Physiological aspects of the determination of comprehensive arterial inflows in the lower abdomen assessed by Doppler ultrasound

Background

The splanchnic circulation has been described as the“blood-giver of circulation”and is believed to play amajor role in overall cardiovascular regulation [1]. Thesplanchnic system receives nearly 30% of the cardiacoutput through three large arteries: the coeliac and thesuperior and inferior mesenteric arteries. Hemodynamics in the splanchnic organs are altered under variousstressful conditions, such as during physical activity andin the postprandial state [2], due to balancing of tonebetween sympathetic and vagus activity; consequently,clinical assessment of the splanchnic circulation couldpotentially provide valuable information regardinghepato-gastrointestinal disease [3-5] and cardiacdysfunction [6].

Previous Doppler ultrasound studies that measuredsplanchnic blood flow in a“single vessel with small sizevolume”, such as the superior mesenteric, coeliac artery,or portal vein, were concerned solely with the targetorgan in the gastrointestinal area [2,7-9]; therefore, evaluation of alterations in these single arterial blood flowsunder the various states were sometimes limited tosmall volumes, even though there was a relatively largechange in flow. Evaluation of the comprehensive arterialblood flow in the lower abdomen (BFAb), including theliver, spleen, gastro-intestine, kidney, and pelvic organsas a multiple arterial function, may potentially be a feasible method of determining the distribution of abdom-inal blood-flow volume or disorder in cases ofsplanchnic or cardiovascular dysfunction, as well as thedistribution following nutritious meal intake or physicalexercise [10,11].

Our previous studies used ultrasonography to assesswhole arterial BFAbhemodynamics: BFAbwas obtainedby subtracting blood flow in the bilateral proximalfemoral arteries [left femoral artery (LFA) and rightfemoral artery (RFA)] from blood flow in the upperabdominal aorta (Ao) above the coeliac artery bifurcation[10-14].

This method of quantitative assessment is a challengingbut unique and non-invasive procedure for determiningthe comprehensive inflows of all abdominal organs, andis a potentially useful indicator of blood flow redistribution in cardiovascular and hepato-gastrointestinal disease,in the postprandial period, and in relation to physicalexercise.

Variability in the hemodynamics (blood velocity, vesseldiameter and blood flow) of the three target arteries isvaluable information for determining BFAb. Therefore,the purpose of the present review is to summarize themethodology for determining BFAbusing validated dataof three target arteries, to discuss methodological consid-erations and limitations in view of previously reportedfindings, and to consider the potential clinical usefulnessand application of measurements for the comprehensivelower abdominal flows.

Methodology

Subjects

The subjects were the participants of three studies:60 healthy males (mean age, 24.1 ± 5.5 years; meanheight, 173.3 ± 7.1 cm; and mean body weight, 68.5 ±9.3 kg), 50 healthy males (mean age, 23.5 ± 4.9 years;mean height, 172.8 ± 7.1 cm; and mean body weight,67.6 ± 9.9 kg) and 10 healthy males (mean age, 25.2 ± 6.6years; mean height, 175.6 ± 7.0 cm; and mean bodyweight, 70.1 ± 7.6 kg). All values are expressed as mean ±standard deviation (SD). Participants had no previoushistory of cardiovascular disease, gastrointestinal disease,hypertension, or anaemia, and no abnormality of the peripheral vasculature. The studies were conducted according to the principles of the Declaration of Helsinki (1976) and with the approval of the Institutional Ethics Committee of the author’s institution. All participants gave theirwritten consent and were informed of the nature andpurpose of the study, as well as potential risks and discomfort. The participants also understood that theycould withdraw from the study at any time without con-sequence. The study populations reviewed in the presentstudy do not include the elderly.

