1079-6061/00/4801-0082 The Journal of Trauma: Injury, Infection, and Critical Care Copyright 2000 by Lippincott Williams & Wilkins,

, Inc.

Vol. 48, No. 1 Printed in the U.S.A.

Anatomic Basis of Safe Percutaneous Subclavian Venous Catheterization

Bien-Keem Tan, MBBS(Sing), FRCS(Ed), Soo-Wan Hong, MBBS(Sing), FRCS(Glasg), FAMS(Plast Surg), Martin H. S. Huang, MBBS(Sing), FRCS(Ed), FRCS(Glasg), MMed Surg(Sing), FAMS(Plast Surg), and Seng-Teik Lee, MBBS(Adel), FRCS(Ed), FAMS(Plast Surg)
Background: The technique of percutaneous catheterization of the subclavian vein by the infraclavicular approach is dependent on the location of the subclavian vein in relation to the clavicle. The purpose of this study was to analyze the anatomic relationship between these two structures and how it is influenced by changes in shoulder positioning. Methods: Dissections of the infraclavicular region were performed in seven fresh cadavers and linear measurements made to determine the extent of overlap between the vein and the clavicle in different shoulder positions. Results: When the shoulder was in neutral position, the subclavian vein was overlapped by the medial third or more of the clavicle and this segment of bone was able to serve as a landmark for the vein. However, shoulder elevation displaced the clavicle cephalad and reduced the degree of overlap. Mild shoulder retraction increased the area of contact between the vein and the undersurface of the clavicle, whereas protraction lifted the clavicle off the vein. Conclusion: Infraclavicular subclavian venipuncture should be performed with shoulders in a neutral position and also in slight retraction. An appreciation of the anatomic relationship between the clavicle and the subclavian vein is the key to successful execution of this technique. Key Words: Subclavian venipuncture, Infraclavicular approach, Cadaveric dissections, Anatomic relationship, Shoulder positioning.

ercutaneous catheterization of the subclavian vein by means of the infraclavicular approach has been established as a relatively safe and reliable method of access to the central veins since it was introduced by Aubaniac in 1952.1 However, it is a blind procedure, because the subclavian vein cannot be visualized or palpated. Serious complications of this technique have been reported in the literature,2 6 the major hazards being pneumothorax and subclavian artery injury. Accidental puncture of the pleura can occur because, beyond the protective edge of the first rib, the pleura lies only 5 mm posterior to the subclavian vein.3 The subclavian artery is also vulnerable to injury because it lies just behind and slightly superior to the path of the subclavian vein. Because the clavicle is used as a landmark for locating the subclavian vein, an understanding of the relationship between these two structures is critical to the successful execution of this technique. This important relationship changes with different shoulder positions. For example, if the shoulder is elevated, the acromial end of the clavicle moves cephalad and the vein assumes a more inferior and medial relationship to it (Fig. 1). This was pointed out in a radiologic study by Land who demonstrated on single-view venograms that the subclavian vein, which was in the path of the needle when the
Submitted for publication January 29, 1998. Accepted for publication October 6, 1999. From the Department of Plastic Surgery, Singapore General Hospital, Singapore. This study was supported by a grant from the Department of Clinical Research, Ministry of Health, Singapore. Address for reprints: Bien-Keem Tan, MBBS(Sing), FRCS(Ed), Department of Plastic Surgery, Singapore General Hospital, Outram Road, Singapore 169608.

shoulder was in neutral position, may be out of its path when the shoulder was abducted or elevated.7,8 By using his observations as a basis for further investigation, anatomic dissections were performed to study the spatial relationship between the clavicle and the subclavian vein, correlating this with surface markings pertinent to the technique of subclavian venipuncture. Changes in the relationship were studied in two planes with the shoulder elevated, depressed, retracted or protracted. Shoulder elevation is defined as shrugging the shoulders. Shoulder depression is the reverse movement. Protraction refers to forward movement of the shoulder, as in pushing, thrusting, or punching. Backward movement, as in bracing the shoulders, is termed retraction.

FIG 1. Positional relationship of the subclavian vein to the clavicle is expressed by the ratio VJ:LC. Comparison of shoulder elevation (A) with neutral position (B) shows that during elevation, the vein is in a more medial and inferior position, hence VJ:LC decreases. Triangles, midpoint of clavicle. The asterisk is the point along the lower border of the clavicle where the center of the vein crosses it. VJ, distance between the sternoclavicular joint and the point on the lower border of the clavicle where the center of the vein crosses it. LC, length of the clavicle.

