Duplex ultrasonography in the diagnosis of celiac and mesenteric artery occlusive disease

Transcription

1 Duplex ultrasonography in the diagnosis of celiac and mesenteric artery occlusive disease Jon C. Bowersox, MD, PhD, Robert M. Zwolak, MD, PhD, Daniel B. Walsh, MD, Joseph R. Schneider, MD, PhD, Anne Musson, RVT, F. Elizabeth LaBombard, RVT, and Jack L. Cronenwett, MD, Lebanon, N.H. Duplex ultrasound criteria for the diagnosis of celiac and superior mesenteric artery (SMA) occlusive disease have not been well defined. We performed a blinded retrospective comparison of mesenteric duplex data with arteriography in 24 consecutive patients who underwent both studies. Arteriography revealed that eight superior mesenteric arteries were normal; five were minimally stenotic; eight had stenoses > 50%, and three were occluded. Nine celiac arteries were normal or minimally stenotic; 12 had stenoses - 50%, and three were occluded. Duplex scans were obtained after an overnight fast. In normal superior mesenteric arteries, peak systolic velocity (PSV) was cm/sec and end-diastolic velocity (EDV) was cm/sec. Superior mesenteric artery PSV in patients with minimal or no stenosis ( cm/sec) was less than PSV in patients with severe ( > 50%) stenosis ( cm/sec,p = 0.006), and less than PSV in patients with patent superior mesenteric arteries who underwent revascularization ( cm/sec, p = 0.017). Similarly, EDV was elevated in superior mesenteric m~teries with severe stenosis ( cm/sec, p = 0.001) and in patients who underwent revascularization ( cm/sec, p < 0.001) compared to those with <50% stenosis ( cm/sec, p = 0.001). An EDV > 45 cm/sec was the best indicator of severe stenosis (sensitivity, 1.0; specificity, 0.92). Peak systolic velocity > 300 cm/sec was less sensitive (0.63), but highly specific (1.0) for severe superior mesenteric artery stenosis. Triphasic superior mesenteric artery Doppler waveforms were present only in normal or minimally stenotic superior mesenteric arteries, making their absence sensitive (1.0), but not specific (0.46) for severe superior mesenteric artery stenosis. Normal superior mesenteric arteries had biphasic low resistance waveforms in the presence of replaced righ t hepatic arteries. Monophasic superior mesenteric arteries were found occasionally in less stenotic arteries in the presence of severe celiac stenosis or occlusion. Celiac arteries that were normal or minimally stenotic had low resistance biphasic waveforms with PSV = and EDV = , whereas stenotic celiacs had monophasic signals, variable velocities, and were often difficult to insonate adequately. Overall, eight patients underwent mesenteric revascularization, and each had an abnormal outcome on preoperative duplex examination. We conclude that mesenteric duplex ultrasonography is an effective diagnostic tool and should be considered early in the evaluation of patients with suspected chronic mesenteric artery occlusive disease. (J VASe SV0RG 1991;14:780-8.) Chronic mesenteric ischemia is a rare disease with no pathognomonic clinical findings. Diagnosis is often delayed because of the lack of an accurate and cost-effective screening test that could be applied safely to an elderly population. Diagnostic arteriog- From the Section of Vascular Surgery, Damnouth-Hitchcock Medical Center, Hanover. Presented at the Fifth Annual Meeting of the Eastern Vascular Society, Pittsburgh, Pa., May 2-5, Reprint requests: Robert M. Zwolak, MD, PhD, Section of Vascular Surgery, Dartmouth Hitchcock Medical Center, One Medical Center Dr., Lebanon, NH /6/ raphy is usually avoided until late in a patient's evaluation since relatively few patients referred for evaluation of chronic abdominal pain and weight loss will actually have significant mesenteric arterial occlusive disease. Duplex ultrasonography can be used to determine velocity and flow characteristics in the celiac and superior mesenteric arteries (SMA), and mesenteric arterial flow velocities have been well described in normal subjects in the fasting state, 1-a after eating, 4-6 and after pharmacologic stimulation.7,8 The use of duplex ultrasonography to evaluate patients for chronic mesenteric ischemia has been described only in anecdotal reports, 9,1 and in a

