Impact of Lesion Location on the Progression of Osteoarthritis in a Rat Knee Model

Abstract

To investigate how surgically created acute full-thickness cartilage defects of similar size and location created on the medial versus lateral femoral condyle influence progression of spontaneous cartilage lesions in a rat model. Full-thickness cartilage defects of 1 mm were surgically created on the medial or lateral femoral condyles on the right leg of 20 rats (n =10/group). Ten rats served as controls. Spontaneous lesion progression on the ipsilateral and contralateral surfaces was examined using a high-resolution digital camera along with H&E and Safranin-O staining. Chondral defects were scored grossly and histologically. Control femur displayed no cartilage disruption. Surgically treated knees exhibited created and spontaneous cartilage defects with no evidence of healing unless subchondral bone was penetrated. Ipsilateral spontaneous lesions on the lateral condyle were significantly more severe on average (p =0.009) compared to medial lesions on gross examination. Histological examination found contralateral lesions on the lateral surface following surgically created medial lesions to be more severe (p =0.057) compared to contralateral lesions. A trend toward more susceptible chondral damage to the lateral condyle was observed following acute lesion creation on either medial or lateral condyles. Mechanisms behind this pattern of spontaneous lesion development are unclear, requring further investigation.

Injuries to the articular cartilage of the knee are exceedingly prevalent in both symptomatic and asymptomatic patient populations and represent one of the most difficult treatment challenges facing orthopaedic surgeons.^1–8 Lesions most frequently involve the medial femoral condyle, trochlea groove, and patellar articular surface.^3,7,8 Such lesions have poor natural healing capacity and current repair techniques yield less than stable regeneration with variable tissue quality, depending on technique.^9,10 While the natural history of chondral defects is poorly understood, acute traumatic, and chronic degenerative lesions are likely to progress over time leading to disabling osteoarthritic changes.^11–13

Several variables in the knee have been studied to determine their role in the progression of focal cartilage damage to osteoarthritis (OA).¹⁴ Variables, such as focal defects, cartilage volume, subchondral bone thickness, presence of osteophytes and the type, and acuity of trauma, have been shown to influence the pathogenesis and progression of cartilage lesions.^{3–5,7,11,15–18} However, a significant gap in knowledge exists concerning the natural history of articular lesions and the impact of defect location on subsequent degeneration and/or healing.

The goal of this study was to create a pilot rodent model to investigate mechanisms of chondral defect progression based on lesion location. This study sought to investigate how the location of an acutely created lesion would impact subsequent cartilage damage in four locations within the knee joint, specifically: (i) around the area of the acute defect; (ii) on the ipsilateral femoral condyle with the acute lesion; (iii) on the opposite (contralateral) condyle surface; and (iv) on the tibial surface. To answer these questions, surgically created lesions of similar diameter, depth, and location on either the medial or lateral femoral condyles were examined in a rat model. We hypothesized that differences in the progression and severity of osteoarthritic changes as measured on gross and histological examination depend on the initial defect location (medial vs. lateral condyle) in all observed locations.

METHODS

Animals and Surgical Procedures

All experimental procedures were pre-approved by the Institutional Animal Care Use Committee at The Ohio State University. Seventeen-week-old Sprague–Dawley female rats (Rattus norvegicus) (n =30) were obtained from Charles River Lab, PA. Rats were randomly divided into three groups (n =10/group). Rats in Group 1 consisted of controls receiving no surgical defects to the cartilage. Rats in Group 2 received acute full-thickness defects on the medial femoral condyle in the front of the right knee using a 1 mm diameter biopsy punch. Lesions were created approximately at the level of the trochlear groove, 0.5–1 mm medial to center on the femoral condyle (Fig. 1). This location on the condyle was chosen in order to exposure the weight-bearing surface of the injured condyle to maximal contact stresses, shown to occur slightly outside the area of the defect rim in small (<2 mm) chondral defects.¹⁸ Rats in Group 3 received the same procedure at the end of the trochlear groove, 1 mm lateral to center on the femoral condyle. Removal of cartilage from each specimen was verified by visually inspecting each biopsy punch for cartilage removal.

