SPECIAL ARTICLE
Guidelines for Performing Ultrasound Guided VascularCannulation: Recommendations of the AmericanSociety of Echocardiography and the Society ofCardiovascular AnesthesiologistsChristopher A. Troianos, MD, Gregg S. Hartman, MD, Kathryn E. Glas, MD, MBA, FASE,Nikolaos J. Skubas, MD, FASE, Robert T. Eberhardt, MD, Jennifer D. Walker, MD,and Scott T. Reeves, MD, MBA, FASE, for the Councils on Intraoperative Echocardiographyand Vascular Ultrasound of the American Society of Echocardiography
TABLE OF CONTENTS PAGE
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .462. Methodology and Evidence Review . . . . . . .463. Ultrasound-Guided Vascular
Cannulation . . . . . . . . . . . . . . . . . . . . . . . . . . . .474. Ultrasound Principles for Needle-Guided
Catheter Placement . . . . . . . . . . . . . . . . . . . . .485. Real-Time Imaging Versus Static Imaging . . . . .506. Vessel Identification . . . . . . . . . . . . . . . . . . . .507. Internal Jugular Vein Cannulation . . . . . . . .52
7.1. Anatomic Considerations . . . . . . . . . . .527.2. Cannulation Technique . . . . . . . . . . . . .527.3. Complications . . . . . . . . . . . . . . . . . . . . .537.4. Recommendation for IJ Vein
Cannulation . . . . . . . . . . . . . . . . . . . . . . .548. Subclavian Vein Cannulation . . . . . . . . . . . .54
8.1. Anatomic Considerations . . . . . . . . . . .548.2. Cannulation Technique . . . . . . . . . . . . .548.3. Complications . . . . . . . . . . . . . . . . . . . . .568.4. Recommendation for SC Vein
Cannulation . . . . . . . . . . . . . . . . . . . . . . .569. Femoral Vein Cannulation . . . . . . . . . . . . . . .57
9.1. Anatomic Considerations . . . . . . . . . . .579.2. Cannulation Technique . . . . . . . . . . . . .579.3. Complications . . . . . . . . . . . . . . . . . . . . .579.4. Recommendation for FV
Cannulation . . . . . . . . . . . . . . . . . . . . . . .5710. Pediatric Ultrasound Guidance . . . . . . . . . . .58
10.1. Cannulation Technique forpediatric patients . . . . . . . . . . . . . . . . . . .5810.1.1. IJ Vein . . . . . . . . . . . . . . . . . . . . .5810.1.2. Femoral Vessels . . . . . . . . . . . . .59
10.2. Recommendations for PediatricPatients . . . . . . . . . . . . . . . . . . . . . . . . . . .60
11. Ultrasound-Guided ArterialCannulation . . . . . . . . . . . . . . . . . . . . . . . . . . . .6011.1. Cannulation Technique . . . . . . . . . . . . .6111.2. Ultrasound-Guided Arterial
Cannulation VersusPalpation . . . . . . . . . . . . . . . . . . . . . . . . .61
11.3. Recommendation for ArterialVascular Access . . . . . . . . . . . . . . . . . . .62
12. Ultrasound-Guided Peripheral VenousCannulation . . . . . . . . . . . . . . . . . . . . . . . . . . . .6212.1. Recommendation for Peripheral
Venous Access . . . . . . . . . . . . . . . . . . . .6213. Vessel Selection . . . . . . . . . . . . . . . . . . . . . . . .6214. Vascular Access Confirmation . . . . . . . . . . . .63
14.1. Recommendations for Vascular AccessConfirmation . . . . . . . . . . . . . . . . . . . . . .63
15. Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6316. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . .64
Notice and Disclaimer . . . . . . . . . . . . . . . . . . . . . . . . . . .64Appendix A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .64Appendix B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .65References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .70
1. INTRODUCTIONThe Agency for Healthcare Research and Quality, in its2001 report Making Health Care Safer: A Critical Analysis ofPatient Safety Practices, recommended the use of ultrasoundfor the placement of all central venous catheters as one ofits 11 practices aimed at improving patient care.1,2 Thepurpose of this document is to provide comprehensivepractice guidelines on the use of ultrasound for vascularcannulation. Recommendations are made for ultrasound-guided central venous access of the internal jugular (IJ)vein, subclavian (SC) vein, and femoral vein (FV) on thebasis of the strength of the scientific evidence present inthe literature (Table 1). The role of ultrasound forvascular cannulation of pediatric patients is discussedspecifically, and the use of ultrasound to facilitate arte-rial cannulation and peripheral venous access is alsodiscussed. Recommendations are made for training, in-cluding the role of simulation.
2. METHODOLOGY AND EVIDENCE REVIEWThe writing committee conducted a comprehensive search ofmedical and scientific literature in the English languagethrough the use of PubMed and MEDLINE. Original researchstudies relevant to ultrasound-guided vascular access pub-lished in peer-reviewed scientific journals from 1990 to 2011
From the Department of Anesthesiology, West Penn Allegheny HealthSystem, Pittsburgh, Pennsylvania (C.A.T.); the Department of Anesthesiol-ogy, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire(G.S.H.); the Department of Anesthesiology, Emory University, Atlanta,Georgia (K.E.G.); the Department of Anesthesiology, Weill-Cornell MedicalCollege, New York, New York (N.J.S.); the Department of Medicine, BostonUniversity School of Medicine, Boston, Massachusetts (R.T.E.); the Depart-ment of Cardiothoracic Surgery, Massachusetts General Hospital, Boston,Massachusetts (J.D.W.) and the Department of Anesthesiology, MedicalUniversity of South Carolina, Charleston, South Carolina (S.T.R.).
The authors reported no actual or potential conflicts of interest in relation tothis document.
Members of the Council on Intraoperative Echocardiography and theCouncil on Vascular Ultrasound of the American Society of Echocardiogra-phy are listed in Appendix A.
This article has been co-published in Anesthesia & Analgesia and the Journalof the American Society of Echocardiography.
Copyright © 2012 the American Society of EchocardiographyDOI: 10.1213/ANE.0b013e3182407cd8
46 www.anesthesia-analgesia.org January 2012 • Volume 114 • Number 1
were reviewed using the Medical Subject Headings terms“ultrasonography,” “catheterization-central venous/adverseeffects/methods,” “catheterization-peripheral,” “jugularveins,” “subclavian vein,” “femoral vein,” “artery,” “adult,”“pediatric,” “randomized controlled trials,” and “meta-analysis.” The committee reviewed the scientific evidence forthe strength of the recommendation (i.e., risk/benefit ratio) assupportive evidence (category A), suggestive evidence (cat-egory B), equivocal evidence (category C), or insufficientevidence (category D). The weight or “level” of evidence wasassigned within each category (Table 1). Recommendationsfor the use of ultra-sound were based on supportive literature(category A) with a level 1 weight of scientific evidence(multiple randomized controlled trials with the aggregatedfindings supported by meta-analysis). The document wasreviewed by 10 reviewers nominated by the American Societyof Echocardiography (ASE) and the Society of CardiovascularAnesthesiologists and approved for publication by the gov-erning bodies of these organizations.
3. ULTRASOUND-GUIDEDVASCULAR CANNULATIONUltrasonography was introduced into clinical practice inthe early 1970s and is currently used for a variety of clinicalindications. Miniaturization and advancements in com-puter technology have made ultrasound affordable, por-table, and capable of high-resolution imaging of both tissueand blood flow.
Cannulation of veins and arteries is an important aspectof patient care for the administration of fluids and medica-tions and for monitoring purposes. The practice of usingsurface anatomy and palpation to identify target vesselsbefore cannulation attempts (“landmark technique”) is
based on the presumed location of the vessel, the identifi-cation of surface or skin anatomic landmarks, and blindinsertion of the needle until blood is aspirated. Confirma-tion of successful cannulation of the intended vascularstructure relies on blood aspiration of a certain characterand color (i.e., the lack of pulsation and “dark” color whencannulating a vein or pulsation and a “bright” red colorwhen cannulating an artery), pressure measurement with afluid column or pressure transducer, or observation of theintraluminal pressure waveform on a monitor. Althoughvascular catheters are commonly inserted over a wire ormetal introducer, some clinicians initially cannulate thevessel with a small caliber (“finder”) needle before theinsertion of a larger bore needle. This technique is mostbeneficial for nonultrasound techniques, because a smallerneedle may minimize the magnitude of an unintendedinjury to surrounding structures. The vessel is then cannu-lated with a larger bore 16-gauge or 18-gauge catheter, aguide wire is passed through it, and a larger catheter isinserted over the wire. The catheter–over–guide wire pro-cess is termed the Seldinger technique.3
Although frequently performed and an inherent part ofmedical training and practice, the insertion of vascularcatheters is associated with complications. Depending onthe site and patient population, landmark techniques forvascular cannulation are associated with a 60% to 95%success rate. A 2003 estimate cited the insertion of �5million central venous catheters (in the IJ, SC, and FV)annually in the United States alone, with a mechanicalcomplication rate of 5% to 19%.4 These complications mayoccur more often with less experienced operators, challeng-ing patient anatomy (obesity, cachexia, distorted, tortuousor thrombosed vascular anatomy, congenital anomalies
Table 1. Categories of Support from Scientific EvidenceCategory A: supportive literatureRandomized controlled trials report statistically significant (P � .01) differences between clinical interventions for a specified clinical outcome.
Level 1: The literature contains multiple randomized controlled trials, and the aggregated findings are supported bymeta-analysis.
Level 2: The literature contains multiple randomized controlled trials, but there is an insufficient number of studies to conduct a viable meta-analysis for the purpose of these guidelines.
Level 3: The literature contains a single randomized controlled trial.Category B: suggestive literatureInformation from observational studies permits inference of beneficial or harmful relationships among clinical interventions and clinical outcomes.
Level 1: The literature contains observational comparisons (e.g., cohort and case-control research designs) of two or more clinical interventionsor conditions and indicates statistically significant differences between clinical interventions for a specified clinical outcome.
Level 2: The literature contains noncomparative observational studies with associative (e.g., relative risk, correlation) or descriptive statistics.Level 3: The literature contains case reports.
Category C: equivocal literatureThe literature cannot determine whether there are beneficial or harmful relationships among clinical interventions and clinical outcomes.
Level 1: Meta-analysis did not find significant differences among groups or conditions.Level 2: There is an insufficient number of studies to conduct meta-analysis, and (1) randomized controlled trials have not found significant
differences among groups or conditions, or (2) randomized controlled trials report inconsistent findings.Level 3: Observational studies report inconsistent findings or do not permit inference of beneficial or harmful relationships.
Category D: insufficient evidence from literatureThe lack of scientific evidence in the literature is described by the following conditions:
1. No identified studies address the specified relationships among interventions and outcomes.2. The available literature cannot be used to assess the relationships among clinical interventions and clinical outcomes. The literature either
does not meet the criteria for content as defined in the “focus” of the guidelines or does not permit a clear interpretation of findingsbecause of methodologic concerns (e.g., confounding in study design or implementation).
Source: American Society of Anesthesiologists and Society of Cardiovascular Anesthesiologists Task Force on Transesophageal Echocardiogra- phy. Practiceguidelines for perioperative transesophageal echocardiography. An updated report by the American Society of Anesthesiologists and the Society of CardiovascularAnesthesiologists Task Force on Transesophageal Echocardiography. Anesthesiology 2010;112:1084–96.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 47
such as persistent left superior vena cava), compromisedprocedural settings (mechanical ventilation or emergency),and the presence of comorbidity (coagulopathy, emphy-sema). Central venous catheter mechanical complicationsinclude arterial puncture, hematoma, hemothorax, pneu-mothorax, arterial-venous fistula, venous air embolism,nerve injury, thoracic duct injury (left side only), intralu-minal dissection, and puncture of the aorta. The mostcommon complications of IJ vein cannulation are arterialpuncture and hematoma. The most common complicationof SC vein cannulation is pneumothorax.4 The incidence ofmechanical complications increases sixfold when morethan three attempts are made by the same operator.4 Theuse of ultrasound imaging before or during vascular can-nulation greatly improves first-pass success and reducescomplications. Practice recommendations for the use ofultrasound for vascular cannulation have emerged fromnumerous specialties, governmental agencies such as theNational Institute for Health and Clinical Excellence5 andthe Agency for Healthcare Research and Quality’s evidencereport.2
4. ULTRASOUND PRINCIPLES FORNEEDLE-GUIDED CATHETER PLACEMENTUltrasound modalities used for imaging vascular structuresand surrounding anatomy include two-dimensional (2D)ultrasound, Doppler color flow, and spectral Doppler in-terrogation. The operator must have an understanding ofprobe orientation, image display, the physics of ultrasound,and mechanisms of image generation and artifacts and beable to interpret 2D images of vascular lumens of interestand surrounding structures. The technique also requiresthe acquisition of the necessary hand-eye coordination todirect probe and needle manipulation according to theimage display. The supplemental use of color flow Dopplerto confirm presence and direction of blood flow requires anunderstanding of the mechanisms and limitations of Dopp-ler color flow analysis and display. This skill set must thenbe paired with manual dexterity to perform the three-dimensional (3D) task of placing a catheter into the targetvessel while using and interpreting 2D images. Two-dimensional images commonly display either the short axis(SAX) or long axis (LAX) of the target vessel, each with itsadvantage or disadvantage in terms of directing the can-nulating needle at the correct entry angle and depth.Three-dimensional ultrasound may circumvent the spatiallimitations of 2D imaging by providing simultaneous real-time SAX and LAX views along with volume perspectivewithout altering transducer location, allowing simultaneousviews of neck anatomy in three orthogonal planes.6 Detailedknowledge of vascular anatomy in the region of interest issimilarly vital to both achieving success and avoiding com-plications from cannulation of incorrect vessels.
Ultrasound probes used for vascular access vary in sizeand shape. Probes with smaller footprints are preferred inpediatric patients. Higher frequency probes ($7 MHz) arepreferred over lower frequency probes (�5 MHz) becausethey provide better resolution of superficial structures inclose proximity to the skin surface. The poorer penetrationof the high-frequency probes is not typically a hindrance,
because most target vascular structures intended for can-nulation are �8 to 10 cm from the skin surface.
