Breaking the blockage by advancing treatment strategies for upper extremity deep vein thrombosis Authors: Suchit Chidurala, Mira Patel, Evan Farrell, Luke D. Perry, Lori L. Pounds, Matthew J. Sideman, James C. Andersen Published: 2026 ABSTRACT BackgroundCatheter-directed thrombolysis (CDT) and percutaneous mechanical thrombectomy (PMT) are established endovascular strategies for the treatment of deep vein thrombosis (DVT). However, comparative outcomes data specific to upper extremity DVT (UEDVT) remain limited. Given that PMT is associated with improved outcomes in lower extremity DVT, we hypothesized that PMT would be associated with improved mortality and morbidity compared with CDT in treating UEDVT. This study evaluates outcomes associated with CDT vs PMT for UEDVT and Paget-Schroetter syndrome (PSS).MethodsA retrospective cohort study was conducted using the TriNetX US Collaborative Network. Patients diagnosed with UEDVT who underwent CDT (n = 1399) or PMT (n = 1406) between 2005 and 2025 were identified. Patients who received both modalities were excluded. Propensity score matching was performed, yielding 1198 patients in each cohort (n = 2396) balanced across demographics, comorbidities (including malignancy, type 2 diabetes mellitus, hypertension, coagulation disorders, and end-stage renal disease), and medication use. A subgroup analysis was performed for patients with PSS who underwent CDT or PMT, with similar propensity score matching (n = 280). Thirty-day and 1-year outcomes were compared using odds ratios (ORs).ResultsAfter matching, pulmonary embolism rates did not differ significantly between groups at 30 days or 1 year. Compared with PMT, CDT was associated with significantly higher OR of 30-day mortality (OR, 1.82), myocardial infarction (OR, 2.43), ischemic stroke (OR, 9.11), transfusion (OR, 1.95), 30-day readmission (OR, 2.34), and intracranial hemorrhage (OR, 6.59). These differences persisted at 1 year, with CDT demonstrating a significantly higher OR for mortality (OR, 1.56), myocardial infarction (OR, 2.01), ischemic stroke (OR, 6.69), and intracranial hemorrhage (OR, 3.93). Repeat intervention occurred more frequently in the PMT cohort at 1 year (12.2% vs 7.68%; OR, 1.67). Among patients with PSS, 1-year mortality was low in both groups. There were no significant differences in pulmonary embolism or subsequent first rib resection rates between CDT and PMT treated patients. However, repeat intervention occurred more frequently in the PMT cohort (21.4% vs 10.7%; OR, 2.27).ConclusionsIn this national propensity-matched analysis, CDT for UEDVT was associated with higher 30-day and 1-year morbidity and mortality compared with PMT, whereas PMT was associated with a greater need for repeat intervention. In patients with PSS, CDT and PMT demonstrated similar clinical outcomes, although PMT was associated with higher reintervention rates. These findings suggest differential risk profiles between the two treatment strategies and may inform procedural selection in the management of UEDVT. Article Highlights •Type of Research: Multicenter retrospective cohort study•Key Findings: We analyzed 2396 patients with upper extremity deep vein thrombosis (1198 treated with catheter-directed thrombolysis [CDT] and 1198 treated with percutaneous mechanical thrombectomy [PMT]). CDT was associated with higher 30-day mortality (odds ratio [OR], 1.82), stroke (OR, 9.11), and intracranial hemorrhage (OR, 6.59), whereas PMT had higher 1-year reintervention rates.•Take Home Message: PMT may offer safer 30-day and 1-year outcomes than CDT for upper extremity deep vein thrombosis, although at the cost of an increased need for reintervention. Article Highlights •Type of Research: Multicenter retrospective cohort study•Key Findings: We analyzed 2396 patients with upper extremity deep vein thrombosis (1198 treated with catheter-directed thrombolysis [CDT] and 1198 treated with percutaneous mechanical thrombectomy [PMT]). CDT was associated with higher 30-day mortality (odds ratio [OR], 1.82), stroke (OR, 9.11), and intracranial hemorrhage (OR, 6.59), whereas PMT had higher 1-year reintervention rates.