Central Venous Catheter Placement

Central Venous Pressure

Central venous pressure (CVP) is the intravascular pressure in the great thoracic veins, measured relative to atmospheric pressure. It is conventionally measured at the junction of the superior vena cava and the right atrium and provides an estimate of the right atrial pressure. The CVP is often used as a marker of volemic status or preload, although the ability of this measurement to provide this information is limited.

The CVP is influenced by the volume of blood in the central venous compartment and also the compliance of that compartment ( Box 60-1-1 ). Starling demonstrated the relationships between CVP and ventricular contraction and Guyton the relationship between venous return and CVP. By plotting the two relationships on the same set of axes, it can be seen that the “ventricular function curve” and the “venous return curve” intersect at only one point, demonstrating that if all other factors remain constant in an individual patient, a given CVP can, at equilibrium, be associated with only one possible cardiac output ( Figure 60-1-1 ). Both curves can of course be affected by a number of factors: total blood volume and distribution of that blood volume between the different vascular compartments (determined by vascular tone). The inotropic state of the right ventricle will affect the shape of the ventricular function curve. When any one of these factors is altered, there will be an imbalance between cardiac output and venous return that will persist for a short time until a new equilibrium is reached at a new central venous blood volume and/or an altered central venous vascular tone.

Normal CVP exhibits a complex waveform, illustrated in Figure 60-1-2 . The a wave corresponds to atrial contraction and the x descent to atrial relaxation. The c wave that punctuates the x descent is caused by the closure of the tricuspid valve at the start of ventricular systole and the bulging of its leaflets back into the atrium. The v wave is due to continued venous return in the presence of a closed tricuspid valve. The y descent occurs at the end of ventricular systole when the tricuspid valve opens and blood once again flows from the atrium into the ventricle. This normal CVP waveform may be modified by a number of pathologic processes ( Box 60-1-2 ).

If the CVP is to be used as an index of cardiac preload, the end-diastolic pressure at end expiration must be identified. The c wave marks the closure of the tricuspid valve at the beginning of ventricular systole, and immediately before its onset, the measured pressure should be equivalent to the right ventricular end-diastolic pressure (except in the case of tricuspid stenosis, in which a pressure gradient will always exist between the two chambers). Where no c wave is clearly visible, it is conventional to take the average pressure during the a wave. Where no a wave is visible (e.g., in atrial fibrillation), the pressure at the Z point (that point on the CVP waveform that corresponds with the end of the QRS complex on the electrocardiogram) should be used.

Taking all these factors into account, it is perhaps not surprising that the CVP will not provide a reliable estimate of preload in critically ill patients. The CVP correlates poorly with overall volemic status, right ventricular end-diastolic volume, stroke index, or an individual patient's response to a fluid challenge. It is perhaps best used in non-critically ill patients when it can provide an estimate of the components to right ventricular filling and venous return because their vasculature is behaving in a normal physiologic manner.

Introduction

The Agency for Healthcare Research and Quality, in its 2001 report Making Health Care Safer: A Critical Analysis of Patient Safety Practices , recommended the use of ultrasound for the placement of all central venous catheters as one of its 11 practices aimed at improving patient care. The purpose of this document is to provide comprehensive practice guidelines on the use of ultrasound for vascular cannulation. Recommendations are made for ultrasound-guided central venous access of the internal jugular (IJ) vein, subclavian (SC) vein, and femoral vein (FV) on the basis of the strength of the scientific evidence present in the literature ( Table 60-2-1 ). The role of ultrasound for vascular cannulation of pediatric patients is discussed specifically, and the use of ultrasound to facilitate arterial cannulation and peripheral venous access is also discussed. Recommendations are made for training, including the role of simulation.

