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


Headache. 2021;61(10):1475-1492. 

In This Article


Of the 3458 studies screened, 27 unique studies met criteria for final inclusion with a pooled sample size of 5036 participants (Figure 1). Seventeen of the studies (63%) included subjects that were 18 years old, with four of those (15%) including subjects older than 18. Therapeutic approaches were broadly characterized into abortive pharmacologic treatments, preventative pharmacologic treatments, procedures, neuromodulation, PT and exercise, and collaborative care and behavioral therapy (Table 2).

Figure 1.

Flow chart of article identification and selection

Abortive Pharmacologic Therapies

We identified six peer-reviewed articles and three conference abstracts that evaluated the use of abortive pharmacologic therapies for the treatment of PTH. This included two prospective RCTs. Abortive treatments were typically administered within the first 2 weeks following mTBI. Both oral and intravenous (IV) therapies were studied. Studies were generally limited by small sample size, lack of comparator group, and retrospective design.

Oral Abortive Therapies. The best evidence comes from a prospective randomized open-label study assessing scheduled acetaminophen, ibuprofen, or both compared to standard of care started within the first week of mTBI to treat PTH. Alternating ibuprofen and acetaminophen resulted in the fewest number of headache days and highest return-to-school rate, followed by ibuprofen alone, then acetaminophen alone.[17] Only median and range values were reported; no formal statistical significance testing was done.

In contrast, a large retrospective cohort study found no difference in the rate of persistent PTH at 3 months in those who used nonsteroidal anti-inflammatory drugs (NSAIDs) and/or triptans in the first 30 days compared to no treatment.[18] In fact, those who used triptans developed persistent PTH at higher rates.[18,19] A potential confounder is that subjects who used triptans may have had more significant headache and/or greater incidence of premorbid migraine, but this was not reported (NR). Additionally, medication doses and frequency were NR, and there was no metric for PTH severity at 3 months. Another study mentioned that triptans provided the most rapid relief for PTH, though more specific data were not included.[20]

A case series reported two subjects who had marked improvement in PTH following a 6-day oral methylprednisolone taper within the first month of sustaining an mTBI.[21] However, one subject required a more prolonged steroid taper due to headache recurrence, and many other co-agents were prescribed.

The potential risks of overusing abortive headache therapies following mTBI were highlighted by a retrospective analysis of adolescents presenting to a headache clinic with persistent PTH. Seventy percent of participants in this cohort met criteria for medication overuse headache (MOH); among those, 68.5% had improvement in headache when over-the-counter analgesics and NSAIDs were discontinued.[22] Notably, many of the subjects in this study were simultaneously receiving additional headache preventative pharmacologic treatments that may have impacted outcomes.

Intravenous Abortive Therapies. One small, RCT provided fair evidence that 3% hypertonic saline (HTS) was significantly more effective at decreasing headache severity than normal saline (NS) in youth with acute PTH immediately and 2–3 days after treatment.[23] This study benefited from a randomized design and a standardized pain scale, but it did not specify the acuity of the mTBI, the follow-up period was short, and the study was stopped early due to challenges with recruitment.

A subanalysis of a prospective cohort compared a single dose of IV metoclopramide given in the emergency department (ED) within 48 h of mTBI to no treatment. They found that IV metoclopramide did not reduce the risk of PTH at 1 and 4 weeks.[24] This study was unique because it looked at the impact ofabortive treatment on long-term headache outcome. However, dosing was NR, treatment was not randomized, and only a very small percentage (3%) were given metoclopramide; it is therefore possible that those with more severe headache at onset were treated.

A retrospective analysis of children ≤14 days of their mTBI who received IV fluids with either IV antiemetics (prochlorperazine or metoclopramide), IV ketorolac, or both found that most participants (80%–93%) experienced a >50% reduction in headache severity immediately following administration.[25] This study was limited by its retrospective design, unreported medication dosages, and lack of long-term outcomes.

Similarly, a retrospective chart review of an outpatient infusion center found that a single infusion of IV ketorolac, prochlorperazine, diphenhydramine, and NS improved headache in 85% of participants.[26] However, this study was limited by retrospective design, undefined mean mTBI duration and follow-up period, small sample size, and unreported dosages.

In summary, oral and IV abortive therapies administered within the first days following mTBI may offer immediate relief for PTH. Fair evidence supports alternating ibuprofen and acetaminophen in the first week,[17] as well as greater efficacy of HTS than NS in alleviating acute PTH.[23] Two studies provided poor evidence for acute benefit of IV medications commonly used to treat migraine.[25,26] In contrast, studies that looked at long-term (>1 week) effects did not find a benefit of NSAIDs, triptans,[18,19] or IV metoclopramide[24] for PTH. It is unclear whether abortive therapies can change the long-term trajectory of headaches following mTBI and whether MOH impacts likelihood of persistent PTH.[22]

Preventative pharmacologic therapies

Five studies have reported on the effectiveness of preventative migraine medications in children with PTH.[27] Quality of evidence was rated as poor for all studies.

