Diagnosing deep vein thrombosis in cancer patients with suspected symptoms: An individual participant data meta-analysis

[Translation into French and republication of: "Recurrent venous thromboembolism in anticoagulated cancer patients: diagnosis and treatment"]

Patients with cancer are at significantly increased risk of venous thromboembolism (VTE), due both to the impact of malignant disease itself and to the impact of certain anticancer drugs on haemostasis. This is true both for first episode venous thromboembolism and recurrence. The diagnosis and management of VTE recurrence in patients with cancer poses particular challenges, and these are reviewed in the present article, based on a systematic review of the relevant scientific literature published over the last decade. Furthermore, it is uncertain whether diagnostic algorithms for venous thromboembolism, validated principally in untreated non-cancer patients, are also valid in anticoagulated cancer patients: the available data suggests that clinical decision rules and D-dimer testing perform less well in this clinical setting. In patients with cancer, computed tomography pulmonary angiography and venous ultrasound appear to be the most reliable diagnostic tools for diagnosis of pulmonary embolism and deep vein thrombosis respectively. Options for treatment of venous thromboembolism include low molecular weight heparins (at a therapeutic dose or an increased dose), fondaparinux or oral direct factor Xa inhibitors. The choice of treatment should take into account the nature (pulmonary embolism or VTE) and severity of the recurrent event, the associated bleeding risk, the current anticoagulant treatment (type, dose, adherence and possible drug-drug interactions) and cancer progression.

Recurrent venous thromboembolism in anticoagulated cancer patients: Diagnosis and treatment

Patients with cancer are at significantly increased risk of venous thromboembolism (VTE), due both to the impact of malignant disease itself and to the impact of certain anticancer drugs on haemostasis. This is true both for first episode venous thromboembolism and recurrence. The diagnosis and management of VTE recurrence in patients with cancer poses particular challenges, and these are reviewed in the present article, based on a systematic review of the relevant scientific literature published over the last decade. Furthermore, it is uncertain whether diagnostic algorithms for venous thromboembolism, validated principally in untreated non-cancer patients, are also valid in anticoagulated cancer patients: the available data suggests that clinical decision rules and D-dimer testing perform less well in this clinical setting. In patients with cancer, computed tomography pulmonary angiography and venous ultrasound appear to be the most reliable diagnostic tools for diagnosis of pulmonary embolism and deep vein thrombosis respectively. Options for treatment of venous thromboembolism include low molecular weight heparins (at a therapeutic dose or an increased dose), fondaparinux or oral direct factor Xa inhibitors. The choice of treatment should take into account the nature (pulmonary embolism or VTE) and severity of the recurrent event, the associated bleeding risk, the current anticoagulant treatment (type, dose, adherence and possible drug-drug interactions) and cancer progression.

Age- versus clinical pretest probability-adjusted D-dimer to rule out lower-extremity deep vein thrombosis in ambulatory patients with active cancer

BACKGROUND

In patients with suspected deep vein thrombosis (DVT), D-dimer thresholds adjusted to age or clinical pretest probability (CPTP) increase the proportion of patients in whom DVT can be safely excluded compared to a standard approach using a fixed D-dimer threshold. Performance of these diagnostic strategies among cancer patients is uncertain.

AIM

To compare the performance of age- and CPTP-adjusted D-dimer approaches among cancer outpatients with clinically suspected DVT, and derive a cancer-specific CPTP rule.

PATIENTS AND METHODS

Consecutive ambulatory patients with active cancer and clinically suspected DVT of the lower extremity underwent CPTP assessment using the Wells rule, D-dimer testing, and whole-leg compression ultrasonography. Patients with normal ultrasonography were followed-up for 3 months for the occurrence of symptomatic venous thromboembolism.

RESULTS

Upon referral, DVT was diagnosed in 48 of 239 (20.1 %) patients. The age-adjusted approach showed higher specificity and efficiency than the standard approach. Compared to the standard and age-adjusted strategies, the CPTP-adjusted approach had 35 % and 21 % higher specificity, and 34 % and 21 % higher efficiency, respectively. Failure rate, sensitivity, and predictive values were similar across strategies. A simplified CPTP score derived from the Wells rule reduced unnecessary imaging with similar accuracy and efficiency, but higher failure rate.

