CGRP-pathway exposure during the pre-EBP interval in spontaneous intracranial hypotension: an exploratory retrospective case series

Abstract

Background

Given hypothesized involvement of calcitonin gene-related peptide (CGRP) in trigeminovascular pain and the frequent migrainous features of spontaneous intracranial hypotension (SIH), we describe temporal associations between initiation of CGRP-pathway therapies and short-term symptom changes during the interval before image-guided epidural blood patch (pre-EBP) in seven consecutively managed SIH patients. Because this is a small retrospective uncontrolled series, we report exploratory observations only.

Case presentation

Work-up included selective head-up tilt testing and heavily T2-weighted MR myelography demonstrating predominantly cervicothoracic and/or multifocal spinal CSF leaks. During the pre-EBP interval patients received CGRP-pathway agents: monoclonal antibodies (fremanezumab n = 1; galcanezumab n = 3) and/or gepants (atogepant n = 4; rimegepant PRN n = 1). Five of seven patients (71%) met a pre-specified rubric of clinically meaningful improvement documented after CGRP initiation; in all seven patients with routinely charted Numeric Rating Scale (NRS, 0–10) scores the median NRS reduction was 3 points (range 1–4). Positional orthostatic features commonly persisted until after EBP. Median interval from CGRP initiation to targeted EBP was 7 days. In this small cohort we did not observe an obvious signal that pre-EBP CGRP exposure impaired subsequent clinical response to EBP; however, the sample size and short follow-up preclude conclusions about rare or delayed interactions.

Conclusions

These observations are hypothesis-generating only and report temporal associations between CGRP-pathway initiation and short-term symptom changes in a highly selected cohort with comorbid chronic migraine. They do not demonstrate efficacy for SIH and should not be interpreted as endorsing CGRP-pathway therapy as a treatment for SIH. Prospective, controlled trials with validated headache and orthostatic endpoints, stratification by migraine status, and peripheral/CSF biomarker assessments are required.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12883-026-05172-1.

Background

Spontaneous intracranial hypotension (SIH) is a secondary headache disorder caused by spontaneous spinal CSF leakage [1, 2]. Although orthostatic headache is the classic presentation, migrainous features (nausea, photophobia, phonophobia) are frequently present and contribute to diagnostic delay [3, 4]. Current management prioritizes timely leak-directed treatment—most commonly image-guided targeted epidural blood patch (EBP)—but access can be delayed and patients remain symptomatic during the pre-procedure interval [5, 6].

SIH headache is hypothesized to arise from a cascade in which spinal CSF leakage leads to CSF hypovolemia and caudal brain displacement, producing traction on the dura and pain-sensitive structures innervated by trigeminal afferents [7, 8]. Mechanical deformation of these afferents could plausibly promote release of vasoactive neuropeptides including CGRP, substance P, and PACAP-38, contributing to peripheral vasodilation, neurogenic inflammation, and central sensitization [8, 9]. This CGRP-centric pathway remains inferential in SIH because we did not measure peripheral or CSF CGRP in this series; prospective biomarker studies are required to validate the mechanism [10].

CGRP signaling is best established in migraine but may act as a convergent nociceptive pathway across related secondary headache disorders (for example, post-traumatic headache [11], post-dural-puncture headache [12], cluster headache [13], and idiopathic intracranial hypertension [14]). Given the migrainous features common in SIH and the practical delays to EBP, we report exploratory, temporally associated observations of CGRP-pathway exposure during the pre-EBP interval.

Methods / case ascertainment / outcome definitions

This single-center retrospective chart-based case series screened consecutive adult patients (≥ 20 years) managed between July and September 2025 and identified those who: (i) met ICHD-3 criterion 7.2.3 for headache attributed to spontaneous low CSF pressure [15]; (ii) had spinal CSF leak confirmed on heavily T2-weighted MR myelography per the 2023 consensus [6]; and (iii) received ≥ 1 dose of a CGRP-pathway agent during the interval between SIH diagnosis and EBP. This case series was reported in accordance with the CARE (CAse REport) guidelines [16]; the completed CARE checklist is provided as Additional file 1.

