Original Paper
Caries Res 2012;46:467–473
DOI: 10.1159/000339664
Received: March 9, 2012
Accepted after revision: May 21, 2012
Published online: July 13, 2012
Dental Caries in Rome, 50–100 AD
O. Fejerskov a P. Guldager Bilde b M. Bizzarro d J.N. Connelly d
J. Skovhus Thomsen a B. Nyvad c
a
Department of Biomedicine – Anatomy, b Department of Culture and Society, and c Department of Dentistry,
Aarhus University, Aarhus, and d Centre for Star and Planet Formation, Natural History Museum of Denmark,
University of Copenhagen, Copenhagen, Denmark
Key Words
Archaeology ⴢ Dental medicine in Imperial Rome ⴢ Dentin
hypermineralization ⴢ Lead isotope ratio ⴢ Micro-computed
tomography
pipe. The Pb isotopic composition of the tooth did not match
that of the tube, suggesting that the subjects were exposed
to different Pb sources during their lifetime other than the
lead tubes.
Copyright © 2012 S. Karger AG, Basel
Abstract
Scarce information exists on the clinical features of dental
caries in the Imperial Roman population and no structural
data on caries lesions from this period have so far been published. We report on the findings of 86 teeth (50–100 AD)
found during archaeological excavations of the temple of
Castor and Pollux in the Forum Romanum. We found that
nearly all teeth had large carious cavities extending into the
pulp. The distribution and size of the caries lesions were similar to those found in contemporary adult populations in
Africa and China living without access to dental care. Most
lesions had a hypermineralized zone in the dentin at the
advancing front of the carious cavities as revealed by microcomputed tomography. This biological dentin reaction combined with the morphology of the cavities might indicate
that some temporary topical pain relief and intervention
treatment slowed down the rate of lesion progression. This
is indirectly supported by examination of cavities of similar
size and depth from a contemporary population without access to dental health care. In contrast to the lesions in the
Roman teeth, these lesions did not exhibit a hypermineralized dentin reaction. We investigated whether the Pb isotopic composition of enamel and/or dentin of a single tooth
matched that of a sample of an ancient Forum water lead
Despite a wealth of information about medicine, diseases, and their treatment in ancient Rome [Plinius, 1951–
1963; Celsus, 1960/61] there is limited knowledge about
the prevalence of dental caries during the Roman Imperial Age (1st–3rd century AD). Studies of skeletal remains
from central Italy (Latinum) suggest that the caries prevalence of slaves and middle-class people such as administrators and merchants was quite low, affecting no more
than 4–6% of the teeth, corresponding to 36–52% of the
population [Manzi et al., 1999]. Similar low caries figures
have been reported for 78 children below the age of 12
[Prowse et al., 2008]. By contrast, excavation of a burial
site for predominantly lower social class males near Rome
showed a ‘high frequency’ of caries [Catalano et al., 2010]
similar to reports from examination of 67 adults from the
necropolis of Quadrella in central Italy originating from
the same time period [Bonfiglioli et al., 2003]. Up until
now there have been no reports on the caries status of the
population from the city of Rome during the Imperial
period, and no studies have described the microscopical
characteristics of dental caries lesions in ancient Rome.
During archaeological excavations in the Forum Romanum in the early 1980s, initiated by the Nordic Insti-
© 2012 S. Karger AG, Basel
0008–6568/12/0465–0467$38.00/0
Fax +41 61 306 12 34
E-Mail [email protected]
www.karger.com
Accessible online at:
www.karger.com/cre
Prof. Ole Fejerskov
Department of Biomedicine – Anatomy
Wilhelm Meyers Allé 3, Aarhus University
DK–8000 Aarhus C (Denmark)
Tel. +45 8716 7638, E-Mail of @ ana.au.dk
tutes in Rome [Guldager and Slej, 1986; Ginge et al., 1989],
it became apparent that the Temple of Castor and Pollux
had a peculiar arrangement with narrow tabernae in the
podium on three sides of the temple. On the floor of one
of these the presence of what was originally thought to be
a grave was found and one of the authors (P.G.B.) became
responsible for this particular excavation site. However, it
appeared to be a channel, which led from the room to another channel in the foundation, eventually opening directly into the Cloaca Maxima [Bilde, 1999]. These channels had been overlooked by previous excavators. They
were packed with soil and remains apparently swept from
the floor of the room when in use around 50–100 AD.
