**3. Discussion**

This is the first meta-analysis aimed to evaluate the effect of anti-inflammatory periodontal therapy on changes of lipid levels in patients with type 2 DM. Periodontal therapy involves the mechanical removal of dental plaque associated with periodontitis. The majority of the studies used non-surgical periodontal treatment as the intervention, however one study [11] used surgical periodontal treatment for select individuals in the intervention arm as well. The analyses demonstrated that total cholesterol and triglycerides were significantly reduced in the intervention arm 3 months after therapy to lower levels, while HDL levels were reduced in the control group. However, no significant differences were observed after 6 months. The studies included in this review showed considerable heterogeneity, which has to be recognized before any conclusion can be drawn. However, this systematic review highlighted the potential benefits of periodontitis therapy to reduce total cholesterol and triglycerides levels. These positive effects may reduce the risk for cardiovascular complications in patients with type 2 DM.

The studies included were selected using stringent selection criteria described in the methods section, however, none of the studies included were designed to analyse lipid profiles as primary outcome measures. This may contribute to the high heterogeneity of the outcomes, as well as factors affecting lipid levels in general, including how long the individuals have had type 2 DM, lifestyle factors, and diet, which have not been assessed or reported in the included studies. All studies demonstrated a substantial reduction of clinical parameters of periodontal disease, including PD and BOP in the intervention groups 3 and 6 months after therapy, indicative of a successful treatment of the inflammatory reaction involved in periodontal disease. By contrast, the control groups did not show obvious changes in assessed oral health parameters. The mean levels of investigated total cholesterol and triglycerides decreased 3 months after periodontitis therapy, however, no differences were observed after 6 months. A possible explanation for this fading effect on lipid profiles after prolonged observation periods is recolonization by the subgingival microbiota and subsequent inflammation [18] if supportive periodontal treatment is not provided. Even though the periodontal parameters were significantly improved at the 6-month follow-up relative to baseline, five out of the seven included studies [10,12,13,15,16] did not provide supportive periodontal therapy to participants in the intervention arm after the initial treatment. In these participants, it is very likely that the recolonization of the microflora re-induced the inflammatory reaction which may have adversely affected lipid parameters. It should also be noted that average periodontal probing depths and bleeding on probing percentages are lower at the 3-month follow-up compared to the 6-month follow-up. This observation indicates the necessity of a regular periodontal maintenance program aimed to minimise the recolonization of tooth surfaces with periodontal pathogens and the concordant inflammation of the adjacent tissues.

Three out of the seven studies included in the current review [14–16] showed a significant reduction in levels of glycated haemoglobin and four studies [10–13] did not show significant changes between baseline and follow-up. In those studies reporting a significant reduction of glycated haemoglobin after periodontal treatment, one study [15] showed improved levels of total cholesterol, one study [16] showed an improvement in levels of HDL, whereas in both studies, other lipid parameters showed no significant changes. The third study [16] did not find any difference in lipid parameters between baseline and follow-up despite the reduction in levels of glycated haemoglobin. Within the limitations of this comparison, a reduction in glycated haemoglobin may not necessarily be accompanied by changes in lipid levels.

Individuals with periodontitis have been noted to have an increased risk of hyperlipidaemia and hypercholesterolaemia [19]. As mentioned previously, periodontitis is a chronic infection of the tooth supporting structures [1], and local chronic infections have been shown to alter concentrations of cytokines and hormones which can result in changes in lipid metabolism [20]. Specifically, systemic exposure to infectious challenges such as bacterial lipopolysaccharide can result in the release of inflammatory cytokines including interleukin-1 (IL-1) and tumour necrosis factor alpha (TNF-α) that alter fat metabolism and promote hyperlipidaemia. Both TNFα and IL-1 inhibit the production of lipoprotein lipase, which causes disturbances of lipid metabolism, including increased amounts of serum cholesterol and LDL [21]. A second mechanism by which bacterial lipopolysaccharides contribute to the development of atherosclerosis is by oxidative modification of increased LDL caused by macrophage activation. Oxidized LDL is taken up by macrophage scavengers, which leads to transformation of macrophages into foam cells, the hallmark of the atherosclerotic process. Oxidized LDL is also cytotoxic to endothelial cells and a potent chemoattractant for circulating human monocytes [22]. Conversely, it has also been demonstrated that a short-term high-fat diet results in prolonged impairment in the antibacterial function of polymorphonuclear leukocytes, which may cause damage of periodontal tissues [23]. Thus, a chronic hyperlipidaemic state may impair the host resistance to bacterial infection.

Cardiovascular disease is a major complication of type 2 DM and lipid abnormalities seen in diabetics are a serious contributor to this complication [6]. Glycaemic control via maintaining adequate levels of HbA1c is considered as an essential way to lower patients' risk of having diabetic complications and each 1% drop in HbA1c levels is associated with a risk reduction of 21% for diabetes-related deaths, 14% for myocardial infarction, and 37% for microvascular complications [24]. Several studies have indicated that periodontal infection caused by gram-negative bacteria had adverse e ffects on diabetic patients' glycaemic control [25,26]. By contrast, improved periodontal conditions following periodontal treatment can significantly improve HbA1c levels [11,27]. A lipid-lowering managemen<sup>t</sup> in type 2 DM patients is also aimed at reducing the incidence of cardiovascular complications, and statins can be very effective in improving the lipid profile and are therefore the first line class of drugs [28]. In general, di fferent statins have varying abilities to improve lipid profiles in patients, e.g., HDL cholesterol levels increase between 5% and 10% with statin therapy, LDL levels reduce, ranging from 27% to 60%, and triglycerides levels reduce between 11% and 40% [29]. The current analysis demonstrated a mean reduction of triglyceride levels by approximately 8% achieved by periodontitis treatment. Within the limitations of the available study data and the heterogeneity of studies, this will not be su fficient to annotate periodontitis treatment as an adjunct to a lipid-lowering managemen<sup>t</sup> in patients su ffering from type 2 DM. However, it may stimulate the interest in further exploring the benefits of good oral health for the prevention of diabetes complications and especially to setup well designed clinical trials with lipid profiles as the primary outcome.

### **4. Materials and Methods**

### *4.1. Types of Studies*

Randomized control trials of 3- or 6-month follow-ups were considered for this review.

### *4.2. Types of Participants*

The participants of the included studies had a diagnosis of type 2 DM and periodontitis. Patients with type I diabetes were excluded.

### *4.3. Types of Intervention*

All periodontal treatments using mechanical debridement (surgical and non-surgical, with and without adjunctive treatment) were included.

### *4.4. Types of Outcome Measures*

Primary outcome measures were total cholesterol, triglycerides, LDL cholesterol, and HDL cholesterol between baseline and 3- or 6-month follow-ups. Secondary outcome measures were periodontal probing depths, clinical attachment loss, and bleeding on probing.