2 Volume 14 Number 6 December 1991 Duplex ultrasonograpy diagnosis of celiac and mesenteric artery occlusive disease 781 Table I. SMA and celiac artery waveform analysis SMA waveform Arteriogram Triphasic Biphasic Monophasic Nonvisualized < 50% stenosis (n = i3) Normal 5 2* 1 + (. = 8) Minimal stenosis (n = 5) -> 50% stenosis (n = 8) Occluded (n = 3) Celiac waveform < 50% stenosis (n = 9) -> 50% stenosis (n = 12) Occluded (n = 3) *Both patients had replaced right hepatic artery from SMA. + Includes one patient with coincident severe celiac stenosis/occlusion. limited number of case studies, s,11 Duplex diagnostic criteria for mesenteric artery stenosis have not been elucidated clearly, nor validated substantially by comparison to arteriographic data. The potential value of duplex ultrasonography as a diagnostic screening test for chronic mesenteric ischemia prompted us to compare the results of mesenteric duplex scanning with arteriographic findings to determine if reliable duplex ultrasound criteria could be developed for the identification of mesenteric arterial stenosis. METHODS This retrospective study compares the results of mesenteric duplex ultrasonography to measurements of celiac and SMA stenoses obtained from biplane arteriograms in 24 patients. The series is consecutive in that it represents all patients at the Dartmouth- Hitchcock Medical Center who underwent both tests within 30 days of each other between March 1988 and November The pretest working diagnoses were possible chronic mesenteric ischemia in 21 patients, celiac axis compression in two patients, and a known aortic dissection in one. All mesenteric duplex scanning was performed by registered vascular technologists in patients prepared with a bowel cathartic and an overnight fhst. A Diasonics DRF.-400 (Diasonics Inc., Milpitas, Calif.) duplex ultrasound instrument with a 3 MHz imaging probe and a 2.25 MHz pulsed Doppler was used in all studies. The celiac artery and SMA were interrogated, and peak systolic velocity (PSV), end-diastolic velocity (EDV), Doppler spectral waveform, and the angle of insonation were recorded. Triphasic waveforms were interpreted as having a forward systolic phase, a distinct: reverse velocity phase, and a tertiary antegrade phase. Biphasic waveforms had two antegrade flow phases with a distinct end-systolic notch but no flow reversal, typical of low resistance outflow beds. Monophasic waveforms had a single phase antegrade flow without an end-systolic notch. The 24 biplane arteriograms were examined by observers unaware of the duplex scan results. Diameters of the celiac artery and SMA were measured from lateral aortograms. The percent stenosis was calculated as the narrowest vessel diameter divided by the diameter of the nearest normal segment distal to the stenosis multiplied by 100. Since the goal of the study was to determine duplex criteria that would disnnguish hemodynamically significant stenoses, the duplex results of patent arteries with 50% or greater diameter reduction on arteriogram were defined as severely stenotic and were compared to the duplex results of arteries that were normal or had stenoses causing less than a 50% diameter reduction. Arteries with total occlusion on artetiogram were also analyzed separately. Data are expressed as mean _+ SEM, and comparisons between groups were performed with the Student t test. Sensitivity and specificity were determined for PSV, EDV, and spectral waveforms. Positive test thresholds were selected by receiver operating characteristic (ROC) analysis to maximize test sensitivity. 12 RESULTS Arteriography revealed the SMA to be normal in eight patients and minimally stenotic in five patients, whereas eight patients had severe stenoses, and three had SMA occlusions. Nine patients had normal or minimally diseased celiac arteries, whereas 12 celiac arteries were severely stenotic, and three were occluded. Ten patients had significant disease in both

3 Journal o f VASCULAR SURGERY 782 Bowersox et al ~,~... ii~ Fig. 1. A, Normal fasting triphasic SMA waveform with PSV = 160 cm/sec, EDV = 15 clyl/sec. Fig. 1. B, Biphasic SMA waveform in patient with 37% diameter reduction by arteriography. PSV = 210 cm/sec, EDV = 28 cm/sec. Fig. 1. C, Monophasic SMA waveform with aliasing and diffuse spectral broadening in patient with a 94% SMA stenosis. PSV estimated at 235 cm/sec by addition of the aliased peak signal to the truncated base. EDV = 88 cm/sec. Fig. 1. D, Reconstituted SMA waveform beyond a complete occlusion reveals a blunted systolic upstroke and marked turbulence. Velocity was not measurable because of inappropriate angle of