Commentary on: Assessing diagnostic certainty for scurvy and rickets in human skeletal remains – An update on Brickley and Morgan (2023)

Within anthropology, bioarchaeologists working in paleopathology are increasingly tackling questions of global significance regarding the past, present, and future of human communities. Greater levels of interdisciplinarity and moves towards transdisciplinary research now typically characterize bioarchaeology and paleopathology (Buikstra et al., 2022). Such developments have helped broaden and clarify long-term perspectives on rickets and vitamin D status, but how these advances take shape requires significant anthropological rigor and care. This commentary proposes a temporary pause on using microscopic mineralization defects in bones and teeth to suggest the highest level of diagnostic certainty in rickets as presented by Brickley and Morgan (2023, p.641). Since the 1990s, there has been a growing interest in metabolic diseases such as rickets and scurvy as important markers of metabolic stress and insufficiency. It has been recognized that such conditions contribute to poor skeletal preservation. There is a need to learn more about life histories and evaluate all individuals, not just those with well-preserved bone and clear-cut cases of disease. However, diagnosing these conditions can be challenging. Following on from the formative work on scurvy led by the late Don Ortner (Ortner & Ericksen, 1997) and work on rickets initiated by Ortner and Mays (1998), there has recently been intense interest in rickets and vitamin D deficiency (Brickley et al., 2020a, pp.121–128). Rickets is a systematic disease of those still undergoing growth and development. In the strictest sense, rickets refers to interference in endochondral bone formation (Glorieux et al., 1998, p.764) with resultant disordered mineralization of bone beneath growth plates (Fischer et al., 2023; Glorieux et al., 1998; Ortner & Mays, 1998). All previously cited authors discuss rickets alongside osteomalacia, reduced and defective mineralization of osteoid at previously formed surfaces, as the two conditions will co-occur. The terminology is, however, messy and inconsistently used in clinical, biomedical, and paleopathology work. Physicians started naming signs of these conditions in the 1600s, long before there was any knowledge regarding vitamin D and the other factors involved in rickets (see discussion in Brickley et al., 2020a, pp.75–76). In paleopathology, clear diagnostic frameworks and consideration of individuals still undergoing growth and development are relatively recent developments. In bioarchaeology and paleopathology, many of the deep-time perspectives on health, and more critically, disease at both global and community levels, rely on knowing the level of certainty in diagnoses. Such an approach was recently proposed for rickets and scurvy (Brickley & Morgan, 2023). Using recommendations on terms used to indicate diagnostic certainty presented by Appleby et al. (2015), a framework for moving from analysis of lesions to suggesting the individual level of diagnostic certainty was provided for rickets, scurvy and co-occurrence cases. Widening the spectrum of cases that could be described, Brickley and Morgan (2023) provided suggestions for the use of terms: “diagnostic of” (the pattern of lesions identified and discussed could not have been caused in any way other than that described), “highly consistent/typical” (the lesions/s could have been caused by the condition described, but there are a few other possible causes that need consideration) and “consistent with” (the lesion/s could have been caused by the condition described, but are nonspecific and there are many other possible causes). When studied carefully, changes to the bone beneath the growth plates and associated structures still offer the possibility of using the “diagnostic of” level of certainty. However, the recommendation that “unambiguous” evidence of systemic microscopic mineralization defects in bones and teeth can contribute to suggesting cases considered “diagnostic of” rickets, as presented by Brickley and Morgan (2023, p.641), needs to be temporarily paused. Pausing would ensure that suggestions stay within the available evidence and, importantly, take into account precisely what can be said about each case and wider interpretations. In tests of the framework proposed by Brickley and Morgan (2023), the term “unambiguous” worked well at all levels of user experience to ensure careful consideration of the underlying pathological process in macroscopic lesions. In particular, “unambiguous” prevented overuse of the “diagnostic of” category for scurvy. “Unambiguous” was therefore also used for microscopic mineralization defects in bones and teeth. However, systemic mineralization defects in teeth (interglobular dentine [IGD]) may be unambiguous, but lower grade (D'Ortenzio et al., 2016) or percentage area (Snoddy et al., 2021); no systems have been developed for scoring microscopic lesions in dry bone, which also span a spectrum of expressions. It should be noted that IGD and microscopic mineralization defects in bone are not specific to vitamin D deficiency. When minimally developed these features may resemble those occurring during normal growth and development. More marked cases can also be caused by conditions that disrupt the vitamin D, calcium, phosphate pathway, or increase the lag time in tissue mineralization (see Section 1.3). The direct association between characteristic changes under growth plates in rickets and microscopic areas of defective and inadequate mineralization in bones, and IGD in teeth is still unclear, and exactly how these lesions relate to serum (25(OH)D) is unknown. Total serum 25-hydroxyvitamin D, (25(OH)D), is the most widely used biomarker for evaluating vitamin D status (Fischer et al., 2023). When only relying upon microscopic mineralization defects as an indicator of rickets, I propose a pause on using the “diagnostic of” level of diagnostic certainty. Although using extremely