Secondary Logo

Journal Logo

Patient Safety: Case Report

Inadvertent Intrathecal Placement of a Pulmonary Artery Catheter Introducer

Schepers, Marcel, MD*; Vercauteren, Marcel, MD, PhD*; De Bock, Dina, MD; Rodrigus, Inez, MD, PhD; Vanderplanken, David, MD; Camerlinck, Michael, MD§

Author Information
doi: 10.1213/ANE.0b013e318269cd55

Internal jugular veins are important routes to secure intraoperative vascular access and to measure the central venous pressure. The right internal jugular vein is the easiest to access, most reliable, and the most used site for central venous cannulation because all structures to be cannulated are in line, and sterile preparation of the patient can proceed during the access procedure. However, because of the proximity of the jugular vein to the carotid artery, heart, cervical spine, major nerves, and pleura, lifethreatening complications are relatively common.1–3 Acute complications such as arrhythmias, bleeding (with arterial puncture in 2.9%, arterial cannulation in 0.2% of cases), pneumothorax in 0.6% of cases, and damage to the esophagus and nerves have been described extensively. The complication rate may be lower in males (2.1%) than in women (3.8%). Even in experienced hands, a considerable number of early complications may occur. After 2 unsuccessful cannulation attempts, associated complications are more likely to occur.

The unintended intrathecal placement of a central venous catheter is a very rare complication that was first reported in 1985.4 We report a case of unintended intrathecal cannulation with a pulmonary artery catheter (PAC) introducer in a 66 yearold man scheduled to undergo coronary artery bypass grafting (CABG) surgery.


A 66yearold man (weighing 82 kg and 178 cm tall) presented with a history of coronary artery disease with severe stenosis of the proximal right coronary circumflex artery and proximal and distal left coronary artery. He was scheduled for elective CABG surgery. Preoperative assessment included echocardiography, a 5-lead electrocardiogram, chest radiograph, and standard laboratory tests.

For induction of general anesthesia, the patient received midazolam 1 mg, remifentanil 0.2 μg/kg, and inhaled sevoflurane 2%, followed by rocuronium 0.9 mg/kg to facilitate tracheal intubation. Anesthesia was maintained with remifentanil 0.2 μg/kg/min and inhaled sevoflurane 2% in oxygen.

After tracheal intubation, the patient was placed in a 10° headdown position and the skin was disinfected and draped for placement of a PAC (Arrow 8.5 Fr percutaneous sheath introducer, Arrow International, Reading, PA) into the right internal jugular vein. The thirdyear resident in anesthesiology had performed >150 similar procedures and was closely supervised by a staff member with 5 years experience in cardiac anesthesia. Our anatomic landmark technique consists of locating the carotid artery and aiming the finder needle lateral to it toward the right nipple away from the artery. At a depth of 1.5 cm, venous blood was aspirated. Using a Seldinger technique, a guidewire (J-wire) was introduced, but resistance was felt during 2 attempts to advance it intravascularly, possibly because of contact with the vessel wall or an obstruction. During the third attempt to introduce the guidewire, the straight soft end of the wire was introduced first, resulting in successful guidewire advancement. After smooth dilation of the wire track, the PAC introducer was inserted. After uneventful removal of the guidewire and dilator, aspiration of the introducer yielded a large amount of clear liquid.

The head and neck were immediately immobilized, because we suspected intrathecal placement of the introducer. Neurosurgeons were consulted, who confirmed by a simple in vitro test that the aspirated fluid was cerebrospinal fluid (CSF). A 3-dimensional computer tomography reconstruction of the cervical spine confirmed that the introducer was placed through the foramen of C6 to 7 up to C7-T1 (Fig. 1). It was positioned on the right side in the spinal canal without visualization of blood. The decision was made to immediately remove the catheter and allow the patient to wake up.

Figure 1
Figure 1:
Noncontrast-enhanced multidetector computed tomography scan with paracoronal curved reconstruction along the axis of the deep venous catheter (A) and left lateral volume rendered reconstruction (B), showing the course of the catheter through the soft tissues of the neck, the right foramen transversarium of C6 (arrowhead), the right C6 to 7 neural foramen (arrow) and the position of the tip on the right side inside the cervical spinal canal at C7–T1 (asterisk).

To achieve faster healing and to monitor CSF pressure and possible presence of blood, the neurosurgeons suggested placement of a lumbar drain (Lumbar drainage Kit II, Codman & Shurtleff Inc, Raynham, MA) for 5 days, together with strict bed rest. With the patient positioned on the left side, the lumbar drain was introduced under sterile conditions at the thirdfourth lumbar (L3 to 4) interspace without difficulties. The drain was placed at the level of the foramen of Monroe, allowing an average draining volume of 10 mL/hour.

