Sphenopalatine Ganglion Nerve Block for Postdural Puncture Headache in Obstetrics

The prime goal of this study is to assess the hypothesis that SPGNB is an effective treatment for PDPH.

Post-dural puncture headache (PDPH) is the most common serious iatrogenic complication after labor epidural analgesia. Accidental dural puncture occurs in approximately 0-2% of obstetrical patients receiving epidural analgesia. The risk of PDPH after inadvertent dural puncture with an epidural needle can be up to 70-80%. It is the most common cause of obstetric claims in the American Society of Anesthesiologists Closed Claims Project database after maternal nerve damage, neonatal brain damage and maternal death. This issue can greatly affect maternal quality of life following delivery as well as cause dissatisfaction with the entire delivery. This data likely underestimates the clinical significance of this complication as the majority of PDPH patients do not progress to litigation and hence do not appear in the ASA Closed Claims Project database.

PDPH is classified as severe (frontal or occipital) headache that appears anywhere from 1-5 days after dural puncture with some case reports of headache presenting up to 12 days after dural puncture. The headache must be present within 15 minutes of an upright position, either standing or sitting, with resolution within 15 minutes of lying flat. Associated symptoms may also include neck stiffness, nausea, vomiting, photophobia, tinnitus, and cranial nerve palsies (especially 6th cranial nerve). Pain can range from a mild headache to a severely incapacitating pain. The mechanism of the headache following a dural puncture is related to the intracranial hypotension following the leakage of cerebrospinal fluid from the dural tear. The Monro-Kellie doctrine states that total volumes of brain, cerebrospinal fluid (CSF) and intracranial blood are constant; thus, CSF reduction causes a compensatory cerebral vasodilation. It is this cerebral vasodilatation that causes headache.

Treatment options for PDPH have been extremely varied and to date the most effective treatment is an epidural blood patch (EBP). Some of the other more conservative treatments studied include caffeine, theophylline, sumatriptan, and adrenocorticotropic hormone. Of these, caffeine has been shown in studies to reduce the severity of PDPH with minimal side effects. The rest of these medications do not have RCTs supporting their efficacy. The EBP success rates vary from 61-98%.

EBP is an invasive procedure and complications include back pain, repeat dural puncture (and blood arachnoiditis if not recognized), and epidural abscess and hematoma. However, there are also complications for persistent intracranial hypotension if not performing EBP, including subdural hematoma (due to traction on and tearing of bridging veins) and dural sinus thrombosis (due to the stasis associated with cerebral venodilation exacerbated by the hypercoagulable state of pregnancy). Furthermore, some of the cerebral nerve palsies can result in more permanent deficits such as diplopia or hearing loss.

Sphenopalatine ganglion nerve block (SPGNB) has been used for treating headaches of various causes since 1908 when the first paper was reported by Sluder, an ENT surgeon, using cocaine to treat headache. SPGNB could present an alternative, less invasive treatment of PDPH. The SPG is located under the medial wall of the posterior to the middle nasal turbinate in the pterygopalatine fossa which represents the largest collection of neurons outside the brain. Stimulation of the SPG activates postganglionic parasympathetic neurons, which cause vasodilation and increase cerebral blood flow. The activation of the postganglionic parasympathetic neurons causes release of acetylcholine, nitrous oxide, and vasoactive intestinal peptide in the blood vessels found in the dura. These postganglionic parasympathetic fibers are also linked to the vascular beds of the cerebral hemispheres and help to regulate vascular tone. Block of the SPG will inhibit this response and produce vasoconstriction of the dural and cerebral vessels. The SPG also receives sympathetic innervation from the superior cervical sympathetic ganglion. The easy and non-invasive access to the SPG makes it an ideal target to treat headache.

