Effects of liquid nitrogen cryotherapy and bone grafting on artificial bone defects in minipigs: a preliminary study

Effects of liquid nitrogen cryotherapy and bone grafting on artificial bone defects in minipigs: a preliminary study

Int. J. Oral Maxillofac. Surg. 2002; 31: 296–302 doi:10.1054/ijom.2001.0210, available online at http://www.idealibrary.com on Research and Emerging ...

271KB Sizes 2 Downloads 53 Views

Int. J. Oral Maxillofac. Surg. 2002; 31: 296–302 doi:10.1054/ijom.2001.0210, available online at http://www.idealibrary.com on

Research and Emerging Technologies Osteobiology

Effects of liquid nitrogen cryotherapy and bone grafting on artificial bone defects in minipigs: a preliminary study

M. A. Pogrel1, J. A. Regezi2, B. Fong1, Z. Hakim-Faal1, M. Rohrer3, C. Tran4, T. Schiff4 1

Department of Oral and Maxillofacial Surgery, Division of Oral Pathology, Department of Stomatology, University of California at San Francisco, San Francisco, CA; 3Department of Oral and Maxillofacial Pathology, University of Oklahoma College of Dentistry, OK; 4 Department of Oral Radiology, University of the Pacific School of Dentistry, San Francisco, CA, USA 2

M. A. Pogrel, J. A. Regezi, B. Fong, Z. Hakim-Faal, M. Rohrer, C. Tran, T. Schiff: Effects of liquid nitrogen cryotherapy and bone grafting on artificial bone defects in minipigs: a preliminary study. Int. J. Oral Maxillofac. Surg. 2002; 31: 283–293.  2002 Published by Elsevier Science Ltd on behalf of the International Association of Oral and Maxillofacial Surgeons. Abstract. Liquid nitrogen cryotherapy has been advocated as an adjunct in the enucleation and curettage of locally aggressive lesions of the jaws. Simultaneous autogenous bone grafting has also been advocated to accelerate bone formation and reduce morbidity. There is, however, relatively little scientific basis for either of these hypotheses. In this study, nine Yucatan minipigs had artificial defects created in the mandible, which were treated with liquid nitrogen spray. Half of the defects were grafted with autogenous bone from the chin and half were closed primarily. Two animals were sacrificed 3 days postoperatively to measure the width of necrosis and the rest were sacrificed at 3 months to assess healing and new bone formation. It was found that drilling the artificial defects alone caused bone necrosis for a mean depth of 0.09 mm. Liquid nitrogen cryospray caused a mean depth of bone necrosis of 0.82 mm (range 0.51–1.52 mm). The defects that were bone grafted healed well clinically. Defects not bone grafted showed a 50% rate of wound breakdown and sequestrum formation with delayed healing. Vital staining showed a non-significantly greater rate of bone formation in the grafted defects. Digitally superimposed radiography showed a non-significantly greater bone density in the non-grafted defects at 3 months postoperatively. It appears that liquid nitrogen cryospray does devitalize an area of bone around defects in the mandible. The width of necrosis is usually less than 1 mm and subsequent healing is enhanced by autogenous bone grafting. This has clinical implications.

Introduction Liquid nitrogen, boiling at 196C, has been advocated as a clinical modality to devitalize tissue20. Sustained temperatures in the tissues below 20C are believed to cause mammalian cell death on a consistent basis18,21. Cell death occurs by a combination of direct damage from intra0901-5027/02/030296+07 $35.00/0

cellular and extracellular ice crystal formation plus osmotic and electrolyte disturbances25. Most protocols advise freezing tissues rapidly to cause the maximum intracellular ice crystal formation and allow the tissues to thaw slowly to cause the maximal electrolyte disturbance since both of these situations result in maximum cell lethal effects10.

