See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/6250290

Radiographic features of periodontal bone defects: Evaluation of digitized

images

Article  in  Dentomaxillofacial Radiology · August 2007

DOI: 10.1259/dmfr/25386411 · Source: PubMed

CITATIONS READS
33 855

8 authors, including:

Isaac S Gomes-Filho Viviane Sarmento

Universidade Estadual de Feira de Santana Universidade Federal da Bahia
124 PUBLICATIONS   1,714 CITATIONS    145 PUBLICATIONS   660 CITATIONS   

SEE PROFILE SEE PROFILE

Soraya Castro Trindade Camila Freitas

Feira de Santana State University Universidade Estadual de Feira de Santana
106 PUBLICATIONS   1,137 CITATIONS    23 PUBLICATIONS   336 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Veterinary Immunology View project

Sickle-cell disease View project

All content following this page was uploaded by Isaac S Gomes-Filho on 24 December 2015.

The user has requested enhancement of the downloaded file.

 Dentomaxillofacial Radiology (2007) 36, 256–262
q 2007 The British Institute of Radiology
http://dmfr.birjournals.org

RESEARCH
Radiographic features of periodontal bone defects: evaluation
of digitized images
IS Gomes-Filho*,1, VA Sarmento1, MS de Castro1, NP da Costa2, SS da Cruz1, SC Trindade1, COT de Freitas1
and J de Santana Passos1
1
Department of Periodontics, Feira de Santana State University, Bahia, Brazil; 2Department of Radiology, School of Dentistry,
Pontifı́cia Universidade Católica, Porto Alegre, Rio Grande do Sul, Brazil

Objectives: Radiographic evaluation of periodontal conditions is a difficult process because of the

many limitations inherent to this complementary examination and the lack of image specificity for
different types of bone defects. The objectives of this study are to utilize digitized imaging resources
to describe the radiographic features of different types of periodontal bone defects produced
artificially in dry mandibles.
Methods: 14 dry mandibles were utilized, in which periodontal bone defects were produced.
Digital photographs and conventional radiographs of each site were obtained in a standardized
manner, before and after producing these defects. The radiographs were then digitized, displayed on
the screen and evaluated by three examiners.
Results: Most of the bone defects presented distinct imaging characteristics that allowed them to
be identified, with the exception of one-, two- and three-wall infrabony defects. The defects that
were most easily interpreted were horizontal and vertical defects and the most difficult were defects
in the radicular septum.
Conclusions: Despite the importance of imaging diagnoses in reaching clinical decisions
regarding treatment, such diagnoses for different types of periodontal bone defects are extremely
difficult to make. In the present study, the utilization of digital tools for interpreting digitized
radiographic images facilitated the process.
Dentomaxillofacial Radiology (2007) 36, 256–262. doi: 10.1259/dmfr/25386411

Keywords: digitized image, periodontal lesions, radiographic interpretation

Introduction

Periodontal disease is characterized by periods of activity
in which the periodontal support structures are destroyed
by the action of chemical mediators of inflammation,
followed by periods of quiescence with remission from the
destructive process. These characteristics, together with
the complexity of the disease, signify that diagnostic
methods play an important role in this specialty. For
periodontal conditions to be defined and hence for
individuals with periodontal disease to be identified,
clinical and radiographic examination need to be utilized
together. Several clinical examinations cited in the
literature have contributed towards determining the
clinical diagnosis for periodontal disease.1 As a comp-
lementary method in the diagnostic process, radiographic
*Correspondence to: Isaac Suzart Gomes-Filho, Av. Getúlio Vargas, 379, Centro.
Feira de Santana, Bahia, 44025-010, Brazil; E-mail: isuzart@gmail.com
Received 13 March 2006; revised 23 June 2006; accepted 28 June 2006
examination has taken on a decisive role in establishing the
state of the bone tissue and, in some situations, the
presence and extent of periodontal disease.2
Even though the indications for radiographic examin-
ation are precise and they are frequently used for evaluating
the periodontal condition, radiographic imaging presents
many limitations. They underestimate the start and extent of
bone resorption3 and produce a two-dimensional represen-
tation of the tissue that has an influence on the apparent size
and shape of the lesion, among other problems.4 – 12 In
addition, the variations in projection angle may affect the
subsequent radiographic quantification of bone height.2
With the advent of digital imaging, radiologists have
adapted to the more complex and versatile tools available,
resolved diagnostic problems and found new and more
efficient solutions for old problems.13 Through the use of
grayscale distribution tools, digitized imaging has become
an examination capable of showing changes in architecture

