Acute and Chronic Management of Posttraumatic Headache in Children

A Systemic Review

Carlyn Patterson Gentile MD, PhD; Ryan Shah BS; Samantha L. Irwin MSc, MB, BCh, BAO, FRCPC; Kaitlin Greene MD; Christina L. Szperka MD, MSCE, FAHS

Disclosures

Headache. 2021;61(10):1475-1492.

Abstract and Introduction

Abstract

Objectives: The goal of this paper is to provide a compilation of the evidence for the treatment of posttraumatic headache (PTH) in the pediatric population. Headache features and timing of therapy were considered.

Background: Headache is the most common symptom following mild traumatic brain injury (mTBI), affecting more than 80% of children and adolescents. It is unclear whether treatment for PTH should be tailored based on headache characteristics, particularly the presence of migraine features, and/or chronicity of the headache.

Methods: Systematic literature searches of PubMed, Embase, Scopus, and Cochrane databases (1985–2021, limited to English) were performed, and key characteristics of included studies were entered into RedCAP® (Prospero ID CRD42020198703). Articles and conference abstracts that described randomized controlled trials (RCTs), cohort studies, retrospective analyses, and case series were included. Participants included youth under 18 years of age with acute (<3 months) and persistent (≥3 months) PTH. Studies that commented on headache improvement in response to therapy were included.

Results: Twenty-seven unique studies met criteria for inclusion describing abortive pharmacologic therapies (9), preventative pharmacotherapies (5), neuromodulation (1), procedures (5), physical therapy and exercise (6), and behavioral therapy (2). Five RCTs were identified. Studies that focused on abortive pharmacotherapies were completed in the first 2 weeks post-mTBI, whereas other treatment modalities focused on outcomes 1 month to over 1-year post-injury. Few studies reported on migrainous features (7), personal history of migraine (7), or family history of migraine (3).

Conclusions: There is limited evidence on the timing and types of therapies that are effective for treating PTH in the pediatric population. Prospective studies that account for headache characteristics and thoughtfully address the timing of therapies and outcome measurement are needed.

Introduction

Headache is the most common symptom following mild traumatic brain injury (mTBI). Posttraumatic headache (PTH) is reported by more than 80% percent of children and adolescents.^[1] Eight percent of children who have had head trauma will have headache that persists beyond 3 months,^[2] fulfilling International Classification of Headache Disorders, 3rd edition (ICHD-3) criteria for persistent headache attributed to trauma or injury to the head and/or neck, here termed persistent PTH.^[3] Migraine is the most common phenotype of PTH in children and is seen in up to 55% of cases.^[4] Prior work in children suggests that PTH with migrainous features^[5–9] or a personal or family history of migraine is associated with prolonged recovery.^[5,6] The PTH "migraine phenotype" may represent a subset of patients who would benefit from more directed headache management.^[9] This is of particular interest in the pediatric population due to the high incidence of migraine in the teenage years and early twenties.^[10] In this age group, individuals prone to developing migraine may not have exhibited symptoms yet and mTBI may act as a triggering event for migraine.^[11] Three reviews have focused on PTH treatment in the past decade.^[12–14] Although informative, only one looked exclusively at children,^[14] and none reported the presence of migrainous features.

The objective of this review was to identify randomized controlled trials (RCTs), cohort studies, retrospective analyses, and case series of youth (<18 years) with acute or persistent PTH that assessed the effect of pharmacologic, procedural, and non-pharmacologic treatments on headache improvement. Headache characteristics, including presence of migrainous features, and the timing of therapies were evaluated when available. This review identifies gaps in knowledge regarding the management of pediatric PTH to help guide future treatment trials.

1 2 3 4 5

Next Section

Abstract and Introduction
Methods
Results
Discussion
Conclusion
References
Appendix

Table 1. Data quality assessment definitions
Data quality	Type of evidence
Good	• Well-designed, well-executed randomized controlled trials
Fair	• Randomized controlled trials with minor limitations including minor flaws in design or execution • Well-designed, well-executed nonrandomized controlled studies, or observational studies
Poor	• Randomized controlled trials with major limitations • Nonrandomized intervention or observational studies with major limitations • Uncontrolled clinical observations without comparison group (e.g., case series, case reports)

