This Article Abstract Full Text (PDF) Data Supplement Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (4) Citing Articles via Google Scholar Google Scholar Articles by Riopelle, J. M. Articles by Miller, M. K. Search for Related Content PubMed PubMed Citation Articles by Riopelle, J. M. Articles by Miller, M. K.

---

CRITICAL CARE AND TRAUMA

Circumferential Adjustment of Ultrasound Probe Position to Determine the Optimal Approach to the Internal Jugular Vein: A Noninvasive Geometric Study in Adults

James M. Riopelle, MD^*, Darren P. Ruiz, MD^*, John P. Hunt, MD, Mark R. Mitchell, MD^*, J. Carlos Mena, MD^¶, Jason A. Rigol, MD^*, Bruno C. Jubelin, PhD, Arthur J. Riopelle, PhD, Valeriy V. Kozmenko, MD^*, and Matthew K. Miller, MD^*

Departments of *Anesthesiology, Surgery, ¶Radiology, and Medicine, Louisiana State University Health Sciences Center at New Orleans, New Orleans, Louisiana; and Department of Psychology, Louisiana State University, Baton Rouge, Louisiana

Address correspondence and reprint requests to James Riopelle, MD, Anesthesia Department, Charity Hospital, 1532 Tulane Ave., New Orleans, LA 70112. Address e-mail to jriope{at}lsuhsc.edu.

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Circumferential adjustment of the position of a two-dimensional ultrasound (US) probe around the neck has been recommended as a strategy for reducing the potential for unintentional common carotid artery puncture during internal jugular venous (IJV) cannulation. We obtained multiple high-resolution US images bilaterally from the necks of 107 adult subjects and analyzed these to determine the degree to which this strategy permits identification of a pathway from the skin to the IJV that minimizes venoarterial overlap while maximizing venous target (angular) width. The method consistently permitted identification of an approach to the IJV superior to that obtainable with any one of four popular surface anatomy-based ("blind") approaches and was even more powerful if used in concert with a US-guided 1) adjustment of the degree of head rotation, 2) choice between a high and low approach, and 3) choice between the right and left IJV. Use of a high-resolution US imaging device also permitted identification of the precise boundaries of additional cervical anatomic structures (nontarget vessels, lymph nodes, and the thyroid gland) potentially relevant to selection of an optimal approach to the IJV.

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Four recent reviews of the pertinent medical literature concluded that the use of real-time two-dimensional ultrasound (US) imaging during internal jugular vein (IJV) cannulation improves the procedural success rate, reduces the need for multiple needle advances, and decreases the rate of inadvertent injury to the common carotid artery (CCA) (1–4) (and, presumably, the attendant risks of hematoma and stroke) (5,6). Our own early (disappointing) experience using US imaging to prevent CCA puncture during US-guided IJV cannulation clearly demonstrated, however, that obtaining full benefit from US imaging entails a learning process that is partly technical (7) but also partly conceptual; i.e., one must develop an understanding of how to make optimal use of the individualized anatomic information being provided (8).

An important mechanism by which CCA puncture can occur despite US guidance of IJV cannulation arises from the coincidence of two factors (9–12): 1) anatomic variation or pathologic change placing some portion of the CCA on the line of needle advancement just beyond the IJV (venoarterial overlap) and 2) through-and-through puncture of the IJV (double-wall puncture or venous transfixation) due to the advancing needle tip traversing the entire IJV rather than entering its lumen. The first of these factors occurs in up to 78% of patients (10) and depends on such variables as patient age (9,13,14), the degree of contralateral head rotation (15), the indication for central venous cannulation (16), and the operator's approach to the IJV (9,10,13). The incidence of the second factor may depend on needle gauge (11,17), needle bevel,¹ speed of needle insertion (10,17), and the operator's level of training or experience (7,8).

