Key words
ACL injury - risk factor - environment - recreational alpine skiing - temperature - prevention
Introduction
Recreational alpine skiing, a popular winter sport activity, is enjoyed annually by
hundreds of millions of skiers worldwide [1]
[2]. In Austria, the overall rate of skiing-related
injuries is less than one injury per 1,000 skier days [3], but the total number of skiing-related injuries is high due to the
large number of recreational alpine skiers [3]. Over
the last few decades, there has been a continuous decrease in alpine skiing-related
injuries, however, the proportion of skiing-related knee injuries has remained
constant [4].
The knee joint remains the most common injured anatomical body location, representing
about one-third of all injuries among female and male recreational alpine skiers
[4]
[5]. Moreover,
50% of recorded knee-injured skiers suffer from an injury to the anterior
cruciate ligament (ACL) [6]. Posch et al. [7] have shown that ACL ruptures during skiing generally
occur in combination with concomitant injuries to the medial collateral ligament
[7].
It is well known that any skiing-related knee/ACL injury is likely the
outcome of a complex interaction of multiple intrinsic and extrinsic risk factors
[8]. While intrinsic risk factors for injury in
skiing include age, sex (biological factors)/gender (psychological and
sociological factors), skill level, etc., extrinsic risk factors are equipment
(ski-boot-binding unit) or environmental-related (snow, slope, and weather
conditions) [4]
[9].
Given the variable conditions that exist in an open-air outdoor environment, such as
alpine skiing, the combination of risk factors related to skier safety and injury
risk are complex [1].
Previous research has shown that knee injury prevalence in female recreational alpine
skiers is two-fold greater than male skiers [4]
[10]
[11]. Furthermore, ACL
injury risk is three times more likely in females [4]
[10]
[11].
This interesting fact may be partly attributed to individual (intrinsic) risk
factors [4]
[8]
[9], e. g. hormonal, anatomical, and
neuromuscular factors and differentiated by sex [12]
[13]. Also, sex-based differences in
equipment-related factors have been investigated, such as ski binding adjustment
regarding knee/ACL injuries in recreational alpine skiing [4]
[14].
Beside sex/gender, skill level is another important intrinsic risk factors
for an ACL injury in skiing. Indeed, a recent study found that beginner skiers are
independently associated with a higher ACL injury risk [15]. Additionally, it is clear that lower skill level of skiers increases
the overall injury risk on alpine ski slopes [16]
[17]. Thus, individual skill level may represent an
important confounder when evaluating injury risks in skiers.
Environmental factors such as snow conditions [18]
[19]
[20], slope difficulty
[21], and weather conditions [4]
[9]
[22] have also been suggested to modify the prevalence
and risk of non-contact knee injuries in recreational alpine skiing.
Specifically, low ambient temperature and snowfall have been shown to be significant
environmental risk factors for knee injuries in female recreational alpine skiers
[13]. In this study, the prevalence of knee
injuries in females was highest when the ambient temperature was low (between
-20° and -8°C), while the number of knee-related injuries in male
skiers did not significantly differ across different ambient temperatures [13]. In addition, knee injuries are two times greater
compared to other injured body parts when skiing during snowfall (15.4% vs.
8.6%, respectively) [13]. In another study, no
significant differences between ACL injured male and female skiers regarding snow
conditions, slope difficulty, or weather conditions were found [23]. Moreover, a further case-control study
investigating multiple risk factors for an ACL injury in female skiers revealed that
icy snow conditions and snowfall increased ACL injury risk by 24-fold and by
17-fold, respectively [24].
However, the above-mentioned studies on the potential impact of environmental factors
on the injury risk do have some limitations due to low sample size [20]
[21]
[24], missing control groups of uninjured skiers [13]
[20]
[23], and/or detailed diagnosis of the knee
injury [13]. Thus, to address this gap in the
literature, we undertook a large case-control study, focusing on ACL-injured female
and male recreational alpine skiers compared to uninjured skiers (matched for sex
and skiing skill level) based on a sufficiently large sample size, i. e.
N>780. It remains unknown if environmental factors are predictive for an ACL
injury in recreational alpine skiing. Therefore, this study aimed to evaluate the
impact of various environmental factors on the ACL injury risk in recreational
alpine skiers across a broad age range.
Materials and Methods
Study design
This study was conducted as a retrospective matched case-control study of
ACL-injured and uninjured male and female recreational alpine skiers during six
consecutive winter seasons (2014/15 to 2019/20), treated in a
single high-patient volume Austrian sports clinic. In order to avoid confounding
by different skiing skill levels between ACL-injured skiers and controls, a sex
and skill-level-matched control group was created.
