Evaluation of gravity effect on inferior vena cava and abdominal aortic flow using multi-posture MRI

Abstract

Background

Inferior vena cava flow (IVCF) and abdominal aortic flow (AAF) are essential components of the systemic circulation. Although postural changes might alter IVCF and AAF by the gravity effect, the exact details are unknown.

Purpose

To evaluate the effect of gravity on IVCF and AAF using a novel magnetic resonance imaging (MRI) system that can image in any position.

Material and Methods

Caval velocity-mapped images were obtained using the cine phase-contrast technique in the upright and supine positions with multi-posture MRI (n = 12). The mean IVCF/AAF velocity, maximum IVCF/AAF velocity, cross-sectional area of IVC/AA, mean IVCF/AAF, maximum IVCF/AAF, and heart rate in the two positions were assessed.

Results

The mean IVCF velocity, maximum IVCF velocity, cross-sectional area of IVC, mean IVCF, maximum IVCF, mean AAF velocity, maximum AAF velocity, mean AAF, and maximum AAF were significantly lower in the upright position compared with the supine position (P < 0.05 for all), with differences of 52% ± 33%, 36% ± 19%, 56% ± 18%, 26% ± 18%, 19% ± 11%, 33% ± 13%, 33% ± 22%, 42% ± 21%, and 37% ± 28%, respectively. Heart rate was significantly higher in the upright position compared with the supine position (116% ± 9.2%; P = 0.003). There were no differences in cross-sectional area of AA between the two positions (108% ± 22%; P = 0.583).

Conclusion

The effect of gravity decreases IVCF and AAF. Clarifying the effect of gravity on IVCF and AAF during a postural change may help to improve the management of patients with circulatory disease.

Keywords

Magnetic resonance imaging vascular aorta vena cava

Introduction

Inferior vena cava flow (IVCF) and abdominal aortic flow (AAF) are essential components of the systemic circulation. These flows affect multiple organs and reflect systemic circulatory dynamics. The circulation system is adjusted by cardiovascular regulatory mechanisms (e.g. cardiac output, heart rate [HR], blood pressure) when faced with the challenges of everyday life, such as postural changes (i.e. differences caused by the effect of gravity) (1). Some circulatory diseases are known as the result of postural change. For example, paroxysmal nocturnal dyspnea in the supine position (2), lower extremity varicose vein in the upright position (3), and postural hypotension (2). Nevertheless, it is difficult to directly evaluate blood flow in the great vessels during postural changes. Understanding these changes is important for the development of treatments for circulatory diseases.

Magnetic resonance imaging (MRI) with phase-contrast sequence can be used to measure the blood flow velocity and volume flow rate of blood vessels. Although this is a non-invasive and reliable method (4,5), general MRI systems can only scan individuals in the supine position, despite the upright position being more common in daily life. However, a novel multi-posture MRI system was recently developed to allow body MRI scans in both the supine and the upright positions (6,7) (Fig. 1). As a preliminary study using MRI, we evaluated the postural change of the portal venous flow which supplies one organ (7). The aim of the present study was to use the multi-posture MRI technique to evaluate differences in IVCF and AAF between the supine and upright positions of healthy volunteers.

Fig. 1.

The multi-posture magnetic resonance imaging system. The system consists a pair of 0.4 T permanent magnets. Images are shown for the (a) upright and (b) supine positions.

Material and Methods

Participants

We enrolled 12 healthy volunteers in the study (12 men, age range = 20–24 years). Approval for the study was obtained from our institutional review board. We explained to all volunteers the purpose and procedures of all investigations and performed the study only after written informed consent was obtained from all volunteers.

Study procedure

Details of the multi-posture MRI were previously reported (6,7). In brief, the multi-posture MRI system consisted of a pair of 0.4 T permanent magnets that were opposed to each other (with the patient space in between) and aligned vertically (field homogeneity of <3.0 ppm over a 170-cm diameter of spherical volume; Hitachi Healthcare, Tokyo, Japan; Fig. 1) (6,7). MRI scans were performed during a breath-hold on inspiration, with a quadrature body coil in the supine and upright positions. The breath-hold scan was chosen to improve the signal-to-noise ratio in the low magnetic field. Scout images (axial, coronal, and sagittal) were used to localize the AA, IVC, and other structures in each position (Fig. 2). Caval velocity-mapped images were then obtained perpendicular to the AA and IVC using an electrocardiography-triggered cine phase-contrast technique in the supine and upright positions. We used the following MRI parameters (as in a previous study (7)): slice thickness = 6 mm; flip angle = 20°; acquisition matrix =128 × 50 (readout × phase encoding); field of view =240 × 70 mm; number of signals averaged = 1; and velocity-encoded gradients (VENC) = 15 cm/s (upright position) and 30 cm/s (supine position) (5,8). The ranges of the VENC were optimized to eliminate aliasing errors.