that are sometimes not detected by conventional radio-
graphic imaging.14 In this light, the present study had the
objective of describing the radiographic features on
digitized images of different types of periodontal bone
defects that were artificially produced, with the aim of
guiding professionals working in this field regarding the
radiographic interpretation of such lesions.
Materials and methods
The design for this research was submitted to and approved
by the Research Ethics Committee of Feira de Santana
State University, Bahia, Brazil (protocol no. 004/2004).
For this study, 14 dry mandibles were utilized. They
all contained at least one posterior tooth that presented
clinically normal periodontal bone tissue (Figures 1 and 2).
The mandibles were made available by the Department of
Morphological Sciences of Feira de Santana State Univer-
sity, and by the Human Anatomy Laboratory of the School
of Agrarian and Health Sciences of the Metropolitan Union
for Education and Culture (UNIME). For each tooth, four
sites could be utilized for producing the bone defects, on the
mesial, distal, vestibular and lingual surfaces.
Eight types of periodontal bone defects were selected:
horizontal, vertical, interdental crater, one-, two- and
three-wall infrabony defects; and defects in the radicular
septum on the vestibular or lingual surface, and in the
radicular septum with apical extension on the vestibular or
lingual surface. A minimum of five defects of each type were
produced.
Figure 1 (a – d) Photographs of the sites before producing the bone defects; (a0 – d0 ) digitized images of the sites before producing the bone defects
Periodontal bone defects on digitized images
IS Gomes-Filho et al 257
The bone defects were produced in a single session by a
single professional who is a specialist in periodontics.
They were made using a slow-rotation micromotor
(Kavow; Lake Zurich, IL) with No. 2 spherical and No.
56 cylindrical burs, under the focused light of the dental
equipment. The simulation of the defects was done in
accordance with the following descriptions:
(a) Horizontal bone defect: bone loss perpendicular to
the long axis of the tooth, along the whole length of
the alveolar bone crest (ABC), with occurrence
of resorption of the vestibular and lingual cortical
laminae, and of the interdental bone (Figure 3a);
(b) Vertical bone defect: located in the interdental bone
and defined by oblique or angular orientation of the
bone resorption in relation to the long axis of the
tooth, with apical direction (Figure 3b);
(c) Interdental crater: depression in the bone crest
between adjacent teeth, composed of vestibular and
lingual walls, and two other walls created by the roots
of the adjacent teeth (Figure 3c);
(d) One-wall infrabony defect: defect limited by one
osseous wall and the tooth surface. Formed when the
mesial or distal portion of the interdental bone
septum is reabsorbed, or likewise the two vestibular
or lingual cortical laminae (Figure 3d);
(e) Two-wall infrabony defect: defect limited by two
osseous walls and the tooth surface. Formed
when the mesial or distal portion of the
interdental bone septum is resorbed, or likewise
one of the vestibular or lingual cortical laminae
(Figure 4a);
Dentomaxillofacial Radiology
 Periodontal bone defects on digitized images
258 IS Gomes-Filho et al

Figure 2 (a – d) Photographs of the sites before producing the bone defects; (a0 – d0 ) digitized images of the sites before producing the bone defects

(f) Three-wall infrabony defect: defect limited by remaining interdental bone septum from the roots
three osseous walls and the tooth surface. In this of the non-involved adjacent tooth, after the
case, the bone defect is surrounded by destruction of the distal or mesial portion of the
the vestibular and lingual laminae, and by the interdental bone (Figure 4b);

