Correlation between Kellgren-Lawrence classification of osteoarthritis and Knee Injury and Osteoarthritis Outcome Score

Abstract

BACKGROUND

Osteoarthritis (OA) is a common degenerative joint disease that considerably affects the quality of life (QoL) of individuals, especially the elderly. The Kellgren-Lawrence (KL) grading system assesses the severity of OA through radiographic evaluation, whereas the Knee Injury and Osteoarthritis Outcome Score (KOOS) measures clinical symptoms and functional status.

AIM

To analyze the relationship between KL grades and KOOS to elucidate the association between radiographic severity and clinical manifestations of knee OA.

METHODS

A cross-sectional study was performed at King Fahd Hospital of the University, involving 164 adult patients diagnosed with knee OA. Patients were evaluated using the KL system, based on standing knee X-rays performed within the last six months. The KOOS questionnaire was utilised for clinical assessment, evaluating five domains: (1) Pain; (2) Symptoms; (3) Activities of daily living; (4) Sport and recreation function; and (5) Knee-related QoL. Analysis of variance was utilised to examine variations in KOOS domain scores among different patient variables. The correlations between KOOS domains were determined using Pearson's correlation.

RESULTS

KOOS scores demonstrated a significant decline in patients with elevated KL grades, with Grade 4 exhibiting the lowest scores and grade 1 the highest (P < 0.001). The correlation between pain and activities of daily living was strong (r = 0.871, P < 0.001), as was the correlation with knee-related QoL (r = 0.754, P < 0.001). Notable age-related disparities were observed, as older patients (≥ 60 years) indicated poorer pain and functional outcomes. Gender differences were noted exclusively in symptoms, with females exhibiting lower scores than males (P = 0.022).

CONCLUSION

The association between radiographic severity and clinical impact is shown by the substantial correlation between KL grading and KOOS results, especially in later stages of OA. The results highlight the necessity for thorough assessments that integrate radiographic and clinical evaluations for individualized therapy choices for OA patients.

Key Words: Function level; Pain severity; Knee injury; Kellgren-Lawrence; Osteoarthritis

Core Tip: This cross-sectional study explores the relationship between radiographic severity and clinical outcomes in knee osteoarthritis using the Kellgren-Lawrence (KL) grading system and the Knee Injury and Osteoarthritis Outcome Score (KOOS). Findings show that patients who have scored grade 4 in the KL radiographic scale had significantly lower KOOS scores across all domains, especially pain and activities of daily living. Pain was strongly associated and observed with both functional limitations and quality of life. Age, gender, and obesity influenced the outcomes. These findings highlight the necessity of the integration of radiographic opinions and patient-reported outcomes to enhance OA management.

INTRODUCTION

Osteoarthritis (OA) is the most common type of degenerative joint disease marked by a breakdown of articular cartilage, remodeling of subchondral bone, and inflammation of the synovium, resulting in pain, stiffness, and diminished mobility[1]. Although younger populations can be affected by trauma or genetic predispositions, this form of arthritis mainly affects older persons and is the most common type of arthritis globally[2]. The complex etiology of OA comprises mechanical stress, metabolic factors, family history, and inflammation, all of which contribute to the progressive weakness of the joint[1,3]. Healthcare systems throughout the world are facing enormous problems due to the growing prevalence of OA, which is exacerbated by an ageing population, increased obesity rates, and sedentary lifestyles[4].

Knee OA, a special subtype, is especially complicated. OA is projected to impact approximately 250 million individuals worldwide, with knee OA being a primary cause of disability[5]. In Saudi Arabia, over 18.9% of individuals aged 18 and above are afflicted by OA, underscoring its extensive prevalence[6]. The problem is not solely a mechanical condition but also a biopsychosocial challenge, since it considerably affects quality of life (QoL) due to pain, decreased mobility, psychological discomfort, and diminished productivity[7]. The economic impact of OA is significant, including direct healthcare expenses such as treatments and long hospital stays, along with indirect expenditures like missed income and productivity[8].

