Limited indication of postoperative anticoagulation based on age reduces the incidence of deep venous thrombosis after anterior cruciate ligament reconstruction: A retrospective study Authors: Masaki Nagashima, Ryo Sasaki, Kentaro Tanaka, Hiroyuki Nagita, Masamitsu Suhara, Mitsuru Yagi Published: 2026 ABSTRACT ObjectiveDeep venous thrombosis (DVT) has the potential to develop into life-threatening pulmonary embolism which is a serious complication after anterior cruciate ligament reconstruction (ACLR). Advanced age (≥40 years) is a known risk factor for DVT. Anticoagulation is an effective pharmacological prophylaxis for DVT, but its indications after ACLR have not been established. This study aimed to investigate the incidence of DVT after ACLR when postoperative anticoagulation is administered to patients ≥40 years of age.MethodsAnticoagulation group comprised 75 consecutive patients who underwent ACLR (including 5 revision surgeries) at our hospital between November 2021 and April 2024. Among those patients, patients ≥40 years old in Anticoagulation group received edoxaban at 30 mg/day on postoperative days 3–7 as pharmacologic prophylaxis for DVT. The historical control group not administered anticoagulation (No anticoagulation group) comprised 56 patients with 56 ACLRs (including 10 revision surgeries) performed at our hospital between April 2017 and September 2018. Propensity score matching was performed for age, and 48 patients were selected from each group and enrolled in this study. There were 16 patients in Anticoagulation group and 17 in No anticoagulation group who were ≥40 years old. DVT was diagnosed using ultrasonography of both legs performed at 1 week postoperatively. Complications associated with anticoagulation and hemoglobin (Hb) levels in patients ≥40 years old were investigated.ResultsAfter ACLR, DVT was detected in 2 patients (4.2%) in Anticoagulation group and 9 patients (18.8%) in No anticoagulation group. The incidence of DVT was thus significantly lower in Anticoagulation group (P = 0.025). All patients with DVT were asymptomatic. With regard to Hb levels in patients ≥40 years old, no significant difference was evident between groups. No complications associated with anticoagulation therapy, such as bleeding, were observed.ConclusionAnticoagulation administered to patients ≥40 years of age significantly reduced the incidence of DVT after ACLR without obvious adverse events. A limited indication of postoperative anticoagulation based on age may be beneficial in preventing DVT after ACLR. Abbreviations anterior cruciate ligament anterior cruciate ligament reconstruction body mass index contrast-enhanced computed tomography confidence interval deep venous thrombosis hemoglobin low-molecular-weight heparin pulmonary embolism postoperative day ultrasonography venous thromboembolism Introduction Arthroscopic anterior cruciate ligament (ACL) reconstruction (ACLR) is widely performed and beneficial for treating rupture of the ACL. However, one potential postoperative complication is venous thromboembolism (VTE), including deep venous thrombosis (DVT) and pulmonary embolism (PE). DVT has the potential to develop into life-threatening PE, and several case reports have described fatal PE following ACLR.1,2 The incidence of DVT after ACLR has been reported as 6.6–16.4% in studies examining DVT with contract venography and ultrasonography (US).3, 4, 5, 6, 7, 8, 9 Although the incidence of PE after ACLR has been reported to range from 0.05% to 0.21% in studies using large databases,10, 11, 12, 13 examination of PE by contrast-enhanced computed tomography (CECT) found an incidence of 7.3% for asymptomatic PE.14 The incidence of PE in patients with DVT appears high, and even asymptomatic DVT requires careful attention. In a previous study, from a total of 55 patients undergoing ACLR, 9 (16.4%) had asymptomatic DVT, of whom 4 (44%) had asymptomatic PE.8 Moreover, asymptomatic DVT is reportedly associated with post-thrombotic syndrome.15 Although, to our knowledge, routine screening for DVT has not been shown to reduce the rate