876 | VOLUME 65 | NUMBER 5 | NOVEMBER 2009
Michael McCrea, Ph.D.
Waukesha Memorial Hospital,
Waukesha, Wisconsin, and
Department of Neurology,
Medical College of Wisconsin,
Kevin Guskiewicz, Ph.D., A.T.C.
Departments of Exercise and
Sport Science and Orthopedics, and
Injury Prevention Research Center,
University of North Carolina
at Chapel Hill,
Chapel Hill, North Carolina
Christopher Randolph, Ph.D.
Department of Neurology,
Loyola University Medical School,
William B. Barr, Ph.D.
Departments of Neurology
New York University
School of Medicine,
New York, New York
Thomas A. Hammeke, Ph.D.
Department of Neurology,
Medical College of Wisconsin,
Stephen W. Marshall, Ph.D.
Departments of Orthopedics
and Epidemiology, and
Injury Prevention Research Center,
University of North Carolina
at Chapel Hill,
Chapel Hill, North Carolina;
James P. Kelly, M.D.
Departments of Neurosurgery and
University of Colorado
School of Medicine,
Michael McCrea, Ph.D.,
Waukesha Memorial Hospital,
721 American Avenue, Suite 406,
Waukesha, WI 53188.
Received, October 14, 2008.
Accepted, April 1, 2009.
Copyright © 2009 by the
Congress of Neurological Surgeons
other abnormalities observed in athletes after
sport-related concussion (3, 6, 12, 15, 16, 20). In
the vast majority of cases, a single concussion
seems to be a relatively benign neurological
event, followed by a rapid and complete recov-
everal prospective studies have recently
clarified the expected course of recovery
in symptoms, cognitive dysfunction, and
ery within several days (3, 7, 19, 20, 24, 25). In
some cases, however, recovery and outcome
can be complicated when an athlete experi-
ences multiple concussions over a short time-
frame. The main risks associated with repeat
concussion are thought to be 2-fold: 1) slowed
clinical recovery characterized by persistent
symptoms or functional impairments, and 2)
extremely rare instances of death or severe,
permanent disability associated with delayed
cerebral swelling (i.e., “second impact syn-
drome”). The pathophysiology of delayed
cerebral swelling remains the subject of great
debate in the sports medicine and scientific
ABBREVIATIONS: BESS, Balance Error Scoring
System; CI, confidence interval; GSC, Graded
Symptom Checklist; SAC, Standardized
Assessment of Concussion; SFWP, symptom-free
EFFECTS OF A SYMPTOM-FREE WAITING PERIOD ON
CLINICAL OUTCOME AND RISK OF REINJURY AFTER
OBJECTIVE:This study is the first to investigate the influence of a symptom-free waiting
period (SFWP) on clinical outcome and risk of repeat injury after sport-related concussion.
METHODS: This was a prospective, nonrandomized study of 16 624 player seasons
from 1999 to 2004, including a cohort of 635 concussed high school and college ath-
letes grouped on the basis of an SFWP or no SFWP observed after their concussion.
Clinical outcome in symptoms, cognitive functioning, and postural stability 45 and
90 days postinjury was compared with preinjury baseline. Data on SFWP and same-
season repeat concussion were recorded.
RESULTS: An SFWP was observed in 60.3% of cases. There were no significant differ-
ences between the SFWP and no SFWP groups in acute injury characteristics or clini-
cal outcome with respect to symptom recovery or postinjury performance on formal
neuropsychological and balance testing. Most repeat concussions (79.2%) occurred
within 10 days of the initial injury. The rate of repeat concussion was actually higher in
the SFWP group (6.49%) than the no SFWP group (0.90%) (P ? 0.005), but the repeat
concussion subgroup’s SFWP was 2.82 days shorter (95% confidence interval, 0.61–5.03;
P ? 0.01) and these athletes resumed participation 3.55 days sooner (95% confidence
interval, 0.06–7.04; P ? 0.05) than those in the SFWP group in which there was no
CONCLUSION: Our findings suggest that an SFWP did not intrinsically influence clin-
ical recovery or reduce risk of a repeat concussion. The overall risk of same-season
repeat concussion seems to be relatively low, but there may be a period of vulnerabil-
ity that increases risk of repeat concussion during the first 7 to 10 days postinjury. Further
study is required to investigate this preliminary finding and help determine whether
this risk can be reduced further with specific injury-management strategies.
KEY WORDS: Brain injury, Concussion, Neuropsychological tests, Sports injuries
Neurosurgery 65:876–883, 2009
SYMPTOM-FREE WAITING PERIOD AND REINJURY RISK AFTER CONCUSSION
VOLUME 65 | NUMBER 5 | NOVEMBER 2009 | 877
communities, because it is unclear whether closely spaced
injuries are the true underlying mechanism (5, 14, 21, 22, 28). A
large prospective study recently suggested that an athlete’s risk
of repeat concussion is increased within the first several days
after an initial concussion (12).
