Hydroxyethyl starch and Ringer’s lactate for volume resuscitation in sepsis and septic shock – The BaSES Trial

doi:10.21203/rs.3.rs-3469382/v1

Download PDF

Article

Hydroxyethyl starch and Ringer’s lactate for volume resuscitation in sepsis and septic shock – The BaSES Trial

https://doi.org/10.21203/rs.3.rs-3469382/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Study objective: Shock develops in one third of ICU patients. Uncertainty remains about the ideal amount and choice of resuscitation fluid to avoid consequences of fluid overload.

Design: The BaSES (Basel Starch Evaluation in Sepsis) Trial is an investigator-initiated, single-centre, randomised controlled clinical trial on volume resuscitation.

Setting: We investigated whether volume resuscitation with Hydroxyethyl starch (HES 130/0.4) compared to normal saline (NaCl 0.9%) reduces ICU length of stay.

Patients: A total of 244 septic shock patients were randomized (119 to HES, 125 to NaCl).

Interventions: Randomization to HES or NaCl for volume resuscitation.

Measurements: ICU length of stay (LOS), hospital LOS, ICU mortality, in-hospital mortality, and 1-year mortality, incidence of acute kidney injury (AKI), fluid balance, trajectory of inflammatory parameters, lactate, and SOFA (sequential organ failure assessment) score.

Main results: Baseline characteristics and ICU management were comparable among groups. ICU LOS was not significantly shorter with HES (hazard ratio, HR, for ICU discharge alive HES vs. NaCl, 1.11,95% CI 0.83 to 1.49). However, the HES group showed a significantly reduced hospital LOS (HR HES vs. NaCl: 1.48, 95 % CI 1.09 to 2.01, p=0.011). HES did not significantly affect other laboratory values, mortality or renal function.

Conclusions: The results of the BaSES trial neither do call for complete suspension of HES from the market, nor do they support a role of HES in volume resuscitation in septic shock.

Trial registration: ClinicalTrials.gov Identifier: NCT00273728

Biological sciences/Physiology/Kidney

Biological sciences/Physiology/Metabolism

Health sciences/Diseases

Health sciences/Medical research

Health sciences/Nephrology

AKI

HES

acute kidney injury

Volume resuscitation is the mainstay of treatment in most types of shock ^{1 2}, especially in haemorrhagic and septic shock. Septic shock is a main cause of morbidity and mortality in Switzerland and worldwide ³.

More than 50% of septic patients suffer from acute kidney injury (AKI). Sepsis thus represents a leading cause of AKI, which often marks beginning multi-organ failure (MOF) ^{4 5}. Septic capillary leakage and microcirculatory impairment due to inflammation-induced injury in capillaries, relatively low perfusion pressure, and high outflow pressure may harm renal function by a decreased glomerular filtration rate. The development of AKI is associated with high in-hospital mortality. ⁶

Although there has been a lot of research to evaluate optimal amount and composition of fluids for volume resuscitation (e.g. colloids, balanced crystalloids, normal saline; red blood cell transfusion, albumin, fresh frozen plasma) particularly in the past two decades ^{7 8}, results remain inconclusive and to some extent contradictive. This includes Hydroxyethyl starch (HES) where further impairment of kidney function and haemostasis is feared ^{9 10}. On the other hand, in patients resuscitated with HES more efficient restoration of circulating blood volume with less interstitial fluid accumulation compared to different crystalloids was observed ¹¹. Moreover, an almost equivalent volume effect can be expected with much less cost compared to albumin ¹¹.

With the BaSES (Basel Starch Evaluation in Sepsis) trial we investigated the impact of volume resuscitation with HES in combination with Ringer's lactate (RL) compared to normal saline (NaCl) in combination with RL in the early phase of septic shock on ICU length of stay (LOS), hospital LOS, mortality and kidney function. We hypothesized that resuscitation with colloids in addition to a balanced crystalloid solution when compared to crystalloids and normal saline would reduce the total amount of fluids with consecutive reduction of ICU and hospital LOS without impairment of renal function.

