Secondary Logo

Journal Logo

Ultrasound-Guided Greater Palatine Nerve Block

A Case Series of Anatomical Descriptions and Clinical Evaluations

Sahar Hafeez, Najmus, MD*; Sondekoppam, Rakesh V., MD; Ganapathy, Sugantha, FRCPC, FRCA; Armstrong, Jerrold E., BSc, DDS, MSc, FRCD(C); Shimizu, Michael, DDS, PhD; Johnson, Marjorie, PhD*; Merrifield, Peter, PhD*; Galil, Khadry A., DDS, PhD*

doi: 10.1213/ANE.0000000000000329
Regional Anesthesia: Research Report

BACKGROUND: Greater palatine nerve (GPN) block is commonly performed for maxillary and palatal anesthesia by using bony landmarks. Ultrasound (US) can be used to consistently identify greater palatine foramen (GPF) as a defect in the bony palate enabling US-guided injections near the foramen.

METHODS: We scanned and injected 16 undissected well-embalmed hemisectioned cadaveric heads after excluding major anatomical malformations. A linear high-frequency hockey stick probe (7–13 MHz) positioned in long axis to the hard palate visualized GPF as a discontinuity in the hard palate. US-guided injections of 0.1 mL India ink were made in an oblique plane. Specimens were dissected immediately after injection, and dye distribution was noted. The success rate of identification of GPF, number of attempts, and number of successful injections were recorded. The technique was evaluated clinically in 7 patients undergoing dental procedures. Five patients had US-guided injections, and 2 patients received US-assisted greater palatine canal blocks.

RESULTS: GPF was successfully identified in 16 hemisectioned heads (n = 16). In 7 of 16 hemisectioned cadaveric specimens (n = 7/16), needle pass was seen on the US and traces of India ink were found within the greater palatine canal and pterygopalatine fossa. In the remaining heads (n = 9/16), the dye was observed in the mucosal tissue of the hard palate anterior to the GPF or in the soft palate. Clinical evaluation reconfirmed successful identification of GPF by US in 6 of 7 patients (n = 6/7). US-guided injections were successful in 6 of the 8 attempted blocks (n = 6/8) with median number (range) of attempts being 2 (1–4). US-assisted injections were successful in 2 patients (n = 2/2).

CONCLUSIONS: US has the potential to successfully locate and characterize GPF in normal and edentulous maxilla. US-guided GPN blocks can be technically challenging. The clinical applicability of US guidance or assistance for GPN block needs further evaluation in a larger sample of patients.

Published ahead of print June 26, 2014.

From the Departments of *Anatomy and Cell Biology, Anesthesia and Perioperative Medicine, and Division of Oral and Maxillofacial Surgery, Western University, London, Ontario, Canada.

Published ahead of print June 26, 2014.

Accepted for publication April 23, 2014.

Funding: None.

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Sugantha Ganapathy, FRCPC, FRCA, Department of Anesthesia and Perioperative Medicine, Western University, Room 3213, LHSC, University Campus, 339, Windermere Rd., London, Ontario N6A 5A5, Canada. Address e-mail to

Greater palatine nerve (GPN) block is commonly performed for maxillofacial, dental, and palatal surgeries to provide anesthesia or for postoperative analgesia in surgeries posterior to canine teeth. The GPN is blocked at its exit from the greater palatine foramen (GPF), which is usually located in the region of the lateral palatal margin close to the third molar. The common description of GPN block involves palpating the GPF as a depression medial to the second or third molar followed by inserting a 27-gauge needle with the bevel pointing toward the palate until contact with the palatal bone is made. The needle tip is “walked off” the bony ledge into the foramen, and anesthetic is deposited after negative aspiration of blood. Sphenopalatine ganglion block can be performed via the GPF similarly, with the needle inserted into the foramen and local anesthetic (LA) deposited after negative aspiration.1 Studies show a success rate of approximately 88% when landmark-based techniques are used.2

The observed rates of failure with landmark-based techniques can be due to the inconsistent location of the GPF between the second molar to behind the third molar or injection into the lesser palatine foramen, which lies just posterior to the GPF. Anatomical studies show the location of the GPF to be opposite to the third molar in approximately 50% of patients.3 The location of the GPF may be further complicated in edentulous patients and those who have undergone previous maxillary or palatal surgeries. Similar problems may be encountered in patients who have had third molar extractions. Variations in anatomy of the foramen such as an anterior direction of opening or bony projections may pose additional problems when inserting the needle into the foramen.

Direct visualization of the GPF using ultrasound (US) before performance of a greater palatine block may improve the safety and efficacy of these blocks. We performed a case study to evaluate the feasibility of using US for locating and characterizing the GPF. After identification, US-guided injections around the GPF were attempted in hemisectioned cadavers. The technique was evaluated for feasibility in 7 patients (11 injections).

