The advantages of ambulatory blood pressure measurement (ABPM) over traditional office measurement (OBP) for the management of hypertensive patients have long been recognized [1–3]. ABPM has been used successfully to identify hypertensive patients at increased risk of developing cardiovascular diseases [1–3]. In addition, the clinical use of ABPM has enabled the identification of parameters other than the 24-h average that might be useful for improving the prognostic precision [1–3]. Several prospective studies have proved the superiority of ABPM to clinic blood pressure (BP) in predicting outcome whereas no study has shown the reverse [2,3]. In this respect, the results of the meta-analysis by Conen and Bamberg  published in this issue of Journal of Hypertension come to no surprise. Figure 2 of their article indicates total agreement among all studies found to be pertinent according to the authors' selection criteria.
A more relevant issue, that is dealt with by the authors in their study is whether ABPM provides prognostic information on top of OBP. International guidelines emphasize that OBP should remain the standard of measurement on which most medication interventions should be based, although they acknowledge that ABPM may be beneficial in certain situations by providing important information not available from OBP . Thus, according to those guidelines ABPM should supplement but not substitute office measurements. Unfortunately, Conen and Bamberg  could identify only four studies in which the association between 24-h blood pressure (BP) and cardiovascular events was assessed after adjusting for OBP. Although all four studies found that ABPM remained a significant predictor of outcome in multivariable analyses which included OBP, there is a series of methodological problems about these studies which should be given due consideration [6–9]. One crucial problem is that in all studies OBP was measured on only one or a few visits. In the study by Staessen et al., six BP readings were collected over three visits performed 1 month apart, but in the other three studies [7–9] BP was measured only on a single visit. In the study by Clement et al. three BP readings were averaged, and in the studies by Bjorklund et al. and by Ohkubo et al., only two readings were used. It is then scarcely surprising that the mean of several dozens of ambulatory readings predicts cardiovascular events better than does the mean of two to six OBP measurements, and it is hard to say whether the improvement in the prediction obtained with ABPM should be attributed to the peculiarity of the ambulatory measurement or to the much larger number of BP readings. International guidelines recommend that treatment should be deferred for several weeks or months in subjects with high–normal BP or stage one hypertension and that, during this time, the decision on whether to start antihypertensive treatment should be based on multiple OBP measurements . The results obtained within the frame of the hypertensive and ambulatory recording Venetia (HARVEST) study highlight the clinical value of prolonged assessment with traditional OBP measurement . We showed that when ABPM was compared with the mean of six OBP readings taken during two baseline visits, the prediction of development of sustained hypertension during the following years was greater for ABPM than OBP, but when the mean of 18 OBP readings obtained over 6 months of observation was taken into account, the prediction was greater for OBP and ABPM provided only marginal additional prognostic information to OBP . Similar results were obtained in a cross-sectional analysis by Fagard et al. who demonstrated that ambulatory BP did not explain the variance in left ventricular mass, in addition to the variance explained by OBP, if the mean of 10 OBP readings taken during two visits was used. In a sample of pediatric patients with treated hypertension and chronic kidney disease, white-coat hypertension showed a prevalence of 17% if only a few OBP readings were used . However, the prevalence of white-coat hypertension declined to 6.3% and the specificity of OBP in detecting ambulatory hypertension increased from 70 to 93% when several clinical readings were averaged . The results of the above studies indicate that by increasing the number of office readings, a prediction even greater than that provided by ABPM can be reached and that the average of several OBP measurements obtained during different days may provide a more precise prognostic information than a larger number of readings obtained during only one day. On the other hand, in their article Conen and Bamberg  themselves point out that the evidence provided by the four longitudinal studies included in their meta-analysis indicates that ABPM is useful in predicting outcome if only a few OBP measurements are taken into account. Had the subjects enrolled in those studies been assessed with multiple readings taken in the office on multiple visits, different results could have been achieved.
In the four cohorts of Figure 3, to assess baseline ambulatory BP only one ABPM was used. Guidelines from the Working Group on BP monitoring of the European Society of Hypertension recommend that in subjects with white-coat hypertension, ABPM should be repeated within 3–6 months to confirm diagnosis . Due to the regression to the mean phenomenon, subjects with white-coat hypertension are likely to have an increase in ambulatory BP when ABPM is repeated, and this may occur in as many as 58% of the subjects initially identified as white-coat hypertensives . This indicates that if white-coat hypertension is diagnosed on the basis of a single ABPM, antihypertensive treatment can be denied to a large proportion of hypertensive patients. None of the four above studies used the suggested approach.
