to the limited dynamic range of the ultrasound scanner.15 In our patients, in only one case of a CA with a 40% angiographic stenosis did the increase in PSV after contrast infusion change the classi®cation of the vessel into one with a hemodynamically signi®cant stenosis. Even here, it may be proposed that the duplex results re¯ect the true anatomic status of the vessel rather than the 2D measurement performed on the angiogram. Not enough clinical experience has yet been attained with echocontrast infusion to recommend changing the already established velocity criteria for diagnosing hemodynamically signi®cant stenosis. We found no change in the physiologic parameters of heart rate or blood pressure, or renal function as evidenced by BUN or creatinine levels, with per¯utren infusion (Figs. 3 and 4). Similarly, no clinically signi®cant changes in vital signs or blood chemistries were previously seen when the same agent was given at twice the dose of this study for evaluation of cardiac left ventricular opaci®cation.5,13 Additionally, no patient suffered any adverse events related to the contrast agent.
CONCLUSION Our study suggests that contrast-enhanced imaging of the mesenteric arteries is safe. Increased velocities are seen but they do not lead to false diagnosis of stenoses. Echocontrast is not routinely required but may be helpful when adequate imaging has not been achieved without it, such as in obese patients, postoperatively, or in the presence of much abdominal gas. Supported in part by an Educational Research Grant from The Peripheral Vascular Surgery REFERENCES 1. Monetta GL, Lee RW, Yeager RA, Taylor LM, Porter JM. Mesenteric duplex scanning: a blinded prospective study. J Vasc Surg 1993;17:79-86.
Contrast imaging of mesenteric arteries 83
2. Zwolak RM, Fillinger MF, Walsh DB, et al. Mesenteric and celiac duplex scanning: a validation study. J Vasc Surg 1998;27:1078-1088. 3. Harward TR, Smith S, Seeger JM. Detection of celiac axis and superior mesenteric artery occlusive disease with use of abdominal duplex scanning. J Vasc Surg 1993;17:738745. 4. Robbin ML, Eisenfeld AJ, et al. Per¯enapent emulsion: a US contrast agent for diagnostic radiology-multicenter doubleblind comparison with placebo. Radiology 1998;207:717722. 5. Kitzman DW, Goldman ME, Gillam LD, Cohen JL, Aurigemma GP, Gottdiener JS. Ef®cacy and safety of the novel ultrasound contrast agent per¯utren (De®nity) in patients with suboptimal baseline left ventricular echocardiographic images. Am J Cardiol 2000;86:669-674. 6. Perko MJ, Just S, Schroeder TV. Importance of diastolic velocities in the detection of celiac and mesenteric artery disease by duplex ultrasound. J Vasc Surg 1997;26:288293. 7. Stoney RJ, Cunningham CG. Acute mesenteric ischemia: clinical update. Surgery 1993;114:489-490. 8. Jager K, Bollinger A, Valli C, Amman R. Measurement of mesenteric blood ¯ow by duplex scanning. J Vasc Surg 1986;3:462-469. 9. Bowersox JC, Zwolak RM, Walsh DB, Schneider JR, et al. Duplex ultrasonography in the diagnosis of celiac and mesenteric artery occlusive disease. J Vasc Surg 1991;14:780788. 10. Healy DA, Neumyer MM, Atnip RA, Thiele BL. Evaluation of celiac and mesenteric vascular disease with duplex ultrasonography. J Ultrasound Med 1992;11:481-485. 11. Blebea J. Duplex ultrasound criteria for diagnosis of splanchnic artery stenosis or occlusion. J Vasc Surg 1992;16: 796-797. 12. Cotter B, Mahmud E, Kwan OL, DeMaria AN. New ultrasound contrast agents: expanding upon existing clinical applications. In: Goldberg BB, ed. Ultrasound Contrast Agents. St. Louis: Mosby, 1997, pp 31-42. 13. Weissman NJ, Cohen MC, Hack TC, Gillam LD, Cohen JL, Kitzman DW. Infusion versus bolus contrast echocardiography: a multicenter, open-label, crossover trial. Am Heart J 2000;139:399-404. 14. Schwarz KQ, Becher H, Schimpfky C, Vorwerk D, Bogdahn U, Schlief R. Doppler enhancement with SH U 508A in multiple vascular regions. Radiology 1994;l93:195201. 15. Forsberg F, Liu JB, Burns PN, et al. Artifacts in ultrasonic contrast agent studies. J Ultrasound Med 1994;13:357365.
