Hrev_master Veins and Lymphatics 2014; volume 3:1863 [Veins and Lymphatics 2014; 3:1863] [page 1] On the consistency of flow rate color Doppler assessment for the internal jugular vein Francesco Sisini,1 Sergio Gianesini,2 Erica Menegatti,2 Angelo Taibi,1 Mirko Tessari,2 Giovanni Di Domenico,1 Anna Maria Malagoni,2 Mauro Gambaccini1 1Department of Physics and Earth Sciences, University of Ferrara; 2Vascular Diseases Center, University of Ferrara, Cona (FE), Italy Abstract Color Doppler methodology to assess the vessel blood flow rate is based on the time averaged velocity of the blood measured in the longitudinal plane and the cross sectional area measurement taken either in the longitudinal plane, by assuming circular cross sectional area, or in the transversal plane. The measure- ment option in longitudinal plane is based on the assumption of circular cross sectional area, while the transversal one needs to evalu- ate both time-averaged velocity and cross sec- tional area in the same vessel point. A precise and validated assessment methodology is still lacking. Four healthy volunteers underwent internal jugular vein colour Doppler scanning. The cross sectional area was assessed by means of B-mode imaging in the transversal plane all along the vessel cervical course. During this assessment, cross sectional area, major and minor axis of the vessel were meas- ured and recorded. The distance between the internal jugular vein wall and the skin surface were measured together with the intra-lumi- nal diameter and statistically correlated with the cross sectional area data. The internal jugular vein cross sectional area measured on the transversal plane were significantly differ- ent from the cross sectional area calculated using the assumption of circular shape. The intra-luminal distance showed high correla- tion with the measured cross sectional area. The proper anatomical point in the cross sec- tional area transversal measurement can be identified by using the internal jugular vein intra-luminal distance as landmark. Introduction Venous flow rate is a main measure scientif- ically recognized for the clinical practice.1-7 It is a derived value coming from the product of time-averaged velocity (TAV) and the cross sectional area (CSA), thus measurable by color Doppler (CD). Nowadays, the available CD software can calculate the flow value only along longitudinal (L) plane. Such value is obtained by contour- ing the Doppler spectrum wave and by calculat- ing the CSA, which is in turn obtained by the measurement of the vessel intra-luminal diameter. The assumption for the flow rate assessment is that the point where such diam- eter is measured has to be precisely positioned in the same anatomical site of the TAV sample volume. Moreover, a circular vessel shape must be assumed for this kind of CSA calculation. This last assumption guarantees that the above-described method can be considered reliable for the arterial evaluation thanks to the circular shape. On the contrary, both venous diameter and shape undergo extreme variations linked to trans-mural pressure and compliance features, together with surround- ing muscles extrinsic compressions,8,9 and thus suggesting the need for a more reliable methodology for the venous flow assessment. Nevertheless, many scientific investigations concerning venous flow rate still refer to a CSA that is obtained by assuming that the vein is a perfect cylinder. Such inaccuracy can deter- mine a considerable loss in the significance of venous flow assessment. Nowadays, the only available methodology to overcome this issue is to assess the TAV along the L plane, while measuring the CSA along the transverse plane (T).1-4,6,10 Of course this method includes a high risk of reliability loss since the TAV and CSA assessment sites could not remain the same when moving from L to T scanning plane. Aim of the present study is to evaluate different ways of CSA measurement and to identify a consistent methodology to assess the flow rate, so as to avoid the signifi- cance loss risk when moving from L to T scan- ning. Materials and Methods The internal jugular vein geometry model We designed a geometry model of the inter- nal jugular vein (IJV), so as to identify anatom- ical landmarks aimed to find corresponding sites between T and L evaluations. IJV anatomic feature can be schematized by considering the ellipse minor (b) and major (a) axis for its shape and the distance from the skin for its location. In L ultrasound images the IJV walls are both visible. We named Lh the distance between the vessel inner linings and Ld the distance between the skin and the superior wall inner lining (Figure 1). In T images, the whole CSA contour is easily traceable. If we consider the ultrasound path crossing the vein center, Td is the distance from the skin to the vein wall and Th is the distance between the vessel walls (Figure 2). In this way Ld, Lh, Td and Th constitute the anatomical landmarks of the investigated vessel. Patient population Four consecutive healthy volunteers (age ranging from 23 to 25 years old, male:female ratio 1:1), were enrolled in this study. All the study participants underwent CD investigation (ESAOTE My-Lab 70, Genoa, Italy) with the same condition of room temperature (23°C). Measurements were all performed in the morning hours following the recommendation to drink 500 mL after waking up, for compara- ble hydration conditions.11 Color Doppler study of the internal jugular veins First subject The first subject underwent a recorded IJV Correspondence: Francesco Sisini, Department of Physics and Earth Sciences, University of