Treponematosis is “a chronic or subacute infection caused by microorganisms called spirochetes of the genus Treponema. On the basis of clinical and geographic variation, the infection is divided into four types: pinta, yaws, bejel (or endemic syphilis) and venereal syphilis” (Aufderheide and Rodríguez-Martín, 1998: 154). There is debate about whether these are different diseases caused by different bacterial species within the genus, or whether they are different clinical manifestations of infection by one species, Treponema pallidum (Aufderheide and Rodríguez-Martín, 1998: 154):
- Pinta is the most geographically restricted of the four syndromes. It is limited to the tropical regions of America, from Mexico to Ecuador. Treponema carateum is responsible for the disease;
- Yaws or frambesia affects especially those populations with a low level of hygiene in tropical or subtropical humid areas. It is caused by Treponema pertenue;
- Bejel (endemic syphilis or non-venereal syphilis) is present in rural populations in temperate and subtropical non-humid regions, and caused by Treponema pallidum endemicum;
- Venereal syphilis is the most ubiquitous of the four syndromes, occurring primarily in urbanised populations in all geographic regions. It is caused by Treponema pallidum pallidum.
Except for pinta, which does not involve bone, the other syndromes produce nearly identical lesions upon the skeleton (Steinbock, 1976: 94). These lesions represent a late stage that marks the generalised spread of bacteria within the body. Inflammatory changes of a mainly destructive nature occur within the skeleton, although considerable repair and regeneration can often also be demonstrated. Venereal syphilis may be either congenital or acquired, congenital syphilis being transmitted in utero via the infected mother. The most severe of the forms, it is characterised not only by inflammatory changes in most tissues of the body, but also by the affection of arterial circulation and the nervous system (Roberts and Manchester, 1995: 151).
The gross bone destruction, called a gumma, is not dissimilar to non-specific osteomyelitis. However, treponemal osteomyelitis is accompanied by extensive regeneration. Consequently, the bone becomes much altered in appearance. The frequency of bone involvement ranges from about 3-5 percent of all cases of yaws to 10-12 percent of all cases of venereal syphilis, with endemic syphilis lying somewhere between the two. In yaws the most commonly affected bone is the tibia, resulting in the distinctive ‘sabre shin’, with the other bones only being affected to a lesser extent, the skull in particular being infrequently involved. Skull involvement is also uncommon in endemic syphilis, but may result in extensive destruction of the nasal and jaw regions. As with yaws, the ‘sabre shin’ shape is often seen, the tibia again being the most commonly affected element (Roberts and Manchester, 1995: 152).
As with the other previously described treponemal diseases, the tibia is the most commonly affected element of venereal syphilis, but multiple bone involvement is frequently noted. It may also result in destructive changes of the joints, such as the characteristic ‘Charcot’s joint’. In contrast to yaws and endemic syphilis, the skull is frequently affected, exhibiting a ‘worm-eaten’ appearance called caries sicca (Roberts and Manchester, 1995: 153).The diagnostic criteria most frequently used to establish the presence of treponemal disease in dry bones are by Hackett (1976).
The epidemiology and spread of treponemal disease has been the subject of much debate, with controversy raging over whether a New or Old World origin for the condition was more likely, the transatlantic voyage of Christopher Columbus being a date of great significance to epidemiologists. It is still unclear when, where and how the currently recognised clinical forms developed, and how they are related to one another from an evolutionary standpoint. Identifying the form of treponemal disease from the physical examination of skeletal remains is often impossible, although attempts have been made to make such a differentiation at population level and it is possible that ancient DNA analysis may aid in this (Mitchell, 2003: 122). For example, the causative agent of venereal syphilis, Treponema pallidum pallidum, has recently been extracted from a two hundred-year-old skeleton from Easter Island using a combination of immunological assay and DNA analysis (Kolman et al., 1999). The oldest dated case of treponematosis in the Middle East comes from Israel, and dates to 1290-1420 AD, during the Mamluk period just after the Crusades (Mitchell, 2003). Pre-Columbian dates have also been given for cases from Britain (Mays et al., 2003) and the United States of America (Hutchinson and Weaver, 1998).