Approach for Doppler ultrasound assessment of threetarget arteries for determining BF_Ab

The target vessels were the following three conduitarteries: 1) the Ao at ~3 cm above the coeliac artery bifur-cation, 2) the proximal LFA, and 3) the proximal RFA(Figure 1). The Ao region was most commonly measuredjust below the diaphragm in longitudinal section view(from the sub-sternal area) to enable Ao sample volume tobe maintained at the end of the expiratory phase duringspontaneous breathing. Detection of the Ao was relativelyconstant and free from interference from intestinal gas.For both femoral arteries, measurement location was chosen to minimize turbulence and the influence of the inguinal region on blood flow above the bifurcation, therebyenabling easy and reliable measurement [10-18]. Bloodvelocity (pulsed wave) and vessel diameter (2-dimensional)measurements were obtained using a curvilinear arrayprobe (3.5 MHz) for Ao and a linear array probe (7.5MHz) for the LFA and RFA. The insonation angle wasmaintained below 60° for each participant and remainedconstant throughout the experiments [10-14,19]. The sample volume was placed in the precise centre of the vesselbefore being adjusted to cover the width of the vessel diameter and blood velocity distribution.

The data in the present review were obtained using anultrasound unit (SONOS 1500, HP77035A; HewlettPackard, Tokyo, Japan) with a real-time two-dimensionalultrasonic imager and a pulsed-Doppler flowmeter forcalculating maximum envelope in the blood velocityprofile. The Doppler instrument used in this review,however, could not determine time- and spatial-averagedand amplitude (signal intensity)-weighted mean bloodvelocities; thus, the measured blood velocity determinedby integration of the outer envelope (maximum velocities) would have reflected the higher (maximum) velocitycomponent at the centre of the vessel through the cardiaccycle. Because this procedure takes no account of thelower velocity component of the flow profile, the bloodvelocity values in this review have potentially been overestimated. On the basis of this physiological phenomenon, the above-mentioned measure of mean bloodvelocity, which expresses the averaged speed for all redblood cells within the vessel, is more precise; however,the present procedure used to determine blood velocityalso provides acceptable data.

Recent developments in ultrasound instrumentationinclude an auto-tracing programme for determiningmean and maximum blood velocity (outer envelope).Maximum (envelope) blood velocity in the femoralartery is previously reported as being ~1.53 [20] and1.3-1.8 [21,22] times higher than mean blood velocity.An in vitro study that used a silicon tube to model theconduit artery found that maximum (envelope) bloodvelocity was approximately 1.75 times higher than meanblood velocity [23].

Measurement procedure of blood velocity, vesseldiameter, and blood flow in the arteries

For each participant, measurement was conductedbetween approximately 7 and 9 am, while fasting. Priorto measurement, colour Doppler was used to check forunsuspected pathology in each conduit artery. For eachof the three conduit arteries, blood velocity was measured for 2-3 minutes and vessel diameter for 1-2 minutes via image observation using a wide expanded viewin longitudinal section, by a single operator (the author).Blood velocity was analysed by integrating the outerenvelope of the maximum velocity values from the flowprofile for each beat, for approximately 20-40 beats[10,24-26]. Blood velocity and vessel diameter analyseswere performed using the phase that demonstrated similar heart rate and blood pressure values among measurements from the three conduit arteries. The systolicand diastolic vessel diameter of each conduit artery wasmeasured in relation to the electrocardiograph displayedon the monitor of the ultrasound unit.

Vessel diameter was also measured under perpendicular insonation and calculated in relation to the temporalduration of the electrocardiography recording curve, asfollows: [(systolic vessel diameter × 1/3) + (diastolic vessel diameter × 2/3)] [15-17]. The mean vessel diameterfor each beat was calculated over approximately 20-40beats. Blood flow was determined by multiplying thecross-sectional area [area =π× (vessel diameter/2)2] bythe amplitude (signal intensity)-weighted blood velocity(time- and spatial-averaged outer envelope of the maximum velocity). To determine BFAbprecisely, blood pressure and heart rate should remain in a steady stateduring measurement of the three target arteries.

Determination of comprehensive BF_Ab

Blood flow in the Ao, LFA, and RFA is defined as BFAo,BFLFA, and BFRFA, respectively. BFAbwas calculated bysubtracting the sum of BFRFAand BFLFAfrom BFAo, as follows: BFAb= [BFAo- (BFLFA+ BFRFA)] [10-14].

