RIETE Score for Risk of Hemorrhage in Pulmonary Embolism Treatment

Use the RIETE Score to predict major bleeding risk during anticoagulant therapy for pulmonary embolism and VTE. A 6-variable score (0-8 points) derived from the 13,057-patient RIETE registry stratifying patients into low, intermediate, and high hemorrhagic risk during the first 3 months of treatment.

RIETE Score Calculator

Indicate which risk factors are present in the patient with acute venous thromboembolism. The RIETE score predicts the risk of major bleeding during the first 3 months of anticoagulant therapy.

Risk factors

Disclaimer: The RIETE score is an educational clinical decision-support tool. It does not replace individualized assessment of bleeding and thrombotic risk, multidisciplinary team discussion, or clinical judgment. Anticoagulation decisions should incorporate the full clinical picture, patient preferences, contraindications, and institutional protocols. Always consult relevant guidelines and specialist expertise.

RIETE Score: formula, variables, and clinical context

Overview

The RIETE Score for Risk of Hemorrhage in Pulmonary Embolism Treatment is a validated prognostic tool derived from the RIETE registry (Registro Informatizado de la Enfermedad TromboEmbólica venosa), an ongoing international registry of consecutive patients with acute venous thromboembolism (VTE). Developed by Ruíz-Giménez and colleagues in 2008 using a derivation cohort of 13,057 patients and internally validated on 6,572 patients, it predicts the risk of major bleeding during the first 3 months of anticoagulant therapy by summing points from six clinical variables.

Scoring variables

Variable	Criteria	Points
Recent major bleeding	Yes	+2
Recent major bleeding	No	0
Anemia	Hgb < 13 g/dL (male) or < 12 g/dL (female)	+1.5
Anemia	Normal hemoglobin	0
Creatinine	> 1.2 mg/dL (106.1 µmol/L)	+1.5
Creatinine	≤ 1.2 mg/dL	0
Malignancy	Yes	+1
Malignancy	No	0
Clinically overt PE	Yes	+1
Clinically overt PE	No (DVT only or incidental PE)	0
Age	> 75 years	+1
Age	≤ 75 years	0
Total score range		0-8

Risk stratification

Score	Risk category	Major bleeding rate
0	Low	0.1%
1	Intermediate	1.4%
1.5-2		2.2%
2.5-3		4.4%
3.5-4		4.2%
4.5-5	High	4.9%
5.5-6		11%
> 6		20%

Major bleeding rates are from the original derivation cohort (n = 13,057). The risk categories (low, intermediate, high) group scores of 0, 1-4, and > 4 respectively.

Variable definitions

Recent major bleeding — 2 points

A major bleeding event occurring before or at the time of VTE diagnosis. Major bleeding is defined in the RIETE registry as overt bleeding accompanied by at least one of the following: requiring transfusion of 2 or more units of packed red blood cells, retroperitoneal haemorrhage, spinal or intracranial haemorrhage, or fatal bleeding. This criterion carries the highest point weight because prior major haemorrhage is among the strongest predictors of subsequent bleeding during anticoagulation.

Anemia (Hgb < 13 g/dL male, < 12 g/dL female) — 1.5 points

Baseline anaemia at the time of VTE diagnosis, using WHO-defined sex-specific thresholds (haemoglobin < 13 g/dL for men and < 12 g/dL for women). Anaemia may reflect occult blood loss, chronic disease, or a haemostatic abnormality that increases the propensity for clinically significant bleeding during anticoagulant therapy. It also reduces the physiological reserve for tolerating any bleeding episode.

Creatinine > 1.2 mg/dL (106.1 µmol/L) — 1.5 points

Elevated serum creatinine indicating impaired renal function at baseline. Renal insufficiency alters the clearance of many anticoagulants (especially low- molecular-weight heparins and direct oral anticoagulants), leading to higher drug exposure and increased bleeding risk. It is also associated with platelet dysfunction and impaired primary haemostasis.

Malignancy — 1 point

Active or known history of cancer. Cancer-associated VTE is a common clinical scenario, and these patients face a dual challenge of high thrombotic risk and elevated bleeding risk during anticoagulation. Tumour-related factors (local invasion, neo-vascularisation, thrombocytopenia from chemotherapy) and treatment factors (surgery, radiation) all contribute to the increased haemorrhagic risk.

Clinically overt pulmonary embolism — 1 point

PE that presented with symptoms such as dyspnoea, chest pain, haemoptysis, or haemodynamic compromise. This is distinguished from isolated deep vein thrombosis (DVT) or incidentally discovered PE. Symptomatic PE may indicate a larger clot burden and is associated with more aggressive initial anticoagulation, which in turn increases bleeding risk.

Age > 75 years — 1 point

Advanced age is a well-established risk factor for bleeding during anticoagulation. Older patients are more likely to have comorbidities (hepatic or renal dysfunction, polypharmacy, falls risk), reduced physiological reserve, and age-related changes in haemostatic pathways. The RIETE registry demonstrated a clear incremental risk of major haemorrhage with age exceeding 75 years.