In current literature the SPGNB has undergone RCT in chronic migraine and tension headache patients. It has been used most effectively at treating cluster headaches and facial pain syndromes including trigeminal neuralgia. In the past decade, its application has been broadened to various other types of headache. In 2014, Cady et al performed SPGNB with 0.5% bupivacaine in a series of 12 blocks with either bupivacaine versus saline twice weekly for 6 weeks in a cohort of 41 chronic migraine patients which found that patients had a statistically significant decrease in pain at 15 minutes, 30 minutes, and 24 hours.

In the obstetric population, only case series have explored the efficacy of SPGNB but no randomized control trials have been performed. In one case series by Kent, three obstetric patients developed PDPH after accidental dural puncture and transnasal SPGNB was performed with 2% viscous lidocaine. This case series reported reduction or elimination of the headache in all the patients including one who had previously failed an epidural blood patch. In another case report, Grant used SPGNB to treat a tension headache in a pregnant patient at 26 weeks. The patient received a total of seven transnasal SPGNBs with 4% lidocaine with improvement in her headache for several weeks. The potential impact of SPGNB for PDPH has far reaching implications including for PDPH, both in obstetric and neurology patients.

The investigator team aims to determine whether SPGNB provides effective pain relief after PDPH as determined by a reduced need for EBP. The team will also explore the duration of effect of SPGNB and EBP in the first month after therapy.

Patient recruitment will occur in the postnatal unit or emergency room, where the patient first presents with PDPH following epidural or CSE for labor or cesarean delivery. The team will endeavor to perform anesthesia ward rounds daily in postnatal wards to assess for possible PDPH. Patients will also be identified who return to labor and delivery or the emergency room with complaints of headache following hospitalization that included epidural or CSE during delivery.

The RCT intervention is sphenopalatine ganglion nerve block (SPGNB). The active drug is 4% lidocaine (LIDO group); placebo control is saline (SALINE group). The sphenopalatine ganglion owing to its location at the posterior wall of the nasopharynx, the SPGNB is readily performed by non-invasive application of topical local anesthetic. Patients are positioned supine with a pillow under their shoulders with their chin raised, so that their neck is extended. Next, 1.5 mL of study drug (4% lidocaine solution or saline) is dripped slowly (over 15 seconds) into each nostril (total 3mL) via a plastic venflon i.v. cannula. The supine, extended-neck position is then maintained for 15 minutes. After that time, the patient is then asked to sit upright to assess the postural headache. If the postural headache is not sufficiently relieved, the procedure is immediately repeated, and may be repeated a third time if needed.

The allocation sequence for the SNOB study is generated using computer-generated random numbers in blocks of ten patients. To reduce predictability of a random sequence, details of the planned restriction by blocking is only provided in a separate document, unavailable to those who enroll participants or assign interventions. Concealment: All patients are given a series of blinded numbered treatment syringes for the SPGNB, prepared by the pharmacy. These will either all be lidocaine 4% (LIDO group) or all saline (SALINE group). In addition there will be a single cross-over rescue syringe, labelled "cross-over rescue", containing the opposite drug, for use prior to EBP. Blinding: The following individuals are blinded to allocation: trial participants, obstetric and anesthesia care providers, outcome assessors, data analysts. A record of allocation is kept in the site pharmacy file which may only consulted in case of medical emergency thought to be possibly related to the anesthesia intervention, or at the judgement of the PI; this will be recorded.

After SPGNB, any patient with inadequate therapy may request to end the study and have an epidural blood patch. On initial enrollment they are advised that all patients requesting an epidural blood patch will be asked to receive an additional cross-over rescue dose of SPGNB which is the opposite of the drug (lidocaine or saline) allocated in their study group. Treatment for up to three 15 min intervals and subsequent assessment is exactly as for the study intervention. If the PDPH is completely relieved by this cross-over medication, the patient will be advised to defer the EBP. If patients refuse the cross-over, this refusal is recorded but their initial request for EBP after the first series of randomized medications is analyzed as our primary endpoint.