Duplex ultrasound of the visceral arteries is a technically challenging procedure. We examined the clinical usefulness of per¯utren intravenous ultrasound contrast to improve the diagnostic accuracy of such studies. Seventeen patients were prospectively studied. A color duplex imaging study of the visceral vasculature was performed with and without the contrast agent. Vessels were imaged and peak systolic velocity and Doppler waveforms of the aorta, celiac artery, superior mesenteric artery, and the inferior mesenteric artery were examined. These results were independently compared to those of contrast angiography. From this analysis we concluded contrast-enhanced duplex imaging of the mesenteric arteries is safe but not routinely required when performed by an experienced sonographer. Ultrasound contrast may be helpful in dif®cult patients when the vessels are not initially successfully visualized.
INTRODUCTION Duplex ultrasound evaluation of the visceral arteries is a technically challenging procedure hampered by the lack of reproducible uniform criteria among different institutions for the diagnosis of hemodynamically signi®cant stenosis.1-3 Intravenous microbubble contrast agents enhance vascular re¯ective acoustic signals and have been found to improve ultrasound diagnostic accuracy in the examination of the heart, liver, and peripheral vas-
1 Division of Vascular Surgery, The Pennsylvania State University College of Medicine, Hershey, PA. 2 Department of Surgery, Thessaloniki School of Medicine, Thessaloniki, Greece.
Presented at the Twenty-sixth Annual Meeting of the Peripheral Vascular Surgery Society, Baltimore, MD, June 9, 2001 Correspondence to: J. Blebea, MD, Division of Vascular Surgery, Penn State College of Medicine, 500 University Drive, M.C. H053, Hershey, PA 17033-0850, USA, E-mail: [email protected]. Ann Vasc Surg 2002; 16: 77-83 DOI: 10.1007/s10016-001-0144-2 Ó Annals of Vascular Surgery Inc. Published online: 21 January 2002
culature.4,5 We prospectively examined the potential clinical usefulness of an ultrasound contrast agent in the duplex examination of the mesenteric vasculature.
PATIENTS AND METHODS During a 9-month period from March to November 2000, a study examining duplex imaging of the mesenteric arteries was performed at the Milton S. Hershey Medical Center of the Pennsylvania State University School of Medicine as part of a prospective multicenter open-label phase 2 clinical trial to assess the usefulness of contrast-enhanced imaging of renal artery stenosis sponsored by DuPont Pharmaceuticals Co. (Billerica, MA). The study was approved by the Institutional Review Board of the Penn State College of Medicine. Adult patients who had undergone abdominal contrast angiography for suspected renal or peripheral atherosclerotic disease within the prior 90 days were eligible for participation. Two patients (12%) had signs and symptoms of chronic mesenteric ischemia. After an 8-h fast, a complete abdominal color 77
78 Blebea et al.
duplex ultrasound scan was performed by an experienced registered vascular technologist (M.N.), blinded to the results of the angiogram, using an ATL HDI 5000 (ATL Ultrasound, Bothell, WA) ultrasound scanner and a curved 2-5 MHz Doppler probe. The scanner's parameters in terms of transmit power, gain, repetition frequency, and dynamic range were individually optimized for each patient. The aorta, celiac artery (CA), and its splenic and hepatic branches, superior mesenteric artery (SMA), and inferior mesenteric arteries (IMA) were imaged in two-dimensional (2-D) grayscale. The peak systolic (PSV) and end diastolic (EDV) velocities were measured and Doppler spectral waveform characteristics evaluated. Care was taken to maintain the angle of insonation between 40° and 60° and obtain Doppler velocity measurements from the center of the vessel. Doppler criteria for hemodynamically signi®cant arterial stenosis were de®ned as a PSV of ³275 cm/sec for the SMA, and ³200 cm/sec for the CA and IMA.1,6 After the initial examination was completed, an identical contrast-enhanced study was performed by the same technologist with the same ultrasound scanner. A vial of per¯utren contrast agent (Definityä, DuPont Pharmaceuticals Co) was agitated for 45 sec at 4500 oscillations/min in a vial shaker. A volume of 1.3 mL was then injected into a 50-mL bag of 0.9% sterile saline solution. This was then infused intravenously at a constant rate of 2 mL/ min via an 18-guage needle in a forearm vein throughout the duration of the duplex examination. All studies were recorded on tape. The duplex results were independently recorded (N.V.) and compared with the results obtained with contrast angiography, by which vessel patency and degree of stenosis were measured. Clinical laboratory examinations, which included serum blood urea nitrogen (BUN) and creatinine, were performed at baseline before the duplex examination and at follow-up 48-72 hr later. Vital signs, consisting of respiratory rate (RR), heart rate (HR), and supine blood pressure, were measured within 15 min prior to the infusion of the contrast agent, during the infusion 30 min later, then at 1 and 3 hr. A follow-up visit was performed 2 to 3 days after the study and a history was compiled and clinical assessment made for possible adverse events. All values are reported as mean standard error of the mean. Statistical analysis was performed using Sigma Stat 2.0 statistical software package (Jandel, San Rafael, CA). Paired Student t-testing was used for data comparisons. A p-value of <0.05 was considered statistically signi®cant.