In accordance with previous studies [7]
[15]
[25], cases were
interviewed annually between December and April over 23 days on average per
winter season using a standardized questionnaire. The recruitment of patients
was dependent upon logistical aspects at the sports clinic (availability of
rooms and personnel) and the readiness of patients to participate. In total,
more than 95% of invited patients agreed to participate [7].
Justification of the sample size is primarily based on the concept of events per
variable (EPV), whereupon Peduzzi et al. [26]
suggested the number of EPV analyzed in logistic regression analysis to be
values of 10 or greater. In the present study, EVP is 26.1 (392 cases and 15
independent predictors, considering the intercept in the logistic
regression).
All study participants were informed about the aims of the study and gave their
written informed consent for taking part in the study. The survey was conducted
according to the ethical guidelines for surveys approved by the Institutional
Review Board (IRB) as well as the Board for Ethical Issues (BfEI) of the
University of Innsbruck (25/2016).
Cases
Inclusion criteria of cases were (i) a skiing-related, non-contact ACL injury
after a fall without the involvement of another skier, (ii) an age>18
years, and (iii) the use of any type of carving skis (in contrast to long and
unshaped traditional skis as well as so-called short ski boards). The ACL injury
was diagnosed by a physician using magnetic resonance imaging (MRI), which is
directly available at the sports clinic [7].
Controls
As described earlier [15], uninjured controls were
simultaneously (i. e. on the same day) chosen at various spots from the
same ski area throughout the whole skiing day on 15 days on average per winter
season. Inclusion criteria were an age > 18 years and the use of any
type of carving skis.
Out of a large pool of controls (N=1425), controls were randomly matched
for sex and skill level to cases (with a one-to-one basis).
Questionnaire
Beside demographic data, sex, age, and skiing skill level (expert vs. advanced
vs. intermediate vs. beginner) and according to Sulheim et al. [27], ACL injured skiers were asked to report
environmental factors like weather conditions (sunny vs. overcast vs. snowfall),
snow conditions (fresh snow vs. grippy snow vs. icy snow vs. slushy/soft
snow), perceived temperature (very cold/cold vs. neutral vs.
warm/very warm), and the difficulty of the downhill slope (easy
– blue vs. moderate – red vs. difficult – black) at the
timepoint of the accident [13]
[24].
According to the methods described in an earlier study, controls rated the
overall weather conditions, snow conditions, perceived ambient temperature and
difficulty of their preferred slopes on the day of the interview [24].
Furthermore, we categorized participants into more skilled (expert and advanced)
and into less skilled (intermediate and beginner) skiers, as a previous study
showed a tendency to underestimate individual skill level, especially among
female skiers [27].
Statistical analysis
All data are presented as means±standard deviations, absolute and
relative frequencies. According to tests on normal distribution
(Kolmogorov-Smirnov), univariate differences among metric data (age) between
cases and controls were evaluated either by independent t-tests or Mann-Whitney
U tests.
Differences in frequencies (weather conditions, snow conditions, perceived
temperature, and difficulty of the downhill slope) were evaluated by chi-square
tests.
Additionally, according to the univariate results, multiple logistic regression
analysis (entering all risk factor variables with p<0.2) was used to
calculate multivariate OR and 95% confidence interval (CI) [14]
[15]. Within
variables with more than two categories, the last category was used as
reference. SPSS 26.0 (IBM Corporation, Armonk, NY) was used for the statistical
analysis. All p-values were two-tailed, and statistical differences were
considered significant at p<0.05.
Results
A total of 392 ACL injured and 392 uninjured skiers (57.9% females) with a
mean age of 39.6±12.2 years participated. Differences between cases and
controls regarding mean age, weather conditions, snow conditions, perceived
temperature, and difficulty of the downhill slope are presented in [Table 1].
