"How Should We Take Blood Pressure in Clinical Practice"
October 2002 BMJUSA Primary Care Medicine for the American Physician How should we take blood pressure in clinical practice? The wider use of home and ambulatory monitoring should be encouraged I t is increasingly clear that the traditional way of sure is persistently elevated. Furthermore, the white measuring blood pressure in the clinic or office coat effect, which is usually defined as the differ- Papers p 549 frequently produces numbers that grossly over- ence between the clinic and daytime ambulatory estimate a patient’s true blood pressure level. This is blood pressure, is present in the majority of hyper- a major problem, since it is one of the most impor- tensive patients. While these facts have generally tant and frequent measurements made by physi- been accepted in the research community, they cians. Two major trends have brought this issue to have not yet exerted much influence on everyday the forefront: first, the development of new tech- clinical practice. nologies for measuring blood pressure; and second, Two recent studies by Little et al, one reprinted the increasing body of evidence that even mild ele- in this issue (p 549), confirm that the white coat ef- vations of blood pressure are associated with in- fect is of substantial magnitude (19/11 mm Hg) in creased cardiovascular risk. the primary care setting4 and The traditional gold stan- that self-measurement of blood dard for evaluating blood pres- sure has been clinic readings Compared with home pressure atby patients, giving a preferred home is the method made by a physician using a and ambulatory much smaller white coat effect mercury sphygmomanometer. (5/6 mm Hg).5 The authors It is hallowed by time, and also readings, it is the concluded, “It is time to stop us- by the fact that it has been the physicians’ readings ing high blood pressure read- standard method for evaluating that are the ings documented by general the risks associated with high practitioners to make decisions blood pressure and the benefits odd man out. about treatment.”4 This is a of treating it. It has been known sweeping statement, and one for more than 50 years that the that will be resisted by many blood pressures recorded in the clinic are substan- physicians who instinctively believe that the read- tially higher than readings taken by the patient at ings that they take in the traditional way are inher- home,1 but this fact was largely ignored until the ently more trustworthy than ones taken with an advent of ambulatory blood pressure monitoring electronic gadget. (ABPM). In the past 20 years a series of publica- However, the hard truth is that whenever physi- tions have shown that cardiovascular risk is predict- cians’ readings are compared with home and ambu- ed better by ambulatory blood pressure than clinic latory readings, it is the physicians’ readings that are pressure.2 the odd man out.6 Given that their measurements This is not surprising, since it is generally as- tend to consistently overestimate a patient’s prevail- sumed that it is not the blood pressure recorded at a ing blood pressure level, some additional method is single point in time that causes damage, so much as clearly needed. One solution would be to advocate the average blood pressure. The main finding has the widespread use of ABPM, but there are other been that patients with “white coat” hypertension, possibilities. The most obvious is self-monitoring, who constitute approximately 20% of the popula- which is relatively cheap and convenient and less tion with mild hypertension,3 are at relatively low burdensome for patients than ABPM.7 Like ABPM, risk in comparison with patients whose blood pres- it can provide large numbers of readings and mini- BMJ USA VOLUME 2 OCTOBER 2002 541 Editorials mize the white coat effect (to allow for this, the with readings taken out of the office will improve “normal” limit of home blood pressure should be patient care. ✦ 135/85 mm Hg). So far, only one study has shown Competing interests: Dr. Pickering is a consultant for Lifeclinic.com, that home monitoring gives a better prediction of a subsidiary of Spacelabs, Inc. risk than clinic pressure,8 but it is particularly useful Thomas G Pickering professor of medicine for monitoring the response to treatment and gives Marie-Josée and Henry R Kravis Center for Cardiovascular Health better correlation with the regression of left ventric- Mount Sinai School of Medicine New York, NY 10029-6574 ular hypertrophy than does clinic pressure. Thomas.firstname.lastname@example.org A practical regimen for patients who present with high clinic