Skip Navigation

Cardiovascular Research 1997 33(1):156-163; doi:10.1016/S0008-6363(96)00175-7
© 1997 by European Society of Cardiology
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Sloth, E.
Right arrow Articles by Hasenkam, J.M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sloth, E.
Right arrow Articles by Hasenkam, J.M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Copyright © 1997, European Society of Cardiology

Where to place the Doppler sample volume in the human main pulmonary artery: evaluated from magnetic resonance phase velocity maps

Erik Slotha,c,1, Kim C. Houlindb,c,1, Erik M. Pedersenb,c,1 and J.Michael Hasenkamb,c,1

aDepartment of Anaesthesia, Skejby Sygehus, Aarhus University Hospital, 8200 Aarhuus N, Denmark
bDepartment of Cardiothoracic and Vascular Surgery T,Skejby Sygehus, Aarhus University Hospital,8200 Aarhuus, Denmark
cInstitute of Experimental Clinical Research and MR-Center, Aarhus Kommune Hospital, Aarhuus, Denmark

Received 20 February 1996; accepted 16 July 1996


   Abstract
Top
 Abstract
1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 
Objective: To give recommendations for the placement of Doppler sample volumes for blood flow assessment in the human main pulmonary artery. Methods: In 10 healthy volunteers MR-phase velocity measurements were obtained and computing of the mean temporal blood velocity data was performed to guide single point Doppler velocity recordings. Results: The mean temporal blood velocity profiles were consistently skewed with the lowest blood velocities towards the inferior/right vessel wall. Blood velocity indices (ratio of point to mean velocities, where a point equals a square of 4 pixels) varied considerably with the lowest indices located towards the inferior/right vessel wall. A centrally located fictive sample volume revealed an average blood velocity index value (average of all 10 subjects) of 1.08 (range 0.99–1.25; s.d. 0.08) where the central point was defined at maximum systole, and a value of 1.13 (range 0.97–1.34; s.d. 0.11) when the central point was defined in end-diastole. The mean of multiple sample volumes along the inferior/right to superior/left diameter revealed a blood velocity index of 1.01 (range 0.87–1.21; s.d. 0.09) in systole and 1.03 (range 0.87–1.19; s.d. 0.09) in diastole. Conclusions: For practical clinical purposes, single point estimation of the mean blood velocity in the pulmonary artery should be performed centrally. The use of multiple sample volumes placed along the inferior/right to superior/left diameter improves the mean velocity estimate in healthy volunteers. Further studies should be conducted to reinforce the basis for Doppler velocity recording in the diseased human pulmonary artery as well as to investigate other important determinants of Doppler-derived CO, namely angle of insonation and assessment of the cross-sectional area.

KEYWORDS Pulmonary artery; Blood velocity profile; Human; NMR; Doppler velocity recording


   1. Introduction
Top
 Abstract
 1. Introduction
2. Methods
 3. Results
 4. Discussion
 References
 
Cardiac output measurements are frequently used for the assessment of cardiac function in the perioperative period and in the intensive care unit. The fact that several methods for cardiac output determination have been applied merely reflects the lack of a true gold standard [1, 2]. So far, the cold water thermodilution technique has been considered the method of choice in clinical practice [1, 2]. Cardiac output can also be assessed as the product of the mean temporal and spatial blood velocities and the cross-sectional area where the blood velocity can be measured with the pulsed Doppler method. The main pulmonary artery has been the subject of several cardiac output studies using a variety of pulsed Doppler ultrasound techniques (e.g., transthoracic echocardiography [3, 4], transesophageal echocardiography [5–7] and intraluminal Doppler [8–10]). Regardless of which acoustic window is used, detailed information about the blood velocity profile is necessary if the vessel cross-sectional area exceeds the area of the Doppler sample volume (SV). Otherwise, substantial errors in blood velocity estimation due to a non-flat mean temporal profile may occur [11, 12]. Such detailed information about the main pulmonary artery (PA) blood velocity profile has been difficult to obtain in humans mainly due to lack of appropriate methods [11, 13–15].

Recently, however, detailed information has become available through the use of high-resolution magnetic resonance (MR) velocity mapping [16–19]. An initial study demonstrated the presence of a relatively large low-velocity area in the inferior/right part of the main PA (minor curvature) [19]. These data were not directly comparable to single point Doppler ultrasound data because the MR data were corrected for cardiac induced main PA movement[19]. MR velocity maps can, however, be analyzed with a fixed reference in space simulating single point pulsed Doppler velocity recordings where the Doppler SV is also fixed in space relative to the moving main PA.