Table 1 Characteristics and univariate odds ratios of
environmental risk factors in uninjured (controls) and in ACL-injured
(cases) recreational alpine skiers.
|
Controls (n=392)
|
Cases (n=392)
|
p-value
|
Age [years]
|
36.4±13.0
|
42.7±10.5
|
<0.001
|
Weather conditions [n, %]
|
|
|
0.167
|
sunny
|
265 (67.6)
|
267 (68.1)
|
|
overcast
|
105 (26.8)
|
91 (23.2)
|
|
snowfall
|
22 (5.6)
|
34 (8.7)
|
|
Snow Conditions [n, %]
|
|
|
<0.001
|
fresh snow
|
40 (10.2)
|
59 (15.1)
|
|
grippy
|
209 (53.3)
|
244 (62.2)
|
|
icy
|
39 (9.9)
|
72 (18.4)
|
|
slushy/soft
|
104 (26.5)
|
17 (4.3)
|
|
Perceived Temperature [n, %]
|
|
|
<0.001
|
Very cold/cold
|
96 (24.5)
|
145 (37.0)
|
|
Neutral
|
164 (41.8)
|
167 (42.6)
|
|
Warm/very warm
|
132 (33.7)
|
80 (20.4)
|
|
Difficulty of the downhill slope [n, %]
|
|
|
<0.001
|
Easy (blue)
|
77 (19.6)
|
210 (53.6)
|
|
Moderate (red)
|
226 (57.7)
|
149 (38.0)
|
|
Hard (black)
|
89 (22.7)
|
33 (8.4)
|
|
Data are presented as mean values±standard deviation, absolute and
relative frequencies
Multiple logistic regression results are shown in [Table
2]. An increasing age is significantly associated with ACL injury. With
regard to snow conditions, ACL injury risk increased by 10.5-fold for fresh snow,
7.8-fold for grippy snow, and 12.4-fold for icy conditions, respectively as compared
to the reference category slushy/soft snow. ACL injury risk increased
1.6-fold for very cold/cold perceived temperature compared to the reference
category warm/very warm. On easy (blue) slopes, ACL injury risk is 6.9-fold
higher when compared to the reference category difficult (black) slope.
Table 2 Adjusted odds ratios (enter method) of individual and
environmental risk factors associated with an ACL injury
Risk factors
|
Coefficient
|
Standard error
|
df
|
p-value
|
Odds ratio (95% CI)
|
Age
|
0.049
|
0.007
|
1
|
<0.001
|
1.05 (1.04–1.07)
|
Weather conditions
|
sunny
|
0.280
|
0.366
|
1
|
0.445
|
1.32 (0.65–2.71)
|
overcast
|
−0.241
|
0.374
|
1
|
0.520
|
0.79 (0.38–1.64)
|
Snow Conditions
|
fresh snow
|
2.351
|
0.423
|
1
|
<0.001
|
10.49 (4.58–24.02)
|
grippy
|
1.956
|
0.289
|
1
|
<0.001
|
7.79 (4.12–14.72)
|
icy
|
2.634
|
0.321
|
1
|
<0.001
|
12.43 (6.03–25.61)
|
Perceived Temperature
|
Very cold/cold
|
0.495
|
0.252
|
1
|
0.050
|
1.64 (1.00–2.69)
|
neutral
|
0.014
|
0.232
|
1
|
0.952
|
1.01 (0.64–1.60)
|
Difficulty of the downhill slope
|
Easy (blue)
|
1.935
|
0.268
|
1
|
<0.001
|
6.93 (4.09–11.71)
|
Moderate (red)
|
0.461
|
0.253
|
1
|
0.069
|
1.59 (0.97–2.60)
|
Notes: All entered factors in the multiple logistic regression analysis were
adjusted for each other. Nagelkerke R-Square=0.356, classification
table – overall percentage=73.3%,
Discussion
The primary aim of this study was to evaluate the potential impact of environmental
factors on the ACL injury risk in recreational alpine skiers. To the best of our
knowledge, this is the first study using a matched (for sex and skill level)
case-control study design, considering both ACL-injured and uninjured male and
female recreational alpine skiers to evaluate the potential impact of environmental
factors on ACL injury risk.
The main findings were twofold: a) increasing age is significantly associated with
ACL injury, and b) with regard to environmental factors, fresh, grippy, or icy snow
conditions, very cold/cold perceived temperatures, and skiing on easy ski
slopes were independent predictors for an increase in ACL injury risk.
ACL injured skiers in this study were five years older on average than uninjured
controls. Results of the multiple logistic regression revealed that an increase of
age by 5 years increased ACL injury risk by 25%. In accordance, other
studies also reported a higher age as a risk factor for knee/ACL injuries in
recreational alpine skiing [9]
[15]. A potential explanation for this result could be that increasing age
is associated with a progressive loss of neuromuscular function due to a reduction
of muscle mass and muscle quality as well as changes in the biology, healing
capacity, and biomechanical function of tendons and ligaments [28]
[29].
