The goal when treating intraarticular fractures is restoration of normal articular cartilage contour. Displaced intraarticular fractures form an incongruent surface that increases local contact stresses.1,3–6,11,12,16,18,22 With time, these increased contact stresses may exceed the tolerance of articular cartilage and result in the development of posttraumatic osteoarthritis.7,9,19 Biomechanical studies have shown significant contact surface pressure changes with as little as 1 mm of articular incongruency3,6,19,22 and studies done on the upper extremity have shown an increased incidence of posttraumatic arthritis with 1 to 2 mm of articular incongruency.7,9,20
Using a rabbit model, Llinas et al11 showed that articular cartilage step-offs can nearly completely remodel if the articular defect is less than the full thickness of articular cartilage. In an effort to confirm these observations and determine how the articular surface and subchondral bone remodels after an intraarticular fracture, joint surface changes were examined in a weightbearing sheep model with an intraarticular fracture.18
MATERIAL AND METHODS
After approval from the University of Washington animal subjects committee, a simulated intraarticular fracture was created in the medial tibial plateau of 12 adult domestic sheep. A sagittal osteotomy was used to create a step-off measuring 1 mm. A sagittal osteotomy was used to allow the incongruent portion of the joint to be loaded regardless of joint angle. The average articular cartilage thickness in the medial plateau of sheep tibia averages 1.5 ± 1 mm. A 1-mm articular cartilage step-off was created to allow contact between the two cartilage surfaces. Llinas et al11 showed improved remodeling potential with contact between the cartilage surfaces on either side of the fracture. The osteotomy was created using an oscillating saw to remove a 1-mm thick transverse section of bone 3 cm below the joint line. The vertical portion of the osteotomy, involving the middle of the articular weightbearing surface, was completed with a straight osteotome to avoid removal of articular cartilage. The fragment then was displaced 1 mm distally and fixed with two 3.5-mm cortical screws using standard lag technique.
The opposite limb was left intact and used as a control. The animals were allowed unrestricted activity after the procedure. In all animals, full weightbearing was observed within 3 days. The animals were euthanized at 12 weeks. Bone activity was determined by labeling with calcein at 2 weeks, and with oxytetracycline at 3 weeks and 3 days before euthanasia.
After the sheep in the experimental group were euthanized, pressure-sensitive film in the shape of the medial tibial plateau was placed above the medial menisci and used to measure pressure changes at the articular surface. The joints were loaded with a material testing device from 0–147 N with a crosshead speed of 1.5 mm per second to approximate normal joint loading forces and position during four-legged stance. In two animals, pressure sensitive strips were placed across and perpendicular to the osteotomy, under the medial menisci. After pressure changes were measured, the medial tibial plateau was removed from the experimental limbs for histologic and microscopic studies. The 12 contralateral unoperated control limbs were divided into two groups. In six limbs, pressure sensitive film and strips were placed separately as above, over and under the medial meniscus, to determine normal joint loading. The articular cartilage of the medial tibial plateau then was examined histologically in these specimens and used as the control. In the other six control limbs, the sagittal osteotomy was created and stabilized as above with lag screws. Joint contact pressures were measured, above and below the meniscus, to determine baseline pressure changes after creation of the osteotomy.
Histology and Electron Microscopy
After mechanical testing was completed, the articular surface of the medial tibial plateau along with approximately 2 cm of underlying bone were resected from the specimen using an oscillating saw. The nondecalcified specimens were placed in 70% ethanol and prepared through the process of sequential dehydration in graded ethyl alcohol solutions followed by embedding in methylmethacrylate. For periods of 48 hours, each of the specimens were placed in ethanol concentrations of 70%, 95%, and 100%. The samples were placed in xylene for 24 hours before embedding in 75% methylmethacrylate (Fisher Chemical, Springfield, NJ) and 25% dibutyl phthalate (Fisher Chemical) for 72 hours before transferring the specimens to a solution containing methylmethacrylate and 1.0 mg/100 mL of benzoyl peroxide (Fisher Chemical). The specimens were polyermized in a 42°C radiant heat oven for 5 days. A Reichert-Jung sledge microtome (Cambridge Instruments, Buffalo, NY) was used to prepare 8-μm thick sections for unstained sections used in fluorescent analysis.
Qualitative assessments of osteoid formation were performed using an Osteometrics Video Digitizer attached to the Zeiss (Munich, Germany) fluorescent microscope. The specimens were decalcified in a formic/hydrochloric acid solution. After dehydration in graded ethyl alcohol baths, the specimens were embedded in paraffin. After the embedding was completed, serial sections were cut and stained using Goldner’s and Masson’s trichrome stains.