We hypothesize that the blood velocity recorded in a spatially fixed SV is able to represent the mean main PA blood velocity, which can then be multiplied by the cross-sectional area to provide reliable cardiac output estimates. Therefore, the purpose of this study was to provide recommendations for single point pulsed Doppler estimates of mean blood velocity in the human main PA based on MR phase velocity maps.


   2. Methods
Top
 Abstract
 1. Introduction
 2. Methods
3. Results
 4. Discussion
 References
 
2.1. Population
Ten healthy, unsedated volunteers, 4 female and 6 male (24–36 years; mean 27.1 years) were studied (Table 1). The investigation was approved by the local ethical committee according to the Helsinki II Declaration and individual informed consent was obtained.


View this table:
[in this window]
[in a new window]

  		Table 1 Demographic and hemodynamic data for all volunteers

 
2.2. Data collection and data analysis
In all 10 subjects MR phase velocity maps were obtained at the mid-main PA level (Fig. GR1). A number of blood velocity indices were calculated to evaluate the impact of sampling from different locations across the main PA (see below). All measurements took place during expiration without any sedation. The total MR measuring time was approximately 40 min for each volunteer depending on heart rate and respiration frequency.


Figure 1
View larger version (64K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR1 Magnetic resonance images of the pulmonary artery. (A) transverse; (B) sagittal. Lines intersecting mid pulmonary artery in both planes are shown. AA = ascending aorta; DTA = descending thoracic aorta; MPA = main pulmonary artery; PV = pulmonary valve; RVOT = right ventricular outflow tract.

 
2.2.1. MR-scanning
The MR scanner used was a 1.5 T 15S Philips GY-ROSCAN HP whole body scanner. A spin echo sequence was used for 2D imaging and a FLAG (FLow Adjusted Gradient) sequence [20] was applied for velocity measurement. Combined respiratory and cardiac triggering were used in all subjects and data acquisition was adjusted to take place during expiration. Cardiac triggering started flow measurements 8 ms after the R-wave in the ECG. From a transversal and sagittal modulus image (see Fig. GR1) a double oblique plane of the mid-main PA was identified for velocity measurements which was performed as 2 averages of 256 phase encoding steps. The following MR-parameters were used: repetition time 25–34 ms, slice thickness 8 mm, in-plane resolution 1.4–2.5 mm2, Vmax = 1.25 m/s.

The velocity data were corrected for background phase error in a dedicated semi-automated software program [21]. In the same program background noise was visually masked by assigning zero velocity to pixels with a low signal amplitude. In order to distinguish the main PA flow from the surrounding cardiac structures, the main PA was manually traced by an experienced user of the semi-automated software program [21]. MR velocity data were transferred to a Macintosh computer for graphic display of the mean temporal blood velocity profiles using the Spyglass data visualization program (Spyglass Inc., Champagne, IL, USA). The principle of calculating the mean temporal blood velocity data is shown in Fig. GR2. Each frame represents a data-storing matrix which is fixed in space. The main PA moves relative to this matrix, schematically illustrated in Fig. GR2 A. Calculation of the mean temporal blood velocity data was performed relative to the matrix by averaging pixels having the same coordinates within each matrix. This simulates single point Doppler velocity recordings, where the transducer represents a fixed reference in space relative to the moving pulmonary artery.


Figure 2
View larger version (13K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR2 Schematic illustration of the method used for processing of the MR mean temporal blood velocity data. (A) Velocity data for different time frames 1-N (N referring to the last frame obtained in each measurement), see Table 1. Note the pulmonary artery movement during the heart cycle relative to the fixed data storing matrix. This simulates single-point Doppler velocity recordings where the transducer also represents a fixed reference in space. A fictive sample volume is indicated as a gray box and is placed centrally (center of the pulmonary artery maps indicated by a dot) during top systole (frame no. 4). The fictive sample volume remains in the same matrix position (same coordinates) throughout the cardiac cycle. (B) Computing of mean temporal blood velocity data. Gray box indicates extraction of temporal mean blood velocity data corresponding to the fictive sample volume in A, see Fig. GR3.

