Table 2.
Subgroup analyses of association between NAFLD and risk of incident AF.
| Subgroup | No. of studies | HR (95%CI) | P association | I2 (%) | P heterogeneity |
|---|---|---|---|---|---|
| Study location | |||||
| Europe | 6 | 1.13 (1.00-1.28) | 0.05 | 55 | 0.05 |
| Asia | 4 | 1.24 (1.06-1.44) | 0.006 | 97 | < 0.00001 |
| North America | 2 | 0.86 (0.67-1.11) | 0.25 | 0 | < 0.00001 |
| Diagnostic methods of NAFLD | |||||
| Imaging techniques | 5 | 1.31 (0.80-2.12) | 0.28 | 59 | 0.04 |
| Diagnostic codes | 2 | 1.00 (0.71-1.40) | 0.99 | 77 | 0.04 |
| FLI | 5 | 1.19 (1.13-1.25) | < 0.00001 | 97 | < 0.00001 |
| Methods of AF verification | |||||
| ECG | 4 | 1.36 (0.79-2.33) | 0.27 | 70 | 0.02 |
| Diagnostic codes | 6 | 1.19 (1.13-1.25) | < 0.00001 | 96 | < 0.00001 |
| Medical records | 2 | 0.82 (0.60-1.11) | 0.20 | 0 | 0.81 |
| Follow-up time | |||||
| ≥ 5 years | 9 | 1.19 (1.13-1.25) | < 0.00001 | 95 | < 0.00001 |
| < 5 years | 3 | 1.12 (1.02-1.23) | 0.02 | 0 | 0.55 |
| Sample size | |||||
| >10000 | 7 | 1.18 (1.12-1.24) | < 0.00001 | 96 | < 0.00001 |
| <10000 | 5 | 1.31 (0.81-2.12) | 0.28 | 59 | 0.04 |
| Average age of participants | |||||
| ≥ 55 years | 6 | 1.10 (1.00-1.20) | 0.05 | 39 | 0.14 |
| < 55 years | 6 | 1.20 (1.04-1.39) | 0.01 | 95 | < 0.00001 |
| Number of gender | |||||
| Male ≥ Female | 6 | 1.22 (1.02-1.47) | 0.03 | 95 | < 0.00001 |
| Male < Female | 6 | 1.12 (1.02-1.23) | 0.02 | 68 | 0.008 |
| Study quality | |||||
| High-quality | 10 | 1.18 (1.12-1.25) | < 0.00001 | 94 | < 0.00001 |
| Moderate-quality | 2 | 1.13 (1.03-1.24) | 0.01 | 0 | 0.88 |
| Adjustment for confounders | |||||
| Adjusted | 11 | 1.18 (1.12-1.23) | < 0.00001 | 93 | < 0.00001 |
| Unadjusted (Crude) | 10 | 1.65 (1.34-2.04) | < 0.00001 | 98 | < 0.00001 |
HR, hazard ratio; CI, confidence interval; NAFLD, nonalcoholic fatty liver disease; FLI, Fatty liver index; AF, atrial fibrillation; ECG, electrocardiogram.