Coins dating back to the Flavian and Trajanic period
were found at the bottom and in the more superficial part
of the material filling the channels, respectively. The material contained more than 13,000 fragments of ceramics
and numerous glass pieces with traces of makeup, medicine, and perfume along with a total of 86 teeth (no bone
fragments were found) scattered throughout the plug.
Consequently, the teeth could be dated to the period from
approximately 50 AD to the end of emperor Trajan’s reign
(117 AD). The material was only made available to the
authors within the last decade and because of restrictions
from the archaeological authorities we could only study
the specimens using nondestructive techniques such as
micro-computed tomography (␮CT).
Having this unique material in hand we also wanted
to investigate whether the Pb isotopic ratio in the teeth
reflects the ratio prevailing in the water lead tubes existing at the time. This question was relevant as the literature contains several suggestions about lead poisoning
from water pipes in antiquity [Nriagu, 1983; Reddy and
Braun, 2010]. However, the analytical methods for these
analyses are necessarily destructive.
Therefore, the aim of the present study was to characterize the clinical and structural features of carious teeth
retrieved from the Temple of Castor and Pollux in the Forum Romanum and to explore whether there were signs
of restorative and/or palliative treatment at the time.
Moreover, we investigated the Pb isotope composition of
two loose fragments obtained from a single tooth.
Material and Methods
Teeth
A total of 86 teeth from the archaeological excavation were
available. Following excavation the teeth were rinsed and brushed
to remove earth, but no attempts were made to clean the teeth
mechanically. The teeth consisted of 81 molars, 3 premolars, and
468
Caries Res 2012;46:467–473
2 deciduous molars. The molars comprised 35 third upper and
lower molars, 17 second molars, and 29 first molars with an equal
distribution between the arches. In addition, 15 extracted teeth
with deep caries lesions and extensive occlusal wear originating
from a contemporary rural Tanzanian population without access
to dental health care [Muya et al., 1984] were included for comparison. The latter teeth were chosen to compare teeth from a
contemporary population exhibiting extensive occlusal wear and
having no access to dental intervention. Moreover, the teeth
should have milder degrees of dental fluorosis, indicating that the
teeth originate from individuals exposed to fluoride levels in the
environment that, from a caries point of view, would most likely
slow down the rate of caries lesion progression. In addition, 1 molar extracted from a present-day adult in Copenhagen with known
environmental Pb contamination served as a control for analyses
of the Pb content and isotope composition of the Roman and Tanzanian teeth. All the teeth were examined under a stereomicroscope.
Micro-Computed Tomography
The teeth were wrapped in foam rubber and placed in a desktop ␮CT scanner (␮CT35, Scanco Medical AG, Brüttisellen, Switzerland). The teeth were scanned in high resolution mode (1,000
projections/180°) with a spatial resolution of 10 ! 10 ! 10 ␮m3,
an X-ray tube voltage of 70 kVp, a current of 114 ␮A, and an integration time of 800 ms. Images were reconstructed using the software supplied with the scanner. It has previously been shown that
the bone morphology obtained by ␮CT closely resembles that obtained by histology [Thomsen et al., 2005]. Subsequently, the image stacks were exported from the ␮CT scanner as 16-bit DICOM
image files for three-dimensional visualization.
Three-dimensional reconstruction and interactive visualization of the image stacks were performed using Amira 4.1.2 (Mercury Computer Systems, Inc., Chelmsford, Mass., USA) running
under Linux (openSUSE 11.2, www.opensuse.org). This three-dimensional visualization tool makes it possible to interactively
scale, rotate, translate, and create virtual sections through the reconstructed three-dimensional image.