A solution to the problem of transvenous puncture of the CCA suggested by some pioneering developers of the technique of US-guided central venous cannulation is circumferential displacement of the site of needle insertion away from a traditional (i.e., surface anatomy-based) location to minimize vascular overlap (9,10,17,18). Because the usefulness of this strategy has not been evaluated quantitatively, we performed this noninvasive anatomic study to measure the degree to which circumferential adjustment of US probe position at the cricoid level permits identification of a cutaneous needle insertion point and a line of needle advancement to the IJV that geometrically avoids CCA injury (by minimizing venoarterial overlap) while favoring internal jugular venipuncture (by maximizing venous target angular width). For purposes of comparison, we also used the same geometric method to study three previously described alternative/incremental strategies for improving the success and safety of IJV cannulation: 1) elimination of head rotation (15), 2) selection of a lower cannulation approach (6,19), and 3) cannulation of the larger or otherwise anatomically preferable (right versus left) IJV (20).

Finally, to increase the clinical relevance of US measurements obtained, we analyzed a simple geometric model of IJV and CCA anatomy to derive the coefficients needed to correct US measurements of vessel depth and angular width for the angle at which a needle approaches the IJV.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
This study was approved by the Louisiana State University Health Sciences IRB and the Medical Center of Louisiana-New Orleans Research Review Committee. One-hundred-fifteen subjects were originally entered: 101 elective and emergency surgical patients, 9 surgical intensive care (SICU) patients (2 receiving mechanical ventilation that included 5 cm H₂O positive end-expiratory pressure (PEEP)), and 5 healthy staff anesthesiologists. Data from 1 elective surgical patient with a thrombosed IJV and from 7 elective surgical patients from whom one or more US images were of insufficient quality to permit performance of all required measurements were excluded from analysis (107 subjects remained; Table 1). In all instances, informed consent was obtained from the subject or a legal representative and, in the case of SICU subjects, from the patient's surgeon.

Subjects were placed in a 15° Trendelenburg position. (A Valsalva maneuver was substituted in four SICU patients at the surgeon's request.) The US imaging device used was a SonoSite (Bothell, WA) Model 180 fitted with a Model L38/10-5 multifrequency vascular probe. Transverse (cross-sectional) US images were recorded bilaterally with the mid probe positioned as follows: 1) at each of three cricoid-level locations chosen (on the basis of external anatomic landmarks) to include cutaneous needle insertion points traditionally associated with the high-anterior, high-middle, and high-posterior approaches to the IJV (6) and 2) at one location at the base of the neck, just cephalad to the sternoclavicular junction (the probe center consistently overlay the mid sternocleidomastoid muscle or was positioned between its sternal and clavicular heads), corresponding to a low-middle approach to the IJV (6,21) (Fig. 1). Images taken from the high-anterior and high-middle probe positions were obtained with subjects' heads directed straight forward and also rotated 45° away from the side of imaging.²

View larger version (48K): [in this window] [in a new window]   Figure 1. Cervical ultrasound anatomy relevant to internal jugular venipuncture. A, Operator's view of the four right-sided ultrasound probe positions; inset shows a magnified view of the region of the sternocleidomastoid triangle and the locations of the six right-sided cutaneous vantage points (Vs). B, Composite diagram of overlapping transverse ultrasonic images obtained from the three right-sided cricoid-level ultrasound probe positions of one subject (head rotated 45° to the left). Key to color coding of rectangles indicating standard ultrasound probe positionings: blue = high anterior; yellow = high middle; orange = high posterior; gray = low middle. Meanings of V subscripts: R = right side of neck; c = cricoid level; b = base of neck; 45 = head rotated 45° away from the side of measurement; aa = medial to SCMM; a = overlying anterior (medial) portion of SCMM; m = overlying the mid portion of the SCMM (transverse plane) or between its sternal and clavicular heads; p = overlying the posterior (lateral) portion of the SCMM; pp = posterior to the SCMM. Anatomic abbreviations: Trach = trachea; StHM = sternohyoid muscle; Thy = thyroid gland; StTM = sternothyroid muscle; OmHM = omohyoid muscle; SCMM = sternocleidomastoid muscle; CCA = common carotid artery; IJV = internal jugular vein; LN = lymph node; ant = anterior; post = posterior; rt = right. IJV = IJV angular width; CCA = CCA angular width; IJV CCA = IJV-CCA angular overlap; * IJV = IJV clear target angle/subangle (un-overlapped portion of IJV); PIJV = IJV depth; DIJV = IJV in-line diameter; TSCMM = in-line thickness of SCMM.

