The effects of fluoride and low calcium on the physical properties of the rat femur.код для вставкиСкачать
The Effects of Fluoride and Low Calcium on the Physical Properties of the R a t Femur DEXTER F. BEARY' Department of Anatomy, Schools of Medicine and Dentistrg, Lorna Linda University, Lorna Linda, California 92354 ABSTRACT The effect of fluoride on the physical properties of bone was investigated with the aid of an Instron materials tester. Two groups of weanling female rats, one on a n adequate (0.6% ) calcium diet and the other on a low (0.1%) diet were given the following doszge levels of fluoride (as NaF) i n their deionized drinking water over a fifteen End one-half week period: 3.4 Fpm, 10.0 ppm and 45.0 ppm. In the adequate calcium group a significant increase i n flexibility in the rat femur was found only a t the 45.0 ppm dosage level. This was not offset by a significant decrease in strength. In the low calcium group a similar significant increase in flexibility appeared at the 10.0 ppm dosage level 2s well as the 45.0 ppm, but a significant decrease in strength at the two dosage levels was observed. These were in direct relation to the amount of fluoride given. This study was designed to determine the influence of fluoride on the physical properties of the fresh rat femur, and indirectly on the common bone disease osteoporosis. The osteopenia produced by a calcium deficient diet in the experimental animal shows essentially the same characteristics. MATERIALS AND METHODS Eighty weanling, 25 day old, 38-40 gm, female, Sprague-Dawley albino rats were employed in the study. They were divided into two main groups, I and 11, of forty rats each. Each main group was in turn divided by random assortment into five subgroups of eight rats each. Subgroups of Group I were designated 1-5; those of Group I1 were designated 6-10. Group I received 0.6% calcium in their diet which was considered to be adequate in other respects. The phosphorus content was 0.358%. Group I1 received the same diet except that calcium was reduced to 0.1% of the diet. In both groups a Jones and Foster ('42) salt mixture without calcium was chosen since it lacks fluoride. ANAT. REC., 164: 305-316. The dietary ingredients were as follows: Diet ingredients Casein Sucrose Vitamin Mix Salt Mixture (without CaC08) Calcium Carbonate Corn Oil Group I Adequate Ca diet Group II Low in Ca % I 24.000 66.250 24.000 65.000 2.000 2.500 1 .SO0 5.000 2.000 2.500 0.250 5.000 It is to be noted that, while the calcium was reduced in the diet of Group 11, the phosphorus content was kept the same in both diets. Thus the calcium to phosphorus ratio of the adequate diet is about 1.7-1.0 while that of the low calcium diet is approximately 0.3-1.0. The diet was fed to the rats ad libitum for a period of 106-110 days depending upon the date of sacrifice. The subgroups consisting of eight rats each received different dosages of fluoride, as sodium fluoride, in their deionized Received Sept. 20, '68; Accepted Feb. 24, '69. 1 Present address: Department of Biology, Southwestern Union College, Keene, Texas 76059. 305 306 DEXTER F. BEARY drinking water (except for the control subgroups) as follows: Group I (Adequate c.cium) Subgroup ppm F Group 11 (LOWcalcium) Subgroup ppm F l(contro1) 0 G(contro1) 0 2 3.4 ( H W ) ' 7 3.4 (HW)1 3 3.4 a 3.4 4 10.0 9 10.0 5 45.0 10 45.0 1 HW, Water from Hereford, Texas, which contained 3.4 ppm F plus other naturally occurring elements. This was used to see if other naturally occurring elements might have an additional effect. The rats were weighed once each week, in order to determine the effect of the low calcium diet and fluoride on growth (see table 1). The amount of diet eaten by each group was weighed and the drinking water consumed by each subgroup was measured. At the end of the seventh week, the rats were anesthetized with ether and x-rayed in order to compare bone den- sities. The rats were again x-rayed at the termination of the experiment. The rats were sacrificed by an overdose of ether over a period of five days beginning with the one hundred sixth day and ending with the one hundred tenth day of the experiment. Their bodies were x-rayed in groups of four and then both femurs were extracted. The femurs were cleaned of muscle tissue, weighed, measured, and placed in air-tight polyethylene containers in order to prevent drying before testing. At the time of sacrifice, the adrenal glands were removed and weighed in order to determine if the rats had been stressed due to the added fluoride in their drinking water. An Instron Universal Testing Instrument was employed to test the femurs for strength