|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
THE JOHN F. MAHER AWARD RECIPIENT LECTURE 2008 |
Center for Health and Biological Sciences, Pontifícia Universidade Católica do Paraná, Curitiba, Brazil
Correspondence to: R. Pecoits–Filho, Center for Health and Biological Sciences, Imaculada Conceição, 1155, Curitiba, PR 80215-901 Brazil. r.pecoits{at}pucpr.br
| ABSTRACT |
|---|
|
|
|---|
The bench-to-bedside approach to translational research is becoming
increasingly important to efficiently advance understanding of the mechanisms
underlying disease and to improve the quality of patient care. Although this
investigation model has been practiced since the early days of the therapy,
robust research platforms built to practice translational research have only
recently been structured in the field of peritoneal dialysis. Experience with
a translational research environment that generated most of the information
cited in this overview is the core of this manuscript. The central
investigation theme described is how to approach the cardiovascular
complications of peritoneal dialysis. The research question was, could the
continuous activation of inflammatory pathways be central in this process and
represent a relevant target for interventions?
KEY WORDS: Inflammation; cardiovascular disease.
In order to have an impact in improving human health, scientific discoveries must reach clinical applications. Such discoveries typically begin at "the bench" with basic research in which scientists study disease at a molecular or cellular level, then progress to the clinical level, or the patient's "bedside." Scientists and clinicians are increasingly aware that this bench-to-bedside approach to translational research is really a two-way road: clinical researchers make novel observations about the nature and progression of disease that often stimulate basic investigations, and basic scientists provide clinicians with new tools for use in patients and for assessment of their impact. The most important part of the definition of translational research is the focus on potential clinical application. Translational research has proved to be a powerful process that drives the clinical research engine; however, a stronger research infrastructure, close interaction, and global collaboration could strengthen and accelerate this critical part of the clinical research enterprise.
Although this investigation model has been practiced since the early days of the therapy, research platforms built to practice translational research have only recently been structured in the field of peritoneal dialysis (PD). Experience with a robust translational research environment that generated the information cited in this overview is the core of this manuscript. The central investigation theme was how to approach the cardiovascular (CV) complications of PD, and the research question was, could the continuous activation of inflammatory pathways be central in this process? The four steps in the translational process are briefly described and illustrated with basic references from our research group.
| STEP 1: BUILDING UP A RATIONALE — INFLAMMATION MAY BE DRIVING A SIGNIFICANT PROPORTION OF THE CV COMPLICATIONS OF CHRONIC KIDNEY DISEASE (CKD) AND LONG-TERM PD |
|---|
|
|
|---|
| STEP 2: SEARCHING FOR CONSISTENT BACKGROUND INFORMATION THAT MAY JUSTIFY THE HYPOTHESIS THAT INFLAMMATION PLAYS A CENTRAL ROLE IN CV COMPLICATIONS OF CKD |
|---|
|
|
|---|
Other sources of stimulating data linking inflammation to CV complications in CKD are based on genetic studies, particularly association studies using single nucleotide polymorphisms (SNPs). Although the nephrology community has only recently started to investigate the impact of various gene polymorphisms in inflammation and oxidative stress-related genes, there is evidence suggesting that genetic variation may indeed affect the phenotype of this patient group. At present, tumor necrosis factor alpha 308, interleukin (IL)-6-174 and IL-10-1082 SNPs seem to be most consistently associated with adverse clinical outcomes in CKD and PD patients (8). Moreover, there are reports suggesting that genetic variations in genes related to oxidative stress [myeloperoxidase (9) and NADPH (10)] are associated with important clinical features in the CKD population. Altogether, these data reinforce that individuals with a genetic predisposition to present an exaggerated inflammatory response to stimuli may be exposed to a high risk of developing CV complications in CKD.