Pb Content and Isotope Composition
Pieces of enamel and dentin from a single Roman tooth that
had broken away due to shrinkage and a sample of a lead water
tube serving the Forum Romanum were analyzed by thermal ionization mass spectroscopy. Two Roman teeth and the above-mentioned fragments were first rinsed with very weak HCl (0.05 M) in
an attempt to remove recently added Pb – the rinse from the 2
teeth was analyzed and is referred to in the text as ‘rinses’. Subsequently, the enamel and dentin fragments were sampled for Pb
isotope analyses with dental drill bits using a computer-guided
microdrilling instrument. The powder collected by this method
was dissolved in 6 M HCl, dried down and converted to 1 M HBr
in preparation for chemical purification. Pb from all samples was
purified by an HBr-HNO3-based chemical separation scheme using anion resin as outlined by Connelly and Bizzarro [2009]. Purified Pb was loaded onto previously outgassed Re ribbon filaments using a high-emitter silica gel and analyzed by a VG Sector
54 thermal ionization mass spectrometer. Standard NBS-981 run
during this project demonstrated linear mass fraction of 0.1%/
amu (80.02%/amu) that was applied to all analyses of unknowns.
External reproducibility of standards for the ratios reported here
Fejerskov /Guldager Bilde /Bizzarro /
Connelly /Skovhus Thomsen /Nyvad
were better than 0.03%. Error bars reported in figure 4 reflect internal errors of the individual analyses.
For comparison we also attempted to determine the Pb isotopic composition of samples from 2 of the Tanzanian (enamel, dentin, and rinse) and 1 of the Danish teeth described above.
Results
The 86 Roman teeth were characterized by rather extensive cracks and partial fractures of both enamel and
dentin as well as of the cementum surfaces at the root as
a result of shrinkage following dehydration for 2,000
years. All but 2 teeth showed deep, rounded spoon-shaped
caries cavities most of which extended close to or into the
pulp chambers (fig. 1). Most lesions (n = 74) were located
on approximal surfaces and in 43 cases the lesions also
involved, or were confined to, the occlusal surfaces. Of
the 74 approximal lesions 28 extended into or were predominantly originating from exposed root surfaces. In 3
cases noncavitated lesions were found below an approximal wear facet (fig. 1a). When examined in the stereomicroscope there were no obvious signs of mechanical
scraping or drilling. However, on several of the molars
there were clear superficial, horizontal traces of tools in
the interproximal, cervical region, which were likely
made by elevators used for extraction.
Most teeth showed marked to extensive occlusal wear
often with exposure of coronal dentin corresponding to
cusp tips (fig. 1d). As a result most occlusal fissures were
partly or totally worn away.
Despite the extensive caries destruction of the tooth
structure it was remarkable that the teeth had been extracted without apparent fractures of the teeth. In 20% of
the teeth small fractures of curved root apices were recorded. Minor aggregations of mineral deposits (calculus) were occasionally found in most cases located close
to the enamel-cementum border. In only 3 molar teeth
such deposits were observed in the middle of the root surfaces at distal roots. In 1 upper third molar part of the
buccal alveolar bone was still attached to the roots.
The ␮CT examinations revealed that in all cavities the
typical extensive lesions were clearly demarcated along
the advancing front by a zone of dentin hypermineralization (fig. 2). Towards the pulp this reaction was evident
even when the lesions were in the outer part of the dentin.
Along the walls of the lesion, where the dentinal tubules
were affected predominantly sideways, the hypermineralization reaction was of limited extent (fig. 2d). At the
interface between the bottom of the cavity and the surrounding dentin – where numerous odontoblast processDental Caries in Rome, 50–100 AD
es had been involved – the hypermineralized reaction
was rather extensive and extended towards the pulp
(fig. 2). In addition to this vital response of the pulpodentinal organ, plugs of tertiary, reactive dentin were recorded in the pulp cavity corresponding to the involved dentinal tubules (fig. 2a). The few noncavitated lesions
showed mineral loss of the outermost dentin with an early hypermineralization response (fig. 2b). The control
teeth from Tanzania with extensive cavitated lesions exhibited a very different structural pattern (fig. 3). The
zone of demineralization at the advancing front of the lesion was clearly visible as a rather narrow band, but in no
case was a hypermineralization reaction observed.
The preliminary Pb data are shown in figure 4. The
overlapping Pb isotopic composition of the three analyses
of the single Copenhagen tooth suggests that the individual was exposed to a common source of Pb during his/
her lifetime and that the tooth preserves a faithful record
of this exposure. The African teeth had insufficient Pb
content to measure the isotopic composition, consistent
with these individuals living in an environment with very
low levels of Pb and in their diet. The samples of the Roman tooth had Pb isotopic compositions of the enamel
and dentin that were different from each other and different from that of the investigated lead pipe from the
Forum Romanum.
