Key words
Sports Injury Rates - Acute Pain - Martial Arts - Return to Play
Introduction
Injury and pain are natural sequela of contact sports. An estimated 8.6 million
people are injured per year through sports and recreational activities in the United
States [1]. Pain from sports injuries has been
discussed in a few studies such as that of soccer, basketball, and rugby, but the
vast majority of these studies have discussed pain as it relates to a surgical
procedure or rehabilitation [2]
[3]
[4]
[5]
[6]. Risk factors for pain have been analyzed in
several descriptive studies and cohort studies [7]
[8]
[9], but these studies do not discuss an
athlete’s perception of pain following different types of acute injury and
how it relates to time loss. With increasing concern over re-injury and concussion,
there is concern with returning to sport too early before underlying physiological
structures have healed.
Taekwondo is a Korean martial art with over 80 million practitioners worldwide [10]. Sparring, a subspecialty of Taekwondo, is
a full-contact sport that permits kicks to the head and torso and punches to the
torso. World Taekwondo (WT), the regulating organization, requires that competitors
wear protective gear to cover the trunk, groin, forearms, shins, gloves, head, and
mouth [11]. Competitors are matched by age,
rank, and weight and fight three 2-minute bouts per match. There have been numerous
studies describing the frequency, severity, type, and location of Taekwondo injuries
[11]
[12]
[13]. In a meta-analysis,
Taekwondo had a competition injury rate of 79.3/1000 athlete-exposures (AE),
which is higher than the rate in men’s American football (35.9/1000
AE), soccer (18.8/1000 AE), and hockey (16.3/1000 AE) [11]
[14].
Moreover, the prevalence of concussion in Taekwondo has been noted to be four-fold
higher than American football over a 15-year span (9.4/1000 AE vs.
2.3/1000 AE). For all these injuries, there is no data yet on whether pain
or injuries have fully resolved prior to returning to sport. In addition, pain is
the body’s failsafe to stop participating in a sport. Certain injuries may
have higher time loss than others, yet there is minimal follow-up on these injuries
and knowledge as to how pain and injury type relate to time loss.
To fill this gap, we documented pain and injury characteristics of 62 athletes in
a
collegiate Taekwondo season and followed up on their pain for six weeks. The purpose
of this study is to investigate how the participant’s self-reported pain
scores change over time following an acute injury. Secondarily, this study explores
the relationship of pain perception, as measured by the Numerical Rating Scale
(NRS), to injury type, injury location, time loss, and athlete experience level.
Finally, this study aims to identify risk factors associated with higher pain
perception and inform guidelines related to injury rehabilitation and return-to-play
standards.
Methods
Data collection
This is a prospective observational study that investigates Taekwondo
athletes’ pain following injury during the 2008–2009 Pacific
West Taekwondo Conference season. Eighty athletes attended at least one of 13
competitions (12 scrimmage-level and 1 multi-state-level) from October 2008
until March 2009. A total of 53 matches between men and 18 matches between women
occurred.
The project staff included study volunteers who were present for all tournaments
of the season. Whenever an injury occurred, the subject was examined by the
tournament healthcare professional and the healthcare professional’s
exam, diagnosis, and treatment were documented by the project staff onto a paper
injury form. Project staff also recorded whether the injured athlete won or lost
the bout, the round in which the injury occurred, and the mechanism of injury
(i. e., punch or kick and which type). Multiple injuries were filled out
on multiple forms. Each athlete was assigned a unique identifier and subsequent
injuries were matched to the same athlete. An injury was defined as a stoppage
of action for medical attention and classified according to the Orchard Sports
Injury Classification System (OSICS-10). This definition included both
sudden-onset and acute exacerbation of slow-onset injuries, but degree of onset
or history of previous injury was not recorded.
Pain was recorded using the Numeric Rating Scale for Pain (NRS) ([Fig. 1]) [15]
[16]. The descriptors ranged
from 0–10, with 0 being no pain and 10 being the worst pain possible and
requiring bed rest. Range 1–3 was defined as mild, 4–6 as
moderate, and 7–10 as severe and prohibitive for competition. The
project staff followed up with all injured athletes at two weeks and six weeks.
At each interval, pain was reassessed using the NRS, which has been shown to be
an efficient and accurate measure of pain assessment and monitoring [15]
[17].
Fig. 1 Numerical rating scale (NRS) for pain.
This paper is a follow-up to the senior author’s previous studies of
injury rates and training habits of collegiate Taekwondo athletes [13]
[18]. Approval to conduct this study was obtained from the
Institutional Review Board (IRB). Informed consent was obtained from the
subjects and this study followed the latest guidelines in Ethical Standards in
Sport and Exercise Science Research [19].
All subjects were given a verbal description of the study and participants were
assured of confidentiality and that their participation in the study would not
jeopardize participation in the competition. Participation was voluntary and a
written description of the study was available upon request.
