Difference in bone ingrowth after one versus two daily episodes of micromotion: Experiments with titanium chambers in rabbits

Stuart good man,'^'^* Jian-Sheng Wang,’ Amol Doshi? and Per Aspenberg’

Stanford, California Department of Orthopedics, University Hospital, Lund, Sweden; and ’Stanford University Medical Center,

Mechanical stimulation has been shown to affect the differentiation and development of mesenchymal tissue. In the present study, we compared the histolopal and histomorphometric results of tissue ingrowth into micromotion chambers that were moved at 0 cycles per day, 20 cycles once per day, and 20 cycles twice per day over 20-30 sec, for 3 weeks. In each case, a chamber having a 1 X 1 X 5 mm square-holed groove for tissue ingrowth was used. The total amplitude of motion was 0.75 mm. Histological sections from nonmoved chambers contained extensive trabecular bone, embedded in a fibrovascular stroma. Histomorphometric analysis dis- closed that bone comprised a mean of 31 k 2% (mean 2 SEM) of the ingrown tissue. Twenty movements per day appeared to further stimulate bone ingrowth (46 2 5%).

Extensive ingrowth of more immature woven and tra- becular bone was noted in a more cellular stroma. In general, increasing the degree of micromotion to 20 movements twice per day resulted in a decreased amount of bone formation (19 k 7%). In several of these specimens, little or no bone could be found. These experiments have demonstrated that, for the parameters chosen in this study, a short daily period of low frequency, micromotion may facilitate bone ingrowth; however, when the same motion is delivered twice daily, bone ingrowth is depressed. Thus a ”window” of externally applied strain appears to exist, which may facilitate or discourage tissue differentiation to bone. 0 1993 John Wiley & Sons, Inc.

INTRODUCTION

Excessive motion at the bone-implant interface has been shown to have a detrimental effect on the amount of bone ingrowth into porous coated i m ~ 1 a n t s . l ~ In an attempt to understand the effects of interfacial micromotion, we have modified the bone harvest chamber5 to form the micromotion chamber (MC), which we then implanted in the proximal tibia of rabbits.Gs Using this model, the quality and quan- tity of tissue ingrowth into a single 1 mm pore can be observed at periodic intervals in the presence of specific, discrete daily applications of motion of a predetermined nature.

Previous studies by the authors using the MC have shown that different parameters of motion can affect

*To whom correspondence should be addressed at Stanford University Medical Center, Division of Orthopaedic Surgery #R171,300 Pasteur Drive, Stanford, CA 94305-5326.

the nature of the tissue growing into the pore. These parameters include the configuration of the hole in the cylinder and the amplitude of motion.6-8 Nonmoved chambers consistently show ingrowth of trabecular bone after 3 weeks.6 Using a chamber with a cylinder containing a round hole, short daily periods of motion (20 cycles per day, 0.5 mm in amplitude, administered over approximately 30 sec) for 3 weeks inhibited the ingrowth of bone into the pore.6 However, when the pore in the outer cylinder was changed from a round configuration to a square configuration, the same motion parameters did not inhibit the ingrowth of bone.7

Studies by other authors have shown that short periods of micromotion can promote bone accretion; excessive microstrain can have the opposite effect.”” In this study, we perform a histomorphologic and morphometric analysis of tissue ingrowth into MCs that were not moved or moved 20 cycles oncejday, or 20 cycles twice/day. In all cases, a chamber with a square holed cylinder and core was used, and the

Journal of Biomedical Materials Research, Vol. 27, 1419-1424 (1993) 0 1993 John Wiley & Sons, Inc. CCC 0021-9304/93/111419-06

1420 GOODMAN ET AL.

amplitude (0.75 mm) and frequency of micromotion were constant.

MATERIALS AND METHODS

The micromotion chamber (MC) has been described elsewhere (Fig. 1).6-8 Briefly, the chamber is implanted in the tibial metaphysis of mature rabbits and is com- posed of a fixed outer chamber (which osseointegrates with the surrounding bone) and an inner removable core. Both the cylinder and the core have a transverse 1 mm wide pore for tissue ingrowth. The inner core contains a 1 X 1 X 5 mm groove that coincides with the bottom plate of the outer cylinder and its holes, providing a continuous canal through the chamber for tissue ingrowth. After a 6 week period of os- seointegration, the inner core can be manually rotated with respect to the fixed, outer cylinder according to a specified protocol. The rotational maneuvers are carried out manually by grasping the subcutaneous horns of the chamber; the horns attach to the chamber cap that attaches to, and moves, the core relative to the fixed outer cylinder. The amplitude of the movement is determined by the size of a stop-screw. Micromotions are applied at a rate of approximately 0.67-1 cycle/sec. Initially, these motions were timed, but as the investigators become more adept with the technique, the manipulations were performed without timing, but with constant counting.

Animals and operations

Seven MCs were inserted into the proximal tibial metaphysis of mature lop-eared rabbits. After the 6 week period of osseointegration, the contents of the core were harvested for the first time. The chambers were then manipulated daily (20 cycles/day over a 20-30 sec period) beginning the third day after harvest. After 3 weeks, the chambers were harvested, then randomly assigned to either another period of manipulation or no manipulation, and harvested again. This protocol was followed repeatedly. Later in the series, another manipulation protocol was included, namely 20 cycles delivered twice/day, once in the morning and once in the afternoon. If infection of the implant site was diagnosed or if the rabbit became ill, the animal was killed.

Evaluation

The harvested tissue was fixed in 10% buffered formalin, decalcified, embedded in paraffin, cut into 5 p m sections and stained with hematoxylin and eosin. Histomorphologic analysis was performed of multiple serial, longitudinal sections, enabling the entire length of the specimen to be visualized on one slide. The Videoplan system (Kontron Bildanalyse GMBH, Germany) was used to perform an analysis of the percentage of trabecular and woven bone in a longitudinal section from center of the specimen.

Statistical analysis

Descriptive statistics were calculated for the per- centage of bone ingrowth in the three different groups. The Kruskall-Wallis test was applied to test whether there were differences in the amount of bone ingrowth between three groups. Intergroup comparisons were then made using the Mann-Whitney test. Nonpara- metric statistics were used in the analysis because the number of animals in the study was only seven and the authors could not assume that the percent- age of bone ingrowth in this model has a normal distribution.

RESULTS Figure 1. The components of the micromotion chamber used in this study include an outer cylinder that is closed at one end and open at the other to receive the inner core. The cylinder is pierced by two 1 mm x 1 mm square holes coinciding with a transverse groove across the inside of the core. When implanted in bone, this provides a continuous canal for tissue ingrowth. The core inside the cylinder is connected to a cover or cap containing two

respect to the fixed cylinder. The spring fits between the

Seven 'pecimens were harvested from five in which there was no micromotion applied. Ten specimens were obtained from six animals having chambers moved at 20 cycles per day. Seven speci- mens were harvested from five animals in which the chambers were moved 2o cycles~ twice I).

The specimens harvested from mnmoved cham- horns, enabling manual manipulation of the core with

- - core and cover. bers contained extensive trabecular bone covered by

TITANIUM CHAMBERS AND BONE INGROWTH 1421

TABLE I Percentage of Bone Ingrowth for the Specimens from Micromotion Chambers with Different Daily Motion

