AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 71:173-183 (1986) Periodontal Disease in Ancient Populations N.G. CLARKE, S.E. CAREY, W. SRIKANDI, R.S. HIRSCH, AND P.I. LEPPARD Department of Dentistry, The Uniuersity of Adelaide, Adelaide, South Australia 5000 KEY WORDS lesions Periodontal disease, Prehistory, Alveolar bone ABSTRACT Recent clinical and anthropological findings indicate that the conventional concept of the pathogenesis of periodontal disease requires review. The periodontal lesion has been defined as a generalised horizontal loss of crestal bone resulting from host immune and inflammatory responses triggered by the action of commensal bacteria, and the extension of gingivitis into the deeper periodontium to become periodontitis has been assumed to occur slowly but steadily over many years. Anthropological and clinical investigations reveal that the widespread loss of crestal tissue is relatively unusual and that lesions of the alveolus are commonly localised and severe. Longitudinal studies have shown that the disease progresses in bursts and is stable in both the gingivitis and periodontal modes in between the burst activity. The findings of the present study demonstrate that generalised horizontal periodontitis has been unusual and has not been responsible for tooth loss. Other factors responsible for deficient alveolar margins in dry bones have been overlooked in most studies, leading to overassessment of the incidence of periodontal disease in postmortem materials; the same assumptions have led to overassessment of periodontal disease in clinical studies and practice. Periodontal disease is considered to be an ancient affliction of humans (Gold, 1985). The traditional concept of the pathogenesis of the disease cites the involvement of commensal oral microbes in the initiation and progression of the disease. Bone lost from the alveolar crest by active resorption is associated with loss of periodontal ligament integrity. In turn, exposed cementum of the root becomes covered by extension of the gingival epithelium to form a more apical attachment and a periodontal pocket. It is thought that over a period of 30-40 years bacterial products provoke the host into a damaging defensive reaction that is slowly progressive (Loe et al, 1978). Extension of the lesion from the gingivae into the alveolus in time was thought to result in reduced tooth support and eventually in tooth loss. Page and Schroeder (1976) stated that the natural history of the disease assumes that some time will be spent in the gingivitis phase of the lesion before the inflammation extends into the deeper periodontal structures, although they later conceded that this 0 1986 ALAN R. LISS, INC may not be the case (Page and Schroeder, 1982). Confusion occurred regarding the interpretation of the results of clinical and epidemiological studies because gingivitis typically does involve wide areas of the mouth, and the assumption was made that the inflammation would extend to the underlying bone. The dental techniques which have evolved over the last two centuries to prevent or minimize the long-term consequences of periodontal disease reflect these assumptions. While there is no evidence to support the assumption that gingival inflammation would extend to the underlying bone, support can be offered for the concept that gingivitis does not usually extend into the underlying bone (Listgarten et al., 1985), and that periodontitis does not progress continuously (Socransky et al., 1984). The findings of Received December 13, 1985; revision accepted May 5, 1986. Address reprint requests to Dr. N. C . Clarke at the above address. 