According to Fitzgerald et al study in Pittsburgh, Pennsylvania, USA(14) who
estimated the prevalence of knee buckling in knee osteoarthritis patients at 63% and
assuming a confidence level of 0.05 and an accuracy of 0.01, the sample size was
The patients who were included in the study signed university-approved written informed
consent forms prior to participation and completed demographic data sheets. All
participants were allowed to be excluded at any time, if they did not want to continue the
study. One hundred and two patients were enrolled to study; 12 subjects did not fulfill the
inclusion criteria thus, the main participants were 90 patients (65 females, 25 males).
Data for each subject were collected during one testing session that lasted approximately
1.5 hours. During the session, subjects first completed a demographic and health history
questionnaire, a self?report measure of function, and rated the severity of knee pain and
knee instability. Following completion of the questionnaires, physical performance
measurements of function were administered.
Outcome measures were consisted of knee pain intensity by Visual Analogue Scale (VAS),
timed up and go test (TUG), six-minute walk test (6MWT) and Western Ontario and
McMaster Universities Osteoarthritis Index (WOMAC). Ebrahimzadeh et al assessed validity
and reliability of WOMAC and demonstrated its suitability for Persian speaking patients with
A self-reported knee instability check list ,based on the study of Felson et al (2),was used to
appraise the episodes of buckling of the knee in the past 3 months. Persons reporting knee
giving way were additionally asked for the number of buckling in the previous 3 months;
whether these episodes concerned the left, right, or both knees and whether knee buckling
had resulted in a fall. We also asked what the patients were doing when their knee buckled
and they were asked to write the name of the activity.
Two functional tests were used for the assessment of performance status in patients with
KOA. In this study the following functional tests were used; three trails of the tests were
performed and the average of their results was recorded.
• The Timed “Up & Go” Test(TUG)
The TUG was used to provide a timed measure of balance and functional mobility in
the KOA patient. The test requires the patient to rise from a standard armchair, walk 3 m at a comfortable pace, walk back to the chair, and sit down. Timing begins when
the person starts to rise from the chair and ends when she returns to the chair and sits down .The time of TUG was recorded by a chronometer Fox 40 (Fox 40 Co, Canada) with 1/100 second precision (16).
• Six-Minute Walk Test (6MWT)
The 6MWT is a safe, easy to administer and good tolerated functional test for KOA patients which reflects activities of daily living. It is a measure of endurance. The primary measurement in 6MWT is the distance covered in 6 min. The 6MWT was completed in an enclosed corridor on a flat course 30 m in length. The object in this test was to walk as quickly as for six minutes around the track (16).
To evaluate the effect of buckling on physical function, we used WOMAC as a self-reported questionnaire consisting of 24 items divided into 3 subscales;
– Pain (5 items): during walking, using stairs, in bed, sitting or lying, and standing upright
– Stiffness (2 items): after first waking and later in the day
– Physical Function (17 items): using stairs, rising from sitting, standing, bending, walking, getting in / out of a car, shopping, putting on / taking off socks, rising from bed, lying in bed, getting in / out of bath, sitting, getting on / off toilet, heavy domestic duties, light domestic duties.
Each item is scored on a scale of 0 to 4 on the basis of the amount of difficulty experienced; the total score ranges from 0 to 96 (15).
Buckling was more common in knees with both involvement of tibiofemoral and
patellofemoral than in those without this condition (10.3% vs. 4.1%). The prevalence of
buckling was 18.5% among patients with knees pain graded more than 4 based on the VAS
with 6.9%among knees with pain less than 4. Patients with knee buckling, also, had higher
WOMAC disability scores than did those without buckling.
Statistical analysis showed the following correlations after verifying the normalization of
data distribution. Chi Square test was used to find a correlation between the knee buckling
and gender which showed a significant relationship between the knee buckling and the
female gender (p = 0.027, ? = 0.225), however, there was no significant correlation between
the knee buckling and BMI and age in the patients (respectively; p = 0.185, ?= 0.000 and p =
0.341, ?= 0.021) in the study. Chi Square test, also, showed a significant correlation between
the knee buckling and history of KOA in the patient (p = 0.033, ?= 0.67).