 
Heart rate and arterial blood pressure were measured before and after the MR investigation to verify stable hemodynamics (before MR: average mean BP 87.2 mmHg, s.d. 5.5; average heart rate 63.8 min–1, s.d. 10.7; after MR: average mean BP 85.8 mmHg, s.d. 6.6; average heart rate 63.4 min–1, s.d. 10.7) (Table 1).

2.2.2. Blood velocity indices
To provide recommendations and estimate potential errors for cardiac output based on single point Doppler estimates, a number of blood velocity indices were calculated.

Nine points (fictive sample volumes) (1–9) along 4 diameters and evenly distributed across the vessel area were selected (Fig. GR3). Since the predominant temporal blood velocity changes are expected to occur along the inferior/right to superior/left vessel diameter [16, 17, 19], 4 additional points (10–13) were selected along this diameter (Fig. GR3). Identification of the 13 points was conducted visually by superimposing a predefined template on the computer screen during analysis (Fig. GR3). Because the main PA moves while the template is fixed, the position of these points within the vessel cross-sectional area varies with its timing during the heart cycle. In order to evaluate this, we superimposed the template on a systolic blood velocity map (frame 4 in Fig. GR2 A) and a diastolic blood velocity map (frame N in Fig. GR2 A) defining the center (point 1) from these two maps. Because the center of the template is different for each map, different pixel values will be extracted from the mean temporal velocity data (Fig. GR2 B).


Figure 3
View larger version (23K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR3 Template superimposed on the computer screen used for selection of the fictive sample volumes on the MR time frames (Fig. GR2). Mean temporal blood velocity data at the different points (1–13) are extracted from Fig. GR2 B but defined from Fig. GR2 A by placing the center (point 1) in a top systolic frame or end-diastolic frame, respectively (Fig. GR2 A). This is possible because the data-storing matrices in both Fig. GR2 A and Fig. GR2 B have the same coordinates. Bold squares represent the location of 4 pixels (1–13) and dotted squares the location of the fictive sample volumes with increasing cross-sectional area. Circles represent 1/4, 2/4, 3/4 and 4/4 of internal vessel radius. S = superior; I = inferior; R = right; L = left.

 
A blood velocity index was calculated as the ratio of the time velocity integral (TVI) (extracted from Fig. GR2 B) in each of the 13 measuring points (Fig. GR3) to the mean time velocity integral (temporal and spatial mean velocity) (TVImean) (averaging all velocities in Fig. GR2 B). An index value of 1 thus indicates that the true temporal and spatial mean blood velocity can be estimated from TVI at that measuring point. Velocity indices measuring points defined at top systole were calculated as:


Formula 1

(1)
Similarly for points defined at end-diastole:


Formula 2

(2)

This was done for a square of 4 pixels at all 13 points (1–13) and also for a square of 9 or 16 pixels at the selected points 1, 2 and 6 in order to evaluate the impact of SV cross-sectional area on the blood velocity index (Fig. GR3). A comparison of indices based on Eq. (1) and Eq. (2) was used to assess the impact of positioning the SV from either a systolic or a diastolic image, respectively (Fig. GR2).

2.3. Statistical methods
Paired t-tests preceded by a probability plot to verify Normal distribution of the variables were used for comparison of blood velocity index values. A P-value of < 0.05 was considered statistically significant.


   3. Results
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
4. Discussion
 References
 
The mean temporal blood velocity profile was consistently skewed for all 10 subjects with a low velocity area at the inferior/right wall (minor curvature) (Fig. GR4). The average blood velocity index distribution (average of all 10 subjects) for points defined in systole and diastole showed the same characteristic pattern (Fig. GR5). Lowest index values were found towards the inferior/right (points 2 and 13). Index values with both systolic and diastolic definition of the fictive SV's were ≤ 1.00 at points 2, 3, 9, 12 and 13 as opposed to the other points where indices were > 1.00.


Figure 4
View larger version (71K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR4 Mean temporal blood velocity surface plot of velocities across the main pulmonary artery for one volunteer showing the low-velocity area towards the inferior/right. S = superior; I = inferior; R = right; L = left.