Although it was initially planned to remove the introducer after performing a venogram, the neurosurgeon, unaware of these plans, removed the introducer. No major bleeding in the surrounding tissue was noted.

After recovering from anesthesia, the patient was able to move his upper and lower extremities without any evidence of motor weakness or decreased sensation, except minor paresthesia of the distal phalanges of the second and third digit of the right hand. The day after the event, a magnetic resonance imaging of the neck could not find evidence of CSF leakage nor injury to the blood vessels or the spinal cord. The paresthesia and hypesthesia resolved 4 days later.

Five days after the event and removal of the external lumbar drain, the patient complained of severe postural headache, nausea, and vomiting possibly because of persistent CSF leak unresponsive to conventional treatment. The patient received a lumbar epidural blood patch with 20 mL of autologous blood, which alleviated all of his complaints. Two days after the blood patch the patient was discharged from the hospital and was readmitted 1 month later for his CABG surgery. Ultrasoundguided puncture (LogiqE, GE Medical Systems China, Wuxi, China) of the internal jugular vein confirmed the location of the internal jugular vein at a depth of 1 cm, followed by easy insertion of the guidewire. Both the anesthetic course and surgery were uneventful.


Several complications of intrajugular vein catheterization such as arterial puncture or cannulation, pneumo- or hemothorax, arrhythmias, and damage to the esophagus or surrounding nerves have been described.1–3 Intrathecal or extradural placement of a central venous catheter is a very rare complication that has been described 3 times in adults4–6 and 3 times in children.7–9 The majority of the patients were discharged from the hospital without neurological sequelae, but in 2 of these 6 cases the catheter was placed extradurally.6,7

Ours is a report of an intrathecal cannulation with a large bore introducer during attempted placement of a PAC.

There are several explanations of how this complication might have occurred. First, the neuraxis may have been directly punctured by the needle before placement of the guidewire. We feel that this was very unlikely because blood was aspirated at a depth of 1.5 cm and backflow of blood did not cease during attempts to advance the soft end of the guidewire. Second, the posterior aspect of the internal jugular vein may have been penetrated by the guidewire or needle. Alternatively, cannulation of a vertebral vein or a small (nonjugular) vein might have occurred, followed by penetration of the vessel wall. The guidewire might have then traveled posterior to the carotid artery, penetrating the prevertebral fascia and scalenus anterior muscle and entering the intervertebral foramen of C6 to 7. Third, an initially created false route or puncture might have facilitated advancement of the rigid dilator through all the tissues surrounding the intervertebral foramen.

Inexperience is one of the major causes for complications and malpositioning of central venous catheters.1,10 However, this was not the case with the trainee performing the procedure in our case. Switching to the straight tip of the guidewire may have facilitated penetration of the vessel wall and subsequently may have created a false route. However, not all published books and articles advise against using the straight end of a Seldinger guidewire, although the risks of such practice are presented.10–12 Despite debates whether a guidewire can penetrate vessel walls or other structures, cardiac, brachiocephalic vein, and even tracheal perforations have been reported during central vein cannulation13–16; even in such cases, however, it was not always clear whether the dilator or the guidewire was responsible for the unintended perforation, nor was it clear, in the case of guidewires, whether the soft or rigid ends of the guidewire were leading during insertion.

The authors of previously described unintended intrathecal insertion of central venous catheters attribute their complication to a number of possible causes. The most important cause is that in most cases, multiple attempts were required.4,7,9 In 2 cases, even though there was no spontaneous return of blood after removing the syringe from the needle, the authors continued to advance the catheter.4,8 Furthermore, retrograde cannulation for jugular vein oxygen saturation measurement was reported in 2 cases; in these reports, the angle necessary to enter the intervertebral foramen was easier to achieve and likely facilitated this complication.5,6 Fumagalli et al. might have punctured directly through the intervertebral foramen but were unable to detect this inadvertent puncture because of the presence of blood in the intrathecal space due to subarachnoid bleeding.5 Other authors also reported the use of the rigid tip of the guidewire as the cause for such serious adverse events.7,9

Because of significant variability in anatomy, the landmark method may have a relatively high complication rate. To improve the success rate, the patient should be positioned correctly aiming to optimize filling of the vascular bed. The use of ultrasoundguided puncture in adults may improve the success rate of cannulation by 11%–25% with each puncture attempt.17 In infants, unfortunately based on limited experience, no improved success rate was noted with ultrasoundguided punctures.18 Although there is a growing trend to use ultrasound for central venous access, it is mostly used to locate the vessel for needle placement and to document the position of the guidewire. If the cannulation procedure, including the insertion of the wire, needs to be visualized, handling of the ultrasound probe requires either a second hand19 or a special needle guidance device to enable realtime imaging, especially of the crucial step of guidewire advancement, which requires both of the operator’s hands.20,21 Although ultrasound guidance might not have avoided penetration of the vessel wall by the guidewire, checking the position of the wire before dilating the catheter track might have prevented further damage.22 Another useful tool might have been verification of the guidewire position with fluoroscopy; however, this method is used less commonly in daily anesthetic practice than for placement of parenteral nutrition, tunneled or port catheters.