174 N.G. CLARKE ET AL TABLE 1. Location of cranial materials Museum Location S. Australia Mus. Bishop Mus. Univ. of Arizona N. Western Univ. Smithsonian Smithsonian Smithsonian Adelaide Hawaii Tucson Evanston Washington Washington Washington Smithsonian Harvard Univ. Harvard Univ. Washington Harvard Harvard Am. M. Nat. Hist. Am. M. Nat. Hist. Am. M. Nat. Hist. Edinburgh Univ. Edinburgh Univ. Brit. M. Nat. Hist. Brit. M. Nat. Hist. Royal Coll. Surg. Duckworth Mus. Duckworth Mus. Mus. de 1’Homme Mus. de 1’Homme Copenhagen Univ. Copenhagen Univ. Copenhagen Univ. Copenhagen Univ. Om Monastery Prague Mus. Prague Mus. Hadassah Univ. Witwatersrand Univ. Otago Univ. Auckland Mus. Aust. Nat. Univ. New York New York New York Edinburgh Edinburgh London London London Cambridge Cambridge Paris Paris Copenhagen Copenhagen Copenhagen Copenhagen Ry (Aarhus) Prague Prague Jerusalem Johannesberg Dunedin Auckland Canberra Sagne and Olsson (1977) and Costa (1982) show that the conventional scenario is not true for a wide range of premodern ethnic groups. Moreover, the causes of bone loss observed in dry skulls frequently have been misinterpreted, resulting in overestimation of both the incidence and severity of periodontal disease (Costa, 1982). In the present study a n attempt was made to discriminate between alveolar modifications resulting from bacterially related crestal bone resorption (periodontal disease) and anatomical, developmental, physiological, and other pathological defects that may be responsible for the increased distances between the cemento-enamel junction (CEJ) and the alveolar crest (AC) in excess of the “normal” 2mm. This discrimination between the variety of causes of deficient alveolar bone probably accounts for the low incidence GrouD No. Aust. Aborigine Mokapu, Oahu Grasshopper Ind. Woodland Ind. Eskimo (Alaska)Tanunuk Egypt (XI1 Dyn) Peru (Chicama, Pachacamac) Terry (modern) Mexico (Yucatan) Yugoslavia (medieval) Panik New Britain (Ralum) Bolivia (Sicasica) Hungary (Keszo Hidegkut) India (Southern Prov) China (Hokien) Romano-Brit. (Poundbury) Egypt (Hawara) 2-3rd cent. Ancient Brit. (Breedon) Somali India (N. W. Prov) Romano-French (Maule) Japan (Izumu) Eskimo (Greenland) Romano-Danish (300 B.C.) ‘modern’ Danish Egypt (Nubia 350 A.D.) Danish (800 A.D.) Czech (Ducove middle ages) Czech (Rajarad 9th cent.) Jerusalem Bantu (modern) Maori Maori Aust. Aborigine 52 49 50 40 40 32 47 23 39 40 20 20 40 20 20 40 20 20 10 15 54 40 22 37 21 33 43 33 40 16 51 46 54 42 - of bone loss attributable to periodontal disease in this study. MATERIALS Sku 11 selection Cranial materials located in 20 museums in ten countries from 34 populations were measured for bone loss believed to be of a periodontal origin. In total, 30,057 teeth from 1,149 skulls were documented to provide the data shown in Tables 2-7. The jaws of two modern populations from South Africa and St. Louis (U.S.A.)were included for comparison with the ancient material. The sample populations are listed in Table 1. Skull selection was based upon maximal development of dental pathology. In all instances, worst-case specimens were diligently sought, and the data reflect the most severe conditions that could be found in each 175 DISEASE IN ANCIENT POPULATIONS group. Where the collections contained a n abundance of materials, the first 40-50 suitable skulls were used (except for the Terry collection where only 23 samples were examined). In smaller collections, or in instances in which the incidence of alveolar pathology was very low, all skulls were examined and the sample size reflects the total number of suitable skulls available for that population. It was not intended that the sample be representative of the population but that the maximum incidence of periodontal pathology should be measured. Most of the skulls selected were in sufficiently good condition to provide confidence in the estimation of bone height and the presence or absence of disease. METHODS Measurement criteria In estimating the alveolus lost as a result of periodontal disease, the general practice of assigning any increase over 2mm in the CEJ to AC distance as periodontal disease was rejected. It was recognised that there is a need to account €or anatomical, developmental, pathological, and physiological causes of increase over 2mm in the CEJ to AC distance. Several reports have shown that there is a progressive increase in the CEJ to AC distance throughout life, and some evidence exists to indicate that this “physiologic” bone loss corresponds with the degree of compensatory process of continuous active eruption (Murphy, 1959; Newman