Study design and population

Sixty-three pregnant women between the ages of 21 and 40 years old with a body mass index of less than 35 kg/m2 and an American Society of Anesthesiology physical status II (ASA II) who were admitted to Obstetric Hospital at Zagazig University Hospitals for an elective cesarean section under spinal anesthesia between June 20, 2021, and June 30, 2022, and diagnosed with PDPH during their post-partum period with visual analog score (VAS) ≥ 4 (Corbey et al. 1997) and modified Lybecker score ≥ 2 (Lybecker et al. 1995) were included in this prospective double-blinded randomized controlled clinical study. Written informed consent was obtained from all participants before their inclusion in this trial, the institutional review board (Research Ethics Committee of the Faculty of Medicine, Zagazig University) authorized this study with the reference number (ZU-IRB#:6868/16–6-2021), and it is identified as (NCT04844229) on ClinicalTrials.gov.

The following individuals were excluded from the study: those who underwent an emergency cesarean section, those with an inadequate temporal window, convulsions, atrial fibrillation, a history of migraines or persistent headaches, hypertensive problems of pregnancy, a cerebrovascular accident, or any condition that would prevent receiving a subarachnoid block, such as coagulopathy, infection at injection site, or history of local anesthetics allergy as well as those who rejected to participate in our trial.

The preoperative visit comprised a comprehensive history taking, physical examination, and review of laboratory testing to rule out any conditions that would pose a contraindication for the study treatments. Each participant provided written informed consent after discussions on the objectives, advantages, and possible disadvantages of the research treatments. In the operating theatre, after securing an 18G intravenous access and attaching the required monitors (electrocardiogram, pulse oximetry, non-invasive blood pressure, and capnograph), an intravenous fluid co-load of 15 mL/kg Lactated Ringer’s solution was given to all individuals. Spinal anesthesia was administered while the patient was seated, and strict aseptic procedures were followed. The procedure was performed in the L3/4 or L4/5 intervertebral spaces using a disposable Quinke spinal needle of 25-gauge by paramedian approach after local anesthetic infiltration of the skin with 3 ml lidocaine 2%. Then, intrathecal injection of anesthetic drugs (12.5 mg hyperbaric bupivacaine 0.5% (2.5 mL) + 25 µg fentanyl) was done after CSF free flow over a 10-s period with no barbotage. For all subjects, spinal anesthesia was done by an anesthetist not participating in this study.

In the post-anesthesia care unit and for the next 5 days during the post-partum period, all patients were questioned and clinically examined twice daily for headaches. The diagnosis of PDPH was determined using the four International Classification of Headache Disorders (ICH-II) guidelines which include a postural headache develops within 5 days after lumbar puncture, worsens within 15 min of sitting or standing, and improves within 15 min of lying down, associated with at least one of these criteria: neck stiffness, nausea, photophobia, and tinnitus, and resolves either spontaneously within 1 week or within 48 h after effective CSF leak treatment. Classically, the patient often reports dull aching, throbbing, or pressure-type fronto-occipital headache (Jabbari and Hasanjani Roushan 2014).

Patients suffering from PDPH during the post-partum period were asked to report their headache severity after 15 min of sitting upright, using the 10-cm visual analog score (VAS) where score 0 is no headache and 10 is the worst headache conceivable (Corbey et al. 1997). Additionally, headache severity was evaluated using a modified Lybecker score (Lybecker et al. 1995) which includes the following:

Grade 1: Mild headache that hardly affects daily activities, no bed-bound patients, no accompanying symptoms, and oral analgesics effectively relieve the pain.

Grade 2: Moderate headaches significantly restrict daily activities and leave patients in bed for most of the day, and they call for injectable analgesia. Other symptoms may or may not be present.

Grade 3: Severe PDPH with accompanying symptoms, complete restriction of daily activities, with patients spending the whole day in bed. Associated symptoms include a feeling of being deaf, tinnitus, dizziness, nausea, and vomiting.

All patients with visual analog score (VAS) ≥ 4 and modified Lybecker score ≥ 2 were enrolled in this study and subjected to baseline trans-cranial Doppler (TCD) evaluation and then were randomly assigned into three equal groups. These patients were hospitalized during the treatment course till the complete cure of PDPH symptoms, then after discharge, they were followed up twice daily by phone call for a total of 5 days from the time of study enrollment.

Randomization was carried out by sealed, opaque envelopes containing random numbers produced by the website (https://www.randomizer.org/).

Control group [Group C (n = 21 participants)]

In addition to increased oral fluid intake and maintained bed rest as part of the conservative PDPH treatment, patients received oral paracetamol 1000 mg/8 h, caffeine 300–500 mg/day (Panadol-Extra tablet, film-coated, GlaxoSmithKline Consumer Healthcare Holdings (US) LLC) (2 tablets were given every 8 h), and a 1000 mL 0.9% normal saline infusion during the first 4 h. Non-steroidal anti-inflammatory drugs (NSAIDs) were added in the form of 30 mg IV ketorolac, which may be repeated every 12 h if required if the aforementioned treatments did not successfully manage pain with the VAS ≥ 4 after 6 h of therapy. The VAS score, modified Lybecker score, and TCD characteristics were assessed after 1, 6, and 24 h. EBP was considered after 24 h of treatment if the pain was still not controlled with VAS ≥ 4 and modified Lybecker score ≥ 2 and after gaining patients’ consent.

