Outerbridge

In 1961, Outerbridge²⁹ introduced a system for grading articular cartilage lesions in the patellofemoral joint. The four-gradient scale was based on the appearance of the cartilage and, in two of the levels, the size of the lesion. A grade 1 lesion was defined as softening and swelling of the cartilage. A grade 2 lesion involved fissuring and fragmentation in an area less than ½” in diameter. A grade 3 lesion involved fissuring and fragmentation in an area greater than ½” in diameter. In a grade 4 lesion, the cartilage was eroded down to subchondral bone. This rating system did not specify the location, depth, or exact size of the lesion.

Insall

Insall and coworkers¹⁸ proposed a classification system for patellofemoral articular cartilage lesions that was composed of four stages based on the cartilage appearance. The first stage consisted of swelling, softening, and blistering of the cartilage. The second stage included lesions with deep fissures that extended to subchondral bone. The third stage involved fibrillation and a “crabmeat” appearance. In the fourth stage, thinning of the cartilage and exposure of subchondral bone in some areas were noted, with the remainder of the cartilage demonstrating a “coarse granular appearance.” This rating system also failed to specify the location, depth, and size of the lesion.

Goodfellow

Goodfellow and associates¹⁵ described two different types of disorders of articular cartilage in the patellofemoral compartment: age-dependent surface degeneration and basal degeneration. Age-dependent surface degeneration was usually found on the odd facet, was often present in middle-aged patients, and became even more frequent with increasing age. These lesions involved surface flaking that progressed to fibrillation and, finally, subchondral bone exposure. The authors believed these lesions were asymptomatic in younger patients, but could become painful as degenerative joint disease in later years.

Basal degeneration involved fasciculation of collagen in the middle and deep zones (usually without involvement of the surface layer) on the crest of the ridge between the medial and the odd facets. Three stages of this type of degeneration were described. In the first stage (fasciculation 1), the articular surface was smooth and intact with a spongy consistency, sometimes exhibiting “pitting edema.” The second stage involved blistering of the cartilage. In the third stage (fasciculation 2), rupture of tangential surface fibers was noted.

Casscells

Casscells⁸ used a rating system to describe lesions located on the undersurface of the patella and femoral condyles in an anatomic cadaveric investigation. The five-level gradient classification included both cartilage appearance and size of the lesion. Grade 0 indicated normal surfaces with no gross lesions present. Grade 1 included small, superficial areas of cartilage erosion, 1 cm or less in diameter. Grade 2 lesions were 1 to 2 cm in diameter and involved the “deeper layers” of cartilage. Grade 3 indicated lesions that were 2 to 4 cm in diameter in which the cartilage was completely eroded and subchondral bone was exposed. Grade 4 designated cartilage that was “completely destroyed,” often involving complete loss of the entire articular surface.

Ficat and Hungerford

Ficat and Hungerford¹² described a two-stage classification system for lesions in the patellofemoral compartment based on the appearance of the cartilage. Stage one was described as closed chondromalacia in which the surface was intact but the degree of severity varied from simple softening to “pitting edema” and loss of elasticity. Stage two was defined as open chondromalacia. This stage was subdivided into two groups: A, single or multiple fissures that were either relatively superficial or extended down to subchondral bone; and B, ulceration or localized loss of cartilage substance, exposed dense subchondral bone that represented the final stage of cartilage destruction.

These authors described other types of patellar articular cartilage defects, including chondrosclerosis, a term used to describe cartilage that was abnormally hard and not depressible. Tuft formation indicated a lesion consisting of multiple deep fronds of cartilage separated from one another by deep clefts that extended to subchondral bone. Superficial surface changes described superficial surface fibrillations that sometimes were not visible. Longitudinal striations could be present in the axis of movement of the joint.

