• Tyler Scaff

The Right Site, the First Time: Which Central Line Insertion Site is Best for EM Docs?

Image: Editorial Team, A. (2020). Care of Central Venous Catheters . Ausmed. Retrieved from https://www.ausmed.com/cpd/articles/-central-venous-catheters

Much dispute exists over the ideal location for a central line. In this review, let’s explore the most recent data regarding which central line location (femoral, internal jugular, or subclavian) is actually the “best” option. I will use the recommendations from Tintinalli’s Emergency Medicine Manual to preface each section before exploring recent literature. I will spoil the ending by stating that Tintinalli et al. strongly recommend the internal jugular site as a general rule unless there is a specific contraindication. Second-line (pun-intended) site is the right subclavian vein, then the left, and finally the femoral approach.

What are the indications for a central venous catheter (CVC)?

First, as a quick review, Tintinalli’s Emergency Medicine Manual describes the indications for a central line as the following:

1. “Inability to obtain peripheral access

2. Access to central circulation needed for procedures such as pulmonary artery catheter placement, transvenous pacemaker placement, or urgent hemodialysis

3. Measurement of central venous pressure (such as in sepsis, congestive heart failure, or pericardial effusion)

4. Administration of sclerosing medications, continuous vasopressors, concentrated ionic solutions, or cytotoxic chemotherapeutic agents”

I would also add to this list the need for therapeutic cooling such as in post-cardiac arrest patients, although this indication will preclude the discussion of optimum location since a standard long cooling catheter can only be inserted in the femoral vein.

What are the potential complications of CVC insertion?

Potential complications of central venous catheterization include:

1. Catheter-Related Bloodstream Infection (CRBI) or local infection

2. Pneumothorax

3. Arterial puncture

4. Malpositioning

5. Catheter-associated infection

6. Thrombosis

7. Chylothorax (injury to thoracic duct from left-sided line placement)

8. Hydrothorax or hydromediastinum

9. Air emboli

10. Great vessel or right atrial perforation (causing hemothorax or tamponade)

11. Airway compromise (tracheal injury, hematoma with airway compression)

Note that many articles I found have lumped together multiple complications into umbrella terms such as “mechanical” vs. “infectious” complications. This does make individual complications, particularly the more rare, difficult to parse.

What does CRBI mean? Is it different from CLABSI?

A side note: CRBI is a different term than the more commonly-known term CLABSI. CLABSI (central-line associated bloodstream infection) is a CDC surveillance term used to describe a bloodstream infection in a patient who had a central line in the 48 hours prior to onset of infection. It does not necessarily stem directly from the catheter itself. A CRBI, on the other hand, requires a lab confirmation that the infection stemmed from the catheter itself. (Source: CDC)

Catheter-associated infection and thrombosis

“Limit femoral vein cannulation because of the higher complication rates (notably infection and thrombosis) and the limits it places on patient mobility. – Tintinalli et al., 9th ed., ch 31 pp. 208

Analysis of these two complication types constitutes the bulk of the associated literature. As we have already seen, Tintinalli’s view on the femoral vein cannulation is clear. Not only does the femoral site have the highest risk of infection, but the authors also cite studies that place femoral line thrombosis rates at 20%! Is the femoral site really that bad, especially since we often use that site in the emergency department for quick access?

The short answer is: yes, femoral central venous catheters (CVCs) have the highest rates of infection and thrombosis according to current literature, but both complications are associated with long-term insertion. Below is a discussion of six recent studies I found which address the topics of catheter-associated infection and thrombosis. Note that some break down the discussion into bloodstream infection, local infection, and positive catheter tip culture.

· Arvaniti, K et al., 2017 – This was a meta-analysis of 11 observational studies, 7 randomized controlled trials (RCTs) which were primarily looking for other outcomes, and 2 RCTs that looked for catheter-associated bacterial infection (CRBI) and colonization primarily. In total this study looked at 18,554 CVCs. The relative risk of bacterial colonization in the internal jugular (IJ) site vs. subclavian (SC) site was 2.25 (95% CI 1.84 – 2.75). The relative risk between the femoral and SC sites was 2.92 (2.11 – 4.04). The relative risk of bacterial infection between the femoral and subclavian sites was 2.44 (1.25 – 4.75), and the relative risk of CRBI between the IJ vs. femoral sites was 0.55 (0.34 – 0.89). Overall, this was a high-power, well-designed meta-analysis which provided a convincing and consistent argument for using the subclavian site to minimize infection. However, I would be interested to know the catheter-day statistics for these catheters and how long specifically it took for each site to become infected. For example, is a femoral site cannulation associated with increased rates of infection even if it is in for less than 24 or 48 hours?