Discussion
The present data are of historical interest because they
are the first to document the pathology of dental caries in
the Roman Empire. In addition, the findings have direct
implications for the ongoing debate among dentists on
the relevance of a more biological approach to treatment
of carious cavities [Gruythuysen et al., 2011; Kidd, personal commun., 2011]. Thus, the treatment philosophy
applied in these Roman teeth bears strong resemblance
with the nonoperative cavity treatment recently ‘rediscovered’ by Gruythuysen [2010].
Our findings clearly show that the Roman teeth had
caries lesions developing for years. Where attrition kept
pace with lesion progression on occlusal surfaces the caries attack was arrested in accord with conditions also prevailing in contemporary populations in Africa and China
without access to dental care [Manji et al., 1989; Luan et
al., 2000]. At surfaces not easily accessible to regular
cleaning, and hence prone to thick microbial deposits for
long periods of time, lesion progression seems to have
continued. However, approximal abrasion gradually wore
Caries Res 2012;46:467–473
469
Fig. 1. Lower right molar: distal (a), buccal
(b), mesial (c) and occlusal (d) surfaces. On
the distal surface a noncavitated brownish
caries lesion is located below a small abrasion facet, which is indicated with arrows.
The occlusal fissure system is partly abraded away exposing the dentin corresponding to the 3 buccal cups.
a
a
b
␮CT images of the mesial carious cavity of figure 1. Note the reactive dentin formed in the pulp corresponding to the deep penetration of the caries lesion into and exposing the pulp (asterisk). The
hypermineralized dentin reaction preceding the dentinal dissolution is marked by arrows and is particularly pronounced in c,
470
d
c
d
c
Fig. 2. a, b Three-dimensional reconstructed images based on
Fig. 3. Three-dimensional reconstructed
images based on ␮CT images of an example of a deep approximal cavity from the
Tanzanian teeth. Note that neither in the
longitudinal nor in the horizontal plane
there is any evidence of a hypermineralized dentin reaction. Instead a distinct
zone of demineralization is seen along the
walls (arrows).
b
a
Caries Res 2012;46:467–473
which represents another cavity. d Horizontal view of the lateral
wall of the cavity in c with a distinct thin zone of hypermineralization (arrows). The frame at the distal surface in b shows the
noncavitated lesion in figure 1 with a demineralization extending
through the enamel and an early demineralization of the dentin.
b
Fejerskov /Guldager Bilde /Bizzarro /
Connelly /Skovhus Thomsen /Nyvad
Pb fragment from
aqueduct
15.70
15.68
Rinse
tooth 1
15.65
Enamel
tooth 1
15.64
15.63
Rinse
tooth 2
15.62
18.42
a
207Pb/204Pb
207Pb/204Pb
15.66
Dentin
tooth 1
15.66
15.64
15.62
Copenhagen
dentin, enamel and rinse
15.60
Copenhagen
rain water
15.58
18.43
18.44
206Pb/204Pb
18.45
15.56
17.8
18.46
Fig. 4. Uranogenic Pb-Pb diagrams. a Pb isotopic analyses of
rinses from 2 Roman teeth and the enamel and dentin from a
single tooth. The Pb isotopic composition of a single fragment
obtained from a Roman aqueduct is also shown. Error bars reflect
2␴ uncertainties of the isotopic analyses. b Pb isotopic analyses of
17.9
b
18.0
18.1
18.2
206Pb/204Pb
18.3
18.4
18.5
rain water collected in Copenhagen in June 2007 and the Pb isotopic analyses of a rinse, enamel, and dentin from a long-term
resident of Copenhagen. The Pb isotopic analyses from a are
shown in the box.
away the partly demineralized enamel and created abrasion facets, but lesion progression continued beneath such
facets once the inflamed gingiva gradually retracted with
age. Hence, with age this resulted in deep caries lesions.