Cincinnati Knee Rating System

In 1989, Noyes and Stabler²⁸ proposed a new articular cartilage classification system that included four variables: the appearance of the cartilage (as visualized arthroscopically), the depth of the lesion, the size (diameter) of the lesion, and the location of the lesion. A point scaling system was devised for research purposes to facilitate statistical analyses (Table 47-2).

The appearance of the cartilage is classified into one of three grades: grade 1, cartilage surface intact, closed lesion; grade 2, cartilage surface damaged, open lesion; or grade 3, subchondral bone exposed. Each grade is divided into subtypes A or B, depending upon the depth of involvement.

Grade 1A corresponds to a moderate degree of softening of the cartilage compared with its normal resilient appearance (Fig. 47-1A). Grade 1B designates complete loss of resilience, resulting in full indentation of the surface when probed (see Fig. 47-1B). This grade is reserved for lesions in which a severe softening of the cartilage exists without gross surface breakage. This description indicates the initial stage of cartilage deterioration characterized by diminished glycoproteins and early collagen fiber deterioration.²⁸ These lesions, which commonly involve the undersurface of the patella, have a high potential to progress to an open lesion.

FIGURE 47-1 A, Arthroscopic photograph of a grade 1A articular cartilage lesion according to the Cincinnati Articular Cartilage Rating System. B, Arthroscopic photograph of a grade 1B articular cartilage lesion.

A grade 2A lesion (Fig. 47-2A) is an open disruption of the cartilage surface, including cracks, surface blisters, fissures, fibrillations, and fragmentations that extend less than one half of the thickness of the cartilage. Grade 2B lesions (see Fig. 47-2B and C) are similar in appearance, but extend greater than one half of the thickness of the cartilage. Superficial lesions with fibrillation-type changes are usually easily distinguishable from deeper lesions that involve lacerations, clefts, and fragmentation down to bone. Chondrosclerosis is classified as a grade 2 lesion, because the surface is abnormal. Lesions that have two or three deep lacerations that extend through the cartilage surrounded by otherwise normally appearing cartilage are graded as 2B.

FIGURE 47-2 A, Arthroscopic photograph of a grade 2A articular cartilage lesion according to the Cincinnati Articular Cartilage Rating System. B, Arthroscopic photograph of a grade 2B articular cartilage lesion. C, A knee with a combined grade 2A lesion on the femoral condyle and a grade 2B lesion on the tibial plateau.

A grade 3A lesion (Fig. 47-3A) indicates a surface with exposed bone in which the normal bony contour remains. The most severe lesion, grade 3B (Fig. 47-3B and C), indicates cavitation or erosion of the bone surface. An example of a grade 3B lesion is an osteochondritis dissecans lesion in which there is loss of the subchondral bone. Other examples are lesions in which grooving of the bony surface is visualized or those involving fractures that extend through the subchondral bone.

FIGURE 47-3 A, Arthroscopic photograph of a grade 3A articular cartilage lesion according to the Cincinnati Articular Cartilage Rating System. B, Arthroscopic photograph of a grade 3B articular cartilage lesion. C, A grade 3A lesion on both the femoral condyle and the tibial plateau in a meniscectomized knee. D, Schematic illustrations of grades according to the Cincinnati Knee Rating System.

The size of all lesions is determined using a calibrated nerve hook probe. The size is recorded according to one of five categories ranging from less than 10 mm to greater than 25 mm (see Table 47-1). A schematic illustration of all grades is shown in Figure 47-3D.

The anatomic location of the lesion is also recorded. The femoral articular surface is divided into the trochlea and anterior, middle, and posterior thirds of the femoral condyles and tibial condyles. The patella is anatomically classified in two ways: in a proximal-to-distal direction and according to facet (odd, medial, lateral). The range of degrees of knee flexion where patellofemoral and tibiofemoral lesions are in contact is documented. This information is provided to the physical therapist postoperatively so that modifications of exercises may be made to avoid placing forces on the affected area. Any lesion in contact from 0° to 45° of flexion may affect normal walking activities and may have more implications than those that contact at greater degrees of knee flexion.

