AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 6023-28 (1983) Cortical Bone Formation and Diet Among Protohistoric lroquoians SUSAN PFEIFFER AND PATRICIA KING School ofHumon Biology, Uniwnrity of Guelph, Guelph,Ontario, C m d a KEY WORDS Cortical bone, Paleonutrition, Ca intake, Ca:P ratio, Huronia ABSTRACT Two samples of protohistoric Iroquoians are found to show low values for bone formation indicators. Two southern Ontario ossuary samples, Kleinburg (1600 AD, N = 561) and Uxbridge (1490 2 80 AD, N = 457) are examined. Mean values for the cortical index (Barnett and Nordin, 1960) and percent cortical area (Garn, 1970)are consistently below expected normal values. Some of the lowest values may reflect the presence of a disease state. Nevertheless, it is argued that the results may indicate chronic dietary insufficiencies. Both absolute calcium intake and the calcium to phosphorous ratio are discussed as possible causative factors. Past researchers have frequently noted differences in adult cortical bone remodeling among prehistoric skeletal samples (Cassidy, 1972;Perzigian, 1973;Eriksen, 1976;Richman et al., 1979).Differences in the amount of cortical bone have been tentatively attributed to dietary differences among groups. It has been recognized that the nutritional influenceson cortical remodeling cannot easily be differentiated from genetic and biomechanical influences. Furthermore, there is no simple relationship between calcium intake and bone formation or maintenance (Garn et al., 1969b). Although calcium intake may influence the incidence of senile osteoporosis (Matkovic et al., 19791,other factors such as meat protein and phosphorus intake complicate the relationship (cf. Chinn, 1981;Draper and Bell, 1979). One reason for the inability to assess accurately the influence of dietary factors on bone kinetics is that researchers must base their observations on generalized dietary reconstructions. We rarely have accurate dietary information and skeletal material from the same population. A general lack of energy intake or a specific dietary deficiency may be postulated, but the data do not facilitate the testing of hypotheses. Documents from the period of historic European contact in southern Ontario furnish relatively thorough dietary information for certain Amerindian groups. The Huron Indians, 0002-9483/83/6001-0023$02.00 0 1983ALAN R. LISS, INC. especially, were the subject of ethnographic documentation by explorers, Recollet and Jesuit missionaries from 1615 to their defeat and dispersal by the Iroquois in 1649-50. These various accounts have fiumished a basis for numerous ethnographic reconstructions (cf. Tooker, 1964; "rigger, 1969, 1976). Heidenreich (1971)approached the records from a cultural geographer's perspective, and established a tentative profile of probable food sources and preferences (see Table 1). It is within this context that two large Iroquoian samples were examined for their relative amount of adult cortical bone. MATERIALS AND METHODS Skeletal samples from two Iroquoian 08suaries were examined. The Kleinburg Ossuary, excavated in 1970 by F. J. Melbye, has been dated by trade goods to ca. 1600 AD. The Uxbridge Ossuary, excavated in 1975-77 by P. Cook, is radiocarbon dated to 1490 2 80 AD. Both sites were located north of Toronto, Ontario, not far from historic Huronia. An Iroquoian ossuary sample consists of the thoroughly mixed skeletal remains of everyone who had died during the 8-12 years preceding ossuary construction. According to ethno- Received March 22,1982;aceepted August 16,1982. 