The case for antihypertensive drug regimens that produce consistent 24-h blood pressure control has largely been founded on various pieces of epidemiologic evidence (1). This evidence is largely dependent on cross-sectional studies (2) or on relatively small-scale follow-up studies (3-5). Therefore, although there is clear evidence that target organ damage associated with hypertension is more closely related to 24-h average rather than to isolated clinic blood pressure measurement, evidence for the clinical superiority of ambulatory blood pressure was lacking. However, recently published data from the SAMPLE Study Group (6) unequivocally demonstrated in a prospective study in hypertensive subjects with left ventricular (LV) hypertrophy that the reduction in LV mass index associated with 1 year of treatment was predicted much more closely by treatment-induced changes in ambulatory blood pressure than in clinic blood pressure. This study clearly provides definitive and confirmatory data to support the aim of achieving blood pressure control on the basis of a smooth and consistent antihypertensive effect over a full 24-h period.
At present, in the developed countries around the world more than 100 different drugs are licensed for the treatment of hypertension. Many of these drugs are licensed and recommended for once-daily administration, but it would be naive for the prescribing physician to assume that all such agents are equally efficacious or, indeed, that they are all appropriate for or optimally suited to once-daily administration. Because there are obvious pharmacologic and formulation differences among different agents, it appears highly unlikely that the profile of blood pressure reduction throughout 24 h will be the same for each drug or drug formulation.
To optimize drug therapy by achieving consistent and sustained blood pressure control, a validated index that defines the duration of action of an antihypertensive drug and that also discriminates among alternative treatments and treatment regimens would be of considerable value.
THE PRINCIPLES UNDERLYING THE TROUGH:PEAK RATIO
After drug administration, the maximal or peak antihypertensive effect usually occurs within 2-8 h of dosing at steady-state, with the time depending on the pharmacologic characteristics of the particular agent. Thereafter, the magnitude of the antihypertensive response gradually declines as the drug is cleared from the body and the "trough" effect is the residual blood pressure-lowering effect at the end of the dosage interval, immediately before administration of the next dose. By definition, the trough blood pressure response is always measured at 24 h post dose with a once-daily drug regimen and 12 h post dose with twice-daily medication. The trough:peak ratio is simply the ratio between the trough and peak blood pressure reductions and is often expressed either as a fraction of 1 or as a percentage (Fig. 1).
The concept of trough:peak ratio was first introduced for reasons related to safety rather than to antihypertensive efficacy when the FDA drafted guidelines in 1988 for the clinical evaluation of new antihypertensive drugs. These guidelines suggested that the trough effect should be no less than one-half to two-thirds of the peak effect, with the result that a trough:peak ratio of 50% is the absolute minimum for satisfying the FDA's recommendation. This aspect of the draft guidelines was formulated largely because of concerns that some new drugs were being administered in inappropriately large doses that produced marked reductions in blood pressure at the time of peak activity. Therefore, a measurable effect at trough was maintained even though there had been a marked and progressive reduction in the magnitude of response across the dosage interval. The rationale for this particular guideline was based largely on concerns that an exaggerated peak effect might lead to impaired critical organ perfusion, particularly to impaired cerebral perfusion, with the risk for symptomatic light-headedness and possible syncope or even, potentially, thrombotic stroke.
Although the FDA made recommendations for the minimal acceptable magnitude of trough:peak ratio, no detailed guidelines were provided concerning the most appropriate methodology. The only definitive aspect that was specifically incorporated into the FDA guidelines was that account must be taken of the placebo effect. Rose and McMahon (7) have clearly demonstrated the importance of placebo correction and have suggested that failure to account for the placebo contribution almost invariably enhances the value of the calculated trough:peak ratio. Circadian variability in blood pressure is a potentially important component of the placebo effect. This is illustrated in Fig. 1, in which the difference in blood pressure at any two time points in the untreated patient may often approach or exceed the decrement in blood pressure associated with the pharmacologic effects of an antihypertensive drug. From this figure it is also apparent that any arbitrary and fixed time point believed to be associated with the peak effect of the drug may also tend to underestimate the peak response and thus result in an inappropriate enhancement of the trough:peak ratio.