Back to the clinical ground, since the first report by Bergström and his co-workers of an association between elevated C-reactive protein (CRP; a marker of systemic inflammation) and increased mortality in CKD patients, several groups have reported almost identical findings in different patient groups, in all stages of CKD, and using other markers of inflammation (11). Although the association between single or multiple (12) measurements of inflammation markers and an increased risk of mortality in CKD patients is a consistent finding, the triggers of the inflammatory response are still only superficially understood. In addition, whether inflammation is a direct causative factor determining target organ damage or represents an epiphenomenon is still an unanswered question. Direct uremic toxicity, accumulation of modified proteins (such as advanced glycation end-products), retention of inflammatory mediators, mechanical stress of the vascular wall as a result of hypertension, comorbidities such as advanced age and diabetes, and disorders of mineral metabolism and anemia are examples of factors potentially triggering the inflammatory response but are unrelated to dialysis treatment (13).
| STEP 3: REBUILDING THE RATIONALE WITH A FOCUS: THE IMPORTANCE OF INFLAMMATION IN CV COMPLICATIONS OF PD |
|---|
|
|
|---|
Inflammation and the Peritoneal Membrane: Inflammatory changes such as increased numbers of macrophages are often seen in the peritoneum of CKD patients. There are consistent data coming from the Peritoneal Biopsy Registry showing that morphological changes are already present before the initiation of PD. Experimental studies confirm that induction of CKD or diabetes can alter the structure and function of the peritoneal membrane not exposed to PD fluids. In addition, peritoneal changes are enhanced with time on PD. This is true even in the absence of peritonitis, indicating that the peritoneum of PD patients is a chronically inflamed organ. Observations of peritoneal tissue from long-term PD patients (and in experimental studies) show a thickening of the submesothelial space and alterations of the microvessel structure [reviewed in Ref. (14)].
These morphological alterations occur concomitantly with functional changes and, over time, both increase peritoneal small-solute transport rate (PSTR). For reasons yet to be clarified, a high PSTR has been identified as an important cause of PD technique failure and, at least in patients under continuous ambulatory PD, represents an independent risk factor for increased mortality in PD. This may be linked to the diminished ultrafiltration capacity during long dwells and may not be true for patients on automated PD. Interestingly, we observed that patients with high PSTR presented higher plasma levels of IL-6 and higher prevalence of comorbidities such as diabetes mellitus and peritonitis (15). Also, in a longitudinal study, intraperitoneal inflammation was associated with high PSTR, but this association was observed predominantly during the early phase of PD (16). Patients with a high PSTR from the beginning of the therapy are typically inflamed and have a high prevalence of comorbidities and high mortality (15). In addition, these patients sometimes present peritoneal albumin leakage, which may reflect endothelial dysfunction (17). On the other hand, patients presenting the "late acquired type" of high PSTR develop high PSTR over time on PD, due primarily to structural changes in the peritoneal membrane caused by continuous exposure to high glucose PD solutions. These patients do not necessarily have a higher prevalence of inflammation or comorbidities, and the high mortality could be prevented by correcting fluid overload (14).
Bioincompatibility of PD Solutions and Inflammation: One potential risk factor for inflammation in PD patients may relate to prolonged exposure to conventional bioincompatible glucose-based PD solutions, that is, solutions that have high osmolarity, glucose, and lactate concentrations, low pH, and high content of glucose degradation products, which enhance the formation of advanced glycation end products.
Many biomarkers of intraperitoneal inflammation have been studied in the clinical setting. We concentrated our studies on the analysis of intraperitoneal IL-6 because its levels are stable in dialysate, the concentration is many-fold higher than in plasma, and the available methods are simple and reproducible (18). However, it must be emphasized that there is no consensus about the ideal biomarker in this area. It is noteworthy that, although IL-6 is a well-recognized marker of systemic inflammation (19), the precise significance of intraperitoneal IL-6 in the setting of PD is still under investigation. It is also important to note that in vitro studies that show IL-6 levels after stimulation reflect cell vitality (18). After such stimulation, the use of more biocompatible solutions is usually associated with higher IL-6 production. In clinical studies, IL-6 levels should represent the effect of local stimulation of IL-6 production, and thus the degree of intraperitoneal inflammation. Dialysate IL-6 increases over time in patients using standard glucose-based PD solutions, and IL-6 effluent levels are related to peritoneal transport and markers of oxidative stress and neoangiogenesis (16). Moreover, in one interventional study, IL-6 levels in the dialysate effluent decreased when patients were shifted to more biocompatible solutions, suggesting less intraperitoneal inflammation with the use of those solutions (20). In addition, an association was found between systemic and intraperitoneal inflammation based on the correlation between plasma and dialysate IL-6 levels (16), although this finding needs to be confirmed in larger studies.
