Hintergrund Sowohl die Anzahl der Patienten, die an Diabetes mellitus Typ II erkrankten als auch die Zahl der Patienten, die sich einer Dialysetherapie unterziehen mussten, stiegen in den letzten Jahren stetig an. Im Vergleich zur Normalbe-völkerung ist die Mortalität bei Dialysepatienten um ein Vielfaches erhöht. Da mit dem Verlust der Nierenfunktion nicht nur die exkretorischen sondern auch die inkretorischen Aufgaben der Nieren, wie beispielsweise die Ausschüttung des Hormons Erythropoie-tin, nicht mehr geleistet werden können, sind diese Patienten auf eine Substitution angewiesen. Einige Dialyse-Patienten benötigen allerdings überdurchschnittlich viel Erythropoietin um den Ziel Hb-Wert zu erreichen und zu halten. Man spricht hier von EPO-Resistenz. Zwei kleine observationelle Studien haben im letzten Jahr suggeriert, dass die Gabe von Atorvastatin eine Verbesserung der EPO Resistenz bewirken könnte. Auch ein Zusam-menhang des erreichten Hämoglobinwertes und Inflammation mit einer EPO Resistenz wurde diskutiert10,11 Diese Studien waren allerdings weder verblindet noch an einem ausreichend großen Patienten-Klientel getestet. Das Ziel der vorliegenden Untersuchung lag daher darin, den Einfluss von Statinen auf die EPO-Resistenz in einer großen ran-domisierten Studie zu untersuchen. Des Weiteren wurden Zusammenhänge von EPO-Resistenz mit klinischen Parametern und Endpunkten evaluiert. Methoden Den Untersuchungen lagen die Daten der 4D-Studie („Die Deutsche Diabetes Dialyse Studie“), einer multizentrischen, randomisierten, prospektiven Doppelblind-Studie mit 1255 an Diabetes mellitus erkrankten Dialysepatienten zu Grunde. Die Patienten wur-den in dieser Studie randomisiert, entweder 20 mg Atorvastatin pro Tag oder das ent-sprechende Placebo über einen mittleren Beobachtungszeitraum von 4 Jahren zu erhalten. 10 Alle Patienten wurden weniger als 2 Jahre dialysiert und waren im Alter zwischen 18 und 80 Jahren alt. Nach einer Anlaufperiode von 4 Wochen erfolgte die Randomisierung: 619 Studienteilnehmer wurden auf Atorvastatin- und 636 auf Placebo randomisiert. Der primäre Endpunkt der 4D-Studie bestand aus Tod aufgrund kardialer Ursa-chen, tödlichem oder nicht tödlichem Schlaganfall und nicht tödlichem Myokardinfarkt. Die sekundären Endpunkte waren definiert als Tod durch spezifische kardiale und zerebrovaskuläre Ereignisse, sowie Tod infolge anderer Ursachen. In regelmäßigen Abständen wurden die Teilnehmer unter Anderem auf Komedikation, Körpergewicht, Blutdruck, Puls und diverse hämatologische und biochemische Laborparameter untersucht. Im Speziellen wurde getestet, ob eine Wirkung von Atorvastatin 20 mg auf die EPO Resistenz bei Dialysepatienten mit Diabetes mellitus besteht. Dafür wurden die durch-schnittlichen Hämoglobinwerte, die mittlere EPO-Dosis und der mittlere EPO Resistenz Index für beide Patienten Gruppen über den gesamten Beobachtungszeitraum der Studie dokumentiert und miteinander verglichen. Weiterhin waren die Zusammenhänge zwischen EPO Dosis in Kombination mit Hb-Wert bezogen auf die Gesamtmortalität, kardiovaskuläre Ereignisse, plötzlichem Herztod, Schlaganfall und Myokardinfarkt ein Hauptpunkt der Analyse. Des Weiteren wurde die Assoziation von Hb-Variabilität zu den klinischen Endpunkten und die Beziehung zwischen Inflammation und EPO- Resis-tenz in der 4D-Studie evaluiert. Ergebnisse Obwohl eine deutliche Senkung der Lipidparameter durch Atorvastatin auch in dieser Studie nachgewiesen werden konnte (Atorvastatin-Gruppe von 121 mg/dl auf 72 mg/dl in vier Wochen, Placebo-Gruppe weitgehend unverändert), zeigte sich in der Primärstu-die keine signifikante Reduktion des primären Endpunktes. In den speziellen Analysen zur EPO Resistenz in der vorliegenden Arbeit konnte im Rahmen einer Post-hoc Auswertung der 4D Studiendaten gezeigt werden, dass Statine die EPO Resistenz, sowie EPO-Dosis und Hb-Wert von Dialysepatienten nicht signifikant beeinflussen. In der Atorvastatin-Behandlungsgruppe war gegenüber der Placebo-Gruppe keine Verbesserung dieser 3 Parameter