 

Figure 5
View larger version (30K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. GR5 Top panel: Distribution of the average blood velocity indices for all volunteers. Each index comprises a square of 4 pixels, and is calculated at several points. These points are defined in Fig. GR3, and are shown in the small circle for easy reference. It is seen that indices derived from data with point 1 placed in the center of the velocity map in systole (left) are different with point 1 placed centrally in the velocity map in diastole (right). White typing indicates index values ≤ 1 in both systole and diastole. Bottom panel: Corresponding graphs showing average blood velocity indices along the inferior/right to superior/left diameter (points 13–11). Note the gradual increase in index value as one approaches the vessel wall. S = superior; I = inferior; R = right; L = left.

 
Along a diameter from the inferior/right towards the superior/left the indices gradually increased (with the exception of point 11) (Fig. GR5). The range in average blood velocity index values was 0.59–1.32 for a systolic placement and 0.73–1.27 for a diastolic placement of the fictive SV corresponding to errors in cardiac output of –41 to +32% and –27 to 27%, respectively. In the most extreme case the blood velocity index value for a systolic placement of the fictive SV was 0.55 for point 13 and 1.88 for point 7. Systolic or diastolic placement of the central pixels had no significant influence on the central point blood velocity indices (P = 0.24). The values were 1.08 (range 0.99–1.25, s.d. 0.08) and 1.13 (range 0.97–1.34, s.d. 0.11) for the systolic and diastolic placements, respectively. Averaging all SV's lying in the inferior/right to superior/left diameter (mean of points 1, 2, 6, 10, 11, 12 and 13) yielded a blood velocity index of 1.01 (range 0.87–1.21, s.d. 0.09) for systole and 1.03 (range 0.87–1.19, s.d. 0.09) for diastolic placement.

3.1. Size of the fictive sample volume
Increasing the size of the fictive SV from a square of 4 pixels (mean 6.2 mm2, range 5.5–10 mm2) to 9 or 16 pixels (mean 21.5 mm2, range 16.8–22.5 mm2) for points 1, 2 and 6 did not change the blood velocity indices significantly (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 Velocity indices for the selected points 1, 2 and 6 (see Fig. GR3) based on Eq. (1)(S = center point defined in a systolic frame) and Eq. (2)(D = center point defined in an end-diastolic frame) for different sizes of the fictive sample volume

 

   4. Discussion
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
References
 
Regardless of whether the transesophageal, transthoracic or intraluminal Doppler modality is applied, it is of paramount importance to acknowledge the mean temporal blood velocity profile. Failure to do so may incur substantial errors in mean blood velocity measurements and hence cardiac output determination. Using MR phase velocity maps and an analysis which simulates Doppler velocity recording, we found a consistently skewed mean temporal blood velocity profile with a low velocity area towards the inferior/right vessel wall. A similar velocity distribution has been described in healthy volunteers [16, 17, 19] and patients with PA hypertension where the reverse flow area is augmented and the velocity distribution more irregular[16, 17, 22]. In a previous work using multiplane transesophageal Doppler echocardiography for PA velocity mapping we found an almost flat mean temporal blood velocity profile in 8 healthy subjects [23]. The reason for this may be that the multiplane Doppler technique is too insensitive to detect the full extent of reverse flow due to limited spatial and temporal resolution [23]. This is supported by a study carried out by Okamoto et al. using transthoracic pulsed Doppler in the PA in healthy volunteers and in patients with PA hypertension [24]. Only in the latter group was prominent reverse flow detected [24]. The presence of a low blood velocity region towards the inferior/right resulting in a skewed mean temporal blood velocity profile is well documented from human [16, 17, 19], animal [15, 25–27] and in vitro studies [28].

The MR phase velocity maps in this study provide a quantitative approach which can be used to estimate errors in cardiac output measurement based on single point Doppler ultrasound SV's in the main PA. The average blood velocity index distributions in Fig. GR5 provide guidelines for placement of the Doppler SV in healthy young persons since the index value represents the factor of mis-estimation from a point which is not temporaly and spatially representative of the mean blood velocity.