Back to Top | Article Outline


After obtaining ethical committee approval, 16 well-embalmed cadaveric hemisectioned heads were selected for the case study from cadavers donated for scientific research. All cadaveric injections were performed in the Department of Anatomy and Cell Biology, Western University. No other consent was necessary for the study. None of the cadavers had any congenital or acquired craniofacial defects. Because the heads were sectioned in a paramedian sagittal plane, one-half of each head was useful because it was necessary to retain enough bony palate for movement of the linear US probe over the surface of the mucosa to be possible. Intact hard palate mucosa was the other criterion of selection. After cleaning the outer surface of the mucosa of the hemisectioned hard palate with a damp towel, an inert transparent US gel was applied over the cleaned mucosal surface.

A linear high-frequency probe (7–13 MHz, Sonosite M-Turbo, SonoSite, Bothell, WA) with a hockey stick configuration was positioned in long axis to the hard palate. The probe was moved laterally until the dental structures were visible as intermittent peaks. The probe then was moved slightly toward the midline until the visualization of the hard palate as a continuous hyperechoic line. The discontinuity in this line closer to the third molar was identified to be the GPF (Fig. 1). We attempted to determine the borders and the direction of the foraminal opening by moving the probe from the lateral to medial side. The probe was finally positioned for the best image of the GPF, after which the needle was inserted in an oblique plane using US guidance watching for tissue and needle movement. The needle tip could not be visualized once it entered the foramen. The oblique plane of needle insertion was used because the space constraints while using US required the probe to be positioned more medially and angled slightly laterally to provide space for needle insertion. The target was to place the needle tip into the opening of the foramen, which required a slight change in the angle of needle insertion to deposit the dye. The zoom feature was used to magnify needle insertion in the oblique plane near the foramen after which black India ink (0.1 mL) was injected into the foramen. Pediatric insulin syringes were used for this purpose. Some tissue movement was observed just outside the foramen as the injection occurred. Each specimen was dissected immediately after the injection to look for traces of ink within the greater palatine canal.

Figure 1

Figure 1

The mucoperiosteum covering the bony lateral wall of the nose was removed with the help of a periosteal elevator. The thin bony plate covering the pterygopalatine fossa and greater palatine canal from the lateral side was carefully fractured using a chisel to expose the inside of the canal.

In the clinical evaluation of the technique, institutional ethics approval (HSREB number 103294, clinical trials registry NCT01870232) and written informed consent were obtained for 7 patients. We identified the defect in the hard palate close to the third molar on palatal scan after which color Doppler scan was used to identify the greater palatine artery seen arising out of the GPF. Patients were in a reclined position with the head resting comfortably on a headrest. Standard monitors (noninvasive arterial blood pressure, pulse oximetry, and electrocardiogram) and supplemental oxygen were used during the procedure. IV sedation was titrated to effect with diazepam and fentanyl by the dentist. The mouth was kept open with a prop. A linear hockey stick probe was positioned in long axis along the palate between the midline and the molars, using an ample amount of sterile US gel. The probe was gradually moved laterally toward the teeth, and the greater palatine artery was identified using color Doppler or color power Doppler. The artery was traced posteriorly until it was seen to enter the GPF (Fig. 2). A 25-gauge 3-cm needle was inserted close to the GPF in oblique plane looking for tissue movements (Fig. 3). One milliliter of 2% lidocaine with 1:100,000 epinephrine was injected while observing deposition of the drug. The blocks were tested by a dentist at 5 and 10 minutes after injection. The blocks were graded as no block, partial block, or complete block on the mesial aspect of the molars and hard palate. Five patients (8 blocks with 3 bilateral injections and 2 unilateral injections) were evaluated for US-guided injections.

Figure 2

Figure 2

Figure 3

Figure 3

The feasibility of US assistance was evaluated in 2 patients (3 greater palatine canal blocks) in whom the location of the GPF was marked on the mucosa using US. The dentist performing needle insertion entered the canal at the premarked site after which 0.5 mL of 2% lidocaine with 1:100,000 epinephrine was injected into the canal. The blocks were tested similar to GPN blocks.

