From the *Department of Anesthesiology, Northwestern Feinberg School of Medicine, Chicago, Illinois; †Department of Anesthesiology, Critical Care and Pain Medicine, Section of Neurocritical Care, “Sapienza” University of Rome, Italy; and Departments of ‡Neurological Surgery and §Neurology, Northwestern University Feinberg School of Medicine, Chicago, Illinois.
Accepted for publication August 7, 2013
Funding: There was no funding for this case series.
The authors declare no conflicts of interest.
Address correspondence to Antoun Koht, MD, Departments of Anesthesiology, Neurological Surgery, and Neurology, Northwestern University Feinberg School of Medicine, 251 E. Huron St., F5-704, Chicago, IL 60611. Address e-mail to email@example.com.
Awake craniotomy is the preferred approach for neurosurgical procedures that require intraoperative monitoring of eloquent areas.1,2 The successful anesthetic is based on patient cooperation, gentle surgical maneuvers and the anesthesiologist’s careful management of sedation, hemodynamics, ventilatory variables and interaction with the patient.1 Local anesthetics are often used to anesthetize the surgical field by individually blocking the nerves that provide sensation to the scalp: the supratrochlear, supraorbital, zygomaticotemporal, auriculotemporal, greater and lesser occipital nerves.3 When scalp nerve blocks are used, the total doses of local anesthetics and sedatives are decreased, and patient comfort is maintained.4,5 As a result, this technique reduces some of the most commonly described complications during anesthesia for awake craniotomies, including pain, intraoperative arterial hypertension, hypoventilation, hypoxemia, nausea, and vomiting.3,5
As with any procedure, scalp blocks have potential complications. Local anesthetic toxicity, hypertension, and inadvertent subarachnoid injection have been reported.6 Other possible but rare complications include hematoma formation, infection, and inadvertent intra-arterial injection, which have been reported in patients receiving other peripheral nerve blocks but not in those receiving scalp nerve blocks.7–9 In an article by Osborn and Sebeo,6 facial nerve palsy after scalp blocks, specifically auriculotemporal nerve block, was listed as a possible but not reported complication. The only case report of postoperative facial nerve palsy attributed to scalp blocks was of a patient who received these blocks while under general anesthesia.10 In this article, we describe 7 patients with transient postoperative facial nerve palsy as a complication of auriculotemporal nerve blockade in patients undergoing awake craniotomy.
The IRBs of both institutions approved the publication of this case series.
Over a 1-year period at our 2 institutions, 42 patients received scalp nerve blocks for awake craniotomies (28 at the University of Rome and 14 at Northwestern University); a total of 81 auriculotemporal nerve blocks were performed. Of these patients, 7 experienced transient postoperative facial nerve palsy (1 at Northwestern University and 6 at the University of Rome). At both institutions, the diagnosis of facial nerve palsy was made in patients who were unable to voluntarily raise their eyebrows, wrinkle their foreheads, completely close their eyes, or smile on the affected side at any time after the scalp nerves were blocked. All of the patients had supratentorial brain lesions adjacent to the primary language and motor areas and were classified as ASA physical status either II or III with normal coagulation studies. Standard ASA monitors were placed on arrival to the operating room. At both institutions, the supratrochlear, supraorbital, zygomaticotemporal, auriculotemporal, greater and lesser occipital nerves were blocked using the technique described by Pinosky et al.,11 with 5 mL local anesthetic used to block the auriculotemporal nerve. Table 1 provides a summary of the patients, the blocks received, and any complications. All patients at the University of Rome received 1 µg/kg fentanyl before bilateral scalp nerve blocks and local infiltration of the Mayfield head frame pin sites, using a total of 40 mL of 0.75% ropivicaine. Immediately before both bladder catheter placement and Mayfield head frame positioning, each patient received a bolus of 0.5 mg/kg propofol IV followed by an infusion during the opening phase at 0.1 to 0.5 mg/kg/h (1.6–8.3 µg/kg/min). Spontaneous ventilation and supplemental oxygen via facemask maintained hemoglobin oxygen saturation above 97%; capnography was used to monitor ventilation.
Patient no. 33 (Northwestern University) received 2 L oxygen via nasal cannula, a propofol infusion was started at 50 µg/kg/min and titrated to 75 to 100 µg/kg/min, in addition to 2 mg of IV midazolam given preoperatively. Fentanyl was given IV in 50 µg increments for a total of 100 µg before the scalp nerve blocks. Individual scalp nerve blocks were performed on the surgical side using 20 mL of 0.25% bupivicaine with epinephrine 1:200,000. This was the only patient at Northwestern University who received bupivicaine; all other patients at this institution received a mixture of 6 mL of 1% tetracaine and 30 mL lidocaine 1% with epinephrine 1:100,000. The 13 Northwestern University patients without complications received IV analgesia and sedation with a remifentanil infusion to maintain a respiratory rate of 8 to 12 breaths per minute supplemented with an infusion of propofol 10 to 25 µg/kg/min after standard monitors and supplemental oxygen were applied before the scalp blocks. All 7 patients with new onset of postoperative facial nerve palsy experienced resolution of their neurologic deficits within 24 hours.