Statistical analysis
Descriptive statistics were used to describe pain scores and missing data values
were excluded from analyses. Linear mixed-effects (LME) models were used to
compare the athletes’ reported NRS pain score at the time of injury with
respect to injury type, location, and types of techniques performed leading to
the injury. This model accounted for correlation within subjects because some
individuals had several injuries (repeated measures). Similarly, LME models were
used to examine associations of age, number of months of training, and belt
level (black belt level 1, 2, 3, and novice) with pain score at the time of
injury. To examine pain and time-loss from training, we fitted an LME model with
the main exposure as the pain score and adjusted for injury type, location, and
whether individuals had multiple injuries. Missing values were excluded on a
complete-case basis. Analyses were performed using SAS version 9.3 with the
significance level set to 0.05.
Results
Participant characteristics
There were 42 men and 20 women injured during our study. Mean age, weight,
height, and months of experience of the athletes are depicted in [Table 1]. Most participants were black
belts (75.8%). ([Fig. 1S])
depicts a flow diagram of our recruitment process.
Table 1 Study participant characteristics for N=62
athletes
|
Variable
|
Mean (SD) or Count (%)
|
|
Age (years)
|
20.90 (2.47)
|
|
Sex (female)
|
20 (32.36%)
|
|
Weight (kg)
|
66.33 (8.89)
|
|
Height (cm)
|
171.85 (7.79)
|
|
Experience (months)
|
92.24 (56.27)
|
SD: standard deviation.
Injuries and pain levels
A total of 111 injuries were documented. The distribution of the number of
injuries were: 56.45% of athletes sustained only one injury,
25.81% sustained two injuries, 8.06% sustained three injuries,
6.45% sustained four injuries, and the remaining two athletes had 6 and
7 injuries, respectively. We examined the association of reported pain levels to
injury type, location, and type of techniques performed leading to the injury.
[Figure 2] presents reported pain
levels at the time of injury (baseline) and follow-up visits.
Fig. 2
a Distribution of injuries. b Pain levels.
Our primary analysis examined the outcome pain at the time of injury. Mean level
of pain scores were higher for sprain/strain (mean 5.4, standard error [SE]
0.47) and contusion (4.9 SE 0.94), which were both significantly higher than the
“other” (3.2 SE 0.53) type of injury (p=0.0062 and
p=0.0117, respectively). Other group differences were not statistically
significant ([Fig. 3]). For location of
injury (head, upper or lower body), the lower body (5.6 SE 0.36) reported the
highest level of pain, followed by the upper body (3.9 SE 0.52) and head (3.1 SE
0.50); the average pain score for the lower body was significantly higher
compared to upper body (p=0.0186) and head (p=0.0014) injuries
([Fig. 4]). With respect to
techniques performed leading to the injury, the mean pain score for fall (5.8 SE
0.67) and roundhouse/double kick (4.9 SE 0.33) were significantly higher
than a punch (2.6 SE 0.76; p=0.0113 and p=0.0062, respectively).
No other significant differences with respect to techniques were found ([Fig. 5]).
Fig. 3 Pain according to type of injury.
Fig. 4 Pain according to injury location.
Fig. 5 Pain according to method of injury.
Pain levels and experience
We also examined athletes’ reported pain score at the time of injury in
association with the level of experience of the injured athlete. We found that
pain was not associated with several indicators of experience level, including
age, number of months of training, and belt level (black level 1, 2, 3, and
novice).
Pain and time loss from training
Median time loss was 2 days and nearly all athletes had a time loss of less than
50 days; however, there were 5 individuals with extreme time loss (one with 90
days, three with 365 days, and one with 120 days). Thus, we examined the time
that the reported pain score at injury was associated with time loss from
training along with other factors of injury (location, type, and whether there
were multiple injuries). There was a significant positive association between
pain and time loss, where an increase in pain score of 1 point was associated
with about 0.85 days (SE 0.37) of time lost from training (p=0.0284)
after controlling for injury type, location, and whether individuals had
multiple injuries ([Table 2]).
Table 2 Pain and time loss from training
|
Variable
|
Group
|
Estimate
|
SE
|
P-value
|
|
Pain at injury
|
–
|
0.849
|
0.372
|
0.0284
|
|
Location
|
Head
|
–1.807
|
2.820
|
0.5361
|
|
Lower body
|
2.819
|
2.269
|
0.2425
|
|
Upper body (reference)
|
–
|
–
|
–
|
|
Type
|
Concussion
|
6.884
|
4.657
|
0.1576
|
|
Contusion
|
–3.042
|
3.005
|
0.4902
|
|
Sprain / Strain (reference)
|
–
|
–
|
–
|
|
Multiple injury
|
No
|
–0.177
|
1.856
|
0.9250
|
|
Yes (reference)
|
–
|
–
|
–
|
SE: standard error.
Discussion
Our study is the first to investigate and document, prospectively, the amount of pain
athletes reported after an injury in competition. Historically, most sports pain
literature describes epidemiological data, pain tolerance, or pain prevalence, and
there are few quantitative studies on the perception of pain as it relates to
different athlete characteristics. Our study was able to uniquely analyze pain
perception as a function of experience level, injury type, and time loss using the
NRS.