Table 2. Synopsis of findings from articles included in the systematic review
First author (date)	Study type	Participant characteristics		Treatment	Outcome	Data quality
Abortive pharmacotherapies (oral)
Petrelli et al. (2017)¹⁷	Prospective randomized open-label control trial	Participants presented to the emergency department within 48 h of mTBI	Migrainous headache: 39%	Treatment with the following analgesic regimens for 1 week: 1. Ibuprofen 10 mg/kg q8 h ×72 h (n = 20) 2. Acetaminophen 10–15 mg/kg q4 h ×72 h (n = 20) 3. Alternating acetaminophen and ibuprofen ×72 h (n = 19) 4. Standard of care: no analgesic recommendations (n = 20)	Outcome measures: Number of headache days, number of headache episodes, headache intensity, and higher return to school rate over 1 week of treatment	Fair
		Age range (mean): 8–18 years (12.9)	Hx of migraine: NR		Outcome: • Patients in any of the treatment arms had fewer headache days, headache episodes, and lower headache intensity, and a greater percentage were back to school compared to the standard of care group. • Subjects in the alternating group had the best response • Ibuprofen.: #headache days 5/7, back to school 61% • Acetaminophen: #headache days 7/7, back to school 33% • Both: #headache days 4/7, back to school 79% • Standard of care: #headache days 7/7, back to school 21% • Mean (SD) headache episodes/patient/day and headache intensity for each of the 7 days in the trial were reported
		Age range (mean): 8–18 years (12.9)	FamHx of migraine: NR
		n = 79 (50% female)
		**Included 18 years old
Heyer et al. (2014)²²	Retrospective cohort	Participants meeting criteria for medication overuse headache (MOH) who presented to a headache clinic >3 months following mTBI	Migrainous headache: NR	Discontinuation of as-needed analgesics (ibuprofen, acetaminophen, naproxen, oxaprozin)	Outcome measure: Headache improvement or resolution within 2 months of discontinuing analgesics	Poor
		Age range (median): 13–17 years	Hx of migraine: 37%		Outcome: • 68.5% of subjects with MOH reported headache improvement after discontinuing analgesics. No further statistical analysis was reported
		(15.2)	FamHx of migraine: NR
		n = 77 (63% female; 54/77 met criteria for MOH)	Note: 75.9% of participants with MOH had daily episodic or continuous headache
Cohen and Conidi (2015)²⁰ (abstract)	Retrospective cohort	Participants were diagnosed with concussion/mTBI at a tertiary care center. Age range: 14–19 years n = 19 (randomly selected)	Migrainous headache: NR	Any reported therapies for PTH	Outcome measure: Headache relief	Poor
		**Included ≥18 years old	Hx of migraine: NR		Outcome: • Triptans and onabotulinumtoxinA reported as providing the most rapid headache relief. No statistical analysis was provided
		**Included ≥18 years old	FamHx of migraine: NR
Tejani et al. (2017, 2019)^{18, 19} (abstract)	Prospective cohort	Participants presented to a concussion clinic within 30 days of concussion/mTBI	Migrainous headache: NR	Effect of NSAID and triptan use within 30 days of concussion/mTBI 1. NSAIDs only (61% of participants) 2. Triptans only (1.1% of participants) 3. NSAIDs + triptans (5.3% of participants) 4. No treatment (32.6% if participants)	Outcome measure: Headache resolution, and headache severity at 3 months	Poor
		Age range: 6–18 years n = 528 (45% female)	Hx of migraine: NR		Outcome: • Reduction in number of subjects reporting headache from 75.4% to 18% regardless of treatment group with no significant difference between the NSAID group and no treatment group (81.4% → 18.6% vs. 86.6% → 13.4%, p = 0.165) • A higher proportion of subjects who used triptans reported greater headache severity compared to the NSAIDs group (5/5 triptans group vs. 10/28 NSAIDs + triptans group vs. 23/172 NSAID group, p < 0.001) • Subjects who used triptans reported greater headache severity than those who used NSAIDs (mean triptan only 1.6 95% CI: 1.0–2.4, mean triptan+NSAID 1.0 95% CI: 0.7–1.4, mean NSAID only 0.3 95% CI: 0.2–0.4, mean no med 0.2 95% CI: 0.1–0.3)
		**Included 18 years old	FamHx of migraine: NR
Bramley et al. (2012)²¹	Retrospective case series	Case series included a 10-year-old male and a 15-year-old female presenting with concussion/mTBI within the first month	Migrainous headache: 100%	4 mg 6-day tapering Medrol dose pack in addition to other abortive and preventative therapies. Other medications included amitriptyline, melatonin, and gabapentin	Outcome measure: Descriptive, headache improvement	Poor
			Hx of migraine: NR		Outcome: • Both subjects reported improvement with Medrol dose pack • One subject completed an additional 14-day taper because of symptom return following completion of the first taper. No significance testing was reported
			FamHx of migraine: NR
Abortive pharmacotherapies (intravenous)