Figure 47-4A illustrates how this classification system is used to record the visual appearance of articular cartilage lesions. The patella has a 10-mm grade 2A lesion that is surrounded by a larger closed area of softening (1B) that extends 20 mm in diameter. The trochlea has a large area of extensive softening as well, graded as a 1B that is 25 mm in diameter. The medial femoral condyle also has a large area of extensive softening, measuring 15 mm in diameter. A “kissing” lesion is found on the medial tibial plateau of a grade 2A lesion that is 10 mm in diameter. The lateral femoral condyle has a complex lesion; 10 mm of exposed bone (3A) is surrounded by an open lesion of fragmentation of 20 mm in diameter (2B), which is in turn surrounded by a closed lesion of soft cartilage that extends to 25 mm (1B). The lateral tibial plateau has a large 2B lesion, 15 mm in diameter. For research purposes, the lesions may be quantified on a scaling system to allow statistical analysis (see Table 47-2). Any lesion that is less than 10 mm in diameter, or graded 1A, is not considered clinically significant and, therefore, no points are given. Figure 47-4B illustrates the points awarded for the lesions noted in the example.

FIGURE 47-4 A, An example of the grading of articular cartilage lesions found at arthroscopy. B, Scoring system for articular cartilage lesions (see Table 47-2).

Occasionally, complex lesions are encountered such as a 2A circumferential fibrillation surrounding a 10-mm 2B lesion. For scoring purposes, additional points for complex lesions may be incorporated so that the total points to be subtracted are increased, but still less than the corresponding 15 mm lesion. Examples for scoring complex lesions are shown in Figure 47-5.

FIGURE 47-5 Three examples of scoring complex lesions (see Table 47-2 and text).

This classification system allows clinical investigations to analyze results according to the condition of the articular cartilage and to determine the difference in outcome between patients with normal cartilage, noteworthy cartilage damage (grade 2B), and severe cartilage damage (grade 3).^24–27 For instance, in a study of 66 patients who had an anterior cruciate ligament (ACL) revision reconstruction,²⁷ the authors found that patients with grades 2B or 3A lesions had significantly poorer subjective and functional outcomes a mean of 42 months postoperatively compared with those with normal or grade 1 surfaces (Table 47-3). The factors included pain (P = .03), stair-climbing (P = .01), kneeling (P = .01), running (P = .001), jumping (P = .01), twisting and turning (P = .01), and the overall Cincinnati Knee Rating Score (CKRS; P = .0001). Analysis of the preoperative data found significant differences between these subgroups as well for giving-way (P = .004), walking (P = .04), stair-climbing (P = .04), kneeling (P = .01), running (P = .004), jumping (P = .001), twisting and turning (P = .01), and the overall CKRS (P = .0001). Differences were also noted in the sports activity levels patients resumed postoperatively (Table 47-4). The stratification of lesions according to the six compartments may be presented as shown in Figure 47-6. These findings allow appropriate patient counseling regarding the expected outcome of the operation according to, among other factors, the condition of the articular cartilage. Larger sample sizes would allow further analysis of the effect of not only the grade of lesions but their locations as well.

TABLE 47-4 Effect of the Condition of the Articular Cartilage at the Index Operation on the Sports Activity Level at the Latest Follow-up Examination after Anterior Cruciate Ligament Revision Reconstruction

FIGURE 47-6 Venn diagrams of the number of patients who had an abnormal articular cartilage surface, according to each separate compartment. Thirty-seven patients (56%) had abnormal articular cartilage, defined as fissuring and fragmentation of more than one half of the involved surface or any exposure of subchondral bone.

(From Noyes, F. R.; Barber-Westin, S. D.; Roberts, C. S.: Use of allografts after failed treatment of rupture of the anterior cruciate ligament. J Bone Joint Surg Am 76:1019–1031, 1994.)

