Neurolysis of the Sphenopalatine Ganglion for Refractory Facial Pain in a Paediatric Cancer Patient: A Case Report

INTRODUCTION

Facial pain with joint cranial or oral involvement represents a diagnostic and therapeutic challenge in the context of oncologic pathologies – and even more so in paediatric patients – and poses significant clinical challenges resulting from nociceptive somatic, neuropathic and inflammatory mechanisms.^[1] It is essential to note that the characteristics of pain can mimic the symptomatology of any orofacial pain disorder, with descriptors ranging from dull and stabbing to burning and pulling pain, and can refer to any craniofacial structure, either regionally or from a distant site.^[2] This symptom has a profound impact on the quality of life of young patients and treatment adherence as it is associated with vital structures and functions such as speech, feeding and psychosocial well-being;^[3] it is an essential marker in the diagnosis of these oncologic diseases.^[4] The presence of orofacial pain is a considerable symptom, affecting approximately 80% of patients diagnosed with head-and-neck cancer due to perineural invasion.^[1,2]

Although the precise prevalence of facial pain in paediatric patients with oncologic pathologies has not yet been fully defined, its frequent occurrence and significantly debilitating impact are widely recognised in the medical literature, where various therapeutic alternatives ranging from pharmacotherapy (opioids, neuromodulators and non-steroidal anti-inflammatory drugs) to analgesic interventionism (local nerve blocks) are used.

Several studies have extensively investigated sphenopalatine ganglion block as an approach to facial pain management in both adult and paediatric patients.^[5] However, in the context of oncologic diseases, a preference for more advanced or aggressive management, such as neurolytic techniques, has been observed.

In this context, chemical neurolysis of the sphenopalatine ganglion has emerged as a potentially effective strategy for the management of facial pain in paediatric oncologic patients, seeking to provide symptomatic relief through the selective interruption of sensitive nerve fibres in the region, offering a therapeutic alternative in cases where conventional treatment options may be limited or inadequate.

In this article, we present a case report illustrating the successful application of chemical neurolysis of the sphenopalatine ganglion in a paediatric patient with oncologic pathology and facial pain. Through this, we seek to contribute to the current knowledge on comprehensive pain management in paediatric patients with oncologic pathology and highlight the relevance of this technique in this specific clinical context.

CASE REPORT

A 9-year-old girl, weighing 26 kg and measuring 133 cm in height, was referred to the Instituto Nacional de Cancerología in Bogotá, Colombia, with a history of severe right-sided otalgia that had been initially managed as an infectious process, without improvement following treatment with antibiotics, oral analgesics or neuromodulators. At the time of admission, she already had a confirmed diagnosis of embryonal rhabdomyosarcoma, based on imaging and biopsy performed at the referring institution. Magnetic resonance imaging revealed a 57 × 37 × 29 mm multilobulated mass with irregular margins involving the right anterolateral skull base, sphenoid sinus, sella turcica and pterygopalatine fossa, with infiltration into adjacent structures.

After admission, she was treated by the paediatric pain and palliative care unit due to the diversity of symptoms present. The patient reported somatic and neuropathic pain on the right side of her face, with greater intensity in the orbital region. It was associated with increased hearing sensitivity and severe pain intensity of 10/10, according to the numeric rating scale (NRS). This situation limited her ability to eat, speak and sleep.

Pain intensity was assessed using the NRS, which is validated for children older than 8 years. Quality of life was evaluated using the PedsQL™ 4.0 Generic Core scales, a validated instrument for paediatric populations aged 2–18 years. A minimal clinically important difference (MCID) of 4.5 points in the total PedsQL™ score is considered clinically meaningful in this population.^[6]

Medical management was initiated using opioids, with a total morphine equivalent daily dose (MEDD) of 42 mg, combined with a neuromodulator and an anticonvulsant (carbamazepine). Carbamazepine was administered at a dose of 200 mg every 8 h. This strategy did not achieve satisfactory pain control. During hospitalisation, the patient experienced persistent episodes of severe pain that negatively impacted her quality of life. An initial assessment using the PedsQL™ 4.0 scale showed an overall score of 32%, with equal scores in both the physical and psychosocial domains.

