Article, Pediatrics

No association between metoclopramide treatment in ED and reduced risk of post-concussion headache

a b s t r a c t

Objective: There is a lack of definitive pediatric literature on effective pharmacotherapy for persistent post- concussion headache symptoms. This study assessed whether acute metoclopramide treatment in the Emer- gency Department (ED) was associated with a reduction in persistent headache in children at 1- and 4-weeks post-concussion.

Methods: Children aged 8-17 years with acute concussion presenting to 9-Canadian pediatric EDs were enrolled in a prospective cohort study, from August 2013-June 2015. Primary and secondary outcomes were persistent headache at 1- and 4-week post-injury respectively. Headache persistence was based on the one and four- week headache scores minus recalled pre-Injury score using the Post-Concussion Symptom Inventory. The asso- ciation between metoclopramide and headache persistence at 1- and 4-weeks were examined using unadjusted and adjusted regression and 1:4 propensity score matching model.

Results: Baseline assessments were completed in 2095 participants; 65 (3.1%) received metoclopramide within 48-hours of injury. At 1- and 4-weeks, 54% (963/1808) and 26% (456/1780) of participants had persistent head- ache relative to baseline respectively. In unadjusted analysis, no association between metoclopramide and head- ache persistence at 1-week was found [treated vs. untreated: 1-week (53% vs. 53%; Relative risk = 1.0 (95% CI: 0.8, 1.3); 4-weeks (27.3% vs. 25.6%; RR = 1.0 (95% CI: 0.9, 1.2)]. Metoclopramide was not associated with lower headache risk on propensity score matching [treated vs. untreated: 1-week, n = 220 (52% vs. 59.4%; RR

= 0.8 (95%CI: 0.6, 1.2) and 4-weeks, n = 225 (27.1% vs. 32.8%; RR = 0.9 (95%CI: 0.8, 1.1)].

Conclusion: Metoclopramide administration was not associated with a reduction in headache persistence in

Abbreviations: PPCS, persistent post-concussion symptom; ED, emergency department; PCSI, post-concussion symptom inventory; 5P, predicting and preventing post-concussive problems in pediatrics study; RCT, randomized clinical trial; PERC, pediatric emergency research Canada.

* Corresponding author at: Department of Pediatrics and Emergency Medecine, Children’s Hospital of Eastern Ontario, 401 Smyth Road, Ottawa K1H 8L1, Ontario, Canada.

E-mail addresses: [email protected], (N. Bresee), [email protected], (M. Aglipay), [email protected], (A.S. Dubrovsky), [email protected], (A.-A. Ledoux), [email protected], (F. Momoli), [email protected], (S.B. Freedman), [email protected], (K. Barlow), [email protected], (L. Richer), [email protected], (N.J. Barrowman), [email protected]. (R. Zemek).

1 Children’s Hospital of Eastern Ontario Research Institute, 401 Smyth Road, Ottawa, Ontario, K1H 8L1.

2 The Montreal Children’s Hospital, McGill University Health Centre, 1001 Boulevard Decarie, Montreal, QC H4A 3J1.

3 Hospital Ste. Justine, 3175 Chemin de la Cote-Sainte-Catherine, Montreal, QC H3T 1C4.

4 Heritage Medical Research Building, 3330 Hospital Dr. NW, Calgary, AB T2N 4N1.

5 Health Research Innovation Centre, Room 1A10, 3330 Hospital Drive NW Calgary, Alberta, Canada, T2N 4N1.

6 Women and Children’s Health Research Institute, University of Alberta, 4-081 Edmonton Clinic Health Academy (ECHA), 11405 87 Avenue NW Edmonton, AB T6G 1C9.

https://doi.org/10.1016/j.ajem.2018.04.007

0735-6757/(C) 2018

children seeking ED care due to a concussion. Further research is necessary to determine which pharmacother- apies may be effective for acute and persistent post-concussive headache.

(C) 2018

Introduction

Pediatric concussions are a common emergency department (ED) complaint with approximately 750,000 visits annually in the USA [1-3] with the annual incidence of concussion in Canada being 0.29% in a re- cent Canadian Community Health Survey [4].

Despite the frequency of such injuries, there is limited literature de- scribing effective pharmacotherapies for treating the signs and symp- toms associated with concussion in the acute or convalescent phase of the injury [2,5,6]. The most commonly reported concussion symptom is headache, occurring in as many as 85% of those with concussion [2,7-10]. Persistent post-Concussion symptoms (PPCS) occur in approx- imately 30% of children and adolescents experiencing a concussion [11]. PPCS is defined by the presence of persistent somatic, cognitive, psycho- logical and/or behavioural changes for 1-month duration or longer fol- lowing a concussion [12]. The burden of PPCS is vast and can impair social, emotional, cognitive and physical functioning [13].