Trial design

The Basel Starch Evaluation in Sepsis (BaSES) trial was an investigator-initiated, single-centre (but including two separate ICUs, i.e., medical and surgical ICU of the University Hospital Basel), quadruple-blind, parallel group, randomised, controlled trial on volume resuscitation in patients suffering from sepsis or septic shock. The study protocol, which is in accordance with Good Clinical Practice standards and the tenets of the Declaration of Helsinki 2002, was approved by the local ethical committees and was registered on ClinicalTrials.gov (NCT00273728). Patients were included from April 21, 2005, to January 25, 2011 Participating patients were randomly allocated in a 1:1 ratio to either HES (molecular weight 130; 0.42, Voluven®) or NaCl (0.9%) for volume resuscitation upon six days after diagnosis of sepsis and consecutive study inclusion. The patient, the treating medical team (i.e., responsible nurse and physician), the investigator, and the outcome assessor were blinded to treatment group.

Institutional Review Boards

All experiments were performed in accordance with relevant guidelines and regulations. The institutional committee approving the experiments (i.e., blood withdrawal) was the University Hospital Basel. No additional blood sampling was done for the study since blood is withdrawn in our ICU on daily basis. The study protocol was approved by the local ethical committee (Ethics Committee reference number 2004/04, approval date 15.11.2004, study title: BaSES_{(130; 0.4)}Trial -BaselStarchEvaluation inSepsis Trial) and by Swissmedic (reference number 2005 DR 3123, approval date 09.05.2005; study title: BaSES_{(130; 0.4)}Trial -BaselStarchEvaluation inSepsis Trial). In any possible case informed consent was obtained from all subjects and/or their legal guardian(s).

Eligibility criteria

Briefly, all adult patients admitted to the medical or surgical ICU of the University Hospital Basel, Switzerland, diagnosed with sepsis were screened for eligibility. The inclusion criteria were age ≥ 18 years and confirmed infection, fulfilling two out of six defined criteria: 1) body temperature below 36°C or above 38.3°C, 2) heart rate above 90/min, 3) respiratory rate > 20/min or arterial pCO₂ below 4.25kPa, 4) white blood cell count higher than 12.000 per microliter or below 4.000 per microliter or more than 10% of immature white cells, 5) systolic blood pressure < 90mmHg or mean arterial pressure < 65mmHg at admission to ICU, and 6) altered mental status or oliguria

The exclusion criteria for the BaSES trial were pregnancy, age below 18 years, known allergy against hydroxyethyl starch or previously known chronic kidney disease (CKD).

Study groups and follow-up

Patients allocated to one of the two trial arms were resuscitated using a strict alternating regimen of 1000mL study infusion (HES or NaCl) and 1000mL RL. Thereby, a maximum dose of study solution of 50ml/kg body weight over 24 hours was allowed during the first 6 days after study inclusion. Thereafter, volume resuscitation was managed upon discretion of the treating physician.

Patients were managed according to ICU standard (i.e., continuous monitoring and maintenance of vital functions). To assess 1-year mortality upon ICU admission patients were followed by telephone interview.

Definition of sepsis and septic shock

Included patients were sub-classified into severe sepsis and septic shock based on definitions for sepsis and organ failure of Sepsis-2 from 2001 ¹². In the present manuscript the term “sepsis” refers to the updated definition of Sepsis-3 from 2016 ¹³.

Outcomes

The primary endpoint of the BaSES trial was ICU length of stay (LOS). Secondary study endpoints were hospital LOS, ICU mortality, in-hospital mortality, and 1-year mortality, incidence of acute kidney injury (AKI), fluid balance, trajectory of inflammatory parameters, lactate, and SOFA (sequential organ failure assessment) score ¹⁴. Renal replacement therapy (RRT) was defined as additional secondary endpoint post-hoc, according to the results of an ancillary study ¹⁵ from the 6S trial ¹⁶.