and Levers, 1979; Tal, 1985; Whittaker et al., 1985). Periodontal disease was recorded only when the bone crest of the alveolar margins showed either the loss of the cortical surface, revealing porous cancellous spaces, or a n altered morphology of the alveolar crest. These changes resulted in the appearance of a porous bony surface or a shelflike margin instead of the normal knife-edged configuration (Fig. 1).Great care was taken with the assessment of the bony status, and when bone was considered to have been lost as a result of periodontal disease, measurements were made on both the buccal and proximal surfaces. A William’s periodontal probe was used to classify the bone loss into three categories: 0-2mm, 2-4mm, and > 4mm. While the difficulties in assessing both quality and quantity of alveolar bone may not have been entirely overcome, the data reflect the best effort to be accurate, objective, and consis- tent. All materials were measured by the same investigator (N.G.C.), with the exception of the South Australian collection, which was measured by N.G.C. and W.S. together. In the present study, periodontal granulomae with origins apical to the alveolar crest were classified as being of pulpal origin. It was possible to record the location of a typical dental abscess a t the root apex and identify the extension of the abscess along the ligament to reach the crest (Fig. 2). Lesions in this category traditionally have been erroneously called periodontal lesions, as clinically they are probed as periodontal pockets. Lesions located in the root fork (furcation lesions) without evidence of adjacent crestal mesial or distal proximal periodontal disease were also considered to be of pulpal origin (Fig. 31, as were lateral wall lesions where discrete granulomae formed without continuity with the crestal tissue (Fig. 4). The present data therefore discriminated between alveolar bone loss of marginal (gingival) origin (Fig. 1)and alveolar bone loss of pulpal origin resulting from the extension of the lesions along the ligament to discharge inflammatory exudate into the gingival sulcus (Figs. 2-4). Only data for periodontal lesions of marginal (gingival) origin are presented here, and in Tables 3, 4, 6 , and 7 the data shown are for the right quadrants only. Data processing The data were processed by a Vax computer of The University of Adelaide using the Biomedical Package (BMDP) software package developed by Statistical Sofeware, Inc., of the University of California, Los Angeles. Statistical tests for significance were considered inappropriate due to the lack of randomness which results from selecting “worst case” skulls, and to the large sample sizes. Data are therefore presented as percentages for comparison and interpretation. RESULTS The data for teeth free from periodontal disease, and for the three categories of bone loss by group, indicate that a marked difference existed between the premodern and modern groups (Table 2). The incidence of teeth free from periodontal disease in the premodern groups ranged from 99.9% (Romano-Danish) to 76% (Mokapu), with twothirds of the groups showing more than 90% 176 N.G. CLARKE ET AL. Fig. 1. Moderate marginal periodontal disease reveals both an altered bone morphology of the crest and “porosity” in the surface of the bone. Fig. 2. Severe localised vertical loss of bone. Severe attrition resulted in the exposure of the pulp chamber and establishes the relationship between heavy tooth wear and periodontal lesions. DISEASE IN ANCIENT POPULATIONS Fig. 3. Alveolar bone was destroyed in the furca of the 36. A pulpal etiology for the furcation lesion may be deduced from the presence of an abscess on the apex of the mesial root. Fig. 4. Lateral root abscesses are present on two adjacent teeth. The crowns show evidence of severe tooth wear. 177 178 N.G. CLARKE ET AL. TABLE 2. Degree ofperiodontal disease by group Alveolar bone loss Buccal (proximal) Normal Aust. Aborigine Mokapu Indian Grasshopper Indian Woodland Indian Eskimo (Alaska) Egypt (XI1 dyn) Peru (Chicama) Mexico (Yucatan) Yugoslav (medieval) New Britain @alum) Bolivia (Sicasica) Hungary (Hidegkut) India (S. Prov) China (Hokien) Rorn-Brit. (P'bury) Egypt (Hawara) British (Breedon) Somali India (N. W. Prov) Rorn-French (Made) Japan (Izumu) Eskimo (Greenland) Rom-Danish (300 B.C.) Modern Danish Egypt (Nubia 350 A.D.) Danish (800 A.D.) Czech (Ducove) Czech (Rajarad) Jerusalem Maori Maori Aust. Aborigine % Total teeth Terry (modern) Bantu (modern) % Total teeth 94.9 (92.1) 70.5 (81.3) 80.7 (82.2) 75.6 (83.6) 88.8 (91.0) 89.2 (90.6) 98.3 (95.4) 90.8 (89.0) 78.7 (79.5) 86.1 (82.7) 93.3 (92.6) 81.3 (80.7) 84.6 (86.4) 84.8 (77.6) 94.6 (95.9) 94.0 (94.4) 97.6 (97.1) 97.7 (95.5) 89.9 (79.3) 99.1 (98.7) 95.2 (90.0) 99.2 (99.0) 100.0 (99.8) 89.9 (86.7) 99.8 (98.3) 94.7 (91.8) 97.5 (97.0) 96.4 (93.2) 86.8 (79.5) 99.1 (98.9) 98.1 (98.0) 97.4 (93.2) 90.8 (90.3) 66.2 (64.4) 78.2 (70.8) 74.7 (68.9) % Mild 4.2 13.5 11.8 16.3 5.8 7.4 1.6 5.9 10.1 9.2 5.4 11.6 10.2 12.2 4.5 3.9 2.0 1.5 5.9 0.4 3.6 0.5 0.0 8.1 0.2 2.9 1.5 2.7 6.3 0.9 0.9 1.8 5.7 13.0 14.1 13.8 (5.6) (6.6) (9.0) (9.6) (5.4) (6.1) (3.8) (6.6) (9.0) (10.3) (5.4) (12.2) (9.1) (18.3) (2.9) (3.6) (2.2) (3.8) (10.1) (0.8) (8.5) (0.2) (0.2) (10.2) (0.8) (5.1) (2.7) (5.3) (11.2) (1.1) (0.0) (3.9) (5.8) (15.6) (21.0) (19.4) % Moderate 0.8 (1.7) 11.6 (8.5) 3.9 (5.7) 7.7 (6.5) 5.1 (2.7) 3.3 (3.1) 0.1 (0.7) 3.1 (3.9) 9.6 (9.8) 4.5 (6.2) 1.3 (1.7) 6.2 (6.3) 3.8 (3.6) 2.0 (3.0) 0.7 (1.1) 2.1 (1.9) 0.2 (0.6) 0.8 (0.4) 3.9 (8.9) 0.5 (0.5) 1.3 (1.5) 0.3 (0.5) 0.0 (0.0) 1.8 (2.6) 0.0 (0.8) 2.1 (2.9) 1.0 (0.2) 0.9 (1.5) 5.2 (7.7) 0.0 (0.0) 0.5 (0.7) 0.8 (2.7) 2.9 (3.1) 14.4 (12.8) 6.8 (7.1) 9.0 (8.8) 9% Severe 0.1 (0.5) 4.4 (3.6) 3.5 (3.1) 0.3 (0.4) 0.3 (0.8) 0.1 (0.1) 0.0 (0.1) 0.1 (0.5) 1.6 (1.6) 0.2 (0.8) 0.0 (0.3) 0.8 (0.8) 1.3 (0.0) 1.0 (1.2) 0.2 (0.1) 0.0 (0.2) 0.2 (0.2) 0.0 (0.4) 0.2 (1.7) 0.0 (0.0) 0.0 (0.0) 0.0 (0.3) 0.0 (0.0) 0.2 (0.5) 0.0 (0.0) 0.3 (0.2) 0.0 (0.0) 0.0 (0.0) 1.6 (1.6) 0.0 (0.0) 0.5 (0.5) 0.0 (0.3) 0.6 (0.7) 6.5 (7.2) 0.8 (1.1) 2.5 (2.9) n 1,487 1,371 1,106 1,097 1,093 733 690 1,038 1,098 595 299 ,056 ,125 597 ,085 534 543 264 406 777 ,094 603 999 607 836 ,263 881 ,152 365 1,008 1,032 1,245 26,079 585 1,393 1.978 n, number of teeth in each group; mild, 0-2mm; moderate, 2-4mm; severe, > 4mm. of teeth to be disease free and nearly one third to be disease free at the 79-90% level. Only one group (Mokapu) showed less than 79% of teeth to be disease free. In the two modern groups, 74.5% (South Africa) and 65.3% (Terry) of teeth were free of periodontal disease. The only sample from the premodern groups that showed a level of periodontal disease comparable to that of the modern groups was Mokapu, and there was doubt concerning whether that group may in fact be relatively modern. to periodontal disease. The incidence of buccal periodontal disease was greater in the mandible than in the maxilla for all categories of severity (Table 3). Mild bone loss was observed adjacent to 5.6% of teeth, moderately involved teeth constituted 2.9% of the total. There was a very low incidence of severely involved teeth (0.6%), and this category was found in only four of the 32 premodern groups examined. In no case was the loss of alveolar tissue which was due to periodontal disease of sufficient severity to jeopardise tooth retention. BUCCAL ALVEOLAR BONE LOSS When buccal bone loss by tooth type was Premodern groups considered in the maxilla, the first and secThe majority of teeth examined (90.8%) ond molars were the most frequently inwere free from buccal alveolar bone loss due volved in all categories of periodontal disease 179 DISEASE IN ANCIENT POPULATIONS TABLE 3. Alveolar bone loss due to periodontal disease in 32 premodern and two modern groups’ buccal surfaces Premodern Modern Maxilla No disease Mild disease (0-2mm) Moderate disease (2-4mm) Severe disease (> 4mm) Total teeth Mandible Maxilla Mandible n (%) n (%) n (%I n (%) 13,525 713 192.2) (4.9) 11,972 875 (89.21 (6.5) 759 121 (77.3) (12.3) 718 152 (72.0) (15.3) 352 (2.4) 466 (3.5) 87 (8.9) 92 (9.2) 76 (0.6) 100 (0.7) 15 (1.5) 34 (3.4) 14,666 13,413 982 996 TABLE 4. Alveolar bone loss due to periodontal disease in 32 premodern groups’ buccal surfaces of individual teeth