Transcript Document

Clinical predictors of adverse
outcome in VTE outpatients –
the VERITY PUSH (Prospective
Follow-Up Survey in Verity
Hospitals) study
Peter Rose, Aidan McManus, Shankaranarayana
Paneesha, Nicholas Scriven, Timothy Farren, Sue
Bacon, Roopen Arya, Olatunde Falode, Denise
O'Shaughnessy & Tim Nokes for the VERITY
Investigators
Background
• Outpatient VTE treatment with LMWH is now
commonplace in the UK
• Factors predictive of VTE recurrence have
been reported including age, cancer,
immobility and thrombophilic mutations, but
there are few data describing risk factors
associated with recurrence, or adverse
outcome in unselected VTE patients treated
in outpatient clinics
PUSH study objectives
• To determine the frequency of major
adverse outcomes (death, recurrence of
VTE, and bleeding) in patients diagnosed
with VTE and treated as outpatients with
low molecular weight heparin
• To establish risk factors for adverse
outcome after VTE
PUSH study design
Patients with suspected VTE
Exclusion algorithms and
diagnostic tests
Timeline
Enrolment
Day 0
Patients negative for VTE
Follow-up
Day 0
Consecutive patients with
confirmed VTE
Day 180
Recurrence
Bleeds
Death
VERITY database
VERITY PUSH database
Day 360
Logistic regression
Features of VERITY
• National registry – outpatient VTE treatment
• Full spectrum of VTE – DVT and PE
• Records information on patients presenting
with suspected and confirmed VTE
• Expanded data on demographics,
presentation, management & outcomes
• Extensive risk factor data
VERITY and PUSH centres
PUSH centres
Enrolment
(Nov 2008 – Apr 2009)
Seven hospitals enrolled 843 consecutive patients
250
214
200
150
131
100
50
0
69
27
74
67
40
Study population
• 221 patients were excluded
– 75 = no follow-up entry
– 50 = no record if treated as an outpatient
– 96 = not treated as outpatient
• Final study population n=622
• Patients were followed for up to 388 days
(mean duration of 195 days)
Baseline characteristics
(risk factors)
90
76.4
80
Patients (%)
70
60
50
36.8
40
30.4
30
20
10
0
9.8 6.8 10.6
15.3
13.7
2.4
3.2 1.3 0.3 1.9
2
RESULTS
Major adverse outcomes
Major bleed
1.2%
(n=34)
(n=16)
(n=36)
Home treatment with LMWH
Levine et al. N Engl J Med. 1996;334:677-81; Koopman et al. N Engl J Med. 1996;334:682-7.
RISK MODELING
Univariate and multivariate logistic regression
analyses were conducted to determine if any of the
known risk factors predicted for recurrence or
adverse outcome.
1. Major surgery (last 4 weeks)
(AND type of major surgery: general/orthopaedic/other)
2. Hormonal risk factor (yes or no)
3. Family history
4. Personal history
5. History of thrombophilia
6. Age (≥50 or ≥70 years on day of diagnosis of VTE)
7. Cancer
8. IVDU
9. Cancer surgery in last 6 months
10. Indwelling catheter
11. Metastatic cancer
12. New cancer diagnosis after VTE
diagnosis
13. Type of VTE (DVT or PE or
DVT+PE)
14. High (quantitative) D-dimer at
diagnosis
15. Gender
Clinical predictors of adverse
outcome in VTE outpatients
• Univariate logistic regression showed that
recurrence was related to younger age
(<50 years, p=0.007) but to none of the 14
other parameters assessed
• Cancer (p<0.001) and a diagnosis of
cancer subsequent to VTE (p=0.037) were
predictive of an adverse event
Clinical predictors of adverse
outcome in VTE outpatients
• Multivariate logistic regression confirmed
these cancer factors were independent
predictors of adverse outcome with high
odds ratios
– Cancer:
– New cancer:
OR 4.3, 95% CI 2.4–7.5
OR 4.3, 95% CI 1.2–15
Clinical predictors of adverse outcome
in VTE outpatients
Non-cancer patients
• Restricting the univariate logistic regression
analysis to non-cancer outpatients:
– age <50 years (p=0.033) was related to the risk of
VTE recurrence
– new cancer diagnosis (p=0.007) was a predictor of
adverse outcome
Clinical predictors of adverse outcome
in VTE outpatients
‘First event’ VTE
• Restricting the univariate logistic regression
analysis to first event VTE outpatients:
– age <50 years (p=0.033) was related to the risk of
VTE recurrence
– cancer (p<0.001), new cancer diagnosis (p=0.008),
metastatic cancer (p=0.02) and high D-dimer at
diagnosis (p=0.023) were all predictors of adverse
outcome
Clinical predictors of adverse outcome
in VTE outpatients
• ‘First event’ VTE
• Multivariate logistic regression confirmed
three factors were independent predictors
of adverse outcome with high odds ratios
– Cancer:
OR 6.3, 95% CI 2.8–14.1
– New cancer: OR 13.0, 95% CI 3.0–57.5
– High D-dimer: OR 2.7, 95% CI 1.0–6.8
Clinical predictors of adverse outcome
in VTE outpatients
Previous history of VTE
• Restricting the univariate logistic regression
analysis to outpatients with previous history
of VTE:
– Cancer (p=0.002) and a hormonal risk factor
(p=0.029) were predictors of adverse outcome
• Cancer was an independent predictor of
adverse outcome on multivariate analysis
– Cancer:OR 6.9, 95% CI 1.8–27.0)
Clinical implication
From the perspective of routine outpatient
treatment of VTE, these results identify
cancer as an overriding risk for adverse
outcome irrespective of VTE history, and
show that high D-dimer at diagnosis is
predictive of adverse outcome in patients
experiencing a first VTE event.
The VERITY PUSH study was
funded by sanofi-aventis