Annals of Vascular Surgery
Table I. Clinical risk factors Factor
Patients (%)
Hypertension Tobacco use Coronary artery disease Peripheral vascular disease Hyperlipidemia Diabetes mellitus
94 88 65 65 56 12
RESULTS There were a total of 17 patients in the study. There were 11 males and 6 females with a mean age of 62 3 years (range 47-79 years) and a body weight of 173 8 pounds. They had the expected risk factors for systemic atherosclerotic disease (Table 1). The rate of contrast infusion was 2 mL/min in all except one patient who had an infusion rate of 4 mL/min. The mean total volume of infusion in each patient was 24 2 mL of contrast solution. For the CA, near equivalent results were obtained with duplex ultrasound alone and with contrast. Both methods correctly identi®ed the two lesions of >70% stenosis but duplex alone could not visualize two vessels and falsely suggested a hemodynamically signi®cant stenosis in a vessel with only a 40% stenosis by contrast angiography. With the addition of contrast, all vessels were successfully identi®ed but two arteries with an angiographic 40% stenosis had a PSV of >300 cm/ sec, falsely suggesting a hemodynamically signi®cant stenoses. The accuracy of the methods for the CA was therefore 82% and 88%, respectively, with and without echocontrast. Examination of the SMA showed an accuracy of 71% for duplex alone, correctly identifying 1 of 2 occlusions, 2 of 3 stenoses, and 9 of 12 normal vessels. With contrast, both occlusions and all the stenosis were correctly identi®ed as well as two normal vessels (Fig. 1), for an accuracy rate of 94%. The diagnostic accuracy in the duplex examination of the IMA was signi®cantly decreased compared to that of the SMA (p < 0.001). In the IMA, duplex alone did not correctly diagnose two of seven occlusions, two of four normal vessels, and ®ve of six vessels with hemodynamically signi®cant stenoses, for an accuracy rate of 35%. The addition of contrast was associated with an accuracy rate of 38%, correctly identifying three of the normal vessels, four occlusions, and one of the three stenoses examined (Fig. 2). Combining all 68 vessels, including the aorta, the accuracy rate for duplex alone was 72% and 85% for those studies performed with the
Vol. 16, No. 1, 2002
Contrast imaging of mesenteric arteries 79
Fig. 1. A Abdominal color ¯ow duplex imaging in an obese patient fails to visualize either the aorta or mesenteric vessels, demonstrating only unidenti®ed branch arteries (arrows). B With the infusion of contrast, both the CA (straight arrow) and SMA (curved arrow) are easily seen at their origins. C Doppler spectra can now be obtained from the arteries.