It is important to appreciate how probe orientationrelates to the image display. Conventions established by theASE for performing transthoracic imaging of the heart, andmore recently epicardial imaging, established that theprobe indicator and right side of the display should beoriented toward the patient’s left side or cephalad.7 In thesesettings, projected images correlate best with those visual-ized by the sonographer positioned on the patient’s left sideand facing the patient’s right shoulder. In contrast, theoperator’s position during ultrasound-guided vascular ac-cess varies according to the target vessel. For example, theoperator is typically positioned superior to the patient’shead and faces caudally during cannulation of the IJ vein.The left side of the screen displays structures toward thepatient’s left side (Figure 1). In contrast, during cannulationof the FVs, the operator is typically positioned inferiorlyand faces cephalad, so that the left side of the screendisplays structures toward the patient’s right side (seesection 9, “Femoral Vein Cannulation”). For SC vein can-nulation, the left and right sides of the screen displaycephalad and caudad structures, depending on laterality(right or left). The changing image orientation is an impor-tant distinction from typical transthoracic, epicardial, ortransesophageal imaging. For ultrasound-guided vascularaccess cannulation, the probe and screen display are bestoriented to display the anatomic cross-section that wouldbe visible from the same vantage point. Therefore, screenleft and right will not follow standard conventions butrather vary with site and needle insertion orientation. Whatis common for all vascular access sites is that it is essential forthe operator to orient the probe so that structures beneath theleft aspect of the probe appear on the left side of the imagingscreen. Although probes usually have markings that distin-guishes one particular side of the transducer, the operatormust identify which aspect of the screen corresponds to themarking on the probe. These markings may be obscure, and arecommended practice is to move the probe toward onedirection or another while observing the screen or apply
Figure 1. Right neck central vein cannulation. The ultrasound probeis held so that each side of the screen displays ipsilateral structures.With the probe mark placed on the upper left corner of the image, thedisplayed structures will move in the same direction with the probe.
SPECIAL ARTICLE
48 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
modest external surface pressure on one side of the trans-ducer to demonstrate proper alignment of left-right probeorientation with image display.
The probe used ultimately depends on its availability,operator experience, ease of use, and patient characteristics(e.g., smaller patients benefit from smaller probes). Someprobes allow the use of a needle guide, which directs theneedle into the imaging plane and defined depth as viewedon the display screen (Figure 2). Needle guides are notavailable from every ultrasound probe manufacturer, but aneedle guide may be a useful feature for the beginner whohas not yet developed the manual dexterity of using a 2Dimage display to perform a 3D task. One study thatevaluated ultrasound-guided cannulation of the IJ veinwith and without a needle guide showed that its usesignificantly enhanced cannulation success after first(68.9%–80.9%, P � .0054) and second (80.0%–93.1%, P �.0001) needle passes.8 Cumulative cannulation success afterseven needle passes was 100%, regardless of technique. Theneedle guide specifically improved first-pass successamong more junior operators (65.6%–79.8%, P � .0144),while arterial puncture averaged 4.2%, regardless of tech-nique (P � .05) or operator (P � .05). A limitation of theneedle guide is that the needle trajectory is limited toorthogonal orientations from the SAX imaging plane. Al-though helpful in limiting lateral diversion of the needlepath, sometimes oblique angulation of the needle path mayfacilitate target vessel cannulation. In addition, there maybe considerable costs associated with the use of needleguides. Depending on the manufacturer, they may cost aslittle as several dollars to �$100 each. Importantly, al-though the needle guide facilitated prompt cannulationwith ultrasound in the novice operator, it offered noadditional protection against arterial puncture.8 However,one in vitro simulation study has refuted these in vivoresults.9
Arterial puncture during attempted venous cannulationwith ultra-sound generally occurs because of a misalign-ment between the needle and imaging screen. It may alsooccur as a result of a through-and-through puncture of thevein into a posteriorly positioned artery. The first scenariois due to improper direction of the needle, while the latteroccurs because of a lack of needle depth control. Needledepth control is also an important consideration becausethe anatomy may change as the needle is advanced deeperwithin the site of vascular access. The ideal probe shouldhave a guide that not only directs the needle to the center ofthe probe but also directs the needle at the appropriateangle beneath the probe (Figure 2). This type of guidecompensates for the limitation of using 2D ultra-sound toperform a 3D task of vascular access. The more experiencedoperator with a better understanding of these principlesand better manual dexterity may find the needle guidecumbersome, choosing instead the “maneuverability” of afreehand technique. Although the routine use of a needleguide requires further study, novice operators are morelikely to improve their first-pass success.
Vascular structures can be imaged in SAX, LAX, oroblique orientation (Figures 3A, 3B, and 3C). The advan-tage of the SAX view is better visualization of surroundingstructures and their relative positions to the needle. There isusually an artery in close anatomic proximity to mostcentral veins. Identification of both vascular structures isparamount to avoid unintentional cannulation of the ar-tery. In addition, it may be easier to direct the cannulatingneedle toward the target vessel and coincidentally awayfrom surrounding structures when both are clearly imagedsimultaneously. The advantage of the LAX view is bettervisualization of the needle throughout its course and depthof insertion, because more of the needle shaft and tip areimaged within the ultrasound image plane throughout itsadvancement, thereby avoiding insertion of the needle
Figure 2. Various needle guides, usedto direct the needle at the center of theprobe (and image) and at an appropri-ate angle and depth beneath the probe.IJV, IJ vein. From Troianos CA. Intraop-erative monitoring. In: Troianos CA, ed.Anesthesia for the Cardiac Patient. NewYork: Mosby; 2002.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 49
beyond the target vessel. A prospective, randomized obser-vational study of emergency medicine residents evaluatedwhether the SAX or LAX ultrasound approach resulted infaster vascular access for novice ultrasound users.10 TheSAX approach yielded a faster cannulation time comparedwith the LAX approach, and the novice operators perceivedthe SAX approach as easier to use than the LAX approach.The operator’s hand-eye coordination skill in aligning theultrasound probe and needle is probably the most impor-tant variable influencing needle and target visibility. Imag-ing in the SAX view enables the simultaneous visualizationof the needle shaft and adjacent structures, but this viewdoes not image the entire needle pathway or provide anappreciation of insertion depth. Although novice users mayfind ultrasound guidance easier to adopt using SAX imag-ing, ultrasound guidance with LAX imaging should bepromoted, because it enables visualization of the entireneedle and depth of insertion, thereby considering ana-tomic variations along the needle trajectory as the needle isadvanced deeper within the site of vascular access. Theoblique axis is another option that may allow better visu-alization of the needle shaft and tip and offers the safety ofimaging surrounding structures in the same view, thuscapitalizing on the strengths of both the SAX and LAXapproaches.11
5. REAL-TIME IMAGING VERSUS STATIC IMAGINGUltrasound guidance for vascular access is most effectivewhen used in real time (during needle advancement) witha sterile technique that includes sterile gel and sterile probecovers. The needle is observed on the image display andsimultaneously directed toward the target vessel, awayfrom surrounding structures, and advanced to an appro-priate depth. Static ultrasound imaging uses ultrasoundimaging to identify the site of needle entry on the skin overthe underlying vessel and offers the appeal of nonsterileimaging, which obviates the need for sterile probe cover-ings, sterile ultrasound gels, and needle guides. If ultra-sound is used to mark the skin for subsequent cannulationwithout real-time (dynamic) use, ultrasound becomes avessel locator technique that enhances external landmarksrather than a technique that guides the needle into thevessel. Both static and real-time ultrasound-guided ap-proaches are superior to a traditional landmark-guided
approach. Although the real-time ultrasound guidanceoutperforms the static skin-marking ultrasound approach,complication rates are similar.12
Venous puncture using real-time ultrasound was fasterand required fewer needle passes among neonates andinfants randomly assigned to real-time ultrasound-assistedIJ venous catheterization versus ultrasound-guided skinmarking.13 Fewer than three attempts were made in 100%of patients in the real-time group, compared with 74% ofpatients in the skin-marking group (P � .01). In this study,a hematoma and an arterial puncture occurred in onepatient each in the skin-marking group.13
One operator can usually perform real-time ultrasound-guided cannulation. The nondominant hand holds theultrasound probe while the dominant hand controls theneedle. Successful cannulation of the vessel is confirmed bydirect vision of the needle entering the vessel and withblood entering the attached syringe during aspiration. Theprobe is set aside on the sterile field, the syringe removed,and the wire is inserted through the needle. Further con-firmation of successful cannulation occurs with ultrasoundvisualization of the guide wire in the vessel. Difficultcatheterization may benefit from a second person withsterile gloves and gown assisting the primary operator byeither holding the transducer or passing the guide wire.
6. VESSEL IDENTIFICATIONMorphologic and anatomic characteristics can be used todistinguish a vein from an artery with 2D ultrasound. Forexample, the IJ vein has an elliptical shape and is larger andmore collapsible with modest external surface pressurethan the carotid artery (CA), which has rounder shape,thicker wall, and smaller diameter (Figure 4). The IJ veindiameter varies depending on the position and fluid statusof the patient. Patients should be placed in Trendelenburgposition to increase the diameter of the jugular veins14,15
and reduce the risk for air embolism when cannulating theSC vein, unless this maneuver is contraindicated. A Val-salva maneuver will further augment their diameter15 andis particularly useful in hypovolemic patients. AddingDoppler, if available, can further distinguish whether thevessel is a vein or an artery. Color flow Doppler demon-strates pulsatile blood flow in an artery in either SAX or
Figure 3. Two-dimensional imaging of theright IJ vein (IJV) and CA from the head ofthe patient over their right shoulder. (A)SAX, (B) LAX, (C) oblique axis. SAX imagingdisplays the lateral-right side of the patienton the right aspect of the display screenand the medial structures on the left as-pect of the display screen. LAX imagingdisplays the caudad structures on the rightaspect of the display screen and cephaladstructures on the left aspect of the displayscreen. If the transducer is rotated counter-clockwise about 30–40 degrees, obliqueimaging displays more lateral-right caudadstructures on the right aspect of the displayscreen, while more medial-left cephaladstructures are on the left aspect of thedisplay screen.
SPECIAL ARTICLE
50 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
LAX orientation. A lower Nyquist scale is typically re-quired to image lower velocity venous blood flows. Atthese reduced settings, venous blood flow is uniform incolor and present during systole and diastole with laminarflow, whereas arterial blood flow will alias and be detected
predominantly during systole (Figure 5) in patients withunidirectional arterial flow (absence of aortic regurgita-tion). A small pulsed-wave Doppler sample volume withinthe vessel lumen displays a characteristic systolic flowwithin an artery, while at the same velocity range displays
Figure 4. Vessel identification. Right IJ vein (top)and CA (bottom) in SAX and LAX orientation. Slightexternal pressure compresses the oval-shapedvein but not the round-shaped artery.
Figure 5. Vessel identification with color flow Doppler. Arterial flow is visible in systole only, irrespective of Nyquist limit. Venous flow is visiblein systole and diastole, but only if the Nyquist limit is sufficiently decreased.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 51
biphasic systolic and diastolic flow and reduced velocity ina vein. A lower pulsed-wave Doppler velocity range makesthis distinction more apparent (Figure 6).
Misidentification of the vessel with ultrasound is acommon cause of unintentional arterial cannulation.Knowledge of the relative anatomic positions of the arteryand vein in the particular location selected for cannulationis essential and is discussed below in the specific sections.Ultrasound images of veins and arteries have distinctcharacteristics. Veins are thin walled and compressible andmay have respiratory-related changes in diameter. In con-trast, arteries are thicker walled, not readily compressed byexternal pressure applied with the ultrasound probe, andpulsatile during normal cardiac physiologic conditions.Obviously, arterial pulsatility cannot be used to identify anartery during clinical conditions such as cardiopulmonarybypass, nonpulsatile ventricular circulatory assistance, andcardiac or circulatory arrest. Confirmation of correct cath-eter placement into the intended vascular structure iscovered later in this document.
7. INTERNAL JUGULAR VEIN CANNULATION7.1. Anatomic ConsiderationsThe IJ is classically described as exiting the external jugularforamen at the base of the skull posterior to the internalcarotid and coursing toward an anterolateral position (inrelation to the carotid) as it travels caudally. Textbookanatomy does not exist in all adult and pediatric patients.Denys and Uretsky16 showed that the IJ was locatedanterolateral to the CA in 92%, �1 cm lateral to the carotidin 1%, medial to the carotid in 2%, and outside of the pathpredicted by landmarks in 5.5% of patients. The anatomy ofthe IJ is sufficiently different among individual patients tocomplicate vascular access with a “blind” landmarkmethod (Figure 7). Therefore at a minimum, a clear andintuitive advantage of using static ultrasound imaging forskin marking is the ability to identify patients in whom thelandmark technique is not likely to be successful.
7.2. Cannulation TechniqueThe traditional approach to IJ vein cannulation uses exter-nal anatomic structures to locate the vein. A commonapproach identifies a triangle subtended by the two headsof the sternocleidomastoid muscle and the clavicle (Figure8). A needle placed at the apex of this triangle and directedtoward the ipsilateral nipple should encounter the IJ 1.0 to1.5 cm beneath the skin surface. The use of externallandmarks to gain access to the central venous system isconsidered a safe technique in experienced hands. A failurerate of 7.0% to 19.4%17 is due partly to the inability ofexternal landmarks to precisely correlate with the locationof the vessel.18 Furthermore, when initial landmark-guidedattempts are unsuccessful, successful cannulation dimin-ishes to �25% per subsequent attempt.19 Additionally,there exists a strong direct correlation between the numberof attempts and the incidence of complications, increasingpatient anxiety and discomfort, and potentially delayingmonitoring and infusion of fluids or medications necessaryfor definitive care. These are important quality of careissues that must be considered when choosing the besttechnique for central venous access.
Many studies have shown a clear advantage of ultra-sound guidance over landmark guidance for IJ centralvenous cannulation.8,12,13,19–22 Troianos et al.19 demon-strated that the overall success rate of central venouscannulation could be improved from 96% to 100% with theuse of ultrasound. This may not seem signifi-cant until oneconsiders the improved first attempt success rate (from 54%to 73%), decreased needle advances (from 2.8 to 1.4 at-tempts), decreased time to cannulation (from 117 to 61 sec),and lower rate of arterial punctures (from 8.43% to 1.39%).