insonation. the celiac artery and SMA, but none had occlusion of both vessels. In two of the three patients with celiac occlusion, retrograde flow was identified in the gastroduodenal artery implying collateral flow from the SMA to the celiac outflow tract. A replaced right hepatic artery was identified in four patients (16%). Duplex SMA waveform analysis revealed that five of the eight normal arteries were triphasic (Table I, Fig. 1, A). Three of the normal SMAs had lowresistance waveforms. This finding occurred in two patients with replaced right hepatic arteries originating from the SMA, and in one patient with coincident severe celiac occlusive disease. O f the five patients with minimal SMA stenosis, one had a triphasic Doppler waveform, three were biphasic, and one was monophasic (Fig. 1, B). Coincident celiac disease was present in one of the patients with a biphasic waveform and the single patient with a monophasic waveform. Among the eight patients with severe SMA stenosis, six had monophasic Doppler waveforms (Fig. 1, C), whereas two had biphasic waveforms. Doppler velocity data in the eight completely normal SMAs revealed a PSV of 134 _ 18 cm/sec and an EDV of 24 _ 4 cm/sec, whereas those with a measurable stenosis causing less than a 50% diameter reduction had a PSV of cm/sec (p = 0.02 compared with normal SMAs) and an EDV of 41 _ 13 cm/sec (p = NS compared with

4 Volume 14 Number 6 December 1991 Duplex ultrasonograpy diagnosis of celiac and mesenteric artery occlusive disease 783 E 700; , > /) 300, n 200, ~ 14o~ ~ 120, 100, > 80-" e~ uj , 0!! Stenosis: < 50% _> 50% < 50% > 50% Fig. 2. Scatter plots of fasting SMA 'PSV, left), and (EDV, right) in 13 normal or minimally stenotic arteries, and 8 severely stenotic ( -> 50%) arteries. Table II. Mesenteric duplex velocity measurements Peak systolic End-diastolic A~.riogram No. velocity cm/sec velocity cm/sec SMA < 50% stenosis 13 _> 50% stenosis 7 p value Celiac < 50% stenosis 6 -> 50% stenosis 6 p value [ ] 30 ± 6 [18-43] 299 ± 40 [ ] 78 _+ 11 [51-105] [50-253] [23-58] 132 ± 27 [66-199] [3-115] Values shown arc mean -+ SEM (95% confidence interval). Stenotic is > 50% diameter reduction. normal SMAs). Considered as a group, the 13 patients with normal or minimally stenotic SMAs had PSV= 171 _ 22 cm/sec and EDV= 30 +_ 6 cm/sec (Table II). Velocity data were excluded in one severely stenotic SMA because of an angle of insonation > 60 degrees, but PSV in the remaining seven stenotic SMAs was 299 _+ 40 cm/sec and EDV was 78 _+ 11 cm/sec, both values being significantly greater than those of patients vcith no or minimal stenosis (PSV, p = 0.006; EDV, p = 0.001; Table II, Fig. 2). The mean angle ofinsonation in the SMAs was 55 degrees _+ 5 degrees, and no difference was observed between groups. Three SMAs were found to be totally occluded at the origin by arteriography. Two of these total occlusions were identified as such by duplex scanning. Monophasic waveforms with severely bhmted systolic upstroke, low velocity, and marked spectral broadening were found distally in these arteries at the region of reconstimtion by collateral vessels (Fig. 1, D). Overall, the most accurate predictor of a significant SMA stenosis was a fasting EDV > 45 cm/sec, with a sensitivity of 1.0, specificity of 0.92, and an accuracy of A fasting PSV > 300 cm/sec was less sensitive (0.63), but highly specific (1.00), with an accuracy of The absence of a triphasic waveform was sensitive for the presence of disease (1.0), but less specific (0.46), with an accuracy of 0.67 (Table III). Arteriography revealed the celiac artery to be normal or minimally stenotic in nine patients. Seven of these were identified by duplex ultrasonography, with a typical low resistance biphasic waveform in six,

5 784 Bowersox et al. Journal of VASCULAR SURGERY Table III. Validity of mesenteric duplex ultrasound in the identification of significant SMA stenosis Test parameter Sensitivity Specificity Accuracy Fasting EDV > cm/sec Fasting PSV > cm/sec Absence of tripha sic fasting SMA waveform and a monophasic waveform in one (Table I, Fig. 3), whereas two normal celiac arteries were not visualized well enough for positive identification. Six of the 12 stenotic arteries were visualized adequately by duplex ultrasonography, and each had a monophasic waveform. One of the three celiac occlusions was also identified as such by duplex ultrasonography, whereas one was not visualized, and