A diagnostic sphenopalatine ganglion block was performed using the anterior nasal approach [Figure 1]. The patient was monitored with continuous cardioscope, non-invasive blood pressure and pulse oximetry. In the supine position, a cotton-tipped applicator soaked with 2 mL of 2% lidocaine without epinephrine was introduced into the right nostril towards the posterior nasal cavity, followed by the instillation of an additional 3 mL of the same local anaesthetic through the applicator. The patient experienced complete pain relief within 5 min, sustained for approximately 1 h, confirming the indication for neurolysis.

Close

Figure 1:

Clinical image of transnasal diagnostic sphenopalatine ganglion block. Clinical image showing the insertion of a cotton-tipped applicator soaked in local anaesthetic through the right nostril, directed towards the posterior wall of the nasal cavity to reach the region of the sphenopalatine foramen for diagnostic block. Patient position: supine, head in neutral position.

Export to PPT

Given the severity of the pain, its mixed somatic-neuropathic characteristics, refractoriness to conventional pharmacologic management and the anatomical extension of the tumour to the pterygopalatine fossa, an infrazygomatic approach for chemical neurolysis of the sphenopalatine ganglion was selected to achieve sustained analgesia. Consequently, a percutaneous chemical neurolysis with alcohol was scheduled at the level of the right sphenopalatine ganglion.

The procedure was performed in the interventional fluoroscopy suite under general anaesthesia with standard American Society of Anaesthesiologists monitoring (electrocardiography, pulse oximetry, non-invasive blood pressure and capnography). In the supine position with the head in a neutral position, an infrazygomatic percutaneous approach was selected. Using anteroposterior (AP) and lateral fluoroscopic guidance, the pterygopalatine fossa was localised. A 22-gauge, 10-cm spinal needle was advanced anterior to the mandibular ramus and inferior to the zygomatic arch in lateral projection, progressing to the pterygopalatine fossa. In AP projection, the needle was advanced towards the lateral border of the nasal pyramid.

Needle placement was confirmed using 0.5 mL of water-soluble iodinated contrast (iohexol 300 mg/mL) in both AP and lateral fluoroscopic views, ensuring correct dispersion within the pterygopalatine fossa and ruling out vascular uptake or extrafusal leakage [Figure 2]. No test dose of local anaesthetic was administered before neurolysis, as the diagnostic block had demonstrated adequate analgesia and general anaesthesia was provided. Once proper needle placement and contrast diffusion were confirmed, 0.5 mL of 99% absolute alcohol was slowly injected under continuous fluoroscopic control, verifying adequate spread in the target area without vascular injection. Intraoperative vital signs remained stable throughout the procedure. No peri-procedural corticosteroids or additional local anaesthetics were administered. Nasal care consisted of isotonic saline irrigation to prevent crusting and local irritation. No prophylaxis for neuritis was provided, given the absence of residual neuropathic symptoms or signs postoperatively. The patient woke up from anaesthesia without complications, with complete pain relief. A summary of the patient’s clinical course is shown in Figure 3.

Close

Figure 2:

Fluoroscopic views of chemical neurolysis of the sphenopalatine ganglion. (a) Lateral fluoroscopic projection showing the needle tip (white arrow) advanced through an infrazygomatic percutaneous approach to the right pterygopalatine fossa. The injection of contrast medium confirms appropriate positioning without vascular uptake. (b) Anteroposterior fluoroscopic view displaying the dispersion of contrast within the right pterygopalatine fossa (white arrow), confirming accurate needle placement before the injection of the neurolytic agent.

Export to PPT

Close

Figure 3:

Clinical timeline. A chronological overview of the patient’s clinical management and outcomes, showing baseline pain and quality-of-life scores, pharmacologic interventions, interventional procedures and follow-up results up to 90 days after neurolysis. NRS: Numerical rating scale (0 = No pain, 10 = Worst imaginable pain); SPG: Sphenopalatine ganglion, PedsQL™: Paediatric quality of life inventory, MEDD: Morphine equivalent daily dose.

Export to PPT

Following the procedure, the patient experienced a significant reduction in pain intensity and opioid requirements. Before the intervention, she was receiving a total oral MEDD of 42 mg. Opioid use decreased to 0 mg MEDD within 7-day post-neurolysis, representing a 100% reduction, while carbamazepine was maintained as monotherapy for neuropathic symptom control.