Lumba-Brown et al. (2014)²³	Prospective randomized controlled trial	Participants were recruited acutely following concussion/mTBI in an emergency department (duration post-injury not specifically reported)	Migrainous headache: NR	Participants were randomized to receive a single dose of 10 ml/kg of one of the following given over 1 h: 1. 3% hypertonic saline (HTS, 52% of participants) 2. Normal saline (NS, 48% of participants)	Outcome measure: Wong-Baker face scale pain score before versus immediately after infusion, and 2–3 days later	Fair
		Age range (mean): 7–16 years	Hx of migraine: NR		Outcome: • Participants who received HTS reported a greater reduction in pain scores compared to NS immediately (3.52 vs. 1.14-point decrease, mean difference between groups −2.38 95% CI: −0.3 to −1.4, p < 0.001) and 2–3 days after infusion (4.61 vs. NS 3.00-point decrease, mean difference between groups −1.61 95% CI: −2.90 to −0.30, p = 0.014) • A multiple-regression model used to adjust for initial higher pain reports in the HTS group and other potential confounders still demonstrated a significantly greater reduction in pain in the HTS compared to NS group (mean difference between groups −1.86 95% CI: 0.92 to 2.79)*95% CI intervals may contain an error, and sign conventions may also be reported in error
		(12)	FamHx of migraine: NR
		n = 44 (37% female)	FamHx of migraine: NR
Bresee et al. (2018)²⁴	Prospective cohort, planned subanalysis	Participants presented to the emergency department within 48 h of concussion/mTBI	Migrainous headache: NR	Participants received one of the following: 1. Single dose of IV metoclopramide (dosage not reported, 3% of participants) 2. No treatment (97% of participants)	Outcome measure: Persistent headache at 1 and 4 weeks	Fair
		Age range: 8–17 years	Hx of migraine: 13%		Outcome: • IV metoclopramide was not associated with lower rate of persistent headaches at 1 week (unadjusted RR 1.0 95% CI: 0.8–1.3; adjusted multivariable logistic regression OR 0.6 95% CI: 0.4–1.1; adjustment with propensity score using 1:4 matching algorithm RR 0.8 95% CI: 0.6–1.2) • At 4 weeks (unadjusted RR 1.0 95% CI: 0.9–1.2; adjusted multivariable logistic regression OR 0.8 95% CI: 0.4–1.5; adjustment with propensity score using 1:4 matching algorithm RR 0.9 95% CI 0.8–1.1)
		n = 2095 (41% female)	FamHx of migraine: 47%
		n = 2095 (41% female)	Note: 75% reported constant headache
Chan et al. (2015)²⁵	Retrospective cross-sectional	Participants presented to the emergency department within 14 days of concussion/mTBI (mean 2 days) and received IV headache treatment	Migrainous headache: NR	Participants received IV NS bolus and the following IV medications (dosages not reported): 1. Ketorolac only (21% of participants) 2. Prochlorperazine, metoclopramide, chlorpromazine (12% of participants) 3. Ondansetron only (15% of participants) 4. Ketorolac + metoclopramide or prochlorperazine (52% of participants)(4)	Outcome measure: Treatment success defined as >50% reduction in headache severity score (0–10) after infusion. Headache resolution was also reported	Poor
		Age range (mean): 8–21 years	Hx of migraine: 21%		Outcome: • 86% (95% CI: 82.0%–90.4%) had treatment success overall • 80% in the ketorolac only group • 93% in the metoclopramide or prochlorperazine only group • 89% in the combined group • 79% in the ondansetron only group • 52.4% (95% CI: 46.3–58.5) reported headache resolution overall • Patients who got a head CT were less likely to have treatment success (80% vs. 91%, p = 0.008) • Patients who did not get a head CT were significantly more likely to have treatment success (multivariable analysis OR 2.95, 95% CI: 1.39–6.56)
		(13.8)	FamHx of migraine: NR
		n = 254 (51% female)
		**Included ≥18 years old
Katz et al. (2018)²⁶ (abstract)	Retrospective cohort	Participants presented with concussion/mTBI to an outpatient infusion center	Migrainous headache: NR	Participants received a single dose of IV ketorolac, prochlorperazine, diphenhydramine, and 20 ml/kg NS	Outcome measure: Headache improvement, headache severity scale (0–6) at follow-up after the infusion	Poor
		Age: ≤18 years	Hx of migraine: NR	IV fluid (medication dosages not reported)	Outcome: • 23/27 (85%) reported headache improvement, 1/27 (4%) found no improvement, 3/27 (11%) were not asked • Mean headache score was reduced from a 4/6 to a 2/6. No significance testing was reported
		n = 27	FamHx of migraine: NR
		**Included 18 years old	FamHx of migraine: NR
Preventative pharmacotherapies
Kuczynski et al. (2013)⁴⁷	Retrospective cohort	Participants presented with concussion/mTBI to a concussion clinic a median of 6.9 months (range 1–29 months) post-injury	Migrainous headache: 39%	Participants were placed on the following medications for a mean duration of 5.5 weeks (SD 4.3 weeks): 1. Amitriptyline, up to maximum of 1 mg/kg/day based on response (n = 18, 41%) 2. Nortriptyline, dosage not given (n = 9, 20%) 3. Topiramate up to 1.5–2 mg/kg/day to a max 200 mg/day (n = 6, 14%) 4. Flunarizine, dosage not given (n = 8, 18%) 5. Melatonin, 3–10 mg nightly (n = 12, 23%) 6. Indomethacin, dosage not given (n = 2, 5%) 39% received more than one treatment	Outcome measure: Significant response defined as ≥50% reduction in headache frequency. Headache resolution was also reported	Poor