Theoretical assumptions that the brain remains in a metabol-
ically challenged (and, therefore, vulnerable) state for some
period of time after concussion have led to the lone constant
across all guidelines for clinical management of sport-related
concussion: that no athlete should return to participation until
completely symptom free (1, 4, 13, 14, 23). Presumably based on
the assumption that this window of cerebral vulnerability per-
sists beyond the point of observable clinical recovery (10, 29),
most guidelines take this recommendation a step further in
requiring that athletes resume participation only after being
completely symptom free for several days. The prescribed
length of this symptom-free waiting period (SFWP), however,
varies across respective guidelines, all of which are based
on expert consensus but not tested by scientific methods.
Moreover, the SFWP approach to injury management in sports
medicine is largely unique to concussion, and in practice, does
not seem to be universally embraced by clinicians or athletes
(8). The current study is the first to prospectively investigate
the extent to which an SFWP was actually implemented by cli-
nicians in the management of sport-related concussion, as well
as the association of an SFWP with clinical outcome and risk of
repeat concussion in a large sample of high school and college
athletes. In keeping with the prevailing theory that a period of
cerebral vulnerability exists after concussion, our hypothesis
was that observance of an SFWP would improve clinical out-
come and reduce the risk of repeat concussion.
PATIENTS AND METHODS
This study combined data sets from 3 parallel, multicenter studies
investigating the effects of sport-related concussion. Across the 3 studies,
16 624 player-seasons were investigated from 1999 to 2004. The current
investigation was focused on the 635 athletes (3.82% of player seasons)
who sustained a concussion during the combined studies. Please see the
notes section of Table 1 for more information on each independent study
and sample characteristics (e.g., sex, age, level of play, and sport). The
principal investigators for all 3 studies secured approval from the
Institutional Review Board for protection of human research subjects at
their host institutions. Written informed consent was obtained from all
participants (or parent/guardian) before enrollment in the study.
All players underwent a preseason baseline evaluation on a battery
of concussion assessment measures. Injured subjects were identified
and enrolled in the study protocol by a team physician or certified ath-
letic trainer present on the sideline during an athletic contest or prac-
tice. Concussion was defined according to the American Academy of
Neurology Guideline for Management of Sports Concussion, which
was the most widely accepted definition at the time these studies were
initiated (i.e., “a trauma-induced alteration in mental status that may or
may not involve loss of consciousness”) (13, 26).
The Graded Symptom Checklist (GSC), Balance Error Scoring
System (BESS), Standardized Assessment of Concussion (SAC), and a
neuropsychological test battery were used to assess postconcussive
symptoms, postural stability, and cognitive functioning. Several stud-
ies on the effects of sport-related concussion have demonstrated the
reliability and accuracy of these measures in correctly classifying
injured and noninjured patients after sport-related concussion (2, 6, 11,
15, 18–20). The athletes were tested on the GSC, SAC, and BESS on the
sideline immediately after injury, 2 to 3 hours later, and again on sev-
eral days during the first week postinjury. Neuropsychological testing
was administered 1 to 2 days and 1 week postinjury. All measures
were then readministered 45 days (Project Sideline) or 90 days postin-
jury (National Collegiate Athletic Association Concussion Study,
Concussion Prevention Initiative Study), the results of which are the
main focus of this study on clinical outcome. Alternate forms of all cog-
nitive outcome measures were used to reduce practice effects over
Two standardized methods were used to determine the point at
which athletes achieved complete symptom recovery. First, clinicians
used a standardized injury tracking system (Concussion Index) to
record detailed information on acute injury characteristics (e.g., uncon-
sciousness, amnesia), recovery time course, SFWP, incidence of repeat
concussion, and other aspects of clinical management. Within this
tracking system, clinicians were required to record the date on which
they thought, based on their overall clinical examination and impres-
sion, that the athlete demonstrated a complete resolution of symptoms.
This methodology was cross-validated using an empirically derived
method that classified an athlete to be symptom free on the date that
the GSC total score returned to the player’s preinjury baseline score.
The empirically derived and clinician impression measures of com-
plete symptom recovery yielded highly consistent results. In the rela-
tively small number of discrepancies (3.74% of all patients) between the
2 methods that changed the assignment of subjects on the study’s main
independent variable (i.e., SFWP or no SFWP group), the more conser-
vative method (i.e., lengthier estimate of complete recovery) was used.
TABLE 1. Statistics for duration of symptoms, symptom-free waiting
period, and total time lost after concussiona
(n ? 187) (n ? 367) Line (n ? 81)
CPI Project Side-Total
(n ? 635)
Time lost from
aNCAA, National Collegiate Athletic Association; CPI, Concussion Prevention Initiative;
SFWP, symptom-free waiting period. Data are mean (standard deviation [SD]) or mean
(95% confidence interval). The NCAA Concussion Study (4251 college player seasons
studied) was sponsored by the NCAA; the CPI Study (9094 high school and college player
seasons) was sponsored by the National Centers for Injury Prevention and Control at the
Centers for Disease Control and Prevention, and coordinated out of the Injury Prevention
Research Center at the University of North Carolina at Chapel Hill; and Project Sideline (3279
high school player seasons) studied high school athletes in suburban Milwaukee, WI. The
athletes from the total injured sample averaged 17.49 (1.62 SD) years of age, 10.14 (1.06
SD) years of education, and 6.72 (2.86 SD) years of organized participation in the sport being
studied. The sample was 88.5% male; there were 327 high school and 308 college athletes
in the concussion sample. Most concussions studied occurred in football (80%), followed
by soccer (13%), lacrosse (6%), and hockey (1%). One-third of athletes (33.5%) reported
a previous concussion, including 6.3% with 2 and 3.1% with 3 or more previous concussions.
