**4. Methods**

### *4.1. Literature Search*

This systematic review considered all studies listed in PubMed until 30 September 2018. For additional studies, we double-checked in EMBASE and supplemented with additional hits obtained from Google Scholar. Grey literature was not part of the review process. It was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement (PRISMA) [43]. For the PRISMA checklist, see http://www.prisma-statement.org. We employed the

Population, Intervention, Control, and Outcome (PICO) format to answer the following PICO questions: Is PD associated with the occurrence of PAOD?

Population = All patients with PD (all degrees of severity) who were detected in the selected literature

Intervention = None

Comparison = Patients with and without PD

Outcome = PAOD of the lower extremity

We first determined a list of synonyms and MeSH-terms (Medical Subject Headings) for PAOD and PD. Using these lists, we defined the following search-formula.

(peripheral arterial disease OR peripheral artery disease OR PAD OR occlusive vascular disease OR IC OR intermittent claudication OR CLI OR peripheral arterial occlusive disease OR peripheral artery occlusive disease OR PAOD OR lower limb ischemia OR DFS OR vascular surgery)

AND (periodontitis OR gum disease OR pyorrhea OR periodontal disease OR periodontal infection OR periodontal conditions OR chronic periodontitis OR periodontal health OR tooth loss OR attachment loss OR probing pocket depth)

### *4.2. Study Selection and Data Extraction*

All studies were reviewed by two independent authors, one with vascular expertise (MK) and one with oral health expertise (US). Reviewers were blinded to each other results. Both reviewers screened all papers selected by the search formula to identify inclusion criteria. Any disagreements between reviewers at each stage of selection were resolved by consensus. Cohen's kappa coe fficient (κ) demonstrated good agreemen<sup>t</sup> between the two reviewers (κ = 0.9) (Appendix A Figure A1). The studies were processed according to the PRISMA flow diagram shown in Figure 1. To sort the studies, we developed a screening-algorithm (Figure 3).

**Figure 3.** Screening algorithm.

Following this algorithm, we decided whether the eligibility criteria were met; first by title, and if the title led to an uncertainty regarding the eligibility criteria, additionally by the abstract. If there was still uncertainty, the whole paper was read. Animal studies, studies that were published in non-English, studies that did not correspond to original papers, case reports, case series, meta-analyses, or systemic reviews were excluded. In addition, studies on coronary or carotidal/cerebral sclerosis or vascular sclerosis were excluded in general because the aim of this systematic review was to focus on PAOD. Studies describing an association between PD (all degrees of severity) and PAOD (all degrees of severity) were included and further processed according to the PRISMA flow diagram (Figure 1). Studies that were identified by additional manual searches were also added. No duplicates were identified. Citations were managed throughout the different stages of preparing the review with the Mendeley reference managemen<sup>t</sup> software.