We found that injuries to the lower extremities and contusions were the most common
injuries by location and type, respectively, which agrees with previous findings.
A
recent review of Taekwondo injuries shows that contusions were the most common
injury and that the lower extremity was the most vulnerable [20]. This has been shown to be true at the
Olympic Games and other major tournaments [11]
[13]
[21]
[22].
Because kicks are a very common method of contact, it seems reasonable that
contusions to the lower extremities are the most common injuries in Taekwondo.
We used the Numerical Rating Scale to assess the level of pain associated with
Taekwondo injuries. Pain is highly subjective and therefore difficult to standardize
even when using the NRS. However, we followed up with our subjects after their
injuries and were able to note their declining pain levels and kept a consistent
record of their subjective level of pain. We found that our most common injuries,
contusions and lower extremity injuries, were also the most painful. Because we
ambulate with our lower extremities and thus place large amounts of pressure on our
feet, it makes sense that a lower extremity injury is more painful. Most studies
have used the NRS to compare post-surgical outcomes or rehabilitation programs, and
thus we were unable to find comparative usage of the NRS for acute injuries.
Interestingly, potentially devastating injuries such as concussion did not rate as
painful to the athletes in our study. However, these head injuries may be more
concerning because of concerns about long-term damage, and more recently, chronic
traumatic encephalopathy (CTE) [23]
[24]. Concussion has been shown to be four times
more prevalent in Taekwondo than in football [21], although many concussed Taekwondo athletes return to play within a
few days [25], which we found to occur among
our athletes as well. Moreover, we were able to note that our athletes’
concussive pain fully dissipated within two weeks, adhering to current
return-to-play protocols that rely on the athlete feeling asymptomatic with activity
[26]. However, recent research shows that
complete recovery of underlying anatomical structures lags behind clinical symptom
resolution even after one year [27]. As our
study shows, concussive pain may be a poor indicator of injury resolution because
most people do not report pain at two weeks post-injury.
There is a need to clarify what concussive pain is, because it often does not feel
acutely painful the same way a bruise does. Neck pain may be the most commonly
conceived notion of pain, but not all concussed patients develop neck pain. In
recent studies, only 68.4% of patients in the emergency room presented with
neck pain within three days of injury, and confusingly, up to 20% of varsity
collision-sport athletes were found to have pre-season neck pain [28]
[29].
Furthermore, a myriad of concussion symptoms are closely intertwined with and
mistaken for other conditions such as depression, anxiety, and PTSD. Because
previous concussions are a risk factor for future concussions, it may be that many
athletes return to play without fully recovering from a previous concussion [26]. This may highlight a lack of clarity in
mapping concussive pain to symptom resolution, which may lead athletes to return to
play too soon.
Our study suggests that although most Taekwondo competition injuries appear mild,
many athletes elect to return to sport before their pain has fully resolved. At two
weeks, 39.6% of our athletes still had pain, yet the median time loss was
only two days. Our time-loss rates are comparable to previous studies which show
that up to 74% of competition injuries result in < 1 day of
time loss, and most of the remaining injuries result in a time loss of 2–7
days [30]. Sprains/strains are common
in Taekwondo [11], yet in our study, pain
actually increased after six weeks. Furthermore, our linear mixed-effects model
showed that injury location, type, and presence of multiple injuries did not affect
time loss, suggesting that players may return soon after injury regardless of the
type of injury they have experienced. Traditionally, it takes contusions four to
seven days to heal and fractures four to eight weeks to heal, yet time loss was
unaffected by injury type, showing how pain may ambiguously map to underlying
physical damage.
It is crucial to standardize return-to-play procedures for athletes, keeping in mind
the limitations of using pain as a gatekeeper for play. Acute pain is the
body’s physiological monitor for whether to restart an activity once the
pain is tolerable or subsides completely, but if it is tolerable, we must ensure
that there is no damage to the underlying structures. Further research needs to be
done to determine the amount of time for optimal injury healing that is easily
measured via radiographical evidence, the development of biomarkers that measure
pain, or recovery from injury, or psychometric analysis.
Strengths and limitations
Our study was able to classify the types, locations, and methods of injury and
compare them to the severity of pain according to the NRS during the
2008–2009 Pacific West Taekwondo Conference season. We followed up with
our subjects until their injuries resolved. Limitations to our study include
small sample size and lack of investigation as to mechanistically why lower
extremity injuries were most painful. Perhaps a future study can use a larger
sample size of athletes across the martial arts or sports spectrum or include a
meta-analysis from multiple competitions.
Conclusion
Our study was able to classify the types, locations, and method of injury and compare
them to the severity of pain according to the NRS. The data and multiple follow-ups
showed that lower extremity contusions were the most painful and common injury.
Importantly, head injuries, although potentially more devastating and of widespread
public concern, were less painful.
