J Clin Periodontol 2016; 43: 839848 doi: 10.1111/jcpe.

12594

Is furcation involvement affected

by adjunctive systemic
amoxicillin plus metronidazole?
A clinical trials exploratory
subanalysis
Eickholz P, Nickles K, Koch R, Harks I, Hoffmann T, Kim T-S, Kocher T,
Meyle J, Kaner D, Schlagenhauf U, Doering S, Gravemeier M, Ehmke B. Is
furcation involvement affected by adjunctive systemic amoxicillin plus
metronidazole? A clinical trials exploratory subanalysis. J Clin Periodontol 2016;
43: 839848. doi: 10.1111/jcpe.12594.

Abstract
Objectives: Evaluation of the clinical effect of systemic amoxicillin and metronidazole adjunctively to mechanical debridement at furcation sites.
Material and Methods: This is an exploratory per-protocol collective subanalysis
from a prospective, randomized, double-blind, multi-centre trial (ClinicalTrials.
gov NCT00707369) on the effect of adjunctive systemic amoxicillin 500 mg plus
metronidazole 400 mg (39/day, 7 days) use on furcation involvement in moderate
to severe periodontitis. Outcome was the change in frequency of classes of furcation involvement after 27.5 months. Therapy comprised mechanical debridement
in conjunction with antibiotic or placebo administration, and maintenance therapy at three months intervals.
Results: Three hundred and forty-five patients (175 placebo, 170 antibiotics) with
6576 furcation sites (class 0 2956; class I 2370; class II 886; class III 364) were
examined (3472 placebo, 3104 antibiotics). Pocket reduction/attachment gain at
the furcation sites was noticeably better after antibiotics (1.2/0.6 mm) than after
placebo (0.7/0.2 mm) 27.5 months after therapy. However, most furcation degrees
were unchanged (placebo 61.5%/antibiotics 62.2%), more sites improved than
deteriorated (20.3%/18.2%, 22.1%/15.7% respectively) and no differences in the
change of furcation degrees between treatments could be detected.
Conclusion: Compared to placebo, prescription of adjunctive systemic antibiotics
failed to show clinically relevant benefit with regard to furcation class involvement.

Peter Eickholz1,*, Katrin Nickles1,*,

Raphael Koch2, Inga Harks3, Thomas
Hoffmann4, Ti-Sun Kim5, Thomas
rg Meyle7, Dog
 an
Kocher6, Jo
8,9
Kaner , Ulrich Schlagenhauf10,
Stephan Doering11, Martina
Gravemeier3 and Benjamin Ehmke3
1

Department of Periodontology, Johann

Wolfgang Goethe-University Frankfurt,
Frankfurt, Germany; 2Institute of Biostatistics
nster,
and Clinical Research, University of Mu
nster, Germany; 3Department of
Mu
nster,
Periodontology, University Hospital Mu
nster, Germany; 4Department of
Mu
Periodontology, TU Dresden, Dresden,
Germany; 5Section of Periodontology,
Department of Conservative Dentistry,
University Hospital Heidelberg, Heidelberg,
Germany; 6Unit of Periodontology, University
Medicine Greifswald, Greifswald, Germany;
7
Department of Periodontology, University of
Giessen, Giessen, Germany; 8Department of
Periodontology, Witten/Herdecke University,
Witten, Germany; 9Department of
Periodontology and Synoptic Dentistry,
 Centrum 3, Charite
Charite
smedizin Berlin, Berlin, Germany;
Universita
10
Department of Periodontology, University
rzburg, Wu
rzburg, Germany;
Hospital Wu
11
Department of Psychoanalysis and
Psychotherapy, Medical University of Vienna,
Vienna, Austria

*Equal contribution to the study.

Key words: amoxicillin and metronidazole;
furcation involvement; mechanical
debridement; periodontitis; systemic
antibiotics
Accepted for publication 7 July 2016

Conflict of interest and source of funding statement

The authors declare that they have no conflict of interest.The study was exclusively supported by a grant from the German
Research Foundation (Deutsche Forschungsgemeinschaft, DFG EH 365 1-1), the ARPA Research Foundation, and from the
authors institutions. No writing assistance other than copy editing was provided.
2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

839

840

Eickholz et al.

In periodontitis-affected multi-rooted
teeth, the periodontal tissue is not only
destroyed vertically but also horizontally between the roots creating furcation involvement. Furcation-involved
molars respond less favourably to
periodontal therapy than molars without furcation involvement or singlerooted teeth and are at greater risk for
further attachment loss (Nordland
et al. 1987, Loos et al. 1989, Wang
et al. 1994) than other teeth. Addressing this issue, Kalkwarf et al. (1988)
reported the success of different surgical and nonsurgical treatment modalities in 158 molars. Irrespective of the
performed therapy, the horizontal
defect in the furcation area increased
during the two year follow-up.
Morphological features such as
enamel projections and accessory pulpal canals into the furcation (Pontoriero et al. 1989), anatomy which
impedes accessibility for individual
oral hygiene in the molar region
(Lang et al. 1973), and professional
root debridement (Fleischer et al.
1989) add up to the factors contributing to a more severe disease progression in furcation involved molars.
These factors cause recurrent periodontal infection, and as a result an
inferior long-term prognosis of these
teeth (McGuire & Nunn 1996, K
onig
et al. 2002, Dannewitz et al. 2006,
2016, Pretzl et al. 2008, Salvi et al.
2014, Graetz et al. 2015).
A clinical benefit regarding
parameters like probing depth reduction, attachment gain and bleeding
on probing is well documented in systematic reviews for the administration
of systemic amoxicillin and metronidazole adjunctively to mechanical
debridement compared to mechanical
debridement alone (Sgolastra et al.
2012a,b, Keestra et al. 2015a,b), but
a recent large controlled study shows
only limited benefit on further attachment loss (Harks et al. 2015).
The impact of a putative additional effect from adjunctive systemic amoxicillin and metronidazole
on furcation defects is, to the best of
our knowledge, unclear. However,
the periodical subgingival application of topical antibiotics adjunctive
to re-instrumentation during supportive periodontal treatment (SPT)
provided conflicting results regarding
effect on furcation involvement
(Tonetti et al. 1998, Dannewitz et al.
2009, Tomasi & Wennstr
om 2011).