US images were analyzed with MagicView 300^TM Dicomm imaging software. Measurements (defined in Table 2 and illustrated in Fig. 1) were obtained bilaterally from five cutaneous cricoid-level US vantage points (Vs) and from an additional V at the base of the neck. In approximately 10% of images, a portion of the IJV cross-sectional image lay outside the image border, so establishing a value for IJV angular width (_IJV) required combining information from images obtained from adjacent Vs.

Student's paired t-test was used to test differences between two matching data sets containing measurements of a continuous variable. Analysis of variance was used to test differences among the three types of subjects on continuous demographic variables. The relationship between subject sex and intensive care unit status was tested with the ² test. Correlations were computed with Pearson's coefficient. Statistical significance was set at P < 0.05. All comparisons were two tailed and were performed separately for right- and left-sided measurements except when determining the incremental geometric advantage of choosing between the right and left IJV.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Because there were no statistically significant differences between subject subgroups on any demographic variable and because there would appear to be no reason why the geometric principles under investigation should not be universally applicable, all subjects were considered as members of a single group. US images obtained during this study were taken in a transverse (cross-sectional) or nearly transverse plane. Because a cannulating needle is inserted toward the IJV from a cutaneous puncture site located cephalad to the plane of US imaging and is advanced at an acute angle to this plane, linear and angular measurements obtained from the US screen (such as those displayed in Fig. 2) must be trigonometrically corrected to accurately predict the distance a needle must be inserted to contact the wall of the target vein and the angle within which such contact will occur. As can be seen from Table 3, the smaller the angle of needle approach to the IJV, the greater the degree of correction required in both vessel depth and vessel angular width. (These coefficients are derived in the supplemental data available at www.anesthesia-analgesia.org.)

View larger version (50K): [in this window] [in a new window]   Figure 2. Unified summary of study findings. Left, Group mean values of geometric variables arrayed by cutaneous ultrasound (US) vantage points (Vs). Right, Percentages of subjects demonstrating the optimum value of a geometric variable at each of the Vs used (separate rankings for right and left sides). The line graph superimposed on (2B') indicates the percentage of subjects from each V demonstrating a nonvascular space between the carotid artery and internal jugular vein of at least 5°. The line graph superimposed on (2D') indicates the percentage of subjects from each V in whom the common carotid artery (CCA) lay closer to V than did the internal jugular vein (IJV): i.e., (PCCA – PIJV) < 0. Meanings of V subscripts: aa = medial to sternocleidomastoid muscle (SCMM); a = overlying anterior (medial) portion of SCMM; m = overlying the mid portion of the SCMM (transverse plane) or between its sternal and clavicular heads; p = overlying the posterior (lateral) portion of the SCMM; pp = posterior to the SCMM. Bars within individual graphs have been vertically aligned so that the cutaneous ultrasound V designations at the bottom of the left- and right-sided graph columns apply to bars in all graphs directly above them. P values are for paired Student's t-tests comparing data from two contiguous vertical bars, i.e., either 1) head straight versus rotated 45° or 2) cricoid level versus the base of the neck, both during 45° head rotation. The total height of each bar in (2A) expresses the total internal jugular vein angular width as the sum of its overlapped and nonoverlapped portions (i.e., IJV = IJVCCA + IJVCCA). IJV = IJV angular width; IJVCCA = IJV-CCA angular overlap; *IJV = IJV clear target angle/subangle; PIJV = depth of IJV; PCCA-IJV = differential depth.

Graphs on the left side of Figure 2 display group mean values of measured and calculated geometric variables from each of the several Vs used in the study. Graphs on the right side of Figure 2 indicate the percentages of subjects who achieved the optimal value of selected geometric variables from each V (separate rankings for right and left sides).

Differences between group mean values of IJV-CCA angular overlap (_IJVCCA) from adjacent cutaneous US Vs at the level of the cricoid cartilage were almost invariably statistically significant (P < 0.001 for each paired Student's t-test except for two comparisons involving more medial closely spaced Vs). Such variation in vascular arterial overlap with circumferential probe movement occurred primarily as a result of the phenomenon of parallax.³

Eliminating head rotation significantly reduced group mean _IJVCCA from only a single V (Fig. 2A, bottom portions of bars). Moving the US probe from the high-middle position to the low-middle position (_c45V_{m b45}V_m) significantly reduced (improved) the group mean _IJVCCA bilaterally.

