Blood pressure (BP) is a physiological entity characterized by significant variations over time. The extent of these variations was initially considered to be mostly a nuisance affecting the accuracy of individual's BP status assessment. However, over the years, BP variability (BPV) has been increasingly acknowledged as a phenomenon with important pathophysiological relevance, reflecting the efficiency of cardiovascular control mechanisms in modulating BP values, including both central neural and humoral ‘rhythms’ (e.g. circadian changes) and the response to environmental perturbations. More recently, based on the consideration that excessive BP oscillations may represent a stressful stimulus for the cardiovascular system, it has also been suggested that an increased BPV might be a risk factor for cardiovascular events, independent of mean BP levels. A large number of studies have indeed supported this possibility [1,2]. Although initially most studies focused on short-term BPV (within 24 h), subsequently the concept of BPV was extended to longer periods of time. Today a considerable amount of evidence is available showing that increased variability of BP values, whether assessed on a beat-by-beat basis, by 24-h monitoring, weekly through repeated home measurements or during and across clinic visits over months or years, independently predicts outcome [2–4].
Nonetheless, the pathophysiological bases underlying increased BPV are far from being adequately understood and the knowledge on BPV predictors is still incomplete . A number of interacting intrinsic and environmental factors may be involved, and to make things more complex, their relative contribution depends on the time frame in which BPV is assessed. As an example, alterations of autonomic neural control of circulation are more likely to be involved in short-term BPV, whereas impact of ambient temperature is particularly pronounced in long-term, seasonal BP changes. Moreover, also the extent to which BPV may independently contribute to outcome, on top of average BP levels, remains a matter of debate .
Chronic kidney disease (CKD) is a particularly interesting condition as far as BPV is concerned, for a number of reasons . One, CKD is associated with an impairment of kidney-related mechanisms controlling volemia and sodium excretion; as a consequence, less effective homeostatic buffering of volemia-related BP changes may lead to increased BPV. Two, CKD is independently associated with arterial aging characterized by an increased arterial stiffness which, in turn, was shown to correlate with increased BPV . Three, CKD is frequently associated with baroreflex impairment and enhanced sympathetic activity, both of which contribute to an increased short-term BPV. Four, CKD is common in diabetic patients in whom autonomic dysfunction may be particularly pronounced, further favouring BPV increase. Five, in patients with advanced CKD undergoing dialysis, important changes in volemia during the intradialytic and interdialytic period may translate into increased BPV. Six, choice of antihypertensive drugs could also play a role. Indeed, current evidence suggests that calcium antagonists could have a favourable impact on BPV but this group of drugs is usually not considered as the preferred choice for controlling hypertension in CKD. Finally, not only increased mean BP levels but also increased BPV might contribute to CKD development and progression, giving birth to a vicious circle in which hemodynamic alterations favour renal damage and vice versa .
A separate mention should be made of a particular feature of 24-h BPV, that is, the circadian BP variation normally characterized by the presence of nocturnal BP fall (dipping). Reduced (less than 10%), absent or reversed dipping is associated with worse outcome and is common among CKD patients, although the underlying mechanisms are far from being clear. Principal pathogenetic mechanisms potentially involved in the relationship between BPV and CKD are summarized in Fig. 1.
In the current issue of the Journal of Hypertension, Sarafidis et al.  report the results of a cross-sectional study on the relationship between BPV and CKD in a large group of patients from general practice and specialist centres included in the Spanish Ambulatory Blood Pressure Monitoring Registry. In this study, almost 40% of participants had CKD, which allowed the authors to analyse BPV across a wide spectrum of CKD severity with the notable exception of dialysed patients who were unlikely to participate. In all participants, ambulatory BP monitoring (ABPM) was available as well as information on renal function (serum creatinine) and urinary albumin excretion, which was used to classify patients according to CKD presence and severity. Short-term BPV was assessed by calculating standard deviation (SD) of daytime and night-time BP, weighted 24-h BP SD (wSD) and average real variability (ARV). Nocturnal BP-dipping pattern was also evaluated.
The main finding of this study is that the presence of CKD was generally associated with unfavourable characteristics of 24-h BP patterns including higher BPV and abnormalities of circadian profile (nondipping or reverse dipping). Moreover, these adverse BPV-related features were increasingly present with advancing CKD stages. Most of this information, in particular, the association of altered 24-h BP variations with the presence of CKD, confirms the data coming from several previously published studies discussed by the authors. The new information coming from this study regards mainly the finding of a progressive increase in BPV quantified by SD, wSD or ARV with worsening CKD severity. The observed differences persisted also after accounting for the association between mean BP and BPV (this was achieved through calculating coefficients of variation or adjusting for mean BP in multivariable models).
Even if the findings of the study are largely expected and partly confirmatory, they are nonetheless of interest for several reasons. One, the information was obtained in a large sample of individuals, representative of the population typically referred for the evaluation of 24-h BP in daily practice (in fact, the patients were included whenever the physician in charge considered the performance of ABPM clinically justified). This ‘clinical’ selection explains the elevated rate of CKD in this sample (the vast majority of participants were hypertensive and over 25% had diabetes). This allows the application of the findings to individuals in whom ABPM is normally performed, but it does not allow generalizing the results to, for instance, normotensive individuals with CKD.
