Hypertension is well recognized for inflicting a considerable deleterious burden upon target organs such as the brain, heart and kidney. There is a clear association between the severity and duration of high blood pressure and changes in the structure of the left ventricle and blood vessels throughout the circulation. Indeed, left ventricular hypertrophy and an increased wall–lumen ratio of small arteries are both strong predictors of increased mortality and morbidity from coronary heart disease and stroke [1,2]. Consequently, there has been intense interest in how homeostatic mechanisms in the circulation may be adversely influenced by an individual's cardiovascular risk profile. The key to understanding why one hypertensive patient might endure a series of vascular events when another with an identical risk profile remains symptom-free becomes paramount when trying to rationalize our management programmes to target the ‘at risk’ individual. The ultimate arbiter of blood flow to target organs must reside in the microcirculation. It is widely accepted that the structure and function of blood vessels less than 75 μm in internal diameter is hugely influenced by the operating conditions in the circulation upstream [3]. In this issue of the Journal, the role of the microcirculation in hypertension is elegantly addressed by Strain et al. [4], and their previous work on larger vessels was also published earlier this year [5]. Both papers report intriguing results, with provocative interpretations. The large vessel data are the opposite of those found in African-Caribbeans measured a decade earlier [6], which may well reflect a changing profile of cardiovascular risk in this ethnic group. It is well recognized that people of Black African descent have an elevated prevalence and therefore incidence, of high blood pressure compared to any other ethnic group. It is commonly assumed, as by Strain et al. [4], that this equates to the blood pressure being more ‘severe'. In other words, the increased risk of vascular target organ damage, particularly among African-Americans, is not explained fully by higher blood pressures. However, that assumption is probably incorrect; for given levels of baseline blood pressure, the evidence of excess vascular damage is weak and likely only reflects the higher prevalence, longer duration, poorer detection and management in African-Americans [7–9]. For example in the largest study of its kind with properly measured initial (baseline) blood pressures and follow-up of the MRFIT screenees [10], rates of total and all cardiovascular mortality did not differ between ethnic groups in each category of blood pressure, were lower for coronary heart disease and increased only for the small number of strokes due to the top category of pressure where there were just many more African-Americans. Indeed, in further analysis of later follow-up, socio-economic differentials entirely accounted for apparent blood pressure-independent differences in target organ damage in that cohort [11]. In the study by Strain et al. [4], the references cited as supporting the contention of a higher risk for given baseline pressures were cross-sectional [12] or did not have measured blood pressures before renal impairment [13], as applies to other studies widely quoted in the field [14]. In Britain, although African-Caribbeans have had excess but recently declining rates of stroke compared to European-born people in Britain, their rates of coronary heart disease have been much lower, such that total vascular (which includes that from excess renal) mortality has persisted lower than the national average [15]. Those data are over a decade old but illustrate the major preventive opportunities provided by good detection and control of hypertension, particularly in this ethnic group. Against this background, in an attempt to examine whether there are pressure-independent ethnic differences in the microvasculature that may contribute to this excess morbidity, Strain et al. [4] used Doppler fluximetry to assess the microcirculation of the skin in African-Caribbean and European men and women. The flux in response to maximal heating as a measure of maximal hyperaemic response and minimal vascular resistance was calculated. Intraventricular septal (IVS) thickness was assessed by echocardiogram and was regarded as an indication of concomitant target organ damage. A large number of baseline characteristics were measured simultaneously. In African-Caribbean men, the maximum hyperaemic response was significantly reduced compared to Europeans, and there was also a significant increase in minimum vascular resistance. Only peak response to ischaemia was different in African-Caribbean women. Importantly, the ethnic difference could not be attributed to standard cardiovascular risk factors. For example, having controlled for differences in blood pressure, lipid profile and insulin resistance plus alterations in waist/hip ratio, there was still evidence of attenuated microvascular structure and function. Although this was a cross-sectional study, the authors suggest that these changes in the microcirculation may contribute to the apparently higher rates of target organ damage seen in such individuals. Because they had measures of aortic pulse wave velocity (PWV), as an index of large vessel function [5], it would have been interesting to know whether this PWV attenuated or ‘accounted for’ the ethnic difference in the microcirculation. The authors go on to conclude that there is a definitive change in microvascular structure and function in people of African origin generally, but the key