Indian Journal of Vascular and Endovascular Surgery

: 2021  |  Volume : 8  |  Issue : 5  |  Page : 41--45

Left common iliac vein compression in patients with may-thurner syndrome: A 10-year retrospective study in an australian cohort

Warren Clements1, Ronny J D Kuang2, Jarrel Seah3, Heather K Moriarty4, Thodur Vasudevan5, Amanda Davis6, Jim Koukounaras4,  
1 Department of Radiology, Alfred Health; Department of Surgery, Central Clinical School, Monash University; National Trauma Research institute, Central Clinical School, Monash University, Melbourne, Australia
2 Department of Radiology, Alfred Health, Central Clinical School, Monash University, Melbourne, Australia
3 Department of Radiology, Alfred Health; Department of Neuroscience, Central Clinical School, Monash University, Melbourne, Australia
4 Department of Radiology, Alfred Health; Department of Surgery, Central Clinical School, Monash University, Melbourne, Australia
5 Department of Vascular Surgery, Alfred Health, Central Clinical School, Monash University, Melbourne, Australia
6 Department of Haematology, Alfred Health, Central Clinical School, Monash University, Melbourne, Australia

Correspondence Address:
Warren Clements
Department of Radiology, Alfred Health; Department of Surgery, Central Clinical School, Monash University; National Trauma Research institute, Central Clinical School, Monash University, Melbourne


Objectives: This study aimed to assess what diameter constitutes clinically-significant left common iliac vein (LCIV) compression in patients with May-Thurner syndrome (MTS). Materials and Methods: Nineteen patients with MTS were over a 10-year period. Minimum LCIV diameter was compared to 100 asymptomatic controls and 27 age- and gender-matched controls. Results: Mean LCIV diameter in MTS group was 3.82 mm (standard deviation [SD] 1.38), control group (mean 7.17 mm SD 3.19, P < 0.0001), and matched control group (mean 6.86 mm SD 3.03, P = 0.007). Statistical threshold analysis showed in MTS patients, a LCIV diameter of 4.7mm or less had an 87.5% sensitivity and 72.7% specificity for the diagnosis. Conclusions: Patients with MTS had a minimum LCIV diameter threshold of 4.7mm, and this can be used in correlating the diagnosis of MTS on computed tomography. However, minimum diameters less than 4.7 mm are also seen in the general population and as such compression alone does not constitute a diagnosis of MTS.

How to cite this article:
Clements W, Kuang RJ, Seah J, Moriarty HK, Vasudevan T, Davis A, Koukounaras J. Left common iliac vein compression in patients with may-thurner syndrome: A 10-year retrospective study in an australian cohort.Indian J Vasc Endovasc Surg 2021;8:41-45

How to cite this URL:
Clements W, Kuang RJ, Seah J, Moriarty HK, Vasudevan T, Davis A, Koukounaras J. Left common iliac vein compression in patients with may-thurner syndrome: A 10-year retrospective study in an australian cohort. Indian J Vasc Endovasc Surg [serial online] 2021 [cited 2022 Jan 28 ];8:41-45
Available from:

Full Text


May-Thurner syndrome (MTS) is a clinical syndrome of left leg iliofemoral deep-venous thrombosis (DVT) associated with an anatomic variation. It was first described by Wanke and Kiel in 1950 as an operative finding in patients with chronic thrombosis of the pelvic veins. In 1957, May and Thurner characterized compression of the left common iliac vein (LCIV) between the crossing right common iliac artery and the vertebral body of L5 in 430 cadaveric specimens, theorising the close topographical relationship of these vessels results in a venous intraluminal fibrous band or “spur.”[1]

Patients with LCIV compression may be asymptomatic; however, the diagnosis of MTS may be made if patients have associated symptomatic DVT in the left leg. The condition is more common in females and in younger patients.[2] Early diagnosis is important as treatments are aimed at preventing postthrombotic syndrome (PTS) which is known to impact quality of life.[2]

When patients present with unilateral left leg swelling, several factors need to be taken into consideration to make a diagnosis of MTS, including excluding other causes for the presenting symptom. In many patients, an ultrasound is appropriate to assess for DVT. However, there may also be a role for computed tomography (CT) to assess for the compression of the LCIV in some patients.