Buckling was also associated with the involvement of both tibiofemoral and patellofemoral
joints in the participants (p = 0.019, ? = 0.825) and the severity of knee pain (p = 0.029, ? =
There was a significant and positive correlation between buckling and the numbers of fall in
the patients with KOA (p = 0.02, ?= 0.87). The Pearson test showed a correlation between
buckling in patients and the results of functional tests. Thus, the results of functional tests in
patients with the knee buckling were worse than those without buckling, the values of p and
r for TUG test were p = 0.0001, r = -0.57 and for the 6MWT test were p = 0.0001, r = – 0.67.
Obviously, the direct or positive correlation means that if one of the variables increases
(decreases), the other also increases (decreases), and the inverse or negative correlation
shows that if one variable is increased, the other variable decreases and vice versa.
Furthermore, an inverse correlation was obtained between the knee buckling and total
WOMAC score (p = 0.02, r = -0.51) and between buckling and the WOMAC pain (p = 0.032,
r = -0.43) and function (p = 0.004, r = -0.43) subscales.
This present study was conducted for the first time in Iran community with the aim of
determining the frequency of the knee buckling and also attempted to designate the
correlation between the buckling with some characteristic in the patients with KOA. The
study was performed on 90 patients with KOA (grade 2 and grade 3 based on Kellgren &
Lawrence criteria), women were the most participants in this study.
The findings of the study indicated a 33.67% prevalence of the knee buckling among the patients. It should be noted that the buckling frequency in female patients was 33.85% and in males 44%.It is not possible to compare our results with other similar studies because no study has been conducted on this topic in Iran. Our findings indicate that a substantial number of individuals with KOA report episodes of giving way during daily activities and many of them considered the knee buckling as a limiting factor in their ability to do physical functions.
According to a study by Fitzgerald et al in Pennsylvania, US, the prevalence of the knee buckling in KOA patients was estimated at 63%(3).
Felson et al in a cohort study published in Massachusetts, US reported the incidence of the knee buckling in patients who experienced it once in the past three months was 11.8% and 78.1% in patients who experienced it more than once in the same period(2). Comparing the results of these two studies, the prevalence of buckling in the Felson et al study was far more than the prevalence of the current study, which is perhaps due to the study type; Felson et al study was a cohort study that lasted for three years, while the present study was a cross-sectional one which was conducted over a 12-month period.
Nguyen et al, in a two-year study, determined the incidence of buckling was 18% in the last three months in patients with KOA (9) which less than our study. The sample size of their study, is far more than ours, but these studies are more or less the same in the participants’ characteristics such as BMI, the history of KOAs, grade of osteoarthritis and pain severity. The reason for the difference in the prevalence of the two studies may be related to different ways of buckling assessments. Nguyen et al evaluated three options; a knee buckling, a knee instability sensation without buckling and a combination of knee buckling and instability, while in our study only the occurrence of buckling was considered.
The activities reported by the patients at the time of buckling according to the number of occurrences were, respectively, walking, going up and down stairs, twisting or turning and others in the present study exactly similar to the findings of Felson et al(2) , Fitzgerald et al (3)and Knoop et al(4) studies. Buckling, which is a sudden loss of postural support on the knee during weight bearing, usually occurs when weight bearing demands are increased, such as ascending or descending stairs. Many patients who have experienced repeated buckling often use a kind of support, such as holding a wall by hand, using the cane while walking or holding onto banisters when coming up or downstairs, to avoid falling and other serious consequences.
There were significant differences between VAS, functional tests and the WOMAC pain, function, and the overall scores between the patients with and without knee buckling. The findings of this study revealed a higher VAS, lower functional tests’ scores, and lower scores for pain and function subscales, as well as the overall score of the WOMAC in the patients with knee buckling.
It seems the sudden giving way can occur because of knee pain or insuf?cient muscle strength to support body weight (2-4, 9, 11). Despite of the findings of this study, which showed a positive correlation between increasing pain intensity and knee buckling, there was not any relation between the knee pain and buckling in the Felson et al’s(2). But, our finding correlated with a study done by van der Esch et al which demonstrated the buckling is strongly associated with increasing knee pain(17). Obviously, Knee pain has an inhibitory effect on the Quadriceps Femoris contraction and causes muscle atrophy and weakness over time(18, 19). Since the muscle has a great impact on knee joint stability; the development of Quadriceps’s weakness due to the KOA and its symptoms may lead to knee giving way(20).
