Central venous cannulation is useful in hemodynamically unstable patients who require critical care or in those undergoing major surgery.1–4 Compared with other central venous approaches, subclavian venous catheterization has some benefits, including greater patient comfort with longterm use, lower infection rate, and decreased collapsibility during hypovolemia.3–5 However, pneumothorax is comparatively frequent in subclavian central access, with an incidence ranging from 0.1% to 6%.2–6
To increase the success rates and avoid potentially dangerous adverse events, ultrasoundguided central venous puncture has become popular recently. Lefrant et al.7 proved that pulsed Doppler ultrasonographic guidance during subclavian venous cannulation reduces malpositioning of catheter tips and decreases the overall rates of complication. Realtime ultrasoundguided cannulation of the subclavian vein (SCV), compared with the landmark (anatomical) method, results in higher success rates and fewer mechanical complications in critical care patients.8 Despite these demonstrated advantages, many physicians still perform the procedure with a landmarkbased technique.9,10
The distance from the SCV to the pleura can be affected by respiration. Because full expiration diminishes the lung volume, it could reduce the likelihood of piercing the lung and prevent pneumothorax during SCV access. In addition, respiration may have an effect on the size of the SCV, which could also influence the overall success rate of the catheterization.
This study was designed to evaluate the effect of full expiration on the distance from the SCV to the pleura and on the crosssectional area (CSA) of the SCV in spontaneously breathing, healthy adult volunteers.
After obtaining the approval of our IRB and written informed consent, we enrolled 22 healthy male volunteers aged between 20 and 30 years. Subjects with chronic disease or who were using medications that might affect vascular tone, and those with a history of clavicle fracture, central venous catheterization, or lung or chest wall surgery were excluded.
Positioning and Examination Protocol
Each subject was placed supine on a horizontal table, with the head rotated 30° to the left and the shoulders arched upon a firm rolled sheet resting between the scapulae. The subject was instructed to breathe naturally, then exhale completely and hold his breath at full expiration for 10 seconds. Ultrasonographic images of the right SCV and adjacent structures were obtained throughout the entire respiration cycle and were stored on digitized videotape for subsequent offline review. After completing recording in the horizontal position, the subject was placed in 15° Trendelenburg position for 2 minutes, and then the recording was repeated.
Equipment and Ultrasound Scanning
Twodimensional ultrasound images of the SCV and the surrounding structures were obtained using an 8 to 13-MHz linear probe (12 LRS; GE Healthcare Systems, Milwaukee, WI), which was placed perpendicularly over the skin beneath the proximal part of the middle third of the clavicle (Fig. 1). Echogenic density, gain, depth, and focus were adjusted to optimize visualization of the SCV and pleura, and to get a circular or nearly circular crosssectional image of the SCV. The same practitioner performed all ultrasound examinations.
Measurements and Statistical Analysis
The recorded videotapes were reviewed and the images at the end of normal inspiration and at 5 seconds after full expiration were captured in both horizontal and Trendelenburg positions. In the freezeframe images of each subject, the distances from the center of the SCV to the pleura (SCVcenPleura distance) and those from the inferior border of the vein to the pleura (SCVinfPleura distance) were determined. The circumferences of the SCVs were outlined with an electronic marker, and the CSAs were calculated using preloaded software installed in the ultrasound machine. The sequence of ultrasound images was randomized, and an independent observer who was blinded to the study protocol examined all images and made measurements. Figure 2 shows how the distances and the CSA were measured on the representative ultrasonographic images of the SCV and the pleura of a subject in the horizontal position.
The sample size was determined based on a pilot study. With a paired t test, 20 participants were required to detect a 20% difference in the CSA with an α of 0.05 and a power of 0.8.
All results were expressed as the mean ± SD and 95% confidence interval (CI). ShapiroWilk test was used for normality of the pairwise differences. If the pairwise difference showed a skewed distribution, 99% CI was additionally reported. Data were analyzed using paired Student t test and a P value <0.05 was considered statistically significant. The 95% CI was calculated using Student 1-group t distribution from the paired Student t test. The percentage change in the CSA was determined by taking an average of each percentage change, which was a quotient of difference between 2 areas divided by an initial area. Based on previous articles in which a 20% change in the CSA was regarded to be clinically significant,11–13 we defined an increase in the SCV CSA >20% as being clinically relevant.