82

Vol. 48, No. 1

Subclavian Venous Catheterization

MATERIALS AND METHODS Anatomic dissections of the infraclavicular region and the root of the neck were performed on seven adult fresh cadavers (four men and three women). A skin incision was made along the entire length of the clavicle, beginning from the sternoclavicular joint to the acromioclavicular joint. The subclavian vein was exposed by detaching the clavicular origins of the pectoralis major and the subclavius muscles. Superiorly, a skin-platysmal flap was raised cephalad to expose the root of the neck. The internal jugular vein was uncovered by dividing the sternal and clavicular heads of the sternocleidomastoid muscle. This vessel was then traced to its junction with the subclavian vein from which the brachiocephalic vein arose, behind the sternoclavicular joint. Two linear measurements were made (Fig. 1): the length of the clavicle (LC), and, the distance between the sternoclavicular joint and the point on the lower border of the clavicle crossed by the center of the subclavian vein (VJ). These measurements were made with the cadaver supine, simulating clinical practice. The shoulders were kept in neutral position and with the arms by the sides. Similar linear measurements were then made with the shoulders maximally elevated (shrugged) for comparison. The effect of shoulder protraction on the positional relationship between the clavicle and the subclavian vein was also examined. All measurements were made within the limits of normal shoulder movement, with all articulations of the shoulder girdle intact to minimize abnormal shoulder mobility.

TABLE 1. Ratio of the distances (VJ:LC) expresses the positional relationship of the subclavian vein to the clavicle a
VJ:LC Cadaver No. Right Shoulder Position Neutral Elevated Left Shoulder Position Neutral Elevated

1 2 3 4 5 6 7

0.32 0.34 0.40 0.41 0.43 0.46 0.33

0.26 0.31 0.34 0.35 0.38 0.35 0.23

0.26 0.34 0.34 0.35 0.34 0.41 0.36

0.23 0.31 0.31 0.32 0.31 0.35 0.26

a VJ, distance between the sternoclavicular joint and the point on the lower border of the clavicle where the center of the vein crosses it. LC, length of the clavicle.

DISCUSSION Several practical points have been highlighted in the literature712 to improve both the safety and success of infraclavicular subclavian venipuncture. Among these, patient positioning and the technique of needle insertion seem to be the most important. However, although these recommenda-

RESULTS The relationship of the subclavian vein to the clavicle at the point where the vein crossed its lower border was expressed as a ratio formulated by Land.7 This was derived by taking the distance from the center of the vein to the sternoclavicular joint (VJ) and dividing this by the length of the clavicle (LC) (Fig. 1). In neutral position, the median VJ:LC ratio on the right side was 0.40 (range, 0.32 to 0.46); whereas on the left, the median VJ:LC ratio was 0.34 (range, 0.26 to 0.41). When the shoulders were maximally elevated, the median VJ:LC ratio for the right was 0.34 (range, 0.23 to 0.38) compared with the left, which was 0.31 (range, 0.23 to 0.35) (Table 1). No gender difference was noted. These results indicate that the point at which the subclavian vein crossed the inferior border of the clavicle shifted medially when the shoulders were elevated (Figs. 1, 2, and 3). The extent of medial displacement of this point ranged from 0.5 to 1.5 cm, depending on the degree of shoulder elevation. Retraction of the shoulders increased the area of contact between the vein and the undersurface of the clavicle (Fig. 4). Conversely, protraction lifted the clavicle off the vein and reduced the segment of vein in contact (Fig. 5). The relationship between bone and vein was most consistent medially where the subclavian vein joined the internal jugular vein.
83

FIG 2. Infraclavicular region (right side, anterior view). The shoulders have been placed in the neutral position. The vein crosses the lower border of the clavicle slightly lateral to its medial third in this specimen. Midpoint of clavicle (arrow).

FIG 3. Infraclavicular region (right side, anterior view). The shoulders have been elevated. The vein crosses the lower border of the clavicle within the medial third of the bone. The midpoint of the clavicle (arrow) has moved cephalad in relation to the path of the vein.

The Journal of Trauma: Injury, Infection, and Critical Care

January 2000

FIG 4. Root of the neck in a decapitated cadaver (right side, superior view). The shoulders have been retracted by placing a sandbag under the vertebral column. The needle tip is resting against the wall of the right subclavian vein. IJV, internal jugular vein; BP, brachial plexus; T, trachea; C6, body of the sixth cervical vertebra.

