Abdom Imaging 23:364–369 (1998)

Abdominal Imaging q Springer-Verlag New York Inc. 1998

Sonographic and Doppler assessment of the inferior mesenteric artery: normal morphologic and hemodynamic features P. Mirk, G. Palazzoni, A. R. Cotroneo, C. di Stasi, A. Fileni Institute of Radiology, Catholic University of the Sacred Heart, Largo Agostino Gemelli 8, 00168 Rome, Italy Received: 20 November 1996/Revision accepted: 16 April 1997

Abstract Background: We wanted to evaluate prospectively the feasibility of sonographic and Doppler assessment of the inferior mesenteric artery (IMA) and to provide data on its normal morphological and hemodynamic characteristics. Methods: Sonography and Doppler study of the IMA were performed on 116 patients without splanchnic vessel pathology. Vessel diameter, systolic, diastolic, and time-averaged mean flow velocities, pulsatility index values, and flow volumes were correlated with patient age (õ50 years vs. ¢50 years) using the Wilcoxon rank-sum test. Findings were verified by splanchnic angiography in 11 cases. Results: Technically valid studies were obtained in 103/ 116 cases (88.8%). Flowmetric data showed high peripheral resistance (mean { SD: systolic flow velocity, 1.41 m/s { 0.48; minimal diastolic flow velocity, 0.10 m/s { 0.16; pulsatility index, 3.49 { 0.49). Mean flow volume calculated in 80 cases was 0.13 L/min { 0.06. Older subjects presented significantly higher time-averaged mean flow velocities and lower resistance than those younger than 50 years. Conclusions: The success rate for sonographic and Doppler study of the IMA is similar to that observed with larger splanchnic vessels. Knowledge of its normal characteristics is necessary for recognition of pathological conditions and for studies of its physiological behavior. Key words: Inferior mesenteric artery, diagnosis— Mesenteric blood flow, Doppler studies—Splanchnic arterial circulation, functional studies—Intestinal ischemia.

Correspondence to: P. Mirk

Most of the Doppler ultrasound studies of splanchnic arteries reported thus far have been limited to the celiac artery and its principal branches (hepatic and splenic arteries) and the superior mesenteric artery (SMA) [1– 7]. Visualization of the inferior mesenteric artery (IMA) is generally felt to be problematic because of its smaller caliber and relatively inaccessible location in the lower left quadrant of the abdomen; in most cases, no attempt is made to include this vessel in studies of the splanchnic circulation. We began to use Doppler ultrasound to examine the IMA in 1993 and found that the success rate and level of accuracy associated with this modality are quite similar to those observed in studies of the celiac and superior mesenteric arteries [8]. The present report describes the techniques used for real-time sonography, color and pulsed-wave Doppler study of this vessel, and its normal morphological and hemodynamic features based on examination of 116 subjects.

Materials and methods Population Real-time sonography and Doppler ultrasound examination of the IMA were performed during scheduled abdominal ultrasound studies in a consecutive series of 181 inpatients (after exclusion of those in critical condition to avoid prolonging the examination and those with suspected chronic intestinal ischemia or acute inflammatory diseases involving the large bowel). Informed consent was obtained in all cases. Of the 181 patients who were studied, 18 were excluded from analysis because both Doppler study and subsequent angiography indicated intrinsic pathology of the IMA itself (obstruction, stenosis, compensatory hypertrophy, etc.). Forty-seven others had vascular disease involving other arteries (aneurysm of the abdominal aorta, peripheral vascular disease, arteriosclerotic changes in carotid vessels). Although none of these subjects presented clinical signs (e.g., angina abdominis) or Doppler evidence of splanchnic involvement, they also were excluded from the present series.

4v12 0019 Mp 364 Tuesday Jun 16 10:36 AM SV-Ab Image (v. 23, #4) 0019 (1515)

P. Mirk et al.: Inferior mesenteric artery: Doppler study Thus, the present report is based on studies performed in 116 patients (53 men, 63 women) ranging in age from 12 to 87 years (mean { SD: 49.5 { 16.8 years). Body weights were 40–97 kg (64.3 { 11.8 kg) and heights were 150–188 cm (167 { 7.8 cm). Sonography of the abdomen had been requested for various reasons, including evaluation of parenchymal abdominal organs, suspected biliary tract or kidney stones, exclusion of hepatic metastases, or lymphadenopathy for tumor staging or hematological disease.

