Inflammation and Arterial Stiffness in Chronic Kidney Disease: Cause or Consequence?

Chronic kidney disease (CKD), which is defined as the presence of (micro)albuminuria and/or a reduced glomerular filtration rate (GFR), constitutes a growing public health concern. Recent epidemiological studies have shown that kidney failure is becoming increasingly prevalent and emerging as a top-10 cause of death.1 In addition, mild-to-moderate (nondialysis-dependent) CKD as well as end-stage renal failure are known as strong risk factors of all-cause and cardiovascular mortality.2 Although the precise mechanism(s) responsible for the predisposition of CKD to cardiovascular disease are not fully elucidated yet, one of the leading candidate causes involves arterial stiffness. Previous studies have demonstrated that aortic stiffness, as measured by carotid-femoral pulse wave velocity (PWV), can predict total mortality and cardiovascular events in CKD.3 Also, aortic PWV is shown to be an independent determinant of GFR and albuminuria, as well as a strong predictor of progressive GFR decline.4–6 Moreover, aortic pulse pressure, which depends on PWV and peripheral wave reflection, widens markedly in CKD7 and correlates closely with the urinary albumin/creatinine ratio.5 All the evidence mentioned above indicates that aortic stiffness and pulsatile hemodynamics play an important role in the pathogenesis and prognosis of CKD. Stiffening of the aorta is usually attributed to gradual degeneration of the elastic media of the aortic wall with advancing age. Owing to the repetitive stretching of the aortic wall induced by the cyclic contraction of the left ventricle, the elastic fibers of the medial layers gradually undergo thinning, splitting, fraying, and fragmentation, and progressively lose the orderly arrangement normally seen in early adulthood.8 Such fracture and degeneration of the elastic fibers are closely linked with reactive increases in collagenous fibers and ground substances and is often accompanied by calcium deposition. The resultant replacement of elastin by collagen leads to structural stiffening of the aorta, because collagen is about 300 times as stiff as elastin. In addition to age, blood pressure contributes to aortic stiffening; acute elevation of blood pressure causes functional aortic stiffening on account of the nonlinear elastic properties, while chronic hypertension accelerates structural aortic stiffening since it increases circumferential stress on the aortic wall according to the law of Laplace. Furthermore, recent studies have been focusing their attention to chronic inflammation, postulating it as another pathogenic factor of arterial stiffening. Such postulation has been made on the 2 premises that (i) the inflammatory process is importantly involved in vascular remodeling, endothelial dysfunction, oxidative stress, and atherosclerosis,9 and (ii) all these pathologies appear to increase stiffening of the aorta and major arteries.10 In fact, a significant relationship between inflammation and arterial stiffness has been demonstrated by several cross-sectional studies showing that PWV is elevated in inflammatory disorders such as rheumatoid arthritis,11 systemic lupus erythematosus,12 aortitis syndrome (Takayasu’s arteritis), as well as inflammatory bowel diseases.13 These observations strongly support the notion that direct inflammation in vasculature (namely, vasculitis) leads to arterial stiffening. However, it remains unknown whether such pathogenic relevance of inflammation to arterial stiffening as observed in established inflammatory diseases can also be applied to the other etiological conditions more commonly seen in CKD (e.g., nephrosclerosis and diabetic nephropathy). In the current issue of the American Journal of Hypertension, Peyster et al. present the largest observational study to date investigating both cross-sectional and longitudinal associations between arterial stiffness and inflammation in 2,933 CKD participants from the Chronic Renal Insufficiency Cohort (CRIC).14 In accordance with previous relevant studies, they confirmed significant univariate correlations between stiffness parameters (including carotid-femoral PWV and central pulse pressure) and inflammatory markers (including interleukin-6, interleukin 1-receptor antagonist, and tumor necrosis factor-α levels) in cross-sectional analyses. They also found independent cross-sectional correlations between the carotid-femoral PWV and inflammatory markers (interleukin-10 and fibrinogen) after multiple adjustments for age, sex, mean arterial pressure, diabetes, estimated GFR, and other potential covariates. Similar multivariate-adjusted correlations were observed between the central aortic pulse pressure and inflammatory markers (high-sensitivity C-reactive protein and fibrinogen). Of note, however, when the longitudinal data were analyzed by comparing baseline inflammation with the changes in arterial stiffness during a 4-year follow-up, there were few meaningful associations between them. The authors concluded that baseline inflammation was not predictive of changes in arterial stiffness over time. The failure of baseline inflammation to predict subsequent aortic stiffening can be interpreted in different ways, as discussed in some part by the authors. First, baseline “snapshot” measurements of inflammatory markers might have been unable to