MCCREA ET AL.
878 | VOLUME 65 | NUMBER 5 | NOVEMBER 2009
The design for each of the 3 studies varied slightly. The National
Collegiate Athletic Association Concussion Study and Concussion
Prevention Initiative Study both involved a multiple arm design strat-
ified based on the intensity of the assessment protocol, from 1 of the fol-
lowing: 1) routine clinical examination plus the GSC, 2) use of a brief
screening battery (GSC, SAC, and BESS), or 3) brief screening battery
plus neuropsychological test battery. Participating institutions were
randomly assigned to a specific arm of the study, so all injured athletes
from that institution underwent the same assessment protocol. Project
Sideline was a single group design in which all assessment instruments
were used for all patients.
Because preliminary analyses demonstrated the consistency of find-
ings across the 3 studies, we restricted statistical analysis to the com-
bined study sample. Cases with missing data that precluded computa-
tion of the SFWP variable or measurement of the study’s main outcome
variables were excluded from analysis. In total, 562 cases (88.5% of the
total sample) were available for analysis. There were no statistically sig-
nificant differences between those cases retained and eliminated from
analysis on the study’s main variables (e.g., acute injury characteristics,
duration of symptoms, clinical outcome measures).
Descriptive statistics were generated on the course of recovery by
category, total duration of symptoms, time lost from competition, and
duration of the SFWP. SFWP was computed based on the total time lost
from sports participation (consecutive days) minus the total duration of
symptoms in each case.
Independent group t tests and χ2analyses were conducted to com-
pare those athletes who observed (SFWP group) and did not observe
an SFWP (no SFWP group) with respect to acute injury severity (e.g.,
frequency of unconsciousness, amnesia, and acute GSC and SAC
scores), as well as total duration of symptoms, SFWP, and time lost
from competition. To control for baseline performance and to stay in
keeping with more advanced methods for statistically measuring
recovery after concussion, the same analyses were then conducted by
comparing the SFWP and no SFWP groups in terms of change from
baseline to day 45 of 90 on the clinical outcome measures. Using base-
line data from the large normative sample of athletes (n ? 2320), 6
main variables derived from the neuropsychological test battery (see
listed with Table 4) were scaled separately and the sum of the scaled
scores was scaled to a global index score (mean, 100; standard devia-
tion, 15) using the same standardized method validated by other well-
established neuropsychological test batteries (27, 30, 31). Scaling tables
derived from this large normative baseline sample were used to calcu-
late global neuropsychological test scores for each subject in the pres-
ent study, at each assessment point.
χ2statistics were generated to compare the SFWP and no SFWP
groups on the frequency of repeat concussion during the same season.
Additionally, descriptive statistics were generated for those patients
who sustained a repeat concussion in the same season, which were
analyzed and compared with those patients who did not sustain a
Across the 3 studies, 85.4% of injured patients reported a full
symptom recovery in less than 1 week, including 21.1% who
reportedly recovered within the first day. Only 2.7% of patients
reported symptoms beyond 1 month postinjury.
Table 1 provides statistics on mean duration of symptoms,
SFWP, and total time lost from competition.
An SFWP of any length was observed for 60.3% of patients,
including 14.6% who had an SFWP period of 7 days or more
(Table 2). There were no differences between the SFWP and no
SFWP groups in demographics (e.g., age, years of education) or
baseline performance on the concussion assessment measures
(GSC, SAC, neuropsychological test battery).
Acute injury severity was similar in the no SFWP and SFWP
groups. There were no statistically significant differences
between the 2 groups in duration of symptoms, performance
on the GSC and SAC immediately or 2 to 3 hours after concus-
sion, or the frequency of unconsciousness and posttraumatic
amnesia (Table 3).
Inherently, patients in the SFWP group had a significantly
longer waiting period than those in the no SFWP group, who
returned to play an average of 1.22 days (95% confidence inter-
val [CI], ?0.89 to ?1.55) before reaching a full symptom reso-
lution (mean difference, 7.10 days; 95% CI, 5.67–8.53; P ?
0.001). Overall, participants in the SFWP group were withheld
from competition significantly longer than those in the no
TABLE 2. Distribution of symptom-free waiting period durationa
line % (n)
?1 d, ?7 d
aSFWP, symptom-free waiting period; NCAA, National Collegiate Athletic Association;
CPI, Concussion Prevention Initiative.
TABLE 3. Acute injury characteristics, duration of symptoms, symp-
tom-free waiting period, and total time lost from competition for
subjects with and without a symptom-free waiting perioda
Loss of consciousness (%) 7.2 10.8
Posttraumatic amnesia (%) 22.927.1
Retrograde amnesia (%)15.7 18.6
21.39 (18.84–23.94)20.63 (18.85–22.41)
23.37 (22.49–24.25) 23.17 (22.28–24.06)
GSC 17.97 (15.61–20.33) 15.96 (14.01–17.91)
SAC25.15 (24.52–25.78)24.89 (24.27–25.51)
aSFWP, symptom-free waiting period; GSC, Graded Symptom Checklist; SAC, Standardized
Assessment of Concussion.Data are mean (95% confidence interval) unless otherwise noted.
bNo statistically significant group differences.
cHigher scores indicate more severe symptoms on the GSC.
dLower scores indicate more severe cognitive dysfunction on the SAC.