The clinical picture of OA is complex and includes symptoms such as oedema, stiffness, joint pain, and crepitus[9,10]. The Kellgren-Lawrence (KL) grading system, which evaluates OA severity using radiographic indicators such as joint space narrowing, osteophyte formation, and sclerosis in standing knee X-rays, is widely utilised but has its drawbacks[11]. The KL system predominantly emphasises structural alterations in the joint and may inadequately reflect the patient's illness experience[11,12]. The disparity between radiographic findings and clinical manifestations continues to pose a considerable problem in the evaluation of OA[12].

Certain individuals with moderate radiographic alterations experience significant pain and functional limitations, whereas others with advanced radiographic OA display few symptoms[3,13]. This discrepancy complicates the diagnosis, management, and prognosis of the condition. The Knee Injury and Osteoarthritis Outcome Score (KOOS) is extensively utilised to evaluate the functional and clinical results of knee OA. It assesses pain, symptoms, daily living activities, sports and leisure pursuits, and knee-related QoL[14]. Correlating KOOS scores with KL grades may yield significant insights into the congruence–or incongruence – between the structural and symptomatic dimensions of the disease[15]. Comprehending this relationship is essential for enhancing patient management, as it highlights the necessity of a patient-centred strategy instead of depending exclusively on imaging results. This study sought to examine the relationship between the KL classification and KOOS scores in patients with knee OA, emphasising the significance of a patient-centred approach in the management of this condition.

MATERIALS AND METHODS

This research utilised a cross-sectional design and was carried out at a tertiary hospital. The target group consisted of adult patients diagnosed with OA. Participants were recruited during follow-up visits to the outpatient orthopaedics department at King Fahd Hospital of the University.

Inclusion criteria included adult patients with a confirmed diagnosis of OA who are actively following up in the orthopaedic outpatient clinics. Patients who had undergone previous knee surgery, received knee injections in the last two years, patients with cognitive impairment, and patients with X-rays older than six months were excluded from the study.

A consecutive sampling technique was used to recruit participants who met the inclusion criteria during the study period. The participants' standing knee X-rays, performed within the past six months, were reviewed and classified using the KL grading system. Data regarding the KOOS were gathered via a questionnaire, which included all sections of the KOOS to assess symptoms and functional limitations comprehensively.

A pilot study was conducted involving 15 OA patients from the outpatient department who were not part of the final sample. Data were collected through the KOOS questionnaire to evaluate the instrument's clarity, the feasibility of the study design, the completion time, and the methodology's overall appropriateness.

The sample size was calculated based on the findings of a previous study by Hulley et al[16] (correlation coefficient r = 0.217; α = 0.05; β = 0.2; Z_α = 1.96 at 95%CI and Z_β = 0.8416 for 80% power). Using these parameters, the total sample size was determined to be 164 cases.

Data were collected using the KOOS questionnaire to evaluate pain, symptoms, activities of daily living, sport and recreation function, and knee-related QoL. The KL grading scale was utilized to assess X-rays for OA severity. Patients were briefed on the study's aims and objectives, and informed consent was obtained before completing the questionnaire.

Before initiating the study, ethical approval was obtained from the Institutional Review Boards at the institute. Verbal consent was obtained from participants, ensuring that their data would be used exclusively for research purposes. Participant data were anonymized, securely stored, and discarded after the study's conclusion. Participation was voluntary, and participants were allowed to withdraw at any time without impacting their treatment or care.

Statistical analysis

The collected data were analyzed using the Statistical Package for the Social Sciences version 26. Descriptive statistics, including mean, standard deviation, and median, were used to summarize numerical variables, while frequencies and percentages described categorical variables. Student's t-test and analysis of variance were employed to compare the mean scores between different groups. Pearson's correlation coefficient was used to determine the correlation between the five outcomes of KOOS.

RESULTS

Our analysis included 164 OA patients, where 37.8% were aged ≥ 60 years and 29.9% were 50-59 years. About 62.2% were female patients, and 62.2% were obese. The KL grading system showed that 34.8% and 33.5% belonged to grade 4 and grade 3, respectively (Table 1).

Table 1 Sociodemographic details.

Sociodemographic details		Number	%
Age	< 18	4	2.4
	19-29	8	4.9
	30-39	12	7.3
	40-49	29	17.7
	50-59	49	29.9
	≥ 60	62	37.8
Gender	Female	102	62.2
	Male	62	37.8
Body mass index	Underweight	2	1.2
	Normal	18	11.0
	Overweight	42	25.6
	Obese	102	62.2
Grading on X-rays for osteoarthritis	1	9	5.5
	2	43	26.2
	3	55	33.5
	4	57	34.8

Data are presented as number (n) and percentage (%). Grading on X-rays is based on Kellgren-Lawrence classification for osteoarthritis.

The five outcomes (pain, symptoms, activities of daily living, sport and recreation function, and knee-related QoL) were measured on a 0-100 scale, with zero representing extreme knee problems and 100 representing no knee problems, as is common in orthopedic scales and generic measures. The mean pain score was found to be 47.6 ± 23.7. The mean symptom score was found to be 58.4 ± 16.3, the activities of daily living score was 55.8 ± 24.1, the sport and recreation function score was 39.6 ± 30.9, and the knee-related QoL score was 42.8 ± 22.3.

The mean pain score was found to be significantly severe in patients aged ≥ 60 years (38.40 ± 21.83 years) and 50-59 years (49.60 ± 23.06 years), when compared to the younger age group (P < 0.001). The symptoms score didn’t show any significant differences between different age groups (P = 0.573). The mean activities of daily living score were found to be significantly lower in patients aged ≥ 60 years (46.54 ± 22.09 years) and among 50-59 years (55.25 ± 24.08 years). At the same time, younger patients (< 18 years and 19-29 years) showed higher scores (P < 0.001). Also, older patients, ≥ 60 years and 50-59 years showed significantly lower scores for both sport and recreation function and the knee-related QoL compared to other age groups (P = 0.001; Table 2).

Table 2 Comparison of age of patients with different domains of Knee Injury and Osteoarthritis Outcome Score.

Domain scale (0-100)	Age	Number	Mean	SD	P value
Pain	< 18	4	75.69	7.98	< 0.001
	19-29	8	68.06	23.00
	30-39	12	53.94	26.34
	40-49	29	51.72	21.38
	50-59	49	49.60	23.06
	≥ 60	62	38.40	21.83
Symptoms	< 18	4	70.54	6.10	0.573
	19-29	8	64.73	7.50
	30-39	12	59.23	6.18
	40-49	29	58.00	17.57
	50-59	49	57.87	15.36
	≥ 60	62	57.20	18.66
Activities of daily living	< 18	4	86.40	10.36	< 0.001
	19-29	8	79.41	18.20
	30-39	12	67.65	25.81
	40-49	29	60.95	20.49
	50-59	49	55.25	24.08
	≥ 60	62	46.54	22.09
Sport and recreation function	< 18	4	70.00	20.00	< 0.001
	19-29	8	68.75	30.68
	30-39	12	55.00	30.90
	40-49	29	43.10	30.37
	50-59	49	38.27	31.55
	≥ 60	62	30.32	26.96
Knee-related quality of life	< 18	4	56.25	13.50	< 0.001
	19-29	8	66.41	30.42
	30-39	12	52.60	25.76
	40-49	29	46.34	23.24
	50-59	49	44.01	19.97
	≥ 60	62	34.38	18.55

The age wise comparison of five outcome scale is shown in Table 2. Scores range from 0 (worst) to 100 (best). Statistical analysis was performed using one-way analysis of variance.

It was found that there were no statistically significant differences observed for the four domains’ scores, pain, activities of daily living, sport and recreation function, and knee-related QoL (P > 0.05). However, the symptoms score was found to be lower in female patients (56.1 ± 16.8) compared to males (62.1 ± 15.6) (P = 0.022; Table 3).

Table 3 Comparison of patients’ gender with different domains of Knee Injury and Osteoarthritis Outcome Score.

Domain scale (0-100)	Gender	Number	Mean	SD	P value
Pain	Female	102	46.4869	23.77010	0.445
	Male	62	49.4176	23.76337
Symptoms	Female	102	56.1275	16.87450	0.022
	Male	62	62.0968	14.60926
Activities of daily living	Female	102	54.2532	23.47466	0.291
	Male	62	58.3729	25.28356
Sport and recreation function	Female	102	37.8922	31.57920	0.364
	Male	62	42.4194	29.70817
Knee-related quality of life	Female	102	41.1152	21.81855	0.216
	Male	62	45.5645	22.87000

Scores range from 0 (worst) to 100 (best). Independent sample t-test was used for comparison.

We found that obese patients showed significantly lower scores, and patients who were normal and underweight showed higher scores for all five domains (P < 0.05; Figure 1A).

Figure 1 Comparison of Knee Injury and Osteoarthritis Outcome Score based on body mass index and the distribution of Knee Injury and Osteoarthritis Outcome Score across different Kellgren-Lawrence grades. A: Comparison of Knee Injury and Osteoarthritis Outcome Score (KOOS) based on body mass index; B: The distribution of KOOS scores across different Kellgren-Lawrence grades. KL: Kellgren-Lawrence; KOOS: Knee Injury and Osteoarthritis Outcome Score.

The KL grading comparison revealed that all five outcomes had significantly lower scores for grade 4, with scores eventually increasing as the grade decreased. Notably, grade 1 showed comparatively significantly higher scores (P < 0.001; Figure 1B).

Pain outcome showed weak correlations with symptoms (r = 0.363, P < 0.001), strong correlation with activities of daily living (r = 0.871, P < 0.001), moderate correlation with sport and recreation function (r = 0.688, P < 0.001), and strong correlation with knee-related QoL (r = 0.754, P < 0.001). Symptoms were weakly correlated with activities of daily living (r = 0.396, P < 0.001), sport and recreation function (r = 0.298, P < 0.001), and moderately correlated with knee-related QoL (r = 0.474, P < 0.001). Activities of daily living showed strong correlations with sport and recreation function (r = 0.752, P < 0.001) and knee-related QoL (r = 0.755, P < 0.001). Lastly, sport and recreation function had a strong positive correlation with knee-related QoL (r = 0.709, P < 0.001; Table 4).

Table 4 Correlation of scores of five outcomes.

		Symptoms	Activities of daily living	Sport and recreation function	Knee-related quality of life
Pain	Rho	0.363	0.871	0.688	0.754
	P value	< 0.001	< 0.001	< 0.001	< 0.001
Symptoms	Rho	-	0.396	0.298	0.474
	P value		< 0.001	< 0.001	< 0.001
Activities of daily living	Rho	-	-	0.752	0.755
	P value			< 0.001	< 0.001
Sport and recreation function	Rho	-	-	-	0.709
	P value				< 0.001

Correlations calculated using Pearson’s correlation coefficient. All P values are two-tailed and statistically significant at P < 0.001.

Logistic regression analysis revealed that age, body mass index (BMI), and KL grading were significant predictors of severe KOOS scores (< 40). Patients aged 50-59 years had more than twice the odds of severe KOOS compared to those aged ≤ 18 years [odds ratio (OR) = 2.43, P < 0.001], while other age groups did not show statistically significant associations. Obese individuals were nearly three times more likely to have severe KOOS scores compared to those with normal BMI (OR = 2.79, P = 0.003), whereas underweight and overweight categories were not significant. Radiographic severity showed a strong stepwise association, with grade 3 (OR = 4.32, P = 0.002) and grade 4 (OR = 7.88, P < 0.001) significantly increasing the odds of severe KOOS relative to grade 1. Gender was not significantly associated with severe KOOS in this model (Table 5).

Table 5 Logistic regression to predict factors related to severe Knee Injury and Osteoarthritis Outcome Score (< 40).

	Predictor variable	Odds ratio	95%CI	P value
Age	19-29	1.38	0.45-4.20	0.769
	40-49	1.94	0.88-4.28	0.574
	50-59	2.43	1.15-5.12	< 0.001
	≥ 60	1.81	0.79-3.09	0.122
	≤ 18	Reference
Gender	Female	1.04	1.03-2.95	0.339
	Male	Reference
Body mass index	Underweight	2.16	0.29-15.91	0.455
	Overweight	1.36	0.65-2.87	0.415
	Obese	2.79	1.41-5.51	0.003
	Normal	Reference
Kellgren-Lawrence grading	Grade 2	2.11	0.85-5.27	0.107
	Grade 3	4.32	1.73-10.78	0.002
	Grade 4	7.88	3.10-20.04	< 0.001
	Grade 1	Reference

