Abstract
Background
To investigate the prevalence, spectrum, and predictors of alternative diagnoses explaining leg symptoms in patients negative for suspected acute deep venous thrombosis (DVT), which can be detected with whole-leg ultrasound.
Methods
We retrospectively analyzed a cohort of 789 patients (median age 70 years, 50.6% women) evaluated with a whole-leg ultrasound examination for suspected acute DVT within one year. All findings in the radiology report were analyzed and electronic chart review was performed to collect clinical information.
Results
Ultrasound was negative for acute DVT in 531 patients (67.3%). Among these, alternative diagnoses explaining leg symptoms were seen in 349 patients (65.7%). The most frequent alternative diagnoses were chronic venous insufficiency (147 patients, 27.7%), followed by lymphedema (48 patients, 9.0%) and chronic post-thrombotic changes (41 patients, 7.7%). Patients with alternative diagnoses were older (median 71 vs. 66 years, p = 0.0226), as well as more likely to present with leg swelling (39.5% vs. 23.1%, p = 0.0002), difference in leg circumference (25.5% vs. 14.8%, p = 0.0055) and redness (7.7% vs. 2.7%, p = 0.0213) than patients without alternative diagnosis. Independent predictors of finding alternative diagnoses on whole-leg ultrasound were older age (odds ratio 1.014 per year, p = 0.0119), leg swelling (OR 1.949, p = 0.0020) and history of previous DVT (OR 2.235, p = 0.0154).
Conclusions
Alternative diagnoses explaining leg symptoms can be detected on whole-leg ultrasound in two thirds of patients with no evidence of acute DVT. Our data supports performing a comprehensive ultrasound evaluation beyond the venous system, particularly, in older patients, who present with leg swelling and a past history of DVT.
Similar content being viewed by others
Explore related subjects
Find the latest articles, discoveries, and news in related topics.Background
Acute deep venous thrombosis (DVT) is a common condition for which patients seek emergency medical care and is associated with pulmonary embolism as a potentially life-threatening complication. Therefore, accurate diagnosis and prompt initiation of anticoagulant therapy is important to reduce morbidity [1]. Clinical signs and symptoms of DVT are often vague [2] and laboratory D-dimer tests, routinely used for DVT screening, tend to have low specificity [3]. Duplex ultrasound of the lower extremity has therefore become the standard radiological screening test for patients with suspected acute DVT [4].
Some centers choose to perform a limited compression ultrasound examination as the initial test focusing primarily on the femoral and popliteal veins [5,6,7,8]. We prefer to perform a whole-leg ultrasound in all patients with suspected acute DVT in our institution, conforming to current multidisciplinary guidelines [4]. The major advantage of this approach, in addition to detection of isolated calf DVT, is that it allows to identify alternative diagnoses which may have caused the patient’s leg symptoms, thereby mimicking acute DVT [9,10,11,12].
In most cohort studies, the majority of focused ultrasound examinations for DVT in patients with suspected acute DVT are negative (56 to 77%) [13,14,15]. However, most of these patients had signs or symptoms that led to the clinical suspicion of acute DVT. For both health care providers and patients, it can be unsatisfactory to rule out DVT and discharge the patient without identifying a cause for the patient’s symptoms. Many alternative diagnoses such as chronic venous insufficiency (CVI) or soft tissue inflammation require specific treatment and it would be prudent to not miss these diagnoses during the initial work-up. Hence, evaluation for alternative diagnoses using whole-leg ultrasound in patients with negative acute DVT scans may enhance the overall quality of patient care as well as patients’ and health care providers’ satisfaction.
Therefore, we performed this study to investigate the prevalence, spectrum and predictors of alternative diagnosis explaining leg symptoms among patients evaluated with whole-leg ultrasound for suspected DVT.