Despite four systematic reviews on concussion treatment, and sev- eral consensus statements, there is little evidence guiding pharmacotherapeutic management of acute post concussive symptoms [5,14-22]. Additionally, there is a paucity of data examining the effects of medications to mitigate post-concussive headache [23,24]. Theoreti- cally, early treatment of headache pain in individuals suffering from a concussion may be associated with improved outcomes if persistent post-concussion headaches adhere to the pain amplification theories described in the ‘wind-up’ or ‘kindling’ hypotheses [25-27]. The kindling hypothesis supports the use of early aggressive abortive therapy for mi- graine, and other chronic pain syndromes. The kindling hypothesis pro- vides a theoretical explanation for the neuroplastic changes that occur in supraspinal structures and the corticolimibic sensitization that may occur in Persistent symptoms of pain. With post-concussive headache, it remains unclear why certain individuals go on to have persistent symptoms of headache and how early therapy may benefit these indi- viduals. It is our hypothesis that persistent post concussive headache may follow other pain amplification syndromes leading to chronic per- sistent symptoms in certain individuals.

Early administration of Metoclopramide may be beneficial at reducing persistent headache symptoms following a concussion. Metoclopramide is a dopamine and serotonin receptor antagonist that treats both symp- toms of nausea and headache, and is the most common medication given for the treatment of headache in pediatric emergency departments across Canada [28]. In a meta-analysis comparing metoclopramide to pla- cebo it was found to be superior for treatment of migraine headache [29]. It is also recommended by the Canadian Headache Society for the treat- ment of acute headache (non migraine) in the emergency department setting for its favorable side effect profile, and moderate evidence of effi- cacy [30]. The American Headache Society also recommends the use of Metoclopramide, as it has been cited as “highly likely to be effective” [31]. In a recent publication by Friedman et al. [32], Metoclopramide was investigated for post concussive headache relief at three time points; in the ED, at 48 h after discharge and at one week. Of the 19 patients with follow-up data 60% reported sustained headache relief at one week. Al- though this study was limited by a small sample size and lack of control arm, it adds to the literature on pharmacotherapies that may be used to treat post concussive headache and persistent post concussive headache. It is hypothesized that early treatment of headache post-injury could po- tentially mitigate the development of persistent post-concussion head- ache. This sub-study sought to assess whether metoclopramide administration is associated with a reduction in ongoing post- concussive headache at different time points after treatment.

Methods

Study design

This was a planned sub-analysis of the Predicting and Preventing Post-concussive Problems in Paediatrics (5P) study [11,12]. The 5P study was a prospective, Multicenter cohort study. Participants were re- cruited in nine Pediatric Emergency Research Canada (PERC) member hospitals. Enrollment occurred from August 2013 through June 2015. The study complied with the TRIPOD statement [33] and was approved by the ethics committees of participating institutions. Written consent and assent was obtained from all participants and their parents/guard- ians as appropriate. A detailed protocol has been published [12].

Inclusion criteria

Patients eligible for this sub-analysis were between the ages of 8.00 and 17.99 years and met criteria for concussion, defined by the Zurich and Berlin consensus statement [14,22].

Only those with concussion and the presence of headache pain at ED presentation, defined as a positive change in headache score from pre- senting symptoms as compared to pre-injury baseline, were included in this study.

Exclusion criteria

Patients experiencing concussion N48 h prior to enrolment or pa- tients with a Glasgow Coma Scale <=13 were excluded. Patients were ex- cluded if there was any abnormality on neuroimaging, including CT findings (though CT or neuroimaging were not necessary for inclusion). Any neurosurgical or operative intervention, Intensive care unit care or intubation was exclusionary as well as multisystem injuries with treatment requiring admission to hospital, or procedures requiring sedation in the ED. Those with chronic neurological delays with resul- tant communication difficulties, evidence of intoxication, and the ab- sence of precipitating trauma were excluded.

Recruitment and follow-up

Patients were identified and recruited by a research assistant in one of the 9 PERC academic pediatric emergency centers. Patients were managed per the treating physician. Physicians completed a brief survey regarding the patient’s medical history, comorbidities, therapies re- ceived in the department and discharge instructions along with prog- nostication regarding symptom duration after discharge. Medications administered during the patients stay in the ED (ibuprofen, acetamino- phen, morphine, ketorolac, IV fluids, odensatron and diphenhydramine) were at the treating physician’s discretion (including metoclopramide). Follow-up was conducted either electronically through a secure link to a web-based questionnaire or on telephone call (as per patient/family preference) at 1- and 4-weeks post-enrolment. Follow-up question- naires used the reliable and validated Post-Concussion Symptom Inven- tory (PCSI) [34-36]. Symptom severities were measured using symptom scales taken from the PCSI self-report forms that are specific for the de- velopment and age of the child/youth; ages 8-12 (17 items, 3-point scale), 13-18 years (20 items, 7-point scale). The PCSI was collected at ED presentation and 1- and 4-weeks post-injury. The PCSI incorporates for each time point a scale measurement of the patient’s recollection of

their baseline symptom prior to the initial concussion.

Exposure

Participants who presented with a headache, defined as a positive change in headache on the PCSI at the initial presentation and were ad- ministered metoclopramide in the ED at presentation were considered the treated group. Participants who presented to the ED with headache but were not administered metoclopramide were considered the un- treated group.