The compression of the LCIV may be primary or secondary. Primary compression occurs when the space between the L5 vertebrae and the right common iliac artery is narrow without an additional extrinsic contributor.[3] Secondary compression occurs when compression is precipitated by an additional lesion, for example, anterior lumbar spine osteophytes, lymphadenopathy, fracture, or an inflammatory process such as osteomyelitis.[3]

Practically, when a CT is performed looking for LCIV compression, there remains no consensus on what constitutes clinically significant compression and what is normal, with acknowledgement that LCIV compression is also seen in patients without MTS.[4] Ou-Yang and Lu retrospectively assessed the relationship of LCIV diameter in 69 Chinese patients with MTS and compared this to patients without symptoms. They showed that in patients with simple or primary MTS, the mean LCIV diameter was 2.52 mm, while in secondary or degenerative MTS, the diameter was 2.29 mm. This compared to the control group of 4.34 mm, P < 0.001.[5] Similar results have been shown in a Turkish cohort where Oguzkurt et al. showed mean percentage compression of the LCIV in 34 patients with MTS of 74% compared with controls where the LCIV was compressed by 28%.[6] However, conflicting evidence has also been presented. Park et al. assessed the LCIV diameter in 87 patients with left leg DVT and PTS. They showed that the mean diameter in patients without PTS (5.89 mm) was similar to the group with PTS (5.56 mm), P = 0.882 and the mean LCIV in these studies was larger than in the Chinese and Turkish cohorts.[4]

This study aimed to assess the association of LCIV diameter in patients with symptoms of MTS in an Australian population, comparing this to the LCIV diameter in asymptomatic patients, in an attempt to define a threshold for the diagnosis on imaging.

 Materials and Methods


Approval was provided by our Institutional Review Board, number 566/20. For this retrospective analysis, individual patient consent was not required.

Patient identification

The study covered a 10-year period from January 1, 2010 to January 1, 2020. Patients were identified through the Radiology Information System when CT scan was performed and the indication for the scan included any of the following: A diagnosis of MTS, left leg swelling due to DVT, left iliac, or femoral DVT. This constituted presentation with acute MTS or symptomatic untreated MTS with recurrent DVT or PTS.

Patients were excluded if they had bilateral leg swelling/DVT, isolated right leg swelling/DVT, or if there was congenital retroperitoneal venous anomaly. Control patients were consecutive patients who presented to the emergency department requiring a CT scan of the abdomen or pelvis from January 1, 2019, for any reason other than left lower limb DVT, for example, assessment of acute abdominal pain.

Measurement of left common iliac vein diameter

Measurements in the study group [Figure 1] and control group [Figure 2] were performed using the measurement tool available in the picture and communications system (PACS) (GE healthcare, Chicago, IL, USA). The LCIV diameter was measured on CT by two vascular radiologists and any major discrepancies mediated by a third independent vascular radiologist. The minimum LCIV diameter was measured in any plane (axial, sagittal, or coronal) as to best assess the diameter based upon the patients' anatomy and was at the discretion of the radiologist.{Figure 1}{Figure 2}

Statistical analysis

Data were deidentified and analyzed using Microsoft Excel (Microsoft, USA) with the Real Statistics Resource Pack software (Release 6.8). Copyright (2013 – 2021) Charles Zaiontz. Data were summarized using mean and standard deviation (SD), median and range, or frequency and percentage where relevant according to the type of data. The control group of 100 patients was age- and gender-matched to the study group generating a matched group of 27 patients. Significance testing was performed using the Mann–Whitney U test and a two-sided P < 0.05 was chosen to indicate statistical significance. Threshold testing was performed by generating a receiver operating characteristic (ROC) curve of the study group and matched controls.