während der 4 jährigen Beobachtungszeit zu verzeichnen. Hohe EPO Dosierungen, insbesondere in Zusammenhang mit nied-rigem Hb waren mit erhöhter Mortalität, kardiovaskulären Endpunkten und plötzlichem Herztod assoziiert. Die EPO-Resistenz, repräsentiert durch den EPO-Resistenz-Index ging mit erhöhten Inflammationswerten einher. Die Hämoglobin-Variabilität über ein Jahr zeigte in unserer Studie allerdings keinen Einfluss auf klinische Endpunkte. Schlussfolgerung Es gilt wissenschaftliche Arbeiten mit kleinen Fallzahlen und observationellem Charak-ter durch randomisierte große Studien kritisch zu beurteilen. Die erhöhte Mortalität und die gesteigerte Rate kardialer Ereignisse und Schlaganfälle bei EPO resistenten Patienten geben Anlass zu erhöhter Vorsicht bei der Gabe hoher EPO-Dosen an Dialysepatienten. Statine üben keinen beweisbaren klinisch bedeutsa-men günstigen Effekt auf die EPO-Resistenz bei diesen Patienten aus. Angesichts des Zusammenhanges zwischen EPO-Resistenz und Inflammation ist der antiinflammatorische Beitrag von Statinen -gemessen an anderen EPO-Resistenz bedingenden Faktoren- möglicherweise relativ zu klein, um sich klinisch bemerkbar zu machen. Weitere Studien zur Erforschung von Risikofaktoren für eine EPO-Resistenz sind daher zwingend notwendig, um weitere Zusammenhänge beleuchten und neue Therapiestrategien etablieren zu können. / Atorvastatin does not improve erythropoietin resistance in hemodialysis patients: results from the 4D study ABSTRACT Background: Statins have been suggested to improve erythropoietin resistance in dialysis patients, but data from randomized clinical studies are missing. Furthermore, little is known about the impact of erythropoietin resistance status on specific cardiovascular outcomes in dialysis patients. Methods: We conducted a post-hoc analysis of the German Diabetes and Dialysis Study (4D Study), a multicenter, randomized, double blind prospective study of 1255 patients with type 2 diabetes mellitus undergoing hemodialysis. Patients were randomly assigned to receive 20mg atorvastatin per day or matching placebo. Erythropoietin resistance status was evaluated every 6 months during a median follow-up of 4 years. By Cox regression analyses, we furthermore determined hazard ratios (HR) for pre-specified, adjudicated endpoints according to baseline erythropoietin resistance status and hemoglobin variability: sudden death (n=160), myocardial infarction (MI, n=200), stroke (n=103), cardiovascular events (CVE; n=469), and all-cause mortality (n=617). Results: Patients had a mean age of 66±8 years (54% male). No differences in hemoglobin, erythropoietin dose requirements, or erythropoietin resistance index were found between the atorvastatin- and the placebo-treated groups for any of the time points analysed during follow-up. Erythropoietin resistance was associated with inflammation, and meaningfully affected adverse outcomes: Patients with Epo doses > 100 IU/week/kg necessary for a sufficient hemoglobin level of 11 g/dl had an adjusted 2.5 fold increased risk of stroke. Similarly, these patients experienced significant, ≥70% higher rates of sudden death, combined cardiovascular events and mortality, while the risk of myocardial infarction was not affected. Higher hemoglobin variability did not increase adverse outcomes in the present analysis. Conclusions: This study in the 4D randomised controlled trial did not suggest any effect of atorvastatin in improving responsiveness to EPO therapy in maintenance hemodialysis patients. Resistance to erythropoietin was strongly associated with stroke, sudden death, combined cardiovascular events and mortality. Further studies are needed to reduce erythropoietin resistance in hemodialysis patients. INTRODUCTION Despite an increase in the use of erythropoietin over the last decade, a substantial proportion of patients with chronic kidney disease (CKD) fail to achieve hemoglobin targets as recommended by the international guidelines (1,2). In this setting, the term epo-resistance is used to define the patients who fail to attain the target Hb despite a higher than usual dose of erythropoietin, or who continuously need a higher dose to maintain it. Approximately 10% of patients with CKD would fulfil this definition, but the true incidence of Epo hyporesponsiveness is not precisely known. The excessive administration of Epo in hyporensponsive patients poses an enormous burden on the social communities. It is accompanied by very high costs requiring large budgets, and competing with other patients needs. Most importantly, Epo resistance has been shown to be associated with an increased mortality (3). Therefore, it is important to reduce Epo resistance, with the potential to improve patient outcomes and support social healthcare systems. Chronic kidney disease, and in particular end-stage renal disease (ESRD) has been recognized as an inflammatory state with increased levels of CRP, IL-6 and TNF alfa. Evidence suggests that the inflammatory state increases the resistance to erythropoietin. Important mechanisms involved may be that cytokines could directly inhibit erythropoiesis and promote apoptosis of erythroid precursors (4-7), decrease iron availability for erythropoiesis and upregulate hepcidine. Likewise, statins have been suggested to decrease or attenuate inflammation in ESRD patients, and potentially reduce Epo resistance (8,9). However, the latter studies were small and uncontrolled, and the anti-inflammatory effect of statins did not translate into an improvement of survival in randomized trials (10,11). In this context, there has furthermore been much discussion and controversy about the impact of target haemoglobin and Epo responsiveness on cardiovascular events and survival in dialysis patients. The effect of statins on Epo resistance remains largely unclear, as the role of Epo resistance and target hemoglobin for the different components of cardiovascular events and survival in these patients. Hence, we conducted a post-hoc analysis of the 4D study, a randomized controlled trial to evaluate the effect of cholesterol lowering therapy in 1255 dialysis patients with type 2 diabetes mellitus. Our primary aim was to investigate the effect of atorvastatin 20mg od on ESA responsiveness in the patients receiving ESA therapy. Second, we hypothesized an association between ESA responsiveness and inflammation in dialysis patients. Finally, we assessed the effect of ESA responsiveness on sudden cardiac death, stroke, myocardial infarction, combined cardiovascular events and mortality, as the effect of haemoglobin variability on these endpoints. Statistical Analysis Patient characteristics are presented according to categories of Erythropoietin resistance status. Patients were divided into the following four groups: 1) Hb 8804; 11g/dl, EPO use 8804; 100 IU/kg/week 2) Hb 8804; 11g/dl, EPO use > 100 IU/kg/week, 3) Hb > 11g/dl, EPO use 8804; 100 IU/kg/week, 4) Hb > 11g/dl, EPO use > 100 IU/kg/week. Continuous variables were expressed as mean with standard deviation or median with interquartile range (IQR) as appropriate, and categorical variables were expressed as percentages. In order to investigate the impact of statins on erythropoietin resistance, the mean haemoglobin,the mean EPO dose,and the mean ESA index per patient for the 5 years of study were determined. The darbepoetin dose (ug/kg/wk) was converted to the equivalent EPO dose by multiplying x 200.The ESA index was computed from the EPO dose (U/kg/wk) divided by the Hb (g/dL). A comparison was made of the values for each of these three measurements between the atorvastatin- and placebo-treated groups. Furthermore, haemoglobin, Epo dose, and Esa index values were investigated longitudinally across the 5 years of study including differences from baseline by treatment. Second, the values of Epo dose and ESA index at baseline were compared between patients with different levels of CRP in order to assess the association of Epo resistance with inflammation. Patients were thereby grouped according to a CRP level of < or 8805;5mg/dl, and a level of < or 8805;10mg/dl in a second step. Third, the effect of erythropoietin resistance on clinical outcomes was assessed by Kaplan-Meier estimates for incidences of the pre-specified endpoints grouped by the categories of Erythropoietin resistance status (see above). Furthermore, relative risks were derived from Cox regression analyses, i.e. hazard ratios (HR) and corresponding 95% confidence intervals. The analyses were adjusted for the potential confounders age, gender, duration of diabetes, duration of haemodialysis, time on EPO therapy, co-morbidity, cholesterol level, CRP level, and ferritin. Similarly, in order to assess the impact of hemoglobin variability on clinical outcomes, Kaplan-Meier and Cox regression analyses were performed. The standard deviation for the first 4 hemoglobin values (baseline, 1, 6, and 12 months) was determined and patients grouped according to quartiles of this parameter. Kaplan-Meier plots were constructed for each of the four groups, as were adjusted hazard ratios and 95% confidence intervals calculated. All p-values are reported two-sided. Analyses were performed using SAS version 9.1.3. RESULTS Patient characteristics Between March 1998 and October 2002, a total of 1255 patients were included into the 4D study. The mean follow-up period was 3.96 years (median 4.0 years) on atorvastatin and 3.91 years (median 4.08 years) on placebo. During follow-up, 617 patients died, of whom 160 patients died of sudden cardiac death. Furthermore, a total of 200 patients experienced a fatal or non-fatal myocardial infarction. In the study population (n=1255), the mean (SD) age was 65.7 (8.3) years, and 54% of the patients were male. A total of 81% of the patients received erythropoietin. The mean (SD) baseline hemoglobin level was 10.9 (1.4) g/dl; with no significant difference between the atorvastatin and placebo groups. The baseline patient characteristics are shown in Table 1. Effect of atorvastatin on erythropoietin resistance There was no difference between the atorvastatin and placebo groups in mean haemoglobin, mean Epo dose, and mean ESA index, when the mean values over 5 years were calculated per patient. The respective values were 11.2 (1.0) vs 11.2 (1.1) g/dl for haemoglobin (p=0.99), 83.6 (48.5) vs 86.6 (49.8) IU/kg/week for Epo dose (p=0.33) and 7.6 (4.9) vs 7.9 (5.0) IU/kg/week/Hb for the ESA index (p=0.32). The results of the longitudinal analyses investigating the single timepoints separately across the 5 years of study including differences from baseline by treatment are shown in Figures 1A-C. Overall, the differences were very small and not statistically significant. Association of inflammation with erythropoietin resistance Patients with a CRP level above 5mg/L had higher doses of erythropoietin and a higher ESA index as compared to the patients with a CRP below 5mg/L. The median level of epo dose was 69 IU/kg/week (IQR 45 - 101) in the patients without, and 76 IU/kg/week (IQR 47 - 108) in the patients with inflammation, as defined by the cutoff of the CRP at 5mg/L. Accordingly, the values for the ESA index were 6,4 (IQR 4,1 - 9,5) and 7,1 (IQR 4,4 -10,8), respectingly, the comparison of which indicated a trend for a difference, as reflected by a p-value of 0.09. Further analyses were performed, using a higher cutoff for the presence of inflammation with a CRP level of 10mg/L. The association of epo resistance with inflammation was more pronounced, as reflected by significantly different levels of the ESA index in patients with and without inflammation. The ESA index was 6,4 (IQR 4,1 - 9,6) in patients with a CRP ≤ 10 mg/L, and 7,2 (IQR 4,5 - 11,1) in patients with a CRP > 10 mg/L (p = 0,034). The doses of erythropoietin were 70 (IQR 45 - 101,5) and 77 (IQR 48-113) IU/kg/week, respectively. Erythropoietin resistance status and risk of adverse clinical outcomes In Kaplan-Meier analyses, the group of patients with a sufficient haemoglobin >11 g/dL and low Epo dose ≤100 IU/week/kg