Differences between group mean values of _IJV obtained from adjacent cricoid-level cutaneous US Vs were invariably statistically significant (P < 0.001 for each paired Student's t-test). This variation arose from the combined influence of two factors: 1) variation in distance between the V and the IJV (the inverse correlation between _IJV and the geometrically predicted value of this variable based on vessel depth was 0.78) and 2) eccentricity or irregularity of shape of the IJV cross-sectional outline.

Elimination of head rotation significantly increased the group mean _IJV from only a single V (Fig. 2A), although this maneuver increased (improved) the value of _IJV in individual subjects by as much as 40°. Moving the US probe from the cricoid level to the base of the neck (_c45V_{m b45}V_m) did not significantly change the group mean _IJV on the right side of the neck and significantly reduced (worsened) the group mean _IJV on the left (Fig. 2A). However, the maximum ipsilateral value of _IJV in an individual subject was often located at the base of the neck (Fig. 2A'), and the value of this measurement increased by as much as 61° on the right and 36° on the left.

Upper portions of the vertical bars of Figure 2A illustrate nonoverlapped portions of the _IJV. Because this "safe" (22) portion of total _IJV was often divided into two component subangles—especially when the IJV was viewed from a V associated with the high-posterior approach (_c45V_p or _c45V_pp; Fig. 1)—a separate graph was created to display the nonoverlapped portion of _IJV available for puncture during a single needle advancement (IJV clear target angle/subangle, *_IJV; Fig. 2B). Differences between group mean values of *_IJV from adjacent cutaneous US Vs at the level of the cricoid cartilage were invariably statistically significant (P for each paired Student's t-test was <0.001).

Head rotation exerted no statistically significant effect on group mean *_IJV from any V. Moving the US probe to the neck base (_c45V_{m b45}V_m) did not significantly change the group mean *_IJV on the left side but produced a statistically significant (Fig. 2B) and probably clinically relevant (Fig. 2B') increase in *_IJV on the right side. This effect originated predominantly from a reduction in venoarterial overlap (_IJVCCA).

Although this study was focused primarily on IJV-CCA overlap and _IJV, skin to vessel distances (depths) were also measured, and they demonstrated a regular pattern of variation with circumferential probe movement (Fig. 2, C and D), two features of which are likely to be of clinical importance. First, the high degree of venoarterial overlap present when the IJV-CCA anatomic relationship is viewed from the high-posterior position occurs in association with a maximization of CCA-IJV differential depth (Fig. 1, views from _c45V_p and _c45V_pp; Fig. 2, A and D).⁴ Second, the vascular overlap present when the IJV-CCA anatomic relationship is viewed from the high-anterior position is often of the (undesirable) type wherein some portion of the artery lies between the skin and the IJV (arteriovenous overlap; cf. IJV views from _c45V_a and _c45V_aa in Fig. 1 and line graph of Fig. 2D').

During the scoring of US images, it was often noted that head rotation could—in an individual subject and from a particular V—either improve or worsen the geometry of the IJV-CCA anatomic relationship with respect to IJV cannulation. To determine the degree to which an operator might be able to take advantage of interpersonal anatomic variation to individually select the degree of head rotation, we (retrospectively) compared subjects' values on three geometric variables (_IJV, _IJVCCA, and *_IJV) obtained under three separate conditions: 1) head facing directly forward, 2) head rotated 45°, and 3) nonrotation or rotation depending upon which of these two positions was shown by US imaging to provide the better score. US-guided selection of either 0° or 45° of head rotation significantly improved the group mean value for all 3 geometric variables from every V from which both views were recorded (P for each paired Student's t-test was <0.001). A statistically significant increment in the group mean value of each of the same three geometric variables was also obtainable when US imaging was used to select between 1) a high versus low approach or 2) the right versus left IJV (P for each paired Student's t-test was <0.001).

Anatomic separation of the IJV and CCA (_IJVCCA) of at least 5°—presumably a favorable circumstance in the setting of IJV cannulation—occurred more often at the base of the neck than at cricoid level and more often on the right side than on the left (line graph of Fig. 2B'). The degree of separation observed was as great as 39° on the right side and 21° on the left.

