Content uploaded by Amir Savage
Author content
All content in this area was uploaded by Amir Savage on Apr 19, 2018
Content may be subject to copyright.
A systematic review of definitions
of periodontitis and methods
that have been used to identify
this disease
Savage Amir, Eaton KA, Moles DR, Needleman I. A systematic review of definitions of
periodontitis and methods that have been used to identify this disease. J Clin
Periodontol 2009; 36:458–467.doi: 10.1111/j.1600-051X.2009.01408.x.
Abstract
Objective: To perform a systematic review and critical analysis of the definitions of
periodontitis and the methods which have been used to identify and measure this
disease.
Material and Methods: Relevant publications were identified after searching
MEDLINE, EMBASE, SCISEARCH and LILACS electronic databases. Screening of
titles and abstracts and data extraction was conducted independently by two reviewers.
To be included in the review, studies were required to define periodontitis and to
indicate how it was measured. Studies that related purely to gingivitis, and/or
intervention studies, and/or studies where prevalence or severity of periodontitis was
not a principal outcome were excluded.
Results: From a total of 3472 titles and abstracts, 104 potentially relevant full text
papers were identified. Of these, 15 met the criteria for inclusion in the final stage of
the review. The survey revealed heterogeneity between the studies in the measurement
tools used, particularly the types of probes and the sites and areas of the mouth that
were assessed. There was also heterogeneity in the use of clinical attachment loss
(CAL) and pocket probing depth (PPD) as criteria for periodontitis. In the 15 studies,
the threshold for a diagnosis of periodontitis when CAL was the criterion ranged from
2toX6 mm and when PPD was used, from 3 to X6 mm.
Conclusions: This review has confirmed previous work which has suggested that
epidemiological studies of periodontal diseases are complicated by the diversity of
methodologies and definitions used. The studies that were reviewed utilized a
minimum diagnostic threshold defining periodontitis, at a given site in terms of CAL of
2 mm and PPD of 3 mm.
Key words: epidemiology; definition;
periodontal disease; periodontitis; systematic
review
Accepted for publication 12 March 2009
From a pathological point of view,
periodontitis can be defined as the pre-
sence of gingival inflammation at sites
where there has been a pathological
detachment of collagen fibres from the
cementum and the junctional epithelium
has migrated apically. Inflammatory
events associated with connective tissue
attachment loss also lead to the resorp-
tion of coronal portions of tooth sup-
porting alveolar bone (Armitage 1995).
The point at which the presence of a
condition can be identified is termed the
diagnostic threshold. Ideally, an appro-
priate diagnostic threshold should be
accurate, consistent, easily diagnosed
and occur as early as practicable in the
life-history of the disease.
The criteria used to define perio-
dontitis must be unequivocal and suita-
ble for any examiner to use such that the
same diagnosis is achieved by other
examiners under identical conditions.
Previous studies have used an array of
clinical signs and symptoms such as
gingivitis, bleeding on probing (BOP),
pocket probing depth (PPD), clinical
attachment loss (CAL), as well as radio-
Amir Savage, Kenneth A. Eaton,
David R. Moles and Ian Needleman
UCL Eastman Dental Institute, London
WC1X 8WD, UK
Conflict of interest and source of
funding statement
None of the authors have any conflict of
interests.
A Savage is employed by the United King-
dom’s Defence Dental Services who
funded his work on this project.
J Clin Periodontol 2009; 36: 458–467 doi: 10.1111/j.1600-051X.2009.01408.x
458 r2009 John Wiley & Sons A/S
graphically assessed alveolar bone loss.
The most commonly used clinical mea-
sures in epidemiological studies for perio-
dontitis are CAL and pocket probing
depth (PPD). The method for measuring
CAL and PPD with a manual probe was
described by Ramfjord (1959). Although
CAL, is a measure of accumulated past
disease at a site rather than current activ-
ity it remains a diagnostic standard for
measuring periodontitis (Goodson 1992),
as it utilizes a fixed reference point rather
than gingival margins whose levels are
subject to alterations with change in
inflammatory state.
The absence of consensus on how
best to incorporate CAL and PPD into
a definition of periodontitis continues to
hamper clinical and epidemiological
research (Burt 2005). These inconsisten-
cies in the use of disease indicators
make large variations in the definition
of periodontitis inevitable (Borrell &
Papapanou 2005).
Previous literature reviews relating to
periodontal epidemiology have all high-
lighted similar diagnostic and methodo-
logical problems. In May 1999, a
MEDLINE search (restricted to the pre-
ceding 5 years) was performed using the
terms ‘‘periodontal disease’’ and ‘‘epi-
demiology’’ as keywords (Papapanou
1999). The conclusions from the review
reflected the wide range of different
methodologies that had been used and
further stressed the obvious lack of
uniformity in a case definition of perio-
dontitis.
One of the main conclusions of the
Borrell & Papapanou (2005) review,
with regard to the definition of perio-
dontitis, was the need to include a range
of several signs and symptoms of the
disease that may be graded to reflect
different severity levels. Some of the
signs and symptoms that can be used
were suggested in an American Academy
of Periodontology position paper (Burt
2005). This paper also reiterated that
historical indices which scored gingivitis
and periodontitis on the same scale
should be considered invalid and should
be discarded. It further suggested that a
current case definition for periodontitis
will not only need to establish how to
include probing measures and BOP but
what depth of CAL, at any one site will
constitute evidence of disease process
and how many of these affected sites
need to be present in order to clinically
establish the presence of disease.
A recent definition of periodontitis
suggested that it should be: the presence
of an inflamed pathological pocket
X4 mm deep in conjunction with attach-
ment loss (Van der Velden 2005); how-
ever, this definition does not take into
account the number of affected sites.
The Group C consensus report of the
5
th
European workshop in perio-
dontology (Tonetti & Claffey 2005)
identified the definition of periodontitis
as being one of the most important issues
that impacted upon data interpretation.
The report emphasized that attachment
loss should be the primary outcome
variable used in studies of risk factors
for periodontitis. However, it stressed
that periodontitis cannot be reflected by
measurements of only a single variable
such as attachment loss or bone loss but
required the additional measurements of
BOP and/or pocket depth.
It is evident from the periodontal
literature that there is a lack of unifor-
mity in the definition of periodontitis
used in epidemiological studies. There
are also inconsistencies in the methodol-
ogies used to obtain information from the
studies (Papapanou 1996). The implica-
tions of this are that findings from studies
using differing case definitions and dif-
fering survey methodologies are not
readily interpretable or comparable
(Gera 2000, Eaton 2002).