Several ultrasound studies have elucidated the anatomicrelation between the IJ and CA, particularly in terms ofvessel overlap.23–27 Sulek et al.25 prospectively examinedthe effect of head position on the relative position of the CAand the IJ. The percentage of overlap between the IJ and theCA increased as the head was rotated contralaterally fromneutral (0°) to 40° to 80°. Troianos et al.23 found �75%overlap among 54% of all patients whose heads wererotated to the contralateral side (image plane positioned inthe direction of the cannulating needle; Figure 9). Addition-ally, two thirds of older patients (age $60 years) had �75%overlap of the IJ and CA. Age was the only demographicfactor that was associated with vessel overlap. The concernis that vessel overlap increases the likelihood of uninten-tional CA puncture by a through-and-through puncture ofthe vein. The accidental penetration of the posterior vesselwall can occur despite the use of ultrasound when the SAXimaging view is used for guidance.26 Typically, the anteriorwall of the vein is compressed as the needle approaches thevein (Figure 10). The compressive effect terminates as theneedle enters the vein (heralded by the aspiration of bloodinto the syringe) and the vessel assumes its normal shape.A low-pressure IJ may partially14 or completely compressduring needle advancement, causing puncture of the ante-rior and posterior walls without blood aspiration into thesyringe.26–28 IJ-CA overlap increases the possibility ofunintentional arterial puncture as the “margin of safety”decreases. Some authors have describe the “margin ofsafety” as the distance between the midpoint of the IJ andthe lateral border of the CA. This zone represents the area ofnonoverlap between the IJ and CA. The margin of safetydecreases, and the percentage overlap increases from 29%to 42% to 72% as the head is turned to the contralateral sidefrom 0° (neutral) to 45° to 90° , respectively.29 Vesseloverlap increasing with head rotation is most apparentamong patients with increased body surface areas (�1.87m2) and increased body mass indexes (�25 kg/m2).27
Ultrasound can be used to alter the approach angle to avoidthis mechanism of CA puncture by directing the advancingneedle away from the CA (Figure 11).30 Vascular anomaliesand anatomic variations of the IJ and surrounding tissueshave been observed in up to 36% of patients.31 Ultrasoundidentifies the vein size and location, anomalies, and vesselpatency, thus avoiding futile attempts in patients withabsent or thrombosed veins and congenital anomalies suchas persistent left superior vena cava. Denys et al.20 ob-served small fixed IJs in 3% of patients. An ultrasound veindiameter �7 mm (cross-sectional area �0.4 cm2) is associ-ated with decreased cannulation success,32,33 prompting
SPECIAL ARTICLE
52 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
redirection to another access site, thus reducing cannula-tion time and patient discomfort.24 Ultrasound also identi-fies disparity in patency and size between the right IJ andthe left IJ (the right IJ usually larger than the left IJ).33–36
Maneuvers that increase the size of the IJ and thus potentiallyimprove the cannulation success include the Valsalva maneu-ver (Figure 12) and the Trendelenburg position.14,15,34
7.3. ComplicationsSeveral factors contribute to the success rate, risk, andcomplications associated with central venous cannulation,including patient characteristics, comorbidities, and accesssite. Although the landmark method is associated with anarterial puncture risk of 6.3% to 9.4% for the IJ, 3.1% to 4.9%
for the SC, and 9.0% 15.0% for the FV,4,19,34 Ruesch et al.37
demonstrated a higher incidence of arterial puncture duringattempted IJ versus SC central venous access. Obese patientswith their attendant short thick necks and others with ob-scured external landmarks derive a particular benefit fromultrasound guidance38 by decreasing the incidence of arterialpuncture, hematoma formation, and pneumothorax.39 The
Figure 6. Vessel identification with pulsed-wave Doppler will distinguish artery (A) from vein (B) at a Nyquist limit of 650 cm/sec. Arterial bloodflow has a predominately systolic component and higher velocity (A) compared with venous blood flow (B, C), which has systolic and diastoliccomponents and much lower velocity, better delineated with a lower Nyquist scale (69 cm/sec) (C).
Figure 7. Variable overlap between CA and IJ vein. RIJV, Right IJ vein.Adapted from J Vasc Interv Radiol.24
Figure 8. External landmarks for IJ cannulation. SCM, Sterno-cleidomastoid muscle. Modified from N Engl J Med.4
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 53
recognition and avoidance of pleural tissue during real-time ultrasound imaging could potentially decrease the riskfor pneumothorax for approaches that involve a needleentry site closer to the clavicle. High-risk conditions includehemostasis disorders,40 uncooperative or unconscious pa-tients, critically ill patients21 who may be hypovolemic,34
and patients who have had multiple previous catheter
insertions. Oguzkurt et al.41 prospectively reviewed 220temporary IJ dialysis catheters placed sonographically byinterventional radiologists in 171 high- risk patients (27.7%with bleeding tendency, 10% uncooperative, 2% obese, 37%with previous catheters, and 21.3% with bedside procedurebecause their medical conditions were not suitable fortransport to the radiology suite). The success rate was100%, with only seven complications among the 171 pro-cedures. The carotid puncture rate was 1.8%, while oozingaround the catheter, small hematoma formation, and pleu-ral puncture without pneumothorax occurred at rates of1.4%, 0.4%, and 0.4%, respectively.
In summary, ultrasound imaging of the IJ and surround-ing anatomy during central venous cannulation both facili-tates identification of the vein and improves first-passcannulation but also decreases the incidence of injury toadjacent arterial vessels.
7.4. Recommendation for IJ Vein CannulationIt is recommended that properly trained clinicians use real-timeultrasound during IJ cannulation whenever possible to improvecannulation success and reduce the incidence of complicationsassociated with the insertion of large-bore catheters. This recom-mendation is based on category A, level 1 evidence.
The writing committee recognizes that static ultrasound(when not used in real time) is useful for the identification ofvessel anatomy by skin-marking the optimal entry site forvascular access and for the identification of vessel thrombosisand is superior to a landmark-guided technique.
8. SUBCLAVIAN VEIN CANNULATION8.1. Anatomic ConsiderationsLandmark-guided SC vein access uses the anatomic land-marks of the midpoint of the clavicle, the junction betweenthe middle and medial border of the clavicle, and the lateralaspect of a tubercle palpable on the medial part of theclavicle. The most common approach is to insert the needle1 cm inferior to the junction of the middle and medial thirdof the clavicle at the deltopectoral groove. The degree oflateral displacement of the entrance point is based on thepatient’s history and anatomic considerations.
8.2. Cannulation TechniqueThe landmark-guided approach to the central venous cir-culation via the SC vein is generally considered by manyclinicians to be the simplest method to access this vein.Several million SC vein catheters are placed each year in theUnited States. The risk factors for complications and failuresare poorly understood, with the exception of physician expe-rience. Advantages of using the SC vein for central venousaccess include consistent surface anatomic landmarks andvein location, patient comfort, and lower potential for infec-tion.42 In contrast to attempted IJ vein cannulation, in whichunintentional injury to the adjacent CA can compromisecirculation to the brain, unintentional injury to the adjacent SCartery during SC vein cannulation carries a less morbidsequela. The physician’s experience and comfort level withthe procedure are the main determinants for successfulplacement of a SC vein catheter, when there are no other
Figure 9. Magnetic resonance imaging of neck anatomy. Contralat-eral turn of the neck increases the overlap between IJ vein (v) and CA(a). From Anesthesiology.23
Figure 10. The anterior wall of the IJ vein (IJV) recesses as theneedle approaches the vein (left). The vein assumes its normalshape after the needle penetrates its wall (right).
Figure 11. Under ultrasound guidance, the needle approach to the IJvein (IJV) can be altered to avoid CA puncture. From CardiovascIntervent Radiol.30
SPECIAL ARTICLE
54 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
patient-related factors that increase the incidence of compli-cations. The SC vein may be cannulated from a supraclavic-ular or an infraclavicular approach. The infraclavicularapproach is the most common approach and hence is thefocus of this discussion. The supraclavicular approach (with-out ultrasound) has largely been abandoned because of ahigh incidence of pneumothorax. As experience withultrasound-guided regional anesthesia for upper extremityblocks has increased imaging and identification of thesupraclavicular vessels and nerves, clinicians are gainingmore familiarity with imaging the supraclavicularapproach to the SC vein using ultrasound for vessel can-nulation. Whether this approach will continue to gainpopularity remains to be demonstrated.
Tau et al.43 analyzed anatomic sections of the clavicleand SC vein and determined that the supine position withneutral shoulder position and slight retraction of the shoul-ders was the most effective method to align the vein for alandmark-based technique. Although many clinicians placepatients in the Trendelenburg (head-down) position todistend the central venous circulation, there is less vesseldistention of the SC vein than the IJ vein because the SCvein is fixed within the surrounding tissue, so relativechanges in size are not realized to the same degree as withthe IJ vein. Thus, the primary reason for the Trendelenburgposition is to reduce the risk for air embolism in spontane-ously breathing patients.
Ultrasound-directed vascular cannulation may lead in-experienced operators to use needle angle approaches thatlead to an increased risk for complications. It is importantthat traditional approaches and techniques are not aban-doned with ultrasound guidance, particularly during can-nulation of the SC vein, in which a steeper needle entryangle may lead to pleural puncture. The needle is directedtoward the sternal notch in the coronal plane. The bevel ofthe needle should be directed anteriorly during insertionand gentle aspiration applied with a syringe, as the needleenters the skin at a very low (nearly parallel) angle to thechest wall. An increased or steeper angle increases thelikelihood of creating a pneumothorax. The needle bevelmay be turned caudally upon venopuncture to direct theguide wire toward the right atrium. The wire is advanced,leaving enough wire outside the skin for advancement of
the entire catheter length over the wire (i.e., the wire shouldextend beyond the catheter outside the skin). The electro-cardiogram should be closely monitored for ectopy thatmay occur when the wire is advanced into the right atriumor right ventricle. Chest radiography is mandatory not onlyto confirm proper line placement but also to rule outpneumothorax.
Similar preparation of the patient occurs withultrasound-guided cannulation as with the landmark-guided approach with respect to positioning, skin prepara-tion, and vascular access kits. The use of a smaller footprinttransducer probe for SC vein access for real-time ultra-sound imaging is recommended because larger probesmake imaging of the vein more challenging. It is generallymore difficult to position the larger footprint probe be-tween the clavicle and rib to obtain an adequate SC veinimage. Despite some loss of resolution in the far field thatinherently occurs with phased-array transducers, smallerprobes may allow better maneuverability underneath theclavicle. Similar to the landmark technique, the middlethird of the clavicle is chosen as the site used for ultrasoundimaging and subsequent needle insertion. The transducer isoriented to image the SC vein in the SAX view with acoronal imaging plane. The vein appears as an echo-lucentstructure beneath the clavicle (Figure 13). It is important todistinguish between pulsatility on the vein due to respira-tory variation and pulsatility of the artery. Confirmation ofthe venous circulation can be facilitated by the injection ofagitated saline “echo contrast” into a vein of the ipsilateralarm (if available) with subsequent imaging of the micro-bubbles in the vein. Confirmation can also be achieved byaddition of color flow Doppler to the ultrasound assess-ment. When positioning the transducer marker toward theleft shoulder (during right SC vein cannulation), arterial flowwill be the color that indicates flow away from the transducer,while venous flow will be the color that indicates flow towardthe transducer. It is important to ensure correct transducerorientation before using color flow Doppler to determine theidentification of artery or vein. Considerable skin pressure isrequired to obtain adequate imaging planes (windows) thatmay incur some patient discomfort.
A prospective randomized SC vein cannulation studyfavored the ultrasound-guided over the landmark-guided
Figure 12. The size of the IJ vein (IJV) isincreased with a Valsalva maneuver (B)compared with apnea (A).
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 55
approach, with a higher success rate (92% vs 44%), fewerminor complications (1 vs 11), and fewer venopunctures(1.4 vs 2.5) and catheter kits (1.0 vs 1.4) per attemptedcannulation.44 A more recent study of 1,250 attemptedcentral venous catheter placements included 354 SC veinattempts. The incidence and success rates with ultrasoundguidance during central venous catheter placement sup-ported the impression of many clinicians that the addedbenefit of ultrasound for cannulation of the SC vein is lessthan the benefit of ultrasound during attempted cannula-tion of the IJ vein. Although ultrasound use was uncom-mon for cannulation of the SC vein, either as the primarytechnique or as a rescue technique, success rates were highwith ultrasound guidance even when surface techniqueswere unsuccessful.45
8.3. ComplicationsComplication rates for the landmark-guided approach toSC vein cannulation are 0.3% to 12% and include pneumo-thorax, hematoma, arterial puncture, hemothorax, air em-bolism, dysrhythmia, atrial wall puncture from the wire,lost guide wire, anaphylaxis in patients who are allergic toantibiotics upon the insertion of an antibiotic-impregnatedcatheter, catheter malposition, catheter in the wrong vessel,and thoracic duct laceration (left side only).46
Kilbourne et al.47 reported the most common errorsduring failed SC vein catheter placement attempts byresident physicians were inadequate landmark identifica-tion, improper insertion position, advancing the needlethrough periosteum, a shallow or cephalad needle angle,and loss of intravenous needle position while attemptingto place the guide wire. Factors associated with cannu-lation failure were previous major surgery, radiationtherapy, prior catheterization, prior attempts at catheter-ization, high body mass index, more than two needlepasses, only 1 year of postgraduate training, lack ofclassic anatomy, and previous first rib or clavicle frac-ture. If only one needle pass was attempted, the failurerate for subsequent catheter placement was 1.6%, com-pared with 10.2% for two passes and 43.2% for three or
more passes. In the 8.7% of patients in whom initialattempts at catheterization failed, subsequent attemptsby second physicians were successful in 92%, with acomplication rate of 8%.47 Similar success was demon-strated in a study of patients undergoing SC vein can-nulation with and without ultrasound guidance.48 Theultrasound group had fewer attempts, better patientcompliance, and a zero incidence of pneumothorax,while the incidence of pneumothorax in the landmarkgroup was 4.8%.
Identification of risk factors before catheter insertionmay decrease complication rates by altering the approachto include ultrasound guidance. Additionally, in patientswith body mass indexes �30 kg/m2 or �20 kg/m2, historyof previous catheterization, prior surgery, or radiotherapyat the site of venous access, experienced physicians shouldattempt catheter placement rather than physicians who arelearning the procedure.
Obese patients with their attendant short thick necksand others with obscured external landmarks derive aparticular benefit from ultrasound guidance38 by decreas-ing the incidence of arterial puncture, hematoma forma-tion, and pneumothorax.39 Mansfield et al.46 noted that abody mass index �30 kg/m2 resulted in a cannulationfailure rate of 20.1% for attempted SC vein cannulation.These investigators found no benefit of ultrasound guid-ance for SC vein catheterization, but in comparing ultra-sound with landmark-guided techniques, Hind et al.’s22
meta-analysis found that the landmark technique had ahigher relative risk for failed catheter placements and meantime to successful cannulation. As operators gain moreexperience with the use of ultrasound for guiding catheter-ization and diagnostic procedures, it is likely that anincremental benefit with the use of ultrasound for SC veincannulation will also be realized. Orihashi et al.49 found abenefit to the use of ultrasound in SC venopuncture in asmall cohort of 18 patients. Although Gualtieri et al.44
demonstrated improved success and fewer minor compli-cations with use of ultrasound for SC vein cannulation,there were no major complications in either group. Theoverwhelming evidence in the literature supports the rou-tine use of ultrasound for IJ access, but the data on the SCapproach warrant consideration of anatomic landmarksand interference of the clavicle as impediments to the use ofreal-time ultrasound for this approach.
8.4. Recommendation for SC Vein CannulationCurrent literature does not support the routine use of ultra-sound for uncomplicated patients undergoing SC vein cannu-lation. Individual operators should not attempt cannulationmore than twice, as the incidence of complication, particularlypneumothorax, rises significantly with additional attempts.High-risk patients may benefit from ultrasound screening of the SCvein before attempted cannulation to identify vessel location andpatency and to specifically identify thrombus before attemptedcannulation. The recommendation for ultrasound guidanceduring SC vein cannulation is based on category A (support-ive), level 3 evidence.
Figure 13. Two-dimensional ultrasound image of the left SC vein andleft SC artery obtained from the left side of the patient duringultrasound-guided cannulation of the left SC vein.
SPECIAL ARTICLE
56 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
9. FEMORAL VEIN CANNULATION9.1. Anatomic ConsiderationsThe femoral vessels are often used to provide access forleft-sided and right-sided cardiac procedures. In addition,the common FV is often used for central venous accessduring emergency situations,50 because of its relative safeand accessible location with predictable anatomic land-marks (i.e., lying within the femoral triangle in theinguinal-femoral region). A detailed understanding of theregional anatomy is important for performing FV cannula-tion using a landmark-guided technique.