Figure 1.

Approximate location of acute created lesions on femoral condyles. Biopsy punch was positioned at the bottom of the trochlear groove and moved 0.5–1.0 mm medial or lateral of center. MC, medial condyle; LC, lateral condyle.

All surgical procedures were carried out using aseptic technique by a single surgeon (RKH). Isoflurane (5%) was used for induction and maintenance of anesthesia. A 2 cm incision was made on the superolateral aspect of the knee on all rats. The underlying soft tissues were dissected bluntly down to the knee joint. The quadriceps and patellar tendons were mobilized, being careful not to disrupt the extensor mechanism. The patella was retracted medially and an arthrotomy was made. Care was taken to prevent iatrogenic joint injury. A full-thickness defect was created with a 1 mm biopsy punch in the locations as described above in Groups 2 and 3. Control rats in Group 1 received no chondral lesion following arthrotomy. Arthrotomy was left open, while skin was closed in interrupted fashion using 3.0 nylon sutures. Postoperatively, all rats received 0.3 mg of Acepromazine subcutaneously and Ibuprofen 200 mg diluted in 25 ml of water via oral lavage for 5 days for pain. Rats remained in their cages as lesions were allowed to progress for 6 weeks (41 days). All rats were sacrificed on day 42.

Gross Morphological Analysis

The right knee joint was dissected and the femoral condyle harvested and fixed in 10% buffered formaldehyde solution. Ligaments and soft tissue were removed and specimens decalcified in Cal-Ex solution. For gross evaluation, condyles were imaged under a stereomicroscope (Zeiss) to record lesion distribution and observable degeneration.

Characteristics of formed lesions of the biopsy site, the ipsilateral condylar surface, the contralateral surface, the ipsilateral tibial surface, and the contralateral tibial surface were assigned numeric scores for severity in a randomized, blinded manner by three observers (RKH, DCF, and SA) of the research team. Using computer images generated from stereoscopic observation, scores were recorded based on the ICRS cartilage repair assessment criteria.^19,20 To correspond with the histologic scoring system by Mankin et al.²¹ of increasing severity corresponding to increasing number score, ICRS scores were reversed for our purposes. Specifically, 0 =intact smooth surface, 1 =fibrillated surface, 2 =small scattered fissure or cracks, 3 =several small or few but large fissures, and 4 =total degeneration (Fig. 2). Observers were encouraged to modify their scores using subscale numbers (e.g., 1.5, 2.5, and 3.5) for specimens not fully conforming to the standardized scoring scale. When more than one spontaneous lesion was found to be present, observers were asked to assess the overall appearance of the condylar surface and grade appropriately. Scores were then analyzed to ensure no significant disagreement (defined as scores >1 point from remaining two observers), then averaged and assigned to each specimen.

Figure 2.

Macroscopic scoring system for gross lesions. Representative examples of each stage of condylar damage that research observers utilized in scoring gross lesion formation on femoral surfaces. Scores were adopted from the ICRS cartilage repair scoring system (as shown above); however number scores were reversed such that increasing score corresponded to increasing lesion severity.

Histological Analysis

Femurs were embedded in paraffin and sectioned at 5 μm thickness in the sagittal plane. Haemotoxylin and eosin (H&E) and 1.5% Safranin O (Fischer Scientific, Pittsburgh, PA) stains were used to examine histological degeneration and glycosaminoglycan (GAG) content. Degenerative changes around the biopsy site, ipsilateral and contralateral surface were analyzed using the structure subset of the Mankin scoring system for H&E stained sections²¹; 0 = normal, 1 =surface irregularities; 2 =pannus and surface irregularities; 3 =clefts to transitional zone; 4 =clefts to radial zone, 5 =clefts to calcified zone; and 6 =complete disorganization. Scores were then averaged among the three observers to ensure no significant outliers as described above. Ipsilateral and contralateral sections were examined at approximately the same distance from center to ensure reliable observation.