Methods
Ethical approval, study design and patient selection
The study was approved by our institutional review board with waiver of informed consent. The investigation was designed as a retrospective, single-centre cohort study. We included all patients (1) who were examined with whole-leg ultrasound (2) at our institution (3) between January 1 and December 31, 2014 (4) for suspected DVT (5) but no findings suggestive of acute DVT were seen on ultrasound. The year 2014 was chosen because during this year venous ultrasound evaluations were almost exclusively performed by two senior radiologists with extensive experience in vascular ultrasound. We excluded patients with (1) other indications for venous ultrasound and (2) follow-up examinations for established acute DVT diagnosed in the past 3 months (Fig. 1). All eligible patients were identified through a retrospective query of our radiology information system (Centricity 5.0, GE Healthcare).
Ultrasound technique
We perform a whole-leg ultrasound of the symptomatic lower extremity in all patients with clinical suspicion of acute DVT in our institution as a combination of color-coded duplex ultrasound, compression ultrasound, and B-mode ultrasound from the groin to the lower leg by using a high-end ultrasound device (Aplio XG, Toshiba) and a linear transducer (PLT 604 AT, 6 MHz, Toshiba). In the case of suspicion of deep vein thrombosis of the iliac veins, the iliac veins are additionally examined using a convex transducer (PVT 375 BT, 3.5 MHz, Toshiba). The images are stored in the digital image archive (PACS, Agfa Impax 6.5.3). The ultrasound evaluation is performed or supervised by board-certified radiologists with subspecialisation in vascular and interventional radiology. Our protocol includes.
-
(a)
compression ultrasound performed at 2 cm intervals from the common femoral vein to the ankle including the peroneal and posterior tibial veins in the calf,
-
(b)
colour-coded duplex evaluation from the common femoral vein to the ankle,
-
(c)
spectral doppler evaluation of the wave form in the common femoral vein (in case of abnormal waveform suggesting obstruction above the inguinal ligament, additional colour-coded duplex evaluation of the iliac veins is performed) and
-
(d)
targeted ultrasound of symptomatic areas if symptoms are not explained by findings on standard thigh-to-ankle examination.
Analysis of radiology reports
Radiology reports of the whole-leg ultrasound examination were retrospectively analyzed for the presence, type, location and presumed etiology of alternative diagnosis.
Analysis of clinical data
Review of electronic patient charts was performed to record age, gender, presenting symptoms, risk factors, Wells scores and D-Dimer levels.
Statistical analysis
Statistical analysis was performed with GraphPad Prism (version 8.4.2, GraphPad Software Inc). Continuous data were presented as median and interquartile range and compared using the nonparametric Mann–Whitney test. Categorical data were displayed as frequencies and proportions and compared between groups using Fisher’s exact test. We performed multiple logistic regression analysis to identify independent predictors of finding an alternative diagnosis in patients without acute DVT on ultrasound. Age and gender were entered into the model as predefined variables. Additionally, we included all variables with significant inter-group differences on univariate analysis. Due to collinearity between both variables, only leg swelling but not measured circumference difference was entered into the model. P values of < 0.05 were regarded as statistically significant.
Results
Patient characteristics
Our final study cohort consisted of 531 of patients with a whole-leg ultrasound negative for acute DVT, of which 269 (50.7%) were women. Patient characteristics are summarized in Table 1. Median age was 70 years (interquartile range 56–78 years). The most common local symptoms were leg pain (35.6%, 189 patients) and swelling (33.9%, 180 patients). 11.1% of patients had active malignancy and 11.9% of patients had a past medical history of DVT.
Prevalence and spectrum of alternative diagnosis
Alternative findings explaining leg symptoms were found on whole-leg ultrasound examination in 349 of 531 patients (65.7%, Fig. 1). The most common alternative findings were chronic venous insufficiency, found in 147 patients (27.7%), followed by lymphedema (48 patients, 9.0%) and post-thrombotic changes (41 patients, 7.7%). The complete spectrum of alternative findings is presented in Table 2.
Comparison of patients with and without alternative diagnosis
Patients with alternative diagnoses were older than patients without alternative findings observed on whole-leg ultrasound (median 71 vs. 66 years, p = 0.0226) and more likely to present with leg swelling (39.5% vs. 23.1%, p = 0.0002), circumference difference (25.5% vs. 14.8%, p = 0.0055) and redness (7.7% vs. 2.7%, p = 0.0213). They were also more likely to have a previous history of DVT (14.3% vs. 7.1%, p = 0.0160). There were no differences in gender, proportion of cancer patients, D-Dimers and Wells score (Table 1).