It is clear that the validity of trough:peak ratio has an index and that duration of action of an antihypertensive drug is pivotally dependent on the accuracy and reproducibility of the approach adopted for quantifying the magnitude of blood pressure reduction. Although a number of different approaches have been adopted for measurement of the trough:peak ratio, only ambulatory blood pressure monitoring and the repeated measurement of blood pressure by conventional sphygmomanometry in the controlled environment of a clinical research unit have been subject to any form of validation (8,9). Both approaches have advantages and disadvantages. In particular, the "research unit" approach offers a higher degree of precision and accuracy. However, no matter which approach is adopted, certain common factors emerge, and these indicate that any study defining trough:peak ratio should incorporate placebo assessment, steady-state treatment and, ideally, randomized crossover design. In the case of the ambulatory blood pressure approach, it appears that a pragmatic approach should be adopted such that trough:peak ratio is calculated only in those patients who achieve a given blood pressure reduction at the time of peak response (8).
Ideally, values of trough:peak ratio should not be expressed as a simple single-point estimate but rather with some measure of the variability around the mean or median value. Therefore, it is important to regard the trough:peak ratio as a mean of ratios rather than as a ratio of mean data. The discrepancies that can arise by the application of different approaches to calculating trough:peak ratio were well illustrated in a study comparing four antihypertensive agents (10). The use of simple estimates from the mean blood pressure profiles provided an illustration of the differential effects of these drugs but could provide no information concerning the variability in these estimates. The alternative approach of analyzing individual data is dependent on whether ambulatory blood pressure data were averaged for each patient over 1- or 3-h intervals. Not surprisingly, averaging over 3-h intervals blunted the estimates of peak effect and thus tended to enhance the trough:peak ratio for all the drugs studied. In general, it is considered that the use of parallel groups to calculate trough:peak ratio is suboptimal, largely because it is not possible to derive trough:peak ratio appropriately in individual patients. However, a method for constructing exact confidence intervals for trough:peak ratio from parallel, placebo-controlled clinical trials has been described (11), and this methodology can also be used to test hypotheses about the absolute value of the trough:peak ratio.
CLINICAL IMPLICATIONS OF A SATISFACTORY TROUGH:PEAK RATIO
Duration of action and blood pressure variability
The first and most important aspect of a satisfactory trough:peak ratio is that the drug is appropriate for its chosen dosage interval. Any value that is less than 100% implies that there is some reduction in the magnitude of the antihypertensive effect towards the end of the dosage interval, but if the trough:peak ratio exceeds 65% this implies that the diminution in effect is not unduly pronounced in the average patient and that the drug will provide a relatively consistent effect throughout 24 h. This consistency of the antihypertensive effect not only is important for ensuring the desired reduction of blood pressure during long-term treatment but also has practical consequences for routine management. Patients typically attend for blood pressure measurement and treatment review at some time during the regular working day, and this is likely to be several hours after the time of morning dosing. Therefore, with most patients, the blood pressure assessment in the clinic is made at or around the time of peak antihypertensive response. Therefore, if the drug has a suboptimal trough:peak ratio, an apparently satisfactory measurement may conceal the fact that towards the end of the dosage interval blood pressure control is lost. Conversely, if the drug is known to have a satisfactory trough:peak ratio, then the routine blood pressure measurement will provide a reasonable index of the level of blood pressure control and the prescribing physician can therefore be reasonably confident of a consistent antihypertensive effect throughout the full 24 h in the absence of dosing-related fluctuations in blood pressure and the associated increase in blood pressure variability.
