We had previously noticed the renal replacement therapy (RRT) data listed under "Hemofiltration" in Table 3 of the Critical Care (CC) publication by Boussekey et al. However, only 8 total patients required RRT according to that table compared with a total 82 of patients requiring RRT in the study's entire cohort.
How could that be? The explanation is that Boussekey and co-authors excluded the majority of their patients before presenting the RRT results in their CC publication: Their study included 363 total patients. The total population was represented in Table 1 and Figures 1-3 of the CC publication. However, in the final paragraph of the Results section they state:
"To evaluate the incidence of AKI with regard to HES administration, we separated the patients into two groups: the patients with normal kidney function and those with a RIFLE class 'risk' on admission."
They do not make clear that they separated the patients into two groups only after first excluding the 199 patients, comprising 55% of the total, who fell into RIFLE categories higher than 'risk'. The RRT data presented in the CC publication were based on only 164 patients with either normal renal function or minimal renal impairment at baseline. Those 164 patients accounted for only 9.8% (8/82) of all RRT cases.
On the other hand, in the first paragraph under Results of their CC publication they did incorporate outcome data pertaining to all 363 patients for several endpoints, namely, incidence and duration of secondary shock, vasopressor usage, frequency and duration of mechanical ventilation, and mortality. In another presentation of the study's results, those same outcome data were accompanied by the RRT results. Yet, in the CC publication the RRT data for the full study population and the significant difference in RRT were omitted.
On this background, the title of the paper is misleading. The entire data set of the study by Boussekey et al. provides evidence of potential deleterious renal effects associated with starch infusion which was statistically significant.
Competing interests
CW received fees for speaking and travel cost reimbursement from CSL Behring, Baxter, Kedrion and PPTA.
Reason for concern
3 September 2013
We had previously noticed the renal replacement therapy (RRT) data listed under "Hemofiltration" in Table 3 of the Critical Care (CC) publication by Boussekey et al. However, only 8 total patients required RRT according to that table compared with a total 82 of patients requiring RRT in the study's entire cohort.
How could that be? The explanation is that Boussekey and co-authors excluded the majority of their patients before presenting the RRT results in their CC publication: Their study included 363 total patients. The total population was represented in Table 1 and Figures 1-3 of the CC publication. However, in the final paragraph of the Results section they state:
"To evaluate the incidence of AKI with regard to HES administration, we separated the patients into two groups: the patients with normal kidney function and those with a RIFLE class 'risk' on admission."
They do not make clear that they separated the patients into two groups only after first excluding the 199 patients, comprising 55% of the total, who fell into RIFLE categories higher than 'risk'. The RRT data presented in the CC publication were based on only 164 patients with either normal renal function or minimal renal impairment at baseline. Those 164 patients accounted for only 9.8% (8/82) of all RRT cases.
On the other hand, in the first paragraph under Results of their CC publication they did incorporate outcome data pertaining to all 363 patients for several endpoints, namely, incidence and duration of secondary shock, vasopressor usage, frequency and duration of mechanical ventilation, and mortality. In another presentation of the study's results, those same outcome data were accompanied by the RRT results. Yet, in the CC publication the RRT data for the full study population and the significant difference in RRT were omitted.
On this background, the title of the paper is misleading. The entire data set of the study by Boussekey et al. provides evidence of potential deleterious renal effects associated with starch infusion which was statistically significant.
Competing interests
CW received fees for speaking and travel cost reimbursement from CSL Behring, Baxter, Kedrion and PPTA.