		Age range (mean): ≤18 years	Hx of migraine: 14%		Outcome: • 64% reported successful response to preventative migraine medications: • 13/18 (72%) on amitriptyline • 1/11 (11%) on nortriptyline • 0/6 (0%) on topiramate • 5/8 (63%) on flunarizine • 9/12 (75%) on melatonin • 2/2 (100%) on indomethacin (had stabbing headache) • 45% reported headache resolution • Length of time for follow-up was not a significant predictor of symptom resolution (χ ² = 0.341, p = 0.559)
		(14.1)	FamHx of migraine: 52%
		n = 44 (66% female)	Note: 61% experienced daily headache at the start of treatment
		**Included 18 years old
Bramley et al. (2015)²⁸	Retrospective cohort	Participants presented to a concussion clinic, "majority within 1–3 weeks of concussion"	Migrainous headache: NR	Amitriptyline started in those with persistent headache at 1 month postconcussion (n = 68; 17% of participants) at 10–100 mg/day (median 25 mg/day) for a median duration of 4 months (range 2–7 months)	Outcome measure: Improvement in headache symptoms	Poor
		Age range: 13–18 years	Hx of migraine: NR		Outcome: • 82% (95% CI 70%–91%) reported improvement in headache symptoms • 33% (95% CI 12%–38%) of those on amitriptyline reported side effects • Females were more likely to be prescribed amitriptyline (24% vs. 13%, p = 0.004)
		n = 400 (38% female)	FamHx of migraine: NR
		**Included 18 years old	FamHx of migraine: NR
Markus et al. (2016)⁴⁶	Retrospective case–control study	Participants with TBI (81% with concussion/mTBI) presented to a headache clinic a median of 12 months (range 2–120 months) after injury	Migrainous headache: 54%	20 participants were prescribed one of the following agents: 1. Amitriptyline (n = 11) 2. Propranolol (n = 6) 3. Topiramate (n = 3) (Dosages not reported)	Outcome measure: Positive response was defined as ≥50% reduction in number of headache days	Poor
		Age range (median): 2–17 years	Hx of migraine: NR		Outcome: • 10/20 (50%) of TBI participants prescribed preventative medication reported a positive response: • 36% improved on amitriptyline • 50% on propranolol • 100% on topiramate • No significance testing was reported
		(9)	FamHx of migraine: NR
		n = 74 (43% female)	FamHx of migraine: NR
Mackie and Kirkham (2017)²⁹ (abstract)	Retrospective cohort	Participants presented with concussion/mTBI to a nurse-led neurology clinic	Migrainous headache: 71%	All participants received school accommodations/decreasing cognitive load. In addition, 16 participants (42%) were treated with topiramate (median dose 2 mg/kg/day)	Outcome measure: Headache improvement, reported reduction in migraine frequency/migraine severity	Poor
		Age range (median): 3–16 years	Hx of migraine: NR		Outcome: • 84% reported headache improvement from the whole cohort • 12/16 participants (75%) prescribed topiramate reported reduction in frequency and severity of migrainous headaches • No significance testing was reported
		(13.8)	FamHx of migraine: NR
		n = 38 (45% female)	FamHx of migraine: NR
Sabo et al. (2017, 2018)^{30, 31}	Retrospective case series	Case series of children and adolescents diagnosed with pseudotumor cerebri following mTBI	Migrainous headache: NR	Acetazolamide 750–1000 mg BID (not all dosing reported) for a treatment interval 6 weeks–4 months	Outcome measure: Headache improvement	Poor
		Age range: 5–16 years n = 6 (66% female)	Hx of migraine: NR	Topiramate (dosing not reported)	Outcome: • 4/5 had improvement with acetazolamide, the nonresponder was found to have craniosynostosis • 1/1 had improvement with topiramate + acetazolamide • Two worsened on amitriptyline, and one worsened on propranolol, and one had only mild improvement on cyproheptadine prior to being diagnosed with pseudotumor cerebri • No significance testing was reported
		Case series of children and adolescents diagnosed with pseudotumor cerebri following mTBI	FamHx of migraine: NR	Topiramate (dosing not reported)
Neuromodulation
Pinchuk et al. (2013)³³	Retrospective cohort	Participants with diagnosis of chronic PTH based on ICHD-2 criteria, 6–9 months following concussion/mTBI	Migrainous headache: NR	Transcranial direct current stimulation (tDCS) for 30–45 min per session with 60–90 μA intensity (number of sessions not reported)	Outcome measure: Headache frequency (days/month), headache severity scale (0–10), % patients with ≥50% reduction in headache frequency, and % patients with headache resolution for ≥4.5 months post-tDCS	Poor