· Camkiran, First A et al., 2016 – This was a single-center RCT focusing on pediatric cardiac surgery patients. The study consisted of 280 patients, and its primary outcome was actually first-attempt success, with a secondary of infectious/mechanical complications (note the generalized term of “mechanical complications”; this pops up a lot in these studies). The first-attempt success rate was not significantly different between the sites, but the overall success rate of catheter placement was 91% for SC vs. 82% in the internal jugular. The rate for positive catheter tip culture was 26.1% in the IJ vs. 3.6% in the SC. The rate of CRBI was 0% in the SC vs. 6.9% in the IJ. Again, another convincing study suggesting that the subclavian catheterization has the lowest infection rates.

· Marik, PE et al., 2012 – This was a meta-analysis consisting of two RCTs (total 1,006 catheters) and 8 cohort studies (total 16,370 catheters). Removing two outlier studies, no significant difference was observed in infection rates. Interestingly, there was a higher risk of femoral site infection in earlier studies that decreased with later studies. There was heterogeneity in the deep venous thrombosis (DVT) analysis but no significant difference in risk of DVT was observed in the femoral site as compared to other sites, according to the two RCTs. This study does suggest that there is no significant difference between infection rates of the three sites.

· McLaughlin, CM. et al., 2019 – Finally, some real talk about thrombosis! This was a retrospective cohort study of 209 pediatric trauma patients. 65% of the catheters were femoral, 19% subclavian, 11% through an arm vein, and 5% internal jugular. The rate of catheter-associated thrombosis was higher in the femoral (18.4 per 1,000 catheter days) vs. other (3.5 per 1,000 catheter days) sites . The femoral catheters were also removed earlier (2-7 days vs. 3-12 days) and were larger in diameter (4-7 French vs. 4 French). The authors suggest that if the femoral site must be used in a pediatric trauma patient, that the smallest useful catheter be utilized to minimize risk of catheter-associated thrombosis.

· Youn, SH et al., 2015 – Another retrospective review of 1,646 CVCs. The local infection rates reported by this study were 4.83 in the subclavian, 9.55 IJ, and 7.93 femoral per 1,000 catheter days. Bloodstream infection rates, the authors note, were primarily associated with catheter insertion duration.

· Parienti, JJ. et al., 2015 – Finishing off with a randomized controlled trial in the adult ICU. Participants were randomized to the subclavian, IJ, or femoral sites in a 1:1:1 ratio if all three sites were suitable (a three-choice scheme), and 1:1 ratio if only two sites were suitable. This study looked at 3,471 catheters for a composite outcome of CRBI and symptomatic DVT. The hazard ratio for femoral vs subclavian for this composite outcome was 3.5 (95% CI 1.5 – 7.8). For IJ vs. SC, the hazard ratio was 2.1 (1.0 – 4.3), and there was no significant difference between the femoral and IJ sites.


Pneumothorax is a possible complication of internal jugular and subclavian central line insertion, manifesting as hypoxia, tachypnea, tachycardia, and sometimes hypotension after attempting to place the line. A chest x-ray demonstrating focal absence of lung markings or ultrasound demonstrating absent pleural motion (the “barcode” sign on M-mode) confirms the diagnosis, and generally these need to be treated with a chest tube or pigtail catheter with or without needle decompression. Obviously in this complication, the femoral site has the advantage because it carries no risk of causing a pneumothorax. This is another reason why the femoral site is preferred for rapid-insertion or “crash” lines; a chest x-ray is not required to confirm position prior to use of the catheter.

Tsotsolis & Zarogoulidis (2015) wrote an excellent article about the incidence of pneumothorax. They note that while the subclavian approach has been reported to have a higher incidence of iatrogenic pneumothorax, their own literature review demonstrated no significant difference in risk between internal jugular and subclavian approaches. What did make the biggest difference was the experience of the provider and number of attempts; i.e., the more attempts at cannulating the vessel, the higher the likelihood of pneumothorax. Specifically, the authors found that performing any more than two attempts at the same site carried the highest likelihood of mechanical complications. Therefore, always set yourself up for first-pass success.