Against this background, the aim of
the current study was to perform a sys-
tematic review of the literature related to
the definitions of periodontitis and the
methods that have been used to identify
and measure this disease. It represents a
step towards the future validation of a
definition for periodontitis and the meth-
od(s) used to measure it.
Material and Methods
Protocol development
The review protocol detailed all the
methods, including aim, search, eligibil-
ity criteria for study inclusion, methods
for screening and data abstraction, and
data analysis.
The types of studies (and search
terms) eligible for inclusion in the
review were those that related to:
population,
screening,
prevalence,
epidemiologic surveys.
The studies also had to include:
a definition of disease,
description of level of periodontal
attachment,
a measure of periodontitis,
a definition of periodontitis.
Studies that were excluded from the
review were those limited to gingivitis
or studies where prevalence or severity
of periodontitis was not a principal out-
come. Intervention studies were also
excluded as the primary aim of the study
was epidemiological.
Search strategy
The search was performed using elec-
tronic databases for studies published up
to August 2006. These were MEDLINE,
Excerpta Medica database (EMBASE),
Science Citation Index (SCISEARCH)
and Latin American & Caribbean Health
Sciences Literature (LILACS).
An electronic search strategy was
developed for MEDLINE via OVID
and revised appropriately for each data-
base. The search also utilized a combi-
nation of controlled vocabulary and free
text terms in the following combinations:
Periodontal Diseases (MeSH term
and Keyword),
Alveolar Bone Loss (MeSH term
and Keyword),
Periodontal Attachment Loss
(MeSH term and Keyword),
Periodontitis (MeSH term and Key-
word),
Periodontal Pocket (MeSH term and
Keyword),
Epithelial Attachment (MeSH term
and Keyword),
Clinical Attachment Loss (Keyword),
Lifetime Cumulative Attachment
Loss (Keyword).
The total number of citations generated
was then combined using the Boolean
variable and with the sum total generated
by the following combinations:
Prevalence (MeSH term),
Epidemiologic Studies (MeSH term),
Cohort Studies (MeSH term),
Epidemiology (MeSH term),
Epidemiologic Research Design
(MeSH term),
Risk (MeSH term).
The citations generated were limited
to human studies and although there
were no specific language restrictions,
only articles for which the reviewers
were able to obtain translations were
included in the review.
Systematic review definitions periodontitis 459
r2009 John Wiley & Sons A/S
All titles and abstracts of publications
were scanned independently by two
reviewers (A. S. and K. E.). Full text
papers were obtained for publications
that appeared to meet the inclusion
criteria, or for which there was insuffi-
cient information in the title and abstract
to make a clear decision. They were
assessed independently, in duplicate,
by the two reviewers to establish
whether the publications met the
required inclusion criteria or not. All
disagreements were resolved by discus-
sion and where necessary, a third
reviewer was consulted.
The references of all publications,
selected for inclusion, were checked
for other additional relevant studies.
The authors tried to identify any unpub-
lished studies by searching abstracts and
conference proceedings. In addition,
reference lists of review papers and
text book chapters were examined to
try to identify additional potentially
relevant studies.
Methodological quality assessment
The methodological quality of the
reviewed studies was assessed by exam-
ining criteria in relation to the studies
and grouping them into generic factors
and factors specific to periodontitis,
respectively.
The main criteria that comprised the
two groups were as follows:
Generic factors
Random selection/representative sam-
ple.
Examiner calibration or training.
Factors specific to periodontitis
Definition of periodontitis used.
Measure and/or index used.
Measurement tools/probe type.
These specific criteria were selected to
assess the potential for bias in the
reviewed studies particularly selection
bias in the case of generic factors and
measurement bias for factors that were
specific to periodontitis.
The data were synthesized qualita-
tively following the production of evi-
dence tables.
Results
Outcome of the search
The search terms yielded 3472 publica-
tions identified through electronic
searches, reference lists and other sources
that were available for the review (Fig. 1).
By using the exclusion criteria, the two
independent reviewers excluded 3368
articles as not relevant to the review
from the review leaving a total of 104
potentially relevant articles that were
chosen for retrieval and evaluation of
thefulltextusingadataextractionsheet.
The authors were unable to obtain trans-
lators for 15 non-English language pub-
lications, at the initial yield, to identify if
they met the criteria for inclusion in the
review. The kscore for agreement
between the reviewers was 0.86 indicat-
ing an excellent level of agreement. Out
of the 104 full text articles retrieved, 89
articles were excluded because they did
not meet the inclusion criteria leaving a
total of 15 relevant articles that attempted
to offer a definition of periodontitis and
measure(s) used to identify it.
Geographic distribution
From Table 1 it can be seen that the
majority (12) of the 15 studies were
undertaken during the past 20 years,
with five studies from the United States,
two Swedish studies and two Chinese
studies. The other six studies were
from New Zealand, Indonesia, Greece,
Argentina, the United Kingdom and
Taiwan.
Sample size and study characteristics
The sample sizes of included studies
ranged from 185 participants (Craig
et al. 2001) to 15,302 participants (Borrell
et al. 2005) (Table 1). This latter study
combined participants from the National
Health and Nutrition Examination Sur-
vey III (NHANES III) and the NHANES
1999–2000 studies. The study of Borrell
et al. (2005) was limited to the records
of adults 18 years or older who self-
ascribed their ethnicity as non-Hispanic
black, non-Hispanic white, or Mexican-
American, and who had a complete
periodontal examination. Four of the
studies were longitudinal. Most of the
studies were cross-sectional of a general
population that encompassed a wide age
range. More specific population groups
were examined by Agerholm & Ashley
(1996), Chiappe et al. (1997), Holmgren
et al. (1994) and Querna et al. (1994),
Inclusion criteria and
search strategy
Search yield including titles
+/– abstracts n = 3472
Potentially relevant publications. n = 104
Ineligible publications
after screening all titles
and abstracts. n = 3368
Excluded ineligible
studies after detailed
assessment of full
text.
n = 89
Included studies in
review n = 15
Fig. 1. Flow summary of the systematic review.
460 Savage et al.
r2009 John Wiley & Sons A/S
who looked at factory workers, dental
students, middle aged/elderly people
and military populations respectively.
With regard to identifying adverse
habits, such as smoking and systemic
conditions, six of the studies identified
smoking status while participants with
medical conditions were excluded from
three studies. One study specifically
excluded participants with gestational
(pregnancy) diabetes.
A summary of the examination
methods used in the studies is shown
in Table 2. Seven of the studies were
described as full-mouth assessments.
Three of the studies reportedly excluded
measurements from the third molar
sites. Two studies were classified as
half-mouth studies, while four assessed
only index teeth.
