Clinical characteristics of pure menstrual migraine and menstrually-related migraine: a cross-sectional online questionnaire study Authors: Yanyan Zhou, Quanquan Zhang, Tian Xia, Mengyi Guo, Hongru Zhao Published: 2026 ABSTRACT AbstractBackgroundMigraine is a highly prevalent neurological disorder which is more common in women. Clinically, there are two subtypes of menstrual migraine, including pure menstrual migraine (PMM) and menstrually-related migraine (MRM). The former experiences migraine attacks only during the perimenstrual period, while the latter experiences migraine attacks during the perimenstrual period and at other times of the menstrual cycle. MRM accounts for 20–25% of female cases and is associated with greater severity and disability. Although hormonal fluctuations are implicated, the pathophysiology of these subtypes remains poorly understood, and treatment options are limited. Whether PMM and MRM differ clinically or in treatment response has not been systematically compared. This study compared their clinical characteristics and acute treatment responses, which may offer insights into menstrual migraine pathophysiology and help with precise treatment.MethodsA cross-sectional study was conducted among 430 female migraine patients recruited from the Headache Specialty Clinic of the First Affiliated Hospital of Soochow University using electronic questionnaires. According to the International Classification of Headache Disorders, 3rd edition (ICHD-3) criteria, participants were divided into PMM (n = 43) and MRM (n = 322). Concurrently, individuals whose migraine attacks were unrelated to menstruation were classified as NMM (n = 65). This analysis focused on the clinical characteristics and acute medication treatment responses between the PMM and MRM groups.ResultsCompared to MRM cases, PMM patients had shorter disease duration (6.8 ± 5.7 vs. 11.1 ± 8.6 years, P < 0.001), fewer prodromal symptoms (58.1% vs. 75.5%, P = 0.016) and accompanying symptoms, including nausea, ocular pain, and osmophobia (all P < 0.05). The efficacy of nonsteroidal anti-inflammatory drugs (NSAIDs) for acute attacks in the PMM group was superior to that in the MRM group, whether perimenstrual period (96.4% vs. 72.8%, P = 0.010) or non-menstrual period (96.4% vs. 78.3%, P = 0.031). Subgroup analysis of patients with low attack frequency (< 4 per month) confirmed that differences in prodromal symptoms and accompanying symptoms remained between PMM and MRM patients, suggesting that attack frequency alone does not explain these differences.ConclusionsPMM was associated with fewer prodromal and accompanying symptoms, while MRM presented with a broader range of symptoms. In the overall analysis, PMM showed a more favorable response to NSAIDs than MRM; however, this difference did not persist after controlling for attack frequency, suggesting it may be partly attributable to differences in migraine burden. By contrast, differences in prodromal and accompanying symptoms remained significant after adjustment, supporting intrinsic phenotypic distinctions between the two subtypes.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12883-026-05025-x.Keywords: Pure menstrual migraine, Menstrually-related migraine, Cross-sectional study, Prodromal symptom, Accompanying symptom Introduction Migraine is a common primary headache disorder with a global average prevalence of approximately 14–15% [1]. According to the latest research based on the 2021 global burden of disease data, migraine is the leading cause of years lived with disability among women of childbearing age worldwide, highlighting its significant disease burden in specific populations [2]. According to ICHD-3, migraine is defined as a primary disorder characterized by episodes lasting 4–72 h, typically presenting as unilateral, moderate to severe pulsating pain that is exacerbated by routine physical activity, accompanied by symptoms such as nausea and/or vomiting, photophobia, and phonophobia [3]. The impact of migraine on daily life extends beyond the headache phase itself, as associated symptoms also contribute significantly to functional impairment. The prevalence of migraine varies significantly across different age groups and genders, with the highest prevalence observed between 25 and 55 years of age. Among adults, the prevalence in females is 3–4 times higher than that in males [2]. Menstrual migraines include pure menstrual migraine (PMM) and menstrually-related migraine (MRM). The former refers to migraines occurring exclusively in association with menstruation, beginning on days − 2 to + 3 of menstruation and at no other times of the cycle, while the latter involves migraines occurring on days − 2 to + 3 of menstruation in at least two out of three menstrual