Diagnostic reasoning required concordant position-dependent clinical features and MR myelography evidence. Alternative causes (post-dural puncture, traumatic temporal bone CSF fistula, primary autonomic disorders) were excluded by history, imaging, and selective ancillary testing.

HUTT was performed selectively to exclude co-existing autonomic disorders; results were ancillary and did not determine SIH diagnosis or influence timing of CGRP therapy or EBP. HUTT results are reported in case narratives only.

Selection considerations

CGRP-pathway therapies were offered preferentially to SIH patients with chronic migraine and substantial pre-EBP disability; the universal chronic migraine comorbidity in this cohort reflects that clinical selection pathway and constitutes the principal competing explanation for observed pre-EBP symptomatic changes.

During the same study window (July–September 2025), four additional SIH patients underwent EBP without preceding CGRP-pathway bridging. All four were managed under the same diagnostic and leak-directed protocol as the bridged cohort (clinical assessment, ICHD-3 criterion 7.2.3, MR myelography review, and standard conservative measures), but were not subjected to the structured rubric-based outcome abstraction and are not part of the prespecified analytic cohort. CGRP-pathway bridging was not offered to these patients for one or more overlapping operational or clinical reasons: immediate (same-day or next-day) availability of targeted EBP, which left no pre-procedure interval in which bridging would have been clinically relevant; and patient preference against, or reimbursement constraints on, CGRP-pathway therapy. Therefore, no formal outcome comparison between the bridged and non-bridged groups is presented.

CGRP-pathway therapy — exposure definition and agent selection

Agent selection was pragmatic but followed a reproducible hierarchy (reimbursement, prior agent response, anticipated onset favoring oral gepants when rapid relief was desired, route preference, availability). The index exposure was the date of the first administered CGRP-pathway agent during the pre-EBP interval; subsequent switches were recorded and annotated in Table 2. The study is observational and not powered for comparative agent inference.

Table 2.

Patient-level imaging, CGRP-pathway exposure during pre-EBP window, and outcomes

Case	Age range	MR myelography (leak levels)	Pre-EBP CGRP-pathway therapy (dose)	NRS pre-CGRP (0–10)	Orthostatic increment pre‑CGRP	NRS post-CGRP / pre-EBP (0–10)	Orthostatic increment pre-EBP	NRS reduction after CGRP initiation	Time to perceived improvement (days)	EBP level (date)	Early NRS post-EBP (0–10)	Safety issues
1	30s	C2–L2 (circumferential epidural collection)	Fremanezumab 225 mg SC	9	Present	7	Reduced	2	3	T1–T2 (Jul 2025)	1	None
2	40s	C7–T1	Atogepant 60 mg QD	7	Present	4	Persistent	3	2	C5–C6 (Jul 2025)	1	None
3	70s	C7–T1; T9–T11; L1	Atogepant 60 mg QD	7	Present	6	Persistent	1	2	T1–T2 (Jul 2025)	2	None
4	20s	C7–T2 (recurrent)	Galcanezumab 240 mg SC	8	Present	5	Persistent	3	3	C7–T1 (Jul 2025)	1	None
5	70s	C2; C4–T3; T6–L1	Galcanezumab 120 mg SC → Atogepant 60 mg QD	8	Present	5	Reduced	3	2	T1–T2 (Aug 2025)	2	None
6	50s	C6–T1	Atogepant 60 mg QD	9	Present	6	Persistent	3	3	C6–C7 (Sep 2025)	3	None
7	50s	T8–L1	Galcanezumab 120 mg SC; Rimegepant 75 mg PRN	8	Present	4	Persistent	4	2	T12–L1 (Aug 2025)	1	None