SYMPTOM-FREE WAITING PERIOD AND REINJURY RISK AFTER CONCUSSION
VOLUME 65 | NUMBER 5 | NOVEMBER 2009 | 879
SFWP group (mean difference, 6.52 days; 95% CI, 4.74–8.30;
P ? 0.001) (Fig. 1).
There were no statistically significant differences between the
SFWP and no SFWP groups in mean change from baseline to
day 45 or 90 postinjury on the GSC, BESS, SAC, and global neu-
ropsychological test scores (Table 4).
SFWP and Repeat Concussion
Overall, 24 patients experienced a repeat concussion in the
same sports season (3.8% of concussed patients, 0.14% of total
sample). These were all male football players (college, n ? 14;
high school, n ? 10). The average interval be tween first and
repeat concussion was 12.75 days (95% CI, 5.41–20.09); 70.8% of
repeat concussions occurred within the first 7 days of initial
injury and 79.2% were within the first 10 days. The average
duration of symptoms after first injury was 3.24 days (95% con-
fidence interval [CI], 1.86–4.62) in those patients who experi-
enced a repeat concussion. In the repeat concussion subgroup,
91.7% returned to competition within 1 week of their first
injury, compared with 59.3% of those with no subsequent
injury (χ2? 9.86; P ? 0.005).
The rate of repeat concussion in the same season was actu-
ally higher in athletes who had observed an SFWP (6.49%) after
their initial concussion than in those who had no SFWP (0.90%)
(χ2? 10.33; P ? 0.005). Ninety-one percent of patients who
experienced a repeat concussion had observed an SFWP after
their first concussion (mean duration, 2.96 days; 95% CI,
1.33–4.29). There was no difference in the amount of time
between preseason baseline testing and in-season date of first
concussion for the SFWP and no SFWP groups.
Within the SFWP group alone, there were no statistically
significant differences in the average duration of symptoms in
those who experienced a repeat concussion and those who
did not (mean difference, 0.73 days; 95% CI, ?0.82 to 2.28;
P ? 0.37). There were also no significant group differences in
acute injury severity (e.g., measured by the GSC and SAC
immediately and 2 to 3 hours postinjury). The repeat concus-
sion subgroup did, however, have a significantly shorter
SFWP (mean difference, 2.82 days; 95% CI, 0.61–5.03; P ?
0.01) and shorter total time withheld from competition (mean
difference, 3.55 days; 95% CI, 0.06–7.04; P ? 0.05) than the
SFWP group in which patients did not have a repeat concus-
sion (Fig. 2). There were no statistically significant differences
between those patients with and without a repeat concussion
in mean change from baseline to day 45 or 90 postinjury on
the main outcome measures (Fig. 3).
This multicenter, prospective study produced data from the
largest sample of athletes with sport-related concussion
reported in the literature, and this study is the first to investi-
gate the longstanding assumption that an SFWP affects clinical
outcome and reduces the risk of repeat injury after sport-
related concussion. Our findings suggest that the incidence of
FIGURE 1. Total duration of symptoms, symptom-free waiting period
(SFWP), and time to return to play (RTP) for the SFWP and no SFWP
TABLE 4. Mean change from baseline to day 45 of 90 on clinical
outcome measures for symptom-free waiting period and no symp-
tom-free waiting period groupsa
Outcome measure No SFWP groupb
GSC–2.36 (–3.39 to –1.33)–2.49 (–3.28 to –1.70)
SAC1.13 (0.65 to 1.61)1.16 (0.81 to 1.51)
BESS –2.68 (–4.33 to –1.03)–2.72 (–3.93 to –1.51)
chological Test Score
11.31 (7.53 to 15.09)12.32 (8.46 to 16.18)
aSFWP, symptom-free waiting period; GSC, Graded Symptom Checklist; SAC, Standardized
Assessment of Concussion; BESS, Balance Error Scoring System. Data are mean (confidence interval).
Negative change score indicates reduction in symptoms on GSC and improved performance on BESS;
positive change scores indicate improved performance on SAC and Global Neuropsychological
Test Score. The Global Neuropsychological Test Score was derived from standardized transformation
of scores on the Hopkins Verbal Learning Test, Trail Making Test Part B, Symbol Digit Modalities Test,
Stroop Color Word Test, and Controlled Oral Word Association Test.
bNo group differences were statistically significant.
FIGURE 2. Total duration of symptoms, symptom-free waiting period
(SFWP), and time until return to play (RTP) for athletes with and with-
out repeat concussion.
MCCREA ET AL.
880 | VOLUME 65 | NUMBER 5 | NOVEMBER 2009
ing which the risks of repeat
concussion are greatest,
because nearly 80% of repeat
concussions occurred within
10 days after initial injury. The
overall number of repeat con-
cussions (n ? 24) in the cur-
rent study was low, however,
which renders this finding
tentative. The low incidence
of repeat concussion in our
study (3.78% of athletes with
initial concussion, 0.14% of
total sample exposed) is also
consistent with previous
reports from prospective
studies (12, 17). Whereas the
overall risk of repeat concus-
sion in the same season was
equivalent to the baseline risk
of initial concussion (3.82%),
repeat concussions were far
more common during the first
10 days postinjury than
beyond the 10-day point (19
of 24 repeat concussions
occurred within the first 10 days of initial injury).