Key words: liquid nitrogen cryotherapy; bone grafting; minipigs. Accepted for publication 12 December 2001

For osseous lesions, cryosurgery offers some unique advantages over other treatment modalities since it will kill cells within the bone but leave the inorganic osseous framework intact so it can remain as a matrix for new bone formation1. In this way, cryosurgery provides a treatment modality for devitalizing bone in situ. This is thought to be an acceptable method for treating locally

 2002 Published by Elsevier Science Ltd on behalf of the International Association of Oral and Maxillofacial Surgeons.

Effects of liquid nitrogen cryotherapy aggressive bone lesions that fall short of true malignancy but have a high recurrence rate following local enucleation or curettage only. The rationale for using this therapy is that recurrences may arise from buds or daughter cells from the lesion remaining in the rim of bone around the lesion. In the field of oral and maxillofacial surgery, there are a number of such lesions, including odontogenic keratocysts, intraosseous ameloblastomas, myxomas, some giant cell lesions and some fibro-osseous lesions. In many of these cases, local enucleation results in an unacceptable recurrence rate23, whereas segmental resection may be considered unnecessarily aggressive treatment3. Protocols involving the treatment of these lesions with aggressive local enucleation and curettage followed by treatment with liquid nitrogen cryotherapy have been described with encouraging results1,6,19 It has, however, been recognized that the cryotherapy treatment may increase postoperative wound breakdown and morbidity and even predispose to mandibular fractures6,19. For this reason, simultaneous bone grafting has been advocated for larger mandibular lesions in order to accelerate new bone formation and prevent mandibular fracture19. However, the exact width of bone necrosis that occurs following liquid nitrogen cryosurgery remains in doubt and the criteria for simultaneous bone grafting have not been validated in research studies. Materials and methods Nine mature (all have lost all deciduous teeth) Yucatan minipigs weighing between 31 kg and 44 kg were used for this study. Surgery was carried out under general anaesthesia utilizing ketamine (20 mg/kg) and xylazine (2 mg/kg) with atropine (0.4 mg/kg) for induction of anaesthesia and isofluorane (2–3%) and oxygen via endotracheal intubation for maintenance of anaesthesia. All animals received cefazolin (22 mg/kg) as a prophylactic antibiotic and ketoralac (1 mg/kg) as an analgesic. Standardized

297

Fig. 1. Standardized 3 cm defect created in the pig mandible.

defects 3 cm in anteroposterior diameter and 2 cm in depth were created in the premolar region on each side of the mandible with a pineapple bur and coolant after removing the associated teeth (Fig. 1). One pig was utilized as a control, having had no treatment on one side of the mandible and on the other side, a standardized 3 cm defect was created but with no cryosurgery or grafting. This animal was sacrificed after 3 days to evaluate the histological structure of a normal mandible and to assess any bone necrosis caused by drilling an artificial defect. Two other animals were sacrificed after 3 days, having had defects created on both sides of the mandible with subsequent liquid nitrogen cryospray in order to measure the width of bone necrosis caused by liquid nitrogen spray. All other animals had artificial defects created on both sides of the mandible with liquid nitrogen cryospray, and perforation of buccal and lingual plates with multiple bur holes to possibly aid revascularization. Half of the defects were grafted (n=6) with autogenous bone obtained bilaterally from the chin via an intraoral approach, while the other half were closed primarily without grafting (n=6). These remaining six

animals were sacrificed at three months. The protocol carried out on each animal is documented in Table 1. The cryosurgery was given by means of a Fridgitronics CS 76 cryospray (Cooper Surgical, Shelton, CT, USA) with the spray pressure set at 10 lb/in2. The protocol consisted of a 1-minute freeze followed by a natural thaw, followed by a second 1-minute freeze and thaw. The timing of the freeze was for 1 min following attainment of the lowest sustained temperature. Thermocouples (Cooper Surgical, Shelton, CT, USA) were placed in the centre of each defect and also 1 mm deep to the defect margins in holes drilled into the bone mesially, distally, buccally and lingually prior to the cryospray (a total of five thermocouples). Temperatures were recorded every 15 s in the thermocouples during the liquid nitrogen cryospray and the lowest sustained temperature was recorded, after which the 1-minute freeze was timed. In the animals sacrificed at 3 days, both sides of the mandible were removed, placed in formalin, decalcified, paraffin embedded and 5  sections were cut and stained with hematoxylin-eosin. Measurements were made every 5 mm

Table 1. Protocol carried out on each of the nine experimental pigs Animal No. 1 2 3 4–9

Right side No treatment 3 cm defect and cryotherapy 3 cm defect and cryotherapy 3 cm defect, cryotherapy and bone grafting

The sides of bone grafted in pigs 4–9 were alternated.