Multivariate analysis
On multiple logistic regression analysis (Table 3), age (Odds Ratio 1.014 for each one-year increment, p = 0.0119), leg swelling (OR 1.949, p = 0.0020) and previous DVT (OR 2.235, p = 0.0154) were independent predictors of finding alternative diagnoses in patients without acute DVT on whole-leg ultrasound.
Discussion
In this cohort study of patients negative for acute DVT on whole-leg ultrasound, alternative diagnoses were found in two thirds of the patients with CVI being the most common alternative diagnosis. Patients with alternative diagnoses were older, more likely to present with leg swelling, difference in leg circumference and redness than patients without an identifiable alternative diagnosis. Independent predictors of finding alternative diagnoses on whole-leg ultrasound were older age, leg swelling and a past history of DVT.
Few prior studies have investigated alternative diagnoses in patients with suspected acute DVT found on whole-leg ultrasound examination [9, 16, 17]. In these studies, alternative diagnoses were observed less often with whole-leg ultrasound, (11% to 31% of all patients with suspected DVT) [9, 16, 17] as compared to our study (44%). In all of these studies, ultrasound scans were either performed by ultrasound technicians [16]/sonographers [17], or by vascular technologists and interpreted by vascular surgeons [9]. This is in contrast to our study, here board-certified radiologists with subspecialisation in vascular and interventional radiology. performed or directly supervised the ultrasound examinations. It can therefore be speculated that the higher prevalence in our study may in part be due to radiologist-performed vs. non radiologist-performed ultrasound, since it is known that duplex ultrasound is highly operator dependent [18, 19]. The most common alternative diagnoses reported in these studies included chronic venous insufficiency [9], lymphadenopathy [16] and old thrombosis/post-thrombotic changes [17] similar to our study.
We assessed predictors for finding an alternative diagnosis on whole-leg ultrasound in patients suspected but negative for acute DVT. In a study to identify the risk factors for the common alternative diagnoses in patients with clinically suspected DVT (CVI, erysipelas, muscle rupture/hematoma and superficial venous thrombosis) [20] Cate Hoek et al. found that patients with CVI reported more often swelling of the entire leg having an insidious onset, as compared to patients with a confirmed diagnosis of acute DVT. Limb redness was seen as a distinctive feature in patients with erysipelas. Patients with muscle rupture/hematoma had neither swelling nor redness. Furthermore, patients with superficial venous thrombosis had tenderness on palpation of the offending vein [20]. However, all these four alternative diagnoses were based on clinical evaluation in a primary care setting and not based on whole-leg ultrasound findings as in our study [20].
In our analysis, older age, leg swelling and previous DVT were independent predictors of finding alternative diagnoses in patients without acute DVT on whole-leg ultrasound. A frequent complication of DVT includes the post-thrombotic syndrome (PTS) with a relatively high number of patients between 20 to 50% within 2 years of DVT diagnosis [21, 22]. Clinical manifestations of PTS typically include limb pain, heaviness, edema and pruritus [23]. PTS is a form of CVI that occurs due to chronic venous obstruction and damaged vein valves [21, 22]. Duplex ultrasound is the first imaging test of choice to evaluate for signs of CVI and post-thrombotic changes [24]. This might explain, why a past history of DVT was identified as risk factor for finding an alternative diagnosis on whole-leg ultrasound.
Since alternative diagnoses are relatively common in patients having a suspicion of acute DVT and these alternative diagnoses have important therapeutic consequences [10], focused sonographic evaluation of the venous system alone may not be sufficient in all patients with suspected acute DVT presenting with lower extremity symptoms. Useche et al. [10] state that in addition to whole-leg sonography in supine position, comparison of both extremities, performing specific maneuvers to elicit symptoms, as well as observing changes due to articular movement and switching to a standing position during the ultrasound examination might also be helpful. However, resources and time required may be limited in most emergency departments [16]. Nevertheless, the effort to establish an alternative diagnosis, when DVT is ruled out, seems particularly advisable in patients, who meet the following criteria: older age, leg swelling and previous history of DVT, which we identified to be independent predictors in our study. It should be kept in mind that patient compliance with follow-up ultrasound examinations after initial ultrasound negative for DVT seems to be extremely low [25], which also advocates the importance of performing a comprehensive initial ultrasound evaluation. As a practical implication of our study, we suggest to add targeted ultrasound of symptomatic areas if no DVT is found on standard thigh-to-ankle venous doppler ultrasound examination; particularly in older patients who present with leg swelling and a previous history of DVT.