of fatal PE, prevention of DVT after ACLR is thus considered an important issue. Anticoagulation, such as with low-molecular-weight heparin (LMWH), reportedly offers effective pharmacological prophylaxis for DVT after ACLR.9,16,17 However, whether any benefits from routine use of LMWH outweigh any risks, such as bleeding, and costs remains controversial.18 According to the American College of Chest Physicians guidelines, pharmacological prophylaxis is recommended for patients with a history of VTE, and may also be effective in the high-risk population for DVT after arthroscopic surgery.19 Risk factors for DVT after ACLR include tourniquet use,6,20 tobacco use,12,21 and additional surgery,4,6,11 as well as advanced age.10,13,20 In patients ≥40 years old, a high odds ratio has been reported for risk of DVT following ACLR.20 Pharmacological prophylaxis for patients with age ≥40 years may be effective, and several studies described that pharmacological prophylaxis should be administered to patients of advanced age.7,13 However, the incidence of DVT under this approach is unknown. The purpose of this study was to investigate the incidence of DVT after ACLR when postoperative anticoagulation is administered to patients ≥40 years of age. We hypothesized that pharmacological prophylaxis limited to patients ≥40 years of age would lead to a lower incidence of DVT after ACLR. Materials and methods Patients This retrospective observational study investigated and compared two groups. Seventy-five consecutive patients underwent single-bundle ACLR (including 5 revision surgeries) at our hospital between November 2021 and April 2024. All of these patients were initially included in this study as Anticoagulation group. In Anticoagulation group, anticoagulation as prophylaxis for DVT after ACLR was conducted only for patients ≥40 years of age. The historical control group (No anticoagulation group) consisted of 56 patients with single-bundle ACLR (including 10 revision surgeries) performed at our hospital between April 2017 and September 2018. In No anticoagulation group, pharmacological prophylaxis for DVT was not administered. The exclusion criteria were the same for both groups, as follows: 1) patients with a history of VTE (n = 1 in No anticoagulation group); 2) patients with DVT diagnosed by preoperative US (n = 1 in Anticoagulation group); 3) patients without adequate data (n = 2 in Anticoagulation group); or 4) patients who required additional treatment for other illnesses within 7 days postoperatively (n = 1 in Anticoagulation group). No patients in either group required additional ligament surgery. Propensity score matching was performed for 71 knees in Anticoagulation group and 55 knees in No anticoagulation group by patient age, resulting in 48 patients being selected from each group and enrolled in this study. There were 16 patients in Anticoagulation group and 17 in No anticoagulation group who were ≥40 years old. This study was approved by our institutional review board (filing no. 5-16-55), and the need to obtain written informed consent was waived by the review board due to the retrospective nature of the study. Surgical procedures and postoperative care All ACLRs were performed by three experienced surgeons, with the patient under general anesthesia with femoral and sciatic nerve blocks. Both groups underwent the same ACLR procedures with use of a tourniquet only during arthroscopic procedures. At first, the graft was harvested and prepared without the use of a tourniquet. The arthroscopic procedure was then started using a tourniquet. Tourniquet pressure was set at 250–280 mmHg, and the affected limb was exsanguinated prior to tourniquet pressurization. ACLs were reconstructed using single-bundle hamstring autografts. For graft preparation, commercially available polyester tape (Leeds-Keio artificial ligament; Neoligaments, Leeds, UK) was connected at the distal end of the graft, and an Endobutton CL BTB or Ultra Button (Smith & Nephew, Andover, MA) was attached at the proximal end of the graft. The graft was introduced through each