Left side 3 cm 3 cm 3 cm 3 cm

defect defect defect defect

only and cryotherapy and cryotherapy only

Sacrificed 3 3 3 3

days (control animal) days (for width of necrosis) days (for width of necrosis) months

298

Pogrel et al.

around the bone defects utilizing a graticule eyepiece on a light microscope. The width of bone necrosis around the defects was calculated utilizing the following criteria diagnostic of cell death5,12: 1. Osteocyte necrosis + More than 50% of lacunae empty + Nuclei not intact in remaining lacunae 2. Marrow non-viability + Capillaries not intact + Collagen not intact or smudgy. The six animals sacrificed at 3 months had vital stains administered intravenously prior to sacrifice. This was given by means of a double-staining technique utilizing tetracycline hydrochloride (Lederle) on days 19 and 18 before sacrifice (1.1% solution, 15 mg/kg) and calcein (fluorescein methyleneiminodiacetic acid–Fisher scientific) on days 5 and 4 before sacrifice (1.1% solution, 15 mg/kg). Both of these compounds cause a staining of the bone that can be visualized under fluorescent light. The tetracycline fluoresces with a lemonyellow colour, while calcein fluoresces with a bright green colour11,15,22,24. At 3-month sacrifice, the portion of the mandible that had contained the artificially created defect was removed. The area containing the artificial defect was sectioned coronally through the centre of the defect. From the area adjacent to the centre of the defect, a 1-mm thick coronal slice was taken through the mandible. This 1 mm slice was fixed in 70% alcohol and dehydrated using increasing concentrations of ethyl alcohol4,7. The specimen was then embedded and ground to a 100  section16. This technique ensures maximal preservation of the vital tetracycline and calcein staining. The 100  section was then polished, viewed under fluorescent light and photomicrographs taken of the tetracycline and calcein staining on each of the 12 specimens at a standardized 10 magnification. These photomicrographs are then projected onto a screen and the distance between the tetracycline and calcein stain is measured every 25 cm on a 6 ft4 ft projection and mean distances between the two stains calculated. This gives a measure of the width of bone generated over the 14-day period between administration of the two stains. The remainder of the specimens removed from the animals sacrificed at 3 months were fixed in formalin, decalcified and 5  sections stained cut

Fig. 2. Radiographic jig positioned over the mandible and attached to the head of the X-ray machine. Table 2. Temperatures reached and width of necrosis

Animal No. 1 left (control) 2 right

2 left

3 right

3 left

Mean temperature at four sites around defect (C)

Mean temperature in centre of defect (C)

21

21

5 10 11 2 5 2 16 2 115 80 120 100 55 90 110 85

50

55

180

160

Width of necrosis (mm) 0.09 (caused by bur alone) 0.076 0.085 0.084 0.076 0.127 0.062 0.173 0.094 0.74 0.51 1.52 1.01 0.63 0.72 0.69 0.74

R= 0.92 (standard error 0.063).

and stained with hematoxylin-eosin for light microscropy. Prior to the initial surgery, after the induction of anaesthesia, standardized radio-opaque markers were placed at three points around the surgical site on all animals to be sacrificed at 3 months. These consisted of size 1/2 round burs, which were driven into the buccal plate of bone, through the soft tissues with a drill, and then broken off. A cold-cured acrylic lower impression was then made of the pig’s occlusion and a customized X-ray jig was attached to it, from which standardized reproducible occlusal-sized films could be taken of the premolar

region of the mandible on each side (Fig. 2). Standardized radiographs were taken prior to surgery, after the extraction of the teeth and creation of the bone defect, and at sacrifice. The radiographs were taken with a Philips Oralix 70 (Philips Medical, Shelton, CT, USA) portable machine. The three occlusal size X-ray films of each defect (12 defects in all–36 radiographs) were captured by means of a standardized scanning and imaging system and the images were then manipulated to be superimposable on the three radio-opaque markers. Once superimposition had been achieved, the dimensions

Effects of liquid nitrogen cryotherapy of the defect were calculated from the X-ray film taken immediately after the defect was created. The mesial-distal and occluso-apical dimensions of the defect were computer calculated directly from the radiograph, while the bucco-lingual dimension of the defect was measured directly from the mandible at the time the defect was created. From these measurements, the volume of the defect created was calculated (in mm3) as was the volume of bone occupying the defect at sacrifice at 3 months. By calculating the amount of bone in the area occupied by the defect before the defect was created, an assessment can be made of the residual bone loss after 3 months. This residual bone loss after 3 months, as calculated from the digitally superimposed radiograph, is utilized to calculate the volume of new bone formation13,14.