24 S. PFEIFFER AND P. KING TABLE 1.Recommended duilv intake of nutrients compared with Humn intake ~~~~ ~~~ ~ Protein Calories Recommended intake Daily food 3,000 Huron intake Corn 1,950 Squash 50 Beans 400 300 Fish Meat 150 Fruit 150 Totals 3,000 gm 70 55.0 1.4 6.0 40.0 25.0 1.6 129.0 ~ Calcium mg Iron mg 800’ 10 120 38 38 14.0 0.6 2.0 1.8 3.2 2.3 23.9 11 34 230 Vitamin A Ascorbic acid Thiamine mg Riboflavin mg Niacin mg 5,000 1.2 1.7 19 70 3,000 9,723 220 240 2.25 0.05 0.14 0.24 0.06 0.06 2.80 0.7 12.0 0.9 1.0 14.7 4.5 0.8 33.9 - IU 10 224 13,417 0.8 0.1 0.1 0.2 1.0 2.9 mg 8 14 - - 32 54 ‘Represents Canadian and American RDA. The FAOiWHO (1962)suggests 400-500 mgiday for adults on cereal diets. Reproduced from Heidenreich (1971)with permission graphic sources, exceptions to universal inclusion may be newborn infants and males killed in battle. Infants are, indeed, underrepresented in most ossuary samples, but there is no evidence that adult males are consistently underrepresented. The Kleinburg sample includes a minimum of 561 individuals, of whom 28% are immature. Most adults died prior to age 40, based on pubic symphysis assessment. The adult sex ratio is approximately balanced. The Uxbridge sample includes a minimum of 457 individuals, of whom 31% are immature. Here, too, most adults died prior to age 40, and there is no pelvic evidence for an unbalanced sex ratio. Analysis of Uxbridge adult age and sex is currently preliminary. From the Kleinburg sample, 96 left femoral midshafts, 22 lumbar vertebrae, and 130 second left metacarpals were examined. From the Uxbridge sample, 90 second left metacarpals were examined. The Kleinburg material was assessed by two separate, but closely correlated, techniques. Followingthe technique of Barnett and Nordin (1960),measurements were taken from metacarpal radiographs and femoral cross sections to assess their relative cortical thickness (the femoral and metacarpal scores), and radiographs of the lumbar vertebrae were used to calculate the relative anterior body height (the lumbar score). The scores are essentially indices of medial and lateral cortical thicknesses over total midshaft breadth, or posterior over anterior body height. These three measurements taken together were proposed by Barnett and Nordin (1960)to be a technique for the diagnosis of osteoporosis. Although this assessment is no longer considered very useful for that purpose (Nordin et al., 1970), i t is still a valuable technique for quantification of cortical bone. Reference values for nonosteopo- rotic Britons are useful for comparison. The technique has been used for previous studies of prehistoric samples (cf. Cassidy, 1972;Cook, 1979),and it correlates well with measures of percent cortical area (r > 0.90)(Garn, 1970). Barnett and Nordin (1960)used the third lumbar vertebra to calculate their lumbar score. We were unable to identify third lumbar vertebrae with certainty, so we used fifth lumbar vertebrae. The second technique used to quantify cortical bone status is the calculation of Garn’s (1970)percent cortical area. The second left metacarpals from both the Kleinburg and Uxbridge samples were measured to this end. All measurements were taken from radiographs, using a Helios needlepoint caliper. Indeed, the same measurements are used to calculate both Barnett and Nordin’s score and Garn’s percent cortical area. The difference between the two measures is that the score is one-dimensional and the area measurement is two-dimensional (Garn, 1970,p 65).The Barnett and Nordin scores are included here because they offer an indication of bone status from multiple skeletal sites, from the trunk as well as appendages. RESULTS A comparison of the ossuary-derived Barnett and Nordin scores with those of normal Britons is displayed in Figures 1-3. Each of the three Kleinburg distributions is skewed to the left. Using Barnett and Nordin’s original cutoff values for the determination of osteoporosis, 24% of the femora, 21.5% of the metacarpals, and 36% of the vertebrae would be classified as “osteoporotic.” The mean values derived for percent cortical areas for Kleinburg and Uxbridge metacarpals are 75.9% (s = 8.22)and 74.5% (s = 10.431, 25 CORTICAL BONE AND DIET FEMORAL SCORE METACARPAL SCORE 30. 