The study by Norderyd & Hugoson
(1998) was a radiographic study which
assessed inter-proximal bone loss. In
this study six bite-wing radiographs
and a dental pantomogram were
assessed for each patient in the 20–30-
year age group and full-mouth intra-oral
radiographs were assessed for each
patient in the 40–80-year-old age group.
Laurell et al. (2003) utilized full-mouth
intra-oral radiographs to assess inter-
proximal bone height during the exam-
ination and based its definition of
periodontitis on this criterion (Table 2).
The most common measurement tool,
used in other studies, was a periodontal
probe. However, the type of probe var-
ied from study to study. Specific types
of probe were identified in 12 of the 15
studies highlighted in Table 2. The force
of probing was identified in three of
the studies. Holmgren et al. (1994)
noted using light probing force consis-
tent with the CPI probe. Mechanical
force controlled probes were used in
two of the studies. Machtei et al.
(1992) utilized a computerized probe
with a 20 gforce while Timmerman
et al. (1998) used a force-controlled probe
(Brodontic
s
Ash/Dentsply 240 N/cm
3
,
BP, Prima, Byfleet, UK) to measure
probing depth, supplemented with a
Williams marking Hu–Friedy
s
(IL,
Chicago, USA) probe to determine
attachment loss.
The number of sites probed and loca-
tions also varied between the studies.
Six studies recorded six-point probing
while three studies utilized four points
per tooth. In one study the specific
probing location points were not
recorded (Laurell et al. 2003). A further
study, Borrell et al. (2005) utilized
mid and mesio-buccal probing sites.
Another study, Timmerman et al.
(1998) used approximal surfaces from
the vestibular aspect of all teeth except
molars, as well as on the vestibular and
lingual surfaces of the Ramfjord teeth
(16, 21, 24, 36, 41 and 44). Chiappe
et al. (1997) and Wang et al. (1987) also
probed the Ramfjord teeth. In four
studies the location of probing was not
clearly described.
Pocket probing depth (PPD) was
recorded in all studies except in the
one by Norderyd & Hugoson (1998)
where the measurements were radio-
graphic. The clinical attachment loss
was recorded in six studies used the
terms periodontal attachment level and
CAL respectively.
Quality appraisal of the studies
The 15 studies revealed a considerable
degree of heterogeneity between the
sample selection processes, and the
recording of these procedures (Table 3).
Five of the studies went through a
random selection process, when select-
ing their study samples. In eight studies,
the level of randomization in the sample
selection was not clear. The samples in
three studies were drawn from specific
sampling frames such as dental students,
factory workers and the military. In
terms of examiner calibration and train-
ing, only two studies, reported examiner
calibration in the methodology used for
data collection. It was not reported in
the other three studies what level of
Table 1. Sample characteristics of included studies listed chronologically
Reference Year of
publication
Country Sample frame Sample
size
Study
design
Age range
(years)
Smoking
status
Medical conditions
Borrell et al. (2005) 2005 United States General population 15,302 CS 181I Diabetes in pregnant
women excluded
Laurell et al. (2003) 2003 Sweden General population 357 L 20–60 I U
Craig et al. (2001) 2001 United States General population 185 CS 19–70 I E
Paidi et al. (1999) 1999 New Zealand General population 240 CS 30–70 I E
Norderyd & Hugoson
(1998)
1998 Sweden University/hospital 547 CS 20–70 I NE
Timmerman et al.
(1998)
1998 Indonesia Untreated general
population
255 L 15–25 NR NE
Chiappe et al. (1997) 1997 Argentina University Dental
Students
475 L 17–25 NR NE
Anagnou-Vareltzides
et al. (1996)
1996 Greece General population 339 CS 25–64 NR NE
Agerholm & Ashley
(1996)
1996 United
Kingdom
Electronic factory
workers
202 L 20–40 NR NE
Holmgren et al. (1994) 1994 China (Hong
Kong)
Middle aged and
elderly
855 CS 35–44 NR NE
65–74
Querna et al. (1994) 1994 United States Military 1334 CS 18–451NR NE
Machtei et al. (1992) 1992 United States University/hospital
referral
and general population
508 CS 25–73 NR NE
Peng et al. (1990) 1990 Taiwan General population 673 CS 15–641NR NE
Brown et al. (1989) 1989 United States General population 1792 CS 19–651NR NE
Wang et al. (1987) 1987 China General population 2284 CS 18–50 NR NE
CS, cross-sectional study; I, identified; L, longitudinal study; NE, not excluded; NR, not recorded; U, unclear; E, excluded.
Systematic review definitions periodontitis 461
r2009 John Wiley & Sons A/S
calibration (if any) was undertaken.
Timmerman et al. (1998) undertook an
examiner calibrated study in an
untreated community in Indonesia.
Definition of periodontitis and/or indices
used
Most of the studies listed in Table 3 also
gave a definition of periodontitis with
thresholds using PPD and CAL. Laurell
et al. (2003) used radiographs and
defined periodontitis as bone loss set at
10% of root length which corresponded
to at least 2–3 mm of CAL.
Paidi et al. (1999) did not state a clear
threshold to define periodontitis while
Norderyd & Hugoson (1998), predomi-
nantly used radiographs, and criteria
from Hugoson & Jordan (1982). How-
ever, although the authors defined perio-
dontitis using criteria from Hugoson &
Jordan (1982), the term ‘‘majority of
teeth’’ in their definition was not clearly
presented in the paper. Chiappe et al.
(1997) defined periodontitis as the sub-
ject having at least one site with loss of
attachment X2 mm.
Wang et al. (1987) used the Ramfjord
Periodontal Disease Index 1959 to define
periodontitis and the criterion selected
was 4 which relates to measurements on
the Ramfjord teeth (16, 21, 24, 34, 31 and
46). The score of 4 is given if gingival
crevice is 43 mm api cal to CEJ, score of
5if3–6mmand6if46mm.
Numerical range of CAL and PPD
thresholds used in the studies
The review illustrated further heteroge-
neity when the individual thresholds
were examined for CAL and PPD as
can be seen from Table 4. When using
PPD, as a threshold for periodontitis, the
minimum PPD defining periodontitis at
an individual or multiple site was a
threshold of X3 mm as seen in (Borrell
et al. 2005 – two sites, Craig et al. 2001
– four sites and Peng et al. 1990 – worst
individual score at any given sextant).
Peng et al. (1990) used 3.5 mm that was
the CPITN threshold. Brown et al.