Regarding weather conditions, about 68% of ACL-injuries occurred on sunny
days, which seems to be in line with findings by Ruedl et al. [24] and Burtscher et al. [9], who reported that 70% and 63% of knee injuries
occurred during sunny conditions, respectively. However, no significant differences
with regard to weather conditions were found between cases and controls neither in
the simple nor in the multiple analysis. Usually, skiers prefer to ski on sunny days
[9]
[22]. Aschauer et
al. [22] found a somewhat higher overall injury risk
when skiing during strong snowfall compared to sunny conditions. However, our
findings do indicate that snow and slope conditions are more important predictors
for ACL injury than weather conditions.
Compared to slushy/soft snow conditions, the risk to experience an ACL injury
increased by 11-fold on fresh snow and by 8-fold on grippy snow in this study.
Interestingly, Moore & Knerl [30] reported a
negative correlation between 24-hour snowfall accumulation and injury severity among
both skiers and snowboarders. Thus, one could assume that fresh snow reduces the
overall severity of injury risk on ski slopes, which makes logical sense considering
the protective effects of falling on soft snow as well as the overall skiing speed
on fresh snow is less than on ice. Given that we did not look at the meteorological
data for snowfall on the days where fresh snow was reported, we can only rely on the
skiers’ self-reported assessments of the snow conditions. It seems
reasonable to speculate that a key reason for our increased ACL risk is due to
catching an edge, which has been shown to be the main kinematic feature of how ACL
ski injuries occur [7]. To explain with fresh snow
accumulation, terrain variations on the slopes are less identifiable, which can lead
to reduced anticipation of bumps and furrows typically evident on ski slopes. Thus,
the opportunity to either catch an edge or a ski tip with enough force that leads to
a fall might result in a subsequent ACL injury.
The higher ACL risk when skiing on grippy snow compared to slushy/soft snow
might be partly caused by a higher mean speed on grippy slopes. Carus &
Castillo [31] reported a mean skiing speed of about
58 km/h on grippy snow and about 49 km/h on wet
snow, which seems comparable to slushy/soft snow conditions in this
study.
Yet the snow condition with the highest risk (OR 12) of an ACL injury while alpine
skiing was icy snow. In this case, icy conditions were about 1.5-fold higher risk
than grippy snow conditions. This aligns with previous research that found that
increased ACL injury risk on icy ski slopes was 24 times greater than
slushy/soft snow conditions in female skiers [24]. Interestingly, Bouter et al. [18]
found that it might not just be the overall self-report of icy conditions but also
exposure to icy spots on slopes that suddenly cause a change in stance and balance,
resulting in falls. Regardless, it can be agreed that skiers have less edge control
on consistently icy snow slopes than grippy snow slopes and that icy spots in mixed
snow conditions are associated with a more likely loss of balance and falling, which
can further result in ACL injuries [24].
In the current study, low perceived temperature turned out to be an independent risk
factor for ACL injuries in recreational alpine skiing. In an earlier study, Ruedl et
al. [13] found a significantly higher prevalence of
knee injuries in female skiers when objectively measured ambient temperature was
low, while male skiers did not show such temperature-dependent differences. It is
important to mention that this temperature dependent ACL injury risk might also be
related to alterations in further environmental conditions that are consequential to
either a rise or fall in temperature [32]. To explain,
previous research has found that colder ambient temperatures are associated with
harder conditions that may progress to icy surface conditions, thereby increasing
risk [18]
[24] of alpine
skiing accidents and associated injuries. In addition, Csapo et al. [32] have suggested that cooling of the thigh and knee
leads to a reduction in the capacity to generate force explosively, particularly
affecting the knee flexor muscles. Moreover, a reduction in hamstring co-activation
during forceful contraction of knee extensors was observed, which may further
contribute to increased ACL injury risk in the cold [32]. It is well known that just one hour of skiing reduces the skin and
intraarticular temperatures of the knee joint [33].
Furthermore, Piedrahita et al. [34] reported a reduced
ability to maintain dynamic balance caused by the cooling of the lower leg and
Dewhurst et al. [35] showed that muscle fiber
conduction velocity of knee muscles was slower in a cold environment. Cooling
increases the latency of soleus H-reflex, which is an important mechanism in
controlling motor behavior of the lower leg [36].
Knowing that losing balance or catching an edge of the ski are the most common
reported circumstances for a fall resulting in an ACL injury in recreational alpine
skiers [23], it is clear that very cold ambient
temperature conditions likely have a direct influence on the control of skiing,
especially when a skier is turning or carving a turn. It is suggested that
preventive measures to counteract cold-related effects of ambient weather conditions
while alpine skiing should include adequate clothing and more frequent breaks to
warm up. This recommendation is likely more important in female skiers [13]. Based on previous research [37], warming up for 15 minutes results in a
higher body temperature for at least 30–45 minutes post warm-up
compared to no warm-up when skiing.
