kowall2015

Accepted Article
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/jcpe.12391
This article is protected by copyright. All rights reserved.
Article Type: Original Article
Prediabetes and well controlled diabetes are not associated with periodontal disease: the
SHIP Trend Study
Running title: Prediabetes and periodontitis
Bernd Kowall
1,2
, Birte Holtfreter
3
, Henry Völzke
4
, Sabine Schipf
4
, Torsten Mundt
5
,
Wolfgang Rathmann
1
, Thomas Kocher
3
1
German Diabetes Center, Institute of Biometrics and Epidemiology, Düsseldorf, Germany
2
Center of Clinical Epidemiology, c/o Institute of Medical Informatics, Biometry and
Epidemiology (IMIBE), University Hospital Essen, Germany
3
Unit of Periodontology, Department of Restorative Dentistry, Periodontology, and
Endodontology, University Medicine, Ernst-Moritz-Arndt-University Greifswald, Germany
4
Institute for Community Medicine, University Medicine, Ernst-Moritz-Arndt-University
Greifswald, Germany
5
Department of Prosthodontics, Gerodontology and Biomaterials, Dental School, University
Medicine, Ernst-Moritz-Arndt-University Greifswald, Germany
Correspondence author:
Dr. Bernd Kowall
Center of Clinical Epidemiology
c/o Institute of Medical Informatics, Biometry and Epidemiology (IMIBE)
Hufelandstraße 55
45147 Essen
Tel: [phone]

Accepted Article
This article is protected by copyright. All rights reserved.
Fax: [phone]
Mail: [e-mail]
Abstract
Aim. To examine associations of prediabetes and well controlled diabetes with periodontitis.
Materials and methods. The Study of Health in Pomerania (SHIP)-Trend is a cross-sectional
survey in North-Eastern Germany including 3,086 participants (49.4% men; age 20–82 years).
Clinical attachment loss (CAL) and periodontal probing depth (PPD) were assessed applying a
random half-mouth protocol. The number of teeth was determined. Prediabetes comprised
impaired fasting glucose and impaired glucose tolerance. Previously known diabetes was
defined as well controlled if glycated hemoglobin (HbA1c) was <7.0%. Participants were
categorized as follows: normal glucose tolerance (NGT), prediabetes, newly detected type 2
diabetes (T2DM), known T2DM with HbA1c<7.0%, known T2DM with HbA1c≥7.0%.
Results. Prediabetes was neither associated with mean CAL and PPD in multivariable
adjusted linear regression models nor with edentulism (OR=1.09 (95%-CI: 0.69-1.71)) and
number of teeth (OR=0.96 (95%-CI: 0.75-1.22), lowest quartile versus higher quartiles) in
logistic regression models. Associations with mean CAL and edentulism were stronger in
poorly controlled previously known diabetes than in well controlled previously known
diabetes (for edentulism: OR=2.19 (95%-CI: 1.18-4.05), and OR=1.40 (95%-CI: 0.82-2.38),
respectively, for comparison with NGT).
Conclusions. Periodontitis and edentulism were associated with poorly controlled T2DM, but
not with prediabetes and well controlled diabetes.
Keywords: periodontitis; periodontal diseases; periodontal attachment loss;; mouth,
edentulous; prediabetic state; hyperglycemia; glucose intolerance; diabetes mellitus
1. Introduction
There is strong evidence that periodontitis is associated with diabetes mellitus (Khader2006;
Chavarry et al 2009; Demmer et al 2010; Lalla & Papapanou 2011; Preshaw et al 2012;

Accepted Article
This article is protected by copyright. All rights reserved.
Borgnakke et al. 2013). Two meta-analyses showed that mean periodontal probing depth
(PPD) and mean clinical attachment loss (CAL) are higher in subjects with than in subjects
without diabetes (Khader et al 2006; Chavarry et al 2009). Because the association between
diabetes and periodontitis is bidirectional, periodontitis also has a negative impact on
glucose regulation. Thus, periodontal treatment such as scaling and root planing was
suggested to improve glycemic control. Indeed, these interventions reduced glycated
hemoglobin (HbA1c) in type 2 diabetes patients by roughly 0.4%, as reported in recent meta-
analyses (Teeuw et al 2010; Engebretson & Kocher 2013; Wang et al 2014). However, little is
known about associations between periodontitis and moderate glycemic disorders such as
prediabetes and well controlled diabetes.
Prediabetes is a state of elevated levels of blood glucose that still are below the threshold
for manifest diabetes (WHO 1999; American Diabetes Association 2014). Prediabetes may be
identified by impaired fasting glucose (IFG) and/or impaired glucose tolerance (IGT), two
conditions that are both highly prevalent in Western countries (International Diabetes
Federation 2014). There are only few studies on the association between periodontal
disease and prediabetes reporting conflicting results. In addition, some studies have
methodical flaws such as small sample sizes or incomplete/insufficient adjustment for
confounders. Results are ambiguous for associations between periodontitis and IGT (Nelson
et al 1990; Emrich et al 1991; Noack et al 2000; Maragumeet al 2003; Saito et al 2004; Saito
et al 2005; Saito et al 2006) as well as for associations between periodontitis and IFG (Zadik
et al 2003; Choi et al 2011; Wang et al 2009). For example, in two reports from the same
research group on the association between IGT and PPD, findings were contrasting: an
association was seen in a mixed-sex Japanese study group (Saito et al 2004), but not in
Japanese women (Saito et al 2005). Likewise, an association between IFG and periodontitis
was reported in only one (Choi et al 2011) of two recent studies from the National Health
and Nutrition Examination Survey (NHANES) (Choi et al 2011; Arora et al 2014).
Moreover, only few studies considered whether periodontal health is associated with
glycemic control in persons with diabetes (Sandberg et al 2000; Tsai et al 2002; Chuang et al
2005; Peck et al 2006; Campus et al 2005; Javed et al 2007 ; Silvestre et al 2009; Preshaw et
al 2010 ; Haseeb et al 2012 ; Demmer et al 2012). Most, but not all, of these studies showed
a worse periodontal status in poorly controlled diabetes. However, several of these studies

