Clinical characteristics, neurovascular compression, and treatment outcomes in SUNCT and SUNA: A contemporary Japanese consecutive case series

Abstract

Objective

This study aimed to characterize the contemporary clinical phenotype of short‐lasting unilateral neuralgiform headache attacks (SUNHA) in Japan, evaluate the prevalence and clinical relevance of neurovascular compression (NVC), and assess real‐world medical and surgical treatment outcomes.

Background

SUNHA, comprising short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) and those with cranial autonomic symptoms (SUNA), is a rare trigeminal autonomic cephalalgia disorder. Although several large cohorts have been reported in Western countries, data from Asia, particularly Japan, remain limited. Moreover, the clinical significance of NVC and the role of microvascular decompression (MVD) in medically refractory cases have not been fully established, and no standardized operational definition of refractory SUNHA exists.

Methods

We conducted a retrospective consecutive case series of 28 patients with SUNHA (14 SUNCT and 14 SUNA) who were evaluated at a tertiary headache center in Japan between 2011 and 2024. Diagnoses were made according to the International Classification of Headache Disorders, 3rd edition. Clinical features, trigger factors, cranial autonomic symptoms, comorbidities, and treatment responses were systematically reviewed and analyzed. High‐resolution constructive interference in steady‐state (CISS) magnetic resonance imaging was used to assess the NVC. Treatment response was defined as a subjective reduction of ≥50% in the attack burden. Refractory SUNHA was operationally defined as failure of or intolerance to at least two of the three principal preventive drug classes (lamotrigine, gabapentinoids, and topiramate). Surgical outcomes following MVD were evaluated in the refractory cases.

Results

SUNHA accounted for 0.2% of all patients with headache evaluated in this study. The mean age at onset was 53.1 ± 20.9 years, with 17 male and 11 female patients. Right‐sided pain (20/28, 71%) and ophthalmic (V1) trigeminal distribution (20/28, 71%) predominated. No statistically significant differences were detected between SUNCT and SUNA with respect to attack duration, frequency, triggers, or treatment responsiveness. NVC was identified in 17 of 27 evaluable patients (17/27, 63%; 95% confidence interval [CI]: 44.3%–78.5%) and was strongly associated with a history of TN–like pain (12/17 vs. 0/10, p < 0.001). Seven of 28 patients (7/28, 25%; 95% CI: 12.7%–43.4%) met criteria for refractory SUNHA, and all refractory cases (7/7, 100%; 95% CI: 64.6%–100%) demonstrated NVC. Five of seven refractory patients (5/7, 71%) underwent MVD and achieved complete and sustained remission.

Conclusion

In this contemporary Japanese SUNHA cohort, no statistically significant clinical differences were detected between SUNCT and SUNA. NVC was relatively common and appeared to be associated with medical refractoriness. In carefully selected patients with refractory SUNHA and convincing NVC, MVD may provide durable clinical benefit.

Plain Language Summary

Short‐lasting unilateral neuralgiform headache attacks (SUNHA) are rare primary headache disorders that are often misdiagnosed because of their overlapping features with other causes of head and facial pain. In this study, we explored the symptoms, imaging findings, treatments, and outcomes of 28 patients with SUNHA seen in our clinic. Our findings suggest that cases of medication‐resistant SUNHA are often associated with nerve compression by a blood vessel, and that surgical treatment of this compression can provide long‐lasting relief in carefully selected patients.

Plain Language Summary

Short‐lasting unilateral neuralgiform headache attacks (SUNHA) are rare primary headache disorders that are often misdiagnosed because of their overlapping features with other causes of head and facial pain. In this study, we explored the symptoms, imaging findings, treatments, and outcomes of 28 patients with SUNHA seen in our clinic. Our findings suggest that cases of medication‐resistant SUNHA are often associated with nerve compression by a blood vessel, and that surgical treatment of this compression can provide long‐lasting relief in carefully selected patients.

Abbreviations

CASs

cranial autonomic symptoms

CI

confidence interval

CISS

constructive interference in steady state

IBM

International Business Machines

ICHD‐3

International Classification of Headache Disorders, 3rd edition

IQR

interquartile range

MRI

magnetic resonance imaging

MVD

microvascular decompression

NVC

neurovascular compression

PH

paroxysmal hemicrania

SD

standard deviation

SPSS

Statistical Package for the Social Sciences

SUNA

short‐lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms

SUNCT

short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing

SUNHA

short‐lasting unilateral neuralgiform headache attacks

TACs

trigeminal autonomic cephalalgias

TN

trigeminal neuralgia

V1

ophthalmic division of the trigeminal nerve

V2

maxillary division of the trigeminal nerve

INTRODUCTION

Short‐lasting unilateral neuralgiform headache attacks (SUNHA), comprising short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) and those with cranial autonomic symptoms (SUNA), are rare trigeminal autonomic cephalalgias (TACs) characterized by strictly unilateral, brief neuralgiform pain accompanied by ipsilateral cranial autonomic symptoms (CASs). ¹ , ² , ³ , ⁴ Attacks typically last 1–600 s and occur at high daily frequencies, distinguishing SUNHA from other TACs and trigeminal neuralgia (TN). ¹ , ² , ³ Their clinical phenotype has been described in several large Western cohorts, which highlighted a short attack duration, high attack frequency, prominent CASs, and occasional overlap with TN. ² , ⁴ , ⁵ , ⁶ , ⁷ According to International Classification of Headache Disorders, 3rd edition (ICHD‐3), SUNCT and SUNA are classified as primary headache disorders. However, neurovascular compression of the trigeminal nerve has been reported in a subset of patients, raising the possibility that some cases may represent secondary SUNHA‐like headache or trigeminal neuralgia with prominent cranial autonomic symptoms. The nosological distinction between primary SUNHA and secondary forms related to vascular–trigeminal conflict remains controversial.

Although several sizeable SUNHA cohorts have been reported in Europe and North America, ² , ⁴ , ⁵ , ⁶ , ⁷ data from Asia are limited. A recent multicenter Chinese cohort provided important region‐specific insights, ⁸ but evidence from Japan is particularly scarce. We previously reported a clinic‐based series of 20 Japanese patients with SUNCT/SUNA ⁹ ; however, that study did not systematically evaluate neurovascular contact (NVC) using high‐resolution magnetic resonance imaging (MRI), nor did it assess outcomes after microvascular decompression (MVD), and the dataset predates recent advances in diagnostic imaging and management.

Consequently, the contemporary clinical characteristics of SUNHA in Japan, including the prevalence and clinical relevance of NVC, real‐world treatment response patterns, and the role of MVD in refractory cases, remain insufficiently understood. Importantly, although several preventive medications are used internationally for SUNHA, no operational definition of medical refractoriness has been established, creating variability in clinical decision‐making and uncertainty regarding the timing of surgical intervention. ⁴ , ⁵ , ⁶ , ⁷ , ¹⁰ Furthermore, the Japanese SUNHA has not yet been contextualized within broader international data, limiting cross‐ethnic comparisons and global interpretations. To address these gaps, we analyzed a contemporary cohort of 28 consecutive Japanese patients with SUNHA—the largest contemporary Japanese cohort—to characterize their clinical phenotype, assess the prevalence and clinical significance of NVC, and evaluate real‐world treatment outcomes, including surgical intervention. Finally, we further contextualized the Japanese SUNHA within major Western and Asian cohorts to provide region‐specific and internationally comparable insights.

We hypothesized that NVC is a prevalent and clinically meaningful feature among Japanese patients with SUNHA, particularly in those with a history of trigeminal neuralgia–like pain. We further hypothesized that the presence of NVC would be associated with medical refractoriness and favorable surgical outcomes following MVD. These hypotheses were formulated a priori to guide the statistical analyses conducted in this study.

METHODS

Subjects

This study was designed as a retrospective, consecutive case series conducted at the Tominaga Hospital Headache Center, a tertiary referral facility in Osaka, Japan. The clinical records of 28 patients diagnosed with SUNHA, including SUNCT and SUNA, who presented to our center between February 2011 and August 2024, were systematically reviewed. All consecutive patients meeting the ICHD‐3 diagnostic criteria for SUNHA during the study period were included without preselection. All patients underwent structured clinical interviews, comprehensive neurological examinations, and neuroimaging during their initial evaluation. One patient with SUNCT had a non–MRI‐compatible cardiac pacemaker and, therefore, did not undergo MRI; consequently, NVC analyses were performed in 27 evaluable patients (13 SUNCT and 14 SUNA).

The diagnoses of SUNCT and SUNA were established according to the ICHD‐3 criteria, ¹ and all clinical diagnoses were made by certified headache specialists from the Japan Headache Society. Because the study period preceded the publication of the ICHD‐3 criteria, some patients were initially diagnosed under ICHD‐2 or ICHD‐3 beta criteria. All cases were retrospectively reviewed and reclassified to confirm that they fulfilled the ICHD‐3 diagnostic criteria at the time of final analysis. Patients were further classified as having episodic or chronic SUNHA according to the ICHD‐3 definitions, based on attack frequency and remission periods specified in the diagnostic criteria. ¹ The presence of neurovascular compression on MRI was not used as an exclusion criterion. Patients with clear alternative secondary structural causes other than neurovascular contact were excluded; therefore, patients were classified according to clinical criteria rather than imaging findings. Several patients included in our earlier case series were also part of the present cohort; however, this study represents an expanded and fully updated dataset, and all analyses were performed exclusively on the entire group of 28 patients included in this study.

No a priori statistical power calculation was conducted prior to this study because of its retrospective and exploratory design. The sample size was determined by the number of consecutive patients who met the inclusion criteria during the study period, reflecting all eligible cases at our tertiary headache center.

Clinical data collection

Clinical data were extracted from electronic medical records using a predefined standardized data collection form by the study investigators and included demographic characteristics (age at onset, age at first visit, and sex), pain location, headache laterality, attack duration and frequency, CASs, migrainous features (photophobia, phonophobia, nausea, and vomiting), presence of a refractory period, occurrence of interictal pain, and identifiable trigger factors. Information on comorbid headache disorders, smoking and alcohol history, prior diagnoses at referring institutions, and all acute and preventive treatments administered before and after presentation was also recorded. TN‐like pain was defined as paroxysmal, electric shock–like pain in the distribution of one or more trigeminal divisions, typically triggered by innocuous stimuli (e.g., touching, chewing, or brushing teeth) and occurring without prominent cranial autonomic symptoms. This phenotype was distinguished from typical SUNHA attacks by the absence of conspicuous cranial autonomic features, as well as by the presence of clear triggerability and a refractory period characteristic of TN.

NVC was evaluated using high‐resolution constructive interference in steady state (CISS) MRI sequences in 27 patients. NVC was defined as vascular contact with or distortion of the cisternal segment of the trigeminal nerve. Imaging findings were reviewed in a multidisciplinary neurology conference by five experienced clinicians, including one board‐certified neuroradiologist, one neurosurgeon specializing in MVD, and three certified headache specialists. All cases were assessed using a standardized imaging protocol. Final determination of NVC was made by consensus after group discussion. Reviewers were not blinded to the clinical diagnosis or the side of headache at the time of imaging evaluation. Except for one patient who did not undergo MRI due to contraindication (1/28, 4%), there were no missing data for the primary clinical variables analyzed in this study. Therefore, all statistical analyses were conducted using complete case analysis without imputation.

Definition of treatment response and refractory SUNHA

Treatment responsiveness was defined according to the criteria of Williams and Broadley, ⁴ whereby a ≥ 50% subjective reduction in attack burden constituted a meaningful therapeutic response. Treatments followed the Japanese Headache Society Clinical Practice Guideline 2021, ¹¹ which recommends lamotrigine, gabapentin, topiramate, and short‐term intravenous lidocaine for SUNCT/SUNA treatment. Pregabalin and mirogabalin were analyzed together with gabapentin as gabapentinoids because of their shared mechanisms and common use in Japan.

Refractory SUNHA was defined as failure to meet the response threshold despite adequate trials of at least two of the three principal preventive classes (lamotrigine, gabapentinoids, and topiramate) at maximally tolerated or clinically appropriate doses, or discontinuation due to unacceptable adverse effects. Because this was a retrospective study, this definition was applied post hoc using predefined operational criteria based on documented medication trials and clinical responses recorded in the medical charts.

While the national guidelines outline a limited set of first‐line agents, real‐world management frequently involves a wider therapeutic range. International cohorts have similarly reported heterogeneous treatment patterns, including the use of pregabalin, carbamazepine, duloxetine, corticosteroids, and indomethacin, particularly in refractory cases. Notably, previous studies have indicated that no pharmacological treatment is universally effective for this condition ⁴ , ¹² and that many patients undergo multiple sequential medication trials before proceeding to surgical intervention. ¹³ , ¹⁴ To ensure comparability with these studies and reflect contemporary clinical practice, we included medications beyond those explicitly listed in the national guidelines.

In all patients (except one with MRI contraindication), high‐resolution MRI was performed to evaluate the presence of NVC as part of routine structural assessment. Regardless of imaging findings, all patients initially underwent optimized medical treatment. MVD or gamma knife radiosurgery was considered as a surgical option only in patients who met the predefined criteria for refractory SUNHA and exhibited NVC on MRI. Patients without NVC were managed medically, and none required surgical intervention during follow‐up.

Preventive pharmacological therapies in this cohort primarily included lamotrigine, gabapentinoids, and topiramate. Treatment response was evaluated based on the clinical follow‐up records. Given the retrospective design, the follow‐up duration varied among patients and was not standardized; treatments were adjusted according to individual clinical responses, and outcomes were assessed using the most recent available clinical data.

Statistical analysis

Comparisons between the SUNCT and SUNA groups were performed according to variable type and distribution. The normality of continuous variables was assessed using the Shapiro–Wilk test and visual inspection of histograms. Continuous variables that did not significantly deviate from normality (e.g., age variables) are presented as mean ± standard deviation (SD) and were analyzed using an independent samples t‐test (unpaired t‐test). Continuous variables that deviated from normality are presented as median with interquartile range (IQR) and were analyzed using the Mann–Whitney U test.

Categorical variables are presented as counts and percentages. Categorical variables with two categories were compared using the chi‐squared test or Fisher's exact test, as appropriate. Fisher's exact test was applied when more than 20% of the expected cell counts were < 5. For categorical variables with more than two categories (including pain distribution, attack duration, and attack frequency), overall group differences were assessed using omnibus chi‐squared tests (Freeman–Halton extension when applicable).

Clinically relevant dichotomized comparisons were also conducted for selected variables (V1 vs. non‐V1 pain distribution, <60 vs. ≥60 s attack duration, and ≤100 vs. >100 attacks/day) to facilitate clinical interpretability. The association between neurovascular compression and a history of TN‐like pain was evaluated using Fisher's exact test. All statistical tests were two‐tailed, with statistical significance set at p < 0.05. Differences in means were estimated using independent samples t‐tests with Welch's correction, when appropriate, and corresponding 95% confidence intervals (CIs). Statistical analyses were performed using International Business Machines (IBM) Statistical Package for the Social Sciences (SPSS) Statistics version 24.0 (IBM SPSS, Tokyo, Japan).

Ethical approval

The study protocol was approved by the Tominaga Hospital Ethics Committee (approval no. 120154). Owing to the retrospective nature of the study, the requirement for written informed consent was waived in accordance with the institutional policy.

RESULTS

Demographic characteristics

A total of 14,000 patients visited the tertiary headache center between February 2011 and August 2024, of whom 28 (28/14,000; 0.2%) fulfilled the ICHD‐3 diagnostic criteria for SUNHA and were included in the final analysis (Table 1). No patients meeting the inclusion criteria were excluded due to incomplete records or other reasons (Figure 1).

TABLE 1.

Baseline demographic and clinical characteristics of patients with SUNCT and SUNA.

Variable	SUNCT (n = 14)	SUNA (n = 14)	Between‐group difference (95% CI)	p‐Value
Female sex, n (%)	4/14 (29%)	7/14 (50%)	21% (−17% to 55%)	0.440
Age at presentation, years; mean (SD)	60.0 ± 19.8	55.5 ± 16.3	4.5 (−9.6 to 18.6)	0.520
Age at onset, years; mean (SD)	56.1 ± 22.5	50.1 ± 19.5	6.0 (−10.4 to 22.4)	0.470
Chronic course, n (%)	2/14 (14%)	2/14 (14%)	0% (−30% to 30%)	>0.999
Migraine comorbidity, n (%)	1/14 (7%)	2/14 (14%)	7% (−20% to 32%)	0.540
Paroxysmal hemicrania, n (%)	0/14 (0%)	1/14 (7%)	−	0.310
Smoking habit, n (%)	4/14 (29%)	2/14 (14%)	15% (−17% to 45%)	0.370
Alcohol consumption, n (%)	4/14 (29%)	3/14 (21%)	8% (−24% to 39%)	0.670
Family history of migraine, n (%)	1/14 (7%)	3/14 (21%)	14% (−12% to 41%)	0.290

Note: Continuous variables were compared using independent samples t‐test or Mann–Whitney U test. Categorical variables were compared using chi‐squared test or Fisher's exact test. All tests were two‐tailed, and p < 0.05 was considered statistically significant.

Abbreviations: CI, confidence interval; SD, standard deviation; SUNA, short‐lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms; SUNCT, short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing.

FIGURE 1.

Flow diagram of patient selection. Among 14,000 consecutive patients with headache evaluated between February 2011 and August 2024, 28 met the International Classification of Headache Disorders, 3rd edition (ICHD‐3) diagnostic criteria for short‐lasting unilateral neuralgiform headache attacks (SUNHA), including short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) (n = 14) and short‐lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms (SUNA) (n = 14). No patients were excluded after inclusion. Magnetic resonance imaging (MRI) was performed in 27 patients; one patient was excluded from MRI and subsequent neurovascular compression (NVC) analysis due to a cardiac pacemaker. The final cohort for NVC analysis comprised 27 patients.

The cohort comprised 17 male and 11 female patients. The mean age at presentation was 60.0 ± 19.8 years in SUNCT and 55.5 ± 16.3 years in SUNA (mean difference 4.5 years; 95% CI: −9.6 to 18.6). The mean age at onset was 56.1 ± 22.5 years in SUNCT and 50.1 ± 19.5 years in SUNA (mean difference 6.0 years; 95% CI: −10.4 to 22.4). Fourteen patients fulfilled the criteria for SUNCT and 14 for SUNA. Most cases exhibited an episodic pattern (24/28, 86%), whereas four patients met the criteria for chronic SUNHA. Comorbid primary headache disorders included migraine in three patients and paroxysmal hemicrania (PH) in one patient. A history of smoking and alcohol use was reported in six and seven patients, respectively.

Initial diagnosis before referral

Before referral to our headache center, 13 of 28 patients (13/28, 46%) had been diagnosed with TN, eight of 28 (8/28, 29%) with cluster headache (CH), and only three of 28 (3/28, 11%) had been correctly identified as having SUNHA (Table 2). No specific diagnosis was made in five of 28 patients (5/28, 18%). These findings underscore the frequent initial misclassification of SUNHA in clinical practice.

TABLE 2.

Previous diagnoses in patients with SUNCT and SUNA.

Initial diagnosis	SUNCT (n = 14)	SUNA (n = 14)	Total (n = 28)
Trigeminal neuralgia	5 (36%)	8 (57%)	13 (46%)
Cluster headache	5 (36%)	3 (21%)	8 (29%)
SUNHA	2 (14%)	1 (7%)	3 (11%)
Migraine	1 (7%)	1 (7%)	2 (7%)
No specific diagnosis	2 (14%)	3 (21%)	5 (18%)

Note: Previous diagnoses assigned to patients with short‐lasting unilateral neuralgiform headache attacks (SUNCT and SUNA) prior to referral to a tertiary headache center. Data are presented as the number of patients (%), highlighting the frequency of initial misclassification before the correct diagnosis of SUNHA was established.

Abbreviations: SUNA, short‐lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms; SUNCT, short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing; SUNHA, short‐lasting unilateral neuralgiform headache attacks.

Clinical features

Right‐sided pain predominated (20/28, 71%), consistent with previous SUNHA cohort studies (Table 3). ⁶ The ophthalmic (V1) division was the most frequently involved trigeminal territory (20/28, 71%), followed by combined V1–V2 and isolated V2 involvement (Table 3). The attack duration showed substantial variability across the predefined categories, ranging from ≤30 s to ≥120 s (Table 3). The daily attack frequency also varied widely, ranging from fewer than 20 to more than 100 attacks per day.

TABLE 3.

Clinical features of patients with SUNCT and SUNA.

Clinical features	SUNCT (n = 14)	SUNA (n = 14)	Total (n = 28)	p‐Value
Right‐sided pain, n (%)	10 (71%)	10 (71%)	20 (71%)	>0.999
Pain distribution
V1	11 (79%)	9 (64%)	20 (71%)
V2	1 (7%)	2 (14%)	3 (11%)
V1–V2	2 (14%)	3 (21%)	5 (18%)
Omnibus p				0.693
Attack duration (s/attack)
≤30 s	5 (36%)	3 (21%)	8 (29%)
30–60 s	4 (29%)	6 (43%)	10 (36%)
60–120 s	3 (21%)	4 (29%)	7 (25%)
≥120 s	2 (14%)	1 (7%)	3 (11%)
Omnibus p				0.711
Attack frequency (attacks/day)
<20	2 (14%)	3 (21%)	5 (18%)
20–100	6 (43%)	5 (36%)	11 (39%)
>100	6 (43%)	6 (43%)	12 (43%)
Omnibus p				0.865
Median daily attack frequency (IQR)	100 (62.5–187.5)	100 (50.0–150.0)		0.386
Cranial autonomic symptoms
Conjunctival injection	14 (100%)	10 (71%)	24 (86%)
Lacrimation	14 (100%)	9 (64%)	23 (82%)
Rhinorrhea	6 (43%)	4 (29%)	10 (36%)
Nasal congestion	4 (29%)	4 (29%)	8 (29%)
Eyelid edema	4 (29%)	5 (36%)	9 (32%)
Sweating	1 (7%)	1 (7%)	2 (7%)
Flushing	7 (50%)	6 (43%)	13 (46%)
Aural fullness	1 (7%)	0 (0%)	1 (4%)
Migrainous features
Photophobia	4 (29%)	3 (21%)	7 (25%)	>0.999
Phonophobia	3 (21%)	1 (7%)	4 (14%)	0.596
Nausea	2 (14%)	1 (7%)	3 (11%)	>0.999
Trigger factors
Cutaneous triggers	13 (93%)	12 (86%)	25 (89%)	>0.999
Eating	2 (14%)	3 (21%)	5 (18%)	>0.999
Talking	2 (14%)	4 (29%)	6 (21%)	0.648
Tooth brushing	5 (36%)	4 (29%)	9 (32%)	>0.999
Cold wind	3 (21%)	4 (29%)	7 (25%)	>0.999
Interictal pain	4 (29%)	3 (21%)	7 (25%)	>0.999

Note: Clinical characteristics of patients with short‐lasting unilateral neuralgiform headache attacks: a comparison of SUNCT and SUNA. Data are presented as numbers (%) unless otherwise indicated. Categorical variables were compared using Fisher's exact test for 2 × 2 tables. For variables with more than two categories (pain distribution, attack duration, and attack frequency), omnibus p‐values were obtained using the chi‐squared test with the Freeman–Halton extension when appropriate. Continuous variables were compared using the Mann–Whitney U test. Clinically meaningful dichotomized comparisons (V1 vs. non‐V1 distribution, <60 vs. ≥60 s attack duration, and ≤100 vs. >100 attacks/day) showed no significant differences between the groups. p values were not calculated for cranial autonomic symptoms included in the ICHD‐3 diagnostic criteria.

Abbreviations: ICHD‐3, International Classification of Headache Disorders, 3rd edition; IQR, interquartile range; SUNA, short‐lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms; SUNCT, short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing; V1, ophthalmic division of the trigeminal nerve; V2, maxillary division of the trigeminal nerve.

Among the CASs, conjunctival injection and lacrimation were common, particularly in patients with SUNCT. Migrainous features, including photophobia, phonophobia, and nausea, were relatively uncommon. A refractory period was documented in five of 28 patients (5/28, 18%), more frequently among those with SUNCT. Interictal pain was reported in seven of 28 patients (7/28, 25%).

Trigger factors were frequently identified in the study. Cutaneous triggers were the most common, followed by tooth brushing, exposure to cold wind, talking, and eating.

No statistically significant differences were detected between the SUNCT and SUNA groups with respect to headache laterality, trigeminal pain distribution, attack duration, or daily attack frequency (all p ≥ 0.690).

Among the CASs, conjunctival injection and lacrimation were universally present in SUNCT, consistent with the ICHD‐3 diagnostic criteria, and were less frequent in SUNA. No statistically significant between‐group differences were observed for the remaining CASs.

NVC

CISS MRI was performed in 27 of 28 patients (27/28, 96%) and demonstrated NVC in 17 of 27 evaluable patients (17/27, 63%; 95% CI: 44.3%–78.5%) (Table 4). The offending vessel was the superior cerebellar artery in 14 of 17 cases (82%) and the anterior inferior cerebellar artery in three of 17 cases (18%). The side of vascular contact corresponded to the side of SUNHA attacks in all cases. In the present study, venous contact was not considered as NVC. The prevalence of NVC was comparable between the SUNCT and SUNA groups, with no statistically significant between‐group difference.

TABLE 4.

Association between NVC and a history of trigeminal neuralgia–like pain in patients with SUNHA.

Parameter	History of TN‐like pain (+)	History of TN‐like pain (−)
NVC present	12 (SUNCT 6; SUNA 6)	5 (SUNCT 2; SUNA 3)
NVC absent	0	10 (SUNCT 5; SUNA 5)

Note: p‐Value (Fisher's exact test, two‐tailed): < 0.001. Association between NVC and a history of TN‐like pain in patients with SUNHA. Data represent the number of patients. The association was evaluated using Fisher's exact test (two‐tailed), and the corresponding p‐value is provided below the table.

Abbreviations: NVC, neurovascular compression; SUNA, short‐lasting unilateral neuralgiform headache attacks with cranial autonomic symptoms; SUNHA, short‐lasting unilateral neuralgiform headache attacks; SUNCT, short‐lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing; TN, trigeminal neuralgia.

In contrast, a history of TN‐like pain was significantly associated with the presence of NVC. In the SUNCT group, six of eight patients with NVC (6/8, 75%) had a history of TN‐like pain, whereas none of the patients without NVC (0/5, 0%) had such a history. A similar distribution was observed in the SUNA group, in which six of nine patients with NVC (6/9, 67%) had a history of TN‐like pain, while none of the patients without NVC (0/5, 0%) reported such a history (Fisher's exact test, two‐tailed p < 0.001).

Treatment response

Intravenous lidocaine was the most effective short‐term preventive therapy, with 9 of 11 treated patients (9/11, 82%) experiencing substantial improvement during hospitalization (Table 5).

TABLE 5.

Treatment responses to acute and preventive therapies in SUNHA.

Parameter	Drug	+++	++	+	−	≥50% responders	Typical dose a (per dose or per day)
Acute treatment	Subcutaneous sumatriptan b		2	7	0	−	0.3 mg per dose
Short‐term preventive therapy	Intravenous methylprednisolone			4	0	−	1000 mg/day for 3 consecutive days
	Continuous intravenous lidocaine	9		1	1	82%	1–2 mg/min (continuous infusion)
Maintenance preventive therapy	Lamotrigine c	10	5	1		94%	12.5–300 mg/day
	Gabapentin	9	2	1	1	85%	200–2400 mg/day
	Pregabalin	5	1		8	43%	150–450 mg/day
	Mirogabalin	1			3	25%	5–20 mg/day
	Carbamazepine	3	4	3	0	70%	200–800 mg/day
	Topiramate	1	2	3	1	43%	12.5–200 mg/day
	Indomethacin d		1		6	14%	75 mg/day
	Duloxetine	1				100%	20 mg/day

Note: Treatment responses to acute, short‐term preventive, and maintenance preventive therapies in patients with SUNHA. Treatment efficacy was graded as +++ (marked improvement), ++ (moderate improvement), + (mild improvement), or – (no improvement). A clinically meaningful response was defined as a ≥ 50% reduction in attack burden.

Abbreviation: SUNHA, short‐lasting unilateral neuralgiform headache attacks.

^a Acute treatments are expressed as per‐dose administration, whereas maintenance preventive therapies are expressed as total daily doses.

^b In Japan, subcutaneous sumatriptan is available at a dose of 0.3 mg per injection.

^c Lamotrigine was discontinued due to skin rash in two patients, including one with Stevens–Johnson syndrome.

^d In Japan, the maximum approved dose of indomethacin under the national health insurance system is 75 mg/day.

Among the oral preventive agents, lamotrigine showed the highest response rate (15/16, 94%), consistent with previous studies conducted in Western cohorts. Gabapentin was effective in 11 of 13 patients (11/13, 85%), and carbamazepine was effective in 7 of 10 treated patients (7/10, 70%).

Other agents, including pregabalin (6/14, 43%), topiramate (3/7, 43%), and mirogabalin (1/4, 25%), showed modest treatment responses. Indomethacin was administered to seven patients (7/28, 25%) as part of the diagnostic evaluation for PH; aside from one patient with comorbid PH (1/7, 14%), no meaningful therapeutic responses were observed. In this patient, two distinct types of periorbital headache were identified based on differences in attack duration, frequency, and treatment response, with PH attacks resolving completely with indomethacin, while SUNA attacks showed only partial improvement and subsequently resolved with lamotrigine, consistent with the ICHD‐3 diagnostic criteria.

Refractory SUNHA and surgical outcomes

Seven of 28 patients (25%, 95% CI: 12.7%–43.4%) met our criteria for refractory SUNHA, defined as an inadequate response to or intolerance of at least two of the three principal preventive medication classes: lamotrigine, gabapentinoids, and topiramate. All refractory patients (7/7; 100%, 95% CI: 64.6%–100%) exhibited NVC on MRI. Five patients subsequently underwent MVD at our institution and achieved complete and sustained remission during a postoperative follow‐up period of 18–135 months (median 42 months). One patient who had previously undergone MVD at an outside hospital experienced only limited improvement and later did not respond to gamma knife radiosurgery. Another patient declined both MVD and gamma knife treatment and instead received a stellate ganglion block, which proved ineffective. Detailed individual clinical characteristics and surgical outcomes are provided in Table S1.

DISCUSSION

This study provides the largest and most up‐to‐date consecutive case series of SUNHA in Japan, offering a comprehensive assessment of clinical features, neurovascular anatomy, treatment responses, and surgical outcomes. Given the rarity of SUNHA and the limited availability of detailed Japanese data, our findings provide important region‐specific information and enable direct comparison with major Western and Asian cohorts. ² , ⁴ , ⁸ , ⁹

A major finding of this study was the high rate of diagnostic misclassification prior to referral. Only three of 28 patients (3/28, 11%) were correctly diagnosed with SUNHA at the referring institutions, with most initially diagnosed with TN or cluster headache. This mirrors observations from Western cohorts and highlights that, despite the clear ICHD‐3 criteria, SUNHA remains challenging to recognize in routine clinical practice. ¹ , ² , ⁴ , ⁵ The combination of very short‐lasting unilateral pain, high daily attack frequency, ophthalmic (V1) trigeminal involvement, and ipsilateral cranial autonomic symptoms can closely mimic both classical TN and other TACs, underscoring the need for increased awareness.

The demographic profile of our cohort, characterized by a mean age at onset in the early 50s and a slight male predominance, is broadly consistent with that reported in major Western series. ² , ⁴ , ⁶ In contrast, a large multicenter Chinese cohort reported a younger age at onset and a higher proportion of female patients with SUNA. ⁸ These findings indicate that the demographic characteristics of Japanese patients with SUNHA more closely align with Western populations. The key similarities and differences between the present cohort and major international series are summarized in Table S2, which provides a detailed comparison with previously reported cohorts.

The core clinical phenotype observed in our patients—including right‐sided predominance, frequent V1 involvement, very short attack duration, high daily attack frequency, and prominent cranial autonomic symptoms—was highly consistent with reports from Europe, North America, and China. ² , ⁴ , ⁶ , ⁸ One potentially distinctive feature of the Japanese cohort was the relatively high prevalence of eating‐induced attacks, which appeared more frequently than in previously reported cohorts. Facial hygiene–related triggers, such as washing or tooth brushing, were also common, consistent with prior studies. Eating and other innocuous facial stimuli have been reported as common triggers in previous SUNHA cohorts. ² , ⁴ , ⁶ These findings suggest that, while the overall phenotype of SUNHA is remarkably conserved across regions, subtle regional variations in trigger patterns and demographics may exist.

Although the general clinical features were comparable with previously reported SUNHA cohorts, our study demonstrated several differences, including a relatively older age at onset, while the prevalence of NVC was broadly comparable to that reported in previous international cohorts. These findings may reflect differences in referral patterns, healthcare systems, or population characteristics between Japanese and Western cohorts.

At the same time, given the possibility of incidental vascular contact, this association should be interpreted cautiously. Rather than fundamentally challenging prior studies, our data reinforce and contextualize existing evidence within a different healthcare system and ethnic background. Further multicenter studies incorporating standardized imaging protocols and longitudinal follow‐ups are important to clarify the pathophysiological role of neurovascular compression and refine treatment strategies for SUNHA.

Importantly, no statistically significant differences were detected between SUNCT and SUNA across major clinical domains not embedded within the ICHD‐3 diagnostic criteria, including attack duration, frequency, triggers, interictal pain, and treatment responsiveness. Given the limited sample size and statistical power of this study, these findings should not be interpreted as definitive evidence of equivalence. The apparent similarity across these non‐defining clinical features is consistent with prior reports suggesting that SUNCT and SUNA may represent closely related phenotypic variants; however, larger studies are required to confirm this interpretation. ² , ⁸ , ¹⁰

High‐resolution CISS MRI demonstrated NVC in 63% of evaluable patients (17/27; 95% CI: 44.3%–78.5%), a prevalence comparable to that reported in Western and Chinese cohorts. ² , ⁸ , ¹³ NVC was more frequently observed in patients with a history of TN‐like pain. Some patients exhibited TN‐like pain characteristics; however, all patients fulfilled the ICHD‐3 diagnostic criteria for SUNCT or SUNA at the time of evaluation, including the presence of CASs. These findings support the concept of clinical and pathophysiological overlap between SUNHA and trigeminal neuralgia, consistent with previous reports. ² , ¹³

These findings suggest a possible association between NVC and treatment resistance. Nevertheless, because vascular contact can also be observed in asymptomatic individuals, imaging findings should be interpreted cautiously and in conjunction with detailed clinical phenotyping. ¹³ The favorable surgical outcomes observed in our cohort are consistent with prior studies reporting the benefits of microvascular decompression in selected patients. ² , ¹³

Our findings regarding medical management closely align with international evidence. Lamotrigine showed the highest response rate, followed by gabapentin, whereas carbamazepine was less consistently effective, highlighting the differences between SUNHA and classical TN. ⁴ , ¹² , ¹⁴ Intravenous lidocaine provided substantial short‐term benefit as a transitional therapy, whereas indomethacin was ineffective except in a patient with comorbid paroxysmal hemicrania, supporting its primary role as a diagnostic probe rather than a treatment for SUNHA. ⁴ , ¹¹ Consistent with prior reports, lamotrigine, gabapentinoids, and topiramate emerged as the principal preventive drug classes and formed a practical basis for defining refractoriness to medical treatment.

Using the operational definition of refractory SUNHA, seven of the 28 patients (25%) in our cohort were classified as having medically refractory SUNHA. Although the number of refractory cases was limited, NVC was observed in all refractory patients (7/7, 100%) compared with 10 of 20 non‐refractory patients (10/20, 50%), suggesting a possible association between NVC and treatment refractoriness.

In our series, MVD resulted in complete and sustained remission in all patients treated at our institution, with postoperative follow‐up ranging from 18 to 135 months. These outcomes are consistent with those reported in Western surgical studies. ² , ¹³

This study has several limitations. First, this was a single‐center retrospective study conducted at a tertiary referral headache center. Consequently, the cohort may preferentially include patients with more severe, atypical, or treatment‐resistant cases. This referral bias may have increased the observed prevalence of neurovascular compression and medical refractoriness compared with the broader SUNHA population.

Second, the study did not include a control group, such as healthy individuals or patients with other headache disorders, examined using the same high‐resolution MRI protocol. Therefore, the prevalence of neurovascular compression observed in this cohort cannot be directly compared with the background rate of incidental vascular contact in the general population.

Third, treatment response and attack burden were assessed retrospectively and relied largely on patient‐reported outcomes rather than prospective headache diaries. This approach may introduce recall bias, particularly given the fluctuating course of SUNHA over long disease durations. In addition, follow‐up duration and treatment adjustments were not standardized and were based on routine clinical practice, which may have introduced further variability in the assessment of treatment response.

Finally, although relatively large for a rare disorder, the overall sample size remained modest and the study was not prospectively powered to detect small differences between SUNCT and SUNA. Therefore, non‐significant findings should be interpreted cautiously.

Prospective multicenter studies including both tertiary and non‐tertiary settings, with standardized imaging protocols and validated outcome measures, will be important to confirm and extend these findings.

CONCLUSION

In this contemporary Japanese cohort, no statistically significant differences were detected between SUNCT and SUNA across major clinical domains. NVC was frequently observed and appeared to be associated with treatment resistance, suggesting a potential role of vascular–trigeminal conflict in a subset of SUNHA. In carefully selected patients with refractory SUNHA and convincing NVC, MVD may provide durable clinical benefit.

AUTHOR CONTRIBUTIONS

Shoji Kikui: Writing – review and editing; writing – original draft; conceptualization; methodology; software; data curation; formal analysis; validation; investigation; visualization. Daisuke Danno: Writing – review and editing; project administration. Masahiro Nawata: Methodology; validation. Yoshiyasu Iwai: Methodology; validation. Takao Takeshima: Project administration; supervision; writing – review and editing; resources.

FUNDING INFORMATION

No funding was received for this research.

CONFLICT OF INTEREST STATEMENT

Shoji Kikui: Speaking fees, Amgen, Daiichi Sankyo, Eli Lilly, and Otsuka Pharmaceutical. Daisuke Danno: Speaking fees; Amgen, Daiichi Sankyo, Eli Lilly, Otsuka Pharmaceutical. Takao Takeshima: Speaking fees, Amgen, Daiichi Sankyo, Eisai, Eli Lilly, and Otsuka Pharmaceutical, Editorial boards, the Japanese Headache Society and the Japanese Society for Neurology. Masahiro Nawata and Yoshiyasu Iwai have no conflicts to declare.

Supporting information

Supplementary Table 1. Summary of refractory factors and surgical outcomes in patients with medically refractory SUNHA.

Supplementary Table 2. Comparison of clinical characteristics between the present Japanese SUNHA cohort and previously reported cohorts.