Are guidelines right to promote lifestyle interventions against hypertension?

From the JNC-VII guidelines [1], ‘Individuals who are normotensive at 55 years have a 90% lifetime risk of developing hypertension. Individuals with a systolic blood pressure (BP) of 120–139 mmHg or a diastolic BP of 80–89 mmHg should be considered as pre-hypertensive and require health-promoting lifestyle modifications to prevent cardiovascular disease'. The ESH–ESC guidelines [2] also recommend initiating lifestyle interventions in individuals with a blood pressure of 130–139 mmHg or 85–89 mmHg, or to carry on these interventions for 1 year even in those individuals with a higher blood pressure when risk is low. The emphasis put by these international guidelines on lifestyle interventions suggests a high potential for public health improvement, and may suffice to convince the naïve reader that they are based on a high level of evidence. Various justifications are used to advocate this strategy: (i) patients may prefer to delay or prevent starting drug treatment and (ii) lifestyle interventions (a) may have fewer adverse effects than drugs; (b) cost less; and (c) make the patient more active and responsible in his own health management. To go beyond that somewhat moral position, it is important to assess whether this strategy is justified by a favourable risk-to-benefit ratio, estimated with an appropriate level of evidence. A lack of solid evaluation Nicolson et al. [3] performed a systematic review of trials comparing drug and lifestyle interventions in hypertension. They identified five reports dealing with heterogeneous interventions that were of poor methodological quality, and which presented inconsistent results, each involving less than 100 people who were followed for less than 1 year without reporting cardiovascular outcome. The authors performed secondary analyses suggesting that dietary lifestyle interventions might be better than blood pressure-lowering drugs for lowering cholesterol levels. This suggestion suffers from some limitations. The secondary position of this intermediate criterion lowers the level of the associated evidence. If the potential of lifestyle interventions in reducing cholesterol is to be compared with that of drugs, the appropriate drugs to use may not be blood pressure-lowering drugs: the proposed answer is not directed towards the appropriate question. Lifestyle interventions and blood pressure: a difficult story Ebrahim and Smith [4] published a systematic review of randomized controlled trials of lifestyle interventions, and similarly concluded that their impact on blood pressure was weak, and likely overestimated in a set of clinical trials of poor methodological quality. Of note is a lack of information on the fewer adverse effects associated with lifestyle interventions. Psychological consequences, such as the disappointment of not reaching the goals assigned by intervention, may be associated with some significant counter-effects and result in a net negative risk-to-benefit balance, at least in some patients. Lifestyle interventions are heterogeneous, including salt restriction, increase in dietary fibre intake, weight reduction diets, stress management, alcohol reduction and exercise. The compliance of patients to lifestyle interventions may be different, more difficult to assess rapidly and conducting to a different withdrawal rate compared to drug intervention. By definition, their assessment is open, which leaves room for some potentially important biases (i.e. placebo effect, impact on the cardiovascular risk management by physician or patient himself, or on the measurement of the outcome criteria). Industrial sponsors (food or pharmaceutical firms) are not prone to promote such an evaluation, due to the low chance of getting their money back. By contrast, some lifestyle interventions may counterbalance the marketing strategies of the food industry, such as high salt levels in canned food to increase its palatability and attractiveness. Are guidelines appropriate? There appears to be a surprising paradox regarding the large systematic support given to lifestyle interventions, which are proposed to delay and/or replace drug prescription in individuals with moderate blood pressure increases when cardiovascular risk is low, and the lack of appropriate evidence on which this is based. The main explanation is probably their presumed benignity and scarcity of side-effects. The delivery of satisfactory information to patients would require large clinical trials to be conducted with relevant methodology, including the use of clinical criteria. The presence of an interaction between lifestyle interventions and a drug would be of great consequence because lifestyle interventions are usually not supposed to be stopped when drug treatment is recommended. In this view, complete factorial design appears to be the most suitable. These large trials should be sponsored by health research institutions, such as UK-MRC, French Inserm, US NIH or their equivalent. At present, simulations may be performed to determine plausible scenarios for the risk-to-benefit ratios associated with the choice of drug or lifestyle intervention, or both.