issue is cause and effect. Such generalizations need to be very carefully analysed before that label becomes widespread. The usual finding of marked differences in average blood pressure between African-Caribbeans and Europeans was not so clear in this subsample of their larger community study. There was no difference in systolic or diastolic blood pressures, in resting values, nor in total mean ambulatory values in men; night-time ambulatory pressure was higher, if not significantly so. All these parameters, except for mean ambulatory pressure, were significantly higher in the women who were much more obese (8 kg more than Europeans of the same height), but in whom microvascular function was generally similar to the Europeans. Could preceding years of higher blood pressure cause these microvascular changes or be the result of them? When, almost 20 years ago, we reported similarity of blood pressures in African-Caribbean (not including West African whose diet and cultural background are quite different) and European factory workers in this journal [16], selection bias for a healthier workforce was the probable explanation. Here, presumably, there was a similar explanation for the lack of difference in men, in whom the microcirculatory changes were most marked. Their current results are from a 50% response rate to a previous 60% response after making contact from General Practitioner lists in the larger community survey, making an overall sample of just over 30%. Thus, this sample is highly unlikely to be representative, and generalized statements of ‘impaired microvascular function’ in this ethnic group are as yet untenable. Furthermore, the authors state that the reproducibility of their method was 10% in one individual studied over 8 years and 8.2% in individuals studied over shorter time periods. These reasonable reproducibility results may again have some bearing on the findings. Could obesity per se confound their microvascular result? This is unlikely because differences were only found in men and not in the much more obese women, with the exception of the peak response to ischaemia. However, the cause or origin of the higher blood pressure itself in the women cannot be due to microvascular dysfunction: perhaps obesity ‘protects’ against such damage, making small vessel dysfunction more likely to be a consequence rather than a precursor of hypertension. Based on the cross-sectional data, the study by Strain et al. [4] cannot solve this issue. Open to further doubt is whether target organ damage is demonstrated by the greater IVS thickness, found only in African-Caribbean women who had the higher pressures. Again, only longitudinal results can answer this question. The reason for a lack of posterior wall data and not calculating left ventricular mass in the study by Strain et al. [4] is unclear. Here, the IVS differences were entirely accounted for by obesity. People of African origin have narrower chest diameters and, as demonstrated in previous data of the authors and many others, slightly larger IVS and left ventricular posterior wall thickness at any level of blood pressure. Studies in earlier life in African-Americans have found larger heart sizes and wall thickness than in European-Americans despite similar blood pressures [17]. Whether this can be called target organ damage is a mute point. The development of people of African origin appears to endow a larger heart size before ‘hypertension’ could lead to any secondary change. Despite these important caveats, the study by Strain et al. [4] raises the possibility that microcirculatory changes are associated with, and thus may either lead to or be the result of, both the apparent IVS hypertrophy and increased blood pressure. Sorting out cause or effect remains a major challenge. The report adds to a growing body of literature suggesting that, in individuals susceptible to target organ damage, there are vital changes in the structure and function of the blood vessels that feed them. In addition, the data offer the intriguing possibility that long-term follow-up of the patients with the greatest reductions in maximum hyperaemic response may yield evidence that these are the individuals who are particularly likely to go on to sustain vascular events. Rizzoni et al. [2] recently demonstrated that patients treated for a variety of causes of hypertension, and who present with the greatest morphological changes in small arteries, sustain the largest number of events. In stroke-prone rats, there is increasing evidence that the breakdown of myogenic responsiveness occurs shortly before cerebrovascular accidents are observed. Such effects of microcirculatory dysfunction on prognosis are within the context of the clear evidence from several separate sources indicating that large vessel stiffness also closely predicts mortality [6,18]. Consequently, we should redouble our efforts to understand the processes that control autoregulation of blood flow to vital organs. There is clear evidence that, in diabetes, the breakdown of these mechanisms leads to a rise in capillary pressure and that this is associated with end-organ damage [19]. Given the close association between diabetes and hypertension, plus the well-recognized increase in mortality and morbidity in diabetic hypertension, efforts directed towards understanding the processes that influence microvascular structure and function, and how these become deranged, may well become the bedrock around which we identify at-risk individuals in the future, target aggressive risk reduction therapies and develop novel treatment modalities for susceptible individuals.