The purpose of this study is to examine if the correct use of antibiotics, as well as the application of one-day-treatment, in the therapy of periodontal diseases. The population of treated patients (184) are divided in 4 groups; Group 1: Metronidazole + mechanical treatment in one day, Group 2: Placebo + mechanical treatment in one day Group 3: Metronidazole + mechanical treatment on day 1 and 21 Group 4: Placebo + mechanical treatment on day 1 and 21 End points, BOP, Plaque, PPD, CAL, RAL, Microbiology will be recorded on all sites at 3 months, 1 -, 3 -, and 5 years post-treatment

This is a clinical research project in the odontologic special field; periodontology. Periodontitis affects > 50% of all persons over 40 years-of-age, whereof 5 - 10% is prone to loose one or more teeth due to the disease. In Norway, less than 40% of the periodontal treatment costs are carried (refunded) by the public. The proportion of personal expenses vs. public spending depends on how serious the periodontitis is diagnosed, and on the individual treatment provided. In addition comes expenses connected to the restoration of the dentition when teeth are lost as a consequence of the disease, and therefore the social, personal and economical consequence of serious tooth loss can be devastating to the patient.

Background and established knowledge Periodontal disease is a family of bacterial infections characterized by the destruction of periodontal supporting tissues. The bacterial flora at the diseased sites is complex, counting > 500 different species in the subgingival dental plaque. Three main hypotheses exist in explaining the infectious nature of periodontal disease; the non-specific infection hypothesis (Theilade 1986), the specific infection hypothesis (Slots and Rams 1991, Loesche 1992) and the ecological plaque hypothesis (Marsh 1994).

Dental biofilm is an important factor in periodontal health and disease (review: Socransky and Haffajee 2002), and is a more correct wording for the term plaque. Most biofilms are biologically compatible with health, as their formation is influenced by the selective pressure of the host organism. It is also of great practical interest that bacteria organized in biofilms are highly resistant to antibiotics as compared to those in a planktonic state. In the development of dental biofilms, the host receives bacteria with different properties from close relatives and others, depending on the living conditions (Tinoco et al 1998). If there are pathogenic bacteria among these, the possibility for developing periodontitis increases (Preus et al 1995). This indicates that periodontal treatment must be aimed at breaking the ecology of the pathogenic biofilm, and then increase the organisms' ability to perform its selective pressure, following treatment, in the making of a new biofilm low of, or free from pathogenic microorganisms.

Traditional Treatment strategies have so far aimed at preventing or retarding the progression of the disease through mechanical, non-specific treatment, and we have succeeded to a certain extent.

Scaling and root planing (SRP) is the standard treatment, aiming to mechanically reduce the number of bacteria present on the affected root surfaces. Representing a non-specific form of treatment to a non-specific infection, it has proven effective in most cases and short follow ups(Hill et al 1981, Pihlstrøm et al 1983, Lindhe et al 1984, Becker et al 1990, Kaldahl et al 1993). However, recurrence rate is fairly high, suggesting that these diseases are life-long traits. Also, a small - still significant - group of patients shows remaining inflammation and continuous destruction of periodontal support despite adequate oral hygiene and proper mechanical treatment (Lang et al 1986, Greenstein & Caton 1990, Walker et al 1993, Bragd et al 1987, Renvert et al 1990, Haffajee et al 1991, Grossi et al 1994, Slots and Rams 1991, Socransky and Haffajee 1992).

Scaling and root planing (SRP) and Adjunctive systemic antibiotics have been suggested in order to prohibit such remaining destructive infection (Van Winkelhoff et al 1996). The antibiotic should preferably be chosen based on microbiological diagnoses and the suggested pathogenic bacterial resistance profile (Mombelli AW & van Winkelhoff AJ 1997, Preus & Laurel 2003). However, such treatment has been reported with varying effects in clinical studies (Joyston-Bechal et al 1986, Lindhe et al 1983a, Lindhe et al 1983b, Bain et al 1994, Tinoco et al 1997, Ciancio et al 1982, Chavez et al 1995, Mahmod & Dolby 1987). Thus, adjunct antibiotics seem as effective/ineffective in reducing the number of recurrent infection as SRP.

Normally the antibiotic is administered at the end of the mechanical treatment, covering sometimes more - sometimes less - of the treatment period. Bearing in mind the biofilms' ability to resist antibiotics, it is clear that in dental biofilms, the bacteria are not easily/not at all affected by treatment where antibiotics are involved. This possibly explains why SRP and adjunct antibiotic treatment - according to the literature - does not give a better result than SRP alone.

SRP and Full-mouth disinfection (FDIS) has been introduced to increase the success rate of SRP. FDIS is an adjunct therapy whereby scaling and root planing is performed within a 48 hour time frame to reduce the possibility for re-infection of the treated root surfaces. Chlorhexidine is used as a disinfectant in addition to the mechanical treatment. In a series of pilot studies on FDIS (Quirynen et al 1995, 1999, 2000, 2006, Bollen et al 1996) reported amazingly good results. On the other hand, Apatzidou et al a, b, c, (2004), Koshy et al (2005) and Wennström et al (2005) could not find any improvement from FDIS over conventional therapy.

New treatment strategy Non-specific part; Traditional SRP to reduce the bacterial load and disrupt/destroy the biofilm, but reduce the size of "the window of opportunity" for re-infection by scaling /removing biofilms in all sites (healthy/not healthy) within 8 hours. All sites are scaled, healthy or diseased. Additional FDIS to reduce amount of bacteria in the oral cavity to levels possible to handle for the antibiotic to be used.

Specific part; Microbiological diagnosis and antibiotic therapy adjunctive to SRP/FDIS.