(1989) using the modified Russell index
defined a threshold for periodontitis at a
single site X4 mm. This threshold value
was shared by Querna et al. (1994) who
also defined early periodontitis based on
the worst individual score at a sextant
occurring at that value. At X5mm
Querna et al. (1994) further defined
moderate to advanced periodontitis.
This threshold was also used by Machtei
et al. (1992) and Peng et al. (1990) as a
definition of periodontitis for their
Table 2. Examination methods used in the studies
Reference Examination area Measurement tool Probing location
(MB, B, DB, ML, L and DL)
PPD CAL
Borrell et al. (2005) Half mouth Probe (type unclear) Mid and MB probing of
randomly assigned quadrants
RR
Laurell et al. (2003) Full mouth Probe (type unclear)
Full-mouth radiographs measuring
inter-proximal bone height
Four-point probing (location NR) R NR
Craig et al. (2001) Full-mouth (third
molars excluded)
North Carolina periodontal probe Six-point probing MB, B DB, DL, L
and ML
RR
Paidi et al. (1999) Half mouth Williams marked probe Six-point probing MB, B DB, DL, L
and ML of randomly assigned
contra-lateral diagonal quadrants
RR
Norderyd & Hugoson
(1998)
Full mouth
(radiographic)
Six Bitewings and one DPT
(20–30 years olds)
Full-mouth intra-oral radiographs
(40–80-year olds)
Location NR NR NR
Timmerman
et al. (1998)
Part mouth Probing depth – force-controlled probe
(Borodontic
s
Ash/Dentsply 240 N/cm
3
Attachment loss – Hu-Friedy
s
probe
(Williams calibration)
MB, DB surfaces of all teeth except
molars, as well as on vestibular and
lingual surfaces of the Ramfjord teeth
(16, 21, 24, 36, 41 and 44)
Specific location NR
RR
Chiappe et al. (1997) Selected index
teeth
Conventional probe (CP12 probe) Four-point probing MB, DB, DL
and ML
RR
Anagnou-Vareltzides
et al. (1996)
Full mouth Calibrated probe with tip diameter of
0.45 mm
Six-point probing MB, B DB, DL, L
and ML
RR
Agerholm & Ashley
(1996)
Full-mouth (third
molars excluded)
PQW periodontal probe with Williams’s
markings (Hu-Friedy
s
)
Four-point probing MB, DB, DL
and ML
RR
Holmgren et al. (1994) Full mouth Community periodontal index – C
periodontal probe using light probing
force consistent with CPI probe
Six-point probing MB, B DB, DL, L
and ML
RR
Querna et al. (1994) Selected index
teeth
Glickman 26-G periodontal probe with
round ended 0.5-mm diameter tip
Six-point probing MB, B DB, DL, L
and ML
RNR
Machtei et al. (1992) Full-mouth (third
molars excluded)
Computerized probe with standardized
20 gforce
Six-point probing MB, B DB, DL, L
and ML
RR
Peng et al. (1990) Full mouth World Health Organization probe Location NR R NR
Brown et al. (1989) Selected index
teeth
Hu-Friedy
s
CP6 round probe with
0.48 mm diameter
Location NR R NR
Wang et al. (1987) Selected index
teeth
Glickman periodontal probe Probing on Ramfjord teeth (16, 11, 24, 36,
41 and 44) Specific location NR
RNR
PPD, pocket probing depth; MB, mesiobuccal; L, mid lingual; CAL, clinical attachment loss; B, mid buccal; DL, distolingual; R, recorded; DB,
distobuccal; NR, not recorded; ML, mesiolingual.
462 Savage et al.
r2009 John Wiley & Sons A/S
Table 3. Sample selection, periodontitis definition and tool used in the studies
Reference Sample selection Examiner
calibration
Definition of periodontitis and/or index
Borrell et al. (2005) Random and representative of
general population
U Threshold – a person who had at least three sites with clinical
attachment loss X4 mm and at least two sites with
PPDX3 mm. However, these conditions did not have to be
present in the same site or tooth
Laurell et al. (2003) Random and representative of
general population
U Threshold – bone loss was set at 10% of the tooth length,
which corresponded to a bone loss of at least 2–3 mm
Craig et al. (2001) Does not appear to be random
and representative of
general population
R Threshold – a periodontal diseased subject was defined as
having at least 20 teeth and at least four sites with pocket
depths 43 mm and at least four sites with attachment loss
43 mm (based on radiographic inter-proximal loss)
Paidi et al. (1999) Does not appear to be random
and representative of
general population
R Threshold – use of three terms. (a) Prevalence of LOA: as the
percentage of LOA of xmm (where x52, 4, 6 and 9 mm).
(b) Extent of LOA: as the mean percentage of sites with LOA
of xmm or more per person. (c) Severity: LOA of xmm or
more per person. Threshold defining periodontitis – NR
Norderyd & Hugoson
(1998)
Does not appear to be random
and representative of
general population
R Threshold – criteria from Hugoson & Jordan (1982). Group 3
– alveolar bone loss around the majority of the teeth not
exceeding 1/3 of the length of roots, Group 4 – alveolar bone
loss around the majority of the teeth ranging between 1/3 and
2/3 of the length of the roots. Group 5 – alveolar bone loss
around the majority of the teeth exceeding 2/3 of the length
of the roots; presence of angular bony defects and/or
furcation defects
Timmerman et al. (1998) Representative study
of community
R Threshold – criteria from Brown et al. (1990). No or minor
periodontitis – 0–2 mm maximum attachment loss, moderate
periodontitis – 3–4 mm maximum attachment loss, advanced
periodontitis X5 mm maximum attachment loss
Chiappe et al. (1997) Does not appear to be random
and representative of
general population
R Threshold – loss of attachment was determined when the
clinical attachment loss was X2mm
Anagnou-Vareltzides
et al. (1996)
Random and representative of
general population
U Threshold – level of X6 mm for pocket probing depth and
periodontal attachment level was utilized as expressing deep
pocketing and advanced attachment losses as an arbitrary
definition of severe periodontal disease
Agerholm & Ashley
(1996)
Does not appear to be random
and representative of
general population
U Threshold – clinical attachment loss at 2, 3 or 4mm. Subsets
made up of 10 index teeth recommended by the WHO for
partial recording (two molars in each quadrant and maxillary
right and mandibular left central incisors) and a subset
comprising maxillary buccal and mandibular lingual sites
(‘‘Pritchard sites’’)
Holmgren et al. (1994) Does not appear to be random
and representative of
general population
R Threshold – definition based on X6 and 9 mm loss of
attachment
Querna et al. (1994) Does not appear to be random a
nd representative of
general population
NR Threshold – subjects with inflammation and PPD of over
3 mm but o5 mm were categorized as early periodontitis
Over 5 mm moderate to advanced periodontitis periodontal
screening exam (PSE) index
Machtei et al. (1992) Does not appear to be random
and representative of
general population. ‘‘Convenience’’
sample also used
R Threshold – the clinical entity of established periodontitis is
suggested based on the presence of clinical attachment level
X6 mm in two or more teeth and one or more sites with
Pocket probing depth X5mm
Peng et al. (1990) Random and representative of
general population
R Threshold – periodontitis included teeth with pocketing and
attachment loss – CPITN 3 – pocket depth between 3.5 and
5.5 mm, CPITN 4 – pocket depth 45.5 mm
Definition made according to the modified method of
Gaengler (1984)
Brown et al. (1989) Random and representative of
general population
R Threshold – periodontitis (pockets X4 mm)
Advanced periodontitis (pockets X6 mm)
Definition based on modified Russell’s periodontal index
Wang et al. (1987) Not random and representative of
general population
NR Threshold – periodontal disease index above 3 was
considered to suffer from periodontitis
Definition based on Ramfjord periodontal disease index 1959
R, recorded; NR, not recorded; U, unclear; PPD, pocket probing depth; CAL, clinical attachment loss; LOA, loss of attachment; CPITN, community
periodontal index of treatment need.