To our knowledge, there is only 1 other prospective study of
same-season repeat concussion that examined the interval
between first and repeat concussion (18). In that study, 2.6% of
college football players who had a concussion later sustained a
same-season repeat concussion. There were only 5 repeat con-
cussions studied, with an average interinjury interval of 33
days (range, 14–70 days). Same-season repeat concussions
would therefore seem to be relatively rare, with an incidence
that is not much different from the base rate of initial concus-
sion, but there may be an increased risk of repeat concussion
during the first 7 to 10 days after concussion. This finding
requires further study.
Our most unexpected and perplexing finding was that the
incidence of repeat concussion in the same sports season was
actually higher among athletes who observed an SFWP before
returning to competition after their concussion, despite the fact
that many athletes in the no SFWP group actually returned to
play while still symptomatic. It is noteworthy, however, that
those athletes who experienced a repeat concussion on average
had an SFWP 3 days shorter and returned to competition
nearly 4 days sooner than others who observed an SFWP with
no subsequent concussion. Also, the repeat concussion sub-
group (92%) was nearly twice as likely as the no repeat group
(59%) to return to competition within the first week after injury,
despite equivalent duration of symptoms and acute injury
characteristics in the 2 groups.
Is it unclear why repeat concussion was observed almost
exclusively in the group of athletes for whom an SFWP was
observed. It seems highly unlikely, of course, that returning to
repeat concussion in the same sports season is very low and
that observation of an SFWP did not intrinsically influence clin-
ical recovery or reduce the risk of a repeat concussion in the
same sports season.
Despite several years of conceptual agreement across sport-
concussion management guidelines on the importance of an
SFWP, this extended recovery time was observed in only 60%
of cases in the current study. Only 15% of athletes observed an
SFWP of 7 days or more, as prescribed by several concussion
management guidelines. When an SFWP was observed, the
average duration was less than 6 days, whereas athletes with
no SFWP actually returned to competition an average of 1 to 2
days before full resolution of their symptoms.
Our findings on clinical recovery closely parallel results
from earlier studies, and were therefore not unexpected (3, 7,
19, 20, 24, 25). Overall, less than 15% of injured athletes
required more than 1 week to achieve full recovery, and less
than 3% experienced symptoms beyond 1 month. A unique
contribution from the current study, however, is the finding
that this rapid recovery from sport-related concussion occurs
in the overwhelming majority of cases, regardless of whether
an SFWP is observed. Athletes who had no SFWP returned to
competition an average of nearly 7 days sooner than those in
the SFWP group, but did not demonstrate any persistent post-
concussive symptoms or residual impairments in cognitive
functioning or postural stability several weeks postinjury rel-
ative to athletes who observed an SFWP.
Our data are also consistent with the theoretical concept sug-
gesting that there is a 7- to 10-day window of cerebral vulner-
ability in the order of several days after concussion (9, 12), dur-
FIGURE 3. Study distribution overview. SFWP, symptom-free waiting period; RTP, return to play.
play while still symptomatic could be protective against repeat
concussion. There were no differences that we could identify,
however, between the 2 groups in acute injury characteristics
(e.g., unconsciousness, amnesia, symptom severity, degree of
cognitive dysfunction) that might have influenced team med-
ical personnel to adopt a more cautious approach for the play-
ers in the SFWP group. It is conceivable that some combination
of factors not easily identified on an objective basis (e.g., player
history, individual playing style) led clinicians to determine
that certain athletes might be at increased risk of repeat concus-
sion and therefore withheld them from play for a longer period.
Merely establishing an SFWP did not, however, protect these
athletes from repeat concussion. The larger group of athletes
initially considered at risk were protected from risk of repeat
concussion while observing a significantly lengthier SFWP and
period of no exposure while withheld from competition for the
first 10 days after concussion. It was when the SFWP was short-
ened and players initially classified in that same at-risk group
were returned to play several days sooner, that the risk of
repeat concussion was greatest. It is conceivable that if the
SFWP had been somewhat longer for these athletes, some of
the repeat concussions would have been avoided, but this also
remains speculative. To ultimately clarify the basis for the cur-
rent findings, further investigation is required to determine
what criteria clinicians rely on most when deciding observance
of an SFWP and an athlete’s eventual readiness to return to
competition after concussion.
In addition to the obvious clinical implications, our findings
have relevance to our understanding of delayed cerebral
swelling. Although several cases of catastrophic outcome after
a sport-related concussion have been reported, it is extraordi-
narily rare and the precise etiology remains unclear. The fact
that nearly 40% of athletes in this study returned to play before
they were symptom free (on average, 1.2 days before symptom
resolution), but experienced no severe adverse events, suggests
that this rare catastrophic outcome might not result from the
mechanistic effects of 2 closely spaced injuries, but is more
likely the result of an unusual physiological vulnerability (e.g.,
an idiosyncratic cerebrovascular autoregulation defect, cellular
disruption, or other factors mediating an unusual vulnerability
to minor brain trauma), independent of the characteristics, mul-
tiplicity, or timing of their concussions (21).