Technique After completion of the requested sonographic examination, all of which were performed after an overnight fast, transverse and longitudinal scans of the IMA were made by the same radiologist (P.M.) using Toshiba SSA 270A and 340 ECC scanners (Tokyo, Japan) with a curved linear 3.75-MHz probe. During the study, the patient remained in the supine position and was allowed to breathe normally. The artery, which is situated near the midline in the lower left quadrant of the abdomen, is often concealed by gas in the overlying intestine. As a consequence, in most patients, increased pressure had to be applied to the probe to displace the overlying intestinal loops by using a technique similar to that which has been described for studies of the appendix [9]. By applying such a graded compression, the IMA in favorable subjects was readily identified simply by B-mode sonography. In difficult (obese or excessively gaseous) subjects, color Doppler was used to assist localization of the IMA. The diameter, origin, and course of the IMA and the length of the segment to be visualized were recorded in all cases in which the artery could be seen (Fig. 1). Doppler scans were made in the longitudinal plane, and the orientation of the probe was modified according to the course of the artery so that flow velocities could be corrected according to the angle of insonation. The patient was instructed to exhale and suspend respiration during recording of the tracings. Velocity waveforms were obtained by sampling the artery within its straight segment, 3–5 cm distal to the origin of the IMA from the aorta, with an angle of insonation less than 607 with respect to the long axis of the artery. Selection of this sampling site rather than close to or at the origin of the aorta was based on technical considerations. After its origin, the artery bends downward and then courses parallel to the aorta. Thus, sampling in its proximal portion makes determining the exact angle of insonation for Doppler analysis difficult. Moreover, aortic pulsations cause strong motion artifacts. In contrast, sampling far from the origin of the IMA allows one to avoid errors in the angle correction, which would affect calculations of velocity and flow volume, and motion artifacts from the aortic walls, which would interfere with pulsed Doppler tracings. The pulse repetition frequency was optimized to record medium to high arterial velocities. The Doppler sample volume was reduced to the diameter of the vessel. The wall filter was set as low as possible to detect slow diastolic flow. Concerning velocity measurements, the entire arterial waveform was measured (by manually tracing the waveform profile) and the angle of insonation was determined to convert the Doppler shift frequencies to true velocities. With this procedure, maximal systolic velocity, minimal diastolic velocity, time-averaged mean velocity, and pulsatility index were obtained and recorded for all patients. The Doppler exam was repeated two or three times in each subject, and the tracing with the best technical characteristics was used for data analysis. In 80/116 cases, IMA blood flow was calculated by measuring the cross-sectional area of the IMA (by measuring the mean vessel diameter) with the built-in software of the computer. The area value obtained was then multiplied for the mean arterial velocity [flow (L/ min) Å cross-sectional area of the artery (assuming a circular crosssection) 1 time-averaged mean velocity] (Fig. 1).

Statistics Morphological and flow parameters were analyzed as functions of patient age (46 patients õ50 years vs. 56 patients ¢50 years) with the

365 Wilcoxon rank-sum test (without the Bonforti correction). No significant differences were observed in the mean body weights (p Å 0.88) or heights (p Å 0.42) of the two age groups.

Confirmation In the majority of cases, there was no clinical or sonographic evidence of splanchnic vessel abnormalities (on the basis of clinical history, physical examination, and radiological examinations such as abdominal computed tomography), and these patients’ physicians did not feel that further diagnostic tests were warranted. In 11 of the 116 patients we examined, angiography was subsequently ordered for various reasons (e.g., tumor staging, characterization of focal hepatic lesions, etc.). Although this examination was not performed specifically to evaluate the splanchnic circulation, the origin and proximal portion of the artery were visualized and proven to be completely normal in all 11 cases.