Fig. 2.

Slice and the ROI position for analysis of IVCF and AAF in the (a–c) supine and (d–f) upright positions. (a, d) The slicing plane (white broken line) to capture the IVC (white arrowheads) and AA (black arrowheads). Examples of the ROIs for the IVC (white circle) and the AA (white broken circle) on the (b, e) magnitude and (c, f) velocity images in each position. AAF, abdominal aortic flow; IVCF, inferior vena cava flow; ROI, region of interest.

Next, we drew manual regions of interest (ROIs) as contours of the cross-sectional areas of the AA and IVC on the magnitude images, and these ROIs were copied onto the corresponding phase images (Fig. 2). We determined the mean flow velocity in the ROI in each cardiac phase. The mean velocity was corrected to subtract the baseline offset caused by eddy currents. The baseline was determined on another ROI in the background stationary region (9). Multiplying the mean velocities by the cross-sectional areas, we obtained a flow curve (Fig. 3) (5,6,10).

Fig. 3.

The scanning and analysis procedure. (a) Capturing of the IVC and AA. (b) Setting of the ROIs for the IVC and AA on the magnitude and velocity images. (c) Representative IVC and AA curves in the supine and upright positions. AA, abdominal aorta; IVC, inferior vena cava; ROI, region of interest.

We assessed the mean IVC velocity (IVC-V_mean), the maximum IVC velocity (IVC-V_max), the mean IVCF (IVCF_mean), the maximum IVCF (IVCF_max), and the cross-sectional area of the IVC (IVC-CA) in the supine and upright positions. In addition, we assessed the mean AA velocity (AA-V_mean), the maximum AA velocity (AA-V_max), the mean AAF (AAF_mean), the maximum AAF (AAF_max), and the cross-sectional area of the AA (AA-CA) in the supine and upright positions. Finally, we assessed HR in the supine and upright positions.

Participants were scanned in the upright position at first. The next scans were performed > 30 min after changing their posture to the supine position. We aimed to evaluate stable circulation in both postures, not immediately after posture change.

Statistical analysis

All statistical analyses were performed using commercially available software (IBM SPSS Statistics, version 23.0; IBM Corp., Armonk, NY, USA). Data are expressed as mean ± SD, and the Wilcoxon signed-rank test was used to determine the significance between the upright and supine positions. A P value of <0.05 was considered to indicate a significant difference.

Results

The IVC-V_mean, IVC-V_max, IVC-CA, IVCF_mean, and IVCF_max in the supine and upright positions are shown in Fig. 4 and Table 1. The IVC-V_mean, IVC-V_max, IVC-CA, IVCF_mean, and IVCF_max were all significantly lower in the upright position than in the supine position, with differences of 52% ± 33% (3.34 ± 2.29 cm/s vs. 8.56 ± 4.46 cm/s, respectively; P = 0.002; Fig. 4a), 36% ± 19% (4.10 ± 2.25 cm/s vs. 13.1 ± 5.09 cm/s, respectively; P = 0.002; Fig. 4b), 56% ± 18% (205 ± 121 mm² vs. 373 ± 207 mm², respectively; P = 0.008; Fig. 4c), 26% ± 18% (345 ± 229 mL/min vs. 1453 ± 462 mL/min, respectively; P = 0.002; Fig. 4d), and 19% ± 11% (432 ± 278 mL/min vs. 2454 ± 926 mL/min, respectively; P = 0.002; Fig. 4e), respectively. However, HR was significantly higher (116% ± 9.2%; P = 0.003) in the upright position (85 ± 8.4 beats/min) compared with the supine position (73 ± 6.9 beats/min) (Fig. 4f).

Fig. 4.

Effect of postural change on IVCF. (a) Mean IVCF velocity (IVC-V_mean), (b) maximum IVCF velocity (IVC-V_max), (c) cross-sectional area of the IVC (IVC-CA), (d) mean IVCF (IVCF_mean), (e) maximum IVCF (IVCF_max), and (f) HR in the upright and supine positions. Each box plot indicates the maximum, first quartile, median, third quartile, and minimum values. AAF, abdominal aortic flow; HR, heart rate; IVCF, inferior vena cava flow; ROI, region of interest.

Table 1.

Results of IVC.

	Upright position	Supine position	Upright/Supine	P*
IVC-V_mean (cm/s)	3.34 ± 2.29	8.56 ± 4.46	52% ± 33%	0.002
IVC-V_max (cm/s)	4.10 ± 2.25	13.1 ± 5.09	36% ± 19%	0.002
IVC-CA (mm²)	205 ± 121	373 ±207	56% ± 18%	0.008
IVCF_mean (mL/min)	345 ±229	1453± 462	26% ± 18%	0.002
IVCF_max (mL/min)	432 ± 278	2454 ± 926	19% ± 11%	0.002
HR (beats/min)	85 ± 8.4	73 ± 6.9	116% ± 9.2%	0.003

Values are given as mean ± SD.