Buckley and Tayles (2003) examined a prehistoric Pacific Island sample for diseases such as yaws, leprosy and malaria. Each type of bone change was recorded, with each type of pathological change also being given a numerical code. This code indicated the type of observed change and whether it was primarily osteoblastic activity, osteoclastic activity, or a mixed response. The presence of lesions was identified macroscopically. Diffuse pitting and/or apposition of new bone on the cortical aspect is an indication of osteoblastic activity as a response to an infectious or non-infectious mechanism. Active (unremodelled) or remodelled lesions were coded as such, the authors stating that “active osteoblastic lesions display a fibrous, vascular, porous, and irregular layer of new bone that has a scab-like appearance over the normal smooth cortex”, in contrast to remodelled lesions that “is usually smooth in appearance, and more organised than new woven bone.” Lesions that developed as a result of osteoclastic activity were identified by lytic foci in the cortical bone and/or trabeculae. Distortion of the normal shape, such as diaphyseal bowing, was also noted as pathological (Buckley and Tayles, 2003: 306). These lesions were then placed into either Osteoblastic (OB) or Osteoclastic (OL) grades 1-4 (Buckley and Tayles, 2003: 306-307), which are described in the article, with photographic illustrations of most, but not all, classes. Hackett’s (1976) coding criteria was used for recording classic gummatous caries sicca lesions of the cranial vault (Buckley and Tayles, 2003: 307).
In an attempt to control some of the limitations of estimating lesion prevalence within a population, the prevalence was estimated using both the individual and skeletal element as denominators. An ‘individual’ was defined by the presence of certain bone elements that are most frequently affected by systemic disease. Those selected, with at least one of each being needed, were the femur, tibia, clavicle, ulna and either the hands or feet for the minimum definition of an ‘observable individual’ (Buckley and Tayles, 2003: 308). This selection criteria was not as strict for the sub-adults due to the fact that sub-adult bones are more readily affected by pathological change. This was useful for increasing the sample size given the fragmentary nature of most of the sub-adult skeletons. Therefore, a sub-adult burial that included the tibia and any two other limb bones were included as individuals (Buckley and Tayles, 2003: 310).
Cranial material was recorded for all observable individuals as complete (presence of all cranial elements, whether articulated or fragmentary), partial (presence of some cranial elements, may be missing the frontal), or absent (no cranial material present). A burial was retained as an individual regardless of cranial preservation, however the presence of the frontal bone and a partial or complete naso-maxillary region was deemed ideal (Buckley and Tayles, 2003: 310).
All bones present were included in the analysis using the skeletal element as the indicator of prevalence, whether complete or partial, and from observable individuals or not. Left and right elements were combined to increase the sample sizes and the bones were sub-divided according to age group. It was considered that sex-related distributions of lesions could be best analysed using observable individuals and so were not studied using skeletal element as the denominator (Buckley and Tayles, 2003: 311). One limitation of the current study was the lack of systematic recording of joint pathology, which may have distorted the prevalence of lytic lesions affecting the joints (Buckley and Tayles, 2003: 310).
The status and severity of lesions in observable individuals was assessed; status 1 indicating several bones were affected and all lesions were active at death, status 2 indicating the presence of active lesions, but with some remodelling at the time of death, and status 3 indicated solely remodelled lesions (Buckley and Tayles, 2003: 310-311). In contrast, the lesions assessed by skeletal element were first classified into three main groups: active OB, remodelled OB, or OL. Cranial lesions were considered separately (Buckley and Tayles, 2003: 311).
It was found that more than half of the observable individuals (57 of 101) had post-cranial lesions, with similar prevalence for all age groups and both sexes. Less than one quarter had cranial lesions and the prevalence decreased with age. However, lesions were present in a low proportion of post-cranial elements overall, although the major long bones, especially the tibia, were most frequently affected. The lesions were effectively all active in sub-adults, whilst in adults there was a mixture of active and remodelled lesions (Buckley and Tayles, 2003: 313-316), a diagnosis of yaws being proposed as the most likely to fit the pattern displayed at the site (Buckley and Tayles, 2003: 321).
