Search outcomes

The Preferred reporting items in systematic reviews and meta-analyses (PRISMA) reporting guideline was followed in this analysis [20]. Our search resulted in a total number of 8620 publications [2432 from MEDLINE, 1745 from EMBASE, 378 from http://www.ClinicalTrials.gov, 2611 from Web of Science, 856 from Google scholar and 598 from the Cochrane databases]. The authors carefully assessed the titles and abstracts. Irrelevant studies were directly eliminated (7890 articles). Seven hundred and thirty (730) full text articles were assessed for eligibility. Duplicated full-text studies were first of all eliminated (493). Based on the criteria for inclusion and exclusion, studies were further eliminated as followed:

(a)

Meta-analyses and systematic reviews (46);

(b)

Repeated studies derived from the same trials (48);

(c)

Did not involve patients with DM (84);

(d)

Absence of a control group (non-diabetes) for comparison (18);

(e)

Did not involve ticagrelor monotherapy after a short course of DAPT (37);

(f)

Case studies (34).

Finally, 4 studies [13,14,15,16] derived from 3 randomized trials were included in this analysis (Fig. 1).

Fig. 1

Flow diagram showing the study selection for this meta-analysis

General features and baseline characteristics of the studies

The general features of the trials have been listed in Table 2. A total number of 22,574 participants were included in this analysis (4693 DM and 17,881 without DM). Participants were enrolled from years 2013 to 2019.

Table 2 General features of the studies

Two studies had a follow-up time period of 12 months whereas one study had a follow-up time period of 24 months. In addition, in three studies, the course of DAPT was 3 months followed by ticagrelor monotherapy while in one study, the course of DAPT was for 1 month followed by ticagrelor monotherapy.

The baseline features of the participants have been listed in Table 3. The mean age of the participants with DM ranged from 63.3 to 68.2 years, whereas the mean age of the participants without DM ranged from 60.0 to 67.4 years as shown in Table 3. Male participants varied from 69.3% to 76.6% in the DM group and 71.1% to 81.8% in the group without DM. The mean percentage of patients with previous MI ranged from 23.6% to 71.9%, previous stoke ranged from 2.80% to 7.40% and the mean body mass index varied from 24.9 to 30.1 kg/m2 as shown in Table 3. The mean percentages of participants with hypertension, dyslipidemia and those who were smokers have been listed in Table 3.

Table 3 Baseline features of the studiesMain results of the analysis

When analyzing for MACEs subgroup, events occurred in a total number of 265 DM participants out of 4668 patients with DM and 669 participants out of 17833 patients without DM and when analyzing for all-cause mortality, events occurred in a total number of 124 DM participants out of 4251 participants with DM and 336 participants out of 16721 participants without DM. When assessing for cardiovascular death, events occurred in a total number of 20 DM participants out of 1726 patients with DM and 13 participants out of 3325 patients without DM. For stroke, the values were 29 participants with DM out of 3058 patients with DM and 105 participants out of 14676 participants without DM.

Results of this analysis showed that in patients with ticagrelor monotherapy after a short course of DAPT, DM was associated with significantly higher risks of MACEs (RR: 1.73, 95% CI: 1.49 – 2.00; P = 0.00001; I2 = 0%), all-cause mortality (RR: 2.15, 95% CI: 1.73 – 2.66; P = 0.00001, I2 = 0%), cardiac death (RR: 2.82, 95% CI: 1.42 – 5.60; P = 0.003, I2 = 23%) and stroke (RR: 1.78, 95% CI: 1.16 – 2.74; P = 0.009, I2 = 24%) when compared to patients without DM as shown in Fig. 2. However, TVR was not significantly different (RR: 1.13, 95% CI: 0.89 – 1.43; P = 0.32, I2 = 0%). MI (RR: 1.63, 95% CI: 1.17 – 2.26; P = 0.004, I2 = 53%) and stent thrombosis (RR: 1.74, 95% CI: 1.03 – 2.94; P = 0.04, I2 = 56%) were also significantly higher in the DM group as shown in Fig. 3.

Fig. 2

Adverse cardiovascular outcomes with Ticagrelor monotherapy after a short course of dual antiplatelet therapy (DAPT) with Ticagrelor plus aspirin following percutaneous coronary intervention (PCI) in patients with versus without diabetes mellitus

Fig. 3

Myocardial infarction and stent thrombosis with Ticagrelor monotherapy after a short course of DAPT with Ticagrelor plus aspirin following PCI in patients with versus without diabetes mellitus

When assessing for the bleeding outcomes, events occurred in 81 participants out of 2154 patients with DM and 126 patients out of 4457 participants without DM for TIMI minor bleeding and 88 DM patients out of 2154 participants with DM versus 123 participants out of 4457 participants without DM were observed for TIMI defined major bleeding. And for BARC 2,3 or 5, events occurred in 256 DM participants out of 3858 participants with DM versus 969 patients out of 15660 participants without DM.

When the bleeding outcomes were compared, TIMI defined minor (RR: 1.25, 95% CI: 0.95 – 1.65; P = 0.11, I2 = 0%) was not significantly different as shown in Fig. 4. In addition, TIMI defined major bleeding (RR: 1.50, 95% CI: 0.90 – 2.49; P = 0.12, I2 = 69%), BARC 3c bleeding (RR: 1.31, 95% CI: 0.14 – 11.90; P = 0.81, I2 = 65%) and BARC 2, 3 or 5 (RR: 1.17, 95% CI: 0.85 – 1.62; P = 0.34, I2 = 82%) were not significantly different among patients with versus without DM as shown in Fig. 5.

Fig. 4

Thrombolysis in myocardial infarction (TIMI) defined minor bleeding observed with Ticagrelor monotherapy after a short course of DAPT with Ticagrelor plus aspirin following PCI in patients with versus without diabetes mellitus

Fig. 5

TIMI defined major bleeding and bleeding defined by the academic research consortium (BARC) observed with Ticagrelor monotherapy after a short course of DAPT with Ticagrelor plus aspirin following PCI in patients with versus without diabetes mellitus

Consistent results were shown throughout during sensitivity analysis. For a better clarity, using a method of exclusion, each study was excluded one by one and a new analysis was therefore carried out each time and the results which were obtained were carefully compared with the main results of this analysis to observe for any significant change. However, no significant change was observed when sensitivity analysis was carried out indicating that the results were not influenced by data from any particular study. For example, when study Global Leaders was excluded from the analysis and a new analysis was carried out, the significances of MACEs (RR: 1.70, 95% CI: 1.38 – 2.09; P = 0.00001), all-cause mortality (RR: 2.13, 95% CI: 1.42 – 3.20; P = 0.0002), MI (RR: 1.73, 95% CI: 1.32 – 2.26; P = 0.0001, TVR (RR: 0.88, 95% CI: 0.18 – 4.36, P = 0.88), stroke (RR: 2.44, 95% CI: 1.03 – 5.75, P = 0.04) and stent thrombosis (RR: 1.72, 95% CI: 1.31 – 2.26; P = 0.0001) were similar to the main results of this analysis, there was no significant difference in the result findings. This could explain the consistence of this sensitivity analysis.

Publication bias was represented by Figs. 6 and 7 whereby the funnel plots which were generated were visually analyzed based on their symmetrical aspect. There was only little evidence of publication bias across the studies which were considered during data analysis. The funnel plots which were generated were almost fully symmetrical and this could explain the low evidence of publication bias.

Fig. 6

Funnel plot showing publication bias (A)

Fig. 7

Funnel plot showing publication bias (B)