The first secondary endpoint was the mean change in hemoglobin from baseline to week five for each subsequent course of ferumoxytol after treatment course one. The group of patients receiving a second (n=244) and third (n=69) course of treatment with ferumoxytol in IDA-303 included patients who had received ferumoxytol in IDA-301 and patients who had received their first course of ferumoxytol in IDA-303. With repeat dosing, ferumoxytol also demonstrated a statistically significant increase in mean hemoglobin from baseline to week five, with a mean hemoglobin increase of 1.5 g/dL following treatment course two and a 1.1 g/dL increase following treatment course three. The mean hemoglobin level achieved by patients at week five following each of the first three treatment courses with ferumoxytol were all comparable, with the mean hemoglobin levels in course one of 11.7 g/dL, in course two of 11.9 g/dL, and in course three of 11.4 g/dL, suggesting the consistent effect of ferumoxytol treatment with repeat dosing. Only a small number of patients (n=18) received four or more courses of ferumoxytol over the 6 month duration of the study, limiting formal analyses of these data.No new safety signals were observed with repeat dosing of ferumoxytol and the types of reported adverse events (AEs) were consistent with those seen in both the previously reported IDA and CKD phase III studies, and those contained in the approved U.S. package insert for Feraheme ; no hypersensitivity reactions or hypotension were reported in the IDA-303 study. Approximately 48 percent of patients who received ferumoxytol treatment course one in this study experienced an adverse event, consistent with the AE rate reported in the IDA-301 study; in each of treatment courses two and three, approximately 38 percent of patients experienced an AE. Approximately 5 percent of ferumoxytol-treated patients in this study experienced a serious adverse event in each of the three treatment courses, none of which were deemed related to treatment by study investigators. Protocol-defined adverse events of special interest (AESI) for the IDA studies were defined to include mild to severe hypotension or hypersensitivity reactions or associated symptoms. There was one AESI, shortness of breath (or dyspnea), which was reported as not related to treatment in one patient who received a second course of ferumoxytol. Cardiovascular AEs, deemed not related to treatment, were reported in approximately 1 percent of ferumoxytol-treated patients in each of the three treatment courses. No deaths were reported in this study.