Regional anaesthesia offers a suitable alternative to local infiltration anaesthesia for facial soft-tissue surgery [1,2]. Regional anaesthesia also presents several advantages over general anaesthesia including smoother recovery, less side-effects, residual analgesia into the postoperative period, earlier discharge from the recovery room and reduced costs. In addition, it avoids the risk of tracheal intubation in patients assumed to have a full stomach in emergency procedures. We have retrospectively analysed patients managed under regional anaesthesia in our institution from July 2003 to June 2004 for facial soft-tissue surgery including plastic procedures, small tumour resection and lacerations repair.
Patients were asked at hospital admission for their consent to have surgery performed under regional block of one to the sensory branches of the 5th cranial (trigeminal) nerve. The ophthalmic nerve provides sensation to eyes and forehead, the maxillary nerve to mid-face and upper jaw and the mandibular nerve to tongue and lower jaw. Regional anaesthesia was carried out by blocking sensory fibres from these three major branches. One or more blocks were inserted unilaterally or bilaterally depending on the anatomical area involved in the surgical field. A supraorbital block was accomplished by palpating the supraorbital foramen, inserting a 30-mm long, 25-G needle perpendicular to the skin until bone contact without entering the canal to avoid direct nerve injury. For a supratrochlear block, a 30-mm long, 25-G needle was advanced at the eyebrow and nose root junction until bone contact. To block the nasociliary nerve or its branches, a 15-mm, 25-G needle was inserted 1 cm above the inner canthus and directed posterolaterally keeping contact with bone. At a depth of 1.5-cm the needle should be at the anterior ethmoidal foramen. This nerve block may also block the infratrochlear nerve. The infraorbital nerve block was performed by using a percutaneous approach. Depression of the infraorbital foramen was fixed with a finger while a 30-mm, 25-G needle was inserted 1 cm inferior to the foramen. The needle tip was tangentially advanced upward to the estimated location of the foramen. A percutaneous technique was used to perform a mental nerve block, a 30-mm, 25-G needle being inserted inferomedially through the skin. For each block, 3 mL of anaesthetic solution was injected after careful negative aspiration. Blocks were carried out with a mixture of equal volumes of 0.25% bupivacaine and 1% lidocaine with epinephrine 1: 200 000. The nasociliary blocks were inserted without epinephrine to eliminate any risk of retinal artery spasm. Intravenous (i.v.) midazolam was used in small doses (0.03 mg kg−1) just before the blocks were performed. We analysed the following variables which were systematically assessed and recorded: failure to obtain full sensory loss 15 min after injection, tolerance of the surgical procedure estimated by the surgeon on a four-point scale (excellent, good, fair or poor), patient satisfaction after surgery on a four-point scale (very satisfied, satisfied, little satisfied or dissatisfied) and side-effects.
Fifty-nine patients, 34 males and 25 females, mean age 56 yr (range 17-92 yr) had surgery under regional block during the defined period. Thirty-five patients were scheduled for elective surgery including excision of cutaneous tumour (23), cosmetic surgery (10) and surgery on the lachrymal system (2). Twenty-four patients were treated as emergencies for isolated facial lacerations. Ten patients were over 80 yr of age (17%) and 11 were American Society of Anesthesiologists (ASA)-III (19%). Number and type of nerve blocks, failures and side-effects are summarized in Table 1. Depending on the location and the size of injury or lesion, the number of blocks performed on each patient was variable. Nineteen patients had 1 block, 21 patients had 2 blocks, 5 patients had 3 blocks, 6 patients had 4 blocks and 6 patients had 5-8 blocks. Onset of sensory loss occurred 2 min after injection with surgical anaesthesia ensuing within 5 min after injection for all the patients. Mean duration of surgery was 58 min (range 10-150 min). The mean duration of block before recovery of light touch sensation was not assessed. No patient received general anaesthesia but additional infiltration of local anaesthetics was necessary in 13 patients (22%) because of incomplete anaesthesia in the surgical area as judged 15 min following block completion. Only minor side-effects were noticed including diplopia or eye akinesia disappearing as the nasociliary block dissipated. Tolerance of the surgical procedure was considered as fair in 3 patients, good in 32 patients or excellent in 24 patients. Twenty-two patients were very satisfied and 29 were satisfied but 8 patients were only little satisfied because of the unpleasant feeling related to the draping over the face, pain at injection of anaesthetic or discomfort during surgery (before additional anaesthetic infiltration) and postural discomfort.
