According to Dr. Bjarne Klausen from Esbjerg, Danish National Clinical Guidelines for the use of antibiotics in dental practice had recently been finalized. See a quick guide here [pdf]. As regards periodontal infection, he writes,
We formulated a focused question that was in line with the Scandinavian consensus: Should prescription of antibiotics be considered in patients with sufficient oral hygiene, if their periodontal condition does not respond to conventional treatment?
Well, they did not find a single relevant study. How’s that? Isn’t it so that dozens if not hundreds of studies had been performed since the early discovery by Jørgen Slots in Copenhagen identifying an unidentified gram-negative rod in abundance in what was then localized juvenile periodontitis? Soon later it was clear that this bug was Actinobacillus (now Aggregatibacter) actinomycetemcomitans, and that it could best be targeted by systemic antibiotics in conjunction with traditional mechanical/surgical periodontal treatment. Soon after a couple of case series by van Winkelhoff et al. in 1989 and 1992, Amoxicillin plus Metronidazole (the infamous “van Winkelhoff cocktail”) became extremely popular and has since then been mentioned in national guidelines for the treatment of severe periodontitis where A. actinomycetemcomitans could be found in abundance.
When having revised a popular textbook on Periodontology for undergraduates for its 3rd edition I slightly updated a table which I had used in teaching for a decade. It was once published in an important article by late Sigmund Socransky and Anne Haffajee  where these two most influential authors in Oral Microbiology made their point that biofilm infections are in fact different (and difficult to target). The table is featured on the first page of the article but may easily be overlooked. But it contains important ideas which are still valid. Socransky and Haffajee differentiate acute, chronic delayed and biofilm infections and present respective examples. While many acute infections are fortunately self-limiting and require just supportive treatment  chronic infections need antibiotics since the host does not properly cope with the causative agents. I have added as typical examples acute necrotizing ulcerative gingivitis, or ANUG, which in most cases resolves under oral hygiene improvement and cautious scaling (supported by antiseptics), and necrotizing ulcerative periodontitis, or NUP, which is longstanding and indicates that the host, for instance due to immune suppression in the course of HIV infection, elicit an improper, impaired response. Antibiotics are not indicated in the former case while they are in the latter; usually metronidazole is the drug of first choice . If the causative agent in delayed infection is known, therapy involving suitable antibiotics is easy. In the past, that was not the case for example for syphilis (Treponema pallidum) where all kinds of mercury treatment had been applied. The disease could only be cured after the detection of penicillin. The same holds, for example, for gastrointestinal ulcers caused by Helicobacter pylori, which has been proven only in the 1980s. Gastric ulcers are now easily treated by high dosages of antbiotics after the bug had been identified.