•Take Home Message: PMT may offer safer 30-day and 1-year outcomes than CDT for upper extremity deep vein thrombosis, although at the cost of an increased need for reintervention. •Type of Research: Multicenter retrospective cohort study Type of Research: Multicenter retrospective cohort study •Key Findings: We analyzed 2396 patients with upper extremity deep vein thrombosis (1198 treated with catheter-directed thrombolysis [CDT] and 1198 treated with percutaneous mechanical thrombectomy [PMT]). CDT was associated with higher 30-day mortality (odds ratio [OR], 1.82), stroke (OR, 9.11), and intracranial hemorrhage (OR, 6.59), whereas PMT had higher 1-year reintervention rates. Key Findings: We analyzed 2396 patients with upper extremity deep vein thrombosis (1198 treated with catheter-directed thrombolysis [CDT] and 1198 treated with percutaneous mechanical thrombectomy [PMT]). CDT was associated with higher 30-day mortality (odds ratio [OR], 1.82), stroke (OR, 9.11), and intracranial hemorrhage (OR, 6.59), whereas PMT had higher 1-year reintervention rates. •Take Home Message: PMT may offer safer 30-day and 1-year outcomes than CDT for upper extremity deep vein thrombosis, although at the cost of an increased need for reintervention. Take Home Message: PMT may offer safer 30-day and 1-year outcomes than CDT for upper extremity deep vein thrombosis, although at the cost of an increased need for reintervention. Upper extremity deep vein thrombosis (UEDVT) typically involves the subclavian (implicated in ≤76% of UEDVT instances), axillary, and brachial veins, comprising the deep venous system. UEDVT can be asymptomatic but clinical manifestations mainly include edema and pain in addition to skin discoloration, numbness, pruritus, and limited range of motion.1 UEDVT represents an increasingly recognized form of VTE. Historically accounting for only 2% of all DVTs, UEDVT now comprises approximately 4% to 10% of cases.2 This upward trend is largely driven by the increased use of central venous catheters, peripherally inserted central catheter lines, implantable ports, and cardiac devices, all of which substantially increase thrombosis risk and now account for the majority of secondary UEDVTs.3,4 In hospitalized populations, the prevalence of UEDVT has been reported at 1% to 2% of admissions, with most events directly associated with intravenous catheters.5 UEDVT demonstrates substantial morbidity and mortality. Pulmonary embolism (PE) occurs in 12% to 36% of patients, whereas post-thrombotic syndrome of the arm, a chronic condition marked by pain, swelling, and functional limitation, develops in 7% to 46% of cases.6,7 Mortality rates are similarly notable. In a cohort of 182 patients, mortality was 16.5% at 30 days and 27.5% at 6 months, emphasizing the severity of this condition.8 Although these deaths frequently involve underlying conditions such as malignancy or critical illness, the data underscore the significant clinical impact of UEDVT. The diagnosis of UEDVT remains challenging due to nonspecific clinical presentations. As a result, clinicians may rely on secondary signs or variable institutional imaging algorithms, contributing to inconsistent diagnostic pathways and potential delays in treatment.3,9 Literature reviews further emphasize that standardized diagnostic criteria remain limited, and heterogeneous imaging practices complicate the timely identification of patients who may benefit from advanced interventions.10,11 Based on etiology, UEDVT is classified as either primary or secondary. Primary UEDVT, often termed Paget-Schroetter syndrome (PSS) or effort thrombosis, results from extrinsic mechanical compression of the subclavian vein, most commonly at the costoclavicular junction, and primarily affects younger, athletic individuals following repetitive upper extremity activity.12 Secondary UEDVT, caused by an underlying medical condition, constitutes the large majority, ranging from 80% to 90% of cases.9 The most common cause of secondary UEDVT in adults is catheter-related thrombosis, accounting for ≤70% of all UEDVTs.3,13 Secondary UEDVT is also strongly associated with malignancy. Management strategies for UEDVT vary widely, reflecting the limited availability of high-quality comparative data. Anticoagulation remains the standard of care for both primary and secondary UEDVT by the CHEST guidelines and demonstrates acceptable safety across agents, including direct oral anticoagulants.14, 15, 16 However, in patients with extensive, symptomatic, or anatomically driven thrombosis, particularly in PSS, endovascular intervention is increasingly pursued.17,18 Catheter-directed thrombolysis (CDT) is a minimally invasive procedure using a catheter to deliver thrombolytics directly to a DVT and has demonstrated favorable technical outcomes. Percutaneous mechanical thrombectomy (PMT) is another minimally invasive procedure that uses a catheter to mechanically remove blood clots from blood vessels, avoiding the systemic hemorrhagic risks associated with thrombolytic therapy. Despite expanding endovascular research, no studies have directly compared CDT and PMT in UEDVT, and current clinical guidelines largely extrapolate recommendations from lower extremity DVT or thoracic outlet syndrome literature rather than robust comparative data. Given the increasing prevalence of UEDVT, the substantial morbidity