Patient characteristics are summarized in Table 1. Of the 22 volunteers who participated in the study, 2 were excluded because of poor image quality.
Table 2 presents the SCVcenPleura distance. The distance did not change secondary to respiration in either position. The difference was 0.005 cm (95% CI −0.04 to 0.05) and 0.02 cm (95% CI −0.005 to 0.05) in the horizontal and Trendelenburg position, respectively.
In the horizontal position, there was a small decrease in the SCVinfPleura distance at endexpiration compared with endinspiration (0.07 cm, 95% CI 0.03–0.11; P = 0.003). However, no difference in the distance was observed in the Trendelenburg position (0.02 cm, 95% CI −0.01 to 0.06) (Table 3).
In comparison with inspiration, the SCV CSA after full expiration increased by 43% (95% CI 21.8%–64.2%, 99% CI 14%–72%) in the horizontal position and by 22% (95% CI 14%–30%) in the Trendelenburg position. The CSA change in the Trendelenburg positioning was 7.5% (95% CI 2.4%–12.6%) at endexpiration and 10.9% (95% CI 3.2%–18.7%) at endinspiration, but these changes did not meet our criteria for clinical relevance (Table 4).
Full expiration during subclavian venipuncture might be expected to increase the distance between the SCV and the lung, thereby reducing the risk of pneumothorax. However, in this study, the distance from the center of the SCV to the pleura (SCVcenPleura distance) did not change after expiration in either the horizontal or the Trendelenburg position. If the pleura had moved away from the SCV because of the reduction in lung volume during expiration, the SCVcenPleura distance would have increased at the end of full expiration.
There are several possible reasons for the lack of significant change in the SCVcenPleura distance. First, the decrease in lung volume produced by expiration may not be sufficient to make a significant difference. Second, the SCV is included in a neurovascular bundle traversing a space between the clavicle and the first rib, and therefore the position of the vein remains relatively constant. In addition, below the clavicle, the pleura is behind the ribs, which have a dense collection of intercostal muscle between them. Consequently, the movement of the thoracic cavity or the chest wall caused by respiration may not have a substantial influence over the position of the SCV or the pleura.14
Our results indicate that the SCV CSA increases significantly at endexpiration as compared with endinspiration irrespective of subject position. These findings are compatible with those of the Hightower and Gooding15 trial, which reported that the anteroposterior diameter of the SCV decreases with inspiration and increases with expiration or a Valsalva maneuver. Inspiration creates a decrease in the intrathoracic pressure, increases venous return, and causes the SCV to collapse, whereas expiration increases intrathoracic pressure, reduces venous return, and distends the vein. The increased dimension of the vein with expiration must result from decreased flow into the superior vena cava (SVC) and backup into the SCV. From reports dealing with changes of the SVC or the internal jugular vein (IJV) during Valsalva maneuver,16,17 it can be inferred that compression of the SVC (an intrathoracic vein) from increased intrathoracic pressure results in resistance toward venous flow from the SCV (an extrathoracic vein) into the SVC. There may be differences in the degree of changes in the pressure or lung volume between active positive intrathoracic pressure caused by a Valsalva maneuver and a full expiration technique conducted to achieve residual volume in this study. Nevertheless, we believe the size of the SCV, another extrathoracic vein, would be influenced by the increased intrathoracic pressure in the same way, although to different degrees.
In this study, although the degree was small, the distance between the inferior wall of the SCV and pleura (SCVinfPleura distance) was reduced, rather than increased, at the end of expiration in the horizontal position. An increase in the SCV CSA during expiration is surely the main cause, because the SCVcenPleura distance remained unchanged by respiration. The diameter of the SCV increased with full expiration and, consequently, the SCVinfPleura distance decreased without a meaningful alteration in the relative position of the SCV to the pleura.
Many practitioners place patients in a slight headdown position during central venous catheterization to facilitate venous filling and avoid air embolism. Some investigations have produced evidence that Trendelenburg positioning increases CSA or diameter of the SCV and the IJV.18–21 Our results agree with these conclusions to a degree. In the present study, the SCV CSA in the Trendelenburg position was larger than that in the horizontal position during both expiration and inspiration. However, these increases were <20%, and thus, not clinically relevant. According to our data, full expiration seems to provide a more effective means of increasing the SCV CSA compared with the Trendelenburg position because it produced increases of >20% in both positions. These results have much in common with previous findings. Verghese et al.19 demonstrated that the Valsalva maneuver, which causes an increase in the intrathoracic pressure and a decrease in the venous return to the heart, produces more increase in the CSA of the IJV compared with liver compression or Trendelenburg position in mechanically ventilated infants and children. Trendelenburg positioning is less effective in children because of their small stature: large head and short legs.