One cohort study evaluated children with PTH who were prescribed different preventative medications for headache for a mean duration of 5.5 weeks. Full treatment response was defined as ≥50% reduction in headache frequency with functional improvement, which was met by 64% of the cohort across all medications; 45% reported headache resolution. Full response occurred in 13/18 prescribed amitriptyline, 1/11 prescribed nortriptyline, 5/8 prescribed flunarizine, 0/6 prescribed topiramate, 9/12 prescribed melatonin, and 2/2 prescribed indomethacin (both with stabbing headache). Limitations of this study include presence of active co-interventions and lack of clarity regarding timing of medication initiation and timing of assessment.

A second retrospective cohort study reported on children with persistent PTH treated with a preventative medication. Positive response to treatment was defined as ≥50% reduction in headache frequency, which was achieved by 10/20 patients overall. Positive response was reported by 4/11 of those prescribed amitriptyline, 3/6 prescribed propranolol, and 3/3 prescribed topiramate. Limitations of this study include unspecified time of initiation of medication and follow-up timing, as well as lack of specification of medication dosing.

Bramley et al. reported on the efficacy of amitriptyline in a subset of adolescents evaluated in a concussion clinic.[28] Participants with PTH ≥ 1 month were started on amitriptyline for headache prevention. Headache improvement was reported by 56/68 patients (82%) at a median of 4.4 months. Side effects were reported by 16/68 (23%). Females were more likely than males to report PTH (90% vs. 79%; p = 0.004) and were more likely to be prescribed amitriptyline (24% vs. 13%; p = 0.004).

Mackie and Kirkham[29] reported on PTH treatment in children presenting to a neurology clinic; 71% of patients had a migrainous phenotype. All patients received school accommodations, and 16 were treated with topiramate. Eighty-four percent of participants in the cohort overall reported headache improvement; of those treated with topiramate, 12/16 reported reduction in frequency and severity of migrainous headache. Limitations of this study include unspecified time of initiation of medication and timing of follow-up. Additionally, there were limited comparisons between those treated with topiramate compared with those treated with headache accommodations only.

One case series reported on six children who were diagnosed with pseudotumor cerebri following mTBI.[30,31] Initiation of acetazolamide resulted in resolution of headache in 5/6 patients; the one case that did not respond was found to have craniosynostosis. In three cases, acetazolamide was used in combination with topiramate when topiramate alone was not effective. Patients were initially treated with other medications including amitriptyline, propranolol, and cyproheptadine for presumed PTH with no or incomplete response. The study was limited because not all patients met diagnostic criteria for pseudotumor cererbri,[32] and it was difficult to ascertain whether pseudotumor cerebri was secondary to PTH given range of time to diagnosis.

In summary, data on preventative medication efficacy in PTH are limited by poor quality of evidence with reliance on retrospective data collection. The most evaluated medication was amitriptyline, with positive responses reported in 36%–82% of patients across three studies. Topiramate use was also described in three studies, but small numbers of patients resulted in variable efficacy outcomes, with response reported as 0%, 75%, and 100%. Notably, amitriptyline and topiramate were found to be ineffective as monotherapy in patients subsequently found to have intracranial hypertension following mTBI, highlighting the importance of identifying headache phenotype or underlying headache syndrome when treating PTH. While several studies reported on the presence of migrainous headache prior to or following mTBI, none evaluated its impact on treatment outcomes.


One retrospective study showed that transcranial direct current stimulation reduced headache frequency and severity in children with PTH. Over half reported headache resolution.[33] This study benefited from clearly defined metrics of headache improvement but was limited by the lack of comparison group, small sample size, and inconsistent number of treatment sessions used.


Occipital Nerve Block. Three retrospective case series reported PTH improvement following occipital nerve blocks. All were limited by small sample size, concomitant therapies, variable number of nerve blocks administered, and lack of comparator group. One study found that nerve blocks were effective at treating youth with acute or persistent PTH, many of whom reported migrainous headache and a personal and/or family history of migraine. In this cohort, 71% of occipital nerve blocks resulted in headache resolution immediately, and 93% of subjects reported a good therapeutic effect defined as >24 h pain reduction or repeat nerve block request.[34] A smaller retrospective study also found benefit of greater occipital nerve blocks to treat adolescents with PTH; in this cohort, 6/15 patients had occipital neuralgia and 8/15 had occipital tenderness on exam. Nerve blocks were effective in reducing headache frequency with 9/15 participants reporting a ≥50% reduction,[35] although time to follow-up was not clear. Finally, a case series described three patients with posttraumatic occipital neuralgia who reported complete but temporary benefit (days) of occipital nerve blocks.[36]

Acupuncture. One case series reported on three youth with PTH who reported a reduction in headache severity following multiple acupuncture sessions.[37] The study is limited by the small case series design and concomitant use of other therapies for PTH in all three subjects.

