In a cross-sectional study, 623 patients with non-valvular atrial fibrillation (NVAF) were examined, of which 59% had an association with inflammation. Systemic inflammation plays a key role in left atrial thrombus (LAT) formation. Multivariable logistic regression and ROC curve analysis were used to evaluate inflammatory indices derived from complete blood counts. Neutrophil-to-lymphocyte ratio (NLR), neutrophil-to-mean platelet volume ratio (NMR), systemic immune inflammation index (SII), systemic inflammatory response index (SIRI) and pan-immune risk index (PIV) were significantly associated with LAT after adjustment. These indices had moderate predictive accuracy with area under the ROC curve ranging from 0.582 to 0.601, with no significant differences between them. Other indexes, such as the ratio of monocytes to lymphocytes, the ratio of blood platelets to lymphocytes, or the ratio of white blood cells to the average volume of blood platelets, did not show predictive value. NLR, NMR, SII, SIRI, and PIV are readily available and cost-effective biomarkers for LAT risk stratification in NVAF patients, but require further validation.