This page will discuss the work and findings that I conducted as a Master’s student at Bristol University. This work investigated a feature of human bone, called a Harris Line (HL), using a new perspective: a micro-CT scanner. Before continuing to read this page I strongly suggest reading my page entitled My Master’s Work – The Background Literature which provides an in-depth literature review surrounding the HL.
HLs are lines of dense, horizontally-orientated trabecular bone that appear on radiographs (Mays 1985, 1995). These features are commonly attributed to trauma and illness and commonly found in long bones (McHenry & Schulz 1976). Although there has been much study of the HL there have been various definitions, classifications and explanations for their formation and identification. The traditional approach for identifying HLs has been to use 2D radiographs (Fig. 1), along with anatomical sections. These methods provide limited information on HL structure and their degree of variability and with the use of micro-CT a 3D perspective may be gained as well as the opportunity to study the micro-architecture of the line.
Throughout the literature there are few clear and concise definitions for a HL, although many refer to them as transverse lines of increased density (e.g. Caffey, 1967; Garn and Schwager, 1967; Gindhart et al., 1969; Mays, 1985; McHenry, 1968). However, when rating and scoring the presence of a HL there does not appear to be one agreed method. Some studies only mark the presence or absence of a line (e.g. Blanco et al., 1974; Hughes et al., 1996) while others produce a criterion (e.g. Piontek et al., 2001). Out of 16 papers reviewed three only counted HLs which were a quarter or more across the bone shaft, seven at least halfway and six all HLs. This indicates that the interpretation of a HL is not well understood and varies between each research project.
This lack of consistency between papers could have an impact in other areas where the HL is used. The feature is associated with illness and malnutrition and therefore the presence of HLs has been used as an indicator of past population health (e.g. Allison et al. 1974, Mays 1995, Piontek et al. 2001, Ribot & Roberts 1996). As there is no agreed method for HL identification the frequency of HLs recorded in any population may differ depending on the chosen criterion. It is therefore necessary to attempt to find which, if any, method for HL identification is the most reliable. To do this a detailed view of the structure is needed. A technique which may allow this is micro-CT.
Traditionally, HL identification relies heavily on radiographs or anatomical sections. With the use of micro-CT more information on structure and appearance may be gained. The main focus of this research will attempt to clarify the definitions and methods for classification of HLs. By observing different ‘classifications’ of HL it may be understood whether a significant difference exists and is therefore valid. This could impact the number of HLs identified within a population, and therefore affect the view on the population’s health.
The aim of this study was to get a better understanding of the HL structure and therefore I wanted to compare an area of bone containing a HL to and area of non-affected bone. This would enable me to draw conclusions and insights around the subject of identification and classification of a HL.
A total of 30 of skeletons were chosen for study from the collection excavated at the priory of St. Peter and St. Paul, Taunton, Somerset (AD 1150–1539). These skeletons consisted of 17 sub-adults and 22 adults and were chosen based on their age and the condition of the tibia (Field 2013). The tibia was chosen for this study as HLs have been successfully identified in this bone in previous studies (e.g. Garn and Schwager, 1967).
These tibias were radiographed and a visual assessment for the presence of Harris Lines (HLs) was then carried out and measurements of the lines were taken. After this a selection of HLs were selected to be observed by micro-CT. The HLs were sorted into categories depending on their transverse length, as measured on the radiograph so a comparison of identification methods could be carried out.
Results and Discussion
With the use of micro-CT every HL was identified in the chosen specimens and consisted of dense trabecular bone (Fig. 2). This area was spread across a proportion of the bone width and transcended the bone shaft but varied between specimens. By comparing Harris Lines (HLs) of different lengths as identified from radiographs the methods used for HL identification and classification could be evaluated.
There was no difference was found between the different categories of HL for any the variables taken. This suggests that the use of HL length as a classification system is not valid. This was true for every HL identified and reinforces that there is little difference in the micro-architecture for the different categories of HL. These results indicate that many of the definitions used in past research are not reliable. The method of excluding some HLs based on their transverse length has the potential to provide false and skewed results, particularly in relation to population health. From the micro-architecture alone it is reasonable to identify any dense, transverse lines on a radiograph as a HL, therefore increasing the frequency of lines recorded. If HLs are an indicator of health then this is likely to show that many past populations appear in poorer health than previously thought. A comparison between a HL observed on a radiograph and the same HL observed with micro-CT can be seen in Figure 3.
With the use of micro-CT a large degree of variation in depth, width and spread across the tibia shaft could be seen. This is not possible on a radiograph which could potentially create issues with identification. When viewed on a radiograph the HL can only be measured in one direction, typically medially-laterally. However, this does not account for any HLs which extend in an anterior-posterior direction. Therefore a HL on a radiograph may appear short in one direction, but longer when viewed from a different side. This further evidence to suggest that all transverse lines identified on a radiograph should be counted as a HL.
From this research it may be necessary to make adjustments to the definitions used to identify HLs to include dense, transverse line, regardless of length, as a HL. Any adjustment to the definition and classification of HLs could impact the results found in previous studies. Due the results from this study if HL were excluded due to their length there may have been an under-estimated the number of HL present in a population. This then affects the view on population health, suggesting that more incidences of ill health occurred than previously thought.
In addition to the results mentioned here I also compared Harris Lines found in sub-adults to those observed in adults, as well as investigation observer error when identifying HLs from a radiograph. If this is of interest to anyone please feel free to get in touch and I will happily discuss it with you. You will also be able to view the abstract to my research thesis at academia.edu.
Allison, M.J., Mendozo, D., Pezzia, A., 1974. A radiographic approach to childhood illness in precolumbian inhabitants of southern Peru. American Journal of Physical Anthropology 40, 409-415.
Blanco, R.A., Acheson, R.M, Canosa, C., 1974. Height, weight, and lines of arrested growth in young Guatemalan children. American Journal of Physical Anthropology 40, 39-47.
Caffey, J., 1967. Transverse lines of Park (Stress lines of Park). Pediatric x-ray diagnosis Year Book Medical Publishers, Chicago, pp.789-795.
Field, S. The Harris Line in 3D; Implications for their identification and future study. Master’s Thesis. University of Bristol 2013.
Garn, S., Schwager, P., 1967. Age dynamics of persistent transverse lines in the tibia. American Journal of Physical Anthropology 27, 375-377.
Gindhart, P.S., 1969. The frequency of appearance of transverse lines in the tibia in relation to childhood illnesses. American Journal of Physical Anthropology 31, 17-22.
Hughes, C., Heylings, D.J., Power, C., 1996. Transverse (Harris) lines in Irish archaeological remains. American Journal of Physical Anthropology 101, 105-131.
Mays, S., 1985. The relationship between harris line formation and bone growth and development. Journal of Archaeological Science 12, 207-220.
Mays, S., 1995. The Relationship between Harris Lines and other Aspects of Skeletal Development in Adults and Juveniles. Journal of Archaeological Science 22, 511-520.
McHenry, H., 1968. Transverse lines in long bones of prehistoric California Indians American Journal of Physical Anthropology 29, 1-17.
McHenry, H., Schulz, P., 1976. The association between Harris lines and enamel hypoplasia in prehistoric California Indians. American Journal of Physical Anthropology 44, 507-511.
Piontek, J., Jerszynska, B., Nowak, O., 2001. Harris Lines in subadult and adult skeletons from the medieval cemetery in Cedynia, Poland. Variability and Evolution 9, 33-43.
Ribot, I. Roberts, C., 1996. A study of non-specific stress indicators and skeletal growth in two mediaeval subadult populations. Journal of archaeological science 23, 67-80.