The common femoral artery and FV lie within thefemoral triangle in the inguinal-femoral region. The supe-rior border of this triangle is formed by the inguinalligament, the medial border by the adductor longusmuscle, and the lateral border by the sartorius muscle.Another important landmark is the femoral artery pulse,because the common FV typically lies medial to the com-mon femoral artery within the femoral sheath. The femoralartery lies at the midpoint of the inguinal ligament connect-ing the anterior superior iliac spine to the pubic tubercle,while the common FV is typically located me-dial to thecommon femoral artery. This side-by-side relationship ofthe common femoral artery and FV occurs in close proxim-ity to the inguinal ligament, but significant vessel overlapmay occur, particularly in children.51,52 In addition, it isessential to understand that the relationship between theinguinal crease and the inguinal ligament is highly vari-able, so the inguinal crease is not always a useful surfacelandmark.53
The femoral site has numerous advantages both withelective vascular access and in critically ill patients. Thefemoral site remains the most commonly accepted site forvascular access for cardiac procedures because of relativelyshort access times and few complications. For critically illpatients, it is relatively free of other monitoring and airwayaccess devices, allowing arm and neck movement withoutimpeding the access line. Femoral access avoids the risks ofhemothorax and pneumothorax, which is particularly im-portant in patients with severe coagulopathy or profoundrespiratory failure. In addition, the femoral site permitscannulation attempts without interruption of cardiopulmo-nary resuscitation during cardiac arrest. However, thefemoral approach is associated with complications, includ-ing bleeding and vascular injury, such as pseudoaneu-rysms, arteriovenous fistulas and retroperitoneal bleeding(see section 9.3, “Complications”).
9.2. Cannulation TechniqueSimilar to other central venous cannulation sites, the modi-fied Seldinger technique is most common method used toaccess the common FV.3 The procedure requires patientpositioning with the hip either in the neutral position orwith slight hip abduction and external rotation. Abductionand external rotation increases the accessibility of thecommon FV from 70% to 83% in adults and increases thevessel diameter in children compared with a straight-legapproach.54,55 The reverse Trendelenburg position in-creases common FV cross-sectional area by �50%.56
The surface landmarks are identified and the FV locatedby palpating the point of maximal femoral artery pulsation
1 to 2 cm below the midpoint of the inguinal ligament.50,57
The FV is located by inserting a needle 1cm medial to themaximal pulsation, directed cephalad and medially at a 45°angle to the skin.
Many clinicians advocate the use of a small (25-gauge)exploratory or “finder” needle to initially identify the veinlocation. A larger 20-gauge to 22-gauge needle is subse-quently placed directly adjacent to the finder needle alonga parallel path to the FV. The vein is normally 2 to 4 cmbeneath the skin in most adults.
9.3. ComplicationsThere are a number of complications associated with FVcannulation.58,59 Infection remains one of the most commonproblems with femoral catheters because of their closeproximity to the perineal region, which is the reason thatthis site is not typically recommended for long-term cath-eters. Some investigators, who have demonstrated that theincidence of catheter-related bloodstream infection withfemoral catheters is not significantly different from theincidence with the supraclavicular access sites, dispute thisrisk.60,61 The number of attempts to gain access mayincrease the risk for infection but seems to be primarilyrelated to the duration of catheter use at the site. Thecomplications of FV cannulation directly related to catheterinsertion technique are most often due to unintentionalfemoral artery puncture. Because of the close proximity tothe common femoral artery, arterial puncture may occur ifthe needle is directed too laterally. This may result inhematoma, retroperitoneal bleeding, pseudoaneurysm, andarteriovenous fistula formation. In addition, thrombus maydevelop within the FV or iliac vein because of the presenceof the catheter or during compression upon removal. If theneedle is directed too laterally, the patient may experienceparesthesia with the potential for femoral nerve injury.Other rare but serious complications include bowel pen-etration and bladder puncture.
Complications occur despite the optimal use of surfacelandmark-guided techniques (Figure 14). Ultrasound imag-ing at the femoral site has demonstrated that surfaceanatomic landmarks are less useful in projecting the under-lying anatomy, although surface anatomy is more reliablewhen the cannulation site is closer to the inguinal liga-ment.21 Ultrasound-guided femoral artery and FV cannu-lation most likely reduces the incidence of complicationsbecause the anatomy is better defined.62 Iwashima et al.63
and Seto et al.64 demonstrated reduction in vascular-relatedcomplications due to inadvertent femoral artery or FVpuncture with the use of ultrasound guidance duringfemoral vessel cannulation.
9.4. Recommendation for FV CannulationThe scientific evidence for real-time ultrasound-guided FVcannulation is category C, level 2: equivocal with insuffi-cient scientific evidence to support a recommendation forroutine use. In addition, complications during FV cannula-tion are less severe than those that occur with SC and IJvein cannulation. It is therefore the recommendation of thiswriting committee that real-time ultrasound be used only forexamining the FV to identify vessel overlap and patency whenfeasible.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 57
10. PEDIATRIC ULTRASOUND GUIDANCEThe United Kingdom’s National Institute for Health andClinical Excellence guidelines recommend real-time use ofultrasound during central vein cannulation in all patients,children and adults.5 Data to support this practice inpediatrics are limited. In a meta-analysis that includedpediatric studies, Hind et al.22 confirmed a higher successrate with 2D ultrasound compared with anatomic land-mark techniques for the IJ vein cannulation among infants.Hosokawa et al.13 demonstrated in a randomized trial of 60neonates weighing �7.5 kg that real-time ultrasound re-duced the cannulation time and needle passes necessary forcannulation of the right IJ vein compared with a surface-marking technique. Grebenik et al.65 demonstrated that useof ultrasound during IJ vein cannulation in children im-proved success rate and lowered the incidence of carotidpuncture. Some have suggested that use of ultrasound byexperienced operators during central venous cannulationin children may be an initial hindrance.13,65 Avoidingcompression of small veins by the ultrasound probe in realtime takes experience to overcome. As noted by Hosokawaet al., most studies demonstrating a positive correlationwith ultrasound use tend to involve operators in training(e.g., fellows), whereas negative correlation studies usuallyinvolves “experienced” anesthesiologists (i.e., the “can’tteach a old dog new tricks” phenomenon).
Despite governmental recommendations and improve-ment in patient safety, the adoption of ultrasound forcentral venous placement by practitioners has been slow.Tovey and Stokes’s66 survey showed that ultrasound wasused in only 25% of pediatric patients undergoing electivesurgery. In addition, three quarters of the respondents were
not specifically trained in the technique. This is consistentwith the National Institute for Health and Clinical Excel-lence guideline 49 follow-up survey, which demonstratedthat only 28% of the anesthesiologists surveyed werecompliant with the guidelines.67 Two years after the guide-lines were instituted, almost half of those surveyed did nothave access to the ultrasound technology, and two thirdslacked necessary training in the technique.
10.1. Cannulation Technique forPediatric Patients10.1.1. IJ VeinThe most frequently accessed central vein using ultrasound inpediatric patients is the right IJ vein. Ultrasound allows easyvisualization of the vessel, demonstrating its position, itspatency and the presence of thrombus.68 Hanslik et al.69 dem-onstrated a 28% incidence of deep venous thrombosis in a seriesof children with short-term central venous line placement. This isproblematic in children requiring frequent central venous access,as in the pediatric cardiac surgical population.
Although one meta-analysis involving five ultrasoundstudies performed solely among infants and children didnot demonstrate an effect on failure rate, nor the rate ofcarotid puncture, hematoma, hemothorax, or pneumotho-rax, the studies included in this meta-analysis used ultra-sound for “prelocation” and/or guidance.70 Ultrasoundwas not used in real time for all patients in this meta-analysis. The council recommends real-time use to derivethe most benefit from ultrasound guidance.
Liver compression may be used to increase IJ size inpediatric patients.34 Alternatively, the Trendelenburg posi-tion can be used. With the patient in this head-down
Figure 14. Femoral vascular anatomy illustrating that the femoral nerve is lateral, while the FV is medial to the femoral artery; top of the figureis cephalad.
SPECIAL ARTICLE
58 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
position, the sterile probe is placed transverse to the neck,creating a cross-sectional view of the vessels. The right IJvein should lie lateral to the right CA and be easilycompressible by the ultrasound probe (Figure 4). The neckshould be scanned with ultrasound to identify the accesspoint that is most conducive to cannulation of the vein,while avoiding the artery. This may or may not be the samepoint identified with landmarks alone. The probe shouldalso be positioned to allow the needle to enter at an angleaway from the carotid. Shorter needles and a more superiorentry point may reduce the risk for pleural or great vesselpuncture, which is a particularly important concern forpediatric patients.
The cannulating needle or catheter should be observedentering the vessel. The technique of observing vessel entryin real time is critical to avoid the complications associatedwith the landmark-guided techniques. The guide wire isinserted using the Seldinger technique, and its presencewithin the vein lumen and its absence within the artery areconfirmed in two image planes, as demonstrated in Figure15, before dilation and placing the central venous catheter.
10.1.2. Femoral VesselsBoth the FV and the femoral artery are frequently used inneonates for access during cardiac procedures. Hopkins etal.55 recently defined the anatomic relationship between thecommon femoral artery and common FV when a child isplaced in a frog-leg position versus a straight-leg positionduring attempted cannulation. They demonstrated that theFV was overlapped by the fem-oral artery in 36% ofpatients in the straight-leg position and in 45% of patientsin the frog-leg position, at the level of the inguinal liga-ment. The frequency of overlap increased as the vesselswere imaged more distally. At 3 cm from the inguinalligament, the incidence of overlap was 93% and 86% in thestraight-leg and frog-leg positions, respectively. This sig-nificant overlap provides credibility to the routine use ofultrasound guidance for cannulation, as vessel overlap mayincrease the risk for complications and is not predictablewith surface landmarks alone. Hip rotation with 60° legabduction decreases fem-oral artery overlap at the level ofthe inguinal crease in both infants and children. Thus, the
optimal place for FV cannulation in pediatric patientsseems to be at the level of the inguinal crease, with 60° legabduction and external hip rotation.71 Another frequentproblem encountered in neonates is the high incidence ofvenous and arterial thrombosis (Figure 16) when multiplecardiac catheterization procedures have been performed.
Visualization of the FV in small neonates is improvedwith several useful maneuvers. First, a small towel or sheetis placed under the child’s buttock; second, the child isplaced in the reverse Trendelenburg position; and finally,abdominal compression to further expand the vein can beused if necessary. A high-resolution linear-array probe ismost frequently used for optimal imaging. Figure 17 dem-onstrates the common anatomy and small vessels encoun-tered among neonates. Because the vein is more superficialin children, it is important to direct the needle at an angleof �30° with the skin when attempting cannulation inpediatric patients.
Many studies have shown a clear advantage of ultrasoundguidance over landmark guidance for FV cannulation. Aouadet al.72 prospectively randomized 48 patients undergoing FVcannulation with a landmark-guided technique compared
Figure 15. The guide wire (arrow) is demonstrated entering the right IJ vein, in SAX (A) and LAX (B) views.
Figure 16. Left FV completely occluded by thrombus (arrow).
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 59
with real-time ultrasound and demonstrated a shorter time tocomplete cannulation with ultra-sound (155 [46–690] vs 370[45–1620] sec, P � .02). The ultrasound group requiredfewer needle passes (1 [1–8] vs 3 [1–21], P � .001) tosuccessful cannulation and had a greater number of suc-cessful cannulations performed on the first needle pass (18[75%] vs 6 [25%], P � .001) compared with the landmarkgroup. The overall success rate was similar in both groups(95.8%), and the incidence of femoral artery puncture wascomparable.72
In another prospective randomized study, Iwashima etal.63 showed no difference in the overall rate of success ofachieving FV access between a landmark-guided approachand an ultrasound approach for pediatric cardiac catheter-ization. The success rate, defined as achieving access withinthe first two attempts without femoral artery puncture, wassimilar in both groups (67.4% for ultrasound guidance vs59.1% for landmark guidance). The procedure time was notsignificantly different between groups. There were two FVocclusions detected in the ultrasound group in patientswith prior vascular entry. In addition, there was a signifi-cant reduction in the complication rate with the use ofultrasound guidance. Unintentional femoral artery punc-ture occurred in three of 43 patients (7%) in the ultrasoundgroup compared with 14 of 44 patients (31.8%) in thelandmark group, for a significantly higher complicationrate in the landmark group (P � .01).63
10.2. Recommendations for Pediatric PatientsIt is the recommendation of this writing committee that trainedclinicians use real-time ultrasound during IJ cannulation when-ever possible to improve cannulation success and reduce theincidence of complications associated with the insertion oflarge-bore catheters in pediatric patients. This recommenda-tion is based on category A, level 1 supportive literature.It is also the recommendation of this council that trainedclinicians use real-time ultrasound during FV cannulationwhenever possible to improve cannulation success and reducethe incidence of complications associated with insertion of
large-bore catheters in pediatric patients. This is a categoryC, level 2 recommendation.
11. ULTRASOUND-GUIDEDARTERIAL CANNULATIONArterial access is an important aspect of vascular access andincludes the radial, brachial, axillary, femoral, and dorsalispedis arteries. The preferred site depends on the experienceof the operator, availability of the site, and expectedduration of access. The advantages of the radial artery areits accessibility, predictable location, and low complicationrates associated with both its access and use. It is usuallypalpable among most patients with a pulsatile circulation.Another advantage to using the radial artery as the cannu-lation site is that this artery is not the sole blood supply tothe distal extremity,73 unlike the axillary, brachial, andfemoral arteries. Ultrasound guidance for arterial cannula-tion improved success and reduced time to cannulationcompared with the palpation method in a prospectivecomparison of ultrasound-guided and blindly placed radialarterial catheters.74
Ultrasound can facilitate access to all these arteriesbut is particularly useful in patients with obesity, alteredanatomy, low perfusion, nonpulsatile blood flow, andpreviously unsuccessful cannulation attempts using alandmark-guided approach.75 Ultrasound-guided arte-rial access can be performed at the traditional locationsused with landmark-guided approaches but has the addedadvantage of allowing the use of nontraditional sites of entrywhere landmarks are not useful. Ultrasound-guided place-ment for femoral artery catheters is more challenging than FVcannulation because the artery is smaller and not amendableto expansion with positioning or volume loading.
Hypotension, low cardiac output, absent or barely pal-pable arterial pulse, the presence of arterial spasm or hema-toma, and excessive limb circumference are reasons for failureor repeated attempts at arterial cannulation of different sites(radial, brachial, axillary, femoral, dorsalis pedis) when usingthe palpation or external landmark–based approach.76 It
Figure 17. (A) Left femoral vessels in a 3-kg neonate demonstrating the small size of the FV (V) and femoral artery (A). (B) Color flow Dopplerdemonstrating a velocity signal in the femoral artery (A) and FV (V). The sonographer is positioned at the patient’s lower extremities and facingcephalad. Note depth scale along side of image: 2.0 cm for 2D image and 1.5 cm for color Doppler image.
