Central Venous Cannulation Should be Performed Using Ultrasound Guidance


There is no question that vascular ultrasound imaging can elucidate difficulties in obtaining central vascular access and may even facilitate the process for difficult patients (Table 1). However, its routine use is unnecessary, and not supported by available data.

The incidence of complications and difficulties during central venous cannulation (CVC), using surface landmarks as a guide, varies from study to study, but is relatively low and probably related to operator experience. The more commonly presented and studied complications include pneumothorax, arterial puncture, prolonged procedural time, and failure to obtain access (Table 2).1-8 In a systematic review of 17 prospective studies using surface landmarks including 2085 internal jugular cannulations and 2428 subclavian cannuluations, the incidence of carotid artery punctures and pneumothoraces ranged from 0-7% and from 0-4.5% respectively.9 The highest incidences were reported when house officers or trainees performed the procedure during urgent or emergent settings. In a large series (n=6245) of attempted CVCs the incidence of carotid artery puncture was 0.9% (120 patients), and pneumothorax was 0.5% (31 patients) .10 Of 1000 patients with known coagulopathy undergoing transvenous liver biopsy there were only 7 unsuccessful attempts (0.8%).11 In this high risk group there were only 10 hematomas reported, and only one required surgical drainage.11 In this case the hematoma was caused by venous bleeding in a patient with a goiter. Other large series report a 95% success rate for internal jugular access (n=2305) and a 74% success rate for external jugular (n=1086).12 The lower value for external jugular attempts was due to difficulty in advancing the guide wire past the clavicle and not due to accessing the external jugular vein. The incidence of requiring multiple needle sticks or advances (> 2) has been reported to be as low as 5.1% (27/500).13

A number of studies have compared the use of vascular ultrasound to surface landmarks to guide CVC.1-8 These studies report a greater percentage of successful cannulation, fewer needle punctures, decreased time required for cannulation, and fewer carotid artery punctures. However, two of these studies did not demonstrate a significant difference between landmark and ultrasound in the incidence of carotid artery puncture.2,3 Several showed no difference in the time required to achieve cannulation,1,3 and one showed no difference in the percent of patients cannulated on the first attempt.2

Several ultrasound studies report a significant decrease in time required for cannulation of the vein.1-8 With the exception of one study, the authors uniformly report the data as the mean, range, and standard deviation. Consistently the standard deviation is much larger reflecting a greater spread of data points. Since CVC was uncomplicated in the majority of cases it is likely that the distribution of data points (times) did not make a normal bell shaped curve and contains significant outliers (ones with very prolonged time for cannulation). These data would be better represented by a median value. If this were done I suspect that the difference between the two techniques would have been less. Furthermore, the reported time differences between landmark and ultrasound only range from 4 to 140 seconds. The clinical significance of these time differences is very small…..if they even exist at all.

Verghese et al reported the median values for cannulation in infants. In this study, the median value for ultrasound guided technique was 4.2 minutes compared to 14 minutes for landmark guided technique.7 These data are significantly prolonged for both study groups when compared to Alderson et al in their study of pediatric patients.8 Verghese et al studied infants who were less than 1 year old compared to Alderson et al who studied children younger than 6 years old.7,8 It is

likely that neonates and infants represent a high risk population for difficult cannulation, and may benefit from ultrasound guided imaging.

Cannulation time in these studies did not include preparation time, but instead measured the time from the initial needle entry into the skin to either aspiration of blood or cannulation of the vein. This does not reflect the amount of time required to prepare the ultrasound probe, and the time needed to visualize the vascular structures. It is important to include these times since a finite period of time is required to prepare the ultrasound device and the vascular structures may be difficult to visualize is as many as 12% of cases.14 Inclusion of preparation time would likely result in ultrasound guided procedures requiring more time than landmark guided CVCs.