Back to Top | Article Outline


We evaluated 6 female and 10 male hemisectioned skulls making a total of 16 specimens of which 7 were edentulous and the other 9 had normal dentition (Table 1). The hard palate was visible under US as a hyperechoic white line on the US scan, and the GPF was identified as an interruption in this hyperechoic line near the region of the maxillary molars in all 16 specimens (100%; Fig. 1). In 7 of the 16 hemisectioned cadaveric specimens (n = 7/16), mild to moderate traces of black India ink were found within the greater palatine canal and pterygopalatine fossa (Fig. 4). In the other 9 specimens (n = 9/16), although dye injection was in the region of the GPF on US, dye was deposited in the mucosal tissue of the hard palate lying anterior to the GPF. In 2 of these specimens, traces of dye were found in the tissue of the soft palate. The number of attempts required to inject dye was variable because tracking of the needle on the US needed frequent adjustments. Dye was injected in the first attempt in 5 cases (n = 5/16), whereas 3 or more attempts were required for the other cases (n = 11/16). With detailed side-to-side scans, we also observed the direction of the GPF opening which was downward in 15 specimens and anterior in 1 specimen showing a lower margin of bone at the GPF (Fig. 5). We could not differentiate the vessels and the nerves from the soft tissue under US imaging in the cadaveric tissue.

Table 1

Table 1

Figure 4

Figure 4

Figure 5

Figure 5

In the clinical evaluation of the technique, 6 of the 8 attempted blocks (n = 6/8) in 5 patients were successful. The block procedure was abandoned in 1 patient because we were unable to position the probe on the palate. This patient differed had sleep apnea and had a large tongue. In another patient receiving bilateral GPN blocks, the artery was not well visualized on 1 side and therefore the block on that side was abandoned. Patient and block characteristics are summarized in Table 2. The indications for block placement were molar and wisdom teeth extractions. In 1 patient who had previous palatal surgery, >1 arterial pulsation was seen on the palate but only 1 artery was noted to enter the GPF. We experienced difficulty in aligning the probe with the artery and maintaining the position during needle insertion in the initial 2 blocks. The US-guided blocks were successfully performed on first attempt in 2 of the 8 attempted blocks (n = 2/8). In 2 additional patients (n = 2/2) in whom US assistance was evaluated for greater palatine canal injections, a single-needle insertion was required to enter the GPF and all the blocks were successful. One patient found the probe to be bulky and painful.

Table 2

Table 2

Back to Top | Article Outline


Our study shows that US is reliable for identifying the GPF in cadaveric specimens. It is evident from anatomical studies regarding the variable location of the GPF that methods to consistently identify the foramen are sparse. US is one such modality that is used extensively for the performance of nerve blocks due to its ability to consistently identify anatomical structures. US can reliably identify the GPF as a defect in the hard palate as was evident from our cadaveric study. We used hemisectioned cadaveric heads to simulate live human heads. US probe insertion and block technique were performed in a manner making the technique applicable to live human subjects.

The GPN enters the oral cavity through the GPF, accompanied by the palatine artery and vein. Greater palatine nerve block is commonly performed by the dentist/surgeon using the second or third molars as the bony landmark against which the GPF is thought to be located. The local anesthetic is commonly injected into a bony depression adjoining the third molar. Alternately, the surgeon or dentist may infiltrate the entire mucosa of hard palate with the local anesthetic, in order to cover all the branches of the greater palatine nerve. Because the mucosa of the hard palate (except for the region around the GPF) is densely textured, these infiltrations may be painful, whereas injections close to the foramen may be less painful due to the presence of loose connective tissue.4

A possible explanation for the high failure rate in the cadaveric component of our study is probably related to the embalming of the cadavers. The loss of natural suppleness of the mucosal tissue in all of the scanned specimens probably contributed to higher failure rates causing difficulty in needle maneuverability within the hardened mucosal tissue around the GPF. US scanning of the foramen also identified anatomical variations in the direction of the opening, such as a bony ledge partially covering the opening of the foramen (Fig. 5). Human anatomical studies have shown the incidence of anterior opening of the foramen to be approximately 9% to 18%,5,6 which is higher than the finding of 1 in 16 cases in our study. This may be due to the lower number of specimens studied. US can identify the direction of the GPF, thereby identifying those cases where a greater palatine canal approach for greater palatine nerve block or maxillary anesthesia may be difficult to perform.

From the clinical component of the study, certain pertinent points need to be stressed. It is challenging to optimally insert and position even the smallest available US probe in the mouth. Probe insertion may be impossible if the oral configuration does not permit it as noted in one of our patients. The limiting factors may include a high arched palate, large tongue, limited mouth opening, and morbid obesity. Inability to see the artery may lead to difficulty in confirming the GPF as noted in one of our blocks where on identifying the GPF as a bony defect, arterial pulsations could not be visualized and hence the block was abandoned. The average block performance time in our study was 5 minutes with an average number of attempts being approximately 2, whereas a trained dentist may accomplish a successful block in <2 minutes.