The aim of the present subanalysis of data from the large clinical
ABPARO trial (ClinicalTrials.gov
NCT00707369, Harks et al. 2015)
was to evaluate the clinical effect of
systemic amoxicillin and metronidazole adjunctively to mechanical
debridement at molar and premolar
furcation sites.
Material and Method
Study design

This is a subanalysis of the per-protocol collective data from the prospective, randomized, stratified, doubleblind, multi-centre ABPARO trial
(ClinicalTrials.gov NCT00707369).
The trial was about the effect of
adjunctive systemic administration of
amoxicillin 500 mg plus metronidazole 400 mg (39/day, 7 days) on furcation involvement in patients
suffering from moderate to severe
periodontitis. Antibiotics were prescribed without detection of any specific bacteria. Thus, this rationale is
called empiric antibiotic therapy.
Patients who followed the study timeline according to the protocol and
took all tablets as scheduled were
included in the per-protocol collective.
The respective protocol and the
trials details and procedures have
been described in detail previously
(Harks et al. 2014, 2015). Thus, in
the following only a brief description
is provided:
Patients with untreated moderate
to severe chronic and aggressive
periodontitis were included. Key
inclusion criteria, were patients age
between 18 to 75 years, a CPITN
(Community Index of Treatments
Needs) of IV in at least one sextant,
at least 10 natural teeth in situ, and
pocket probing depths of 6 mm at
a minimum of four teeth. Key exclusion criteria were confirmed or
assumed allergies or former hypersensitive skin reactions to amoxicillin
and/or metronidazole, systemic medications affecting periodontal health,
and pregnancy.
The institutional review boards
(IRB) of the participating centres
approved the protocol and all
patients provided written informed
consent, and an independent data and
safety monitoring board reviewed the
safety data throughout the trial.

Participants were divided into

four strata according to the extent of
periodontal disease [localized: <38%;
generalized: 38% of teeth with
pocket probing depths (PPD)
6 mm] and smoking habit [non-/
light smoker: <7 ppm CO in exhaled
air; moderate to heavy smoker:
7 ppm
(Bedfont-Smokerlyzer,
Bedfont, UK)]. Quad-block patients
randomization lists were computer
generated for each stratum per centre (1:1 allocation ratio) by a statistician otherwise not being involved in
trial affairs. Randomization lists for
participating centres were stored
exclusively at the study centre.
Periodontal therapy

Per
patient,
12
visits
over
27.5 months were scheduled. Within
one and a half months after baseline
examination (visit 2), patients
received supra- and subgingival
debridement in up to two sessions on
two consecutive days (visit 3). All
mechanical therapy was performed
with hand instruments and/or
machine-driven scalers. After completion of mechanical debridement, the
antibiotic
user-group
patients
received two empirically chosen
antibiotics [amoxicillin 3H2O 574 mg
(Amoxicillin-ratiopharm
500 mg,
Ratiopharm, Germany); metronidazole 400 mg (Flagyl 400, SanofiAventis, Germany)], and placebo
group patients got two placebo drugs,
each to be taken three times a day for
seven days. The University pharmacy
in Dresden, Germany repacked the
medication in neutral capsules with
identical appearance. Each patient
received two medication packages
with consecutive numbers according
to the randomization list. Re-evaluation (visit 4) was performed three and
a half months after baseline and at
least two months after mechanical
debridement. Thereafter, all patients
received maintenance therapy, including full-mouth supragingival debridement and oral hygiene instruction at
three month intervals. Sites with PPD
4 mm also received subgingival redebridement.
Examinations and endpoints

The degree of furcation involvement

was assessed at all furcation sites of
maxillary first premolars (mesial,

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Amoxicillin/metronidazole in furcations
distal), maxillary molars (buccal, distooral, mesiooral), and mandibular
molars (buccal, lingual) using a
curved, scaled Nabers probe (PQ2N,
Nabers colour coded, Hu-Friedy,
Chicago, IL, USA) and the defect
characterized according to the following classification (Hamp et al.
1975, Eickholz & Staehle 1994):
Degree 0: the furcation is not
probable
Degree I: horizontal loss of periodontal tissue support up to 3 mm
Degree II: horizontal loss of support exceeding 3 mm, but not
encompassing the total width of
the furcation area
Degree III: horizontal throughand-through-destruction of the
periodontal tissue in the furcation.
Horizontal furcation measurements were performed at visit 2 (baseline), after 3.5 months (re-evaluation,
visit 4), and at 9.5, 15.5, 21.5 and
27.5 months (visit 12) follow-ups.
In the same intervals, full-mouth clinical periodontal measurements were
carried out at six sites of each tooth
[relative attachment level, pocket
probing depth, recession, bleeding on
probing (Lang et al. 1990), and plaque index (plaque control record,
O0 Leary et al. 1972)]. Measurements
were carried out by masked examiners not involved in periodontal therapy (Harks et al. 2014, 2015).
Statistical analysis

Statistical analyses were performed

using SAS software, version 9.4 of
the SAS System for Windows (SAS
Institute, Cary, NC, USA). Inferential statistics were intended to be
exploratory, not confirmatory. pvalues represent a metric measure of
evidence against the respective null
hypothesis and were used only to
generate new hypotheses. Therefore,
neither global nor local significance
levels were determined, and no
adjustment for multiplicity was
applied. p-values 0.05 were considered as noticeable.
Standard univariate statistical
analyses were performed to describe
demographic and clinical parameters. Categorical variables are shown
as absolute and relative frequencies.
Normal-distributed continuous variables are shown as mean  standard
deviation. Not normal-distributed

continuous variables are reported as

median [25% quantile75% quantile]. Patients characteristics were
compared
between
antibiotics
assigned group and placebo assigned
group patients using Fishers exact
tests for categorical variables, Students t-test for normally distributed
data and MannWhitney U-tests for
non-normally distributed data.
Site-level analysis (multiple sites
within one tooth and teeth within
each patient) was performed using
mixed regression models (Zeger &
Liang 1986, Verbeke & Molenberghs
2000, Molenberghs & Verbeke 2005)
to account for dependencies across
observations due to the nested data
structure. Therefore, p-values for the
site-based comparison of clinical
parameters between treatment groups
were calculated using either linear or
generalized linear mixed models
(LMM/GLMM). Binary variables
gingival bleeding and plaque control
record were compared fitting marginal
generalized linear mixed models with
treatment group as fixed effect, a
compound symmetric covariance
structure for the residuals and patient
as subject. Probing depth (mm),
attachment level (mm) at each time
point and their change were compared using linear mixed models with
treatment group as fixed factor, a random intercept for patient and random
intercept for tooth within patient.
The main outcome variables were
the frequency of change of furcation
degrees from visit 2 (baseline) to reevaluation (visit 4; three and a half
months) and from
baseline to
27.5 months (visit 12) respectively.
Change of furcation degree (0, I, II,
III) at site-level was described with
contingency tables and analysed
using marginal homogeneity tests
(Bhapkar 1966) separately within the
placebo and antibiotics group.
Change of furcation degree was
recoded into binary variables measuring improvement or impairment
of furcation degree. Mixed effect
logistic regressions were applied to
account
for
the
dependencies
between multiple sites within the
same patient. To fit the marginal
model, a compound symmetry
covariance structure for the residuals
was applied with patient as subject.
The following influencing factors
were entered into the analysis to
explain the dependent variables