Consideration of the physiological aspect of thethree arterial blood flows and BF_Ab

Day-to-day reliability and variability in hemodynamics ofthe three arteries and BFAbvia repeated measurements

Hemodynamic measurements (blood velocity, vessel diameter, and blood flow) in the three arteries were performed on three consecutive days by a single operator.

BFAbcan then be determined for each day by the formula BFAo- (BFLFA+ BFRFA), with blood flow calculatedby multiplying blood velocity by the cross-sectional area.Variability in the values of blood velocity, vessel diameter, and blood flow in the three arteries may be valuable information for determining BFAb.

First, relative reliability was estimated by analysing thethree hemodynamic measurements repeated on three different days, for 60 healthy male participants [12]. Asshown in Table 1, F-test revealed no significant differencein blood velocity, vessel diameter, or blood flow valuesamong the three measurements. Consequently, the single-measure intra-class correlation coefficient was significantlyhigh for relative reliability estimated by repeated hemodynamics measurements [27]. This indicates that Dopplerassessment of hemodynamic parameters in the three targetarteries has potential as a stable and acceptable procedurefor determining BFAb.

Second, Bland-Altman analysis [28] was used for statistical analysis of mean values (x-axis) and difference values(y-axis) in hemodynamics (blood velocity, vessel diameter,and blood flow) between two measurements over threedifferent days (Figure 2). Figure 2 shows that no systematicbias (fixed bias and proportional bias) was found betweenany two measurements over three different days. The limits of agreement (mean ± 1.96 SD) in terms of the difference between two hemodynamics measurements and the95% confidence interval also indicate validity in the present study population within an acceptable and permissible range and true mean values (Table 2). Bland-Altmananalysis revealed that the acceptable range in difference(bias) between two measurements may be less than 8.9cm/s for Ao and less than 5.3 cm/s for blood velocity inthe femoral arteries; less than 1.5 mm for Ao and less than0.95 mm for vessel diameter in the femoral arteries; andless than 1.0 l/min and less than 0.18 l/min for blood flowin the Ao and femoral arteries, respectively. It is considered that this range takes into account day-to-day physiological variation as well as measurement error. Thepresent results for repeated measurements on three different days reveals that the range in blood flow values in thethree arteries remained similar for individual participantsunder similar testing conditions; thus, mean BFAbis considered a reliable value with the acceptable range in difference (bias) between two measurements may be less than0.9 l/min for blood flow in the Ab. (Tables 1, 2; Figure 3).

Validity of target arterial blood flows and BFAbcomparedwith previous findings

Abdominal aorta

Values for vessel diameter, estimated cross-sectional area,and blood flow of the upper abdominal aorta obtained inthe present review were 15.6 ± 1.2 mm, 1.91 ± 0.29 cm2,and 2951 ± 767 ml/min, respectively (Table 3).

The present values for diameter and cross-sectional areaare similar to those of 15.5-17.6 mm and 1.88-2.43 cm2,respectively, measured by Gabriel and Kindermann [29].Nimura et al. [7] reported blood flow values obtained byDoppler ultrasound of the upper abdominal aorta and thesum total blood flow of the coeliac, superior mesentericand both renal arteries as 2470-3246 ml/min and 2450-3549 ml/min, respectively. These values are similar tothose for BFAoexpressed in the present review.

Femoral artery

The present values for the diameter of the femoralarteries were 9.0 ± 0.7 mm in the LFA and 9.1 ± 0.7 mmin the RFA (Table 3), which is in the same range as thepreviously reported values of 7.5 ± 0.3 mm measured

using angiography [30] and 8.1 ± 0.11 mm measured byDuplex Doppler [31]. In the present review, cross-sectional area of the femoral arteries was 0.65 ± 0.1 cm2inthe LFA and 0.65 ± 0.11 cm2in the RFA. Mean bloodvelocity in the femoral arteries was 8.3 ± 2.4 cm/s in theLFA and 8.1 ± 2.1 cm/s in the RFA; these values are inthe same range as that of 10.2 ± 0.39 cm/s previously

