Acute Concussion Symptom Severity and DelayedSymptom Resolution

WHAT’S KNOWN ON THIS SUBJECT: Children are often evaluatedin the emergency department after a concussion. Althoughprolonged symptoms are associated with higher initial symptomseverity when measured 2 to 3 weeks after injury, a similarassociation with acute symptom severity has not beendemonstrated.

WHAT THIS STUDY ADDS: Higher acute symptom severity is notassociated with development of persistent post-concussionsymptoms 1 month after injury, but persistent post-concussivesymptoms affect a significant number of children afterconcussion. Outpatient follow-up is essential to identify childrenwho develop persistent symptoms.

abstractBACKGROUND AND OBJECTIVES: Up to 30% of children who have concus-sion initially evaluated in the emergency department (ED) display delayedsymptom resolution (DSR). Greater initial symptom severity may be aneasily quantifiable predictor of DSR. We hypothesized that greater symp-tom severity immediately after injury increases the risk for DSR.

METHODS: We conducted a prospective longitudinal cohort study of chil-dren 8 to 18 years old presenting to the ED with concussion. Acute symp-tom severity was assessed using a graded symptom inventory. Presence ofDSR was assessed 1 month later. Graded symptom inventory scores weretested for association with DSR by sensitivity analysis. We conducted a sim-ilar analysis for post-concussion syndrome (PCS) as defined by theInternational Statistical Classification of Diseases and Related HealthProblems, 10th revision. Potential symptoms characteristic of DSR wereexplored by using hierarchical cluster analysis.

RESULTS: We enrolled 234 subjects; 179 (76%) completed follow-up. Thirty-eight subjects (21%) experienced DSR. Initial symptom severity was notsignificantly associated with DSR 1 month after concussion. A total of 22subjects (12%) had PCS. Scores .10 (possible range, 0–28) wereassociated with an increased risk for PCS (RR, 3.1; 95% confidenceinterval 1.2–8.0). Three of 6 of the most characteristic symptoms of DSRwere also most characteristic of early symptom resolution. However,cognitive symptoms were more characteristic of subjects reporting DSR.

CONCLUSIONS: Greater symptom severity measured at ED presentationdoes not predict DSR but is associated with PCS. Risk stratification there-fore depends on how the persistent symptoms are defined. Cognitivesymptoms may warrant particular attention in future study. Follow-up isrecommended for all patients after ED evaluation of concussion tomonitor for DSR. Pediatrics 2014;134:54–62

AUTHORS: Joseph A. Grubenhoff, MD,a,b Sara J. Deakyne,MPH,c Lina Brou, MPH,a,b Lalit Bajaj, MD, MPH,a,b R. DawnComstock, PhD,a,d and Michael W. Kirkwood, PhDe

Departments of aPediatrics, and ePhysical Medicine andRehabilitation, University of Colorado, Aurora, Colorado;bEmergency Department, and cDepartment of ResearchInformatics, Children’s Hospital Colorado, Aurora, Colorado;and dDepartment of Epidemiology, Colorado School of PublicHealth, Aurora, Colorado

KEY WORDSbrain concussion, brain injury, acute, brain injury, traumatic,post-concussion symptoms, post-concussion syndrome,emergency medicine

ABBREVIATIONSAUC—area under the curveCI—confidence intervalDSR—delayed symptom resolutionED—emergency departmentGCS—Glasgow Coma ScaleIQR—interquartile rangeLOC—loss of consciousnessICD-10—International Statistical Classification of Diseases andRelated Health Problems, 10th revisionOR—odds ratioPCS—post-concussion syndromeRR—relative risk

Dr Grubenhoff conceptualized and designed the study, designedthe database, oversaw data collection and analysis, drafted theinitial manuscript, and reviewed and revised the manuscript; MsDeakyne performed the primary statistical analysis, assistedwith study design and database design, managed studypersonnel in recruitment and data acquisition, and co-authored,reviewed, and revised the manuscript; Ms Brou assisted withthe cluster analysis and drafting and reviewed and revised themanuscript; Dr Bajaj assisted with study design and studypersonnel management, oversaw data analysis, and criticallyreviewed and revised the manuscript; Dr Comstock assistedwith data analysis and interpretation and critically reviewed andrevised the manuscript; Dr Kirkwood conceptualized anddesigned the study, assisted with data instrument design, andco-authored, reviewed, and revised the manuscript; and allauthors approved the final manuscript as submitted.

www.pediatrics.org/cgi/doi/10.1542/peds.2013-2988

doi:10.1542/peds.2013-2988

Accepted for publication Apr 10, 2014

Address correspondence to Joseph A. Grubenhoff, MD, 13123 East16th Ave, B-251, Aurora, CO 80045. E-mail: joe.grubenhoff@childrenscolorado.org

(Continued on last page)

54 GRUBENHOFF et al

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

There are 630 000 emergency depart-ment (ED) visits annually for mild trau-matic brain injury (concussion) amongchildren ages 0 to 19 years.1 The ma-jority of those who have concussionexperience symptom resolution in a fewweeks.2 However, a notable minorityexperience persistent post-concussivesequelae.

