Lifestyle Modification and Blood Pressure Control: Title and subTitle BreakIs the Glass Half Full or Half Empty?

The idea that lifestyle modifications can ameliorate mild hypertension has been popular for many years and was given a huge boost by the publications of results obtained with the Dietary Approaches to Stop Hypertension (DASH) diet.^{1 - 2} At last, there was convincing evidence that a nondrug form of treatment could reduce blood pressure (averaging decreases of 11.4/5.5 mm Hg in hypertensive patients and 5.5/3.0 mm Hg in those with borderline blood pressure) as much as some drugs.¹ A second study (DASH Sodium) combined the DASH diet with sodium restriction and reported a blood pressure reduction of 5.9/2.9 mm Hg with the DASH diet in patients with borderline hypertension and 8.9/4.5 mm Hg with a combination of the DASH diet and sodium restriction.² In a subsequent study, the Diet, Exercise, and Weight Loss-Intervention Trial (DEW-IT),³ overweight hypertensive individuals who had been treated with drugs reported a net change of 9.5/5.3 mm Hg in ambulatory pressure and 7.4/5.7 mm Hg in clinic pressure using a low-calorie version of the DASH diet in combination with weight loss.

Now, a fourth study (PREMIER), reported in this issue of THE JOURNAL by the Writing Group of the PREMIER Collaborative Research Group,⁴ has investigated the effects of combining the DASH diet with "established" recommendations, comprising weight loss, exercise, and restriction of sodium and alcohol. Since all these interventions have been shown to lower blood pressure individually, it might be anticipated that the effects of combined interventions on blood pressure would be additive. In actuality, the net changes (blood pressure change from baseline to 6 months in each treatment group minus that in the advice only group) were smaller than expected. In the group assigned to lifestyle modification only, the established group, the mean net reduction in blood pressure was 3.7/1.7 mm Hg, whereas for the group that followed the established recommendations together with the DASH diet, the net mean blood pressure was reduced by 4.3/2.6 mm Hg. Thus, the addition of the DASH diet in the PREMIER trial produced an incremental decrease of blood pressure of only 0.6/0.9 mm Hg (1.7/1.6 mm Hg in the individuals with hypertension). What happened?

Several explanations need to be considered. Patients in the PREMIER study were not following the DASH diet as closely as in the 3 earlier DASH studies, in which the participants were provided with prepared meals. PREMIER is the first study to intentionally investigate the effects of the DASH diet when patients actually purchased their own food. Thus, in the original DASH study, patients were provided with an average of 9.6 daily servings of fruits and vegetables, whereas in PREMIER, the intake increased from 4.8 servings at baseline to 7.8 servings. In the original DASH study, urine potassium increased by 105% with the change of diet, whereas in the combined group in the PREMIER study, it increased by only 28%. The same explanation could account for the results of the DASH Sodium study, in which the effects of sodium restriction (a reduction of blood pressure of 6.7/3.5 mm Hg) greatly exceeded the changes observed in almost all other studies in which participants prepared their own low-salt meals.

Another possibility is that the beneficial effects of the interventions were masked by parallel reductions of blood pressure in the control group. In the PREMIER study, the control group was given advice on diet and lifestyle changes in a single 30-minute session, but they showed a fairly large absolute decrease in blood pressure (absolute change from baseline to 6 months of 6.6/3.8 mm Hg, compared with absolute changes of 10.5/5.5 mm Hg in the established group and 11.1/6.4 mm Hg in the established plus DASH group). In the original DASH study, patients consuming the control diet showed no significant change of blood pressure.¹ Likewise, in the DEW-IT study, the control group showed no change of ambulatory blood pressure.³ However, potential similarities in diet and lifestyle would not explain the lack of effect of adding the DASH diet.

Yet another potential explanation is what the authors term "subadditivity of intervention effects," ie, the combination of 2 or more interventions has a smaller effect on blood pressure than the sum of the effects of the individual interventions. In the DEW-IT study, in which obese hypertensive patients were given a hypocaloric DASH diet, the net reduction of clinic blood pressure was 7.4/5.7 mm Hg, comparable to the effects of the DASH diet alone. However, these patients lost 5.5 kg of body weight, which might have been expected to result in a further blood pressure decrease of around 6/5.5 mm Hg (the decrease reported in the Trials of Hypertension II [TOHP II] study, in which the weight loss was 4.4 kg).⁵ In the TOHP II study, the effect of adding sodium restriction to weight loss produced no further decrease of blood pressure, even though sodium restriction alone produced a modest but significant decrease. Similarly, in the DASH Sodium study, the combined effects of the DASH diet and sodium restriction were less than the effects of either intervention on its own.² So far, therefore, no study has demonstrated any additive effect of combining lifestyle interventions on blood pressure.

Why should this be so? Two possibilities must be considered. First, it may be that most individuals are not capable of changing more than 1 lifestyle factor at a time. This explanation does not hold up, however. The TOHP II study had 3 groups: weight loss, sodium restriction, and both in combination. The decreases of body weight and sodium excretion were only marginally less in the combined group than in the 2 individual intervention groups, but the reduction of blood pressure was no greater than with weight loss alone.⁶ Since the DASH Sodium trial was a feeding study it seems unlikely that the participants were not consuming the diets they were prescribed, an observation verified by measuring urinary excretion of key minerals.

A second possibility is that the different lifestyle interventions may act through the same physiological mechanism, and that as with antihypertensive drugs, the dose-response relationship is nonlinear. For most drugs, doubling the dose produces only a small further decrement of blood pressure.⁷ In contrast, combining 2 different drugs often has an additive effect,⁸ unless the 2 drugs have a similar mechanism of action. Thus, adding a diuretic to the treatment of patients whose blood pressure is not controlled with a combination of amlodipine and lisinopril is more effective than adding a β-blocker.⁹ Unfortunately, relatively little is known about the mechanisms by which lifestyle factors such as obesity increase blood pressure.^{10 - 11}

Perhaps the most important issue is how these new findings should be interpreted and put into practice. While it is often stated that nondrug treatment is less expensive and has fewer adverse effects than drug treatment, this is not necessarily the case. Interventions such as those studied in the PREMIER trial require numerous counseling sessions to achieve their results and are not feasible in everyday practice. Nevertheless, for whatever reason, the advice only group in the PREMIER trial did have a substantial decrease in blood pressure. Physicians should certainly continue to counsel patients using the type of advice provided to the advice only group of the PREMIER study and advocated in the sixth report of the Joint National Committee on Detection, Evaluation, and Treatment of High Blood Pressure (JNC VI).¹² Beyond this, it is worth focusing on weight, the risk factor most likely contributing most to the patient's level of risk. Weight loss has many benefits for obese and overweight individuals other than lowering blood pressure. For normal-weight individuals with hypertension, the DASH diet also may be beneficial.¹³

Based on studies such as DASH, DEW-IT, and now PREMIER, clinicians should consider that the glass (or perhaps the plate) is indeed "half-full" when it comes to the potential benefit of promoting lifestyle changes for patients with hypertension.