SPECIAL ARTICLE
60 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
should be noted, however, that these cannulation condi-tions may be equally challenging despite ultrasound guid-ance, because the application of the probe may compressvenous structures in the hypovolemic patient.
11.1. Cannulation TechniqueAs described in detail in the sections describing the tech-nique of venous structure cannulation, arteries appear to bepulsatile on 2D echocardiography and are not fully com-pressible with external pressure from the transducer (Fig-ures 18A and 18D). The addition of a color flow Dopplersector should demonstrate phasic blood flow in eitherthe SAX (Figure 18B) or the LAX orientation (Figure 18E).The placement of a small sample volume (�0.5 cm)within the lumen of the artery will demonstrate a typicalsystolic-diastolic pattern of arterial blood flow (Figures 18Cand 18F). Scanning both arteries before attempted cannu-lation should identify the artery with the largest diameter.The real-time guided insertion of the catheter (with orwithout a guide wire) is preferred over a skin-markingstatic imaging technique. The transducer is placed inside asterile sheath, and the operator should observe sterile oraseptic technique. The nondominant hand of the operatorholds the ultrasound transducer, while the dominant handholds the arterial catheter. The catheter-needle system isinserted at an angle of 45° to the skin and is advanced
under ultra-sound guidance until it is observed enteringthe vessel, in either the SAX or the LAX view. The catheteris inserted over the needle or over a guide wire.
11.2. Ultrasound-Guided Arterial CannulationVersus PalpationThe first-attempt success rate during arterial cannulation ishigher when using ultrasound-guided approach comparedwith palpation alone. In either the emergency room or theoperating theater setting, the success rate for theultrasound-guided approach is in the range of 62% to 87%in adults (compared with 34%–50% for palpation)74,77,78
and 14% to 67% in the pediatric population (compared with14%–20% for palpation).79,80 A recent meta-analysis thatincluded four controlled trials of radial artery cannulationwith a total of 311 adult and pediatric patients demon-strated an overall 71% improvement in first-attempt suc-cess (relative risk, 1.71; 95% confidence interval [CI],1.25–2.32).81
Seto et al.64 randomized 1,004 patients undergoing ret-rograde femoral artery cannulation to either fluoroscopic orultrasound guidance. There was no difference in the pri-mary end point, with similar common femoral arterycannulation rates with either ultrasound or fluoroscopicguidance (86.4% vs 83.3%, P � .17). The exception was inthe subgroup of patients with common femoral artery
Figure 18. Surface ultrasound imaging of the radial artery with a small probe. Two-dimensional ultrasound in SAX (A) and LAX (D) orientations.Doppler ultrasound demonstrating systolic flow with color Doppler (B, E) and pulsed-wave Doppler in SAX (C) and LAX (F) orientations,respectively.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 61
bifurcations occurring over the femoral head (82.6% vs69.8%, P � .01). Ultrasound guidance resulted in an im-proved first-pass success rate (83% vs 46%, P � .0001), areduced number of attempts (1.3 vs 3.0, P � .0001), reducedrisk for venipuncture (2.4% vs 15.8%, P � .0001), andreduced median time to access (136 vs 148 sec, P � .003).Vascular complications occurred in seven of 503 patients inthe ultrasound group and 17 of 501 in the fluoroscopygroup (1.4% vs 3.4%, P � .04). Thus, ultrasound guidanceimproved common fem-oral artery cannulation rate only inthe subset of patients with high common femoral arterybifurcations but reduced the vascular complications infemoral arterial access.
11.3. Recommendation for ArterialVascular AccessAlthough ultrasound may identify the presence, location,and patency of arteries suitable for cannulation or vascularaccess, the council does not recommend routine real-time ultra-sound use for arterial cannulation in general. However, forradial artery cannulation, there is category A, level 1support for the use of ultrasound to improve first-passsuccess.81
Ultrasound is most effectively used as a rescue tech-nique for arterial access and to identify the location andpatency of suitable arteries for cannulation or proceduralaccess. LAX imaging is particularly useful to identify vesseltortuosity, atheromatous plaques, and difficulties withcatheter insertion.
12. ULTRASOUND-GUIDED PERIPHERALVENOUS CANNULATIONPeripheral venous access is usually performed by cannu-lating super-ficial veins that are directly visualized withinthe dermis. Intravenous access can be difficult in obesepatients, chronic intravenous drug abusers, edematouspatients, and long-term hospitalized patients. Ultrasoundfacilitates access to anatomically deeper veins not directlyor easily visible within the dermis. Reports in both theanesthesiology and emergency medicine literature describethe use of ultrasound to facilitate peripheral access in thesedifficult patient populations.82,83 Keyes et al.82 had successusing a SAX view but encountered some arterial puncturesand early catheter failures due to catheter dislodgement.Sandhu and Sidhu83 advocated a LAX ultrasound approachand the placement of a catheter with $2.5 cm of catheter in thevein. Placement of a shorter catheter should be converted to aSeldinger technique to minimize inadvertent dislodgement.
A follow-up study of emergency medicine physiciansplacing peripheral intravenous catheters in difficult-accesspatients compared the use of real-time ultrasound guidancewith traditional approaches of palpation and landmarkguidance.84 Cannulation was more successful for the ultra-sound group (97%) than the control group (33%). Theultrasonographic group required less overall time (13 vs 30min, for a difference of 17 min [95% CI, 0.8–25.6 min]), lesstime to successful cannulation from first percutaneouspuncture (4 vs 15 min, for a difference of 11 min [95% CI,8.2–19.4 min]), and fewer percutaneous punctures (1.7 vs3.7, for a difference of 2.0 [95% CI, 1.27–2.82]) and hadgreater patient satisfaction (8.7 vs 5.7, for a difference of 3.0
[95% CI, 1.82–4.29]) than the traditional landmark ap-proach. It is important to note that all sonographers in thisseries participated in 15 hours of didactic lectures related toultrasound and 100 ultrasound exams during their trainingor practice.84 Another prospective emergency medicinestudy did not demonstrate a decrease in the number ofattempts or the time to successful cannulation with ultra-sound or improved patient satisfaction compared with thenonultrasound group.85 A comparison of skin markingwith static imaging to real-time ultrasound for peripheralvein cannulation in an adult patient population did notdemonstrate improve success rates but decreased the timeto successful cannulation when ultrasound was used in realtime.86
Percutaneous intravenous central catheterization (PICC)is a similar but distinct procedure and patient subset. PICClines are placed for long-term intravenous access for anti-biotic or chemotherapy administration or in long-termacute care patients in need of intravenous access. Venogra-phy was the standard access method before two reportsdescribing the use of ultrasound. Sofocleous et al.87 pro-moted the use of sonography over venography for centralaccess with a series of 355 patients and a 99% success rate.Parkinson et al.88 described the success of ultrasoundversus blind cannulation at their facility, with 100% successfor ultrasound-guided versus 82% for blind procedures.Robinson et al.89 showed that a dedicated PICC team, usingultra-sound guidance, increased the success rate from 73%to 94%, reduced the wait time for a catheter and overallplacement costs, and reduced the overall usage of cathetersby disapproving inappropriate requests.
12.1. Recommendation for PeripheralVenous AccessAlthough ultrasound may identify the presence, location,and patency of peripheral veins, the council does not recom-mend routine real-time ultrasound use for peripheral venouscannulation, although there is category B, level 2 (suggestiveobservational studies) support for the use of ultrasound forPICC insertion. Ultrasound is most effectively used toidentify the location and patency of suitable veins forperipherally inserted central venous catheters.
13. VESSEL SELECTIONThe benefit of ultrasound guidance for improving cannu-lation success and reducing complications varies accordingto the site selected. The risk for thrombosis and infectionalso varies according to the access site chosen for cannula-tion and is an important consideration when choosing aparticular site. Prior cannulation and radiation exposureare specific to the affected area. Femoral access has thehighest incidence of infection and thrombosis at 19.8% and21.5%, respectively.4 It is often used for emergent accessand secondary line access. The infection rate for the IJ veinranges between 4% and 8.6%, and the thrombosis rate is7.6%. SC access is favored for longer dwelling catheters,with the lowest infection rate (1.5%–4%) and thrombosisrate (1.2%–1.9%). The risk for infection and thrombosis fortunneled IJ vein catheters is similar to that of tunneled SCcatheters.60 Larger catheters with more lumens are associ-ated a higher risk for infection.90 Patient factors such as
SPECIAL ARTICLE
62 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
thrombocytopenia, obesity, chronic obstructive pulmonarydisease, myocardial infarction, sepsis, and malnutritionincrease the risk for infection at all access sites. Hyperco-agulation disorders such as heparin-induced thrombocyto-penia with thrombosis and factor V Leiden, catheter length,inability to anticoagulate, malignancy, and the duration ofcatheter indwelling increase the risk for thrombosis.91
14. VASCULAR ACCESS CONFIRMATIONThe complications arising from the incorrect cannulation ofan artery with a large bore catheter intended for an adjacentvein have signifi-cant morbidity and mortality. This isparticularly true for unintentional CA cannulation during IJvein cannulation attempts but also holds true for uninten-tional arterial puncture at other sites. Ultrasound reliablydetects the guide wire in the target vessel before dilationand catheter insertion92 but is not a substitute for roentgen-ography to verify catheter location and course or toidentify complications such as pneumothorax or hemo-thorax. Other confirmation techniques of central venouscannulation and wire passage via the Seldinger tech-nique include fluoroscopy, visualization of the wire withtransesophageal echocardiography in the superior venacava or inferior vena cava, manometry with a fluidcolumn connected to a catheter, blood gas analysis, anddirect pressure transduction.
14.1. Recommendations for VascularAccess ConfirmationThe council recommends that real-time ultrasound be used forconfirmation of successful vessel cannulation. It is vitally impor-tant for the guide wire to be visualized in the target vessel andthat the adjacent structures be visualized to confirm the absenceof the guide wire. Because there may be ambiguity of the guidewire tip with SAX ultrasound imaging alone, manometry with afluid-filled catheter through a flexible catheter in the vessel isrecommended when LAX imaging is not used for confirmation ofvenous catheter placement.93 When available, transesopha-geal echocardiographic or fluoroscopic imaging of theguide wire in the superior vena cava or inferior vena cavaprovides definitive confirmation of placement into thecentral venous system (Figure 19).
15. TRAININGMultiple training techniques have recently been describedusing ultrasound for central venous cannulation.94–96 Allforms of training must emphasize the importance of devel-oping proficiency in both cognitive and psychomotor skillsets. Training must include image acquisition, interpreta-tion, real-time use of ultrasound for vessel puncture andcannulation, and an experienced instructor who demon-strates to the trainee how to translate 2D imaging toperform a 3D task. The techniques used to enhance thesafety of the procedure using landmark guidance shouldnot be abandoned during ultrasound, but rather ultrasoundimaging should enhance the safety of the techniques usedduring landmark-guided training. Comprehensive educa-tion should include a combination of didactic lectures, liveor simulated demonstrations, and mentoring by a skilledsonographer. Formal training will reduce the failure rate ofultrasound-guided cannulation and ultimately improvepatient safety.
There is a lack of scientific literature to specificallydelineate the number of procedures necessary to developcompetence in performing real-time ultrasound cannula-tion because clinicians acquire knowledge and developdexterity for the technique at different rates. The opinionamong expert users with �10 years of experience with thistechnique has suggested that training include a minimumof 10 procedures performed under the guidance of anexperienced user. It is the recommendation of this council thatindividuals gain the requisite knowledge, develop the requireddexterity, and perform 10 ultrasound-guided vascular accessprocedures under supervision to demonstrate competence toindependently practice this technique (Table 2). A portion ofthis training can also be accomplished in a simulatedenvironment that allows a trainee to develop the dexterityneeded for simultaneous probe manipulation and needleinsertion. It is preferable that training occurs at one par-ticular site, so that learning the ultrasound technique maybe a priority over learning the approach to different sites.However, once the ultrasound technique is mastered, theprinciples can be used to access vessels at other siteswithout additional ultrasound specific supervision.
Proper training that imparts the cognitive knowledgeand technical skills to perform ultrasound-guided cannula-tion is outlined Table 2. This training is necessary to realizethe clinical outcomes supported by the literature. Mostimportant, the operator must possess an appreciation of theultrasound anatomy surrounding the target vessel, theability to identify the optimal entry site and needleangulation, and an understanding of the limitations ofthe ultrasound-guided technique. The safety techniquesused for landmark-guided approaches, such as a laterallydirected needle angulation, should not be abandonedwhen ultrasound is used but rather enhanced withultrasound imaging. For example, if ultrasound imagingreveals significant vessel overlap, an entry site with amore side-by-side vessel orientation should be selectedas a direct response to the ultrasound information toenhance cannulation safety and reduce the likelihood ofcomplications.
Figure 19. Successful cannulation of the IJ vein with a guidewireshown entering the right atrium (RA) via the superior vena cava (SVC)(midesophageal bicaval view). LA, Left atrium.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 63
16. CONCLUSIONSIt is the recommendation of this council, on the basis oflevel 1 scientific evidence, that properly trained cliniciansuse real-time ultrasound during IJ cannulation wheneverpossible to improve cannulation success and reduce theincidence of complications associated with the insertion oflarge-bore catheters. Despite fewer scientific studies, thecouncil also recommends the use of real-time ultrasoundfor the cannulation of the IJ and FV in pediatric patients.Complications during FV cannulation in adults are lesssevere than those that occur with SC and IJ vein cannula-tion, and therefore, ultrasound guidance is recommendedonly for identifying vessel overlap and patency whenfeasible for FV cannulation. Obese and coagulopathic pa-tients should have ultrasound screening of the SC veinbefore attempted cannulation to identify vessel locationand patency. If real-time ultrasound is not used as theinitial technique for SC vein cannulation, it should be usedas a rescue device. It is also an effective rescue device forarterial cannulation.
Proper training is necessary to realize the clinical out-comes supported by the literature, to gain an appreciationof the ultrasound anatomy, identify the optimal entry site
and needle angle, and understand the limitations of theultrasound-guided technique. Precannulation or static ul-trasound with skin marking is useful for identifying vesselanatomy and thrombosis but may not improve cannulationsuccess or reduce complications, as does real-time ultra-sound needle guidance.
NOTICE AND DISCLAIMERThis report is made available by the ASE as a courtesyreference source for its members. This report containsrecommendations only and should not be used as the solebasis to make medical practice decisions or for disciplinaryaction against any employee. The statements and recom-mendations contained in this report are based primarily onthe opinions of experts, rather than on scientifically verifieddata. The ASE makes no express or implied warrantiesregarding the completeness or accuracy of the informationin this report, including the warranty of merchantability orfitness for a particular purpose. In no event shall the ASE beliable to you, your patients, or any other third parties forany decision made or action taken by you or such otherparties in reliance on this information. Further, the use ofthis information does not constitute the offering of medicaladvice by the ASE or create any physician-patient relation-ship between the ASE and your patients or anyone else.
