Technique of Mesenteric Duplex Scanning

To minimize bowel gas interference, mesenteric duplex scanning is best performed early in the day with the patient fasting. The patient is supine, and the head of the bed is elevated to 30 degrees. B-mode imaging and Doppler insonation of the visceral vessels is required. Doppler frequencies are typically in the range of 2.0 to 5.0 MHz. Gray-scale and color-flow imaging (Figure 1) are used to identify and follow the selected vessel segments and to note the presence or absence of any disease process within the vessel lumen. Doppler evaluation quantifies the severity of the disease and should include assessment for the presence or absence of flow and, when flow is present, evaluation of peak systolic velocity (PSV) and end diastolic velocity (EDV). Spectral analysis should be obtained in all vessel segments. Measurements should be obtained proximal, throughout, and distal to any flow disturbance identified. All spectral derived velocity information should be derived with a Doppler insonation angle between 45 and 70 degrees. Angles of insonation exceeding 70 degrees result in significant artificial elevation of velocities.

Fasting Doppler Mesenteric Artery Waveforms

Fasting SMA and CA waveforms differ. Fasting peak-systolic velocities in the SMA tend to be higher than those in the celiac artery. Fasting end-diastolic velocities also tend to be lower in the SMA than in the CA. Peak systolic velocities in the SMA in angiographically normal vessels vary from 125 to 170 cm/sec. A reverse flow component is often present at the end of systole in the SMA but not in the CA (Figure 2). The increased diastolic flow present in the CA compared to the SMA likely reflects the decreased end organ resistance of the hepatic and splenic circulations versus that of the fasting intestinal circulation.

Fasting Duplex Criteria for Mesenteric Artery Stenosis

Establishment of criteria for diagnosis of CA and SMA stenosis began in the late 1980s. The initial study, a retrospective review of 34 patients, compared mesenteric duplex scans and arteriograms of these arteries. This study suggested that a SMA peak systolic velocity (PSV) of at least 275 cm/sec or no flow signal and a CA PSV at least 200 cm/sec or no flow signal were reliable indicators of a 70% or greater angiographic stenosis of the SMA and CA, respectively.

Based on these initial observations, investigators at the Oregon Health and Science University designed a prospective blinded study to compare mesenteric duplex scanning and lateral arteriograms to validate proposed duplex criteria for splanchnic artery stenosis. During an 18-month period, 100 consecutive patients underwent routine mesenteric artery duplex scanning and lateral abdominal aortography. Aortograms were evaluated for the presence or absence of a 70% or greater stenosis in the CA or SMA.

A 70% to 100% CA stenosis was found in 24 patients, and 76 patients had less than 70% CA stenosis. There were 14 patients with 70% to 100% lesions in the SMA, and 85 SMAs had less than 70% angiographic stenosis. Comparing PSVs and EDVs in the patients with less than 70% angiographic stenosis versus those with more than 70% angiographic stenosis, PSVs and EDVs were found to be higher in the patients with greater than 70% stenosis in the CA and/or SMA (Figures 3 and 4).

There were 24 angiographic high-grade CA lesions (>70%). Duplex scanning identified 20 out of 24 of these lesions for a true positive rate of 83%. There were three false negatives (13%). This yielded a sensitivity of 87%, a specificity of 80%, a positive predictive value of 63%, and a negative predictive value of 94% for a PSV of at least 200 cm/sec as representing a 70% or greater angiographic stenosis of the CA. The overall accuracy was 82%.

There were 14 high-grade SMA angiographic stenoses, and 12 of these were identified by duplex scanning for a true positive rate of 86%. There was one false negative (7%). This yielded a sensitivity of 92%, a specificity of 96%, a positive predictive value of 80%, and a negative predictive value of 99% for a PSV of at least 275 cm/sec as representing a 70% or greater angiographic stenosis of the SMA. The overall accuracy for detecting a 70% or greater angiographic SMA stenosis was 96%.

There are other duplex criteria for stenosis in CAs and SMAs. The Section of Vascular Surgery at the Dartmouth Hitchcock Medical Center conducted a study to validate the accuracy of their previously established duplex ultrasound criteria for a 50% or greater SMA or CA stenosis by comparison with arteriography. Duplex criteria established retrospectively in their laboratory in 1991 identified an end-diastolic velocity of 45 cm/sec or greater, or no detectable flow in the SMA as highly sensitive (100%) and specific (92%) for an SMA 50% or greater angiographic stenosis or occlusion. For the CA, accurate velocity thresholds were not identified, but it was noted that retrograde common hepatic artery flow was highly predictive of severe celiac artery stenosis or occlusion.

To validate the accuracy of their proposed duplex ultrasound criteria for a 50% or greater mesenteric artery stenosis, the Dartmouth group reviewed 243 mesenteric duplex scans. Angiographic confirmation was available for 46 patients. SMA and CA stenoses were measured on lateral aortograms, and the original duplex diagnostic criteria evaluated for accuracy. Receiver operator characteristic curve analysis was also performed on the velocity data to identify the most accurate velocity thresholds in the new data. For the SMA, an EDV of 45cm/sec or greater again provided the best sensitivity (90%) and specificity (91%) for detecting a greater than 50% stenosis. The positive predictive value was 90%, the negative predictive value was 91%, and the overall accuracy 91%. With respect to the celiac artery, retrograde common hepatic artery flow direction was 100% predictive of severe celiac artery stenosis or occlusion. EDVs at least 55 cm/sec or no flow signal in the artery had the best overall accuracy (95%), with high sensitivity (93%) and specificity (100%). A PSV of greater than 200 cm/sec or no flow signal also had an excellent accuracy (93%), sensitivity (93%), and specificity (94%). The Dartmouth group noted mesenteric anatomic variants could be identified by duplex ultrasonography, with three of four anatomic anomalies correctly identified.