Methodology and Evidence Review

The writing committee conducted a comprehensive search of medical and scientific literature in the English language through the use of PubMed and MEDLINE. Original research studies relevant to ultrasound-guided vascular access published in peer-reviewed scientific journals from 1990 to 2011 were reviewed using the Medical Subject Headings terms “ultrasonography,” “catheterization-central venous/adverse effects/methods,” “catheterization-peripheral,” “jugular veins,” “subclavian vein,” “femoral vein,” “artery,” “adult,” “pediatric,” “randomized controlled trials,” and “meta-analysis.” The committee reviewed the scientific evidence for the strength of the recommendation (i.e., risk/benefit ratio) as supportive evidence (category A), suggestive evidence (category B), equivocal evidence (category C), or insufficient evidence (category D). The weight or “level” of evidence was assigned within each category ( Table 60-2-1 ). Recommendations for the use of ultrasound were based on supportive literature (category A) with a level 1 weight of scientific evidence (multiple randomized controlled trials with the aggregated findings supported by meta-analysis). The document was reviewed by 10 reviewers nominated by the American Society of Echocardiography (ASE) and the Society of Cardiovascular Anesthesiologists and approved for publication by the governing bodies of these organizations.

Ultrasound-Guided Vascular Cannulation

Ultrasonography was introduced into clinical practice in the early 1970s and is currently used for a variety of clinical indications. Miniaturization and advancements in computer technology have made ultrasound affordable, portable, and capable of high-resolution imaging of both tissue and blood flow.

Cannulation of veins and arteries is an important aspect of patient care for the administration of fluids and medications and for monitoring purposes. The practice of using surface anatomy and palpation to identify target vessels before cannulation attempts (“landmark technique”) is based on the presumed location of the vessel, the identification of surface or skin anatomic landmarks, and blind insertion of the needle until blood is aspirated. Confirmation of successful cannulation of the intended vascular structure relies on blood aspiration of a certain character and color (i.e., the lack of pulsation and “dark” color when cannulating a vein or pulsation and a “bright” red color when cannulating an artery), pressure measurement with a fluid column or pressure transducer, or observation of the intraluminal pressure waveform on a monitor. Although vascular catheters are commonly inserted over a wire or metal introducer, some clinicians initially cannulate the vessel with a small caliber (“finder”) needle before the insertion of a larger bore needle. This technique is most beneficial for nonultrasound techniques, because a smaller needle may minimize the magnitude of an unintended injury to surrounding structures. The vessel is then cannulated with a larger bore 16-gauge or 18-gauge catheter, a guide wire is passed through it, and a larger catheter is inserted over the wire. The catheter–over–guide wire process is termed the Seldinger technique.

Although frequently performed and an inherent part of medical training and practice, the insertion of vascular catheters is associated with complications. Depending on the site and patient population, landmark techniques for vascular cannulation are associated with a 60% to 95% success rate. A 2003 estimate cited the insertion of >5 million central venous catheters (in the IJ, SC, and FV) annually in the United States alone, with a mechanical complication rate of 5% to 19%. These complications may occur more often with less experienced operators, challenging patient anatomy (obesity, cachexia, distorted, tortuous or thrombosed vascular anatomy, congenital anomalies such as persistent left superior vena cava), compromised procedural settings (mechanical ventilation or emergency), and the presence of comorbidity (coagulopathy, emphysema). Central venous catheter mechanical complications include arterial puncture, hematoma, hemothorax, pneumothorax, arterial-venous fistula, venous air embolism, nerve injury, thoracic duct injury (left side only), intraluminal dissection, and puncture of the aorta. The most common complications of IJ vein cannulation are arterial puncture and hematoma. The most common complication of SC vein cannulation is pneumothorax. The incidence of mechanical complications increases sixfold when more than three attempts are made by the same operator. The use of ultrasound imaging before or during vascular cannulation greatly improves first-pass success and reduces complications. Practice recommendations for the use of ultrasound for vascular cannulation have emerged from numerous specialties, governmental agencies such as the National Institute for Health and Clinical Excellence and the Agency for Healthcare Research and Quality's evidence report.

Ultrasound Principles for Needle-Guided Catheter Placement

Ultrasound modalities used for imaging vascular structures and surrounding anatomy include two-dimensional (2D) ultrasound, Doppler color flow, and spectral Doppler interrogation. The operator must have an understanding of probe orientation, image display, the physics of ultrasound, and mechanisms of image generation and artifacts and be able to interpret 2D images of vascular lumens of interest and surrounding structures. The technique also requires the acquisition of the necessary hand-eye coordination to direct probe and needle manipulation according to the image display. The supplemental use of color flow Doppler to confirm presence and direction of blood flow requires an understanding of the mechanisms and limitations of Doppler color flow analysis and display. This skill set must then be paired with manual dexterity to perform the three-dimensional (3D) task of placing a catheter into the target vessel 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 its advantage or disadvantage in terms of directing the cannulating needle at the correct entry angle and depth. Three-dimensional ultrasound may circumvent the spatial limitations of 2D imaging by providing simultaneous real-time SAX and LAX views along with volume perspective without altering transducer location, allowing simultaneous views of neck anatomy in three orthogonal planes. Detailed knowledge of vascular anatomy in the region of interest is similarly vital to both achieving success and avoiding complications from cannulation of incorrect vessels.