tibial tunnel and the femoral tunnel, which was made with the transportal technique. The diameter of the bone tunnel was matched to the diameter of the hamstring autograft, which was 8–10 mm. Graft fixation was performed with the Endobutton or Ultra Button on the femoral side and two staples on the tibial side. In revision ACLR, the hamstring autograft was harvested from the contralateral side, and ACLR was performed in the same manner as the primary ACLR. If meniscus tears existed and were repairable, repair was performed using an all-inside technique with the FAST-FIX360 meniscal repair system (Smith & Nephew) or an outside-in technique. Middle to posterior meniscus injuries were repaired with the all-inside technique, and anterior meniscus injuries were repaired with the outside-in technique. A closed-suction drain was installed and left in the knee for 24 h. The knee was postoperatively immobilized with a brace for 1 day for patients without meniscus repair, and for 7 days for patients with meniscus repair. After knee immobilization, range-of-motion and muscle-strengthening exercises were started. On postoperative day (POD) 7, full weight-bearing with a hinged knee brace was permitted. For DVT prophylaxis, both groups were treated similarly except for pharmacological prophylaxis. Intermittent pneumatic compression for the foot was applied to the contralateral limb during ACLR. Intermittent pneumatic compressions for bilateral feet were used for 1 day, then compression stockings covering the entirety of both lower limbs were used for 7 days after ACLR. In Anticoagulation group, edoxaban was administered at 30 mg once daily from POD 3 to POD 7 as oral pharmacologic prophylaxis for DVT only for those patients ≥40 years old. Diagnosis and management of DVT US was routinely performed on both legs both preoperatively and 1 week after surgery for all patients using a TUS-X100 system (CANON MEDICAL SYSTEMS, Tochigi, Japan) to detect DVT. In this study, DVT was defined as venous thrombus located inside the deep fascia. DVT was diagnosed using both a compression technique and color Doppler evaluation. US was performed and evaluated by experienced vascular surgeons. Any DVTs diagnosed were treated with oral anticoagulation therapy with edoxaban at 60 mg once daily, with the dose reduced to 30 mg once daily if the patient weighed less than 60 kg or showed creatinine clearance under 50 ml/min. Edoxaban treatment was continued for at least 3 months. Outcomes and data collection The primary outcome of this study was the incidence of DVT. As hemoglobin (Hb) levels might be affected by anticoagulation in patients ≥40 years old, the change in Hb (ΔHb) was defined as a secondary outcome in this study. The ΔHb from POD 1 to POD 7 was defined as follows: ΔHb = Hb on POD 1 - Hb on POD 7. Preoperative clinical and surgical data including age, sex, body mass index (BMI), current smoking status, interval from injury to ACLR (less than 3 months or equal to or more than 3 months), primary or revision ACLR, operative time, tourniquet time, and presence or absence of meniscus repair were recorded. In addition, complications associated with anticoagulation, such as bleeding or delayed rehabilitation, were investigated. Statistical analysis In addition to the primary and secondary outcomes, clinical factors were compared between groups using Student's t-test, the Mann-Whitney U test, and the χ2 test. The comparisons were conducted for both all patients and patients ≥40 years old. Values of P < 0.05 were considered significant. These statistical analyses were performed using BellCurve for Excel software version 4.06 (Social Survey Research Information, Tokyo, Japan). A power analysis was performed for the primary outcome with a significance level of α = 0.05. The power analysis and propensity score matching were performed with R version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria). Results For preoperative clinical data, no significant differences were seen between groups other than for BMI (Table 1). In patients ≥40 years old, revision ACLRs were