Figure 3 (a – d) Photographs of the periodontal bone defects. (a0 – d0 ) Digitized images of the periodontal bone defects: (a,a0 ) horizontal bone defect;
(b,b0 ) vertical bone defect; (c,c0 ) interdental crater; (d,d0 ) one-wall infrabony defect

Dentomaxillofacial Radiology

Figure 4 (a –d) Photographs of the periodontal bone defects. (a0 –d0 ) Digitized images of the periodontal bone defects: (a,a0 ) two-wall infrabony defect;
(b,b0 ) three-wall infrabony defect; (c,c0 ) radicular septum bone defect; (d,d0 ) radicular bone defect with apical extension
(g) Radicular septum bone defect: non-uniform radi-
cular septum resorption. It compromises both the
alveolar bone cortex as well as the trabecular
bone, on the lingual or vestibular surface, leaving
a portion of the tooth root exposed to the
periodontal pocket (Figure 4c);
(h) Radicular septum bone defect with apical exten-
sion: progression of a radicular septum bone
defect. Characterized by resorption of the radicu-
lar septum, extending towards the apex of the
tooth. May also occur both on the lingual surface
and on the vestibular surface (Figure 4d).
Before and after producing the bone defects, the
mandibles were photographed and radiographed in a
standardized manner. Initially, the mandibular teeth
were scaled to remove possible calculi present in the
areas selected for the study. The photographs were
taken using a digital still camera (Olympus ER 100Sw;
Olympus Corp., Japan), with a matrix of 1360 £ 1024
pixels and resolution of 96 ppi, and the images were
saved in JPEG format. Both the vestibular and the
lingual surface of each study area were photographed.
Next, the mandibles were radiographed using radio-
graphic film of size 2 (AGFA Dentus M2 Comfort; Agfa,
Belgium), which was attached parallel to the lingual
surface of the mandible using adhesive tape, at the sites
selected. The lingual surface of the mandible was marked
out so that another sheet of film could be positioned there
for the radiograph to be repeated after creating the bone
defects. A soft tissue simulator was utilized (4 cm3 of
water) and was positioned in front of the mandible.
Periodontal bone defects on digitized images
IS Gomes-Filho et al 259
The locating cylinder for the X-ray apparatus, operating at
70 kVp and 8 mAs (Spectro 70X; Dabi Atlante SA,
Ribeirão Preto, Brazil), was positioned so as to have
perpendicular incidence on this setup, with a focal distance
of 20 cm. The exposure utilized was 0.4 s. The films were
processed using the temperature-time method, in a
labyrinthine darkened chamber, with a safety lamp
(GBX-2; Kodak) and new processing solutions (Kodak
Company, New York, NY). The radiographs were dried in
a hot-air chamber (EMB, Brazil).
When ready, the radiographs were digitized using a
scanner with a transparency reader (HP PrecisionScan Pro
2.5, ScanJet XPA; Hewlett Packard Company, CO), with a
spatial resolution of 600 ppi, in grayscale mode and at 8
bits. The positioning of the radiograph on the active
surface of the scanner and also the exposure adjustments
(half-tones, shadows and highlights) were standardized.
All the images were saved in BMP format and stored on
CD-ROM.
For interpretation, the digitized radiographic images
were first opened within the ImageTool program (Univer-
sity of Texas, Health Science Center, San Antonio, TX).
After calibration of the linear distance-measuring tool, the
distance from the cemento-enamel junction (CEJ) to the
ABC was established for each interdental septum to be
evaluated. These measurements were noted and made
available to the examiners. Next, a presentation in