associated with delayed or inadequate treatment, and heterogeneous utilization of endovascular therapies, comparative studies investigating optimal treatment is needed. In studies analyzing PMT vs CDT for the treatment of PE and lower extremity DVT, PMT was found to have improved outcomes compared with CDT.19,20 Therefore, we hypothesized PMT would similarly have improved outcomes compared with CDT in the treatment of UEDVT. Specifically, this study uses a nationwide billing database (TriNetX) to explore the implementation of CDT and PMT in the treatment of UEDVT, focusing on 30-day and 1-year outcomes. Methods The data used in this study were queried from the TriNetX US Collaborative Network (TriNetX), which provides access to electronic medical records from 68 large health care organizations across the United States. The available data includes demographics, diagnoses (based on International Classification of Diseases, 10th edition, coding), procedures (based on International Classification of Diseases, 10th edition, Procedure Coding System or Current Procedural Terminology coding), and medications (based on RxNorm and VA National Formulary coding). All data available through the TriNetX database, up to and including patient-level data, are deidentified and conform to the standard defined in the Health Insurance Portability and Accountability Act. This study used deidentified patient data and did not collect, use, or disseminate individually identifiable patient data. This study was also classified as exempt from review by the University of Texas Health and Science Center at San Antonio Institutional Review Board. A propensity score-matched retrospective cohort study of patients found within the TriNetX US Collaborative Network was conducted. Post-op outcomes from two cohorts were compared: patients receiving CDT for UEDVT (n = 1399) and patients receiving PMT for UEDVT (n = 1406). Patients receiving both CDT and PMT were excluded from the study due to confounding influence. Patients diagnosed with UEDVT within seven days before undergoing CDT or PMT were identified between December 21, 2005, and December 21, 2025. The seven-day time frame between UEDVT diagnosis and subsequent treatment was included to ensure treatment was specifically for UEDVT. Specific codes and parameters used for the CDT and PMT cohorts are included in Supplementary Tables I and II, (online only) respectively. Cohorts were built based on diagnoses, medication, and procedure codes published in studies using the TriNetX database.20, 21, 22 Propensity score matching of the CDT and PMT cohorts were performed based on 16 characteristics. In the demographics category, patients were matched on race, age at index, and sex. In the diagnosis category, patients were matched on nicotine dependence (unspecified, associated with cigarettes, chewless tobacco, and other tobacco products), type 2 diabetes mellitus, essential hypertension, hyperlipidemia, other coagulation defects, neoplasms, cerebrovascular diseases, acute myocardial infarction (MI), and end-stage renal disease. In the medication category, patients were matched on ethinyl estradiol, and oral progestins (norethindrone, levonorgestrel, and norgestimate). Cohort outcomes were based on diagnostic, procedural, and medication codes along with instances of mortality and inpatient readmission within a period of 30 days (perioperative) and 1 year post intervention with either CDT or PMT. Outcomes included: open surgery, repeat vascularization by CDT or PMT, mortality, PE, MI, ischemic cerebrovascular accident (CVA), transfusion, 30-day readmission, and intracerebral hemorrhage (ICH). Outcome instances and risk were analyzed as an odds ratio (OR) with 95% confidence interval (CI) for perioperative and 1-year postoperative outcomes. Specific codes used to identify each outcome are included in Supplementary Table III (online only). The validity of cohort creation was tested by determining subsequent CDT and PMT operations as an outcome. Additional investigation of CDT and PMT in the management of primary UEDVT (PSS) was conducted. Because PSS is specifically due to compression of upper extremity deep veins, it was defined as requiring the diagnosis of UEDVT and compression of a vein. Two cohorts were analyzed after propensity score matching using the same characteristics previously listed using the TriNetX US Collaborative Network: patients receiving CDT for PSS (n = 149) and patients receiving PMT for PSS (n = 389). CDT and PMT must have occurred within 7 days of the diagnosis of PSS. One-year outcomes after the initial intervention were analyzed, including reintervention with CDT or PMT, subsequent first rib resection, PE, MI, CVA, ICH, and mortality. The PSS specific criteria