significantly more frequently performed in No anticoagulation group (Table 2). In terms of operative data, no significant differences in operative time or tourniquet time were evident, but meniscal repairs were significantly more frequent in Anticoagulation group (Table 3). After ACLR, DVT was detected in 2 patients (4.2%) in Anticoagulation group and 9 patients (18.8%) in No anticoagulation group, representing a significant difference (P = 0.025) (Power = 0.602). In patients ≥40 years old, the incidence of DVT was significantly lower in Anticoagulation group, with 1 patient (6.3%) in Anticoagulation group and 6 patients (35.3%) in No anticoagulation group (P = 0.041) (Power = 0.460) (Table 4). On the other hand, there was no significant difference between the two groups in patients <40 years old (P = 0.286).Table 1Preoperative clinical and surgical demographics of both groups.Table 1Anticoagulation group (n = 48)No anticoagulation group (Control; n = 48)PAge, years32.6 ± 14.632.4 ± 15.20.935Age, n<40 years32310.830≥40 years1617Sex, nFemale19220.536Male2926BMI (kg/m2)24.7 ± 4.622.8 ± 2.50.012Current smoking status, nNonsmoker43450.460Smoker53Interval from injury to ACLR, n<3 months27200.153≥3 months2128Type of ACLR, nPrimary44380.083Revision410ACLR = anterior cruciate ligament reconstruction; BMI = body mass index. For preoperative clinical and surgical data, no significant differences were seen between groups other than for BMI.Table 2Preoperative clinical and surgical demographics in patients ≥40 years old for both groups.Table 2Anticoagulation group (n = 16)No anticoagulation group (Control; n = 17)PAge, years50.6 ± 6.250.9 ± 5.80.903Sex, nFemale670.829Male1010BMI (kg/m2)24.2 ± 3.823.3 ± 2.80.431Current smoking status, nNonsmoker14160.509Smoker21Interval from injury to ACLR, n<3 months640.383≥3 months1013Type of ACLR, nPrimary16120.019Revision05ACLR = anterior cruciate ligament reconstruction; BMI = body mass index. In patients ≥40 years old, no significant differences were noted between groups except for the type of ACLR.Table 3Incidence of deep venous thrombosis and surgical data for both groups.Table 3Anticoagulation group (n = 48)No anticoagulation group (Control; n = 48)PDVT, n (%)2 (4.2)9 (18.8)0.025Operative time (min)78.7 ± 16.882.0 ± 19.00.380Tourniquet time (min)67.4 ± 12.369.3 ± 15.00.509Meniscal repair, nYes3917<0.001No931DVT = deep venous thrombosis. No significant differences in operative time or tourniquet time were evident, but meniscal repairs were significantly more frequent in Anticoagulation group. The incidence of DVT was significantly lower in Anticoagulation group.Table 4Incidence of deep venous thrombosis, surgical data, and perioperative Hb changes in patients ≥40 years old for both groups.Table 4Anticoagulation group (n = 16)No anticoagulation group (Control; n = 17)95% CI for the differencePDVT, n (%)1 (6.3)6 (35.3)(3.4 – 54.7)0.041Operative time (min)79.4 ± 13.283.0 ± 24.8−10.6 – 17.80.607Tourniquet time (min)70.4 ± 10.172.4 ± 19.8−9.2 – 13.30.715Meniscal repair, n (%)Yes13 (81.2)2 (11.8)(9.4 – 45)<0.001No3 (18.8)15 (88.2)Preoperative Hb (g/dL)13.8 ± 1.514.3 ± 1.5−0.6 – 1.50.384Hb on POD 1 (g/dL)12.5 ± 1.513.2 ± 1.7−0.4 – 1.90.189Hb on POD 7 (g/dL)12.8 ± 1.613.0 ± 1.4−0.8 – 1.30.672ΔHb (g/dL)−0.4 ± 0.70.2 ± 0.60.1 – 1.00.026CI = confidence interval; DVT = deep venous thrombosis; Hb = hemoglobin; POD, postoperative day; ΔHb = POD 1 Hb – POD7 Hb. In patients ≥40 years old, the incidence of DVT was significantly lower in Anticoagulation group. With regard to Hb levels, ΔHb was significantly lower in Anticoagulation group. Table 1Preoperative clinical and surgical demographics of both groups.Table 1Anticoagulation group (n = 48)No anticoagulation group (Control; n = 48)PAge, years32.6 ± 14.632.4 ± 15.20.935Age, n<40 years32310.830≥40 years1617Sex, nFemale19220.536Male2926BMI (kg/m2)24.7 ± 4.622.8 ± 2.50.012Current smoking status, nNonsmoker43450.460Smoker53Interval from injury to ACLR, n<3 months27200.153≥3 months2128Type of ACLR, nPrimary44380.083Revision410ACLR = anterior cruciate ligament reconstruction; BMI = body mass index. For preoperative clinical and surgical data, no significant differences were seen between groups other than for BMI. Preoperative clinical and surgical demographics of both groups. ACLR = anterior cruciate ligament reconstruction; BMI = body mass index. For preoperative clinical and surgical data, no significant differences were seen between groups other than for BMI. Table 2Preoperative clinical and surgical demographics in patients ≥40 years old for both groups.Table 2Anticoagulation group (n = 16)No anticoagulation group (Control; n = 17)PAge, years50.6 ± 6.250.9 ± 5.80.903Sex, nFemale670.829Male1010BMI (kg/m2)24.2 ± 3.823.3 ± 2.80.431Current smoking status, nNonsmoker14160.509Smoker21Interval from injury to ACLR, n<3 months640.383≥3 months1013Type of ACLR, nPrimary16120.019Revision05ACLR = anterior cruciate ligament reconstruction; BMI = body mass index. In patients ≥40 years old, no significant differences were noted between groups except for the type of ACLR. Preoperative clinical and surgical demographics in patients ≥40 years old for both groups. ACLR = anterior cruciate ligament reconstruction; BMI = body mass index. In patients ≥40 years old, no significant differences were noted between groups except for the type of ACLR. Table 3Incidence of deep venous thrombosis and surgical data for both groups.Table 3Anticoagulation group (n = 48)No anticoagulation group (Control; n = 48)PDVT, n (%)2 (4.2)9 (18.8)0.025Operative time (min)78.7 ± 16.882.0 ± 19.00.380Tourniquet time (min)67.4 ± 12.369.3 ± 15.00.509Meniscal repair, nYes3917<0.001No931DVT = deep venous thrombosis. No significant differences in operative time or tourniquet time were evident, but meniscal repairs were significantly more frequent in Anticoagulation group. The incidence of DVT was significantly lower in Anticoagulation group. Incidence of deep venous thrombosis and surgical data for both groups. DVT = deep venous thrombosis. No significant differences in operative time or tourniquet time were evident, but meniscal repairs were significantly more frequent in Anticoagulation group. The incidence of DVT was significantly lower in Anticoagulation group. Table 4Incidence of deep venous thrombosis, surgical data, and perioperative Hb changes in patients ≥40 years old for both groups.Table 4Anticoagulation group (n = 16)No anticoagulation group (Control; n = 17)95% CI for the differencePDVT, n (%)1 (6.3)6 (35.3)(3.4 – 54.7)0.041Operative time (min)79.4 ± 13.283.0 ± 24.8−10.6 – 17.80.607Tourniquet time (min)70.4 ± 10.172.4 ± 19.8−9.2 – 13.30.715Meniscal repair, n (%)Yes13 (81.2)2 (11.8)(9.4 – 45)<0.001No3 (18.8)15 (88.2)Preoperative Hb (g/dL)13.8 ± 1.514.3 ± 1.5−0.6 – 1.50.384Hb on POD 1 (g/dL)12.5 ± 1.513.2 ± 1.7−0.4 – 1.90.189Hb on POD 7 (g/dL)12.8 ± 1.613.0 ± 1.4−0.8 – 1.30.672ΔHb (g/dL)−0.4 ± 0.70.2 ± 0.60.1 – 1.00.026CI = confidence interval; DVT = deep venous thrombosis; Hb = hemoglobin; POD, postoperative day; ΔHb = POD 1 Hb – POD7 Hb. In patients ≥40 years old, the incidence of DVT was significantly lower in Anticoagulation group. With regard to Hb levels, ΔHb was significantly lower in Anticoagulation group. Incidence of deep venous thrombosis, surgical data, and perioperative Hb changes in patients ≥40 years old for both groups. CI = confidence interval; DVT = deep venous thrombosis; Hb = hemoglobin; POD, postoperative day; ΔHb = POD 1 Hb – POD7 Hb. In patients ≥40 years old, the incidence of DVT was significantly lower in Anticoagulation group. With regard to Hb levels, ΔHb was significantly lower in Anticoagulation group. All patients with DVT were asymptomatic. In Anticoagulation group, DVT was found in the soleal vein on the operated side in both patients. In No anticoagulation group, DVT was found on the operated side in 8 of the 9 patients and was located at the soleal vein in 5 patients, peroneal vein in 1 patient, soleal and peroneal veins in 1 patient, and soleal, peroneal, posterior tibial, and popliteal veins in 1 patient. In the 1 patient in No anticoagulation group who underwent revision ACLR, the DVT was found in both soleal and