Keywords
running - Lysholm knee scoring scale - body mass index - sport-related injuries -
knee injuries - patellofemoral pain syndrome
Introduction
Running is the sport that most contributes to the occurrence of injuries in physically
active adult individuals.[1] The incidence of lower limb injuries in runners ranges from 19.4 to 92.4%, affecting
mainly the knees, with a specific incidence from 7.2 to 50%;[2] 30 to 70% of these injuries require training reduction and > 79% require medical
attention.[3] Anterior knee pain, also called patellofemoral pain (PFP),[5] is a frequent cause for medical care.[4]
Short and long distance recreational runners report mainly knee injuries,[6] with 50% of them resulting from excessive use.[3] In addition, these lesions may be associated with risk factors, such as body mass
index (BMI)[7] and advanced age.[2]
[3]
[8] Since injuries are multifactorial, studies on running-related risk factors must
present a high quality to allow precise conclusions.[9] For Powers et al.,[10] failure to treat this lesion is constant, and it can be attributed to the lack of
understanding of its causes.
The diagnosis is based on the history and physical examination of the patient, since
imaging tests, including radiography and magnetic resonance imaging (MRI), do not
provide specific findings.[11] As such, qualitative and quantitative assessment tools are required.[12] These tools include the Lysholm questionnaire, due to its reliability and validity
in athletes and patients with joint cartilage conditions,[13]
[14] and the Patellofemoral Disorders Scale (Kujala Anterior Knee Pain Scale), which
is a specific tool for anterior knee pain evaluation.[15]
[16]
[17]
Due to the diverse etiology, the diagnosis is complex and susceptible to interpretation
errors.[18] Therefore, the Lysholm and Kujala questionnaires can provide additional information
to the history and physical examination of the patient, reducing the inaccuracy in
clinical evaluation; in addition, these are easily applied, low-cost tools. The main
objective of the present study was to compare the subjective perception of pain and
anterior knee pain symptoms in people with different BMI classifications. The secondary
objective was to verify the association between biological and anthropometric variables
with subjective questionnaires scores. Our initial hypothesis is the existence of
an association between biological (age) and anthropometric (BMI) variables with pain
perception and PFP symptoms.[7]
[8]
Material and Methods
Study Design
The sample consisted of 126 recreational runners of both genders, aged between 20
and 59 years old. All volunteers were recruited by invitation and declared they did
not run competitively. An Informed Consent Form (ICF) was signed; this document provided
the telephone number of the researchers in charge to resolve possible doubts, since
there was no direct contact with the volunteers. The document was in a digital format,
according to a project approved by the Research Ethic Committee (CEP, in the Portuguese
acronym) under the number 2.774.475/2018.
The study was conducted through questionnaires digitally available on a single link,
via the internet; a brief explanatory text about these tools was also provided. In
addition, data regarding age, knee pain intensity according to the visual analog scale
(VAS), weight and height for BMI calculation were collected. Next, recruited individuals
were encouraged to answer the Lysholm and Kujala questionnaires on their computers,
notebooks, cell phones, tablets, or other electronic devices.
Body mass index
Body mass index was used to assess the subject's weight in relation to height, with
the following classification: < 18.5, low weight; from 18.5 to 24.9, normal weight;
from 25 to 29.9, overweight; ≥ 30, obesity. The BMI was calculated by dividing the
body mass in kilograms (kg) by the squared height (m2). Data were provided by the subjects, who were instructed to weight themselves on
a digital or analogic scale, and to measure their heights before answering the online
questionnaires.
Lysholm questionnaire
The Lysholm questionnaire is a specific knee questionnaire, which was translated and
validated in Portuguese.[19] The questionnaire was answered by the volunteers, who chose only one answer per
item. Items are divided into limping, support, locking, instability, pain, swelling,
climbing stairs and squatting. The score 5 refers to the maximum in the items support,
limping and squatting, the score 10 refers to the maximum in the items swelling and
climbing stairs, the score 15 refers to the maximum in the item locking and the score
25 refers to if the maximum in the items instability and pain. The total score is
classified as excellent (≥ 95 points), good (94 to 84 points), fair (83 to 65 points)
and poor (≤ 64 points).[20]
Kujala questionnaire
The Kujala questionnaire (Patellofemoral Disorders Scale) is used to assess anterior
knee pain and functional limitations. It was validated and translated into Portuguese,
and it is the only questionnaire that concomitantly evaluates anterior knee pain,
patellofemoral joint function and patellar alignment.[12] It scores from 0 to 100 points, where 0 represents the absence of pain and/or functional
limitations, and 100 points corresponds to constant pain and several functional limitations.