Restorative care has been known for thousands of
years [Bennike and Fredebo, 1986], but from the extensive medical literature of Plinius and Celsus from ancient
Rome [1951–1960] it is apparent that as long as pain could
be relieved it was recommended to preserve the teeth in
a functional state. The present sample set seems to indicate that the pain relief treatments must have been fairly
successful in ancient Rome. Based on this sample of teeth,
we believe that if successful pain relief was provided,
teeth were not extracted before lesions penetrated into
the pulp, resulting in necrosis and most likely painful
periapical inflammatory reactions. According to the records of Celsus [1960/61] pain control might have been
achieved by treating the lesion with opium, saffron, and
pepper or more ‘exotic’ components, such as fried worms
(Lumbricus terrestris), Nardus paste, or spider eggs!
To obtain efficient pain relief, it might have been necessary to remove part or most of the soft content of the
carious cavities prior to ‘pharmaceutical’ treatment. This
is most likely the explanation for the rounded and almost
circular cavities observed in this Roman material as small
spoon-shaped metal instruments were available at the
time [Milne, 1907]. This process, if performed regularly
for months or years, might in and of itself have had a dramatic influence on the rate of caries lesion progression
[Anderson, 1938; Kidd et al., 2008]. However, as such cavities have not been easily accessible for removal of the
microbial mass by the patient, no long-lasting effect on
further lesion progression could be expected. The present
␮CT investigations strongly support this interpretation.
When caries lesions progress slowly, the cellular response
from odontoblasts in the pulpodentinal tissue will try to
‘seal off’ the mineral dissolution resulting in mineral deposits within the dentinal tubules as well as in the peritubular dentin at the advancing front of the lesion [Frank
and Voegel, 1980]. This might be the explanation for the
finding in the present teeth of a distinct hypermineralized zone deep to the lesion, even in lesions, which finally extended into the tertiary dentin and pulp and resulted
in extraction. Similar phenomena are not likely to happen in carious cavities of similar size, had they not been
temporarily ‘arrested’ during years of progression, as
demonstrated by the ␮CT findings in the carious cavities
in the teeth from Tanzania. The contemporary control
teeth from Tanzania all exhibited various degrees of dental fluorosis, and hence the individuals must have been
exposed to relatively high levels of fluoride, which was
incorporated into the dentin and, thus, was available in
the oral cavity during the long-lasting periods of lesion
development. Under such conditions, without regular removal of the microbial biomass from the cavities, even a
constant fluoride exposure may not be strong enough to
reduce the rate of lesion progression and hence stimulate
a vital dentin response.
Dental Caries in Rome, 50–100 AD
Caries Res 2012;46:467–473
471
Moreover, it is remarkable that the treatment for pain
relief in ancient Rome probably resulted in preserving the
pulp vital until the moment when the pulp was exposed
to the oral environment. This is indicated by the excessive
formation of reactive dentin along the pulpodentinal interface corresponding to the dentinal tubules involved by
the caries attack as well as in the adjacent parts of the pulp.
The distribution of cavities according to tooth types is
similar to observations reported from excavations in central Italy [Manzi et al., 1999] and in a Romano-British
population from 320 to 420 AD [Whittaker et al., 1981].
Moreover, it is consistent with observations in contemporary populations around the world with no access to dental care and exhibiting marked occlusal wear [Manji et al.,
1991].
It is striking that despite the commonly very large cavities the teeth had been extracted without breaking the
crowns. Moreover, in only 22 teeth fractures of minute
fragile root tips could be observed. This demonstrates
that those removing the teeth were very skilled. There
was no evidence of any restorative material that had been
stuffed into the cavities prior to extraction to avoid fracture during removal. Several of the molar teeth exhibited
obvious signs of ‘elevators’ being applied, which most
likely have been widely used at the time [Milne, 1907].
Thus, medical and dental care must have been available
to part of the population of Rome [Plinius, 1951–1963;
Celsus, 1960/61].
The teeth did not show evidence of much calculus
around the neck of the teeth and as such there is little
evidence that extractions were eased because the teeth
were loose as a result of periodontal breakdown of the
marginal alveolar bone. With increasing age a gradual
loss of bone around molar teeth is well documented in
contemporary populations exhibiting excessive occlusal
tooth wear without loosening of the teeth [Baelum and
Fejerskov, 1986; Baelum et al., 1988; Manji et al., 1991;
Baelum et al., 1996]. There is no reason to expect that it
should be different in this ancient population and the
findings of a number of the caries lesions developed from
exposed interproximal cervical areas support this assumption.