for both cohorts are listed in Supplementary Table IV (online only). Results Based on these criteria, we included 1406 patients in the PMT cohort and 1399 patients in the CDT cohort. Propensity score matching was then performed to yield a total of 2396 patients with 1198 patients from each cohort. The matched comorbidities, demographics, and medications are shown in Table I and are comparable between the two groups.Table ISummarized characteristics of the cohorts before and after propensity score matchingCharacteristicBefore matchingAfter matchingCDT (n = 1399)PMT (n = 1406)P valueCDT (n = 1198)PMT (n = 1198)P valueDemographics Age at index, years51.5 ± 20.050.5 ± 18.8.18449.4 ± 19.850.4 ± 19.1.218 White924 (66.0)979 (69.7).039830 (69.3)834 (69.6).859 Male727 (52.0)750 (53.4).453627 (52.3)629 (52.5).935Diagnosis Nicotine dependence287 (20.5)290 (20.6).934233 (19.4)232 (19.4).959 Type 2 diabetes mellitus375 (26.8)365 (26.0).620281 (23.5)302 (25.2).317 Essential hypertension647 (46.2)628 (44.7).410492 (41.1)518 (43.2).282 Hyperlipidemia416 (29.7)431 (30.7).588330 (27.5)342 (28.5).585 Coagulation defects175 (12.5)162 (11.5).426139 (11.6)136 (11.4).848 Neoplasms386 (27.6)456 (32.5).005348 (29.0)356 (29.7).720 Cerebrovascular diseases272 (19.4)170 (12.1)<.001142 (11.9)161 (13.4).243 MI129 (9.2)101 (7.2).05091 (7.6)88 (7.3).816 End-stage renal disease166 (11.9)268 (19.1)0.001164 (13.7)180 (15.0).351Medication Ethinyl estradiol51 (3.6)43 (3.1).39044 (3.7)42 (3.5).826 Norethindrone32 (2.3)35 (2.5).72431 (2.6)30 (2.5).897 Levonorgestrel19 (1.4)15 (1.1).48218 (1.5)15 (1.3).599 Norgestimate41 (2.9)39 (2.8).80533 (2.8)33 (2.8)1.000CDT, Catheter-directed thrombolysis; MI, myocardial infarction; PMT, percutaneous mechanical thrombectomy.Values are number ± standard deviation or number (%). Table ISummarized characteristics of the cohorts before and after propensity score matchingCharacteristicBefore matchingAfter matchingCDT (n = 1399)PMT (n = 1406)P valueCDT (n = 1198)PMT (n = 1198)P valueDemographics Age at index, years51.5 ± 20.050.5 ± 18.8.18449.4 ± 19.850.4 ± 19.1.218 White924 (66.0)979 (69.7).039830 (69.3)834 (69.6).859 Male727 (52.0)750 (53.4).453627 (52.3)629 (52.5).935Diagnosis Nicotine dependence287 (20.5)290 (20.6).934233 (19.4)232 (19.4).959 Type 2 diabetes mellitus375 (26.8)365 (26.0).620281 (23.5)302 (25.2).317 Essential hypertension647 (46.2)628 (44.7).410492 (41.1)518 (43.2).282 Hyperlipidemia416 (29.7)431 (30.7).588330 (27.5)342 (28.5).585 Coagulation defects175 (12.5)162 (11.5).426139 (11.6)136 (11.4).848 Neoplasms386 (27.6)456 (32.5).005348 (29.0)356 (29.7).720 Cerebrovascular diseases272 (19.4)170 (12.1)<.001142 (11.9)161 (13.4).243 MI129 (9.2)101 (7.2).05091 (7.6)88 (7.3).816 End-stage renal disease166 (11.9)268 (19.1)0.001164 (13.7)180 (15.0).351Medication Ethinyl estradiol51 (3.6)43 (3.1).39044 (3.7)42 (3.5).826 Norethindrone32 (2.3)35 (2.5).72431 (2.6)30 (2.5).897 Levonorgestrel19 (1.4)15 (1.1).48218 (1.5)15 (1.3).599 Norgestimate41 (2.9)39 (2.8).80533 (2.8)33 (2.8)1.000CDT, Catheter-directed thrombolysis; MI, myocardial infarction; PMT, percutaneous mechanical thrombectomy.Values are number ± standard deviation or number (%). Summarized characteristics of the cohorts before and after propensity score matching CDT, Catheter-directed thrombolysis; MI, myocardial infarction; PMT, percutaneous mechanical thrombectomy. Values are number ± standard deviation or number (%). All 1198 patients in the CDT cohort underwent CDT without mechanical thrombectomy, whereas all patients in the PMT cohort underwent PMT without thrombolytic exposure. The need for open surgery within 30 days and 1 year after the procedure was attempted to be analyzed between the two groups, but ≤10 patients were identified as requiring open surgery after initial intervention in both the CDT and PMT cohorts, which is too limited for TriNetX to report detailed analysis on. To preserve patient privacy, TriNetX does not report any patient counts <10. During the 30-day perioperative period, the OR of PE occurring was not significantly different between the two groups (OR, 1.16; 96% CI, 0.90-1.50). However, significant differences were observed across multiple adverse outcomes. Compared with PMT, CDT was associated with a significantly higher odds of 30-day mortality (OR, 1.82; 95% CI, 1.24-2.67), MI (OR, 2.43; 95% CI, 1.51-3.91), ischemic CVA (OR, 9.11; 95% CI, 6.05-13.73), transfusion (OR, 1.95; 95% CI, 1.33-2.85), 30-day readmission (OR, 2.34; 95% CI, 1.99-2.76), and intracranial hemorrhage (OR, 6.59; 95% CI, 3.37-12.91). The repeat intervention rates were not significantly different between the two groups; 6.8% of patients who received CDT needed repeat CDT, whereas 8.2% of patients who received PMT needed repeat PMT within 