When examining post-concussive se-quelaewithstandardizedperformance-based cognitive and behavioral tests,most prospective studies indicate that by2 to 3 months post-injury, deficits are nolonger apparent.2–6 Fewer studies havesystematically examined outcomes usingpost-concussive symptom reports fromchildren. However, available researchsuggests that some pediatric patientsdisplay more persistent symptoms thanmight be expected if examining perfor-mance-based test results alone.7,8 Theacute injury risk factors predictive ofdelayed symptom resolution (DSR) inchildren are poorly understood.

Traditionally, determining the severity ofconcussion was predicated on the pre-sence of certain signs and symptoms atthe time of injury, most notably loss ofconsciousness (LOC).9,10 However, LOCoccurs relatively infrequently after con-cussion and is no longer used to defineinjury severity as it is not consistentlyassociated with neuropsychological de-ficits or DSR.11,12 In contrast, researchin adults who have concussion hasfound that post-traumatic amnesia aswell as higher overall symptom levels(ie, both number and severity of symp-toms) are associated with DSR.13–17 Inpediatric patients, greater symptomlevels present a few weeks after injuryare associated with a longer duration ofpost-concussive symptoms.18 Greatersymptom levels have also been associ-ated with objective signs of alteredmental status (eg, post-traumatic am-nesia) in pediatric ED patients immedi-ately after injury.19 Taken together, thesefindings suggest that the number and

severity of acute concussion symptomsmay be a useful indicator of overall se-verity, and therefore may constitute aneasily measurable risk factor to predictDSR. Recent research among youthathletes who have sports-related con-cussion supports this concept.20

DSR is a defining feature of post-concussive syndrome (PCS). However, itis important to highlight that there is nouniversally accepted definition of PCS.Indeed, whether the nonspecific symp-toms typically attributed to this conditionconstitute a syndrome with a commonpathophysiological explanation is contro-versial.21,22 Nonetheless, DSR affects chil-dren who have concussion, and evidencesuggests that the risk for experiencingpersistent symptoms is modifiable.23–25

Identifying children at increased risk forDSR at the time of injury would allowselective implementation of interventionsearlier in the recovery phase.

The primary objective of this study wastodeterminewhethergreatersymptomseverity measured immediately afterinjury is associated with DSR. We hy-pothesized that higher scores ona graded symptom inventory immedi-ately after injury would be associatedwith theDSRat 1month inapediatric EDcohort presenting for acute evaluationof concussion. Given the lack of a uni-versally accepted definition of PCS, wealso evaluated the performance ofa graded symptom inventory for iden-tifying the risk for meeting clinicalcriteria for PCS laid out in the In-ternational Statistical Classification ofDiseases and Related Health Problems,10th revision (ICD-10).26 Although ICD-10 has not yet been adopted in theUnited States for coding purposes, theclinical criteria for PCS have been in-vestigated in concussion research.14,27

METHODS

Study Design

WeconductedaprospectivecohortstudyfromOctober 1, 2010 toMarch31, 2013 of

a convenience sample of children ages8 to 18 years who sustained con-cussions no .6 hours before present-ing to Children’s Hospital Colorado’strauma center ED, which has ∼65 000annual visits. Patients identified on theED electronic track board presentingwith complaints of head injury or symp-toms associated with concussion werescreened for enrollment 16 hours perday, 7 days per week by professionalresearch assistants who enrolled sub-jects and administered all study proce-dures. Subjects were contacted bytelephone 30 days after injury to com-plete follow-up procedures. Subjectswere considered lost to follow-up if theyfailed to respond after 3 attempts. Thestudy was approved by the ColoradoMultiple Institutional Review Board.

Subjects

Children were considered to haveconcussion if they had a Glasgow ComaScale (GCS) score of 13 or 14 or at least2 of the following symptoms occurr-ing after a direct blow to or rapidacceleration/deceleration of the head:bystander-witnessed LOC; post-traumaticamnesia; disorientation to person,place, or time; subjective feelings ofslowedthinking;perseveration;vomiting/nausea; headache; diplopia/blurry vision;dizziness; or somnolence. This clinicaldefinition of concussion has been usedelsewhere.28,29 Children who had openhead injuries, intoxication with alcoholor controlled substances, receipt ofnarcotics for pain control, injuries re-sulting from child abuse, multisysteminjuries, or underlying central nervoussystem abnormalities were excluded.

Measurements

At the ED enrollment visit, the followingdemographic and injury character-istics were obtained: mechanism ofinjury; parental report of previousconcussion; GCS score as determinedby the treating provider; and presenceof abnormalities on head CT scan as

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 55

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

reported by a board-certified pediatricneuroradiologist if obtained.