DISCUSSION

The KOOS is a patient-reported outcome measure designed to capture an individual’s perception of knee function, pain, and symptoms[17]. Given that pain in OA is inherently subjective, patient-reported scores can vary considerably even in individuals with similar structural findings[10]. In this study, OA patients reported a mean pain score of 47.6, consistent with the chronic, often persistent nature of OA pain[18]. The mean symptom score was slightly higher at 58.4, reflecting the contribution of stiffness, oedema, and crepitus to the disease burden. These symptoms can interfere with daily life and may coexist with pain as drivers of disability. The Activities of Daily Living Score of 55.8 underscores the impact of OA on self-care and functional independence, including tasks such as walking, dressing, and bathing[19,20]. Sport and recreation function had the lowest subscale mean score (39.6), illustrating the pronounced limitation in physical activity, though it should be noted that this domain may be less relevant to some patients, depending on lifestyle and baseline activity levels[21,22]. The QoL subscale mean of 42.8 reflects the cumulative impact of pain, mobility restriction, and functional decline on broader aspects of life over time.

The KL classification system is a widely accepted radiographic grading method for OA severity, ranging from grade 0 (no radiographic features) to grade 4 (severe OA with marked joint space narrowing, osteophytes, and bone sclerosis)[23]. This method has been validated with good inter-observer and intra-observer reliability, particularly for the tibiofemoral joint[24,25]. In our analysis, all KOOS subscales were significantly lower in patients with KL grade 4 OA, showing a progressive decline in scores with increasing radiographic severity. This finding is in line with prior studies that have reported moderate to strong negative correlations between KL grade and KOOS scores[26-28]. These results reinforce the construct validity of both tools: KL grading reflects structural joint changes visible on radiographs, while KOOS captures the patient’s subjective experience of pain, symptoms, and functional impairment. However, the observed association between higher KL grades and lower KOOS scores is not absolute, and this “structure – symptom discordance” warrants further discussion. In clinical practice, some patients with severe radiographic OA report minimal symptoms, while others with mild to moderate OA on imaging experience profound pain and disability. Several mechanisms may account for this discrepancy. Neuropathic pain mechanisms, including central sensitization, may amplify pain perception independently of structural severity, leading to disproportionately severe symptoms in some individuals. Bone marrow lesions (BMLs), which are detectable on magnetic resonance imaging (MRI) but not on plain radiographs, have been shown to correlate strongly with pain severity and may explain cases where radiographic changes underestimate symptom burden. Similarly, synovitis – a marker of active joint inflammation – has been associated with pain and functional decline regardless of radiographic stage. These pathophysiological processes may act independently or synergistically with cartilage loss and osteophyte formation, influencing KOOS outcomes beyond what KL grading alone can predict.

This interplay between structural and non-structural factors underscores the importance of integrating both objective radiographic findings and subjective patient-reported measures in OA assessment. A purely radiographic approach risks underestimating disease impact in patients with significant symptoms but minimal visible joint changes, while relying solely on patient reports may overlook progressive joint damage that could inform treatment planning.

From a surgical perspective, the relationship between pre-operative OA severity and post-total knee arthroplasty (TKA) outcomes remains contentious. Scott et al[29] reported no association between KL grade and patient-reported outcomes following TKA, whereas Rehman et al[30] observed a significant relationship. These conflicting findings highlight the need for future prospective studies with larger, more diverse cohorts that also incorporate advanced imaging (e.g., MRI for BMLs and synovitis) and pain phenotyping to better understand predictors of both pre-operative and post-operative function.

CONCLUSION

The findings of this study demonstrate a significant association between radiographic severity and clinical outcomes in knee OA. Patients with higher KL grades exhibited markedly lower KOOS scores across all domains, particularly in pain and activities of daily living, highlighting the escalating clinical burden in advanced OA. Strong correlations were observed between pain and activities of daily living and knee-related QoL. Older patients (≥ 60 years) experienced poorer pain and functional outcomes, while females exhibited lower symptom scores compared to males. Future research should explore interventions tailored to patients with advanced OA and investigate the role of additional biomarkers in predicting disease progression. More longitudinal studies are recommended to validate these findings and evaluate the impact of targeted treatments on long-term outcomes.