This study implies that optimal conditions for SCV cannulation in conscious patients can be provided by asking the patients to breathe out fully in the headdown position. Although this method is not likely to prevent pneumothorax by increasing the distance between the SCV and the pleura, it may increase overall success rates by enlarging the CSA of the vein.
There are some limitations in the current study. First, because the ultrasound provided only singleplane images of the SCV and the pleura, our sonographic examination could not cover the complete trajectory of the needle, and therefore we cannot be certain what part of the vessel our imaginary needle would have entered. The needle entry site to the SCV was probably a proximal portion of the vein, and it could not be readily imaged with ultrasound. By placing the probe below the proximal part of the middle third of the clavicle, we gained the images of the middle or distal portion of the SCV. The vein on our image was located between the skin insertion site and the proximal portion of the SCV, where the needle might touch the pleura before reaching the proximal portion during subclavian venous catheterization. Therefore, using these images, we could partly estimate the possible risk of pneumothorax by assessing the relative position of the SCV and the pleura according to respiration. In addition, the pattern of changes of the CSA would be similar between the proximal and middle portion of the SCV because of their close proximity. If the CSA of the middle portion of the SCV is increased, the proximal one must be distended because the mechanism of increased dimension with expiration is a reduced flow into the SVC and backup into the SCV. Second, we could not confirm the hypothesis that full expiration increases the success rate of subclavian venous catheterization or decreases the risk of pneumothorax, because this study was not designed to evaluate clinical outcomes such as success or complication rates of the procedure. Further studies are needed in this regard. Third, all volunteers who participated in this study were healthy young males. Although subclavian venous cannulation is primarily applied to critically ill or hypovolemic patients, we believe our standardized results will be the basis for further investigations designed for ill patients who require central venous catheterization. Another concern is the duration of the breathholding time. Most ill patients are unable to maintain full expiration for a prolonged period. If conscious patients cannot hold their breath any longer during the venous access time, we can instruct them to breathe and then hold their breath again while stopping the advancement of the needle. This approach may be easier for ill patients, because only 5 second expiration is enough for the enlarging effect of the SCV, and prolongation of the apneic period is unnecessary after the entry of the needle into the vein. Lastly, the ability of an untrained patient to perform a complete exhalation maneuver so as to achieve a true residual volume should also be considered during the procedure.
In conclusion, the distance between the SCV and the pleura did not change after full expiration in conscious, spontaneously breathing adults. However, the increase in the CSA of the vein was clinically relevant. This simple “full expiration maneuver” can still be considered beneficial during the placement of a subclavian venous catheter. In this study, these enlarging effects on the CSA occurred in both horizontal and Trendelenburg positions.
Name: Kyung-Jee Lim, MD.
Contribution: This author helped analyze the data and prepare the manuscript.
Name: Jung-Man Lee, MD.
Contribution: This author helped analyze the data.
Name: Hyo-Jin Byon, MD.
Contribution: This author helped conduct the study.
Name: Hee-Soo Kim, MD, PhD.
Contribution: This author helped design the study.
Name: Chong-Sung Kim, MD, PhD.
Contribution: This author helped prepare the manuscript.
Name: Soo-Kyung Lee, MD, PhD.
Contribution: This author helped prepare the manuscript.
Name: Jin-Tae Kim, MD, PhD.
Contribution: This author helped design the study and conduct the study.
This manuscript was handled by: Sorin J. Brull, MD, FCARCSI (Hon).