299

Fig. 3. A mandible at sacrifice, showing good healing on the grafted left side and a large sequestrum (arrow) on the right side.

Results The mean width of bone necrosis at the different sites for each defect is shown in Table 2. This also shows the correlation with the mean lowest temperatures reached at each site once a steady state has been achieved. The mean width of necrosis caused by the saline cooled bur alone in the control animal (animal No. 1 in Table 2) was 0.09 mm. When the results are examined for animal No. 2 in Table 2, it can be seen that, due to technical problems, the lowest temperatures achieved on the thermocouples in the centre of the defect were 50C on the right side and 55C on the left. This translated into temperatures 1 mm from the edge of the lesion in the bone of between of +5C and 16C. These temperatures are generally not considered low enough to cause tissue necrosis. This is confirmed by examining the width of necrosis caused at these sites which varies between 0.076 mm (which is less than the control animal that received no cryosurgery) and 0.17 mm recorded at the site that reached 16C. This indicates that 16C may cause a small width of bone necrosis. In animal No. 3 in Table 2, where an efficient freeze was achieved with temperatures of 180C and 160C in the centre of the defects (liquid nitrogen boils at 196C), temperatures were achieved in the surrounding bone at a depth of 1 mm, which varied from a high of 55C to a low of 120C. This

Fig. 4. A typical ground section viewed in fluorescent light showing calcein (green) and tetracycline hydrochloride (orange) stain. The distance between the two stains represents the bone formed over a 2-week period.

produced widths of necrosis in the surrounding bone that varied from 0.51 mm adjacent to the 80C thermocouple to 1.52 mm adjacent to the 120C thermocouple. These widths are significantly different from those in the control animal and in the inadequately frozen defects in animal No. 2 shown in Table 2. There is a positive correlation between the temperature achieved and the width of the resulting necrosis in the bone (r= 0.92; standard error [SE] 0.063, P<0.0001). For the six animals sacrificed at 3 months, of those that received simul-

taneous bone grafts (n=6), no defect showed any signs of wound breakdown or dehiscence. Of the six defects that did not receive bone grafts, three showed evidence of wound breakdown and sequestration of bone (Fig. 3). A typical fluorescent histomicrograph is shown in Fig. 4, showing the lines of tetracycline and calcein deposition. Mean distances between these lines are shown in Table 3. In arbitrary units, the mean distance between the two vital stains in the grafted defects was 2.72 and that in the non-grafted defects was 2.39. This difference is not statistically significant.

300

Pogrel et al.

Table 3. Mean width of bone formation over a 14-day period as measured from vital staining

Animal No. 4 5 6 7 8 9

Side

Graft status

Mean width of bone formed over 2-week period (arbitrary units)

Left Right Left Right Left Right Left Right Left Right Left Right

Grafted Non-grafted Non-grafted Grafted Grafted Non-grafted Non-grafted Grafted Grafted Non-grafted Non-grafted Grafted

2.98 2.54 1.76 3.31 2.22 2.64 2.90 2.31 2.33 1.43 3.09 3.22

Units are arbitrary from projected slides and are for comparative purposes only.

graphs, is shown in Table 4. Of the 12 sides available on the six animals, six were available for analysis. Three defects could not be accurately subtracted since teeth had erupted into the defects during the postoperative 3-month period and the other three were the non-grafted defects that developed sequestrae, and so bone volumes could not be calculated. Of the three defects into which teeth erupted, one had been grafted and two had not been grafted. Presumably, these three defects had refilled with bone to a suitable extent for teeth to migrate and erupt into them. Of the six defects that were available at 3 months for subtraction radiographic analysis, four sides had been bone grafted and two had not (Fig. 6). In the two animals where both sides were available for analysis (Nos 4 and 7), in both cases the non-grafted side had a greater bone volume than the grafted side at 3 months. In no case had the bone volume at 3 months reached preoperative levels. Also, if the mean volumes at 3 months are taken, then the density in the non-grafted sides is greater than that in the grafted sides (2274.13 mm3 vs 1798.58 mm3). These differences do not approach statistical significance. Discussion

Fig. 5. Ground section showing native mandible (black arrow) and new bone (white arrow) formed in defect after grafting at 3 months (hematoxylin-eosin 10). Grafted and non-grafted sides were similar.