Ultrasound probes used for vascular access vary in size and shape. Probes with smaller footprints are preferred in pediatric patients. Higher frequency probes (≥7 MHz) are preferred over lower frequency probes (<5 MHz) because they provide better resolution of superficial structures in close proximity to the skin surface. The poorer penetration of the high-frequency probes is not typically a hindrance, because most target vascular structures intended for cannulation are <8 to 10 cm from the skin surface.

It is important to appreciate how probe orientation relates to the image display. Conventions established by the ASE for performing transthoracic imaging of the heart, and more recently epicardial imaging, established that the probe indicator and right side of the display should be oriented toward the patient's left side or cephalad. In these settings, projected images correlate best with those visualized by the sonographer positioned on the patient's left side and facing the patient's right shoulder. In contrast, the operator's position during ultrasound-guided vascular access varies according to the target vessel. For example, the operator is typically positioned superior to the patient's head and faces caudally during cannulation of the IJ vein. The left side of the screen displays structures toward the patient's left side ( Figure 60-2-1 ). In contrast, during cannulation of the FVs, the operator is typically positioned inferiorly and faces cephalad, so that the left side of the screen displays structures toward the patient's right side ( “Femoral Vein Cannulation” ). For SC vein cannulation, the left and right sides of the screen display cephalad and caudad structures, depending on laterality (right or left). The changing image orientation is an important distinction from typical transthoracic, epicardial, or transesophageal imaging. For ultrasound-guided vascular access cannulation, the probe and screen display are best oriented to display the anatomic cross-section that would be visible from the same vantage point. Therefore, screen left and right will not follow standard conventions but rather vary with site and needle insertion orientation. What is common for all vascular access sites is that it is essential for the operator to orient the probe so that structures beneath the left aspect of the probe appear on the left side of the imaging screen. Although probes usually have markings that distinguish one particular side of the transducer, the operator must identify which aspect of the screen corresponds to the marking on the probe. These markings may be obscure, and a recommended practice is to move the probe toward one direction or another while observing the screen or apply modest external surface pressure on one side of the transducer to demonstrate proper alignment of left-right probe orientation 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). Some probes allow the use of a needle guide, which directs the needle into the imaging plane and defined depth as viewed on the display screen ( Figure 60-2-2 ). Needle guides are not available from every ultrasound probe manufacturer, but a needle guide may be a useful feature for the beginner who has not yet developed the manual dexterity of using a 2D image display to perform a 3D task. One study that evaluated ultrasound-guided cannulation of the IJ vein with and without a needle guide showed that its use significantly enhanced cannulation success after first (68.9%–80.9%, P = .0054) and second (80.0%–93.1%, P = .0001) needle passes. Cumulative cannulation success after seven needle passes was 100%, regardless of technique. The needle guide specifically improved first-pass success among more junior operators (65.6%–79.8%, P = .0144), while arterial puncture averaged 4.2%, regardless of technique ( P > .05) or operator ( P > .05). A limitation of the needle guide is that the needle trajectory is limited to orthogonal orientations from the SAX imaging plane. Although helpful in limiting lateral diversion of the needle path, sometimes oblique angulation of the needle path may facilitate target vessel cannulation. In addition, there may be considerable costs associated with the use of needle guides. Depending on the manufacturer, they may cost as little as several dollars to >$100 each. Importantly, although the needle guide facilitated prompt cannulation with ultrasound in the novice operator, it offered no additional protection against arterial puncture. However, one in vitro simulation study has refuted these in vivo results.