Effect of dose on trough:peak ratio
The ideal antihypertensive agent should produce a consistently high trough:peak ratio over the recommended therapeutic dose range. This has not invariably been the case and, for example, a study of the angiotensin-converting enzyme (ACE) inhibitor lisinopril demonstrated that the trough:peak ratio was dose-dependent (Fig. 2A)(12). It is apparent that at the lower doses the trough:peak ratio was less than 30%, but the ratio was improved to an acceptable level of greater than 50% at relatively high doses of 20 mg or greater. This characteristic is clearly undesirable because it demands the use of higher doses simply to extend duration of action and, in addition, it also creates a potential problem in therapeutic practice. For example, at a dose of 10 mg of lisinopril once daily, the blood pressure fall in the clinic, which is measured around the time of peak response, may be entirely satisfactory and the assumption may well be made that this is a reflection of the blood pressure control throughout the 24-h period, which clearly is unlikely to be so. It should not, however, be assumed that this phenomenon will be exhibited by all drugs in a given class of antihypertensive agent and it is almost certainly a reflection of the relationship between antihypertensive effect and drug concentration/dose (13). Thus, in contrast to lisinopril, the ACE inhibitor trandolapril shows a clear dose-response relationship for both trough and peak blood pressure responses but does not exhibit dose dependence in trough:peak ratio at a range of doses below and through the therapeutic range (Fig. 2B)(14).
It is well recognized that hypertensive patients are often poorly compliant with treatment, either by being late in taking their morning dose or by omitting doses entirely (15). An antihypertensive agent with a high trough:peak ratio that is dependent on the intrinsic properties of the drug rather than on its formulation will, in theory, retain some residual effectiveness despite a belated or missed dose and will therefore retain much of its effectiveness as antihypertensive prophylaxis despite the patient's failings. These theoretic considerations have been supported by evidence from a series of studies examining the control of blood pressure beyond the dosage interval, after the deliberate insertion of a placebo dose to mimic the dosage omission of poor compliance (16-18). The findings of a study in which enalapril and trandolapril (18) were compared are summarized in Fig. 3. It is clear that the two agents are disparate, not only in their ability to control blood pressure over the dosage interval, in which trandolapril has a significantly higher trough:peak ratio, but also in their ability to control blood pressure beyond the end of the dosage interval, such that 48 h post dose enalapril shows little residual effect in terms of lowering blood pressure, whereas the diminution in effect with trandolapril is much less marked and a clinically useful blood pressure reduction is still apparent 48 h post dose.
TROUGH:PEAK RATIO AND COMBINATION THERAPY
The clinical benefits of combination drug regimens are well recognized and, indeed, a substantial proportion of hypertensive patients require combination treatment to achieve blood pressure control. Although the trough:peak ratios and duration of action of the individual components of combination regimens may be known, there is a relative paucity of information about the trough:peak ratio with different combination regimens. From first principles, a number of theoretical scenarios may arise. First, in the unlikely event that the two components of the combined product exhibit their peak effects at or around the same time and the effects are additive, it would be anticipated that the net trough:peak ratio would be a simple arithmetical average derived from the two components. However, in practice this is improbable, and it is more likely that the two components would not elicit a peak response at the same time and in that, on the basis of established concentration-effect relationships, the effects would not be additive, particularly at peak. With either of these eventualities, it would be anticipated that the trough:peak ratio of the combination product would be greater than that of either of the single components. In practice, this theoretical consideration appears to hold good, and this has been demonstrated with the combination of losartan and hydrochlorothiazide (19). The same pattern has been observed in the evaluation of the combination of verapamil SR (180 mg) and trandolapril (1 mg) (20). The placebo-corrected trough:peak values are summarized in Fig. 4, and it is apparent that the combination product not only enhances the absolute blood pressure response but also has a higher trough:peak ratio than either of the individual components.
The available epidemiologic data support the argument that optimal approaches in the treatment of hypertension depend primarily on achieving a reduction in blood pressure that is smooth and consistent over a full 24-h period. In assessing which drug and drug regimens will achieve such a therapeutic goal, the prescribing physician is faced with a wide range of options with regard to antihypertensive drugs. The evidence clearly indicates that the use of an appropriately characterized trough:peak ratio provides useful information that is clinically relevant as an index of the likelihood of producing full and consistent 24-h blood pressure control. A "satisfactory" trough:peak ratio indicates that the duration of antihypertensive effect is appropriate for the chosen dosage interval and further indicates that the control of blood pressure will be consistently maintained and that there will be some maintenance of antihypertensive effect beyond the dosage interval in the eventuality that the patient is irregularly compliant. Many established once-daily antihypertensive treatment regimens are suboptimal, either because they do not meet the minimal requirement of a trough:peak ratio in excess of 50% or because they do so only by virtue of increments in dosage. Although these contentions remain to be confirmed in prospective studies, it appears likely that drug treatments that can consistently produced trough:peak ratios in excess of 70% throughout the recommended dosage range are to be preferred.