Individual case mapping of MR myelography findings, CGRP-pathway exposure (agent, dose, and timing), descriptive NRS values, orthostatic increment coding, time to perceived improvement, EBP level, and early post-EBP course. NRS values were abstracted retrospectively from routine clinical documentation as secondary descriptive metrics and were not protocolized primary endpoints. Early post-EBP NRS values are reported for temporal context and do not isolate CGRP-pathway effects from EBP effects. Concurrent conservative measures were initiated after SIH diagnosis in all patients and are therefore treated as potential confounders rather than analytically separable interventions. All EBPs were image-guided and targeted to regions of suspected leak and/or epidural fluid collection based on imaging findings and procedural considerations

Orthostatic increment = documented worsening of headache on assuming upright posture; ‘Reduced’ = documented attenuation but not resolution; ‘Persistent’ = unchanged positional worsening

CGRP calcitonin gene–related peptide, EBP (image-guided targeted) epidural blood patch, MR magnetic resonance, NRS Numeric Rating Scale (0–10), PRN as needed, QD once daily, SC subcutaneous

Outcome assessment — operational rubric, timing, blinding, adjudication

We operationalized a three-domain rubric:

• Headache intensity: ‘Improved’ = median NRS decrease ≥ 2 points comparing the 24-hour window pre-index exposure vs. the 24-hour window pre-EBP; absent NRS required explicit contemporaneous clinician/nurse documentation of marked reduction.

• Orthostatic increment: ‘Improved’ required explicit documentation of reduced positional worsening or disappearance of a previously recorded orthostatic exacerbation.

• Functional capacity: ‘Improved’ required documented enhancement in upright tolerance, ambulation, reduced PRN analgesia, or discharge readiness.

Case-level ‘clinically meaningful improvement’ = ‘Improved’ in ≥ 2 of 3 domains within the pre-EBP interval after index exposure; improvements could occur on different days within that interval. Two clinicians (C.-Y.Y., F.-C.Y.) independently abstracted outcomes; agent identity (name/dose) was redacted on abstraction forms when feasible; reviewers required time-stamped notes for pre/post comparisons and therefore were not blinded to timing. Disagreements were resolved by consensus; persistent disagreements were adjudicated by a third senior author.

NRS values were abstracted retrospectively as secondary descriptive metrics (medians of 24-hour windows) and are not protocolized primary endpoints.

We pre-specified capture and coding of conservative measures, including recumbency, oral or intravenous hydration, caffeine, abdominal binder use, theophylline, and short-course corticosteroids. Because these measures were initiated after SIH diagnosis in all patients, were not standardized, and could not be analytically separated from CGRP-pathway exposure, they were treated as potential confounders rather than formal analytic variables; this is stated in the Table 2 legend and addressed in the Discussion.

Case presentations

Cohort overview

Seven adults with SIH and comorbid chronic migraine received CGRP-pathway therapy pre-EBP (Tables 1 and 2). Key aggregated findings: median age 52 (range 25–74); 6/7 female; median interval from CGRP initiation to targeted EBP 7 days; 5/7 met rubric for clinically meaningful improvement. Median NRS reduction after CGRP initiation was 3 points (range 1–4). Positional orthostatic features typically persisted until after targeted EBP, while reductions in overall pain/function were often documented earlier. By rubric domain, the headache-intensity criterion (a Numeric Rating Scale reduction of at least 2 points; Table 2) was met in six of seven patients and the orthostatic criterion in two (Cases 1 and 5; Table 2); the composite threshold of improvement in at least two of three domains, which additionally incorporates the functional-capacity domain abstracted from clinical documentation, was met in five of seven. Table 2 provides detailed timelines and annotations.

Table 1.

