Background: Heparin-induced thrombocytopenia (HIT) remains a challenging diagnosis especially in surgical intensive care unit (SICU) patients. The aim of the study was to evaluate for the first time the diagnostic accuracy of the HIT Expert Probability (HEP) score in the early identification of HIT in SICU patients. Methods: The HEP and 4Ts scores were calculated in all patients with suspected HIT during their stay in our SICU. The diagnosis of HIT was finally confirmed (HIT+ group) or excluded (HIT− group) by an independent committee blinded to the HEP and 4Ts score values. The primary outcome was the sensitivity and specificity of a HEP score ≥ 5 for the diagnosis of HIT. The secondary outcome was the area under the ROC curve (AUC) of the HEP and 4Ts scores in the diagnosis of HIT. Results: Respectively 6 and 113 patients were included in the HIT+ and HIT− groups. A HEP score value ≥ 5 had a sensitivity (95% confidence interval (95% CI)) of 1.00 (0.55–1.00), and a specificity (95% CI) of 0.92 (0.86–0.96). The AUC (95% CI) was significantly higher for the HEP score versus for the 4Ts score (0.967 (0.922–1.000) versus 0.707 (0.449–0.965); p = 0.035). Conclusions: A HEP score value < 5 could be helpful to rule out HIT in SICU patients.