PowerPoint (Microsoft Officew, Redmond, WA) was
created in which the digitized radiographic images and
the corresponding photographs, obtained before and after
producing the bone defects, were sequentially arranged for
comparison. The file was displayed on a 17 inch super
Dentomaxillofacial Radiology
 Periodontal bone defects on digitized images
260 IS Gomes-Filho et al
VGA flat-screen monitor and the images were analysed by
three professionals: two radiologists and one specialist in
periodontics and radiology. These examiners described
characteristic radiographic features of each type of bone
defect on a specific form (Figure 5). The images could, if
necessary, be freely adjusted in size, brightness and
contrast by the examiners to facilitate the evaluation, by
means of the figure tools in the PowerPoint software. In the
event of disagreement, the feature cited by the majority
would prevail. The level of agreement achieved among the
three examiners varied from 0.67 to 0.78 (kappa test). These
interpretations were made in a darkened environment with the
examiners positioned approximately 50 cm from the monitor.
Results
The radiographic features observed by the examiners in the
different types of periodontal bone defects generated the
following descriptive analysis after data compilation:
(a) Horizontal bone defect: by tracing out two imaginary
lines between the CEJs of the adjacent teeth and over
the ABC, it was observed that the distance between
them was greater than 2 mm. In addition to this, it
was noted that the radiopacity of the ABC had
disappeared (Figure 3a0 );
(b) Vertical bone defect: by tracing out two imaginary lines
between the CEJs of the adjacent teeth and over the
ABC, it was observed that there was an intersection
between them. It was also observed that the radiopacity
of the ABC and the lamina dura in the area close to the
defect had disappeared (Figure 3b0 );
Figure 5 Specific data collection form used by the examiners to record the description of the bone defects
Dentomaxillofacial Radiology
(c) Interdental crater: partial disappearance of the radio-
pacity of the ABC and subtle alteration in the trabecular
bone texture in the region of the defect were observed.
On the other hand, by tracing out two imaginary lines,
between the CEJs of the adjacent teeth and over
the ABC, it was seen that the distance between
them remained within normal values (up to 2 mm)
(Figure 3c0 );
(d) One-wall infrabony defect: almost complete disappear-
ance of the radiopacity of the ABC of the interdental
septum was observed, along with partial disappearance
of the lamina dura and alteration of the trabecular bone
texture in the region of the defect (Figure 3d0 );
(e) Two-wall infrabony defect: there was almost complete
disappearance of the radiopacity of the ABC of the
interdental septum, along with partial disappearance of
the lamina dura and alteration of the trabecular bone
texture in the region of the defect (Figure 4a0 );
(f) Three-wall infrabony defect: almost complete disap-
pearance of the radiopacity of the ABC of the
interdental septum was observed, along with partial
disappearance of the lamina dura and alteration of the
trabecular bone texture in the region of the defect
(Figure 4b0 );
(g) Radicular septum bone defect: subtle alteration of the
trabecular bone texture in the region of the defect was
observed (Figure 4c0 );
(h) Radicular septum bone defect with apical extension:
there was subtle alteration of the trabecular bone texture
in the region of the defect that extended to the periapical
region, suggesting a radiolucent image of the periapical
lesion (Figure 4d0 ).