30 days of the initial procedure (OR, 1.23; 95% CI, 0.90-1.67; P = .19). These perioperative outcomes are summarized in Table II.Table IISummary of the 30-day post-procedural outcomes in patients with upper extremity deep vein thrombosis (UEDVT) treated with catheter-directed thrombolysis (CDT) (n = 1198) and percutaneous mechanical thrombectomy (PMT) (n = 1198) after propensity score matchingOutcomeCDTPMTOR [95% CI]P valueRepeat CDT81 (6.8)0N/AN/ARepeat PMT098 (8.2)N/AN/AMortality76 (6.3)43 (3.6)1.82 [1.24-2.67].002PE143 (11.9)125 (10.4)1.16 [0.90-1.50].243MI59 (4.9)25 (2.1)2.43 [1.51-3.91]<.001Ischemic stroke208 (17.4)27 (2.3)9.11 [6.05-13.73]<.0001Transfusion81 (6.8)43 (3.6)1.95 [1.33-2.85]<.00130-Day readmission786 (65.6)538 (44.9)2.34 [1.99-2.76]<.0001Intracranial hemorrhage63 (5.3)10 (0.8)6.59 [3.37-12.91]<.0001Open surgery<10<10N/AN/ACI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism.Instance of outcome (% of cohort developing outcome) is reported in the first two columns.Values are number (%) unless otherwise noted. Table IISummary of the 30-day post-procedural outcomes in patients with upper extremity deep vein thrombosis (UEDVT) treated with catheter-directed thrombolysis (CDT) (n = 1198) and percutaneous mechanical thrombectomy (PMT) (n = 1198) after propensity score matchingOutcomeCDTPMTOR [95% CI]P valueRepeat CDT81 (6.8)0N/AN/ARepeat PMT098 (8.2)N/AN/AMortality76 (6.3)43 (3.6)1.82 [1.24-2.67].002PE143 (11.9)125 (10.4)1.16 [0.90-1.50].243MI59 (4.9)25 (2.1)2.43 [1.51-3.91]<.001Ischemic stroke208 (17.4)27 (2.3)9.11 [6.05-13.73]<.0001Transfusion81 (6.8)43 (3.6)1.95 [1.33-2.85]<.00130-Day readmission786 (65.6)538 (44.9)2.34 [1.99-2.76]<.0001Intracranial hemorrhage63 (5.3)10 (0.8)6.59 [3.37-12.91]<.0001Open surgery<10<10N/AN/ACI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism.Instance of outcome (% of cohort developing outcome) is reported in the first two columns.Values are number (%) unless otherwise noted. Summary of the 30-day post-procedural outcomes in patients with upper extremity deep vein thrombosis (UEDVT) treated with catheter-directed thrombolysis (CDT) (n = 1198) and percutaneous mechanical thrombectomy (PMT) (n = 1198) after propensity score matching CI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism. Instance of outcome (% of cohort developing outcome) is reported in the first two columns. Values are number (%) unless otherwise noted. At 1 year, the OR of PE occurring remained not significantly different between the two groups (OR, 1.18; 95% CI, 0.94-1.49). CDT also remained associated with significantly higher odds of mortality (OR, 1.56; 95% CI, 1.21-2.01), MI (OR, 2.01; 95% CI, 1.40-2.89), ischemic stroke (OR, 6.69; 95% CI, 4.84-9.24), and intracranial hemorrhage (OR, 3.93; 95% CI, 2.39-6.49). The odds of requiring repeat intervention at 1 year were significantly greater in the PMT cohort, with 12.2% undergoing repeat thrombectomy compared with 7.7% of CDT patients undergoing repeat thrombolysis (OR, 1.67; 95% CI, 1.26-2.18; P = .0002). One-year postoperative outcomes are summarized in Table III.Table IIISummary of 1-year postoperative outcomes in patients with upper extremity deep vein thrombosis (UEDVT) treated with catheter-directed thrombolysis (CDT) (n = 1198) and percutaneous mechanical thrombectomy (PMT) (n = 1198) after propensity score matchingOutcomeCDTPMTOR [95% CI]P valueRepeat CDT92 (7.7)0N/AN/ARepeat PMT0146 (12.2)N/AN/AMortality166 (13.9)112 (9.3)1.56 [1.21-2.01].001PE180 (15.0)156 (13.0)1.18 [0.94-1.49].158MI91 (7.6)47 (3.9)2.01 [1.40-2.89]<.001Ischemic stroke257 (21.5)47 (3.9)6.69 [4.84-9.24]<.001Intracranial hemorrhage75 (6.3)20 (1.7)3.93 [2.39-6.49]<.001Open surgery<10<10N/AN/ACI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism.Instance of outcome (% of cohort developing outcome) is reported in the first two columns.Values are number (%) unless otherwise noted. Table IIISummary of 1-year postoperative outcomes in patients with upper extremity deep vein thrombosis (UEDVT) treated with catheter-directed thrombolysis (CDT) (n = 1198) and percutaneous mechanical thrombectomy (PMT) (n = 1198) after propensity score matchingOutcomeCDTPMTOR [95% CI]P valueRepeat CDT92 (7.7)0N/AN/ARepeat PMT0146 (12.2)N/AN/AMortality166 (13.9)112 (9.3)1.56 [1.21-2.01].001PE180 (15.0)156 (13.0)1.18 [0.94-1.49].158MI91 (7.6)47 (3.9)2.01 [1.40-2.89]<.001Ischemic stroke257 (21.5)47 (3.9)6.69 [4.84-9.24]<.001Intracranial hemorrhage75 (6.3)20 (1.7)3.93 [2.39-6.49]<.001Open surgery<10<10N/AN/ACI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism.Instance of outcome (% of cohort developing outcome) is reported in the first two columns.Values are number (%) unless otherwise noted. Summary of 1-year postoperative outcomes in patients with upper