Statistical methods

The sample size was calculated based on an expected mean difference in ICU LOS of 1 day, a standard deviation of 2.5 days, and a power of 80% at a two-sided significance level of 5%. Assuming normally distributed ICU LOS (for a two-sample t-test), this would result in 100 patients per group. We then added 15% in order to have a sufficient sample size for a non-parametric Wilcoxon test, based on the asymptotic relative efficiency compared to the t-test ¹⁷, which resulted in 118 patients per group (236 in total). We rounded this number to 240 to allow for some drop-out.

All analyses were performed using the R system for statistical computing and graphics (R Core team, 2022 ¹⁸). ICU LOS was calculated as hours between ICU admission and discharge alive and was analysed with a cause-specific Cox proportional hazards model with treatment as explanatory variable. Death during ICU stay was considered as competing risk, i.e., patients who died were considered as right-censored in the Cox model. Because all patients either died or were discharged alive (both from ICU or hospital), no further censoring occurred. In line with the intention-to-treat principle, the analysis was done using all randomized patients. Hazard ratio (HR) estimates and 95% confidence intervals (CI) for the effect of treatment (HES vs. NaCl) are reported together with p-values to test the null hypothesis that the HR equals 1 (Wald test). Time to ICU discharge or death are graphically displayed by treatment group using cumulative incidence curves. As sensitivity analyses, the cause-specific Cox model was once applied to per protocol set and once using study entry as time origin instead of ICU admission. Two subgroup analyses were performed for the primary endpoint, comparing the effect of HES between patients under shock (blood pressure Riva-Rocci syst < 90 at study entry) vs. patients not under shock as well as patients with high (> 120) vs. low levels (≤ 120) of serum creatinine at study entry. We added “shock” or “high creatinine” to the Cox model together with the corresponding interaction with treatment. The secondary endpoint hospital LOS was calculated as days between ICU entry and hospital discharge alive and was analysed like ICU LOS with (in-hospital) death as competing risk. Death within one year from ICU admission was analysed with a Cox proportional hazards model with censoring after 365 days or at the last contact (if before 365 days). AKI (binary) was analysed with a generalised linear model with binomial error distribution and logit link function. The continuous secondary endpoint fluid balance was analysed with linear models. For serum creatinine, C-reactive protein and procalcitonin we calculated the area under the curve (AUC) using the composite trapezoid rule over all available measurements on days 1–6. AUC was then analysed as continuous outcome, with the respective baseline measurement as covariate (in addition to treatment). The secondary outcomes creatinine clearance, urea excretion, lactate and SOFA score were analysed as repeated measurements (on day 1, 3, 5 and 6, lactate on days 1–6) using linear mixed-effects models with a random intercept per patient and treatment, the respective baseline value, day of the measurement (numeric, 1–6), as well as the interaction between treatment and day of the measurement as fixed explanatory variables. To account for temporal autocorrelation of residuals, an autoregressive process of order 1, AR(1), was modelled. All secondary endpoint analyses were done using all n = 244 patients. However, missing values were not imputed, resulting in < 244 patients in several analyses.

A total of 244 patients were included in the BaSES trial, of which 119 were randomized to HES and 125 to NaCl. Five patients had major protocol violations and were excluded from the PP set (Fig. 1). Baseline characteristics were comparable among groups, with slight imbalance regarding the origin of sepsis (e.g., less pneumonia, more peritonitis) and slightly lower platelet counts but higher procalcitonin levels in HES patients. ICU management was also similar among groups, except for a lower proportion of non-invasive ventilation in the HES group (Table 1).