Discussion
According to the descriptive analysis of the periodontal
bone defects that were artificially produced for this study,
the results showed that most defects presented distinct
radiographic features of differing levels of interpretational
difficulty. Most of the defects presented characteristic
images that differentiated one from another, except for the
one-, two- and three-wall infrabony defect types, for which
the radiographic features were very similar.
Among the most easily interpreted periodontal bone
defect types, horizontal and vertical bone defects can be
highlighted. The possibility of tracing out an imaginary
line joining the CEJs of adjacent teeth and another line
over the ABC of the interdental septum, before evaluating
the positioning of the lines and the distance between them,
made the radiographic interpretation of these defects easier
to interpret. Normally, these lines should be positioned
parallel to each other, approximately 1 – 2 mm apart. In
addition to this radiographic finding, the loss of radiopacity
of the ABC and the lamina dura was an important feature
in arriving at a final definition of these types of bone
defects. These findings corroborate those reported in the
literature.12,15 – 17
On the other hand, among the types of periodontal bone
defects investigated in this study, the most difficult to
identify and interpret was bone defects in the radicular
septum, when created on either vestibular or lingual
surfaces. The image of this type of defect is the product
of superimposition of the tooth and the remaining bone
structure. Normally the radicular bone septum has less
thickness than the radicular structure and is superimposed
on it. Consequently, it is almost impossible to detect the
radiographic image of this bone defect.
Next in decreasing order of difficulty for identifying the
periodontal bone defects were crater defects, which
presented subtle features upon evaluation. These defects
are characterized by destruction of the ABC along its entire
mesial – distal extent. For this reason, it was observed that
the radiopacity of the ABC had disappeared and, because
of the advance of the destruction within the spongy bone
tissue, radiographic alteration of the trabecular texture of
the interdental bone tissue was also seen.
Finally, the one-, two- and three-wall infrabony period-
ontal defects had radiographic features that were very
similar to each other. Once again, superimposition of the
remaining bone structure made it difficult to define the
specific type of bone alteration. Radiographically, it was
observed that there was an alteration in the image when
complete radiopacity of the ABC, lamina dura and
trabecular texture in the interdental bone were not
identified. Of these, one-wall infrabony defect is the
most easily detectable type, because of the non-existence
of external bone cortex. On the other hand, the radio-
graphic images of the remaining bone tissue in two- and
three-wall defects are confounded due to superimposition
of the external bone cortex. In this light, it was not possible
to define separate, specific radiographic features for each
of these defect types.12
Periodontal bone defects on digitized images
IS Gomes-Filho et al 261
In the present study, one type of bone alteration was
created that is widely encountered in clinical practice at the
time of surgical access to periodontal lesions: radicular
septum bone defects with apical extension. This type of
defect is characterized by the advance of the bone
destruction process to the areas around the dental apex.
Because of the superimposition of the images, the radio-
graphic appearance seen is suggestive of periapical lesions.
Comparison of the photograph of the defect with the
radiographic image facilitated the interpretation of the
defect in the present study. However, in vivo evaluation of
such defects requires the utilization of complementary tests
such as the pulp sensitivity test, which will define whether
or not necrosis of the pulp tissue is present and thus suggest
what type of lesion is really present.
The different characteristics presented by each type of
bone defect arise as a consequence of each individual’s
structural determinants, in each region of the mouth and,
more specifically, each dental site. For example, it is known
that the trabecular texture of the spongy bone in a person of
small stature differs from that of a taller individual with
longer bones. In the former, there is a tendency towards a
trabecular bone texture that is more radiolucent and, in the
latter, more radiopaque. This difference is probably due to
the layout of the medullary spaces, in quantity and size.
From this perspective, spongy bone tissue with a smaller
quantity of medullary spaces of larger size will, when
affected by an inflammatory process, tend to provoke a bone
defect more rapidly than in the opposite situation, in which
