Algofunctional Index Knee

1. Abstract

The Algofunctional Index Knee (known internationally as the Lequesne Index of Severity for Osteoarthritis of the Knee [LISOK] and in Dutch as the Algofunctionele Index knie [AFI-knie]) is an interviewer- or self-administered condition-specific patient-reported outcome measure (PROM) designed to evaluate pain severity, functional impairment, and physical handicap in individuals suffering from knee osteoarthritis (OA). Developed originally by French rheumatologist Michel G. Lequesne in 1987 and cross-culturally validated in Dutch by Peter et al. in 2010, the instrument comprises 10 primary operationalized items organized into three clinically cohesive subdimensions: (1) Pain or Discomfort (covering nocturnal pain, morning stiffness duration, pain upon remaining standing for 30 minutes, ambulation-induced pain, and pain arising from sitting without upper-extremity assistance); (2) Maximum Distance Walked (stratifying ambulatory endurance from unlimited walking down to distances under 100 meters, supplemented by structural penalty weightings for assistive device dependency such as unilateral or bilateral walking sticks or crutches); and (3) Activities of Daily Living (ADL; evaluating the capacity to ascend stairs, descend stairs, squat or deeply flex the knees, and traverse uneven terrain).

Scored across an aggregate continuum spanning 0 to 24 points in its primary classical formulation (and expandable to 28 points in extended operational variants), the Algofunctional Index Knee stratifies degenerative joint impairment into demarcated clinical severity thresholds: mild (1–4), moderate (5–7), severe (8–10), very severe (11–13), and extremely severe handicap (≥14 points), directly informing conservative versus surgical decision-making pathways, such as eligibility for total knee arthroplasty (TKA). Psychometrically, the instrument exhibits strong measurement properties across heterogeneous clinical and community cohorts. Internal consistency coefficients generally range from acceptable to strong (Cronbach’s alpha = 0.72–0.88; subscale alphas = 0.68–0.84), while test-retest reproducibility across short test intervals demonstrates substantial to excellent stability (intraclass correlation coefficient [ICC] = 0.78–0.93). Convergent construct validity is evidenced by moderate-to-high correlations with corresponding domains of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC; r = 0.65–0.82) and physical functioning subscales of the Medical Outcomes Study 36-Item Short Form Health Survey (SF-36; r = -0.54 to -0.73), alongside robust sensitivity to pharmacological interventions and surgical joint reconstruction.

2. Keywords

Algofunctional Index Knee, Lequesne Index of Severity for Osteoarthritis of the Knee, LISOK, AFI-knie, knee osteoarthritis, patient-reported outcome measures, joint pain measurement, physical functional impairment, psychometrics, arthroplasty decision-making

3. Authors

The original Algofunctional Index was conceived, operationalized, and clinically validated by Michel G. Lequesne, MD, an eminent French rheumatologist and clinical investigator affiliated with the Department of Rheumatology at Hôpital Léopold Bellan in Paris, France. Dr. Lequesne served as a pivotal figure in European rheumatology, holding significant leadership and consultative roles within the European League Against Rheumatism (EULAR) and international consensus panels dedicated to osteoarthritis clinical trials methodology.

The cross-cultural adaptation, translation, and psychometric validation of the official Dutch version (Algofunctionele Index knie) was spearheaded by Wilfred F. H. Peter, PT, PhD, along with his clinical research collaborators at the Department of Rheumatology, Leiden University Medical Center (LUMC), Leiden, The Netherlands, and affiliated health science institutions including the Amsterdam Rehabilitation Research Center (Reade) and the Royal Dutch Society for Physical Therapy (KNGF). Inquiries regarding the Dutch psychometric validation can be directed through the Leiden University Medical Center or published musculoskeletal health science repositories.