Table 1

**Demographic and baseline characteristics of study patients upon ICU admission**
	All n = 244	HES n = 119	NaCl 0.9% n = 125	MV (%)
Demographics
Age (years)	64.8 ± 15.0	63.6 ± 16.2	65.85 ± 13.7	0.0
Females (n, %)	89 (36.5)	47 (39.5)	42 (33.6)	0.0
Origin of sepsis				0.0
Meningitis	4 (1.7)	3 (2.6)	1 (0.8)
Pneumonia	79 (32.2)	32 (27.6)	47 (38.5)
Peritonitis	33 (13.9)	18 (15.5)	15 (12.3)
Urinary tract	25 (10.5)	13 (11.2)	12 (9.8)
Endocarditis	3 (1.3)	1 (0.9)	2 (1.6)
Unknown	31 (13.0)	16 (13.8)	15 (12.3)
Other	63 (26.5)	33 (28.4)	30 (24.6)
Positive blood cultures	104 (42.6)	52 (43.7)	52 (41.6)	0.0
Medical history**
CAD	71 (29.1)	35 (29.4)	36 (28.8)	0.0
Chronic Heart Failure	27 (11.1)	13 (10.9)	14 (11.2)	0.0
COPD	38 (15.6)	21 (17.6)	17 (13.6)	0.0
Active/recent malignant tumors	66 (27.0)	29 (24.4)	37 (29.6)	0.0
Hematological disease	21 (8.6)	11 (9.2)	10 (8.0)	0.0
Laboratory values at admission
Na (mmol/L)	137 [134,141]	137 [134,140]	138 [134,142]	3.7
K (mmol/L)	4.0 [3.6,4.6]	4.0 [3.4,4.7]	4.1 [3.6,4.6]	4.1
Cl (mmol/L)	105 [100,109]	105 [101,108]	105 [100,109]	4.9
Osmolality (mmol/kg; mOsm/kg)	300 [291,313]	299 [293,309]	303 [291,315]	25.4
Lactate (mmol/l)	2.3 [1.3,4.6]	2.6 [1.4,4.6]	2.3 [1.2,4.5]	9.0
Arterial pH	7.34 [7.23,7.41]	7.33 [7.24,7.39]	7.36 [7.23,7.41]	5.3
paO₂ (mmHg)	100 [73,132]	98 [72,132]	104 [75,131]	8.6
ScvO₂ (%)	73 [65,80]	74 [65,81]	72 [64,80]	17.2
Hematocrit (vol%)	29 [26, 34]	29 [26, 34]	29 [26, 34]	6.1
Platelets (10⁹/L)	198 [113,280]	183 [101,249]	212 [127,290]	5.7
Creatinine (µmol/L)	128 [87,186]	132 [87,202]	128 [89,182]	4.5
Creatinine > 120µmol/L (%)	126 (54.1)	59 (53.2)	67 (54.9)	4.5
CrCl (ml/min/1.73m²)	47 [28,76]	47 [26,76]	48 [30,76]	4.1
Urea (mg/dL)	11 [7, 19]	11 [7, 19]	11 [7, 18]	4.9
FEUrea	24.6 [16.4,37.6]	23.8 [14.1,40.1]	24.7 [18.6,36.3]	31.1
Albumin (g/L)	19 [14, 23]	19 [14, 24]	19 [14, 22]	5.3
CRP (mg/L)	194 [122,273]	191 [123,273]	194 [121,276]	5.3
PCT (ng/mL)	12.5 [3.3,37.2]	13.4 [3.8,37.8]	11.5 [3.1,34.4]	20.5
White blood count (per mm³)	12185 [5988,19550]	11910 [5690,19060]	12320 [6630,19570]	5.7
BNP (pg/mL)	560 [245,1063]	481 [212,969]	595 [264,1234]	19.7
Management
Furosemide on day one (mg)	0 [0,0]	0 [0,0]	0 [0,0]; n = 109	13.1
Organ dysfunction scores
SOFA (points)	8 [4, 11]	8 [4, 12]	8 [5, 11]; n = 115	8.2
SAPS II (points)	48 [32,62]	48 [36,64]	48 [30,62]; n = 115	7.4
Mean ± standard deviation or median [interquartile range] are presented for continuous variables and frequency (percentage) for categorical variables. Hematological disease, haemato-oncological malignancies; BNP, brain natriuretic peptide; CAD, coronary artery disease; CrCl; creatinine clearance, assessed by the Cockcroft-Gault Equation; CRP, C-reactive protein; FEUrea, fractional excretion of urea; HES, Hydroxyethyl starch 130/0.42, Voluven®); MV, missing values; PCT, procalcitonin; SAPS, simplified acute physiology score; SOFA, sequential organ failure assessment.