SFA System

Dougados and colleagues¹¹ described a rating system for assessing the severity of articular cartilage lesions by determining the location, size, and appearance of the lesions. The location of the lesions was drawn on a diagram for the patella, trochlea, medial femur, medial tibia, lateral femur, and lateral tibia. The size of the lesion was recorded in terms of percentage of the joint surface affected. The appearance was graded according to a system proposed by Beguin and Locker² in which grade I represents closed lesions; grade II, superficial fissuring; grade III, deep fissuring, reaching subchondral bone that is not visualized but may be probed; and grade IV, subchondral bone exposure.

The authors assessed two different scoring systems. The first was a 100-mm long visual analog scale that the surgeon completed to represent the overall cartilage damage, in which 0 reflected no damage and 100, the most severe damage possible. This scale was completed for each compartment of the knee. The second scoring method was the more qualitative grading system (grades I–IV) described previously. Whereas both scoring systems appeared to be “of interest,” further studies are required to validate the proposed means of scoring, determine reliability issues, and assess whether this system can detect changes in lesion severity over time.

ICRS

The ICRS was founded in 1997 to develop a standardized system for the evaluation of articular cartilage injury and repair.^4,5 The system focuses on the depth of lesions (graded from 0–4; see Table 47-1), the size of the lesion (recorded in millimeters), and the area of damage (documented on a mapping system; Fig. 47-7). The results are also recorded as either normal, nearly normal, abnormal, or severely abnormal for use with the International Knee Documentation Committee (IKDC) knee rating system.¹⁶

FIGURE 47-7 International Cartilage Repair Society knee cartilage mapping system.

(Document reprinted with permission from the International Cartilage Repair Society, Switzerland.)

Normal cartilage is graded as ICRS 0, or normal on the IKDC system (see Chapter 45, The International Knee Documentation Committee Rating System). Intact lesions with slight softening or mild fibrillation are classified as ICRS 1A and, if additional superficial lacerations and fissures are present, are classified as ICRS 1B (nearly normal on the IKDC system for both ICRS 1A and 1B). Lesions that extend deeper than 1B defects that involve less than 50% of the cartilage thickness are classified as ICRS 2, or abnormal on the IKDC system.

Lesions that extend through greater than 50% of the cartilage thickness are classified as ICRS 3, or severely abnormal on IKDC and classified into one of four subgroups. ICRS 3A indicates defects that extend greater than 50% of the cartilage depth, but not to the calcified layer. ICRS 3B indicates defects that extend greater than 50% of the cartilage depth and that reach the calcified layer. Defects that extend to but not through the subchondral bone plate are classified as ICRS 3C, and blisters are classified as ICRS 3D. Full-thickness cartilage defects resulting from joint trauma that extend into the subchondral bone are classified as ICRS 4 (severely abnormal on IKDC). Defects of osteochondritis dissecans have a separate classification system of their own.

The ICRS mapping system requires validation because it contains 46 zones that have minimal anatomic and functional referencing.¹⁷ Brittberg and Winalski⁶ expressed that “If the total joint involvement can be assessed with percentages and/or with such a scoring system, perhaps with the aid of a computer-based system, the progression or improvement of cartilage disease can be followed.”⁶

The ICRS proposed a rating for the assessment of cartilage resurfacing procedures such as mosaicplasty, osteochondral autograft transfer, and ACI. The assessment includes consideration of the volume of the defect that is filled with repair tissue, the integration of repair tissue with surrounding articular cartilage, and the macroscopic surface appearance (Table 47-5). A grading system is incorporated that allows both statistical analysis of the total points obtained (scale, 1–12 points) and placement of patients into IKDC overall rating categories.