During the 10-year period, 19 patients met inclusion criteria. This included 14 females (73.7%) of mean age 45.8 years (SD 14.9). Primary MTS was identified in 14 patients (73.7%) and secondary MTS in 5 patients (26.3%). This compared to the control group of 100 patients consisting of 45% females of mean age 55.7 years (SD 19.0). The matched control group consisted of 27 patients, 74% females, mean age of 47.5 years (SD 14.0) [Table 1].{Table 1}

[Figure 3] shows a box-plot comparing the LCIV diameters in the MTS, control, and matched control groups. While the minimum value was low in both groups, the control and matched control groups showed a higher Q1, median, Q3, and maximum value indicating a larger variability of the diameters in the control population which is supported by the larger SD. In the study group, the mean LCIV diameter was 3.81 mm (SD 1.38) and median 4.05 mm (interquartile range [IQR] 3.5–4.7) compared to the control group mean of 7.17 mm (SD 3.19) and median 6.70 mm (IQR 4.9–9.0). The median difference of 2.65 mm generated a P value of P < 0.0001, 95% confidence interval [CI] (1.82, 4.85). The mean LCIV diameter of the matched control group was 6.86 mm (SD 3.03) and median 6.70 mm (IQR 4.9–9.0). The difference of medians of 2.65 mm generated a P value of P = 0.007 and 95% CI (1.54, 4.55).{Figure 3}

A ROC curve was used to assess the diagnostic ability of a LCIV threshold, as shown in [Figure 4]. The total area under the curve was 0.802. At a threshold LCIV value of less than or equal to 4.7 mm, the sensitivity for diagnosis was 87.5% and the specificity 72.7%.{Figure 4}

Age group was outlined on a histogram in [Figure 5]. The trend line shows a bimodal distribution with distinct peaks above and below 45 years. For the 11 patients below 45 years old, 91% were female and the median LCIV diameter was 3.75 mm (IQR 2.7–4.5), 100% were primary MTS, and 0% secondary MTS. For the eight patients above 45 years old, 50% were female, the median LCIV diameter was 4.42 mm (IQR 3.9–5.3), 37.5% were primary MTS, and 62.5% secondary MTS.{Figure 5}


In patients where LCIV compression is seen on CT, the significance of such a finding harbours debate. While it is acknowledged that LCIV compression is an important component of the development of MTS, the syndrome was initially described in cadaveric specimens and there have been only limited studies correlating the syndrome with radiologic findings.[3],[4],[5],[6] Given the advent of high accuracy and readily available CT, assessment of LCIV diameter on CT has practical use for radiologists, vascular surgeons, and hematologists who are planning appropriate management.

This study showed that the minimum diameter of the LCIV in patients with symptoms of MTS was significantly lower than in asymptomatic controls, with a mean diameter of only 3.82mm. In previous studies, Ou-Yang and Lu suggested that mean LCIV is as narrow as 2.29mm in MTS, while Oguzkurt et al. suggested a mean diameter of 3.5 mm.[3,5-6] Park et al. found the mean diameter to be larger in their cohort ranging from 5.56 to 5.89 mm.[4] By setting a range of thresholds using a ROC curve against the data, choosing a threshold of 4.7 mm or less as the level of stenosis yields a high sensitivity and specificity in correlating imaging findings with a diagnosis of MTS.

There were two distinct groups of patients who presented with symptoms of MTS. In the group of 45 years old or less, there were more females, and none had and additional anatomic factor contributing to extrinsic compression. This is in keeping with the demographic of traditional MTS which is said to occur in younger females (primary MTS).[2]

However, this study showed that in patients who present with MTS over the age of 45 years, there was an equal incidence of males and females, and 62.5% had an additional anatomic factor contributing to compression (secondary MTS). This was most commonly anterior L5 disc-osteophyte complex and/or common iliac artery ectasia/aneurysm. This has practical implications in helping triage the potential cause of LCIV compression, in that for patients over 45 with a new diagnosis of MTS, it should be considered that the condition is likely secondary MTS and that a centrally obstructing lesion should be assessed with CT. Magnetic resonance imaging venography may represent an alternative cross-sectional imaging modality for this assessment, however was not possible in our country due to Medicare funding models which preclude its use.