The primary outcome is initial success assessed after completion of the first series of SPGNB treatment with the study medication (LIDO vs SALINE) - irrespective of the number of treatments required (1, 2 or 3 applications, ie after 15, 30 or 45 min). There are two possible options at this stage: Adequate analgesia; patient does not wish to proceed to EBP; or Inadequate analgesia; patient wishes to proceed to EBP.

Secondary outcomes: 1) Success or failure of the cross-over dose to treat PDPH and prevent EBP - irrespective of the number of treatments required (1, 2 or 3 applications, ie after 15, 30 or 45 min); 2) Number of repeat SPGNB treatments required for success in either the initial treatment series or the rescue cross-over series; 3) Time to headache on assuming upright posture; 4) Return to function using the Brief Pain Inventory http://www.npcrc.org/files/news/briefpain_short.pdf - using the pain interference section (questions 9 A-G). 5) Standardized questionnaire (after each intervention): following questions - each with a binary answer Yes or No: Head/neck pain, Nausea, Photophobia, Tinnitus, Ear stuffiness, Diplopia. If the answer to head/neck pain is "yes", then there are supplementary questions: a) location, b) postural element to pain, c) time pain appears after upright posture, d) severity assessed by VAS 0-10 (11-point verbal scale), where 0=none and 10=worst imaginable.

Patients with adequate therapy move to the follow up phase of the study. We use the same questionnaire as in secondary outcomes above, and in addition ask if they are back to daily activities assessed by VAS 0-10. Patients at home are assessed by telephone daily for the first week and at 1 month. If symptoms recur at home following successful analgesia after the use of active drug, the patient will be offered to come to the hospital together with her partner or other responsible adult for reassessment and the partner instructed how to administer the nasal drops by SPGNB at home. Active drug will be prescribed to allow up to 10 repeated SPGNB applications per patient. The number of SPGNB applications used at home are recorded in follow up. At any stage the patient is free to request to return to the hospital for EBP.

Pragmatic trial and standard care: Epidural blood patch is performed as standard for each institution. This will typically be 20-22mL blood, inserted at the same site as the epidural dural puncture site, with the patient sitting or lying on her side, with 2hr lying in the supine position after epidural blood patch. Some units may use fluoroscopy, most probably will not. As this is a pragmatic trial, apart from the administration of SPGNB with lidocaine 4% or saline, all other decisions about analgesic and other management will be made by the treating physician and by patient request according to standard care protocols. The intention of pragmatic trials is to interfere as little as possible with routine care, increasing the applicability of the study findings to the real care setting. Thus oral fluid intake, caffeine management, over-the-counter analgesics (paracetamol and ibuprofen) and narcotics (typically percocet) will follow standard clinical management for that medical center in both groups. All drugs used are recorded. Standard care typically includes caffeine (typically 300mg in morning, possibly repeated once at lunch time), acetaminophen (typically prn 1g q 8hrs), ibuprofen (typically prn 400mg q 8 hrs after meals), increased prn oral fluid intake and rarely an opioid (second line therapy), such as oxycodone (or percocet - oxycodone with acetaminophen), typically prn 2.5-5mg q 4hrs). Drugs administered are recorded. Standard care may vary between institutions.

Interim sample size analysis will be preformed after the completion of 40 patients.

Statistical assessments: The primary endpoint (initial success) and secondary outcome #1 are assessed using chi-squared test; the number of subsequent repeat SPGNB treatments required will not be assessed inferentially, only descriptively; time to pain on assuming upright posture will be compared between the treatment and blinded control using t-test or Mann Witney test depending on normal distribution of data; return to function using the Brief Pain Inventory (questions 9 A-G) will be compared at days 1, 7 and 30 with respect to pre-intervention; data will be compared using repeated measures analysis of variance to compare those patients who ultimately received an epidural blood patch versus those patients who opted not to get a blood patch due to successful SPGNB (irrespective of whether this was in the treatment or cross-over attempt).