Protocols: Data on Each Animal in Chronological Order

3

Daily Motion Animal # (cycles/day) Percent Bone Ingrowth

1 20 26 2 20 65

20 54 20 65 20 36 0 32

20 x 2 23 0 31

20 61 20 44

0 36 20 x 2 0

0 20 4 20 22

20 x 2 6 0 33

20 x 2 50 5 0 33

20 47 6 0 35

20 42 20 x 2 6 20 x 2 28

7 20 x 2 24

plump osteoblasts (indicating an active rather than a resting metabolic activity of these cells) embed- ded in a fibrovascular stroma (Fig. 2). The trabecula were generally aligned parallel to the long axis of the chamber. Woven bone constituted a minor por- tion of the bone present. The sections from chambers moved 20 cycles once/day contained a more extensive amount of both trabecular and woven bone compared to the 0 cyclesjday (cpd) group (Fig. 3). This bone appeared to be thicker compared to the bone in non- moved chambers. Furthermore, the bone was aligned in a more random fashion within the chamber. The osteoblasts surrounding or embedded within the bone were plump. The fibrous interstitial tissue surround- ing the trabecular and woven bone was extremely cellular and contained numerous plump mesenchy- ma1 cells with prominent nuclei and nucleoli. The mesenchymal cells were frequently located adjacent to pink-staining, amorphous material. The tissue was ex- tremely vascular, with numerous capillaries coursing throughout the fibrous stroma. There was no evidence of acute or chronic inflammation. The specimens from the group subjected to 20 cycles twice daily contained a less extensive amount of bone compared to the 0 or 20 cpd groups (Fig. 4). Indeed, in several of the specimens, little or no bone could be found. In gen- eral, the specimens contained longitudinally oriented fibrous tissue with many intervening capillaries. The stroma exhibited numerous mesenchymal cells. The

(B) Figure 2. Specimen from a nonmoved micromotion chamber. Note the extensive amount of bone ingrowth in a fibrovascular stroma. Hematoxylin and eosin stain. A: 40X magnification originally. B: lOOX magnification originally.

bone present was generally oriented in a longitudinal fashion and was both woven and trabecular. The trabeculae seemed shorter and thinner compared to the other two groups. There was no evidence of inflammation or foreign body reaction. In one section from this group, the bone ingrowth was as extensive as that in the other groups.

The histomorphometric results are listed in Table I. The amount of bone ingrowth (percent of section area) averaged 31 2 2% (mean -+ S.E.M.) with a range of 20-36% for the nonmoved group (“0”), 46 2 5% (range = 22-65% for the 20 cycles/day group (“20”), and 20 5 7% (range 0-50%) for the 20 cycles twice/day group (“20 x 2”). The Kruskall-Wallis test was significant at P = .011. There was more bone in the ” 2 0 group compared to the ”0” group ( P = .032) or the “20 X 2“ group ( P = .015). The “20 X 2” group specimens showed a tendency to contain less bone than the ”0” group ( P = .064). Three of the seven specimens in the “20 X 2” group contained almost

1422 GOODMAN ET AL.

(B) Figure 4. Specimen from a micromotion chamber moved 20 cycles twice/day. The section consists pri- marily of fibrous tissue. Hematoxylin and eosin stain. A: 40X magnification originally. B: 10OX magnification originally.

Figure 3. Specimen from a micromotion chamber moved at 20 cycles oncelday. Note that bone ingrowth is more extensive; the trabeculae of bone are also thicker than in Figure 2. Hematoxylin and eosin stain. A: 40X magnification originally. 8: l00X magnification originally.

no bone (< 10%); all specimens in the "0" and "20" groups contained at least 20% bone in the section.

DISCUSSION

Theoretical and experimental studies have impli- cated a relationship between the mechanical envi- ronment and the differentiation, development, and remodeling of mesenchymal t i s ~ u e . ~ - ~ Superim- posed, exogenous loads can alter the structure of these tissues dramatically. These externally applied loads need only be administered intermittently, for short periods of time each day.6-9,12,16 It is our hypothesis that a shear stimulus is applied to the ingrowing tissue at the interface between the cylinder

and core of the MC. Using a square-holed cylinder, the absence of motion allowed the ingrowth of trabecular bone longitudinally within the chamber. When 20 cycles of 0.75 mm amplitude were delivered to this chamber once daily, the ingrowth of a more immature type of bone was stimulated. It would appear that this imposed mechanical shear strain directly or indirectly stimulated the mesenchymal cells/osteoblasts to form more new bone rather than fibrous tissue. These results should be contrasted with those found when a round-holed cylinder with 0.5 mm of micromotion was utilized.6 In the latter study, bone ingrowth was decreased when 20 cycles/day were applied to a chamber with a round-holed cylinder. This discrepancy may be due to the less congruous interface in the previously used chamber, and/or the larger area for ingrowth