Self-reported concussion symptomswere quantified by using a gradedconcussion symptom inventory. Thesymptom inventory included the 12items from the Concussion SymptomInventory30 plus 2 additional items re-garding feeling irritable and sad. Sub-jects verbally rated to what degree theywere experiencing 14 symptoms com-mon to concussion. We modified theinstrument for our pediatric popula-tion from a 0 to 6 point scale to a 0 to 2point scale to ensure understanding(range, 0–28). Parents rated theirchild’s symptom severity in the weekbefore injury using the same in-strument to provide a pre-injury base-line for these nonspecific symptoms.The primary outcome, DSR, was de-fined as the presence of 3 or moresymptoms 1 month after injury thatwere absent or less severe in the weekbefore injury reflective of findings insimilar cohorts.18

ICD-10 criteria for PCS require thepresence of 3 ormore of the following 8symptoms 1 month after injury: head-ache, dizziness, fatigue, irritability, dif-ficulty in concentration or performingmental tasks, impairment of memory,insomnia, and reduced tolerance tostress, emotional excitement, or alco-hol.31 The criteria do not adjust forsymptoms present before injury as ourclinical definition did. Patients wereconsidered to have PCS if they reported3 ormore symptoms on the ConcussionSymptom Inventory that aligned withPCS diagnostic criteria. There is nocorollary for “reduced tolerance tostress, emotional excitement, or alco-hol” on the Concussion Symptom In-ventory, so that symptom was excludedduring analysis.

Statistical Analysis

Based on a previous study examiningsymptom severity in ED patients who

have concussion, we planned to dividethe cohort into low and high acutesymptom groups based on a definedcut-off score on a graded symptom in-ventory.32 We hypothesized that DSRwould be more prevalent in the highsymptom group. Therefore, we esti-mated that a sample size of 202 subjectswould be necessary to demonstratea 15% absolute difference in prevalenceof DSR between the low and highsymptom group, using 90% power and2-tailed a of 0.05.

Pre-injury scores for individual symp-toms were subtracted from both theinitial and 30-day follow-up scores toaccount for the presence of these non-specific symptoms before injury. Symp-toms present before injury but absentpost-injury were scored as 0 (ie, a neg-ative score was not assigned). De-scriptive statistics for demographic andacute injury data were calculated asproportions or medians with inter-quartile ranges and compared by usinga x2 test and Wilcoxon rank sum asappropriate. Because there is no de-fined point separating low from highsymptoms, a sensitivity analysis wasperformed by using the x2 statistic todetermine the best cut-point for initialsymptom severity scores to divide thelow and high groups using DSR as theoutcome. After sensitivity analysis, thebest cut-point score was used for mul-tiple logistic regression, adjusting forgender and age, as symptom reportmay vary by both age and gender.32–34

The same methods were usedsubstituting our clinical definition ofDSR with the PCS criteria. Results of x2

analysis were considered significant if P, .05. Relative risk and odds ratios (OR)were considered significant if the 95%confidence interval (CI) did not include1. Lastly, we conducted hierarchicalclustering with average linkage analysisto determine if certain symptoms weremore characteristic of delayed versusearly symptom resolution.35 Analyses

were conducted by using SAS 9.3 (SASInstitute, Inc, Cary, NC) and hierarchicalclustering was conducted by usingSPSS 22.0 (IBM SPSS Statistics, IBMCorporation, Chicago, IL).

RESULTS

Research assistants screened 1253patients for participation; 273 met in-clusion criteria and 234 subjects con-sented to participate in the study. Ofthose enrolled, 179 subjects (76%)completed the 30-day follow-up call andcomprised the study cohort (Fig 1).Subjects who did not complete follow-up were similar to those who did inage, gender, initial GCS, mechanism ofinjury, and history of previous concus-sion. Subjects lost to follow-up hadsignificantly lower initial graded symp-tom inventory scores (median score, 7;interquartile range [IQR], 4–12) com-pared with subjects completing thestudy (median score, 10; IQR, 7–13;P = .01).

Thirty-eight children (21%) from thestudycohortmet thestudydefinition forDSR. The pre-injury baseline symptomscore differed significantly in bothgroups,but thescoreswere lowforbothgroups (Table 1). Two subjects in theearly symptom resolution (ESR) groupand 1 in the DSR group experienceda subsequent concussion in the follow-up period. Forty subjects underwenthead CT scan with only 5 abnormalfindings, all found in the ESR group(Table 2).

The results of sensitivity analysis ofinitial symptom inventory scores rang-ing from 8 to 14 are shown in Table 3.The receiver-operator characteristiccurve is shown in Fig 2 (area under thecurve [AUC], 0.508; 95% CI, 0.475–0.683;P = .14, rounded). The best cut pointwas a score of 11 with a sensitivity of63% and a specificity of 50% for DSR.However, a score of 11 was not signif-icantly associated with DSR (P = .46). Inmultivariate analysis, adjusting for age

56 GRUBENHOFF et al

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

and gender, a score of 11 was still notassociated with DSR (OR, 1.4; 95% CI,0.7–2.8).