Lumbar subarachnoid drainage has been widely recommended for treating CSF fluid leaks after skull base, sinus, and cervical spine surgery.23,24 To maintain a low pressure at the level of C7-T1, a volume of 10 mL/h of CSF was drained in an attempt to facilitate healing of the dural perforation at this level.

The patient’s reduced sensitivity and minor paresthesia of the distal phalanges of the second and third digits of the right hand recovered spontaneously within a few days after the event, suggesting only minor irritation and swelling around the foramen of C6 to 7. The complaint of headache was consistent with persistent CSF leak, which may have occurred at 2 levels, either at C7-T1, or at L3 to 4, which was the site of placement of the spinal drain. As conservative therapy was ineffective, an epidural blood patch at the lumbar level (close to the original puncture site) resulted in almost immediate relief and in longterm cure of the complaints. In some studies, epidurally injected blood was shown by magnetic resonance imaging studies to move preferentially in a cephalad, rather than caudad, direction.25 One hour after a lumbar blood patch, blood could be detected in the cervical epidural area.26,27

The nearinstantaneous relief of headache complaints may be explained by the increase in CSF pressure and the effect upon the intracranial pressure. However, with respect to a cervical dural perforation, it may be questioned whether despite immediate improvement, this may be effective for longterm relief of the headache. Nevertheless, several authors have reported successful treatment of proven cervical leaks with a simple lumbar patch without subsequent exacerbation of headaches.28,29 If our patient had still suffered headaches after the first blood patch, repeating the blood patch at the cervical level would have been considered. However, it may be questioned whether such a large anterior dural defect might be sealed by a (posteriorly placed) blood patch. Surgical repair might have been the only remaining solution.

In summary, internal jugular vein cannulation difficulties and complications may occur for many reasons. Using the straight end of the guidewire should be avoided even if placement of the soft Jwire end is unsuccessful. Ultrasonography is considered the best aid during central venous catheterization, particularly when vein cannulation or guidewire advancement is difficult; in such cases, ultrasound guidance should be mandatory. However, it should be recognized that even ultrasound use may not have avoided the unintended penetration of the vessel wall; however, ultrasound visualization of the advancing dilator may have prevented its intrathecal placement.


Name: Marcel Schepers, MD.

Contribution: This author helped as the attending resident during surgery and wrote the case description.

Name: Marcel Vercauteren, MD, PhD.

Contribution: This author helped with the case report supervision and performed EBP.

Name: Dina De Bock, MD.

Contribution: This author helped with the surgical procedure.

Name: Inez Rodrigus, MD, PhD.

Contribution: This author helped supervise the cardiac surgery and helped with perioperative care.

Name: David Vanderplanken, MD.

Contribution: This author helped with lumbar drain placement and supervision.

Name: Michael Camerlinck, MD.

Contribution: This author helped with imaging.

This manuscript was handled by: Sorin J. Brull, MD, FCARCSI (Hon).