extremity deep vein thrombosis (UEDVT) treated with catheter-directed thrombolysis (CDT) (n = 1198) and percutaneous mechanical thrombectomy (PMT) (n = 1198) after propensity score matching CI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism. Instance of outcome (% of cohort developing outcome) is reported in the first two columns. Values are number (%) unless otherwise noted. Propensity score matching yielded a total of 280 patients who had PSS, with 140 of them treated with CDT and 140 treated with PMT. The odds of repeat intervention at 1 year were significantly greater in the PMT cohort, with 21.4% undergoing repeat thrombectomy compared with 10.7% of CDT patients undergoing repeat thrombolysis (OR, 2.27; 95% CI, 1.16-4.43; P = .016). Furthermore, the odds of PE occurring 1 year after the operation was not significantly different between the two groups (OR, 1.12; 95% CI, 0.58-2.16). The number of instances of mortality, MI, CVA, and ICH were too small for TriNetX to report meaningful analysis on. Finally, at 1 year after the intervention, 52 patients with PSS who underwent CDT (37.1%) required subsequent first rib resection, whereas 43 patients with PSS who underwent PMT (30.7%) required it. The odds of subsequent first rib resection were not significantly different between the two groups (OR, 1.33; 95% CI, 0.81-2.19). One-year postoperative outcomes for patients with PSS are summarized in Table IV.Table IVSummary of 1-year postoperative outcomes in patients with Paget-Schroetter syndrome (PSS) treated with catheter-directed thrombolysis (CDT) (n = 140) and percutaneous mechanical thrombectomy (PMT) (n = 140) after propensity score matchingOutcomeCDTPMTOR [95% CI]P valueRepeat CDT15 (10.7)0N/AN/ARepeat PMT030 (21.4)N/AN/AMortality<10<10N/AN/APE22 (15.7)20 (14.3)1.12 [0.58-2.16]073MI<10<10N/AN/AIschemic stroke<10<10N/AN/AIntracranial hemorrhage<10<10N/AN/AFirst rib resection52 (37.1)43 (30.7)1.33 [0.81-2.19]0.26CI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism.Instance of outcome (% of cohort developing outcome) is reported in the first two columns.Values are number (%) unless otherwise noted. Table IVSummary of 1-year postoperative outcomes in patients with Paget-Schroetter syndrome (PSS) treated with catheter-directed thrombolysis (CDT) (n = 140) and percutaneous mechanical thrombectomy (PMT) (n = 140) after propensity score matchingOutcomeCDTPMTOR [95% CI]P valueRepeat CDT15 (10.7)0N/AN/ARepeat PMT030 (21.4)N/AN/AMortality<10<10N/AN/APE22 (15.7)20 (14.3)1.12 [0.58-2.16]073MI<10<10N/AN/AIschemic stroke<10<10N/AN/AIntracranial hemorrhage<10<10N/AN/AFirst rib resection52 (37.1)43 (30.7)1.33 [0.81-2.19]0.26CI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism.Instance of outcome (% of cohort developing outcome) is reported in the first two columns.Values are number (%) unless otherwise noted. Summary of 1-year postoperative outcomes in patients with Paget-Schroetter syndrome (PSS) treated with catheter-directed thrombolysis (CDT) (n = 140) and percutaneous mechanical thrombectomy (PMT) (n = 140) after propensity score matching CI, Confidence interval; MI, myocardial infarction; N/A, not available; OR, odds ratio; PE, pulmonary embolism. Instance of outcome (% of cohort developing outcome) is reported in the first two columns. Values are number (%) unless otherwise noted. Discussion This study is a large-scale propensity score-matched analysis looking into the comparative outcomes between CDT and PMT for the treatment of UEDVT and PSS. These findings reveal valuable differences in short- and long-term outcomes that have the potential to positively inform clinical decision-making. Studies report that CDT for axillosubclavian thrombosis leads to high rates of complete or partial lysis and substantial symptom relief, especially when followed by careful reassessment of residual venous obstruction.23 Another case series of UEDVT treated with CDT and balloon angioplasty similarly describe high technical success and restoration of luminal patency in patients with symptomatic, occlusive disease.24 In primary UEDVT such as PSS, Lugo et al25 evaluated patients treated with CDT with or without first rib resection and showed that first rib resection after thrombolysis is associated with superior long-term symptom resolution (95%) and venous patency (98%) compared with CDT only (54% and 48%, respectively), noting that a substantial proportion of patients initially managed without decompression ultimately required delayed surgery. In a single-center retrospective analysis of 14 UEDVT patients, Pebror et al26 reported 100% technical success using a mechanical thrombectomy device, the ClotTriever system, with all procedures completed in a single session and no major bleeding events; at