This methodology seems particularly useful as it does not require any specialist, or expensive equipment, and would be readily adoptable by most osteoarchaeologists. The categories seem a little crude at times, however this is not uncommon with those methodologies based upon macroscopic examination, and the classes do at least provide a baseline for comparison between populations. Better illustration of the various classes, by photographs of all grades for example, would have been useful in order to help decrease inter-observer variation.
It is worth noting, however, that bone lesions caused by treponemal disease are frequently very similar to non-specific bone diseases, particularly in the long bones. The use of polarised light in light microscopy can illuminate elements of the internal bone structure such as the arrangement of collagenous fibres, and this can make visible structures typically built up by different diseases. At the micro-level, telltale signs of treponemal disease are polsters and grenzstreifen (Schultz, 2001: 126).
Polsters are frequently found in the chronic treponemal diseases of the long bones, and consist of “parallel lamellae arranged at the periosteal level in the form of pillow-like newly built bone formations demarcated by periosteal blood vessels developed during the course of the inflammatory process” (Schultz, 2001: 126). These show a homogeneous structure due to the generally slow growth, in contrast to haematogenous osteomyelitis, which has a much more rapid growth pattern, and thus a more irregular structure (Schultz, 2001: 126). ‘Grenzstreifen’ or ‘grenzlinie’ can be observed in chronic treponemal diseases. This is a “very fine line or a narrow, band-like structure that represents the original external surface of the bone shaft (remains of external circumferential lamellae) and newly built lamellae that originate during the first infection of the periosteum due to the pathological process” (Schultz, 2001: 126).
In contrast to leprosy, treponemal diseases such as endemic syphilis include not only alterations in the subperiosteal bone, but also osteoclastic changes in the endosteal bone and the bony trabeculae of the medullary cavity, as well as in the compact bone of the affected elements (Schultz, 2001: 128). Palaeohistopathological analysis can, therefore, be an important tool for diagnosing conditions such as this in archaeological specimens, helping to discount non-specific conditions that are similar at a macromorphological level.
Aufderheide, A. C and Rodríguez-Martín, C. 1998. The Cambridge Encyclopedia of Human Paleopathology. Cambridge: Cambridge University Press.
Buckley, H. R, and Tayles, N. 2003. Skeletal pathology in a prehistoric pacific island sample: issues in lesion recording, quantification, and interpretation. American Journal of Physical Anthropology 122: 303-324.
Hackett, C. 1976. Diagnostic criteria of syphilis, yaws and treponarid (treponematosis) and of some other diseases in dry bone (for use in osteo-archaeology). Berlin: Springer-Verlag.
Hutchinson, D. L, and Weaver, D. S. 1998. Two cases of facial involvement in probable treponemal infection from Late Prehistoric coastal North Carolina. International Journal of Osteoarchaeology 8: 444-453.
Kolman, C. J, Centurion-Lara, A, Lukehart, S. A, Owsley, D. W, and Tuross, N. 1999. Identification of Treponema pallidum subspecies pallidum in a 200-Year-Old skeletal specimen. Journal of Infectious Diseases 180: 2060-2063.
Mays, S, Crane-Kramer, G, and Bayliss, A. 2003. Two probable cases of treponemal disease of medieval date from England. American Journal of Physical Anthropology 120: 133-143.
Mitchell, P. D. 2003. Pre-Columbian treponemal disease from 14th century AD Safed, Israel, and implications for the medieval Eastern Mediterranean. American Journal of Physical Anthropology 121: 117-124.
Roberts, C., and Manchester, K. 1995. The Archaeology of Disease. 2nd Edition. Stroud: Alan Sutton Publishing Ltd.
Schultz, M. 2001. Paleohistopathology of bone: a new approach to the study of ancient diseases. Yearbook of Physical Anthropology 44: 106-147.
Steinbock, R. T. 1976. Paleopathological Diagnosis and Interpretation: Bone Diseases in Ancient Human Populations. Springfield: Thomas Books.