Autogenous cultured growth plate chondrocyte transplantation in the treatment of physeal injury in rabbits

Surgical procedures

A total of 14 New Zealand white rabbits at the age of five weeks were selected for the experiment. Each animal was assigned a number from one to 14 (experimental rabbits in the subsequent part of the article will be referred to as samples). The surgery was performed twice on both the left and right tibias of each rabbit on two different days. Every animal of the group was operated on within one day. The surgery was performed under general anaesthetic with the implementation of subcutaneous atropine sulphate (0.1 mg). After ten minutes, an intramuscular injection of xylazine hydrochloride (0.2 mg/kg) was given. The puncture of the marginal ear vein cannulation, the cannula (0.7) and administration of a thiopental intravenous drip (25 mg/kg) were used, and both the respiratory rate and the heart rate were monitored (Datex Ohmeda Inc., Madison, Wisconsin). The skin over the knee was shaved and prepared using Skinsept Ecolab (Düsseldorf, Germany) and the area was draped. A 1.5 cm medial incision over the proximal part of the tibia was made in order to get to the right and the left physes. The periosteum was incised with a surgical scalpel blade and elevated gently from the medial aspect of the proximal part of the tibia. After a visual localisation of the growth plate at the right tibia, bone needles were used to remove the growth plate cartilage. A standard defect was thus created by removing a quarter of the proximal tibial physis, and inserting a standard drainage tube (Table I). The normal scale of the growth plates fuses at six to seven months of age.¹⁰ The culture medium is standard for chondrocyte culture as described by Peterson et al,¹¹ however, this has since been modified.¹²

At the left tibia, the same procedure of the physis excision was made (Table II). The left tibia, however, (samples 1 to 7) was filled with a drainage tube, (samples 8 to 11) the site of the physis resection was left without a drain, (samples 12 to 14) and no defect in the physis was created. The wound was closed with an interrupted suture using PDS 3-0 (Ethicon, Johnson & Johnson, Norderstedt, Germany). The mean time of the surgery was 25 minutes (12 to 37). The animals were allowed to bear weight and walk freely in the cages without a splint. A microscopic control (Olympus type BX51, Olympus America Inc., Melville, New York) was used in all cases to ensure that the physeal cartilage had been removed, completely exposing the epiphyseal and metaphyseal bone using haematoxylin and eosin staining. In all cases, the presence of the growth plate cartilage was confirmed.

A total of 22 days after the first procedure caused damage to the physis, the second surgery, using the same method of anaesthetic and the same incision, was performed. A procedure was then developed to remove the drainage tube and to transplant autologous physeal cells into the focal defect in the medial part of the proximal right tibia. Each pre-implantation graft was controlled by surgeons in order to maintain accuracy of the numbering of the graft and the animal. After tube drain ablation, the shape of the graft was contoured to match the shape in the tibia’s defect. The graft was not fixed in situ, but a gentle periosteum suture was made using PDS 4-0 (Ethicon). Each graft was controlled for stability of fixation before the skin was sutured. At the left tibia, the graft of autologous growth plate cells was not used, however, the same type of incision was made and the wound closure was the same as at the right tibia. In samples 1 to 7 the drainage tubes were removed.

After the second surgery, the animals were allowed to walk freely in their cages. After 60 days the experimental animals were killed by intravenous administration of Morbital (200 mg/kg) (Biowet Ltd, Puławy, Poland).