US imaging also revealed several types of variant or pathologic anatomy of potential importance during IJV cannulation: 1) nontarget veins of diverse size and orientation, often emptying into the medial side of the IJV; 2) a more-cephalad-than-expected position of the subclavian artery; 3) IJV and CCA thrombosis; 4) lymph nodes (which appeared similar to veins in cross-section but neither were compressible nor formed confluences with true blood vessels); and 5) goiter.

Finally, although the color Doppler function on the US imaging device used did not indicate the direction of blood flow, its display of flow speed and flow pattern (pulsatility and respiratory variation) was occasionally helpful in confirming vessel type (artery versus vein) and patency.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Data obtained during this study provide strong geometric confirmation of the value of circumferential cervical adjustment of US probe position to identify a high approach to IJV cannulation that minimizes venoarterial overlap (parallax effect), maximizes target venous angular width (proximity effect), or optimizes a combination of the two. Elimination of head rotation generally did little to improve the view of the IJV-CCA anatomic relationship, although individualized adjustment of head position was found to be a useful incremental strategy. Changing from a high-middle to a low-middle approach to the IJV decreased venoarterial overlap bilaterally and increased _IJV on the right side. Any geometric benefit gained from selection of a low approach to IJV cannulation, however, must be weighed against its potentially greater risk of pneumothorax (6) or needle injury to the subclavian artery.

Collecting and analyzing geometric data necessitate the use of precise mathematical definitions and careful performance of linear and angular measurements. Fortunately, the method of using the principles of parallax and proximity to optimize the view of the IJV-CCA anatomic relationship before IJV cannulation is no more complex than the method used by a wedding photographer to get a good picture of the faces of both newlyweds. At the chosen axial level, the operator 1) transversely images the IJV-CCA relationship, 2) identifies the point along the probe-skin interface from which he or she projects there will be no or minimal venoarterial overlap (preferably close to the IJV to maximize venous angular width), and then 3) moves the US probe circumferentially around the neck so that its center lies over this point, meanwhile keeping the IJV cross-sectional image in the center of the US screen. The geometrically optimized line of needle advancement will then cross the viewing screen's vertical midline, and the probe's longitudinal axis can serve as a needle-aiming guide.⁵

Should a satisfactory approach to the IJV not be identifiable with this method, the operator can adjust the degree of patient head rotation or expand the search for a safe path from the skin to the IJV proximal and distal to the original axial plane of US imaging. With experience, one learns to mentally compile anatomic information obtained from multiple axial levels, generate a three-dimensional model of relevant cervical anatomic structures, and then analyze this model to identify a line of needle advancement to the IJV that will miss nontarget vessels, lymph nodes, and the thyroid gland and/or minimize needle (and subsequent catheter) passage through superficial cervical musculature. Should the IJV initially selected for cannulation be found to be thrombosed or of very small diameter, US imaging can be used to assess the contralateral IJV (and also the subclavian and femoral veins).

Provided an acceptable line of needle advancement is found to either the right or the left IJV, the operator can augment a patient's intrathoracic pressure to enhance IJV distention and resistance to compression by the advancing needle tip (Valsalva maneuver, sigh breath, or PEEP) and then advance the (finder or cannulating) needle toward the target vessel. He or she can then select the angle of needle approach to the IJV after considering 1) the proximity of the cutaneous puncture site to the subclavian artery and/or lung, 2) the influence of this angle on skin-to-vessel distance and angular venous target size, and 3) any perceived advantage from having the cannulating needle (and, later, the central venous cannula or introducer) enter the IJV at an acute (rather than right) angle.

Unfortunately, this method of using US imaging to facilitate IJV cannulation cannot, by itself, eliminate all possibility of needle injury to the CCA or guarantee successful IJV cannulation. At least the following technical and conceptual pitfalls of US-guided vascular cannulation must also be avoided: 1) mistaking the CCA for the IJV and so targeting the wrong blood vessel, 2) failing to properly aim the needle along a well chosen line of needle advancement, 3) falsely assuming that the anatomic relationship between the IJV and CCA at the level of venipuncture is identical to that at the axial level of (transverse) US imaging,⁶ 4) confusing the needle shaft for its tip on the US screen (23), and 5) paying insufficient attention to the US device screen (relative to the puncture site) during needle advancement.