CONCLUSIONS

In this prospective cohort of ambulatory cancer patients with clinically suspected DVT, the CPTP-adjusted D-dimer approach held the highest specificity and efficiency, potentially safely reducing unnecessary ultrasonography examinations compared to other approaches. Additional studies are warranted to evaluate the use of a simplified clinical prediction rule in this setting.

Deep Vein Thrombosis in Intravenous Drug Users: An Invisible Global Health Burden

The prevalence of intravenous drug use has increased in the past decade and it represents an important risk factor for deep vein thrombosis. Intravenous drug use is a global problem, with the main culprit being heroin. Peer pressure and poverty in high-risk groups such as sex workers, females, and young adults raise the risk of intravenous drug use, which expresses itself in the form of venous thromboembolism eventually. Deep vein thrombosis typically manifests itself eight years after the initial intravenous drug administration, rendering it a silent killer. Aiming to review and summarize existing articles in this context, we performed an exhaustive literature search online on PubMed and Google Scholar indexes using the keywords "Deep Venous Thrombosis (DVT)" and "Intravenous Drug Users (IVDU)." English articles that addressed epidemiology, pathogenesis, clinical manifestations, diagnosis, differential diagnosis, management, and outcomes of DVT, including those in IVDU, were selected and analyzed. The pathogenesis of DVT development in IVDU is mainly attributed to the interplay of trauma to the vessel by repeated injection and the injected drug itself. The right-sided femoral vein is the most common vein affected. Prevalent clinical presentations include local pain, swelling, and redness with typical systemic symptoms including fever, cough, dyspnea, and chest pain on top of addiction features. There appeared to be a delay in reporting symptoms, which was most likely due to the social stigma attached to IVDU. There are over 50 conditions that present with swollen and painful limbs comparable to DVT in IVDU, making precise diagnosis critical for timely treatment. Venous ultrasound is the method of choice for diagnosing DVT. Extended anticoagulant therapy with low-molecular-weight heparin combined with warfarin is the recommended treatment. Intravenous drug abusers having DVT are affected by multiple complications and poorer outcomes such as slower recovery, recurrent venous thromboembolism (VTE), and a longer hospital stay, which put them at higher risk of morbidity, mortality, reduced productivity, and economic burden.

Cost-Effective Machine Learning Based Clinical Pre-Test Probability Strategy for DVT Diagnosis in Neurological Intensive Care Unit

In order to overcome the shortage of the current costly DVT diagnosis and reduce the waste of valuable healthcare resources, we proposed a new diagnostic approach based on machine learning pre-test prediction models using EHRs. We examined the sociodemographic and clinical factors in the prediction of DVT with 518 NICU admitted patients, including 189 patients who eventually developed DVT. We used cross-validation on the training data to determine the optimal parameters, and finally, the applied ROC analysis is adopted to evaluate the predictive strength of each model. Two models (GLM and SVM) with the strongest ROC were selected for DVT prediction, based on which, we optimized the current intervention and diagnostic process of DVT and examined the performance of the proposed approach through simulations. The use of machine learning based pre-test prediction models can simplify and improve the intervention and diagnostic process of patients in NICU with suspected DVT, and reduce the valuable healthcare resource occupation/usage and medical costs.

Current Challenges in Diagnosis of Venous Thromboembolism

In patients with suspected venous thromboembolism, the goal is to accurately and rapidly identify those with and without thrombosis. Failure to diagnose venous thromboembolism (VTE) can lead to fatal pulmonary embolism (PE), and unnecessary anticoagulation can cause avoidable bleeding. The adoption of a structured approach to VTE diagnosis, that includes clinical prediction rules, D-dimer testing and non-invasive imaging modalities, has enabled rapid, cost-effective and accurate VTE diagnosis, but problems still persist. First, with increased reliance on imaging and widespread use of sensitive multidetector computed tomography (CT) scanners, there is a potential for overdiagnosis of VTE. Second, the optimal strategy for diagnosing recurrent leg deep venous thrombosis remains unclear as is that for venous thrombosis at unusual sites. Third, the conventional diagnostic approach is inefficient in that it is unable to exclude VTE in high-risk patients. In this review, we outline pragmatic approaches for the clinician faced with difficult VTE diagnostic cases. In addition to discussing the principles of the current diagnostic framework, we explore the diagnostic approach to recurrent VTE, isolated distal deep-vein thrombosis (DVT), pregnancy associated VTE, subsegmental PE, and VTE diagnosis in complex medical patients (including those with impaired renal function).