Twenty-two subjects (12%) met criteriafor PCS. Sensitivity analysis for initialgraded symptomscores ranging from8to 14 is shown in Table 4. The receiver-operator characteristic curve is shown

in Fig 2 (AUC, 0.629; 95% CI, 0.509–0.748;P = .03). The best cut point was 10 witha sensitivity of 77% and specificity of51% for PCS (P = .02). The relative riskfor PCS in subjects who had an initialsymptom score .10 was 3.1 (95% CI,1.2–8.0). In multivariate analysis, ad-justing for age and gender, the OR for

PCS in the high symptom versus lowsymptom group was 3.7 (95% CI, 1.3–10.6).

Figures 3 and 4 show the results ofcluster analysis. Three of the 6 mostcharacteristic initial symptoms in theDSR and ESR groups (those in whichthe relative linkage distance on thex-axis are shortest) were similar (pho-nophobia, photophobia, blurred, or dou-ble vision). However, cognitive symptoms(difficulty remembering, difficulty con-centrating, or “feeling foggy”) weremorecharacteristic of the DSR group.

DISCUSSION

Our study of children 8 to 18 years oldpresenting to an ED ,6 hours afterconcussion demonstrated that initialsymptom severity is not associated withDSR. This is an important finding givenevolving knowledge of concussion symp-tom resolution. In 1988, Lishman pro-posed that symptoms appearing shortlyafter a concussion were primarily theresult of physiologic derangements di-rectly related to the injury, whereasprotracted symptoms were more likelyrelated to latent psychological factors.36

Twodecades later, accumulated researchsuggests that “physiogenic” and “psy-chogenic” factors contribute to the con-stellation of symptoms present bothimmediately after injury as well asthroughout recovery.37

Intuitively, it is reasonable to assumethat more severe acute physiologic in-jury will manifest as more severesymptomatology and likely requirea longer recovery period. Recent re-search supports this assumption. Aprospective cohort study of youth ath-letes evaluated in sports concussionclinics in the first 3 weeks after a con-cussion demonstrated that increasinginitial graded symptom inventoryscores were associated with increasedodds of symptom resolution occurringbeyond 28 days.20 Similarly, pediatricED patients who had high symptom

FIGURE 1Study Participant Flow Diagram.

TABLE 1 Demographic and Injury Characteristics for Early and Delayed Symptom ResolutionGroups

ESR (n = 141) DSR (n = 38) P a

DemographicsMean age, years (SD) 12.6 (2.5) 13.4 (2.2) .79Male, % 70 66 .69History of previous concussion, % 24 29 .53

Injury characteristicsMechanism, % — — .79Sport 48 53 —

Fall 43 34 —

Assault 3 5 —

Motor vehicle collision 1 3 —

Other 5 5 —

LOC, % 26 29 0.68Post-traumatic amnesia, % 26 34 0.42Received head CT scan in ED, % 21 26 0.52Abnormal head CT scan results, % 4 0 0.02Initial GCS, medianb 15 15 0.99Preinjury graded symptom score, median (IQR) 1 (0–2) 2 (1–4) 0.002Initial ED graded symptom score, median (IQR) 9 (6–13) 10.5 (7–15) 0.14

a x2 was used to compare proportions and Wilcoxon rank sum to compare medians. x2 analysis compared the overalldifference among mechanism for the ESR and DSR groups and only this single p-value is provided.b IQR for GCS was 15–15 for both groups.

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 57

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

levels measured in the early weeksafter a concussion had significantlyhigher odds of symptoms persistingfor up to 1 year.18 In both studies, initialsymptom inventories were obtained anaverage of 11 days after concussion.

Although both physiologic and psycho-logical factors contribute to manifes-

tation of concussion symptoms, there issome evidence that the acute injuryfactors are stronger determinants ofsymptom reports early in recovery,whereas non-injury factors contributemore to persistent symptoms.38 Astrength of our study is that we en-rolled subjects within 6 hours of injury,

but our results differed from theseprevious reports. This suggests thatacute symptom report alone is not anaccurate reflection of the physiologicand psychological factors that ulti-mately lead to DSR.

We defined DSR in terms relevant toclinical practice. Specifically, we de-veloped a definition that would likelypromptaprimarycareprovider torefera child for specialist evaluation (at least3 symptoms that are worse 1 monthafter injury than they were before in-jury). Although the clinical criteria forPCSalsorequire thepresenceofat least3 symptoms 1 month after concussion,the diagnostic accuracy of this defini-tion is a topic of scientific debate as it isboth subjective and imprecise.39–41

There is also significant controversy asto whether the term “syndrome” isappropriate, given that common con-cussion symptoms are also found inpatients who do not have concussion.42

Regardless of these shortcomings, PCS

TABLE 2 Characteristics of Subjects Who Had Abnormal Head CT Scans

Age (y) Gender GCS LOC (+/2) Injuries

10.8 F 15 – Cerebral contusion10.0 M 14 – Subarachnoid and epidural hematoma, skull fracture14.0 M 15 + Subdural hematoma9.4 M 14 – Subarachnoid hematoma, skull fracture10.1 M 15 + Subarachnoid hematoma

TABLE 3 Sensitivity Analysis of Graded Symptom Inventory Scores for Identifying DelayedSymptom Resolution

Cut Point Sensitivity, % Specificity, % NPV, % PPV, % P

8 73 33 79 26 .669 71 33 81 22 .4510 71 35 82 23 .1411 63 50 84 26 .4612 53 55 81 24 .1313 47 66 82 27 .1914 37 74 81 27 .10

FIGURE 2ROC curves displaying sensitivity analysis of concussion symptom inventory scores for identifying delayed symptom resolution or post-concussive syndrome.Optimal scores (closest to upper left of graph) for each outcome are shown along with the AUC and associated P value for each curve.