Outcomes and outcome measures

The primary outcome was persistence of headache at 1 week. Pres- ence of headache, a dichotomous variable, was based on the 1-week

Not Eligible (n = 3520) Eligible but cannot complete follow-up (n = 72)

Screened (n = 8046)

PCSI headache score minus the recalled pre-injury score. In other words, headache symptoms at 1 week that were not present prior to the concussion (and/or headaches rated worse than perceived pre- injury headache ratings) were considered ongoing headache. The sec- ondary outcome was the presence of persistent headache symptoms at 4 weeks. Presence of headache, was based on the 4-week PCSI head- ache score minus the recalled pre-injury score.

Statistical analyses

Frequencies and descriptive statistics were used to summarize pa- tient baseline characteristics for the overall sample and by study group. The proportion of participants with headache persistence in

Did not consent (n = 1172) Declined participation (n = 969)

RA not available to consent family (n = 167) Other reason for non- consent (n = 36)

Logistic regression model (n = 1779)

Did not receive metoclopramide (n = 1722)

Received metoclopramide (n = 57)

Completed follow-up at 1 week (n = 1808)

Lost to follow-up (n = 287)

Baseline (n = 2095) [Headache at baseline & >= age 8 years]

Consented but not seen by RA (n = 73)

Withdrew (n = 133) Ineligible after physician assessment (n = 13)

No headache at baseline or age <8 years (n = 968)

Consented (n = 3282)

Eligible Patients (n = 4454)

Matched on propensity score (n = 170)

Did not receive Maxeran

(n = 1750)

Matched on propensity score (n = 50)

Received Maxeran (n = 58)

Fig. 1. Flow diagram of study participants.

each group was computed with a Wilson score 95% CI [37]. The associ- ation of metoclopramide with 1 and 4-week persistence was estimated by 3 separate models: 1) unadjusted analysis; 2) adjusted multivariable

Table 1

Baseline characteristics of the unadjusted regression analysis sample and propensity- matched participants.

logistic regression; and 3) 1:4 propensity score matching modelling.

Selection of pretreatment characteristics variables

Unadjusted regression analysis sample

Metoclopramide given in the ED

Propensity-matched participants

Metoclopramide given in the ED

sion, hospital site, early signs of concussion (this includes appearing dazed and confused, confused about events, answers questions slowly, repeats questions, forgetful of recent information), sensitivity to light and noise [8,11,38,39]. Lastly, medication given for pain (i.e. acetamino- phen, ibuprofen) and nausea (i.e. dimenhydrinate and ondansetron) prior and during the ED visit were adjusted for in the analyses, as

Hours between ED visit and injury (hours) median (IQR) Nausea (%)

Previous number of concussionsa

The following pretreatment characteristics were identified by our

concussion panel as a-priori confounders; that is, they were likely indi- cators of later concussion symptoms or recovery and likely influenced

the decision to prescribe metoclopramide in the ED: age, sex, prior his-

Characteristic

Agea

8 to 12

No

N = 1750

913 (52.2)

Yes

N= 58

21 (36.2)

No

N = 170

69 (40.6)

Yes

N= 50

17 (34.0)

tory of migraine headache, Family history of MIgraine headache, head-

13 to 18

837 (47.8)

37 (63.8)

101 (59.4)

33 (66.0)

ache severity, duration of prior concussion (no prior concussion/

Female (%)a

711 (40.6)

28 (48.3)

74 (43.5)

24 (48.0)

concussion lasting b1 week, prior concussion lasting >=1 week), depres-

History (%)

3.1

(1.5-16.3)

5.9

(1.7-23.3)

3.0

(1.4-16.3)

3.7

(1.7-22.8)

1065

41(70.7)

101 (59.4)

36 (72.0)

(60.9)

0 1315

(75.5)

these variables may be associated with both treatment choice and 1-

1

278 (16.0)

10 (17.2)

28 (16.5)

8 (16.0)

week persistence. We restricted ourselves to those specific medications

2

94 (5.4)

5 (8.6)

16 (9.4)

4 (8.0)

as other medications were rarely prescribed.

3+

54 (3.1)

5 (8.6)

9 (5.3)

3 (6.0)

38 (65.5) 117 (68.8) 35 (70.0)

Adjusted multivariable logistic regression

A multivariable logistic regression model was used to estimate the

Longest symptom duration of previous concussiona

b1 week 169 (40.0) 6 (30.0) 15 (28.3) 5 (33.3)

1-4 weeks

164 (38.8)

10 (50.0)

24 (45.3)

6 (40.0)

5+ weeks

90 (21.3)

4 (20.0)

14 (26.4)

4 (26.7)

odds ratio and associated 95% CI of headache based on metoclopramide

Migrainea

223 (12.8)

18 (31.6)

35 (20.6)

12 (24.0)

administration, adjusting for pretreatment characteristics.