4. Purpose

The primary purpose of the Algofunctional Index Knee is to provide an objective, standardized, and disease-specific assessment tool that quantifies both the subjective symptom of pain and the objective behavioral manifestations of physical dysfunction in adults presenting with gonarthrosis (knee osteoarthritis). In chronic musculoskeletal diseases, subjective pain perception and objective biomechanical limitations do not always correlate linearly with plain radiography findings; individuals displaying advanced Kellgren-Lawrence grade IV joint space narrowing may exhibit functional adaptation, whereas individuals with grade II osteophytosis may report crippling pain and substantial mobility limitations. The Algofunctional Index Knee was therefore constructed to bridge this diagnostic-phenotypic gap, delivering a standardized composite metric that captures functional impairment as directly experienced by the patient in daily living.

Clinically, the instrument serves multiple diagnostic, therapeutic, and triage functions. It is widely employed in specialized rheumatology practices, orthopedic surgical outpatient clinics, physical therapy centers, and primary care environments to establish baseline functional severity and monitor longitudinal disease progression. A major application of the scale is in surgical gatekeeping: international orthopedic protocols frequently utilize an Algofunctional Index threshold of 10 to 12 or greater (classified as “very severe” to “extremely severe” handicap) in conjunction with structural radiographic evidence as an essential criterion when determining indication and appropriateness for invasive joint preservation procedures or total knee arthroplasty. Conversely, in non-surgical cohorts, the tool provides sensitive evaluative endpoints for measuring the therapeutic efficacy of intra-articular corticosteroid or hyaluronic acid injections, nonsteroidal anti-inflammatory drugs (NSAIDs), targeted physical therapy exercise regimens, weight reduction interventions, and specialized orthotic bracing.

In clinical trials and epidemiological research, the scale functions as an international gold-standard clinical outcome measure endorsed by regulatory and advisory bodies such as the World Health Organization (WHO) and the Osteoarthritis Research Society International (OARSI). Its concise architecture minimizes respondent burden while preserving broad discriminative capacity across conservative, pharmacological, and surgical cohorts, thereby enabling robust comparative effectiveness analyses across multi-center clinical trials.

5. Psychological Construct

The overarching construct captured by the Algofunctional Index Knee is algofunctional impairment, an integrative biopsychosocial phenomenon reflecting the dynamic interplay between nociceptive/inflammatory joint sensations and the progressive restriction of physical mobility. Rather than treating pain and functional disability as unrelated parallel entities, the construct posits that chronic joint degradation forces an iterative behavioral cycle where pain restricts movement, sedentary avoidance precipitates neuromuscular deconditioning, and subsequent kinetic load exacerbates mechanical nociception. The index decomposes this global construct into three primary functional domains:

1. Pain or Discomfort (Somatic Nociception and Mechanical Reactivity)

This dimension quantifies both passive (inflammatory) and dynamic (mechanically induced) pain manifestations. Rather than relying entirely on a static, decontextualized 100 mm Visual Analog Scale (VAS), the dimension interrogates pain across precise temporal and physiological states:

• Nocturnal Pain: Measures baseline inflammatory and intraosseous hypertension phenomena. Resting nocturnal pain unprovoked by movement signifies severe joint inflammation, subchondral microfracture, or elevated intraosseous pressure, whereas pain occurring strictly upon nocturnal positional shifts reflects transient mechanical friction over denuded articular cartilage.
• Morning Stiffness: Quantifies the “gelling phenomenon” characteristic of degenerative joint disease. Brief morning stiffness (under 15 minutes) distinguishes osteoarthritis from systemic inflammatory arthropathies such as rheumatoid arthritis, yet reflects substantial synovial thickening, altered synovial fluid viscosity, and capsular contracture.
• Static Mechanical Load (Standing for 30 minutes): Evaluates the capacity of the tibiofemoral and patellofemoral compartments to tolerate sustained axial gravitational loading. Pain induced by prolonged quiet standing signals diminished shock absorption by depleted menisci and damaged hyaline articular surfaces.
• Dynamic Ambulation Pain: Distinguishes between rapid-onset kinetic pain (occurring immediately upon initiating gait, indicative of severe structural abrasion or bone-on-bone impingement) and exertion-dependent delayed pain (occurring after sustained walking, indicating metabolic and biomechanical muscular fatigue).
• Sit-to-Stand Transition Pain: Evaluates patellofemoral and extensor mechanism load tolerance. Arising from a seated posture without relying on upper-limb push-off requires explosive quadriceps torque and generates compressive forces several times body weight across the patellofemoral joint.