Similarly, the number of needle sticks or advances required for cannulation is presented as the mean value, which may not be as accurate a reflector as the median value. On average the difference in number of needle advances was between 1 and 2, which has minimal clinical significance. These studies report a range of successful cannulation on the first attempt from 26-54% with landmark guidance and from 43-78% for ultrasound guided procedures.1-8 These data are inferior to that reported by Dee Boyd et al in which CVC using landmark guidance was successful on the first attempt in 95%, suggesting that operator experience contributed more significantly than the guidance technique used.1-8,13

It is interesting that the ultrasound studies tended to report a higher failure rate for cannulation using landmark guided procedures than previously reported series.11,12 Failure to cannulate ranged from 4 to 50% with landmark guidance, and from 0 to 11% for ultrasound guided procedures.1-8 This likely represents inadequate operator experience, and small numbers of patients studied. At most, failure rate of CVC should not exceed 5% in unselected patients.11,12

While ultrasound imaging is not routinely necessary, there is a definite niche for these devices. Complicated patients (Table 1) may require significantly greater effort to perform CVC. For these patients, especially with a previous history of difficult CVC, an ultrasound device is likely to be beneficial.

Another niche for ultrasound imaging is for teaching. The newer technology may help teach us to better perform CVC using the older technology landmark. Learning the effects of head rotation, valvsalva, Trendelenberg, and/or external compression on vessel position and caliber will improve the success rate with any technique. With the head in neutral position or 30 degrees rotation, the internal jugular was anterior and lateral to the carotid artery in 183/200 (92%) of patients.15 In 5 patients (2.5%) no internal jugular vein was visualized suggesting the increased likelihood that the vessel was thrombosed. All had had previous procedures or the internal jugular vein. Six patients (3%) had a small internal jugular vein (< 0.5 cm) which did not enlarge with valvsalva or Trendelenberg position suggesting scarring of the surrounding tissue or the vein itself. Again, all six of these patients had previous procedures. Finally, in 2 patients (1%) the IJ was significantly lateral to the CA, and in 4 patients (2%) the IJ was more medial and overlapped the carotid artery more than would have been suspected. These data may explain why difficulties accessing the internal jugular occurs in approximately 5% of patients. They also demonstrate that previous central venous procedures increase the difficulty in obtaining future access at the same site.

Other studies showed variations in the relationship of the internal jugular and the carotid artery with changes in head position.14,16 When the head is rotated to the contralateral side, at least 50% of the cases show overalapping of the carotid artery and internal jugular vein by > 50%. In 6.4% the internal jugular vein was significantly lateral and did not overlap at all. Sulek et al reported minimal overlap when the head was in neutral position or rotated to 40 degrees to the contralateral side.16 However, with greater head rotation (>40 degrees) the internal jugular vein tended to lie over the carotid artery. This occurred more frequently on the left side.

These data help to explain the difficulties encountered in attempted central venous access in a small number of patients. The data also remind us that significant head rotation (>40 degrees) and/or extension will not only alter the relative position of the major vascular branches but also increase the chance of compressing the vein.

In the studies comparing ultrasound to landmark, the approach to internal jugular cannulation proceeded from the apex of the middle cervical triangle (formed by the anterior and posterior bellies of the sternocleidomastoid (SCM) muscle.1-8 It is possible that a greater success may be achieved with another approach such as the anterior approach in which the needle entry is anterior to the anterior SCM muscle and directed laterally. Using this approach, rotation of the head is minimized and vascular relations and caliber are left unchanged.

There is a finite cost for the use of ultrasound imaging. Current ultrasound machines range from several thousand to greater than 100 thousand dollars which may not include the costs of the transducer, technical maintenance, and costs (time and/or money) necessary to train operators to become proficient at using ultrasound. There is also the need for assistance during the ultrasound guided procedure to either help pass the probe in a sterile manner or for assistance in imaging the vascular structures if the operator is inexperienced. To date, there is no data showing cost savings over a large series of patients when using the ultrasound device.

Routine use of ultrasound technology may create an unwanted dependency on this technology. If this were the case then the ultrasound machine and probe would need to be readily available twenty-four hours per day. Our efforts would be better directed at educating ourselves and our students at being proficient using landmark guided techniques. Although its routine use is not necessary, ultrasound technology would be a useful part of our armamentarium for these difficult cases.