Results In 13 of the 116 patients examined (11.2%), the IMA could not be adequately visualized on real-time sonography or color Doppler scans, or the technical quality of the scans was considered to be suboptimal. Most of these failures were related to an excessive amount of intestinal gas or patient obesity, which made it difficult to exert sufficient probe pressure to displace the gascontaining intestinal loops. In other cases, flowmetric data were considered to be invalid because of artifacts due to the transmission of aortic pulsation, an excessively wide angle of incidence, or respiratory movements. Thus, the normal findings presented are based on successful examination of 103 patients. On real-time sonography and color Doppler, the IMA could be visualized at its emergence from the left anterior margin of the distal aorta. The proximal segment (3–6 cm) of the artery was linear or anteriorly concave, with a diameter of 1.3–4.3 mm (2.8 { 0.55 mm) (Fig. 1). The Doppler waveforms were characterized by a high systolic peak, early diastolic reflux (in most cases), and low or absent diastolic flow. Analysis of absolute velocity measurements (i.e., with appropriate angle correction) revealed values of 1.41 m/s { 0.48 for maximal systolic rate, 0.10 m/s { 0.16 for minimal diastolic velocity, and 0.43 m/s { 0.19 for time-averaged mean flow velocity. Because velocity waveforms from the IMA often exhibited reverse diastolic flow, the minimal diastolic velocities, as obtained by digitizing the entire waveforms, often had negative values (corresponding to maximum reverse flow velocity). Pulsatility index was 1.60–6.23 (3.49 { 0.49). Flow volume calculated in 80 cases was 0.05–0.29 L/min (0.13 { 0.06 L/min) (Fig. 1). Analysis of findings according to age revealed that the diameter of the IMA was greater in the ¢50 age group (õ50 years: 2.4 mm { 0.5; ¢50 years: 3.3 mm { 0.7; p Å 0.05). Among hemodynamic parameters,

4v12 0019 Mp 365 Tuesday Jun 16 10:36 AM SV-Ab Image (v. 23, #4) 0019 (1515)

366

P. Mirk et al.: Inferior mesenteric artery: Doppler study

Fig. 1. A Transverse and left sagittal sonograms of the infrarenal aorta in a 48-year-old man. The IMA (arrows) originates from the left anterior aspect of the aorta and descends along its left side, with a straight course in its proximal portion. B Doppler waveforms of the IMA in the same patient show a high systolic peak, early diastolic reflux, and low velocity diastolic flow; the pulsatility index is high (5.46). C Flow volume calculation (by combining mean area of vessel lumen and mean flow velocity) produces an estimated flow volume of 0.12 L/min. D Digital subtraction angiography in the same patient confirms a normal size and appearance of the IMA (arrowhead).

4v12 0019 Mp 366 Tuesday Jun 16 10:36 AM SV-Ab Image (v. 23, #4) 0019 (1515)

P. Mirk et al.: Inferior mesenteric artery: Doppler study

367

peak systolic and minimal diastolic flow velocities did not differ significantly between the two age groups (systolic flow velocity: p Å 0.43; diastolic flow velocity: p Å 0.26). Time-averaged mean flow velocity increased in subjects ¢50 years (0.38 m/s { 0.15 vs. 0.47 m/s { 0.21; p õ 0.05) (Fig. 2). In contrast, the pulsatility index tended to decrease with advancing age (õ50 vs. ¢50 years: 3.77 { 0.90 vs. 3.33 { 0.92; p Å 0.01), and, as a result, higher IMA blood flow volumes were observed in the older age group (õ50 vs. ¢50 years: 0.11 L/min { 0.04 vs. 0.15 L/min { 0.05; p Å 0.01) (Figs. 3, 4). The difference between the two age groups for these three parameters is statistically significant.