*Wilcoxon signed-rank test.

HR, heart rate; IVC-CA, cross-sectional area of IVC; IVCF, inferior vena cava flow; IVC-V, IVCF velocity.

The AA-V_mean, AA-V_max, AA-CA, AAF_mean, and AAF_max in the supine and upright positions are shown in Fig. 5 and Table 2. The AA-V_mean, AA-V_max, AAF_mean, and AAF_max were all significantly lower in the upright position than in the supine position, with differences of 33% ± 13% (2.10 ± 0.79 cm/s vs. 6.71 ± 2.30 cm/s, respectively; P = 0.002; Fig. 5a), 33% ± 22% (5.22 ± 2.39 cm/s vs. 17.9 ± 6.52 cm/s, respectively; P = 0.002; Fig. 5b), 42% ± 21% (236 ± 113 mL/min vs. 694 ± 312 mL/min, respectively; P = 0.002; Fig. 5d), and 37% ± 28% (572 ± 303 mL/min vs. 1823 ± 630 mL/min, respectively; P = 0.003; Fig. 5e), respectively. However, there were no differences in AA-CA between the upright position and the supine position (184 ± 50.1 mm² vs.175 ± 43.6 mm², respectively; P = 0.583; Fig. 5c).

Fig. 5.

Effect of postural change on AAF. (a) Mean AAF velocity (AA-V_mean), (b) maximum AAF velocity (AA-V_max), (c) cross-sectional area of the AA (AA-CA), (d) mean AAF (AAF_mean), and (e) maximum AAF (AAF_max) in the upright and supine positions. Each box plot indicates the maximum, first quartile, median, third quartile, and minimum values. AAF, abdominal aortic flow.

Table 2.

Results of AA.

	Upright position	Supine position	Upright/Supine	P*
AA-V_mean (cm/s)	2.10 ± 0.79	6.71 ± 2.30	33% ± 18%	0.002
AA-V_max (cm/s)	5.22 ± 2.39	17.9 ± 6.52	33% ± 22%	0.002
AA-CA (mm²)	184 ± 50.1	175 ± 43.6	108% ± 22%	0.583
AAF_mean (mL/min)	236 ± 113	694 ± 312	42% ± 21%	0.002
AAF_max (mL/min)	572 ± 303	1823 ± 630	37% ± 28%	0.003

Values are given as mean ± SD.

*Wilcoxon signed-rank test.

AA-CA, cross-sectional area of AA; AAF, abdominal aortic flow; AA-V, AAF velocity.

Discussion

In the present study, we evaluated IVCF and AAF in two positions using a multi-posture MRI system. We found that IVC-V_mean, IVC-V_max, IVC-CA, IVCF_mean, and IVCF_max were all significantly decreased in the upright position, suggesting that IVCF may be affected by gravity in the upright position.

Systemic circulatory changes, such as fluid shifts in the upright position, may affect IVCF as blood pressure and vascular resistance in the lower body increase in the upright position. Thus, there is a fluid shift effect that corresponds to an approximate 30% decrease in cardiac output (1,11). The effects of postural changes on IVCF remain unclear. In a previous study, postural changes in portal venous flow assessed using multi-posture MRI showed similar trends to our IVCF data (7). IVCF reflects a wider circulation than the portal venous flow, which supplies one organ. Thus, in the present study, we can infer that similar blood flow change due to postural change is occurring in a wide area of the lower body. Further, in that study and in a phantom study (6), it was confirmed that multi-posture MRI provides an equivalent estimate of blood flow to that using conventional MRI.

Several ultrasound (US) studies have reported changes in IVC diameter with postural changes, although the findings were not constant. For example, Mookadam et al. reported that IVC diameter was larger in the supine compared with the left lateral position (12). By contrast, Kundra et al. reported that IVC diameter was significantly higher in the left lateral position compared with supine position (13). These studies suggest the difficulty in objective evaluation with US. Thus, we consider that vessel morphology evaluation by MRI is more useful as it requires less dependence on the skill of the operator and the physique of the individual.

With respect to blood flow, phase-contrast MRI was reported to have less variability and greater reproducibility for flow values than that using US (10,14). Thus, we suggest that MRI measurement is better suited to detecting blood flow alterations caused by postural changes. We also found that AA-V_mean, AA-V_max, AAF_mean, and AAF_max were significantly decreased in the upright position, despite the direction of the vector power between the gravity and AAF in the upright position being the same. These AAF changes in the upright position may be affected by a decrease of IVCF and the fluid shift effect. Further, in the upright position, the gravity may make the pump effect of the heart in ascending aorta be more difficult. However, AA-CA was not changed significantly in each position. These different responses of the AA-CA and IVC-CA may relate to differences in the vessel wall structure.

