The study aimed to revisit the in-hospital predictors of shunt thrombosis (ST) in the foreground of the pulmonary artery size in patients who received modified Blalock-Taussig shunt (mBTS) as the first-stage palliation. Data from 80 patients who received mBTS as their initial palliative procedure between February 2012 and January 2017 was retrospectively collected. The median age and weight of the patients at the time of their mBTS procedure was 4 days (IQR 2-22 days) and 3.2 kg (IQR 2.8-3.7 kg), respectively. Of the 80 patients in the study, 11 (13.8%) developed ST. The diameter and corresponding z scores of the pulmonary arteries were significantly lower in patients with ST. The median shunt size/shunted pulmonary artery size (S/PA) ratio was considerably higher in patients with ST. In logistic regression analysis, pulmonary artery hypoplasia (PAH) [odds ratio (OR) = 13.7 (0.06-0.21), p < 0.001], S/PA ratio ≥ 0.9 [OR = 8.1 (0.03-0.53), p = 0.03], prematurity [OR = 9.5 (0.05-0.33), p = 0.003], and shunt size/weight (S/W) ratio ≥ 1.3 [OR = 6.4 (0.04-0.67), p = 0.012] were found to have a significant impact on ST. The best combination of sensitivity and specificity of the S/W (0.73 and 0.75) and the S/PA ratio (0.73 and 0.80) were achieved at the cut-off value of 1.3 and 0.9, respectively. The Youden index of S/PA was 0.52. While the area under the curve (AUC) of the S/W ratio was 0.686 ± 0.12 (p = 0.049), the AUC of the S/PA ratio was 0.791 ± 0.08 (p = 0.002). In conclusion, instead of weight, considering the size of the target pulmonary artery and thereby, the S/PA ratio would be more instructive in determining shunt size. There were a high number of patients in our study who showed PAH having received a shunt size based on their body weight. By contrast, our results showed that the S/PA ratio of ≥ 0.9 would be a good predictor of in-hospital ST.