Taken together, the available data suggest that PD solutions may induce intraperitoneal inflammation, and the generation of a cellular response in the peritoneal cavity may affect systemic inflammation. Whether systemic inflammation causes local inflammation and/or intraperitoneal inflammatory activation leads to a systemic response requires further evaluation. Also, the impact of intraperitoneal inflammation on CV risk remains to be investigated.
Fluid Overload and Inflammation: Volume overload is a frequent complication in PD patients and may itself be associated with immune activation. This activation may occur because of bacterial or endotoxin translocation in patients with severe gut edema as a result of severe volume overload, which in turn may lead to increased production of proinflammatory cytokines (21). Indeed, we showed that CKD patients with signs of fluid overload present higher circulating endotoxin levels (22). A recent study found that endotoxemia is common in PD patients and that the degree of circulating endotoxemia might be related to the severity of systemic inflammation and features of CV disease. These results suggest that endotoxemia may play a contributory role in the systemic inflammatory state and accelerated CV risk in PD patients. In summary, fluid overload is common among PD patients and may potentially be an important modifiable contributor to inflammation and CV risk among PD patients.
Altered Carbohydrate Metabolism, Body Composition, and Inflammation: Carbohydrate metabolism is severely altered in CKD, even before the initiation of PD. Nondiabetic patients with CKD often have mild fasting hyperglycemia and abnormal glucose tolerance, although most patients can maintain normoglycemia at the expense of hyperinsulinemia. Particularly in PD patients using glucose-based solutions, systemic glucose load may have an impact on the disorders described above (5). Indeed, there is increasing concern about the adverse systemic effects of absorbed glucose, especially with respect to the CV risk profile. Interestingly, we recently described a state of insulin resistance, mild increase in fasting glycemia, dyslipidemia, altered glycated hemoglobin (HbA1c), and inflammation in a group of nondiabetic PD patients that was markedly worse in PD than in hemodialysis patients (23). These preliminary data point to a complex metabolic disorder of carbohydrate metabolism present in PD patients, a disorder that might have potential effects on the CV system.
The mechanism by which altered carbohydrate metabolism may lead to an increase in CV risk is only recently being understood, but emerging evidence points to an important role for the central obesity–insulin resistance–inflammation axis. Several studies have demonstrated that fat tissue, particularly the visceral fat accumulated in central obesity, is immunologically active and it is able to attract macrophages and initiate a process of production of inflammatory mediators that ultimately lead to insulin resistance, hepatic steatosis, and accelerated atherogenesis. Added to hypertension, dyslipidemia, and central obesity — all common features of PD patients — a metabolic syndrome (a cluster of risk factors related to increased morbidity and mortality in the general population) potentially develops in a high proportion of these of patients. In a recent cross-sectional study, we observed that 53% of our PD patients filled the criteria for metabolic syndrome, a proportion that was significantly higher than that observed in hemodialysis patients (23). Accordingly, we observed in a large cohort of dialysis patients that high plasma levels of monocyte chemoattractant protein-1 (a central mediator of the early atherogenic response) were significantly associated with hypertension, abdominal circumference, homeostasis model assessment (HOMA) index, and fasting glucose. Although there is no description of the impact of a metabolic syndrome on clinical outcome in PD patients, a few studies have described the relationship between abnormal fasting glycemia and HbA1c and increased mortality in groups of nondiabetic patients. Based on these observations, we hypothesize that the continuous absorption of glucose during PD will lead to a progressive increase in central fat accumulation, inducing a worsening of insulin resistance, hepatic steatosis, and endothelial dysfunction, as well as leading to an increase in the concentration of plasma adhesion molecules, chemokines, and surrogate markers of inflammation. Those inflammatory mediators will orchestrate organ damage in the vascular wall as well as in other target organs.