and flexibility (fig. 1 ) . A load cell was chosen which measures forces from 0-200 pounds with chart recording setting at 0-20 and 0-50 pounds respectively. A high-speed electronic graphic re- Figure 1 307 FLUORIDE AND PHYSICAL PROPERTIES OF BONE corder, driven synchronously with the tester, recorded the forces and deflections exerted on each femur. The accuracy of the system is better than 0.25% at all load ranges. A special metal holding device for the femurs was used so that the diaphysis was loaded anteriorly at the midpoint, after the method of Bell, Cuthbertson and Om (’41), and Weir, Bell and Chambers (’49) (see fig. 2a). The adjustable, metal supports, A and A’, were so adjusted that points of contact or distance, between supports was fixed at 0.75 inch (19.05mm) apart on the posterior aspect of the bone. The force, F, was applied midway between these supports on the anterior surface. The metal supports and metal used for applying the force were rounded slightly in order to avoid cutting into the bone as the force was being applied. Figure 2b is a representation of a photographically enlarged view of the femur at the fracture site. The crosshead or screw driven force was applied at a constant rate of 0.1 inch (2.54 mm) per minute. The chart speed was set at ten inches (254 mm) per minute. Every inch (25.4 mm) of chart on the y axis represents 0.01 inch (0.254 mm) deflection (see fig. 3). The Instron machine allows the testing speed to be precisely controlled, thus giving a constant rate of specimen deformation independent of the load applied. The synchronously driven chart provided a permanent record of the change in resistance to the load applied and the corresponding deflection. The load at the elastic limit, maximum load, and load at the breaking point are measured on the x axis, the amounts of deflection at these loads are measured on the y axis. The amount of energy absorbed is calculated by dividing the area under the recorded stress-strain curve by the bone volume. Based on the preceding data Young’s modulus of elasticity was applied. From the latter, stiffness and flexibility were determined, the one being the reciprocal of the other. Young’s modulus of elasticity, E, is a measure of the stiffness of the material being tested. It is the coefficient of the elasticity or the ratio of unit stress-unit deformation. The formula for E, for a beam loaded at the center, is E = where “ W is the load at the elastic limit, “L” is the length between supports A and A‘, “y” is the amount of strain or deflection at the elastic limit and “I” is the moment of inertia (see figs. 2a, 3, table 3 ) . The latter concerns the bending moment about the neutral axis and is related to the configuration of the bone at the point of breaking. Since the cortex is nearly elliptical on cross-section, “I” can be derived from the formula I = 4$ (BD3-bd3) (see fig. 2b). The formula for E assumes that the configuration of the cortex of the femur between supports and its cross-sectional area are constant. Admittedly the femoral I Force(W) I I t t Figure 2 I* b -I I I ! 308 DEXTER F. BEARY DEFLECTION Y, =elastic limit rn axi rn u m Y3=breaking point Y2 LOAD Load -deflection recording Wl= elastic limit W2= maximum W3: breaking point Figure 3 shaft changes slightly in configuration along the shaft, being somewhat greater in diameter towards the ends where the cortex is proportionately thinner. However, comparison of computer-calculated and actual volumes of corresponding lengths of eight femurs, as described below, suggests that one is justified in handling the calculations as though the configuration and surface area of the cortex remain constant. The stiffness of the femoral shaft, or its resistance to bending is proportional to EI. The flexibility, or ease of bending would be be proportional to The stress, or force per unit area, on the bone substance is greater on the outer surface of the cortex where the load is applied. The amount of stress per unit area of bone is derived from the formula S = y ( w h e r e D is the outside diameter of the bone in the direction of the applied force - see fig. 2b). The next step was to derive the bone volume. For greater accuracy the surface areas of the bones at the fracture site were computed by taking graph paper tracings of the outer and inner circumferences of the cortex of the bone which had been photographically enlarged. A Photo- &. cell Scanner and a Calcomp 685 Plotter in conjunction with an IBM 1620 computer were employed for this purpose. The bone volume was calculated from the area of each femur and the distance, L, between supports. This bone volume was