It consists of 13 multiple choice items, and 1 answer per item is allowed. Items are
divided into limping, supporting body weight, walking, going up and down stairs, squatting,
running, jumping, sitting for a long time with bent knees, pain in the affected knee,
swelling, subluxations, loss of muscle mass and difficulty flexing the injured knee.
A maximum score of 5 points is attributed to limping, sustaining body weight, walking,
squatting, loss of muscle mass and difficulty flexing the injured knee, while a maximum
score of 10 points is given to going up and down stairs, running, jumping, sitting
for a long time with bent knees, pain in the affected knee, swelling and subluxations.
Scores are classified as excellent (≥ than 95 points), good (94 to 85 points), fair
(84 to 65 points) and poor (≤ 64 points).[15]
Visual analog scale
The VAS was used to subjectively measure the level of knee pain in recreational runners.
The classification goes from 0 to 10 points, where 0 to 2 corresponds to mild pain,
3 to 7 equates to moderate pain and 8 to 10 represents severe pain. The VAS was answered
according to current pain during the application of the questionnaire.[21]
Statistical analysis
Descriptive data were presented as mean ± standard deviation (SD). Data normality
was examined using the Shapiro-Wilk test. A paired sample t-test compared mean values
from parametric data, while the Wilcoxon test was used for nonparametric data. Variables
were analyzed by Pearson linear correlation. The 95% confidence interval (CI) for
variables association was calculated. The magnitudes of the correlation adopted were
(r): Trivial when less than or equal to 0.1; small when greater than 0.1 to 0.3; moderate
when greater than 0.3 to 0.5, large when greater than 0.5 to 0.7, very large when
greater than 0.7 to 0.9 and almost perfect when greater than 0.9 to 1.0. In case of
95%CI overlapping, small positive and negative magnitude values were considered unclear;
otherwise, the observed magnitude was considered.[22] Significance was adopted at p ≤ 0.05. Analyzes were performed using IBM SPSS Statistics for Windows, Version 22
(IBM Corp., Armonk, NY, USA). Figures were generated with GraphPad Prism software,
version 6.0 (San Diego, CA, USA).
Results
A total of 138 questionnaires were responded, with 126 considered viable and included
in the analyzes. Twelve questionnaires were excluded: 5 due to the duplicate participation,
2 from subjects younger than the age stipulated by our study, 1 from a volunteer older
than required, and 4 for not fully completing the questionnaire.
Descriptive data regarding age, height and body mass from recreational runners are
presented in [Table 1]. A significant height difference was observed between overweight and grade 1 obesity
subjects (p = 0.029) ([Table 1]); body mass was significantly different when comparing the normal weight group to
the overweight (p < 0.001) and grade 1 obesity groups (p < 0.001) ([Table 1]).
Table 1
|
Normal weight group
|
Overweight group
|
Grade 1 obesity group
|
|
Age (years old)
|
33.83 ± 7.98
|
34.10 ± 8.23
|
39.22 ± 8.84
|
|
Height (m)
|
1.67 ± 0.08
|
1.71 ± 0.09
|
1.68 ± 0.08[**]
|
|
Weight (kg)
|
63.26 ± 8.22
|
78.64 ± 9.48[*]
|
91.55 ± 11.87[*]
|
The mean BMI values from overweight subjects were significantly different to those
from normal weight subjects; the mean BMI values from grade 1 obesity subjects were
significantly different from normal weight and overweight subjects ([Figure 1A]). The VAS, Kujala and Lysholm mean scores presented no significant difference between
groups ([Figure 1B], [1C], [1D]).
Fig. 1 (A) The black bar refers to the average body mass index (BMI) of normal weight subjects,
the light gray bar represents the average BMI of overweight subjects, and the dark
gray bar indicates the average BMI grade 1 obesity subjects. (B) The black bar corresponds to the average visual analog scale (VAS) score of normal
weight subjects, the light gray bar shows the average VAS score of overweight participants,
and the dark gray bar corresponds to the average VAS score of grade 1 obesity patients.
(C) The black bar symbolizes the average Kujala score of normal weight subjects, the
light gray bar refers to the average Kujala score of overweight subjects, and the
dark gray bar represents the average Kujala score of grade 1 obesity individuals.
(D) The black bar refers to the average Lysholm score of normal weight participants,
the light gray bar refers to the average Lysholm score of overweight subjects, and
the dark gray bar represents the average Lysholm score of with grade 1 obesity individuals.
*, significant difference compared with normal weight subjects, p ≤ 0.05; **, significant difference compared with overweight subjects, p ≤ 0.05.
Significant correlations between BMI and the VAS (r = 0.18; p = 0.04), Kujala score (r = - 0.17; p = 0.05) and Lysholm score (r = - 0.22; p = 0.01) were observed ([Figure 2]); however, there were no significant correlations between age and specific questionnaires
and VAS scores ([Figure 3]).
