Flow Dynamics Affecting Blood Draws

For CT angiographic studies, optimal contrast opacification of the vasculature depends on several factors, including injection delay and duration, scan duration, injection rate, total volume of contrast medium administered [1], and concentration or iodine content of the contrast medium. Poor contrast opacification can result in an examination that does not yield enough information for diagnosis or lead to misdiagnosis. Optimal opacification of the pulmonary arteries is necessary for confident CT evaluation of pulmonary emboli. In 37% of CT pulmonary angiograms, however, varying degrees of a transient decrease in pulmonary artery opacification can occur during scanning [2] despite use of a contrast injection rate and scan delay that should have been sufficient. This transient interruption of contrast enhancement has been described as a physiologic artifact associated with initial inspiration immediately before the scan.

Changes in intrathoracic pressure during the respiratory cycle can affect the diameter of and flow velocity in the inferior vena cava (IVC) [3-5]. An increase in unopacified venous return from the IVC during inspiration resulting in dilution of the contrast column may be an explanation for poor contrast opacification during some pulmonary CT angiographic studies [2].

Noninvasive quantitative measurement of blood flow through a vessel can be performed with velocity-encoded cine MRI [6-9]. The purpose of our observational study was to use MRI to examine changes in blood flow in the superior vena cava (SVC) and IVC during respiratory maneuvers to improve understanding of the possible effects of respiration on contrast dynamics and of transient interruption of contrast artifact on pulmonary CT angiographic studies.

Study Subjects

Ten healthy subjects (six men, four women) with no history or symptoms of heart disease participated in the study. Their median age was 30 years (range, 25-55 years). The study was approved by the institutional review board, and informed consent was obtained from the subjects. Blood flow in the SVC and IVC was measured in each subject during five respiratory maneuvers: free breathing (baseline), continuous inspiration, breath-hold at end inspiration, Valsalva maneuver, and breath-hold at end expiration. To perform the Valsalva maneuver, the subjects were instructed to perform a forced expiration against a closed glottis. To prolong continuous inspiration for the duration of MRI data acquisition, the subjects were instructed to hold their lips loosely apposed and to breathe in continuously against this resistance. The respiratory maneuvers were practiced outside of the MRI unit before the MRI examination.

View larger version (193K)

Fig. 1 —28-year-old woman in good health. Axial phase image obtained between azygous vein confluence and right atrium with region of interest placed on superior vena cava.

View larger version (208K)

Fig. 2 —30-year-old man in good health. Axial phase image obtained below level of right atrium with region of interest placed on inferior vena cava.

Velocity-Encoded Cine MRI

Quantitative measurements of blood flow in the SVC and IVC per cardiac cycle were performed with a retrospectively ECG-gated, velocity-encoded cine gradient-echo sequence on a 1.5-T MRI unit (Espree, Siemens Medical Solutions) with a phased-array torso coil.

Blood flow in the SVC was measured on transaxial slices obtained at a level between the azygous confluence and the right atrium (Fig. 1 ). Blood flow in the IVC was measured on transaxial slices obtained at the level of the diaphragm between the right atrium and hepatic vein confluence (Fig. 2 ). The orientation of the slices was perpendicular to the direction of flow. For both locations, the table was repositioned so that the slice level was within 3 cm of isocenter.

For the baseline free-breathing measurements, a non-breath-hold retrospectively gated sequence was used with the following parameters: TR/TE, 42.0/3.3; matrix size, 192 × 256; flip angle, 30°; number of segments, three; number of signals averaged, three; slice thickness, 6 mm; velocity encoding, 100 cm/s; average scan duration, 2 minutes 56 seconds to ensure adequate sampling of the entire respiratory cycle. For the respiratory maneuver measurements, the breath-hold retrospectively gated imaging sequence parameters were 74.0/2.8; matrix size, 96 × 256; flip angle, 30°; number of segments, 6; number of signals averaged,1; slice thickness, 6 mm; velocity encoding, 100 cm/s. The data were acquired over 16 cardiac cycles, and each measurement lasted approximately 12-16 seconds depending on the subject's heart rate. These sequences provided both phase and magnitude images. If aliasing was observed on the images, the measurement was repeated with velocity encoding of 150 cm/s.

Each measurement was performed twice, and the average of the two measurements was used to represent venous blood flow during each respiratory maneuver at each anatomic location. To allow hemodynamic equilibration after each respiratory maneuver, 2 minutes was allowed before each new measurement.

Image Analysis

The images were sent to an independent workstation (Leonardo, Siemens Medical Solutions), and flow measurement data were obtained with commercially available software (Argus, Siemens Medical Solutions). A region of interest was drawn manually on one image with visible flow and then propagated across all of the phases. The region of interest was then manually corrected on each phase to account for changes in vessel position and contour during the cardiac cycle.