utilized in computing the amount of energy absorbed before breaking. The reliability of the computer method of determining the bone volume was confirmed by the water displacement method which gave figures averaging only 3% higher than those calculated by the computer. One could never be sure that all the air bubbles had been cleaned from the specimens and it is difficult to accurately read the volumetric scale. The close agreement coupled with difficultly in actually measuring volumes favored the computer method. If any error is introduced, it would be consistent with all bones tested, since they were compared with those of the control animals. The bone between supports was cut out carefully, degreased and weighed. Finally the fluoride uptake by the femoral shaft tested between supports A and A' (see fig. 2a) was calculated by dissolving this section of bone in 4 cm3 of a 1 :1 dilution of concentrated HCl solution. This was then diluted to 50 cm3 using a 2% NaOH solu- 309 FLUORIDE AND PHYSICAL PROPERTIES OF BONE tion. The pH was adjusted to 2.07 2 0.03, since this is considered a desirable range for reading fluoride ion activity. An Orion Fluoride Activity Electrode was used in conjunction with a reference electrode for the fluoride determination according to the method devised by Frant and Ross ('66). The accuracy of this method has been confumed by Lingane ('67). The calculations were recorded as mg of F per bone segment and percentage F by weight (see table 1). A statistical analysis was done to determine the validity of the conclusions reached. perimental period was 2.3 liters per rat and was essentially the same for the animals in each subgroup. Consequently, the fluoride assay was directly proportional to the dosage given. The percentage content of fluoride by weight was strikingly different in Groups I and I1 at the 10 ppm F and 45 ppm F dosage levels, there being a percentage increase per bone segment in the low calcium group. There is also a difference between the Hereford water subgroup (3.4 ppm F) and the corresponding 3.4 ppm F in deionized water subgroup, the former containing less fluoride than the latter. O n adrenal weight. Metabolic stress, RESULTS due to possible toxic effects of the higher The principal data is recorded in tables fluoride levels used, was excluded with regard to having an influence on the results 1-5. since there was no significant difference Effects of the diet and fluoride in the adrenal weight between Group I On body weight. As shown in table 1 and Group I1 or between any of the subthe rats on the low calcium regimen (Group groups (table 1). Because of this nega11) whose diet contained only 0.1% of cal- tive finding a dry weight determination cium show 7% less weight gain than those on the thymus was not done. O n linear bone measurements. No apon an adequate calcium diet (Group I). Group I1 consumed 43.8 kg of diet over the preciable differences in bone lengths were experimental period; Group I, 47 kg. The seen between Group I and Group I1 or amount of weight gained was in direct pro- between any of the subgroups. Outside portion to the amount of diet consumed. diameters of the femoral cortex likewise It bore no relationship to the amount of were unchanged but the inside diameters fluoride administered during the experi- showed a marked increase in those rats maintained on a low calcium diet resulting mental period. 072 amount of jhoride per bone segment in a marked thinning of the cortex. This between supports and its percentage value. was reflected in the volume of the cortical The average water consumed over the ex- bone between supports A and A' (fig. 2a), TABLE 1 Means of rat femur analysis ppm Bqdy weight x;tGE between supports I (Adequate Ca> Group I1 (LowCa) 1 Hereford water. 2 I and I1 z:$z per "f:l;: supports (dry, fat free) bone segment P~~~~~~ Adrenal adrenal Ratio weight weight wt,body wt (mg) (mg/100 pm) <mm3) (pm) <mg) 0 3.4 1 3.4 10.0 45.0 251.250 257.125 248.125 258.625 254.375 102.846 106.310 99.935 102.930 103.775 0.225 0.223 0.218 0.226 0.230 0.014 0.045 0.059 0.148 0.693 0.01 0.02 0.03 0.07 0.30 56.437 52.387 65.362 59.637 55.975 22.520 20.363 26.376 23.151 21.969 0 3.4 1 3.4 10.0 45.0 237.875 239.625 231.428 238.375 230.285 66.981 66.751 65.138 67.211 67.157 0.133 0.144 0.129 0.141 0.130 0.006 0.035 0.041 0.149 0.671 0.02 0.03 0.16 0.76 57.325 59.500 59.042 56.550 60.014 24.041 24.849 25.463 23.891 26.271 (pm) Group - groups Weight 2 Bone weights obtained from corresponding segment of opposite femur. 