Definition of major bleeding

In the RIETE registry, major bleeding is defined as overt bleeding accompanied by at least one of the following:

Required transfusion of 2 or more units of packed red blood cells
Retroperitoneal haemorrhage
Spinal haemorrhage
Intracranial haemorrhage
Fatal bleeding

This is typically associated with a decline in haemoglobin concentration of at least 20 g/L (2 g/dL).

Study details

Registry: RIETE (Registro Informatizado de la Enfermedad TromboEmbólica venosa), an ongoing international, multicentre prospective registry of consecutive patients with objectively confirmed acute VTE.
Derivation cohort: 13,057 patients with acute VTE (DVT, PE, or both) enrolled in the RIETE registry.
Validation cohort: 6,572 patients from the same registry (internal temporal validation).
Outcome: Major bleeding events during the first 3 months of anticoagulant therapy.
Statistical method: Multivariate logistic regression identified six independent predictors. Points were assigned based on beta coefficients.
Performance: The score demonstrated good discrimination between risk categories in both derivation and validation cohorts. In the derivation cohort, major bleeding rates ranged from 0.1% (score 0) to 7.3% (score > 4).

Limitations

Internal validation only: The RIETE score was validated within the same registry. Independent external validation in different healthcare settings and populations has been limited, though subsequent studies using the RIETE registry have confirmed its utility.
Registry-based derivation: As an observational registry, the RIETE database may not capture all confounders, and treatment practices may vary across participating centres and over time.
Anticoagulant era: The score was derived primarily in the era of vitamin K antagonists (VKA) and low-molecular-weight heparins (LMWH). Its applicability to patients treated with direct oral anticoagulants (DOACs) has not been fully established.
Does not include all bleeding risk factors: Variables such as concomitant antiplatelet therapy, liver disease, alcohol use, prior stroke, and thrombocytopenia are not included in the score. Clinicians should consider these factors alongside the RIETE score.
3-month timeframe: The score predicts bleeding risk during the initial 3 months of anticoagulation. It does not address bleeding risk during extended or indefinite anticoagulation.

Key references

Ruíz-Giménez N, Suárez C, González R, Nieto JA, Todolí JA, Samperiz AL, Monreal M; RIETE Investigators. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism. Findings from the RIETE Registry. Thromb Haemost. 2008;100(1):26-31.
Guijarro R, Montes J, Sanromán C, Monreal M; RIETE Investigators. Venous thromboembolism in Spain. Comparison between an administrative database and the RIETE registry. Eur J Intern Med. 2008;19(6):443-446.
Laporte S, Mismetti P, Décousus H, et al; RIETE Investigators. Clinical predictors for fatal pulmonary embolism in 15,520 patients with venous thromboembolism: findings from the RIETE Registry. Circulation. 2008;117(13):1711-1716.
Trujillo-Santos J, Prandoni P, Rivron-Guillot K, et al; RIETE Investigators. Clinical outcome in patients with venous thromboembolism and hidden cancer: findings from the RIETE Registry. J Thromb Haemost. 2008;6(2):251-255.
Klok FA, Hösel V, Engelen DJPG, et al. Prediction of Major Bleeding in Anticoagulated Patients for Venous Thromboembolism: Comparison of the RIETE and the VTE-BLEED Scores. Thromb Haemost. 2021;121(10):1373-1381.

Practice caveats

The RIETE score was derived from a large registry predominantly reflecting European and Latin American practice patterns. Its performance in other populations and healthcare settings should be interpreted with caution.
The score should be used as one input among many when weighing the benefits and risks of anticoagulant therapy. It does not supersede comprehensive clinical assessment.
For patients with cancer-associated VTE, dedicated guidelines (e.g., ISTH, ASCO) provide specific anticoagulation recommendations that should be considered alongside the RIETE score.
In the era of DOACs, the absolute bleeding rates may differ from those derived in the VKA/LMWH era. The relative risk stratification is likely preserved, but updated validation is warranted.

Introduction

Anticoagulant therapy is the cornerstone of treatment for acute venous thromboembolism (VTE), encompassing both deep vein thrombosis (DVT) and pulmonary embolism (PE). While anticoagulation dramatically reduces the risk of thrombus propagation, recurrent VTE, and death from PE, it does so at the cost of an inherent risk of hemorrhagic complications. Major bleeding during anticoagulant therapy occurs in approximately 1% to 4% of VTE patients during the initial three months of treatment, with rates as high as 10% to 20% in patients with multiple risk factors. The clinical consequences of major bleeding are substantial: it can necessitate blood transfusion, surgical intervention, or interruption of anticoagulation (which in turn increases the risk of recurrent thromboembolism), and in the most severe cases, it is fatal.

The challenge facing clinicians is that the patients who are at highest risk for recurrent VTE are often the same patients who are at highest risk for bleeding. Advanced age, malignancy, renal insufficiency, and recent surgery are risk factors for both thrombosis and hemorrhage. This creates a clinical tension in which the decision to initiate, continue, or intensify anticoagulation must be weighed against the individualized bleeding risk. Tools that quantify the risk of major bleeding help clinicians navigate this balance, enabling more informed shared decision-making with patients, more targeted selection of anticoagulant agents and doses, and more appropriate monitoring strategies.