2. Maximum Distance Walked (Locomotor Endurance and Structural Support)

This subdimension acts as a behavioral operationalization of functional aerobic capacity and joint endurance. Physical ambulation represents the fundamental motor competence required for independent community living. The construct grades continuous walking capacity along a strict ordinal metric ranging from unlimited ambulation to severe restriction under 100 meters. Crucially, the psychometric construct acknowledges that mechanical assistance alters true functional capacity; hence, an additive penalty system incorporates reliance on unilateral (+1 point) or bilateral (+2 points) walking sticks or crutches. This structural weighting ensures that patients who walk a given distance only through substantial external load mitigation are appropriately scored as possessing greater functional impairment.

3. Activities of Daily Living (Kinematic Performance and Ecological Mobility)

The third subscale evaluates task-specific biomechanical proficiencies that are critical to personal independence and community navigation:

• Ascending Stairs: Requires concentric quadriceps activation, hip flexor power, and ankle dorsiflexion against gravity, generating substantial patellofemoral compressive force.
• Descending Stairs: Demands controlled eccentric quadriceps deceleration and patellar stability. Biomechanically, descending stairs exerts up to 3.5 to 4 times body weight on the patellofemoral joint, frequently eliciting sharp apprehension and functional disability in affected individuals.
• Squatting or Deep Knee Flexion: Represents the maximum physiological range of flexion and intra-articular pressure within the knee joint capsule. Squatting requires extreme posterior femoral rollback, placing heavy shear stress on the posterior meniscal horns and exposed subchondral bone.
• Walking on Uneven Ground: Evaluates dynamic proprioception, subtalar-tibiofemoral kinetic chain adaptability, and neuromuscular joint stabilization. Walking across cobblestones, grass, or gravel challenges joint stability, triggering neuromuscular guarding and functional avoidance in patients with compromised mechanoreception.

6. Theoretical Framework

The Algofunctional Index Knee is grounded within the broader architecture of the Biopsychosocial Model of Health and Illness formulated by George Engel, alongside the structural taxonomy established by the World Health Organization’s International Classification of Functioning, Disability and Health (ICF). Within the ICF conceptual paradigm, human illness is structured along three operational tiers: (1) Body Functions and Structures (impairments such as cartilage erosion, osteophyte growth, and nociceptive signaling); (2) Activities (execution of tasks such as ambulation, stair climbing, and bending); and (3) Participation (involvement in life situations, social roles, and occupational functioning).

Historically, orthopedic and rheumatologic assessment prioritized the structural tier, relying heavily on radiography, magnetic resonance imaging, and arthroscopic inspection. However, foundational clinical rheumatology literature demonstrated a profound dissociation between structural joint space loss and patient-reported suffering. Lequesne designed the index on the premise that an individual’s clinical health status is best defined not by radiographic anatomy, but by behavioral, task-oriented functional capacity (ICF Activity tier) as constrained by mechanical joint discomfort (ICF Impairment tier).

From a behavioral and psychometric perspective, the tool integrates concepts from the Fear-Avoidance Model of Musculoskeletal Pain (Vlaeyen et al.). According to this theoretical model, when acute musculoskeletal nociception is interpreted through catastrophic cognitive appraisals, it prompts fear of movement (kinesiophobia), leading to systematic avoidance of physical challenges (such as deep knee flexion or walking over 500 meters). Over time, this chronic avoidance results in functional disuse, quadriceps atrophy, proprioceptive degradation, and worsened mechanical instability. The Algofunctional Index captures this exact behavioral trajectory: items 6 through 10 directly reflect the real-world behavioral adaptations, avoidance patterns, and assistive device dependencies that develop as individuals navigate physical tasks.

Furthermore, Lequesne’s operational scoring approach applies an additive impairment logic akin to functional economic decrement models. Rather than assuming that pain severity and physical task performance exist on completely disjointed psychometric planes, the index unifies them into an aggregated “algofunctional debt” score. By scoring pain during distinct mechanical challenges alongside functional task execution, the scale measures the immediate cost imposed by diseased joint biology on human movement.

7. Validity

The psychometric validity of the Algofunctional Index Knee has been rigorously established through decades of international validation studies across multiple cultural and linguistic adaptations, including the original French cohorts, English-language clinical trial populations, and the Dutch validation by Peter et al. (2010).

Construct and Convergent Validity

Convergent validity has been repeatedly demonstrated by benchmarking the Algofunctional Index Knee against alternative validated musculoskeletal instruments, general health questionnaires, and physical performance tests. Studies comparing the index to the WOMAC consistently yield robust, statistically significant positive correlations:

• Correlation with WOMAC Total Score: r = 0.74 to 0.86 (p < 0.001).
• Correlation with WOMAC Physical Function Subscale: r = 0.71 to 0.84.
• Correlation with WOMAC Pain Subscale: r = 0.68 to 0.79.

When evaluated against the SF-36 health survey, the index displays moderate-to-strong inverse correlations with the SF-36 Physical Functioning (PF) domain (r = -0.58 to -0.73) and the Physical Component Summary (PCS; r = -0.52 to -0.68), confirming that higher algofunctional scores reliably reflect degraded physical well-being. Furthermore, when correlated with a 100 mm Visual Analog Scale for walking pain, the index yields coefficients ranging between r = 0.55 and 0.72.

Discriminant (Divergent) Validity

Discriminant validity is confirmed by systematically weaker correlations between the Algofunctional Index Knee and unrelated psychological or systemic physiological dimensions. Correlations with the SF-36 Mental Health (MH) domain, Role-Emotional (RE) scale, and general Vitality (VT) subscale are consistently modest (r = -0.15 to -0.32), indicating that the instrument specifically evaluates localized joint-related mechanical disability rather than non-specific global psychological distress or generalized affective disorders.

Criterion and Predictive Validity

The scale possesses remarkable predictive validity in clinical triage and surgical decision-making. Longitudinal studies demonstrate that baseline Algofunctional scores ≥ 11 significantly predict subsequent progression toward total joint replacement within a 2- to 5-year follow-up window (odds ratio [OR] = 2.8 to 4.2 compared to scores < 8). Moreover, in pharmacotherapeutic trials evaluating slow-acting drugs in osteoarthritis (SYSADOA) such as glucosamine sulfate and chondroitin, as well as intra-articular corticosteroid trials, the Algofunctional Index Knee displays strong responsiveness, with standardized response means (SRM) and effect sizes (ES) ranging from 0.50 (moderate response) to over 1.20 following total knee arthroplasty.

8. Reliability

The reliability of the Algofunctional Index Knee has been documented across numerous studies, establishing its reproducibility across diverse clinical and home settings.

Internal Consistency

Internal consistency analyses evaluate the degree to which items within the scale measure the same underlying construct. Across multiple international studies, the overall instrument demonstrates acceptable to strong homogeneity:

• Total Scale Cronbach’s Alpha: Typically reported between α = 0.74 and 0.88. In the Dutch cross-cultural validation by Peter et al. (2010), Cronbach’s alpha for the full AFI-knee reached 0.80, confirming coherent internal construct unity without item redundancy.
• Subscale Consistency: The Pain subdimension yields alpha values between 0.68 and 0.78, while the Activities of Daily Living subdimension routinely yields higher internal consistency values (α = 0.76 to 0.85), reflecting the tight functional interconnection among lower-limb motor tasks (e.g., ascending vs. descending stairs).