Several limitations to the current study warrant acknowl-
edgment, the most significant of which is that this was an
observational study and not a randomized controlled trial.
Given the existence of published clinical practice guidelines
prescribing the SFWP and expert consensus on the assumed
clinical benefit of the SFWP before this study, it is unlikely that
a study with a random assignment of an SFWP (versus no
SFWP) could have been ethically feasible. This study repre-
sents the largest prospectively studied sample of sport-related
concussion accumulated to date, and these data may be the
most informative that we can expect to obtain on the topic.
Despite this large sample of athletes exposed to injury risk, the
sample of repeat concussions in this study was very small and
highly consistent across all 3 studies. Sizable resources would
SYMPTOM-FREE WAITING PERIOD AND REINJURY RISK AFTER CONCUSSION
VOLUME 65 | NUMBER 5 | NOVEMBER 2009 | 881
be required to mount a controlled study of sufficient size to
yield a large enough subsample of repeat concussions to
explore the effects of management strategy (presuming that
one could obtain Institutional Review Board approval for a
randomized study of this nature). For instance, a study of
70 000 athletes would be required to accrue 100 repeat concus-
sions within the same sports season based on the rate in our
study and the only other prospective study on the rate of same-
season repeat concussion (17).
Another potential limitation is that symptom recovery in
this study was largely based on self-report, and it has been
suggested that athletes may be motivated to deny or mini-
mize their symptoms in order to return to competition. If any-
thing, however, this would suggest that a higher percentage of
athletes were returned to play while still symptomatic than we
have reported, which would not seem likely to significantly
alter our overall findings. In addition, the other main outcome
measures in the study (balance testing, neurocognitive test-
ing) are performance-based and provide a more objective
indication of recovery, especially when compared with each
individual player’s preinjury baseline score. Missing data
eliminated approximately 10% of cases from study. However,
separate analysis indicated that the eliminated cases did not
differ in any systematic way from the remaining subjects.
Finally, this study does not address the potential long-term
consequences of multiple concussions, which constitutes a sep-
arate focus of concern.
Findings from the current study suggest that the incidence of
repeat concussion in the same sports season is relatively low,
and that adherence to an SFWP did not intrinsically influence
short-term clinical outcome or reduce the risk of repeat injury
after sport-related concussion. These data also provide tenta-
tive support for the theory that there is a window of increased
vulnerability during the first 7 to 10 days after concussion, and
suggest that the risk of repeat concussion may be more closely
associated with that initial period of vulnerability than with
apparent symptom resolution or the observance of an SFWP.
These findings indicate the need for conservative injury man-
agement and return to play decision making by clinicians dur-
ing the first 7 to 10 days after initial sport-related concussion,
but further study is required to more precisely quantify the
true risks associated with sport-related concussion, and the
extent to which these risks can be modified by an SFWP or
other preventative injury management strategies.
This research was funded in part by the National Collegiate Athletic Associa -
tion, the National Operating Committee on Standards for Athletic Equipment,
Waukesha Memorial Hospital Foundation, National Academy of psychology,
National Federation of State High School Associations, National Football League
Charities, Green Bay Packers Foundation, Waukesha Service Club, and the
Medical College of Wisconsin General Clinical Research Center (M01 RR00058).
Major funding came from the National Centers for Injury Prevention and Control
(for the Concussion Prevention Initiative Study, and the University of North
MCCREA ET AL.
882 | VOLUME 65 | NUMBER 5 | NOVEMBER 2009
sion in collegiate football players: The NCAA Concussion Study. JAMA
21. McCrory P: Does second impact syndrome exist? Clin J Sport Med 11:144–
22. McCrory PR, Berkovic SF: Second impact syndrome. Neurology 50:677–683,
23. McCrory P, Johnston K, Meeuwisse W, Aubry M, Cantu R, Dvorak J, Graf-
Baumann T, Kelly J, Lovell M, Schamasch P: Summary and agreement state-
ment of the 2nd International Conference on Concussion in Sport, Prague
2004. Br J Sports Med 39:196–204, 2005.
24. Pellman EJ, Lovell MR, Viano DC, Casson IR, Tucker AM: Concussion in
professional football: Neuropsychological testing—Part 6. Neurosurgery
25. Pellman EJ, Viano DC, Casson IR, Arfken C, Powell J: Concussion in profes-
sional football: Injuries involving 7 or more days out—Part 5. Neurosurgery
26. Quality Standards Subcommittee of the American Academy of Neurology:
Practice parameter: The management of concussion in sports (summary state-
ment). Neurology 48:581–585, 1997.
27. Randolph C: Repeatable Battery for the Assessment of Neuropsychological Status
(RBANS). San Antonio, The Psychological Corporation, 1998.
28. Saunders RL, Harbaugh RE: The second impact in catastrophic contact-sports
head trauma. JAMA 252:538–539, 1984.
29. Slobounov S, Slobounov E, Sebastianelli W, Cao C, Newell K: Differential rate
of recovery in athletes after first and second concussion episodes. Neurosurgery
30. Wechsler D: The Wechsler Adult Intelligence Scale. San Antonio, The Psycho -
logical Corporation, 1997, ed 3.
31. Wechsler D: The Wechsler Memory Scale. San Antonio, The Psychological
Corporation, 1997, ed 3.