Systematic review definitions periodontitis 463
r2009 John Wiley & Sons A/S
study. However, Peng et al. (1990) used
5.5 mm corresponding to the CPITN
value to register this threshold value.
Severe and advanced periodontitis was
defined by Anagnou-Vareltzides et al.
(1996) at mean site and Brown et al.
(1989) at an individual site respectively
at X6 mm. Brown et al. (1989) used the
modified Russell index to define this
threshold value corresponding to
X6 mm.
The use of CAL at an individual site
provided an even greater spectrum in the
results. Timmerman et al. (1998) used
42 mm as the threshold of no to minor
periodontitis which was the same
threshold as 2 mm used by Chiappe et
al. (1997). Three different threshold
values of X,X4 and X6 mm were
used by Agerholm & Ashley (1996) in
order to determine which individuals in
their sample had the worst periodontitis
based on CAL at different diagnostic
thresholds. Timmerman et al. (1998)
defined moderate periodontitis at CAL
of X3 mm. This threshold value was
shared by Craig et al. (2001) based on
four sites. Borrell et al. (2005) utilized a
threshold of X4 mm at three sites.
Advanced periodontitis was defined by
Timmerman et al. (1998) using a thresh-
old value of X5 mm while Anagnou-
Vareltzides et al. (1996) defined severe
periodontitis with a mean threshold
value of X6 mm which was the same
as Holmgren et al. (1994). Machtei
et al. (1992) also used a threshold
value of X6 mm that was based on
two sites.
Discussion
This systematic review investigated the
literature related to definitions of perio-
dontitis and methods that have been
used to identify and measure this dis-
ease. The reason for selecting this meth-
odology was to find and appraise the
totality of the evidence, in this case,
which definitions of periodontitis had
been used. The reason for excluding
many records was that the search was
deliberately made to be highly sensitive
i.e. a high probability of finding poten-
tially eligible studies, but with the risk
of finding many ineligible papers. All
records were screened in order to find
the eligible ones. This is standard pro-
cedure where indexing of potential stu-
dies is not efficient and therefore the
search is made to be inclusive to avoid
missing relevant papers. Therefore, the
benefits of using a systematic approach
are important and they are, searching
and retrieval of totality of evidence,
objective critical appraisal of the evi-
dence and structured synthesis leading
to minimization of bias.
Therefore this systematic review
documented that the reviewed studies
utilized a minimum diagnostic threshold
defining periodontitis, at a given site in
terms of CAL of 2 mm and PPD of
3 mm. The review has additionally con-
firmed previous findings of a consider-
able lack of methodological consistency
in periodontal epidemiology.
Only 15 of the 104 publications,
identified through the search, gave a
quantitative definition of periodontitis
and it was noted that smoking or med-
ical conditions were not clearly identi-
fied in most of the reviewed articles.
Furthermore, the methods of sample
selection and examiner calibration
were also not clearly highlighted in all
the studies. More specific findings
revealed heterogeneity between the stu-
dies in the measurement tools, particu-
larly the types of probes used.
Differences also existed in the areas
surveyed (full/part mouth), indices
used and location of probing between
the various studies. It was further noted
that the use of CAL and/or PPD as
single, multiple or mean scores at any
given site varied between the studies
resulting in heterogeneity in the thresh-
old(s) by which periodontitis was
defined.
A number of specific weaknesses
were apparent. The first weakness was
that smoking and/or medical conditions
were not clearly identified in most of the
papers that were reviewed. This is
important because smoking and certain
medical conditions are associated with
increased bone loss resulting in greater
PPDs and clinical attachment loss which
Table 4. Numerical ranges of CAL and PPD thresholds used in the studies
42mm X2mm X3mm X4mm X5mm X6mm
CAL study
Timmerman
et al.
(1998) (no
or minor
periodontitis)
at one or
more sites
Agerholm &
Ashley (1996)
at one or two
approximal sites
Chiappe et al.
(1997) at one site
Craig et al. (2001)
at four sites
Timmerman
et al. (1998)
(moderate
periodontitis)
at one or more sites
Agerholm & Ashley
(1996) at one or two
approximal sites
Borrell et al. (2005)
at three sites
Agerholm & Ashley
(1996)
at one or two
approximal sites
Timmerman et al.
(1998) (advanced
periodontitis)
at one or more sites
Anagnou-Vareltzides
et al. (1996) (severe
periodontitis) at
mean sites
Holmgren et al. (1994)
at mean sites
Machtei et al. (1992)
at two or more teeth
PPD study
Borrell et al. (2005)
at two sites
Craig et al. (2001)
at four sites
Peng et al. (1990)
(3.5 mm CPITN)
worst individual
score at sextant
Querna et al. (1994)
(early) worst individual
score at sextant.
Brown et al. (1989)
(modified Russell index)
at one site
Querna et al. (1994)
(moderate – advanced)
at the highest score
per sextant
Machtei et al. (1992)
at one or more teeth
Peng et al. (1990)
(5.5 mm CPITN) worst
individual score at sextant
Anagnou-Vareltzides
et al. (1996)
(severe periodontitis)
at mean sites
Brown et al. (1989)
advanced periodontitis
(modified Russell
index) at one site
CAL, clinical attachment loss; CPITN, community periodontal index of treatment need; PPD, pocket probing depth.