In summary, this noninvasive US study of cervical anatomy confirms a geometric advantage of circumferential cervical adjustment of the position of a two-dimensional US imaging probe for identifying an optimal cutaneous puncture site and line of needle advancement to the IJV (especially in a patient with unsuspected cervical variant anatomy or pathology). This circumferential searching strategy appears to be even more powerful, moreover, if used in concert with US-guided selection of the side of cannulation, choice between a high and low approach, and individualized adjustment of the degree of head rotation. Determination of this method's true value during US-guided IJV cannulation, however, will require performance of a controlled clinical study.

The authors appreciate the loan of the US imaging device used in this study by Mike Vega of Sonosite, Inc., and Sonosite technical support from Jason Hysom, Scott Temonia, and Lee Word. Acquiring and processing of US images were facilitated by Diane Benavids, Rebeca Mercer, RT, Irma Cordova, LRT, Lisa Lemen, PhD, Susheela Viswanathan, MD, Dawn Galliano, MD, Jane Clayton, MD, JoAnn Tierney, RT, Debbi Sibley, Amanda Walker, Joyce Adams, and Rose Jones. Karen Grady and Eugene New (Louisiana State University Health Sciences Center Department of Learning Resources) created the figures. Jane Eyrich, MD, and Jean LaCour, MD, helped with manuscript preparation. Andrew Jones (Department of Mathematics, Kings College, London, UK) checked the trigonometry in the Appendix. W. J. Hansche, PhD (Tulane University emeritus professor of psychology and statistics), advised regarding the performance and interpretation of statistical correlation. The authors also wish to express their appreciation to the journal reviewers and editors whose many helpful suggestions greatly improved this article's readability.

	Footnotes

 
¹ Ukranian anesthesiologists sometimes substitute a sharper-beveled needle to cannulate the IJV than is customarily included in a standard central venous cannulation kit (V. V. Kozmenko, MD, personal communication, 2004).

² The influence of head rotation was not investigated 1) from the high-posterior position because this approach requires some degree of head rotation or 2) at the base of the neck because of the minimal effect of rotation on internal cervical anatomic relationships at this axial level.

³ Parallax: "the apparent displacement or the difference in apparent direction of an object as seen from two different points not on a straight line with the object" (Merriam-Webster Online Dictionary; http://www.m-w.com).

⁴ This anatomic fact could at least partially explain the established clinical safety (6) of this approach to the IJV.

⁵ The needle can be advanced through a physical guide attached to the US probe to ensure a consistent line of needle advancement with respect to the US probe central axis (7), but at the cost of some loss in the ability to make fine adjustments in the line of needle advancement and to adjust the angle of the needle approach to the IJV.

⁶ That is, a geometrical model (such as that in the supplemental data available at www.anesthesia-analgesia.org.) describing the neck as a cylinder containing smaller, parallel IJV and CCA cylinders is only approximately true.

Supplemental data available at www.anesthesia-analgesia.org.

The investigators performed this study during the course of their clinical duties supplemented with personal time. There were no financial incentives for participation from any source. During the 6-month period of data collection, SonoSite Inc. (Bothell, WA) provided the ultrasound imaging device, multifrequency solid-state probe, and technical support. The Medical Center of Louisiana-New Orleans Radiology Department provided unrestricted access to their ultrasonic imaging resources.