Statistical Analysis

Numeric data were summarized with the sample median and range. A paired Student's t test was used to make all pairwise comparisons in blood flow between the stages of breathing. For comparisons of blood flow at more than two stages of breathing, a mixed-effects model was used with a fixed effect included for stage of breathing and a random effect included for patient. The following comparisons were considered: total systemic venous return during free breathing was compared with the total systemic venous return during the other four respiratory maneuvers; blood flow in the SVC and IVC at baseline was compared with blood flow in the SVC and IVC during the other four respiratory maneuvers; changes in blood flow from baseline to the other four respiratory maneuvers, expressed both as a difference and as a ratio, were compared between the SVC and the IVC; ratio between blood flow in the IVC and blood flow in the SVC at baseline was compared with the IVC to SVC blood flow ratio during the other four respiratory maneuvers. To partially account for the number of tests performed, only p <0.01 was considered statistically significant.

Total Systemic Venous Return

Table 1 shows systemic venous return during the five respiratory maneuvers. Total systemic venous return was higher (p < 0.001) than at baseline during continuous inspiration and was lower than at baseline during the Valsalva maneuver (p = 0.007) and during breath-hold at end inspiration (p = 0.003).

TABLE 1: Total Systemic Venous Return Measured by Velocity-Encoded MRI in 10 Healthy Volunteers

View larger version (15K)

Fig. 3 —Graph shows blood flow measured with velocity-encoded MRI in superior vena cava (black bars) and inferior vena cava (white bars) during respiratory maneuvers in 10 healthy volunteers. Asterisks indicate p < 0.01 versus baseline (free breathing).

Blood Flow in the SVC and IVC

Figure 3 shows blood flow in the SVC and the IVC during the five respiratory maneuvers. Blood flow in the SVC was higher during continuous inspiration than at baseline (p <0.001). Blood flow in the IVC was higher (p < 0.001) than at baseline during continuous inspiration and lower (p = 0.004) than at baseline during the Valsalva maneuver. A trend toward lower blood flow in the IVC during breath-hold at end inspiration than during baseline was not statistically significant (p = 0.026).

Changes in Blood Flow Between SVC and IVC

Table 2 shows the changes in blood flow from baseline to the other four respiratory maneuvers represented as both differences and as ratios for the SVC and the IVC. The increase in blood flow from baseline to continuous inspiration was higher (p < 0.001) in the IVC than in the SVC. The ratio of blood flow during continuous inspiration to blood flow during baseline was higher for the IVC than for the SVC (p = 0.009).

TABLE 2: Changes in Blood Flow from Baseline to Other Respiratory Maneuvers Between Superior and Inferior Venae Cavae

Ratio Between Blood Flow in the IVC and Blood Flow in the SVC

Table 3 shows the IVC to SVC blood flow ratio for different respiratory maneuvers. IVC to SVC blood flow ratio was higher (p = 0.009) than at baseline during continuous inspiration and lower than at baseline during the Valsalva maneuver (p = 0.008). A trend toward lower IVC to SVC blood flow ratio during breath-hold at end inspiration than during baseline was not statistically significant (p = 0.016).

TABLE 3: Ratio of Blood Flow in the Inferior Vena Cava (IVC) to Blood Flow in the Superior Vena Cava (SVC) During Respiratory Maneuvers

In this study we used MRI for noninvasive evaluation of the manner in which the SVC and IVC contributions to systemic venous return to the thorax vary with different respiratory maneuvers. When intrathoracic pressure remains constant, as during breath-holds at end inspiration and end expiration, the IVC to SVC flow ratio is very similar to that during free breathing, the ratio of IVC flow to SVC flow being approximately 2:1. During continuous inspiration, however, intrathoracic pressure becomes negative. This change in pressure gradient increased venous return to the right atrium nearly 50% according to our measurements. The percentage increase in IVC flow during inspiration is significantly greater than the increase in SVC flow. Therefore the contribution of IVC flow to total systemic venous return is transiently greater during inspiration, the IVC to SVC flow ratio being 2.4:1 according to our measurements. Because contrast material is almost always administered through an arm vein, this relative increase in unopacified blood from the IVC dilutes contrast medium injected at a constant rate.

Pulmonary CT angiograms are typically obtained with one or more preparatory breaths and then a deep inspiration a few seconds after injection of contrast medium has begun, just before image acquisition begins. The blood volume with a lesser concentration of contrast medium produced by these initial deep inspirations travels to the pulmonary arteries by the time scans at the level of the pulmonary arteries are acquired resulting in the observed decrease in pulmonary artery opacification while opacification in the right atrium and aorta may be satisfactory.

View larger version (116K)

Fig. 4A —42-year-old man with anxiety and dyspnea. Initial pulmonary CT angiogram obtained with deep prescan inspiration shows poor pulmonary artery opacification but dense contrast enhancement in superior vena cava and aorta consistent with transient interruption of contrast artifact.

View larger version (157K)

Fig. 4B —42-year-old man with anxiety and dyspnea. Pulmonary CT angiogram obtained 5 minutes after A with same contrast injection volume, rate, and timing delay but in expiration with no preliminary inspiration shows opacification of pulmonary artery is markedly better than in A.