Statistical Analysis

Statistical analysis was performed by calculating a two-tailed Student’s t-test to identify significant differences between medial versus lateral scoring of the biopsy site, as well as spontaneous lesions on the ipsilateral and contralateral surfaces. The Satterthwaite approximation was used to account for unequal samples sizes when necessary. Statistical significance was defined as p <0.05.

RESULTS

Rats in surgical groups were successfully subjected to creation of acute lesions on either the medial or lateral femoral condyles while controls received no defects. No complications were encountered during surgeries or in the postoperative period. The activity scale and behavior of all rats postoperatively was no different than that observed pre-operatively with normal feeding and behavior towards handlers with no impaired or slowed movement. All rats were sacrificed and their knees harvested on day 42 and evaluated and scored grossly and histologically.

Control Knees

Scores in all control knees were zero on both gross and histologic examination on the condylar and tibial surfaces, indicating healthy, undisturbed cartilage in all compartments analyzed.

Acute Created Lesions

Lesion location was not shown to result in significant differences in progression of defects based on average severity scores on gross and histologic examination (p =0.233, p =0.193, respectively). Gross examination revealed evidence of cartilage damage at the area of the biopsy punch in all surgical specimens (Fig. 3). Areas corresponding to biopsy lesions grossly were identified and confirmed on histological examination (Figs. 4B and 5A). However, one medial and two lateral lesions were found to penetrate the subchondral bone with resultant fibrocartilage formation. These specimens were excluded from scoring. Safranin O staining revealed reduced staining indicating GAG loss around the borders and depth of acute lesions, with loss extending into the immediately adjacent cartilage in both medial and lateral defects (Figs. 6A and 7A). On gross examination, acute lesions exhibited primarily superficial fibrillations with some partial thickness erosions not extending to the subchondral bone. Medial lesions had an average defect score of 1.415, while lateral lesions had a score of 1.766 (Table 1). On histologic examination, acute lesions were predominately identified by cartilage loss extending into the transitional and radial zones. Medial lesions had an average defect score of 2.795, while lateral lesions had a score of 3.375 (Table 2).

Figure 3.

Gross examination of femoral condyle lesions. Acute created lesions using biopsy punch were identified by circular impressions created on the cartilage surface. Spontaneous lesions were identified as areas of fraying and fibrillation to the cartilage surface, identified on both the ipsilateral and contralateral femoral condyles.

Figure 4.

Histological examination of acute created medial lesions. Acute created lesions using biopsy punch (A) were identified by areas of cartilage disruption extending into the cartilage tissue, but not penetrating into the subchondral bone. Specimens exhibiting penetration of the subchondral bone with fibrocartilage formation were excluded. Spontaneous lesions were identified by areas of surface disruption identified on both the ipsilateral (B) and contralateral (C,D) femoral surfaces.

Figure 5.

Histological examination of acute created lateral lesions. Acute created lesions using biopsy punch (A) were identified by areas of cartilage disruption extending into the cartilage tissue, but not penetrating into the subchondral bone. Specimens exhibiting penetration of the subchondral bone with fibrocartilage formation were excluded. Spontaneous lesions were identified by areas of surface disruption identified on the ipsilateral surface (B), but not on the contralateral (C,D) femoral surfaces.

Figure 6.

Safranin O staining of acute created medial lesions. Acute lesions created using biopsy punch (A) were identified by areas of cartilage disruption extending into the cartilage tissue with corresponding loss of staining, representative of GAG loss. Spontaneous lesions were identified by areas exhibiting a reduction of stain beginning at the surface with variable extension into the superficial layers (B). Contralateral defects following medial lesions exhibited more pronounced reduction in staining at the surfaces (C,D) with minimal preservation of GAG content in the deeper cartilage layers.