In this case series, we describe postoperative facial nerve palsy in patients after awake craniotomy using selective scalp nerve blocks. In a 1-year period, 7 of the 42 patients receiving scalp nerve blocks at our institutions (1 at Northwestern University and 6 at the University of Rome) developed this complication. Specifically, transient facial nerve palsy was seen after 10.7% of the 56 auriculotemporal nerve blocks at the University of Rome and 4% of the 25 auriculotemporal nerve blocks at Northwestern University. Overall, 8.6% of the 81 total auriculotemporal nerve blocks performed were associated with transient facial nerve palsy. This is significant because scalp blocks are considered relatively safe with few reported complications, of which transient facial nerve palsy was reported only once previously after scalp blocks performed under general anesthesia for a craniotomy.10 Therefore, possible etiologies of transient postoperative facial nerve palsy after auriculotemporal nerve block need to be explored, and the technique may need to be refined to avoid such complications.
Adequate local anesthesia to block the nerves supplying sensory innervation to the scalp is essential to provide adequate regional anesthesia during awake craniotomies and can be achieved by 1 of 3 methods.6 The first method is the ring block, where local anesthetic is infiltrated circumferentially around the scalp. The second method is local anesthetic infiltration of the incision line and head-holder pin sites. We do not believe that any cases of transient facial nerve palsy have been associated with either of these methods; however, both techniques require larger doses of local anesthetics and therefore increase the risk of local anesthetic toxicity. These methods may also provide incomplete sensory blockade.5 In contrast to these 2 methods, local anesthetics can be used to individually block the 6 nerves that provide sensation to each side of the scalp, affording a longer block time with less local anesthetic and fewer side effects. In addition, scalp blocks have been shown to be superior to local infiltration in blunting the hemodynamic and stress responses to pinning and reducing postoperative pain.4,12
The auriculotemporal nerve is commonly blocked via the technique described by Pinosky et al.11: the needle is introduced perpendicular to the skin 1 to 1.5 cm anterior to the ear at the level of the tragus, posterior to the superficial temporal artery; 5 mL local anesthetic is infiltrated deep to the fascia and superficially as the needle is withdrawn. The facial nerve trunk is usually 1 cm deep and slightly inferior and medial to the tragus (Figure 1).13 With this anatomical relationship in mind, one can imagine that a low-placed injection or the relatively large volume of local anesthetic typically used could spread to block both the auriculotemporal nerve and the nearby facial nerve, leading to facial nerve palsy. The duration of the postoperative nerve palsy in our cases, <24 hours, support such cause.
A second hypothesis is that facial nerve palsy after auriculotemporal nerve blockade may be related to a nerve injury. The needle used to block the auriculotemporal nerve could mechanically disrupt the nerve itself or its blood vessels, resulting in a reduction or loss of neural blood flow and subsequent neural ischemia.14 Considering the location of the facial nerve in relation to the site of auriculotemporal nerve block, this is unlikely. Similarly, stretching of the facial nerve from properly placed Mayfield pins or surgical retraction are also unlikely. Compression of the nerve from a hematoma, edema, or the pressure of a local anesthetic injection may also cause neural ischemia and injury. Vasoconstriction-induced neural ischemia could potentially result from the addition of epinephrine to the local anesthetic solution, especially in higher concentrations.14 In the case of patient no. 33, 0.25% bupivicaine with epinephrine 1:200,000 was used while epinephrine 1:100,000 was used in all of the other scalp blocks performed at Northwestern University, and no epinephrine was used for the scalp blocks done at the University of Rome. Thus, the role of epinephrine in the development of an ischemic nerve injury in these particular cases is inconclusive.
Direct neurotoxic effects of local anesthetics could be another cause of nerve injury after auriculotemporal nerve blockade. However, the neurotoxic potential of one anesthetic over another is unknown, and there are no clinical studies that relate nerve injury to a particular local anesthetic dose or concentration.14 If facial nerve palsy after auriculotemporal nerve block were due to local anesthetic neurotoxicity alone, sensory deficits lasting longer than the expected local anesthetic duration in the distribution of both the auriculotemporal nerve and the facial nerve would have been observed. In addition, the auriculotemporal nerve was blocked using the same volume of local anesthetic and the same technique, described above, at both institutions, but the local anesthetic and concentration varied. At the University of Rome, 0.75% ropivicaine was used for all scalp nerve blocks, and 6 (10.7%) of the 56 blocks resulted in postoperative facial nerve palsies. In contrast, the only 1 (4%) of 25 blocks at Northwestern University resulted in this deficit. Interestingly, this patient had unilateral scalp blocks using 0.25% bupivicaine with epinephrine 1:200,000 rather than the previously described mixture of tetracaine and lidocaine with epinephrine 1:100,000 used for all other scalp blocks at that institution. It is possible that the difference in the incidence of postoperative facial nerve palsy between the 2 institutions is due to different local anesthetics, concentrations, and doses used for the scalp blocks, although the possibility of operator dependence cannot be excluded.