accurately reflect true inflammation status over time owing to the fluctuating nature of the circulating biomarkers.15 Second, since both arterial stiffness and inflammation are known to increase in the presence of many common risk or mediating factors (such as hypertension, smoking, diabetes, obesity, and metabolic disorders),16 the statistical elimination of confounding influences in this study might have caused a partial underestimation of inflammation-induced change in arterial stiffness. The third interpretation, which is probably of the most importance, is “reverse causality”; namely, the overall results of this study might be explicable on the basis that arterial stiffening causes inflammation.17 Aortic stiffening induces renal damage/dysfunction through hemodynamic (pressure and flow) mechanisms. More specifically, aortic stiffening widens aortic pulsatile pressure, which is transmitted deep into the vulnerable microvasculature in high-flow organs such as the kidney and brain.8 The resultant increase in pulsatile tensile stress can cause microvascular damage as well as endothelial dysfunction, oxidative stress and chronic inflammation. This is compatible with microscopic findings of renal tissue, which often reveal inflammatory changes around the microvasculature, glomeruli, or tubules not only in accelerated hypertension (malignant nephrosclerosis) but also in milder hypertension (benign nephrosclerosis) and diabetic nephropathy. Also clinically shown is that increased aortic PWV and widened pulse pressure are independent predictors of albuminuria,5 which is associated with elevated C-reactive protein levels.18 Similar microvascular injury and reactive inflammation resulting from aortic stiffening are also noted in the brain (i.e., neuroinflammation in “pulse-induced encephalopathy”),19 and likely seen elsewhere in the body. Moreover, aortic stiffening causes an increase in aortic blood flow reversal (i.e., backward flow) that leads to reductions in renal artery inflow and thus GFR;6 such flow abnormality potentially induces renal ischemia, which may cause tissue damage provoking an inflammatory response and apoptosis. All these arguments mentioned above, together with the current findings of Peyster et al.,14 are in good agreement with the results of a Mendelian randomization study,20 suggesting that high pulse pressure is not a consequence but rather a cause of C-reactive protein elevation. The present study by Peyster et al.14 has several strengths. The study cohort included a substantial number of CKD patients, who were followed up with for as long as 4 years. Arterial stiffness and inflammation were extensively investigated using different markers. The statistical results based on univariate and multivariate analyses seem to be solid. There are, however, some limitations also. First, the inflammatory markers were examined only once at baseline so that it was unclear whether inflammation (if present) was transient or persistent over the follow-up period. Second, the CRIC was composed of CKD patients with various etiologies. Therefore, additional sensitivity analyses, if done, may have clarified whether different underlying diseases had different influences on the inflammation–stiffness relationship. More concretely, given the above-mentioned evidence, it seems likely that inflammation-induced aortic stiffening can be observed only in patients with some established inflammatory diseases (e.g., lupus nephritis or rheumatoid arthritis), but not in others. Finally, a considerable number of patients in this cohort were receiving various drug treatments. This may have biased the overall results on inflammation and stiffness change, particularly if treatment was required to be changed in the follow-up period. The present observations by Peyster et al.14 provide an important insight into therapeutic targets in CKD. There is no doubt in established inflammatory diseases affecting large arteries (e.g., rheumatoid arthritis) that anti-inflammatory drugs would help to delay arterial stiffening or even reduce arterial stiffness.11 Nevertheless, in most CKD with other etiologies (e.g., hypertensive nephrosclerosis and diabetic nephropathy), inflammation can occur secondarily to microvascular injury or tissue ischemia caused by large artery stiffening and resultant pulsatile hemodynamic abnormality, causing tissue damage, with inflammation as part of the repair process. Could widespread microvascular disease caused by stiffened aorta be the reason for increased inflammatory markers in the elderly? This in turn implies that therapeutic reductions in aortic pulse pressure and/or backward flow, which might be achievable by certain types of antihypertensive vasodilator drugs, would be more effective in preventing microvascular damage and inflammation to improve cardiovascular, cerebrovascular, and renal outcome. Compatible with this view, some vasodilators (such as angiotensin receptor blockers) have been shown to reduce aortic stiffness, aortic pulse pressure, albuminuria, and even inflammatory markers. Obviously, the validity of such therapeutic considerations needs to be tested further by clinical and pathophysiological studies on the causal relationship between arterial stiffening and inflammation. Certainly there is no basis at present for use of anti-inflammatory drugs in persons with unexplained increase of inflammatory markers.