Primary endpoint

There was no statistically significant effect of treatment with HES vs. NaCl on ICU LOS, analysed as time to discharge alive from the ICU with death as competing risk (HR HES vs. NaCl: 1.11, 95% CI 0.83 to 1.49, p = 0.49; Table 2, Fig. 2). The results of the sensitivity analyses were similar.

Table 2

**Patient management characteristics during ICU stay**.
	All n = 244	HES n = 119	NaCl 0.9% n = 125	MV (%)
Intubation (n, %)	122 (50)	63 (52.9)	59 (47.2)	0
Non-invasive ventilation (n, %)	89 (36.5)	36 (30.3)	53 (42.4)	0
PEEP (max, mbar)	8 [5,, 12]; n = 146	10 [5, 12]; n = 70	8 [5, 10]; n = 76	40.2
Noradrenalin (max, µg/kg/min)	11 [0,33]; n = 221	11 [0,33]; n = 108	11 [0,32]; n = 113	9.4
Adrenalin (max, µg/kg/min)	0 [0,0]; n = 219	0 [0,0]; n = 107	0 [0,0]; n = 112	10.2
Dobutamin (max, µg/kg/min)	0 [0,300]; n = 220	0 [0,300]; n = 107	0 [0,200]; n = 113	9.8
Fluid amount, (total, ml)	3775 [2000,6403,]; n = 240	3526 [1946, 5667]; n = 116	4000 [2388, 6540]; n = 124	1.6
RL (total, ml)	5649 [3080, 9700]; n = 227	5313 [3060, 8730); n = 109	5810 [3246, 9885]; n = 118	7.0
Fluid and RL (total, ml)	10021 [5663, 15975]	9395 [5600, 14564]	10934 [5850, 17068]	7.0
EC transfusion (total, bags)	1 [0, 2]; n = 227	0 [0, 2]; n = 109	1 [0, 2]; n = 118	7.0
Platelet transfusion (total, bags)	0 [0,0]; n = 224	0 [0, 0]; n = 106	0 [0, 0]; n = 118	8.2
FFP transfusion (total, bags)	0 [0,0]; n = 223	0 [0, 0]; n = 106	0 [0, 0]; n = 117	8.6
Median [interquartile range] are presented for continuous variables and frequency (percentage) for categorical variables. AUC, area under the curve; CrCl, creatinine clearance, assessed by the Cockcroft-Gault Equation; the term “fluid” refers to the total amount of study-specific volume given (HES or NaCl); the term “fluid and RL” refers to the sum of the amount of study-specific volume and RL; CRP, C-reactive protein; EC, erythrocyte; FEUrea, fractional excretion of urea; FFP, fresh frozen plasma; Fluid amount, amount of study-specific fluid administered; HES, Hydroxyethyl starch 130/0.42, Voluven®); max, maximum value during the investigated period; PCT, procalcitonin; RL, Ringer’s lactate.

Subgroup analyses of patients with shock vs. no shock and with high vs. low serum creatinine levels did not reveal significant differences in the effect of HES between subgroups, neither on the time to discharge alive from the ICU nor on death on the ICU (data not shown).

Secondary endpoints

Hospital LOS and mortality

Hospital LOS, analysed as time to discharge alive from the hospital with death as competing risk was significantly shorter in the HES group compared to the NaCl group (HR HES vs. NaCl 1.48, 95% CI 1.09 to 2.01, p = 0.011; Fig. 3, Table 2), with median time to hospital discharge estimated as 34 days with HES vs. 50 days with NaCl (based on the cumulative incidence function, as for the curves in Fig. 3). Of the 244 patients, 64 patients died during ICU stay (26.2%) and eleven patients died during the remaining hospital stay, resulting in 75 patients who died during hospital stay (30.7%; Table 2). Among the 169 hospital survivors, a total of 15 patients died within 365 days from study inclusion, resulting in 90 patients (36.9%) who died within a year from ICU admission. Mortality did not differ significantly between trial arms, neither during ICU or hospital stay nor during one year of follow-up (Table 3).

Table 3

**Outcome variables: descriptive statistics and estimates for the effect of HES vs. NaCl.**
	All n = 244	HES n = 119	NaCl 0.9% n = 125	Effect size estimate (95% CI)	p-value	N in analysis
Length of stay (LOS)
ICU LOS (hours), primary outcome	121 [57, 267]	121 [55, 276]	123 [62, 247]	HR 1.11 (0.83 to 1.49)¹	0.49	244
Hospital LOS (days)	28[16, 46]	23 [14–35]	33 [19, 50]	HR 1.48 (1.09 to 2.01)¹	0.011	244
Mortality
Death during ICU stay	64 (26.2)	29 (24.4)	35 (28)	HR 0.93 (0.57 to 1.52)¹	0.77	244
Death during hospital stay	75 (30.7)	35 (29.4)	40 (32)	HR 1.12 (0.71 to 1.76)¹	0.64	244
Death during 1-year follow-up	90 (36.9)	43 (36.1)	47 (37.6)	HR 0.96 (0.63 to 1.45)	0.843	244
Renal function
AUC Serum creatinine (µmol/L)	283 [145–521]	283 [165–467]	284 [129–566]	-41.2 (-124.8 to 42.4)²	0.332	218
AUC CRP (mg/L)	451 [227–864]; n = 225	384 [214–723]; n = 107	494 [264–959]; n = 118	-131.9 (-246.0 to -17.7)²	0.0237	216
AUC PCT (ng/mL)	23.3 [6.6–59.9]; n = 207	23.7 [6.8–54.5]; n = 98	22.9 [6.6–70.6]; n = 109	6.52 (-24.51 to 37.54)²	0.679	172
AKIN (n, %)	9 (4.1)	3 (2.9)	6 (5.2)	OR 0.56 (0.12 to 2.16)	0.415	218
Need for RRT within 1 year (%)	57 (23.4)	30 (25.2)	27 (21.6))	OR 1.22 (0.68 to 2.23)	0.506	244
Fluid balance days 1–6 (ml)	6343 [3312–11179]	5650 [3273–11069]	6825 [3399–11275]	-544.6 ( -2661.1 to 1572.0)	0.613	224
Reduction of body weight³	0.34 [0.16–0.60]	0.32 [0.17–0.60]	0.38 [0.16–0.59]	-0.04 ( -0.22 to 0.14)	0.669	80
CrCl (ml/min/1.73m²), day 1	56.8 [35.3–88.8]; n = 220	54.9 [33.9–82.7]; n = 104	58.2 [35.8–95.5]; n = 116	-6.45 (-14.98 to 2.08)⁴	0.138	217
CrCL (ml/min/1.73m²), day3	65.6 [37.8, 98.6], n = 185	63.2 [39.8, 97.2], n = 97	67.6 [35.8, 95.5], n = 88
CrCl (ml/min/1.73m²), day 6	71 [47–110]; n = 92	69 [47–113]; n = 39	72 [48–107]; n = 53
FEUrea, day 1	36.4 [25.3–45.5]; n = 144	36.1 [26.4–47.0]; n = 72	36.8 [25.1–44.8]; n = 72	0.24 (-3.95 to 4.43)⁴	0.9101	121
FEUrea day 3	38.0 [25.6, 47.2]; n = 84	30.3 [22.1, 46.8]; n = 47	40.6 [30.8, 47.5]; n = 37
FEUrea, day 6	38.6 [32.0-46.4]; n = 48	42.8 [35.7–48.8]; n = 18	37.8 [31.1–45.7]; n = 30
Metabolism
Lactate (mmol/l), day 1	1.7 [1.1–3.1]; n = 215	1.9 [1.0-3.8]; n = 104	1.5 [1.1–2.9]; n = 111	0.44 (-0.26 to 1.15)⁴	0.219	209
Lactate (mmol/l), day 3	1.3 [0.9–1.7]; n = 133	1.4 [0.9-2.0]; n = 62	1.2 [0.9–1.5]; n = 71
Lactate (mmol/l), day 6	1.3 [1.0-1.6]; n = 83	1.4 [1.0-1.7]; n = 35	1.2 [0.9–1.4]; n = 48
Disease severity
SOFA day 1 (points)	8 [4–11]; n = 209	7 [4–13]; n = 101	8 [4–11]; n = 108	0.15 (-0.69 to 0.99)⁴	0.723	207
SOFA day 3 (points)	6 [3–12]; n = 175	6 [3–12]; n = 85	6 [2–11]; n = 90
SOFA day 6 (points)	7 [4–10]; n = 89	8 [4–12]; n = 42	6 [4–9]; n = 47
Median [interquartile range] are presented for continuous variables and frequency (percentage) for categorical variables. Effects sizes were estimated using the statistical models described in the Statistical methods section and are given together with corresponding 95% confidence intervals, p-values, and number of patients included in the analysis (N). AKIN, acute kidney injury; ICU, intensive care unit.
¹ ICU and hospital length of stay were analyzed as time to discharge alive with death during ICU stay and death during hospital stay as competing risk. Descriptive statistics are given for LOS of patients discharged alive only.
² Estimate adjusted for baseline measurement
³ Reduction in body weight during the stay at the hospital after ICU discharge as quotient (body weight at ICU discharge – body weight at hospital discharge)/length of stay at the hospital after ICU discharge ; >50% missing values.
4 Estimated from linear mixed-effects models on measurements on days 1, 3, 5 and 6 or days 1–6 (for lactate only) with a random intercept per patient and the baseline measurement and time (study day) as covariate in addition to treatment (HES vs NaCl).