follow-up, most patients remained free from rethrombosis or PE. Additional experience with the ClotTriever system demonstrated successful thrombus extraction in all eight reported patients, with immediate symptomatic improvement, restoration of venous outflow, and no major periprocedural complications.27 Furthermore, CDT and PMT can be used together to treat UEDVT. Lungren et al28 reported nine children with catheter-associated UEDVT treated with catheter-directed alteplase, balloon angioplasty, and mechanical thrombectomy, achieving preservation of central venous access in all cases and low complication rates. Collectively, these case series highlight the feasibility and safety of PMT and CDT for UEDVT across varied anatomical and clinical contexts, although data are limited by sample size and retrospective design. As a validity check, the occurrence of subsequent CDT and PMT was evaluated within the respective treatment cohorts to confirm that patients were appropriately categorized. Our results verify that patients in the CDT cohort received CDT exclusively, whereas patients in the PMT cohort received only PMT. Patients who received both treatment modalities were excluded during cohort construction, consistent with the initial study design. These findings further support the integrity of the cohort definitions and confirm that the TriNetX database appropriately generated the study groups based on the specified inclusion and exclusion criteria. Our data suggest that PMT is associated with a significantly higher need for long-term reintervention. At 1 year, patients treated with PMT required repeat thrombectomy at higher rates than those needing repeat thrombolysis when treated with CDT, with 12.2% of PMT patients undergoing repeat PMT compared with 7.68% of CDT patients needing repeat CDT. This finding suggests that, although PMT achieves technical success at the time of the procedure, long-term patency may be inferior to that of CDT. This finding provides more insight into the results of the PMT case series done by Pebror et al26 and Patel et al,27 which reported 100% technical success rates and favorable short-term outcomes at 3 and 6 months, respectively. These reports, however, included small numbers of patients and lacked longitudinal follow-up, limiting their ability to detect later reinterventions or loss of patency. Conversely, CDT was associated with significantly higher rates of serious perioperative and long-term complications. Significantly increased perioperative rates of mortality, ischemic stroke, intracranial hemorrhage, MI, transfusion, and 30-day readmission were found in patients undergoing CDT. Significant differences in rates of mortality, ischemic stroke, intracranial hemorrhage, and MI persisted at 1 year, with the OR of ischemic stroke rates being almost seven times higher in the CDT group compared with PMT. These findings align with established risks of thrombolytic therapy; the existing literature reviews report ICH and major bleeds requiring transfusions as risks for CDT.29, 30, 31 When contextualized within the existing literature, the findings of this study reinforce that patient selection plays a central role in determining outcomes. Prior positive outcomes associated with CDT largely derive from younger, healthier patients with primary UEDVT who underwent thrombolysis followed by thoracic outlet decompression.25 In contrast, the national cohort evaluated in this study included a diverse population diagnosed with significant comorbidities. The higher mortality, stroke, and bleeding rates observed in the CDT cohort likely reflect, at least in part, the vulnerabilities of this real-world population. This study remains limited by the nature of the data. Important clinical variables cannot be reliably captured, including clot chronicity, thrombus burden, specific device used, and imaging-confirmed patency. The dataset also cannot distinctly distinguish complications as directly linked to the initial treatment patients received. Despite the time window set to increase the likelihood of capturing procedure related complications, the nature of the TriNetX database prevents distinguishment between a coincidental instance of a complication happening in the defined time range or a true complication resulting directly from the procedure. Additionally, the retrospective design of the study further limits the ability to establish causality in treatment complications. Another limitation of this study is difficult stratification of UEDVT as primary (PSS) vs secondary given the lack of specific diagnosis codes for the two subtypes. This point is important to mention because treatment differs between the types of UEDVT. In PSS, CDT and first rib resection have been shown to restore venous