Baseline demographic, testing, and imaging characteristics (N = 7)

Characteristic	Value
Age, years — median (range)	52 (25–74)
Female sex — n (%)	6 (85.7)
Chronic migraine comorbidity — n (%)	7 (100)
Prior onabotulinumtoxin A exposure — n (%)	3 (42.9)
MR myelography: leak distribution — n (%)
• Any cervical involvement	6 (85.7)
• Any thoracic involvement	7 (100)
• Any lumbar involvement	4 (57.1)
Interval from CGRP therapy start to EBP — median days (range)	7 (5–7)
Duration from SIH symptom onset to CGRP initiation — days, median (range)	14 (7–21)
Prior EBP for SIH — n (%)	1 (14.3) (Case 4)

Baseline characteristics were obtained from seven consecutively managed patients with spontaneous intracranial hypotension and comorbid chronic migraine. The leak distribution categories reflected the involvement of the indicated spinal region on MR myelography. HUTT results are reported in individual case narratives and are not included in this summary table

CGRP calcitonin gene–related peptide, EBP (image-guided targeted) epidural blood patch, HUTT head-up tilt testing, MR magnetic resonance

Patient perspective

Routine clinical notes documented that the pre-EBP interval was highly disabling; patients reported that any symptomatic relief improved upright tolerance and daily function. Prospective patient-reported outcome instruments were not administered in this retrospective series; future studies should incorporate validated patient-perspective measures.

Representative cases

See Table 2 for numeric/timing details; Cases 1 and 4 retained as illustrative vignettes.

Case 1

A patient in their 30s with a history of chronic migraine presented with a new-onset, severely disabling orthostatic headache beginning in July 2025, characterized by immediate worsening on standing or coughing and complete relief when supine. Head-up tilt testing (HUTT) revealed no autonomic dysfunction. MR myelography revealed extensive circumferential epidural fluid collection consistent with CSF leakage spanning the C2 through L2 vertebral levels (Fig. 1). Given the extensive leak and a delay in procedural scheduling, the patient received fremanezumab 225 mg subcutaneously, with partial headache intensity reduction within 72 h, although orthostatic features persisted. A computed tomography (CT)-guided targeted EBP at the T1–T2 level performed 1 week later (July 2025) resulted in substantial additional improvement with resolution of orthostatic symptoms. No CGRP therapy- or EBP-related adverse events were recorded.

Fig. 1.

Representative imaging in spontaneous intracranial hypotension (SIH) with targeted epidural blood patch (EBP). A–B Heavily T2-weighted magnetic resonance myelography showing a circumferential epidural fluid collection with cerebrospinal fluid leakage spanning C2–L2. C Computed tomography image of targeted EBP at T1–T2 with contrast delineating epidural spread (arrow). All imaging was fully de-identified before submission: patient name, medical record number, date of birth, examination date, and institutional headers were removed; no facial features or other directly identifying anatomical landmarks are visible

Case 4

A patient in their 20s with chronic migraine and prior SIH successfully treated with targeted EBP at C7–T1 (2023) re-presented in July 2025 with recurrent orthostatic headache and dizziness. HUTT was normal, and MR myelography confirmed a recurrent leak from C7 to T2. Inpatient galcanezumab 240 mg loading dose produced noticeable headache improvement within 3 days, while the positional component persisted; repeat targeted EBP at C7–T1 five days later resulted in complete symptom resolution, with no adverse events from either intervention.

Discussion

Central interpretive caveat — chronic migraine biology as the dominant competing explanation. The most important interpretive consideration in this series is that all seven participants had comorbid chronic migraine, with medication-overuse headache present in two. CGRP-pathway agents — including the monoclonal antibodies and gepants administered here — are established, guideline-endorsed preventive therapies for migraine; for example, the monoclonal antibody fremanezumab reduces migraine burden and acute medication use in randomized controlled trials of episodic migraine [17]. The pre-EBP partial reductions in overall headache intensity and functional burden observed in our cohort are therefore biologically and temporally compatible with a treatment effect on the comorbid migraine substrate, without any need to invoke an SIH-specific mechanism. We acknowledge this as the leading — not secondary — competing explanation for the temporal associations reported here, and we emphasize that robust attribution to an SIH-specific nociceptive pathway would require prospective study designs that stratify enrollment by migraine comorbidity and, ideally, include a parallel non-migrainous SIH comparator arm. In contemporaneous chart documentation, a baseline migrainous headache component was generally distinguishable from a superimposed positional orthostatic component: the latter typically persisted until after targeted EBP, whereas the former appeared to attenuate following CGRP-pathway initiation. This temporal dissociation is consistent with CGRP-pathway agents acting predominantly on the migraine substrate while the SIH-specific orthostatic component remained unaffected pending mechanical leak closure — an interpretation that reinforces, rather than weakens, the primacy of migraine biology as the dominant competing explanation.