Individual variation in the manner in which patients follow breathing instructions and the forcefulness of their inspiratory effort may explain why this artifact is seen on some but not all pulmonary CT angiograms. For example, in two of our subjects, inspiratory IVC flow increased only 15%, but in one subject it increased 140%. The physiologic reason for this relative increase in IVC flow is not explained by our results. A possible mechanism may be that diaphragmatic descent during inspiration increases intraabdominal pressure [10]. With positive intraabdominal pressure and negative intrathoracic pressure, the greater pressure gradient along the IVC than along the SVC results in greater augmentation of IVC flow.

The breath-hold at end inspiration and the end-expiration maneuvers were included because they correspond with verbal instructions that typically are given to patients for routine clinical CT examinations, such as "take a deep breath in and hold it" and "breath out and hold your breath out." Our study subjects were all medical professionals who understood the concept and performance of the Valsalva maneuver. The Valsalva maneuver produced the greatest decrease in venous return and therefore would be expected to result in the best contrast opacification of the pulmonary arteries. It would be difficult, however, for most patients to sustain a Valsalva maneuver for the entire length of a pulmonary CT angiogram, from the beginning of contrast injection to completion of the scan.

The small number of study subjects limited the power of our study. Our findings in healthy volunteers may not apply to patients with elevated right atrial pressure, such as patients with tricuspid disease, pericardial disease, or pulmonary hypertension. The extent of changes in venous flow during respiratory maneuvers varies with respiratory effort, but we did not use a spirometer to quantify or normalize for the respiratory effort. The MR sequence we used gave an average of flow over several cardiac cycles and did not provide instantaneous, real-time measurements. To generate negative intrathoracic pressure continuously for the length of the acquisition for the inspiration measurements, study subjects were asked to breath in slowly against resistance. This maneuver may not accurately simulate the physiologic mechanism of the initial inspiration before thoracic CT scans.

The results of our velocity-encoded MRI study of blood flow in the SVC and IVC during different respiratory maneuvers support the hypothesis that dilution of contrast material resulting from a greater increase in blood flow in the IVC than in the SVC during inspiration can cause transient interruption of the contrast bolus on pulmonary CT angiograms. On the basis of these findings, we recommend that pulmonary CT angiography be performed with only a small inspiration or no inspiration before scanning. Deep inspiration immediately before scanning should be avoided. If the transient interruption of contrast artifact is recognized by the technologist at the time of the examination, the examination can be repeated with no inspiration (Fig. 4A , 4B ) to improve opacification of the pulmonary artery.

Address correspondence to R. S. Kuzo ([email protected]).

1. Bae KT. Peak contrast enhancement in CT and MR angiography: when does it occur and why? Pharmacokinetic study in a porcine model. Radiology 2003; 227:809-816 [Google Scholar]

2. Gosselin MV, Rassner UA, Thieszen SL, Phillips J, Oki A. Contrast dynamics during CT pulmonary angiogram: analysis of an inspiration associated artifact. J Thorac Imaging 2004; 19:1-7 [Google Scholar]

3. Natori H, Tamaki S, Kira S. Ultrasonographic evaluation of ventilatory effect on inferior vena caval configuration. Am Rev Respir Dis 1979; 120:421-427 [Google Scholar]

4. Grant E, Rendano F, Sevinc E, Gammelgaard J, Holm HH, Gronvall S. Normal inferior vena cava: caliber changes observed by dynamic ultrasound. AJR 1980; 135:335-338 [Abstract] [Google Scholar]

5. Wexler L, Bergel D, Gabe I, Makin G, Mills C. Velocity of blood flow in normal human venae cavae. Circ Res 1968; 23:349-359 [Google Scholar]

6. Mohiaddin RH, Wann SL, Underwood R, Firmin DN, Rees S, Longmore DB. Vena caval flow: assessment with cine MR velocity mapping. Radiology 1990; 177:537-541 [Google Scholar]

7. Eyskens B, Mertens L, Kuzo R, et al. The ratio of flow in the superior and inferior caval veins after construction of a bidirectional cavopulmonary anastomosis in children. Cardiol Young 2003; 13:123-130 [Google Scholar]

8. Lotz J, Meier C, Leppert A, Galanski M. Cardiovascular flow measurement with phase-contrast MR imaging: basic facts and implementation. RadioGraphics 2002; 22:651-671 [Google Scholar]

9. Eichenberger AC, Schwitter J, McKinnon GC, Debatin JF, von Schulthess GK. Phase-contrast echoplanar MR imaging: real-time quantification of flow and velocity patterns in the thoracic vessels induced by Valsalva's maneuver. J Magn Reson Imaging 1995; 5:648-655 [Google Scholar]

10. Takata M, Robotham JL. Effects of inspiratory diaphragmatic descent on inferior vena caval venous return. J Appl Physiol 1992; 72:597-607 [Google Scholar]

Flow Dynamics Affecting Blood Draws

Source: https://www.ajronline.org/doi/10.2214/AJR.06.5035

Share this post