Tooth 18 17 16 15 14 13 12 11 Total No. of teeth 758 878 860 929 972 1,002 973 950 7,322 Upper right quadrant Severe (%) 0.7 Moderate (%) 3.7 Mild (%) 7.1 Normal (%) 88.5 0.9 4.4 10.0 84.6 1.4 5.0 7.7 85.9 0.2 2.0 5.1 92.8 0.2 2.2 4.3 93.3 0.2 1.2 2.9 95.7 0.2 0.9 2.5 96.4 0.0 1.1 1.6 97.2 0.4 2.5 5.0 92.1 47 46 45 43 42 41 Total Tooth 48 44 No. of teeth 725 773 722 867 914 929 908 873 6,711 Lower right quadrant Normal 1%) 91.4 5.4 Mild (%I Moderate (%a) 2.8 Severe (%) 0.4 84.9 9.8 4.7 0.6 80.0 13.3 5.3 1.5 93.0 4.4 2.2 0.5 93.4 3.5 2.4 0.7 93.3 3.7 1.7 1.3 89.4 6.4 3.4 0.8 89.2 6.5 3.7 0.6 89.6 6.4 3.2 0.8 Mild, 0-2mm; moderate, 2-4mm; severe, >4mm. severity, with involvement approximately twice as frequent as in the central incisors. Mandibular first molars were more vulnerable than any other unit in the mandible or maxilla for all categories (Table 4). Modern groups The most notable finding from the combined results of the two twentieth century groups was that, while the incidence of periodontal disease was greater than in that of the premodern specimens, the incidence of severe alveolar bone loss resulting from periodontal disease was still very low. Seventyfive percent of teeth were free from periodontal disease (Table 3); mild periodontal disease was observed in 13.8% of the buccal surfaces; and moderate periodontal loss was estimated to have occurred in 9.0% of the surfaces. This compares with 90.8, 5.6, and 2.9%, respectively, in the combined premodern groups. The pattern of bone loss by tooth type was less regular than in the premodern groups of skulls, and the severity of the periodontal loss was greater. The areas of greatest damage in the maxilla were in the molar regions in all categories of disease (Table 4). In contrast, the mandibular patterns of bone loss were the most severe in the incisior region for all categories, but in the posterior quadrant the first molar was the more severely involved. The overall percentages of involve- 180 N.G. CLARKE ET AL. TABLE 5. Alveolar bone loss due to periodontal disease in two modern groups’ (American and South African) buccal surfaces of individual teeth Tooth 18 17 53 Upper right quadrant Severe (%) 0.0 Moderate (%I 15.1 18.9 Mild (%I Normal (%) 66.0 16 15 14 13 12 11 Total 69 62 61 498 65 61 63 No. of teeth 64 1.5 13.8 12.3 72.3 6.5 13.1 13.1 67.2 1.6 3.2 17.4 77.8 1.6 7.8 15.6 75.0 0.0 4.4 11.6 84.1 0.0 4.8 11.3 83.9 0.0 3.8 9.8 86.9 1.4 8.1 13.7 76.9 47 46 45 44 Tooth 48 50 Lower right quadrant Normal (%) 84.0 Mild (%) 8.0 Moderate (%) 6.0 Severe (%) 2.0 55 57 67 No. of teeth 69 72.7 14.5 12.7 0.0 68.4 15.8 14.0 1.8 82.1 13.4 3.0 1.5 82.6 11.6 4.3 2.9 43 42 41 Total 70 68 60 496 75.7 15.7 5.7 5.9 55.8 20.6 17.6 5.9 55.0 21.7 16.6 6.7 72.0 15.3 9.9 2.8 Mild, 0-2mm; moderate, 2-4mm; severe, >4mm. TABLE 6. Alveolar bone loss due to periodontal disease in 32 premodern and two modern groups’ proximal surfaces Pre-modern Maxilla No disease Mild disease 0-2mm) Moderate disease (2-4mm) Severe disease (> 4mm) Total teeth Modern Mandible Maxilla Mandible n (%) n (%I n (%) n (%) 13,459 753 (91.8) (5.1) 11,904 884 (88.7) (6.6) 668 197 (68.0) (20.1) 695 187 (69.8) (18.8) 375 (2.5) 507 (3.8) 96 (9.8) 78 (7.8) 75 (0.6) 120 (0.7) 21 (2.1) 36 (3.6) 14,666 13,413 982 996 The proximal surfaces most frequently inment for mandibular teeth were higher than for the maxillary units, especially in the in- volved were found in the molar regions for all disease categories (Table 7). The maxilcisor region (Table 5). lary molars were notably more involved than PROXIMAL ALVEOLAR BONE LOSS any other maxillary teeth. The pattern of Premodern groups bone loss was more regular in the mandible Less than 10% of total teeth showed evi- than in the maxilla and slightly more severe. dence of proximal bone loss due to periodonModern groups tal disease (Table 6). The incidence of proximal periodontal disease was higher in The frequency of proximal alveolar bone the mandible than the maxilla for all cate- loss for periodontal reasons was approxigories of disease but the differences were mi- mately three times greater in the modern nor. Mild disease accounted for 