Secondary Logo

Journal Logo


A severe complication after ultrasound-guided thoracic paravertebral block for breast cancer surgery

total spinal anaesthesia

A case report

Albi-Feldzer, Aline; Duceau, Baptiste; Nguessom, Williams; Jayr, Christian

Author Information
European Journal of Anaesthesiology: December 2016 - Volume 33 - Issue 12 - p 949-951
doi: 10.1097/EJA.0000000000000536
  • Free


Thoracic paravertebral block (PVB) is used for breast surgery as it induces prolonged ipsilateral block of several thoracic afferent nerves and their corresponding dermatomes. We describe the first case of total spinal anaesthesia after ultrasound-guided PVB. Written informed consent was obtained from the patient to publish this report.

A 48-year-old woman (63 kg, 170 cm) was scheduled to undergo mastectomy for carcinoma of the left breast.

For the PVB, the patient was placed in the right lateral position. After ultrasound detection of the paravertebral space at the T3 level, the injection was guided by out-of-plane ultrasound (i.e. the needle was perpendicular rather than parallel to the ultrasound probe). Hydro-localisation using saline 0.9% was performed to show the position of the needle tip. Appropriate needle tip position was assumed because of displacement of the pleura with widening of the intercostal space after injection of the saline. Aspiration of clear liquid was interpreted as being because of the saline used for hydro-localisation. In total, 20 ml of 0.75% ropivacaine (150 mg) was injected and the patient was turned into the supine position. The PVB was performed by a trainee anaesthetist and, unusually, without supervision by a senior anaesthetist.

Five minutes after the injection, the patient complained of headache, general numbness and her blood pressure (BP) was 80/44 mmHg with a pulse rate of 65 bpm. She was placed head down and 500 ml of Voluven (6% hydroxyethyl starch 130/0.4 in 0.9% sodium chloride injection, Fresenius Kabi France, Sevres Cedex, Paris, France) was commenced to infuse over 20 min. Repeated bolus doses of ephedrine totalling 15 mg were administered. Despite correction of haemodynamic parameters (BP: 131/60 mmHg, pulse: 135 bpm), the patient appeared to lose consciousness for about 10 min after the injection and, on neurological assessment, the Glasgow coma score (GCS) was 3: eye, 1/4; visual, 1/5; motor, 1/6). The respiratory rate was 20 breaths min−1 with paradoxical thoraco-abdominal movements and pulse oximetry revealed desaturation to 77%. There were no other signs of major respiratory distress. Because there was now a suspicion of local anaesthetic toxicity, an intravenous infusion of intralipid (20%; Fresenius Kabi France, Sevres Cedex, Paris, France) was commenced, to infuse over 10 to 15 min. Positive pressure face mask ventilation with oxygen (9 l min−1) was commenced and this stimulation resulted not only in recovery of consciousness within about 3 min, but also recovery of effective spontaneous ventilation, and avoided further need for mechanical ventilation. Nasal oxygen was then commenced.

Thirty minutes later, the patient opened her eyes in response to a command and was able to answer questions by blinking. She now presented a normal level of consciousness but there were no verbal or motor responses (GCS 5: eyes, 3/4; visual, 1/5; motor, 1/6). Clinical findings were indicative of total spinal anaesthesia with an upper level now at the 3rd cervical segment (C3). The patient was transferred to the ICU (in a separate building) by ambulance. One hour after the accident, she was haemodynamically stable with a BP of 94/52 mmHg and a pulse rate of 76 bpm, a GCS score still at 5, symmetrical intermediate reactive pupils and a slowly regressing quadriplegia. By 5 h after the injection of local anaesthetic, she had regained full neurological function. During her hospital stay, she experienced disabling headache, nausea and neck stiffness for 5 days. She reported onset the of unilateral tinnitus on day 8 that resolved over 48 h. Although an epidural blood patch was considered, there was considerable anxiety at the prospect of performing one at the T2/T3 level, but the patient also refused consent. In summary, this patient received an accidental intrathecal injection of an unknown volume of 0.75% ropivacaine during a PVB for postoperative analgesia after mastectomy. She made a full recovery with no neurological sequelae.

PVB can be performed by various techniques.1,2 However, according to clinical practice guidelines, ultrasound is ‘probably recommended’ when performing space blocks because it allows more precise anatomical localisation and administration of local anaesthetic.

The complications reported in the literature essentially concern PVB performed without ultrasound guidance and it is reported that accidental epidural or intrathecal extension of a PVB may occur in up to 1% of cases.3,4 Several cases of total spinal anaesthesia have been reported.5,6 When the diagnosis is made rapidly, full recovery can be expected in 3 to 6 h. Jouve et al.6 proposed two main mechanisms for accidental spinal anaesthesia: insertion of a needle through the intervertebral foramen and through the dura, and protrusion of a dural cuff through the intervertebral foramen into the paravertebral space, where it may be encountered by a needle tip. A recent study of 856 patients involving1427 ultrasound-guided PVBs reported six complications (a rate of 0.7%). These included symptomatic bradycardia and hypotension (n = 3), vasovagal episode (n = 1) and evidence of possible local anaesthetic toxicity (n = 2).7 The authors concluded that the routine use of a single-injection, transverse, in-plane ultrasound-guided technique for thoracic PVB is associated with very few complications.7

In our case report, we describe the first case of apparent total spinal anaesthesia in association with out-of-plane ultrasound-guided PVB for breast surgery. Although the PVB was performed by an anaesthetist with reasonable experience in ultrasound-guided blocks, she was in the early stages of learning the PVB technique. A good understanding of ‘in-plane’ (Fig. 1) and ‘out-of-plane’ (Fig. 2) needle guidance techniques is a prerequisite for the safety and success of this block. In our everyday practice, we teach trainee anaesthetists to perform the in-plane technique in contrast to the out-of-plane technique as performed in this case.

