Prehypertension in China

The paper by Gu et al. ^[1] published in this issue of the journal is a major and scientifically very well documented confirmation of the public health impact of prehypertension. We now know that prehypertension is widely prevalent in China and that it is associated with excess cardiovascular mortality. The impact of prehypertension on cardiovascular morbidity and mortality is similar in north versus south and urban versus rural China. This is remarkable, as one has to assume that living conditions, habitual diets and social circumstances in these disparate communities are substantially different. This ought to give pause to scientists interested in behavioral and psychosocial causes of hypertension; although the evidence for these aspects of hypertension in individual patients is strong, there is something inherent in hypertension that overcomes environmental conditions and mandates a similar worldwide prevalence of the disease. In western countries, being overweight is one of the strongest correlates of hypertension. In the Tecumseh study ^[2], compared with normotensive individuals, there was 16.5% excess of being overweight in people with prehypertension. That excess in China was only 5%, but this did not preclude a wide prevalence of prehypertension.

The impact of prehypertension on cardiovascular outcomes in China is impressive and not dissimilar to that observed in the USA. Admittedly, comparisons of different investigations have serious limitations, but it is worth noting that the long-term MRFIT follow-up study in the USA ^[3] and the study by Gu et al. ^[1] produced similar ‘ballpark’ figures for coronary heart disease and stroke mortality. The Chinese study used the current JNC 7 classification of prehypertension, which includes both the ‘high normal’ (130–139/85–89 mmHg) and the ‘normal but not optimal’ (120–129/80–84 mmHg) categories, and the adjusted relative risk of coronary heart disease for this combined group was 1.47. In the MRFIT report, the relative risk of the ‘high normal’ group was 1.66 and that of the normal but not optimal group was 1.27; thus, the relative risk of the combined group in China falls in the middle of the two components groups shown in MRFIT. The relative risk of death due to strokes was higher in both studies than that due to coronary heart disease. However, the relative risk in China (1.67) was somewhat lesser than in the USA (2.14 for high normal and 1.73 for the non optimal group).

Gu et al. ^[1] decided to further accentuate the impact of prehypertension by calculating the number needed to be treated to prevent an event. Although this figure illustrates the problem in easier-to-understand terms, their calculation is a major departure from the usual usage of this term. The number needed to be treated is typically derived from actual data in intervention studies. Not having such data, the authors calculated a ‘what if’ variable that they term the optimal number to be treated; what would be the difference if all prehypertensive patients were treated and if the treatment were to bring the blood pressure (BP) into the normal range in each patient. We all know this will not happen; the observed rates of BP control in various countries are between 6% and a maximum of 40% of treated patients. The BP control rates for various BP categories are rarely reported, but it is logical to assume that they would be lesser in prehypertension than in other forms of hypertension. People who feel good and are considered to have only ‘a small’ BP increase are less likely to adhere to medication, physician enthusiasm for aggressive treatment without outcome data is likely to be lesser, and we frequently ignore the fact that the BP response to drugs is positively correlated with baseline values and is likely to be lesser in patients with lower baseline values. Thus, in real life, the number needed to be treated is likely to be much larger. I understand the intent is to illustrate the problem; but I doubt the utility of setting such an ideal goal. Personally, when someone asks me to do the impossible, I get depressed rather than motivated.

The authors are right to point out that, presently, we cannot recommend drug treatment of prehypertension and their suggestion to consider treatment in prehypertensive diabetic patients is well supported by their findings. If we were to treat all patients with prehypertension and hypertension, 58% of the general population in China ^[1] would receive life-long antihypertensive medication. However, reducing the number by treating only diabetic prehypertension may not be the way forward. The underlying idea with respect to the treatment of prehypertension comprises the exciting possibility that early treatment might prevent the evolution of future vascular disease. Although diabetic prehypertensive patients indeed might benefit from treatment, most of them already have microvascular and macrovascular disease. We must develop better predictors of future events and identify a high-risk subpopulation among persons with prehypertension. A good first step would be to focus only on prehypertensive patients in the 130–139 or 85–89 mmHg range and, among them, seek to identify those at high risk. Such patients could be selected for future controlled trials.

Everybody interested in prehypertension will wholeheartedly agree with Gu et al. ^[1] that ‘randomized clinical trials’ are needed, but will also note that their statement that such trials ‘will provide more definitive evidence’ is too optimistic. The ‘will’ means that such trials are in the offing but, to the best of my knowledge, nobody has yet succeeded in obtaining funding for such a study. We must sincerely thank the authors for their excellent paper and hope that their contribution will rekindle interest in outcome studies in prehypertension.