severe cases of microscopic mineralization defects to ascribe “diagnostic of” may prove technically correct, additional data are needed for diagnostic certainty. Changes to the bone beneath the growth plates and associated structures still offer the possibility of using the “diagnostic of” level of certainty when used carefully. It is however advisable that mineralization defects in both bones and teeth are still recorded whenever possible. Far more will be learned about changes in health through the life course and the heterogeneity in the experience of conditions causing rickets if researchers can recognize and interpret these features. The period of life or age at which an individual was affected by a condition and whether it was an isolated event or recurrent can be determined. Such knowledge also facilitates a greater understanding of the mother-infant dyad (Brickley, Kahlon, & D'Ortenzio, 2020b). During the pause on use of “diagnostic of” level of diagnostic certainty, the terms “highly consistent/typical” and “consistent with” can still be used. Microscopic mineralization defects that are clearly part of systemic events and have multiple clear features across a thin section of bone, or higher grades or percentage areas of IGD will often be highly consistent/typical of rickets in those still undergoing growth and development. IGD visible on most micro-CT scans at the date of writing is also highly consistent with rickets (active or healed). Where microscopic mineralization defects of bones and teeth are not clearly part of a systemic condition, the term “consistent with” is more appropriate and could replace the term “developmental” used for IGD (see Section 1.3). Work is underway to obtain more evidence on how microscopic mineralization defects of bones and teeth relate to signs used in clinical work. A partnership of researchers, including paleopathologists, imaging specialists, and those with expertise in artificial intelligence machine learning, is being established to link microscopic mineralization defects throughout the skeleton and dentition to changes in the bone beneath the growth plates. Others have demonstrated the utility of such approaches in preliminary studies, establishing that object detection software can identify rickets from wrist radiographs, the traditional clinical tool used to diagnose rickets (Meda et al., 2021). Initial work undertaken at the author's institution indicates that in dry bone, multi-slice analysis of micro-CT scans is needed to provide robust information on mineralization defects; recording such features consistently was found to be exceptionally time-consuming. However, preliminary collaborative investigations indicate that machine learning and artificial intelligence can be applied to micro-CT scans of bones and teeth and would assist greatly in consistently evaluating mineralization defects. Factoring in scoring that equates diagnostic certainty to the quality of the bone beneath growth plates should be achievable. Rickets was relatively common in many recent (e.g., Lapatsanis et al., 1968) and distant past contexts. Brickley and colleagues reviewed cases of IGD and paleopathological studies of rickets. They demonstrated that cases likely to meet the “diagnostic of” threshold were present in contexts far beyond the boundaries of northern industrial centers (Brickley et al., 2017, p.421). Public health campaigns in many countries have decreased the number of modern individuals with vitamin D deficiency and frank rickets. However, the condition remains a global issue, with the lowest estimates for deficiency still at 10% in the USA and much higher prevalences currently found in Black Americans and some groups in the Middle East (Fischer et al., 2023). Effective, timely treatment is the priority in clinical medicine; directly investigating the quality of bone beneath the growth plates and dental signs with no associated symptoms in those without frank rickets has no immediate patient benefits, so it is not part of routine clinical work. The retrospective study by D'Ortenzio et al. (2018) linked serum 25(OH)D and the presence/absence of IGD, but it involved just two individuals. Using extracted third permanent molars and 25(OH)D data, D'Ortenzio and colleagues linked the presence of IGD to deficient 25(OH)D obtained during the appropriate tooth formation stage. IGD was absent in the individual with sufficient 25(OH)D (D'Ortenzio et al., 2018). At 31 and 127 nmol/L, respectively, these results would be defined as deficient and sufficient, according to most studies discussed by Fischer et al. (2023, p.490), but there is no consensus on range of values required for optimal bone and nonskeletal health. Brickley et al. (2020b) explained that any condition that disrupts the vitamin D calcium phosphate pathway has the potential to cause mineralization defects to develop in teeth. This principle also applies to bone. Current knowledge on conditions with the potential to result in IGD and gaps in available information were tabulated (Brickley, Kahlon, & D'Ortenzio, 2020). Warning against overdiagnosed of vitamin D deficiency recommended that only cases that were part of systemic events were categorized as IGD following D'Ortenzio et al. (2018). These authors applied developmental IGD to slight defects that do not clearly follow incremental development. However, as there are many possible causes of such features, the term “consistent with” is more appropriate. Where aspects of conditions are not studied clinically, but lesions visible to those working directly with bone may be present, the biological approach set out by Mays (2018) is particularly valuable. The biological approach involves “Careful observation of the morphology and location of the lesions encountered in skeletal remains, coupled with a detailed reading of the biomedical literature while paying close attention to relationships between hard and soft tissue anatomical structures at the sites in the skeleton where lesions were characteristically seen” (Mays, 2018, p.18). Factoring in that