464 Savage et al.
r2009 John Wiley & Sons A/S
affects disease severity. Given that
smoking (Haber et al. 1993) and sys-
temic diseases such as diabetes are well
known risk factors for periodontitis
(Genco 1996), and that the search term
RISK encompassed the studies with
these factors, the small number of
papers that were found was perhaps
surprising. It may be that because the
majority of the studies reported were
performed in the early 1990s and during
that period the primary outcomes of
these epidemiological surveys were not
generally smoking and systemic health;
this aspect may not have been incorpo-
rated into the protocols for the studies.
The second major weakness was that
methods of sample selection were not
clearly defined in many of the studies
that were included in the review. In a
simple prevalence study, an epidemiol-
ogist merely needs to determine if a
disease or condition is present or not.
If the sample in question is small the
epidemiologist may examine all the
people. With a larger sample frame,
the epidemiologist will need to examine
a representative sample. Examiner cali-
bration was also not clearly defined and
should ensure repeatability and consis-
tency of all the measurements. In addi-
tion, clearly defined and reported
methods of calibration allow not only
for reproducible results but also for the
results of the studies to be analysed and
compared with other studies.
Perhaps unsurprisingly, the third
weakness revealed by the review was
heterogeneity in the measurement tools
used, particularly the types of probes
used in the 15 studies. It has been
suggested that ideally probe tips need
to have a diameter of 0.6 mm and a
probing force of 0.2 g. is required to
reach the base of the pocket (Garnick
et al. 2000). The shape of the probe tip
(round, parallel sided and tapered) can
contribute to significant differences.
There is a tendency for the parallel-
sided tine to yield deeper readings in
some cases (Atassi et al. 1992). Differ-
ent probe designs therefore give slightly
different readings which can have a
significant influence on the results. Dis-
ease definitions are based on a single
clinical sign (e.g. deep pockets or
attachment loss measurements) or on a
combination of different signs and
symptoms which can be influenced by
the performance properties of the probe.
Most of the cross-sectional and long-
itudinal epidemiological studies in this
review used different diagnostic tools
and criteria. Consequently, such find-
ings are difficult to compare (Gera
2000). The measurement of periodontal
attachment loss is a valuable measure of
the severity of periodontal tissue loss,
and may be indicative of a previous
destructive disease process, as such it
is a historic measure of the extent and
severity of past disease activity (Eaton
et al. 2001, Susin et al. 2005).
The conversion of the junctional
epithelium to pocket epithelium has
been regarded as a hallmark in the
progression of gingivitis to periodontitis
(Bosshardt & Lang 2005). This break-
down in attachment is usually measured
by probing. However, the ability to
detect change due to breakdown may
vary. The probe tine diameter and cali-
bration should also therefore be consid-
ered in addition to other variables of
periodontal probing (Van der Zee et al.
1991). The position of the probe tip may
be affected by the probing force. Prob-
ing force can be achieved either by a
manual or electronic pressure sensitive
probe. Some of the studies that were
reviewed such as those of Holmgren et
al. (1994), Machtei et al. (1992) and
Timmerman et al. (1998) made an
attempt to control this by the use of a
consistent force probe. Other difficulties
associated with probing include angula-
tion of the probe, shape of the tooth,
subgingival deposits and cooperation of
the patient Haffajee et al. (1983).
Despite these shortcomings in the
attachment level measurement, at pre-
sent this technique appears to provide an
indication and estimate of periodontal
breakdown.
Methods aimed at improving the
reliability of recordings, such as use of
calibrated instruments (Philstrom 1992),
stents (Clark et al. 1987) and electronic
probes. (Gibbs et al. 1988) have been
reported in the literature (Alves et al.
2005). The use of periodontal probes is
further supported by an expert opinion
position paper prepared by the research,
science and therapy committee of the
American Academy of Periodontology
(Academy Report 2003), which sup-
ported the view that probing depth and
clinical attachment loss measurements
obtained with periodontal probes are
practical and valid methods for asses-
sing periodontal status. A universal
periodontal probe would not only make
studies and surveys more comparable
but would also allow for better and
more accurate statistical analysis of
any data generated in the future by
reducing measurement bias. This goal
of uniformity may however be difficult
to achieve given the different probes
commercially available and individual
operator preference.
Differences existed in the area of the
mouth surveyed (full/part), indices used
and location of probing between the
various studies. Eaton (2002) stated
that it was perhaps unsurprising, as
concepts of the nature of chronic inflam-
matory periodontal diseases have chan-
ged over the years and many methods
and indices have been developed and
used to assess the disease, both clini-
cally and epidemiologically.
Along with the various indices the
other factor that needs to be taken into
account is the range and scope of the
survey (i.e. full mouth or part mouth).
Part-mouth assessments (Wang et al.
1987, Brown et al. 1989, Querna et al.
1994, Timmerman et al. 1998, Paidi
et al. 1999, Laurell et al. 2003, Borrell
et al. 2005) have the distinct advantages
of being quick and can be duplicated in
large populations. Nevertheless, they do
have the potential to underestimate the
prevalence of periodontal breakdown in
populations with less susceptibility
(Beck & Lo
¨e 1993) or overestimate the
prevalence when the particular teeth
selected are first molars and lower inci-
sors (Carlos et al. 1986).
Full-mouth assessments (Peng et al.
1990, Agerholm & Ashley 1996, Ana-
gnou-Vareltzides et al. 1996, Laurell
et al. 2003) provide the optimal exam-
ination of periodontal conditions. How-
ever, important priorities when
examining periodontal status of subjects
in population studies include short
examination time and avoidance of sub-
ject discomfort, in order to maximise
numbers people examined in the time
available and to encourage subject com-
pliance (Agerholm & Ashley 1996).
Although it is desirable to monitor as
many sites as possible to increase the
probability of detecting disease preva-
lence, one of the main drawbacks of
full-mouth assessments is that it can be
time consuming. Therefore if part-
mouth assessments are performed a cor-
rection factor should be calculated by
performing full-mouth assessments on a
certain percentage of subjects and com-
paring the results with those obtained
from part-mouth assessments.
In the studies reviewed, the use of
CAL and/or PPD at individual, multiple
or mean site(s) in the assessment
of periodontal breakdown varied. Al-
Systematic review definitions periodontitis 465
r2009 John Wiley & Sons A/S
though a commonly observed sign of
periodontal disease is an increased
pocket depth, this may occur due to
conditions other than periodontitis,
such as delayed passive eruption and
inflammatory or fibrous gingival enlar-
gement. Deep pockets, therefore, may
not be diagnostic for either gingivitis or
periodontitis. Indeed, periodontitis may
develop and progress without significant
increase in pocket depth if gingival
recession occurs (Jenkins et al. 2001).