1. Meredith PA, Perloff D, Mancia G, Pickering T. Blood pressure variability and its implications for antihypertensive therapy. Blood Pressure
2. Mancia G, Di Rienzo M, Parati G. Ambulatory blood pressure monitoring: use in hypertension research and clinical practice. Hypertension
3. Perloff D, Sokolow M, Cowsan R. The prognostic value of ambulatory blood pressures. JAMA
4. Frattola A, Parati G, Cuspidi C, Albini F, Mancia G. Prognostic value of 24-hour blood pressure variability. J Hypertens
5. Verdecchia P, Porcellati C, Schillaci G, et al. Ambulatory blood pressure: an independent predictor of prognosis in essential hypertension. Hypertension
6. Mancia G, Zanchetti A, Agebiti-Rosei E, et al., for the SAMPLE Study Group. Ambulatory blood pressure is superior to clinic blood pressure in predicting treatment-induced regression of left ventricular hypertrophy. Circulation
7. Rose M, McMahon FG. Some problems with antihypertensive drug studies in the context of the new guidelines. Am J Hypertens
8. Omboni S, Parati G, Zanchetti A, Mancia G. Calculation of trough to peak ratio of antihypertensive treatment from ambulatory blood pressure: methodological aspects. J Hypertens
9. Elliott HL, Meredith PA. Methodological considerations in calculation of the trough:peak ratio. J Hypertens
10. Gama G, Santos A, Polónia J. Measurement of trough-to-peak ratios of four antihypertensive drugs on the basis of 24 hour ambulatory blood pressure monitoring: different methods may give different results. J Hum Hypertens
11. Stewart WH, Hafner KB. Statistical analysis of trough/peak ratios. Drug Inform J
12. Ménard J, Bellet M, Brunner HR. Clinical development of antihypertensive drugs: can we perform better? In: Laragh JH, Brenner BM, eds. Hypertension: pathophysiology, diagnosis and management.
1st ed. New York: Raven Press, 1990:2331-50.
13. Meredith PA, Elliott HL. Antihypertensive treatment and trough:peak ratio: general considerations. J Hypertens
14. Meredith PA. Importance of trough:peak ratios in treatment of blood pressure in post-myocardial infarction and heart failure settings. J Hum Hypertens
15. Magometschnigg D. Compliance von Hypertonikern in der ärzlichen Praxis. In: Hitzenberger G, ed. 19. Symposium Teil 2: compliance. Österreich Arbeitsgemeinsch Klin Pharmakol.
Wien: Blackwell Wissenschafts-Verlag, 1995:47-54.
16. Johnston BF, Whelton A. A study design for comparing the effect of missing daily doses of antihypertensive drugs. Am J Ther
17. Hernandez-Hernandez R, de Hernandez MJ, Armas-Padilla MC, Carvajal AR, Guerro-Pajuelo J. The effects of missing a dose of enalapril versus amlodipine on ambulatory blood pressure. Blood Pressure Monit
18. Vaur L, Dutrey-Dupagne C, Boussac J, et al. Differential effects of a missed dose of trandolapril and enalapril on blood pressure control in hypertensive patients. J Cardiovasc Pharmacol
19. Elliott HL, Meredith PA. Clinical implications of the trough:peak ratio. Blood Pressure Monit
20. Mancia G, Agabiti Rosei E, Benetti G, et al. The effects of verapamil SR, trandolapril, and their fixed combination on 24-h blood pressure. Am J Hypertens
Proceedings of satellite symposium of the 8th European Meeting on Hypertension June 13, 1997; Milan, Italy