|
|
NewsletterOctober 2002 Newsletter Ultrasound-Guided Central Venous Cannulation
William Vernick, MD and John G. Augoustides, MD
Central venous catheter placements number approximately 3 million annually (1). Most catheters are inserted percutaneously using the anatomic landmark technique. In 1984, two-dimensional ultrasound and Doppler techniques were first reported for internal jugular vein (IJV) cannulation (2). The recent development of portable lightweight ultrasound machines designed specifically for central venous cannulation has made them practical for routine clinical use. Two different ultrasound techniques can be employed. In the first technique, ultrasound is used to locate the vessel relative to anatomic surface landmarks followed by percutaneous cannulation. The second technique involves imaging the great vessels during cannulation. This allows for visualizing any distortion of the vein during needle pressure on the skin as the needle indents and enters the vein. This second technique requires a sterile sheath to cover the ultrasound probe, gel to maintain acoustic coupling, and an assistant to place the probe into the sterile sheath. Success of internal jugular cannulation using the anatomic landmark technique has been found to be around 95% (3). In a study of 1,125 patients by Schwartz et al., cannulation was achieved in 95.3% using the landmark technique (4). Goldfarb and Lebrec reported a success rate of 99.3% in 1,000 patients using the landmark technique but 4.3% required left IJV cannulation following multiple unsuccessful attempts on the right (5). While these high rates of success were impressive, they often required multiple attempts and were associated with a significant complication rate. In Schwartz's study, there was a carotid artery puncture rate of 4.2% with 5 patients having an 8F sheath placed in the carotid artery (4). Only 57.3% of patients in Goldfarb and Lebrec's study were cannulated on the first or second attempt and the carotid puncture rate was 7.4%, hematoma rate was 1%, hemothorax rate was 0.2%, Horners' syndrome rate was 0.2%, and dysphagia rate was 0.1% (5). Sznajder et al. showed that complication rates for the anatomic landmark technique between house officer vs. attendings for arterial puncture was 6.7% vs. 3.3%, for pneumothorax was 1.7% vs. 0.8%, and for hematoma was 2.6% vs. 1.1% (6). Factors that may complicate cannulation using the anatomic landmark technique include obesity, neck deformity or rigidity, previous surgery at the cannulation site, IJV thrombosis, hypovolemia, or the inability to lie flat. Using ultrasound, Denys and Uretsky found that 8.5% of 200 patients had abnormal IJV anatomy with a small fixed IJV in 3%, no right IJV at all in 2.5%, an IJV medial to the carotid in 2%, and an IJV lateral to the carotid with no overlap in 1% (7). These common anatomic variants may explain the complication rate and need for multiple attempts to achieve success using the landmark technique alone.
Given the added cost associated with ultrasound for central venous cannulation, studies were necessary to justify its routine use. Troianos et al. in 1991 compared real-time ultrasound to the landmark technique for IJV cannulation (8). The ultrasound group had a success rate of 100% with 56 of 77 (73%) cannulations on the first attempt and an average time from local anesthetic until IJV entry of 61+/- 46 seconds. In the landmark group, 80/ 83 (96.4%) cannulation attempts were successful, but in only 45/ 83 (54%) was cannulation achieved on the first attempt. The incidence of carotid puncture was 1/ 77 (1.3%) in the ultrasound group and 7/ 83 (8.4%) in the landmark group. Of the 3 patients who could not be cannulated using external landmarks, two of the 3 were cannulated on the first attempt with ultrasound and the third patient was cannulated via the external jugular vein. Hatfield and Bodenham studied the use of ultrasound in patients who were predicted to be difficult to cannulate (9). They chose 33 patients, 23 of whom had previous cannulation failure or complications from previous cannulation attempts and the other ten were simply predicted to be difficult to cannulate. In 16 of these 23 difficult patients, ultrasound imaging demonstrated an anatomical reason for the difficulty, including the presence of a small vein or venous thrombosis. Using real-time ultrasound guidance, cannulation was successful in 22 of 22 (100%) attempted with 20 of 22 (90.9%) being successful on the first needle pass and the remaining two successful on the second pass. Armstrong et al. used real-time ultrasound to delineate vessel anatomy compared to using only anatomic landmarks in 115 patients. They found an increase in speed of cannulation, a decrease in the number of attempts, and a reduced failure rate after defining the vessel anatomy