Conclusion: As discussed above, the significance of any clinical benefits of routine Uultrasound imaging have not been demonstrated. Furthermore, given the time to prepare the ultrasound equipment, the total procedural time may be longer when compared to landmark guided CVC. Finally, there is still a finite failure and complication rate with the ultrasound device, which may not be different than that reported by experienced technicians using landmark guidance for CVC. While ultrasound may help facilitate CVC in difficult cases, its routine use is not justified.

Table 1: List of complications during central venous cannulation

Multiple sticks
Arterial punctures
Brachial Plexus injury
Failure to Cannulate

Table 2: List of risk factors predicting difficulty in obtaining central venous access.

Previous central venous catheter at same site
Clavicular injury
Previous sternotomy
Venous thrombosis at site
Restless Patient
Morbidly obese
Lack of experience
Scarred Vessel


  1. Randolph AG, Cook DJ, Gonzales CA, Pribble CG: Ultrasound guidance for placement of central venous catheters: A meta-analysis of the literature. Crit Care Med 1996;24:2053-2058.
  2. Slama M, Novara A, Safavian A, Ossart M, Safar M, Fagon J-Y: Improvement of internal jugular vein cannulation using an ultrasound-guided technique. Crit Care Med 1997;23:916-919.
  3. Armstrong PJ, Cullen M, Scott DHT: The "SiteRite" ultrasound machine-an aid to internal jugular vein cannulation. Anaesthesia 1993;48:319-323.
  4. Denys BG, Uretsky BF, Reddy PS: Ultrasound-assisted cannulation of the internal jugular vein. A prospective comparison to the external landmark-guided technique. Circulation 1993;87:1557-1562.
  5. Koski EMJ, Suhonen M, Mattila MAK: Ultrasound-facilitated central venous cannulation. Crit Care Med 1992;20:424-426.
  6. Troianos CA, Jobes DR, Ellison N: Ultrasound-guided cannulation of the internal jugular vein. A prospective, randomized study. Anesth Analg 1991;72:823-826.
  7. Verghese ST, McGill WA, Patel RI, Sell JE, Midgley FM, Ruttimann UE: Ultrasound-guided internal jugular venous cannulation in infants. Anesthesiology 1999;91:71-77.
  8. Alderson PJ, Burrows FA, Stemp LI, Holtby HM: Use of ultrasound to evaluate internal jugular vein anatomy and to facilitate central venous cannulation in paediatric patients. Br J Anaesth 1003;70:145-148.
  9. Ruesch S, Walder B, Tramer MR: Complications of central venous catheters: Internal jugular versus subclavian access- A systematic review. Crit Care Med 2002;30:454-460.
  10. Shah KB, Rao TLK, Laughlin S, El-Etr AA: A review of pulmonary artery catheterization in 6,245 patients. Anesthesiology 1984;61:271-275.
  11. Goldfarb G, Lebrec D: Percutaneous cannulation of the internal jugular vein in patients with coagulopathies: An experience based on 1,000 attempts. Anesthesiology 1982;56:321-323.
  12. Jobes DR, Schwartz AJ, Greenhow E, Stephenson LW, Ellison N: Safer jugular vein cannulation: Recognition of arterial puncture and preferential use of the external jugular route. Anesthesiology 1985;59:353-355.
  13. Dee Boyd K, Thomas ST, Gold J, Boyd AD: A prospective study of complications of pulmonary artery catheterizations in 500 consecutive patients. Chest 1983;84:245-249.
  14. Troianos CA, Kuwik RJ, Pasqual JR, Lim AJ, Odasso DP: Internal jugular vein and carotid artery anatomic relation as determined by ultrasonography. Anesthesiology 1996; 85:43-48.
  15. Denys BG, Uretsky BF: Anatomical variations of internal jugular vein location: Impact on central venous access. Crit Care Med 1991;19:1516-1519.
  16. Sulek CA, Gravenstein N, Blackshear RH, Weiss L: Head rotation during internal jugular vein cannulation and the risk of carotid artery puncture. Anesth Analg 1996;82:125-128.

Andrew D. Maslow, MD
Rhode Island Hospital
Providence, RI

Contains 0 items
Subtotal: $0.00