patency, decrease clot burden, and improve long-term functional outcomes.32,33 With secondary UEDVT being associated with a provoking factor like intravenous devices or malignancy, endothelial injury and chronic inflammation lead to thrombus development rather than extrinsic compression. Therefore, anticoagulation remains the primary therapy, whereas CDT may be considered in severe or refractory cases, although it is associated with increased bleeding risk and no clear long-term benefit.34 PMT may be considered in PSS or secondary UEDVT refractory to anticoagulation and thrombolysis.34 Because CDT offers longer patency but carries a higher risk of bleeding, its utility in treating PSS, where patients tend to be younger and healthier, is apparent.35 CDT would allow for fewer future interventions in younger patients while bleeding risk in minimized due to the lack of comorbidities. Our results appear to support this as ≤10 of the 140 patients with PSS who were treated by CDT developed MI, CVA, ICH, or died. However, this limited patient count cannot derive a meaningful comparison. Furthermore, PSS treated with PMT was associated with higher odds of requiring reintervention compared with CDT, despite no significant difference in the odds of requiring subsequent first rib resection between the two groups. Because first rib resection is typically performed in patients with persistent or recurrent symptoms due to thoracic outlet obstruction after endovascular therapy, the need for rib resection may serve as an indirect marker of vessel patency after the initial intervention. Although limited by sample size, these findings suggest CDT may represent a safe treatment option for PSS that is associated with lower rates of reintervention without a significant difference in patency, consistent with prior literature. The interpretation of these findings supports a risk-based approach to endovascular management. CDT may remain appropriate for young, otherwise healthy patients with primary effort thrombosis. However, for patients with an elevated bleeding risk, PMT may offer a safer alternative given its avoidance of thrombolytic exposure. Clinicians should recognize that PMT may carry a greater likelihood of requiring subsequent interventions, and long-term patency may be less durable without adjunctive therapy. These observations highlight the importance of individualized therapy based on symptom duration, thrombus extent, comorbidity burden, and anatomical considerations. Furthermore, future studies are needed to elucidate specific comorbidity profiles that impact outcomes associated with being treated with PMT or CDT for UEDVT. Additionally, new intravascular technologies like pharmacomechanical thrombectomy, which combines mechanical thrombectomy with thrombolysis, need to be studied in the treatment of UEDVT. Finally, optimal treatment for secondary UEDVT vs primary UEDVT needs to be specifically investigated due to their drastically different etiologies. Conclusions This study provides comparative data on the outcomes and patency of endovascular treatment options for UEDVT. CDT is associated with lower 1-year reintervention rates, but it appears to carry higher risks of MI, ischemic stroke, intracranial hemorrhage, and mortality. In contrast, PMT seems to demonstrate more favorable perioperative and 1-year safety profiles, but is associated with a greater need for subsequent intervention. These findings support a risk-stratified treatment approach, in which CDT may be most appropriate for younger patients with a lower bleeding risk, particularly those with PSS, whereas PMT may be advantageous in older, medically complex patients at increased risk for hemorrhage. Further investigation is needed to better define how comorbidity profiles influence treatment safety and effectiveness. Prospective studies are also warranted to validate these observations and guide standardized treatment algorithms. Author Contributions Conception and design: SC, LDP, JA Analysis and interpretation: SC, MP, EF, LDP, LLP, MS, JA Data collection: SC Writing the article: SC, MP, EF Critical revision of the article: SC, LDP, LLP, MS, JA Final approval of the article: SC, MP, EF, LDP, LLP, MS, JA Statistical analysis: Not applicable Obtained funding: SC, LDP Overall responsibility: SC Funding Funded by the 10.13039/100030579Society for Vascular Surgery Foundation 2025 Student Research Fellowship. Declaration of generative AI and AI-assisted technologies in the writing process During the preparation of this work, the authors used ChatGPT Model 5.2 to assist with formatting and clarity. After using this tool, the authors reviewed and edited the content and take full responsibility for the content of the publication. Disclosures None.