This study has some limitations, which include its retrospective single-center nature. Symptomatic patients who presented to a university hospital were included in our analysis. This patient population may be different as compared to patients seen by primary care physicians. No external reference standard was available to confirm the findings. A more comprehensive prospective study is needed to determine the impact of alternative findings on subsequent diagnostic algorithm in patients and the benefits including costs in patient management.
Conclusions
In summary, we found that alternative findings explaining leg symptoms can be detected on whole-leg ultrasound in two thirds of patients negative for acute DVT. Therefore, our study supports performing a detailed whole-leg ultrasound evaluation beyond the venous system (including targeted ultrasound of symptomatic areas) when negative for acute DVT. This may be particularly relevant in older patients, who present with leg swelling and who have a previous history of DVT, which were identified as independent predictors in our patient cohort.
Availability of data and materials
The datasets used and analyzed during this study are available from the corresponding author on reasonable request.
Abbreviations
- CVI:
-
Chronic venous insufficiency
- DVT:
-
Deep venous thrombosis
- OR:
-
Odds ratio
References
Karande GY, Hedgire SS, Sanchez Y, Baliyan V, Mishra V, Ganguli S, et al. Advanced imaging in acute and chronic deep vein thrombosis. Cardiovasc Diagn Ther. 2016;6(6):493–507.
Tovey C, Wyatt S. Diagnosis, investigation, and management of deep vein thrombosis. BMJ. 2003;326(7400):1180–4.
Kelly J, Rudd A, Lewis RR, Hunt BJ. Plasma D-dimers in the diagnosis of venous thromboembolism. Arch Intern Med. 2002;162(7):747–56.
Needleman L, Cronan JJ, Lilly MP, Merli GJ, Adhikari S, Hertzberg BS, et al. Ultrasound for lower extremity deep venous thrombosis: multidisciplinary recommendations from the society of radiologists in ultrasound consensus conference. Circulation. 2018;137(14):1505–15.
Zuker-Herman R, Ayalon Dangur I, Berant R, Sitt EC, Baskin L, Shaya Y, et al. Comparison between two-point and three-point compression ultrasound for the diagnosis of deep vein thrombosis. J Thromb Thrombolysis. 2018;45(1):99–105.
Jahanian F, Khatir IG, Bani-Mostafavi E-S, Moradi S, Aghamalaki FH. Diagnostic accuracy of a three-point compression ultrasonography performed by emergency medicine resident for the diagnosis of deep vein thrombosis: a prospective diagnostic study. Acta Informatica Medica. 2019;27(2):119–22.
Zitek T, Baydoun J, Yepez S, Forred W, Slattery DE. Mistakes and pitfalls associated with two-point compression ultrasound for deep vein thrombosis. West J Emerg Med. 2016;17(2):201–8.
Adhikari S, Zeger W, Thom C, Fields JM. Isolated deep venous thrombosis: implications for 2-point compression ultrasonography of the lower extremity. Ann Emerg Med. 2015;66(3):262–6.
Adhikari S, Zeger W. Non-thrombotic abnormalities on lower extremity venous duplex ultrasound examinations. West J Emerg Med. 2015;16(2):250–4.
Useche JN, Castro AMFD, Galvis GE, Mantilla RA, Ariza A. Use of US in the evaluation of patients with symptoms of deep venous thrombosis of the lower extremities. Radiographics. 2008;28(6):1785–97.
Naringrekar H, Sun J, Ko C, Rodgers SK. It’s not all deep vein thrombosis: sonography of the painful lower extremity with multimodality correlation. J Ultrasound Med. 2019;38(4):1075–89.