one was felt to contain a very blunted low velocity monophasic waveform. Doppler velocity data were excluded in one of the visualized nonstenotic celiac arteries for an insonation angle >60 degrees, leaving six celiac arteries without severe stenosis and six stenotic celiac arteries for analysis. The mean insonation angle for these 12 celiac arteries was 43 degrees _+ 6 degrees without difference between groups. Peak systolic velocity in the normal and minimally stenotic celiac arteries was cm/sec, and EDV was cm/sec. Most of the severely stenotic celiac arteries had elevated PSV and EDV, but two had low velocities. Although these waveforms had a poststenotic appearance, no high velocity region could be identified proximally. Thus mean PSV (132 +_ 27 cm/sec) and EDV ( cm/sec) in severely diseased celiac arteries was no different from the comparison group (PSV,p = 0.69; EDV,p = 0.42; Table II). No value for celiac PSV or EDVcould be identified that would discriminate clearly between minimal and severe stenosis. The most sensitive predictor of celiac stenosis was the absence of a low resistance biphasic waveform, but our inability to visualize several stenotic celiac arteries precluded application of numeric accuracy methods. Based on a global clinical picture of severe symptoms and documented mesenteric stenoses and occlusions, eight patients underwent mesenteric revascularization. Each of these patients was considered abnormal on preoperative duplex ultrasonography. Five of the eight had patent SMAs before operation with mean PSV = cm/sec and EDV = cm/sec, values that were significantly greater than those with no or minimal stenosis (PSV p = 0.017, EDV p < 0.001). No patient in the group operated on had either an SMA or a celiac artery, which was called normal by duplex ultrasonography, although four of the celiac arteries were not identified clearly enough to derive velocity information. Six revascularized patients were discharged from the hospital and had relief of symptoms at follow-up, whereas two patients died after operation, one of a massive stroke and one of cardiac dysrrthymia. DISCUSSION Although the natural history of chronic mesenteric ischemia is not well defined, its potential lethality in untreated patients mandates use of safe, cost-effective, and accurate diagnostic tests. 13 Noninvasive methods reported previously for the assessment of chronic mesenteric ischemia have had insufficient accuracy to be useful as screening tests for this disease. Arteriography is sensitive and specific for identifying mesenteric arterial occlusive disease, but cost and risks associated with its use in an older, debilitated patient population preclude its application as a screening technique. The application of duplex technology has been extended to intraabdominal imaging. Investigators have shown that mesenteric arterial blood flow can be measured accurately in healthy volunteers by duplex scanning. >6 In this study we documented the feasibility and accuracy of mesenteric duplex scanning in a generally ill and elderly patient population. The data presented here deal only with duplexderived Doppler waveform and velocity. The low frequency transducers required for insonation of these arteries provided little information regarding plaque morphology, and B-mode was used primarily as a guide for placement of the Doppler sample volume. Superior mesenteric artery velocity data obtained from patients with normal mesenteric arteriograms (PSV = cm/sec, EDV = 24 _+ 4 cm/sec) were slightly greater than those reported by Flinn et al., 1 Moneta et al., 6 and Jager et al., 4 but less than those reported by Nicholls et al.s or Lilly et al. 7 The explanation for this interlaboratory variability is probably muttifactorial. As emphasized by Rizzo et al. a4 maintenance of an insonation angle < 60 degrees is important in preventing artifactual elevation of flow velocities, and we adhered to this principle during analysis of our data. In addition, we found that mean PSV from minimally stenotic SMAs was significantly greater than normal PSV,