showed the best survival and lowest rates of adverse cardiovascular outcomes (Figures 2A-D). When investigating relative risks by Cox regression analyses, these patients were used as the reference group. Compared to those, patients with sufficient hemoglobin >11 g/dl, but higher Epo use > 100 IU/week/kg had an adjusted 2.5 fold higher risk of stroke. Furthermore, the risk of sudden death was increased by 79%, and the risk of combined cardiovascular events by 73%, respectively. Similarly, these patients experiened a 70% higher mortality as compared to the reference group. Of note, low hemoglobin levels also affected outcomes, both with the use of low or high Epo doses. These patients with low hemoglobin levels < 11g/dl showed higher risks of sudden death, combined cardiovascular events and mortality as compared to the reference group. In contrast, myocardial infarction was not affected by neither combination of hemoglobin and Epo use (Table 2). Hemoglobin variability and risk of adverse clinical outcomes Overall, the risks of sudden death, stroke and myocardial infarction did not meaningfully differ between the quartiles of hemoglobin variability. Surprisingly, there was a lower incidence of cardiovascular events in the highest as compared to the lowest hemoglobin variability group, when cardiovascular events were analyzed in the combined endpoint. However, the risk of all-cause mortality did not differ across the quartiles of hemoglobin variability. DISCUSSION This study investigated the effect of statin use on erythropoietin resistance in hemodialysis patients, using data from the randomized controlled German Diabetes and Dialysis Study (4D Study). We have shown in a large group of 1255 patients that atorvastatin did not meaningfully affect erythropoietin resistance, the dose of erythropoietin, and the level of hemoglobin during a median of 4 years follow-up. There was an association between erythropoietin resistance and inflammation, as patients with a higher ESA index had higher levels of C-reactive protein. Furthermore, patients with higher Epo doses necessary for a sufficient hemoglobin level of 11 g/dl had an adjusted 2.5 fold increased risk of stroke. Similarly, these patients experienced significantly higher rates of sudden death, combined cardiovascular events and mortality. Hemoglobin variability did not meaningfully impact on adverse outcomes in the present analysis. - comparison with literature - mechanisms This study had certain limitations. It was a post-hoc analysis within a selected cohort of German patients with type 2 diabetes mellitus on hemodialysis. Therefore, the associations found may not be generalisable to other patient populations. The double-blind treatment and setting of a randomized clinical trial was the major strength of this study. Further important strengths include the specific clinical outcomes to be analyzed. In this context, the long-term follow-up, adequate sample size and high incidence of pre-specified and centrally adjudicated endpoints are further to be mentioned. In conclusion, this post hoc analysis of 1255 patients in the 4D randomised controlled trial did not suggest any effect of atorvastatin in improving responsiveness to EPO therapy in maintenance HD patients. Resistance to erythropoietin was a strong risk factor for stroke, and furthermore increased the risks of sudden death, combined cardiovascular events and mortality. Further studies with novel treatments are needed in order to address the problem of erythropoietin resistance in hemodialysis patients.
| Identifer | oai:union.ndltd.org:uni-wuerzburg.de/oai:opus.bibliothek.uni-wuerzburg.de:4236 |
| Date | January 2010 |
| Creators | Fischer, Martina |
| Source Sets | University of Würzburg |
| Language | deu |
| Detected Language | German |
| Type | doctoralthesis, doc-type:doctoralThesis |
| Format | application/pdf |
| Rights | https://opus.bibliothek.uni-wuerzburg.de/doku/lic_ohne_pod.php, info:eu-repo/semantics/openAccess |
Page generated in 0.0068 seconds