58 GRUBENHOFF et al

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

has been studied as an outcome mea-sure in studies of persistent symp-toms.27,43,44 Therefore, we repeated ouranalysis using PCS as the outcomerather than our clinical definition ofDSR. We noted 2 important findings.First, we found a 43% relative decreasein prevalence in the outcome (9% ab-solute difference) when applying thisalternate definition. Second, we showedthat a graded symptom inventory score

.10 was associated with a threefoldincreased risk for PCS in our cohort,whereas there was no association withour clinical definition.

One may conclude from these findingsthat the smaller subset of symptomsthat meet criteria for PCS are morerepresentative of a specific clinical entitycharacterized by persistent symptomsthan those found on broader symptominventories. An analysis comparing 3

different criteria for PCS showed thata subset of 6 symptoms common to all 3criteria was specific to PCS owing toconcussion among adults.44 In contrast,other work calls into question whetherthe symptoms included in various di-agnostic criteria are specific to con-cussion. Our results resemble those ofMcCauley, Boake, and colleagues, whofound wide variations in the prevalenceof PCS, depending on the criteriaemployed as well as a lack of specificity,because many adult patients who didnot have head trauma also met PCScriteria.45,46 Additionally, although thePCS ROC curve showed a statisticallysignificant association (P = .034) be-tween an initial symptom severity scoreof 10 and PCS, the absolute AUC of 0.629suggests no more than a modest re-lationship. Therefore, concluding that

TABLE 4 Sensitivity Analysis of Graded Symptom Inventory Scores for Identifying ICD-10 Post-Concussion Syndrome

Cut Point Sensitivity, % Specificity, % NPV, % PPV, % P

8 77 33 91 14 .469 77 36 92 14 .3410 77 51 94 18 .0211 64 55 92 17 .1112 55 66 91 18 .1013 41 73 90 18 .2114 36 80 90 21 .10

FIGURE 3Hierarchical cluster analysis for participants who had early symptom resolution. Shorter relative cluster linkage distances on the x-axis indicate symptoms(y-axis) that are more characteristic of the group, whereas longer distances indicate symptoms that are less characteristic.

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 59

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

PCS criteria are more representative ofa unique clinical syndrome is difficult tojustify and suggests that accurate riskstratification is heavily dependent onhow the outcome is defined.

Hierarchical cluster analysis is an ex-ploratory method that aims to dem-onstrate which features are mostcharacteristic of a group. Three of the 6most characteristic symptoms for boththe DSR and ESR groupswere identical inour cohort, suggesting that these symp-toms (phonophobia, photophobia, blur-red or double vision)may not be useful inidentifying those at risk for DSR. However,we found it interesting that cognitivesymptoms were more characteristic ofsubjects who had DSR. Although someauthorshavefoundastrongerassociationbetween the number of initial symptomsand DSR,17,47 others have shown that

specific symptoms are more closelyassociated.14,48 The exploratory na-ture of cluster analysis preventsdrawing firm conclusions regardingthe ability of these symptoms to pre-dict DSR. However, cognitive symp-toms may warrant particular scrutinywhen present.

We experienced a lost-to-follow-up rateof 24%. The subjects lost to follow-uphad a significantly lower median ini-tial symptom score than the final studycohort. It is plausible that most of thesesubjects would have fallen into the ESRgroup but did not complete the studyowing to resolution of symptoms.49 Iftrue, the absence of these patientsfrom analysis would tend to bias ourresults toward the null hypothesis.

Our follow-up period was limited to 30days, as we were interested in identi-

fying children who, at the time of theirconcussion, are at risk for DSR. We didnot perform serial assessments todetermine the precise day of symptomresolution or the range of symptomduration. It is possible that some sub-jects in the DSR group had resolution ofsymptoms shortly after their 30-dayfollow-up call and were misclassified.However, the prevalence of DSR in thisstudy was 21% compared with studiesin other US pediatric ED cohorts 3months after concussion, which rangedfrom 15% to 29%, so this limitation isunlikely to have hada significant impacton our results.50,51

Finally, we did not include a controlgroup who had injuries to body regionsother than the head. Although childrenwho have concussion tend to reportmore post-concussive symptoms than

FIGURE 4Hierarchical cluster analysis for participantswho had delayed symptom resolution. Shorter relative cluster linkage distances on the x-axis indicate symptoms(y-axis) that are more characteristic of the group, whereas longer distances indicate symptoms that are less characteristic.