Accepted for publication July 28, 2004.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Randolph AG, Cook DJ, Gonzales CA, Pribble CG. Ultrasound guidance for placement of central venous catheters: a meta-analysis of the literature. Crit Care Med 1996;24:2053–8.[Web of Science][Medline]
2. Keenan SP. Use of ultrasound to place central lines. J Crit Care 2002;17:126–37.[Web of Science][Medline]
3. Rothschild JM. Ultrasound guidance of central vein catheterization. Evidence report/technology assessment No. 43. Making health care safer. A critical analysis of patient safety practices. Agency for Healthcare Research and Quality Publication No. 01-E508. 2001:245–53. Available at: http://www.ahrq.gov/clinic/ptsafety/chap21.htm.
4. Hind D, Calvert N, McWilliams R, et al. Ultrasonic locating devices for central venous cannulation: meta-analysis. BMJ 2003;327:361–7.[Abstract/Free Full Text]
5. Reuber M, Dunkley LA, Turton EPL, et al. Stroke after internal jugular venous cannulation. Acta Neurol Scand 2002;105:235–9.[Web of Science][Medline]
6. Latto IP. The internal jugular vein. In: Latto IP, Ng WS, Jones PL, Jenkins B, eds. Percutaneous central venous and arterial catheterization. 3rd ed. London: WB Saunders, 2000:135–95.
7. Augoustides JG, Diaz D, Weiner J, et al. Current practice of internal jugular venous cannulation in a university anesthesia department: influence of operator experience on success of cannulation and arterial injury. J Cardiothorac Vasc Anesth 2002;16:567–71.[Web of Science][Medline]
8. Jobes DR. The element of experience in internal jugular vein catheterization. Anesth Analg 1992;75:643.
9. Troianos CA, Kuwik RJ, Pasqual JR, et al. Internal jugular vein and carotid artery anatomic relation as determined by ultrasonography. Anesthesiology 1996;85:43–8.[Web of Science][Medline]
10. Gordon AC, Saliken JC, Johns D, et al. US-guided puncture of the internal jugular vein: complications and anatomic considerations. J Vasc Interv Radiol 1998;9:333–8.[Web of Science][Medline]
11. Koski EMJ, Suhonen M, Mattila MA. Ultrasound-facilitated central venous cannulation. Crit Care Med 1992;20:424–6.[Web of Science][Medline]
12. Mangar D, Turnage WS, Mohamed SA. Is the internal jugular vein cannulated during insertion or withdrawal of the needle during central venous cannulation? Anesth Analg 1993;76:1375.
13. Mallinson C, Bennett J, Hodgson P, Petros AJ. Position of the internal jugular vein in children: a study of the anatomy using ultrasonography. Paediatr Anaesth 1999;9:111–4.[Web of Science][Medline]
14. Alderson PJ, Burrows FA, Stemp LI, Holtby HM. Use of ultrasound to evaluate internal jugular vein anatomy and to facilitate central venous cannulation in paediatric patients. Br J Anaesth 1993;70:145–8.[Abstract/Free Full Text]
15. Sulek CA, Gravenstein N, Blackshear RH, Weiss L. Head rotation during internal jugular vein cannulation and the risk of carotid artery puncture. Anesth Analg 1996;82:125–8.[Abstract]
16. Lin BS, Kong CW, Tarng DC, et al. Anatomical variation of the internal jugular vein and its impact on temporary haemodialysis vascular access: an ultrasonographic survey in uraemic patients. Nephrol Dial Transplant 1998;13:134–8.[Abstract/Free Full Text]
17. Mauro MA, Jaques PF. Radiologic placement of long-term central venous catheters: a review. J Vasc Interv Radiol 1993;4:127–37.[Medline]
18. Denys BG, Uretsky BF, Reddy PS. Ultrasound-assisted cannulation of the internal jugular vein: a prospective comparison to the external landmark-guided technique. Circulation 1993;87:1557–62.[Abstract/Free Full Text]
19. Silberzweig JE, Mitty HA. Central venous access: low internal jugular vein approach using imaging guidance. AJR Am J Roentgenol 1998;170:1617–20.[Abstract/Free Full Text]
20. Lichtenstein D, Saifi R, Augarde R, et al. The internal jugular veins are asymmetric: usefulness of ultrasound before catheterization. Crit Care Med 2001;27:301–5.
21. Oshima E, Ishizu K, Urabe N. A newly designed curved needle for percutaneous cannulation of the internal jugular vein. Anesthesiology 1993;78:792–3.[Web of Science][Medline]
22. Daniel M. Head rotation during internal jugular vein cannulation and risk of carotid artery puncture. Anesth Analg 1996;83:660.
23. Hull JE, Hunter CS, Luiken GA. The Groshong catheter: initial experience and early results of imaging-guided placement. Radiology 1992;185:803–7.[Abstract/Free Full Text]

This article has been cited by other articles:

				E. Saade Error in BMI Calculations Anesth. Analg., September 1, 2005; 101(3): 924 - 924. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Data Supplement Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Web of Science (4) Citing Articles via Google Scholar Google Scholar Articles by Riopelle, J. M. Articles by Miller, M. K. Search for Related Content PubMed PubMed Citation Articles by Riopelle, J. M. Articles by Miller, M. K.