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

841

improvement and impairment of furcation degree after 3.5 and

27.5 months: therapy (placebo/antibiotics), gender (female/male), smoker
(yes/no), tooth group (maxillary premolars/maxillary molars/mandibular
molars), age at screening visit (years),
baseline absolute attachment level
(pocket probing depth plus recession
measured in mm). Additionally, all
interactions between the covariates
and therapy were examined. Results
are reported as odds ratio (OR) and
corresponding 95% confidence limits.
Results

The patient characteristics from the

345 per-protocol patients are shown
in Table 1.
Clinical parameters from furcation sites at baseline (visit 2), at reevaluation after 3.5 months (visit 4),
and after 27.5 months (visit 12) are
given in Table 2: regarding bleeding
on probing (BOP) (placebo: 22.9%;
24.8%/antibiotics 13.7%; 15.7%), PPD
reduction (placebo: 0.5  1.3 mm;
0.7  1.6 mm/antibiotics 0.9 
1.4 mm; 1.2  1.7 mm) and attachment gain (placebo: 0.3  1.4 mm;
0.2  1.8 mm/antibiotics: 0.7 
1.5 mm; 0.6  1.7 mm); the antibiotics group provided statistically
noticeable more favourable results
than the placebo group 3.5 and
27.5 months after therapy.
The distribution of furcation
involvement at baseline according to
the jaw (maxilla/mandible) and number of premolar furcation sites is
provided in Table 3. A total of 6576
furcation sites (class 0: 2956 [45.0%];
class I: 2370 [36.0%]; class II: 886
[13.5]; class III: 364 [5.5%]) were
treated. A total of 3229 (49.1%) furcation sites were treated at maxillary
molars, 1036 (15.8%) at maxillary
premolars and 2311 (35.1%) at
mandibular molars. The percentages
of the different degrees of furcation
involvement were similar in upper
and lower molars. In contrast, first
upper premolars exhibited less frequent
furcation
involvement
(Table 3, italic numbers).
Overall, 6180 furcation sites were
followed up to re-evaluation after
three and a half months (placebo
group: 3249 and antibiotics group:
2931; Table 4). After three and a
half months on average, the majority
of
furcation
sites
remained

842

Eickholz et al.

Table 1. Patient characteristics

Placebo group
n = 175
Sex; n (%)
Male
Female
Age; years
Active smokers; n (%)
CO non-smoker; ppm
CO smoker; ppm
Body mass index; kg/m
HbA1c; %
HbA1c status
0% HBA1c 5.7%
5.7% < HBA1c 6.4%
6.4% < HBA1c
CRP, mg/l

88
87
52.3
44
0.0
13.0
25.2
5.3
116
53
6
1.4

(50.3%)
(49.7%)
 10.8
(25.1%)
(0.0, 1.0)
(9.0, 20.0)
(22.9, 28.4)
 0.8
(66.3%)
(30.3%)
(3.4%)
(0.6, 3.0)

Antibiotics group
n = 170

p-value

85
85
53.5
49
0.0
12.0
24.9
5.5

0.999F

105
52
12
1.1

(50.0%)
(50.0%)
 10.1
(28.8%)
(0.0, 1.0)
(7.0, 18.0)
(22.9, 27.5)
 0.8
(62.1%)
(30.8%)
(7.1%)
(0.6, 2.8)

0.2710T
0.4681F
0.5590U
0.5686U
0.4752U
0.1900T
0.3035F
0.4118U

Categorical variables are reported as absolute and relative frequencies. Continuous variables
are shown as mean  standard deviation or median (25% quantile, 75% quantile) due to
their distributional properties. p-values are from FFishers exact test, UMannWhitney
U-test, Tt-test for independent groups.
CO, carbon monoxide; CRP C-reactive protein.

unchanged (placebo group: 79.8%/

antibiotics group: 79.4%). The proportion of furcation sites showing
improvement, i.e. showing a lower
grade of furcation involvement at reevaluation after three and a half
months, was 10.0% in the placebo
group and 11% in the antibiotics
group. On the contrary, deterioration
in furcation sites was detected at
10.2% and 9.7% of sites respectively.
These changes from baseline to reevaluation were only statistically
noticeable in the placebo group
(p = 0.028).
A total of 5641 furcation sites
were followed up to 27.5 months
after baseline (placebo group: 2970,
antibiotics group: 2671; Table 5).
Over the 27.5 months period, most
furcation sites remained unchanged
in the placebo group (61.5%) and in
the antibiotics group (62.2%). However, the percentage of sites with
improved furcation class was 20.3%
in the placebo and 22.1% in the
antibiotics group. In contrast to the
re-evaluation measurements, the percentage of furcation sites showing a
deterioration of furcation class was
higher after 27.5 months (18.2%/
15.7% respectively) in both groups.
These changes after 27.5 months
were statistically noticeable in both
groups (p < 0.001).
From the different influencing
factors included in the generalized
linear mixed models (GLMM) to
explain improvement or impairment

of the degree of furcation involvement

at
re-evaluation
after
3.5 months, only the comparison of
maxillary premolars versus mandibular molars resulted in a statistically
noticeable lower odds ratio of 0.36
for furcation deterioration (p < 0.001,
Table 6). That implies maxillary premolars tend to have a higher degree
of deterioration of furcation degree
than mandibular molars. However,
initially higher attachment loss at the
furcation site resulted in the ambivalent finding that this may lead to
greater improvement (OR 1.07,
p = 0.007) as well as to deterioration
(OR 1.07, p = 0.006).
For the changes of furcation
degree from baseline to 27.5 months
follow-up (Table 7), female gender
compared to male gender (OR 0.75,
p = 0.039) and higher age (per year
older OR 0.99, p = 0.03) were statistically associated with a noticeably
decreased chance for improvement.
Initially, higher vertical attachment
loss was associated with a higher statistically noticeable probability of furcation degree improvement. Per one
millimetre higher vertical attachment
loss, the odds ratio was 1.14
(p < 0.001). Similar as from the baseline to three and a half months measurements, higher initial attachment
loss and maxillary premolars compared to mandibular molars provide
the probability of furcation class
impairment (OR 1.05, p = 0.015, and
OR 0.33, p < 0.001 respectively).