		Age range (median): 11–16 years	Hx of migraine: NR		Outcome: • 52% reported headache resolution, with an additional 29% reporting ≥50% reduction in headache frequency • Headache severity significantly improved (mean 5.4 SD 1.8 → mean 1.9 SD 1.2; p < 0.001) • Headache frequency significantly improved (mean 10.3 SD 6.5 → mean 4.1 SD 2.2; p < 0.001)
		(13.6)	FamHx of migraine: NR
		n = 44	FamHx of migraine: NR
Procedures
Dubrovsky et al. (2014)³⁴	Retrospective case series	Participants with mean of 70 days post-injury with a diagnosis of acute or persistent PTH based on ICHD-3	Migrainous headache: 83%	Greater occipital nerve block ±lesser occipital nerve block ± supraorbital nerve block with 2% lidocaine + epinephrine; participants received between 1 and 6 nerve blocks with median duration between repeated nerve blocks of 18.5 days. Many subjects received abortive therapies, preventative therapies, or both concomitant to their nerve block	Outcome measure: Good therapeutic effect was defined as decrease in pain for >24 h or requested repeat nerve block, headache pain severity immediately after nerve block (0–10 scale)	Poor
		Age range (mean): ≤18 years	Hx of migraine: NR		Outcome: • 93% reported good therapeutic effect, and 71% of nerve blocks resulted in 0/10 headache pain • Headache intensity decreased pre- versus post-headache (mean = 5.6, SD = 1.6 vs. mean = 0.4, SD = 0.9); % reduction mean = 93%, SD = 13%
		(14.6)	FamHx of migraine: 77%
		n = 28	Note: history of headache was reported to be 70%
		**Included 18 years old	Note: history of headache was reported to be 70%
Seeger et al. (2015)³⁵	Retrospective case series	Participants treated in concussion and TBI clinic who received a nerve block for PTH, mean of 5.6 months post injury (range 1–12 months)	Migrainous headache: 40%	Greater occipital nerve block with 2% lidocaine + steroid (triamcinolone or methylprednisolone). Study did not comment on concomitant use of other therapies	Outcome measure: Full response was defined as >50% reduction in headache frequency, mean headache frequency (days/month) at follow-up was also reported. Routine follow up was via telephone 2–4 weeks after the procedure and 2–3 months for an in-person visit. Secondary outcomes included Rivermead Post-concussion Symptoms Questionnaire (RPSQ), and the Pediatric Quality of Life Inventory (PedsQL)	Poor
		Age range (mean): 13–17 years	Hx of migraine: 6%		Outcome: • 64% of children had full response • Headache frequency decreased from mean 26 SD 7 days/month to mean 18 SD 12 days/month (p = 0.014) • RPSQ scores significantly decreased following treatment (mean 33.67 SD 13.47 vs. 24.33 SD 18.54; t(5) = 3.04, p = 0.029) • PedsQL improved post-treatment (mean 52.04 SD 21.36 vs. 68.75 SD 27.50; t(7) = −2.62, p = 0.035)
		(15.5)	FamHx of migraine: NR
		n = 15 (14 presented for follow up), (67% female)	Note: 13/15 (87%) had daily headaches at time of procedure
		n = 15 (14 presented for follow up), (67% female)	6/15 (40%) met ICHD-3 criteria for occipital neuralgia
Zaremski et al. (2015)³⁶	Retrospective case series	Participants (16, 17, and 28 years old) presented with symptoms of occipital neuralgia following concussion/mTBI	Migrainous headache: NR	Variable: Greater and lesser occipital nerve block, trigger point injections, cervical medial branch block of C2 to C3, third occipital nerve blocks	Outcome measure: Headache improvement	Poor
		**Included ≥18 years old	Hx of migraine: NR		Outcome: • Both participants under 18 years old experienced complete but temporary relief from nerve blocks. No significance testing was reported
		**Included ≥18 years old	FamHx of migraine: NR
Lin and Tung (2016)³⁷	Retrospective case series	Participants (8 yM, 15yF, and 18 yF) with PTH	Migrainous headache: NR	Acupuncture for 6 to >12 sessions initiated 4–6 weeks following concussion/mTBI for persistent headache	Outcome measure: Headache frequency, headache severity (10-point scale)	Poor
		**Included 18 years old	Hx of migraine: NR		Outcome: • All subjects reported reduction in headache frequency and a 3–7 point reduction in headache severity. No significance testing was reported
		**Included 18 years old	FamHx of migraine: NR
Physical therapy and exercise
Schneider et al. (2014)³⁸	Prospective randomized controlled trial	Participants with persistent postconcussion symptoms a mean of 50 days (range 8–276) following concussion/mTBI	Migrainous headache: NR	Participants were randomized into the following groups: 1. Intervention group (n = 15): cervical spine and vestibular rehabilitation 2. Standard of care (n = 14)	Outcome measure: Numeric pain rating scale for headache (010) compared between those who were cleared to return to sport vs. those who were not cleared in the treatment and control groups. Primary outcome measure was return to sport within 8 weeks of treatment	Fair