Migraine familya history

815 (46.7)

35 (60.3)

92 (54.1)

28 (56.0)

Learning disabilities

139 (8.0)

4 (6.9)

15 (8.9)

2 (4.0)

ADD/ADHD

147 (8.4)

8 (13.8)

15 (8.9)

7 (14.0)

2.7.3. Propensity analysis

Propensity score analyses were used to establish balance among the

Other developmental disorder

Anxiety

64 (3.7)

152 (8.7)

0 (0.0)

8 (13.8)

7 (4.2)

21 (12.4)

0 (0.0)

6 (12.0)

large set of potential confounders in the comparison of those treated

Depressiona

53 (3.0)

4 (6.9)

11 (6.5)

3 (6.0)

and untreated group [40-42]. The propensity score is the probability

Loss of consciousness

234 (13.4)

11 (19.0)

38 (22.4)

9 (18.0)

that a participant received metoclopramide in the ED based on that participant’s observed set of pretreatment characteristics. The analysis involves a two-step procedure. First, pretreatment characteristics were included in a logistic regression model, with a dichotomous variable for having been administered metoclopramide as the dependent vari- able. This step provides the propensity to have been treated with metoclopramide for each person in the cohort. The propensity score

Duration (minutes) median

(IQR)

0.25

(0.1-1.0)

0.25

(0.1-0.9)

0.3

(0.20-1.0)

0.50

(0.1-1.0)

Seizure

31 (1.8)

0 (0.0)

3 (1.8)

0 (0.0)

Any early signs of concussiona

748 (42.7)

34 (58.6)

101 (59.4)

30 (60.0)

Mechanism of injury

Sports/recreational play

1284

47 (81.0)

122 (71.8)

40 (80.0)

(73.4)

Non-sport related injury/fall

334 (19.1)

5 (8.6)

33 (19.4)

4 (8.0)

Motor vehicle collision

29 (1.7)

2 (3.4)

6 (3.5)

2 (4.0)

was then used in place of the large set of covariates.

Assault

22 (1.3)

0 (0.0)

3 (1.8)

0 (0.0)

Patients in the treated and the untreated group were matched 1:4 on

Other

80 (4.6)

4 (6.9)

6 (3.5)

4 (8.0)

the logit of the propensity scores with a greedy algorithm and nearest- neighbor approach using the ‘MatchIt’ package in the R statistical soft-

helmet use

Pain management

535 (41.7)

18 (38.3)

47 (38.5)

15 (37.5)

Received prior to ED

ware. This allowed for treated and untreated patients to be matched

Acetaminophena

465 (26.6)

23 (39.7)

50 (29.4)

19 (38.0)

based on closest propensity score, where closest was within a caliper

Ibuprofena

651 (37.2)

29 (50.0)

86 (50.6)

24 (48.0)

distance of 0.1. Four matches were attempted for each treated

Other

81 (4.6)

16 (27.6)

10 (5.9)

13 (26.0)

(metoclopramide) participant at a time in random order. At each step, a patient without ED metoclopramide administration was chosen that had not yet been matched. Balance between groups was assessed in the matched data visually and by testing for differences between treated and untreated participants in pretreatment characteristics using chi- squared and t-tests where appropriate. After obtaining a matched dataset with appropriate covariate balance relative risk were calculated. Effects of metoclopramide after a 4-week follow-up post-ED was exam- ined using the same modelling approaches.

Received prior to ED

Dimenhydrinatea

24 (1.4)

2 (3.4)

2 (1.2)

2 (4.0)

Given within the ED

Dimenhydrinatea

2 (0.1)

0 (0.0)

0 (0.0)

0 (0.0)

Participants who did not complete the 1-week follow-up were con- sidered lost to follow-up and were not included in the analysis. To ex-

Given within the ED

Acetaminophena

265 (15.1)

12 (20.7)

24 (14.1)

10 (20.0)

Ibuprofena

417 (23.8)

11 (19.0)

46 (27.1)

8 (16.0)

Morphine PO

4 (0.2)

0 (0.0)

0 (0.0)

0 (0.0)

Other PO

10 (0.6)

1 (1.7)

0 (0.0)

1 (2.0)

Ketorolac IV

12 (0.7)

18 (31.0)

3 (1.8)

15 (30.0)

Morphine IV

4 (0.2)

0 (0.0)

1 (0.6)

0 (0.0)

IV Fluids

6 (0.3)

10 (17.1)

3 (1.8)

8 (16.0)

Other IV

1 (0.1)

2 (3.4)

1 (0.6)

2 (4.0)

Nausea management

amine the impact of missing outcomes on our data, we conducted

Ondansetrona

70 (4.0)

1 (1.7)

8 (4.7)

1 (2.0)

sensitivity analyses with two scenarios: 1) assuming all participants

missing the primary outcome had headache at one weeks and 2) assum-

Other PO Other IV

9 (0.5)

6 (0.3)

1 (1.7)

4 (6.9)

2 (1.2)

2 (1.2)

1 (2.0)

4 (8.0)

ing all participants missing the primary outcome did not have headache at one weeks.

All analyses were performed using IBM SPSS Statistics 23 and R ver- sion 3.0.2.

a Pretreatment variables identified as priori confounders, PO: oral administration.

Results

Of the 2095 (65 were prescribed metoclopramide) patients that completed baseline data and had headache at inclusion, 1808 com- pleted 1-week follow-up. At one and four weeks respectively, 53% (963/1808) and 26% (456/1780) of participants had persistent head- ache. Fig. 1 provides a flow diagram of the 2095 participants selected from the 5P cohort.