Co-equal alternative explanations

Beyond migraine comorbidity, four further non-causal mechanisms of essentially equal a priori probability must be considered.

First, spontaneous symptom fluctuation in acute SIH is well documented: a systematic review and meta-analysis estimated that conservative management alone is sufficient to resolve SIH symptoms in 28% (95% CI 18–37%) of patients [4]. Our entire cohort lay within the acute phase of SIH, during which CSF dynamics are most markedly disturbed and clinical phenotype evolution is most pronounced [18]; therefore, a 71% rubric-defined partial-improvement rate observed over a 1-week interval in seven patients is fully compatible with the upper bound of natural-history fluctuation alone.

Second, all patients concurrently received some combination of recumbency, oral hydration, and caffeine — interventions widely used in SIH despite an evidence base limited to small uncontrolled series [6]. In the parallel low-CSF-pressure literature on post-lumbar-puncture headache, fluids, bed rest, and caffeine have been formally classified as interventions for which the supporting evidence remains weak and observational [12]. Because these conservative measures were neither standardized nor analytically separable from CGRP-pathway exposure, any independent symptomatic contribution they may have made is methodologically indistinguishable from a putative CGRP effect.

Third, placebo and contextual effects in headache care are substantial. A meta-analysis reported a pooled placebo responder rate of approximately 21%, with higher rates in studies employing a parallel design [19]. This placebo response has shown an increasing temporal trend in more recent migraine prevention trials [20]. In anti-CGRP monoclonal antibody trials specifically, the placebo ≥ 50% responder rate in chronic migraine was 23.6% (95% CI 11.2–38.8%) [21]. Our patients were acutely hospitalized, highly symptomatic, and receiving a novel parenterally administered therapy — a clinical context expected to amplify placebo and expectation effects.

Fourth, the referral and scheduling timing inherent in the bridging-therapy design creates an unavoidable temporal coincidence between CGRP-pathway exposure and the acute-phase natural-history window. The median time from SIH symptom onset to CGRP initiation was 14 days (range 7–21), and the median time from CGRP initiation to EBP was 7 days (range 5–7); both intervals fall within the ≤ 10-week acute-phase window in which the most pronounced spontaneous evolution of CSF dynamics and the orthostatic headache phenotype has been documented [18]. This overlap is an inherent property of any bridging-therapy case series and cannot be resolved by retrospective analysis.

Each of these four alternative explanations is, on the evidence available, individually sufficient to account for the temporally associated partial symptom reductions reported in this series; collectively, together with the dominant migraine-biology explanation, they provide a parsimonious non-CGRP account of the entire observed signal.

Mechanistic framing (speculative)

The biological rationale that motivated CGRP-pathway bridging in our clinical service is as follows: spinal CSF leakage produces CSF hypovolemia and caudal brain displacement, generating sustained traction on meningeal structures densely innervated by trigeminal Aδ- and C-fibers [7, 8]. Mechanical deformation of these afferents may trigger axonal release of CGRP and other vasoactive neuropeptides, promoting dural vasodilation, neurogenic inflammation peripherally, and second-order sensitization within the trigeminocervical complex centrally [8, 9] (Fig. 2). However, this CGRP-centric cascade remains inferential in SIH because peripheral and CSF CGRP levels have not been directly measured in this condition; even in the adjacent disorder of idiopathic intracranial hypertension, published CGRP studies are conflicting [22, 23]. The absence of direct biomarker data is not merely a minor methodological gap — it is a core limitation that keeps the mechanistic rationale at the level of hypothesis and prevents any mechanistic claim from being made on the basis of this case series. A case report of onabotulinumtoxin A for SIH headache provides indirect support for a CGRP-linked symptomatic mechanism, albeit also without biomarker confirmation [24]. Prospective studies incorporating peripheral and CSF CGRP sampling — ideally in supine versus upright conditions and before versus after EBP — are essential to validate or refute this hypothesized pathway [10].