In summary, there is consistent evidence that occipital nerve blocks may be helpful for treating PTH in youth, but evidence is limited by small sample sizes and lack of comparator groups. Very limited data suggest acupuncture may also be helpful.

Physical Therapy and Exercise

We identified six studies (three prospective,[38–40] three retrospective[41–43]) that evaluated PT and timing of increased activity on PTH in youth. Some reported headache improvement, whereas others found modest or no effect.

Physical Therapy. A small prospective RCT of youth with postconcussion symptoms compared cervical spine and vestibular physical rehabilitation to standard of care. Both groups worked with a physiotherapist for 8 weeks or until medical clearance. Eleven of fifteen participants in the intervention group were medically cleared to return to sport versus 1/14 in the control group. None of the medically cleared patients reported persistent headache, although there was no significant difference in the reduction of headache symptom scores between those who were cleared or not cleared in either group.[38] Conclusions for this study are limited by small sample size, and most comparisons that were made were between those who were cleared or not cleared for sports, not intervention versus standard of care. Retrospective analysis of a large prospective cohort provided fair evidence that a 4-week active rehabilitation intervention improved headache and postconcussion symptom scores for youth with persistent symptoms at 4 weeks post-injury.[41] A small retrospective cohort study looking at the impact of PT, cardiovascular exercise, vestibular and oculomotor exercise, cervicothoracic manual therapy, and sports-specific training on postconcussion symptoms persisting beyond 3 weeks found a modest decrease in headache severity following intervention over 7 days to up to 6 months, though significance testing was not done. They did find a significant reduction in postconcussion symptoms overall (mean total symptom score pre-intervention 18.2 vs. post-intervention 9.1 p < 0.01).[42]

Exercise. A pilot, prospective RCT compared those who participated in a subsymptom threshold exercise program to a control group of youth who had sustained a concussion and were at least 4 weeks post-injury. Following intervention, both groups reported a higher percentage with mild or no headache. There was a significant effect in the intervention group when total postconcussion symptoms scores were evaluated, but significance testing comparing headache outcomes in the two groups was NR.[39] A prospective study found that completing diagnostic exertional testing did not significantly reduce headache symptom scores 24 h after testing. This study was limited by a high drop-out rate (31%).[40] One retrospective study looked at the impact of early physical activity (<3 weeks following mTBI) on the presence of PTH at the initial follow-up visit and found that those who did not report early physical activity were significantly more likely to report headache.[43] Interpretation of this study is limited by the possibility that those who did not report early physical activity may have been experiencing more significant and more debilitating postconcussion symptoms at onset.

In summary, there is poor to fair evidence to support physical and exercise therapy as a treatment for PTH, although not all studies reported a benefit. Notably, most studies were designed to look at postconcussion symptoms overall and not headache specifically.

Collaborative Care, Cognitive Behavioral Therapy, and Biofeedback

A large prospective RCT reported that collaborative care including cognitive behavioral therapy (CBT), care management, and medication consultation as needed significantly improved patient satisfaction and postconcussion symptoms compared to standard of care. However, there was no significant improvement in PTH,[44] and details about medication used and for what purpose were not described.

Another retrospective analysis evaluated biofeedback as a treatment for PTH in youth. Forty-six percent of the cohort reported an improvement in PTH, with 35% reporting reduced frequency and 23% reporting reduced severity. Participants who reported a decrease in headache severity or frequency following biofeedback were less likely to be on antidepressant medications and more likely to have stayed in school compared to those who did not respond.[45] Sample size was NR and use of concomitant therapies limits interpretation.

In summary, collaborative care including CBT was shown to improve overall quality of life and postconcussion symptoms but did not specifically improve headache outcomes.[44] There is limited support for biofeedback improving PTH.

Headache Risk Factors and Characteristics

Of 27 studies, 7 reported headache with migrainous features,[6,17,21,29,34,35,46] 7 reported on premorbid migraine history,[6,22,24,25,35,39,42] and 3 reported on migraine family history.[6,24,34] When reported, the frequency of headache was high, with 61%–87% of patients described as having daily or constant headache.[6,22,24,35,45] When considered as a confounding variable, presence of prior headache history appeared to impact treatment outcomes,[24,41] although this was not consistent across studies.[25]

Four retrospective studies (all poor data quality) reported targeted treatments to headache phenotype. One study evaluating the efficacy of topiramate as a preventative specifically commented on reduction of "migrainous" headaches.[29] Two patients with stabbing headache were effectively treated with indomethacin.[47] Case series that reported on posttraumatic occipital neuralgia[36] and posttraumatic pseudotumor cerebri[30,31] reported high response rates to targeted treatment.

Timing of Therapies

Most abortive pharmacologic therapies were studied between 48 h and 14 days after mTBI, and their efficacy was measured immediately to days after treatment. All other treatment modalities were studied in youth with at least 3–4 weeks of persistent symptoms. In some cases, symptoms were persistent for >1 year. The duration of preventative pharmacologic treatment, when reported, ranged from 5.5 weeks to 4 months on average.