We thank the certified athletic trainers, team physicians, and student athletes
from the high schools and universities that participated in this study.
n this prospective study of more than 16 600 high school and college
football players, 635 players sustained a concussion, and, of these, 24
sustained a second concussion. Notably, in 79% of these 24 players, the
second concussion occurred within 10 days of the initial concussion
and occurred at a much higher rate (6.5% versus 1%) in the group of
players who observed a symptom-free waiting period (SFWP) than in
the group who did not observe an SFWP. As the authors indicate, this
important and somewhat surprising finding suggests that the absolute
time for return to play is a more important parameter than concussion-
related symptoms in determining the window of vulnerability to a sec-
ond concussion. Furthermore, it suggests that the cerebral metabolic
pathophysiology that underlies this vulnerability has a relatively con-
sistent timeline that largely resolves within 10 days of impact in most
individuals. Until further data are available, the authors’ recommenda-
tion of a minimum waiting period of 10 days postconcussion before
return to athletic play appears to be prudent, simple, and safe.
Daniel F. Kelly
Santa Monica, California
sion. Curiously, the incidence of repeat concussion in the same season
was increased in athletes who underwent an SFWP before returning to
competition. However, SFWP athletes who experienced a repeat con-
cussion had a significantly shorter SFWP (almost 3 days shorter) and
earlier return to competition (3.5 days sooner) than those who under-
he authors investigate the widespread belief that mandating an
SFWP reduces the risk of repeat injury after sports-related concus-
Carolina Injury Prevention Research Center contributed to the execution and suc-
cess of this project. Michael McCrea, Ph.D., had full access to all of the data in the
study and takes responsibility for the integrity of the data and the accuracy of the
data analysis. None of the authors has any conflict of interest to declare with
respect to this study, nor personal financial or institutional interest in any of the
drugs, materials, or devices described in the article. The funding sponsors did not
have a role in the design or execution of the study beyond funding, and had no
involvement in the preparation of the manuscript.
1. Aubry M, Cantu R, Dvorak J, Graf-Baumann T, Johnston K, Kelly J, Lovell M,
McCrory P, Meeuwisse W, Schamasch P: Summary and agreement statement
of the First International Conference on Concussion in Sport, Vienna 2001.
Recommendations for the improvement of safety and health of athletes who
may suffer concussive injuries. Br J Sports Med 36:6–10, 2002.
2. Barr WB, McCrea M: Sensitivity and specificity of standardized neurocogni-
tive testing immediately following sports concussion. J Int Neuropsychol
Soc 7:693–702, 2001.
3. Belanger HG, Vanderploeg RD: The neuropsychological impact of sports-
related concussion: A meta-analysis. J Int Neuropsychol Soc 11:345–357,
4. Cantu RC: Cerebral concussion in sport. Management and prevention. Sports
Med 14:64–74, 1992.
5. Cantu RC: Second-impact syndrome. Clin Sports Med 17:37–44, 1998.
6. Collins MW, Grindel SH, Lovell MR, Dede DE, Moser DJ, Phalin BR, Nogle
S, Wasik M, Cordry D, Daugherty KM, Sears SF, Nicolette G, Indelicato P,
McKeag DB: Relationship between concussion and neuropsychological per-
formance in college football players. JAMA 282:964–970, 1999.
7. Echemendia RJ, Putukian M, Mackin RS, Julian L, Shoss N: Neuropsycho -
logical test performance prior to and following sports-related mild traumatic
brain injury. Clin J Sport Med 11:23–31, 2001.
8. Ferrara MS, McCrea M, Peterson CL, Guskiewicz KM: A survey of practice
patterns in concussion assessment and management. J Athl Train 36:145–149,
9. Giza CC, Hovda DA: The neurometabolic cascade of concussion. J Athl Train
10. Gosselin N, Thériault M, Leclerc S, Montplaisir J, Lassonde M: Neurophysio -
logical anomalies in symptomatic and asymptomatic concussed athletes.
Neurosurgery 58:1151–1161, 2006.
11. Guskiewicz KM: Assessment of postural stability following sport-related con-
cussion. Curr Sports Med Rep 2:24–30, 2003.
12. Guskiewicz KM, McCrea M, Marshall SW, Cantu RC, Randolph C, Barr W,
Onate JA, Kelly JP: Cumulative effects associated with recurrent concussion
in collegiate football players: The NCAA Concussion Study. JAMA 290:2549–
13. Kelly JP, Rosenberg JH: Diagnosis and management of concussion in sports.
Neurology 48:575–580, 1997.
14. Kelly JP, Nichols JS, Filley CM, Lillehei KO, Rubinstein D, Kleinschmidt-
DeMasters BK: Concussion in sports. Guidelines for the prevention of cata-
strophic outcome. JAMA 266:2867–2869, 1991.
15. Lovell MR, Collins MW, Iverson GL, Field M, Maroon JC, Cantu R, Podell K,
Powell JW, Belza M, Fu FH: Recovery from mild concussion in high school
athletes. J Neurosurg 98:296–301, 2003.
16. Macciocchi SN, Barth JT, Alves W, Rimel RW, Jane JA: Neuropsychological
functioning and recovery after mild head injury in collegiate athletes.