5.8% of the groups than in the premodern groups; howtotal of marginally damaged sites, moder- ever, the overall incidence was still relaately involved locations comprised 3.1% of tively low, with 68.9% of teeth free from the the total, while severely involved sites were evidence of periodontal disease (90.3% in the less than 1%of the total (Table 6). premodern groups). The severity of proximal DISEASE IN ANCIENT POPULATIONS 181 TABLE 7. Alveolar bone loss due to periodontal disease in 32 premodern groups’ proximal surfaces of individual teeth Tooth 18 17 16 15 14 13 12 11 Total No. of teeth 758 878 860 929 972 1,002 973 950 7,322 Upper right quadrant Severe (%) 0.8 Moderate (%) 5.1 Mild (%) 7.3 Normal (%) 86.8 1.3 6.3 8.7 83.8 0.8 4.7 7.3 87.2 0.0 2.2 4.6 93.2 0.1 2.3 4.2 93.5 0.1 1.2 3.2 95.4 0.1 1.0 3.0 96.0 0.0 0.6 2.6 96.7 0.4 2.6 5.0 91.9 47 46 45 43 42 41 Total Tooth 48 44 No. of teeth 725 773 722 867 9 14 929 908 873 6,711 Lower right quadrant Normal (%j 87.6 Mild (%) 7.3 Moderate (%) 4.4 Severe (%j 0.7 84.7 9.2 4.9 1.2 85.7 8.7 4.2 1.4 91.8 5.2 2.5 0.5 92.8 3.7 2.8 0.6 91.4 5.2 2.5 1.0 88.0 7.0 4.0 0.9 88.7 6.6 3.8 0.9 88.9 6.5 3.6 0.9 Mild, 0-2mm; moderate, 2-4mm; severe, >4mm. TABLE 8. Alveolar bone loss due to periodontal disease in two modern groups’ (American and South African) proximal surfaces of individual teeth Tooth 18 17 16 15 53 65 61 63 1.5 13.8 12.3 72.3 6.5 13.1 13.1 67.2 1.6 3.2 17.4 77.8 48 47 46 45 50 55 57 67 72.7 14.5 12.7 0.0 68.4 15.8 14.0 1.8 82.1 13.4 3.0 1.5 13 12 11 Total 64 69 62 61 498 1.6 7.8 15.6 75.0 0.0 4.4 11.6 84.1 0.0 4.8 11.3 83.9 0.0 3.8 9.8 86.9 2 .o 8.2 21.9 67.9 44 43 42 41 Total 69 70 68 60 496 82.6 11.6 4.3 2.9 75.7 15.7 5.7 5.9 55.8 20.6 17.6 5.9 55.0 21.7 16.6 6.7 71.0 17.5 8.1 3.4 14 No. of teeth Upper right quadrant Severe (%) 0.0 15.1 Moderate (%) 18.9 Mild (%) 66.0 Normal (%I Tooth No. of teeth Lower right quadrant 84.0 Normal (%I 8.0 Mild (%) 6.0 Moderate (%I 2.0 Severe (%j Mild, 0-2mm; moderate, 2-4mm; severe, 24mm. alveolar bone loss was also notably greater than in the premodern groups, with 19.4% mildly involved teeth in the modern groups compared with 5.8% in the premodern groups; 8.8%moderately affected teeth in the modern groups compared with 3.1% in the premodern groups; and 2.9% severe bone loss in the modern groups compared to 0.7% of the premodern groups (Table 6). Maxillary molars and mandibular incisors were more frequently affected than other units (Table 8). DISCUSSION The findings of the present study suggest that several important dental concepts require careful review. One important finding was the very low incidence of periodontal 182 N.G. CLARKE ET AL. disease which could be attributed to a spread of gingival inflammation. The jaws examined in the premodern groups had not been influenced by sophisticated oral hygiene practices and yet, despite the presence of massive calculus deposits in many instances, there was no discernible loss of bone from a periodontal origin in over 90% of the teeth. The data for periodontal disease was similar for both buccal and proximal surfaces in most groups, although some variation was present. While the concept of a bacterial etiology for the disease would suggest that the proximal surfaces should be more heavily involved, the present data do not show the expected bias toward the disease at proximal sites. Another important finding was the marked difference between the premodern and modern groups in the incidence of periodontal disease. In the premodern groups, approximately 10% of subjects were found to have experienced loss of crestal bone as a n extension of a gingival lesion, while approximately 30% of the modern groups demonstrated some evidence of crestal resorption. The present findings also suggest that where progression of a superficial lesion was evident in the bone crest, the attack was not aggressive and usually resulted in only minor degrees of lost tissue. None of the teeth examined had experienced a degree of bone loss that could have resulted in eventual tooth loss. The resistance to the extension of gingivitis in the population groups examined may have