TITANIUM CHAMBERS AND BONE INGROWTH 1423

provided by a square rather than a round-holed ~ y l i n d e r . ~ Using a micromotion chamber with a round-holed cylinder (with a radius of 0.5 mm) and the square-holed core, if the total amplitude is 0.5 mm (0.25 mm maximal deflection to each side), the actual area for ingrowth at maximal deflection would be 0.63 mm2 by trigonometric calculation. Using a square-holed cylinder with a total amplitude of 0.75 mm (maximal deflection = 0.375 mm), the area for ingrowth at maximal defection would be 0.625 mm2 (1 mm2 - 0.375 mm2). Thus, despite the larger amplitude of motion in the present study, the actual area for ingrowth at maximal deflection was the same in this study compared to when a round- holed chamber with a smaller amplitude was used.6

It may also be hypothesized that the total daily time that micromotion is administered is an impor- tant determinant of tissue differentiation. If this is so, then this may provide a further explanation of the different results using a round-holed versus a square-holed cylinder. The former series was per- formed earlier, when the authors were not as adept at manipulating the chamber as they were when the latter series was performed. Consequently, the manipulations may have taken longer to perform in the “round hole” series. Whatever the cause, in the present study 20 cycles delivered once daily appeared to stimulate ingrowth of bone into the chamber.

When the motion was increased further to 20 cycles delivered twice daily, bone ingrowth was depressed. Thus, there may be a critical amount of strain that mitigates the differentiation of mesenchymal cells to osteoblasts, or modulates the osteoblast’s ability to form bone. Studies by Rubin and Lanyon12 and Ken- Wright and G o o d ~ h i p ~ , ~ ~ have implicated a threshold effect by which low levels of externally applied strain may facilitate bone f o r m a t i ~ n . ~ - ’ ~ Our present study using the micromotion chamber extends this concept by suggesting the existence of a “window” of exter- nally applied strain that may facilitate or discourage bone formation.

Although the ”20 X 2“ manipulations were initi- ated later in the series (usually 1 or 2 months after the “0” and ”20” series), the decreased amount of bone ingrowth in this group was not due to an ”exhaustion” effect. In ongoing studies by the authors that include some rabbits with over 20 harvests, an exhaustion effect for bone ingrowth does not occur as long as the animal is healthy. It is recognized, however, that the in vivo time of the chamber and the fact that each animal did not receive an identical treatment regimen may have affected the results of the present experiment.

Using the micromotion chamber, parameters such as the amplitude, the total number of daily manipula- tions and the configuration of the hole in the cylinder have been shown to have an important effect in

determining the differentiation of tissue into the pore. Further studies are in progress to ascertain whether the total time that the stimulus is applied is important, and to determine the relationship of some of the other parameters of stimulation with tissue differentiation.

This study was financially supported by Greta och Johan Kocks Stiftelse, Alfred Osterlunds Stiftelse, Stiftelsen Bis- tand at Vanfora i Skane, Tore Nilssons Fund for Medical Research, Craafords Stiftelse, the Swedish Medical Research Council 2031 and 09509-01, and the Medical Faculty of Lund University.

References 1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

H. U. Cameron, R.M. Pilliar, and I. Macnab, “The effect of movement on the bonding of porous metal to bone,” J. Biomed. Muter. Res., 7, 301-310 (1973). S. D. Cook, ”Clinical, radiographic and histologic evaluation of retrieved human noncemented porous coated implants,“ J . Long-term Effects of Medical Im- plants, 1(1), 11-51 (1991). S.D. Cook, K. A. Thomas, and R. J. Haddad, “His- tologc analysis of retrieved human porous-coated total joint components,” Clin. Orthop., 234,90 (1988). R.M. Pillar, J.M. Lee, and C. Maniatopoulos, ”Observation on the effect of movement on bone ingrowth into porous surfaced implants,” Clin. Ortkop., 208, 108-113 (1986). T. Albrektsson, M. Jacobson, and P. Kalebo, “The harvest chamber- A newly developed implant for analysis of bone remodeling in situ,” in Biomaterials and Biomechanics 1983 P. Ducheyne, G. Van der Perre, and A.E1 Ubert, eds., Amsterdam, Elsevier Science Publishers (1984), pp. 283-288. P. Aspenberg, S. B. Goodman, S. Toksvig-Larsen, L. Ryd, and T. Albrektsson, ”Intermittent micro- motion inhibits bone ingrowth: Experiment using titanium implants in rabbits,” Acta Ortkop. Scand.,