Re-establishing a beneficial and protective biofilm: FDIS and rinsing with CHX and brushing tongue with CHX gel for 7 days after treatment. Change tooth brush the day of treatment to restrict possible re-inoculation from this, that particular day.

Working hypothesis There is no difference in long term clinical and microbiological outcome between groups of patients treated with SRP/FDIS and SRP with or without Metronidazole. Treatment outcome is measured by the recurrence rate on an individual as well as on a group level. (see definitions at end-point evaluation - p6), at included sites at 1, 2, 3, 4 and 5 years of follow up. "No difference" is consistent with nil-hypothesis.

Power calculation: This calculation is based on the end point "attachment level (AL)", either it is measured on roentgenograms (RAL) or clinically (CAL) which again is the measurable basis for "recurrence of periodontal disease" = further loss of attachment after treatment. Based on results from similar published studies we assume that the standard deviation of AL in each treatment group will be approximately 1.5 mm. A two-sample t-test will be applied when comparing average AL in two treatment groups at a given time point, with a 5% significance level. Further, we consider differences in average AL between two treatment groups to be clinically significant if it is at least 1.0 mm.

In order to have at least 80% probability to demonstrate a statistically significant difference in average AL between two treatment groups, given that the true difference is at least 1.0 mm, at least 36 patients must be available for statistical analysis in each group. We expect that about 10% of the patients will drop out of the study during the 5 years follow-up period, but we cannot exclude the possibility of a drop-out rate as high as 20%.

At least 44 patients should be included in each treatment group.

176 (i.e.180) males and females 30 - 60 years-of-age referred to specialist clinic in Bø in Telemark. No interactive diseases with periodontal disease and willing to participate by signing an informed consent. Patients should have >4 pockets 5 - 8mm deep, bacterial diagnosis which indicates the use of Metronidazole only (to prohibit differences due to antibiotic choice). Included patients should not have known allergy or ill-reaction to this drug. Microbiological diagnosis will be performed by PCR and culture and a second microbiological "shadow-test" will be performed at a parallel microbiological service. The primary microbiological diagnosis will be based on both culture and PCR on samples obtained by curettes.

Therapy (appendix 2 - Time Table) Patient will be screened, and their oral hygiene will be brought up to standard by hygienist (TP). Eligibility, Indexes and microbiological diagnosis will then be recorded by project leader (HP). If willing to participate in the study, the patients will randomly be assigned to the 4 different groups by secretary/ project coordinator (N). Group 1; One-day SRP/FDIS with antibiotics, Group 2; One-day SRP/FDIS with placebo, Group 3; SRP over 2 weeks (scaling day 1 and 14) with antibiotics, Group 4; SRP over 2 weeks (scaling day 1 and 14) with placebo. Treatment at baseline will be performed after protocol (TP)l, and maintenance controls with recording of indexes and microbiological diagnosis will be performed at 3 months, 1, 3 and 5 year post-treatment (HP). All data, Measurements, microbiological diagnoses and response to questionnaires will be sent to CodeBookMaster (PG) following each measuring session. Reliability testing of the measurer (HP) every 12 month.

Microbiology I; Microbiological diagnosis will be performed for treatment and for follow up. This will be done at a clinical laboratory (Telelab, Skien) which offers a service for Oral microbiology.

Microbiology II: Concomitantly with every sample taken and processed an parallel sample will be obtained in PBS. These samples will be processed at a by Dr Gunnar Dahlen in Gothenburg using chequerboard.

End Point evaluation - Choice of Method/indexes Periodontal disease is an inflammatory, infectious disease which destroys tooth attachment including bone and fibres. Treatment is aimed at halting or retarding this destruction. At 3 months, 1, 3 and 5 years the CAL and RAL as a parameter for periodontal disease and PD/BoP as parameters for inflammation will be evaluated and statistically treated between the groups. Pocket Depth (PD) and bleeding on probing (BoP) are only surrogate parameters as the disease is defined as "loss of attachment which includes bone". Periodontitis is measured in CAL or on Xrays = RAL.

Since periodontal disease seems to recur after 3 - 4 years, we have suggested a 5 year follow-up-study. Blinding I: To the Project Leader who is the one who performs all measurements, the randomization to groups will be hidden since it is the research Coordinator (N) who does this. N will also keep track of time schedules, appointments etc. A separate record with the patient names and study number (1 - 160) will be kept by N locked down in a separate cabinet for such patient records. At measuring appointments the measurements are sent by N to the code book master who will log all measurements into SPSS program for statistical evaluation.

Blinding II: To the patient as well as the hygienist, the mechanical treatment the patient gets will be revealed since it is physically impossible to hide the fact that some are treated in one day, others in 14 days. However, the placebo/antibiotic treatment ensures blinding of both patient and hygienist.

Thus total blinding will only apply to the project leader/measurer.

Inclusion Criteria:

• Females and males with advanced periodontal disease referred between November 2007 and January 2009.
• No diseases known to be associated with periodontal diseases, i.e. diabetes mellitus, rheumatoid arthritis, pregnancies.
• No continuous medication known to affect the severity or progression of periodontal diseases.
• No previous systematic periodontal treatment.
• More than 5 pockets 5mm < PPD < 9 mm after a 3 month hygiene phase - i.e. before inclusion into study.
• Oral hygiene; < 5% of surfaces with plaque at inclusion time.
• Signing an informed consent.

Exclusion criteria:

• diseases known to be associated with periodontal diseases, i.e. diabetes mellitus, rheumatoid arthritis.
• pregnancy
• continuous medication known to affect the clinical expression of periodontal diseases
• subgingival microbiological diagnosis which reveals bacteria with reduced/no sensitivity to metronidazole