Fig. 2 Correlation between body mass index (BMI) and subjective scales scores. The black
circle corresponds to the correlation with the visual analog scale (VAS), the white
circle represents the correlation to the Kujala score, and the gray circle shows the
correlation with the Lysholm score. The black line represents the limit between positive
or negative correlation. The gray area shows the trivial correlation threshold, while
the dotted lines represent small, moderate, large, very large or almost perfect correlation
thresholds. *, significant difference, p ≤ 0.05.
Fig. 3 Correlation between participants' age and subjective scales scores. The black circle
corresponds to the correlation with the visual analog scale (VAS), the white circle
represents the correlation to the Kujala score, and the gray circle shows the correlation
with the Lysholm score. The black line represents the limit between positive or negative
correlation. The gray area shows the trivial correlation threshold, while the dotted
lines represent small, moderate, large, very large or almost perfect correlation thresholds.
*, significant difference, p ≤ 0.05.
The virtual questionnaire inquired the running experience of participants, with three
possible answers: 1 - < 6 months; 2 - > 6 months; 3 - ≥ 1 year. The weekly frequency,
referring to how many times a week the subject does street running, was also questioned,
with three possible answers: 1 - Once a week; 2 - Twice a week; 3 - ≥ 3 times a week.
Regardless of their nature, all answers were included in our analysis; this information
was collected to better understand the characteristics from our volunteers.
Discussion
Our main findings were the following: 1-) There is a significant difference between
mean BMI values. 2-) BMI has a significant correlation with the VAS, Kujala and Lysholm
scores. 3-) There is no significant correlation between age, subjective pain scale
and specific questionnaires. The significant difference observed between BMI classifications
([Figure 1A]) is due to the difference between the body weight from normal weight, overweight
and grade 1 obesity subjects ([Table 1]) and the height difference between the overweight and grade 1 obesity groups ([Table 1]).
Our data suggest that the BMI is associated with the level of pain and PFP symptoms
([Figure 2]). Linton et al.[7] observed that injured subjects have a higher BMI compared with noninjured individuals,
so BMI can be a risk factor for running injuries; in addition, runners from the injured
group reported both a knee injury in the previous 12 months and a current lesion.
Kastelein et al.[23] detected an association between persistent knee pain in subjects with BMI values > 25 kg/m2 during the 1st year of follow-up; those same patients, at a 6-year follow-up, presented bilateral
symptoms, including reports of knee swelling and locking sensation in the Lysholm
questionnaire. Similarly, Nielsen et al.[24]
[25] highlighted that an increased BMI consequently increases the risk of running-related
injuries, and that BMI values < 20 kg/m2 are considered protective factors for the development of lesions.[24]
Neal et al.,[26] demonstrated that BMI is not a risk factor for injuries in runners since the evaluated
papers show evidence that subjects both above or within an ideal weight are predisposed
to PFP development; furthermore, these authors state that the risk of having this
type of pain is present regardless of the type of runner. Nevertheless, these results
are not yet fully elucidated in the literature. Vitez et al.[8] and Linton et al.[7] observed that overweight runners are more susceptible to injuries than those with
normal weight, corroborating our findings, which demonstrate an unclear significant
correlation between BMI and VAS, Kujala and Lysholm scores.
Our results indicate that age has a trivial correlation with specific knee and pain
questionnaires. On the contrary, Gion-nogueron et al.,[3] Van Gent et al.,[2] and Vitez et al.[8] pointed out that advanced age is a risk factor for lower limb injury. A recent systematic
review and meta-analysis found moderate evidence that age is not a risk factor for
patellofemoral pain in runners, including recreational runners.[26] According to Nielsen et al.,[24] middle-aged runners between 45 and 65 years old are more susceptible to running-related
injuries. This observation justifies the trivial correlation found by our study, in
which most volunteers were young adults ([Table 1]), since few symptoms are reported by this age group.
Our study had two main limitations: Indirect contact with volunteers and the lack
of distinction between pain and injury, mainly due to the difficulty in diagnosing
and controlling the injury factor. Future studies must attempt to control variables
that the literature proposes as risk factors for PFP, including sport experience,
flexibility, patellar alignment, quadriceps muscles strength, weekly training volume,
running speed, running shoes and mileage covered with them, step type, guidance and
periodization by a professional, as well as face-to-face questionnaire application
and the differentiation between pain and injury.
Conclusion
We conclude that a high BMI value can be a causative factor for knee pain in recreational
runners; therefore, the weight of such subjects must be controlled to minimize the
occurrence of injuries. Lysholm and Kujala questionnaires can be used to assess knee
symptoms in this population, providing additional information to the physical evaluation
and assisting in preventive strategies, as they enable the characterization of current
symptoms.