		Age range (median): 12–30 years	Hx of migraine: NR	Both groups worked with a physiotherapist ×8 weeks or until medical clearance (mean number of sessions for each group = 6)	Outcome: • There was no significant difference in reduction of headache severity between cleared and not cleared subjects in the treatment or the control groups. Change in severity score (post – pre-treatment): • Control group: Cleared: −7 (n = 1); Not cleared: −2.5 (range −7 to 1, n = 12^a one participant in this group did not complete the follow-up questionnaire), p > 0.05 • Intervention group: Cleared: −3 (−8 to 0, n = 11); Not cleared: −2.5 (range −4 to 1, n = 4), p > 0.05 • Treatment group was 10.27 times more likely to be medically cleared to return to sport within 8 weeks (χ ² = 13.08, p < 0.001)
		(15)	FamHx of migraine: NR	Both groups received postural education, range of motion exercises, cognitive and physical rest until asymptomatic followed by graded exertion
		n = 31 (42% female; 2 withdrew)
		**Included ≥18 years old
Gauvin-Leplage et al. (2019)⁴¹	Retrospective analysis of a prospective cohort	Participants reported ongoing symptoms at 4 weeks following concussion/mTBI	Migrainous headache: NR	Exercise based active rehabilitation intervention ×4 weeks	Outcome measure: Post-concussion symptom scale (PCSS) headache symptom score (0–6) compared from baseline to 2 and 4-weeks f/u. Primary outcome was total PCSS score at these visits	Fair
		Age range (mean): 8–17 years	Hx of migraine: NR		Outcome: • PCSS headache scores decreased over the 4 weeks of treatment. PCSS headache score: Girls 2.5 at baseline, 2.1 at 2-weeks f/u, 1.5 at 4-weeks f/u; Boys 1.9 at baseline, 1.3 at 2-weeks f/u, 1.1 at 4-weeks f/u; ANOVA, p < 0.001 for headache score change over time • More girls reported headache, and had higher PCSS toral scores at baseline, and experienced a faster recovery compared to boys. Repeated-measures ANOVA, sex-specific difference for headache (p = 0.046). Generalized mixed regression model total PCSS by gender: Boys visit OR (95% CI): Visit 1: −8.81 (−12.37 to −5.26); Visit 2: −7.37 (−9.6- to −5.14); Visit 3: −11.40 (−14.33 to −8.48)
		(14.3)	FamHx of migraine: NR
		n = 355 (53% female)	FamHx of migraine: NR
De Matteo et al. (2015)⁴⁰	Prospective cross-sectional study	Participants presented to a concussion clinic a median of 4.1 months (range 0.7–35) post-injury	Migrainous headache: NR	Exertional testing ×1 session	Outcome measure: Reduction in headache symptom score (0–6) at 5-min, 30-min, and 24-h post-exertional testing, headache improvement. Primary outcome was total decrease in symptom scores	Poor
		Age range (mean): 8–18 years	Hx of migraine: NR		Outcome: • There was no significant difference in headache scores before vs. after exertional testing. Headaches worsened for 5 subjects and improved for 9 subjects. Mean headache symptom score decreased from 0.97 pre-exertion to 0.78 at 24 h post-exertion (Z = −1.10, p = 0.27), 30-min after exertion was also not significant (p = 0.39, Friedman test) • Overall, the total number of symptoms and symptom severity significantly decreased. There was a significant total decrease in symptoms at 30 min and 24 h post-exertion testing (30 min: severity χ ² = 14.00, p = 0.001; number χ ² = 8.7, p = 0.01; 24 h: severity Z = −3.47, p < 0.01)
		(14.8)	FamHx of migraine: NR
		n = 54 (41% female; 69% completed follow-up)
		**Included 18 years old
Grabowski et al. (2017)⁴²	Retrospective cohort	Participants were youth athletes with concussion symptoms persisting beyond 3 weeks (80% reported headache at initial visit)	Migrainous headache: NR Hx of migraine: 20% (also included chronic tension-type headache)	PT, cardiovascular exercise, vestibular/oculomotor exercise, cervicothoracic manual therapy and exercise, and sports specific training. Mean therapy duration was 84 days (range 7–266)	Outcome measure: Headache score of how bothersome headaches were from post-concussion symptom scale (PCSS, 0–6). Primary outcome measure was change in total PCSS	Poor
		Age range (mean): 12–20 years	FamHx of migraine: NR		Outcome: • Patients reported significantly decreased PCSS score post-PT. Total PCSS scores: Pre-PT mean 18.2 (SD = 14.2); Post-PT mean 9.1 (SD = 10.8); p < 0.01 • Significance testing was not done on the difference in headache scores pre- versus post-PT. Headache score: Pre-PT median 2 (range 0–5); Post-PT median 1 (range 0–5) • Athletes with a history of chronic migraine or tension-type headache had a mean 10.22 point higher post-PT score compared to athletes with no headache history (p = 0.05)
		(15)
		n = 25 (56% female)
		**Included ≥18 years old