Chondrocyte culture

The growth plate samples were placed in a sterile tube containing Ham's F12 (Mediatech Inc., Herndon, Virginia) with 50 g/ml gentamycin sulfate (PAA, Linz, Austria) and 2 g/ml amphotericin B (PAA). All samples were transported to the Cell Culture Laboratory Tissue Bank in Katowice, Poland. The next day a fragment of the cartilage was washed and cut into small pieces. The specimen was left in Ham's F12 (Mediatech) medium, supplemented as described above with 0.1% collagenase type II (Worthington, Lakewood, New Jersey), for six to seven hours. After digestion, the cell suspension was washed in the Ham’s F12 medium (Mediatech) once, and inoculated into flasks in a DMEM/HamF12 (PAA) medium containing 10% NCS (PAA), with 50 g/ml gentamycin sulfate (PAA), 2 g/ml amphotericin B (PAA) and 50 g/ml L-ascorbic acid (PLIVA, Kraków, Poland). Cells were cultured in an incubator at 37°C in a 5.5% CO₂ environment. Within a few hours, seeding cells adhered to the bottom, and about three days later they started to divide. The culture medium was exchanged every 48 hours. Proliferation was monitored using inverted microscopy (Carl Zeiss, Oberkochen, Germany). When the cells were 80% confluent they were passaged to new culture flasks. The day before transplantation into the growth plate defect, cell grafts in fibrinogen were prepared. Human fibrinogen was prepared from Regional Blood Center voluntary donors. Plasma was frozen at -80°C and thawed twice in a water bath. Concentrated fibrinogen was centrifuged at 4°C at 1000 x g in a Heraeus Cryofuge (Thermo Scientific, Wilmington, Delaware) for 15 minutes. Supernatant was discarded and the pellet of concentrated fibrinogen was used for the preparation of fibrin glue. The donor’s blood (plasma) was then tested for the presence of infectious agents including HIV, Hepatitis B, Hepatitis C and TP-syphilis. Before surgery the chondrocyte suspension was mixed with fibrin glue to prepare the fibrograft. Cultured cells were trypsinised, and then washed twice in DMEM/Ham’s F12 (Mediatech) medium supplemented with 10% NCS serum. The cell pellet was mixed with fibrinogen and tranexamic acid, then thrombin dissolved in calcium chloride (10%) was added. The graft was adjusted to fit the size and thickness of the growth plate defect in the tibia. Prepared grafts were then transported to the operating room. The time of culture was 22 days, and 14 autologous chondrocyte growth plate cultures were made. In all 14 samples, the presence of growth plate chondrocytes was confirmed before implantation. Before grafting, each was controlled in a histological examination using S-100 (Fig. 1). When the cells were 80% confluent they were passaged to new culture flasks. The day before transplantation into the growth plate defect, cell grafts in fibrinogen were prepared. Human fibrinogen was prepared from Regional Blood Center voluntary donors. Plasma was frozen at -80°C and thawed twice in a water bath. Concentrated fibrinogen was centrifuged at 4°C at 1000 x g in a Heraeus Cryofuge (Thermo Scientific, Wilmington, Delaware) for 15 minutes. Supernatant was discarded and the pellet of concentrated fibrinogen was used for the preparation of fibrin glue. The donor’s blood (plasma) was then tested for the presence of infectious agents including HIV, Hepatitis B, Hepatitis C and TP-syphilis. Before surgery the chondrocyte suspension was mixed with fibrin glue to prepare the fibrograft. Cultured cells were trypsinised, and then washed twice in DMEM/Ham’s F12 (Mediatech) medium supplemented with 10% NCS serum. The cell pellet was mixed with fibrinogen and tranexamic acid, then thrombin dissolved in calcium chloride (10%) was added. The graft was adjusted to fit the size and thickness of the growth plate defect in the tibia. Prepared grafts were then transported to the operating room. The time of culture was 22 days, and 14 autologous chondrocyte growth plate cultures were made. In all 14 samples, the presence of growth plate chondrocytes was confirmed before implantation. Before grafting, each was controlled in a histological examination using S-100 (Fig. 1) for 15 minutes. Supernatant was discarded and the pellet of concentrated fibrinogen was used for the preparation of fibrin glue. The donor’s blood (plasma) was then tested for the presence of infectious agents including HIV, Hepatitis B, Hepatitis C and TP-syphilis. Before surgery the chondrocyte suspension was mixed with fibrin glue to prepare the fibrograft. Cultured cells were trypsinised, and then washed twice in DMEM/Ham’s F12 (Mediatech) medium supplemented with 10% NCS serum. The cell pellet was mixed with fibrinogen and tranexamic acid, then thrombin dissolved in calcium chloride (10%) was added. The graft was adjusted to fit the size and thickness of the growth plate defect in the tibia. Prepared grafts were then transported to the operating room. The time of culture was 22 days, and 14 autologous chondrocyte growth plate cultures were made. In all 14 samples, the presence of growth plate chondrocytes was confirmed before implantation. Before grafting, each was controlled in a histological examination using S-100 (Fig. 1) was then tested for the presence of infectious agents including HIV, Hepatitis B, Hepatitis C and TP-syphilis. Before surgery the chondrocyte suspension was mixed with fibrin glue to prepare the fibrograft. Cultured cells were trypsinised, and then washed twice in DMEM/Ham’s F12 (Mediatech) medium supplemented with 10% NCS serum. The cell pellet was mixed with fibrinogen and tranexamic acid, then thrombin dissolved in calcium chloride (10%) was added. The graft was adjusted to fit the size and thickness of the growth plate defect in the tibia. Prepared grafts were then transported to the operating room. The time of culture was 22 days, and 14 autologous chondrocyte growth plate cultures were made. In all 14 samples, the presence of growth plate chondrocytes was confirmed before implantation. Before grafting, each was controlled in a histological examination using S-100 (Fig. 1) medium supplemented with 10% NCS serum. The cell pellet was mixed with fibrinogen and tranexamic acid, then thrombin dissolved in calcium chloride (10%) was added. The graft was adjusted to fit the size and thickness of the growth plate defect in the tibia. Prepared grafts were then transported to the operating room. The time of culture was 22 days, and 14 autologous chondrocyte growth plate cultures were made. In all 14 samples, the presence of growth plate chondrocytes was confirmed before implantation. Before grafting, each was controlled in a histological examination using S-100 (Fig. 1) was added. The graft was adjusted to fit the size and thickness of the growth plate defect in the tibia. Prepared grafts were then transported to the operating room. The time of culture was 22 days, and 14 autologous chondrocyte growth plate cultures were made. In all 14 samples, the presence of growth plate chondrocytes was confirmed before implantation. Before grafting, each was controlled in a histological examination using S-100 (Fig. 1). Articular chondrocytes are released from the surrounding matrix by enzyme isolation and then expanded during cell culture in a special cell medium as primary cells in culture are chondrocyte- differentiated somatic cells.