To prepare the specimens for scanning electron microscopic evaluation, decalcified and dehydrated articular blocks were frozen in liquid nitrogen-cooled isopentane (−165°C).14 The tissue samples were fixed at low temperatures by substitution of the frozen tissue using alcohol-based fixatives. The initial fixative solution was acrolein and 0.2% tannic acid in a mixture of methanol (20%) and acetone (80%) cooled to −79°C in a dry-ice/isopropyl alcohol bath. After 48 hours, the specimen was wished three times with cold acetone and placed in a second fixative solution of 10% glutaraldehyde, 20% methanol, and 70% acetone at −21°C of 48 hours (the stock glutaraldehyde solution was a 50% aqueous solution). The specimens were washed with precooled 98% ethanol solution and slowly warmed to room temperature. Freeze-fracturing (cryofracturing) then was done perpendicular to the plane of the articular surface, yielding samples with surfaces undamaged by the oscillating saw during harvesting. Critical-point drying was done by immersion in t-butyl alcohol followed by lyophilization. The dried samples then were sputter coated with gold for 4 minutes using a sputter coater (Polaron, Halfield, PA).
A repeated measures ANOVA was used to determine whether significant differences existed between any of the three groups tested in the pressure study: control, Time 0 osteotomy, and 12 weeks after osteotomy. Fisher’s PLSD test was used to establish significant differences in the mean values at p < 0.05.
In the control group at Time 0, there were significant changes in contact pressure below the meniscus but not above the meniscus (p < 0.05), with joint contact pressures made before and after the osteotomy. Above the meniscus and before the osteotomy, the pressure averaged 1.20 ± 0.16 MPa adjacent to the osteotomy and the pressure average decreased to 1.04 MPa ± 0.09 after the osteotomy (n = 5). Below the meniscus and before the osteotomy, the pressure averaged 2.11 ± 0.08 MPa and the pressure average decreased to 0.67 ± 11 MPa after the osteotomy (n = 5) (Table 1).
At euthanasia, in the experimental limbs, the average joint contact pressures above the meniscus (1.11 ± 0.12 MPa) were decreased compared with the control group (1.20 ± 0.16 MPa). After the initial data suggested a pressure damping effect of the meniscus, two animals were tested with the pressure strips below the meniscus. A substantial difference was observed between the depressed fracture surface (1.67 MPa) compared with the normal side (2.11 MPa) (Table 1). A statistical comparison was not possible, however, because only two specimens were compared.
Light microscopic evaluation of the five experimental animals showed union and remodeling of the subchondral bone; however, there was no observed healing of the articular cartilage and a persistent discontinuity of the calcified zone and tidemark remained. On the high side of the osteotomy, the articular cartilage thinned as it neared the osteotomy. Compared with sites distant to the osteotomy, the thickness of the articular cartilage on the high side was reduced by approximately 20% (Fig 1). On the low side, the articular cartilage compensated for the step-off with increased chondrocyte cellularity and hypertrophy. The thickness of the articular cartilage on the low side of the osteotomy was 0.38 mm more than the thickness observed on the high side and more than 0.18 mm thicker than the cartilage observed on the low side of the control osteotomy group.
In the experimental group, fluorescent microscopic analysis showed increased bone production on both sides, adjacent to the osteotomy (1.84 μm per day), compared with bone formation distant to the osteotomy (1.42 μm per day). At 2.0 mm from the osteotomy the bone turn-over rate on the low side of the osteotomy (1.85 μm per day) was not significantly different from the bone turnover rate on the high side (1.67 μm per day).
Even though the specimens were fixed in an unloaded state, electron microscopic evaluation in the experimental group showed progressively greater bending of the collagen fibrils toward the osteotomy on the high side (Fig 2). On the low side of the osteotomy, the collagen fibrils were aligned vertically. On higher view, the collagen fibrils on the high side bent beneath the articular surface (Figs 3–5), eventually resulting in a smooth overlapping shelf at the osteotomy site.
In the current study, and in the study of Llinas et al11 of a rabbit medial femoral condyle osteotomy model, articular pressure changes across a 1-mm osteotomy were observed. This is important, because several investigators have shown, in biomechanical models, a correlation between increased articular cartilage pressures after an articular injury and the development of degenerative changes.3,16,22
Using an in vivo, nonweightbearing medial femoral condyle rabbit model, Llinas et al11 showed that in 0.5-mm articular cartilage step-offs that did not exceed the thickness of articular cartilage, the cartilage and subchondral bone remodeled during 12 weeks. Under light microscopy, the cartilage on the high side of the healed osteotomy decreased in thickness, displaced, and tapered toward the depressed side, and ended in a hypocellular tissue flap. In contrast, the cartilage on the depressed side increased in thickness as a result of chondrocyte hyperplasia and cartilage hypertrophy. This remodeling partially restored articular incongruity and improved the pressure differential across the healed osteotomy site. Llinas et al concluded that articular cartilage incongruency can remodel to a nearly normal articular surface if the articular defect is less than the full thickness of articular cartilage. However, the impact of osseous remodeling was not evaluated and the actual contact pressures in the joint were not measured.
