Association of postoperative hemoglobin and lower extremity deep vein thrombosis after total knee arthroplasty: a focus on isolated intramuscular venous thrombosis

Abstract Background Total knee arthroplasty (TKA) is the standard treatment for end-stage knee osteoarthritis, but lower extremity deep vein thrombosis (DVT) remains a significant complication. While various DVT risk factors are established, the relationship between postoperative hemoglobin (Hb) levels and postoperative DVT risk remains controversial. This study investigated the association between postoperative Hb levels and lower extremity DVT following TKA. Methods This retrospective study included 832 TKA patients classified by postoperative anemia severity: moderate-to-severe (postoperative Hb ≤ 90g/L, n = 42), mild (90g/L < postoperative Hb ≤ 120g/L, n = 552), and non-anemia (postoperative Hb > 120 g/L, n = 238). Ultrasound-detected postoperative DVT was the primary endpoint and included isolated intramuscular venous thrombosis (IMVT), deep calf vein thrombosis (DCVT), and proximal deep vein thrombosis (PDVT). As the predominant phenotype, IMVT was the principal focus of the analysis. Clinically significant DVT, defined as DCVT or PDVT, was analyzed as a secondary endpoint reflecting more severe thrombotic involvement. Multivariate logistic regression assessed the postoperative Hb-DVT association with adjustments for confounders. Results The overall incidence of postoperative DVT was 21.1%, predominantly consisting of IMVT (19.8%), whereas clinically significant DVT was rare (DCVT, 1.0%; PDVT, 0.2%). The highest incidence (45.2%) occurred in the moderate-to-severe anemia group. Each 1 g/L increase in postoperative Hb was associated with a 3% decreased postoperative DVT risk (adjusted OR = 0.97, 95% CI 0.96–0.99, p = 0.011). Compared to moderate-to-severe anemia, mild anemia (OR = 0.29) and non-anemia (OR = 0.21) had significantly lower postoperative DVT risk. A linear dose–response relationship was observed between postoperative Hb and postoperative DVT risk, with no significant subgroup interactions. Adding postoperative Hb to the baseline risk model provided significant but limited improvement in postoperative DVT prediction (ΔAUC = 0.010, p = 0.339; NRI: 0.081, p = 0.015; IDI: 0.009, p = 0.021). Conclusion In this retrospective cohort, higher postoperative Hb levels were associated with a lower risk of lower-extremity DVT after TKA in patients with osteoarthritis, mainly driven by IMVT. Exploratory findings for clinically significant DVT should be interpreted cautiously because of the limited number of events.