Fig. 2.

Pathophysiological cascade in spontaneous intracranial hypotension (SIH) and therapeutic intervention points. This diagram illustrates a hypothesized pathway by which spinal cerebrospinal fluid (CSF) leakage may contribute to headache generation in patients with SIH. The proposed cascade begins with CSF leakage, often at cervicothoracic or multifocal spinal levels, followed by CSF volume depletion and caudal brain displacement. This stretches the pain-sensitive meninges innervated by trigeminal nerve fibers containing calcitonin gene-related peptide (CGRP) and other neuropeptides. Mechanical stimulation may promote the release of CGRP and other neuropeptides in the dural vessels (peripheral) and trigeminal nucleus caudalis (central), potentially contributing to vasodilation, neurogenic inflammation, and sensitization, which may underlie orthostatic headaches with migraine features. These two therapeutic approaches target different cascade points. An epidural blood patch (EBP; green pathway) provides leak-directed treatment by sealing the leak and restoring intracranial pressure. CGRP-targeted therapies (blue pathway), including monoclonal antibodies (fremanezumab, galcanezumab, eptinezumab, and erenumab) and gepants (atogepant, rimegepant, and ubrogepant), are hypothesized to modulate CGRP signaling and may reduce symptom burden while awaiting EBP, although this mechanism has not been directly demonstrated in SIH. Red inhibition symbols (⊘) mark therapeutic blockade sites

Contribution to the literature

Based on targeted literature searches conducted up to May 2026, we did not identify any prior clinical series describing CGRP-pathway exposure during the pre-EBP interval in SIH. This series provides four descriptive — not efficacy — contributions: (1) a structured real-world description of CGRP-pathway agents used as a short-term bridge before targeted EBP in SIH; (2) a preliminary observation that CGRP-pathway exposure was not temporally associated with apparent attenuation of subsequent EBP response in this small cohort, although the sample size and follow-up preclude definitive conclusions; (3) explicit conceptualization of CGRP-pathway agents as candidate adjunctive modulators of trigeminovascular pain rather than as alternatives to leak-directed care; and (4) suggested methodological elements — standardized orthostatic and headache metrics, time-to-EBP, rescue medication use, and migraine-stratified enrollment — that can be incorporated into prospective, controlled study designs.

Differential diagnostic context

For the general neurology and neuroradiology readership, complementary CT and MRI play important roles in distinguishing spontaneous spinal CSF leak from post-traumatic temporal-bone CSF leak [25, 26]. The former typically manifests as orthostatic headache, is best demonstrated by heavily T2-weighted spinal MR myelography, and is treated primarily by image-guided targeted EBP; the latter presents with CSF rhinorrhea or otorrhea, is characterized by high-resolution temporal-bone CT, and is frequently self-limiting.

Limitations

Several principal limitations temper any causal inference and deserve explicit enumeration. First, the series is small (n = 7), single-center, and retrospective, with heterogeneous CGRP-pathway agents, doses, timing, and concurrent non-standardized conservative measures; accordingly, findings are descriptive and cannot establish causality. Second, outcomes were abstracted from routine clinical documentation rather than prospectively collected with validated headache or orthostatic instruments; although the rubric was operationalized a priori with explicit scoring rules, dual independent blinded abstraction, and prespecified consensus adjudication, the NRS values and rubric scores remain descriptive. Third, universal comorbid chronic migraine — with medication-overuse headache in two patients — is the dominant confounder and, as discussed above, the leading competing explanation for observed pre-EBP symptom changes. Fourth, spontaneous symptom fluctuation, placebo response, and concurrent conservative measures may each have contributed to the temporal associations. Fifth, the short bridging interval (median 7 days, range 5–7) substantially constrains the window within which any sustained pharmacodynamic effect of CGRP-pathway therapy could be observed, particularly for anti-CGRP monoclonal antibodies whose full pharmacodynamic effect plateaus over 4–12 weeks. Finally, we did not measure CGRP levels in plasma or CSF; direct biomarker evidence in SIH is currently absent, and this constitutes a core — not peripheral — limitation.