A stimulating aspect of this study is related to the different data reported for systolic and diastolic BPV: the trend for BPV to increase across CKD stages was only evident for systolic BPV, whereas diastolic BPV, if anything tended to decrease. This was found after adjusting for mean BP levels, and thus cannot be explained by a similar pattern observed for mean ambulatory SBP and DBP (reflecting the aging-related stiffening of large arteries). This finding should be regarded in the light of the prognostic relevance of systolic and diastolic BPV. In fact, whereas in most studies on organ damage associations with systolic BPV were reported, in several large outcome studies diastolic BPV was the stronger predictor of events [9–11]. The interpretation of these differences is not an easy task but one could hypothesize that systolic BPV could be an epiphenomenon reflecting the aging of the cardiovascular system, whereas diastolic BPV may reflect more specifically the risk associated with derangement of cardiovascular control mechanisms (e.g. of autonomic control of circulation), although it is not clear whether it is merely a risk indicator or rather a true causal factor for cardiovascular complications.
Another interesting result of the study by Sarafidis et al.  is the exception to the overall trend (more severe CKD stage – higher systolic BPV) represented by patients with stage I CKD (abnormal albuminuria with normal glomerular filtration rate) in whom systolic BPV levels did not differ from those observed in individuals without CKD. Interestingly, this subgroup was slightly younger than non-CKD group and much younger than patients with more advanced CKD while sharing with the latter other adverse clinical characteristics including higher mean ambulatory systolic BP, higher rate of diabetes, preexisting cardiovascular disease, left ventricular hypertrophy and dyslipidaemia. The rate of active smokers was also much higher in this group compared with all other groups (with and without CKD). Taken together these data would suggest that aging plays a major role in the interaction between CKD and BPV, as supported by the results of multivariable analysis where age was among the strongest independent predictors of BPV in the CKD group.
The principal limitation of the study by Sarafidis et al.  is related to the cross-sectional study design. Although multivariable statistical analyses may limit the impact of confounding factors, it cannot fully exclude the possibility that some hidden or nonlinear relationships affected the results. As mentioned previously, generalizability of the results may also be an issue. Nevertheless, the present data from the Spanish ABPM Registry provide a valuable contribution to the state of the art in the field and may stimulate further research aimed at finding direct clinical applications of BPV assessment.
Conflicts of interest
There are no conflicts of interest.
1. Parati G, Pomidossi G, Albini F, Malaspina D, Mancia G. Relationship of 24-h blood pressure mean and variability to severity of target-organ damage in hypertension. J Hypertens
2. Parati G, Ochoa JE, Lombardi C, Bilo G. Assessment and management of blood-pressure variability. Nat Rev Cardiol
3. Webb AJS, Mazzucco S, Li L, Rothwell PM. Prognostic significance of blood pressure variability on beat-to-beat monitoring after transient ischemic attack and stroke. Stroke
4. Stevens SL, Wood S, Koshiaris C, Law K, Glasziou P, Stevens RJ, McManus RJ. Blood pressure variability and cardiovascular disease: systematic review and meta-analysis. BMJ
5. Parati G, Ochoa JE, Bilo G, Agarwal R, Covic A, Dekker FW, et al. European Renal and Cardiovascular Medicine (EURECA-m) working group of the European Renal Association-European Dialysis Transplantation Association (ERA-EDTA). Hypertension in chronic kidney disease part 2: role of ambulatory and home blood pressure monitoring for assessing alterations in blood pressure variability and blood pressure profiles. Hypertension
6. Schillaci G, Bilo G, Pucci G, Laurent S, MacQuin-Mavier I, Boutouyrie P, et al. Relationship between short-term blood pressure variability and large-artery stiffness in human hypertension: findings from 2 large databases. Hypertension
7. Chia YC, Lim HM, Ching SM. Long-term visit-to-visit blood pressure variability and renal function decline in patients with hypertension over 15 years. J Am Heart Assoc
8. Sarafidis PA, Ruilope LM, Loutradis C, Gorostidi M, de la Sierra A, de la Cruz JJ, et al. Blood pressure variability increases with advancing chronic kidney disease stage: a cross-sectional analysis of 16 546 hypertensive patients. J Hypertens
9. Hansen TW, Thijs L, Li Y, Boggia J, Kikuya M, Bjorklund-Bodegard K, et al. International Database on Ambulatory Blood Pressure in Relation to Cardiovascular Outcomes Investigators. Prognostic value of reading-to-reading blood pressure variability over 24 hours in 8938 subjects from 11 populations. Hypertension
10. Palatini P, Reboldi G, Beilin LJ, Casiglia E, Eguchi K, Imai Y, et al. Added predictive value of night-time blood pressure variability for cardiovascular events and mortality: the Ambulatory Blood Pressure-International Study. Hypertension
11. Mancia G, Bombelli M, Facchetti R, Madotto F, Corrao G, Trevano FQ, et al. Long-term prognostic value of blood pressure variability in the general population: results of the Pressioni Arteriose Monitorate e Loro Associazioni Study. Hypertension