Cardiovascular risk associated with use of NSAIDs controversy

1. Diana Girnita, MD PhD
UC Rheumatology Fellow

2. Nonsteroidal anti-inflammatory
drugs (NSAIDs)
• Effects
– anti-inflammatory
– analgesic
– antipyretic
• CDC reported: 52.5 million
US people with arthritis
• NSAIDs Indications:
– Gout
– Spondylarthropathies
(Ankylosing
Spondylitis)
– Rheumatoid Arthritis
– Osteoarthritis
– Acute or Chronic
Musculoskeletal Pain

3. NSAIDs Usage
• $5 billion/year - drug costs
in prescribed NSAIDs
• $3 billion additional
spending/yearly for OTC
analgesics including
acetaminophen.
Irena Melnikova Nature Reviews Drug
Discovery 9, 589-590;2010

4. CDC data on NSAIDs
prescriptions
Percent of NSAID visits
1999-2000
18 - 44 years 2000-2001
45 – 64 years
65 -74 years
> 75 years
Centers for Disease
Control and Prevention,
National Center for Health
Statistics. Health, US,2004

5. Pain assessment scale
P<0.001
McKenna F, Borenstein D, Wendt H et al. Celecoxib versus diclofenac in the
management of osteoarthritis of the knee. Scand J Rheumatol 2001;30:118.

7. 1899, Bayer
Archives

8. History of Aspirin

9. NSAIDs
Traditional (starting 1960’s)
Salicylic acids Aspirin
Acetic acids Diclofenac
Etodolac
Indometacin
Sulindac
Ketorolac
Propionic acids Ketoprofen
Ibuprofen
Naproxen
Pyrazolones Phenilbutazone
Oxicams Pyroxicam
Meloxicam
Nonacidic Relafen
Coxibs –COX2 selective
(approved 1998)
Sulfonamide Celecoxib
(celebrex)
Sulfonylurea Etoricoxib
(arcoxia)
Nonacid Lumaricoxib

10. Mechanism of action
STIMULUS
Phospholipids –cellular
membrane
Phospholipase A2
Lipoxygenase COX
Leukotrienes Arachidonic acid
COX = cyclo-oxygenase
TX2 =thromboxane
PG =Prostaglandines
TX2
PG
Prostacyclin

11. COX-1 versus COX-2
COX-1 COX-2
Expression
Tissue
localization
Ubiquitous Inflammatory and
Role “Housekeeping” and
maintenance
neoplastic sites (small
amounts in kidney,
uterus, ovary, CNS
[neocortex,
hippocampus])
Pro-inflammatory and
mitogenic functions (?
neuronal plasticity)
Constitutive
(activated by
physiologic stimuli)
Inducible by pro-inflammatory
stimuli (LPS,
TNFa, IL-2, IFNg etc)

12. Mechanism of
action
Vasoconstriction
Platelet aggregation
Vasodilatation
Hyperalgesia
Fever
Diuresis
Immunomodulation
Smooth muscle
contraction
Bronchoconstriction
Vasodilatation
Inhibits platelet
aggregation
Smooth muscle
contraction
Inhibits platelet
aggregation
Hydrolysis
Oxygenation
NSAIDs

13. NSAIDs Classification by COX selectivity
In vitro
assessment
of COX-1/ COX-2
activity
Non-specific
inhibition of
COX-1 -
gastrointestinal
and platelet-adverse
Sleisenger & Fordtran's Gastrointestinal and Liver Disease, 8th ed., Copyright © 2006 Saunders
effects

14. Are they safe medication?
Wolfe MM et al. N Engl J Med 1999;340:1888-1899.

15. Common Side Effects
• Gastrointestinal
– Dyspepsia
– Esophagitis
– Ulcers
– Perforations
– Small bowel erosions/ strictures
– Colitis
• Renal
– sodium retention
– Edema
– HTN
– RTA, AKI, AIN
– Hyperkalemia
– Papillary necrosis
• Hepatic
– Elevated AST, ALT
• Asthma
-Aspirin induced, rash
• Hematologic -cytopenia
• Nervous -dizziness,
confusion, seizures, aseptic
meningitis
• Cardiac –will discuss later….

16. Since the reported risk of GI
bleed was high with the
traditional NSAIDs (tNSAIDs),
the COXIBS - approved 1998

17. RA- RHEUMATOID ARTHRITIS
OA-OSTEOARTHRITIS

18. STUDY Participants RESULT
VIGOR, 2000 8076 COX-2 -Increased CV risk, no difference
in mortality
CLASS, 2000 8095 COX-2 -No increase in CV risk
Metaanalysis, 2001 18000 COX-2 Increased CV risk
TARGET, 2004 18325 COX-2 No difference in CV risk,
increase in blood pressure
APPROVE, 2005 2586 COX-2 Increased CV risk (thrombotic
events, CHF, HTN )
APC+PreSAP 3800 COX-2 Increased CV risk
Metaanalysis, 2006 145343 Increased CV risk for both COX-2 and
tNSAIDs, not naproxen
MEDAL, 2009 34701 COX-2 No difference in CV risk
Metaanalysis, 2013 124513 Increased CV risk for both COX-2 and
tNSAIDs, not naproxen
Danish, 2014 83677 Increased CV risk for COX2 and
Diclofenac

19. VIGOR STUDY
NEJM 2000; 343:1520-8

20. VIGOR study
• 8076 patients > 50 yo / 40 yo on long term
steroids with Rheumatoid arthritis
• Double blinded, RTC
Rofecoxib (Vioxx)
50 mg QD
(n=4047)
Naproxen 500mg BID
(n=4029)
• vs
• Primary endpoint: upper GI events
(perforation, bleeding, obstruction, ulcers)

21. VIGOR - Confirmed Thrombotic
Patients with Events (Rates per 100 Patient-Years)
Event Category
Rofecoxib
N=4047
Naproxen
N=4029
Relative Risk
(95% CI)
Confirmed
CV events
45 (1.7) 19 (0.7) 0.42
(0.25, 0.72)
Cardiac
events
28 (1.0) 10 (0.4) 0.36
(0.17, 0.74)
Cerebrovascular
events
11 (0.4) 8 (0.3) 0.73
(0.29, 1.80)
Peripheral
vascular events
6 (0.2) 1 (0.04) 0.17
(0.00, 1.37)
Cardiovascular Events
Source: Data on file, MSD

22. No difference in the mortality rate
• 0.5 % for rofecoxib vs 0.4% in naproxen –
seven patients died from CV event in each
group
• MI less common in the naproxen vs rofecoxib
group (0.1% vs 0.4% -95%CI 0.1-0.6; RR =0.2)
• MERK allow patients to use LOW –DOSE
aspirin (history of myocardial infarction,
angina, stroke, CABG/PCI)

23. JAMA 2000; 284: 1247-1255

24. CLASS =Celecoxib long term
arthritis safety study
• Prospective, double-blind, RTC (1998-2000)
• 8095 patients ≥ 18yo with OA or RA (27%);
• Aspirin use was permitted ≤325 mg/day (20%
of patients)
Celecoxib 400 mg BID
(n=3987)
• vs
tNSAIDs
Ibuprofen 800 mg TID
(n=1985)
Diclofenac 75 mg BID
(n=1996)

25. NO increase risk in CV
thromboembolic events
No report about the CV events in patients that
WERE TAKING aspirin

26. ARE COX-2 INHIBITORS
ASSOCIATED WITH
INCREASED
CARDIOVASCULAR (CV) RISK?

27. Increased CV risk with COX-2
Mukherjee D , Topol E. Risk of CV events associated with selective
COX2 inhibitors JAMA 2001, 286 954-959

28. Increased Risk of Myocardial
Infarction
• MI= myocardial
infarction

29. TARGET STUDY
Lancet 2004; 364: 675–84

30. TARGET study
• 18325 patients ≥ 50 yo with OA for 52 weeks
• Double blinded, RTC
Lumiracoxib 400mg QD
(n=9158)
Naproxen 500mg BID
(n=4754)
Ibuprofen 800 mg TID
(n=4415)
• 24% of patients (n=4326) on low dose ASA
• Primary endpoints
• Nonfatal and silent MI, stroke and CV death

31. No Difference in The Composite
Primary Endpoint
•Primary endpoints
• Nonfatal and
silent MI, stroke
and CV death
•No difference
between tNSAIDs
and Cox-2
Farkouh M et al; TARGET study, The Lancet, Volume 364, Issue 9435, 2004, 675 - 684

32. No Difference in Confirmed and
Probable MI
• Overall, no significant
difference
• Ibuprofen sub-study:
rates did not differ
irrespective of Aspirin use
• Naproxen sub-study: 4
events, 0.11% vs Coxib
group -10 events; 0.28%
in patients NOT taking
Aspirin

33. Change in baseline
blood pressure
(BP)
• Systolic BP (Least square
means change) was +0.4
mmHg for Coxib vs
+2.1mmHg for NSAIDS
(p<0.001)
• Diastolic BP (least square
means) was -0.1 mmHg
for Coxib vs +0.5 mmHg
for NSAIDS (p<0.001)

34. APPROVE STUDY
N Engl J Med 2005;352:1092-102.

35. APPROVE STUDY
• 2586 patients-history of colorectal
adenoma
• Double blinded RTC, 3 years (2000-2003)
Rofecoxib 25 mg QD
(n=1287) Placebo (n=1299)
• vs
• Primary endpoint: Thrombotic CV events
• 20% were on low dose aspirin

36. Increased incidence of
thrombotic events

37. Increased incidence of CV
events after 18 months

38. Increased incidence for CHF

39. Increased Risk for
CHF and HTN

40. 2004 FDA Warning for all NSAIDs
Public Health Advisory concerning the use NSAIDs,
including COX-2 selective agents:
-COX-2 selective agents (Vioxx, Celebrex, and
Bextra) may be associated with an increased risk of
serious CV events (heart attack and stroke)
-Long-term use of a non-selective NSAID, naproxen
(sold as Aleve, Naprosyn) may be associated with an
increased cardiovascular risk compared to
placebo.
December 23, 2004
http://www.fda.gov/cder/drug/infopage/cox2/default.htm

41. 2005- Worldwide withdrawal of
Rofecoxib

42. Questions
• Is it a dose related risk?
• Is it a time relation between Coxibs
and increase the CV risk?
• Is the association with Aspirin of any
benefit?

43. Aspirin-the
only
Cardio-protective
NSAID
287 trials
200,000
patients
Lancet. May 30, 2009; 373(9678): 1849–1860. Aspirin in the primary and
secondary prevention of vascular disease: collaborative meta-analysis - participant
data from Antithrombotic Trialists' (ATT) Collaboration

44. How does Aspirin work?
•Irreversibly inhibits COX1
(selectivity COX1150-200 x
>COX2), change catalitycal site
of COX2
•COX1 enzyme is inhibited for
the lifetime of the platelet (~8
-11 days).
•Effect achieved at very low
dose.
•Thromboxane A2 synthesis is
inhibited in platelets  no
Platelet aggregation

45. Lipoxins –antiinflammatory

46. HOW ARE NSAIDS AFFECTING
ASPIRIN EFFECT?

47. Ibuprofen before aspirin limits
cardioprotective effect of aspirin
• Ibuprofen (COX-1) inhibits thromboxane B2 and platelet
aggregation
Lawson FC et al “Cyclooxigenase inhibitors and the antiplatelet
effect of aspirin; NEJM 2001;345:1809-17

48. Rofecoxib and diclofenac (>COX2)- do
not limit aspirin effect
Lawson FC et al “Cyclooxygenase inhibitors and the antiplatelet
effect of aspirin; NEJM 2001;345:1809-17

49. WHAT DO THE CLINICAL
TRIALS REPORT?

50. APC
(adenoma
prevention with
Celecoxib)
Celecoxib -200mg
and 400mg BID
vs placebo
PreSAP
(prevention of
spontaneous
adenomatous
polyps)
-to reduce colorectal
adenoma recurrence
after polypectomy
Celecoxib 400mg
QD vs placebo
• Independent committee of both studies
evaluated CV events
Solomon S et al. Circulation. 2006;114:1028-1035

51. Doubled risk for CV events
• 83 patients with CV death, nonfatal MI,
nonfatal stroke, heart failure
• Prespecified composite endpoint – overall HR
=1.9 (95%CI 1.1-3.1)
• Significant increase in systolic BP at 1 and 3
years in the APC study (2-3mmHg for 200mg
BID and 3-6mmHg for 400mg BID)

52. Higher the dose…higher the CV risk
Solomon S et al. Circulation. 2006;114:1028-1035

53. Time to the composite end point
Solomon S et al. Circulation. 2006;114:1028-1035

54. HR for CV death, nonfatal MI,
stroke, heart failure

55. Meta-analysis of randomized
trials
Kearney PM et al. Do selective cyclo-oxygenase-
2 inhibitors and traditional non-steroidal
anti-inflammatory drugs increase the
risk of atherothrombosis? BMJ. 2006;332:1302.

56. Metaanalysis, 2006
• 145 373 participants
• 138 randomised trials involving a
comparison of a selective COX 2 inhibitor
versus placebo or versus a traditional
NSAID (or both)

57. tNSAIDs vs Placebo
NSAID RR 95% CI
Diclofenac 1.63 1.12 - 2.37
Ibuprofen 1.51 0.96 - 2.37
Naproxen 0.92 0.67 - 1.26
Kearney PM et al. BMJ. 2006;332;1302.

58. COX-2 inhibitors vs Placebo
Kearney et al. BMJ. 2006;332:1302.

59. COX-2
inhibitors
vs
NSAIDs
Kearney et al. BMJ. 2006;332:1302.

60. Antman EM et al. Use of nonsteroidal antiinflammatory drugs: an update for
clinicians: a scientific statement from the American Heart Association.
Circulation. 2007 Mar 27;115(12):1634-42.

61. American College Rheumatology- Drug Safety
Committee members
“While we appreciate the documented CV toxicity associated
with selective and non-selective NSAIDs, the current AHA
statement does not reflect the proven benefits of these agents
in patients with arthritis and other chronic inflammatory
conditions. Sufficient treatment options are critical for patients
with arthritis. Just as collaborative treatment plans for a given
patient result in improved care, we urge the cardiology
community to engage their rheumatology colleagues in
formulating optimal recommendations for the use of NSAIDs”
Neal S. Birnbaum, MD, ACR President Daniel J. Lovell,
MD, MPH, and Daniel H. Solomon, MD, MPH-

62. RA and Cardiovascular disease (CVD)
• The absolute CV risk in RA patients has been
estimated to be similar to that in non-RA
patients that are 10 years older.
• The risk of CV events and death in RA is
similar to that seen in patients with type 2
diabetes
Kremers HM et al. High ten-year risk of CV disease in newly diagnosed rheumatoid arthritis
patients: a population-based cohort study. Arthritis Rheum 2008; 58:2268–2274.
Peters MJ et al. Does rheumatoid arthritis equal diabetes mellitus as an independent risk factor
for cardiovascular disease? A prospective study. Arthritis Rheum 2009; 61:1571–1579.

63. RA and cardiovascular disease (CVD)
• People with RA are at excess risk for CVD.
• Systemic inflammation and its interplay with traditional
and nontraditional CV risk factors appear to have a major
role.
• Cardiovascular risk scores (e.g. Framingham) developed for
the general population are unlikely to accurately estimate
cardiovascular risk in RA.
Sherine G et al. Risk factors for cardiovascular disease in rheumatoid
arthritis. Curr Opin Rheumatol 2012, 24:171–176

64. MEDAL STUDY -2009

65. MEDAL Program Trials
3 Trials 46 countries 1380 sites
 EDGE (OA): N=7,111
 EDGE II (RA): N=4,086 34,701 patients
 MEDAL (OA/RA): N=23,504
R n=17,412
A
NDOMIZE
Etoricoxib
60 or 90 mg/d (OA)
90 mg/d (RA)
Diclofenac
150 mg/d
(50 mg tid or 75 mg bid)
n=17,289
 ≥ 50 years of age
 OA of knee, hip, hand,
or spine; or RA
 Require long-term therapy
with traditional NSAID or
COX-2 inhibitor
 Broad CV risk
 Allowed aspirin and PPI
use in appropriate patients
Mean duration of therapy=18 months

66. Patient Characteristics
No. of Patients (%)
Etoricoxib
(n=17,412)
Diclofenac
(n=17,289)
Mean age, years (SD) 63.2 (8.5) 63.2 (8.5)
Female 12,925 (74.2) 12,823 (74.2)
Osteoarthritis 12,533 (72.0) 12,380 (71.6)
Rheumatoid arthritis 4878 (28.0) 4909 (28.4)
Diabetes 1810 (10.4) 1855 (10.7)
Hypertension 8109 (46.6) 8221 (47.6)
Dyslipidemia 5097 (29.3) 5034 (29.1)
Current smoker 2034 (11.7) 2037 (11.8)
Established atherosclerotic CV
disease 2014 (11.6) 2010 (11.6)
≥2 CV risk factors* or established
6586 (37.8) 6639 (38.4)
atherosclerotic CV disease
*Included hypertension, diabetes mellitus, dyslipidemia, family history of CV disease, or smoking.

67. Primary Endpoint: Cumulative Incidence
of Thrombotic CV Events
Etoricoxib (320 events)
Diclofenac (323 events)
Months
No. of patients at risk*
Etoricoxib
Diclofenac
16,819
16,483
13,359 10,733 8277 6427 4024 805
12,800 10,142 7901 6213 3832 815
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0
0 6 12 18 24 30 36 42
Cumulative incidence (%) with 95% CI
Etoricoxib vs diclofenac
HR = 0.95
95% CI = (0.81-1.11)
*Per protocol population.

68. Cumulative Incidence of Arterial
Thrombotic CV and APTC Events
Arterial Thrombotic CV Events CVD/MI/Stroke
Etoricoxib (216 events)
Diclofenac (216 events)
Etoricoxib (272 events)
Diclofenac (272 events)
7.0
6.0
5.0
4.0
3.0
2.0
1.0
16,819
16,483
13,362
12,801
10,735
10,144
8277
7903
6427
6214
4024
3832
805
815
No. of patients at risk
Etoricoxib
Diclofenac
Months
Etoricoxib vs diclofenac
HR = 0.96
95% CI = (0.79-1.16)
0
0 6 12 18 24 30 36 42
Cumulative incidence (%) with 95% CI
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0
0 6 12 18 24 30 36 42
Cumulative incidence (%) with 95% CI
Etoricoxib vs diclofenac
HR = 0.96
95% CI = (0.81-1.13)
Months
16,819
16,483
13,366
12,814
10,745
10,155
8282
7906
6429
6218
4026
3832
805
816
No. of patients at risk
Etoricoxib
Diclofenac

69. 2013 Meta-analysis
• 280 trials of NSAIDs vs placebo (124513 patients, 68342
person years)
• 474 trials one NSAID vs another NSAID (229296
patients, 165456 person-years)
• Main outcomes
– Major vascular events (nonfatal MI/ stroke or vascular death)
– Major coronary events (nonfatal MI or coronary death)
– Upper GI complications (perforation, obstruction, bleed)

70. Drug Major vascular
events
Major coronary
events
Coxibs RR -1.37
95% CI -1.14-1.66
P=0.009
RR – 1.76
95% CI 1.31–2.37
P=0·0001
Diclofenac RR-1.41
95% CI - 1.12- 1.78
P =0.036
RR - 1·70
95% CI – 1.19–2·41
P=0·0032
Ibuprofen RR- 1·44,
95% CI -0.89–2.33
P>0.05
RR - 2.22,
95% CI 1.10–4.48
P=0·0253
Naproxen RR- 0.93
95% CI -0.69-1.27
P-0.66
RR- 0.84
95% CI -0.52-1.35
P= 0.48
!!!ALL NSAIDS doubled the risk of heart failure
RR-rate ratio

71. Coxibs and traditional NSAID Trialist’s collaboration. Vascular and upper GI effects of
NSAIs: meta-analyses from randomized trails. Lancet 2013; 382 769-79

72. The Impact of NSAID Treatment
on Cardiovascular Risk –
Insight from Danish Observational
Data (1997-2006)
Anne-Marie Schjerning Olsen et al. Basic & Clinical Pharmacology &
Toxicology, 2014, 115, 179–184

73. Increased CV risk with NSAID
use in healthy people
• the risk might be
the same in
healthy people
• most NSAIDs
were associated
with increased CV
risk (worse for
diclofenac and
rofecoxib)

74. Diclofenac and Rofecoxib
increased risk of MI/ re-MI

75. The RR of
NSAID use was
associated with
persistently
increased CV
risk throughout all 5
years after
discharge from
hospital
after the first MI.

76. NSAIDs and
Platelets/Endothelial Cells

77. CONCLUSION

79. 2014 – FDA recommendations
NSAID medicines may increase the chance of a heart
attack or stroke that can lead to death. This chance
increases:
•with increasing doses of NSAID medicines
•with longer use of NSAID medicines
•in people who have heart disease
NSAID medicines should never be used right before or
after a heart surgery called a “coronary artery bypass
graft (CABG)."
http://www.fda.gov/downloads/drugs/drugsafety/ucm387559.pdf

80. 2014-FDA recomendations
NSAID medicines should only be used:
• exactly as prescribed
• at the lowest dose possible
• for the shortest time needed
http://www.fda.gov/downloads/drugs/drugsafety/ucm387559.pdf

81. Some practical advises
• NSAIDs (Coxibs) - viable and effective
options to treat pain, fever and inflammation
• They are associated with increased CV risk
(varying degrees)
• Consider co-morbidities
• Elderly patients are at increased risk of
adverse events

82. Practical advises for patients
with high CV risk
• Counsel patients about NSAIDs cardiovascular
risks.
• Low dose coxibs and Naproxen seem to be
safer than Diclofenac, rofecoxib, and ibuprofen
that seem to have increased risk
• Regularly review for the need of continuing this
medication

83. Association with Aspirin does not
confer cardioprotection
• The use of low dose aspirin does not
consistently abrogate the potential CV risk
of a NSAID or COX-2 inhibitor.
• Patients who require the cardioprotective
effects of aspirin may not be ideal
candidates for NSAID therapy.

84. The Future….
• Await results of trials such as PRECISION
(Prospective Randomised Evaluation of Celecoxib
Integrated Safety versus Ibuprofen or Naproxen)-
phase IV
Celecoxib 100 to 200 mg BID vs
Ibuprofen 600 mg to 800 mg TID or
Naproxen 375mg to 500 mg BID
• http://clinicaltrials.gov/ct2/show/results/NC
T00346216

85. THANK YOU!

86. VIGOR - GI Endpoints
5
4
3
2
1
0
Confirmed Clinical
Upper GI Events
†p < 0.001. * p = 0.005.
Confirmed
Complicated
Upper GI Events
All Clinical
GI Bleeding
RR: 0.46†
(0.33, 0.64)
RR: 0.43*
(0.24, 0.78)
RR: 0.38†
(0.25, 0.57)
Rates per 100 Patient-Years
Rofecoxib
Naproxen
( ) = 95% CI.
Source: Bombardier, et al. N Engl J Med. 2000.

87. CLASS -Incidence of upper GI
events
Numbers above bars indicated events per patient-years
exposure
NSAIDS- nonsteroidal anti-inflammatory drugs

88. Aspirin- dose dependent effects
Low: < 300mg blocks platelet aggregation
Intermediate: 300-2400mg/day
antipyretic and analgesic effect
High: 2400-4000mg/day
anti-inflammatory +/- uricosuric effect

89. Effect of NSAID’s on Platelet-Endothelial Interactions

90. Inhibiting COX-2
• COX-2 derived PGI2 can antagonize catecholamine- and
angiotensin II-induced vasoconstriction that will elevate
blood pressure
• Destabilize atherosclerotic plaques (due to its anti-inflammatory
actions)
• COX-1 and COX-2 – generated PGs (TxA2, PGF2 , PGI2
(glom), PGE2 (medulla), powerful vasodilators) affect Na+
retention at kidney level

91. Role of Lipoxins in Anti-inflammatory effects of Aspirin