Finally, in the GLMMs, the comparison of systemic use of amoxicillin/metronidazole versus placebo
with regard to furcation class
improvement and deterioration was
not statistically different, neither in
the time span from baseline through
three and a half months nor through
27.5 months (Tables 6 and 7).
Discussion

Three
hundred
and
forty-five
patients (175 placebo, 170 antibiotics) with 6576 furcation sites were
examined (3472 placebo, 3104 antibiotics). BOP, PPD and attachment at
the furcation sites improved on average noticeably better after antibiotics
(BOP
15.7%,
PPD
reduction
1.2 mm; attachment gain 0.6 mm)
than after placebo (BOP 24.8%,
PPD reduction 0.7; attachment gain:
0.2 mm) 27.5 months after therapy.
Most
furcation
degrees
were
unchanged (placebo 61.5%/antibiotics 62.2%) and more sites
improved than impaired (placebo:
20.3%/18.2%; antibiotics 22.1%/
15.7% respectively). However, no
differences in the change of furcation
degrees between treatments could be
detected.
The primary outcome of this
study was the percentage of sites
showing further attachment loss
(PSAL) 1.3 mm 27.5 months after
therapy. PSAL was chosen as surrogate variable most reliably predicting
tooth loss. From 506 participating
patients, 406 were included in the
intent to treat analysis. The median
PSAL observed in the placebo group
was 7.8% compared to 5.3% in the
antibiotics group (p < 0.001). Both
treatments resulted in significant
PPD reduction, attachment gain and
were effective in preventing disease
progression (PSAL 1.3 mm). Compared to placebo, the administration
of empirically chosen adjunctive systemic antibiotics resulted in statistically noticeable more favourable
PPD reduction, attachment gain and
reduction of further attachment loss.
This is in accordance with this analysis of the respective parameters that
were measured only at the furcation
sites of multi-rooted teeth. However,
the difference between the antibiotic
and placebo group did not reach the
threshold of clinical relevance: i.e.
<50% of sites with PSAL 1.3 mm

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Amoxicillin/metronidazole in furcations
Table 2. Clinical measurements at furcation sites
Placebo group
BOP (%, n/N)
Baseline
3.5 months
27.5 months
PI (%, n/N)
Baseline
3.5 months
27.5 months
PPD (mm)
Baseline
3.5 months
Change (3.5 month Baseline)
27.5 months
Change (27.5 month Baseline)
Attachment Level (mm)
Baseline
3.5 months
Change (3.5 month Baseline)
27.5 months
Change (27.5 month Baseline)

Antibiotics
group

p-value
(LMM/GLMM)

36.7%
1307/3566
22.9%
805/3517
24.8%
819/3304

41.4%
1352/3266
13.7%
442/3228
15.7%
483/3082

0.0461*

40.3%
971/2412
29.4%
692/2350
40.0%
909/2274

40.9%
895/2190
26.8%
576/2152
38.9%
812/2086

0.9923*

3.9  2.0
3.4 (2.4, 5.2)
3.4  1.8
3.0 (2.0, 4.4)
0.5  1.3
0.4 (1.0, 0.2)
2.6  1.8
2.6 (1.8, 4.0)
0.7  1.6
0.6 (1.6, 0.2)

3.9  2.0
3.4 (2.4, 5.2)
3.0  1.4
2.6 (2.0, 3.8)
0.9  1.4
0.6 (1.6, 0.0)
2.4  1.4
2.4 (1.8, 3.4)
1.2  1.7
0.8 (2.0, 0.0)

0.7603**

4.5  2.1
4.0 (3.0, 5.8)
4.2  2.1
3.8 (2.6, 5.6)
0.3  1.4
0.2 (1.0, 0.4)
4.2  2.2
3.8 (2.6, 5.4)
0.2  1.8
0.2 (1.2, 0.8)

4.5  2.1
4.0 (2.8, 5.8)
3.7  1.9
3.4 (2.4, 4.8)
0.7  1.5
0.6 (1.4, 0.2)
3.8  1.9
3.4 (2.4, 4.8)
0.6  1.7
0.6 (1.6, 0.4)

<0.0001*
<0.0001*

0.2675*
0.8118*

<0.0001**
<0.0001**
<0.0001**
<0.0001**
0.7394**
0.0002**
<0.0001**
0.0026**
<0.0001**

Results are from n = 345 patients (placebo n = 175, antibiotics n = 170) and reported as
absolute and relative frequencies for binary variables, and mean  standard deviation and
median (25% quantile, 75% quantile) for continuous variables.
*p-value (Wald) from LMMs with treatment group as fixed factor, random intercept for
patient and random intercept for tooth within patient.
**p-values from marginal GLMM with treatment group as fixed factor, compound symmetric covariance structure for the residuals and patient as subject.
BOP, bleeding on probing; GLMM, generalized linear mixed model; LMM, linear mixed
model; mm, millimetre; n, number of sites; N, total number of sites within the treatment
group; PI, Plaque index (OLeary Plaque control record); PPD, pocket probing depth;
Attachment Level, vertical attachment level.

in the antibiotic than in the placebo

group (Harks et al. 2015).
Furcation defects are a frequent
finding in periodontitis patients
(Dannewitz et al. 2006, 2016).
Molars affected with furcation
involvement respond less favourably
to periodontal therapy than molars
without furcation involvement or
single-rooted teeth. They also exhibit
a greater risk for further attachment
loss than other teeth (Nordland
et al. 1987, Kalkwarf et al. 1988,
Loos et al. 1989, Wang et al. 1994).

Thus, more severe disease progression in furcation lesions results in an

inferior long-term prognosis of these
teeth (McGuire & Nunn 1996,
K
onig et al. 2002, Dannewitz et al.
2006, 2016, Pretzl et al. 2008, Salvi
et al. 2014, Graetz et al. 2015).
These observations may be explained
by the bizarre furcation morphology
which impedes accessibility for individual oral hygiene (Lang et al.
1973), and professional mechanical
therapy (Fleischer et al. 1989, Dannewitz et al. 2009). Therefore,

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

843

following a pathophysiological logic,

systemic amoxicillin and metronidazole could serve as an effective
adjunct to mechanical debridement
of furcation sites because via gingival crevicular fluid, antibiotics get
into areas inaccessible to the different types of mechanical approaches.
The main problem of furcation
involvement is periodontal destruction between the roots of multirooted teeth, i.e. horizontal attachment loss. Measurement of horizontal probing attachment levels as a
continuous variable, i.e. measurements in millimetres, would provide
the most precise measurement of
horizontal attachment loss. However, continuous horizontal probing
attachment levels cannot be measured at sites with degree III furcation involvement (Eickholz &
Staehle 1994). Thus, using horizontal
probing attachment levels as outcome variable would not cover class
III furcation sites adequately. Hence,
measurement of horizontal attachment loss was not performed in this
study. By using furcation classes as
main outcome variable, through and
through furcation lesions are covered
and a relation to prognosis may be
deduced (Dannewitz et al. 2009).
Additionally, it was shown that different classes of furcation involvement are related to different levels of
prognosis (McGuire & Nunn 1996,
Dannewitz et al. 2006, 2016, Salvi
et al. 2014, Graetz et al. 2015).
In this sample of 345 periodontitis patients, 45% of all furcation
sites were without furcation involvement which approximately confirms
the frequency reported for molars
(Dannewitz et al. 2006). Dannewitz
et al. (2006) observed degree I furcation lesions less frequently (23%)
than this study (36%). However,
degree II and III furcation lesions
were reported more frequently
(24%/13%) than in this study
(13.5%/5.5%). For a sample of 71
patients with periodontal disease,
Dannewitz et al. (2006) reported furcation involvement tooth-based and
not site-based. They assigned to each
molar the most severe furcation
involvement that was observed.
Using this mode, the information
relevant for prognosis is given for
each molar. However, the frequency
of less severe furcation degrees is
underestimated. This study reports

844

Eickholz et al.