6 Volume 14 Number 6 December 1991 Duplex ultrasonograpy diagnosis of celiac and mesenteric artery occlusive disease 785 i ~ ~L~ "~ i~i/i l~r~w~ GP t'l 3 5 ':' 2C41 ~e~ ; i r: r:? : ~RF ~;~[Kz ~PF 208 HZ 44 k~ F~ Fig. 3. A, Normal biphasic celiac waveform consistent with low resistance outflow. PSV = 178 cm/sec, EDV = 44 cm/sec. Fig. 3. B, Monophasic celiac waveform with signal aliasing and marked spectral broadening in the presence of a 73% diameter reduction. PSV estimated at 250 cm/sec, EDV = 126 cm/sec. Fig. 3. C, Monophasic celiac waveform with blunted systolic upstroke, spectral broadening and low PSV, felt suspicious for a poststenotic signal. However, no high velocif signal could be identified proximally. and inclusion of EDV from minimally stenotic SMAs influenced the mean EDV of the normal group upward, emphasizing the importance of' arteriographic confirmation when defining normal values. However, precise insonation of deep abdominal arteries offers still other potential sources of error, including variable sampling position on the artery, poor signal separation when SMA and celiac origins are closely apposed, influence of anatomic variants and collateral flow, and even misidendficadon of vessels. Although further efforts at standardization are needed, the apparent variations in published normal SMA and celiac velocities emphasize the importance of ongoing quality control and arteriographic correlation by individual laboratories performing these studies. In evaluation of patients with possible mesenteric artery occlusive disease, one must distinguish severely stcnotic SMAs from those with low resistance anatomic variants, and those in whom increased SMA velocity is due to collateral flow. We observed spectral broadening, loss of diastolic reversed flow, and substantial elevations of PSV and EDV in patients with significant SMA stenosis, whereas loss of diastolic reversed flow with modest velocity elevations and less spectral broadening were more characteristic of less stenotic SMAs with replaced right hepatic arteries or collateral flow to the celiac outflow bed. Although previous authors have suggested useful diagnostic information could be gained from postprandial studies, we found that data derived from the SMA in fasting patients was adequately sensitive and specific to identify patients with clinically significant mesenteric occlusive disease. This

7 786 Bowersox et al. Journal of VASCULAR SURGERY finding is useful since patients with severe postprandial pain are often hesitant to drink a test meal. In addition, it has reduced the time required to complete the study. Specific diagnostic velocity criteria for the identitication of celiac artery stenosis were not developed in this study. In contrast to the SMA, where tight stenoses were uniformly indicated by elevated velocities, several tightly stenotic celiac stenoses actually had blunted arterial waveforms and diminished velocities. Although these may have been poststenotic waveforms, we were unable to locate a high velocity focus more proximally. Alternatively, it may be that generous collateral circulation provided adequate inflow pressure from the SMA, resulting in a low pressure gradient as an explanation of the apparently paradoxical absence of increased velocity across a tight celiac stenosis. In support of this concept, we observed reversed flow in the common hepatic artery by duplex insonation in several patients with complete celiac occlusion. We are investigating further the efficacy of hepatic artery duplex evaluation in this patient cohort. In conclusion, we found that duplex ultrasonography of the mesenteric arteries is a useful screening test for patients with symptoms suggesting chronic mesenteric artery occlusive disease. The absence of a triphasic SMA waveform, a fasting EDV >45 cm/sec, and a fasting PSV > 300 cm/sec were useful thresholds in the identification of patients with severe SMA stenosis. Interpretation of waveform and velocity information must be considered in light of possible anatomic variants and the influence of collateral blood flow as a result of celiac stenosis or occlusion. These conclusions are based on a retrospective analysis of 24 duplex ultrasound examinations, and prospective validation studies will be necessary to determine their true clinical usefulness. REFERENCES 1. Taylor KJW, Burns PN, Woodcock JP, Wells PNT. Blood flow in deep abdominal and pelvic vessels: ultrasonic pulsed- Doppler analysis. Radiology 1985;154: Qamar MI, Read AE, Skidmore R, Evans JM, Williamson RCN. Transcutaneous Doppler ultrasound measurement of coeliac axis blood flow in man. Br J Surg 1985;72: Qamar MI, Read AE, Skidmore R, Evans JM, Wells PNT. Transcutaneous Doppler ultrasound measurement of superior mesenteric artery blood flow in man. Gut 