measured by pulsed Doppler [31]. Blood flow in thefemoral arteries was 316 ± 97 ml/min in the LFA and313 ± 83 ml/min in the RFA. These values are in thesame range as those of 450-886 ml/min [32], 301 ± 81ml/min [33], and 390 ± 20 ml/min [34] measured usingindicator dilution; and 376 ± 154 ml/min [35], 226.5 ±28.6 ml/min [31], 344 ml/min [36], and 350-367 ml/min[37] measured by Doppler ultrasound. Furthermore,Ganz et al. [38] reported a value of 383-766 ml/min bythermodilution, and Vänttinen [39] reported a value of239 ml/min using electromagnetic flowmetry. Thesevalues are in the same range in those of the presentreview, despite differences in the method of measurement. Blood flow may also be influenced by the subject’sposition during measurement and local blood flow perbody weight, as well as thigh muscle mass [40].

Blood flow in the lower abdomen

There is lack of comparative BFAbdata measured byother valid methods (gold standard) such as the thermodilution technique or the cardiovascular magnetic resonance method. However, ultrasound Doppler is also anacceptable valid measure for determining blood velocity/flow in the conduit artery.

Including the results of previous reports [12], the rangeof BFAbover the three different days was 1153-4401 ml/min in the 60 participants. Furthermore, the mean valueof BFAbwas 2630 ± 649 ml/min in 18 of the participants(age range, 20-38 years) in the previous reports [10].

Based on the general anatomical features shown inFigure 1, the BFAbvalues are considered to indicate thesum of blood flow to the coeliac artery; mesentericarteries; the bilateral renal, suprarenal, gonadal, and internal iliac arteries; and some lumbar arteries.

Previous studies [41-44] reported average splanchnicblood flow (including that of the coeliac trunk, superiormesenteric, and inferior mesenteric arteries) as approximately 1500 ml/min, corresponding to 20%-30% of cardiacoutput. The sum of the blood flow values in the two renalarteries is approximately 1000-1200 ml/min, whichcorresponds to 20% of cardiac output [45]. In addition,blood flow is 1400 ml/min in the liver, gastro-intestine,and spleen (the so-called splanchnic organs), and 1100 ml/min in the kidney [46]. The range of values for the sum ofblood flow in the“splanchnic”and the“two renal arteries”reported in previous studies is similar to the BFAbvaluesobtained using the present method. However, the widerange in BFAbmay also be related to individual physicalfeatures such as body surface area and body weight.

Relationship of blood flows to body surface area and tobody weight

Previous studies have shown that cardiac outputincreases in proportion to body surface area [46,47],which means that cardiac output is regulated throughout life in almost direct proportion to overall metabolicactivity. Furthermore, a positive correlation has beendemonstrated between cardiac output and abdominal-splanchnic blood volume, using whole-body scintigraphy[48]. BFAbis expected to be closely related to body surface area, because the splanchnic system receives ~30%of cardiac output.

A significant positive relationship exists between BFAband both body surface area and body weight (Figure 4).The formula used to calculate body surface area is widelyused in the target population [49]. Furthermore, anincrease in BFAbwith increasing body weight may be reasonable, taking into consideration the total weight of thelower abdomen. This relationship is based on the conceptthat blood flow distribution is associated with a higherflow per weight to the liver and intestine compared withskeletal muscle at rest [50]. An expected additional findingwas that peripheral blood flow at each conduit artery alsohad a positive linear relationship (p< 0.05) with body surface area as well as with body weight (Figure 4). This correlation is in agreement with evidence concerning therelationship between cardiac output supply and peripheralarterial blood flow, with cardiac output being closelyrelated to body surface area [46,47].