We do not know the identity of the persons from whom
these teeth originate. It is tempting to suggest that the
teeth belonged to persons of a certain social standing taking the location of the taberna into consideration. If correct, this sample of Roman teeth may reflect persons that,
in contrast to previous findings from poor populations
from rural areas [Manzi et al., 1999], had access to pain
relief treatment and dental care.
472
Caries Res 2012;46:467–473
That the Pb isotope composition of the teeth was different from that of the lead pipe may best be explained by
the fact that the drinking water in Rome was isolated
from the inner side of the lead pipes by thick incrustations of calcium carbonates [Hodge, 1981]. The difference
between dentin and enamel may suggest that this person
was exposed to different Pb sources during his lifetime.
The fact that the rinse of 1 tooth contained Pb with an
isotopic composition that may have been a mixture of Pb
from the enamel and the lead pipe indicates that a component of Pb in or on the tooth came from the aqueduct.
However, we cannot determine whether this Pb was incorporated when the subject was alive or after the tooth
was extracted and lay in the channel exposed to the local
water for several years coming from the lead pipes in
Rome. In fact it is more likely that the Pb intoxication
commonly claimed to be a problem in ancient Rome
[Nriagu, 1983; Reddy and Braun, 2010] may have originated from food and wine sweetened with lead salt. Anyhow, an in-depth analysis of the Pb isotopic composition
of the teeth would require that portions of the unique archaeological material be destroyed, which was not allowed by the Italian authorities and could consequently
not be performed in the present study.
We conclude that it is likely that this taberna had functioned as a small pharmacy and beauty shop where it was
possible also to seek pain relief treatment for toothache.
In addition to pain relief many of the remedies used have
most likely also had a certain bacteriostatic effect. This
combined with the removal of the soft content of the cavities with small spoon-shaped instruments could explain
the temporary arrest of lesion progression. As a protective zone of hypermineralization in the dentin was attainable relief of pain was obtained and the teeth remained functional for a longer period. This most likely
explains why Celsus [1960/61] did not recommend tooth
extraction as the first choice as long as the tooth could
remain functional. With present-day knowledge about
the role of ‘caries control’ in modern populations [Fejerskov, 2004] these data are highly relevant and may help
us justify when and how to apply nonoperative versus
classical restorative approaches to dental care.
Acknowledgments
The Bagger-Sørensen Foundation and The Velux Foundation
are thanked for economic support. Special thanks are due to the
Danish Institute in Rome, the Ministero per i Beni e le Attività
Culturali, Rome, and Dr. Kaj Josephsen for help in preparing the
figures.
Fejerskov /Guldager Bilde /Bizzarro /
Connelly /Skovhus Thomsen /Nyvad
References
Anderson BG: Clinical study of arresting dental
caries. J Dent Res 1938;17:443–452.
Baelum V, Chen X, Manji F, Luan WM, Fejerskov
O: Profiles of destructive periodontal disease
in different populations. J Periodontal Res
1996;31:17–26.
Baelum V, Fejerskov O: Tooth loss as related to
dental caries and periodontal breakdown in
adult Tanzanians. Community Dent Oral
Epidemiol 1986;14:353–357.
Baelum V, Luan WM, Fejerskov O, Chen X:
Tooth mortality and periodontal conditions
in 60–80 year-old Chinese. Scand J Dent Res
1988;96:99–107.
Bennike P, Fredebo L: Dental treatment in the
Stone Age. Bull Hist Dent 1986; 34:81–87.
Bilde PG: Hos tandlægen: en butik i Castor og
Pollux’ tempel; in Bilde PG, Nørskov V, Pedersen P (eds): Hvad fandt vi? En gravebog
fra Institut for Klassisk Arkæologi, Aarhus
Universitet, 1999, pp 135–140.
Bonfiglioli B, Brasili P, Belcastro MG: Dento-alveolar lesions and nutritional habits of a Roman Imperial age population (1st–4th c.
AD): Quadrella (Molise, Italy). Homo 2003;
54:36–56.
Catalano P, Benassi V, Caldarini C, Cianfriglia
L, Mosticone R, Nava A, Pantano W, Porreca
F: Health status and life style in Castel Malnome (Rome, I–II cent. A.D.). Med Secoli
2010;22:111–128.
Celsus, translated by Spencer WG: The Loeb
Classical Library. London, 1960/61, vol I–III:
On Medicine.