Pain scores decreased from a baseline mean of 10 ± 0 to 4 ± 1.1 at 24 h, 0.5 ± 0.7 at 7 days and 0 ± 0 at 3-month post-procedure. Quality of life, assessed using the PedsQL™ 4.0 scale, improved from 32% at baseline to 73.9% at 3 months, with increases in the physical health subscale from 32% to 93% and in the psychosocial health subscale from 32% to 63%. These changes exceeded the MCID of 4.5 points, indicating clinically meaningful improvement. No pain recurrence or late complications were documented during the 3-month follow-up [Table 1].

The patient’s mother expressed that after the procedure, her daughter was able to eat, sleep and interact with family members with minimal discomfort, which had not been possible for several weeks. She described the intervention as “a turning point in her child’s recovery, providing relief and significantly improving her quality of life.”

Written informed consent for the procedure and for publication of this case report was obtained from the patient’s mother. The institutional review board of Instituto Nacional de Cancerología waived formal review for this single-case report.

DISCUSSION

The sphenopalatine ganglion, located in the pterygopalatine fossa, is one of the four parasympathetic ganglia in the cranial region. In addition to parasympathetic fibres, it carries sympathetic and sensory fibres, gives rise to the nasopalatine nerve, the greater and lesser palatine nerves and branches of the pharyngeal nerve through the maxillary division of the trigeminal nerve.^[7]

Given its complex innervation and strategic anatomical location, the sphenopalatine ganglion has been targeted for pain management in a wide range of conditions, including migraines, cluster headaches, trigeminal neuralgia, post-herpetic neuralgia and oncologic head-and-neck pain. It has also been used to manage post-dural puncture headaches and as an adjunct for haemodynamic control during nasal endoscopic procedures.^[8]

In oncologic pathologies, pain often emerges as one of the earliest and most disabling symptoms, resulting from tumour infiltration of surrounding tissues, perineural invasion, metastasis or local inflammatory mediators.^[9] Given the ganglion’s role in modulating nociceptive and autonomic pathways, sphenopalatine ganglion neurolysis offers a valuable therapeutic option in managing refractory facial pain in paediatric oncology, particularly when standard pharmacologic therapies are insufficient. Among available approaches, three main techniques have been described: Transnasal, transoral and transfacial (infrazygomatic). Of these, the infrazygomatic approach is the most commonly used, although it is the most invasive, allowing direct application of neurolytic agents to the pterygopalatine fossa for more effective and sustained pain control, potentially reducing opioid use and related side effects.^[9]

It is important to distinguish between a diagnostic or therapeutic sphenopalatine ganglion block and a neurolytic procedure. A block involves the temporary interruption of nerve conduction through the administration of local anaesthetic, with or without corticosteroids, providing transient analgesia and diagnostic information. In contrast, chemical neurolysis with absolute alcohol is a destructive intervention intended to produce long-lasting or permanent nerve disruption. While highly effective for sustained analgesia, it carries inherent risks such as irreversible sensory deficits, transient or persistent dysesthesia, epistaxis, vascular injection, chemical neuritis and inadvertent injury to adjacent structures. In particular, the risk of chemical neuritis and vascular injury with alcohol neurolysis warrants caution, as these complications, though uncommon, may result in prolonged morbidity. Compared to pulsed radiofrequency – a non-destructive, reversible technique with a favourable safety profile in paediatric populations, particularly in abdominal neuralgias – chemical neurolysis entails a higher risk of permanent sensory dysfunction. However, the absence of established efficacy of pulsed radiofrequency for head-and-neck cancer pain and its technical limitations under general anaesthesia in paediatric patients limited its use in this case.^[10] Consequently, the decision to perform a neurolytic procedure must be preceded by a positive diagnostic block and balanced against potential complications.

Previous reports have described the use of sphenopalatine ganglion interventions in both paediatric and adult populations. In paediatric patients, Peck et al. (2023))^[5] reported its application for refractory migraines and nononcologic facial pain, while in adult oncology, Peña et al. (2019)^[9] described the use of sphenopalatine ganglion blocks for cancer-related head-and-neck pain, although without neurolytic techniques. However, to our knowledge, no published reports have addressed the use of chemical neurolysis of the sphenopalatine ganglion specifically for oncologic facial pain in paediatric patients. This case contributes to the literature by documenting the first reported experience of chemical neurolysis in this clinical setting, highlighting its potential role in the management of refractory cancer pain in children where conventional strategies fail.