The promise with US in this area might be in facilitating greater palatine canal blocks and sphenopalatine blocks with preprocedural scanning as noted in the 3 greater palatine canal injections in our study. US assistance for these blocks may result in fewer punctures in a sensitive area. The success rate of the technique might improve with increased experience and the availability of high-frequency probes with suitable curvature to apply over the curvature of the palate.

One of the limitations of our study was that it was not randomized. We needed to evaluate the feasibility before launching prospective randomized studies. The numbers of patients in this pilot study are too small to be analyzed statistically. Although not all clinically performed blocks were successful (6/8 attempted injections were successful), the success rate of GPN block can be increased with an increase in anesthetic volume because flooding the vicinity of the GPF with LA can also result in anesthesia without the needle entering the canal. Because the needle tip could not be visualized once it entered the canal due to the bony margins and the oblique approach of needle insertion, we could not be sure of the exact injection point within the canal, but injection outside of the GPF can be excluded if there is no visualization of the injectate using US. To our surprise, because of the patients’ IV sedation, gagging was not a major problem with probe insertion and movement. These technical issues related to real-time US guided injections suggest that US assisted blocks with pre-procedural localization of GPF may be better suited for clinical application. US may have other applications such as identification of occult submucous cleft palate in children7 and identification and grading of submucosal fibrosis with concomitant use of Doppler8 and may also help to prevent iatrogenic injuries after maxillofacial surgeries. GPF canal injections with US assistance could also be used to reduce bleeding during sinus surgery.9

To conclude, US can identify the GPF reliably in human cadavers and in live subjects and may be useful in guiding the performance of GPN blocks.

Back to Top | Article Outline


Name: Najmus Sahar Hafeez, MD.

Contribution: This author helped in the anatomical component of the study as a master’s degree student.

Name: Rakesh V. Sondekoppam, MD.

Contribution: This author helped in design and conduct of the anatomical and clinical conduct of the study, data analysis, and manuscript preparation.

Name: Sugantha Ganapathy, FRCPC, FRCA.

Contribution: This author is the senior investigator for the study and was crucial for study design, generation of the technique, conduct of the anatomical and clinical component of the study, data collection, data analysis, and manuscript preparation.

Attestation: Sugantha Ganapathy approved the final manuscript and is the archival author for the clinical component.

Name: Jerrold E. Armstrong, BSc, DDS, MSc, FRCD(C).

Contribution: This author helped identify patients and verify the onset of block and performed US-assisted blocks in 2 patients.

Name: Michael Shimiz, DDS, PhD.

Contribution: This author helped identify patients and verify the onset of block.

Name: Marjorie Johnson, PhD.

Contribution: This author helped in study design and facilitated the anatomic component of the study.

Name: Peter Merrifield, PhD.

Contribution: This author was involved in mentoring the master’s student.

Name: Khadry A. Galil, DDS, PhD.

Contribution: This author helped in generation of the idea and facilitated the conduct of the anatomic component of the study.

Attestation: Khadry A. Galil is the archival author.

This manuscript was handled by: Terese T. Horlocker, MD.

Back to Top | Article Outline


1. Wong JD, Sved AM. Maxillary nerve block anaesthesia via the greater palatine canal: a modified technique and case reports. Aust Dent J. 1991;36:15–21
2. Kamath MR, Mehandale SG, Us R. Comparative study of greater palatine nerve block and intravenous pethidine for postoperative analgesia in children undergoing palatoplasty. Indian J Anaesth. 2009;53:654–61
3. Saralaya V, Nayak SR. The relative position of the greater palatine foramen in dry Indian skulls. Singapore Med J. 2007;48:1143–6
4. DuBrul EL Sicher and DuBrul’s Oral Anatomy. 19888th ed Louis, MO: Ishiyaku EuroAmerica, Inc.,:273
5. Wang TM, Kuo KJ, Shih C, Ho LL, Liu JC. Assessment of the relative locations of the greater palatine foramen in adult Chinese skulls. Acta Anat (Basel). 1988;132:182–6
6. Westmoreland EE, Blanton PL. An analysis of the variations in position of the greater palatine foramen in the adult human skull. Anat Rec. 1982;204:383–8
7. Meier JD, Banks CA, White DR. Ultrasound imaging to identify occult submucous cleft palate. Laryngoscope. 2013;123:1285–8
8. Manjunath K, Rajaram PC, Saraswathi TR, Sivapathasundharam B, Sabarinath B, Koteeswaran D, Krithika C. Evaluation of oral submucous fibrosis using ultrasonographic technique: a new diagnostic tool. Indian J Dent Res. 2011;22:530–6
9. Douglas R, Wormald PJ. Pterygopalatine fossa infiltration through the greater palatine foramen: where to bend the needle. Laryngoscope. 2006;116:1255–7
© 2014 International Anesthesia Research Society