cycles, and additionally at other times of the cycle [3]. Abnormal activity of the central nervous system and activation of the peripheral trigeminovascular system [4, 5], fluctuations in estrogen levels [6], prostaglandin release [7], and genetic mechanisms [8] play significant roles in the pathogenesis of menstrual migraine [9]. Previous studies from non-Asian populations have shown that compared to attacks at other times of the menstrual cycle, perimenstrual migraine attacks are more severe, more disabling, and respond less well to acute treatments [10], further increasing the disease burden of migraine. First proposed in the 1970s, the estrogen withdrawal hypothesis has been the primary theory explaining menstrual migraine [11]. Although the association between estrogen fluctuations and the pathophysiology of migraine seems evident, its exact mechanism remains controversial. This study focused on Chinese female patients with migraine and compared the differences in clinical characteristics and responses to acute treatment between PMM and MRM. It aims to provide evidence for in-depth exploration of the pathophysiological mechanisms and precise treatment strategies of menstrual migraine in the future. Methods Study design and population This study continuously enrolled female migraine patients who visited the Headache Specialty Clinic of the First Affiliated Hospital of Soochow University from May 2023 to May 2025. A total of 468 electronic questionnaires were collected. According to the inclusion and exclusion criteria, 430 patients were finally included for analysis. Inclusion criteria were: (1) Meeting the diagnostic criteria for migraine in the third edition of the ICHD; (2) Female patients with migraine; (3) Voluntarily participate and complete the questionnaire survey. Exclusion criteria included: (1) Pregnant or lactating women; (2) Combined with other severe primary or secondary headaches; (3) Severe mental illness or cognitive dysfunction; (4) The questionnaire is incomplete or has logical contradictions. Based on the temporal relationship between migraine attacks and the menstrual cycle, the enrolled patients were further classified into two groups: (1) PMM group: migraines occurring exclusively in association with menstruation, beginning on days − 2 to + 3 of menstruation and at no other times of the cycle; (2) MRM group: migraines occurring on days − 2 to + 3 of menstruation in at least two out of three menstrual cycles, and additionally at other times of the cycle. The grouping criteria were based on the diagnostic recommendations of the International Headache Society regarding menstrual migraine. Patients whose headache attacks were not related to the menstrual cycle were classified as the non-menstrual migraine (NMM) group (n = 65), which was used for exploratory comparisons of acute treatment response. See Fig. 1 for the specific procedure. Patient disposition and group creation. ICHD-3, International Classification of Headache Disorders, 3rd edition; MM, menstrual migraine; PMM, pure menstrual migraine; MRM, menstrually-related migraine; NMM, non-menstrual migraine Study outcomes and assessments In this study, all patients completed the following contents through electronic questionnaires: (1) Demographic and clinical characteristics: general characteristics recorded included age, disease duration, family history, and presence of dysmenorrhea. Anxiety and depression were assessed using the Generalized anxiety disorder scale (GAD-7) [12] and the patient health questionnaire (PHQ-9) [13], respectively; (2) Clinical features of migraine: patients reported the monthly frequency of migraine attacks; triggers in MRM, including environmental factors, fatigue, sleep deprivation and emotional excitement. Prodromal symptoms such as stiff neck, dizziness, and photophobia were also recorded. Auras were classified as visual, sensory, or verbal. Additionally, we evaluated accompanying symptoms, which included gastrointestinal discomfort (nausea, vomiting), sensory sensitivities (photophobia, phonophobia, osmophobia), and autonomic symptoms like tearing and nasal congestion. (3) Acute phase treatment effect: acute treatment responses were evaluated based on the use of NSAIDs and triptans. The NSAIDs included ibuprofen, diclofenac sodium, and celecoxib. The triptans included zolmitriptan and rizatriptan. Treatment efficacy was categorized as good response (effective) or insufficient response (ineffective), defined according to criteria from previous studies. Effectiveness was defined as achieving a pain-free state within 2 h after acute medication use in at least 4 out of 5 attacks, while ineffective was defined as achieving pain-free status within 2 h in no more than 3 out of 5 attacks [14, 15]. For