The majority of ACL-injuries (54%) occurred on easy (blue) slopes compared to
uninjured controls, who reported to be mainly skiing on moderate (red) slopes. Due
to matching for skiing skills, the preference of slope type should not differ much
between ACL-injured and control skiers. According to the results of the logistic
regression analysis, skiing on easy ski slopes is associated with a 7-fold higher
risk for an ACL injury. In contrast, Demirag et al. [21] reported that hard and steep slopes increased risk for any knee
ligament injury. We would surmise that a greater number of beginners choose flat
slopes, especially when learning how to ski. Thus, the skill level of skiers
choosing flat and moderate slopes rather than the slope itself may explain our
findings. This would align with Sulheim et al. [17],
who found that the overall injury risk of skiing-related injuries increases among
beginners as well as total knee injuries. When considering only less skilled
ACL-injured recreational alpine skiers (data not shown), about 60% reported
that the accident happened on easy slopes. In addition, carelessness and higher
skiing speed of more skilled skiers skiing on easy slopes could provoke catching an
edge, which potentially results in a subsequent fall, leading to an injury. Finally,
there are fewer steep slopes compared to easy and intermediated slopes in most ski
resorts, including those used in this study, thus the influence of skill level,
total time on moderate and flat slopes, and reckless behavior of skilled skiers on
moderate slopes potentially contribute to the ACL risk found in this study.
To sum up, this is the first large, matched, case-control study (N=784) that
has evaluated the potential impact of various environmental factors on the ACL
injury risk in recreational alpine skiing. Our results revealed 5 environmental
factors to be predictive for an ACL injury. While some intrinsic risk factors, such
as sex/gender and age are not modifiable [8],
extrinsic – i. e. environmental risk factors can be considered
strategically by the individual skiers, e. g. avoiding icy spots on ski
slopes and very cold ambient air temperatures. Recreational alpine skiers should
consider adequate clothing and use more frequent breaks to warm up [13].
However, it seems difficult to avoid easy slopes, which are often leading the skier
to the chair lift station at the bottom of the slope. Thus, the impact of
environmental factors in this study seems limited, also due to a Nagelkerke R-Square
of 0.356 as Ruedl et al. [38] recently reported,
individual factors such as age, skill level, and risk-taking behavior, as well as
equipment-related factors such as ski length, tip, waist width of the ski, and
standing heights are predictive for an ACL injury on ski slopes with a Nagelkerke
R-Square of 0.706. Nevertheless, an improved understanding of current environmental
risk factors for an ACL injury could provide a valuable basis for future research,
especially for preventive interventions. As these results have important
implications for skiers, these findings should be used in knowledge translation to
raise the importance of specific environmental factors that may cause ACL injury
risk in recreational alpine skiing. In addition, as a practical preventive measure,
ski lessons will be useful to improve skill level and in order to become familiar
with the slide characteristics of the skis on different slopes and in different
environmental conditions.
Strengths and Limitations
Strengths and Limitations
Due to the restriction of our patients to the ski clinic, we cannot entirely exclude
a possible selection bias of ACL-injured skiers. However, a major part of knee
injuries occurring in the study area were treated in the ski clinic and there are no
obvious indications of any source of selection. With regard to the factor difficulty
of the ski slope, some uncertainties might arise as we compared self-reported data
on preferred slope difficulties for the uninjured control group and the actual slope
difficulty where cases suffered the ACL injury. The inclusion of a control group
(matched for sex and skiing skill level) clearly strengthens the study results, as
the impact of risk factors on the onset of an injury event can be investigated using
an appropriate case-control study design. Another strength of the study is that
uninjured controls were interviewed mostly (95%) on the same days when
ACL-injured cases were treated in the ski clinic to enable the analysis of
investigating the impact of environmental factors on the ACL injury risk.
Additionally, in certain cases, poor snow and weather conditions may have been
misjudged as ACL injured skiers may look for an explanation as to why the injury
occurred [24].
Conclusion
An increasing age is a significant individual factor associated with an ACL injury in
recreational alpine skiing. Regarding environmental factors, icy and easy slopes,
fresh and grippy snow, and low ambient temperatures are associated with an increased
ACL injury risk in recreational alpine skiing. Those factors are at least partly
modifiable and should be taken into consideration for the implementation of
preventive strategies.
Disclosure statement
The authors report that there are no competing interests to declare.