Discussion The IMA is the last collateral artery of the aorta included in the splanchnic circulation. It arises from the left side of the aorta at the level of the third lumbar vertebra (between the lower edge of L2 and the upper border of L4 in 87% of all subjects) [10]. It passes downward behind the parietal peritoneum before branching into the left colic, sigmoid, and superior hemorrhoidal arteries, which supply the distalmost portion of the transverse colon, the descending colon and sigmoid, and the upper part of the rectum [11]. Because of its small caliber and its position behind loops of the small intestine, sonographic visualization of the IMA has usually been considered difficult, and it is not included in most ultrasonographic studies of the splanchnic circulation. The literature contains numerous reports describing the normal sonographic appearance and Doppler features of the larger celiac and superior mesenteric arteries in both adults [1–7] and newborns [12–14]. Studies also have been conducted to assess the variability of Doppler findings related to larger splanchnic vessels (both arterial and venous) and to compare the accuracy of these findings with those obtained by angiography [15–17]. The first report of this type regarding the IMA was published by our group following a preliminary study of normal and pathological cases [8]. These data have been confirmed by others in a more recent study of the same type [18]. The present results confirm that the initial segment of the IMA can be visualized with fairly constant features in the majority of all patients. Based on angiographic studies, the diameter of the IMA in normal adults is 1.2–5.5 mm, with a mean of 3.3 mm [10]. The fact that this figure is somewhat higher than that observed with sonography in the present study (i.e., 2.8 mm) probably can be attributed to a magnification effect on angiography. In the fasting state, the Doppler tracings indicated relatively high systolic flow velocities with little or no diastolic flow (Fig. 1). Transient retrograde flow during

Fig. 2. Graph showing mean velocity values (range and SD) in the IMA with respect to patient age (õ50 years or ¢50 years) (p õ 0.05).

Fig. 3. Graph showing pulsatility index values (range and SD) for the IMA with respect to patient age (õ50 years or ¢50 years). The difference between the two age groups is statistically significant (p Å 0.01).

protodiastole also was fairly common. The high peripheral resistance indicated by these findings is to be expected during intestinal inactivity. Blood flow in general and diastolic flow in particular are higher in the celiac and superior mesenteric arteries, which can be explained in part because of their larger size (which allows higher systolic velocities), and also because the metabolic needs of organs they supply (i.e., liver, pancreas, and spleen by the former and the head and body of the pancreas by the latter) [11] require more continuous perfusion throughout the cardiac cycle. Besides reporting the results relative to a larger population, the present study differs from the previously published papers [8, 18] in that the sonographic and

4v12 0019 Mp 367 Tuesday Jun 16 10:36 AM SV-Ab Image (v. 23, #4) 0019 (1515)

368

P. Mirk et al.: Inferior mesenteric artery: Doppler study

Fig. 4. Graph showing estimated flow volumes (range and SD) for the IMA with respect to patient age (õ50 years or ¢50 years). The difference between the two age groups is statistically significant (p Å 0.01).

Doppler findings were verified by splanchnic angiography in a significant percentage of the subjects. Furthermore, flow volume calculations for the IMA have not been published by others. IMA flow volume was calculated in 80 cases based on the estimated cross-sectional area of the vessel and mean flow velocity (Fig. 1). The values reported in the present study, however, have an indicative value only. Indeed, we are aware of the potential sources of error that are inherent to the Doppler techniques, especially with regard to the determination of calculated values such as flow volume (e.g., errors in determinating the angle of insonation, errors in the measurement of vessel diameter). Notwithstanding such technical limitations, data concerning the estimated flow volume in the SMA by previous investigators are nevertheless in good agreement between different studies (517 { 19 mL/min in [4] and 538 { 37 mL/min in [19]). Compared with these values for the SMA, our estimated flow volume for the IMA (130 { 0.06 mL/min) seems reasonably acceptable when considering the smaller caliber of the vessel and the more limited extent of the gut it supplies. Unfortunately, a direct comparison of our results with flow volume calculations obtained by other non-Doppler methods is not available. The arterial inflow for the SMA was determined with a dye dilution technique [20] or video dilution technique [21], but there are no similar studies regarding the IMA. We sought to detect possible age-related differences in arterial flow pattern. Age-related differences were noted in most hemodynamic parameters. Subjects 50 years of age and older presented higher time-averaged mean velocities (p õ 0.05) and in particular lower vascular resistance (pulsatility index, p Å 0.01) with respect to younger individuals (Figs. 2, 3). These variations