Table 3. Distribution of furcation involvement (sites) at baseline according to jaw (maxilla/

mandible) and tooth type (premolar/molar)
Degree of furcation
involvement

II

III

Total

Maxilla

Mandible

Molars

Premolars

Molars

1172
(17.8%)
(36.3%)
1244
(18.9%)
(38.5%)
556
(8.5%)
(17.2%)
257
(3.9%)
(8.0%)
3229
(49.1%)
(100%)

755
(11.5%)
(72.9%)
227
(3.5%)
(21.9%)
42
(0.6%)
(4.1%)
12
(0.2%)
(1.2%)
1036
(15.8%)
(100%)

1029
(15.6%)
(44.5%)
899
(13.7%)
(38.9%)
288
(4.4%)
(12.5%)
95
(1.4%)
(4.1%)
2311
(35.1%)
(100%)

Total

2956
(45.0%)
2370
(36.0%)
886
(13.5%)
364
(5.5%)
6576
(100%)

Results are reported as absolute frequencies, percentage of the total, and column percentage
(italic) within maxillary molars/premolars and mandible.
Table 4. Frequency of degrees of furcation involvement at baseline (visit 2) and three and
a half months later (visit 4)
Furcation involvement

Placebo group
Three and a half months after therapy

Baseline

Degree 0

Degree I

Degree II

Degree III

Total

Degree
Degree
Degree
Degree
Total

1163/35.8%
148/4.6%
23/0.7%
14/0.4%
1348/41.5%

185/5.7%
965/29.7%
107/3.3%
16/0.5%
1273/39.2%

10/0.3%
86/2.6%
323/9.9%
17/0.5%
436/13.4%

5/0.2%
16/0.5%
28/0.9%
143/4.4%
192/5.9%

1363/42.0%
1215/37.4%
481/14.8%
190/5.8%
3249/100%

0
I
II
III

Furcation involvement

Antibiotics group
Three and a half months after therapy

Baseline

Degree 0

Degree I

Degree II

Degree III

Total

Degree
Degree
Degree
Degree
Total

1108/37.8%
194/6.6%
12/0.4%
4/0.1%
1318/45.0%

193/6.6%
833/28.4%
87/3.0%
8/0.3%
1121/38.2%

15/0.5%
54/1.8%
263/9.0%
16/0.5%
348/11.9%

5/0.2%
3/0.1%
13/0.4%
123/4.2%
144/4.9%

1321/45.1%
1084/37.0%
375/12.8%
151/5.2%
2931/100%

0
I
II
III

Placebo: Improvement: 10.0%, No change 79.8%, Deterioration 10.2%, p = 0.028 (Bhapkar

Test).
Antibiotics: Improvement: 11.0%, No change 79.4%, Deterioration 9.7%, p = 0.084 (Bhapkar Test).
Results are reported as absolute frequencies and percentage of total.
Blue: improvement; red: deterioration.

the degree of furcation involvement

of all sites enlarging the proportion
of less severe furcation lesions. Considering this and the fact that Dannewitz et al. (2006) did not report
premolars, the actual data on frequency of furcation lesions roughly
confirm earlier findings in a much
larger sample.
To the best of our knowledge up
to now, there is no study addressing
the putative additional effects of

systemically administered amoxicillin

and metronidazole adjunctively to
mechanical debridement on furcation
defects. Regarding BOP, reduction
of PPD and vertical attachment gain
at furcation sites, the antibiotic
group responded statistically noticeably better than the placebo group.
These results confirm the observation at all sites (Harks et al. 2015).
However, regarding frequency of
furcation degrees, the subanalysis

failed to find statistically noticeably

differences between both groups. In
the present subanalysis, more than
half of the furcation sites have been
kept stable in terms of furcation
classes (about 62%) over the
27.5 months period. Improvements
appeared in 21% of furcation sites,
and deterioration of furcation classes
were measured at less than 17% of
furcation sites. Overall, therapy was
successful in both groups. After the
27.5 months period in the present
patient collective, an amount of furcation sites had changed comparable
to a study on subgingival re-instrumentation during supportive periodontal therapy with (73.4%) or
without (66.7%) topical subgingival
14% doxycycline gel (doxycyclinehyclat in a polyethylene glycollactid/glycolid copolymer gel [Ivoclar
Vivadent AG, Schaan, Liechtenstein]) after 3 months (Dannewitz
et al. 2009). The rates of improvement (i.e. reduction of furcation
class) were higher 3 months after
both treatments (mechanical debridement: 20.6%; mechanical debridement plus local doxycycline: 19.6%)
than after 3.5 months in this study.
However, mechanical debridement
alone exhibited a significant higher
rate of impairment, i.e. increase of
furcation involvement class, of
12.7% compared to 7.0% after
adjunctive topically applied doxycycline, which was interpreted as a
benefit of topical subgingival doxycycline. Overall, due to the short
duration of trial and the small absolute difference in deterioration, the
clinical relevance is difficult to judge.
In the present subanalysis, after
27.5 months, deterioration of furcation sites was higher after mechanical debridement alone (18.2%) than
with adjunctive systemic amoxicillin
plus metronidazole administration
(15.7%). However, due to long
observation period, the even smaller
absolute difference between groups
and no statistically noticeably effect
of the therapy mode on furcation
class improvement or deterioration,
the prescription and use of systemic
amoxicillin plus metronidazole is of
limited clinical value. Evidently, furcation sites without furcation
involvement can only change if deterioration occurs, i.e. increase of furcation involvement. However, at
least 74% of degree 0 furcation sites

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Amoxicillin/metronidazole in furcations
Table 5. Frequency (n/%) of degrees of furcation involvement at baseline (visit 2) and
27.5 months later (visit 12)
Furcation involvement

Placebo group
27.5 months after therapy

Baseline

Degree 0

Degree I

Degree II

Degree III

Total

Degree
Degree
Degree
Degree
Total

919/30.9%
320/10.8%
60/2.0%
9/0.3%
1308/44.0%

274/9.2%
632/21.3%
160/5.4%
22/0.7%
1088/36.6%

42/1.4%
129/4.3%
181/6.1%
31/1.0%
383/12.9%

8/0.3%
44/1.5%
44/1.5%
95/3.2%
191/6.4%

1243/41.9%
1125/37.9%
445/15.0%
157/5.3%
2970/100%

0
I
II
III

Furcation involvement

Antibiotics group
27.5 months after therapy

Baseline

Degree 0

Degree I

Degree II

Degree III

Total

Degree
Degree
Degree
Degree
Total

907/34.0%
344/12.9%
69/2.6%
3/0.1%
1323/49.5%

266/10.0%
555/20.8%
145/5.4%
17/0.6%
983/36.8%

28/1.0%
71/2.7%
103/3.9%
12/0.4%
214/8.0%

8/0.3%
12/0.4%
34/1.3%
97/3.6%
151/5.7%

1209/45.3%
982/36.8%
351/13.1%
129/4.8%
2671/100%

0
I
II
III

Placebo: Improvement: 20.3%, No change 61.5%, Deterioration 18.2%, p < 0.001 (Bhapkar
Test).
Antibiotics: improvement: 22.1%, No change 62.2%, Deterioration 15.7%, p < 0.001 (Bhapkar Test).
Results are reported as absolute frequencies and percentage of total.
Blue: improvement; red: deterioration.

in both groups did not change

through the 27.5 months observation
period. Vice versa furcation sites
with class III involvements at baseline may only change if furcation
involvement class improves. In class
III furcation, improvements were
observed in about 39.5% of the placebo group furcation sites and in
24.8% of the antibiotic group sites.
Both therapy options resulted in
similar rates of improvement after
27.5 months.