The RIETE Score for Risk of Hemorrhage in Pulmonary Embolism Treatment was developed from the RIETE registry (Registro Informatizado de la Enfermedad Tromboembólica venosa), one of the largest prospective registries of acute VTE worldwide. Published by Ruíz-Giménez and colleagues in Thrombosis and Haemostasis in 2008, the score uses six readily available clinical and laboratory variables to stratify patients into low, intermediate, and high risk categories for major bleeding during the first three months of anticoagulant therapy.

The RIETE Registry

The RIETE registry is an ongoing, international, multicenter, prospective registry that has been enrolling consecutive patients with objectively confirmed acute VTE since March 2001. It is one of the largest and most comprehensive real-world databases of VTE management and outcomes, with contributions from over 200 hospitals, predominantly in Spain and other European and Latin American countries, and increasingly from North America and Asia.

Registry Design and Data Collection

Patients enrolled in the RIETE registry are those with objectively confirmed symptomatic or incidentally discovered acute DVT or PE. Objective confirmation requires venous ultrasound or venography for DVT, and computed tomography pulmonary angiography (CTPA), ventilation-perfusion scanning, or pulmonary angiography for PE. At enrollment, the registry captures a comprehensive set of demographic, clinical, laboratory, and treatment variables. Patients are followed for at least three months, with systematic recording of outcomes including recurrent VTE, major bleeding, clinically relevant non-major bleeding, and all-cause mortality. The prospective design and systematic outcome ascertainment distinguish the RIETE registry from administrative databases and provide the granular clinical detail needed for prognostic model development.

Derivation and Validation Cohorts

The RIETE Score was derived from 13,057 patients enrolled in the registry and internally validated in a subsequent temporal cohort of 6,572 patients from the same registry. The derivation cohort included patients with both DVT and PE, reflecting the full spectrum of acute VTE. The large sample size provided robust statistical power for multivariable logistic regression modeling, allowing the identification of six independent predictors of major bleeding from a broad set of candidate variables.

Outcome Definition: Major Bleeding

A precise and consistent definition of the outcome being predicted is essential for the clinical utility of any prognostic score. The RIETE registry defines major bleeding as overt bleeding accompanied by at least one of the following criteria:

Requirement for transfusion of 2 or more units of packed red blood cells
Retroperitoneal hemorrhage
Spinal hemorrhage
Intracranial hemorrhage
Fatal bleeding

This definition is broadly consistent with the International Society on Thrombosis and Haemostasis (ISTH) definition of major bleeding, which also includes bleeding into a critical anatomic site, bleeding causing a hemoglobin drop of 20 g/L (2 g/dL) or more, and bleeding leading to death. The RIETE definition is pragmatic and captures clinically consequential bleeding events that alter patient management. The three-month timeframe aligns with the initial treatment phase of VTE, during which anticoagulant intensity is typically highest and the risk of early bleeding is greatest.

The Six Scoring Variables

Variable 1: Recent Major Bleeding (2 Points)

A major bleeding event occurring before or at the time of VTE diagnosis carries the highest point weight in the RIETE Score. This variable captures patients who present with VTE in the context of a recent hemorrhagic complication, placing them in a uniquely challenging clinical situation where the indication for anticoagulation (acute VTE) directly conflicts with the contraindication (active or recent bleeding).

The pathophysiology underlying this high risk is multifactorial. Patients with recent major bleeding may have an underlying hemostatic defect (coagulopathy, thrombocytopenia, vascular malformation, peptic ulcer disease, or other structural lesion) that predisposes them to recurrent hemorrhage. The bleeding source may not be fully resolved. Additionally, anticoagulation impairs the normal hemostatic mechanisms that would otherwise limit bleeding from any existing or developing vascular injury. In the RIETE derivation cohort, recent major bleeding approximately doubled the odds of subsequent major hemorrhage during anticoagulant therapy, justifying its 2-point weight.

Clinical management of patients with recent major bleeding and concurrent acute VTE requires careful multidisciplinary deliberation. Options include delaying anticoagulation until the bleeding source is controlled, using reduced-intensity anticoagulation regimens, placement of an inferior vena cava (IVC) filter as a temporary measure (recognizing that IVC filters do not treat the underlying VTE and carry their own complications), or catheter-directed therapies in selected cases of massive PE. The RIETE Score does not prescribe a specific management approach but rather quantifies the risk to inform these complex decisions.

Variable 2: Anemia (1.5 Points)

Baseline anemia at the time of VTE diagnosis, defined using World Health Organization sex-specific thresholds (hemoglobin below 13 g/dL for men or below 12 g/dL for women), is assigned 1.5 points. Anemia is a marker of reduced physiological reserve for tolerating hemorrhage and may also indicate an underlying condition that predisposes to bleeding.