Neurosurgery 39:510–514, 1996.
17. Macciocchi SN, Barth JT, Littlefield L, Cantu RC: Multiple concussions and
neuropsychological functioning in collegiate football players. J Athl Train
18. McCrea M: Standardized mental status testing on the sideline after sport-
related concussion. J Athl Train 36:274–279, 2001.
19. McCrea M, Barr WB, Guskiewicz K, Randolph C, Marshall SW, Cantu R,
Onate JA, Kelly JP: Standard regression-based methods for measuring recov-
ery after sport-related concussion. J Int Neuropsychol Soc 11:58–69, 2005.
20. McCrea M, Guskiewicz KM, Marshall SW, Barr W, Randolph C, Cantu RC,
Onate JA, Yang J, Kelly JP: Acute effects and recovery time following concus-
SYMPTOM-FREE WAITING PERIOD AND REINJURY RISK AFTER CONCUSSION Download full-text
VOLUME 65 | NUMBER 5 | NOVEMBER 2009 | 883
went an SFWP without subsequent concussion. The majority of repeat
concussions occurred within 10 days of the initial injury. The authors
conclude that the simple use of an SFWP, in and of itself, did not influ-
ence clinical recovery or reduce the risk of a repeat concussion in the
same season. Instead, the time between injury and return to competi-
tion may have been more important, and it may be more appropriate
for trainers and team physicians to wait for a defined period before
allowing return to play after a concussion, instead of waiting only for
symptom resolution with or without a brief SFWP.
It is also noteworthy that almost 40% of concussed players in this
study returned to play before they were symptom-free (on average, 1.2
days before symptom resolution) yet experienced no major adverse
events. As the authors discuss, this finding suggests that the second-
impact syndrome may have more to do with an aberrant physiological
response than to the time interval between injuries.
Within the limits of study methodology, the authors could not iden-
tify any inherent differences in baseline characteristics between the
SFWP and non-SFWP groups. Perhaps trainers and team physicians
had subtle reasons to delay return to play in those athletes who ended
up in the SFWP group, but the answers to such questions await further
Alex B. Valadka
cCrea et al. report on a cohort of 635 high school and collegiate
athletes who sustained concussions during the years 1999–2004;
they focus on the contribution of an SFWP and the likelihood of recur-
rent injury. Although 60% of the injured players had an SFWP, this
group actually had a higher risk of recurrent concussion. This counter-
intuitive finding should be considered within the context of several
limitations of the study. First, the number of second concussions was
low, at only 24 athletes. In addition, the data were self-reported in
nature and observational, as the authors note. Also, the potential for the
SFWP to affect the potential for cumulative brain injury cannot be fully
assessed in a contemporaneous study.
The findings nevertheless highlight some important features of this
population. It is not universal that an SFWP is observed across the
sports medicine community. The occurrence of second concussion is
unusual within the same season, and some players were still sympto-
matic when allowed to return to play. The authors believe that there
may be a period of increased susceptibility for repeat injury within the
7 to 10 days after a concussion. Rather than assume that this means that
a management style that minimizes an SFWP is safe, it likely affirms
that other factors may be involved with an obvious, currently deleteri-
ous outcome, such as acute cerebral edema (second impact syndrome)
or noticeable diminution in playing ability or in mental functioning.
The harmful effects of second concussion more likely are related to
factors such as genetic predisposition, the kinetic energy or biome-
chanical factors of each particular impact (neck strength, being “blind-
sided,” g forces, etc.), multiple blows within the same contest (e.g.,
boxing), style of play, severity (grading) of each concussion, and others.
The authors’ detailed work and collection of extensive data that
should add to our understanding of the myriad features of concussion.
This study is provocative in encouraging us to continue to analyze all
aspects of traditional management strategy, and it affirms the impor-
tance of individualization of the care and decision-making process for
these athletes. The above being stated, I do not think that we should
assume that disregarding an SFWP is advisable or desirable. As I like
to emphasize, these patients are the only ones who present for the neu-
rosurgeon’s evaluation and recommendation and basically say,
“Doctor, I have injured my brain, and it has healed; I now would like
to go back and sustain innumerable more brain impacts. Is that fine?”
Julian E. Bailes
Morgantown, West Virginia
ver since the seminal papers by Cantu in the 1980s, the hallmark of
sports concussion management has required a symptom-free
period before allowing a return to competition. This tenet is now being
challenged with reasonably robust scientific data by McCrea et al. As
the authors point out, however, data derived from self-reporting—
especially from highly competitive athletes—is always open to ques-
tion. That, 40% of the time, presumably knowledgeable athletic train-
ers and sports medicine clinicians, especially those at the National
Collegiate Athletic Association level, chose to ignore this basic tenet is
alarming, although it was apparently not harmful to the 19 players
who experienced a second concussion while still symptomatic.
Although I would agree with the authors that the second-impact
syndrome likely represents a highly individualized selective vulnera-
bility, the one constant across all reports of its occurrence is repeat con-
cussion in athletes with ongoing symptoms. Since it remains impossi-
ble to prospectively identify such vulnerable athletes, it is difficult to
support a non-SFWP policy, in spite of this study’s conclusions. This is
especially true because the data appear to show a 7- to 10-day period
of increased vulnerability to a second concussion.
Jack E. Wilberger