not all IGD is linked to vitamin D deficiency, Brickley and Morgan (2023, pp.641–642) advise considering conditions such as fluorosis as an alternate diagnostic option. Toxic fluoride levels cause an increased lag time in the mineralization of tissues in bones and teeth, resulting in microscopic mineralization defects (Brickley et al., 2020a, p.87). Brickley and Morgan (2023) highlight that in addition to considering genetic conditions, malabsorption, and various forms of toxicity are relevant to IGD formation. These authors also clearly state that some changes seen in bone that could have been caused by conditions that lead to rickets have many other potential causes and so are “consistent with” rickets, a low level of diagnostic certainty. Brickley et al. (2020b) identified antibiotics as an area warranting further investigation as a cause of IGD, as it was found in exposure to antibiotics in experimental animals (Kumazawa et al., 2012). If future studies involving living individuals are designed to include data on relevant health conditions and medications are available, there is clear potential to provide evidence of the real-world effects of standard doses of various antibiotics and other health interventions and medications. In the future, it should be possible to better link clinical definitions to paleopathological research, enhancing what can be said regarding globally significant issues. IGD will likely be the microstructural indicator that could be developed as a diagnostic biomarker, allowing the transdisciplinary linking of data. Many individuals with low serum 25(OH)D levels likely also have mineralization defects in their bones and teeth. However, given the prompt treatment of those identified with rickets and/or deficient 25(OH)D in clinical medicine, these studies will not provide data on cut-off levels of serum associated with microscopic mineralization defects and associated macroscopic skeletal lesions. However, they can provide data on what microscopic mineralization defects occur, if any, in people with sufficient 25(OH)D levels; this will be immensely valuable in clarifying the diagnostic certainty terminology used when describing levels and patterns of mineralization defects in bones and teeth (IGD). Collaborative interdisciplinary partnerships between paleopathologists and contemporary clinicians/biomedical researchers involving retrospective and prospective cohort studies are underway and planned by various teams and can potentially provide valuable data. Data that can assist in establishing links between serum 25(OH)D, medical history, and dental changes are amongst the information such studies might provide. Using human skeletal material, it should be possible to determine links between IGD and bone lesions associated with rickets observed across the skeleton at the macroscopic and microscopic levels in those undergoing growth and development. Regarding microscopic mineralization defects in bone, micro-CT evaluation provides broadly equivalent data to that obtained from thin sections, but currently, less data can be accessed nondestructively from teeth (Welsh et al., 2020). At present, many micro-CT instruments and modalities only detect more marked instances of IGD (e.g., Veselka et al., 2019). Such micro-CT data can be carefully integrated with dental thin section evaluation. However, this will likely change in the future as instruments develop. Future studies of teeth need to include larger samples (or incorporate greater data sharing) and involve a more comprehensive analysis of tooth roots alongside a range of tooth types, which have different rates of growth and mineralization defect development (D'Ortenzio et al., 2018). Naturally exfoliated primary teeth hold significant potential, but due to the resorption of roots before exfoliation, relatively few such teeth retain much data on the post-birth period (Brickley, et al., 2020b, see fig. 3a). Information on maternal health should also be factored in, allowing for a deeper investigation of the mother-infant dyad through IGD relative to 25(OH)D. For others planning research involving living individuals, it is worth considering that in North America and some other places, extracting third molars, which develop during adolescence, is relatively routine. While (25(OH)D) is an immensely valuable public health tool, several limitations need to be kept in mind when using it in research (Binkley et al., 2017; Sempos et al., 2024). The estimated 3-week half-life of 25(OH)D must be considered when linking IGD to serum test results to other variables, such as sunlight exposure, season and latitude (Fischer et al., 2023). Considerable strides have been made in understanding vitamin D and the consequences of a deficiency in clinical and biomedical work (Jones, 2022) and paleopathology (Brickley et al., 2020a, pp.75–128), but areas in which knowledge can be improved remain in both fields. In clinical and biomedical research, more work on the nonskeletal effects of vitamin D is a priority (Fischer et al., 2023). The potential benefits of being aware of the nonskeletal effects of vitamin D to learn more about past communities were highlighted by Snoddy et al. (2016). For paleopathologists, a clearer understanding of how mineralization defects in bones and teeth are linked to serum 25(OH)D data will enable researchers working with skeletal remains to make more consistent assessments and comparisons between sites. Although different, this work could also tie into the need to know more about asymptomatic children and adolescents in clinical work and public health initiatives (Fischer et al., 2023). The areas covered in this commentary offer insights into how increased Interdisciplinarity and moves towards transdisciplinarity research within anthropology will assist in tackling these and other questions of global significance. Megan B. Brickley: Conceptualization (lead); funding acquisition (lead); writing – original draft (lead); writing – review and editing (lead).

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.