The systolic point 3 revealed an average index value of 1.0 which is optimal since correction due to a non-flat mean temporal blood velocity profile can be omitted. It is, however, associated with a considerable standard deviation of 0.20. Point 12, assessed during diastole, also reveals an average index value of 1.0 with an acceptable standard deviation of 0.11. Point 12 is however situated close to the area where retrograde flow occurs [16, 17, 19] and only subtle movements towards the inferior/right vessel wall may dramatically reduce the index value (Fig. GR5). On that basis we recommend placing a single SV centrally during systole where the major contribution of total volume flow takes place. The recorded velocity should, for quantitative purposes, be multiplied by a factor 0.93 (1/1.08) according to the average blood velocity index from this study. It is important to acknowledge that the indices which are referred to above represent an average of 10 healthy persons and may not be valid in patients where the velocity distribution may be different. For the entire study population the central point may account for an underestimation of 1% to an overestimation of 25% for systolic placement of the central fictive SV and 3% underestimation to 34% overestimation for a diastolic placement of the central SV. For quantitative purposes, this is not acceptable for a clinical tool where the overall error is further augmented because of difficulties in angle correction. However, this method does not address the problems in assessing vessel cross-sectional area, another large source of error in cardiac output determination. For individual monitoring purposes, on the other hand, this may be acceptable.

In this study we used MR phase velocity maps to obtain information about the impact of different sampling sites in the main PA. The data are analyzed in such a way that they simulate pulsed Doppler velocity recordings. In several validation studies using different reference methods, MR phase velocity measurement correlated well with other flow measurement techniques [29–31]. Despite the limit value of r-values in assessing agreement [32, 33] it allows some comparison between different investigations [7]. Studying the x-y plots (cardiac output determined by MRI plotted against cardiac output determined by the reference method) in the papers mentioned, good agreement is made probable [29–31]. In the present study we did not calculate volume flow but used the velocity maps. No repeatability studies were conducted for ethical reasons due to time constraints in the investigational procedure. The positioning of the scanning planes was done from a sagittal and transversal image guided by the pulmonary valve and bifurcation making use of the ability to look at successive slices. Because no repeatability study was conducted, repeatability of the assessment of imaging planes was not evaluated. Tracing of the PA is rather subjective but necessary to avoid flow from surrounding structures, especially the aorta. Tracing was performed by an experienced user of the semi-automated software program. Because of the relatively small pixel size compared to the PA cross-sectional area, partial volume effects are expected to be of minor importance and we did not evaluate the impact of this in the present study. Since the main PA is rather short, the slice thickness should be kept small. The slice thickness in the present study was 8 mm, still big enough to provide adequate signals. Other groups have used a larger slices for main PA phase velocity mapping [16, 18].

Our results reinforce the importance of spatial handling of the SV in single point cardiac output determinations (Fig. GR5). Individually the error may be substantial (e.g., in volunteer No. 4, as the worst case, where the velocity index range is 0.55–1.88, giving an overall error range in velocity estimation of 133%). The finding of a mean blood velocity index of 1.01–1.03 across the inferior/right to superior/left diameter accompanied by a low standard deviation supports the need for multiple SV's. Newer and faster color Doppler algorithms may, in the near future, enable precise sampling and processing from multiple SV's without any compensatory measures due to suboptimal color frame rates.

4.1. Size of the fictive sample volume
One could speculate that an increased SV cross-sectional area would optimize the blood velocity index values in terms of minimizing interindividual variation, and give values close to 1.00 or even both. An increase from a square of 4 pixels to 9 or 16 pixels did not alter the index values significantly for the selected points 1, 2 and 6 (Fig. GR3 and Table 2). The ultrasound SV cross-sectional area for conventional use varies between 5 and 6 mm2, which corresponds nicely with the fictive SV of 4 pixels.

The potential error in cardiac output estimation which may occur due to a non-flat temporal mean blood velocity profile is substantial. For practical clinical purposes estimation of pulmonary artery blood flow by means of single point Doppler should be done by placing the SV in the central part of the main PA. The use of multiple SV's is highly recommended for optimizing the velocity estimate. We would emphasize, however, that this study was conducted in healthy volunteers. Available qualitative knowledge about the blood velocity distribution in PA hypertension[16, 17, 22] suggests that the velocity indices may not be applicable under pathological conditions. Further studies should be conducted to reinforce the basis for Doppler velocity recording in the diseased human PA as well as other important determinants of Doppler-derived CO, namely angle of insonation and assessment of the cross-sectional area.


   Acknowledgements
 
This work was supported by the Karen Elise Jensen Foundation, the John and Birthe Meyer Foundation, and the Danish Heart Foundation. We thank Lars Rybro M.D. for statistical advice.