Do the statistically noticeably

better results regarding BOP, reduction of PPD and vertical attachment
gain at furcation sites justify the use
of systemic amoxicillin and metronidazole? Are these statistically noticeably larger improvements in the
antibiotic group clinically relevant?
A statistically significant weighted
mean difference of 0.53 mm in PPD
reduction between quadrant scaling
and full-mouth disinfection (FMD)
in baseline pockets of 56 mm was

845

found in a systematic review and

judged modest and the implications
for periodontal care are not profound (Eberhard et al. 2008). In
this study, the difference between
antibiotics and placebo regarding
PPD reduction is 0.5 mm after
27.5 months. Having the conclusion
of Eberhard et al. (2008) in mind,
the difference between antibiotics
and placebo in this study has to be
judged modest as well. This is particularly true considering that this difference was achieved by systemic
administration of antibiotics and not
topical use of chlorhexidine.
Less
inflammation
as
was
obtained in the adjunctive antibiotics
group may be very useful for a subsequent regenerative procedure, e.g.
at class II furcation defects. However, are systemic antibiotics affecting the whole body appropriate to
additionally reduce inflammation at
a few sites that may qualify for a
regenerative procedure of a furcation
defect (Sanz et al. 2015)? Topical
application of antibiotics at those
furcation sites qualifying for regeneration may be considered the more
appropriate option.
The effect of initial attachment
loss at furcation sites leading simultaneously to more improvement and
deterioration, can be interpreted as
an increasing variability in the
change of furcation degree in each
direction associated with increasing
attachment loss. At sites with severe
attachment loss, treatment may
cause decreased inflammation which

Table 6. Generalized linear mixed model estimates for the change of furcation degree from baseline to three and a half months
Target variable

Independent variables
Therapy: AMOX/METR versus placebo
Sex: female versus male
Smoker: yes versus no
Tooth group:
(maxillary molars) versus (mandibular molars)
(maxillary premolars) versus (mandibular molars)
Age at visit 1(x + 1 versus x years)
Attachment level at baseline (x + 1 versus x mm)

Improvement of furcation degree

(baseline to three and a half months)

Impairment of furcation degree

(baseline to three and a half months)

OR

95% confidence
limits

p-value

OR

95% confidence
limits

p-value

1.123
0.951
1.187

0.788
0.668
0.755

1.602
1.354
1.866

1.063
0.814
0.998
1.066

0.844
0.568
0.983
1.018

1.339
1.166
1.014
1.117

0.5195
0.7800
0.4576
0.3461
0.6053
0.2614
0.8308
0.0069

0.863
1.063
0.953

0.614
0.755
0.639

1.212
1.496
1.422

1.010
0.363
1.013
1.073

0.801
0.251
0.997
1.020

1.274
0.525
1.029
1.128

0.3937
0.7253
0.8132
<0.0001
0.9322
<0.0001
0.1099
0.0063

Result from the generalized linear mixed model (345 patients, 6166 sites). Estimates of fixed effects on the improvement and impairment of
furcation degree are reported as odds ratio, 95% confidence limits and p-values of the Wald-Tests. A logistic random effect regression was
fitting to account for the dependencies between multiple sites within the same patient. To fit the marginal model, a compound symmetry
covariance structure for the residuals was applied with patient as subject. Additionally, all interactions between the independent variables
and the therapy were examined, but because of non-statistically noticeable result (p > 0.05) not included in the final model.
OR, odds ratio.
2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

846

Eickholz et al.

Table 7. Generalized linear mixed model estimates for the change of furcation degree from baseline to 27.5 months
Target variable

Improvement of furcation degree

(baseline to 27.5 months)
OR

Independent variables
Therapy: AMOX/METR versus placebo
Sex: female versus male
Smoker: yes versus no
Tooth group:
(maxillary molars) versus (mandibular molars)
(maxillary premolars) versus (mandibular molars)
Age at visit 1(x + 1 versus x years)
Attachment level at baseline (x + 1 versus x mm)

Impairment of furcation degree

(baseline to 27.5 months)

95% confidence
limits

p-value

OR

95% confidence
limits

p-value

1.167
0.746
0.995

0.883
0.565
0.705

1.542
0.985
1.404

0.916
0.814
0.986
1.136

0.757
0.623
0.973
1.093

1.109
1.065
0.999
1.181

0.2755
0.0387
0.9774
0.2985
0.3664
0.1337
0.0334
<0.0001

0.815
0.918
1.090

0.626
0.704
0.792

1.061
1.197
1.501

1.031
0.328
1.010
1.054

0.838
0.239
0.997
1.010

1.267
0.450
1.024
1.099

0.1273
0.5262
0.5943
<0.0001
0.7743
<0.0001
0.1198
0.0152

Result from the generalized linear mixed model (345 patients, 5628 sites). Estimates of fixed effects on the improvement and impairment of
furcation degree are reported as odds ratio, 95% confidence limits and p-values of the Wald-Tests. A logistic random effect regression was
fitting to account for the dependencies between multiple sites within the same patient. To fit the marginal model, a compound symmetry
covariance structure for the residuals was applied with patient as subject. Additionally, all interactions between the covariates and the therapy were examined, but because of non-statistically noticeable result (p > 0.05) not included in the final model.
OR, odds ratio.

results in higher resistance against

vertical and horizontal probing (improvement of furcation involvement). However, therapy also may
cause recession which causes easier
access to horizontal probing. Both
will result in improvement of PPD.
However, both also will result in
controversial furcation scores. It
may be that a factor that was not
considered in this analysis makes the
difference in change of furcation
involvement at sites with severe vertical attachment loss.
This subanalysis bares a few limitations. Although, data came from a
prospective clinical trial, all inference
statistics have only an exploratory
interpretation. The statistical models
were fitted data-driven and were not
determined before data collection.
Therefore, the applied statistical
models are not valid to make
assumptions about population inference.
Assessment of furcation degrees
is reproducible and reliable (Eickholz
& Staehle 1994, Eickholz 1995, Eickholz et al. 1999). However, it has
been shown that assessments at
disto-oral furcation sites are less reliable than at other sites particularly
in the presence of an adjacent tooth.
Further, it has been shown that a
rigid curved probe (e.g. Nabers
probe) provides more valid measurements than flexible and rigid straight
probes (Eickholz & Kim 1998, Eickholz et al. 1999). Thus, a colourcoded Nabers probe was used for all

furcation
measurements.
Even
though all examiners were trained
and experienced periodontal specialists, furcation diagnosis is tricky and
bears some risk for error. This is
particularly true for the distinction
of degree II and III furcation lesions
(Eickholz 2010).