1. Ruesch S, Walder B, Tramèr MR. Complications of central venous catheters: internal jugular versus subclavian access—a systematic review. Crit Care Med. 2002;30:454–60
2. McGee DC, Gould MK. Preventing complications of central venous catheterization. N Engl J Med. 2003;348:1123–33
3. Rebecca AS, Atilio B, Shahar B, Jonathan BM. Cardiovascular monitoring. In: Miller RD, ed. Miller’s Anesthesia. 7th ed. Philadelphia: Elsevier Churchill Livingstone. 2009:1285–328
4. Merrer J, De Jonghe B, Golliot F, Lefrant JY, Raffy B, Barre E, Rigaud JP, Casciani D, Misset B, Bosquet C, Outin H, BrunBuisson C, Nitenberg G. French Catheter Study Group in Intensive Care. Complications of femoral and subclavian venous catheterization in critically ill patients: a randomized controlled trial. JAMA. 2001;286:700–7
5. Citak A, Karaböcüo˘glu M, Uçsel R, Uzel N. Central venous catheters in pediatric patients: subclavian venous approach as the first choice. Pediatr Int. 2002;44:83–6
6. Purdue GF, Hunt JL. Placement and complications of monitoring catheters. Surg Clin North Am. 1991;71:723–31
7. Lefrant JY, Cuvillon P, Bénézet JF, Dauzat M, Peray P, Saïssi G, de La Coussaye JE, Eledjam JJ. Pulsed Doppler ultrasonography guidance for catheterization of the subclavian vein: a randomized study. Anesthesiology. 1998;88:1195–201
8. Fragou M, Gravvanis A, Dimitriou V, Papalois A, Kouraklis G, Karabinis A, Saranteas T, Poularas J, Papanikolaou J, Davlouros P, Labropoulos N, Karakitsos D.. Realtime ultrasound-guided subclavian vein cannulation versus the landmark method in critical care patients: a prospective randomized study. Crit Care Med. 2011;39:1607–12
9. Girard TD, Schectman JM. Ultrasound guidance during central venous catheterization: a survey of use by house staff physicians. J Crit Care. 2005;20:224–9
10. Bailey PL, Glance LG, Eaton MP, Parshall B, McIntosh S. A survey of the use of ultrasound during central venous catheterization. Anesth Analg. 2007;104:491–7
11. Kim JT, Kim HS, Lim YJ, Bahk JH, Lee KH, Kim CS, Kim SD, Jeon Y. The influence of passive leg elevation on the crosssectional area of the internal jugular vein and the subclavian vein in awake adults. Anaesth Intensive Care. 2008;36:65–8
12. Marcus HE, Bonkat E, Dagtekin O, Schier R, Petzke F, Wippermann J, Böttiger BW, Teschendorf P. The impact of Trendelenburg position and positive endexpiratory pressure on the internal jugular cross-sectional area. Anesth Analg. 2010;111:432–6
13. Kim JT, Park CS, Kim HJ, Lee JM, Kim HS, Kim CS, Kim SD. The effect of inguinal compression, Valsalva maneuver, and reverse Trendelenburg position on the crosssectional area of the femoral vein in children. Anesth Analg. 2009;108:1493–6
14. Lim KJ, Kim JT, Kim HS, Byon HJ, Lee SK, Lee JM. The effect of lung deflation on the position and size of the subclavian vein in mechanically ventilated infants and children. Anesth Analg. 2011;112:1448–51
15. Hightower DR, Gooding GA. Sonographic evaluation of the normal response of subclavian veins to respiratory maneuvers. Invest Radiol. 1985;20:517–20
16. Attubato MJ, Katz ES, Feit F, Bernstein N, Schwartzman D, Kronzon I. Venous changes occurring during the Valsalva maneuver: evaluation by intravascular ultrasound. Am J Cardiol. 1994;74:408–10
17. Gindea AJ, Slater J, Kronzon I. Doppler echocardiographic flow velocity measurements in the superior vena cava during the Valsalva maneuver in normal subjects. Am J Cardiol. 1990;65:1387–91
18. Fortune JB, Feustel P. Effect of patient position on size and location of the subclavian vein for percutaneous puncture. Arch Surg. 2003;138:996–1000
19. Verghese ST, Nath A, Zenger D, Patel RI, Kaplan RF, Patel KM. The effects of the simulated Valsalva maneuver, liver compression, and/or Trendelenburg position on the crosssectional area of the internal jugular vein in infants and young children. Anesth Analg. 2002;94:250–4
20. Armstrong PJ, Sutherland R, Scott DH. The effect of position and different manoeuvres on internal jugular vein diameter size. Acta Anaesthesiol Scand. 1994;38:229–31
21. Clenaghan S, McLaughlin RE, Martyn C, McGovern S, Bowra J. Relationship between Trendelenburg tilt and internal jugular vein diameter. Emerg Med J. 2005;22:867–8