1986;27: Jager K, Bollinger A, Valli C, Ammann R. Measurement of mesenteric blood flow by duplex scanning. J VASC SURe 1986;3: Nicholls SC, Kohler TR, Martin RL, Strandness DE Jr. Use of hemodynamic parameters in the diagnosis of mesenteric insufficiency. J VASC SUR6 1986;3: Moneta GI, Taylor DC, Helton WS, Mulholland MW, Strandness DE Jr. Duplex ultrasound measurement of postprandial intestinal blood flow: effect of meal composition. Gastroenterology 1988;95: Lilly MP, Harward TRS, Flinn WR, Blackburn DR, Astleford PM, Yao JST. Duplex ultrasound measurement of changes in mesenteric flow velocity with pharmacologic and physiologic alteration of intestinal blood flow in man. J VAsc SURG 1989;9: Van Bel F, Van Zoeren D, Schipper J, Guit GL, Baan J. Effect of indomethacin on superior mesenteric artery blood flow velocity in preterm infants. J Pediatr 1990;116: Harmer GG. Persistent mesenteric ischaemia in a young woman. Br Med J 1987;295: Flinn WR, Sandager GP, Lilly M, Yao JST, Bergan JJ. Duplex scan of celiac and mesenteric arteries. In: Bergan JJ, Yao JST, eds. Arterial surgery: new diagnostic and operative techniques. Orlando: Grune and Stratton, 1988: Jagar ICA, Former GS, Thiele BI, Strandness DE. Noninvasive diagnosis of intestinal angina. J Clin Ultrasound 1984;12: McNeil BJ, Keeler E, Adelstein SJ. Primer on certain elements of medical decision making. N Engl J Med 1975;293: Dunphy JE. Abdominal pain of vascular origin. J Med Sci 1936;192: Rizzo RJ, Sandager G, Astleford P, et al. Mesenteric flow variations as a function of angle of insonation. J VASC SURG 1990;11: Submitted May 13, 199i; accepted Aug. 20, DISCUSSION Dr. Michael Zatina (New Brunswick, N.J.). This paper describes the problem ofmesenteric occlusive disease as a rare entity, and I think it might be better termed an underdiagnosed or underrecognized entity. Therefore, I think that any work done to evaluate a noninvasive modality that will help us determine whether or not this patient may have mesenteric occlusive disease is important. In fact, our own interest in magnetic resonance angiogra- phy developed because of this very problem, being frustrated with finding people with acute or chronic mesenteric disease and having physicians reluctant to obtain invasive arteriograms to make the diagnosis. Mthough we think that the presence of a triphasic waveform in the superior mesenteric artery indicates a normal vessel, in fact it indicates that the vessel has only less than a 50% stenosis. And in this arterial bed with a celiac

8 Volume 14 Number 6 December 1991 Duplex ultrasonograpy diagnosis of celiac and mesenteric artery occlusive disease 787 access, an SMA, and an inferior mesenteric artery, all having the ability to give arterial supply to the gut, I am not so sure that the presence or absence of a stenosis of 40% is not all that important. So I would just use some caution there in that interpretation. They do show very nicely that the presence of an EDV of greater than 45 cm/sec has a 95% accuracy. So I t~hink that there are some good points and I think that you have shown that this has a potential to be a very worth~a&ile noninvasive study. I have three questions. Number one, where do you insonate? You have shown us exactly what angle we have to use on our probes, but is this a situation analogous to carotid artery disease where we have to do our duplex measurements just distal to the stenosis? Is there a certain part of the vessel that we have to examine or can we examine any part of the SMA? Number two, have you correlated these data with the presence or absence of collateral flow seen on your arteriograms? ][ am a little nervous about coming out with specific numbers for the presence or absence of disease in the superior mesenteric artery when a rich collateral network exists from, the celiac access, a rich collateral network, a potential network from the inferior mesenteric artery and the meandering mesenteric artery. Do you have data, have you looked at those arteriograms and then looked at your numbers to determine what the presence or absence of those collamrals do to these numbers? And finally, having operated on several patients that have been missed because of acute SMA occlusion, I think that is a very important problem as well, and that is, the patients come in, they are very ill, the internist is reluctant to obtain the arteriogram; that is the patienfs only chance for survival. Have you looked