The literature search and include studies
A flowchart of the literature search was shown in Figure 1. Initially, in the primary search from the major databases, a total of 674 studies were included. After removing duplicates and screening title and abstracts, a total of 132 papers were remained, but 118 of them did not meet our purpose. The remaining 14 articles were assessed for eligibility based on full-text review, 9 were deemed ineligible. After qualitative and quantitative analysis, according to the inclusion criteria, only 5 studies published from 2015 to 2019 were selected for our meta-analysis [18-22].
The main characteristics of the included studies were listed in Table 1. A total of 3122 patients were included. These studies were all observation researchers and two conducted in Turkey [18, 19], three conducted in China [20-22]. The mean age of the patients ranges from 58.9 years old to 63.5 years old. One studies explicitly stated that there were no statistically significant differences of age between LMR groups [20], but four studies showed statistically significant differences of age [18, 19, 21, 22]. Three studies enrolled STEMI patients and all the patients underwent percutaneous coronary intervention (PCI) [18-20]. Two studies enrolled NSTEMI patients [21, 22]. One of the studies explicitly stated that the enrolled NSTEMI patient underwent PCI [21], while the other did not specify if enrolled patient underwent PCI [22]. Three studies reported the mortality [18, 19, 21], and two studies reported MACE [20, 22]. According to the Newcastle-Ottawa scale (NOS) [17], all cohort studies were of high quality and had scores of seven or more.
LMR and Mortality/MACE
The short-term was included in-hospital and 30-day. Others were defined as long-term. The combined analysis of 4 studies covering 2444 patients described the relationship between LMR and short-term mortality/MACE [18-20, 22], the result showed that LMR predicted short-term mortality/MACE (OR = 2.61, 95% CI: 1.15–5.94, P = 0.022, Figure 2A), with high heterogeneity among studies (I2 = 85.1%, p < 0.001). The combined analysis of 3 studies covering 1302 patients described the relationship between LMR and long-term mortality/MACE. The pooled outcome for low LMR value compared with high LMR value group was found to be 2.10 (95% CI: 1.06– 4.19, P = 0.035, Figure 2B), also with high heterogeneity among studies (I2 = 93.0%, p < 0.001).
Subgroup analysis
We conducted subgroup analysis to further analyze the association between LMR and mortality/MACE according to country of patients (Turkey and China), diseases of patients (STEMI and NSTEMI), sample size (≥600 and <600) and mean age (≥62 and <62). For 4 studies that researched LMR and short-term mortality/MACE [18-20, 22]. According to the results of subgroup analysis (Table 2), Low LMR predicted short-term mortality/MACE showed a statistical significance in Turkey researches (OR = 4.16, 95% CI: 2.32–7.46, p < 0.001), large sample size researches (≥600, OR = 3.50, 95% CI: 1.84–6.67, p < 0.001), younger patients researches (< 62, OR = 3.76, 95% CI: 2.29–6.18, p < 0.001). But there didn’t show a statistical significance in China ACS patients, STEMI patients and small sample size researches. According to the change of I2, the sources of heterogeneity among studies might be country, sample size and mean age of patients. For 3 studies that researched LMR and long-term mortality/MACE [19-21]. According to the results of subgroup analysis (Table 3), Low LMR predicted long-term mortality/MACE didn’t show a statistical significance in any subgroup. According to the change of I2, the sources of heterogeneity among studies might be country and mean age of patients.