Bailey et al. (2019)³⁹	Pilot, prospective randomized controlled trial	Participants with concussion at least 4 weeks post-injury (mean 56 days)	Migrainous headache: NR	Participants were randomized into the following groups: 1. Sub-threshold exercise program (n = 7): targeted 80% of heart rate that correlated with symptom exacerbation with physiotherapist ×6 weeks 2. Control group (n = 8): instructed on daily stretching activities at home ×3 weeks, then daily walking ×3 weeks	Outcome measure: % of participants with mild or no headaches defined as a Post-concussion Scale-Revised (PCS-R) headache score of ≤ 2 (0–6 scale). Primary outcome measure was change in total PCS-R score	Poor
		Age range (mean): 14–18 years	Hx of migraine: 37%		Outcome: • Total PCS-R scores decreased more in the intervention than in the control group when depression was accounted for (PCS-R % change: intervention group mean = −63.3%, SD = 17.4; control group mean = −56.8%, SD = −27.8; F _2,13 = 5.20, p < 0.05, partial η² = 0.32) ANOVA (depression was a covariate). *1 subject from the intervention group was removed prior to analysis due to worsening PCS-R score and high baseline depression score • % with mild or no headache was compared between baseline and post-treatment (intervention group 33% → 67%; control group 33% → 63%). Significance testing was not done
		(15.75)	FamHx of migraine: NR
		n = 16 (44% female; 1 withdrew due to repeat injury)
		**Included 18 years old
Wilson et al. (2020)⁴³	Retrospective cohort	Participants seen by a concussion specialist within 21 days of concussion stratified by early physical activity (PA)	Migrainous headache: NR	Participants were stratified based on reports of early physical activity (prior to initial evaluation): 1. PA group (n = 69): Reported early physical activity 2. No PA group (n = 506): Reported that they did not have early physical activity	Outcome measure: Reporting persistent headache at the initial visit. Primary outcome was symptom resolution times	Poor
		Median days post-injury was 8 for the early PA group and 12 for the no PA group (p < 0.001)	Hx of migraine: NR		Outcome: • The no PA group had significantly longer recovery times than the PA group (median 16 IQR 8–24 vs. 10.5 IQR 4–17, p = 0.02) • Subjects in the PA group were significantly less likely to report persistent headache at their initial visit when adjusting for pre-existing headache history (adjusted OR = 0.14, 95% CI: 0.07–0.26)
		Age range (mean): 8–18 years	FamHx of migraine: NR
		(14.8)	Note: History of pre-existing headache was reported (28% across the cohort, no significant difference between groups)
		n = 575 (34% female)
		**Included 18 years old
Collaborative care, cognitive-behavioral therapy, and biofeedback
McCarty et al. (2021)⁴⁴	Prospective randomized controlled trial	Participants presented to a concussion clinic within 9 months of concussion/mTBI with ≥3 persistent symptoms at least 1 month after concussion	Migrainous headache: NR	Treatment: Cognitive behavioral therapy and care management with enhanced medication consultation if warranted ×6 months	Outcome measure: Traumatic Brain Injury-Quality of Life Headache Pain (13-item Likert-based scale). Primary outcome measure was improvement in postconcussive symptoms	Fair
		Age range (median): 11–18 years	Hx of migraine: NR	Standard of care: Standard management by subspecialty clinics ×6 months	Outcome: • There was a significant difference between postconcussive symptom outcomes between the collaborative care group and standard of care group (net difference from baseline between groups mean 3 months −3.4 95% CI: −6.6 to −0.1; mean 6 months −3.0 95% CI: −6.4 to 0.3; mean at 12 months −4.1 95% CI: −7.7 to −0.4; effect size Cohen d at 3 months 0.26 and 12 months 0.32) • No difference in headache outcomes between collaborative care group and standard of care group (net difference from baseline between groups mean 3 months 0.2 95% CI: −1.9 to 2.3; mean 6 months −0.4 95% CI: −2.6 to 1.8; mean at 12 months −1.6 95% CI: −3.9 to 0.7)
		(14.7)	FamHx of migraine: NR
		n = 200 (62% female)	Note: 25% reported a history of chronic headaches
		**Included 18 years old	Note: 25% reported a history of chronic headaches
Schwarz et al. (2017)⁴⁵ (abstract)	Retrospective cohort	Participants received PTH treatment and were evaluated a median of 5.7 months following mTBI	Migrainous headache: NR	≥2 biofeedback sessions for treatment of PTH. Compared responders and non-responders	Outcome measure: % reporting reduced headache frequency, % reporting reduced headache severity	Poor
		Age range (mean): 8–18 years	Hx of migraine: NR		Outcome: • 46% of participants responded to therapy, with 35% reporting reduced headache frequency and 23% reporting reduced severity • Responders were significantly more likely to have stayed in school (χ ²: 5.52, p = 0.02) • Responders were significantly less likely to be on serotonin reuptake inhibitors or tricyclic anti-depressants (χ ²: 3.86, p = 0.05)
		(15.5)	FamHx of migraine: NR
		n = NR (77% female)	Note: 66% had daily headaches
		**Included 18 years old	Note: 66% had daily headaches