Leow KS, Chew KM, Chawla A, Lim TC. Sonographic assessment of musculoskeletal causes of calf pain and swelling. Emerg Radiol. 2019;26(3):349–59.
Dehbozorgi A, Damghani F, Mousavi-Roknabadi RS, Sharifi M, Sajjadi SM, Hosseini-Marvast SR. Accuracy of three-point compression ultrasound for the diagnosis of proximal deep-vein thrombosis in emergency department. J Res Med Sci. 2019;24:80.
Schellong S, Schwarz T, Halbritter K, Beyer J, Siegert G, Oettler W, et al. Complete compression ultrasonography of the leg veins as a single test for the diagnosis of deep vein thrombosis. Thromb Haemost. 2003;89(02):228–34.
Schutgens RE, Haas FJ, Biesma DH. Reduced efficacy of clinical probability score and D-dimer assay in elderly subjects suspected of having deep vein thrombosis. Br J Haematol. 2005;129(5):653–7.
Sutter ME, Turnipseed SD, Diercks DB, Samuel P, White RH. Venous ultrasound testing for suspected thrombosis: incidence of significant non-thrombotic findings. J Emerg Med. 2009;36(1):55–9.
de Oliveira A, Franca GJ, Vidal EA, Stalke PS, Baroncini LA. Duplex scan in patients with clinical suspicion of deep venous thrombosis. Cardiovasc Ultrasound. 2008;6:53.
Shabani Varaki E, Gargiulo GD, Penkala S, Breen PP. Peripheral vascular disease assessment in the lower limb: a review of current and emerging non-invasive diagnostic methods. Biomed Eng Online. 2018;17(1):61.
Caronia J, Sarzynski A, Tofighi B, Mahdavi R, Allred C, Panagopoulos G, et al. Resident performed two-point compression ultrasound is inadequate for diagnosis of deep vein thrombosis in the critically III. J Thromb Thrombolysis. 2014;37(3):298–302.
ten Cate-Hoek AJ, van der Velde EF, Toll DB, van Weert HC, Moons KG, Buller HR, et al. Common alternative diagnoses in general practice when deep venous thrombosis is excluded. Neth J Med. 2012;70(3):130–5.
Thukral S, Vedantham S. Catheter-based therapies and other management strategies for deep vein thrombosis and post-thrombotic syndrome. J Clin Med. 2020;9(5):1439.
Kahn SR. The post-thrombotic syndrome. Hematology. 2016;2016(1):413–8.
Makedonov I, Kahn SR, Galanaud J-P. Prevention and management of the post-thrombotic syndrome. J Clin Med. 2020;9(4):923.
Farrell JJ, Sutter C, Tavri S, Patel I. Incidence and interventions for post-thrombotic syndrome. Cardiovasc Diagn Ther. 2016;6(6):623–31.
McIlrath ST, Blaivas M, Lyon M. Patient follow-up after negative lower extremity bedside ultrasound for deep venous thrombosis in the ED. Am J Emerg Med. 2006;24(3):325–8.
Acknowledgements
Not applicable.
Funding
Open Access funding enabled and organized by Projekt DEAL. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Author information
Authors and Affiliations
Contributions
FGM, JCK and TH conceived and designed the study. EB, TH, and MB analyzed and interpreted the data. MAW, RR and RH contributed analysis tools and interpretation of the data. MB, EB, FGM and TH were major contributors in writing the manuscript. JCK, RR, RH and MAW have substantively revised the manuscript. All authors revised the manuscript critically for important intellectual content and approved the version to be published. All authors read and approved the final manuscript.
Corresponding author
Ethics declarations
Ethics approval and consent to participate
The study was approved by our institutional review board at the Rostock University Medical Center. Due to the retrospective nature, written informed consent was waived by the institutional review board.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
About this article
Cite this article
Beller, E., Becher, M., Meinel, F.G. et al. Prevalence and predictors of alternative diagnoses on whole-leg ultrasound negative for acute deep venous thrombosis. BMC Med Imaging 20, 127 (2020). https://doi.org/10.1186/s12880-020-00527-7
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/s12880-020-00527-7