VTE - Assessing risk of recurrence
A knotty topic that I have been avoiding for some time. Does it really belong on a revision wesbite? On balance I think so as I could imagine examiners expecting candidates to be able to discuss some of the competing factors listed below even if any particular case may not have a ‘right’ answer.
A patient 3 months post vte comes to clinic
Clinic Agenda: Are you going to recommend to this patient that they stop or continue their anticoagulation?
Some Questions:
VTE - location? extent? First event or a recurrence?
Cause - provoking factors? If so, minor or major? transient or persistent?
Risk of recurrence - how big a risk? consequences if it does occur?
Risk of bleeding - greater or lesser than risk of recurrence? consequences if it does occur?
Patient preferences - attitudes to risk? atttiudes to medication? employment / contact sports?
Other - male patients? D-dimer results? Use of scoring tools?
quick aside: why minimum 3 months ac? (BMJ 2011)
Summary:
Recurrence was higher if AC stopped at 1-1.5 months vs 3 months, but similar if stopped at 3 months vs 6 months
Meta-analysis of 7 trials, BMJ 2011
3000 patients with 1st VTE who did not have cancer, followed for 24 months after stopping AC
Recurrence was lower in distal vs proximal DVT, similar for proximal DVT vs PE
Recurrence was lower if temporary provoking factors vs unprovoked
Recurrence was higher if AC stopped at 1-1.5 months vs 3 months, but similar if stopped at 3 months vs 6 months
Note: useful graphs in included in results of this paper
Site of 1st vte (JTH 2010)
Summary:
The risk of recurrence as PE was 3.1 fold greater in patients with a first PE compared to a first proximal DVT
First proximal DVT 4.8 fold higher cumulative recurrence rate than distal DVT
Site of 1st VTE Predicting Site of Recurrence, JTH 2010
Meta-analysis, >2500 patients with a first symptomatic VTE who were followed after AC was stopped.
5-year cumulative rate of PE or DVT recurrence = 22.6%
If PE first, 5-year cumulative rate of recurrence as PE or DVT 22%, and for PE 10.6%
If proximal DVT first, 5-year cumulative rate of recurrence as PE or DVT 26.4%, and for PE 3.6%
If distal DVT first, 5-year cumulative rate of recurrence as PE or DVT 7.6%, and for PE 1.2%
provoking factors (ISTH guidance 2016)
Summary:
The presence and nature of provoking factors is the single variable most influencing risk of recurrence
Risk of recurrence from highest to lowest: persistent provoking factors > unprovoked VTE > transient provoked VTE
Transient Provoking Factors
Major: General anaesthetic >30 minutes, Hospitalisation 3+ days, Cesarean section
Minor: General anaesthetic <30 minutes, Hospitalisation <3 days, Oestrogen therapy, Pregnancy, Confined to bed for 3+ days at home due to acute illness, leg injury with reduced mobility for 3+ days
Unprovoked VTE
Neither transient nor persistent provoking factors identified.
Persistent Provoking Factors
Active cancer, ongoing non-malignant conditions associated with increased VTE risk (e.g. inflammatory bowel disease)
Effect of Provoking factors on recurrence risk
Major Transient: Half the risk of VTE recurrence after stopping anticoagulation compared to an unprovoked VTE, when the risk factor occurred up to 3 months prior to VTE diagnosis (alternative: >10-fold increased risk of recurrence compared to the risk of having a first VTE)
Minor Transient: Half the risk of VTE recurrence after stopping anticoagulation compared to an unprovoked VTE, when the risk factor occurred up to 2 months prior to VTE diagnosis (alternative: 3-10-fold increased risk of recurrence compared to the risk of having a first VTE)
estimating Risk of recurrence (ESC Guideline 2019, BMJ 2019)
Summary:
For unprovoked or minimally provoked first VTE treated with 3 months AC then stop, the risk of fatal recurrent PE is 1.5% at 10 years.
The cumulative risk of recurrence after stopping AC at 3 months is 10% at 1 year, rising to 36% at 10 years.
The European Society of Cardiology suggests 3 risk categories (see page 575):
Low risk (<3% per year recurrence risk): major transient provoking factor (see ISTH definition above)
Intermediate (3-8% per year): minor transient provoking factor, non-malig persistant provoking factor, unprovoked cases
High risk (>8% per year): Active cancer, Antiphospholipid Syn, 1+ prev VTE in absence of major transient factors
Long-term risk of recurrent VTE in patients receiving extended AC for 1st unprovoked VTE 2021
Meta-analysis. 15,000 patients
5-yr cumulative risk of recurrent VTE = 7%
5-yr cumulative risk of fatal PE = 1.2%
European Society of Cardiology Guidelines on management of PE 2019
Pages 575 onwards discuss risk of recurrence and risk of bleeding, and include very helpful traffic light chart on page 575, summarised above
Long Term Risks of Recurrent VTE, BMJ 2019
Meta-analysis, >7500 patients with Unprovoked or Minimally Provoked first VTE who received a minimum of three months treatment before stopping anticoagulation.
Cumulative incidence of recurrent VTE (time after discontinuing AC):
10% at 1 year
16% at 2 years
25% at 5 years
36% at 10 years
Cumulative incidence of fatal recurrent PE:
0.7% at 2 years
1% at 5 years
1.5% at 10 years
Pooled fatality rate if recurrent VTE occurs = 3.8%
Sex: Men 1.4x rate of recurrence compared to women
Site: Distal DVT less likely to recur than other sites
Incidence of recurrent VTE in relation to clinical and thrombophilic risk factors, Lancet 2003
570 unselected patients with first VTE, prospectively followed up (exc. active cancer and APS)
11% recurrence at 2 years
0% recurrence after surgery-related VTE, 19% after unprovoked VTE
Testing for heritable thrombophilia does not predict recurrence
estimating risk of bleeding
Summary:
Annual risk of major bleeding on AC = 1.2%. Converts to a risk of 1.3% at 10 years, compared to the fatal recurrent VTE of 1.5%
Other analyses have suggested a 3% annual incidence of major bleeding in patients on VKAs
40% reduction in risk of major bleeding with DOACs compared to VKAs
Long-term bleeding risk in patients receiving extended AC for a 1st unprovoked VTE 2021
Meta-analysis. 17,000 patients on VKA or DOAC
5-yr cumulative incidence of major bleeding on VKA = 6% (insufficient data for DOACs)
Bleeding statistically higher with: Age >65, CrCl <50, concomittant antiplatelets, history of bleeding & Hb <100
Long-term bleeding risk in patients who have stopped AC after an unprovoked VTE 2021
Meta-analysis. 8000 patients
5-yr cumulative incidence of major bleeding after stopping AC = 1%
European Society of Cardiology Guidelines on management of PE 2019
Risk of major bleeding highest in first month of AC treatment
3% annual incidence of major bleeding in patients on VKAs
40% reduction in risk of major bleeding with DOACs comapred to VKAs
Risk factors include: Age >75, Prev. bleeding, Active cancer, Prev. stroke, CKD, liver disease, concurrent antiplatelets, other serious chronic illness
Consider use of bleeding scores (e.g. HAS-BLED)
Clinical impact of bleeding in patients taking AC for VTE, meta-analysis, Ann Intern Med 2003
4300 patient-years of anticoagulation (VKA - this is 2003)
Case-fatality rate due to major bleeding for patients receiving >3 months AC = 9.1%
Intracranial bleeding rate 0.65 per 100 patient-years
Efficacy and Safety of NOACs in treatment of VTE, meta-analysis, ESVS 2014
38,000 patients
DOACs equally effective as VKAs in preventing recurrent VTE
Major bleeding occurred less often in DOAC vs VKA (1% vs 1.7%, and fatal bleeding 0.09% vs 0.18%)
predictive scoring systems/models
Many available - DASH, DAMOVES, HERDOO2, Vienna
All have limitations, excluding different combinations of patient groups
Attempts to prospectively validate these systems have failed so far - e.g. VISTA 2020 found no difference in rate of recurrence with use or non-use of the Vienna model.
Additional factors included in these models that are not discussed above:
Male sex is assciated with a higher risk of recurrence (JTH 2004)
A raised D-dimer whilst off anticoagulation is associated with increased risk of recurrence (Circulation 2003) (NB performing this test potentially exposes patient to risk in time spent off AC)
conclusions
Practice will vary, in line with the ESC 2019 Guideline, could use the following to discuss cases:
Low risk (<3% per year recurrence risk): Recommend stop after 3 months
Intermediate (3-8% per year): Individualised shared decision making with patient
High risk (>8% per year): Recommend long-term anticoagulation
Predictive scoring systems mostly fallen out of favour, not been performed well in prospective validation studies
Final Thought
Still not sure what you would do with the next intermediate risk patient to enter your clinic? Don’t worry, you’re not alone - just this year an international survey of >300 clinicians found a lack of consensus on the management of these patients (JTH 2021).