80 Blebea et al.
contrast agent (Table II). This was associated with a somewhat higher sensitivity for detecting stenosis or occlusion with the addition of contrast compared to that of duplex alone (81% vs. 55%, respectively) and similar speci®city (84% vs. 79%, respectively). Although duplex alone falsely indicated a hemodynamically signi®cant stenosis in 4 vessels and was falsely normal in another 3, 12 vessels (12/ 68; 18%) were not successfully visualized. This accounted for 63% (12/19) of misdiagnoses with duplex alone as compared to only 3 vessels (5%) not visualized when contrast was infused. Of the 12 arteries not seen by duplex alone, 9 (75%) were successfully visualized with the addition of contrast. The vessels not visualized by duplex ultrasound alone, but seen with contrast, included an occluded and a stenotic SMA, a stenotic IMA, along with a normal aorta and two normal inferior mesenteric and celiac arteries. The arteries with stenosis, both with and without contrast, had PSV signi®cantly higher than those without stenosis (Table III). Although there was a trend toward increasing PSV after the addition of contrast, this did not reach statistical signi®cance. In only one patient, with a 40% stenosis of the CA, did the addition of contrast result in an increase in the PSV, suggesting a hemodynamically signi®cant stenosis not ful®lling such velocity criteria without the presence of contrast infusion. There were no adverse physiologic effects seen with contrast infusion. Respiratory rate, heart rate, and blood pressure were all without change (Figs. 3 and 4). The BUN at baseline was 24 4 compared to 22 3 mg/dL after contrast. Similarly, creatinine was 1.2 0.2 before and 1.3 0.2 mg/dL after contrast. No serious adverse effects were seen with administration of contrast or related to the echocontrast during the study.
DISCUSSION Chronic mesenteric ischemia is an uncommon clinical condition usually seen in association with advanced systemic atherosclerotic occlusive disease. If untreated, this chronic condition can lead to acute mesenteric ischemia and bowel infarction, with a mortality rate as high as 70%.7 Because of the inaccuracy of clinical signs and symptoms in diagnosing this condition, duplex ultrasound was introduced to measure splanchnic blood ¯ow8 and the detection of stenosis of the mesenteric arteries in patients suspected of having chronic mesenteric ischemia.9 Unlike the experience with extracranial carotid arterial disease, however, it has been much more
Annals of Vascular Surgery
dif®cult to establish duplex ultrasound criteria for the diagnosis of hemodynamically signi®cant stenosis in the splanchnic circulation.Because of the depth of these intraabdominal vessels and the overlying bowel gas, visualization of the mesenteric arteries is technically challenging. In order to see them, a low-frequency transducer is needed, which results in deterioration of the B-mode image resolution. With this decreased visualization, inaccurate placement of the Doppler sample volume for velocity measurements can become a signi®cant source of error.1,8 Different institutions have also de®ned varying criteria for diagnosis. Harward et al. employed peak systolic frequencies3 while most other investigators have used velocity criteria. Within this group, Zwolak et al.2,9 found EDV more useful while Monetta et al. have utilized PSV.1 The latter group also de®ned a hemodynamically signi®cant lesion as one with a >70% diameter reduction on contrast angiograph, while others have used the very same velocity criteria for a 50% stenosis.6 Differences in threshold criteria used among diverse institutions may also be partially explained by different patient populations, prevalence of the disease, and intrinsic variation in velocity measurements among different brands of ultrasound scanner.2,10,11 In this study, we wished to examine the clinical usefulness of an ultrasound contrast agent in improving the diagnostic accuracy of duplex ultrasound imaging of the mesenteric arteries. Ultrasound contrast agents are stabilized gas microbubbles that enhance ultrasound signals through the difference between them and the surrounding blood in terms of compressibility and density. Although originally described in the context of room air rapidly agitated in a syringe of normal saline before injection, a variety of agents ranging from albumin to carbohydrates and ¯uorocarbons, have subsequently been developed to coat the microbubbles.12 These second-generation agents have been developed for increased density, compressibility, intravascular persistence, and acoustic backscatter. They re¯ect transmitted ultrasound strongly, at both fundamental and harmonic frequencies, which provides improved vascular imaging and tissue differentiation. In the present study, the per¯utren contrast agent was used.5,13 This is composed of three phospholipids and a per¯uoropropane gas which, when mixed together, form lipid encapsulated gas-®lled microbubbles. They are <5 lm in diameter, thus smaller than red blood cells, which allows them to pass easily through the pulmonary microcirculation. They are rapidly cleared from the systemic circu-
Vol. 16, No. 1, 2002
Contrast imaging of mesenteric arteries 81
Fig. 2. A The IMA is identi®ed without contrast but its origin is not seen and Doppler velocities of 109 cm/sec are measured beyond the area of greatest stenosis. B The infusion of ultrasound contrast better visualizes the focal ori®cial stenosis and allows measurement of PSV of 352 cm/sec, correctly diagnosing a hemodynamically signi®cant stenosis.
lation without metabolism through expired air, with none being detected within 5 min of termination of the infusion. In our study, we found a tendency toward increased visualization and identi®cation of the mesenteric vessels and improved diagnosis of hemodynamically signi®cant stenoses (Table II).