25- N.125 25 % N.125 20- 20. 1510. 10. 5. 5 SCOR 5- 5- 10- % 10. 15 15 20- 20. 25- N = 96 3oJ 25 4 Fig. 1. Femoral cortex scores for Barnett and Nordin’s British reference sample, all over 50 years of age (above), as compared to Kleinburg Ossuary femora (below). A low score indicates a low amount of cortical bone per unit of area. Fig. 2. Second left metacarpal cortex scores for Barnett and Nordin’s British reference sample, all over 50 years of age (above),as compared to Kleinburg Ossuary metacarpals (below). A low score indicates a low amount of cortical bone per unit of area. respectively.These are substantially lower than the population means summarized by Garn (1970), which range from 84.2% (Costa Rican males) to 91.0% (El Salvador females). Garn’s weighted pooled means are 85.7% for males, and 89.7%for females. One proposed statistical limit of osteoporosis is a percent cortical area value that is 2 SD below the mean. Using the lowest population mean (84.2, s = 6.8) the osteoporosis cutoff is 70.6%. The cortical area values of 25.6% of the Kleinburg metacarpals and 26.7% of the Uxbridge metacarpals fall at or below this value. The distributions of metacarpal cortical areas for Kleinburg and Uxbridge are plotted in Figure 4. The plot of the Uxbridge values is markedly bimodal, with approximately 10%of the area values lying below 60. This group of low values explains the lower mean value for Uxbridge as compared to Kleinburg. It has been argued elsewhere that the Uxbridge material shows strong evidence of tuberculosis or some similar condition (Pfeiffer, 1981). Evidence of premortem vertebral destruction and sacroiliac deterioration has been used to estimate that a minimum of 20-26 adults, or 6.5-8%, were affected. Similar cases of premortem deterioration are not present at Kleinburg, at least not in substantial numbers. The lowest area values from the Uxbridge sample are therefore likely to be linked to a disease state. Figure 4 also illustrates that the modal class for the two samples is around 80, a value much closer to, but still below, Garn’s range of population means. DISCUSSION The two protohistoric Iroquoian populations examined here demonstrate low values for amounts of cortical bone, as expressed by Barnett and Nordin scores or the calculation of percent cortical area. Results have been expressed in terms of osteoporosis not because we wish to argue that a disease state is present, but because we wish to emphasize the general trend toward low values for amounts of cortical bone. Finer levels of analysis are not possible using these ossuary samples, because the age and sex of the individual specimens are unknown. The sex of complete Kleinburg femora has been estimated using discriminant function analysis S.PFEIFFER AND P. KING 26 LUMBAR SCORE 30 N=125 % sco % N=22 25 Fig. 3. Lumbar body scores for Bamett and Nordin’s British reference sample, all over 50 years of age (above), aa compared to Kleinburg os~uaryvertebrae (below).A OW score indicates some degree of anterior body wedging. % 1 (Pfeiffer, 1979).This represents only a fraction of the total sample, however. We also tried to discriminate the sexes statistically by comparing Mexican-American metacarpal measurements (Garn et al., 1973), but, because of their gracility, results were unrealistically skewed toward females. We can only say that the samples are composed primarily of young adults, and probably include approximately equal representation of males and females. Given these demographic characteristics, cortical remodeling appears to be markedly deficient. Heidenreich argued that the Huron diet (Table 1)was likely to have been adequate in both energy content and nutrients. Intake of calcium and vitamin C appears to have been below current recommended allowances, but “additional calcium was probably obtained by gathering calcium-rich foods such as clams, nuts, and crabs.” Furthermore, “the water supply in Huronia is high in calcium carbonate” (Heidenreich, 1971, p 165). Meat protein intake was probably low, and exposure to sunlight was adequate. Both these factors reduce the need for dietary calcium. There are no docIUnented Ca8es Of faders Heidenreich to conclude that dietary calcium intake was marginally adequate. 