310 DEXTER F. BEARY which averaged 35% lower in the low example of the graph recording with the calcium group (table 1 ) . Since the com- load points at which the load at elastic puted bone volume in the subgroups of limit, m a x i m u m load, and load at breahGroup I1 is essentially the same in every ing point were taken for each femoral instance, it is evident that it is the cal- shaft tested. The corresponding defleccium deficiency and not fluoride admin- tions are given for each load. The results, istration or the lack of it that was re- as seen in the tables, show significant differences when comparing the variables of sponsible. O n loads and deflections. The loads corresponding subgroups of Groups I and and deflections listed in table 2 were cal- I1 and also when comparing certain subculated primarily for use in formulas em- groups within each of these same groups. ployed in computing the results of the vari- This difference is reflected by the amount ables listed in table 3. Figure 3 shows an of calcium and/or fluoride given. TABLE 2 Means o f rat femur analysis - group I and 11 2t$gt limit ppm F ( ~ ~ Deflection elastic limit ~ t Maximum Deflection load maxlmum ~(Newtons) ~ ~ )load (mm) Group I (Adequate ca> Group I1 (Low Ca) 1 Hereford Load at bT: EFg (Newtons) Deflection breaking point (mm) 0 3.4' 3.4 10.0 45.0 88.946 85.386 82.744 84.550 82.355 0.255 0.273 0.261 0.263 0.303 108.416 106.106 104.104 107.943 101.100 0.385 0.425 0.401 0.413 0.471 108.416 106.106 104.104 107.943 101.100 0.385 0.425 0.401 0.413 0.471 0 3.4 3.4 10.0 45.0 37.465 35.851 39.129 34.911 23.617 0.276 0.253 0.294 0.274 0.306 57.406 57.802 57.217 53.040 34.432 0.637 0.654 0.704 0.750 1.038 50.815 52.623 48.856 43.889 29.942 0.868 0.854 0.895 1.020 1.591 water. TABLE 3 Rat femur analysis group I and I1 - ppmF Moment in2ia (mm4) Group I (Adequate Ca 1 Group I1 (Low Ca) Stress elasticat limit (N/=S) Streps at load (N/mmz) ¶Young% IFlexImodulus bility Stiffnew 1 1Energy absorbed (N/mm2) (:l/Nmmz ) (N/mmz) (N-M/mma) x 10-2 x 105 45.0 3.629 3.648 3.470 3.463 3.671 248.385 236.950 240.453 242.660 226.782 302.887 294.280 302.773 309.919 278.445 142.815 125.327 133.683 134.677 107.986 1.984 2.233 2.193 2.172 2.578 517.308 455.230 460.363 463.099 393.696 2.273 2.350 2.264 2.40 2.59 0 3.4' 3.4 10.0 45.0 2.673 2.876 2.714 3.057 3.019 141.766 129.146 146.796 121.186 81.610 216.900 207.245 211.620 183.190 118.507 73.038 72.187 72.864 61.740 37.546 5.185 4.913 5.225 5.469 9.184 194.633 204.742 195.719 185.712 113.065 5.283 5.287 5.370 5.704 6.409 0 3.4' 3.4 10.0 x10-2 Xlor Note: 1. The figures have been altered by powers of ten in order to conserve space., e.g.. Young's modulus mean of subgroup 1 which is actually 14181.500 N/mmZ has been reduced to 142.815 N/mmZ The loads have all been expressed h terms of Newtons (N), t h e international measuring standard of force 1 lb. wt. = 4.45 Newtons (N). 3. To convert to kg/cms for E, (Young's modulus) multiply above values by 100/9.8 (approximately 10). e.g.. 14281.5 x 10 = 142815 kg/cmS. 4. The range of standard deviations of data (not shown) from each rat femur, gi*g v&ables of each group, was not wide enough as to SignMcantly affect the analysls of the means. 2. FLUORIDE AND PHYSICAL PROPERTIES OF BONE On moment of inertia. The moment of inertia was naturally lower for the calcium group as the cortex was much thinner. The fluoride had no effect on this calculation since the thickness of the cortex was unaffected by the uptake of fluoride. As previously shown in the section on methods and materials, the moment of inertia is necessary in applying the formula for Young's Modulus of Elasticity, E. On the variables. (Stress at Elastic Limit, Stress at Maximum Load, Young's Modulus of Elasticity, Flexibility, and Stiffness)These variables all show highly significant differences between the corresponding subgroups of Group I and Group 11 and therefore for the groups as a whole. All except flexibility are greater for the adequate calcium group, flexibility being greater for the low calcium group. To determine the effect of fluoride on the rat femur, Anova (analysis of variance) was run to determine differences between means of the first five subgroups and the last five subgroups for each of the 13 variables listed in tables 1 and 3. Table 4 gives the variables for which there were significant differences among means of the different subgroups 1-5 and 6-10 and their level of significance. For these variables, t tests were performed between the control groups and each of the fluoride treatment groups (see table. 5). 