FIG 6. Infraclavicular approaches to the subclavian vein. 1: Aubaniac (1952)1; Wilson et al. (1962)13; 2: Davidson et al. (1963)15, Corwin and Moseley (1966)16, Morgil et al. (1967)9; 3: Tofield (1969) 10.

tions have been widely accepted, their anatomic basis remains unclear. Patient Positioning It is usually recommended that the patients shoulders be placed in an anatomically neutral position and the arms kept by the side. In addition, a small sandbag or pillow should be inserted underneath the vertebral column between the scapulae. The findings of the present study suggest that placing the shoulders in a neutral position brings the vein and the clavicle together (Figs. 1 and 2). However, it is a common mistake to assume that the shoulders are in neutral position as long as the arms are by the side. This is because there is a natural tendency for the shoulders to slide cephalad when the body is in the Trendelenburg position, a position routinely used for the procedure. Therefore, a conscious effort should be made to counter this tendency by applying gentle traction in a caudal direction on both shoulders. Another recommendation is to place a small sandbag be-

neath the vertebral column between the scapulae to allow the shoulders to fall back. This maneuver serves two purposes: first, it prevents interference with the path of needle insertion by the humeral head,9 thus ensuring that the needle and syringe are always parallel to the coronal plane. Second, it brings the subclavian vein into close contact with the undersurface of the clavicle (Fig. 4), which is desirable for accurate identification of the vein. Technique of Needle Insertion Three techniques have been described for the infraclavicular approach. They differ in the point of insertion of the needle in relation to the midpoint of the clavicle (Fig. 6). Most authors3,13,14 use the midpoint approach, which was originally described by Aubaniac1 in 1952 (Fig. 6). The surface marking of the entry point of the needle is 1 to 2 cm inferior to the lower border of the clavicle along the midclavicular line. The needle tip is directed at the upper border of the suprasternal notch, keeping parallel to the coronal plane. If this landmark is not clearly defined, a finger tip can be placed in the suprasternal notch as a target. Constant negative pressure is applied on the syringe as the needle is advanced posterior to the clavicle, keeping close to its undersurface. A flashback of blood indicates that the vein has been entered, and this entrance should occur at the junction between the middle and medial thirds of the clavicle. The technique aims to enter the subclavian vein just when it begins its passage underneath the clavicle. Neutral shoulder positioning sets the correct bone-vein relationship for this point of entrance to occur. On the other hand, shoulder elevation and protraction disrupt this relationship, thereby reducing the reliability of anatomic landmarks. Tofield10 in 1969 advocated a more lateral approach in which the point of needle insertion was lateral (precise distance was not given) to the midclavicular line. The author stated that this approach improved the safety of the procedure but did not provide the anatomic basis of the recommendation. The rationale for this approach can be understood by
84

FIG 5. Root of the neck (right side, superior view). The shoulders have been protracted. The clavicle has been lifted off the subclavian vein. The needle passed along the undersurface of the bone has missed the vein. IJV, internal jugular vein; SCV, subclavian vein; SA, scalenus anterior; BP, brachial plexus.

Vol. 48, No. 1

Subclavian Venous Catheterization

considering the infraclavicular contour of the upper chest. The clavicle is a long bone whose shaft has a double curve in the horizontal plane. The medial two-thirds of the bone is convex anteriorly, whereas the lateral third is concave anteriorly. A needle inserted along this concavity and maintained level will automatically find the vein along the undersurface of the bone. Although this lateral approach is safe, the needle may not be sufficiently long in patients who have thick chest walls or who are obese. Morgil et al.,9 Davidson et al.,15 and Corwin and Moseley16 (Fig. 6) described a medial approach in which the needle was inserted at the junction between the middle and inner thirds of the lower border of the clavicle. Christensen et al.4 described a point of insertion located even more medially, which was adjacent to the lateral palpable margin of the costoclavicular ligaments. The needle was directed at a point just above the suprasternal notch, and aimed to penetrate the vein as it runs in the groove between the clavicle and the first rib. The main advantage of this medial approach is the relatively constant course of the vein in relation to the bone. Because the site of the costoclavicular ligaments is the fulcrum of clavicular movements,17 bone-vein relationships remain fairly constant at this point. Furthermore, a broad target formed by portions of the three great veins, i.e., the subclavian, the internal jugular, and the innominate, increases the likelihood of success. But, the following anatomic caveats should be recognized when attempting to puncture the vein by using the medial approach: It is more difficult to maintain the needle in a horizontal position when the point of insertion is medial; The thickness of the clavicle increases from lateral to medial. (The diameter of the sternal end is 1.5 to 2 times the diameter of the lateral half of the clavicle). Thus, it is more difficult to insinuate the tip of a straight needle under the medial part of the clavicle because the increased thickness of the bone mandates a steeper approach; The needle has to pass through a thicker layer of intervening soft tissue, including the costoclavicular ligaments. A large-bore needle can sometimes cut a core of tissue in its lumen. The cannulation failure rate has been reported to be between 6 and 20% by authors who use the medial approach.4,15 It seems logical that these factors could be contributory to failure. Christensen et al.,4 Morgil et al.,9 and Corwin and Moseley16 have attributed their failures to obesity and chest wall thickness. Clearly, cannulation failure in these cases is due to insufficient cannula length for a lateral and level approach to the vein. Furthermore, excessive fat distributed around the arms and upper chest in grossly obese patients may prevent neutral positioning of the shoulders. The consistent relationship between the vein and clavicle when the shoulder is in neutral position was documented by Land.7 In this study, right subclavian venograms were performed in 70 adult patients undergoing intravenous urography. Land demonstrated that in 67% of patients examined in the neutral position, the vein passed beneath the clavicle. In 30% of the 70 cases, the vein ran more cephalad. Only in one patient did the vein not pass beneath the clavicle. In another
85