Bilateral sphenopalatine ganglion block group [Group S (n = 21 participants)]

This got the same conservative management as in the control group together with bilateral transnasal sphenopalatine ganglion (SPG) block.

SPG block was done with the patient lying in a supine position using 6 in. (15 cm) cotton-tipped plastic hollow applicator inserted in the nose with the swab soaked in 1.5 ml 10% lignocaine solution. The applicator was inserted parallel to the floor of the nose until resistance was encountered. The swab was rested in the pterygopalatine fossa superior to the middle turbinate and removed after 10 min. This procedure was done again in the other nostril as well (Kumar et al. 2020). After 1 h, patients were assessed for severity of headache by VAS score, modified Lybecker score, and other successful block indictors such as bilateral lacrimation and nasal congestion as well as TCD evaluation was performed. Then, the headache severity assessment by the VAS score and modified Lybecker score were repeated at 6 h and 24 h of the block as well as the TCD measurements. If the VAS is still ≥ 4 and modified Lybecker score ≥ 2, EBP was indicated and performed after gaining patients’ consent.

Combined nerve and ganglion block group [Group NAG (n = 21 participants)]

The same conservative management as in the control group together with a combined block of bilateral transnasal SPG block and bilateral ultrasound-guided greater occipital nerve block (GONB) was received.

While the patients were lying prone, bilateral ultrasound-guided greater occipital nerve block was done using a high frequency (6–13 MHz) probe of Siemens Acuson X300 machine placed in transverse orientation lateral to external occipital protuberance parallel to the superior nuchal line to detect occipital artery where the nerve is located medial to it 1.5 in.; then a 20-gauge needle was inserted out of plane to avoid vascular injury. Four milliliters of treatment solution containing 2.5 mg/ml bupivacaine and 1 mg/ml dexamethasone (prepared by adding 2 ml bupivacaine 0.5% + 1 ml dexamethasone + 1 ml saline) was injected on each side. Block was confirmed by ipsilateral anesthesia of the scalp area supplied by greater occipital nerve (Türkyilmaza et al. 2016), and then, these patients were assessed for VAS score, modified Lybecker score, and TCD after 1, 6, and 24 h of the block. If the VAS is still ≥ 4 and modified Lybecker score ≥ 2, EBP was indicated and performed after gaining patients’ consent.

Transcranial Doppler was performed to measure mean flow velocity (MV), and the Gosling pulsatility index (PI) for all included patients in a supine position by using Siemens Acuson X300 ultrasound machine equipped with 2–5 MHz probe. The probe was placed after gel application in the right temporal window which is positioned between the ear and lateral orbital margin above the zygomatic bone on the temporal squma to identify the right middle cerebral artery (MCA), and then, the probe was fixed in the same place and measurements (MV, PI) were repeated for three times with the third reading was taken and recorded. All the measurements were taken by the same anesthesiologist who is experienced in neurosonology (Bathala et al. 2013).

The severity of the headache was assessed by VAS score and modified Lybecker score, as well as TCD measurements were performed at 0 (baseline), 1, 6, and 24 h where 0 is the time of enrollment before receiving any medication, and all other evaluations were performed 1 h after starting the treatment either conservative medications alone or with the block and was repeated at 6 and 24 h after treatment.

Our study is double-blind research. Participants were not aware of their group allocation, and the experienced operator who performed all TCD measurements was not aware of the patient group.

Sample size calculation

The sample size is calculated using (open Epi) program, assuming that the visual analog score (VAS) was (2.2 ± 1.14) in PDPH patients who received medical treatment combined with SPG block versus (4 ± 0.67) in medical treatment group (Yılmaz et al. 2020). The sample was found to be 63 subjects allocated into three groups (21 patients in each group), at a confidence interval of 95% and power of test 80.

Data collection

Patient age, body mass index, preoperative hemoglobin level, intra-operative estimated blood loss, and number of spinal attempts, as well as the time interval between lumbar puncture and PDPH occurrence were recorded.

The VAS score, modified Lybecker score, and TCD measurements at 0 (baseline), 1, 6, and 24 h where 0 is the time of enrolment before getting any medication and all subsequent assessments were carried out after beginning treatment. Additionally, the three groups’ need for EBP were noted and recorded.

Our primary outcome was to explore the analgesic efficacy of the SPG clock either alone or in combination with the GONB in PDPH treatment using the VAS and modified Lybecker scores, and the secondary outcomes include the need for EPB as well as the effects of these interventions on cerebral blood flow using TCD measurements.

Statistical analysis

The information was input into a computer and analyzed using IBM SPSS software, version 20.0 (Armonk, NY: IBM Corp.). The phrase number and percentage were used to convey qualitative data. To establish if the distribution was normal, the Shapiro–Wilk test was utilized. The range (minimum and maximum), mean, standard deviation, and median were used to characterize quantitative data. The results were found to be significant at the 5% level. The chi-square test was performed to compare categorical variables across various groups, and the Monte Carlo correction for chi-square was applied when more than 20% of the cells had an anticipated count less than 5, F: F. to compare between more than two periods or phases, an ANOVA with repeated measurements for pairwise comparisons, a post hoc test (Bonferroni corrected) is used for normally distributed quantitative variables. For quantitative variables with abnormally distributed distributions, apply the Kruskal–Wallis test to compare more than two studied groups. Use the one-way ANOVA test for quantitative variables with normally distributed data to compare more than two groups. P values of 0.05 or less and 0.001 or less, respectively, were considered statistically significant.