at any patients with acute mesenteric arterial occlusion to see if you can make similar determinations about the presence or absence of that problem? Dr. Ion Bower sox. With regard to the first point, where we chose to insonate the artery, we did approach it in a similar manner to carotid artery disease. We were dealing primarily with orificial lesions, we identified the aorta, and then began insonating with a postoperative probe from that point. I think one of our key points, as was pointed out by the Northwestern group, was to maintain an angle of insonation less than 60 degrees, and although I did not describe it in my presentation, in the manuscript the angle of insonation was kept at 55 degrees, plus or minus 5 degrees for the SMA, and 43, plus or minus 6 degrees for the celiac artery. And indeed we proceeded using those criteria until we identified turbulent flow with increased PSVs and EDVs. We did not specifically obtain images further distal to that point where we identified disease. With regard to the second point, the identification of collaterals and how that influences the data, we did re,aew the arteriograms with regard to this matter, not so much regarding the inferior mesenteric artery collateral circulation, which I do not know the value of, but particularly with regard to the celiac SMA collateral situation. And as I touched on briefly, in the presence of a replaced right hepatic artery or severe celiac disease with retrograde flow in the gastroduodenal artery, it does seem to affect the SMA waveform. And the final point I think is probably the most significant one, and that is the application of the technique to patients with potentially acute mesenteric ischemia. Although the use of duplex ultrasonography has been well integrated into our practice at Dartmouth with regard to evaluating patients with chronic mesenteric ischemia, we specifically have not used this in clinical evaluations of patients with acute mesenteric ischemia for several reasons. When the patients come in, the physician is usually working against the clock at that point. Unlike ultrasound evaluation in the emergency room for ruptured aneurysms that takes just a few minutes, the evaluation of the patient with mesenteric ischemia can take anywhere from 30 minutes to an hour and a half to thoroughly insonate the mesenteric vessels. It is hard to determine a priori the length of time that will be required. Second, these patients have not undergone a bowel preparation, which I think would make the technique significantly more difficult and potentially miss significant findings. The patients are uncomfortable. And finally, a significant number of patients come in with SMA embolus distal to the origin or the take-off from the aorta, and we are concerned about missing this group of patients. So I think this clearly has to be emphasized that this is an investigational application if it were to be applied to the acutely ischemic patient, but I think it is one that merits application in a research setting potentially. Dr. Brian Thiele (Hershey, Pa.). I have a couple of critical comments and a word of caution. We in fact have been interested in this problem as well and published a series of 25 cases reported at the American Heart Association meeting last year. I think the title of this paper perhaps should be altered because in fact what it addresses is the presence of disease in the SMA and not the presence of chronic mesenteric ischemia. In fact we have been unable to correlate the distribution of disease with the presence of truly classic intestinal symptoms. The word of caution I raise is that in the series we have looked at we have had a great deal of difficulty because of the issue of the interaction between these two circulations. In fact what we found was that as the SMA velocity increased, so did the celiac artery velocity increase, even in normal arteries, and that led to us overcalling the presence of disease in the celiac artery. So the status of the collateral circulation is really a major factor, and we have not been able to identify any way of evaluating this carefully to give us accurate information about what is really going on at the origin of the visceral artery. I agree with you, and our findings are almost identical, that if you look at isolated SMA disease, this technique is very usefill, but if you want to look at the whole mesenteric circulation, some major problems are still to be veorked out. I noted that you deleted the celiac data in the abstract,