^aData quality measures were based on the National Institutes of Health (NIH) Quality Assessment Tools. Details can be found at https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools.

Authors and Disclosures

Carlyn Patterson Gentile MD, PhD^1,2, Ryan Shah BS³, Samantha L. Irwin MSc, MB, BCh, BAO, FRCPC⁴, Kaitlin Greene MD⁵ and Christina L. Szperka MD, MSCE, FAHS^1,2

¹Pediatric Headache Program, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
²Department of Neurology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, Pennsylvania, USA
³School of Arts and Sciences, University of Richmond, Richmond, Virginia, USA
⁴UCSF Child & Adolescent Headache Program, San Francisco, California, USA
⁵Doernbecher Children's Hospital Child and Adolescent Headache Program, Division of Pediatric Neurology, Oregon Health & Science University, Portland, Oregon, USA

Correspondence
Carlyn Patterson Gentile, Pediatric Headache Program, Children's Hospital of Philadelphia, 3501 Civic Center Boulevard, Philadelphia, PA 19104, USA. Email: pattersonc@chop.edu

Conflict of Interest
CPG: Dr. Patterson Gentile has received grant support through the American Academy of Neurology Clinical Research Training Scholarship and the International Headache Academy Research Award. RS: Nothing to disclose. SLI: Dr. Irwin receives honoraria for authoring a chapter for the Canadian Pharmacy Association (CPhA) and from NeuroDiem, compensation for scientific consulting (Impel NeuroPharma, Inc.; Biohaven Pharmaceuticals; and Lundbeck A/S), and for research support from the Duke Clinical Research Institute. KG: Dr. Greene has received grant support from the International Headache Academy through the American Headache Society. She is site PI for a clinical trial of Qudexy XR sponsored by USB for which her institution receives financial support. CLS: Dr. Szperka has received research/grant support from the National Institutes of Health/National Institute of Neurological Disorders and Stroke (K23 NS102521). Dr. Szperka or her institution has received compensation for her consulting work for Allergan/AbbVie Inc; Impel NeuroPharma, Inc.; Eli Lilly; Lundbeck; Teva Pharmaceutical Industries Ltd; and Upsher-Smith Laboratories, LLC.

Author Contributions
CPG was responsible for study conception, and design, designed and performed literature searches and organized data, was involved in abstract review, and article review and data input into RedCAP®, and wrote the manuscript. RH was involved in abstract review, article review and data input into RedCAP®, and reviewed the manuscript. SLI and KG provided critical feedback to study design, were involved in abstract review, and article review, and provided text and critical revisions to the manuscript. CLS was involved in study conception, study design, article review, and provided critical revisions to the manuscript. All authors read and approved the final manuscript.