In addition, the change-ratios of the mean values (IVC-V_mean, IVCF_mean, AA-V_mean, and AAF_mean) were similar to those of the maximum values (IVC-V_max, IVCF_max, AA-V_max, and AAF_max), indicating that the shape of the flow curves in the cardiac cycle was unchanged, even in the upright position. Further, HR was significantly increased when changing from the supine to the upright position, which may be explained by the increased sympathetic tone in response to the associated fluid shift.

In the present study, we used a conventional phase-contrast sequence and a novel MRI scanner. In the future, it is expected that more knowledge will be obtained by using advanced sequences, such as 4D flow sequences with the multi-posture MRI.

The present study has several limitations. First, we had a small number of participants, all of whom were healthy adult men. Thus, there may be differences within other groups of individuals or in patients with heart disease. Second, the magnetic field in the multi-posture MRI system was only 0.4 T, although a prior phantom flowmetry study confirmed the high accuracy and precision of this MRI system (6). Further, the low magnetic field needed breath-hold scan to improve signal-to-noise ratio. A breath-hold in inspiration may reduce IVCF (1); however, postural change is an independent factor for IVCF.

In conclusion, our data suggest that the effect of gravity during postural change from a supine to an upright position significantly decreases IVCF and AAF. Multi-posture MRI provides more detailed information on the circulation system. As IVCF and AAF are clinically important, clarifying the effect of gravity on IVCF and AAF during a postural change from the supine to the upright position may help to improve the management of patients with circulatory disease, such as heart failure. Furthermore, the multi-posture MRI can help diagnosing diseases caused by postural change, such as postural hypotension. Specifically, the increase in IVCF and AAF while recumbent may help maintain systemic circulation when compared with the upright position.

Footnotes

Acknowledgements

We appreciate Dr Hiroshi Furusho for his valuable advices for systemic circulation. We thank Edanz Group () for editing a draft of this manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Yoshisuke Kadoya

Satoshi Kobayashi

References

Ganong

WF.

Review of Medical Physiology. 22nd edn. New York, NY: McGraw-Hill Companies, 2005.

Longo

Fauci

Kasper

, et al. Harrison’s Principles of Internal Medicine. 18th edn. New York, NY: McGraw-Hill Companies, 2012.

Browse NL, Burnand KG, Thomas ML, editors. Disease of Veins. 1st ed. London: Edward Arnold, 1988.

O’Donnell

NMR blood flow imaging using multiecho, phase contrast sequences. Med Phys 1985; 12:59–64.

Lotz

Meier

Leppert

, et al. Cardiovascular flow measurement with phase-contrast MR imaging: basic facts and implementation. Radiographics 2002; 22:651–671.

Ohno N, Miyati T, Hiramatsu Y, et al. Quantitation of Venous Blood Flow in Gravity MRI: A Phantom Study. Med Imag and Info Sci 2017;34:141–143.

Kadoya Y, Miyati T, Kobayashi S, et al. Effect of gravity on portal venous flow: Evaluation using multiposture MRI. J Magn Reson Imaging 2019;50:83–87.

Evans

Iwai

Grist

, et al. Magnetic resonance imaging of blood flow with a phase subtraction technique. In vitro and in vivo validation. Invest Radiol 1993; 28:109–115.

Alperin

Lee

Bagci

AM.

MRI measurements of intracranial pressure in the upright posture: The effect of the hydrostatic pressure gradient. J Magn Reson Imaging 2015; 42:1158–1163.

10.

Yzet

Bouzerar

Allart

, et al. Hepatic vascular flow measurements by phase contrast MRI and doppler echography: a comparative and reproducibility study. J Magn Reson Imaging 2010; 31:579–588.

11.

Smith

Porth

Erickson

Hemodynamic response to the upright posture. J Clin Pharmacol 1994; 34:375–386.

12.

Mookadam

Warsame

Yang

, et al. Effect of positional changes on inferior vena cava size. Eur J Echocardiography 2011; 12:322–325.

13.

Kundra

Arunsekar

Vasudevan

, et al. Effect of postural changes on inferior vena cava dimensions and its influence on haemodynamics during caesarean section under spinal anaesthesia. J Obstet Gynaecol 2015; 35:667–671.

14.

Lycklama a Nijeholt

Burggraaf

Wasser

, et al. Variability of splanchnic blood flow measurements using MR velocity mapping under fasting and post-prandial conditions–comparison with echo-Doppler. J Hepatol 1997; 26:298–304.