Figure 7.

Safranin O staining of acute created lateral lesions. Acute lesions created using biopsy punch (A) were identified by areas of cartilage disruption extending into the cartilage tissue with corresponding loss of staining, representative of GAG loss. Spontaneous lesions were identified by areas exhibiting a reduction of stain beginning at the surface with variable extension into the superficial layers (B). Contralateral defects following lateral lesions exhibited preservation of Safranin O staining with mild reduction of GAG content on the surfaces (C,D).

Table 1.

Gross Scoring Results

Lateral condyle				Medial condyle
Specimen #	Acute	Spontaneous		Specimen #	Acute	Spontaneous
		Ipsilateral	Contralateral			Ipsilateral	Contralateral
1	2.0	1.5	0	11	2.0	0.83	1.0
2	3.0	3.83	0	12	1.0	0.167	0
3	1.0	0.66	0	13	1.83	1.0	1.0
4	2.0	2.83	1.0	14	1.0	1.0	1.0
5	2.0	2.0	0	15	1.0	0.3	1.167
6	2.83	2.667	0	16	2.0	1.83	1.83
7	1.0	2.16	1.0	17	2.0	1.0	0
8	1.0	0.83	0	18	1.0	1.16	0
9	1.0	1.0	0	19	1.16	1.0	0
10	1.83	2.0	0.33	20	1.16	0.67	0
Avg.	1.766	1.947	0.776	Avg.	1.415	0.895	1.199

Score for each specimen is average of three observers.

Table 2.

Histological Scoring Results

Lateral condyle				Medial condyle
Specimen #	Acute	Spontaneous		Specimen #	Acute	Spontaneous
		Ipsilateral	Contralateral			Ipsilateral	Contralateral
1	4.0	1.83	1.0	11	X	5.0	2.0
2	X	2.0	1.16	12	3.0	1.16	2.0
3	4.0	3.0	1.0	13	3.0	1.16	1.0
4	2.0	1.67	0	14	3.0	1.83	2.3
5	3.0	1.16	0	15	2.0	1.0	0
6	X	1.0	1.0	16	3.16	2.16	2.0
7	2.0	1.16	0	17	2.0	1.0	0
8	4.0	2.16	1	18	4.0	2.0	3.0
9	3.0	2.0	0	19	3.0	1.0	0
10	5.0	2.0	1.16	20	2.0	1.0	2.16
Avg.	3.375	1.798	0.632	Avg.	2.795	1.731	1.446

Score for each specimen is average of three observers.

Spontaneous Ipsilateral Lesions

Spontaneous lesions on the same condyle receiving acute created lesions were identified on gross (Fig. 3) and histologic (Figs. 4B and 5B) examination on all specimens. Spontaneous lesions were significantly more severe based on average severity score on the lateral surface when compared to spontaneous medial lesions (1.947 vs. 0.895, p =0.009) on gross examination (Table 1). No cartilage disruption was observed on the tibial surfaces regardless of the severity of acute lesions (Fig. 8). Histological examination revealed no significant difference in spontaneous ipsilateral lesion severity (Table 2). Lateral lesions had an average severity score of 1.798, while medial lesions had a score of 1.731 (p =0.879). Spontaneous lesions were located approximately 0.25–1 mm anteriorly on the condylar surface relative to the edge of the acute defects, primarily within the intercondylar ridge. Safranin O staining showed reduction of staining isolated to the surfaces of spontaneous ipsilateral defects in both medial and lateral defects (Figs. 6B and 7B).

Figure 8.

Gross examination of right tibial plateaus. Regardless of severity of acute created lesion, no surface changes or evidence of cartilage were identified on either surfaces of the right tibia.