60 GRUBENHOFF et al

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

childrenwith orthopedic injuries, thereis considerable overlap in symptom re-port, highlighting the non-specific na-ture of post-concussive symptoms.29 Weare therefore unable to evaluate whatproportion of symptoms is simply at-tributable to injury in general and whatproportion is attributable to concus-sion specifically. Our objective, how-ever, was to determine whether acutesymptom severity could be used forrisk-stratification among head-injuredchildren rather than as a diagnostictool. Nonetheless, the issue of symptom

specificity further emphasizes the needfor more accurate definitions related tosequelae after concussion.

CONCLUSIONS

Greater symptom severity at the time ofinjury does not predict DSR amongchildren presenting to the ED for eval-uation of concussion, but it is a riskfactor for meeting criteria for PCS asdefined by ICD-10. These findings un-derscore the need to refine the defini-tionofpost-concussivesyndrometoonethat is truly representative of concussion

sequelae and that accounts for thecontribution of both physiologic andpsychological processes. Given the in-ability to predict the resolution of post-concussive symptoms at the time ofinjury, outpatient follow-up and serialsymptom assessment should be a cor-nerstone of concussion managementfor all children after ED discharge.

ACKNOWLEDGMENTSWe thank Kendra Kocher, BS, who pro-vided study coordination and databasemanagement.

REFERENCES

1. Faul MXL, Wald MM, Coronado VG. Trau-matic Brain Injury in the United States:Emergency Department Visits. Hospital-izations and Deaths 2002–2006. Atlanta, GA:Centers for Disease Control and Pre-vention, National Center for Injury Pre-vention and Control; 2010

2. Carroll LJ, Cassidy JD, Peloso PM, et al;WHO Collaborating Centre Task Force onMild Traumatic Brain Injury. Prognosis formild traumatic brain injury: results of theWHO Collaborating Centre Task Force onMild Traumatic Brain Injury. J Rehabil Med.2004; (43, suppl)84–105

3. Satz P, Zaucha K, McCleary C, Light R,Asarnow R, Becker D. Mild head injury inchildren and adolescents: a review ofstudies (1970-1995). Psychol Bull. 1997;122(2):107–131

4. Satz P. Mild head injury in children andadolescents. Curr Dir Psychol Sci. 2001;10:106–109

5. Babikian T, Asarnow R. Neurocognitiveoutcomes and recovery after pediatric TBI:meta-analytic review of the literature.Neuropsychology. 2009;23(3):283–296

6. Belanger HG, Vanderploeg RD. The neuro-psychological impact of sports-relatedconcussion: a meta-analysis. J Int Neuro-psychol Soc. 2005;11(4):345–357

7. Barlow KM, Crawford S, Stevenson A,Sandhu SS, Belanger F, Dewey D. Epide-miology of postconcussion syndrome inpediatric mild traumatic brain injury. Pe-diatrics. 2010;126(2). Available at: www.pediatrics.org/cgi/content/full/126/2/e374

8. Yeates KO. Mild traumatic brain injury andpostconcussive symptoms in children andadolescents. J Int Neuropsychol Soc. 2010;16(6):953–960

9. Leclerc S, Lassonde M, Delaney JS, LacroixVJ, Johnston KM. Recommendations forgrading of concussion in athletes. SportsMed. 2001;31(8):629–636

10. Kelly JP, Rosenberg JH. The development ofguidelines for the management of concus-sion in sports. J Head Trauma Rehabil.1998;13(2):53–65

11. Lovell MR, Iverson GL, Collins MW, McKeagD, Maroon JC. Does loss of consciousnesspredict neuropsychological decrements af-ter concussion? Clin J Sport Med. 1999;9(4):193–198

12. Erlanger D, Kaushik T, Cantu R, et al. Symptom-based assessment of the severity of a con-cussion. J Neurosurg. 2003;98(3):477–484

13. King NS, Crawford S, Wenden FJ, CaldwellFE, Wade DT. Early prediction of persistingpost-concussion symptoms following mildand moderate head injuries. Br J ClinPsychol. 1999;38(pt 1):15–25

14. Yang CC, Hua MS, Tu YK, Huang SJ. Earlyclinical characteristics of patients withpersistent post-concussion symptoms:a prospective study. Brain Inj. 2009;23(4):299–306

15. Sheedy J, Geffen G, Donnelly J, Faux S.Emergency department assessment of mildtraumatic brain injury and prediction ofpost-concussion symptoms at one monthpost injury. J Clin Exp Neuropsychol. 2006;28(5):755–772

16. Drake AI, McDonald EC, Magnus NE, Gray N,Gottshall K. Utility of Glasgow Coma Scale-Extended in symptom prediction followingmild traumatic brain injury. Brain Inj. 2006;20(5):469–475

17. Lundin A, de Boussard C, Edman G, Borg J.Symptoms and disability until 3 monthsafter mild TBI. Brain Inj. 2006;20(8):799–806

18. Yeates KO, Taylor HG, Rusin J, et al. Longi-tudinal trajectories of postconcussivesymptoms in children with mild traumaticbrain injuries and their relationship toacute clinical status. Pediatrics. 2009;123(3):735–743