with ultrasound, but found no difference in the carotid artery puncture rate. Inability to visualize the catheter entering the vein and going through the vein into the underlying artery may explain the reason for arterial puncture using this variation of the ultrasound technique. The application of ultrasound for subclavian vein cannulation was demonstrated by Gualtieri et al, in 52 ICU patients having subclavian catheters placed by first or second year house officers (11). Ultrasound was used in 25 patients and success was achieved in 23 patients with an average of 1.4 venipunctures per patient. There was one arterial puncture in this group. In the landmark group, cannulation was achieved in 12 of 27 patients with an average venipuncture rate of 2.5 per patient. There were 3 arterial punctures, 5 hematomas, and 3 catheter malpositions in this group. Of the 15 patients who could not be cannulated by the landmark technique, 12 of them were cannulated using ultrasound. Of note, the landmark group required 40% more catheter kits then the ultrasound group. A meta-analysis in 1996 evaluated 8 randomized controlled trials involving 513 patients from 1989 to 1995 comparing landmark versus real-time ultrasound for central vein cannulation (12). Complications were defined as arterial puncture, hematoma, nerve injury, pneumothorax, or catheter malposition. Placement failure was defined as unsuccessful cannulation after five attempts or more in one study, greater then 4 attempts in another, and greater then 3 in another, requiring greater then 30 minutes in another, and was undefined in the other four studies. Six studies examined IJV cannulation, one study examined subclavian vein cannulation, and one study examined both. The results of this meta-analysis found real-time ultrasound to decrease both IJV and subclavian vein cannulation failures, complications, and need for multiple attempts, but the time to achieve cannulation was not different with ultrasound. Review of published studies indicates an advantage of ultrasound for decreasing the number of cannulation attempts, the number of catheter placement failures, and the number of complications associated with IJV and subclavian vein catheterization. These advantages have been demonstrated in different patient populations and among operators with different experience levels. The question then becomes in whom should ultrasound guidance be used and how can the added cost of ultrasound equipment be justified. Some might argue that ultrasound is unneces-sary because the complication rate is still very low when using surface landmarks. In addition, if the approach to cannulation of the first vessel is unsuccessful, success will be achieved by cannulation of another central vein. To date, no study quantified the added costs of complications incurred by using the landmark technique versus the added costs of the ultrasound technique. It seems reasonable to speculate that the costs of the ultrasound equipment the added cost of sterile sheaths and gel could be recurred from the savings associated with a decreased rate of complication, decreased time to achieve successful cannulation, and decrease in the number of catheter kits used. Even though the external landmark technique is usually successful, one most also consider patient discomfort caused by multiple attempts at cannulation. The ultrasound technique also requires an assistant to place the probe in the sterile sheath, but this assistant requires only a minimal amount of training. Ultrasound guidance could be recommended for use in selected patients with known complications associated with cannulation or who may be at increased risk of complications associated with cannulation. The difficulty and risks associated with cannulation may be greater in patients with obesity, carotid artery disease, previous neck surgery, indwelling cardiac pacemakers, coagulopathy, anticoagulation or antiplatelet therapy, or the inability to lie flat. While this seems a legitimate strategy, if the fixed cost has already been made to purchase the ultrasound, then why not use an available resource on every patient given that it has been shown to be superior. An argument often cited against using the ultrasound in every patient is that the operator will lose his or her skills in landmark techniques, which could be a problem in situations when ultrasound is not available. This belief seems somewhat unreasonable, because the dexterity and knowledge required for cannulation remains virtually the same with either technique. In conclusion, the advantages of ultrasound guidance for central venous cannulation is making it more difficult to justify not using this technique in all patients.
References:
© Society of Cardiovascular Anesthesiologists Questions or comments? Please send email to webmaster@scahq.org |
