Objective:To analyze risk factors for the occurrence and progression of isolated distal deep vein thrombosis (IDDVT) of lower limbs, and to explore the predictive value of dynamic changes of coagulation index D-dimer on the occurrence and progression of IDDVT in acute brain injury (ABI) patients during perioperative period.Methods:A retrospective case-control study was conducted. Perioperative ABI patients admitted to department of neurocritical care unit (NCCU) of the First Affiliated Hospital of University of Science and Technology of China from September 2019 to May 2020 were enrolled. Patients' baseline characteristics, disease characteristics, treatment approaches, outcomes and coagulation function index at 1, 2-4, 5-7 and > 7 days post operation were analyzed between patients with IDDVT and patients with progressive IDDVT. Risk factors for IDDVT occurrence and progression were identified by multivariate Logistic regression. Receiver operating characteristic curve (ROC curve) were drawn to assess the predictive value of coagulation indexes for IDDVT occurrence and progression.Results:A total of 164 ABI patients were enrolled. Most of the patients were elderly [age was 60 (51, 69) years], male [99 cases (60.4%)], and severe cases [Glasgow coma score (GCS) at admission was 6 (5, 8)]. The rates of IDDVT occurrence and progression were 61.6% (101 cases) and 16.8% (17 cases), respectively, the rate of proximal deep venous thrombosis (DVT) was 12.8% (21 cases). Compared with the IDDVT group (101 patients), patients without IDDVT group were younger (years: 55±13 vs. 62±13), length of intensive care unit (ICU) stay were shorter (days: 12±6 vs. 15±7), body mass index (BMI) and GCS at admission were higher [59 patients, BMI (kg/m 2): 23±5 vs. 19±8, GCS scores: 7±2 vs. 6±2], the differences were statistically significant (all P < 0.05). Compared with patients with IDDVT progression group, male patients were fewer [61.9% (52/84) vs. 88.2% (15/17)], the proportion of transfusion of red blood cell and anticoagulant therapy were lower [8.3% (7/84) vs. 29.4% (5/17) and 47.6% (40/84) vs. 94.1% (16/17)], the proportion of cerebral herniation was higher [42.9% (38/84) vs. 11.8% (2/17)] in patients without IDDVT progressive group. All of the differences were statistically significant (all P < 0.05). D-dimer were increased in two groups of whether IDDVT occurrence or not over time. D-dimer peaked on 5-7 days after surgery in IDDVT occurrence group, and then decreased. D-dimer peaked at > 7 days after surgery in patients without IDDVT. With time, D-dimer were increased in groups of whether IDDVT progression or not, both peaked at 5-7 days postoperation, and then decreased. Compared with non-IDDVTgroup, D-dimer was significantly increased in IDDVT group from 2-4 days after surgery [mg/L: 4.1 (2.3, 8.0) vs. 2.4 (1.7, 3.4), P < 0.05], and lasted until 5-7 days [mg/L: 5.5 (3.3, 11.4) vs. 3.9 (2.6, 5.8), P < 0.05]. Compared with IDDVT group, D-dimer was significantly increased in IDDVT progressive group from 2-4 days [mg/L: 11.2 (4.7, 20.0) vs. 3.7 (2.1, 6.8), P < 0.05], and lasted until 7 days [mg/L: 11.0 (3.0, 18.9) vs. 4.1 (2.6, 6.5), P < 0.05]. Multivariate Logistic regression analysis showed that age > 60 years [odds ratio ( OR) = 3.43, 95% confidence interval (95% CI) was 1.69-6.96, P = 0.001], GCS score at admission > 8 ( OR = 0.35, 95% CI was 0.17-0.76, P = 0.008), length of ICU stay > 13 days ( OR = 2.25, 95% CI was 1.08-4.70, P = 0.031) were risk factors for IDDVT. Gender ( OR = 0.19, 95% CI was 0.02-0.71, P = 0.019), transfusion of red blood cell ( OR = 6.50, 95% CI was 1.33-31.94, P = 0.021), cerebral herniation ( OR = 0.18, 95% CI was 0.37-0.90, P = 0.036) were risk factors for IDDVT profression. ROC curve analysis showed that age and D-dimer at 5-7 days were predicators of IDDVT [the area under curve ROC (AUC) were 0.68 and 0.72, 95% CI were 0.60-0.75 and 0.64-0.80, both P value were 0.000 1]. When the cut-off value of age was 60 years old and the D-dimer was 5.4 mg/L, the sensitivity were 60.6% and 54.4%, specificity were 71.2% and 80.9%, respectively, positive predictive value were 78.7%, 84.5%, negative predictive value were 51.2%, 48.1%, respectively. The elevation of D-dimer to 3.9 times at days 5-7 compared with day 1 of NCCU stay was a predicator of IDDVT progression (AUC = 0.81, 95% CI was 0.71-0.88, P = 0.000 1). The sensitivity, specificity, positive predictive value and negative predictive value were 76.5%, 74.6%, 41.9% and 93.0%, respectively. Conclusions:IDDVT occurrence and progressiveare common in severe ABI patients during perioperative period. The dynamic change of D-dimer, especially at days 5-7, is a valuable predictor of IDDVT progressionin ABI patients, which is helpful for guiding implementation of deep vein ultrasound of lower limb.