been due to the level of efficiency of the host defence systems that operate in the gingival crevice and the gingivae (Clarke and Carey, 1985). In modern society these defence mechanisms may be compromised by prolonged or combined environmental factors such as stress, smoking, and diet. In prehistory, the seasonal availability of food may have resulted in periodic nutritional inadequacies (Gilbert and Mielke, 1985), but the periodontal structures appear to be unaltered. Ruffer (1920) found, in a study of ancient Egyptian skulls, that severe suppurative periodontitis was often associated with marked attrition of the teeth. The question was posed, is periodontal disease caused or made worse by very heavy tooth wear? It is unfortunate that this early observation has not been investigated further, since heavy tooth attrition has the capacity to injure the dental pulp. Subsequent diffusion of inflammatory mediators from the dental pulp to the periodontal ligament via the dentinal tubules and cementum may result in inflammation of the periodontal ligament (Torobinejad and Bakland, 1980). Larato (1970) also found that abscesses were commonly seen in relation to teeth with severe caries or tooth wear in dry skulls. There is no correlation between attrition and gingivitis, but a strong correlation does exist between attrition and pathology of the pulp, culminating in the most severe condition with exposure of the organ. There is also a strong correlation between pulp pathology and the formation of ligamental granulation tissue (Simon, 1983). Where the bone is thin on buccal or lingual surfaces, total destruction of the process results in a suprabony lesion. Such lesions occurring in the periodontium as a result of pulpal pathology are generally deep, destructive of alveolar bone, and can seriously compromise tooth support (Simring and Goldberg, 1964). In teeth with sound crowns, pulpal disease may result from disturbed vascular supply to the most peripheral aspect of the pulpal microcirculation. Causes of such a disturbance are conjectural but could correlate with age changes or any one of the environmental factors involved in peripheral vascular disease. Periodontal lesions of pulpal origin (furcation, hemiseptal, vertical, and deep, isolated bony defects) eventually discharge into the gingival crevice and may be probed, thereby mimicking a periodontal pocket. In the present study such dental abscesses were frequently observed to cause alveolar bone loss. Nevertheless, periodontal lesions of pulpal origin other than those at the apex of the root are not generally recognised in either the dental literature (Schafer et al., 1983)or the anthropological literature (Molnar and Molnar, 1985). The low incidence of periodontal disease found in the present study is consistent with the findings of a number of recent studies. One epidemiological survey of a n adult Dutch population found that periodontal disease did not generally occur in the older age group and only a few cases of severe periodontal destruction were found, these being in many cases limited to single teeth (Pilot and Schaub, 1985). A comprehensive longitudinal survey of periodontally involved patients found that most periodontal lesions are stable (Socransky et al., 1984). Similarly, a longitudinal study of patients with gingivitis DISEASE IN ANCIENT POPULATIONS (Listgarten et al., 1985) found that little, if any, periodontal breakdown occurred in the study population over a 3-year period. Another study found that the most apical point of bony craters in 148 skulls was located adjacent to the tooth rather than in the interdental supporting bone, and clinical observations of angular bony defects were found to be consistent with the skeletal materials (Saari et al., 1968). These findings are consistent with the concept of a tooth-derived lesion arising in the periodontal ligament, rather than the commonly postulated site-specific bacteria attacking a localised sector of the periodontium. It would appear that severe isolated periodontal lesions have a pulpal etiology and should not be confused with more moderate and generalised periodontal