19. Grubenhoff JA, Kirkwood MW, Deakyne S,Wathen J. Detailed concussion symptomanalysis in a paediatric ED population.Brain Inj. 2011;25(10):943–949

20. Meehan WP III, Mannix RC, Stracciolini A,Elbin RJ, Collins MW. Symptom severitypredicts prolonged recovery after sport-related concussion, but age and amnesiado not. J Pediatr. 2013;163(3):721–725

21. Iverson GL. Misdiagnosis of the persistentpostconcussion syndrome in patients withdepression. Arch Clin Neuropsychol. 2006;21(4):303–310

22. Garden N, Sullivan KA. An examination ofthe base rates of post-concussion symp-toms: the influence of demographics anddepression. Appl Neuropsychol. 2010;17(1):1–7

23. Ponsford J, Willmott C, Rothwell A, et al.Impact of early intervention on outcomefollowing mild head injury in adults.J Neurol Neurosurg Psychiatry. 2002;73(3):330–332

24. Wade DT, King NS, Wenden FJ, Crawford S,Caldwell FE. Routine follow up after headinjury: a second randomised controlledtrial. J Neurol Neurosurg Psychiatry. 1998;65(2):177–183

25. Mittenberg W, Tremont G, Zielinski RE,Fichera S, Rayls KR. Cognitive-behavioralprevention of postconcussion syndrome.Arch Clin Neuropsychol. 1996;11(2):139–145

26. Boake C, McCauley SR, Levin HS, et al.Limited agreement between criteria-based

ARTICLE

PEDIATRICS Volume 134, Number 1, July 2014 61

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

diagnoses of postconcussional syndrome. JNeuropsychiatry Clin Neurosci. 2004;16(4):493–499

27. Iverson GL, Lange RT. Examination of “post-concussion-like” symptoms in a healthysample. Appl Neuropsychol. 2003;10(3):137–144

28. Taylor HG, Dietrich A, Nuss K, et al. Post-concussive symptoms in children withmild traumatic brain injury. Neuropsychol-ogy. 2010;24(2):148–159

29. Moran LM, Taylor HG, Rusin J, et al. Dopostconcussive symptoms discriminate in-jury severity in pediatric mild traumaticbrain injury? J Head Trauma Rehabil. 2011;26(5):348–354

30. Randolph C, Millis S, Barr WB, et al. Concus-sion symptom inventory: an empirically de-rived scale for monitoring resolution ofsymptoms following sport-related concussion.Arch Clin Neuropsychol. 2009;24(3):219–229

31. World Health Organization. The ICD-10Classification of Mental and BehavioralDisorders: Clinical Description and Di-agnostic Guidelines. Geneva: World HealthOrganization; 1992

32. Grubenhoff JA, Kirkwood M, Gao D, DeakyneS, Wathen J. Evaluation of the standardizedassessment of concussion in a pediatricemergency department. Pediatrics. 2010;126(4):688–695

33. Frommer LJ, Gurka KK, Cross KM, IngersollCD, Comstock RD, Saliba SA. Sex differencesin concussion symptoms of high schoolathletes. J Athl Train. 2011;46(1):76–84

34. Berz K, Divine J, Foss KB, Heyl R, Ford KR,Myer GD. Sex-specific differences in the se-verity of symptoms and recovery rate fol-lowing sports-related concussion in youngathletes. Phys Sportsmed. 2013;41(2):58–63

35. Beckstead JW. Using hierarchical clusteranalysis in nursing research. West J NursRes. 2002;24(3):307–319

36. Lishman WA. Physiogenesis and psycho-genesis in the ’post-concussional syn-drome’. Br J Psychol. 1988;153(10):460–469

37. Silverberg ND, Iverson GL. Etiology of thepost-concussion syndrome: physiogenesisand psychogenesis revisited. Neuro-Rehabilitation. 2011;29(4):317–329

38. McNally KA, Bangert B, Dietrich A, et al.Injury versus noninjury factors as pre-dictors of postconcussive symptoms fol-lowing mild traumatic brain injury inchildren. Neuropsychology. 2013;27(1):1–12

39. Frances A. The new crisis in confidence inpsychiatric diagnosis. Ann Intern Med.2013;159(3):221–222

40. Kendell R, Jablensky A. Distinguishing be-tween the validity and utility of psychiatricdiagnoses. Am J Psychiatry. 2003;160(1):4–12

41. Carroll CP, Cochran JA, Guse CE, Wang MC.Are we underestimating the burden oftraumatic brain injury? Surveillance of se-vere traumatic brain injury using Centersfor Disease Control International Classifi-cation of Disease, 9th revision, ClinicalModification, Traumatic Brain Injury Codes.Neurosurg. 2012;71(6):1064–1070; discus-sion 1070

42. Meares S, Shores EA, Taylor AJ, et al. Mildtraumatic brain injury does not predictacute postconcussion syndrome. J NeurolNeurosurg Psychiatry. 2008;79(3):300–306