(notwithstanding that peak systolic and minimal diastolic velocities were almost unchanged) can be accepted when considering that the spectral waveform is the result of several variables, whereas peak systolic and minimal diastolic velocities alone produce quite limited information about vascular hemodynamics. To fully evaluate the shape of the curve of the Doppler tracing, other parameters must be analyzed such as mean velocity and resistance to flow. The pulsatility index involves the measure of entire waveforms and relates peak systolic and minimal diastolic shift to mean Doppler shift. Thus, we can accept that variations in the waveform pattern may occur (as expressed by mean velocity and resistance to flow), although its extreme values (peak systolic and minimal diastolic values) remain almost unchanged. The diameter of the IMA also tended to increase with age (p Å 0.05). The combination of these factors should account for the higher flow volumes found in older individuals (p Å 0.01) (Fig. 4). These age-related differences are somewhat difficult to explain. The only study that has addressed this topic [4] found no correlation between flow and age in the SMA. However, the population examined by these investigators was younger (mean age Å 37 years), and this may have prevented age-related differences to become apparent. The increase in vessel caliber can be interpreted as an effect of loss of elasticity in the vessel walls due to aging. However, arterial blood pressure also increases with age, and both factors could have contributed to the hemodynamic changes in the older patients. Splanchnic angiography is undoubtedly the most sensitive and specific method for studying the morphology of the intestinal vessels, and it is the only method capable of demonstrating the more distal branches of the arterial circulation. With Doppler sonography, even under the best of circumstances, only the most proximal segments of the splanchnic vessels can be visualized. However, the Doppler approach offers the possibility to obtain qualitative and quantitative data on the functional aspects of these vessels in a totally noninvasive manner; for this reason, it is particularly suitable for investigating physiological splanchnic hemodynamics. It has already been used to investigate blood flow in the SMA under fasting and postprandial conditions [19, 22–24], during acute hypoglycemia [25], and after physical exercise [26]. Drug-induced modifications also have been studied in both adults and newborns [14, 27]. Knowledge of the normal morphological and hemodynamic characteristics of the IMA are fundamental for identifying pathological conditions affecting this vessel. Arterial pathology may be associated with abdominal pain, which often occurs with chronic intestinal ischemia due to generalized atheromatous stenosis or occlusion of the splanchnic vessels. However, in most cases,

4v12 0019 Mp 368 Tuesday Jun 16 10:36 AM SV-Ab Image (v. 23, #4) 0019 (1515)

P. Mirk et al.: Inferior mesenteric artery: Doppler study

even in the presence of significant arterial disease (e.g., stenosis, compensatory hypertrophy or occlusion involving the IMA alone), the patient may be totally asymptomatic. Routine inclusion of the IMA in sonographic and Doppler studies of the splanchnic circulation should provide more complete information in cases of intestinal vascular pathology. Moreover, this noninvasive technique also may be used for investigating the physiological behavior of the splanchnic circulation (e.g., fasting vs. postprandial hemodynamics, responses to various drugs). References 1. Qamar MI, Read AE, Skidmore R, et al. Transcutaneous Doppler ultrasound measurement of celiac axis blood flow in man. Br J Surg 1985;72:391–393 2. Taylor KJW, Burns PN, Woodcock JP, et al. Blood flow in deep abdominal and pelvic vessels: ultrasonic pulsed Doppler analysis. Radiology 1985;154:487–493 3. Jager K, Bollinger A, Valli C, et al. Measurement of mesenteric blood flow by Duplex scanning. J Vasc Surg 1986;3:462–469 4. Qamar MI, Read AE, Skidmore R, et al. Transcutaneous Doppler ultrasound measurement of superior mesenteric artery blood flow in man. Gut 1986;27:100–105 5. Sato S, Ohnishi K, Sugita S, et al. Splenic artery and superior mesenteric artery blood flow: nonsurgical Doppler US measurement in healthy subjects and patients with chronic liver disease. Radiology 1987;164:347–352 6. Nakamura T, Moriyasu F, Ban N, et al. Quantitative measurement of abdominal arterial blood flow using image directed Doppler ultrasonography: superior mesenteric, splenic, and common hepatic arterial blood flow in normal adults. J Clin Ultrasound 1989;17:261–264 7. De Franco A, Mirk P, Brizi MG, et al. Anatomy of splanchnic vessels by combining imaging procedures and Doppler US. Rays 1991;16:227–246 8. Mirk P, Cotroneo AR, Palazzoni G, et al. Valutazione con ecoDoppler dell’arteria mesenterica inferiore: studio di fattibilita` e definizione dei caratteri morfologici e flussimetrici. Radiol Med 1994;87:275–282 9. Puylaert JB. Acute appendicitis: US evaluation using graded compression. Radiology 1986;158:355–358 10. Abrams HL. Inferior mesenteric arteriography. In: Abrams HL, ed. Abrams angiography: vascular and interventional radiology. Boston: Little, Brown and Company, 1991:1701–1705 11. Kadir S, Lundell C, Saeed M. Celiac, superior, and inferior mesenteric arteries. In: Kadir S, ed. Atlas of normal and variant angiographic anatomy. Philadelphia: WB Saunders, 1991:297–307