Arterial Puncture

The arterial puncture is one of the highest occurring complications although often the least harmful. The importance is recognizing the arterial access, because arterial dilation can be devastating. While often the artery will vasoconstrict and occlude a needlestick (and direct pressure by the operator will also help), dilation of the artery invariably requires vascular surgery consultation, and possibly repair in the OR. It is for this reason that many proceduralists will take a moment to confirm using ultrasound that their guidewire is in the vein prior to dilation.

A single-center review of 280 pediatric cardiac surgery patients found an incidence of arterial puncture to be 8% in the internal jugular site and 2% in the subclavian site, suggesting that the internal jugular may be riskier for arterial puncture than subclavian (Camkiran et al., 2016).


Malpositioning essentially refers to a catheter going somewhere it shouldn’t. This umbrella term includes lines that go too deep, too shallow, or off into an unintended vessel. It even can include arterial cannulation (discussed above) depending on the article you read. Regardless, often malpositioning can be detected through bedside x-ray, although CT and/or ultrasound may offer additional information. As many case reports are available, I will abbreviate the discussion to three papers.

First, the aforementioned study by Camkiran et al. noted a malposition rate of 17% in subclavian catheters in their pediatric patients vs. 1% in the internal jugular catheters.

Second, Vigo et al. (2018) describe an interesting case of internal jugular CVC malpositioning producing a finding known as Lancisi’s sign, which is a large V wave during venous pulsation normally seen in tricuspid regurgitation. However, this case of Lancisi’s sign was iatrogenic; an internal jugular central line was placed too deep and artificially stented open the tricuspid valve. This led to significant tricuspid regurgitation that self-resolved when the line was pulled back.

Finally, Bekele et al. (2017) described a case in which after multiple attempts, a left subclavian catheter inadvertently found its way up the left internal jugular vein. This was seen on x-ray. (ed. Note: I have seen this complication several times, often associated with patient history of pulmonary hypertension)

Chylothorax, Hydrothorax, or Hydromediastinum

Because of the risk of damage to the thoracic duct on the left side of the chest, Tintinalli recommends attempting cannulation of the right subclavian vein before the left. With the use of x-ray to confirm catheter tip placement prior to delivery of infusate, the risk of hydrothorax or hydromediastinum is minimized.

Air Emboli

McCarthy et al. (2016) provides a great review of techniques to mitigate the dreaded risk of air embolism. This complication is technique-dependent and very rare. Some ways to reduce the chances of air embolism are by flushing the catheter thoroughly before insertion, placing the patient in Trendelenburg for SVC/IJC cannulation, and always flushing vertically to keep air bubbles from entering the line.

Great Vessel or Right Atrial Perforation

One case report I found describes a 69-year-old male in ESRD who was admitted for intracerebral hematoma. He received a left subclavian dialysis catheter that inadvertently perforated the superior vena cava. This resulted in a right-sided hemothorax requiring a chest tube that drained 1500 mL of blood and he required surgery for repair (Turkyilmaz et al., 2018).

Airway Compromise

I did not find any case reports or literature specifically discussing airway compromise such as tracheal perforation or compression from hematoma. However, it does stand to reason that like all complications, the risk of these rare complications increases with multiple attempts and inexperienced hands.


To summarize, Tintinalli suggests the internal jugular site as the “safest” option for central venous catheterization, but the literature shows that each site has risks and benefits depending on which complication you are concerned about. For example, while the femoral site does indeed appear to post the greatest risk for catheter-associated infection, it will not cause a pneumothorax and can be used immediately after successful placement. The subclavian site, conversely, carries a higher risk for pneumothorax but the lowest rate of infection per catheter-days. The internal jugular route appears to carry the highest risk of arterial puncture but the anatomy lends itself well to an experienced ultrasonographer. Ultimately, the emergency provider must know their options, inspect each site of potential cannulation, and most importantly, maximize their chance of first-pass success.