In this study, histologic analysis showed thinning of the articular cartilage on the high side of the osteotomy with collagen compression and bending of the collagen fibrils in an attempt to provide a smooth contour at the osteotomy site. On the low side of the osteotomy, increased chondrocyte cellularity and hypertrophy were observed. The resultant relative changes in the articular cartilage thickness, 12 weeks after a 1-mm osteotomy, were substantial. The difference in cartilage thickness between the two sides of the fracture measured 0.38 mm, which is approximately 40% of the thickness of the normal sheep articular cartilage. In addition, the bending of the collagen fibrils was observed in an unloaded state. In normal cartilage, bending of collagen fibrils only occurs when the cartilage is fixed with loads placed on the articular surface using a technique that freezes the joint while the joint surface is compressed. Because the collagen fibrils in this study already were abnormally deformed in the unloaded state, normal joint loading may cause excessive collagen bending that may result in early degeneration of the articular surface. This suggests that the cartilage in our animals may have less resistance to joint wear. Fluorescent microscopic evaluation did not show subchondral bone changes, suggesting that the chondrocyte compression and collagen fibril bending on the high side, in conjunction with chondrocyte hypertrophy and increased cellularity on the low side account for the joint surface leveling.
Although Tang and Chai17 reported extensive chondrocyte death in a rabbit model with a 2.0-mm step-off of the femoral condyle, cellular necrosis was not evident on histologic analysis in this study.
This study showed the important stress-shielding effects of the meniscus and may help explain why fractures in some joints, or in particular knee locations, may be more resistant to developing symptoms and radiographic evidence of degeneration. Although not statistically significant, the joint contact pressures measured above the meniscus during loading were higher on the high side of the 1-mm osteotomy step-off. With only two specimens for comparison, it seemed that below the meniscus, the pressure difference was even greater.
Bai et al4 showed the importance of the meniscus in decreasing contact pressure in a cadaver model. A simulated split fracture of the lateral tibial plateau, with resultant intraarticular step-off heights measuring 1–6 mm, was created. Knee alignment and contact pressures were measured. They showed that with progressively greater articular step-off, valgus angle and contact pressures increased. With a meniscectomy, these valgus angle and contact pressures increased even more.
The sensitivity of a joint to articular incongruity may vary from joint to joint, depending on the modulus of elasticity, the thickness of articular cartilage, and the joint congruity.8 This may explain why recommended maximal values for joint surface incongruity have ranged from 1–10 mm in the lower extremity,2,10,15,21 to 1–2 mm in the radiocarpal joint.7,9,20 It may also explain why patients seem to tolerate a greater amount of displacement in the weightbearing knee than in the wrist.8,13
The relationship between articular surface incongruity and the development of posttraumatic arthritis in the upper extremity was examined by Knirk and Jupiter9 in a retrospective analysis of 43 intraarticular distal radius fractures in young adults, treated by various methods. They reported that 91% of wrists that healed with > 1 mm of incongruity and 100% of wrists that healed with ≥ 2 mm of incongruity had radiographic evidence of radiocarpal osteoarthritis. In contrast, only 11% of wrists that healed with a congruent joint surface had radiographic evidence of radiocarpal osteoarthritis.
Likewise, in a retrospective review of displaced intraarticular distal radius fractures, Trumble et al20 reported a substantial decrease in patients’ functional results at a mean followup of 38 months when a residual gap on postoperative radiographs was greater than 1 mm. Fernandez and Geissler7 reported similar findings in their series of 37 displaced intraarticular distal radius fractures that healed with less than 1 mm of joint incongruency. Only one patient had radiographic evidence of posttraumatic osteoarthritis at a mean followup of 4 years.
Limitations of this study include the short followup, the limited number of specimens, and the single type of displaced fracture pattern in the osteotomy model. The limbs in the experimental group were examined only at one time. With the osteotomy model, there was no articular impaction at the time of injury, as often is seen with clinical intraarticular fractures. Finally, the dampening effects of the meniscus made it difficult to measure articular contact pressures.