Accepted Article
This article is protected by copyright. All rights reserved.
suffered from drawbacks such as lack of adjustment for confounders (Haseeb et al 2012;
Campus et al 2005), small sample size (Javed et al 2007; Peck et al 2006), and lack of a
constant, globally accepted HbA1c cut-off value to define poorly controlled diabetes.
Thus, there is still a need for studies with a large sample size and well characterized
participants to clarify the association between periodontal disease and moderate glycemic
disorders. The aim of the present study is to assess whether periodontitis, low number of
teeth and edentulism are more prevalent in subjects with prediabetes and well controlled
diabetes than in subjects with normal glucose tolerance using data from the large,
population-based Study of Health in Pomerania (SHIP)-Trend study.
2. Materials and Methods
Study population
SHIP-Trend is a cross-sectional survey conducted in West Pomerania, a region in the North-
Eastern part of Germany (Völzke et al 2011) – a follow-up of the study will run from 2016 -
2018. Examinations were conducted between 2008 and 2012. A stratified random sample of
10,000 adults aged 20–82 years was drawn from population registries. Inclusion criteria
were residency in the study region and - to ensure the individual had lived in the area (or in
Germany) for a long time - German citizenship. The only exclusion criterion was participation
in SHIP-0, another population study started in 1997. Sample selection was facilitated by
centralization of local population registries in the Federal State of Mecklenburg/West
Pomerania. Stratification variables were age, sex and city/county of residence. Subjects were
sampled from the regional strata with a probability proportional to size design. After
exclusion of migrated (N=851) and deceased (N=323) persons, the net sample included 8,826
persons. Because of several reasons (241 did not answer, 3367 declined participation, 549
did not keep the appointment and 249 agreed without an appointment) 4,420 subjects were
finally enrolled in the study, corresponding to a response rate of 50.1%.
Dental examination
Periodontal probing depth (PPD) and clinical attachment loss (CAL) were assessed at four
sites per tooth (mesiobuccal, midbuccal, distobuccal and midlingual/midpalatinal) according
to the random half-mouth method, excluding third molars. A manual periodontal probe

Accepted Article
This article is protected by copyright. All rights reserved.
(PCPUNC 15, Hu-Friedy, Chicago, IL, USA) was used. Measurements were mathematically
rounded to the nearest (or closest) whole millimeter. PPD was measured as the distance
between the free gingival margin (FGM) and the bottom of the sulcus or periodontal pocket.
If the cemento-enamel junction (CEJ) was located subgingivally, CAL was calculated as PPD
minus the distance between free gingival margin (FGM) and CEJ. If recession was present at
the examined site, CAL was directly measured as the distance between CEJ and the pocket
base. Where the determination of the CEJ was impossible due to wedge-shaped defects or
restorations, attachment level was not recorded. The number of all natural teeth present
(excluding third molars) was recorded, and hence bridge abutments, prostheses, and dental
implants were excluded.
Every 6–12 months, all five examiners performed calibration exercises on subjects not
associated with the study. Intra-rater correlations for CAL measurements ranged from 0.67
to 0.89 and inter-rater correlation was 0.70. For PPD measurements, the examiners yielded
intra-rater correlations between 0.68 and 0.88 and an inter-rater correlation of 0.72.
Definition of periodontitis
We used one periodontitis case definition (Tonetti & Claffey 2005) and various continuous
and dichotomous measures to assess the periodontal and dental status. According to the 5
th
European Workshop in Periodontology (EWP) (Tonetti & Claffey 2005), we defined incipient
periodontitis as presence of proximal CAL of ≥3 mm in ≥2 non-adjacent teeth. Manifest
periodontitis was defined as presence of proximal CAL of ≥5 mm in ≥30% of teeth, thereby
identifying “cases with substantial extent and severity” (Tonetti & Claffey 2005). For the EWP
case definition, analyses were restricted to those with at least two sites with CAL
measurements, as it necessitates the presence of at least two teeth (Tonetti & Claffey 2005).
Mean CAL/PPD equals the average CAL/PPD over all assessed sites.
Laboratory measurements
HbA
1c
was measured by high-performance liquid chromatography with spectrophotometric
detection (Diamat Analyzer; Bio-Rad, Munich, Germany) and a coefficient of variation of
1.5%. For the oral glucose tolerance test (OGTT), fasting venous blood is sampled, and 75 g
of anhydrous glucose with blackcurrant flavor (Dextro OGT, Boehringer Mannheim,