The causes of anemia in VTE patients are varied. Iron deficiency anemia may reflect chronic gastrointestinal blood loss from peptic ulcer disease, malignancy, arteriovenous malformations, or nonsteroidal anti-inflammatory drug use, all of which represent potential bleeding sources that can be unmasked or worsened by anticoagulation. Anemia of chronic disease may accompany cancer, chronic kidney disease, or chronic inflammatory conditions. Myelosuppressive chemotherapy can cause anemia and concurrent thrombocytopenia, compounding the bleeding risk in cancer-associated VTE. Regardless of the etiology, anemic patients have a lower baseline hemoglobin, meaning that even a modest hemorrhage can produce a clinically significant drop in oxygen-carrying capacity and hemodynamic compromise.

From a practical standpoint, the identification of anemia at VTE diagnosis should prompt two actions: first, investigation of the underlying cause (including gastrointestinal evaluation if iron deficiency is suspected, as this may reveal a treatable bleeding source), and second, closer monitoring during anticoagulation, with a lower threshold for investigating symptoms that may indicate bleeding.

Variable 3: Creatinine Above 1.2 mg/dL (1.5 Points)

Elevated serum creatinine above 1.2 mg/dL (106.1 micromol/L) at baseline receives 1.5 points, reflecting the impact of impaired renal function on bleeding risk during anticoagulation. The relationship between renal insufficiency and anticoagulant-related bleeding is well established and operates through multiple mechanisms.

Pharmacokinetically, many anticoagulants depend on renal clearance. Low-molecular-weight heparins (LMWHs) are predominantly renally excreted; in patients with creatinine clearance below 30 mL/min, standard LMWH doses produce higher peak and trough anti-Xa levels, prolonging anticoagulant exposure and increasing bleeding risk. Among direct oral anticoagulants (DOACs), dabigatran is approximately 80% renally cleared, rivaroxaban approximately 33%, apixaban approximately 27%, and edoxaban approximately 50%. Dose adjustments are required for most DOACs in moderate-to-severe renal impairment. Unfractionated heparin (UFH) is cleared predominantly by the reticuloendothelial system and is less affected by renal function, though its dosing can be unpredictable in critically ill patients with fluctuating renal function.

Pathophysiologically, uremia impairs platelet adhesion and aggregation (uremic platelet dysfunction), reduces von Willebrand factor activity, and disrupts the interaction between platelets and the subendothelium. These defects in primary hemostasis compound the effect of therapeutic anticoagulation, which impairs secondary hemostasis (the coagulation cascade). The net result is a "double hit" to the hemostatic system that disproportionately increases bleeding risk compared to patients with normal renal function.

The creatinine threshold of 1.2 mg/dL was selected based on its statistical association with major bleeding in the RIETE derivation cohort. It is a simplified marker of renal function that does not account for age, sex, or body mass. In clinical practice, estimated glomerular filtration rate (eGFR) calculated using the CKD-EPI equation provides a more accurate assessment of renal function and should be used to guide anticoagulant dosing decisions, even though the RIETE Score itself uses the simpler creatinine cutpoint.

Variable 4: Malignancy (1 Point)

Active or known history of cancer receives 1 point. Cancer-associated VTE is a major clinical challenge because these patients face simultaneously elevated risks of both recurrent thromboembolism and major bleeding during anticoagulation. The elevated bleeding risk in cancer patients during anticoagulation is driven by multiple factors: tumor invasion of vascular structures, tumor neovascularization with fragile, easily disrupted blood vessels, chemotherapy-induced thrombocytopenia and mucosal damage, radiation-induced tissue fragility, surgical wounds, hepatic synthetic dysfunction from liver metastases, and the frequent co-occurrence of other RIETE Score risk factors (anemia, renal impairment, advanced age).

The management of cancer-associated VTE has been the subject of dedicated guidelines from the International Society on Thrombosis and Haemostasis (ISTH), the American Society of Clinical Oncology (ASCO), the American Society of Hematology (ASH), and the National Comprehensive Cancer Network (NCCN). Historically, LMWH was the preferred agent for cancer-associated VTE based on the CLOT trial (dalteparin vs. warfarin). More recently, the SELECT-D trial (rivaroxaban), the Hokusai VTE Cancer trial (edoxaban), the CARAVAGGIO trial (apixaban), and the ADAM VTE trial (apixaban) have established DOACs as acceptable alternatives for many cancer patients, though with caveats regarding increased gastrointestinal bleeding with rivaroxaban and edoxaban in patients with gastrointestinal malignancies.

The RIETE Score identifies cancer as a bleeding risk factor with a 1-point weight. While this is the lowest individual weight among the six variables, the clinical significance is amplified by the fact that cancer patients frequently accumulate additional points from anemia, renal impairment, and advanced age, producing aggregate scores in the intermediate-to-high risk range. The RIETE Score can help frame the bleeding risk discussion for cancer patients, but treatment decisions should also incorporate tumor-specific and treatment-specific factors not captured by the score.

Variable 5: Clinically Overt Pulmonary Embolism (1 Point)

Clinically overt PE, defined as PE presenting with clinical symptoms (dyspnea, chest pain, hemoptysis, tachycardia, or hemodynamic compromise), as distinguished from isolated DVT or incidentally discovered PE, receives 1 point. This variable captures the observation that symptomatic PE is associated with a higher risk of major bleeding during anticoagulation compared to isolated DVT.