lesions resulting from host interaction with microbes. The dental literature asserts that periodontal disease is responsible for the majority of tooth loss after the age of thirty. Costa (1982) disagrees with this concept and stated that periodontal disease was not a major cause of tooth loss in Eskimoes at any age. The findings of the present study strongly support those of Costa (1982) since no skulls were examined that demonstrated a degree of bone loss of marginal (periodontal) origin that could be considered capable of causing tooth loss. Periodontal disease appears to be minimally destructive, predominantly located in molar and mandibular incisor regions, and present in only a very small proportion of the populations examined. The findings of the present study, coupled with other paleopathological and clinical studies, contradict the traditional view of the incidence and severity of periodontal disease, and they suggest that the traditional concept of the etiology of severe periodontal lesions requires a reorientation toward a pulpal origin. 183 110:689-691. Costa, RL (1982) Periodontal disease in the prehistoric Ipiutak and Tigara remains from Point Hope, Alaska. Am. J. Phys. Anthropol. 59:97-110. Gilbert, RI, and Mielke, JH (1985)The analysis of prehistoric diets. Orlando: Academic Press, Inc., p. 164. Gold, SI (1985) Periodontics. The past. J. Clin. Periodont. 12:79-97. Larato, DC (1970) Intrabony defects in the dry human skull. J. Periodont. 41:496-498. Listgarten, MA, Schifter, CC, and Laster, L (1985)3 year longitudinal study of the periodontal status of an adult population with gingivitis. J. Clin. Periodont. 12:225238. Loe, H, Anerud, A, Boysen, H, and Smith, M (1978) The natural history of periodontal disease in man. J. Periodont. 49:607-620. Molnar, S, and Molnar, I(1985) Observations of dental diseases among prehistoric populations of Hungary. Am. J. Phys. Anthropol. 67.51-63. Muruhv, T (1959) Comuensatorv mechanisms in facial heigdt adjustment to'functionh tooth attrition. Aust. Dent. J. 41312-323. Newman, HN, and Levers, BGH (1979) Tooth eruption and function in early Anglo-Saxon population J. Roy. SOC.Med. 72341-350. Page, RC, and Schroeder, H (1976) Pathogenesis of inflammatory periodontal disease. A summary of current work. Lab. Invest. 33:235-249. Page, RC, and Schroeder, H (1982) Periodontal Disease in Man and Other Animals. Basel: Karger, pp. 5-17. Pilot, T, and Schaub, RMH (1985) Reappraisal of periodontal treatment needs. IADWAADR Abstracts No. 770. J. Dent. Res. 64:260. Ruffer, A (1920)Study of abnormalities and pathology of ancient Egyptian teeth. Am. J. Phys. Anthropol. 3t335382. Saari, JT,Hurt, WC, and Biggs, NL (1968) Periodontal defects on the dry skull. J. Periodont. 39:278-283. Sagne, S, and Olsson, G (1977)Studies of the periodontal status of a medieval population. Dentomaxillofac. Radiol. 6:46-52. Simon, JH (1983) Pathology. In B Cohen and RC Burns (ed): The Pathways of the Pulp. St. Louis: Mosby, p. 390. Simring, M, and Goldberg, M (1964) The pulpal pocket approach: retrograde periodontitis. J. Periodont. 352248. Schafer, WG, Hine, MK, and Levy, BM (1983)A textbook of oral pathology. Philadelphia: W.B. Saunders Co., p. 497. Socransky, SS, Haffajee, AD, Goodson, JM, and Lindhe, J (1984) New concepts of destructive periodontal disease. J. Clin. Periodont. 11:21-32. Tal, H (1985) Periodontal bone loss in dry mandibles of ACKNOWLEDGMENTS South African blacks: a biometric study. J. Dent. Res. The hospitality and the access t o the skel- 64:925-929. etal materials offered by the curatorial au- Torobinejad, M, and Bakland, LK (1980)Prostaglandins: their possible role in the pathogenesis of pulpal and thorities a t all locations is gratefully re- periapical diseases. J. Endo. 6:733-739. corded. Whittaker, DK, Molleson, T, Daniel, AT, Williams, JT, Rose, P, Resteghini, R (1985) Quantitative assessment LITERATURE CITED of tooth wear, alveolarcrest height and continuing eruption in a Romano-British population. Arch. Oral Clarke, NG, and Carey, SE (1985) The aetiology of ginBiol. 30:493-501. givitis: a n alternative explanation. J. Am. Dent. Assn.