What is also evident is that some compression of the LCIV is seen in normal asymptomatic patients and does not alone constitute disease: the minimum value in the matched control group was only 1.8 mm and 22.2% of matched patients had a minimum LCIV diameter less than the threshold value of 4.7mm. This suggests that LCIV compression plays a part in this condition; however, the inference of physiologically-relevant compression cannot be made alone without appropriate history of iliac or femoral DVT. This holds true for other vascular compression syndromes in the abdomen.[7] There are associated downstream implications for any potential treatment of LCIV compression in patients without symptoms, where compression may be a normal finding. Further studies are warranted to identify if there are any long-term ramifications of this finding in asymptomatic patients, for example, if these patients are at higher risk of future DVT.

This study is limited by a small sample size of only 19 patients. However, most symptomatic patients were <45 years of age, and in the authors' experience, CT is often avoided and the diagnosis made clinically whenever possible in this cohort. This also reflects the low prevalence of the condition and is supported by similar sample sizes in the prior literature. The single center and retrospective nature of this study is also acknowledged with the resultant potential for selection bias inherent in a study of this design.

The authors also acknowledge that a more sensitive measurement would be to perform cross-sectional diameter measurement with a CT workstation rather than measuring minimum diameter on PACS. In addition, the minimum diameter could have been compared to the prestenotic size or to the right common iliac vein to generate and compare a compression percentage. However, minimum diameter was chosen given it correlated with methods from previous studies and this provides standardization. It is also a quick and easily reproducible measurement even for a nonradiologist. In addition, it is acknowledged that the LCIV diameter will vary according to patient position and other factors such as systemic venous pressure changes and hydration. The authors also had concerns regarding the accuracy of percentage stenosis measurements in the context of prestenotic dilatation and/or distension from an acutely thrombosed vein.


This study shows that in patients with MTS, findings of LCIV compression on CT with a minimum diameter of <4.7 mm correlates with both high sensitivity and specificity. For patients diagnosed over 45 years old, a diagnosis of secondary MTS was more likely than primary MTS and CT should be considered as a part of disease workup in this group.

In addition, this study shows that LCIV compression alone in patients without symptoms does not constitute a diagnosis of MTS, as some of the matched controls had LCIV diameter less than the MTS threshold. Placing clinical significance on such a finding when there are no symptoms should be exercised with caution, and further studies on how and whether to follow-up such patients are warranted.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1May R, Thurner J. The cause of the predominantly sinistral occurrence of thrombosis of the pelvic veins. Angiology 1957;8:419-27.
2Harbin MM, Lutsey PL. May-Thurner syndrome: History of understanding and need for defining population prevalence. J Thromb Haemost 2020;18:534-42.
3Oguzkurt L, Ozkan U, Ulusan S, Koc Z, Tercan F. Compression of the left common iliac vein in asymptomatic subjects and patients with left iliofemoral deep vein thrombosis. J Vasc Interv Radiol 2008;19:366-70.
4Park SJ, Park HJ, Kwon EK, Kim SJ, Cho HR. Extrinsic compression of left iliac vein does not predict the development of post thrombotic syndrome in left side deep venous thrombosis. J Korean Surg Soc 2012;82:370-3.
5Ou-Yang L, Lu GM. Underlying anatomy and typing diagnosis of may-thurner syndrome and clinical significance: An observation based on CT. Spine (Phila Pa 1976) 2016;41:E1284-91.
6Oguzkurt L, Tercan F, Pourbagher MA, Kizilkilic O, Turkoz R, Boyvat F, et al. Computed tomography findings in 10 cases of iliac vein compression (May-Thurner) syndrome. Eur J Radiol 2005;55:421-5.
7Fong JK, Poh AC, Tan AG, Taneja R. Imaging findings and clinical features of abdominal vascular compression syndromes. Am J Roentgenol 2014;203:29-36.