The mechanistic basis for this association is multifactorial. Patients with symptomatic PE often have a larger clot burden, which leads to more aggressive initial anticoagulation (higher initial heparin doses, more frequent dose adjustments, longer parenteral anticoagulation before transition to oral therapy). The hemodynamic compromise associated with PE may reduce hepatic and renal blood flow, potentially altering anticoagulant metabolism and clearance. Patients with PE are more likely to require intensive care, invasive procedures (central venous catheterization, mechanical ventilation), and concurrent medications (vasopressors, inotropes) that can independently increase bleeding risk. Additionally, the initial management of massive or submassive PE may include systemic thrombolysis, catheter-directed therapy, or surgical embolectomy, all of which carry significant hemorrhagic risk beyond that of standard anticoagulation alone.

The distinction between clinically overt PE and incidental PE is clinically important. Incidental PE, discovered on imaging performed for other indications (typically staging CT scans in cancer patients), is increasingly recognized as a common finding. While most guidelines recommend anticoagulation for incidental PE, the bleeding risk profile may differ from that of symptomatic PE, and the RIETE Score appropriately assigns the additional point only for clinically overt presentations.

Variable 6: Age Above 75 Years (1 Point)

Age greater than 75 years at the time of VTE diagnosis receives 1 point. Advanced age is one of the most consistently identified risk factors for bleeding during anticoagulant therapy across virtually all bleeding risk prediction models and clinical studies.

The age-related increase in bleeding risk during anticoagulation reflects a convergence of physiologic, pharmacologic, and clinical factors. Physiologically, aging is associated with increased vascular fragility, reduced hepatic synthetic function, decreased renal clearance, and changes in the balance of pro-coagulant and anti-coagulant pathways. Pharmacologically, elderly patients are more sensitive to anticoagulant effects at standard doses due to altered body composition (reduced lean mass, increased body fat, altered volume of distribution), decreased hepatic metabolism, and reduced renal clearance. Clinically, elderly patients are more likely to have comorbidities that independently increase bleeding risk (hypertension, cerebrovascular disease, falls risk, peptic ulcer disease, hepatic dysfunction) and to be taking concomitant medications that potentiate bleeding (antiplatelet agents, nonsteroidal anti-inflammatory drugs, selective serotonin reuptake inhibitors).

For elderly patients with VTE, the bleeding risk must be carefully weighed against the thrombotic risk, which is also elevated in this population. The decision to anticoagulate is almost always affirmative for acute VTE (the thrombotic risk of untreated PE far exceeds the bleeding risk of anticoagulation in most scenarios), but the choice of agent, dose, duration, and monitoring intensity should be adjusted for the increased hemorrhagic risk. DOACs, which have more predictable pharmacokinetics and lower rates of intracranial hemorrhage compared to vitamin K antagonists, are generally preferred in elderly patients without severe renal impairment, though dose reductions are often required.

Score Calculation and Risk Stratification

The RIETE Score is calculated by summing the points for all six variables. The total score ranges from 0 to a maximum of 8 (if all six variables are present). The score includes half-point increments because anemia and creatinine elevation each contribute 1.5 points.

Score	Risk Category	Major Bleeding Rate (3 months)
0	Low	0.1%
1	Intermediate	1.4%
1.5 - 2	Intermediate	2.2%
2.5 - 3	Intermediate	4.4%
3.5 - 4	Intermediate	4.2%
4.5 - 5	High	4.9%
5.5 - 6	High	11%
> 6	High	20%

The three-tier risk stratification groups scores into low (score 0), intermediate (score 1 to 4), and high (score above 4). This framework reflects a clinically meaningful gradient: patients with a score of 0 have a negligible 3-month major bleeding rate (0.1%), while those with scores above 6 face a 20% probability of major hemorrhage, representing a 200-fold difference in absolute risk.

Clinical Application by Risk Category

Low Risk (Score 0)

A RIETE Score of 0 identifies patients with the lowest predicted risk of major bleeding (0.1%). These patients have none of the six risk factors: no recent major bleeding, no anemia, normal renal function, no malignancy, no symptomatic PE (DVT only or incidental PE), and age 75 or younger. Standard anticoagulation for the recommended duration is appropriate, with routine monitoring of hemoglobin, renal function, and clinical signs of bleeding per institutional protocol. The low bleeding risk may favor the use of standard-dose anticoagulation without dose reduction and supports confidence in the safety of the treatment regimen. Patient education should still include warning signs of bleeding, instructions for when to seek medical attention, and avoidance of activities with high injury risk.

Intermediate Risk (Score 1 to 4)

The majority of VTE patients fall into the intermediate-risk category, with 3-month major bleeding rates ranging from 1.4% to 4.4%. Anticoagulation remains clearly indicated for acute VTE in this group, but the elevated bleeding risk warrants several adjustments to the standard management approach.

Closer surveillance is appropriate, including more frequent monitoring of hemoglobin (e.g., at 1 week, 2 weeks, and monthly thereafter during the first 3 months), periodic renal function assessment (particularly in patients with baseline creatinine elevation or those receiving renally cleared anticoagulants), and clinical assessment for signs of occult bleeding (fatigue, dizziness, dark stool, hematuria). Modifiable risk factors should be optimized: gastrointestinal protection (proton pump inhibitor therapy for patients at risk of upper GI bleeding), discontinuation or minimization of concurrent antiplatelet agents and NSAIDs when clinically feasible, blood pressure optimization, and correction of coagulopathy if present.