43. Yang CC, Tu YK, Hua MS, Huang SJ. Theassociation between the postconcussionsymptoms and clinical outcomes forpatients with mild traumatic brain injury. JTrauma. 2007;62(3):657–663

44. Laborey M, Masson F, Ribéreau-Gayon R,Zongo D, Salmi LR, Lagarde E. Specificity ofpostconcussion symptoms at 3 months af-ter mild traumatic brain injury: resultsfrom a comparative cohort study. J HeadTrauma Rehabil. 2014;29(1):E28–E36

45. McCauley SR, Boake C, Pedroza C, et al.Postconcussional disorder: Are the DSM-IVcriteria an improvement over the ICD-10? JNerv Ment Dis. 2005;193(8):540–550

46. Boake C, McCauley SR, Levin HS, et al. Di-agnostic criteria for postconcussionalsyndrome after mild to moderate trau-matic brain injury. J Neuropsychiatry ClinNeurosci. 2005;17(3):350–356

47. Chrisman SP, Rivara FP, Schiff MA, Zhou C,Comstock RD. Risk factors for concussivesymptoms 1 week or longer in high schoolathletes. Brain Inj. 2013;27(1):1–9

48. De Kruijk JR, Leffers P, Menheere PP,Meerhoff S, Rutten J, Twijnstra A. Predictionof post-traumatic complaints after mildtraumatic brain injury: early symptoms andbiochemical markers. J Neurol NeurosurgPsychiatry. 2002;73(6):727–732

49. Blinman TA, Houseknecht E, Snyder C, WiebeDJ, Nance ML. Postconcussive symptoms inhospitalized pediatric patients after mildtraumatic brain injury. J Pediatr Surg. 2009;44(6):1223–1228

50. Eisenberg MA, Andrea J, Meehan W, MannixR. Time interval between concussions andsymptom duration. Pediatrics. 2013;132(1):8–17

51. Babcock L, Byczkowski T, Wade SL, Ho M,Mookerjee S, Bazarian JJ. Predicting post-concussion syndrome after mild traumaticbrain injury in children and adolescentswho present to the emergency department.JAMA Pediatr. 2013;167(2):156–161

(Continued from first page)

PEDIATRICS (ISSN Numbers: Print, 0031-4005; Online, 1098-4275).

Copyright © 2014 by the American Academy of Pediatrics

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

FUNDING: Funding for the conduct of this study was provided by a Thrasher Research Fund Early Career Award to Dr Grubenhoff. Use of REDCap Database wassupported by NIH/NCATS Colorado CTSI grant UL1 TR000154. Contents are the authors’ sole responsibility and do not necessarily represent official NIH or ThrasherResearch Fund views.

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

62 GRUBENHOFF et al

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

DOI: 10.1542/peds.2013-2988 originally published online June 23, 2014; 2014;134;54Pediatrics 

and Michael W. KirkwoodJoseph A. Grubenhoff, Sara J. Deakyne, Lina Brou, Lalit Bajaj, R. Dawn Comstock

Acute Concussion Symptom Severity and Delayed Symptom Resolution

ServicesUpdated Information &

http://pediatrics.aappublications.org/content/134/1/54including high resolution figures, can be found at:

Referenceshttp://pediatrics.aappublications.org/content/134/1/54.full#ref-list-1This article cites 48 articles, 8 of which you can access for free at:

Subspecialty Collections

brain_injury_subhttp://classic.pediatrics.aappublications.org/cgi/collection/traumatic_Traumatic Brain Injury_subhttp://classic.pediatrics.aappublications.org/cgi/collection/concussionConcussiondicine:physical_fitness_subhttp://classic.pediatrics.aappublications.org/cgi/collection/sports_meSports Medicine/Physical Fitnessfollowing collection(s): This article, along with others on similar topics, appears in the

Permissions & Licensing

https://shop.aap.org/licensing-permissions/in its entirety can be found online at: Information about reproducing this article in parts (figures, tables) or

Reprintshttp://classic.pediatrics.aappublications.org/content/reprintsInformation about ordering reprints can be found online:

ISSN: . 60007. Copyright © 2014 by the American Academy of Pediatrics. All rights reserved. Print American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since . Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from

DOI: 10.1542/peds.2013-2988 originally published online June 23, 2014; 2014;134;54Pediatrics 

and Michael W. KirkwoodJoseph A. Grubenhoff, Sara J. Deakyne, Lina Brou, Lalit Bajaj, R. Dawn Comstock

Acute Concussion Symptom Severity and Delayed Symptom Resolution

http://pediatrics.aappublications.org/content/134/1/54located on the World Wide Web at:

The online version of this article, along with updated information and services, is

ISSN: . 60007. Copyright © 2014 by the American Academy of Pediatrics. All rights reserved. Print American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois,has been published continuously since . Pediatrics is owned, published, and trademarked by the Pediatrics is the official journal of the American Academy of Pediatrics. A monthly publication, it

by guest on May 20, 2018http://pediatrics.aappublications.org/Downloaded from