Self measurements of BP at home (HBPM) have been advocated in place of ABPM for the assessment of the hypertensive patient  and cross-sectional  as well as longitudinal  studies have suggested that the association of HBPM with outcome may be as good as that of ABPM. However, the validity of HBPM relative to ABPM and routine OBP has not been assessed systematically in patients with hypertension. Clearly, the cost-effectiveness of ABPM in clinical practice should be established in studies in which also HBPM is available because self-BP measurement is inexpensive and readily available to almost all patients . On the other hand, Sega et al. recently found in the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA) population, that the overall ability to predict death was not greater for ambulatory or home BP than for OBP and that it was increased by the combination of office and out-of-office measurement.
Another important methodological problem that may hamper generalizability of Conen and Bamberg's data  is the hetereogenity of the four samples considered for their meta-analysis. Two studies included only elderly subjects [6,7], and the other two studies elderly as well as younger subjects [8,9]. A more important confounding factor, as recently shown by a cohort study encompassing a large number of subjects , is represented by antihypertensive treatment. The study by Staessen et al. was performed in patients with systolic hypertension untreated at baseline assessment. During the follow-up, 50% of the patients were treated and 50% remained untreated . The study by Clement et al. was made in treated subjects with diastolic hypertension, and the studies by Bjorklund et al. and by Ohkubo et al. in a mix of normotensive and treated or untreated hypertensive subjects. The main goals of ABPM are
- to assess whether in an untreated subject antihypertensive treatment should be started
- to verify whether in a treated subject blood pressure is well controlled. Results from populations with different characteristics and in which different goals are pursued can hardly be put together. The results obtained in the Systolic Hypertension in Europe (Syst-Eur)  study are particularly relevant to this point. They showed that in the untreated arm of the study ABPM was an independent predictor of stroke after OBP was taken into account. In contrast, in the treated arm the predictive power of baseline ABPM for outcome disappeared after adjustment in multivariable regression and only OBP remained an independent predictor of total mortality . A limitation of the Syst-Eur study is that ABPM was not used to document treatment-induced BP changes during the follow-up that might have been more informative than those of OBP.
Another limitation of Conen and Bamberg's study,  acknowledged by the authors, is that the meta-analysis was restricted to systolic BP. Diastolic BP was not tested because in most ABPM studies, effect estimates were not provided for diastolic BP due to the fact that they did not reach statistical significance. The authors' conclusions cannot thus be extrapolated to diastolic BP.
Controversy still remains as to what ABPM parameter should be used to diagnose hypertension or to define BP control. That is to say, which ABPM value is the best predictor of cardiovascular risk. One of the main advantages of ABPM is that it allows the measurement of BP during sleep. A preponderance of results seems to indicate that nocturnal BP and the day/night difference are a key element of ABPM [6,20,21]. However, the comparative prognostic value of night-time and daytime blood pressure is still a matter of certain debate. Recent analysis in five populations provided greater value to night-time BP, mainly when both (daytime and night-time values) were included simultaneously in the predicting models . Table 2 of Conen and Bamberg's  meta-analysis shows higher hazard ratios for night-time BP, except for combined cardiovascular endpoint. However, this potentially relevant issue is not addressed to full extent in this meta-analysis and whether night-time BP is a true risk factor for mortality or it is merely a marker of poor outcome remains a matter of debate.
Despite the increased use of ABPM in studies over the last 20 years, it still generates much controversy . Expert working population has not universally endorsed ABPM for use in the adult population, outside of diagnosing white-coat hypertension. When a new diagnostic procedure becomes available, the most efficient match between the new technology and the patients' need must be determined. Given the relatively high cost of routine ABPM application, advocating screening requires cost-effectiveness studies, which are largely missing. No trial has been designed to investigate whether a strategy based on ABPM is more cost-effective than a strategy based on multiple OBP and home BP readings. Rather than reiterating that ABPM has a great potential it may be timely to identify the key issues that remain unsolved and focus the research efforts accordingly. Before ABPM can be applied to the large majority of prehypertensive and hypertensive subjects, prospective studies should be implemented using OBP, HBPM, and ABPM combined. More evidence is needed to clarify whether HBPM and ABPM can be regarded as mutually interchangeable, or whether they provide complementary, nonsuperimposable information . Diagnostic algorithms and intervention strategies based on different combinations of the above methods of measurement should be tested. Studies should be addressed also to assess the longitudinal relationship between the loss of the diurnal BP rhythm and cardiovascular morbidity and to determine whether the restoration of a normal BP fall at night actually improves prognosis.
1 Zanchetti A. The role of ambulatory blood pressure monitoring in clinical practice. Am J Hypertens 1997; 10:1069–1080.
2 White WB. Ambulatory blood pressure monitoring in clinical practice. N Engl J Med 2003; 348:2377–2378.
3 Mancia G, Parati G. Guiding antihypertensive treatment decisions using ambulatory blood pressure monitoring. Curr Hypertens Rep 2006; 8:330–337.