Fig. 1
Fig. 1:
In-plane ultrasound-guided technique. Patient lying in the right lateral position. Direction of needle (→). The ultrasound probe is in a transverse position, perpendicular to the longitudinal axis of the spine at the level of the transverse processes. The needle is directed in the plane of the ultrasound beam. Ca, caudal; Cr, cranial; CTL, costo-transverse ligament; Lat, lateral; Med, medial; Pl, pleura; PVS, paravertebral space; TP, transverse process.
Fig. 2
Fig. 2:
Out-of-plane ultrasound-guided technique. Patient lying in the right lateral position. Position of the needle (→). CTL, Costo-Transverse Ligament; PVS, Paravertebral Space; Pl, Pleura; Cr, Cranial; Ca, Caudal. The ultrasound probe is in a sagittal position, parallel to the longitudinal axis of the spine, at the level of the transverse processes. The needle is directed out of the plane of the ultrasound beam. Ca, caudal; Cr, cranial; CTL, costo-transverse ligament; L, left; Pl, pleura; PVS, paravertebral space; R, Right.

When performing the out-of-plane ultrasound-guided technique, it is recommended that the needle tip be continuously visualised by using hydro-localisation as far as the paravertebral space. As it is out-of-plane, only a short section of the needle can be seen and, without hydro-localisation to identify it, this may not be the tip. Thus the needle tip may be situated beyond the field of the ultrasound beam and may therefore extend beyond the paravertebral space, with the risk of entering the subarachnoid space through the intervertebral foramen or of penetrating a lateral projection of the dura mater. Only hydro-localisation indicates the position of the needle tip with any certainty and PVB injections should only be performed after ensuring that the needle tip is within the paravertebral space. This procedure was not followed in the case reported here.

We report a case of total spinal anaesthesia in association with PVB despite the use of ultrasound guidance. It is not the ultrasound guidance per se that makes PVB well tolerated but the experienced user practising regional anaesthesia with appropriate ultrasound guidance. Although the out-of-plane ultrasound-guided technique is not contraindicated, it does require expertise and an understanding of its limitations. As there is the potential for serious complications in association with PVB, we would argue in favour of the in-plane technique. For training in our hospital, only the in-plane approach is taught under senior supervision. Finally, this complication occurred as a result of four separate deviations from normal practice: an unsupervised trainee performing the block, an out-of-plane ultrasound technique, incorrect interpretation of the aspiration test and the injection of the entire dose in one bolus. The development of such a serious complication following a procedure that had become part of our experienced team's everyday practice provided a trigger to review several important issues: our clinical practice, continuing education for recovery ward nurses concerning the complications of regional anaesthesia, how to improve the training of anaesthetists, how to evaluate an anaesthetist's performance and the provision of better patient information.

Acknowledgements relating to this article

Assistance with the case report: none.

Financial support and sponsorship: none.

Conflicts of interest: none.


1. Bouzinac A, Tournier JJ, Delbos A, et al. Interest of ultrasound-guided lateral pectoral nerve block associated with paravertebral block for complete mastectomy pain management. Ann Fr Anesth Reanim 2014; 33:548–550.
2. Marhofer P, Kettner SC, Hajbok L, et al. Lateral ultrasound-guided paravertebral blockade: an anatomical-based description of a new technique. Br J Anaesth 2010; 105:526–532.
3. Naja Z, Lonnqvist PA. Somatic paravertebral nerve blockade. Incidence of failed block and complications. Anaesthesia 2001; 56:1184–1188.
4. Schnabel A, Reichl SU, Kranke P, et al. Efficacy and safety of paravertebral blocks in breast surgery: a meta-analysis of randomized controlled trials. Br J Anaesth 2010; 105:842–852.
5. Piccioni F, Colombo J, Fumagalli L, et al. Inadvertent high central neuraxial block and possible total spinal anaesthesia occurring after nerve stimulation-guided thoracic paravertebral block. Anaesth Intensive Care 2014; 42:270–271.
6. Jouve P, Nathan N, Suzanne F, et al. Accidental spinal anesthesia following intercostal paravertebral block. Cah Anesthesiol 1987; 35:121–124.
7. Pace MM, Sharma B, Anderson-Dam J, et al. Ultasound-guided thoracic paravertebral blockade: a retrospective study of the incidence of complications. Anesth Analg 2016; 122:1186–1191.
© 2016 European Society of Anaesthesiology