Unadjusted regression sample participant baseline demographics are summarized in Table 1. The mean age of the participants in this study was 13.0 years. Females represented 40.9% (739/1808) of partici- pants, and the majority of participants had no prior concussion [75.2% (1353/1808)]. Thirteen percent of participants (241/1808) had a per- sonal history of migraine headache and 47% (850/1808) had a family history. Nausea was present in over half of those at initial presentation [61.2% (1106/1808)]. A total of 55 patients (that completed 1- and 4- week outcomes) were prescribed metoclopramide in the ED. Other pain (i.e. ibuprofen, acetaminophen) and nausea relief medications (i.e. dimenhydrinate and ondensatron) were given to the treated [n

= 46 (79.3%); n = 3 (5.2%)] and the untreated group [n = 1172

(67%); n = 87 (5%)].

Our 1:4 matched analysis resulted in 50 treated participants matched to 170 untreated participants. Because of the higher proportion of un- treated participants compared to treated ones, there were 7 unmatched treated participants and 1552 unmatched untreated participants. The seven unmatched treated participants were those with the highest pro- pensity for treatment with metoclopramide (propensity score N 0.3). Sup- plemental Fig. 1 demonstrates the distribution of propensity scores of matched and unmatched participants in the metoclopramide-treated and untreated groups. The mean propensity score among matched treated participants was 0.1 (SD = 0.1) and the mean propensity score among matched untreated participants was 0.1 (SD = 0.1).

Of the 1808 patients with post-concussive headache at inclusion, there were 963 (53%) patients that had persistent headache symptoms at 1 week. As seen in Table 2, non-significant effects were found with the unadjusted regression in the treated and untreated groups for the primary outcome at 1 week [unadjusted RR: 1.0; 95% CI: 0.8-1.3], When adjusting for possible confounding variables, the odds ratio of the multivariable logistic regression was 0.6 (95% CI: 0.4-1.1), adjust- ment with propensity score matching using a 1:4 matching algorithm yielded a RR of 0.8 (95% CI: 0.6-1.2).

In secondary outcome analysis, there were 456 (26%) patients that continued to have persistent headache symptoms at 4 weeks. As seen in Table 3, in the unadjusted analyses there was no significant difference in the treated and the untreated groups [unadjusted RR: 1.0; 95% CI: 0.9-1.2]. When adjusting for possible confounding variables, the odds ratio of the multivariable logistic regression was 0.8 (95%CI: 0.4, 1.5), ad- justment with propensity score matching using a 1:4 matching algorithm yielded a RR of 0.9 (95%CI: 0.8, 1.1). Finally, no significant differences were found for the two sensitivity analyses. (Appendix Tables S1 and S2).

Discussion

In this study a single dose of IV Metoclopramide administration for concussion associated headaches within the first 48 h of injury did not

appear to offer protective effects on the presence of post-concussion headache at either 1- or 4-weeks. These findings should be considered in light of our limited number of metoclopramide exposed individuals despite this study being the largest pediatric concussion study to date. Our low range of propensity score in this population potentially reflects a reluctance to initiate pharmacotherapy with post concussive head- ache. Despite the small treatment effect, these findings are important and cannot be disregarded, as there is a paucity of data currently for the use of effective pharmacotherapy in concussion and post- traumatic headaches acutely.

While the concussion literature includes multiple studies recommending trigger reduction and graduated Return to play/rest, there is a paucity of evidence to guide effective pharmacotherapy for headache pain reduction in the ED or for interventions to prevent persistent post-concussive headaches. In a recent systematic review by Gravel et al., multiple randomized control trials (RCT) were reviewed where any intervention was provided within the first week of a mild traumatic brain injury and assessed for its effective- ness on the main outcome of somatic symptoms (headache, dizzi- ness, fatigue sleep problems) [5]. Only one of the 17 studies identified, specifically looked at pharmacotherapy. This adult study (n = 17) compared nasal administration of desmopressin (DDAVP) 10 micrograms twice daily for 5 days with placebo. The result indi- cated an improvement in performance on two memory tests on the third day of treatment but no other improvement on other testing for cognition. However, this study did not look specifically at the symptoms of headache [5]. Since the publication of this systematic review, a prospective pediatric RCT by Lumba-Brown et al., exam- ined 10 ml/kg of hypertonic saline compared to normal saline ad- ministered over 1 h on headache in which a significant difference in self-reported headache pain reported after fluid administration and at 2-3 days post-treatment in the hypertonic saline group [43]. However, the long-term post-concussive symptoms were unknown as patients only received telephone follow-up three days after treat- ment. Despite four systematic reviews, there is little evidence guid- ing pharmacoTherapeutic treatments for concussion [14-18,22]. Overall, there is a lack of RCT data to support evidence-based phar- macotherapies for the treatment of headache symptoms of concus- sion acutely and the potential effects on PPCS [5,19-21]. To our knowledge our study has been the first of its kind to look at medium and long-term symptoms of headache in pediatric patients after treatment given in the ED. As such we sought to test the hypothesis that early treatment may reduce persistent headache pain.

