The risk of pulmonary thromboembolism (PTE) after total knee arthroplasty (TKA) with tourniquet is still controversial. In this study, we determined whether plasma D-dimer could be used as a marker of PTE, and whether intraoperative use of tourniquet is a risk factor for symptomatic PTE after TKA. We divided 86 patients (12 males and 74 females) who underwent primary TKA into four groups based on (i) operation method (unilateral/bilateral) and (ii) use of pneumatic tourniquet during surgery (with/without). Plasma D-dimer values were measured preoperatively and on Day 1 postoperatively. "D1-value" was defined as the difference between postoperative and preoperative plasma D-dimer levels. We determined the cutoff level of "D1-value" for symptomatic PTE, monitored by O(2) saturation (SAT) and lung scintigraphy. "D1-values" for patients with PTE symptoms were higher than those without symptoms (p<0.001). Furthermore, "D1-values" for TKA without tourniquet procedures were significantly lower than those with tourniquet both in unilateral (p=0.003) and bilateral groups (p=0.004). At a cutoff level of >or=8.6 mug/mL, the sensitivity, specificity, and positive and negative predictive values were 100%, 82.5%, 30.0% and 100% for symptomatic PTE, respectively. The results of our pilot study showed that elevated plasma "D1 value" is a sensitive marker for detection of PTE. Further studies will be needed, in order to evaluate the risk of tourniquet, and to reduce PTE after TKA.