The subgroup analyses of patients with shock vs. no shock and with high vs. low serum creatinine levels (> 120µmol/L vs. 120 < µmol/L) did not reveal a difference in the effect of HES between subgroups, neither on the time to discharge alive from the ICU nor on death on the ICU.

Markers of kidney function

HES had no significant effect on the AUC of serum creatinine when adjusted for serum creatinine at baseline, but the latter was strongly positively associated with the AUC of serum creatinine. HES neither significantly affected creatinine clearance nor fractional excretion of urea (FEUrea), both analysed as repeated measurements, nor the time course of these endpoints (no significant interactions between treatment and time). Again, the corresponding measurements at baseline were strongly and positively associated with measurements on days one to six. Moreover, creatinine clearance and FEUrea increased over time independent of the trial arm. AKI was comparable among groups. Unfortunately, there were more than 50% missing values for FEUrea already by day one on the ICU.

Overall, 57 patients (23.4%) underwent RRT, 30 (25.2%) within the HES group, and 27 (21.6%) within the NaCl group (OR HES vs. NaCl 1.22, 95% CI 0.68 to 2.23, p = 0.506; Table 2). Of note, three patients included in BaSES (1 with HES, 2 with NaCl) required renal replacement therapy after the one-year follow-up.

Finally, no significant difference could be shown for fluid balance and reduction in body weight. There were more than 50% of missing values for reduction of body weight.

Inflammatory and metabolic markers, SOFA score

HES significantly reduced the AUC of CRP levels (estimated difference in AUC HES vs. NaCl − 131.884, 95% CI from − 246.019 to -17.748], p = 0.0237), while it had no significant effect on the AUC of PCT. Baseline measurements showed strong positive association with the AUC for both measures. HES did not significantly affect lactate, SOFA score (all analysed as repeated measurements), or the time course of these endpoints (no significant interactions between treatment and time). Also here, the corresponding measurements at baseline were strongly and positively associated with measurements on days 1–6. Moreover, lactate and the SOFA score decreased over time, independent from the trial arm.

