High occurrence of osteoarthritic histopathological features unaccounted for by traditional scoring systems in lateral femoral condyles from total knee arthroplasty patients with varus alignment

Abstract

Background and purpose - A better understanding of the patterns and variation in initiation and progression of osteoarthritis (OA) in the knee may influence the design of therapies to prevent or slow disease progression. By studying cartilage from the human lateral femoral condyle (LFC), we aimed to: (1) assess specimen distribution into early, mild, moderate, and severe OA as per the established histopathological scoring systems (HHGS and OARSI); and (2) evaluate whether these 2 scoring systems provide sufficient tools for characterizing all the features and variation in patterns of OA. Patients and methods - 2 LFC osteochondral specimens (4 x 4 x 8 mm) were collected from 50 patients with idiopathic OA varus knee and radiographically preserved lateral compartment joint space undergoing total knee arthroplasty. These were fixed, sectioned, and stained with HE and Safranin O-Fast Green (SafO). Results - The histopathological OA severity distribution of the 100 specimens was: 6 early, 62 mild, 30 moderate, and 2 severe. Overall, 45/100 specimens were successfully scored by both HHGS and OARSI: 12 displayed low OA score and 33 displayed cartilage surface changes associated with other histopathological features. However, 55/100 samples exhibited low surface structure scores, but were deemed to be inadequately scored by HHGS and OARSI because of anomalous features in the deeper zones not accounted for by these systems: 27 exhibited both SafO and tidemark abnormal features, 16 exhibited only SafO abnormal features, and 12 exhibited tidemark abnormal features. Interpretation - LFC specimens were scored as mild to moderate OA by HHGS and OARSI. Yet, several specimens exhibited deep zone anomalies while maintaining good surface structure, inconsistent with mild OA. Overall, a better classification of these anomalous histopathological features could help better understand idiopathic OA and potentially recognize different subgroups of disease.

Figure 1.

Sample of lateral femoral condyle (black star) (right knee shown in example) was obtained (A, B). The orientation of the condyle was noted in the operating room (A = anterior; P = posterior; M = medial; L = lateral) and placed in an in-house fabricated miter box in AP orientation and cut into 4 mm thick arches through the region of the femur that is weight bearing in extension (C). The central region of 1 arch was cut into 5 specimens (4 mm x 4 mm), the second (Lateral) and fourth (Medial) were processed for histology (D). Each specimen was paraffin embedded, sectioned, and stained with HE (E), and SafO-FG (F) for analysis.

Figure 2.

Correlation plot of average HHGS versus average OARSI scores for specimens from lateral and medial locations. Moderate correlation was observed between the 2 systems (Spearman’s coefficient =0.806). The total HHGS and OARSI scores were divided into 5 equal bins as indicated by the dotted lines, and classified as early, mild, moderate, and severe to assess specimen distribution. The shaded regions on the bottom and left represent the regions containing the 50% of samples with scores below the median for each scoring system (median HHGS =4.8, OARSI =5.5) where correlation between the systems is less robust. For illustration, the trend is represented by a red line from a fitted linear regression of OARSI on HHGS.

Figure 3.

Representative HE and SafO stained images of cartilage specimens obtained from the lateral femoral condyle. Top panel images indicate histopathological features found in “normal” cartilage, showing normal surface and cells (A), uniform safraninO staining (B) and single undulating tidemark not breached by blood vessels (C). Bottom panel images suggest typical osteoarthritic cartilage features, where surface shows fissures (D), hypercellularity, and cloning associated with fissures (D, E) and tidemark breached by blood vessels (F).

Figure 4.

Representative images of cartilage obtained from the lateral femoral condyle that indicate the unaccounted histopathological safraninO features: (A) loss of SafO stain in the top half of the tissue that is not associated with much surface erosion or fissures; (B) tissue necrosis/degradation in the radial zone, accompanied by some loss of SafO stain in the inter-territorial matrix region; (C) loss of SafO stain in the inter-territorial matrix, mainly confined to the bottom half of the tissue section; (D) varying staining patterns seen in the territorial matrix region and no SafO stain observed in some inter-territorial regions in the radial zone; (E) SafO staining loss near the tidemark even when the rest of the cartilage features appear relatively normal. The table indicates the total HHGS scores for each individual specimen, along with structure score (S), cell score (C), safraninO/fast green score (Saf), tidemark score (T). 43% of sample cohort presented with at least one feature.

Figure 5.

Representative images of cartilage obtained from the lateral femoral condyle that indicate the unaccounted tidemark-related features as seen in the HE stained sections: (A) multiple tidemarks; (B) multiple tidemarks that are breached by multiple blood vessels; (C) unknown tissue composition deposited near the tidemark, which stains significantly differently from normal hyaline cartilage and bone tissue; (D) formation of bone tissue well within the hyaline cartilage region resulting in the appearance of cartilage–bone–cartilage–bone interleaved tissue layers accompanied by multiple tidemarks. The table indicates the total HHGS scores for each individual specimen, along with structure score (S), cell score (C), safraninO/fast green score (Saf), tidemark score (T). 39% of the sample cohort presented with at least 1 feature.