Editor Commentary from Dr. Dzurik:

What Dr. Scaff teases out through excellent literature review is the cultural shift in central lines through the evolution of medicine. Femoral lines have traditionally been considered “dirty” lines, but as central line care evolves both in nursing care of line sites and in operator sterile prep, these differences in infection narrows and disappears. Subclavians have been vanishing from education because we have transitioned to ultrasound guided lines as a standard of care. That approach is often faster, cleaner, but is intensely operator variable based on experience and often represents more complications to the younger generation of physicians. While the introduction of ultrasound to line placement reduced complications, that benefit has vanished because we become too dependent on ultrasound and lost some of our anatomical knowledge needed for “blind” lines. Essentially preferences and opinions on lines are often formed by when you trained, and the literature has shifted as the focuses in resident education has transitioned. For example: I trained in the second half of 2010s, and essentially the only subclavians I do are on people who already have pneumothorax on that side. Far more have I done IJs (my preference) and femoral lines as I trained entirely during the ultrasound era of procedures. My partners who trained less than 10 years ahead of me has radically different ratios of procedural numbers. The answer is what Dr. Scaff summarizes: know the indications, the anatomy, and recognize when not to use a site. Also; pull the line as soon as they don’t need it.

Suggested Reading:

  1. American College of Emergency Physicians., Tintinalli, J. E., Ma, O. John et al. (2020). Tintinalli's emergency medicine manual (9th ed.). New York: McGraw-Hill Medical. Chapter 31.

  2. Arvaniti K, Lathyris D, Blot S, Apostolidou-Kiouti F, Koulenti D, Haidich AB. Cumulative Evidence of Randomized Controlled and Observational Studies on Catheter-Related Infection Risk of Central Venous Catheter Insertion Site in ICU Patients: A Pairwise and Network Meta-Analysis. Crit Care Med. 2017;45(4):e437-e448. doi:10.1097/CCM.0000000000002092

  3. Bekele NA, Abebe WA, Shifa JZ. Misplaced subclavian central venous catheter. Pan Afr Med J. 2017;27:59. Published 2017 May 26. doi:10.11604/pamj.2017.27.59.9532

  4. Camkiran Firat A, Zeyneloglu P, Ozkan M, Pirat A. A Randomized Controlled Comparison of the Internal Jugular Vein and the Subclavian Vein as Access Sites for Central Venous Catheterization in Pediatric Cardiac Surgery. Pediatr Crit Care Med. 2016;17(9):e413-e419. doi:10.1097/PCC.0000000000000878

  5. Marik PE, Flemmer M, Harrison W. The risk of catheter-related bloodstream infection with femoral venous catheters as compared to subclavian and internal jugular venous catheters: a systematic review of the literature and meta-analysis. Crit Care Med. 2012;40(8):2479-2485. doi:10.1097/CCM.0b013e318255d9bc

  6. McCarthy CJ, Behravesh S, Naidu SG, Oklu R. Air Embolism: Practical Tips for Prevention and Treatment. J Clin Med. 2016;5(11):93. Published 2016 Oct 31. doi:10.3390/jcm5110093

  7. McLaughlin CM, Barin EN, Fenlon M, et al. Symptomatic catheter-associated thrombosis in pediatric trauma patients: Choose your access wisely. Surgery. 2019;166(6):1117-1121. doi:10.1016/j.surg.2019.05.018

  8. Parienti JJ, Mongardon N, Mégarbane B, et al. Intravascular Complications of Central Venous Catheterization by Insertion Site. N Engl J Med. 2015;373(13):1220-1229. doi:10.1056/NEJMoa1500964

  9. Tsotsolis, N., & Zarogoulidis, C. (2015, March). Pneumothorax as a complication of central venous catheter insertion. Retrieved July 28, 2020, from http://atm.amegroups.com/article/view/5829/6569

  10. Turkyilmaz A, Karapolat S, Kilic M, Tekinbas C. The Perforation of the Superior Vena Cava Secondary to the Left Subclavian Dialysis Catheter. Vasc Endovascular Surg. 2017;51(2):95-97. doi:10.1177/1538574416689427

  11. Vigo V, Lisi P, Galgano G, Lomonte C. Lancisi's sign and central venous catheter tip position: a case report. J Vasc Access. 2018;19(1):92-93. doi:10.5301/jva.5000760

  12. Youn SH, Lee JC, Kim Y, Moon J, Choi Y, Jung K. Central Venous Catheter-Related Infection in Severe Trauma Patients. World J Surg. 2015;39(10):2400-2406. doi:10.1007/s00268-015-3137-y

Post peer-reviewed by: Alex Dzurik, MD, FAAEM; Assistant Program Director, Mercy Health St. Vincent Medical Center EM Residency Shyam Murali, MD; Academic Chief Resident, Website Editor-in-Chief, Mercy Health St. Vincent Medical Center EM Residency

  • Twitter
  • Instagram

© 2020 by Mercy St. Vincent Emergency Medicine Residency, Toledo, OH. Proudly created with wix.com.