Arterial puncture during attempted venous cannulation with ultrasound generally occurs because of a misalignment between the needle and imaging screen. It may also occur as a result of a through-and-through puncture of the vein into a posteriorly positioned artery. The first scenario is due to improper direction of the needle, while the latter occurs because of a lack of needle depth control. Needle depth control is also an important consideration because the anatomy may change as the needle is advanced deeper within the site of vascular access. The ideal probe should have a guide that not only directs the needle to the center of the probe but also directs the needle at the appropriate angle beneath the probe ( Figure 60-2-2 ). This type of guide compensates for the limitation of using 2D ultrasound to perform a 3D task of vascular access. The more experienced operator with a better understanding of these principles and better manual dexterity may find the needle guide cumbersome, choosing instead the “maneuverability” of a freehand technique. Although the routine use of a needle guide requires further study, novice operators are more likely to improve their first-pass success.

Vascular structures can be imaged in SAX, LAX, or oblique orientation ( Figures 60-2-3A, 60-2-3B, and 60-2-3C ). The advantage of the SAX view is better visualization of surrounding structures and their relative positions to the needle. There is usually an artery in close anatomic proximity to most central veins. Identification of both vascular structures is paramount to avoid unintentional cannulation of the artery. In addition, it may be easier to direct the cannulating needle toward the target vessel and coincidentally away from surrounding structures when both are clearly imaged simultaneously. The advantage of the LAX view is better visualization of the needle throughout its course and depth of insertion, because more of the needle shaft and tip are imaged within the ultrasound image plane throughout its advancement, thereby avoiding insertion of the needle beyond the target vessel. A prospective, randomized observational study of emergency medicine residents evaluated whether the SAX or LAX ultrasound approach resulted in faster vascular access for novice ultrasound users. The SAX approach yielded a faster cannulation time compared with the LAX approach, and the novice operators perceived the SAX approach as easier to use than the LAX approach. The operator's hand-eye coordination skill in aligning the ultrasound probe and needle is probably the most important variable influencing needle and target visibility. Imaging in the SAX view enables the simultaneous visualization of the needle shaft and adjacent structures, but this view does not image the entire needle pathway or provide an appreciation of insertion depth. Although novice users may find ultrasound guidance easier to adopt using SAX imaging, ultrasound guidance with LAX imaging should be promoted, because it enables visualization of the entire needle and depth of insertion, thereby considering anatomic variations along the needle trajectory as the needle is advanced deeper within the site of vascular access. The oblique axis is another option that may allow better visualization of the needle shaft and tip and offers the safety of imaging surrounding structures in the same view, thus capitalizing on the strengths of both the SAX and LAX approaches.

Real-Time Imaging Versus Static Imaging

Ultrasound guidance for vascular access is most effective when used in real time (during needle advancement) with a sterile technique that includes sterile gel and sterile probe covers. The needle is observed on the image display and simultaneously directed toward the target vessel, away from surrounding structures, and advanced to an appropriate depth. Static ultrasound imaging uses ultrasound imaging to identify the site of needle entry on the skin over the underlying vessel and offers the appeal of nonsterile imaging, which obviates the need for sterile probe coverings, sterile ultrasound gels, and needle guides. If ultrasound is used to mark the skin for subsequent cannulation without real-time (dynamic) use, ultrasound becomes a vessel locator technique that enhances external landmarks rather than a technique that guides the needle into the vessel. Both static and real-time ultrasound-guided approaches are superior to a traditional landmark-guided approach. Although the real-time ultrasound guidance outperforms the static skin-marking ultrasound approach, complication rates are similar.

Venous puncture using real-time ultrasound was faster and required fewer needle passes among neonates and infants randomly assigned to real-time ultrasound-assisted IJ venous catheterization versus ultrasound-guided skin marking. Fewer than three attempts were made in 100% of patients in the real-time group, compared with 74% of patients in the skin-marking group ( P < .01). In this study, a hematoma and an arterial puncture occurred in one patient each in the skin-marking group.

One operator can usually perform real-time ultrasound-guided cannulation. The nondominant hand holds the ultrasound probe while the dominant hand controls the needle. Successful cannulation of the vessel is confirmed by direct vision of the needle entering the vessel and with blood entering the attached syringe during aspiration. The probe is set aside on the sterile field, the syringe removed, and the wire is inserted through the needle. Further confirmation of successful cannulation occurs with ultrasound visualization of the guide wire in the vessel. Difficult catheterization may benefit from a second person with sterile gloves and gown assisting the primary operator by either holding the transducer or passing the guide wire.