Non-steroidal anti-inflammatory drugs and antihypertensive drug therapy

With the exception of low-dose aspirin [1,2] non-steroidal anti-inflammatory drugs (NSAIDs) have been shown to be capable of increasing blood pressure in both treated and untreated hypertensives. This effect has been observed with both COX1 and COX2 inhibitors, although it is the non-specific COX inhibitors that have been studied most extensively. Blunting of blood pressure control has been seen when drugs such as indomethacin have been used in patients treated with diuretics, beta-blockers, angiotensin-converting enzyme (ACE) inhibitors or varying combinations of these and older agents. Some trials suggest that the effect may be less with sulindac than with other NSAIDs [3–5]. Both COX1 and COX2 are expressed constitutively in renal and vascular tissue and the effect on blood pressure has been ascribed to inhibition of vasodilator and natriuretic prostaglandins in the kidney and/or decreased synthesis of vascular or endothelial prostaglandin synthesis with consequent salt retention and vasoconstriction. Average blood pressure increases of the order of 5 mmHg systolic have been reported in treated hypertensives but there is considerable individual variation in the pressor response, and severe hypertension, cardiac and renal failure may occur in susceptible individuals. The question as to whether blood pressure control by newer antihypertensive agents, such as the angiotensin II receptor blockers, shows a similar response to NSAIDs is addressed in this issue of the journal by Fogari et al. [6] with regard to the effect of indomethacin on the antihypertensive efficacy of valsartan and lisinopril. They randomized 128 patients to receive one or other of these antihhypertensives for 10 weeks, at the end of which they were further randomized to receive indomethacin or placebo for 2 weeks each in a cross-over fashion with a 1-week wash out in between. They found that indomethacin significantly blunted the blood pressure-lowering effect of both antihypertensives. Although the pressor effect of indomethacin was on average greater in the lisinopril-treated patients (5.45/3.22 mmHg) than those on valsartan (2.12/1.87 mmHg), this difference was not statistically different. Nevertheless, from the data, the authors postulate that bradykinin-stimulated prostaglandin synthesis might play some role in the antihypertensive action of angiotensin receptor inhibitors but to a lesser extent than for ACE inhibitors. Given the widespread use of anti-inflammatory agents and their frequent use in older patients in the treatment of hypertension, it is valuable to have comparative data from controlled trials. However, some caveats and comments are in order. First, the blood pressures were clinic recordings and thus give no indication of the effects of NSAIDs on 24-h or diurnal blood pressure control. Furthermore, clinic pressures may underestimate the pressor effects of NSAIDs. Second, although the average blood pressure rise with indomethacin was greater in the lisinopril-treated group, it would be hazardous to assume that the risk was less for patients on angiotensin receptor blockers in the absence of a significant difference between the patient groups; particularly as Conlin et al. [7] showed that 1 week of indomethacin 75 mg daily attenuated the 24-h ambulatory blood pressure falls seen with captopril and losartan to a similar extent. Indomethacin has also been shown to cause comparable blunting of the natriuresis induced by valsartan and enalapril in a single-blind, cross-over study in healthy subjects [8]. However, support for the concept that NSAIDs may have a greater effect on blood pressure control by ACE inhibitors than with some dihydropyridine calcium channel blocking agents comes from a trial comparing the effects of indomethacin in 42 elderly hypertensives who were well controlled on either amlodipine or enalapril [9]. When indomethacin 50 mg twice daily or placebo were given for 3 weeks in a double-blind, cross-over design, 24-h ambulatory blood pressure rose substantially and pulse rates fell in the enalapril group with little effect in those on amlodipine, with the differences averaging 10.1/4.9 mmHg and 5.6 beats/min. Similarly, in a smaller controlled study, Polonia et al. [10] showed indomethacin 75 mg daily for 1 week caused a 45% blunting of the effects on ambulatory blood pressure reduction with enalapril but not nifedipine. There has been no controlled study of the effects of non-specific NSAIDs on cardiovascular outcome in trials of antihypertensive therapy, with the exception of the low-dose aspirin arm of the HOT study [11], which showed a reduction in major cardiovascular events compared to placebo. In this respect, a post-hoc analysis of the nitrendipine-based Syst-Eur trial [12] is of interest in that the 861 patients reported as taking NSAIDs on at least 50% of occasions showed no difference in cardiovascular outcome compared to the 2882 patients who were never taking NSAIDs. It would be interesting to see similar post-hoc analyses in relation to NSAID use from outcome studies using an ACE inhibitor, angiotensin II receptor blocker or other antihypertensive agent. This issue is one of particular interest with respect to COX2 inhibitors in the light of the unconfirmed suggestion that they might increase the risk of thrombosis [13]. Although the risks of NSAIDs with calcium channel blocking agents may be less than for other antihypertensive drugs, pending further data, clinicians should continue to exercise particular caution and minimize the use of NSAIDs in patients with hypertension regardless of the type of antihypertensive therapy. The risk of aggravating hypertension appears to be greater in older subjects, in those with higher pressures and in the presence of a high salt intake [5]. When NSAIDs are required, a patient's blood pressure should be carefully checked in the weeks following initiation of therapy, while those with decreased myocardial reserve or renal impairment should be made aware of possible deterioration and be monitored appropriately. The other main risk of NSAIDs in relation to antihypertensive therapy is the development of acute renal failure. This is most likely to occur following severe hypovolaemia as a result of dehydration, salt depletion or bleeding in patients on a combination of diuretics and drugs that block the renin–angiotensin system. Early data on the COX2 inhibitors suggest that similar precautions should apply to these as for the non-specific agents [14,15].