In a recent publication by Friedman et al. [32], metoclopramide was

investigated for post concussive headache relief. In this exploratory study of 21 patients, they found of the 19 patients with follow-up data at one week, 60% reported sustained headache relief. To our knowledge, this is one of the first studies, in the adult population, to specifically look at pharmacotherapy for treatment of post-concussion headaches. Friedman’s study differs in that they do not specifically examine pediat- ric patients and their therapeutic success is defined as having a reduc- tion of headache severity (not necessarily absence). This study was limited by a small sample size and lack of control arm; however, adds to the literature on possible pharmacotherapies for post concussive headache pain in the ED.

Table 2

Comparison of the estimated treatment effect of metoclopramide on headache at 1 week.

Type of analysis No. (Absolute risk %)

Treated group Untreated group Estimate (95%CI)

Unadjusted model 58 (53) 1750 (53) RRa = 1.0 (0.8, 1.3)

Logistic regression model 57 (-) 1722 (-) ORb = 0.6 (0.4, 1.1)

Matched on propensity score 1:4 Matchit 50 (52) 170 (59.4) RRa = 0.8 (0.6, 1.2)

a RR = relative risk.

b OR = odds ratio.

Table 3

Results from different modelling strategies for the association of metoclopramide treatment and headache at 4 weeks.

No. (Absolute risk, %)

Treated group

Untreated group

Estimate (95%CI)

Unadjusted model Logistic regression model

Matched on propensity score 1:4

55 (27.3)

54 (-)

48 (27.1)

1725 (25.6)

1703 (-)

177 (32.8)

RRa=1.0 (0.9, 1.2)

ORb=0.8 (0.4, 1.5)

RRa=0.9 (0.8, 1.1)

a RR = relative risk.

b OR = odds ratio.

Despite the benefits of metoclopramide for use in treatment of acute headache relief in the emergency department, we did not find a differ- ence with its use for prevention of persistent post concussive headache. This may be a result of the limitations of our study or a lack of efficacy for metoclopramide use for long term headache relief. To our knowledge metoclopramide’s efficacy has not previously been studied in the pre- vention of Headache recurrence, nor is it well known to be effective in post- traumatic headache other than the recent study by Friedman et al. There are several limitations of this study. Although propensity score matching is methodologically an attempt to compare patients with sim- ilar baseline characteristics, this was a secondary data analysis and not a randomized control trial. Second, immediate post-treatment headache scores were not obtained in the ED. This may have affected the results when recalling pain around the time of the initial concussion. If tempo- rary headache relief was provided in the ED with metoclopramide, this could have affected subsequent reports of pain in the weeks that followed. We did attempt to adjust for recall bias by having patients rate their current presence of headache compared to their pre-injury headache in attempts to limit this. Third, we were not able to control for use of over-the-counter medications at home, nor further treatment with these after discharge from the emergency department. However, on phone follow-up no patient had received further physician pre- scribed treatment for headache as an outpatient. Fourth, we did not look at headache severity but the presence of any headache pain greater than their preinjury status; therefore, cannot comment on reduction of severity. In addition, we could not account for any potential synergist medication effects if family members administered ibuprofen or acet- aminophen. Finally, the metoclopramide treatment effect estimated by our model refers to the average treatment effect among the treated. In other words, our results refer only to a patient population with a distri- bution of pretreatment characteristics (those measured at baseline)

similar to those ultimately administered metoclopramide.

Conclusion

Among children and adolescents aged 8 to 18 years with acute con- cussion, a one time IV dose of metoclopramide administration was not associated with a reduction in the risk of ongoing headache symptoms at 1- and 4-weeks. Further research is necessary to determine which pharmacotherapies may be effective in acute post-traumatic headache treatment and prevention of persistent headaches following pediatric concussion.

Supplementary data to this article can be found online at https://doi. org/10.1016/j.ajem.2018.04.007.

Authors disclosure

Funding source

This study was supported by the Canadian Institutes of Health Re- search (CIHR) operating grant (MOP 126197); the CIHR-Ontario Neurotrauma Foundation Mild Traumatic Brain Injury team grant (TM1 127047); and CIHR planning grant (MRP 119829).

Financial disclosure

Dr. Stephen Freedman is supported by the Alberta Children’s Hospi- tal Foundation Professorship in Child Health and Wellness. Dr. Roger Zemek is supported by a University of Ottawa Brain and Mind Research Institute Clinical Research Chair in Pediatric Concussion.

Natalie Bresee wrote the first draft of the manuscript. No honorar- ium, grant or other form of payment was given to anyone to produce the manuscript. All authors have no financial disclosures relevant to this article.

Conflict of interest

No competing financial interests exist.

Acknowledgements

We would like to acknowledge the dedication and efforts of all re- search coordinators and research assistants across the nine sites respon- sible for patient recruitment, enrollment and follow-up. student volunteers at CHEO, ACH, CHUSJ and HSC provided invaluable assistance in patient screening at the Emergency Department. We would also like to acknowledge the Predicting Persistent Post-concussive Problems in Pediatrics (5P) team: Peter Anderson, PhD; Miriam H. Beauchamp, PhD; Darcy Beer, MD; Kathy Boutis, MD; Brian L. Brooks, PhD; Emma Burns, MD; William Craig, MDCM; Carol DeMatteo, MSc; Ken J. Farion, MD; Isabelle Gagnon, PhD; Gerard Gioia, PhD; Blaine Hoshizaki, PhD; Yael Kamil, BSc; Michelle Keightley, PhD; Terry Klassen, MD; Candice McGahern, BA; William P. Meehan III, MD; Willem Meeuwisse, MD, PhD; Angelo Mikrogianakis, MD; Gurinder Sangha, MD; Martin H. Osmond, MDCM; Michael Vassilyadi, MDCM, MSc; Keith Owen Yeates, PhD.