| STEP 4: BACK TO THE PATIENT |
|---|
|
|
|---|
Statins have consistent background information generated from studies in the general population pointing to a reduction in CV risk (24). Ongoing trials that will be finished in the near future may answer the question of whether statins will be universally applied to reduce risk in dialysis patients, since the results of the 4D Trial were not positive. There is an urgent need to perform trials with statins in the PD population, in which dyslipidemia is more prevalent and severe.
Angiotensin-converting enzyme inhibitors may prove beneficial in reducing inflammation and CV disease in CKD patients. Clearly, the renin–angiotensin system may contribute to inflammatory processes within the vascular wall and contribute to the development of plaque instability and acute coronary syndromes. Inhibiting the formation of angiotensin II by ACEi and angiotensin II blockers may reduce the damaging effects on endothelial function, vascular smooth muscle cells, and inflammatory processes. Preliminary data in PD are promising (25), but prospective studies will determine if ACE inhibition may lead to a decrease in CV mortality in PD patients.
Although studies in the nonrenal population have failed to show that vitamin E supplementation improves CV outcomes, it is reasonable to speculate that populations exposed to high levels of oxidative stress, such as malnourished and inflamed CKD patients, would benefit the most from antioxidant intervention. Indeed, in the CKD population, the SPACE study (26) demonstrated a reduction in myocardial infarction and other CV events (but not overall mortality) in the vitamin E-treated group compared to patients receiving placebo. Moreover, a recent study (27) showed that n-acetylcysteine (a thiol-containing antioxidant) reduces composite CV end points in hemodialysis patients.
The insulin-sensitizing thiazolidinediones, which are selective ligands of
the nuclear transcription factor PPAR
, are the first drugs to address
the basic problem of insulin resistance. PPAR
is expressed most
abundantly in adipose tissue but is also found in pancreatic beta cells,
vascular endothelium, and macrophages. Rosiglitazone is a PPAR
agonist
that has anti-inflammatory properties and was administered to type 2 diabetes
patients undergoing PD. There was a significantly greater decrease in insulin
dosage in the treatment group compared to the control group. At the end of the
study, patients treated with rosiglitazone also had significantly lower CRP
levels than the control group
(28). Whether long-term use of
rosiglitazone reduces CV risk needs further investigation.
Finally, the impact of low glucose regimens, different electrolyte compositions, better ultrafiltration capacity, and alternative solutions with a better biocompatibility profile will need to be analyzed in the clinical setting to test the hypothesis that they can reduce CV mortality in PD patients (7). Background results from animal studies, small interventions leading to a reduction of surrogate markers, and relatively large cohort studies are stimulating but large randomized controlled trials designed to reduce CV end points and, ultimately, mortality are lacking. The PD community needs to join forces to induce these studies, which are expensive and hard to perform but are absolutely necessary to advance the knowledge of how to manage these patients.
| SUMMARY AND CONCLUSIONS |
|---|
|
|
|---|
|
| DISCLOSURE |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
The author thanks the research group from the Karolinska Institute (Bengt Lindholm, Peter Stenvinkel, Bjorn Anderstam, and Olle Heimbürger), co-workers at the research laboratory at PUC-PR (Andrea Stinghen, Aline Hauser, Lia Nakao, Paulo Aveles, and Debora Grahl), and those at the nephrology clinic involved in these studies (Miguel Riella, Paulo Fortes, Thyago Moraes, Silvia Ribeiro, Simone Gonçalves, and Lucimary Sylvestre).
Received 31 December 2008; accepted 3 March 2009.
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |