Hydroxyethyl starch in sepsis.
Author(s): Haase NR.
Affiliation(s): Author information:
Department of Intensive Care, Copenhagen University Hospital - Rigshospitalet,
Blegdamsvej 9, 2100 Copenhagen, Denmark.
nicolai.rosenkrantz.segelcke.haase@regionh.dk.
Publication date & source: 2014, Dan Med J. , 61(1):B4764
BACKGROUND: Hydroxyethyl starch (HES) is a colloid that has been widely used for
fluid resuscitation for decades. The newest generation of HES, tetrastarch, was
believed to provide an efficient volume expansion without causing the side
effects observed with former HES solutions. However, this belief was based on
physiological models and small studies rather than on firm clinical evidence. Our
aim was to assess the safety and efficacy of tetrastarch in a randomised clinical
trial and in a systematic review.
METHODS: We first conducted a blinded, clinical trial, in which we randomly
assigned patients with severe sepsis in the intensive care unit to fluid
resuscitation with either 6% HES 130/0.42 (Tetraspan) or Ringer's acetate. The
primary outcome measure was death or dialysis-dependency at 90 days after
randomisation. Secondary outcomes described kidney function and serious adverse
reactions. Secondly, we systematically identified all randomised clinical trials
comparing tetrastarch with either crystalloid or albumin in patients with sepsis
and pooled their results in meta-analyses and trial sequential analyses.
RESULTS: Of the 804 patients who underwent randomisation, 798 were included in
the modified-intention-to-treat population. At 90 days after randomisation, 201
of 398 patients (51%) assigned to HES 130/0.42 had died, as compared with 172 of
400 patients (43%) assigned to Ringer's acetate (relative risk 1.17, p=0.03); 1
patient in each group was dialysis-dependent at 90 days. In the 90 day
observation period, 87 patients (22%) assigned to HES received renal replacement
therapy vs. 65 patients (16%) assigned to Ringer's acetate (relative risk 1.35,
p=0.04), and 38 patients (10%) vs. 25 patients (6%) had severe bleeding (relative
risk 1.52, p=0.09). Post-hoc sensitivity analysis showed a strongly significant
increased risk of any bleeding with HES vs. Ringer's acetate (relative risk 1.56,
p=0.003). In the systematic review, we identified nine trials that randomised
3,456 patients with sepsis. In meta-analyses, tetrastarch vs. crystalloid or
albumin lead to increased use of renal replacement therapy (relative risk 1.36,
p=0.009) and red blood cells (relative risk 1.29, p=0.0002) and to more serious
adverse events (relative risk 1.30, p=0.03). Trials with low risk of bias
suggested 11% increased risk of death. After adjusting the results with trial
sequential analysis signals for harm persisted.
CONCLUSION: Our randomised clinical trial is one of several high-quality trials
in critically ill patients with and without sepsis that now provide evidence that
the use of tetrastarch impairs kidney function and haemostasis and may even
increase mortality. Whether the results can be extrapolated to other types of
patients is unclear, but so far no group of patients with an overall benefit of
HES beyond surrogate markers has been identified. In line with this, the European
Medicines Agency's Pharmacovigilance Risk Assessment Committee now recommends
that the marketing authorisations of all HES solutions are suspended in the
European Union.
|