study, Land showed that on the left side, the vein was sited more medially in relation to the clavicle.8 The findings of the present study suggest that gentle shoulder retraction brings the clavicle and the subclavian vein together, whereas shoulder protraction lifts the clavicle off the subclavian vein. One should be aware of this in patients who have neck or shoulder stiffness caused by underlying joint pathologic conditions. In such instances, the clavicle may be a poor landmark for the subclavian vein because neutral positioning is difficult to achieve. On the other hand, excessive shoulder retraction may result in compression of the vein in the groove between the first rib and the clavicle. This was demonstrated by Jesseph et al.12 who showed on magnetic resonance imaging in five individuals that there was a decrease in the anteroposterior diameter of the subclavian vein when the shoulders were braced backward. However, this change should not be an issue if the attempt to cannulate the vein is made lateral to this narrow groove. In clinical situations in which correct shoulder positioning cannot be achieved, such as the presence of cervical spine and shoulder joint pathologic conditions, obesity, or edema of the chest wall and upper limbs attributable to extensive burns, a more medial approach may be attempted. Although the vein may bear a more constant relationship with the clavicle medially, the hazards of the medial approach should be recognized; in attempting to negotiate the needle under a thick clavicular head, there is a risk of injury to deep structures. Subclavian Venipuncture in Children Eichelberger et al. reported that central venous catheterization in neonates and children by using the infraclavicular route is safe.18,19 According to the authors technique, the skin puncture is made at the deltopectoral groove, and by advancing the needle at a gradual angle, the subclavian vein is located in the groove between the clavicle and the first rib. In neonates the subclavian vein position is more cephalad, hence the needle is directed at a point midway between the chin and the sternal notch. In older children, the course of the subclavian vein becomes less cephalad and the syringe system is directed more medially and toward the sternal notch.18,19 Despite anatomic differences attributable to growth and development, Eichelberger et al. have shown that the vein can be consistently located within the medial third of the clavicle. This finding concurs with our observation that the relationship between the vein and clavicle is more constant medially, and consequently less affected by shoulder positioning. Supraclavicular Approach to the Subclavian Vein Since its original description by Yoffa in 1965,20 this technique has gained popularity because of the relatively short and direct approach to the vein.21,22 The needle is inserted in the angle formed by the clavicular head of the sternomastoid muscle and the upper border of the clavicle. Once the skin is entered, the syringe is depressed 15 degrees below the coronal plane and the needle is directed at an angle 45 degrees to the sagittal plane. The vein is met at an average depth of 1 to 1.5 cm from the skin. From an anatomic standpoint, this

The Journal of Trauma: Injury, Infection, and Critical Care

January 2000

approach is reliable because of the constant relationship between the subclavian vein and the medial third of the clavicle. And as long as the depth and direction of needle penetration are controlled, the method is safe. CONCLUSIONS The findings of this anatomic investigation indicate that the subclavian vein passes posterior to the clavicle close to the junction between the inner and middle thirds of the bone. Elevation of the shoulder alters this relationship. The vein is more caudad in these situations, and the point at which it crosses the clavicle is more medial. Therefore, during percutaneous puncture of the subclavian vein, the vein which is in the path of the puncture needle with the arm in the neutral position may be out of its path when the shoulder is elevated. The subclavian vein is closely positioned to the posterior surface of the inner third of the clavicle as the vein passes medially to join the internal jugular vein. Protraction of the shoulders moves the clavicle anteriorly away from the vein. In this position, a needle which is advanced along the posterior surface of the medial third of the bone will not find the subclavian vein. Thus, infraclavicular subclavian venipuncture should be performed with shoulders in the neutral position and in slight retraction. Understanding the variable anatomic relationship between the clavicle and the subclavian vein is critical to the success and safety of subclavian venipuncture. Acknowledgments
The authors thank Professor T. C. Chao, Director of the Institute of Science and Forensic Medicine, Ministry of Health, Singapore for providing the specimens; Dr. Peter Mack, Director of the Department of Experimental Surgery, Singapore General Hospital, for the use of facilities; Mr. Robert Ng, for technical support, and Dr. S. C. Aung for critique of the manuscript. REFERENCES Aubaniac R. Linjection intraveineuse sous-claviculaire; advantages et technique. Presse Med. 1952;60:1456. 2. Malinak LR, Gulde RE, Faris AM. Percutaneous subclavian catheterization for central venous monitoring. Am J Obstet Gynecol. 1965;92:477 482. 3. Smith BE, Modell JH, Gaub M, Moya F. Complications of subclavian vein catheterization. Arch Surg. 1965;90:228 229. 1.

4. 5. 6.

7. 8. 9. 10. 11. 12. 13. 14.

15. 16. 17. 18. 19.

20. 21. 22.

Christensen KH, Nerstrom B, Baden H. Complications of percutaneous catheterization of the subclavian vein in 129 cases. Acta Chir Scand. 1967;133:615 620. Eerola R, Kaukinen L, Kaukinen S. Analysis of 13,800 subclavian vein catheterizations. Acta Anaesthesiol Scand. 1985;29:193197. Casado-Flores, Valdivielso-Serna A, Perez-Jurado L, et al. Subclavian vein catheterization in critically ill children: analysis of 322 cannulations. Intensive Care Med. 1991; 17:350 354. Land RE. Anatomic relationships of the right subclavian vein. Arch Surg. 1971;102:178 180. Land RE. The relationship of the left subclavian vein to the clavicle. J Thorac Cardiovasc Surg. 1972;163:564 568. Morgil RA, Delaurentis DA, Rosemond GP. The infraclavicular venipuncture. Arch Surg. 1967;95:320 324. Tofield JJ. A safer technique of percutaneous catheterization of the subclavian vein. Surg Gynecol Obstet. 1969; 128:1069 1070. Feiler EM, DeAlva WE. Infraclavicular percutaneous vein puncture: a safe technic. Am J Surg. 1969;118:906 908. Jesseph JM, Conces DJ, Augustyn GT. Patient positioning for subclavian vein catheterization. Arch Surg. 1987; 122:12071209. Wilson JN, Grow JB, Demong CV, Prevedel AE, Owens JC. Central venous pressure in optimal blood volume maintenance. Arch Surg. 1962;85:563577. Blackett RL, Bakran A, Bradley JA, Halsall A, Hill GL, McMahon MJ. A prospective study of subclavian vein catheters used exclusively for the purpose of intravenous feeding. Br J Surg. 1978;65:393395. Davidson JT, Ben-Hur N, Nathen H. Subclavian venipuncture. Lancet. 1963;2:1139 1140. Corwin JH, Moseley T. Subclavian venipuncture and central venous pressure. Am Surg. 1966;32:413 415. Last RJ. Upper limb. In: McMinn RMH, ed. Lasts Anatomy. 8th ed. Edinburgh, PA: Churchill Livingstone; 1990:53144. Eichelberger MR, Rous PG, Hoelzer DJ, Garcia VF, Koop CE. Percutaneous subclavian venous catheters in neonates and children. J Pediatr Surg. 1981;16:547553. Eichelberger MR, MacDonald MG. Percutaneous central venous catheterization. In: Fletcher MA, MacDonald MG, Avery GB, eds. Atlas of Procedures in Neonatology. Philadelphia: JB Lippincott; 1983:179 185. Yoffa D. Supraclavicular subclavian venipuncture and catheterization. Lancet. 1965;2:614 617. Defalque RJ. Subclavian venipuncture: a review. Anesth Analg Curr Res. 1968;47:677 682. James PM Jr, Myers RT. Central venous pressure monitoring: complications and a new technic. Am Surg. 1973;39:75 81.

86