**2. Results**

### *2.1. Selection of Studies*

Six hundred and eighty-two studies were retrieved from the electronic databases PubMed, MEDLINE via Ovid, EMBASE via Ovid, and Web of Science. Three hundred and sixty-nine duplicates were removed, and the remaining abstracts were screened for eligibility, resulting in sixty-eight studies being further excluded. Two hundred and forty-five studies were then selected for full-text analysis. After reading of the full texts, seven studies [10–16] were included in the systematic review (Figure 1). Full text studies were excluded if they did not report lipid levels as an outcome or studies did not have appropriate intervention and control arms.

**Figure 1.** Study selection flow diagram.

### *2.2. Characteristics of Studies*

The characteristics of the included studies are described in Table 1. In total, there were 411 individuals who underwent periodontal treatment (intervention arm) and 341 individuals who did not or received post-trial intervention (control arm). The total number of participants in each study ranged from 40 to 264. The follow-up times for the studies ranged from 3 to 12 months, however, for this review only the 3- and 6-month follow-up data has been analysed. Among seven studies, five studies reported 3-month follow-up [10,12,13,15,16] and three studies reported 6-month follow-up data [10,11,14].

The mean age of individuals in the studies ranged from 45.5 to 63.2 years old. Three out of seven studies [11,12,16] excluded patients with cardiovascular disease. Three studies [10,14,15] excluded patients with uncontrolled systemic diseases, however, did not specify whether cardiovascular disease was amongs<sup>t</sup> this exclusion criteria. One study [13] did not specify whether participants with any uncontrolled systemic diseases were excluded or not. Two studies [11,14] specified that participants taking anti-hypertensive/cholesterol medications were included whereas five studies [10,12,13,15,16] did not specify whether patients taking anti-hypertensive/cholesterol medications were included or not. Six out of the seven studies [10–12,14–16] presented periodontal inclusion criteria and, among them, four different criteria were identified. All studies [10–16] reported diabetes inclusion criteria, and seven different criteria were identified (Table 1).



SPT: surgical periodontal treatment; NSPT: non-surgical periodontal treatment; SRP: scaling and root planing; DM: diabetes mellitus; HbA1c: glycated haemoglobin; WHO: World Health Organization; PD: pocket depth; CAL: clinical attachment loss.

### *2.3. Risk of Bias within Studies*

The Cochrane Collaboration's tool for assessing risk of bias [17] was used to assess the risk of bias within the included studies which have been summarized in Figure 2. Of the seven clinical trials included [10–16], five studies described the methods of randomisation [10–12,14,15]. Three studies [11,12,14] used a computer-generated table for allocation concealment, two studies [10,15] assigned an independent individual to allocate participants, and two studies [13,16] did not specify allocation concealment. Blinding of patients and personnel was not possible due to the nature of the periodontal treatment. Blinding of investigators was conducted in four studies [10–12,14], two did not specify whether investigators were blinded [13,15], and one study was non-blinded [16]. Four studies had minimal/no participant drop out or the data for the participants who dropped out was excluded [13–16], two studies had a moderate number of participants drop out [10,11] and one study [12] had a high participant dropout rate indicating attrition bias. There was no evidence of selective reporting or any other form of bias in any of the studies.

**Figure 2.** Risk of bias analysis of individual studies.