the MRM group, treatment responses were further analyzed separately for attacks occurring during menstruation and those occurring outside of menstruation. Statistical analysis Statistical analyses were performed using SPSS 26.0 (IBM Corp., Armonk, NY, USA). Categorical variables were presented as counts and percentages, and comparisons between groups were performed using the chi-square test or Fisher’s exact test (when the expected frequency was less than 5). Continuous variables were expressed as mean±standard deviation. For comparisons between two groups, the independent samples t-test or Mann–Whitney U test was applied, as appropriate. For comparisons involving multiple groups, one-way analysis of variance or the Kruskal–Wallis test was used. All statistical tests were two-sided, and a P value < 0.05 was considered statistically significant. Results From May 2023 to May 2025, a total of 468 electronic questionnaires from female migraine patients were collected. Based on inclusion and exclusion criteria, questionnaires from 430 patients were included in the study. Based on questionnaire data, patients were classified into a PMM group (n = 43), an MRM group (n = 322), and an NMM group (n = 65) (Fig. 1). Differences were found among these groups in prodromal symptoms, accompanying symptoms, and acute treatment efficacy. Demographic and clinical characteristics Regarding demographic and clinical characteristics, the disease duration, anxiety (GAD-7) and depression (PHQ-9) scores of the PMM and MRM groups were shown in Table 1. The duration of disease in the PMM group was significantly shorter than that in the MRM group (6.8 ± 5.7 years vs. 11.1 ± 8.6 years, P < 0.001), and the incidence of dysmenorrhea was significantly lower than that in the MRM group (32.6% vs. 50.0%, P = 0.032). There were no significant differences in genetic background, GAD-7 score and PHQ-9 score between the two groups (P > 0.05). Table 1Demographic and clinical characteristics of PMM and MRM groupsCharacteristicPMM (n = 43)MRM (n = 322)P valueDysmenorrhea, n (%)0.032*Yes14 (32.6)161 (50.0)No29 (67.4)161 (50.0)Heredity, n (%)0.53Paternal5 (11.6)32 (9.9)Maternal9 (20.9)89 (27.6)None29 (67.4)201 (62.4)Disease duration, years6.8 ± 5.711.1 ± 8.6< 0.001*GAD-7 score5.0 ± 5.34.9 ± 4.20.97PHQ-9 score4.7 ± 4.96.0 ± 5.00.12Data were presented as n (%) or mean ± SD. P values were calculated using Chi-square test, Fisher’s exact test, or independent t-test as appropriatePMM pure menstrual migraine, MRM menstrually-related migraine, GAD-7 Generalized Anxiety Disorder-7, PHQ-9 Patient Health Questionnaire-9*p ≤ 0.05 Demographic and clinical characteristics of PMM and MRM groups Data were presented as n (%) or mean ± SD. P values were calculated using Chi-square test, Fisher’s exact test, or independent t-test as appropriate PMM pure menstrual migraine, MRM menstrually-related migraine, GAD-7 Generalized Anxiety Disorder-7, PHQ-9 Patient Health Questionnaire-9 *p ≤ 0.05 In contrast to PMM patients, whose attacks were triggered by menstruation alone, MRM patients could be provoked by multiple common factors, such as cold wind (54.7%), hot environment (46.0%), fatigue (69.3%), sleep deprivation (75.5%), emotional stress (53.7%) and alcohol (12.7%). There was no significant difference in the overall incidence of aura between the two groups (P = 0.059). The incidence of sensory aura in the PMM group was significantly lower than that in the MRM group (0% vs. 10.2%, P = 0.031). There was no significant difference between the two groups in the incidence of visual aura and language aura (P > 0.05). Prodromal and accompanying symptoms Compared with the MRM group, the PMM group had a lower overall incidence of prodromal symptoms (58.1% vs. 75.5%, P = 0.016) in Table 2. The PMM group also showed lower incidence of neck stiffness (20.9% vs. 41.3%, P = 0.012), photophobia (14.0% vs. 33.2%, P = 0.013), fatigue (32.6% vs. 50.3%, P = 0.029), difficulty concentrating (14.0% vs. 30.7%, P = 0.022), and mood changes (16.3% vs. 38.5%, P = 0.004). The incidence of prodromal dizziness was similar between the two groups (30.2% vs. 46.0%, P = 0.054). The remaining prodromal symptoms (yawning, drowsiness) did not differ significantly between the two groups. Compared with the MRM group, the PMM group had lower incidences of ocular pain (27.9% vs. 47.5%, P = 0.015), nausea (53.5% vs. 73.9%, P = 0.005), osmophobia (9.3% vs. 27.0%, P = 0.012) and nasal congestion (0% vs. 10.2%, P = 0.028). The incidence of vomiting was similar between the two groups (41.9% vs. 55.9%, P = 0.086). The remaining accompanying symptoms (fear of light, fear of noise, tears, and runny nose) were not significantly different between the two groups. Table 2Headache characteristics and associated symptomsPMM (n = 43)MRM (n = 322)P valueHeadache location, n (%)Unilateral30 (69.8)192 (59.6%)0.20Bilateral13 (30.2)130 (40.4%)Headache quality, n (%)Pressing31 (72.1)223 (69.3)0.71Pulsating24 (55.8)188 (58.4)0.75Explosive17 (39.5)126 (39.1)0.96Prodromal symptoms, n (%)Any prodromal symptoms25 (58.1)243 (75.5)0.016*Neck stiffness9 (20.9)133 (41.3)0.012*Dizziness13 (30.2)148 (46.0)0.054Photophobia6 (14.0)107 (33.2)0.013*Fatigue14 (32.6)162 (50.3)0.029*Mood changes7 (16.3)124 (38.5)0.004*Poor concentration6 (14.0)99 (30.7)0.025*Yawning8 (18.6)101 (31.4)0.091Drowsiness10 (23.3)116 (36.0)0.109Aura, n (%)Any aura7 (16.3)97(30.1)0.059Visual7 (16.3)60 (18.6)0.71Sensory0 (0.0)33 (10.2)0.031*Language0 (0.0)4 (1.2)1.00Accompanying symptoms, n (%)Ocular pain12 (27.9)153 (47.5)0.015*Nausea23 (53.5)238 (73.9)0.005*Vomiting18 (41.9)180 (55.9)0.086Osmophobia4 (9.3)87 (27.0)0.012*Nasal congestion0 (0.0)33 (10.2)0.028*Photophobia16 (37.2)140 (43.5)0.44Phonophobia23 (53.5)180 (55.9)0.77Tearing5 (11.6)57 (17.7)0.33Rhinorrhea1 (2.3)33 (10.2)0.10Data are presented as n (%). P values were calculated using Chi-square test or Fisher’s exact test as appropriate*p ≤ 0.05 Headache characteristics and associated symptoms Data are presented as n (%). P values were calculated using Chi-square test or Fisher’s exact test as appropriate *p ≤ 0.05 Acute treatment efficacy of NSAIDs In Table 3, analysis of response to NSAIDs showed that the efficacy rate in the PMM group was higher than that in the MRM group whether for perimenstrual attacks (96.4% vs. 72.8%, P = 0.010) or for non-menstrual attacks (96.4% vs. 78.3%, P = 0.031). For triptans, the efficacy rate in the PMM group was comparable to that in the MRM group, whether for perimenstrual attacks (100% [4/4] vs. 76.7% [33/43], P = 0.564) or for non-menstrual attacks (100% [4/4] vs. 80.5% [33/41], P = 0.583) ( Table 3). However, the sample size for triptan use in the PMM group was very small (n = 4), therefore, the statistical comparison between the two groups should be interpreted with caution. This finding may represent a potential clinical trend that warrants further investigation in large sample study. Within the MRM group, there was no significant difference in the acute treatment efficacy rates of NSAIDs or triptans between perimenstrual and non-menstrual attacks (P > 0.05). There were also no significant differences in the efficacy rates of these two drug classes between non-menstrual attacks in the MRM group and the NMM group (P > 0.05) (not shown in the tables). Table 3Acute treatment responsePMM groupMRM group(menstrual)P valueMRM group(non-menstrual)P valueNSAIDs efficacy, n/N (%)27/28 (96.4)107/147 (72.8)0.010*116/148 (78.3)0.031*Triptans efficacy, n/N (%)4/4 (100)33/43 (76.7)0.56433/41 (80.5)0.583Data are presented as n/N (%). P values for each comparison (PMM vs. menstrual MRM; PMM vs. non-menstrual MRM) were calculated using Chi-square test for 2 × 2 contingency tablesOnly patients with evaluable efficacy data were included in the analysis for each medication. Sample sizes reflect medication users with valid data, not the total group populationPMM pure menstrual migraine, MRM menstrually-related migraine, NSAIDs non-steroidal anti-inflammatory drugs*p ≤ 0.05 Acute treatment response Data are presented as n/N (%). P values for each comparison (PMM vs. menstrual MRM; PMM vs. non-menstrual MRM) were calculated using Chi-square test for 2 × 2 contingency tables Only patients with evaluable efficacy data were included in the analysis for each medication. Sample sizes reflect medication users with valid data, not the total group population PMM pure menstrual migraine, MRM menstrually-related migraine, NSAIDs non-steroidal anti-inflammatory drugs *p ≤ 0.05 Subgroup analysis controlling for migraine frequency In the overall analysis, MRM patients exhibited a significantly higher migraine burden than PMM patients, with a greater proportion experiencing ≥ 4 attacks per month and a longer disease duration (11.1 ± 8.6 vs. 6.8 ± 5.7 years, P < 0.001). To reduce the potential confounding effect of migraine frequency, a subgroup analysis was performed in patients with < 4 migraine attacks per month. In this restricted cohort, 197 MRM patients were included for comparison. This subgroup analysis aimed to reassess the previously observed clinical differences between pure menstrual migraine (PMM) and menstrually-related migraine (MRM) under more balanced levels of migraine burden. The analysis helps to determine whether the reported differences reflect intrinsic phenotypic distinctions between PMM and MRM or are mainly attributed to imbalances in attack frequency. After restricting the analysis to patients with < 4 migraine attacks per month of MRM, most baseline characteristics were comparable between the two groups, although disease duration remained longer in MRM patients (P = 0.005), as shown in Supplementary Table S1. In terms of headache characteristics, MRM patients showed susceptibility to several common triggers, including cold wind, hot environment, fatigue, sleep deprivation, and emotional stress. The overall prevalence of prodromal symptoms was also higher in the MRM group (P = 0.030), particularly for neck stiffness, photophobia, fatigue, mood changes, and poor concentration (all P < 0.05). Regarding aura, the overall prevalence did not differ significantly between the two groups; however, sensory aura was more frequently reported in MRM patients (P = 0.016). In addition, certain accompanying symptoms, including ocular pain, nausea, and nasal congestion, remained more common in MRM (all P < 0.05), whereas most other clinical features were comparable between the two groups. Details are presented in Supplementary Table S2. No statistically significant differences were observed in acute treatment response between the two groups. However, the comparison of triptan efficacy should be interpreted with caution due to the very small sample size in the PMM group (n = 4). See Supplementary Table S3 for details. Discussion Through the analysis of clinical data from 430 Chinese female migraine patients, this study provides evidence suggesting differences between PMM and MRM in terms of clinical symptom profiles and acute treatment responses. These findings offer clinical insights to further understand the heterogeneity of menstrual migraine and its underlying pathophysiological mechanisms. Previous researches on menstrual migraine often combined PMM and MRM into a homogeneous group, focusing on their differences from non-menstrual migraine, but lacking direct comparisons of clinical features and treatment responses between the two subtypes. Reviews and meta-analyses by Głogowska et al. [16] and Khoo et al. [15] did not systematically differentiate between them; even studies like Chalmer et al. [17], while distinguishing subtypes, still focused on horizontal comparisons with non-menstrual migraine. This study has systematically compared PMM and MRM in a Chinese female population, and found significant differences in prodromal symptoms (photophobia, mood changes, fatigue, neck stiffness) and accompanying symptoms (nausea, ocular pain, osmophobia, nasal congestion), as well as divergent treatment responses. This provides clinicians with actionable reference indicators for subtype differentiation based on symptom characteristics. The dual differences in clinical symptoms and treatment responses suggest that the two subtypes may have different pathophysiological mechanisms, not only extending beyond the classic “estrogen withdrawal” hypothesis but also incorporating central sensitization and peripheral inflammation into the explanatory framework. This provides an actionable basis and a more complete theoretical explanation for clinical subtype differentiation and personalized treatment. In this study, the disease course in the PMM group was significantly shorter than that in the MRM group (6.8 ± 5.7 years vs. 11.1 ± 8.6 years, P < 0.001), which may reflect differences in the clinical presentation of the two subtypes [3, 10]. The migraine attacks of PMM are strictly confined to the menstrual period, and the clinical manifestations are more regular, making it easier to identify at an early stage. In addition to menstrual episodes, MRM patients also have non-menstrual episodes, and the disease manifestations are more complex, which may lead to delayed diagnosis or longer accumulation of the disease course. Prodromal symptoms represent reversible manifestations of central nervous system dysfunction occurring hours to days before a migraine attack, reflecting the functional state of higher-order centers like the cortex and thalamus [18]. We found significantly higher incidences of several prodromal symptoms in the MRM group compared to the PMM group: photophobia (33.2% vs. 14.0%, P = 0.013), mood changes (38.5% vs. 16.3%, P = 0.004), fatigue (50.3% vs. 32.6%, P = 0.029), and neck stiffness (41.3% vs. 20.9%, P = 0.012). These differences suggest a more pronounced degree of central sensitization in MRM patients. Neuroimaging studies provide mechanistic evidence: photophobia is a classic manifestation of hyperexcitability in visual pathways, closely related to abnormal activation of thalamocortical circuits [19]. A 2025 study by May’s team found that migraine patients exhibited significantly decreased pain thresholds for electrical, heat, and cold stimuli as early as 48 h before an attack, while pain thresholds in healthy controls remained stable throughout the observation period [20]. A 2025 study by Tang et al. [21] systematically compared, for the first time, differences in large-scale brain network functional connectivity between MRM and non-menstrual migraine patients, finding abnormal connectivity among the default mode network, salience network, and dorsal attention network in MRM patients, and a negative correlation between the functional connectivity strength of the dorsal attention network and the auditory network with disease duration. This finding provides a neuroimaging basis for the prominence of multisensory symptoms (including photophobia) in MRM patients. Mood changes are closely related to the functional state of the limbic system, particularly the amygdala and anterior cingulate cortex [22]. Huang et al. [23] confirmed that MRM patients exhibit functional network abnormalities centered on the right insula, with the degree of abnormality correlating with disease duration and anxiety scores, indicating greater involvement of the limbic system. The differences in fatigue and neck stiffness suggest broader and more sustained nociceptive input and central modulation processes in MRM patients [24]. A 2025 longitudinal fMRI study by Li et al. [25] dynamically monitored MRM patients, finding that sharp declines in hormone levels can modulate the height of the hemodynamic response function, and MRM patients exhibited a pattern of hemodynamic response width changes opposite to that of healthy controls, with significantly reduced perimenstrual insular phase synchrony. In summary, the predominance of prodromal symptoms in MRM patients reflects a more significant gain in the central nervous system’s processing of multisensory stimuli, along with broader and more sustained nociceptive input and central modulation, suggesting that its pathophysiology involves more extensive central sensitization and engagement of emotional regulatory circuits. Accompanying symptoms reflect direct activation of the trigeminovascular system and central sensitization during the migraine attack. This study reveals that the MRM group has significantly higher incidences of several accompanying symptoms compared to the PMM group, directly pointing to different emphases in the pathophysiological mechanisms of the two subtypes. Nausea showed the most significant difference between the groups (MRM 73.9% vs. PMM 53.5%, P = 0.005) and is closely related to activation of the nucleus tractus solitarius in the medulla oblongata [22]. Kuruvilla DE et al. [26] pointed out that estrogen withdrawal can relieve inhibition of the CGRP signaling pathway, leading to increased Calcitonin Gene-Related Peptide (CGRP) release, and CGRP has potent effects on brainstem regions, suggesting significant brainstem activation during attacks in MRM patients. Ocular pain (MRM 47.5% vs. PMM 27.9%, P = 0.015) is associated with activation of the ophthalmic division of the trigeminal nerve and periorbital vasodilation. Kuruvilla DE et al. [26] confirmed that estrogen fluctuations can affect neurovascular coupling, suggesting a broader or more intense activation of the trigeminovascular system in MRM patients. Osmophobia (MRM 27.0% vs. PMM 9.3%, P = 0.012) is a specific manifestation of central sensitization in olfactory pathways. The abnormal multisensory network integration found by Tang et al. [21] in MRM provides indirect evidence, indicating that central sensitization extends beyond the pain modality to involve chemosensory pathways. Particularly specific was nasal congestion, observed only in the MRM group (10.2% vs. 0%, P = 0.028). As a typical manifestation of autonomic dysfunction, this suggests a more pronounced trigeminal-autonomic reflex hyperactivity in MRM patients, offering a new entry point for future mechanistic exploration. In summary, the richer array of accompanying symptoms in MRM reflects more extensive activation of the trigeminovascular system, multisensory central sensitization, and significant autonomic dysfunction, contrasting sharply with the relatively limited, peripherally-driven pattern hypothesized for PMM. With respect to acute treatment response, this study observed that the efficacy rate of NSAIDs was higher in the PMM group than in the MRM group in the unadjusted analysis. NSAIDs exert anti-inflammatory and analgesic effects by inhibiting cyclooxygenase and reducing prostaglandin synthesis. Prostaglandin release, originating from endometrial shedding entering the systemic circulation during menstruation, has been considered an important mechanism in menstrual migraine pathophysiology [27]. The numerically higher response to NSAIDs observed in PMM may be compatible with relatively greater involvement of peripheral prostaglandin-mediated mechanisms involving estrogen withdrawal, prostaglandin release, and trigeminovascular activation [4–7]. In contrast, the numerically lower response observed in MRM may reflect greater clinical heterogeneity and the possible involvement of additional mechanisms beyond prostaglandin-related pathways, including central sensitization and neuropeptide-mediated mechanisms such as CGRP signaling. However, after adjustment for attack frequency, the differences in NSAID response between the two groups were no longer statistically significant and should therefore be interpreted cautiously. Regarding triptan response, no statistical difference was found between the two groups, consistent with the conclusion from the meta-analysis by Khoo et al. [15] that triptans were generally effective for menstrual migraine. This suggests that the vasoconstrictive and neurotransmitter inhibition mechanisms mediated by 5-hydroxytryptamine 1B/1D (5-HT1B/1D) receptors are operative in both subtypes, with the differences primarily originate from upstream triggering mechanisms and the degree of central sensitization, rather than fundamental differences in downstream effector pathways [28]. These findings may provide preliminary observations relevant to individualized treatment considerations. PMM patients showed a numerically higher response to NSAIDs, which may warrant further investigation in future treatment studies. Compared with PMM patients, MRM patients, who exhibited broader and more severe clinical manifestations, including symptoms associated with central sensitization, such as photophobia, osmophobia, and mood changes, may show a less favorable response to purely peripheral anti-inflammatory drugs. They may require migraine-specific treatments acting on the calcitonin gene-related peptide‌ (CGRP) pathway or 5-HT1B/1D receptors, or combination therapy with NSAIDs. Future prospective studies are needed to further clarify whether treatment responses differ meaningfully between these subtypes and whether distinct therapeutic strategies may be beneficial. A key methodological concern is whether the observed clinical differences between PMM and MRM reflect differences in migraine burden rather than intrinsic phenotypic distinctions. In the overall analysis, MRM patients had higher attack frequency and longer disease duration. To address this potential confounding, we performed a subgroup analysis that included MRM patients with fewer than 4 migraine attacks per month and PMM patients, to reduce the imbalance in migraine frequency between the two groups. Notably, even after this restriction, MRM patients continued to show a higher prevalence of several prodromal symptoms (e.g., photophobia, mood changes, fatigue, neck stiffness, and poor concentration), accompanying symptoms (e.g., nausea, ocular pain, and nasal congestion), and sensitivity to multiple non-hormonal triggers (e.g., cold wind, hot environment, fatigue, sleep deprivation, and emotional stress) compared with PMM patients (Supplementary Table S2). In contrast, differences in acute treatment response were no longer statistically significant after this adjustment, which may reflect reduced statistical power due to smaller sample sizes or, alternatively, suggest that the previously observed differences in NSAID response may be partly influenced by differences in overall migraine burden. These findings suggest that the observed differences in symptom burden may not be fully explained by differences in migraine frequency alone, but may also be associated with differences in disease characteristics. One possible interpretation is that MRM may be associated with a broader pattern consistent with central sensitization, whereas PMM may be more closely linked to hormonally driven and peripheral inflammatory mechanisms. However, these interpretations remain speculative and require further validation. Based on the observed clinical patterns, it may be hypothesized that PMM could be characterized by a predominantly hormonal and peripheral inflammatory profile, in which estrogen withdrawal may trigger prostaglandin release and activate the peripheral trigeminovascular system, with a relatively limited degree of central sensitization. Clinically, this may manifest as relatively fewer associated symptoms and a more favorable response to NSAIDs. In contrast, MRM may involve a more complex profile incorporating both hormonal influences and central sensitization, potentially reflecting a heightened state of central sensory gain and a broader range of associated symptoms, such as photophobia, osmophobia, and mood changes. From a therapeutic perspective, these patterns may suggest that patients with MRM could potentially benefit from treatment strategies beyond peripheral anti-inflammatory approaches, including migraine-specific therapies targeting the CGRP pathway or 5-HT1B/1D receptors, or combination approaches. However, these considerations should be interpreted with caution and should not be taken as direct evidence to guide treatment decisions without further prospective validation. Overall, these findings may contribute to a more nuanced understanding of menstrual migraine heterogeneity and provide a preliminary framework for future research into subtype-specific mechanisms and individualized management strategies. However, it is important to note that multiple statistical comparisons were performed without formal adjustment (e.g., Bonferroni correction). The risk of type I error cannot therefore be excluded, and the findings should be regarded as exploratory and hypothesis-generating, requiring confirmation in independent cohorts. Conclusions This study suggests that PMM and MRM may exhibit distinct clinical characteristics and acute treatment responses, indicating that the previously held notion of uniformly poor treatment response in menstrual migraine may be an oversimplification. These findings advance our understanding from a generalized perspective toward more nuanced subtype differentiation. Specifically, PMM patients tend to have fewer prodromal and accompanying symptoms and respond well to NSAIDs, consistent with a pathophysiological model involving hormonal fluctuations and peripheral inflammation. In contrast, MRM patients exhibit more extensive prodromal and accompanying symptoms, with relatively lower NSAID efficacy, which may reflect a complicated pathophysiological mechanism involving hormonal influence and central sensitization. Notably, even after controlling for attack frequency, MRM patients still showed a higher prevalence of prodromal symptoms (e.g., photophobia, mood changes) and specific accompanying symptoms (e.g., nausea, nasal congestion) compared with PMM patients (Supplementary Table S2). These differences are therefore unlikely to be driven solely by attack frequency and may instead reflect an inherent central sensitization in MRM. This observation provides preliminary insight into the heterogeneous mechanisms of menstrual migraine and supports the rationale for developing precise, personalized treatment strategies based on subtype. In conclusion, PMM and MRM appear to differ in several clinical features, particularly in symptom profiles, and treatment responses. However, these findings should be interpreted as exploratory and hypothesis-generating, and they require validation in larger, prospective, multicenter studies. Further research is needed to better elucidate the clinical and pathophysiological heterogeneity of perimenstrual migraine and to guide the development of more targeted therapeutic strategies. Supplementary Information Supplementary Material 1. Acknowledgments Consent to participate Participants were informed about the project both verbally and in writing, and enrollment was based on informed consent. Abbreviations Calcitonin gene-related peptide‌ Generalized anxiety disorder scale International Classification of Head Disorders-3rd Edition Menstrually-related migraine Non-menstrual migraine Non-steroidal anti-inflammatory Drugs Patient Health Questionnaire Pure menstrual migraine 5-hydroxytryptamine 1B/1D receptors Authors’ contributions ZHR had the original idea for the project. ZHR and GMY contributed to design of the study. ZHR, GMY, ZYY and ZQQ performed all interviews, made the analyses and drafted the manuscript. ZYY, ZQQ and XT reviewed the first and subsequent drafts. All authors read and approved the final manuscript. Funding This work was supported by the funding from the Natural Science Foundation of Jiangsu Province (BK20240367), the Pilot Project for Basic Research of Suzhou City (Project Number: SSD2025050) and the Horizontal Project of Soochow University (Project Number: H240792). Data availability The dataset used during the current study is available from the corresponding author on reasonable request. Declarations Ethics approval and consent to participate The study was approved by the Ethics Committee of the First Affiliated Hospital of Soochow University (Approval No. 2024454; approved on 21 November 2024), with all relevant guidelines and regulations being observed. Written informed consent was obtained from all participants in accordance with the Declaration of Helsinki. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Yanyan Zhou and Quanquan Zhang contributed equally to this work. Contributor Information Mengyi Guo, Email: guomengyi277@163.com. Hongru Zhao, Email: tiantan11@163.com. References Associated Data Supplementary Materials Supplementary Material 1. Data Availability Statement The dataset used during the current study is available from the corresponding author on reasonable request.