Figure 5 shows the typical interface of normal bone and the bone formed in the defect at 3 months. There was no subjective difference between the appear-

ance of the bone on the grafted and non-grafted sides. The amount of bone formed in the defects, as calculated from the radio-

The minipig animal model was selected for this study because the bone turnover rate and occlusion are similar to those in humans and the anatomical and physiological parameters of minipigs closely resemble those of the human. Additionally, the mandible is of a sufficient size to allow realistic studies to be performed8,9,17. By sacrificing the animals at 3 months, the defects had almost totally refilled with new bone to such an extent that it was not possible to determine a statistically significant difference in the bone formation rate between those

Table 4. Amount of bone removed at surgery (mm3) and amount regenerated in the 3-month postoperative period on six animals sacrificed at 3 months Amount of bone (mm3)

Pig No. 4 5 6 7 8 9

Side

Preoperative quantity of bone

Amount of bone removed

Postoperative quantity of bone

Amount of bone formed

Quantity of bone at 3 months

Net bone loss

Bone grafted

L L L L L L

2320.5 4263 1790.25 3472 2352 2072

1326 2436 1023 1984 1344 1184

994.5 1827 767.25 1488 1008 888

748 N/A 838.2 1600 857.6 N/A

1742.5 N/A 1605.45 3088 1865.6 N/A

578 N/A 184.8 384 486.4 N/A

Yes No Yes No Yes No

There are no results for animal No. 5 (both sides) and No. 9 (right side) since teeth erupted into the space and made digital superimposition impossible. No. 6 (right side), No. 8 (right side) and No. 9 (left side) had sequestra and could not be evaluated. NA: not applicable.

Effects of liquid nitrogen cryotherapy

Fig. 6. A typical standardized occlusal-sized film of the defect at 3 months. Two of the three radio-opaque markers are seen.

defects that were grafted and those that were not. It is possible that sacrifice at an earlier age may have revealed a difference. This study does indicate that liquid nitrogen cryosurgery will devitalize a rim of bone around a defect and that once a steady state is achieved, temperatures in the order of 55–120C can be achieved a millimeter or so from the edge of the defect. Bone necrosis, however, only appears to extend between 0.51 and 1.52 mm from the edge of the defect. This is less than previous studies have suggested (B & F 19751, B 19862) and may cause some revision in the use of this technique for malignant lesions. However, a margin of 0.51–1.52 mm may be adequate to prevent recurrence of locally aggressive lesions where any seedlings may be within this distance from the edge of the defect. This study also shows that autogenous bone grafts placed in defects treated by cryosurgery will incorporate and become revascularized. Clinically, bone grafting did improve the healing of these defects in that none of the grafted defects broke down, whereas 50% of the non-grafted defects broke down, leading to bony sequestrae. The vital staining did not indicate any statistically significant difference in the rate of bone formation. The superimposition radiography did not indicate any greater bone density at 3 months in the sides that were grafted vs the nongrafted sides. This is difficult to explain since clinically the grafted sides did better. It may, however, be that healing

may be complete at this stage in the minipig model and that a significant difference in one direction or the other may be apparent at an earlier stage of healing. This study should be considered preliminary and further studies need to be carried out on this animal model and the effects of cryotherapy on bone and its healing. Nevertheless, the clinical implications of this study are: 1. Liquid nitrogen cryotherapy does devitalize bone around a defect. The width of bone necrosis is between 0.51 and 1.52 mm with the freezing protocol employed. 2. Simultaneous bone grafting does improve clinical healing of these defects. 3. Simultaneous bone grafting does not appear to increase the bone density or rate of bone formation in these defects at three months postoperatively in the minipig animal model. Acknowledgments. We acknowledge research assistant Hari Prasad, BS and Tom Sprowl of the Hard Tissue Research Laboratory at the University of Oklahoma College of Dentistry for the undecalcified histotechnology. This research was funded by an American Association of Oral and Maxillofacial Surgery (AAOMS) research support grant and NIDR grant R03-DE11189. References 1. B PF. The cryosurgery of bone in the maxillofacial region. In: Cryosurgery of the Maxillofacial Region. Vol. 2. Boca Raton, FL: CRC Press 1986: 55–91.