APPENDIX AMembers of the Council on IntraoperativeEchocardiography
Kathryn E. Glas, MD, MBA, FASE, ChairScott T. Reeves, MD, MBA, FASE, Vice ChairStanton Shernan, MD, FASE, Immediate Past ChairMadhav Swaminathan, MD, FASE, Vice Chair ElectGregg S. Hartman, MDAlan Finley, MDMarsha RoellerDavid Rubenson, MDDoug Shook, MDNikolaos J. Skubas, MD, FASEChristopher A. Troianos, MDJennifer Walker, MDWill Whitley, MD
Members of the Council on Vascular UltrasoundRobert T. Eberhardt, MD, ChairVijay Nambi, MD, Co-ChairHeather L. Gornik, MDEmile R. Mohler, III, MDTasneem Naqvi, MD, FASEMargaret M. Park, BS, RDCS, RVT, FASEJoy R. Peterson, RVS, RVTKathleen Rosendahl-Garcia, BS, RVT, RDCS
Table 2. Recommended Training Objectives forUltrasound-Guided Vascular CannulationCognitive skills
1. Knowledge of the physical principles of ultrasound2. Knowledge of the operation of the ultrasound equipment,
including the controls that affect the imaging display3. Knowledge of infection control standards for performing
vascular access and sterile preparation of the ultrasound probefor real-time use
4. Knowledge of the surface anatomy specific to the access siteand ultrasound anatomy that allows identification of the targetvessel and structures that are to be avoided
5. Ability to recognize the location and patency of the targetvessel
6. Ability to recognize atypical anatomy of vessel location andredirect the needle entry to minimize complications
7. Knowledge of the color flow and spectral Doppler flow patternsthat identify arterial and venous flow characteristics
Technical skills1. Ability to operate the ultrasound equipment and controls to
produce quality information to identify the target vessel2. Dexterity to coordinate needle guidance in the desired direction
and depth on the basis of the imaging data3. Use of needle guides for coordination of needle insertion with
imaging data when operator dexterity is lacking or clinicalconditions make dexterity coordination challenging
4. Ability to insert the catheter into the target vessel usingultrasound information
5. Ability to confirm catheter placement into the target vessel andthe absence of the catheter in unintended vessels andstructures
SPECIAL ARTICLE
64 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
App
endi
xB
.Sum
mar
yof
Ran
dom
ized
Clin
ical
Tria
lsof
USG
Cen
tral
Ven
ous
Can
nula
tion
Stu
dySet
ting
Par
tici
pant
sC
ompa
riso
n(e
ntry
site
)O
utco
mes
mea
sure
dO
pera
tor
expe
rien
ceFi
ndin
gsM
allo
ryet
al.
(1990)1
4U
Ste
rtia
ryca
re,
teac
hing
hosp
ital
Crit
ical
lyill
adul
tpa
tient
sin
inte
nsiv
eca
re;
high
and
low
risk
(dis
ease
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
failu
reon
first
atte
mpt
Sen
ior
ICU
staf
fan
dcr
itica
lca
refe
llow
s;nu
mbe
rno
tre
port
ed;
mea
n6
yex
perie
nce
Suc
cess
100%
vs65%
Troi
anos
etal
.(1
991)1
9U
Ste
rtia
ryca
re,
teac
hing
hosp
ital
Car
diot
hora
cic
surg
ical
patie
nts
(age
,di
seas
e,an
dris
kfa
ctor
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,fa
ilure
onfir
stat
tem
pt,
num
ber
ofat
tem
pts
tosu
cces
sful
cath
eter
izat
ion,
time
tosu
cces
sful
cath
eter
izat
ion
Not
repo
rted
Suc
cess
100
vs96%
Alde
rson
etal
.(1
993)
Can
adia
nur
ban
child
ren’
sho
spita
lIn
fant
s(a
ged
�2
y)un
derg
oing
card
iac
surg
ery;
dise
ase
and
risk
not
repo
rted
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
Expe
rienc
edca
rdia
can
esth
etis
tD
eter
min
edab
norm
alan
atom
yin
18%
Soy
eret
al.
(1993)
Fren
chho
spita
lAd
ult
patie
nts
with
liver
dysf
unct
ion
requ
iring
tran
sjug
ular
liver
biop
sy(r
isk
asse
ssm
ent
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,nu
mbe
rof
atte
mpt
sto
succ
essf
ulca
thet
eriz
atio
n,tim
eto
succ
essf
ulca
thet
eriz
atio
n
2ra
diol
ogis
tsw
ithsa
me
expe
rienc
e(n
otqu
antifi
ed)
Suc
cess
100%
vs74%
Bra
nger
etal
.(1
994)
Fren
chte
achi
ngho
spita
lPa
tient
sne
edin
gce
ntra
lve
nous
cath
eter
izat
ion
for
hem
odia
lysi
s,ap
here
sis,
orpa
rent
eral
nutr
ition
(dis
ease
not
repo
rted
),lo
wris
kfo
rco
mpl
icat
ions
(hig
h-ris
kpa
tient
sex
clud
ed)
Dop
pler
US
Gvs
LMK
met
hod
(IJV
and
SC
V)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofat
tem
pts
tosu
cces
sful
cath
eter
izat
ion,
time
tosu
cces
sful
cath
eter
izat
ion
14
juni
orpo
stgr
adua
test
uden
tsw
ith�
5y
clin
ical
expe
rienc
ean
d8
seni
orst
aff
with
�5
yex
perie
nce,
from
neph
rolo
gy,
emer
genc
y,an
din
tens
ive
care
;ta
ught
the
Dop
pler
tech
niqu
eov
er2
wk,
achi
eved
�1
veno
usca
thet
eriz
atio
nbe
fore
ente
ring
stud
y
Sal
vage
of4
of12
failu
res
ofLM
Kat
tem
pts
Gra
tzet
al.
(1994)
US
tert
iary
care
,te
achi
ngho
spita
lPa
tient
sfo
rca
rdio
thor
acic
orva
scul
arsu
rger
y(a
gean
ddi
seas
eno
tre
port
ed)
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,fa
ilure
onfir
stat
tem
pt,
num
ber
ofat
tem
pts
tosu
cces
sful
cath
eter
izat
ion,
time
tosu
cces
sful
cath
eter
izat
ion
Num
ber
not
repo
rted
;“e
xper
ienc
edan
esth
esio
logi
sts”
Suc
cess
84%
vs55%
(Con
tinue
d)
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 65
App
endi
xB
.C
onti
nued
Stu
dySet
ting
Par
tici
pant
sC
ompa
riso
n(e
ntry
site
)O
utco
mes
mea
sure
dO
pera
tor
expe
rien
ceFi
ndin
gsVu
cevi
cet
al.
(19
94)
Brit
ish
hosp
ital
Car
diac
surg
ery
and
ICU
patie
nts
(age
,di
seas
e,an
dris
k-as
sess
men
tno
tre
port
ed)
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,tim
eto
succ
essf
ulca
thet
eriz
atio
n
2co
nsul
tant
anes
thet
ists
;10
proc
edur
esN
odi
ffer
ence
;sm
art
need
leav
oide
dca
rotid
punc
ture
in2
case
s
Gilb
ert
etal
.(1
99
5)
US
tert
iary
care
,te
achi
ngho
spita
lAd
ult
patie
nts
(dis
ease
not
repo
rted
)at
high
risk
from
com
plic
atio
ns(o
besi
tyor
coag
ulop
athy
)
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,n
umbe
rof
com
plic
atio
ns,
failu
reon
first
atte
mpt
,tim
eto
succ
essf
ulca
thet
eriz
atio
n
Num
ber
not
repo
rted
;ju
nior
hous
est
aff
“rel
ativ
ely
inex
perie
nced
inus
ing
eith
erte
chni
que”
Suc
cess
84.4
%vs
61.4
%;
com
plic
atio
ns2%
vs16.3
%
Gua
ltier
iet
al.
(1995)4
4U
Sur
ban
teac
hing
hosp
ital
Crit
ical
care
patie
nts
unde
rgoi
ngno
nem
erge
ncy
proc
edur
es(a
ge,
dise
ase,
and
risk
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(SC
V)N
umbe
rof
faile
dca
thet
erpl
acem
ents
;nu
mbe
rof
com
plic
atio
ns
18
phys
icia
nsw
ith�
30
proc
edur
esS
ucce
ss92%
vs44%
;co
mpl
icat
ions
sam
e
Hilt
yet
al.
(1997
)U
Sur
ban
teac
hing
hosp
ital
Patie
nts
unde
rgoi
ngca
rdio
pulm
onar
yre
susc
itatio
n(a
ge,
dise
ase,
and
risk
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(FV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
failu
reon
first
atte
mpt
,nu
mbe
rof
atte
mpt
sto
succ
essf
ulca
thet
eriz
atio
n,tim
eto
succ
essf
ulca
thet
eriz
atio
n
2em
erge
ncy
med
icin
ere
side
nts
inPG
Ys3
and
4;
15–2
0pr
oced
ures
usin
gLM
Km
etho
d;6–1
0pr
oced
ures
usin
gul
tras
onog
raph
y
Suc
cess
90%
vs65%
;co
mpl
icat
ions
0%
vs20%
Sla
ma
etal
.(1
99
7)
Fren
chun
iver
sity
hosp
ital
Adul
tsin
inte
nsiv
eca
rere
quiri
ngca
nnul
atio
nof
IJV
(dis
ease
and
risk
asse
ssm
ent
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,fa
ilure
onfir
stat
tem
pt;
time
tosu
cces
sful
cath
eter
izat
ion
Juni
orho
use
staf
f(in
tern
sor
resi
dent
s)un
der
the
dire
ctsu
perv
isio
nof
seni
orph
ysic
ian
afte
r�
3de
mon
stra
tions
byex
perie
nced
oper
ator
and
3at
tem
pts
ofrig
htIJ
Vus
ing
LMK
met
hod
Suc
cess
100%
vs76%
Teic
hgra
ber
etal
.(1
997)
Ger
man
univ
ersi
tyte
achi
ngho
spita
lPa
tient
sun
derg
oing
rout
ine
cath
eter
izat
ion
ofIJ
V(a
ge,
dise
ase,
and
risk
asse
ssm
ent
not
repo
rted
)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
Phys
icia
ns;
num
ber
and
expe
rienc
eno
tre
port
edS
ucce
ss96%
vs48%
Bol
det
al.
(1998)
US
tert
iary
care
,ou
tpat
ient
onco
logy
cent
re
Adul
tch
emot
hera
pypa
tient
s(c
ance
rty
pes
not
repo
rted
);hi
ghris
kfo
rfa
ilure
orco
mpl
icat
ions
Dop
pler
US
Gvs
LMK
met
hod
(SC
V)N
umbe
rof
faile
dca
thet
erpl
acem
ents
18
surg
ical
onco
logy
fello
ws
(PG
Y6–1
0);
inst
ruct
ion
inus
eof
smar
tne
edle
and
“dem
onst
rate
dco
mpe
tenc
e”in
use
ofD
oppl
erpr
obe
No
differ
ence (C
ontin
ued)
SPECIAL ARTICLE
66 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
App
endi
xB
.C
onti
nued
Stu
dySet
ting
Par
tici
pant
sC
ompa
riso
n(e
ntry
site
)O
utco
mes
mea
sure
dO
pera
tor
expe
rien
ceFi
ndin
gsLe
fran
tet
al.
(19
98)
Fren
chte
achi
ngho
spita
lC
ritic
ally
illad
ults
unde
rgoi
ngno
nem
erge
ncy
proc
edur
es(d
isea
sean
dris
kno
tre
port
ed)
Dop
pler
US
Gvs
LMK
met
hod
(SC
V)N
umbe
rof
faile
dca
thet
erpl
acem
ents
,nu
mbe
rof
com
plic
atio
ns,
failu
reon
first
atte
mpt
1st
aff
anes
thes
iolo
gist
,un
trai
ned
inD
oppl
ergu
idan
cebe
fore
stud
y
Suc
cess
:no
differ
ence
;co
mpl
icat
ions
5.6
%vs
16.8
%N
adig
etal
.(1
99
8)
Ger
man
teac
hing
hosp
ital
Dia
lysi
spa
tient
s(a
ge,
dise
ase,
and
risk
leve
lnot
repo
rted
)
2D
US
Gvs
2D
US
Gfo
rve
ssel
loca
tion
follo
wed
bybl
ind
veni
punc
ture
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,fa
ilure
onfir
stat
tem
pt,
time
tosu
cces
sful
cath
eter
izat
ion
Phys
icia
ns;
clin
ical
expe
rienc
e1–7
yS
ucce
ss100%
vs70%
Verg
hese
etal
.(1
999)
US
univ
ersi
tyte
achi
ngho
spita
lIn
fant
ssc
hedu
led
for
card
iova
scul
arsu
rger
y,ag
ed�
12
mo,
wei
ght
�10
kg(d
isea
sean
dris
kas
sess
men
tno
tre
port
ed)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,nu
mbe
rof
atte
mpt
sto
succ
essf
ulca
thet
eriz
atio
n,tim
eto
succ
essf
ulca
thet
eriz
atio
n
Num
ber
not
repo
rted
;bo
ard-
elig
ible
anes
thes
iafe
llow
sw
hoha
dco
mpl
eted
resi
denc
ytr
aini
ngin
anes
thes
ia
Suc
cess
100%
vs77%
;co
mpl
icat
ions
(car
otid
punc
ture
s)0%
vs25%
Sul
eket
al.