Furthermore, although chemical neurolysis offers prolonged analgesia, alternative, less-destructive modalities exist. Botulinum toxin infiltration has been used in paediatric populations for various conditions, with good tolerability but limited evidence for craniofacial pain and analgesic efficacy comparable to triptans.^[11] Intranasal ketamine represents a promising option for acute pain relief in children, though its role in chronic or oncologic pain remains poorly studied and is primarily supported for intravenous administration. While unavailable at our institution, it could be considered for moderate-to-severe cancer pain as an opioid-sparing agent, although with higher sedation rates as a known side effect.^[12] Pulsed radiofrequency has also been reported in paediatric pain management, though the current evidence is limited to isolated case series, predominantly in abdominal neuralgias, with no validated use for head-and-neck cancer pain.^[10]

In this case, a comprehensive multimodal pain management strategy was initially implemented, including opioids, neuromodulators and anticonvulsants, alongside nonpharmacological interventions such as psychological support, relaxation techniques, sleep hygiene recommendations and caregiver-guided distraction methods. Despite these measures, the patient continued to experience severe, refractory pain with significant functional impairment. Although repeated sphenopalatine ganglion blocks could have been considered, this approach was deemed impractical due to the frequency and intensity of the pain episodes, high opioid requirements and the potential need for multiple anaesthetic procedures in a paediatric oncology patient.

Given the irreversible nature of chemical neurolysis and its potential risks – including sensory deficits, neuritis and inadvertent injury to adjacent structures – a thorough risk-benefit analysis was performed. This considered the severity and persistence of the pain, cumulative opioid exposure, functional impairment and the lack of effective alternative treatments within our institution. A multidisciplinary pre-procedural evaluation, including paediatric psychology and palliative care specialists, was conducted to assess the patient’s psychological and developmental status, coping capacity and family support network. The decision to proceed was made collaboratively with the clinical team and the patient’s family, who were fully informed of the procedure’s risks, benefits and alternatives. Written informed consent was obtained from the patient’s legal guardian. Consequently, chemical neurolysis was selected to provide definitive, sustained pain relief with a single intervention. In broader clinical practice, chemical neurolysis of the sphenopalatine ganglion should be considered in paediatric oncologic patients presenting with severe, refractory facial pain characterised by mixed somatic and neuropathic components, particularly when conventional pharmacological management – including opioids and adjuvant analgesics – fails to achieve adequate symptom control. Additional criteria supporting its indication include anatomical tumour extension towards the pterygopalatine fossa, a positive response to a diagnostic sphenopalatine ganglion block and a significant impact on quality of life, notably in essential functions such as feeding, sleep and social interaction.

In our institution, interventional modalities such as pulsed radiofrequency, intranasal ketamine and botulinum toxin are not available. Therefore, based on institutional experience and the absence of comparable options providing a similar duration of effect, chemical neurolysis with alcohol was selected. Despite its invasive nature, this procedure offered effective, sustained analgesia, allowing opioid withdrawal and significant functional recovery.

It is important to acknowledge that these institutional limitations restrict the generalisability of our findings, as centres with broader interventional resources might prioritise other techniques with different safety and efficacy profiles. Further studies in such settings are warranted to compare long-term outcomes and procedural risks. Nevertheless, the absence of long-term follow-up beyond 3 months represents a limitation of this report. Longer follow-up would be necessary to better characterise the long-term safety profile, possible late complications and the durability of analgesic benefit associated with this intervention in paediatric oncologic patients.

CONCLUSION

This case report demonstrates the successful use of chemical neurolysis of the sphenopalatine ganglion as a promising alternative approach for managing complex facial pain in paediatric patients with head-and-neck oncologic diseases. The intervention provided significant pain relief, reduced opioid requirements and improved the patient’s quality of life, highlighting the potential of neurolytic techniques in cases where conventional therapies may be insufficient. The findings underscore the complexity of facial pain in paediatric oncology and the importance of multimodal approaches tailored to address this challenge comprehensively. While these results are promising, further studies are necessary to confirm the long-term efficacy and safety of sphenopalatine ganglion neurolysis in paediatric oncology populations.