311 tent of bones in fluorotic subjects may have been due to subnormal living standards and a low nutritional status. McClure et al. ('58) who examined the bones of a subject with a history of prolonged exposure to water containing 8 ppm fluoride found the amount of Ca F in the skeleton to be quite uniform being within the narrow range of 0.517and 0.653% in the dry fat-free samples. Soriano ('65) found, in human skeletal fluorosis, considerable variation in the bone F content, with an observed high of 1.34% in the coccyx and a low of 0.12% in the tibia. Experiments with rats have clearly shown that when fluoride is added to the drinking water there is an increase in the size of the apatite crystals and they are fewer in number (Menczel et al., '63 and Eanes et al., '65). The fluorapatite crystals in bone while they have a close structural similarity to the bone salt, hydroxyapatite, appear to be less reactive with body fluids, resulting in decreased resorption (Rich and Ivanovich, '65). Fluoride in the amount of 50 ppm in the drinking water of female rats was shown by Gedalia et al. ('64) to significantly increase the calcium content of the bone ash without a corresponding increase in phosphorus. It is noted, however, that McCann and Bdloch ('57) found no such increase on a 100 ppm dosage level. Bell et al. ('41), who used a bending DISCUSSION test, found that the midpoint of the rat There have been numerous reports con- femoral shaft is the weakest. They atcerning the physiology, crystalline stmc- tribute this to the smaller diameter and ture and ash content of bone on adequate greater bending moment. According to and low calcium diets, with and without their findings, bone is elastic up to the fluoride (Boelter and Greenberg, '40; Light moment of breaking. Weir et al. ('49), and Frey, '41; Harrison and Fraser, '60; after further testing, revised this estimate El-Maraghi et al., '65). Studies in humans stating that the dry rat femur is elastic as by Leone et al. ('55) have suggested only up to 79% of its breaking stress. They that what is generally considered in this calculated the modulus of elasticity (E) to country to be an excessive fluoride content be 1.6 X lo6 lbs/ing. Evans ('64) finds in the water ( 8 ppm) and which causes that the method they used for determining mottling of the teeth, has a beneficial ef- extreme or outer fiber stress is only valid fect in counteracting osteoporosis. Zipkin at the elastic limit. Evans and Lebow et al. ('58) found that fluoride in the ('51) using wet human bone samples drinking water of human subjects up to showed that it is the middle third of the 4 ppm increases their fluoride content femoral shaft that has the highest modulus proportionally to the amount of exposure of elasticity. The figure they give is 2.11 but does not affect the ash content of the x lo6 lbs/ina. Drying was found to inbones. In their opinion previously re- crease the modulus of elasticity by 17.6%. ported very high values for fluoride con- Currey ('59) found the value of "E" for 312 DEXTER F. BEARY TABLE 4 Significant Variables: Means and Levels of Significance Group I Subgroups 2 (HW-3.4 pprn F) Mean (X) 5z P 4 3 (3.4ppm F) P P (10.0 ppm F) -X P 5 (45.0 ppm F) X P Stress at Elas. Limit (N/mm2) 248.4 237.0 0.058 240.5 0.148 242.7 0.241 226.8 0.000 Stress at Max. load (N/mm2) 302.9 294.3 0.256 302.8 0.987 309.9 2.76 278.4 0.002 Young’s Mod. of Elas. (N/ mm2 x 142.8 125.3 0.019 133.7 0.169 134.7 0.238 108.0 0.000 Flexibility (1 / Nmmt x lo5) 1.98 2.23 0.038 2.19 0.040 2.17 0.050 2.58 0.000 Stiffness (Nmmz x 10-2) 517.3 455.2 0.028 460.4 0.036 463.1 0.036 393.7 0.000 0.014 0.043 0.000 0.060 0.000 0.145 0.000 0.669 0.000 9 (10.0 ppm F) 10 (45.0 ppm F) Fluoride Assay (mg/mm3 x 102) Group I1 Subgroups 6 (Control) 7 (HW-3.4 pprn F) Mean (X) -X P 8 (3.4 ppm F) X P F P -X P Load a t Elas. Limit (Newtons) 37.47 35.85 0.274 39.12 0.334 34.91 0.141 23.62 0.000 Maximum Load (Newtons) 57.41 57.80 0.797 57.22 0.699 53.04 0.010 34.43 0.000 Stress a t Elas. Limit ( N/mm2 ) 141.8 129.1 0.080 146.8 0.547 121.2 0.013 81.6 0.000 Stress at Max. Load (N/mm2 1 216.9 207.2 0.150 211.6 0.578 183.2 0.000 118.5 0.000 Young’s Mod. of Elas. (N/mm2 x 10-2) 73.0 72.2 0.788 72.9 0.963 61.7 0.003 37.5 0.000 Flexibility (1/ N m m 2 x 105 5.19 4.91 0.112 5.23 0.875 5.47 0.213 9.18 0.000 Stiffness (Nmm2 x 10-2) 194.6 204.7 0.116 195.7 0.905 185.7 0.246 113.1 0.000 Energy Absorbed (N-M/rnms x 104) 5.283 5.287 0.991 5.370 0.836 5.704 0.324 6.409 0.008 0.009 0.054 0.000 0.064 0.000 0.225 0.000 0.991 0.000 Fluoride Assay (mg/mm3 x 102) Note: X. mean of each variable. p. level of significance of each variable. 313 FLUORIDE AND PHYSICAL PROPERTIES OF BONE TABLE 5 A list of the correlation coeficients betureen the fluoride assay and the significant variables of subgroups 1-5 and subgroups 6-10 ~- ~~ Group I Subgroups 1-5 Variable Assay - Stress at Elas. Limit Assay - Young’s Mod. of Elas. Assay - Flexibility Assay - Stiffness n r P 40 40 40 40 -0.3528 - 0.4827 0.4772 - 0.4256 0.0256 0.0016 0.0019 0.0062 n r P 38 38 38 38 38 38 38 38 -0.7702 - 0.8571 -0.7323 - 0.8573 -0.8359 0.8440 -0.8307 0.4532 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0043 Group I1 Subgroups 6-10 Variable Assay - Load at Elas. Limit Assay - Maximum Load Assay - Stress at Elas. Limit Assay - Stress at Max. Load Assay - Young’s Mod. of Elas. Assay - Flexibility Assay - Stiffness Assay - Energy Absorbed Note: n number of samples. r,’correlation coefficient. p. level of si@cance. cortical bone from the ox femur to be about 2.5 X 10‘ lbs/inz or approximately ten times the value for collagen. Consequently any change in the crystalline structure would significantly alter “E.” Semb (’66) found no significant difference in the breaking strength of longitudinal cortical bone sections of equal dimensions from femurs of normal and osteoporotic dogs. He states that any weakness of the osteoporotic bone is due to the thinning of the cortex. His data is recorded in Newtons (the metric unit of force). Ascenzi and Bonucci (’67) studied individual osteons, wet and dry, from both human and ox cortical bone. The tensile strength and modulus of elasticity were found to be greater in the dry specimens, particularly in those in which there was a marked longitudinal arrangement of collagen fibers in successive lamellae. The “ E for wet human specimens with longitudinally disposed collagen fibers was 119400 kg/cmz (1.7 X lo6 lbs/in2). For corresponding ox specimens, the “ E was 149600 kg/cmz (2.12 X lo6 Ibs/ine). Vincentelli (’68) is doing similar work in this field. Experimental studies on the effects of fluoride on breaking strengths and elastic- ity of bone have been few. Jovanovits (’44) whose work is summarized by Uehlinger (’64) found that in both the very young and the older growing rabbit, fluorine in the form of various salts in the amounts of 5mg/kg and 50mg/kg of body weight per day respectively, over a 150 day period, diminished the modulus of elasticity proportional to the amount of fluoride administered. He also noted a significant lowering of the breaking strength. The effects were more marked in the first group. Gedalia et al. (’64) also found that, in young growing rats given 50 ppm F in their drinking water for four weeks, the breaking strengths of their dry defatted femurs were significantly lower than those of the control animals. It is yet to be confirmed that fluoride administration is accompanied by the increased calcium content of the bone they have reported. Normally bone strength is directly proportional to its calcium content (Shah et al., ’67) and this would not be consistent with their findings. Where lesser doses of fluoride have been employed no effects on the physical properties of bone have been noted. Smith and Keiper ( ’ 6 5 ) , in testing the viscoelastic properties of the femora of Beagle dogs, found that 1.0 mg F in 314 DEXTER F. BEARY their diet over a period of 122 days had no demonstrable effect on bone elasticity. Saville ('67) found no significant difference between the breaking strength of the femur taken from weanling male rats given 20 ppm M in the drinking water for 123 days and those of the controls. The results of this study show that, when the diet of weanling rats is low in calcium (as seen in the controls), for fifteen and one-half weeks, there is a slight loss in body weight (table 1) as also noted by Boelter and Greenberg ('40). Fluoride administration at the dosage levels employed did not appear to be the cause. On the low calcium diet the outer dimensions of the femur remain unchanged but the inner diameter is increased, thus producing a thinner cortex and decrease in bone volume (table 1). This observation agrees with that found by Light and Frey ('41) and is not unlike that seen in the metabolic disease osteoporosis. There is abundant evidence that a low calcium diet can induce this disease (McClendon et al., '62; Whedon et al., '63). Each of the subgroups within each major group drank essentially the same amount of water (18.6 liters rt 0.4). In the low calcium group, at the 10 ppm and 45 ppm fluoride administration levels, the percentage of fluorine in the bone is seen to be higher than in the adequate calcium group given the same dosage levels of fluoride. Comparing the subgroups within the two major groups given drinking water from Hereford, Texas, containing 3.4 ppm F, with those subgroups with the same amount of fluoride in their deionized drinking water, it