9 788 Bowersox et al. Journal of VASCULAR SURGERY and I guess it is partly because you could not get good data or it did not correlate, and that is really my question. Dr. Bowersox. I appreciate those comments. I agree, Dr. Thiele, that it should be emphasized that it is difficult to correlate the degree of mesenteric stenosis with symptoms, but I emphasize that this duplex ultrasound examination was applied only in patients with symptoms on admission. It was not a population screening technique. And our real thrust was developing a screening tool that could reduce the number of patients who underwent mesenteric arteriography and perhaps bring those patients who deserved or required mesenteric arteriography to the definitive procedure earlier. So in its use as a screening technique, I think we were most concerned about detecting the presence of SMA stenosis, and I think it does accomplish that objective. The celiac data, as you pointed out, did not add, in our study, the ability to do that. Dr. All Aburahma (Charleston, W. Va.). The first concern I had is what Dr. Thiele already indicated about the correlation of celiac and SMA. That has already been answered. The second one is, I do not know whether you had better hands or better technicians or better physicians, but for over 2 or 3 years we have been unable to master the technique. Do you have any recommendations regarding this? Dr. Bowersox. The technologists who do the mesenteric examinations in our laboratory are very experienced in abdominal ultrasound imaging, and dearly it is a different technique. Only two of our six technologists routinely do the abdominal imaging, and it is truly an art. I think that will be one of the limitations in widely applying this technique. Dr. Michael Silane (New York, N.Y.). I would also like to inject a word of caution. In my opinion, looking at the origin of the SMA or the celiac artery, whether you do it by duplex scanning or arteriography, does not necessarily rule out mesenteric ischemia. I have seen chronic visceral ischemia with distal SMA disease that was surgically corrected. So until you get a fiall arteriograrn and look at the entire mesenteric vessel, I do not think you can rule out mesenteric ischemia. Dr. Bowersox. Of the 12 of our 25 patients that we evaluated arteriographically and with duplex examinations that did have significant disease, it was all orificial in nature, but I think that your comment is valid. A number of people have disease more distal in the mesenteric vessels, and this is probably a limitation of that technique. Dr. Joseph Van DeWater (New Hyde Park, N.Y.). As is being emphasized again and again today, it is important that we treat patients and not lesions. I think you have the beginning of something here, and I urge you to take it to the laboratory and see if you can come up with a good noninvasive test and learn what this means to actual intestinal integrity. Dr. Bowersox. A number of groups have been approaching this, and that is where the wealth of data, parti~larly on normal subjects and changes that occur with feeding, are derived from that. We were specifically interested in applying these data that had come out of the investigative laboratory to a subset of patients, and those are people who were admitted with symptoms of chronic mesenteric ischemia and how it could be applied specifically in that setting. BOUND VOLUMES AVAILABLE TO SUBSCRIBERS Bound volumes of the JOURNAL OF VASCULAr, SuRa~r,y for 1991 are available to subscribers only. They may be purchased from the publisher at a cost of $ for domestic, $ for Canadian, and $75.00 for international subscribers for Vol. 13 (January to June) and Vol. 14 (July to December). Price includes shipping charges. Each bound volume contains a subject and author index, and all advertising is removed. Copies are shipped within 60 days after publication of the last issue in the volume. The binding is durable buckram with the journal name, volume number, and year stamped in gold on the spine. Payment must accompany all orders. Contact Subscription Services, Mosby-Year Book, Inc., Wesdine Industrial Drive, St. Louis, MO , USA. In the United States call toll free: (800) , ext In Missouri call collect: (314) , ext Subscriptions must be in force to qualify. Bound volumes are not available in place of a regular Joue, N~ subscription.