4 Conen D, Bamberg F. Noninvasive 24-h ambulatory blood pressure and cardiovascular disease: a systematic review and meta-analysis. J Hypertens 2008; 26:1290–1299.
5 Mancia G, De Backer G, Dominiczak A, Cifkova R, Fagard R, Germanò G, et al
. ESH-ESC Practice Guidelines for the Management of Arterial Hypertension: ESH-ESC Task Force on the Management of Arterial Hypertension. J Hypertens 2007; 25:1751–1762.
6 Staessen JA, Thijs L, Fagard R, O'Brien ET, Clement D, de Leeuw PW, et al
. Predicting cardiovascular risk using conventional and ambulatory blood pressure in older patients with systolic hypertension. Systolic Hypertension in Europe Trial Investigators. JAMA 1999; 282:539–546.
7 Bjorklund-Bodegard K, Lind L, Zethelius B, Andren B, Lithell H. Isolated ambulatory hypertension predicts cardiovascular morbidity in elderly men. Circulation 2003; 107:1297–1302.
8 Clement DL, De Buyzere ML, De Bacquer DA, de Leeuw PW, Duprez DA, Fagard RH, et al
. Office versus ambulatory pressure study investigators. Prognostic value of ambulatory blood-pressure recordings in patients with treated hypertension. N Engl J Med 2003; 348:2407–2415.
9 Ohkubo T, Kikuya K, Metoki H, Asayama K, Obara T, Hashimoto J, et al
. Prognosis of ‘masked’ hypertension and ‘white-coat’ hypertension detected by 24-h ambulatory blood pressure monitoring. 10-year follow-up from the Ohasama study. J Am Coll Cardiol 2005; 46:508–515.
10 Palatini P. Too much of a good thing? A critique of overemphasis on the use of ambulatory blood pressure monitoring in clinical practice. J Hypertens 2002; 20:1917–1923.
11 Fagard R, Staessen J, Thijs L. Prediction of cardiac structure and function by repeated clinic and ambulatory blood pressure. Hypertension 1997; 29:22–29.
12 Wuhl E, Hadtstein C, Mehls O, Schaefer F. Home, clinic, and ambulatory blood pressure monitoring in children with chronic renal failure. Pediatr Res 2004; 55:492–497.
13 O'Brien E, Asmar R, Beilin L, Imai Y, Mallion JM, Mancia G, et al
. European Society of Hypertension Working Group on Blood Pressure Monitoring. European Society of Hypertension recommendations for conventional, ambulatory and home blood pressure measurement. J Hypertens 2003; 21:821–848.
14 Palatini P, Dorigatti F, Roman E, Giovinazzo P, Piccolo D, De Venuto G, et al
. White-coat hypertension: a selection bias? HARVEST Study Investigators. Hypertension and Ambulatory Recording Venetia Study. J Hypertens 1998; 16:977–984.
15 Parati G, Hernandez-Hernandez R, Velasco M. Home blood pressure monitoring in general practice: expectations and concerns. J Hypertens 2006; 24:1699–1701.
16 Mule G, Caimi G, Cottone S, Nardi E, Andronico G, Piazza G, et al
. Value of home blood pressures as predictor of target organ damage in mild arterial hypertension. J Cardiovasc Risk 2002; 9:123–129.
17 Asayama K, Ohkubo T, Kikuya M, Obara T, Metoki H, Inoue R, et al
. Prediction of stroke by home ‘morning’ versus ‘evening’ blood pressure values: the Ohasama study. Hypertension 2006; 48:737–743.
18 Sega R, Facchetti R, Bombelli M, Cesana G, Corrao G, Grassi G, et al
. Prognostic value of ambulatory and home blood pressures compared with office blood pressure in the general population: follow-up results from the Pressioni Arteriose Monitorate e Loro Associazioni (PAMELA) study. Circulation 2005; 111:1777–1783.
19 Boggia J, Li Y, Thijs L, Hansen TW, Kikuya M, Bjorklund-Bodegard K, et al
. Prognostic accuracy of day versus night ambulatory blood pressure: a cohort study. Lancet 2007; 370:1219–1229.
20 Ohkubo T, Hozawa A, Yamaguchi J, Kikuya M, Ohmori K, Michimata M, et al
. Prognostic significance of the nocturnal decline in blood pressure in individuals with and without high 24-h blood pressure: the Ohasama study. J Hypertens 2002; 20:2183–2189.
21 Kario K, Pickering TG, Matsuo T, Hoshide S, Schwartz JE, Shimada K. Stroke prognosis and abnormal nocturnal blood pressure falls in older hypertensives. Hypertension 2001; 38:852–857.
22 Parati G, Valentini M. Do we need out-of-office blood pressure in every patient? Curr Opin Cardiol 2007; 22:321–328.