Clinical implications and future directions

CGRP-pathway therapy should not replace or delay leak-directed evaluation or timely image-guided targeted EBP. Prospective trials should pre-specify validated headache and orthostatic endpoints (including supine versus upright metrics), stratify enrollment by migraine comorbidity, record conservative-care timing and dosing, randomize or time-stratify CGRP-pathway exposure to permit analytic separation from concurrent conservative care, and incorporate peripheral and CSF CGRP biomarker sampling in supine and upright conditions before and after EBP.

Conclusions

In this small retrospective series of SIH patients with comorbid chronic migraine, CGRP-pathway initiation was temporally associated with short-term reductions in overall headache intensity and functional burden during the pre-EBP interval, while positional orthostatic features typically persisted until after targeted EBP. These observations are hypothesis-generating only; no causal inference can be drawn. The observed pattern is most parsimoniously explained by the established effect of CGRP-pathway agents on chronic migraine biology, and is additionally compatible with spontaneous fluctuation in acute SIH, concurrent conservative care, and substantial placebo effects in headache populations. Prospective, controlled trials with validated endpoints, migraine-stratified enrollment, and mechanistic biomarker assessments are required to determine whether these temporal associations reflect any true therapeutic benefit specific to SIH.

Abbreviations

SIH

Spontaneous intracranial hypotension

CGRP

Calcitonin gene–related peptide

EBP

(image guided targeted) epidural blood patch

CSF

Cerebrospinal fluid

IIH

Idiopathic intracranial hypertension

MR

Magnetic resonance

HUTT

Head-up tilt testing

CT

Computed tomography

mAbs

Monoclonal antibodies

NRS

Numeric Rating Scale (0–10)

ICHD-3

International Classification of Headache Disorders, 3rd edition

PACAP-38

Pituitary adenylate cyclase-activating polypeptide-38

PDPH

Post-dural-puncture headache

PTH

Post-traumatic headache

CARE

CAse REport

Authors’ contributions

FCY collected clinical data; KHC and YCH performed the image-guided epidural blood patch procedures; CYY and FCY drafted the manuscript and interpreted clinical data; YCA and IFY joined the discussion and helped data interpretation; FCY supervised the project; all authors critically revised the manuscript and approved the final version.

Funding

This project was supported in part by grants from the Tri-Service General Hospital, Taiwan (TSGH-D-114105 and TSGH-D-115115); Cheng Hsin General Hospital, Taiwan (CHNDMC-114-03 and CHNDMC-115-12).

Data availability

The datasets generated and analyzed during the current study are not publicly available because they contain individual-level clinical information and sharing them openly could compromise participant privacy. De-identified data are available from the corresponding author upon reasonable request, subject to Institutional Review Board approval and applicable data use requirements.

Declarations

Ethics approval and consent to participate

This retrospective case series was approved by the Institutional Review Board of Tri-Service General Hospital, Taiwan, which included a waiver of written informed consent for study participation and retrospective data use given the retrospective, observational design. The study was conducted in accordance with the Declaration of Helsinki.

Consent for publication

Written informed consent for publication of anonymized clinical details and imaging was obtained from all participants. To further protect anonymity, exact ages and individual-level sex information have been removed from the case descriptions and patient-level table; ages are presented only as broader age ranges. All presented imaging has been de-identified: the MR myelography and CT images contain no facial features, names, dates of birth, medical record numbers, or other directly identifying information, and other clinical details have otherwise been de-identified throughout, with only outcome data and clinically important differences reported. Documentation of written consent for publication is retained by the corresponding author and can be provided to the journal upon request.

Competing interests

The authors declare no competing interests.