The diagnostic improvements with echocontrast were most evident in the celiac and superior mesenteric arteries, while examination of the IMA did not bene®t markedly by the infusion of contrast. The lack of statistical signi®cance most likely re¯ects the small number of patients in this study. Such results, however, are consistent with other
Table II. Comparison of accuracy of mesenteric artery imaging Angiogram
Duplex
Accuracy
Duplex + contrast
Accuracy
Normal (48) Stenosis ³50% (11) Occlusions (9) Accuracy (68)
38 5 6 49
79% 45% 67% 72%
44 7 6 57
90% 78% 67% 85%
82 Blebea et al.
Annals of Vascular Surgery
Table III. Doppler peak systolic velocities
Normal arteries Stenosis ³50% All arteries a b
Fig. 3. Respiratory rate in breaths per minute (dotted line) and heart rate expressed as beats per minute (solid line) at baseline just before injection (time 0) and following the infusion of contrast. There were no changes in either parameter.
Fig. 4. Contrast infusion induced no changes in either systolic or diastolic blood pressure.
series that have demonstrated the usefulness of echocontrast in the evaluation of cardiac and peripheral vascular circulation.5,13,14 In a multicenter placebo-controlled trial that included 140 patients with initially suboptimal non-contrast echocardiograms, Kitzman et al. found that with per¯utren, diagnostic studies were achieved in 48% of Patients.5 Using per¯enapent bubbles (EchoGen), Robbin et al. found that 93% of patients had vascular enhancement as compared to placebo, converting 15% of the nondiagnostic studies into complete studies and improving diagnostic con®dence in 54% of patients.4 Similarly, in the peripheral circulation, there was a marked increase in the intensity of pulsed-wave Doppler signals obtained in the femoral artery after a bolus injection of galactose-covered microbubbles (Levovist), with an attendant improvement in clinical diagnostic con®dence when used in initially suboptimal clinical exams.14
Duplex alone
Duplex + Contrast
110 9 190 2a 142 14
120 8 214 34b 157 15
p < 0.01 vs. normal. p < 0.001 vs. normal.
Even when performed by an experienced ultrasonographer, identifying the mesenteric arteries can be dif®cult (Fig. 1). Ten vessels (15%) were not successfully visualized in our study. This is not an unusual occurrence. In a series with 38 patients, Harward et al. noted that 18% of studies did not have a clear image of the mesenteric vessels, thereby precluding accurate positioning of the pulsed-wave Doppler sample volume for measurement of velocity.3 In 100 patients, Monetta et al. failed to visualize 7% of SMAs and 17% of celiac vessels.1 Artery identi®cation may be most problematic with diseased vessels. Bowersox et al. failed to visualize one-third (2 of 6) of CA or SMA occlusions and one-half (6/12) of the stenoses by duplex ultrasound.9 By contrast, Perko et al. found that in 10% (7/78) of occluded arteries, unenhanced duplex ultrasound falsely suggested ¯ow.6 The addition of echocontrast may be particularly applicable in this subset of patients with diseased vessels. In our experience, of the 10 nonvisualized vessels, 6 were successfully seen following the infusion of per¯utren ultrasound contrast. In addition, the overall rate of nonvisualization was only 7% when contrast was employed. Better visualization of the mesenteric vessels will allow more precise measurement of velocity and classi®cation of stenosis (Fig. 2, Table II). Accurate velocity measurements are extremely dependent on the angle of insonation of the Doppler beam in relation to the vessel axis, with a 2° change from 60° inducing a 7% error.8 Contrast may enable improved visualization of the site of stenosis and therefore a more accurate placement of the sample volume to allow determination of a hemodynamically signi®cant lesion. Like other investigators, we noted a 10% increase in PSV after per¯utren infusion (Table III). The increase in PSV does not appear to re¯ect real changes in ¯ow. Such changes in velocity measurements were documented in animal studies, even though measured blood ¯ow did not change.15 Similarly, in a ¯ow phantom under controlled conditions, there was a 17% increase in the Doppler shift velocity without changes in actual ¯ow volumes. Such increases in measured blood velocities are most likely artifactual and re-
Vol. 16, No. 1, 2002
lated to the limited dynamic range of the ultrasound scanner.15 In our patients, in only one case of a CA with a 40% angiographic stenosis did the increase in PSV after contrast infusion change the classi®cation of the vessel into one with a hemodynamically signi®cant stenosis. Even here, it may be proposed that the duplex results re¯ect the true anatomic status of the vessel rather than the 2D measurement performed on the angiogram. Not enough clinical experience has yet been attained with echocontrast infusion to recommend changing the already established velocity criteria for diagnosing hemodynamically signi®cant stenosis. We found no change in the physiologic parameters of heart rate or blood pressure, or renal function as evidenced by BUN or creatinine levels, with per¯utren infusion (Figs. 3 and 4). Similarly, no clinically signi®cant changes in vital signs or blood chemistries were previously seen when the same agent was given at twice the dose of this study for evaluation of cardiac left ventricular opaci®cation.5,13 Additionally, no patient suffered any adverse events related to the contrast agent.