is seen that the fluoride percentage is less in the Hereford water subgroup. This could be due either to the fact that the drinking water supplied additional calcium, or that other elements such as magnesium influence fluoride absorption. Such influence finds corroboration in the review article of Marier et al. ('63). Hereford water contains 61 mg/ liter of magnesium in addition to 64mg/ liter of calcium. With respect to the effects of fluoride administration on the physical properties of the femora of subgroups 2-5 of Group I (adequate calcium diet) it is noted that there is no significant difference in strength (maximum load and/or load at breaking point) at any of three fluoride dosage levels. A significant increase in flexibility was apparent only in the 45 ppm F subgroup (tables 3 and 4). Such an increase in flexibility was also noted by Jovanovits ('44). While he recorded a lowering of the breaking strength it was not proportional to the amount of fluoride given. Further, the dosages he employed were much larger than any used in this study. Gedalia et al. ('64), like the lastmentioned, also record a lowering of the breaking strengths in femurs of rats given 50.0 ppm F in their drinking water and for only four weeks. They give no indication as to flexibility. The 20.0 ppm F employed by Saville ('67) over a 14week period had no effect on the breaking strength of the rat femur. The effects of fluoride administration in Group I1 (low calcium diet) are more apparent even at lower dosage levels. A decrease in strength and an increase in flexibility are significant in both the 10.0 ppm F and 45 ppm F subgroups which are related proportionately to the greater percentage of fluoride per bone segment (tables 1, 4, 5). The significant increase in the amount of energy absorbed before breaking found only in the 45 ppm F subgroup indicates that the bone is able to withstand a greater bending force at this dosage level, even though it is weaker from the standpoint of load bearing. Osteoporosis-like changes (osteopenia) will develop in the rat if calcium is deficient in spite of fluoride administration in the drinking water at the relatively high dosage of 45.0 ppm F. This is demonstrable radiologically as early as the eighth week. Fluoride in both the adequate and low calcium subgroups will increase flexibility. However, in the low calcium subgroups the increased flexibility is associated with a significant diminution in bone strength. Whether or not extra fluoride prevents fractures is still a matter of conjecture. Since fluoride in high doses weakens bone, it might tend to predispose to compression fractures of vertebrae as is not uncommon in areas where endemic fluorosis FLUORIDE AND PHYSICAL PROPERTIES OF BONE is present. Shambaugh and Petrovic ('68) believe fluoride availability to be a major contributory factor in the prevention of osteoporosis as seen in the human since they believe the fluoride ion slows the resorptive phase of the remodeling process and promotes calcification. The calcium levels employed in this experiment were not such as to test this hypothesis. It should be noted that Adams and Jowsey ('65) have stressed the importance of concomitant calcium supplementation in high fluoride therapy for osteoporosis. The results of this study, as shown in tables 1, 2 and 3, also clearly indicate that adequate calcium is needed in the diet in order to give the bone strength and that fluoride, at the 45 ppm range, will increase the flexibility. Finally, in this study there is some indication that elements other than fluorine, such as occur naturally in the drinking water at Hereford, Texas, may influence the physical properties of bone. One of these is possibly magnesium as has been suggestea by Barnetc('54) and Marier et al. ('63). ACKNOWLEDGMENTS Computation for this research was performed by the Lorna Linda University Scientific Computation Facility supported in part by NIH grant FR 00276-03. Special thanks is given to Professor Harold Shryock, Head of the Department of Anatomy, Lorna Linda University, for the facilities provided for this project and for his encouragement. The following individuals, University staff members, were especially generous of their time: Jerry Chrispens of the computational facility; Drs. U. D. Register, J. Riggs, and W. H. Roberts; and Mrs. Lucille Innes of the Audiovisual Department did the art work. Dr. Earl S. Gerard of the Upjohn Company, Kalamazoo, Michigan, suggested the subject of the investigation and gave many helpful suggestions. LITERATURE CITED Adams, P. H. and J. Jowsey 1965 Sodium fluoride in the treatment of osteoporosis and other bone diseases. Ann. Int. Med., 63: 11511155. Ascenzi, A. and E. Bonucci 1967 The tensile properties of single osteons. Anat. Rec., 158: 375-386. 315 Barnett, L. B. 1954 New concepts in bone healing. J. Applied Nutr., 7: 318-323. Bell, G. H., D. P. Cuthbertson and J. OR 1941 Strength and size of bone in relation to calcium intake. J. Physiol., 100: 299-317. Boelter, M. D. D. and D. M. Greenberg 1941 Severe calcium deficency in growing rats. J. Nutr., 21: 61-84. Currey, J. D. 1959 Difference in the tensile strength of bone of different histological types. J. Anat., 93: 87-94. Eanes, E. D., I. Zipkin, R. A. Harper and A. S. Posner 1965 Small angle x-ray diffraction analysis of the effect of fluoride on hum a n bone apatite. Arch. Oral Biol., 10: 161173. El-Maraghi, N. R. H., B. S. Platt and R. J. C. Stewart 1965 The effect of the interaction of dietary protein and calcium on the growth and maintenance of the bones of young, adult and aged rats. Brit. J. Nutr., 19: 491-507. Evans, F. G. 1961 Relation of the physical properties of bone to fractures. Instr. Course Lectures, Amer. Acad. Orthop. Surg., 18: 110121. Evans, F. G. and M. Lebow 1951 Regional differences in some of the physical properties of the human femur. J. Appl. Phys., 3: 563572. Frant, M. S. and J. W. Ross 1966 Electrode for sensing fluoride ion activity in solution. Science, 154: 1553-1555. Gedalia, I., A. Frumkin and H. Zukerman 1964 Effects of estrogen on bone composition in rats at low and high fluoride intake. Endocrinol., 75: 201-205. Harrison, M. and Russell Fraser 1960 Bone structure and metabolism in calcium deficient rats. J. Endocrinol., 21: 197-207. Jones, J. H., and C. Foster 1942 A salt mixture for use with basal diets either low or high in phosphorus. J. Nutr., 24: 245-256. Jovanovits, J. 1944 Beitrag zur experimentallen fluorosteopathie. Archiv. of Geweberpath Geweberhygiene, 12: 233-276. Leone, N. C., C. A. Stevenson, T. F. Hilbish and M. C. Sosman 1955 A roentgenologic study of a human population exposed to high fluoride domestic water. Amer. J. Roentgenol., 74: 874-885. Light, R. F. and C. N. Frey 1941 Bone fractures due to low calcium diets. Proc. SOC.Expo Bio. Med., 48: 256-258. Lingane, J. J. 1967 A study of the lanthanum fluoride membrane electrode for end point detection in titrations of fluoride with thorium, lanthanum, and calcium. Analytical Chem., 39: 881-887. Marier, J. R., D. Rose and M. Boulet 1963 Accumulation of skeletal fluoride and its implications. Arch. Environm. Health, 6: 664-671. McCann, H. G., and T. A. Bulock 1957 The effect of fluoride ingestion on the composition and solubility of mineralized tissues of the rat. J. Dental Res., 36: 391-398. McClenden, J. F., J. Jowsey, J. Gershon-Cohen and W. C. Foster 1962 The curative effect 316 DEXTER F. BEARY of a high calcium diet o n senile osteoporosis. J. Nutr., 77: 299-302. McClure, F. J., H. G. McCann and N. C. Leone 1958 Excessive fluoride in water and bone chemistry. Public Health Reports, 73: 741-746. Menczel, J., R. Schraer, G. Pakis,A. S . Posner and R. C. Likins 1963 Effect of a low calcium diet on bone crystallinity and skeletal uptake of Ca in rats. Proc. SOC.Exp. Bio. Med., 112: 128-131. Rich, C. and P. Ivanovich 1965 Response to sodium fluoride in severe primary osteoporosis. Ann. Int. Med., 63: 1069-1074. Saville, P. D. 1967 Water fluoridation effect on bone fragility and skeletal calcium content in the rat. J. Nutr., 91: 353-357. Semb, H. 1966 The breaking strength of normal and immobilized cortical bone from dogs. Acta Orthop. Scand., 37: 131-140. Shah, B. G.,G. V. G. Krishnarao and H. H. Draper 1967 The relationship of calcium and phosphorus nutrition during adult life and osteoporosis in aged mice. J. Nutr., 92: 3042. Shambaugh, G. E. and A. Petrovic 1968 Effects of sodium fluoride on bone. J.A.M.A., 204: 964-973. Smith, R. W. and D. A. Keiper 1965 Dynamic measurement of viscoelastic properties of bone. Amer. J. Med. Electronics, 4: 15G160. Soriano, M. 1965 La fluorosis vinico. Revista Clin. Espanola, 97: 375. Uehlinger, E. 1962 The Toxicology of Fluorine. Edited by T. Gordonoff. Schwabe and Co., Basel, pp. 77-79. Vincentelli, P. 1968 Relation of collagen fiber orientation to some mechanical properties of human cortical bone. Anat. Rec., 160: 445. Weir, J. B., G. H. Bell and J. W. Chambers 1949 The strength and elasticity of bone in rats on a rachitogenic diet. J. Bone Joint Surg., 31-B: 444-451. Whedon, G. D. and L. Lutwak 1963 Symposium on human calcium requirements. J. Amer. Med. Assoc., 185: 591-592. Zipkin, I., F. J. McClure, N. C. Leone and W. A. Lee 1958 Fluoride deposition in human bones. Public Health Reports, 73: 732-740.