Our results demonstrated that many facial surgical procedures can be performed with trigeminal nerve peripheral blocks. These blocks are safe and provide good intraoperative conditions. However, the failure rate (22%) remains high for a regional technique even in regard to the total number of blocks carried out (n = 152). In 4 patients, failure to obtain a full sensory loss in the surgical area with a supraorbital block may be explained by ignorance of sensory innervation. The lateral and medial aspect of the upper eyelid and brow and a triangular-shaped area bordered by the zygoma, temporal hair line and lateral orbital rim are, respectively, innervated by the lachrymal and zygomaticotemporal nerve . The lachrymal nerve is a terminal branch of the ophthalmic nerve in the orbit and the zygomaticotemporal nerve is a branch of the maxillary nerve leaving the lateral orbit to emerge in the anterior temporal fossa. These nerves are not blocked by a supraorbital approach  and a surgical procedure in their territory requires additional local anaesthetic infiltration and/or a block of the zygomaticotemporal nerve at the level of the lateral canthus . Similarly, two failures concerning the infraorbital nerve block were related to the lack of nasociliary nerve block for surgery involving the ala and the apex of the nose; this area being innervated by its nasal branches . Finally six partial failures (46%) can be explained by ignorance of facial innervation and the direct failure rate was 12%.
The infraorbital block was associated with the highest rate of failure in our study. Nevertheless, this block is considered as technically easy to perform in other studies where the intraoral route was used [5,6]. We suggest that the higher frequency of insufficient blocks measured with the percutaneous technique was due to inappropriate placement of the needle secondary to our tangential percutaneous approach having regard to nerve distribution. When the infraorbital nerve emerges through the infraorbital foramen, it divides into four groups of branches supplying four different cutaneous areas including the skin of the lateral nose, the lower eyelid, the cheek and the upper lip. These areas matched perfectly with those where we failed to obtain a full sensory loss in our study. This observation suggests a partial blockade of the nerve. The intraoral route is considered to be more reliable than the external percutaneous approach [5,6]. However, a comparative study where bilateral infraorbital nerve blocks were performed using the intraoral technique on one side and the percutaneous technique on the other, failed to find any difference between the two approaches . 12 of the 13 failed blocks in our study were performed by residents. Their reported experience was <5 with the different blocks, whereas trained anaesthetists have carried out at least 20 of each block. The exact number of each procedure required before attaining acceptable failure rates remains unknown and may have a wide inter-individual variability. However, it must be emphasized that the failure rate included only partial failure allowing surgery to be proceeded with single local anaesthetic infiltration.
Peripheral blocks of the trigeminal nerve can achieve safe and effective regional anaesthesia for facial soft-tissue surgery. Although easy to perform, the success of these techniques depends on the experience of anaesthetist and requires a detailed knowledge of facial innervation, especially with overlapping sensory areas of nose apex and orbital structures and surroundings. Regarding rate of failure and side-effects, we recommend particularly the use of supratrochlear, nasociliary and mental nerve blocks.
*Département d'Anesthésie et de Réanimation Hôpital Bellevue, CHRU de Saint-Etienne, France
1. Kays CR. Local infiltration versus regional anaesthesia of the face: case report and review. J S C Med Assoc
2. Eaton JS, Grekin RC. Regional anaesthesia of the face. Dermatol Surg
3. Knize DM. A study of the supraorbital nerve. Plast Reconstr Surg
4. Molliex S, Navez M, Baylot D, Prades JM, Elkhoury Z, Auboyer C. Regional anaesthesia for outpatient nasal surgery. Br J Anaesth
5. Lynch MT, Syverud SA, Schwab RA, Jenkins JM, Edlich R. Comparison of intraoral and percutaneous approaches for infraorbital nerve block. Acad Emerg Med
6. Hanke CW. The tumescent facial block: tumescent local anaesthesia and nerve block anaesthesia for full-face laser resurfacing. Dermatol Surg