Anticoagulant selection should favor agents with the most favorable safety profile for the individual patient. DOACs are generally preferred over vitamin K antagonists in eligible patients due to lower rates of major bleeding, particularly intracranial hemorrhage. Among DOACs, apixaban has demonstrated the lowest rates of major bleeding in the pivotal VTE treatment trials (AMPLIFY trial: 0.6% vs. 1.8% for conventional therapy), though head-to-head comparisons between DOACs for VTE are lacking.

High Risk (Score Above 4)

Patients with RIETE Scores above 4 face 3-month major bleeding rates of 4.9% to 20%, representing a clinically concerning hemorrhagic risk. The management of these patients requires careful weighing of the bleeding risk against the thrombotic risk, ideally in a multidisciplinary setting involving hematology, pulmonary medicine, and, when applicable, oncology and interventional radiology.

In most cases, anticoagulation remains necessary because the risk of untreated PE (particularly massive or submassive PE) is life-threatening. However, several strategies can mitigate the bleeding risk. Shorter anticoagulation duration (3 months rather than extended therapy) may be appropriate if the VTE was provoked by a transient risk factor. Dose-adjusted regimens, such as reduced-intensity DOAC dosing after the initial treatment phase, have been evaluated in clinical trials (e.g., the EINSTEIN-CHOICE trial evaluated rivaroxaban 10 mg for extended VTE prevention) and may provide a favorable balance of efficacy and safety. IVC filter placement may be considered as a temporary measure in patients with acute PE who cannot receive anticoagulation due to active bleeding, though filters should be retrieved once anticoagulation can be safely resumed. Multidisciplinary consultation can help identify individualized strategies that balance thrombotic and hemorrhagic risks.

For patients at the highest end of the scale (score above 6, with an estimated 20% major bleeding rate), the bleeding risk may approach or exceed the thrombotic risk in certain clinical scenarios. These cases demand the most careful deliberation, with explicit documentation of the risk-benefit analysis and shared decision-making with the patient and family. In some cases, observation without anticoagulation (for subsegmental or incidental PE with low clot burden) or abbreviated anticoagulation courses may be appropriate, though these decisions should be made on a case-by-case basis.

Anticoagulant Selection in the Context of Bleeding Risk

Unfractionated Heparin (UFH)

UFH has the advantage of a short half-life (approximately 60 to 90 minutes), rapid reversibility with protamine sulfate, and renal-independent clearance. These properties make it the preferred initial anticoagulant for patients at high bleeding risk, those requiring potential procedural intervention, and those with severe renal impairment (creatinine clearance below 30 mL/min). The disadvantage is the requirement for continuous intravenous infusion and frequent aPTT monitoring, necessitating inpatient administration.

Low-Molecular-Weight Heparins (LMWHs)

LMWHs (enoxaparin, dalteparin, tinzaparin) offer more predictable pharmacokinetics, subcutaneous administration, and generally do not require routine monitoring in patients with normal renal function. However, they are predominantly renally cleared and accumulate in patients with creatinine clearance below 30 mL/min, increasing bleeding risk. Anti-Xa level monitoring is recommended in patients with significant renal impairment, extreme body weight, or pregnancy. Protamine provides only partial reversal of LMWH anticoagulant effect (approximately 60%), which is a disadvantage in the setting of major bleeding.

Vitamin K Antagonists (VKAs)

Warfarin, the most widely used VKA, has been the standard oral anticoagulant for VTE treatment for decades. Its advantages include extensive clinical experience, low cost, and complete reversibility with vitamin K and prothrombin complex concentrate (PCC) or fresh frozen plasma (FFP). Its disadvantages are numerous: narrow therapeutic index, extensive drug and food interactions, requirement for regular INR monitoring, slow onset and offset of action, and a relatively high rate of intracranial hemorrhage compared to DOACs. For patients at elevated bleeding risk by the RIETE Score, VKAs are generally less preferred than DOACs unless specific circumstances (mechanical heart valve, antiphospholipid syndrome, severe renal impairment, or drug interactions precluding DOAC use) necessitate their use.

Direct Oral Anticoagulants (DOACs)

DOACs (rivaroxaban, apixaban, edoxaban, dabigatran) have transformed VTE management over the past decade. Compared to conventional therapy (LMWH/warfarin), DOACs as a class have demonstrated non-inferior efficacy for VTE recurrence prevention and superior or comparable safety for major bleeding. Key safety advantages include significantly lower rates of intracranial hemorrhage and fatal bleeding. Among the DOACs evaluated in the pivotal VTE treatment trials, apixaban (AMPLIFY trial) demonstrated the most favorable bleeding profile, with a major bleeding rate of 0.6% compared to 1.8% for conventional therapy, and edoxaban (Hokusai VTE trial) showed a similarly favorable safety signal, particularly when dose-reduced in patients with renal impairment or low body weight.

For patients with elevated RIETE Scores, the DOAC safety advantage is particularly relevant. The lower intracranial hemorrhage rate is critical for elderly patients (who are at highest risk of this devastating complication), and the more predictable pharmacokinetics reduce the risk of supratherapeutic drug levels. Specific reversal agents are available for dabigatran (idarucizumab) and for the factor Xa inhibitors (andexanet alfa), providing a safety net in the event of life-threatening bleeding, though availability varies by institution.

Comparison with Other Bleeding Risk Scores

Feature	RIETE Score	HAS-BLED	VTE-BLEED	ACCP Criteria
Target population	Acute VTE (DVT + PE)	Atrial fibrillation	VTE on anticoagulation	VTE
Timeframe	First 3 months	1 year	6+ months (extended therapy)	Not specified
Number of variables	6	9	6	Multiple (not scored)
Derivation cohort	13,057 VTE patients	AF cohorts	8,240 VTE patients	Expert consensus
Validated for VTE	Yes (internal)	Limited in VTE	Yes (internal)	N/A
Includes renal function	Yes (creatinine > 1.2)	Yes	Yes	Yes
Includes malignancy	Yes	No	Yes	Yes
Score range	0 - 8	0 - 9	0 - 8.5	Not scored

The RIETE Score is purpose-built for the acute VTE population and specifically addresses the initial 3-month treatment period, when bleeding risk is highest. The HAS-BLED score, widely used in atrial fibrillation, was not derived from VTE patients and has shown inconsistent performance when applied to this population. The VTE-BLEED score, developed by Klok and colleagues, is complementary to the RIETE Score but focuses on bleeding risk during extended (beyond 6 months) anticoagulation rather than the initial treatment phase. The ACCP (American College of Chest Physicians) bleeding risk criteria provide a qualitative framework (listing risk factors for bleeding) rather than a quantitative score. The RIETE Score fills a specific niche: quantitative, scored, and validated for the first 3 months of VTE treatment.

Special Populations

Cancer-Associated VTE

Patients with cancer-associated VTE represent a particularly challenging subgroup in which the RIETE Score is frequently applied. These patients accumulate points from malignancy (1 point) and often from anemia (1.5 points due to myelosuppressive chemotherapy or chronic disease), elevated creatinine (1.5 points from nephrotoxic chemotherapy or tumor-related renal impairment), and advanced age (1 point). A cancer patient with anemia and renal impairment already has a RIETE Score of 4, placing them at the upper end of the intermediate-risk category.

Dedicated cancer-associated VTE guidelines from ISTH, ASCO, and NCCN provide specific anticoagulant recommendations that should be integrated with the RIETE bleeding risk assessment. Low-molecular-weight heparin (particularly dalteparin) was historically the standard based on the CLOT trial. More recently, DOACs (apixaban and rivaroxaban) have been shown to be non-inferior to LMWH for recurrent VTE prevention in cancer patients, but with an increased risk of major bleeding, particularly gastrointestinal bleeding in patients with gastrointestinal or genitourinary malignancies. The RIETE Score helps quantify this bleeding risk and may support the decision between LMWH (generally safer from a GI bleeding perspective) and DOACs (more convenient, better tolerated) in individual cancer patients.

Elderly Patients

Patients over 75 years receive 1 point for age alone and frequently accumulate additional points from anemia (common in the elderly due to chronic disease, nutritional deficiency, or myelodysplasia), elevated creatinine (age-related decline in renal function), and malignancy (cancer incidence increases with age). An 80-year-old patient with moderate renal impairment and anemia has a RIETE Score of 4 before considering other factors. Anticoagulant selection and dosing in elderly patients should account for the RIETE-estimated bleeding risk, with a preference for DOACs with favorable bleeding profiles (apixaban is often preferred), appropriate dose adjustments, and enhanced monitoring. Fall risk assessment is important, as falls in anticoagulated elderly patients can result in traumatic intracranial hemorrhage.

Patients with Renal Impairment

Renal impairment contributes 1.5 points to the RIETE Score and has additional implications for anticoagulant pharmacokinetics. For patients with creatinine clearance below 30 mL/min, LMWH requires dose reduction or monitoring with anti-Xa levels, dabigatran is contraindicated or used at reduced dose depending on the jurisdiction, and rivaroxaban and edoxaban require dose adjustment. Apixaban is less dependent on renal clearance (approximately 27%) and has been studied in patients with end-stage renal disease on hemodialysis, though data in this population remain limited. UFH is the safest initial parenteral option in severe renal impairment. The RIETE Score's identification of creatinine elevation as a risk factor should prompt careful anticoagulant selection and dose optimization based on measured or estimated renal function.

Applicability in the DOAC Era

The RIETE Score was derived primarily from patients treated with vitamin K antagonists and low-molecular-weight heparins, reflecting the anticoagulant landscape of the early to mid-2000s. The widespread adoption of DOACs since 2010 raises the question of whether the score's absolute bleeding rates and relative risk stratification remain valid for DOAC-treated patients.

Several considerations are relevant. First, the risk factors identified by the RIETE Score (recent major bleeding, anemia, renal impairment, malignancy, symptomatic PE, advanced age) are biologically plausible predictors of bleeding regardless of the anticoagulant used. They reflect underlying patient characteristics (hemostatic fragility, reduced drug clearance, structural bleeding sources) that increase hemorrhagic risk with any anticoagulant. Second, the relative risk stratification (low vs. intermediate vs. high) is likely preserved even if the absolute bleeding rates are lower with DOACs. A patient with a RIETE Score of 6 will still be at substantially higher bleeding risk than one with a score of 0, even if both are treated with apixaban rather than warfarin. Third, the absolute bleeding rates reported by the original RIETE study may overestimate the risk for DOAC-treated patients, particularly for intracranial hemorrhage and fatal bleeding, which are consistently lower with DOACs across all major VTE treatment trials.

Updated validation of the RIETE Score in DOAC-treated populations would strengthen its clinical utility in contemporary practice. In the interim, clinicians can apply the score as a relative risk stratification tool, recognizing that the absolute bleeding rates may be somewhat lower than those from the original VKA/LMWH era.

Integration into Clinical Decision-Making

The RIETE Score is most effectively used as one component of a comprehensive bleeding risk assessment, not as the sole determinant of anticoagulant management. Practical integration involves several steps:

At the Time of VTE Diagnosis

Calculate the RIETE Score using readily available clinical and laboratory data (age, hemoglobin, creatinine, cancer status, bleeding history, VTE presentation type).
Document the score and risk category in the medical record.
Communicate the estimated bleeding risk to the patient as part of the anticoagulation consent discussion.

Anticoagulant Selection

For low-risk patients (score 0): Standard anticoagulation with any guideline-recommended agent is appropriate.
For intermediate-risk patients (score 1-4): Favor agents with lower bleeding profiles (apixaban is often preferred). Optimize modifiable risk factors. Plan closer monitoring.
For high-risk patients (score above 4): Select the anticoagulant with the most favorable safety profile. Consider dose-adjusted regimens after the acute phase. Plan intensive monitoring. Involve multidisciplinary consultation for complex cases.

Monitoring and Follow-up

Low-risk patients: Standard monitoring per institutional protocol.
Intermediate-risk patients: Hemoglobin check at 1 to 2 weeks and monthly. Renal function reassessment at 1 month and quarterly. Clinical assessment for bleeding at each visit.
High-risk patients: Hemoglobin monitoring weekly during the first month, then biweekly. Frequent renal function checks. Low threshold for imaging or endoscopy if occult bleeding is suspected. Consider stool occult blood testing.

Duration of Anticoagulation Decisions

The RIETE Score specifically addresses the first 3 months of anticoagulation. When considering extended or indefinite anticoagulation (for unprovoked VTE or cancer-associated VTE), the VTE-BLEED score may provide more relevant bleeding risk assessment for the extended treatment phase. The RIETE Score remains useful at the 3-month reassessment point: a high RIETE Score that contributed to a decision for close monitoring during the initial phase should also factor into the extended-duration decision, particularly if the underlying risk factors (anemia, renal impairment, malignancy) persist.

Limitations and Considerations

Internal Validation Only

The RIETE Score was validated within the same registry from which it was derived (temporal internal validation). While the large sample size and prospective design strengthen this validation, independent external validation in different healthcare systems and populations has been limited. Subsequent analyses within the RIETE registry have confirmed the score's utility, and comparative studies (e.g., Klok et al., comparing RIETE and VTE-BLEED scores) have provided additional validation, but prospective external validation in North American and Asian populations would strengthen the evidence base.

Registry-Based Derivation

As an observational registry, the RIETE database reflects real-world practice patterns that may vary across participating centers and over time. Treatment decisions (anticoagulant choice, dose, duration) were at the discretion of individual clinicians, introducing potential confounding. Centers participating in a registry may not be representative of all healthcare settings, potentially introducing selection bias.

Incomplete Coverage of Bleeding Risk Factors

The RIETE Score does not include several recognized bleeding risk factors: concomitant antiplatelet therapy (aspirin, clopidogrel), liver disease, alcohol use, prior hemorrhagic stroke, thrombocytopenia, uncontrolled hypertension, and specific medications that potentiate bleeding (NSAIDs, SSRIs). Clinicians should consider these factors alongside the RIETE Score when assessing overall bleeding risk. The score provides a structured starting point, not an exhaustive assessment.

Three-Month Timeframe

The RIETE Score predicts bleeding risk during the initial 3 months of anticoagulation only. It does not address the bleeding risk during extended or indefinite anticoagulation, which may be relevant for patients with unprovoked VTE, recurrent VTE, or ongoing cancer-associated VTE requiring long-term treatment. For extended-duration bleeding risk assessment, the VTE-BLEED score is a more appropriate tool.

DOAC Era Applicability

As discussed, the absolute bleeding rates derived from the VKA/LMWH era may overestimate risk for patients now treated with DOACs. Updated absolute risk estimates for DOAC-treated populations would enhance the clinical utility of the score in contemporary practice. The relative risk stratification (low vs. intermediate vs. high) is expected to remain valid.