Agerholm & Ashley (1996) report that
in their population if screening had been
based on PPDX6 mm, as an approxima-
tion for CALX4 mm, then the test
would have overestimated those aff-
ected in the youngest two age groups
(aged 20–30 years) and underestimated
those affected in the group aged 36–40
years. This cast doubts on conclusions
drawn from surveys where reliance has
been placed on PPD measurement or
categorization.
Collectively, the findings of this
review indicate a strong need for future
uniformity in the design of epidemiolo-
gical periodontal studies. Such studies
need to be designed to allow the result-
ing data to be compared with those
obtained from other studies. The first
step in this process, as mentioned in
previous reviews, should be the use of
a uniformly agreed measure and a mea-
suring tool that clearly defines the dis-
ease threshold and the surveyed area
(full mouth/part mouth). This may be
difficult to achieve due to the wide
variation of probes available and indivi-
dual operator preference. If part-mouth
assessments are made then a correction
factor should be calculated to account
for possible differences between part-
mouth and full-mouth assessments.
Further more there should be a require-
ment for training/calibration of the
examiners and the nature of that training
to be clearly reported in the study.
Finally, in order to truly reflect disease
activity, and reiterating previous reviews
on the subject, the combined use of CAL
and PPD should be considered as the two
key variables to be assessed in future
epidemiological studies.
Conclusions
A number of conclusions can be drawn
from this systematic review.
The first is that this review has con-
firmed the view that epidemiological
studies of periodontal diseases are com-
plicated by the diversity of definitions
and measurements used to describe and
quantify these diseases. The second is
the lack of consensus as to a uniform
definition and classification of perio-
dontal disease (Kingman & Albandar
2002). The third is a further confirma-
tion of the heterogeneity between all the
studies and the indices that were used
and the areas of the mouth surveyed,
making direct comparisons and specific
answers to the focused questions diffi-
cult. The fourth point relates to the
methodological inconsistency, particu-
larly in terms of clearly identifying the
methods used in selecting and generat-
ing the sample to ensure it is represen-
tative of the population studied and
clearly recording the level of examiner
training and calibration received.
The fifth conclusion which signifi-
cantly adds to the existing evidence
base in defining periodontitis is that
the studies reviewed utilized a minimum
diagnostic threshold, at a given site in
terms of CAL of 2 mm and PPD of
3 mm.
References
Academy Report. (2003) Diagnosis of Perio-
dontal Diseases. Journal of Periodontology
74, 237–1247.
Agerholm, D. M. & Ashley, F. P. (1996)
Clinical assessment of periodontitis in young
adults – evaluation of probing depth and
partial recording methods. Community Den-
tistry and Oral Epidemiology 24, 56–61.
Alves, R. V., Machion, L., Andia, D. C., Casati,
M. Z., Sallum, A. W. & Sallum, E. A. (2005)
Reproducibility of Clinical Attachment Level
and Probing Depth of a Manual Probe and a
Computerised Electronic Probe. Journal of
the International Academy of Periodontology
7, 27–30.
Anagnou-Vareltzides, A., Diamanti-Kipioti, A.,
Afentoulidis, N., Moraitaki-Tsami, A.,
Lindhe, J., Mitsis, F. & Papapanou, P. N.
(1996) A clinical survey of periodontal con-
ditions in Greece. Journal of Clinical Perio-
dontology 23, 758–763.
Armitage, G. C. (1995) Clinical evaluation of
periodontal diseases. Periodontology 2000 7,
39–53.
Atassi, F., Newman, H. N. & Bulman, J. S.
(1992) Probe tine diameter and probing
depth. Journal of Clinical Periodontology 5,
301–304.
Beck, J. D. & Lo
¨e, H. (1993) Epidemiological
principles in studying periodontal diseases.
Periodontolgy 2000 2, 34–45.
Borrell, L. N., Burt, B. A. & Taylor, G. W.
(2005) Prevalence and trends in periodontitis
in the USA: from the NHANES III to the
NHANES, 1988 to 2000. Journal of Dental
Research 84, 924–930.
Borrell, L. N. & Papapanou, P. N. (2005)
Analytical epidemiology of periodontitis.
Journal of Clinical Periodontology 32
(Suppl. 6), 132–158.
Bosshardt, D. D. & Lang, N. P. (2005) The
junctional epithelium: from health to disease.
Journal of Dental Research 84, 9–20.
Brown, L. J., Oliver, R. C. & Loe, H. (1989)
Periodontal diseases in the US in 1981:
prevalence, severity, extent, and role in tooth
mortality. Journal of Periodontology 60,
363–370.
Brown, L. J., Oliver, R. C. & Loe, H. (1990)
Evaluating perodontal status of US employed
adults. Journal of the American Dental Asso-
ciation 121(2), 226–232.
Burt, B. (2005) Position paper: epidemiology of
periodontal diseases. Journal of Perio-
dontology 76, 1406–1419.
Carlos, J. P., Wolfe, M. D. & Kingman, A.
(1986) The extent and severity index: a
simple method for use in epidemiologic stu-
dies of periodontal disease. Journal of Clin-
ical Periodontology 13, 500–505.
Chiappe, V., Gomez, M., Pedreira, P., Galeano,
A., Grinfeld, A., Viale, J. & Sznajder, N.
(1997) Longitudinal study of periodontal
condition in students of the Dental School
the University of Buenos Aires Argentina.
Acta Odontologica Latinoamericana 10,
117–132.
Clark, D. C., Chin Quee, T., Bergeron, M. J.,
Chan, E. C., Lautar-Lemay, C. & de Gruchy,
K. (1987) Reliability of attachment level
measurement using the cementoenamel junc-
tion and a plastic stent. Journal of Perio-
dontology 58, 115–118.
Craig, R. G., Boylan, R., Yip, J., Bamgboye, P.,
Koutsoukos, J., Mijares, D., Ferrer, J., Imam,
M., Socransky, S. S. & Haffajee, A. D. (2001)
Prevalence and risk indicators for destructive
periodontal diseases in 3 urban American
minority populations. Journal of Clinical
Periodontology 28, 524–535.
Eaton, K. A. (2002) Factors affecting commu-
nity oral health care needs and provision.
PhD Thesis, University of London.
Eaton, K. A., Duffy, S., Griffiths, G. S.,
Gilthorpe, M. S. & Johnson, N. W. (2001)
The influence of partial and full-mouth
recordings on estimates of prevalence and
extent of lifetime cumulative attachment loss:
a study in a population of young male
military recruits. Journal of Periodontology
72, 140–145.