Conclusion

Within the limitations of this explorative analysis, the following conclusions may be drawn: the use of
empirically chosen systemic amoxicillin plus metronidazole as an
adjunct to mechanical debridement
Provided statistically noticeably better BOP, PPD reduction and
attachment gain at the furcation
sites than placebo three and a half
and 27.5 months after therapy.
Frequency of change of furcation
degrees was small and the study
failed to show any benefits of
antibiotics compared to placebo
with regard to change of class of
furcation involvement three and a
half and 27.5 months after therapy.

Acknowledgements

We thank the members of the Center

for Clinical Trials, Medical Faculty
M
unster, Germany, for supporting
the trial: Sonja Baier, Trude Butterfa-Bahloul, J
urgen Grebe, Kerstin Hovestadt, Heidi Oellers, Anita

Ripkens-Reinhard, and Gudrun

W
urthwein.
We thank Dr. Almut von Schwedler, MBBS (Tas), Matcham,
NSW, Australia and Ilona Wall, Salt
Lake City, UT, USA for their valuable and competent support in the
preparation of this manuscript.
Last but not least, the authors
are greatly indebted to collaborators
and staff members represent the
ABPARO-Group for their successful
work on this project. The members
of the ABPARO-Group were: Study
center, University Hospital M
unster:
Christina Elberg, Heike FrielingBraithwaite, Anna-Maria Marx,
Marie Christin Ohlmeier, Martin
Sachs, and Thomas Weniger. Charit
e
- Universit
atsmedizin Berlin: Peter
Purucker, Marta Czownicka, Kathleen Kraatz, Nicole Pischon, and
Bernd-Michael Kleber. University
Hospital Dresden: Gerlinde Bruhn,
Ihssan Khallili, and Katrin Lorenz.
Johann Wolfgang Goethe-University
Frankfurt: Bettina Dannewitz, Lasse
R
ollke, Susanne Scharf, and Martin
Wohlfeil. University Hospital Giessen: Heidi Fastnacht, Jose Roberto
Gonzales, and Tomas CabreraChica. University Hospital Greifswald: Jutta Daus and Jutta
Fangh
anel. University Hospital Heidelberg: Raluca Cosggarea, Amelie
Meyer-B
aumer, Nihad El Sayed,
Sven Zehaczek, and Nils Zimmermann. University Hospital W
urzburg:
Markus Bechtold, Yvonne JockelSchneider, and Simone Veihelmann.

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

Amoxicillin/metronidazole in furcations
Institute of Biostatistics and Clinical
Research, Medical Faculty Muenster:
Andreas Faldum, Joachim Ger,
and Achim Heinecke. Clinical Pharmacy, University Hospital Dresden:
Ina-Maria Klut and Madeleine
Schubert. Institute of Clinical Chemistry and Laboratory Medicine,
University Hospital Greifswald: Matthias Nauck, Astrid Petersmann and
Helma Preez. Data Monitoring and
Safety Board: Guido Knapp, Gregor
Petersilka, and Anne Sonntag.
References
Bhapkar, V. P. (1966) A note on the equivalence
of two test criteria for hypotheses in categorical
data. Journal of the American Statistical Association 61, 228235.
Dannewitz, B., Krieger, J. K., H
using, J. & Eickholz, P. (2006) Loss of molars in periodontally
treated patients: a retrospective analysis five
years or more after active periodontal treatment. Journal of Clinical Periodontology 33,
5361.
Dannewitz, B., Lippert, K., Lang, N. P., Tonetti,
M. S. & Eickholz, P. (2009) Supportive periodontal therapy of furcation sites: non-surgical
instrumentation with or without topical doxycycline. Journal of Clinical Periodontology 36,
514522.
Dannewitz, B., Zeidler, A., H
using, J., Saure, D.,
Eickholz, P. & Pretzl, B. (2016) Loss of molars
in periodontally treated patients: results ten
years and more after active periodontal therapy.
Journal of Clinical Periodontology 43, 5362.
Eberhard, J., Jerve-Storm, P. M., Needleman, I.,
Worthington, H. & Jepsen, S. (2008) Full-mouth
treatment concepts for chronic periodontitis: a
systematic review. Journal of Clinical Periodontology 35, 591604.
Eickholz, P. (1995) Reproducibility and validity
of furcation measurements as related to class
of furcation invasion. Journal of Periodontology
66, 984989.
Eickholz, P. (2010) Glossar der Grundbegriffe f
ur die
Praxis: Parodontologische Diagnostik 6: Furkationsdiagnostik. Parodontologie 21, 261266.
Eickholz, P. & Kim, T.-S. (1998) Reproducibility
and validity of the assessment of clinical furcation parameters as related to different probes.
Journal of Periodontology 69, 328336.
Eickholz, P. & Staehle, H. J. (1994) The reliability
of furcation measurements. Journal of Clinical
Periodontology 21, 611614.
Eickholz, P., Steinbrenner, H., Lenhard, M., Marquardt, M. & Holle, R. (1999) Interexaminer
reliability of the assessment of clinical furcation
parameters as related to different probes. European Journal of Oral Sciences 107, 28.
Fleischer, H. C., Mellonig, J. T., Brayer, W. K.,
Gray, J. L. & Barnett, J. D. (1989) Scaling and
root planing efficacy in multirooted teeth. Journal of Periodontology 60, 402409.
Graetz, C., Sch
utzhold, S., Plaumann, A., Kahl,
M., Springer, C., S
alzer, S., Holtfreter, B.,
Kocher, T., D
orfer, C. E. & Schwendicke, F.