10 5 0 4i-4‘5 5b $5 40 45 o; 7’5 s’o s’5 do 915 id0 PCA (Percent Cortical Area) Fig. 4. Distrihution of percent cortical area values, second left metacarpals from the Kleinburg and Uxhridge ossuaries. 27 CORTICAL BONE AND DIET It is now recognized, however, that availability of calcium for bone tissue formation depends not only on dietary calcium but also the ratio of phosphorus to calcium in the diet (cf. Draper and Scythes, 1981). The optimal ratio varies according to the absolute calcium content, but generally, as the amount of phosphorus increases beyond a Ca:P ratio of 1:1, there is a rise in plasma phosphorus, a decline in plasma calcium, and a consequent increase in parathyroid hormone synthesis and bone resorption. A 6-month study on adult pigs (whose bones undergo cortical remodeling) demonstrated that a Ca:P ratio of 1:3, with moderate calcium intake, was associated with decreased rib ash, decreased femoral bone formation, and increased lumbar vertebral marrow width (DeLuca et al., 1976). Such effects are most marked when calcium concentration in the diet is high. Because the efficiency of calcium absorption decreases at high intakes and the efficiency of phosphorus absorption remains unchanged, both absolute and relative calcium intake must be considered when evaluating the possible effects of the Ca:P ratio. By all accounts, the staple food in the Huron diet was corn. Corn is relatively low in calcium and high in phosphorus. Using a standard table of food composition values (Health and Welfare Canada, 19711, the reconstructed Huron diet shows a calculated Ca:P ratio of 1:3.8, due mainly to the corn component. However, nearly half the phosphorus in corn is contained in insoluble phytin. This means that the ratio of available nutrients may be closer to 1:2. The effect of this ratio on bone formation, assuming low absolute calcium levels, should not be very great for adults. Animal experimentation indicates that the effect on rapidly growing juvenile bone could be marked, however. Protein-calorie malnutrition in the growing child has been shown to result in low percent cortical areas (Garn et al., 1964,1969a). Childhood differences in the cortical index have been used to try to discriminate paleonutrition among prehistoric Illinois Valley groups (Cook, 1979). We know of no evidence, however, that such childhood cortical deficiencies are normally retained into adulthood. The low values for adult bone formation in the Kleinburg and Uxbridge samples are not clearly explained by any single dietary variable. A number of chronic factors may, however, be implicated. These include low calcium intake, excess phosphorus relative to calcium, and protein-calorie malnutrition. Genetic and biomechanically based hypotheses are less at- tractive. Although it is clear that there are differences in bone dynamics among major racial groups, there is thus far no evidence for genetic differences at the more local level. We also have no reason to believe that the Iroquoian horticulturists were markedly inactive. Further work should compare corn-baaed horticulturists with related groups depending on other dietary regimes. This approach may differentiate any unique bone formation characteristics associated with a corn-based, low meat diet from those possibly associated with protein-calorie insufficiency. ACKNOWLEDGMENTS We wish to thank Dr.F. J. Melbye of Erindale College, University of Toronto, for access to the Kleinburg sample. Radiography was kindly provided by Mr. Carl Basiano and Mr. Alex Nelson of the Ontario Veterinary College, University of Guelph. We thank Dr. H. H. Draper for his review of the manuscript. Funding for this research was partially furnished by a grant from the Social Sciences and Humanities Research Council of Canada. LITERATURE CITED Barnett, E, and Nordin, BEC (1960) The radiological diagnosis of osteoporosis: A new approach. Clin. Radiol. 