   Notes
 
1 All authors are affiliated to the Cardiovascular Research Center, Aarhus University. Back


   References
Top
 Abstract
 1. Introduction
 2. Methods
 3. Results
 4. Discussion
 References
 

  1. Moore FA, Haenel JB, Moore EE. Alternatives to Swan-Ganz cardiac output monitoring. Surg Clin North Am (1991) 71:699–721.[Web of Science][Medline]
  2. Taylor SH, Silke B. Is the measurement of cardiac output useful in clinical practice? Br J Anaesth (1988) 60:90S–98S.[Medline]
  3. Labovitz AJ, Buckingham TA, Habermehl K, Nelson J, Kennedy HL, Willams GA. The effects of sampling site on the two-dimensional echo-Doppler determination of cardiac output. Am Heart J (1985) 109:327–333.[CrossRef][Web of Science][Medline]
  4. Ferrazza A, Marino B, Giusti V, Affinito V, Ragonese P. Usefulness of left and right oblique subcostal view in the echo-Doppler investigation of pulmonary arterial blood flow in patients with chronic obstructive pulmonary disease. The subxiphoid view in the echo-Doppler evaluation of pulmonary blood flow. Chest (1990) 98:286–289.[CrossRef][Web of Science][Medline]
  5. Muhiudeen IA, Kuecherer HF, Lee E, Cahalan MK, Schiller NB. Intraoperative estimation of cardiac output by transesophageal pulsed Doppler echocardiography. Anesthesiology (1991) 74:9–14.[CrossRef][Web of Science][Medline]
  6. Roewer N, Bednarz F, Schulte am Esch J. Continuous measurement of intracardiac and pulmonary blood flow velocities with transesophageal pulsed Doppler echocardiography: technique and initial clinical experience. J Cardiothorac Anesth (1987) 1:418–428.[CrossRef][Medline]
  7. Savino JS, Troianos CA, Aukburg S, Weiss R, Reichek N. Measurement of pulmonary blood flow with transesophageal two-dimensional and Doppler echocardiography. Anesthesiology (1991) 75:445–451.[Web of Science][Medline]
  8. Segal J, Nassi M, Ford AJ Jr, Schuenemeyer TD. Instantaneous and continuous cardiac output in humans obtained with a Doppler pulmonary artery catheter. J Am Coll Cardiol (1990) 16:1398–1407.[Abstract]
  9. Sega J, Gaudiani V, Nishimura T. Continuous determination of cardiac output using a flow-directed Doppler pulmonary artery catheter [see comments]. J Cardiothorac Vasc Anesth (1991) 5:309–315.[CrossRef][Medline]
  10. Segal J, Pearl RG, Ford AJJ, Stern RA, Gehlbach SM. Instantaneous and continuous cardiac output obtained with a Doppler pulmonary artery catheter. J Am Coll Cardiol (1989) 13:1382–1392.[Abstract]
  11. Somod L, Hasenkam JM, Kim WY, Nygaard H, Paulsen PK. Three dimensional visualisation of velocity profiles in the normal porcine pulmonary trunk. Cardiovasc Res (1993) 27:291–295.[Abstract/Free Full Text]
  12. Kim WY, Bisgaard T, Nielsen SL, et al. Two-dimensional mitral velocity profiles in pig models using epicardial Doppler echocardiography. J Am Coll Cardiol (1994) 24:532–545.[Abstract]
  13. Wolstenholme GEW, Knight J. Circulatory and Respiratory Mass Transport. London: J. & A. Churchill Ltd. 1969:172–199.
  14. Reuben SR, Swadling JP, Lee GdeJ. Velocity profiles in the main pulmonary artery of dogs and man, measured with a thin-film resistance anemometer. Circ Res (1970) 27:995–1001.[Abstract/Free Full Text]
  15. Somod L, Pedersen EM, Kim WY, Hasenkam JM, Nygaard H, Paulsen PK. Axial development of velocity fields in the porcine main pulmonary artery system. Heart Vessels (1994) 9:67–78.[CrossRef][Web of Science]
  16. Bogren HG, Klipstein RH, Mohiaddin RH, et al. Pulmonary artery distensibility and blood flow patterns: a magnetic resonance study of normal subjects and of patients with pulmonary arterial hypertension. Am Heart J (1989) 118:990–999.[CrossRef][Web of Science][Medline]
  17. Kondo C, Caputo GR, Masui T, et al. Pulmonary hypertension: pulmonary flow quantification and flow profile analysis with velocity-encoded cine MR imaging. Radiology (1992) 183:751–758.[Abstract/Free Full Text]