(2015) Prognostic factors for the loss of molars

an 18-years retrospective cohort study. Journal of Clinical Periodontology 42, 943950.
Hamp, S.-E., Nyman, S. & Lindhe, J. (1975) Periodontal treatment of multirooted teeth. Results
after 5 years. Journal of Clinical Periodontology
2, 126135.
Harks, I., Harmsen, D., Gravemeier, M., Prior,
K., Koch, R., Doering, S., Petersilka, G.,
Weniger, T., Eickholz, P., Hoffmann, T., Kim,
T.-S., Kocher, T., Meyle, J., Purucker, P., Schlagenhauf, U. & Ehmke, B. (2014) A concept
for clinical research triggered by suggestions
from systematic reviews about adjunctive
antibiotics. Applied Clinical Research, Clinical
Trials Regulatory Affairs 1, 4350.
Harks, I., Koch, R., Eickholz, P., Hoffmann, T.,
Kim, T.-S., Kocher, T., Meyle, J., Kaner, D.,
Schlagenhauf, U., Doering, S., Holtfreter, B.,
Gravemeier, M., Harmsen, D. & Ehmke, B.
(2015) Is progression of periodontitis relevantly
influenced by systemic antibiotics? A clinical
randomized trial. Journal of Clinical Periodontology 21, 332342.
Kalkwarf, K. L., Kaldahl, W. B. & Patil, K. D.
(1988) Evaluation of furcation region response
to periodontal therapy. Journal of Periodontology 59, 794804.
Keestra, J. A. J., Grosjean, I., Coucke, W., Quirynen, M. & Teughels, W. (2015a) Non-surgical
periodontal therapy with systemic antibiotics in
patients with untreated chronic periodontitis: a
systematic review and meta-analysis. Journal of
Periodontal Research 50, 294314.
Keestra, J. A. J., Grosjean, I., Coucke, W., Quirynen, M. & Teughels, W. (2015b) Non-surgical
periodontal therapy with systemic antibiotics in
untreated aggressive periodontitis patients: a
systematic review and meta-analysis. Journal of
Periodontal Research 50, 689706.
K
onig, J., Plagmann, H. C., R
uhling, A. &
Kocher, T. (2002) Tooth loss and pocket probing depths in compliant periodontally treated
patients: a retrospective analysis. Journal of
Clinical Periodontology 29, 10921100.
Lang, N. P., Adler, R., Joss, A. & Nyman, S.
(1990) Absence of bleeding on probing. An
indicator of periodontal stability. Journal of
Clinical Periodontology 17, 714721.
Lang, N. P., Cumming, B. & L
oe, H. (1973)
Toothbrushing frequency as it relates to plaque
development and gingival health. Journal of
Periodontology 44, 396405.
Loos, B., Nylund, K., Claffey, N. & Egelberg, J.
(1989) Clinical effects of root debridement in
molar and non-molar teeth. A 2-year follow-up.
Journal of Clinical Periodontology 16, 498504.
McGuire, M. K. & Nunn, M. E. (1996) Prognosis
versus actual outcome. III. The effectiveness of
clinical parameters in developing an accurate
prognosis. Journal of Periodontology 67, 666674.
Molenberghs, G. & Verbeke, G. (2005) Models
for discrete longitudinal data. New York, NY:
Springer.
Nordland, P., Garrett, S., Kriger, R., Vanooteghem, R., Hutchens, L. H. & Egelberg, J.
(1987) The effect of plaque control and root
debridement in molar teeth. Journal of Clinical
Periodontology 14, 231236.
O0 Leary, T. J., Drake, R. B. & Naylor, J. E.
(1972) The plaque control record. Journal of
Periodontology 43, 3839.

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd

847

Pontoriero, R., Lindhe, J., Nyman, S., Karring,

T., Rosenberg, E. & Sanavi, F. (1989) Guided
tissue regeneration in the treatment of furcation
defects in mandibular molars. A clinical study
of degree III involvement. Journal of Clinical
Periodontology 16, 170174.
Pretzl, B., Kaltschmitt, J., Kim, T.-S., Reitmeir,
P. & Eickholz, P. (2008) Tooth loss after
active periodontal therapy. 2. Tooth-related
factors. Journal of Clinical Periodontology 35,
175182.
Salvi, G. E., Mischler, D. C., Schmidlin, K., Matuliene, G., Pjetursson, B. E., Br
agger, U. &
Lang, N. P. (2014) Risk factors associated with
the longevity of multi-rooted teeth. Long-term
outcomes after active and supportive periodontal therapy. Journal of Clinical Periodontology
41, 701707.
Sanz, M., Jepsen, K., Eickholz, P. & Jepsen, S.
(2015) Clinical concepts for regenerative therapy
in furcations. Periodontology 2000 68, 308332.
Sgolastra, F., Gatto, R., Petrucci, A. & Monaco,
A. (2012b) Effectiveness of systemic amoxicillin/metronidazole as adjunctive therapy to
scaling and root planing in the treatment of
chronic periodontitis: a systematic review and
meta-analysis. Journal of Periodontology 83,
12571269.
Sgolastra, F., Petrucci, A., Gatto, R. & Monaco,
A. (2012a) Effectiveness of systemic amoxicillin/metronidazole as an adjunctive therapy to
full-mouth scaling and root planing in the
treatment of aggressive periodontitis: a systematic review and meta-analysis. Journal of Periodontology 83, 731743.
Tomasi, C. & Wennstr
om, J. L. (2011) Locally
delivered doxycycline as an adjunct to mechanical debridement at retreatment of periodontal
pockets: outcome at furcation sites. Journal of
Periodontology 82, 210218.
Tonetti, M. S., Cortellini, P., Carnevale, G., Cattabriga, M., de Sanctis, M. & Pini Prato, G. P.
(1998) A controlled multicenter study of
adjunctive use of tetracycline periodontal fibers
in mandibular class II furcations with persistent
bleeding. Journal of Clinical Periodontology 25,
728736.
Verbeke, G. & Molenberghs, G. (2000) Linear
mixed models for longitudinal data. New York,
NY: Springer.
Wang, H. L., Burgett, F. G., Shyr, Y. & Ramfjord, S. (1994) The influence of molar furcation involvement and mobility on future
clinical periodontal attachment loss. Journal of
Periodontology 65, 2529.
Zeger, S. L. & Liang, K.-Y. (1986) Longitudinal
data analysis for discrete and continuous outcomes. Biometrics 42, 121130.

Address:
Peter Eickholz
Department of Periodontology
Center for Dentistry and Oral Medicine
(Carolinum)
Johann Wolfgang Goethe-University
Frankfurt am Main
Theodor-Stern-Kai 7 (Haus 29)
60596 Frankfurt am Main, Germany
E-mail: eickholz@med.uni-frankfurt.de

848

Eickholz et al.

Clinical Relevance

Scientific rationale for this analysis:

Furcation-involved
multi-rooted
teeth are reported to respond less
favourably to periodontal therapy
and are at greater risk for disease
progression than molars without
furcation involvement or single
rooted teeth. Administration of

empirically chosen adjunctive systemic

amoxicillin and metronidazole adjunctively to mechanical debridement may
improve furcation involvement.
Principal
findings:
Empirically
administration of empiric systemic
amoxicillin
and
metronidazole
adjunctively to mechanical debridement failed to show a difference with

regard to average improvement of

furcation involvement.
Practical implications: Furcationinvolved molars may not benefit
from empiric administration of systemic amoxicillin and metronidazole adjunctively to mechanical
debridement.

2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd