Abstract
Although it is recognized that a systolic blood pressure (SBP) increase ≥2 mm Hg or a diastolic blood pressure (DBP) increase ≥1 mm Hg increases the risk of heart attacks and strokes in middle-aged adults, the Food and Drug Administration (FDA) lacks an adequate policy for regulating medications that increase blood pressure (BP). Some FDA reviewers consider a clinically significant increase in BP to occur only if a drug raises SBP ≥20 mm Hg or if a drug raises DBP ≥10 to 15 mm Hg. In recent years, numerous drugs have been regulated or taken off the market due to cardiovascular safety concerns. The list includes rofecoxib (Vioxx), valdecoxib (Bextra), nonselective nonsteroidal anti-inflammatory drugs, sibutramine (Meridia), and phenylpropanolamine. It is probable that the hypertensive effect of these drugs explains why they increase the risk of adverse cardiovascular events. Other drugs, notably serotonin–norepinephrine reuptake inhibitors and drugs used to treat attention deficit hyperactivity disorder, were approved without cardiovascular safety data despite the fact that they raise BP comparable to valdecoxib and sibutramine. It is the responsibility of the FDA to ensure that drugs are properly labeled regarding risk. Even if a drug raises BP only modestly, FDA guidelines for new drug approvals should include a requirement for cardiovascular safety data. However, such guidelines will not address the problem of how to obtain cardiovascular safety data for the many already approved drugs that increase BP. The FDA should play a role in obtaining cardiovascular safety data for such drugs.
Keywords
Rofecoxib (Vioxx), valdecoxib (Bextra), nonselective nonsteroidal anti-inflammatory drugs, sibutramine (Meridia), and phenylpropanolamine all have 3 things in common. First of all, they were all initially approved by the Food and Drug Administration (FDA) without any warnings about adverse cardiovascular outcomes. Second, all these drugs have been regulated or taken off the market in recent years after data demonstrated that they either increase the risk of heart attacks and strokes or are associated with an increased risk of heart attacks and strokes. Third, they all raise blood pressure (BP).
It is likely that the reason these drugs increase the risk of serious adverse cardiovascular events is because of their hypertensive side effect, and the longer the duration of use, the greater the risk.
A perusal of FDA medical reviews for new drug approvals reveals that different FDA reviewers have utilized different criteria to determine when medications raise BP to an extent that is clinically significant. In 1997, an FDA reviewer specified that in order to be considered clinically significant, the FDA Neuropharmacology guidelines require that a medication raises systolic BP (SBP) to a level ≥180 or raises the SBP from baseline ≥20 mm Hg, or else the medication raises diastolic BP (DBP) to a level ≥105 or raises the DBP ≥15 mm Hg. 1
An FDA reviewer in 1999 wrote that a clinically significant increase in BP means that the drug raises the SBP to a level ≥140 mm Hg and increases the SBP ≥20 mm Hg. Correspondingly, a clinically significant increase in DBP means that the drug raises the DBP to a level ≥90 mm Hg and increases the DBP ≥10 mm Hg. 2
The FDA reviewers also compare the difference in mean change in BP between drug and placebo. When there are small differences in the mean BP effect between drug and placebo (in the range of 1-4 mm Hg) that are not statistically significant, it is commonplace for FDA officials to comment that the differences are clinically irrelevant. 3
On the other hand, when small but statistically significant differences between the BP effect of a drug versus placebo exist, even if the differences are in the 1 to 4 mm Hg range, FDA officials acknowledge that a study with a larger sample size would be necessary to determine whether or not the drug increases the risk of cardiovascular/thrombotic events. 4,5 There appear to be no instances in which approval of a drug was delayed due to a hypertensive effect.
Cardiovascular experts recognize that small elevations in BP increase the risk of adverse cardiovascular events. 6 An epidemiological study published in the Lancet in 2002 demonstrated that there is a continuous relationship between BP and adverse cardiovascular outcomes for SBP ranging between 115 and 180 mm Hg and for DBP ranging between 75 and 105 mm Hg. 7 Thus, for middle-aged individuals, a 2-mm Hg increase in SBP or a 1-mm Hg increase in DBP increases the risk of a heart attack by 10% and increases the risk of a stroke by 7%. 7
According to Norman Stockbridge, MD, PhD, Director, Division of Cardiovascular and Renal Products, FDA Center for Drug Evaluation and Research, the FDA recognizes that a modest elevation in BP, even if it is undetectable in any given individual, increases the risk of heart attacks and strokes (Stockbridge, MD, PhD, e-mail communication, June 2011). Nonetheless, according to Dr Stockbridge, in the absence of unfavorable safety data for a specific medication, the FDA’s stance is that no regulatory action is indicated.
When rofecoxib was approved in 1999, FDA officials were aware that it increased SBP at the approved doses (both clinically and statistically), yet the FDA failed to require the manufacturer to provide cardiovascular safety data. 8 Only after being marketed for years was it recognized that rofecoxib increases the risk of heart attacks and strokes. 9
The FDA officials approved valdecoxib in 2001without a requirement for cardiovascular safety data, despite being aware that valdecoxib significantly increases SBP by 2 mm Hg at the doses for which it was approved (Table 1). 10 The FDA officials approved sibutramine in 1997 although data indicated that it raises SBP by 1.7 to 2.7 mm Hg and that it raises DBP by 2.0 to 2.1 mm Hg at the doses it was approved (Table 1). 11
Blood Pressure Effect of Valdecoxib and Sibutramine.
Abbreviations: N, total number of participants; SBP, systolic blood pressure; DBP, diastolic blood pressure; ND, no data.
In fairness to the FDA, the cardiovascular risk of small increases in BP may not have been understood prior to the epidemiological study published in 2002. Furthermore, the FDA eventually required the manufacturer of sibutramine to provide cardiovascular safety data.
For drugs submitted for approval prior to 2002 which exhibit modest pressor effects, it might be reasonable to overlook FDA approval in the absence of cardiovascular safety data. However, the FDA has approved 3 serotonin–norepinephrine reuptake inhibitor antidepressants since 2002, despite being aware that these drugs raise SBP ≥2 mm Hg and/or raise DBP ≥1 mm Hg. Table 2 summarizes the BP properties of duloxetine (Cymbalta), milnacipran (Savella), and levomilnacipran (Fetzima). Duloxetine, approved in 2004, raises SBP by 1.6 to 3.7 mm Hg and raises DBP by 1.1 mm Hg at approved doses. 12 Milnacipran, approved in 2009, raises SBP by 3.1-3.2 mm Hg and raises DBP by 2.2-2.7 mm Hg at approved doses. 13 Levomilnacipran, approved in 2013, raises SBP by 3.4 to 3.9 mm Hg and raises DBP by 3.1 to 3.2 mm Hg at approved doses. 14
Blood Pressure Effect of Serotonin–Norepinephrine Reuptake Inhibitor Antidepressants Approved Since 2002.
Abbreviations: N, total number of participants; SBP, systolic blood pressure; DBP, diastolic blood pressure; ND, no data. * Since there was no placebo group, the change in mean SBP and DBP is the difference between treatment group and baseline.
There is a dose–response relationship for duloxetine and BP whereby higher doses of the drug cause greater elevations in BP. If cardiovascular safety data were available for duloxetine, it is to be expected that higher doses (daily doses of 80 mg or more) would be associated with an increased risk of adverse cardiovascular events.
For milnacipran and levomilnacipran, all approved doses cause SBP elevations ≥3 mm Hg and DBP elevations ≥2 mm Hg. These levels of BP elevation exceed the hypertensive effect of sibutramine. Since sibutramine was taken off the market due to adverse cardiovascular outcome risk, we predict that safety data, should it become available, will demonstrate that milnacipran and levomilnacipran increase the risk of heart attacks and strokes.
In the FDA’s medical review of lisdexamfetamine (Vyvanse), approved in 2007, DBP was elevated 1.3 mm Hg and 1.7 mm Hg at the 50 mg/day and 70 mg/day doses, respectively, in comparison to placebo.15 In the same medical review, there is mention that the average BP increase with stimulant drugs is about 2-4 mm Hg. 15
Other already approved drugs raise BP but lack adequate cardiovascular safety data. Although there are data posted at the FDA Web site for venlafaxine extended release (XR), 16 there are no regulatory documents posted at the FDA Web site for venlafaxine immediate release (IR), bupropion, and amphetamine salts combination. Nonetheless, as summarized in Table 3, data from the research literature indicate that venlafaxine IR, bupropion, and amphetamine salts combination each cause SBP elevations ≥2 mm Hg and/or DBP elevations ≥1 mm Hg at approved doses. 17 –19 Moreover, the medical review for venlafaxine XR mentions that premarketing studies revealed that the mean DBP increase for venlafaxine IR at the maximally approved daily dose of 375 mg is 7.2 mm Hg. 16
Blood Pressure Effect of Some Medications Approved Prior to 2002.
Abbreviations: N, total number of participants; SBP, systolic blood pressure; DBP, diastolic blood pressure; ND, no data; avg, average; XR, extended release; IR, immediate release.
It is curious for a regulatory agency such as the FDA to claim that, in the absence of unfavorable safety data, no regulatory action is indicated for medications that raise BP. After all, the reason that there are no safety data is because the FDA failed to require it.
The FDA ought to play a role in collecting cardiovascular safety data for drugs already on the market that raise BP, even those approved prior to 2002. For generically available drugs, the FDA can sponsor retrospective studies to ascertain whether these drugs are associated with an increased risk of adverse cardiovascular events. The FDA has employed this approach with some medications used to treat attention-deficit hyperactivity disorder. 20,21 For drugs already approved but not available generically, as well as for drugs not yet approved, the FDA can require manufacturers to obtain cardiovascular safety data.
To this day, FDA reviewers use outmoded criteria to determine what constitutes a clinically significant increase in BP. 22 In our opinion, this situation exists because FDA reviewers lack proper guidance. A 2005 FDA guideline for reviewers makes no mention as to what constitutes a clinically significant increase in BP. 23 Members of the Cardiac Safety Research Consortium (CSRC) published a White Paper in 2013 with the purpose of assisting industry, clinicians, and regulatory authorities regarding the issue of medications that increase BP. 24 The members of the CSRC made no recommendation as to what threshold elevation in BP ought to be used such that a BP increase above this level warrants additional study to clarify cardiovascular risk.
Because small elevations in BP appear to increase serious adverse cardiovascular events, the FDA ought to require cardiovascular safety data for drugs that modestly increase BP. Although this might slow the development of new pharmaceutical products, we believe that the FDA ought to err on the side of patient safety. Accordingly, we recommend that if a drug increases SBP > 2 mm Hg or if a drug increases DBP > 1 mm Hg, then there should be a requirement for cardiovascular safety data.
In conclusion, it is unwise to allow drugs that predictably increase risk to be marketed without adequate safety data, regardless of when the drugs were approved. When drugs modestly increase BP, the FDA should require the collection of cardiovascular safety data for new drug approvals. In addition, the FDA should take a role in collecting cardiovascular safety data for already approved drugs that modestly increase BP. For drugs that increase risk, it is not necessary that they be taken off the market, but it is the duty of the FDA to ensure that drugs are properly labeled regarding risk.
Footnotes
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: RPB has a patent pending for the concept that one can use markers of fluid retention to gauge the cardiovascular risk of a drug. RPB and IHI have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.