Relationship of BFAbto BFAo, BFLFA, and BFRFAinestimating BFAb

The distribution of BFAbmay be influenced by the magnitude of both cardiac output and limb blood flows. Specifically, it is speculated that alterations in limb bloodflow may play an important role in regulating BFAbviachanges in the tone in the vascular bed of abdominalorgans during low-intensity exercise when there is littlefluctuation in the magnitude of BFAo[10]. Similarly, it isunclear whether BFAbor cardiac output at rest has amajor impact on limb blood flow in a steady state ofneural response.

Day-to-day coefficients of variation in blood flow wererelatively high in the femoral arteries compared withBFAo, even though the absolute BFAovalues wereapproximately 10 times higher than those of blood flowin both femoral arteries. Accordingly, BFAbwas morestrongly related to BFAo(r= 0.966) than to BFLFA(r=0.303) or BFRFA(r= 0.281) (Figure 5). Alterations inBFAothat are closely related to cardiac output (exceptcerebral and arm blood flows) potentially have the greatest influence on BFAb, even if blood flow in the femoralarteries has less influence on BFAb, at least at rest;accordingly, BFAopotentially has the largest influence onBFAbas a central hemodynamic factor. Figure 5 showsthat precise BFAbvalues may be unreliable when thereare large variations in both BFLFAand BFRFA. Thus, evaluation of BFAbmay be better expressed by the followingformula: BFAb(l/min) = 0.85 × BFAo- 0.19, if Ao measurement alone is performed (Figure 5).

Effect of respiration and posture

Deep thoracic breathing in inspiration produces rapidacceleration of blood flow in veins located near thethorax, such as the hepatic vein, jugular vein, and inferiorvena cava [51], while the blood velocity in these veins isreduced just as rapidly at the start of expiration [52].Mechanical ventilation causing a higher positive end-expiratory pressure-induced increase in lung volume could impede venous return, thereby altering systemichemodynamics and hepatic venous outflow [53]. In animal experiments, portal vein blood velocity and hepaticarterial blood velocity were shown to decrease with positive end-expiratory pressure as a result of a simpleincrease in the downstream pressure [54].

Also relevant to the present method are respiratoryphase and posture, which are related to alteration inBFAb, as shown in Figure 6. The difference in BFAbwasapproximately 550 ml/min between inspiration andexpiration in the sitting position; in the supine position,the difference was 480 ml/min. Blood flow was significantly less in inspiration compared with expiration in Ao,LFA, and RFA, in both the sitting and supine positions.BFAbwas found to be lower in inspiration than in expiration, in both the sitting and supine positions. Respiration-related changes in the hemodynamics of the threeconduit arteries potentially lead to alterations in BFAb.

This result may be in partial agreement with the theorythat respiratory-induced alteration of BFAboccurs withimpedance of venous return in the splanchnic area.

The change in intra-abdominal pressure during breathing (thoracic-abdominal movement) possibly reflectstransient changes in blood velocity in the Ao and femoralarteries. Higher values of venous outflow are found in thehepatic and portal veins in the supine rather than uprightposition, due to the effect of gravity [51]. In contrast, thereduction of BFAbin the inspiratory phase is similarbetween sitting and supine, and thus no postural effecton BFAbis seen in Figure 6. As the splanchnic circulationis intrinsically susceptible to the adverse effects of hydrostatic force [51,55], redistribution due to postural changebetween sitting and supine may differ between venousand arterial sites. Respiratory effects should be taken intoaccount in evaluation of BFAbdetermined by measurements obtained in the three arteries. The present studydemonstrated that changes in blood velocity betweenexpiration and inspiration in the three conduit arteriesmay potentially indicate alterations in BFAb, and are onlyminimally influenced by posture. This phenomenoncould be due to mechanical compression of vascular flowperfusion in comprehensive BFAbor via the vasovagalresponse. Because respiration and posture effects have aneffect when organ perfusion is adequate, it is importantnot to confuse these effects as a sign of impaired organperfusion. Evaluation of BFAbhemodynamics in the threeconduit arteries should take respiratory effects intoaccount.

