This study is reported with guidance of the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement [14].

The data was prospectively collected from four medical alliance hospitals (Shanghai Ninth People’s Hospital, Fengcheng Hospital, Zhoupu Hospital, and Putuo Hospital). between January 2016 and December 2021. Acute, transiently provoked DVT patients between 18 and 45 years old with iliac stenosis who underwent AngioJet rheolytic thrombectomy (ART) and met the criteria were included. This study was approved by the Ethics Committee of Shanghai JiaoTong University School of Medicine.

The inclusion criteria were as follows: [1] age 18–45 years; [2] acute-phase DVT (within 2 weeks of symptoms onset) with iliac vein stenosis and received ART; [3] a clear provoking factor of DVT; [4] iliac-femoral DVT diagnosed for the first time, with or without femoropopliteal DVT and pulmonary embolism, and no history of superficial varicose veins or leg swelling. The exclusion criteria were as follows: [1] congenital diseases, such as Klippel-Trenaunay syndrome; [2] contraindications to anticoagulation or abnormal coagulation function; [3] incomplete treatment or missing data, and [4] iliac vein stenosis less than 50%, or contrast agent stasis time more than 6 s after ART and possible balloon dilation or catheter-directed thrombolysis (CDT).

The ART procedures have been described previously [5]. Briefly, patients received 200 U/kg low-molecular-weight heparin or 1.0 mg/kg enoxaparin following diagnosis. During the procedure, after the guidewire and the AngioJet catheter (Solent/Zelante, Boston Scientific, USA) were successfully advanced through the thrombosed vein segment, a power pulse lytic model was used with 0.25 × 106 units of urokinase in 100 mL of saline. Fifteen minutes later, the AngioJet catheter was placed in standard rheolytic thrombectomy mode. This sequence was repeated if significant residual thrombi remained on subsequent venograms.

Some patients underwent balloon dilation and CDT after ART. The balloon dilation was performed using balloon catheters (Mustang, Boston Scientific, USA) with a diameter of 8–16 mm for serial dilation from the femoral vein to the common iliac vein. The procedure of CDT was inserting a multiple-side-hole infusion catheter (Angiodynamics, Queensbury, USA) into the thrombotic segment, and injecting urokinase (30,000–50,000 U/100 mL/h). When the fibrinogen level reached 1.5 g/L, the urokinase dosage was halved, and CDT would be finished when the fibrinogen level was decreased to 1.0 g/L. After ART and possible balloon dilation or CDT, venogram was performed to exam the iliac vein stenosis rate and contrast agent stasis time. The representative DSA sequence of contrast agent stasis time examination is shown in Supplemental file 1. The contrast agent stasis time is defined as the interval from the arrival of contrast agent in the iliac vein to the disappearance of contrast at the same level.

As described before [15], for patients underwent stenting, self-expanding stents Wallstent (Boston Scientific, USA) or a Lifestar (Bard Medical, USA) were implanted after balloon dilation. The stent diameters ranged from 12 to 14 mm for the iliac-femoral vein and from 14 to 16 mm for the common iliac.

After the procedure, patients of both groups received the same anti-coagulation therapy. Low molecular weight heparin was used during hospitalization. After discharge, warfarin or rivaroxaban were prescribed for at least 6 months since patients underwent endovascular intervention according to CIRSE standards of practice guidelines [8]. For patients taking warfarin, the international normalized ratio was maintained between 2 and 3. Patients received no postoperative anti-platelet treatment. All the patients used compression stockings (class II, 30 mmHg) as a standard adjunct treatment.

The outcomes included severity of lower limb symptoms, quality of life, free from re-thrombosis rate and complications. The severity of lower limb symptoms was evaluated using the Villalta score. The VEINES-QOL questionnaire was translated into local language from the appendix of Kahn 2006 [16]. Clinical follow-up visits were scheduled at 1, 3, 6, and 12 months after the procedure, and then yearly. A duplex scan or venography was performed to assess patency, and questionnaire follow-up was used to record the Villalta score and the VEINES-QOL score. Adverse events were continuously monitored during follow-up.

Binary data are expressed as absolute or relative frequencies and were compared by the chi-square or Fisher exact tests. Ranked data are expressed as absolute values and were compared by the Cochran-Armitage trend test. Continuous data are expressed as the mean ± standard deviation and were compared using Student’s t-test. The free from re-thrombosis rate was assessed by the Kaplan–Meier estimator. For all tests, p < .05 was defined as significant. All analysis was performed using SAS 9.2 (Stat, SAS Institute, Cary, NC, USA).