(2000)
US
univ
ersi
ty-a
ffilia
ted
hosp
ital;
oper
atin
gro
om
Adul
tpa
tient
ssc
hedu
led
for
elec
tive
abdo
min
al,
vasc
ular
,or
card
ioth
orac
icpr
oced
ures
with
gene
ral
anes
thes
iaan
dm
echa
nica
lven
tilat
ion
inw
hom
cent
ralv
enou
sca
nnul
atio
nw
ascl
inic
ally
indi
cate
d(d
isea
sean
dris
kas
sess
men
tno
tre
port
ed)
2D
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,nu
mbe
rof
atte
mpt
sto
succ
essf
ulca
thet
eriz
atio
n,tim
eto
succ
essf
ulca
thet
eriz
atio
n
Anes
thet
ist;
allo
pera
tors
expe
rienc
edin
cann
ulat
ion
ofIJ
V(�
60
cath
eter
plac
emen
ts)
with
know
nex
pert
ise
inus
eof
US
GIJ
Vte
chni
que
Suc
cess
95%
vs91%
Verg
hese
etal
.(2
000)
US
univ
ersi
tyte
achi
ngho
spita
l45
infa
nts
sche
dule
dto
unde
rgo
IJca
nnul
atio
ndu
ring
card
iac
surg
ery
(dis
ease
and
risk
asse
ssm
ent
not
repo
rted
)
2D
US
Gvs
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
Num
ber
offa
iled
cath
eter
plac
emen
ts,
num
ber
ofco
mpl
icat
ions
,tim
eto
succ
essf
ulca
thet
eriz
atio
n
Num
ber
not
repo
rted
;fe
llow
sin
pedi
atric
anes
thes
ia
Suc
cess
(ultr
asou
nd,
Dop
pler
,LM
K)
94%
,77%
,81.3
%;
com
plic
atio
ns(c
arot
idpu
nctu
re)
6%
,15%
,19%
Hay
ashi
(2002)
Uni
vers
ityho
spita
lIn
trao
pera
tive
patie
nts
unde
rge
nera
lane
sthe
sia
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
;pr
esen
ceof
resp
irato
ryju
gula
rpu
lsat
ions
used
tost
ratif
y
Suc
cess
rate
,fir
sttim
esu
cces
s,co
mpl
icat
ions
,pr
esen
ceof
jugu
lar
puls
atio
ns
Anes
thes
iolo
gist
Ifpu
lsat
ions
pres
ent,
nodi
ffer
ence
:su
cces
sra
te95.6
%vs
96.9
%,
first
atte
mpt
85.7
%vs
83.5
%;
ifno
puls
atio
ns(2
2%
),ac
cess
86.2
%vs
30.4
%,
succ
ess
rate
100%
vs78.3
%;
arte
rialp
unct
ures
0%
vs13% (C
ontin
ued)
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 67
App
endi
xB
.C
onti
nued
Stu
dySet
ting
Par
tici
pant
sC
ompa
riso
n(e
ntry
site
)O
utco
mes
mea
sure
dO
pera
tor
expe
rien
ceFi
ndin
gsB
ansa
l(2005)
Uni
vers
ityho
spita
lN
ephr
olog
ist
2D
US
Gvs
LMK
met
hod:
IJV
for
hem
odia
lysi
sca
thet
er
Suc
cess
rate
,fir
stat
tem
ptsu
cces
sra
te,
com
plic
atio
ns
Nep
hrol
ogis
tS
ucce
ssra
te100%
vs6.7
%;
first
-tim
esu
cces
s86.7
%vs
56.7
%;
adve
rse
outc
omes
0%
vs16.7
%K
arak
itsos
(2006)
Uni
vers
ityho
spita
lIC
Upa
tient
sD
oppl
erU
SG
vsLM
Km
etho
d(IJ
V)O
vera
llsu
cces
s,tim
eto
inse
rtio
n,nu
mbe
rof
atte
mpt
s,co
mpl
icat
ions
Uni
vers
ityfa
culty
mem
bers
expe
rienc
edin
both
tech
niqu
es
Suc
cess
rate
s100%
vs94%
(ultr
asou
ndvs
LMK
);ac
cess
times
17
vs44
sec;
com
plic
atio
ns4%
vs23%
;nu
mbe
rof
atte
mpt
s1.1
vs2.6
Leun
g(2
006)
Tert
iary
care
ERER
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
Suc
cess
rate
,nu
mbe
rof
atte
mpt
s,ac
cess
times
,co
mpl
icat
ions
ERph
ysic
ians
Suc
cess
rate
93.9
%vs
78.5
%(u
ltras
ound
vsLM
K);
first
atte
mpt
s82%
vs70.6
%;
acce
sstim
eno
tdi
ffer
ent;
com
plic
atio
ns4.6
%an
d16.9
%S
chw
emm
er(2
00
6)7
9U
nive
rsity
hosp
ital
Ope
ratin
gro
omTr
aditi
onal
vsU
SG
cath
eter
izat
ion
ofra
dial
arte
ryin
smal
lchi
ldre
n(�
6yr
s)
Suc
cess
rate
,nu
mbe
rof
atte
mpt
s,ef
fect
sof
posi
tioni
ng
Anes
thes
iolo
gist
sin
oper
atin
gro
omS
ucce
ss100%
vs80%
;at
tem
pts
1.3
3vs
2.3
per;
dors
iflex
ion
redu
ced
cros
s-se
ctio
nala
rea
Kor
oglu
(2006)
Uni
vers
ityho
spita
lIn
terv
entio
nalr
adio
logy
Com
bine
dre
al-ti
me
ultr
asou
ndan
dflu
oros
copy
vsLM
Kfo
rem
erge
nthe
mod
ialy
sis
cath
eter
s
Suc
cess
rate
,nu
mbe
rof
atte
mpt
s,pu
nctu
reof
back
wal
lof
vess
el
Inte
rven
tiona
lrad
iolo
gyS
ucce
ssra
te100%
vs97.5
%;
com
plic
atio
ns0%
vs14%
;at
tem
pts
not
differ
ent
Hos
okaw
aet
al.
(2008)
Uni
vers
ityho
spita
lIn
fant
sw
eigh
ing
�7.5
kgU
SG
skin
mar
king
vsre
al-ti
me
cann
ulat
ion
Tim
esto
punc
ture
and
cath
eter
izat
ion,
num
ber
ofat
tem
pts,
com
plic
atio
ns
Uni
vers
ityfa
culty
mem
bers
Rea
l-tim
eca
nnul
atio
nim
prov
edsp
eed
topu
nctu
rean
dca
thet
eriz
atio
n,nu
mbe
rof
atte
mpt
s;1
arte
rialp
unct
ure
inm
arki
nggr
oup
(Con
tinue
d)
SPECIAL ARTICLE
68 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
App
endi
xB
.C
onti
nued
Stu
dySet
ting
Par
tici
pant
sC
ompa
riso
n(e
ntry
site
)O
utco
mes
mea
sure
dO
pera
tor
expe
rien
ceFi
ndin
gsTu
rker
etal
.(2
009)
Turk
ish
depa
rtm
ent
ofm
edic
ine
Spo
ntan
eous
lybr
eath
ing
patie
nts
Dop
pler
US
Gvs
LMK
met
hod
(IJV)
Ove
rall
succ
ess,
num
ber
ofat
tem
pts,
time
toca
nnul
atio
n,co
mpl
icat
ions
,ac
cess
time
Uni
vers
ityfa
culty
mem
bers
Suc
cess
97.4
%vs
99.5
%;
acce
sstim
e236
�110
vs95
�136
seco
nds;
com
plic
atio
ns8.4
2%
vs1.5
7%
;nu
mbe
rof
atte
mpt
s1.4
2vs
1.0
8Ev
ans
etal
.(2
010)
Tert
iary
teac
hing
hosp
ital
ERpa
tient
sD
idac
tics
plus
com
pete
ncy-
base
dsi
mul
atio
nvs
trad
ition
alte
achi
ng;
blin
ded
obse
rver
toou
tcom
e
Suc
cess
atfir
stat
tem
ptan
dov
eral
lcan
nula
tion
wer
epr
imar
you
tcom
es;
seco
ndar
you
tcom
esw
ere
erro
rsan
dco
mpl
icat
ions
115
resi
dent
sS
imul
atio
nim
prov
edtr
aditi
onal
:fir
stat
tem
ptO
R1.7
(95%
CI1.1
–2.8
),ov
eral
lOR
1.7
(95%
CI1.1
–2.8
)
Prab
huet
al.
(2010)6
2Te
rtia
ryte
achi
ngho
spita
lD
ialy
sis
patie
nts
2D
US
Gvs
LMK
met
hod:
FVS
ucce
ssfu
lcan
nula
tion,
num
ber
ofat
tem
pts,
com
plic
atio
ns
89.1
%fo
rLM
K,
98.2
%fo
rU
SG
;fir
st-ti
me
succ
ess
54.5
%,
85.5
%;
com
plic
atio
ns18.2
%,
5.5
%;
OR
for
succ
ess
with
US
G13.5
(95%
CI
1.7
–10.7
)M
itre
etal
.(2
010)
Rom
ania
nop
erat
ing
room
and
ICU
Hos
pita
lized
patie
nts
Dop
pler
US
Gvs
LMK
met
hod
(EJ
vein
)O
vera
llsu
cces
s,nu
mbe
rof
atte
mpt
s,tim
eto
cann
ulat
ion,
com
plic
atio
ns
Sec
ond-
year
resi
dent
sS
ucce
ssra
tes
for
punc
ture
ofEJ
:80%
and
73%
for
US
Gvs
LMK
;no
differ
ence
intim
ean
dnu
mbe
rof
atte
mpt
s;su
cces
sful
cann
ulat
ion
30%
and
20%
Set
o(2
010)6
4M
ultic
ente
rIn
terv
entio
nalr
adio
logy
patie
nts
for
retr
ogra
defe
mor
alar
tery
cann
ulat
ion
2D
ultr
asou
ndvs
fluor
osco
pyS
ucce
ssra
tes,
time
tosh
eath
inse
rtio
ns,
need
lepa
sses
,co
mpl
icat
ions
Inte
rven
tiona
lcar
diol
ogis
tsU
ltras
ound
vsflu
oros
copy
,su
cces
sra
tes:
nodi
ffer
ence
exce
ptin
popu
latio
nw
ithfe
mor
albi
furc
atio
nov
erfe
mor
alhe
ad;
first
-pas
ssu
cces
s82.7
%vs
46.4
%;
time
185
vs213
sec;
com
plic
atio
ns(a
ny)
1.4
%vs
3.4
%
EJ,
Exte
rnal
jugu
lar;
ER,
emer
genc
yro
om;
ICU
,in
tens
ive
care
unit;
IJV,
IJve
in;
LMK
,la
ndm
ark;
OR
,od
dsra
tio;
PGY,
post
grad
uate
year
;S
CV,
SC
vein
;U
SG
,ul
tras
ound
-gui
ded.
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 69
REFERENCES1. Bodenham AR. Can you justify not using ultrasound guidance
for central venous access? Critical Care 2006;10:175–62. Rothschild JM. Ultrasound guidance of central vein catheter-
ization. In: Making healthcare safer: a critical analysis ofpatient safety practices. AHRQ Publication No 01-E058. Rock-ville, MD: Agency for Healthcare Research and Quality; 2001.pp. 245–53
3. Seldinger SI. Catheter replacement of the needle in percutane-ous arteriography; a new technique. Acta Radiologica1953;39:368–76
4. McGee DC, Gould MK. Preventing complications of centralvenous catheterization. N Engl J Med 2003;348:1123–33
5. National Institute for Health and Clinical Excellence. NICETechnology Appraisal No 49: guidance on the use of ultra-sound locating devices for placing central venous catheters.Available at: http://www.nice.org.uk/nicemedia/live/11474/32461/32461.pdf. Accessed October 17, 2011
6. Dowling M, Jlala HM, Hardman JG, Bedforth NM. Real-timethree-dimensional ultrasound-guided central venous catheterplacement. Anesth Analg 2011;112:378–81
7. Cerqueira MD, Weissman NJ, Dilsizian V, Jacobs AK, Kaul S,Laskey WK, et al; American Heart Association Writing Groupon Myocardial Segmentation and Registration for CardiacImaging. Standardized myocardial segmentation and nomen-clature for tomographic imaging of the heart: a statement forhealthcare professionals from the Cardiac Imaging Committeeof the Council on Clinical Cardiology of the American HeartAssociation. Circulation 2002;29(105):539–42
8. Augoustides JG, Horak J, Ochroch AE, Vernick WJ, GamboneAJ, Weiner J, et al. A randomized controlled clinical trial ofreal-time needle-guided ultrasound for internal jugular venouscannulation in a large university anesthesia department. J Car-diothorac Vasc Anesth 2005;19:310–5
9. Beach ML, Spence BD, Sites BD. A needle guide device is better thana free hand technique for ultrasound guided cannulation of theinternal jugular vein: results from a simulation study. Internet J MedSimul. Available at: http://www.ispub.com/journal/the_internet_journal_of_medical_simulation/volume_2_number_2_62/article/a_needle_guide_device_is_better_than_a_free_hand_technique_for_ultrasound_guided_cannulation_of_the_internal_jugular_vein_results_from_a_simulation_study.html, 2009;2. Accessed October 9,2011
10. Blaivas M, Brannam L, Fernandez E. Short-axis versus long-axis approaches for teaching ultrasound-guided vascular ac-cess on a new inanimate model. Acad Emerg Med2003;10:1307–11
11. Phelan M, Hagerty D. The oblique view: an alternative ap-proach for ultrasound-guided central line placement. J EmergMed 2009;37:403–8
12. Milling TJ Jr, Rose J, Briggs WM, Birkhahn R, Gaeta TJ, Bove JJ,et al. Randomized, controlled clinical trial of point-of-carelimited ultrasonography assistance of central venous cannula-tion: the Third Sonography Outcomes Assessment Program(SOAP-3) trial. Crit Care Med 2005;33:1764–9
13. Hosokawa K, Shime N, Kato Y, Hashimoto S. A randomizedtrial of ultra-sound image-based skin surface marking versusreal-time ultrasound-guided internal jugular vein catheteriza-tion in infants. Anesthesiology 2007;107:720–4
14. Mallory DL, Shawker T, Evans RG, McGee WT, Brenner M,Parker M. Effects of clinical maneuvers on sonographicallydetermined internal jugular vein size during venous cannula-tion. Crit Care Med 1990;18:1269–73
15. Armstrong PJ, Sutherland R, Scott DH. The effect of positionand different manoeuvres on internal jugular vein diametersize. Acta Anaesthesiol Scand 1994;38:229–31
16. Denys BG, Uretsky BF. Anatomical variations of internaljugular vein location: impact on central venous access. CritCare Med 1991;19:1516–9
17. Sznajder JI, Zveibil FR, Bitterman H, Weiner P, Bursztein S.Central vein catheterization: failure and complication ratesby three percutaneous approaches. Arch Intern Med1986;146:259 – 61
18. Gallieni M. Central venous catheterization of dialysis patients.J Vasc Access 2000;1:10–4
19. Troianos CA, Jobes DR, Ellison N. Ultrasound-guided cannu-lation of the internal jugular vein. A prospective, randomizedstudy. Anesth Analg 1991;72:823–6
20. Denys BG, Uretsky BF, Reddy PS. Ultrasound-assisted cannu-lation of the internal jugular vein—a prospective comparisonto the external landmark guided technique. Circulation1993;87:1557–62
21. Karakitsos D, Labropoulos N, De Groot E, Patrianakos A,Kouraklis G, Poularas J, et al. Real-time ultrasound-guidedcatheterisation of the internal jugular vein: a prospectivecomparison with the landmark technique in critical care pa-tients. Crit Care 2006;10:R162
22. Hind D, Calvert N, McWilliams R, Davidson A, Paisley S,Beverley C, et al. Ultrasonic locating devices for central venouscannulation: meta-analysis. BMJ 2003;327:361–8
23. Troianos CA, Kuwik RJ, Pasqual JR, Lim AJ, Odasso DP.Internal jugular vein and carotid artery anatomic relation asdetermined by ultrasonography. Anesthesiology 1996;85:43–8
24. Gordon AC, Saliken JC, Johns D, Owen R, Gray R. US-guidedpuncture of the internal jugular vein: complications and ana-tomic considerations. J Vasc Interv Radiol 1998;9:333–8
25. Sulek CA, Gravenstein N, Blackshear RH, Weiss L. Headrotation during internal jugular vein cannulation and the riskof CA puncture. Anesth Analg 1996;82:125–8
26. Blaivas M, Adhikari S. An unseen danger: frequency of poste-rior vessel wall penetration by needles during attempts toplace internal jugular vein central catheters during ultrasoundguidance. Crit Care Med 2009;37:2345–9
27. Lieberman JA, Williams KA, Rosenberg AL. Optimal headrotation for internal jugular vein cannulation when relying onexternal landmarks. Anesth Analg 2004;99:982–8
28. Docktor BL, Sadler DJ, Gray RR, Saliken JC, So CB. Radiologicplacement of tunneled central catheters: rates of success and ofimmediate complications in a large series. Am J Roentgenol1999;173:457–60
29. Wang R, Snoey ER, Clements RC, Hern G, Price D. Effect ofhead rotation on vascular anatomy of the neck: an ultrasoundstudy. J Emerg Med 2006;31:283–6