Contrary to the findings of this study which indicated a significant association between buckling and female gender, there was no significant relationship between buckling and female gender in van der Esch et al study(17).It may be due to the weakness of the quadriceps femoris in the female of our study. According to the results of Slementa et al (21), Palmieri-Smith et al (22)and Segal et al(23), there was a significant decrease in Quadriceps Femoris strength in the women with KOA which has a significant impact on knee stability. But since the quadriceps strength has not been evaluated in the study, one cannot confirm the reason for and there may be another reason for the different results.
Felson et al also estimated the rate of falls was 12.6% in the last three months (2), while this rate was 39.39% in the patients over the same period which was almost three times more than the results of Felson et al. Some of the reasons for this difference are less age of the patients in Felson et al study, more muscular weakness or more severe pain of our participants or uneven levels of walking in our community. Fortunately, there was no serious lesion due to the falls requiring hospitalization in the study.
In the present study, all patients have experienced the knee buckling more than 2 times in the past three months and similar to the results of Felson et al (2) buckling was more common in knees with osteoarthritis in both tibiofemoral and patellofemoral joints.
The history of buckling in the right knee (36/36%) was higher than the left knee
(24.24%) in the patients, however, 39.39% of patients had a history of knee buckling on both
knees. In none of the existing studies, the prevalence of buckling has not been taken into
consideration based on the knee side.
Accordingly to the assumptions of the research, it seemed that there should be a correlation between the increase in BMI and knee buckling, while our results did not show such a correlation similarly to the findings of van der Esch et al (17). Therefore, the increase in BMI may be considered as a risk factor for the development of KOA(24, 25); but the results of the study did not reveal any positive correlation between obesity and knee buckling.
However, there was a significant correlation between knee buckling and history of KOA. The recent finding was consistent with Felson et al results(2). On the contrary, the findings of Schmitt et al’s study showed no significant relation between knee buckling with the history of osteoarthritis(7). Similarly to Schmitt et al’s study, van der Esch et al (17) did not find any significant correlation buckling and history of KOA.
It looks like a long history of KOA is often associated with more and more pain, increasing of muscle weakness, decreasing of dynamic stabilization and enhancing of knee buckling episodes.
One of the notable points in our findings was a positive correlation between knee buckling and increasing of fall rates in the patients. In the other words, the results of the present study confirmed that 39.39% of patients with knee buckling have mentioned the greater number of falls during the past three months. Since the most patients with KOA were elderly people, a fall can lead to fractures (especially femoral neck), soft tissue damage, bruising, subarachnoid hematoma, and so on. Moreover, the fall in the elderly could cause fear of falling, inability to work, loss of self-esteem, anxiety, depression, rejection, withdrawal, dependence, reduced of quality of life, etc. (26-28). Elderly people with KOA, due to decreased muscle strength, neuromuscular, coordination and balance dysfunction, are at greater risk of falling, which requires interventions to prevent falls in them(29, 30).
The patients with buckling have gained lower scores in the functional tests in the study which suggests that a reverse correlation between knee buckling and the results of TUG and Six MWT tests. In the same way, Nguyen et al. has obtained a positive correlation between buckling and impaired physical function of patients with KOA(9). Similarly, the findings of Sharma et al.(10)was in consistent with ours. This finding may be attributed to knee pain, reduction of muscle strength and endurance, proprioception disorders, Quadriceps Femoris atrophy as well as fear of falling (2, 11, 16, 18, 23, 26, 27, 29, 31, 32).
Also reverse correlations were revealed between buckling and the score of total WOMAC scores as well as WOMAC pain and function subscales in the study. This means the scores of the questionnaire and its subscales declined in the patient with buckling history, which was confirmed by the previous findings of Felson et al (2), Schmitt et al (7), Nguyen et al(9) , and Sharma et al (10). KOA symptoms such as pain, effusion, muscle weakness, function disability and joint instability were among the most common factors effective in reducing WOMAC questionnaire score.
The most important limitations of this study were its low sample size especially male
samples, lack of sufficient studies in Iran and the world to compare the results and also
failure to check all risk factors associated with the knee buckling. Therefore, conducting
similar studies with higher sample sizes, considering other factors related to the knee giving way, studying the knee buckling in athletes with secondary KOA and …could promote our knowledge in this field.
The findings demonstrated a 33.67% prevalence of knee buckling in the patients with KOA. There was also significant differences in VAS, functional tests results and the score of the WOMAC questionnaire and its subscale between the patients with and without history of knee buckling. Furthermore, there was a significant and positive correlation between knee buckling and female gender, history of KOA, and the fall in the patients.