1. Schummer W, Schummer C, Norman R. Mechanical complications and malpositions of central venous cannulations by experienced operators. Intensive Care Med. 2007;33:1055–9
2. Karapinar B, Cura A. Complications of central venous catheterizations in critically ill children. Pediatrics International. 2007;49:593–9
3. Pikwer A, Baath L, Davidson B. The incidence and risk of central venous catheter malpositioning: a prospective cohort study in 1619 patients. Anaesth Intensive Care. 2008;36:30–7
4. Nagai K, Kemmotsu O. An inadvertent insertion of a SwanGanz catheter into the intrathecal space. Anesthesiology. 1985;62:848–9
5. Fumagalli P, Lusenti F, Martini C. Retrograde cannulation of the jugular vein: erroneous positioning of the catheter in the subarachnoid space. Br J Anaesth. 1995;74:345–6
6. Gemma M, Tommasino C, Cipriani A. Cannulation of the cervical epidural venous plexus: a rare complication of retrograde internal jugular vein catheterisation. Anesthesiology. 1999;90:308–11
7. Skinner TA, Mather SJ. Inadvertent extradural insertion of an internal jugular catheter in an infant. Br J Anaesth. 1995;75:790–3
8. Miyamoto Y, Kinouchi K, Hiramatsu K. Cervical dural puncture in a neonate: A rare complication of internal jugular venipuncture. Anesthesiology. 1996;84:1239–42
9. Fujita Y, Sobue K, Hattori T, Takeuchi A, Tsuda T, Katsuya H. Inadvertent intrathecal cannulation in an infant, demonstrated by threedimensional computed tomography: a rare complication of internal jugular vein catheterization. J Anesth. 2006;20:122–5
10. Polderman KH, Girbes AJ. Central venous catheter use. Part 1: mechanical complications. Intensive Care Med. 2002;28:1–17
11. Schroeder RA, Barbeito A, BarYosef S, Mark JBMiller RD. Cardiovascular monitoring. In: Anesthesia. 2010 Philadelphia Churchill Livingstone:1287
12. Antonelli D, Freedberg NA, Turgeman Y. Pacing and defibrillation lead exchange without vein puncture. Pacing Clin Electrophysiol. 2009;32:588–90
13. Blake PG, Uldall R. Cardiac perforation by a guide wire during subclavian catheter insertion. Int J Artif Organs. 1989;12:111–3
14. Dimitriou V, O’Leary AM. Perforation of the trachea by a central venous catheter guidewire. Can J Anaesth. 1993;40:903–4
15. Porter JM, Page R, Wood AE, Phelan D. Ventricular perforation associated with central venous introducerdilator systems. Can J Anaesth. 1997;44:317–20
16. Innami Y, Oyaizu T, Ouchi T, Umemura N, Koitabashi T. Lifethreatening hemothorax resulting from right brachiocephalic vein perforation during right internal jugular vein catheterization. J Anesth. 2009;23:135–8
17. Augoustides JG, Horak J, Ochroch AE, Vernick WJ, Gambone AJ, Weiner J, Pinchasik D, Kowalchuk D, Savino JS, Jobes DR. A randomized controlled clinical trial of realtime needleguided ultrasound for internal iugular venous cannulation in a large university anesthesia department. J Cardiothorac Vasc Anesth. 2005;19:310–5
18. Mitre CI, Golea A, Acalovschi I, Mocan T, Caea AM, Ruta C, Mariana M. Ultrasoundguided external iugular vein cannulation for central venous access by inexperienced trainees. Eur J Anaesthesiol. 2010;27:300–3
19. Sigaut S, Skhiri A, Stany I, Golmar J, Nivoche Y, Constant I, Murat I, Dahmani S. Ultrasound guided internal jugular vein access in children and infant: a metaanalysis of published studies. Paediatr Anaesth. 2009;19:1199–206
20. Dowling M, Jlala HA, Hardman JG, Bedforth NM. Realtime threedimensional ultrasoundguided central venous catheter placement. Anesth Analg. 2011;112:378–81
21. Ball RD, Scouras NE, Orebaugh S, Wilde J, Sakai T. Randomized, prospective, observational study comparing residents’ needleguided vs freehand ultrasound techniques for central venous catheter access. Br J Anaesth. 2012;108:72–9
22. Moak JH, Lyons MS, Wright SW, Lindsell CJ. Needle and guidewire visualization in ultrasoundguided internal iugular vein cannulation. Am J Emerg Med. 2011;29:432–6
23. Allen KP, Isaacson B, Purcell P, Kutz JW Jr, Roland PS. Lumbar subarchnoid drainage in cerebrospinal fluid leaks after lateral skull base surgery. Otol Neurol. 2011;32:1522–4
24. Joseph V, Kumar GS, Rajshekhar V. Cerebrospinal fluid leak during cervical corpectomy for ossified posterior longitudinal ligament: incidence, management, and outcome. Spine. 2009;34:491–4
25. Szeinfeld M, Ihmeidan IH, Moser MM, Machado R, Klose KJ, Serafini AN. Epidural blood patch: evaluation of the volume and spread of blood injected into the epidural space. Anesthesiology. 1986;64:820–2
26. ChenSan S, Min-Yu L, YungYee C. Neuroimaging and treatment of spontaneous intracranial hypotension and magnetic resonance imaging evidence of blind epidural blood patch. Eur Neurology. 2009;61:301–7
27. Ferrante E, Arpino I, Citterio A. Is it a rational choice to treat with lumbar epidural blood patch headache caused by spontaneous cervical CSF leak? Cephalalgia. 2006;26:1245–6
28. ColonnaRomano P, Linton P. Cervical dural puncture and lumbar extradural blood patch. Can J Anaesth. 1995;42:1143–4
29. Förderreuther S, Yousry I, Fuhry L. Partial improvement of headache in a patient with spontaneous cervical cerebrospinal fluid leakage after a lumbar bloodpatch. Cephalalgia. 2000;20:674–6
© 2013 International Anesthesia Research Society