301

2. B PF, F AD. The cryosurgery of bone, an experimental and clinical assessment. Brit J Oral Surg 1975: 13: 111–127. 3. B PA. Treatment of cysts of the jaws. Proc R Soc Med 1971: 64: 547–550. 4. C RF, L J. Tetracycline bone labelling: An improved technique using incident fluorescence. Ann Otol 1982: 91: 160–162. 5. E WED. The Chemistry of Death. Springfield, IL, USA: Charles C. Thomas 1963. 6. F AD, W DF, B PF. The effect of cryosurgery on the strength of bone. Br J Oral Surg 1977: 15: 215–222. 7. F HM. Tetracycline based analysis of bone remodelling. Calif Tissue Res 1969: 3: 211–217. 8. F K, H-S Y, C MV, H BR, P RJ, J IT. The viability of revascularized calvarial bone graft in a pig model. Am Plast Surg 1992: 29: 136–142. 9. G D, G JE, MK H, C D. Is RAP responsible for higher bone activity around implants? J Dent Res 1992: 71(Suppl 1 Abstract 1466: 289. 10. L P PJ. Basic consideration of the cryolesion. In: Ablin RJ, ed: Handbook of Cryosurgery. New York: Marcel Dekker Inc., 1980: 22. 11. L RD, F HM. Histopathologic Technique and Practical Histochemistry. New York: McGraw Hill Book, Co, 1976. 12. M TI, P VP. A review of tissue and organ viability assay. Cryobiology 1967: 4: 106–115. 13. O T, O M, M T, S H, I I, Y N. Quantitative evaluation of proximal bone lesions using digital subtraction radiography. Dentomaxillofac Radiol 1988: 17: 99–103. 14. R UE, W RL. Volumetry of localized bone lesions by subtraction radiography. J Perio Res 1987: 22: 215–216. 15. R RR. Bone biopsy and histomorphometry in clinical practice. Rheum Dis Clin N Am 1994: 20: 609–627. 16. R MD, S CC. The cuttinggrinding technique for histologic preparation of undecalcified bone and bone-anchored implants: Improvements in instrumentation and procedures. Oral Surg, Oral Med, Oral Pathol 1992: 74: 73–78. 17. R EH, V LM, K GA. The panniculus carnosus in the domestic pig. Plast Reconst Surg 1977: 59: 94–97. 18. R G, V IY. Cryosurgery for basal cell carcinoma of the head and neck. S Afr Med J 1979: 56: 26–27. 19. S DA, P MA. Liquid nitrogen cryosurgery and immediate bone grafting in the management of aggressive

302

Pogrel et al.

primary jaw lesions. J Oral Maxillofac Surg 1995: 53: 784–790. 20. S A, B D. A historical account of the technical means used in cryotherapy. Min Med 1974: 65: 3718–3722. 21. S JJ, F J. An estimation of tissue damage and thermal history in the cryolesion. Cryobiology 1974: 11: 139–147. 22. S AJ, S DA. Use of the fluorochromes xylenol orange, calcein green, and tetracycline to document bone deposition and remodeling in healing

fractures in chickens. Avian Dis 1992: 36: 447–449. 23. T PA. Newer concepts of odontogenic cysts. Int J Oral Surg 1972: 1: 3–16. 24. T RW, B CT, K E, P SR. An in vitro study of electrical osteogenesis using direct and pulsating currents. Clin Orthop Rel Res 1979: 145: 30–36. 25. W DK. Mechanisms of tissue destruction following cryosurgery. Ann R Coll Surg Engl 1984: 66: 313–318.

Address: M. Anthony Pogrel Department of Oral and Maxillofacial Surgery University of California, San Francisco 521 Parnassus Avenue, C-522 San Francisco, CA 94143-0440 USA Tel: +1 (415) 476 8226 Fax: +1 (415) 476 6305 E-mail: [email protected]