Ferrara, via Saragat 1, 44122 Ferrara, Italy. E-mail: ssf@unife.it Key words: echo-color-Doppler, ultrasound, flow rate, internal jugular vein. Contributions: FS developed the geometrical model and the equations, collected and analysed the raw data, performed the statistical analysis and wrote the paper. EM performed the color- Doppler scannings and collected the data. SG, GD and AT wrote the paper and revised it critically. AMM, MT contributed to collect and analyse the data. MG provided scientific supervision and founded the study. All authors participated in the design study, read and approved the final manu- script. Conflict of interests: the authors declare no potential conflict of interests. Funding: this study was partially supported by the Italian Ministry of Education, University and Research (MIUR Programme PRIN 2010-2011), Grant no. 2010XE5L2R. Received for publication: 5 August 2013. Revision received: 22 November 2013 Accepted for publication: 23 Dcember 2013. This work is licensed under a Creative Commons Attribution 3.0 License (by-nc 3.0). ©Copyright F. Sisini et al., 2014 Licensee PAGEPress, Italy Veins and Lymphatics 2014; 3:1863 doi:10.4081/vl.2014.1863 No n- co mm er cia l u se on ly Article [page 2] [Veins and Lymphatics 2014; 3:1863] transverse scanning from J1 to J3:11 the assessment was performed at a constant veloc- ity, for a total length of 12 cm, and lasted 5 sec- onds. The total number of analyzed frames was 38. The analysis of each frame image included a, b, Th, Td and CSA measurements. The CSA was measured by manually tracking the bound- aries of the lumen. In the following we call this measure contoured CSA (conCSA). For relia- bility purposes, the operator repeated the same scanning 5 times. The reliability of the con- stant velocity assumption was previously test- ed in the laboratory by asking the operator to scan a phantom having a thickness linearly related to the displacement (PMMA step wedge). After the operator training, the chi square test showed a P>0.9 indicating a uni- form motion of the probe. By dividing the 12 cm total length (L) by the number (N) of acquired images, we calculated the sampling distance and we named this value as unit length (UL). Then we multiplied the image number (In) by UL to obtain the exact observa- tion site along the IJV (y). y = In ¥ UL (1) Besides, we obtained the manually traced conCSA, the vessel circular CSA (cirCSA) and the elliptical CSA (ellCSA) by using the meas- ured Th, a and b values. The circular area can be calculated using the radius r=b/2 as follows: cirCSA = π �× r2 (2) While the elliptical one can be calculated using maximum (R=a/2) and minimum (r) radii: ellCSA = π �× r×� R (3) We measured the mean conCSA from all the acquired frames. The Pearson’s r coefficient was calculated to determine the linear correla- tion between the conCSA and the Td and Th parameters. Three control subjects The subjects were placed in supine position and underwent IJV transverse scanning at level of J1, J2 and J3. Each assessment (total of 9) was performed by keeping the probe in a fixed position among J1, J2 and J3. The acquired image sequences were processed by considering a number of frames within at least one cardiac cycle. For each analyzed frame we measured the major and minor axis (a and b), while the conCSA was obtained by manually tracking the boundaries of the lumen. Total of 370 data record (corresponding to about 40 records for each IJV segment) was collected, and the analyses were performed either sepa- rately for each subject or as a whole. Statistical analysis The statistical difference among the conCSA, the ellCSA and the cirCSA was tested by using appropriately homoscedastic and het- eroscedastic Student’s t test; P value<0.05 was considered significant. Homoscedasticity and heteroscedastic hypothesis of the variances has been tested by F Test. The Pearson’s r coef- ficient was calculated to measure the linear correlation between the conCSA and the Td and Th parameters. The reported uncertainties refer to one standard deviation. Table 1. P-value of T-test for elliptical and circular cross sectional area hypothesis. P-value Subject no. conCSA vs ellCSA conCSA vs cirCSA 1 0.95 <0.001 2 0.35 <0.001 3 0.09 <0.001 4 0.11 <0.001 conCSA, contoured cross sectional area; ellCSA, elliptical CSA; circCSA, circular CSA. Figure 1. Schematic representation of an ultrasound probe insonating an internal jugular vein (IJV) in the L plane. The IJV’s walls are represented. The distance Ld between the IJV and the skin and the wall distance Lh of the IJV are shown. Figure 2. Schematic representation of ultrasound probe insonat- ing an internal jugular vein (IJV) in the T plane. The IJV is rep- resented as an ellipse of major and minor axis a and b respective- ly. The distance Td between the IJV and the skin and the depth Th of the IJV are shown. No n- co mm er cia l u se on ly Article [Veins and Lymphatics 2014; 3:1863] [page 3] Results First subject The measured conCSA data range from a maximum of 0.82 cm2 in J1 to a minimum of 0.14 cm2 in J3 with a mean value of 0.36±0.17 cm2 (Figure 3). The mean cirCSA was 0.22±0.16 cm2, while the mean ellCSA was 0.36±0.16 cm2. In Figure 4 elliptical and circu- lar IJV CSA are plotted as a function of the con- toured CSA. It shows a significant lack of over- lapping between conCSA and cirCSA while there is a good agreement between ellCSA and conCSA. This also results in a significant sta- tistical difference between these quantities (see the P value in Table 1). The correlation coefficient r between Td and conCSA resulted 0.5 while it resulted 0.95 between