peroneal veins of both legs. Patients with DVTs in Anticoagulation group were 25- and 44-year-old healthy men with no risk factors for DVT other than age. With regard to Hb levels in patients ≥40 years old, ΔHb was significantly lower in Anticoagulation group with pharmacological prophylaxis (Table 4). No complications associated with anticoagulation were observed in Anticoagulation group. Discussion The results of the present study supported our hypothesis that pharmacological prophylaxis limited to patients ≥40 years old would lead to a lower incidence of DVT after ACLR. All detected DVTs were asymptomatic and most were distal. In addition, patients ≥40 years old, comprising slightly less than half of the patients in this study, did not experience significant Hb reductions or major bleeding events due to this pharmacological prophylaxis. Edoxaban for pharmacological prophylaxis was administered orally for as short as 5 days postoperatively. Pharmacological prophylaxis limited to patients ≥40 years of age may be a beneficial method for applying DVT prophylaxis after ACLR. Pharmacological prophylaxis such as with LMWH is reportedly effective for reducing the risk of DVT after ACLR.9,16,17 Zhu et al. conducted a meta-analysis of 8 randomized controlled trials and reported that LMWH reduced the risk of VTE in patients undergoing ACLR, but did not increase the risk of bleeding events.22 However, whether the benefits of LMWH outweigh the risks and costs remains controversial.18 Some reports have described negative findings for routine pharmacological prophylaxis.23,24 According to the American College of Chest Physicians guidelines, pharmacological prophylaxis after arthroscopic surgery is suggested for patients at high risk of DVT, such as those with a history of VTE.19 A recent systematic review by Dave et al. stated that factor Xa and LMWH administration may be appropriate for patients at high risk of DVT.16 In this study, pharmacological prophylaxis is initiated on POD 3 to minimize complications such as postoperative bleeding. However, the anticoagulation is generally administered earlier. Currently, we aim to initiate anticoagulation even earlier to enhance its efficacy. The duration of the anticoagulation in this study was five days, shorter than the standard prophylactic period. On POD 7 of the final administration, the presence of DVT was examined. If DVT was detected, anticoagulation was increased to a therapeutic dose. Conversely, if no DVT was observed, anticoagulation was considered unnecessary and discontinued at this point, as sufficient activity of patients, such as walking, was achieved. Risk factors for DVT after ACLR include tourniquet use,6,20 tobacco use,12,21 non-weight-bearing status,25 hypertension requiring medication,11 wound infection,11 and additional surgery,4,6,11,12,26 as well as advanced age.3,4,8,10, 11, 12, 13,20,26 Regarding tourniquet use, the incidence of DVT after ACLR reportedly improved from 16.4% to 1.7% after shifting from tourniquet use to non-tourniquet use.20 A large database study by Cancienne et al. found that patients using tobacco products had a significantly higher incidence of DVT after ACLR compared to matched controls, with an odds ratio of 1.9. They also reported that tobacco use significantly increased the rates of infection and subsequent ACLR along with DVT.21 Kanaya et al. reported that DVT occurred in 29.5% of 112 patients who underwent ACLR, and non-weight-bearing status at 1 week after ACLR was an independent risk factor for DVT.25 Hypertension requiring medication and wound infection were reported as risk factors for DVT after ACLR along with older age in a study by Bokshan et al. using a database of 9146 cases.11 Additional surgeries that significantly increase the incidence of DVT after ACLR include additional ligament reconstructions,6,12,26 microfracture,11 and high tibial osteotomy.11,12 Although advanced age has been reported as a risk factor for DVT after ACLR in many studies, categories have varied widely, including ≥30 years,3,11 ≥35 years,12 ≥40 years,10,13,20 and ≥45 years.26 Pharmacological prophylaxis was administered to patients ≥40 years old in the present study, considering the cost and the reported odds ratio of 8.3 for risk of DVT after ACLR for patients ≥40 years of age.20 Edoxaban is a factor Xa inhibitor and has been reported to be effective and safe for preventing DVT after total knee arthroplasty and total hip arthroplasty.27,28 A systematic review by Feng et al. found that compared to enoxaparin, an LMWH, edoxaban was more effective at preventing DVT with a similar risk of bleeding.27 Further, while most LMWHs are administered subcutaneously, edoxaban can be administered orally, has a fixed dose, and does not require routine coagulation monitoring.29 Given these advantages, edoxaban was used for pharmacological prophylaxis in this study. Among the risk factors for DVT mentioned above, whether ACLR is performed with or without a tourniquet is a factor that can be determined by the surgeon. A tourniquet is useful for securing the surgical field and was used in this study. Use of a tourniquet during ACLR has been reported to reduce soft tissue damage and total blood loss compared to no tourniquet use.30 In addition, Zaid et al. performed a randomized clinical trial with and without tourniquet use during ACLR and reported that tourniquet use significantly improved intraoperative visualization, shortened operative time, and decreased intraoperative sterile saline consumption and blood loss without serious adverse events.31 However, the use of a tourniquet is a risk factor for DVT,20 and we believe that ACLR without tourniquet should be considered if the surgical field has been secured. This study included patients of both primary and revision ACLRs as research subjects. The main differences between primary and revision ACLR at our hospital were that the graft tendon was harvested from the contralateral side and the staples placed during the primary ACLR needed to be extracted in revision ACLR. Postoperative care is identical for both. The procedures associated with DVT occurrence were considered to be almost the same between primary and revision ACLR in this study. Revision ACLR was not reported to influence the risk of VTE.13 This study showed some limitations that should be considered when interpreting the results. First, the number of patients was small and the study was not conducted as a randomized controlled trial. In addition, historical patients were used as controls, and the study periods for the two groups were separated by about three years. There is a potential for bias due to differences in the time period. Second, in the preoperative clinical and operative data, BMI, meniscal repairs, and type of ACLR were significantly different between the two groups and may have influenced the results. In particular, the postoperative care for meniscus repair was slightly different. Although we should have performed propensity score matching including the above variables in addition to age, the small sample size of this study made this difficult. However, neither of these has been reported as a significant factor in VTE development after ACLR in previous comparisons between groups with and without VTE.5,7,8,11, 12, 13 Third, US evaluation was conducted only once at one week after the ACLR. Additionally, although the site of DVT was diagnosed by US, the length of the DVT was not measured. Fourth, postoperative development of hemarthrosis is another concern regarding safety for patients receiving anticoagulation for VTE, but was not evaluated in this study. Although there was no significant difference in ΔHb between the two groups, further safety evaluation of anticoagulation is necessary. Conclusion In this study, pharmacological prophylaxis limited to patients ≥40 years of age significantly reduced the incidence of DVT after ACLR without obvious adverse events. A limited indication of postoperative anticoagulation based on age may be beneficial in preventing DVT after ACLR. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest The authors have no conflicts of interest relevant to this article.