Spontaneous Contralateral Lesions

Spontaneous contralateral lesions were identified on gross examination (Fig. 3) on five lateral condyles following acute medial defects and on three medial condyles following acute lateral defects. On histological examination, seven specimens exhibited defects on the lateral condyles following acute medial defects, while six had lesions on the medial condyle following acute lateral defects (Figs. 4C,D and 5C,D). Scoring on gross examination revealed no significant differences in average lesion severity (p =0.165, Table 1). Gross examination of contralateral tibial plateaus revealed no evidence of cartilage damage (Fig. 8). Histological examination found the differences in lesion severity approaching statistical significance (p =0.057, Table 2). Spontaneous lesions on the lateral surface (acute created medial lesions) had an average defect score of 1.446, while spontaneous lesions on the medial surface (acute created lateral lesions) had a score of 0.632. Safranin O staining similarly exhibited more extensive reduction of GAG staining to the contralateral surface of medial versus laterally created lesions (Figs. 6C,D and 7C,D). Spontaneous contralateral lesions were located at approximately the same level on the intercondylar ridge as spontaneous ipsilateral lesions, anterior to the surgical defects.

DISCUSSION

The progression of an isolated chondral defect to osteoarthritis is a multifactorial process involving patient and defect specific characteristics.^16,22–24 The current gap in knowledge exists in understanding the exact mechanisms and risk factors by which such chondral defects progress. In this study, we found that acute created lesions of similar depth and location on either the medial or lateral femoral condyle increased the risk of OA development, with slight differences in OA development depending on defect location. Specifically, we observed: (i) more severe spontaneous defects on the lateral surface following acute defect creation on the lateral femoral condyle on gross examination; and (ii) a trend toward more severe spontaneous defects with increased GAG loss on the lateral surface following acute defect creation on the medial femoral condyle on histologic examination.

To our knowledge, this is the first study utilizing direct, full-thickness chondral trauma in a rat model to study patterns of arthritis progression in rodent knees.²⁵ As a result, the mechanisms behind our findings of more severe defects on average on the lateral femoral condyle in our model remain unknown. While other animal studies have found similar patterns of spontaneous lesion development with respect to lesion location,¹⁸ explanations such as differences in peak pressures,^18,26,27 condylar shape/incongruity²⁸ or meniscal morphology¹⁷ have not been substantiated through biomechanical analysis in the rat model.

There were a number of limitations identified in this pilot study. The creation and placement of biopsy lesions during surgery was guided simply by human approximation of the distance from the trochlear groove and the pressure applied to each specimen in removing cartilage was guided by touch alone. While care was taken to assess the appropriate distance between the end of the trochlea and the condylar ridges, there was no standardized method to ensure uniform depth and location. This resulted in acute lesions located outside the desired area as well as partial-thickness lesions. Meniscal and patellar pathology were not examined in the current model, hindering our study from accessing the impact of lesion location on total joint pathology. In addition, scoring the severity of cartilage damage was performed individually and subjectively based on the observer’s personal interpretation of the scoring system. Furthermore, this study did not take into account the initial amount of cartilage volume in the knee or the weight of the individual rats, factors which could potentially confound our results.

This pilot study was successful in creating a rodent model that can be used for understanding the mechanisms of OA progression. While utilization of a rat model provides information less applicable to the progression of OA in humans than larger animal models,^29,30 these findings lay groundwork for future translational studies in a cost effective model to study the effects of location on isolated chondral defect progression. However, further analyses of contact pressures and gait mechanics before and after defect creation will be required to help explain the mechanisms behind our observed pattern of spontaneous lesion formation in future studies.

CONCLUSION

Acute created defects on the lateral condyle led to more severe spontaneous ipsilateral lesions, while acute medial created lesions resulted in a trend towards more severe contralateral (lateral) lesions. This study suggests that the location of an acute traumatic lesion on the femoral condyles influences the development of spontaneous lesions within the knees of rats. Further investigation is warranted to better understand the mechanisms influencing spontaneous lesion formation.