Patients under general anesthesia or deep sedation are unable to report pain and paresthesias during local anesthetic injection, which may be an indication of nerve injury and a reason to abort the procedure. This could have been a contributing factor in the previously reported case by Harbers et al.10 where the patient was under general anesthesia or in our own cases where deep sedation was used during the scalp blocks. For this reason, the American Society of Regional Anesthesia and Pain Medicine recommends against routinely performing peripheral nerve blocks on deeply sedated patients or those under general anesthesia.15 Performing scalp nerve blocks on patients who are awake or lightly sedated and able to communicate facilitates obtaining a neurological examination after the scalp nerves are blocked and before the start of surgery. If a facial nerve deficit is present at this time, it would likely be attributable to the auriculotemporal nerve block rather than a surgical cause.
In conclusion, scalp nerve blocks can be a useful component of a balanced anesthetic for patients undergoing awake craniotomies, particularly when performed by a physician experienced with this technique. Though relatively safe, care must be taken when performing scalp nerve blocks to decrease the possibility of developing postoperative facial nerve palsy after auriculotemporal nerve blockade. We recommend performing these blocks on patients who are relaxed but not deeply sedated, limiting the local anesthetic volume for auriculotemporal nerve block to 3 mL, and staying above rather than below the level of the tragus. We also suggest that the patient be examined for the presence or absence of facial nerve palsy before the start of surgery when the auriculotemporal nerve is blocked. A larger prospective study may be needed to evaluate the differences between drugs and the preferred location of the injection site.
1. Bilotta F, Rosa G. ‘Anesthesia’ for awake neurosurgery. Curr Opin Anaesthesiol. 2009;22:560–5
2. Shinoura N, Yoshida M, Yamada R, Tabei Y, Saito K, Suzuki Y, Takayama Y, Yagi K. Awake surgery with continuous motor testing for resection of brain tumors in the primary motor area. J Clin Neurosci. 2009;16:188–94
3. Piccioni F, Fanzio M. Management of anesthesia in awake craniotomy. Minerva Anestesiol. 2008;74:393–408
4. Geze S, Yilmaz AA, Tuzuner F. The effect of scalp block and local infiltration on the haemodynamic and stress response to skull-pin placement for craniotomy. Eur J Anaesthesiol. 2009;26:298–303
5. Costello TG, Cormack JR. Anaesthesia for awake craniotomy: a modern approach. J Clin Neurosci. 2004;11:16–9
6. Osborn I, Sebeo J. “Scalp block” during craniotomy: a classic technique revisited. J Neurosurg Anesthesiol. 2010;22:187–94
7. Meyers EF, Ramirez RC, Boniuk I. Grand mal seizures after retrobulbar block. Arch Ophthalmol. 1978;96:847
8. Ben-David B, Stahl S. Axillary block complicated by hematoma and radial nerve injury. Reg Anesth Pain Med. 1999;24:264–6
9. Nseir S, Pronnier P, Soubrier S, Onimus T, Saulnier F, Mathieu D, Durocher A. Fatal streptococcal necrotizing fasciitis as a complication of axillary brachial plexus block. Br J Anaesth. 2004;92:427–9
10. Harbers JB, Beems T, Hoen MB, Bleeker CP. A case of temporary facial nerve palsy after regional anesthesia of the scalp. Anesth Analg. 1998;87:1375–6
11. Pinosky ML, Fishman RL, Reeves ST, Harvey SC, Patel S, Palesch Y, Dorman BH. The effect of bupivacaine skull block on the hemodynamic response to craniotomy. Anesth Analg. 1996;83:1256–61
12. Bala I, Gupta B, Bhardwaj N, Ghai B, Khosla VK. Effect of scalp block on postoperative pain relief in craniotomy patients. Anaesth Intensive Care. 2006;34:224–7
13. Myckatyn TM, Mackinnon SE. A review of facial nerve anatomy. Semin Plast Surg. 2004;18:5–12
14. Liguori GA. Complications of regional anesthesia: nerve injury and peripheral neural blockade. J Neurosurg Anesthesiol. 2004;16:84–6
15. Neal JM, Bernards CM, Hadzic A, Hebl JR, Hogan QH, Horlocker TT, Lee LA, Rathmell JP, Sorenson EJ, Suresh S, Wedel DJ. ASRA practice advisory on neurologic complications in regional anesthesia and pain medicine. Reg Anesth Pain Med. 2008;33:404–15