S.B. Goodman, S. Toksvig-Larson, and P. Aspen- berg, “Ingrowth of bone into pores in titanium chambers implanted in rabbits: effect of pore cross- sectional shape in the presence of dynamic shear,” J. Biomed. Muter. Res., 27, 247-253 (1993). S.B. Goodman and P. Aspenberg, ”The effect of amplitude of micromotion on bone ingrowth into titanium chambers implanted in the rabbit tibia,” Biomaterials, 13, 944-948 (1992). A.E. Goodship and J. Kenwright, ”The influence of induced micromovement upon the healing of experimental tibial fractures,” J. Bone It. Surg., 67-B,

J. Kenwright and A. E. Goodship, “Controlled me- chanical stimulation in the treatment of tibial frac- tures,” Clin. Ortkop., 241, 36-47 (1990). L.E. Lanyon, “Functional strain in bone tissue as an objective, and controlling stimulus for adaptive bone remodeling,” J . Biomechanics, 20 (ll), 1083-1093 (1987). C. T. Rubin and L. E. Lanyon, ”Regulation of bone formation by applied dynamic loads,” J. Bone Jt. Surg., 66-A, 397-402 (1984). D. R. Carter, ”Mechanical loading history and skele- tal biology,” 1. Biomeck., 20, 1095-1109 (1987).

63(2), 141-145 (1992).

650-655 (1985).

1424 GOODMAN ET AL.

14.

15.

16.

17.

18.

19.

20.

D. R. Carter, M. Wong, and T. E. Orr, ”Musculoskele- tal ontogeny, phylogeny, and functional adapta- tion,” J. Biomech., 24, Suppl. 1, 3-16 (1991). R. Huiskes, H. Weinans, H.J. Grootenboer, M. Dalstra, B. Fudala, and T. J. Sloof, ”Adaptive bone- remodeling theory applied to prosthetic-design analysis,” J . Biomech., 1135-1150 (1987). C. T. Rubin and K. J. McLeod, “Promotion of bony ingrowth by frequency specific, low amplitude me- chanical strain,” Transactions of the 38th Annual Meet- ing of the Orthopaedic Research Society, 69 (1992). C.T. Rubin, K. J. McLeod, and S.D. Bain, ”Func- tional strains and cortical bone adaptation: Epige- netic assurance of skeletal integrity,” J. Biomech., 23,

C. T. Rubin and L. E. Lanyon, ”Dynamic strain simi- larity in vertebrates; an alternative to allometric limb bone scaling.” J. Theor. Biology, 107, 321-327 (1 984). C.T. Rubin and L.E. Lanyon, “Regulation of bone mass by mechanical stimulation,” Calcified Tissues International, 37, 411-417 (1985). A. Sarmiento, J. F. Schaeffer, L. Beckerman, L. Latta, and J.E. Enis, ”Fracture healing in rat femora as

SUPPI. 1, 43-54 (1990).

21.

22.

23.

24.

affected by functional weight-bearing,” J . Bone It.

E.A. Williams, J .A. Rand, E.Y.S. Chao, and P. J. Kelly, ”The early healing of tibia1 osteotomies stabi- lized by one-plane or two-plane external fixation,” 1. Bone Jt. Surg., 69-A, 355-365 (1987). J.W. Wolf, A.A. White, M.M. Panjabi, and W.O. Southwick, ”Comparison of cyclic loading versus constant compression in the treatment of long-bone fractures in rabbits,” J. Bone Jt. Surg., 63-A, 805-810 (1981). J.L. Wolff, ”The law of bone remodeling,” (“Das Gesetz der Transformation der Knochen,” 1892), translated by P. Maquet and R. Furlong, Berlin, Springer-Verlag (1986). M. Wong and D. R. Carter, “A theoretical model of endochondral ossification and bone architectural re- construction in long bone otogeny,” Anat. Embryol.,

Surg., 59-A, 369-375 (1977).

181, 523-532 (1990).

Received December 29, 1992 Accepted June 8, 1993