Limitations

The disadvantages of the present methods are that measurement of the three target arteries cannot easily beperformed in a short period of time, and that blood flowin the pelvis and other organs (except for the targetsplanchnic area) cannot be excluded. To avoid over- orunderestimation of BFAb, measurement of the three target arteries should be performed under steady-state conditions at rest, with only minor changes in heart rateand blood pressure.

Potential clinical usefulness and application

Evaluation of BFAbas a quantitative assessment, encompassing physiologic flow, is a potentially useful indicatorof 1) reserve blood volume and 2) blood flow in redistribution in the lower abdominal circulation in cardiovascular and hepato-gastrointestinal disease, shock, multipleorgan failure, and stressful conditions such as followingphysical exercise and in the postprandial period. Theadvantage of the present method is that it enables evaluation of comprehensive BFAbwithout interference fromintestinal bowel gas, because the three target conduitarteries can be detected relatively easily. It may also beuseful for examining pathological hemodynamics, whichmay influence the abdominal circulation under the conditions of 1) extraordinary hemodynamics associatedwith abdominal aneurysm; 2) collateral circulation inabdominal-iliac peripheral arterial disease, as well ascomparison with the post-operative state; and 3) intestinal neurological dysfunction associated with spinal disorder, in cerebrovascular disease, and in orthostatichypotension with vasovagal syncope.

Although Doppler methods are less commonly usedfor quantifying flow in the three target arteries compared with other techniques, the measurement procedure used in the present method is potentially clinicallyviable. However, because it can be time-consuming toperform routine hemodynamic measurements for thethree conduit arteries, measurement may be limited tothe region of the femoral arteries around the inguinalligament close to the genital area, which may not be anacceptable method for general use. It is possible thatDoppler ultrasound evaluation of BFAbusing a singlevessel (Ao) may enable the necessary information to beobtained (Figure 5).

Also of note, because BFAbvalues are potentiallyrelated to many factors, including mean arterial bloodpressure and/or cardiac index, it has potential use as asurrogate parameter for central venous saturation in theclinical setting.

Conclusions

The advantage in the described procedure for the determination of splanchnic hemodynamics is that it maypotentially enable evaluation of the whole lower abdominal blood flows, assessed by non-invasive measurementusing cardiovascular ultrasound. In contrast, it has thedisadvantage that measurement of the three targetarteries during steady heart rate and blood pressure canbe time-consuming, and in measuring blood flow in thetarget splanchnic area, blood flow of pelvic and otherorgans cannot be excluded. Respiration and posturerelated to alterations in BFAbshould be taken intoaccount when measuring the three arteries. Determination of BFAbby evaluating three-conduit arterial hemodynamics using the technique described in this reviewmay provide a valid measurement that encompasses thecomprehensive physiologic arterial blood inflow to multiple abdominal organ systems.

Abbreviations

BF: Blood flow; Ao: Upper abdominal aorta; LFA: Left femoral artery; RFA:Right femoral artery; BFAo: Blood flow in the upper abdominal aorta; BFRFA:Blood flow in the right femoral artery; BFLFA: Blood flow in the left femoralartery; FAs: Both femoral arteries; BFAb: Comprehensive blood flow in thelower abdomen.

Acknowledgements

The author would like to express gratitude to the deceased Professor HisaoIwane, as a previous supervisor; Professor Takafumi Hamaoka, as previous cosupervisor at Ritsumeikan University; Dr. Ayumi Sakamoto, as the director ofthe Tokyo Therapeutic Institute; Mr. Yukihiro Yamamoto of GE HealthcareJapan for his kind support in the initial studies; colleagues Dr. Norio Muraseand Dr. Ryotaro Kime; and Professor Toshihito Katsumura, as the presentdepartmental supervisor. I would also like to thank all of the participants inthe studies.

Author details

Department of Sports Medicine for Health Promotion, Tokyo MedicalUniversity, Tokyo, Japan.Cardiac Rehabilitation Centre, Tokyo MedicalUniversity Hospital, Tokyo, Japan.

Competing interests

The author declares that they have no competing interests.

Received: 12 December 2011 Accepted: 26 March 2012

Published: 26 March 2012