References

  1. National Center for Injury Prevention. Report to congress on mild traumatic brain in- jury in the United States: steps to prevent a serious Public health problem. Atlanta, GA; 2003.
  2. Meehan III WP, Mannix R. Pediatric concussions in United States emergency depart- ments in the years 2002 to 2006. J Pediatr 2010;157(6):889-93.
  3. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of trau- matic brain injury: a brief overview. J Head Trauma Rehabil 2006;21(5):375-8.
  4. Gordon KE, Kuhle S. Validation of “reported concussion” within a national health survey. Brain Inj 2018;32(1):41-8. https://doi.org/10.1080/02699052.2017. 1403046.
  5. Gravel J, D’Angelo A, Carriere B, et al. Interventions provided in the acute phase for mild traumatic brain injury: a systematic review. Syst Rev 2013;2:63.
  6. Emery C, Kang J, Shrier I, et al. Risk of injury associated with bodychecking experi- ence among youth hockey players. Can Med Assoc J 2011;183(11):1249-56.
  7. Seifert TD. Sports concussion and associated post-traumatic headache. Headache 2013;53(5):726-36. https://doi.org/10.1111/head.12087.
  8. Babcock L, Byczkowski T, Wade SL, Ho M, Mookerjee S, Bazarian JJ. Predicting Postconcussion syndrome after mild traumatic brain injury in children and adoles- cents who present to the emergency department. JAMA Pediatr 2013;167(2): 156-61.
  9. Makdissi M, Darby D, Maruff P, Ugoni A, Brukner P, McCrory PR. Natural history of concussion in sport: markers of severity and implications for management. Am J Sports Med 2010;38(3):464-71.
  10. Guskiewicz KM, Weaver NL, Padua DA, Garrett Jr WE. Epidemiology of concussion in collegiate and high school football players. Am J Sports Med 2000;28(5):643-50.
  11. Zemek R, Barrowman N, Freedman SB, et al. Clinical risk score for persistent postconcussion symptoms among children with acute concussion in the ED. JAMA J Am Med Assoc 2016;315(10):1014-25. https://doi.org/10.1001/jama.2016.1203.
  12. Zemek R, Osmond MH, Barrowman N, Team. PERC (PERC) C. Predicting and preventing postconcussive problems in paediatrics (5P) study: protocol for a pro- spective multicentre clinical prediction rule derivation study in children with con- cussion. BMJ Open 2013;3(8).
  13. Yeates KO, Kaizar E, Rusin J, et al. Reliable change in postconcussive symptoms and its Functional consequences among children with mild traumatic brain injury. Arch Pediatr Adolesc Med 2012;166(7):615-22.
  14. McCrory P, Meeuwisse WH, Aubry M, et al. Consensus statement on concussion in sport: the 4th International Conference on Concussion in Sport held in Zurich, No- vember 2012. Br J Sports Med 2013;47(5):250-8.
  15. Marshall S, Bayley M, McCullagh S, Velikonja D, Berrigan L. Clinical practice guide- lines for mild traumatic brain injury and persistent symptoms. Can Fam physician Medecin Fam Can 2012;58(3):257-67 (e128-40).
  16. Harmon KG, Drezner JA, Gammons M, et al. American Medical Society for Sports Medicine position statement: concussion in sport. Br J Sports Med 2013;47(1): 15-26.
  17. Herring SA, Cantu RC, Guskiewicz KM, et al. Concussion (mild traumatic brain in- jury) and the team physician: a consensus statement–2011 update. Med Sci Sports Exerc 2011;43(12):2412-22.
  18. Giza CC, Kutcher JS, Ashwal S, et al. Summary of evidence-based guideline update: evaluation and management of concussion in sports: report of the Guideline Devel- opment Subcommittee of the American Academy of Neurology. Neurology 2013;80 (24):2250-7.
  19. Comper P, Bisschop SM, Carnide N, Tricco A. A systematic review of treatments for mild traumatic brain injury. Brain Inj 2005;19(11):863-80.
  20. King D, Brughelli M, Hume P, Gissane C. Assessment, management and knowledge of sport-related concussion: systematic review. Sports Med 2014;44(4):449-71. https://doi.org/10.1007/s40279-013-0134-x.
  21. Borg J, Holm L, Peloso PM, et al. Non-surgical intervention and cost for mild trau- matic brain injury: results of the WHO Collaborating Centre Task Force on Mild Traumatic Brain Injury. J Rehabil Med 2004(43 Suppl):76-83.
  22. McCrory P, Meeuwisse W, Dvorak J, et al. Consensus statement on concussion in sport–the 5th international conference on concussion in sport held in Berlin, Octo- ber 2016. Br J Sports Med April 2017:bjsports-2017-097699. https://doi.org/10. 1136/bjsports-2017-097699.
  23. Dubrovsky AS, Friedman D, Kocilowicz H. Pediatric post-traumatic headaches and Peripheral nerve blocks of the scalp: a case series and patient Satisfaction survey. Headache J Head Face Pain 2014;54(5):878-87. https://doi.org/10.1111/head. 12334.
  24. Pinchefsky E, Dubrovsky AS, Friedman D, Shevell M. Part II–management of pediat- ric post-traumatic headaches. Pediatr Neurol 2015;52(3):270-80. https://doi.org/ 10.1016/j.pediatrneurol.2014.10.015.
  25. Rome HP, Rome JD. Limbically augmented pain syndrome (LAPS): kindling, corticolimbic sensitization, and the convergence of affective and sensory symptoms in chronic pain disorders. Pain Med 2000;1(1):7-23. https://doi.org/10.1046/j.1526- 4637.2000.99105.x.
  26. Walker WC, Marwitz JH, Wilk AR, et al. Prediction of headache severity (density and functional impact) after traumatic brain injury: a longitudinal multicenter study. Cephalalgia 2013;33(12):998-1008. https://doi.org/10.1177/0333102413482197.
  27. Post RM, Susan R, Weiss B. Sensitization, kindling, and carbamazepine: an update on their implications for the course of affective illness. Pharmacopsychiatry 1992;25

    (1):41-3. https://doi.org/10.1055/s-2007-1014386.

    Richer LP, Laycock K, Millar K, et al. Treatment of children with migraine in emer- gency departments: national practice variation study. Pediatrics 2010;126(1): e150-. https://doi.org/10.1542/peds.2009-2337.

  28. Colman I, Brown MD, Innes GD, Grafstein E, Roberts TE, Rowe BH. Parenteral metoclopramide for acute migraine: meta-analysis of Randomised controlled trials. BMJ 2004;329(7479):1369-73. https://doi.org/10.1136/bmj.38281.595718.7C.
  29. Orr SL, Aube M, Becker WJ, et al. Canadian headache society systematic review and recommendations on the treatment of migraine pain in emergency settings. Cephalalgia 2015;35(3):271-84. https://doi.org/10.1177/0333102414535997.
  30. Orr SL, Friedman BW, Christie S, et al. Management of adults with acute migraine in the emergency department: the American headache society evidence assessment of parenteral pharmacotherapies. Headache 2016;56(6):911-40. https://doi.org/10. 1111/head.12835.
  31. Friedman BW, Babbush K, Irizarry E, White D, John Gallagher E. An exploratory study of IV metoclopramide + diphenhydramine for acute post-traumatic headache. Am J Emerg Med 2018;36(2):285-9.
  32. Bouwmeester W, Zuithoff NPA, Mallett S, et al. Reporting and methods in clinical prediction research: a systematic review. PLoS Med 2012;9(5):1-12.
  33. Gioia GA, Schneider JC, Vaughan CG, Isquith PK. Which symptom assessments and approaches are uniquely appropriate for paediatric concussion? Br J Sports Med 2009;43(Suppl. 1):i13-22. https://doi.org/10.1136/bjsm.2009.058255.
  34. Gioia G, Isquith P, Schneider J, Vaughan C. New approaches to assessment and mon- itoring of concussion in children. Top Lang Disord 2009;29:266-81.
  35. Gioia GA, Janusz JA, Isquith PK, Vincent D. Psychometric properties of the parent and teacher Post-Concussion Symptom Inventory (PCSI) for children and adolescents. J Int Neuropsychol Soc 2008;14(Suppl. 1):204.
  36. Newcombe RG. Two-sided confidence intervals for the single proportion: compari- son of seven methods. Stat Med 1998;17(8):857-72.
  37. Zemek RL, Farion KJ, Sampson M, McGahern C. Prognosticators of persistent symp- toms following pediatric concussion: a systematic review. JAMA Pediatr 2013;167 (3):259-65. https://doi.org/10.1001/2013.jamapediatrics.216.
  38. Kienbacher C, Wober C, Zesch HE, et al. Clinical features, classification and prognosis of migraine and tension-type headache in children and adolescents: a long-term follow-up study. Cephalalgia 2006;26(7):820-30. https://doi.org/10.1111/j.1468- 2982.2006.01108.x.
  39. Kurth T, Walker AM, Glynn RJ, et al. Results of multivariable logistic regression, pro- pensity matching, propensity adjustment, and propensity-based weighting under conditions of nonuniform effect. Am J Epidemiol 2006;163(3):262-70. https://doi. org/10.1093/aje/kwj047.
  40. Rosenbaum PR, Rubin DB. Constructing a control group using multivariate matched sampling methods that incorporate the propensity score. Am Stat 1985;39(1).
  41. Rubin DB, Thomas N. Matching using estimated propensity scores: relating theory to

    practice. Biometrics 1996;52(1):249-64.

    Lumba-Brown A, Harley J, Lucio S, Vaida F, Hilfiker M. Hypertonic saline as a therapy for pediatric concussive pain: a randomized controlled trial of symptom treatment in the emergency department. Pediatr Emerg Care 2014;30(3):139-45. https://doi. org/10.1097/PEC.0000000000000084.