  18. Paz R, Mohiaddin RH, Longmore DB. Magnetic resonance assessment of the pulmonary arterial trunk anatomy, flow, pulsatility and distensibility. Eur Heart J (1993) 14:1524–1530.[Abstract/Free Full Text]
  19. Sloth E, Houlind KC, Oyre S, Kim WY, Pedersen EM, Jørgensen HS, Hasenkam JM. Three-dimensional visualization of the velocity profiles in the human main pulmonary artery using magnetic resonance phase velocity mapping. Am Heart J (1994) 128:1130–1138.[CrossRef][Web of Science][Medline]
  20. Groen JP, van Dijk P, In den Kleef JTE. Design of flow adjustable gradient waveforms. In: Abstracts, Society of Magnetic Resonance in Medicine, 6th Annual Meeting, New York, 1987;868.
  21. Walker PG, Cranney GB, Scheidegger MB, Waseleski G, Pohost GM, Yoganathan AP. Semiautomated method for noise reduction and background phase error correction on MR phase velocity data. JMRI (1993) 3:521–530.[CrossRef][Medline]
  22. Mohiaddin RH, Yang GZ, Kilner PJ. Visualization of flow by vector analysis of multidirectional cine MR velocity mapping. J Comput Assist Tomogr (1994) 18:383–392.[Web of Science][Medline]
  23. Sloth E, Pedersen EM, Nygaard H, Hasenkam JM, Juhl B. Multiplane transesophageal Doppler-echocardiographic measurements of the velocity profile in the human pulmonary artery. J Am Soc Echocardiogr (1994) 7:132–140.[Medline]
  24. Okamoto M, Miyatake K, Kinoshita N, Sakakibara H, Nimura Y. Analysis of blood flow in pulmonary hypertension with the pulsed Doppler flowmeter combined with cross sectional echocardiography. Br Heart J (1984) 51:407–415.[Abstract/Free Full Text]
  25. Frantz EG, Henry GW, Lucas CL, et al. Characteristics of blood flow velocity in the hypertensive canine pulmonary artery. Ultrasound Med Biol (1986) 12:379–385.[CrossRef][Web of Science][Medline]
  26. Katayama H, Henry GW, Lucas CL, et al. Three-dimensional visualization of pulmonary blood flow velocity profiles in lambs. Jpn Heart J (1992) 33:95–111.[Medline]
  27. Lucas CL, Henry GW, Ferreiro JI, Ha B, Keagy BA, Wilcox BR. Pulmonary blood velocity profile variability in open-chest dogs: influence of acutely altered hemodynamic states on profiles, and influence of profiles on the accuracy of techniques for cardiac output determination. Heart Vessels (1988) 4:65–78.[CrossRef][Medline]
  28. Lynch PG, Saylor A, Ha B, et al. The effects of curvature on fluid flow fields in pulmonary artery models: flow visualization studies. J Biomech Eng (1993) 115:97–103.[Web of Science][Medline]
  29. Møgelvang J, Stahlberg F, Søndergard L, Thomsen C, Lindvig K, Henriksen O. Determination of cardiac output by magnetic resonance imaging. Am J Noninvas Cardiol (1992) 6:201–204.
  30. Maier SE, Meier D, Boesiger P, Moser UT, Vieli A. Human abdominal aorta: comparative measurements of blood flow with MR imaging and multigated Doppler US. Radiology (1989) 171:487–492.[Abstract/Free Full Text]
  31. Firmin DN, Nayler GL, Klipstein RH, Underwood SR, Rees RS, Longmore DB. In vivo validation of MR velocity imaging. J Comput Assist Tomogr (1987) 11:751–761.[Web of Science][Medline]
  32. Altman DG, Bland JM. Measurement in medicine: the analysis of method comparison studies. Statist (1983) 32:307–317.[CrossRef]
  33. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet (1986) 1:307–310.[CrossRef][Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow E-letters: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when E-letters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Sloth, E.
Right arrow Articles by Hasenkam, J.M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sloth, E.
Right arrow Articles by Hasenkam, J.M.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?