the tissue is formed with a larger quantity of medullary
spaces of smaller size. In the latter, the larger quantity of
bone trabeculae makes it more difficult for the bone tissue
to be destroyed, giving rise to a slower rate of progression.
Together with these features, there are other important
factors that determine different types of bone defects, such
as the shape of the alveolar bone crest, the thicknesses of
the interdental and radicular bone septa, the position and
calibre of the arteries in bone septum, the eruption location
of the tooth within the arch and the presence of bone
dehiscence and fenestration.15
Knowledge of these factors facilitates comprehension of
the mechanisms through which periodontal disease pro-
gresses among the periodontal support structures. In
periodontics, the type of surgical procedure is determined
by means of prior evaluation of the clinical indicators, in
association with features viewed radiographically. Regen-
erative techniques using biomaterials exist, and the sine
qua non for using these is that the bone defect should be
retentive. The possibility, prior to surgical therapy, of
defining the type of bone defect that exists and whether it
is retentive or not may dictate which periodontal surgical
procedure is utilized. It will at least make it easier or
improve the detail in the definitive planning for the surgery
to be utilized, for example in acquiring and typing the
biomaterials needed.
In conclusion, this in vitro investigation makes knowl-
edge of bone alterations more accessible, since in vivo
studies on this subject are increasingly rare considering the
ethical commitment involved in research. On the other
hand, digitized imaging has become an important tool in
Dentomaxillofacial Radiology
 Periodontal bone defects on digitized images
262 IS Gomes-Filho et al
determining the subtle alterations seen on images of bone
defects. This is not only because of the magnification of the
image on the screen, but also because of the possibility of
adjusting its brilliance and contrast. Moreover, with
References
1. Gomes-Filho IS, Sarmento VA, Cerqueira EMM, Sampaio FP,
Rösing CK, Vianna MIP, et al. Critérios para o diagnóstico clı́nico da
doença periodontal. JBC Jornal Brasileiro de Clı́nica Odontológica
Integrada & Saúde Bucal Coletiva 2005; 9: 88 – 89.
2. Greenfield DS, Williams RC, Goldhaber P. Radiographic measure-
ment of alveolar bone loss: a perspective in vitro. J Clin Perio 1981;
8: 474 –480.
3. Albandar JM. Validity and reliability of alveolar bone level
measurements made on dry skulls. J Clin Perio 1989; 16: 575 –579.
4. Forsberg J. A comparison of the paralleling and bisecting angle
radiographic technique in endodontics. Int Endod J 1987; 20:
177 –182.
5. Bender IB. Factors influencing the radiographic appearance of bone
lesions. J Endod 1982; 8: 161 – 170.
6. Aun CE, Freitas A. Radiologia em endodontia. In: Paiva JA,
Antoniazzi JH (eds). Endodontia: bases para a prática clı́nica, 2nd
edn. São Paulo: Artes Médicas, 1993, pp 331 – 364.
7. Serson D, Nunes JEO. Cintilografia. In: Freitas A, Rosa JE, Souza IF
(eds). Radiologia odontológica, 3rd edn. São Paulo: Artes Médicas,
1994, pp 631 –637.
8. Fuhrmann RA, Wehrbein H, Langen HJ, Diedrich PR. Assessment of
the dentate alveolar process with high resolution computed
tomography. Dentomaxillofac Radiol 1995; 24: 50 – 54.
Dentomaxillofacial Radiology
View publication stats
improvements in the resources of digitized imaging
programs, many further advances in this field of research
can be achieved, with verification of the incipient radio-
graphic features of periodontal alterations.
9. De Deus QD. Endodontia, 5th edn. Rio de Janeiro: Medsi, 1992,
pp 125 – 188.
10. Rosa JE, Tavares D. Princı́pios de interpretação radiográfica. In:
Freitas A, Rosa JE, Souza IF (eds). Radiologia odontológica, 3rd edn.
São Paulo: Artes Médicas, 1994, pp 339 – 365.
11. Mol A, van der Stelt PF. Locating the periapical region in dental
radiographs using digital image analysis. Oral Surg Oral Med Oral
Pathol 1993; 75: 373 –382.
12. White SC, Pharoah MJ. Periodontal diseases. In: White SC, Pharoah
MJ (eds). Oral radiology – principles and interpretation, 5th edn. St
Louis, MO: Mosby, 2004, pp 314 –329.
13. Webber RL. Computers in dental radiography: a scenario for the
future. JADA 1985; 111: 419 – 424.
14. Sarmento VA, Pretto SM, Costa NP. Entendendo a imagem
digitalizada. Rev Odonto Ciência 1999; 14: 171 – 178.
15. Ramfjord SP, Ash MM Jr. Exame e diagnóstico. In: Ramfjord SP, Ash
MM Jr (eds). Periodontologia e periodontia: Teoria e prática
moderna, 1st edn. São Paulo: Santos, 1991, pp 129 – 133.
16. Lascala NT, Lascala CA. Exame radiográfico em periodontia. In:
Lascala NT, Moussalli NH (eds). Compêndio terapêutico period-
ontal, 2nd edn. São Paulo: Artes Médicas, 1995, pp 42 – 57.
17. Freitas A, Rosa JE, Souza IF. Radiologia odontológica, 4th edn. São
Paulo: Artes Médicas, 1998, pp 411 –435.