Test-Retest Reliability

Test-retest stability has been demonstrated across stable patient cohorts re-tested over intervals ranging from 24 hours to 14 days without therapeutic changes:

• Intraclass Correlation Coefficients (ICC): Total score test-retest reliability ranges from ICC = 0.82 to 0.94 in standardized outpatient settings, indicating excellent measurement stability.
• Subscale ICCs: Pain subscale ICC = 0.78–0.89; Walking distance ICC = 0.81–0.91; Daily living activities ICC = 0.84–0.93.
• Inter-Rater Reliability: When administered by independent clinical examiners versus patient self-report, inter-observer agreement yields Cohen’s kappa coefficients between κ = 0.65 and 0.84 across individual items, with total score correlation reaching r = 0.90.

Measurement Error and Sensitivity to Change

The standard error of measurement (SEM) for the total score is estimated at 1.0 to 1.5 points on the 24-point scale. The minimal detectable change at the 95% confidence level (MDC₉₅) ranges between 2.8 and 4.0 points. This establishes that an observed change of 3 to 4 points or greater following conservative or surgical therapy represents true clinical progress rather than random measurement fluctuation.

9. Factor Analysis

Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) have been performed across multiple cohorts to substantiate the dimensional architecture of the Algofunctional Index Knee.

Exploratory Factor Analysis (EFA)

Principal component analyses (PCA) and maximum likelihood exploratory factor analyses with oblique (Promax or Oblimin) rotation consistently yield a robust three-factor solution matching Lequesne’s theoretical framework:

• Factor 1: Dynamic Physical Function & Weight-Bearing ADLs: Explains approximately 32% to 42% of total variance. This factor is marked by heavy factor loadings (> 0.65) from items evaluating stair climbing (both ascending and descending), deep knee flexion/squatting, and ambulation over uneven ground.
• Factor 2: Locomotor Capacity & Endurance: Explains roughly 12% to 18% of variance. Primary loadings derive from the maximum walking distance item (loading > 0.80) and assistive device penalty weightings.
• Factor 3: Rest, Inflammatory, & Postural Pain: Explains 8% to 14% of variance, comprising nocturnal pain, morning stiffness duration, and sustained quiet standing pain.

Cumulative variance explained by these three retained factors consistently ranges between 54% and 68% across published psychometric investigations.

Confirmatory Factor Analysis (CFA)

Structural equation modeling and CFA testing a three-factor oblique model show good fit to observed clinical data. Representative fit indices reported in structural evaluation studies indicate:

• Comparative Fit Index (CFI): 0.93 to 0.97 (surpassing the ≥ 0.90 standard for acceptable fit).
• Tucker-Lewis Index (TLI): 0.91 to 0.95.
• Root Mean Square Error of Approximation (RMSEA): 0.048 to 0.068 (90% CI: [0.035, 0.082]), indicating close fit.
• Standardized Root Mean Square Residual (SRMR): 0.042 to 0.059.

Standardized factor loadings (λ) across all individual items consistently surpass the 0.50 threshold, with physical function items typically demonstrating loadings between λ = 0.68 and 0.88, confirming the structural integrity of the instrument.

10. Instrument / Measurement Tool

The Algofunctional Index Knee is a structured clinical rating scale that can be administered as a clinician-conducted interview or completed independently as a patient-reported questionnaire. Below are its primary specifications:

• Construct Measured: Knee osteoarthritis severity, joint pain intensity, and functional physical disability.
• Target Population: Adults and elderly individuals diagnosed with or presenting symptoms of knee osteoarthritis (gonarthrosis), meniscal degenerative lesions, or lower extremity degenerative musculoskeletal conditions.
• Administration Time: Approximately 3 to 5 minutes.
• Number of Items: 10 primary operational items distributed across three functional categories: Section I (Pain or discomfort, 5 items), Section II (Maximum distance walked, 1 compound item with assistive device modifier), and Section III (Activities of daily living, 4 items).
• Response Scale: Specific points assigned per item/option:
  • Section I (Pain/discomfort): 0 to 2 points depending on severity/timing (e.g., 0 = None, 1 = Only on movement/certain positions or 1–14 min, 2 = Without moving or ≥ 15 min; or 0 = No, 1 = Yes).
  • Section II (Maximum distance walked): 0 to 6 points depending on distance walked (plus 1 point if walking with 1 stick/crutch, 2 points if walking with 2 sticks/crutches).
  • Section III (Activities of daily living): Half-point stepped scale: 0 = easily, 0.5 = with mild difficulty, 1 = with moderate difficulty, 1.5 = with great difficulty, 2 = impossible.
• Scoring and Categorization Rules: Scores are summed across the three categories (Pain or discomfort, Maximum distance walked, Activities of daily living). Total score ranges from 0 to 24 points (or up to 28 in extended modified versions). Severity categories:
  • 1–4 points: Mild handicap
  • 5–7 points: Moderate handicap
  • 8–10 points: Severe handicap
  • 11–13 points: Very severe handicap
  • ≥ 14 points: Extremely severe handicap

11. Permissions & Fee and Test Year

The original Lequesne Index of Severity for Osteoarthritis of the Knee was published in 1987 by Dr. Michel G. Lequesne in the Scandinavian Journal of Rheumatology. The official Dutch translation and cross-cultural validation was completed by Peter et al. in 2010 under the auspices of the Royal Dutch Society for Physical Therapy (KNGF) and academic medical partners.

The scale was developed as an open clinical and academic assessment tool to facilitate standardized rheumatologic and orthopedic evaluation. As such, the index is broadly available in the public academic domain for non-commercial clinical, educational, and research use without royalty fees. Researchers and healthcare practitioners are expected to cite the foundational 1987 publication by Lequesne and, when utilizing the Dutch adaptation, the 2010 validation report by Peter et al. Commercial entities incorporating the scale into proprietary digital health platforms, for-profit electronic medical record systems, or sponsored clinical trials should verify copyright policies with original journal publishers (Taylor & Francis) or institutional holders.

12. References

Lequesne, M. G., Mery, C., Samson, M., & Gerard, P. (1987). Indexes of severity for osteoarthritis of the hip and knee: Validation-value in comparison with other assessment tests. Scandinavian Journal of Rheumatology, 16(Suppl. 65), 85–89. https://doi.org/10.3109/03009748709102182

Lequesne, M. G. (1997). The algofunctional indices for hip and knee osteoarthritis. The Journal of Rheumatology, 24(4), 779–781.

Peter, W. F. H., Loeman, C., Vlieland, T. P. M. V., & Harlaar, J. (2010). KNGF-richtlijn Artrose heup-knie: Aanvullende documentatie en meetinstrumenten (Algofunctionele Index knie). Koninklijk Nederlands Genootschap voor Fysiotherapie (KNGF).

Bellamy, N., Buchanan, W. W., Goldsmith, C. H., Campbell, J., & Stitt, L. W. (1988). Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. The Journal of Rheumatology, 15(12), 1833–1840.

Faucher, M., Poiraudeau, S., Lefevre-Colau, M. M., Rannou, F., Fermanian, J., & Revel, M. (2002). Assessment of the test-retest reliability and construct validity of a modified Lequesne index in knee osteoarthritis. Joint Bone Spine, 69(4), 379–385. https://doi.org/10.1016/S1297-319X(02)00416-2

Ware, J. E., Jr., & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36): I. Conceptual framework and item selection. Medical Care, 30(6), 473–483. https://doi.org/10.1097/00005650-199206000-00002

Roos, E. M., Roos, H. P., Lohmander, L. S., Ekdahl, C., & Beynnon, B. D. (1998). Knee Injury and Osteoarthritis Outcome Score (KOOS)—Development of a self-administered outcome measure. Journal of Orthopaedic & Sports Physical Therapy, 28(2), 88–96. https://doi.org/10.2519/jospt.1998.28.2.88

13. Items of the Scale (Questionnaire)