11:166-174. Cassidy, CM (1972) A comparison of nutrition and health in pre-agricultural and agricultural Amerindian skeletal populations. Ph.D. Thesis, University of Wisconsin. Chinn, HI (1981) Effects of dietary factors on skeletal integrity in adults: Calcium, phosphorus, vitamin D and protein. Life Sciences Research OtXce, Fed. Am.Soc. Ex. Biol., 9650 Rockville Pike, Bethesda, MD. Cook,DC (1979) Subsistence base and health in prehistoric nlinois valley: Evidence from the human skeleton. Med. Anth. 4:109-124. DeLuca, HF, Castillo, L,Jee, W,Oldham, S,and Grummer, RH (1976) Studies on high phosphate diets. Food Research Institute Annual Report, University of Wisconsin-Madison, pp. 394498. Draper, HH, and Bell, RD (1979) Nutrition and osteoporosis. In HH Draper (ed):Advancea in Nutritional Research, Vol. 2. New York: Plenum, pp. 79-106. Draper, HH, and Scythes, CA (1981) Calcium, phosphorus and osteoporosis. Fed. Proc. 40:9:2434-2438. Eriksen, MF (1976) Cortical bone loas with age in three native American populations. AJF’A 453.443452. Food and Agriculture OrganizationWorld Health Organization (1962) Calcium requirements. Report of a FAOl WHO expert committee on calcium requirements. WHO Tech. Rep. Ser. 230. Rome: FAO, pp. 1-54. Garn, SM (1970) The Earlier Gain and the Later Loss of Cortical Bone; In Nutritional Perspective. Springfield, Illinois: Thomas. Garn, SM,GuzmBn, MA, and Wagner, B (1969a) Subperiwteal gain and endosteal loss in protein-calorie malnutrition. AJPA 30:153-6. Garn, SM, Rohmann, CG,Wagner, B, Dairla, HG,and Ascoli, W (1969b) Population similarities in the onset and rate of adult endosteal bone I-. Clin. Orthop. 65:51-60. 28 S.PFEIFFER AND P. KING Garn, SM, Poznanski, AK, and Larson, K (1973)Metacarpal lengths, cortical diameters and areas from the 10-state nutrition survey. In ZFG Jaworski (4) Pmceedings : of the First Workshop on Bone Morphometry. Ottawa, Canada: University of Ottawa Press. Garn, SM, Rohmann, CG, Behar, M,Viteri, F,Gunnan, MA (1964)Compact bone deficiency in protein-calorie malnutrition. Science 145:144&1445. Health and Welfare. Canada (1971)Nutrient Value of Some Common Foods. Ottawa: Health and Welfare Canada. Heidenreich, C (1971)Huronia: A history and geography of the Huron Indians 1600-1650.Toronto: McClelland and Stewart. Matkovic, V, Kostial, K, Simonovic, I, Buzina, R,B d a r e c , A, and Nordin, BEC (1979)Bone status and fracture rates in two regions of Yugoslavia. Am. J. Clin. Nutr. 32:540549. Nordin, BEC, Young, MM, Bulusu, L, and Horsman, A (1970) Osteoporosis re-examined. In U. B m l (ed): Osteoporosis. New York: Grune and Stratton, pp. 47-67. Penigian, AJ (1973)Osteoporotic bone loss in two prehistoric Indian populations. AJPA 39:1:87-96. F'feiffer, S (1979)Assignment of sex to adult femora from an oasuary population. Can. Rev. Phys. Anthropol. 156-62. PfeiEer, S (1981)Preliminary report of paleopathology from the Uxbridge ossuary. Paper presented to CAPNAAPC, Banff, Alberta, December 1981. Richman, EA, Ortner, DJ, and Schulter-Ellis, FP (1979) Differences in intracortical bone remodelling in three aboriginal American populations: Possible dietary factors. Calcif. " h u e Int. 28:209-214. Tooker, E (1964)An Ethnography of the Huron Indians, 1616-1649.Bur. Am. Ethnol. Bul. No. 190.Washington: Smithsonian Institution. Trigger, BG (1969)The Huron Farmers of the North. New York Holt, Rinehart and Winston. Trigger BG (1976)The Children of Aataentsic, a History of the Huron People to 1660. Montreal: McGillQueen's University Press.