CONCLUSION Our study suggests that contrast-enhanced imaging of the mesenteric arteries is safe. Increased velocities are seen but they do not lead to false diagnosis of stenoses. Echocontrast is not routinely required but may be helpful when adequate imaging has not been achieved without it, such as in obese patients, postoperatively, or in the presence of much abdominal gas. Supported in part by an Educational Research Grant from The Peripheral Vascular Surgery REFERENCES 1. Monetta GL, Lee RW, Yeager RA, Taylor LM, Porter JM. Mesenteric duplex scanning: a blinded prospective study. J Vasc Surg 1993;17:79-86.
Contrast imaging of mesenteric arteries 83
2. Zwolak RM, Fillinger MF, Walsh DB, et al. Mesenteric and celiac duplex scanning: a validation study. J Vasc Surg 1998;27:1078-1088. 3. Harward TR, Smith S, Seeger JM. Detection of celiac axis and superior mesenteric artery occlusive disease with use of abdominal duplex scanning. J Vasc Surg 1993;17:738745. 4. Robbin ML, Eisenfeld AJ, et al. Per¯enapent emulsion: a US contrast agent for diagnostic radiology-multicenter doubleblind comparison with placebo. Radiology 1998;207:717722. 5. Kitzman DW, Goldman ME, Gillam LD, Cohen JL, Aurigemma GP, Gottdiener JS. Ef®cacy and safety of the novel ultrasound contrast agent per¯utren (De®nity) in patients with suboptimal baseline left ventricular echocardiographic images. Am J Cardiol 2000;86:669-674. 6. Perko MJ, Just S, Schroeder TV. Importance of diastolic velocities in the detection of celiac and mesenteric artery disease by duplex ultrasound. J Vasc Surg 1997;26:288293. 7. Stoney RJ, Cunningham CG. Acute mesenteric ischemia: clinical update. Surgery 1993;114:489-490. 8. Jager K, Bollinger A, Valli C, Amman R. Measurement of mesenteric blood ¯ow by duplex scanning. J Vasc Surg 1986;3:462-469. 9. Bowersox JC, Zwolak RM, Walsh DB, Schneider JR, et al. Duplex ultrasonography in the diagnosis of celiac and mesenteric artery occlusive disease. J Vasc Surg 1991;14:780788. 10. Healy DA, Neumyer MM, Atnip RA, Thiele BL. Evaluation of celiac and mesenteric vascular disease with duplex ultrasonography. J Ultrasound Med 1992;11:481-485. 11. Blebea J. Duplex ultrasound criteria for diagnosis of splanchnic artery stenosis or occlusion. J Vasc Surg 1992;16: 796-797. 12. Cotter B, Mahmud E, Kwan OL, DeMaria AN. New ultrasound contrast agents: expanding upon existing clinical applications. In: Goldberg BB, ed. Ultrasound Contrast Agents. St. Louis: Mosby, 1997, pp 31-42. 13. Weissman NJ, Cohen MC, Hack TC, Gillam LD, Cohen JL, Kitzman DW. Infusion versus bolus contrast echocardiography: a multicenter, open-label, crossover trial. Am Heart J 2000;139:399-404. 14. Schwarz KQ, Becher H, Schimpfky C, Vorwerk D, Bogdahn U, Schlief R. Doppler enhancement with SH U 508A in multiple vascular regions. Radiology 1994;l93:195201. 15. Forsberg F, Liu JB, Burns PN, et al. Artifacts in ultrasonic contrast agent studies. J Ultrasound Med 1994;13:357365.