TABLE 47-5 International Cartilage Repair Society Cartilage Repair Assessment

Reliability Studies of Cartilage Rating Systems

Only a select number of intraobserver and interobserver reliability studies of various articular cartilage rating systems have been conducted to date.^3,7,22,23 Brismar and associates³ measured intraobserver and interobserver reliability among four orthopaedic surgeons who rated lesions according to Outerbridge, SFA, and a modified Collins⁹ classification system. Videotapes of 19 knees with a total of 114 surfaces were rated by each surgeon during two separate sessions that were at least 2 months apart. The authors reported that the coefficient kappa statistic (measurement of agreement beyond that expected by chance alone) demonstrated only moderate intraobserver and interobserver agreements. Disagreement was evident both between and within observers in the rating of lesions; however, only 6% to 8% of the paired intraobserver classifications differed by more than one grade or category within each rating system. The investigators concluded that, for research purposes, an improved classification system was required.

Cameron and colleagues⁷ investigated six cadaveric knees aged 50 to 79 years in which first standard arthroscopy was performed and then arthrotomy and measurement of lesions was accomplished using calipers to assign the appropriate grade according to the Outerbridge classification system. A total of nine orthopaedic surgeons reviewed each videotape twice during the same session and graded any lesions viewed.

The results demonstrated moderate accuracy (68%) and agreement (kappa score, 0.602) of the observers in correctly grading the lesions from the arthroscopic videotapes compared with the grade assigned during the arthrotomy. The lower-grade lesions were diagnosed with less accuracy than the higher-grade lesions. The observers tended to grade lesions higher than that assigned during the arthrotomy. The authors concluded that orthopaedic surgeons can

accurately grade chondral lesions” with the Outerbridge classification system. The number of years of orthopaedic experience did not significantly influence the study findings.

Marx and coworkers²² measured the interobserver variability in the grading of articular cartilage lesions using the Outerbridge system as observed on surgical videotapes. Six surgeons provided grades for each articular surface in 53 knees. In the first 31 knees, grades 2 and 3 were combined, whereas in the remaining 22 knees, these grades were not combined. Multirater kappa statistics demonstrated high interrater agreement for all joint surfaces except the medial and lateral tibial plateau when grades 2 and 3 were combined. However, in the knees in which grades were not combined, only moderate interrater agreement was reported.

Smith and associates³⁰ tested the validity and reliability of the ICRS system and the previously unpublished Oswestry Arthroscopy Score (OAS), as used to rate repair of articular cartilage using procedures such as osteochondral autograft transfer and ACI. The OAS score rates five macroscopic parameters of graft level with surrounding cartilage, integration with surrounding cartilage, appearance of surface, color of graft, and stiffness on probing to reveal an overall point score from 0 to 10 points. Six orthopaedic surgeons viewed five arthroscopic videotapes in patients who had undergone ACI and scored the chondral surfaces according to both rating systems twice, 6 weeks apart.

The intraclass correlation coefficient demonstrated adequate reliability for the test-retest data (0.94 for both rating systems) and interrater reliability (0.83 for ICRS; 0.76 for OAS). Both systems had satisfactory internal consistency for use in research investigations. The OAS score was slightly superior in terms of content validity because it measured stiffness and provided an indication of hypertrophy of a grafted area, whereas the ICRS score did not measure these parameters and could therefore induce a positively biased score.

Correlation of ICRS Rating and Changes in Tissue Mechanical Properties

Kleemann and colleagues¹⁹ conducted an investigation to determine the relationship between various stages of articular cartilage deterioration, as rated macroscopically by the ICRS system (grades 0–4) and changes in tissue mechanical propeties. Tibial plateaus were collected in 21 patients who underwent total knee arthroplasty and ICRS grades assigned by three surgeons. The specimens were examined biomechanically and histologically using a modified Mankin scoring system (Table 47-6).²¹ The mechanical properties were measured with a custom-made testing material testing device that determined the Young modulus of the cartilage by unconfined compression. Histologic analysis used hematoxylin and eosin staining and safranin-O staining for the evaluation of the Mankin score.

TABLE 47-6 Mankin Histologic and Histochemical Grading System for Evaluation of Articular Cartilage Degeneration