Th and conCSA. Three control subjects In Figure 5, elliptical and circular IJV CSA are plotted as function of the contoured CSA. Again, the plot shows a significant lack of over- lapping between conCSA and cirCSA while there is a good agreement between ellCSA and conCSA. P values are reported separately for each subject in Table 1. This finding is very interesting because it shows that the elliptical shape assumption remains valid also in the vein pulsation phase, and thus suggesting that the use of the Th and Lh landmarks is reliable. Discussion The main outcome of a cirCSA lower than the conCSA, in the observed T scanning, demonstrates the risk of a flow rate underesti- mation, under the assumption of a cylindrical venous shape (Figures 4 and 5). Moreover, the significant conCSA variability along the vein tract points out the consistency loss when the US probe is positioned in a different anatomi- cal site between T and L scannings. The pres- ent study offers an assessment methodology, endowed with anatomical landmarks (Ld, Lh, Td and Th) that can become extremely useful for a correct sample volume positioning, in particular thanks to the high Th correlation coefficient. Conversely, the lack of correlation between Td and conCSA indicates that this parameter is unsuitable for our purpose. It is worth noting that although this method helps to identify the same anatomical site in both L and T planes, the actual CSA of the vein depends on the probe compression applied by the operator. Anyway, this does not affect the flow calculation because also the blood veloci- ty is affected by the CSA change hence, if the velocity and the CSA are measured by using the same pressure, the resulting calculated flow is virtually invariant (high compression can alter the haemodynamics and the flow). Since there are no warranties that the operator applies the same pressure on both the T and L plane, there is no certainty that the CSA assumes the same value when measured in the T and L plane hence, there is the risk to use the landmarks in a misleading way. For this reason we are investigating the use of a correction factor that allows to calculate the elliptical CSA based on the Lh parameter that has to be measured on the same plane where the TAV is also measured. The true reliability of the novel measuring method has not been assessed in this paper since no direct flow measurement is done. Actual flow measure- ment is not possible on human subjects with Figure 3. The measured cross sectional area is plotted as function of the position along the internal jugular vein (IJV) from J1 to J3. Figure 4. Elliptical and circular internal jugular vein (IJV) cross sectional area (CSA) of one subject are plotted as function of the measured CSA. Different measured CSA corre- spond to different assessment along the IJV length. No n- co mm er cia l u se on ly Article [page 4] [Veins and Lymphatics 2014; 3:1863] non-invasive methods. For this reason we are planning to verify this methodology by labora- tory in vitro investigation. Finally, if a quanti- tative flow calculation of the IJV is requested for a clinical reason, we believe that its calcu- lation based on the vein diameter acquired in the L plane is not correct. Conclusions The flow value assessment is often based on the assumption of constant CSA, which can generate a possible significant inaccuracy. We have demonstrated that, thanks to vali- dated longitudinal and transversal landmarks, the TAV can be measured in the L plane in a point where the Lh is equal to Th. References 1. Valdueza JM, von Münster T, Hoffman O, et al. Postural dependency of the cerebral venous outflow. Lancet 2000;355:200-1. 2. Schreiber SJ, Lurtzing F, Gotze R, et al. Extrajugular pathways of human cerebral venous blood drainage assessed by duplex ultrasound. J Appl Physiol 2003;94:1802-5. 3. Gisolf J, van Lieshout JJ, van Heusden K, et al. Human cerebral venous outflow pathway depends on posture and central venous pressure. J Physiol 2004;560:317-27. 4. Doepp F, Schreiber SJ, von Münster T, et al. How does the blood leave the brain? A sys- tematic ultrasound analysis of cerebral venous drainage patterns. Neuroradiology 2004;46:565-70. 5. Zamboni P, Menegatti E, Pomidori L, et al. Does thoracic pump influence the cerebral venous return? J Appl Physiol 2012;112: 904-10. 6. Yamaki T, Nozaki M, Fujiwara O, Yoshida E. Comparative evaluation of duplex-derived parameters in patients with chronic venous insufficiency: correlation with clinical man- ifestations. J Am Coll Surg 2002;195:822-30. 7. Griffin M, Nicolaides AN, Bond D, Geroulakos G, Kalodiki E. The efficacy of a new stimulation technology to increase venous flow and prevent venous stasis. Eur J Vasc Endovasc Surg 2010;40:766-71. 8. Fung YC. Biomechanics circulation. 2nd ed. Berlin: Springer; 1997. 9. Attinger EO. Wall properties of veins. Biomedical engineering. IEEE Trans 1969; 16:253-61. 10. Zamboni P, Sisini F, Menegatti E, et al. An ultrasound model to calculate the brain blood outflow through collateral vessels: a pilot study. BMC Neurol 2013;13:81. 11. Zamboni P, Morovic S, Menegatti E, et al. The Intersociety Faculty: screening for chronic cerebrospinal venous insufficiency (CCSVI) using ultrasound. Recommen - dation for a protocol. Int Angiol 2011; 30:571-97. Figure 5. Manually contoured, elliptical and circular internal jugular vein (IJV) cross sec- tional area (CSA) of three subject are plotted as function of the contoured CSA. Different measured CSA correspond to different assessment along the IJV length and to a different assessment during the cardiac cycle. No n- co mm er cia l u se on ly