Connelly JN, Bizzarro M: Pb/Pb dating of chondrules from CV chondrites by progressive
dissolution. Chem Geol 2009;259:143–151.
Fejerskov O: Changing paradigms in concepts
on dental caries – consequences for oral
healthcare. Caries Res 2004;38:182–191.
Dental Caries in Rome, 50–100 AD
Frank RM, Voegel JC: Ultrastructure of the human odontoblast process and its mineralization during dental caries. Caries Res 1980;14:
367–380.
Ginge B, Becker M, Guldager P: Of Roman extraction. Archaeology 1989;42:34–37.
Gruythuysen RJM: Niet-Restaurative Caviteitsbehandeling. Cariësactiviteit beteugelen in
plaats van maskeren. Ned Tijdschr Tandheelkd 2010;117:173–180.
Gruythuysen RJM, van Strijp AJP, van Palenstein Helderman WH, Frankenmolen FW:
Niet-restaurative behandeling van cariës in
het melkgebit: doelmatig en kindvriendelijk.
Ned Tijdschr Geneeskd 2011;155:A3489.
Guldager P, Slej K: Gli scavi scandinavi nel Foro
Romano. Il tempio di Castore e Polluce. Archeologia Viva 1986;5.4:24–37.
Hodge AT: Vitruvius, lead pipes and lead poisoning. Am J Archaeol 1981;85: 486–491.
Kidd EAM, Bjørndal L, Beighton D, Fejerskov O:
Caries removal and the pulpo-dentinal complex; in Fejerskov O, Kidd EAM (eds): Dental
Caries. The Disease and Its Clinical Management, ed 2. Oxford, Blackwell, 2008, pp 367–
382.
Luan WM, Baelum V, Fejerskov O, Chen X: Tenyear incidence of dental caries in adult and
elderly Chinese. Caries Res 2000; 34: 205–
213.
Manji F, Fejerskov O, Baelum V: Pattern of dental caries in an adult rural population. Caries
Res 1989;23:55–62.
Manji F, Fejerskov O, Baelum V, Luan WM, Chen
X: The epidemiological features of dental
caries in African and Chinese populations:
implications for risk assessment; in Johnson
NW (ed): Risk Markers for Oral Diseases.
Cambridge, Cambridge University Press,
1991, vol 1, Dental Caries: Markers of High
and Low Risk Groups and Individuals, pp
62–99.
Manzi G, Salvadei L, Vienna A, Passarello P:
Discontinuity of life conditions at the transition from the Roman Imperial age to the early middle ages: example from central Italy
evaluated by pathological dento-alveolar lesions. Am J Hum Biol 1999; 11:327–341.
Milne JS: Surgical Instruments in Greek and Roman Times. Oxford, Claredon Press, 1907.
Muya RJ, Rambusch E, Fejerskov O, Hobdell
MH, Nørmark S: Changing and developing
dental health services in Tanzania 1980–
2000. Ministry of Health, Dar-es-Salaam,
Tanzania and DANIDA, Copenhagen, 1984.
Nriagu JO: Lead and Lead Poisoning in Antiquity. New York, Wiley & Sons, 1983.
Plinius, translated by Jones WHS: The Loeb
Classical Library. London, 1951, 1956, 1963,
vol VI, VII, VIII: Natural History.
Prowse TL, Saunders SR, Schwarcz HP, Garnsey
P, Macchiarelli R, Bondioli L: Isotopic and
dental evidence for infant and young child
feeding practices in an imperial Roman skeletal sample. Am J Phys Anthropol 2008;137:
294–308.
Reddy A, Braun CL: Lead and the Romans. J
Chem Educ 2010;87:1052–1055.
Thomsen JS, Laib A, Koller B, Prohaska S,
Mosekilde LI, Gowin W: Stereological measures of trabecular bone structure: comparison of 3D micro computed tomography with
2D histological sections in human proximal
tibial bone biopsies. J Microsc 2005;218:171–
179.
Whittaker DK, Molleson T, Bennett RB, Edwards I, Jenkins PR, Llewelyn JH: The prevalence and distribution of dental caries in a
Romano-British population. Arch Oral Biol
1981;26:237–245.
Caries Res 2012;46:467–473
473