Fig. 1

Histological examination of the S-100 cultured chondrocytes (x 400).

Tissue examination

The tissues were taken for histological and radiological examination. The tibia and femur were excised and fixed in 10% formalin. The radiograph was performed on the day the animals were killed using an X-ray machine (Siemens type Multix Top, Munich, Germany). Both an anteroposterior and lateral view allowed for the bone bridge’s appearance and angular deformation evaluation. The angle between the longitudinal axis of the tibia and a line parallel to the tibial condyles was measured for each (n = 28). All downloaded bone material was decalcified using TBD2 (Shandon, Thermo Scientific, Waltham, Massachusetts). Sections 5 µ thick were frontally cut through the grafted area of the tibia before visual localisation of the grafted area was established. The histological examination was undertaken by three pathologists (RT, AG, AW) and haematoxylin and eosin staining was made.

Histological findings

Following the grafting of the autologous physeal cartilage (samples 1 to 5, 11, 13), the right tibia presented a regular shape of plate growth with hypertrophic maturation, chondrocyte columniation and endochondral calcification. No inflammatory reaction was observed.

Following the grafting of the autologous physeal cartilage (samples 7 to 9), the right tibia presented an irregular physeal shape with the lacunae but with normal morphological changes during the process of cartilage differentiation and no bone bridge formation. After the grafting of the autologous physeal cartilage (sample 14), the right tibia presented with bone bridge formation at the operated medial side of approximately 15% of the growth plate width, however, the remaining grafted cartilage on another 10% of the growth plate width showed an irregular growth plate with the lacunae.

In those samples without grafting of the autologous physeal cartilage (1 to 11), the left tibia displayed bone bridge formation over growth plate resection. However, the left tibia in samples 12 to 14 (no surgery) presented no changes in plate growth (Fig. 2), the left tibia displayed bone bridge formation over growth plate resection. However, the left tibia in samples 12 to 14 (no surgery) presented no changes in plate growth (Fig. 2) presented no changes in plate growth (Fig. 2). The difference between the mean values of the number of cells in both the right tibias and the left was 475.17. In the right tibia there were 66% more cells in 1 mm² of the plate than there were in the left tibia.

Fig. 2

Microscopic image of a tibia after grafting of an autologous cultured growth chondrocyte.

Radiological findings

A radiograph of the tibia at both the anteroposterior and the lateral view presented a regular visible growth plate at the site of grafting at the right tibia (samples 1 to 6, 11, 13). The growth plate was irregular with a folded shape (samples 7 to 9). The growth plate in sample 14 presented the appearance of an irregular growth plate (Fig. 3). The growth plate was irregular with a folded shape (samples 7 to 9). The growth plate in sample 14 presented the appearance of an irregular growth plate (Fig. 3). The growth plate in sample 14 presented the appearance of an irregular growth plate (Fig. 3) with bone bridge formation at the medial side covering about 15% of the growth cartilage (Fig. 4).

Fig. 3

Radiograph showing the right tibia after autologous cultured growth chondrocytes with no angular deformity, and the left tibia after partial resection of the plate growth without grafting varus deformity.

Fig. 4

Diagram showing how the angular deformity is measured between the long axis of the tibia and the tibial plateau (anteroposterior view on the left).

In samples 1 to 11 at the left tibia a bone bridge was presented and it filled a mean of 45% (35% to 70%) of the growth plate. The growth plate in samples 12 to 14 was observed without abnormality. The results of the right and left tibial angular deformity are listed in Table III, and differences between them in Figure 5.

Fig. 5

Diagrams of the results of right (top graph) and left (bottom graph) tibial angular deformity

Statistical analysis

The means and standard deviations were calculated for tibial right and left varus deviation and angular deformity at the lateral view. The values were statistically analysed using STATISTICA version 10 (StatSoft Inc., Tulsa, Oklahoma). A brief analysis of the data reveals some significant results (Table IV).

The mean tibial deformity of the left angle was 20.29° and the right was 7.21°. The median value in the left tibia was 30º, which indicates that 50% of the results were above this value, and the other half was below. The median of the right tibial angle was 5° when compared with the angular deformity of the left tibia. The high rate of standard deviation shows differentiation of the obtained results. Of the samples of the left tibia, 67% show the reduction in the plateau in different parts of the value of 50°. The right reduction occurred in 78% of the tested samples and showed a considerably lower value, with an average of 12°. Results are presented in Table V.