pressures and in whom treatment decisions are unclear is as follows: If there is evi- 1. Ayman P, Goldshine AD. Blood pressure determinations by pa- tients with essential hypertension I. The difference between clinic dence of blood pressure–related target organ dam- and home readings before treatment. Am J Med Sci 1940;200:465- age, treatment is indicated. If such evidence is not 474. present, home readings may be helpful to confirm 2. Verdecchia P. Prognostic value of ambulatory blood pressure: cur- rent evidence and clinical implications. Hypertension 2000;35:844- an elevation. If these are high (above 135/85 mm 851. Hg), treatment is indicated. If they are normal and 3. Pickering TG. White coat hypertension. Curr Opin Nephrol Hypertens 1996;5:192-198. if there is a persistent discrepancy between clinic 4. Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, and home readings, ABPM may be useful to make Mant D. Comparison of agreement between different measures of the final decision. blood pressure in primary care and daytime ambulatory blood pressure. BMJ 2002;325:254. These considerations are not intended to mean 5. Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, that physicians should throw away their sphygmo- Mant D. Comparison of acceptability of and preferences for differ- manometers, but they should come to accept that ent methods of measuring blood pressure in primary care. BMJ 2002;325:258-259. they do not have a monopoly on accurate blood 6. Pickering TG, James GD. Some implications of the differences be- pressure measurement. Patients should be encour- tween home, clinic and ambulatory blood pressure in normoten- sive and hypertensive patients. J Hypertens Suppl 1989;7:S65-S72. aged to monitor their own blood pressure regularly, 7. Yarows SA. Home blood pressure monitoring in primary care. and their monitors should be checked for accuracy. Blood Press Monit 1998;3(suppl 1):S11-S17. In circumstances where small differences of blood 8. Imai Y, Ohkubo T, Tsuji I, Nagai K, Satoh H, Hisamichi S, Abe K. Prognostic value of ambulatory and home blood pressure measure- pressure may alter treatment decisions, more is bet- ments in comparison to screening blood pressure measurements: a ter, and supplementing physician measurements pilot study in Ohasama. Blood Press Monit 1996;1(suppl 2):S51-S58. Prevention and cure of type 2 diabetes Weight loss is the key to controlling the diabetes epidemic T ype 2 diabetes is reaching epidemic propor- associated with diabetes, and waist circumference, a tions, and epidemics are seldom controlled measure of intra-abdominal fat, is the strongest pre- unless their causes are addressed. Obesity is dictor of glucose tolerance. Similarly, obesity relat- strongly and causally linked to type 2 diabetes. ed diabetes in childhood is common worldwide.4 Recent data suggest that the prevention of diabetes is So, could we prevent type 2 diabetes? In a feasible if weight management is addressed ade- prospective study of 84 941 female nurses followed quately in individuals at high risk. More controver- for 16 years, a combination of five modifiable risk sially, weight management also has the potential to factors related to dietary behavior, physical activity, make a significant impact in those with established weight, and cigarette smoking was identified that type 2 diabetes. was associated with a remarkable 91% reduction in The most common definition of obesity is a the risk of developing diabetes.5 Even with a family body mass index greater than 30 kg/m2. In the history of diabetes the risk reduction was 88%. In Nurses’ Health Study the risk of type 2 diabetes in theory, therefore, most diabetes could be pre- women with an index of 29-31 was 28-fold increased ventable, largely irrespective of genetic back- compared with women with an index lower than 22, ground. and an index greater than 35 carried a 93-fold in- Two pioneering studies show that this is feasible. creased risk.1 In the Finnish diabetes prevention study weight loss The overall prevalence of self reported diabetes in overweight subjects with impaired glucose toler- in the United States has reached 7.3%, and 15% in ance, averaging just 3-4 kg over 4 years, led to a 58% This editorial originally people over 60 years of age, driven by epidemic obe- reduction in incident diabetes.6 A similar result was appeared in the August 3, 2002 issue of the sity,2 and is higher among ethnic groups.3 Obesity achieved by the diabetes prevention program in the BMJ (325:232–233). and physical inactivity are the principal risk factors United States, in which lifestyle intervention involv- 542 BMJ USA VOLUME 2 OCTOBER 2002