Gaengler, P. (1984) Prevalence and distribution
of gingivitis, periodontitis and missing teeth
in adolescents and adults according to GPM/
T index. Community Dent Oral Epidemiol 12,
255–259.
Garnick, J. J. & Silverstein, L. (2000) Perio-
dontal probing: probe tip diameter. Journal of
Periodontology 71, 96–103.
Genco, R. J. (1996) Current view of risk factors
for periodontal disease. Journal of Perio-
dontology 67, 1041–1049.
Gera, I. (2000) Periodontal treatment needs in
Central and Eastern Europe. Journal of the
466 Savage et al.
r2009 John Wiley & Sons A/S
International Academy of Periodontology 2,
120–128.
Gibbs, C. H., Hirschfeld, J. W., Lee, J. G., Low,
S. B., Magnusson, Thousand, R. R., Yemeni,
P. & Clark, W. B. (1988) Description and
clinical evaluation of a new computerized
periodontal probe - the Florida Probe. Jour-
nal of Clinical Periodontology 15, 137–144.
Goodson, J. M. (1992) Diagnosis of perio-
dontitis by physical measurement: interpreta-
tion from episodic disease hypothesis.
Journal of Periodontology 63, 373–382.
Haber, J., Wattles, J., Crowley, M., Mandell, R.,
Joshipura, K. & Kent, R. L. (1993) Evidence
for cigarette smoking as a major risk factor
for periodontitis. Journal of Periodontology
64, 16–23.
Haffajee, A., Socransky, S. & Goodson, M.
(1983) Comparison of different data analyses
for detecting changes in attachment level.
Journal of Clinical Periodontology 10, 298–
310.
Holmgren, C. J., Corbet, E. F. & Lim, L. P.
(1994) Periodontal conditions among the
middle-aged and the elderly in Hong Kong.
Community Dentistry and Oral Epidemiology
22, 396–402.
Hugoson, A. & Jordan, T. (1982) Frequency
distribution of individuals aged 20–70 years
according to severity of periodontal disease.
Community Dental Oral Epidemiology 10,
187–192.
Jenkins, N.M. & Papapanou, P.N. (2001) Perio-
dontology 2000 26, 16–32.
Kingman, A. & Albandar, J. (2002) Methodo-
logical aspects of epidemiological studies of
periodontal diseases. Periodontology 2000
29, 11–30.
Machtei, E. E., Christersson, L. A., Grossi, S.
G., Dunford, R., Zambon, J. J. & Genco, R. J.
(1992) Clinical criteria for the definition of
established periodontitis. Journal of Perio-
dontology 63, 206–214.
Norderyd, O. & Hugoson, A. (1998) Risk
of severe periodontal disease in a Swedish
adult population. A cross-sectional study.
Journal of Clinical Periodontology 25,
1022–1028.
Paidi, S., Pack, A. R. & Thomson, W. M. (1999)
An example of measurement and reporting of
periodontal loss of attachment (LOA) in
epidemiological studies: smoking and perio-
dontal tissue destruction. New Zealand Den-
tal Journal 95, 118–123.
Papapanou, P. N. (1996) Annals of Perio-
dontology 1, 1–36.
Papapanou, P. (1999) Epidemiology of perio-
dontal diseases: an update. Journal of the
International Academy of Periodontology 4,
110–116.
Peng, T. K., Yao, J. H., Shih, K. S., Dong, Y. J.,
Chen, C. K. & Pai, L. (1990) Assessment of
periodontal disease in an adult population
survey in Taipei city using CPITN and
GPM/T indices. Chung-Hua Ya i Hsueh Hui
Tsa Chih [Chinese Dental Journal] 9, 67–74.
Philstrom, B. L. (1992) Measurement of attach-
ment level in clinical trials: probing methods.
Journal of Periodontology 63, 1072–1077.
Querna, J. C., Rossmann, J. A. & Kerns, D. G.
(1994) Prevalence of periodontal disease in
an active duty military population as indi-
cated by an experimental periodontal index.
Military Medicine 159, 233–236.
Ramfjord, S. P. (1959) Indices for prevalence
and incidence of periodontal disease. Journal
of Periodontology 30, 51–59.
Susin, C., Valle, P., Oppermann, R. V., Hauge-
jorden, O. & Albandar, J. M. (2005) Occur-
rence and risk indicators of increased probing
depth in an adult Brazilian population.
Journal of Clinical Periodontology 32,
123–129.
Timmerman, M. F., van der Weijden, G. A.,
Armand, S., Abbas, F., Winkel, E. G., Van
Winkelhoff, A. J. & van, D. V. (1998)
Untreated periodontal disease in Indonesian
adolescents. Clinical and microbiological
baseline data. Journal of Clinical Perio-
dontology 25, 215–224.
Tonetti, M. S. & Claffey, N. (2005) Advances in
the progression of periodontitis and proposal
of definitions of a periodontitis case and
disease progression for use in risk factor
research. Group C consensus report of the
5th European Workshop in Periodontology.
Journal of Clinical Periodontology 32, 210–
213.
Van der Velden, U. (2005) Purpose and pro-
blems of periodontal disease classification.
Periodontology 2000 39, 13–21.
Van der Zee, E., Davies, E. H. & Newman, H.
N. (1991) Marking width, calibration from tip
and tine diameter of periodontal probes.
Journal of Clinical Periodontology 18, 516–
520.
Wang, W. J., Liu, C. Y., Liu, D. Z. & Lee, C. J.
(1987) Survey of periodontal disease among
workers in Tianjin using Ramfjord’s Perio-
dontal Disease Index (PDI). Community Den-
tistry and Oral Epidemiology 15, 98–99.
Address:
Kenneth A. Eaton
Old Saddlers
Kempe’s Corner
Canterbury Road
Ashford
Kent TN25 4EW
UK
E-mail: k.eaton@eastman.ucl.ac.uk
Clinical Relevance
Scientific rationale for the study:Itis
evident from the periodontal litera-
ture that there is a lack of uniformity
in the definition of periodontitis used
in epidemiological studies. The aim
of the current study was to perform a
systematic review of the literature
related to the definitions of perio-
dontitis and the methods that have
been used to identify and measure
this disease.
Principal findings: The studies that
were reviewed utilized a minimum
diagnostic threshold defining perio-
dontitis, at a given site in terms of
CAL of 2 mm and PPD of 3 mm.
Practical implications: This review
represents a step towards the future
validation of a definition for perio-
dontitis and the method(s) used to
measure it.
Systematic review definitions periodontitis 467
r2009 John Wiley & Sons A/S