30. Turba UC, Uflacker R, Hannegan C, Selby JB. Anatomicrelationship of the internal jugular vein and the commoncarotid artery applied to percutaneous transjugular proce-dures. Cardiovasc Intervent Radiol 2005;28:303–6
31. Benter T, Teichgraber UK, Kluhs L, Papadopoulos S, KohneCH, Felix R, et al. Anatomical variations in the internal jugularveins of cancer patients affecting central venous access. Ana-tomical variation of the internal jugular vein. Ultraschall Med2001;22:23–6
32. Mey U, Glasmacher A, Hahn C, Gorschluter M, Ziske C,Mergelsbery M, et al. Evaluation of an ultrasound-guidedtechnique for central venous access via the internal jugularvein in 493 patients. Support Care Cancer 2003;11:148–55
33. Lichtenstein D, Saifi R, Augarde R, Prin S, Schmitt JM, Page B,et al. The internal jugular veins are asymmetric. Usefulness ofultrasound before catheterization. Intensive Care Med2001;27:301–5
34. Lobato EB, Sulek CA, Moody RL, Morey TE. Cross-sectionalarea of the right and left internal jugular veins. J CardiothoracVasc Anesth 1999;13:136–8
35. Khatri VP, Wagner-Sevy S, Espinosa MH, Fisher JB. Theinternal jugular vein maintains its regional anatomy andpatency after carotid endarterectomy: a prospective study.Ann Surg 2001;233:282–6
36. Botha R, Van Schoor AN, Boon JM, Becker JM, Meiring JH.Anatomical considerations of the anterior approach for centralvenous catheter placement. Clin Anat 2006;19:101–5
37. Ruesch S, Walder B, Tramer M. Complications of centralvenous catheters: Internal jugular versus subclavian access—asystematic review. Crit Care Med 2002;30:454–60
38. Beaulieu Y, Marik PE. Bedside ultrasonography in the ICU.Part 2. Chest 2005;128:1766–81
SPECIAL ARTICLE
70 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
39. Giacomini M, Iapichino G, Armani S, Cozzolino M, BrancaccioD, Gallieni M. How to avoid and manage a pneumothorax.J Vasc Access 2006;7:7–14
40. Tercan F, Ozkan U, Oguzkurt L. US-guided placement ofcentral vein catheters in patients with disorders of hemostasis.Eur J Radiol 2008;65:253–6
41. Oguzkurt L, Tercan F, Kara G, Torun D, Kizilkilic O, YildirimT. US-guided placement of temporary internal jugular veincatheters: immediate technical success and complications innormal and high-risk patients. Eur J Radiol 2005;55:125–9
42. Goetz AM, Wagener MM, Miller JM, Muder RR. Risk ofinfection due to central venous catheters: effect of site ofplacement and catheter type. Infect Control Hosp Epidemiol1998;19:842–5
43. Tau BK, Hong SW, Lee ST. Anatomic Basis of Safe Percutane-ous Subclavian venous catheterization. J Trauma Injury InfectCrit Care 2000;48:82–6
44. Gualtieri E, Deppe SA, Sipperly ME, Thompson DR. Subcla-vian venous catheterization: greater success for less experi-enced operators using ultra-sound guidance. Crit Care Med1995;23:692–7
45. Balls A, Lovecchio F, Kroeger A, Stapczynski JS, Mulrow M,Drachman D, et al. Ultrasound guidance for central venouscatheter placement: results from the central line emergencyaccess registry database. Am J Emerg Med 2010;28:561–7
46. Mansfield PF, Hohn DC, Fornage BD, Gregurich MA, Ota DM.Complications and failures of subclavian-vein catheterization.N Engl J Med 1994;331:1735–8
47. Kilbourne MJ, Bochicchio GV, Scalea T, Xiao Y. Avoidingcommon technical errors in subclavian central venous catheterplacement. J Am Coll Surg 2009;208:104–9
48. Orsi F, Grasso RF, Arnaldi P, Bonifacio C, Biff R, DeBraud F, etal. Ultra-sound guided versus direct vein puncture in centralvenous port placement. J Vasc Access 2000;1:73–7
49. Orihashi K, Imai K, Sato K, Hamamoto M, Okada K, Sueda T,et al. Extra-thoracic subclavian venipuncture under ultrasoundguidance. Circ J 2005;69:1111–5
50. Dailey RH. Femoral vein cannulation: a review. J Emerg Med1985;2:367–72
51. Hughes P, Scott C, Bodenham A. Ultrasonography of thefemoral vessels in the groin: implications for vascular access.Anaesthesia 2000;55:1198–202
52. Warkentine FH, Pierce MC, Lorenz D, Kim IK. The anatomicrelationship of femoral vein to femoral artery in euvolemicpediatric patients by ultra-sonography: implications for pedi-atric femoral central venous access. Acad Emerg Med2008;15:426–30
53. Lechner G, Jantsch H, Waneck R, Kretschmer G. The relation-ship between the common femoral artery, the inguinal crease,and the inguinal ligament: a guide to accurate angiographicpuncture. Cardiovasc Intervent Radiol 1988;11:165–9
54. Werner SL, Jones RA, Emerman CL. Effect of hip abductionand external rotation on femoral vein exposure for possiblecannulation. J Emerg Med 2008;35:73–5
55. Hopkins JW, Warkentine F, Gracely E, Kim IK. The anatomicrelationship between the femoral artery and common femoralvein in frog leg position versus straight leg position in pediat-ric patients. Acad Emerg Med 2009;16:579–84
56. Stone MB, Price DD, Anderson BS. Ultrasonographic investi-gation of the effect of reverse Trendelenburg on the cross-sectional area of the femoral vein. J Emerg Med 2006;30:211–3
57. Abboud PA, Kendall JL. Ultrasound guidance for vascularaccess. Emerg Med Clin North Am 2004;22:749–73
58. Eisen LA, Narasimhan M, Berger JS, Mayo PH, Rosen MJ,Schneider RF. Mechanical complications of central venouscatheters. J Intensive Care Med 2006;21:40–6
59. Karapinar B, Cura A. Complications of central venous cathe-terization in critically ill children. Pediatr Int 2007;49:593–9
60. Parienti JJ, Thirion M, M egarbane B, Souweine B, Ouchikhe A,Polito A, et al. Femoral vs jugular venous catheterization andrisk of nosocomial events in adults requiring acute renalreplacement therapy: a randomized controlled trial. JAMA2008;299:2413–22
61. Deshpande KS, Hatem C, Ulrich HL, Currie BP, Aldrich TK,Bryan-Brown CW, et al. The incidence of infectious complica-tions of central venous catheters at the subclavian, internaljugular, and femoral sites in an intensive care unit population.Crit Care Med 2005;33:13–20
62. Prabhu MV, Juneja D, Gopal PB, Sathyanarayanan M, Subhra-manyam S, Gandhe S, et al. Ultrasound-guided femoral dialy-sis access placement: a single-center randomized trial. ClinJ Am Soc Nephrol 2010;5:235–9
63. Iwashima S, Ishikawa T, Ohzeki T. Ultrasound-guided versuslandmark-guided femoral vein access in pediatric cardiaccatheterization. Pediatr Cardiol 2008;29:339–42
64. Seto AH, Abu-Fadel MS, Sparling JM, Zacharias SJ, Daly TS,Harrison AT, et al. Real-time ultrasound guidance facilitatesfemoral arterial access and reduces vascular complications:FAUST (Femoral Arterial Access With Ultrasound Trial). J AmColl Cardiol Intv 2010;3:751–8
65. Grebenik CR, Boyce A, Sinclair ME, Evans RD, Mason DG,Martin B. NICE guidelines for central venous catheterization inchildren: is the evidence base sufficient? Br J Anaesth2004;92:827–30
66. Tovey G, Stokes M. A survey of the use of 2D ultrasoundguidance for insertion of central venous catheters by UKconsultant paediatric anaesthetists. Eur J Anaesthesiol2007;24:71–5
67. Howard S. A survey measuring the impact of NICE guideline49: the use of ultrasound locating devices for placing centralvenous catheters. Available at: http://www.nice.org.uk
68. Pirotte T. Ultrasound-guided vascular access in adults andchildren: beyond the internal jugular vein puncture. ActaAnaesth Belg 2008;59:157–66
69. Hanslik A, Thom K, Haumer M, Kitzmuller E, Albinni S,Wolfsberger M, et al. Incidence and diagnosis of thrombosis inchildren with short-term central venous lines of the uppervenous system. Pediatrics 2008;122:1284–91
70. Sigaut S, Skhiri A, Stany I, Golmar, Nivoche Y, Constant I, et al.Ultrasound guided internal jugular vein access in children andinfant: a meta-analysis of published studies. Paediatr Anaesth2009;19:1199–206
71. Suk EH, Lee KY, Kweon TD, Jang YH, Bai SJ. Ultrasonographicevaluation of the femoral vein in anaesthetised infants andyoung children. Anaesthesia 2010;65:895–8
72. Aouad MT, Kanazi GE, Abdallah FW, Moukaddem FH, Tur-bay MJ, Obeid MY, et al. Femoral vein cannulation performedby residents: a comparison between ultrasound-guided andlandmark technique in infants and children undergoing car-diac surgery. Anesth Analg 2010;111:724–8
73. Brzezinski M, Luisetti T, London MJ. Radial artery cannula-tion: a comprehensive review of recent anatomic and physi-ologic investigations. Anesth Analg 2009;109:1763–81
74. Shiver S, Blaivas M, Lyon M. A prospective comparison ofultrasound-guided and blindly placed radial arterial catheters.Acad Emerg Med 2006;13:1257–79
75. Sandhu NS, Patel B. Use of ultrasonography as a rescuetechnique for failed radial artery cannulation. J Clin Anesth2006;18:138–41
76. Shiloh AL, Eisen LA. Ultrasound-guided arterial catheteriza-tion: a narrative review. Intens Care Med 2010;36:214–21
77. Levin PD, Sheinin O, Gozal Y. Use of ultrasound guidance inthe insertion of radial artery catheters. Crit Care Med2003;31:481–4
78. Tada T, Amagasa S, Horikawa H. Usefulness of ultrasonictwo-way Doppler flow detector in routine percutaneous arte-rial puncture in patients with hemorrhagic shock. J Anesth2003;17:70–1
79. Schwemmer U, Arzet HA, Traunter H, Rausch S, Rower N,Greim CA. Ultrasound-guided arterial cannulation in infantsimproves success rate. Eur J Anaesthesiol 2006;23:476–80
80. Ganesh AG, Kaye R, Cahill AM, Stern W, Pachikara R,Gallagher PR, et al. Evaluation of ultrasound-guided radialartery cannulation in children. Pediatr Crit Care Med2009;10:45–8
Vascular Cannulation Guidelines (JASE)
January 2012 • Volume 114 • Number 1 www.anesthesia-analgesia.org 71
81. Shiloh AL, Savel RH, Paulin LM, Eisen LA. Ultrasound-guidedcatheterization of the radial artery: a systematic review andmeta-analysis of randomized controlled trials. Chest2011;139:524–9
82. Keyes LE, Frazee BW, Snoey ER, Simon BC, Christy D.Ultrasound-guided brachial and basilic vein cannulation inemergency department patients with difficult intravenous ac-cess. Ann Emerg Med 1999;34:711–4
83. Sandhu NPS, Sidhu DS. Mid-arm approach to basilic andcephalic vein cannulation using ultrasound guidance. Br JAnaesth 2004;93:292–4
84. Costantino TG, Parikh AK, Satz WA, Fojtik J.Ultrasonography-guided peripheral intravenous access versustraditional approaches in patients with difficult intravenousaccess. Ann Emerg Med 2005;46:456–61
85. Stein J, George B, River G, Hebig A, McDermott D. Ultrasono-graphically guided peripheral intravenous cannulation inemergency department patients with difficult intravenous ac-cess: a randomized trial. Ann Emerg Med 2009;54:33–40
86. Resnick JR, Cydulka RK, Donato J, Jones RA, Werner SL.Success of ultrasound-guided peripheral intravenous accesswith skin marking. Acad Emerg Med 2008;15:723–30
87. Sofocleous CT, Schur I, Cooper SG, Quintas JC, Brody L, ShelinR. Sonographically guided placement of peripherally insertedcentral venous catheters: review of 355 procedures. AJR Am JRoentgenol 1998;170:1613–6
88. Parkinson R, Gandhi M, Harper J, Archibald C. Establishing anultrasound guided peripherally inserted central catheter(PICC) insertion service. Clin Radiol 1998;53:33–6
89. Robinson MK, Mogensen KM, Grudinskas GF, Kohler S, JacobsDO. Improved care and reduced costs for patients requiringperipherally inserted central catheters: the role of bedsideultrasound and a dedicated team. JPEN J Parenter Enteral Nutr2005;5:374–9
90. Merrer J, De Jonghe B, Golliot F, Lefrant JY, Raffy B, Barre E, etal. Complications of femoral and subclavian venous catheter-ization in critically ill patients: a randomized controlled trial.JAMA 2001;286:700–7
91. Hamilton HC, Foxcroft DR. Central venous access sites for theprevention of venous thrombosis, stenosis and infection inpatients requiring long-term intravenous therapy. CochraneDatabase Syst Rev 2007;(3). CD004084
92. Stone MB, Nagdev A, Murphy MC, Sisson CA. Ultrasounddetection of guidewire position during central venous cathe-terization. Am J Emerg Med 2010;28:82–4
93. Ezaru CS, Mangione MP, Oravitz TM, Ibinson JW, Bjerke RJ.Eliminating arterial injury during central venous catheteriza-tion using manometry. Anesth Analg 2009;109:130–4
94. Feller-Kopmann D. Ultrasound-guided internal jugular accessa proposed standardized approach and implications for train-ing and practice. Chest 2007;132:302–9
95. Skippen P, Kissoon N. Ultrasound guidance for central vascu-lar access in the pediatric emergency department. PediatrEmerg Care 2007;23:203–7
96. Chen J, Lee S, Huynh T, Bandiera G. Procedures can be learnedon the Web: a randomized study of ultrasound-guided vascu-lar access training. Acad Emerg Med 2008;15:949–54
SPECIAL ARTICLE
72 www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA