Interventional procedure overview of transcutaneous electrical stimulation of the supraorbital nerve for treating and preventing migraine

Pain reduction

In a double-blind RCT of 106 people with acute migraine attacks with or without aura, eTNS for 1 hour (eTNS, n=52) was compared with sham stimulation (n=54). In the ITT analysis, there was a statistically significantly greater reduction in pain intensity (measured using a 10‑point VAS, with the highest score representing maximum pain) at 1 hour with eTNS than with sham (eTNS 5.92 to 2.46, sham 6.17 to 4.39; mean change: -3.46 compared with -1.78, p<0.0001). Decrease in pain score was also higher with eTNS than with sham at 2 hours (eTNS 5.92 to 3.06, sham 6.17 to 4.31; mean change -2.87 compared with -1.85, p=0.028) and 24‑hour time points (eTNS 5.92 to 2.46, sham 6.17 to 3.79; mean change -3.46 compared with -2.38, p=0.062). In the modified ITT analysis, pain intensity reduced statistically significantly more with eTNS than with sham at 1 hour compared with baseline (65% compared with 32%, p<0.0001). In the subgroup analysis, there was a statistically significant difference in pain reduction at 1 hour between eTNS and sham for people with migraine without aura (mean VAS reduction was 3.3 compared with 1.7, p=0.0006). For people with migraine with aura, pain reduction at 1 hour was greater with eTNS than with sham but did not reach statistical significance (mean VAS reduction was 4.3 compared with 2.6, p=0.060; Chou 2019).

In a case series of 35 people who had acute treatment with eTNS for episodic or chronic migraine with or without aura, there was a statistically significant decrease in pain intensity (measured using a 10‑point VAS, with the highest score representing maximum pain) from a baseline score of 5.63 to 2.42 after 1 hour and to 2.66 after 2 hours (p<0.001; Chou 2017).

Freedom from pain

In an RCT of 538 people (ITT analysis) having 2‑hour eTNS (n= 259) or sham stimulation (n= 279) for acute treatment of migraine attacks at home, the percentage of people with no pain at 2 hours was statistically significantly higher with eTNS (26% [66/259]) than with sham (18% [51/279], p= 0.043). Pain freedom at 24 hours was also statistically significantly higher with eTNS (23% [59/259]) than with sham (16% [44/279], p= 0.039); Kuruvilla 2022).

In the RCT of 106 people (ITT analysis), the proportion of people free from pain at 1 hour was statistically significantly higher with eTNS than with sham (29% [15/52] compared with 6% [3/54], p=0.0016) but not at 2 hours (p=0.15) or at 24‑hour follow up (p=0.056). In the modified ITT analysis, the proportion who were pain free at 24 hours was statistically significantly higher with eTNS than with sham (32% compared with 13%, p=0.032; Chou 2019).

In the case series of 35 people using eTNS, freedom from pain was reported in 20% (6/30) at 1-hour follow up, and 13% (4/30) at 2-hour follow up (Chou 2017).

In a case series of 59 people using eTNS for treating acute migraine at home for 2 hours, freedom from moderate or severe pain (measured on a scale of 0 to 3, with higher scores indicating more severe pain) was reported in 35% (17/48) of people at 2 hours and 25% (12/48) had sustained freedom from pain at 24-hour follow up (Kuruvilla 2019).

Pain or headache relief

In the RCT of 538 people, the percentage of people with pain relief at 2 hours was statistically significantly higher with eTNS (70%) than with sham (55%, p= 0.001). Sustained pain relief at 24 hours was also statistically significantly higher with eTNS (46%) than with sham (35%, p=0.006; Kuruvilla 2022).

In the RCT of 106 people (ITT analysis), the proportion who had more than 30% pain relief was statistically significantly higher with eTNS than with sham at 1 hour (79% [41/52] compared with 39% [21/54], p<0.0001) but not at 2 hours (65% [34/52] compared with 52% [28/54], p=0.17) or at 24-hour follow up (73% [38/52] compared with 61% [33/54], p=0.22). The proportion of people who had more than 50% pain relief was statistically significantly higher with eTNS than with sham at 1 hour (63% [33/52] compared with 31% [17/54], p=0.0017) but not at 2 hours (54% [28/52] compared with 41% [22/54], p=0.24) or at 24-hour follow up (60% [31/52] compared with 48% [26/54], p=0.25). In the modified ITT analysis, the proportion of people who had more than 30% sustained pain relief were statistically significantly higher in the eTNS group compared with the sham stimulation group (43% compared with 21%, p=0.030; Chou 2019).

In the case series of 35 people, more than 50% pain relief was reported by 77% (23/30) at 1 hour and 57% (17/30) at 2 hours, and more than 30% pain relief was reported by 83% (25/30) at 1 hour and 70% (21/30) at 2 hours (Chou 2017).

In the case series of 59 people, pain relief (measured on a scale 0 to 3, with higher scores indicating severe pain) was reported in 71% (34/48) of people at 2 hours (Kuruvilla 2019).

In a retrospective case series of 19 people using eTNS for acute VM attacks for 20 minutes, complete relief of headache was reported by 57% (8/14) of people who had headache at baseline. Seven people had complete resolution of vertigo. Mean headache severity (assessed using a 10-point VAS, with higher scores representing more severe headache or vertigo) improved from 4.8 at baseline to 1.4 at 15 minutes with eTNS. Mean improvement in headache was 77%. Everyone reported improvement in vertigo severity. Mean vertigo severity improved from 6.6 at baseline 6.6 to 2.7 at 15 minutes with eTNS. Mean improvement in vertigo was 61%. Other symptoms such as head and eye pressure, ear and facial pain also improved in some people (Beh 2020).

Rescue antimigraine medication use

In the RCT of 538 people, there was no statistically significant difference between the groups for use of rescue medication between 2 and 24 hours post treatment (32% [82/259] compared with 38% [105/279], p= 0.15; Kuruvilla 2022).

In the RCT of 106 people (ITT analysis), there was no statistically significant difference in rescue medication intake in the eTNS group compared with the sham stimulation group at 2 hours (6% [3/52] compared with 4% (2/54), p=0.66) or at 24-hour follow up (35% [18/52] compared with 39% [21/54], p=NS; Chou 2019).

In the case series of 59 people, 50% (24/48) of people reported that they used medication between 2 and 24 hours (Kuruvilla 2019).

In the case series of 35 people, rescue medications were not used by anyone at 2 hours nor by 65% (17/26) at 24 hours (Chou 2017).

Migraine associated symptoms

In the RCT of 538 people, the percentage of people experiencing absence of most bothersome migraine-associated symptoms at 2 hours after treatment was statistically significantly higher with eTNS (57% [146/259]) than with sham (42% [118/279], p= 0.001). The percentage of people experiencing absence of all migraine-associated symptoms at 2 hours after treatment was statistically significantly higher with eTNS (43% [110/259]) than with sham ( 34% [95/279], p= 0.044) (Kuruvilla 2022).

In the case series of 59 people, freedom from MBS at 2 hours was reported in 60% (29/48) (nausea 46% [5/11], photophobia 63% [17/27] and phonophobia 78% [7/9]; Kuruvilla 2019).

Reduction in monthly migraine days

In a multicentre double-blind RCT of 67 people, tSNS (n=34) was compared with sham treatment (n=33). In the ITT analysis, there was a statistically significant decrease in the mean number of monthly migraine days between baseline and after 3 months with tSNS (6.94 to 4.88, p=0.023), but not with sham (6.54 to 6.22, p=0.608). However, the difference between the 2 groups was not statistically significant (p=0.054; Schoenen 2013, 2016).

In a multicentre RCT of 90 people with at least 2 migraine attacks per month comparing STNS (n=45) with PMES (n=45), there was a statistically significant reduction in migraine days at 3-month follow up in both the groups (STNS: from 6.50 days at baseline to 3.00 days, change -3.50, p=0.012; PMES: from 4.71 days at baseline to 1.86 days, change -2.85, p=0.027). The percentage difference from baseline between the 2 groups was not statistically significant (61% compared with 54%, p=0.88). There were statistically significant decreases in average migraine days (from baseline to 3 months) in both groups but it was not statistically significant (39% compared with 44%, p=0.45; Deng 2020).

In an RCT of 165 people with episodic migraine comparing tSNS plus flunarizine with tSNS or flunarizine alone, monthly migraine days statistically significantly decreased in all 3 groups from baseline to 3‑months follow up (tSNS from 5.92 to 3.73; flunarizine from 5.68 to 3.43; tSNS plus flunarizine from 5.17 to 2.00, p<0.001 for all). There was a greater reduction in the tSNS plus flunarizine group compared with tSNS alone (p=0.039) or flunarizine alone (p=0.041; Jiang 2019).

In a case series of 100 people with migraine with and without aura, and at least 2 attacks per month, the average number of migraine days at 12 weeks in 83 people statistically significantly decreased (from baseline 8.16 to 6.84, p=0.0035), with an average reduction of 1.32 days. When people with chronic migraine (n=23) were compared with those who had episodic migraine (n=60), number of migraine days reduced more for chronic migraine than for episodic migraine at 12 weeks, but the difference was not statistically significant (0.90 compared with 1.06, p=0.543; Danno 2019).

In a case series of 73 people with chronic migraine who had eTNS to prevent migraine, the frequency of migraine days in an ITT analysis (n=58) statistically significantly decreased by 19% from baseline to 3 months (19.02 to 15.67, p<0.001). In the subgroup analysis, number of migraine days was statistically significantly reduced by 23% (15.21 to 11.76, p<0.001) in people with non-continuous headache (n=34), and by 13% (24.42 to 21.21, p<0.05) in people with continuous headache (n=24; Birlea 2019).

Responder rate (reduction in monthly migraine days)

In the RCT of 67 people the 50% responder rate (defined as the percentage of people having a greater than 50% reduction in monthly migraine days) was statistically significantly higher in the tSNS group than in the sham stimulation group (38% [n=13] compared with 12% [n=4], p=0.023) in the ITT analysis. The percentage of people with at least a 25% reduction (moderate improvement) in migraine days was also statistically significantly higher in the tSNS group than in the sham group (59% [n=20] compared with 27% [n=9], p=0.014) (Schoenen 2013, 2016).

In the RCT of 90 people, the 50% responder rate (defined as the percentage of people having a greater than 50% reduction in monthly migraine days) was not statistically significantly different between the STNS and PMES group (62% compared with 78%, p=0.070; Deng 2020).

In the RCT of 165 people, the 50% responder rate (defined as more than a 50% reduction in number of migraine days per month) was statistically significantly higher in the tSNS plus flunarizine group than with flunarizine or tSNS alone (79% compared with 46% compared with 39%, p<0.001; Jiang 2019).

In the case series of 100 people, 50% responder rate (defined as the percentage of people having at least 50% reduction of migraine days between baseline period and after 12 weeks) was 19%, and the 30% responder rate was 29% (Danno 2019).

In the case series of 73 people, the 30% responder rate was 24% and the 50% responder rate was 19% in all people analysed (n=58). In the subgroup analysis, the 30% and 50% responder rates were higher in people with non-continuous headache (n=34) than those with continuous headache (n=24; 30% response rate: 38% compared with 4%; 50% response rate: 29% compared with 4%; Birlea 2019).

Reduction in monthly migraine attacks

In the RCT of 67 people (ITT analysis), the monthly migraine attack frequency was 4.37at baseline and 3.55 at 3 months (p=0.058) in the tSNS group, and 4.04 at baseline and 3.89 at 3 months (p=0.516) in the sham stimulation group. The difference between the 2 groups was statistically significant (p=0.044; Schoenen 2013, 2016).

In the RCT of 90 people, frequency of monthly migraine attacks statistically significantly decreased in both the STNS and PMES groups at 3-month follow up (STNS: from baseline 4.50 to 2.96, p=0.0017; PMES from baseline 3.29 to 1.86, p=0.021). However, the percentage difference between the 2 groups was not statistically significant (34% compared with 44%, p=0.87; Deng 2020).

In the case series of 100 people, the average number of migraine attacks in 83 people statistically significantly decreased from baseline (5.33 to 3.94, p=0.0002). When people with chronic migraine (n=23) and episodic migraine (n=60) were compared, the number of monthly migraine attacks reduced more in people with chronic migraine than those with episodic migraine at 12 weeks, but the difference was not statistically significant (0.75 compared with 0.96, p=0.173; Danno 2019).

Reduction in headache days

In the RCT of 67 people (ITT analysis), there was a statistically significant decrease in monthly days with any headache between baseline and 3 months after treatment in the tSNS group (7.78 to 5.27, p=0.011), but not in the sham stimulation group (6.72 to 6.49, p=0.674). The difference between the 2 groups was statistically significant (p=0.041; Schoenen 2013, 2016).

In an RCT of 91 people (61 with tSNS and pharmacotherapy compared with 30 with pharmacotherapy), there was a statistically significant reduction in the frequency of average monthly headaches after tSNS in all 3 subgroups at 30 days follow up: migraine with aura (from 7.5 to 4.0, p<0.01); migraine without aura (from 7.5 to 4.0, p<0.05); and other primary headaches (from 18 to 10, p<0.001). In the control pharmacotherapy group, there was no statistically significant reduction in frequency of headaches at 30 days in any sub-groups: migraine with aura (from 11 to 7); migraine without aura (from 10.0 to 7.5); and other primary headaches (from 17.0 to 12.5; Przeklasa-Muszynska 2017).

In the case series of 100 people, the average number of headache days in 83 people statistically significantly decreased from baseline (11.48 to 9.81, p=0.009; Danno 2019).

In the case series of 73 people (ITT analysis, n=58), the frequency of headache days (days with at least 1 headache episode) statistically significantly decreased by 16% from baseline to 3-month follow up (22.55 to 19.43, p<0.001). The frequency of moderate or severe headache days (days with at least 1 headache episode with intensity of 2 or 3) also statistically significantly decreased by 22% from baseline to 3-month follow up in all people (from 16 to13, p<0.001). In a sub-group analysis, frequency of headache days reduced by 24% in people with non-continuous headache (n=34) (from 18.71 to 14.29) but only by 5% (from 28.00 to 26.71) in people with continuous headache (n=24; (Birlea 2019).

Reduction in mean headache severity

In the RCT of 67 people (ITT analysis), the mean headache severity per migraine day (on a 4-point scale, with 0 indicating no pain and 3 indicating severe pain prohibiting daily activities) was 1.96 at baseline and 1.8 at 3 months (p=0.131) in the tSNS group, and 1.78 at baseline and 1.73 at 3 months (p=0.443) in the sham group. The difference between the 2 groups was not statistically significant (p=0.301; Schoenen 2013, 2016).

In the RCT of 90 people, the change in the headache impact test-6 decreased in both the STNS and PMES groups from baseline 3-month follow up (STNS: 63.50 to 40.33, p=0.007; PMES: 69.29 to 51.57, p=0.028). The change was higher in the STNS group than that in the PMES group (37% compared with 26%, p=0.041; Deng 2020).

In the RCT of 165 people, migraine severity (assessed using a 10‐point VAS, with higher scores representing severe pain prohibiting daily activities) decreased statistically significantly in all 3 comparison groups from baseline to 3-month follow up (tSNS from 6.75 to 5.53; flunarizine from 6.96 to 4.82; tSNS plus flunarizine from 6.57 to 3.32, p<0.001 in all). The decrease was greater in the tSNS plus flunarizine group compared with tSNS alone (p<0.001) or flunarizine alone (p=0.030; Jiang 2019).

In the RCT of 91 people, a statistically significant reduction in pain intensity during a headache episode (assessed using 10 point NRS, with higher scores indicating strong pain) was reported after tSNS and pharmacotherapy in all 3 subgroups at 30-day follow up: migraine with aura (from 9.0 to 5.5, p<0.001); migraine without aura (from 8.5 to 5.5, p<0.001); and other primary headaches (from 7.5 to 4.5, p<0.001). In the control pharmacotherapy group, there was no statistically significant reduction in pain intensity at 30 days in any subgroups: migraine with aura (from 8.8 to 7.8); migraine without aura (from 9.0 to 7.6); and other primary headaches (from 8.5 to 7.2). In people with strong pain (NRS 7 to10), pain reduction after tSNS treatment was 33% to 34% in migraine with aura, migraine without aura, and other primary headaches. In people who only had pharmacological treatment, the pain reduction was about 10% to 13%.Subjective improvement (assessed by people) after tSNS was about 40% to 47% compared with about 23% to 29% in the control group (Przeklasa- Muszynska 2017).

In the case series of 100 people, there was no statistically significant difference in average severity of headache in 83 people from baseline (4.5 to 4.11, p=0.122; Danno 2019).

In the case series of 73 people, there was a statistically significant reduction in average headache intensity (assessed on a scale 1 to 3) of 5% from baseline to 3-month follow up (1.98 to 1.87, p<0.05). In the subgroup analysis, headache intensity statistically significantly decreased by 10% in people with continuous headache (n=24; 2.08 to 1.87, p<0.05), but not in those with non-continuous headache (n=34; 1.92 to 1.87, p=NS; Birlea 2019).

Reduction in headache episodes and cumulative hours of headache

In the RCT of 91 people, there was a statistically significant reduction in pain duration after tSNS in all 3 subgroups 30-day follow up: migraine with aura (from 27 to 13 hours, p<0.05); migraine without aura (from 23 to 9 hours, p<0.05); and other primary headaches (from 12 to 7 hours, p<0.001). In the control pharmacotherapy group, there was no statistically significant reduction in pain duration at 30 days: migraine with aura (from 15 to 12 hours); migraine without aura (from 18 to 13 hours); and other primary headaches (from 7 to 6 hours, Przeklasa-Muszynska 2017).

In the case series of 73 people ( ITT analysis; n=58), the frequency of headache episodes (number of patient-reported headache attacks) statistically significantly decreased by 16% from baseline to 3 months (22.66 to 19.53, p<0.001). The monthly cumulative headache hours also statistically significantly decreased by 15% (from 249 to 222, p<0.05). In the subgroup analysis, the mean changes were higher in people with non-continuous headache (n=34; 166.00 to 128.92, mean change 23%, p<0.01) than those with continuous headache (n=24; 368.14 to 354.88, mean change 4%; Birlea 2019).

Reduction in monthly acute migraine drug intake

In the RCT of 67 people (ITT analysis), there was a statistically significant decrease in the monthly intake of migraine drugs for acute attacks in the tSNS group (11.45t o 7.25, p=0.0057), but not in the sham group (9.24 to 9.28, p=0.822). The difference between the 2 groups was statistically significant (p=0.0072; Schoenen 2013, 2016).

In the RCT of 90 people, monthly intake of acute antimigraine drugs statistically significantly decreased in both the STNS and PMES group from baseline to 3-month follow up (STNS: 6.2 to 2.43, p=0.004; PMES 5.5 to 1.9, p=0.003). However, the percentage difference between the 2 groups was not statistically significantly different (66% compared with 61%, p=0.71; Deng 2020).

In the RCT of 165 people, the monthly intake of acute antimigraine medication statistically significantly reduced during attacks in all 3 comparison groups (tSNS group from 4.88 to 2.78; flunarizine group from 4.96 to 2.89; tSNS plus flunarizine group from 4.15 to 1.06, p<0.001 for all). This was more statistically significant with tSNS plus flunarizine than with tSNS (p<0.013) or flunarizine (p=0.02; Jiang 2019).

In the case series of 100 people, the average intake of acute antimigraine drugs in 83 people statistically significantly decreased from baseline (8.75 to 7.83, p=0.0166). The number of migraine days were more reduced in people with chronic migraine (n=23) than those with episodic migraine (n=60) at 12 weeks, but the difference was not statistically significant (0.85 to 0.96, p=0.406; Danno 2019).

In the case series of 73 people who had eTNS, mean monthly acute antimigraine medication intake was statistically significantly reduced by 31% from baseline to 3-month follow up (26.33 to 18.22, p<0.001). This reduction was higher in the subgroup of people with continuous headache (n=24; mean change -36%, p<0.01) than in those with non-continuous headache (n=24; mean change -25%, p<0.05; Birlea 2019).

Patient satisfaction

In the RCT of 67 people, the percentage of very or moderately satisfied people was higher in the tSNS group (71%) than in the sham group (39%; Schoenen 2013, 2016).

In the RCT of 165 people, 84% (43/51) of people who had tSNS plus flunarizine were satisfied with treatment and expressed a desire to continue the treatment compared with 55% (28/51) of people who had tSNS alone and 50% (26/52) of people who had flunarizine alone (p=0.001; Jiang 2019).

In a case series of 2,313 people, 53% (1,236/2,313) of people were satisfied and wanted to continue the treatment (Magis 2015).

In the case series of 100 people, 66% (63/95) of people who answered the questionnaire were satisfied, and the reasons for satisfaction were improvement in headache in 45% and reduction in acute medications in 36% (Danno 2019).

Device compliance

In the RCT of 67 people, device compliance (mean numbers of sessions assessed by a built-in system in the device) was 62% (56 sessions) in the tSNS group and 55% (49 sessions) in the sham stimulation group. The difference between the 2 groups was not statistically significant (Schoenen 2013, 2016).

In the case series of 2,313 people, 49% of people used the device for the recommended period of time (Magis 2015).