369 12. Leidig E. Doppler analysis of superior mesenteric artery blood flow in preterm infants. Arch Dis Child 1989;64:476–478 13. Van Bel F, Van Zwieten PHT, Guit GL, et al. Superior mesenteric artery blood flow velocity and estimated volume flow: duplex Doppler US study of preterm and term neonates. Radiology 1990;174:165–169 14. Hsu CH, Lee HC, Huang FY. Duplex ultrasonographic assessment of gut blood flow velocity: effect of meal composition in normal full-term newborns after first feed. J Ultrasound Med 1994;13:15–18 15. Sabba´ C, Ferraioli G, Sarin SK, et al. Feasibility spectrum for Doppler flowmetry of splanchnic vessels in normal and cirrhotic populations. J Ultrasound Med 1990;9:705–710 16. Perko MJ, Just S. Duplex ultrasonography of superior mesenteric artery: interobserver variability. J Ultrasound Med 1993;5:259– 263 17. Mallek R, Mostbeck G, Walter RM, et al. Duplex Doppler sonography of celiac trunk and superior mesenteric artery: comparison with intra-arterial angiography. J Ultrasound Med 1993;12:337–342 18. Denys AL, Lafortune M, Aubin B, et al. Doppler sonography of the inferior mesenteric artery: a preliminary study. J Ultrasound Med 1995;14:435–439 19. Moneta GL, Taylor DC, Helton WS, et al. Duplex ultrasound measurement of postprandial intestinal blood flow: effect of meal composition. Gastroenterology 1988;95:1294–1301 20. Norryd C, Dencken H, Lundenquist A, et al. Superior mesenteric artery blood flow in man studied with a dye-dilution technique. Acta Chir Scand 1974;141:109–118 21. Lantz M, Link D, Holcroft J, et al. Video dilution technique: angiographic determination of splanchnic blood flow. In: Granger D, Bulkeley G, eds. Measurement of blood flow applications to the splanchnic circulation. Baltimore: Williams & Wilkins, 1981:425–37 22. Qamar MI, Read AE. Effects of ingestion of carbohydrate, fat, protein, and water on mesenteric blood flow in man. Scand J Gastroenterol 1988;23:26–29 23. Lilly MP, Harward TRS, Flinn WR, et al. Duplex ultrasound measurement of changes in mesenteric flow velocity with pharmacologic and physiologic alteration of intestinal blood flow in man. J Vasc Surg 1988;9:18–23 24. Moneta GL, Cummings C, Caster J, et al. Duplex ultrasound demonstration of post-prandial mesenteric hyperemia in splanchnic circulation collateral vessels. J Vasc Tech 1991;15:37–39 25. Braatvedt GD, Newrick PG, Halliwell M, et al. Splanchnic haemodynamic changes during acute hypoglycemia in man. Clin Sci 1991;81:519–523 26. Qamar MI, Read AE. Effects of exercise on mesenteric blood flow in man. Gut 1987;28:583–587 27. Cooper AM, Braatvedt GD, Qamar MI, et al. Fasting and postprandial splanchnic blood flow is reduced by a somatostatin analogue (octreotide) in man. Clin Sci 1991;81:169–172

4v12 0019 Mp 369 Tuesday Jun 16 10:36 AM SV-Ab Image (v. 23, #4) 0019 (1515)