Twelve weeks after an intraarticular osteotomy in the midweightbearing portion of the medial tibial plateau in a sheep model, the 1-mm step-off improved secondary to changes in the cartilage thickness and shape. This resulted in a decrease in differential pressure distribution between the normal articular cartilage and the depressed articular cartilage. Despite this improvement in congruency however, the collagen fibrils remained bent and deformed. Because intraarticular incongruency, with resultant joint contact pressure differences, seems to be involved in the development of traumatic arthritis, methods to improve cartilage remodeling potential and restore its structure may improve outcome after intraarticular fractures.
We thank Christopher H. Allan, MD; John Miyano, MD; John M. Clark, MD, PhD; Susan Ott, MD; D. E. Casey Jones, MD; Lt.Col., Patrick Fernacola, MD; Maj., Marteinn Magnusson, MD; and Allan Tencer, PhD for investigative work related to this project.
1. Anderson DD, Bell AL, Gaffney MB, Imbriglia JE: Contact stress distribution in malreduced intraarticular distal radius fractures. J Orthop Trauma 10:331–337, 1996.
2. Anglen J, Healy W: Tibial plateau fractures. Orthopedics 11:1527–1534, 1988.
3. Ausberger S, Necking L, Horton J, Bach AW, Tencer AF: A comparison of scaphoid-trapezium-trapezoid fusion and four bone tendon weave for scapholunate dissociation. J Hand Surg 7A:360–369, 1992.
4. Bai B, Kummer FJ, Sala DA, Koval KJ, Wolinsky PR: Effect of articular step-off and meniscectomy on joint alignment and contact pressures for fractures of the lateral tibial plateau. J Orthop Trauma 15:101–106, 2001.
5. Brown TD, Anderson DD, Nepola JV, et al: Contact stress aberrations following imprecise reduction of simple tibial plateau fractures. J Orthop Res 6:851–862, 1988.
6. Brown TD, Pope DF, Hale JE, Buckwalter JA, Brand RA: Effect of osteochondral defect size on cartilage contact stress. J Orthop Res 9:559–567, 1991.
7. Fernandez D, Geissler W: Treatment of displaced articular fracture of the radius. J Hand Surg 16A:375–383, 1991.
8. Huber-Betzer H, Brown T, Mattheck C: Some effects of global joint morphology on local stress aberrations near imprecisely reduced intra-articular fractures. J Biomech 23:811–822, 1990.
9. Knirk J, Jupiter J: Intra-articular fractures of the distal end of the radius in young adults. J Bone Joint Surg 68A:647–659, 1986.
10. Lansinger O, Bergman B, Köner L, et al: Tibial condylar fractures. J Bone Joint Surg 62A:628–634, 1986.
11. Llinas A, McKellop HA, Marshall GJ, et al: Healing and remodeling of articular incongruities in a rabbit fracture model. J Bone Joint Surg 75A:1508–1523, 1993.
12. Lovasz G, Llinas A, Benya PD, et al: Cartilage changes caused by a coronal surface stepoff in a rabbit fracture model. Clin Orthop 354:224–234, 1988.
13. Marsh JL, Buckwalter J, Gelberman R, et al: Articular fractures: Does an anatomic reduction really change the result? J Bone Joint Surg 84A:1259–1271, 2002.
14. Nöttzli H, Clark J: Deformation of loaded articular cartilage prepared for scanning electron microscopy with rapid freezing and freeze substitution fixation. J Orthop Res 15:76–86, 1997.
15. Rasmussen P: Tibial condylar fractures as a cause of degenerative arthritis. Acta Orthop Scand 43:566–575, 1972.
16. Sangeorzan BJ, Wagner UA, Harrington RM, Tencer AF: Contact characteristics of the subtalar joint: The effect of talar neck misalignment. J Orthop Res 4:544–551, 1992.
17. Tang X, Chai B: Effect of intra-articular fracture on articular chondrocytes: A transmission electron microscopic study. Chin Med J 102:207–213, 1989.
18. Trumble TE, Allan CH, Miyano J, et al: A preliminary study of joint surface changes after an intraarticular fracture: A sheep model of a tibia fracture with weight bearing after internal fixation. J Orthop Trauma 15:326–332, 2001.
19. Trumble TE, Culp RW: Intraarticular fractures of the distal aspect of the radius. Instr Course Lect 48:465–480, 1999.
20. Trumble TE, Schmitt SR, Vedder NB: Factors affecting functional outcome of displaced intra-articular distal radius fractures. J Hand Surg 19A:325–340, 1994.
21. Waddell J, Johnston D, Neidre A: Fractures of the tibial plateau: a review of ninety-five patients and comparison of treatment methods. J Trauma 21:376–381, 1981.
© 2004 Lippincott Williams & Wilkins, Inc.
22. Wagner WF, Tencer AF, Kiser P, Trumble TE: Effects of intra-articular distal radius depression on wrist joint contact characteristics. J Hand Surg 21A:554–560, 1996.