The BaSES (Basel Starch Evaluation in Sepsis) trial investigated the impact of volume resuscitation with HES (Hydroxyethyl starch 130/0.42, Voluven®) in combination with RL (Ringer's lactate) compared to normal saline (NaCl 0.9%) in combination with RL in the early phase of sepsis and septic shock on ICU length of stay (LOS), hospital LOS, mortality, kidney function, inflammatory and metabolic parameters, as well as on SOFA score. Based on our results for the primary outcome and most secondary outcomes, the hypothesis that HES is superior to normal saline for volume resuscitation in sepsis was not confirmed. However, patients in the HES group stayed significantly shorter in the hospital. Further, we found significantly reduced CRP levels in the HES group, but without a significant effect on PCT. Importantly, baseline SOFA and SAPS II scores were comparable among groups. Moreover, gender ratio in BaSES was typical compared to other studies reporting two thirds of male patients within their ICU population ^19–21, except for studies reporting on sepsis ^{22 23}.

Safety of HES administration in the critically ill in the context of kidney function in particular was questioned based on the results of several studies ^24–26. In addition, negative effects of positive fluid balance on kidney function due to organ oedema have been discussed in many studies ^27–31. According to the revised International Guidelines for Management of Sepsis and Septic Shock of 2016, the administration of Hydroxyethyl starches (HES) meanwhile is contraindicated in case of septic shock, burn injury ³², acute or chronic kidney injury, and in the management of short-term oliguria unresponsive to fluids ³³. Interestingly, our RCT did not reveal an increased incidence of AKI in the HES group ^{9 10}, and the number of patients included in BaSES requiring RRT within the one-year follow-up period was comparable among groups. Possible explanations could be that HES was always given with RL in an alternating manner, and/or because the colloidal volume effect, which would even be accentuated by impaired kidney function ³⁴ was prolonged because HES was given slowly. Strikingly, the administration of HES did not reduce the amount of volume given since there was no difference of fluid balance and reduction in body weight among groups. Moreover, the three patients included in BaSES requiring renal replacement therapy after the one-year follow-up are difficult to connect to an influence based on the study regime.

The effect of HES on CRP levels remains unclear. A protective effect of HES on splanchnic microcirculation was proposed in several studies ³⁵. A study published in 2005 comparing HES to gelofusine suggested damping of the systemic inflammatory response and reduction of endothelial dysfunction with the administration of HES ³⁶, which was confirmed in cases of acute severe pancreatitis ³⁷. Of note, CRP and PCT are two very different biomarkers with different properties. CRP is an acute phase protein that is synthesised in the liver under the stimulation of IL-6 during infections or other inflammatory conditions. PCT is the precursor protein of calcitonin, which is normally synthesized by the C-cells of the thyroid gland, but in response to bacterial infections, multiple cell types throughout the body produce PCT. According to our data, PCT levels were comparable, making a direct effect on the inflammatory response unlikely or which could be explained by the differing production pathways, respectively.

Additional adverse effects associated with HES due to tissue storage and coagulopathy were reported ³⁸. The previously described coagulopathy in the context of HES administration leading to increased risk of bleeding thus an increased need for blood products in sepsis ³⁹ underlines the results of the BaSES trial with a significantly higher platelet transfusion rate in the HES group. However, sample size was not calculated for this endpoint, and the amounts of transfused erythrocytes and fresh frozen plasma were similar among groups.

The results of the BaSES trial do not support the complete suspension of HES from the market especially in this context to avoid further discrimination of middle-income and low-income countries ⁴⁰. However, current data do not support reliable colloid alternatives when a more efficient restoration of circulating blood volume with less interstitial fluid accumulation is warranted ^{41 42}.

We acknowledge that the study has several limitations. First, it should be noted in general, that all secondary endpoints with the exception of hospital LOS, in-hospital mortality, and 1-year mortality had missing values. For some secondary endpoints (i.e., reduction in body weight, FEUrea) the proportion of missing values even exceeded 50%. Second, we only included patients with sepsis and septic shock and results cannot be generalized to a general ICU population. Third, data were assessed in two ICUs only. Strengths are the fact that this was a prospective, randomized controlled study and that 1-year mortality was assessed as a long-term outcome.

According to our results, the administration of HES indeed does not go along with a clear advantage for the critically ill, but just as little seems to be an obvious threat when managing vulnerable patients. The reduction in hospital LOS suggests that HES may be kept as an option for the choice of fluids to manage volume resuscitation in ICU patients when given as proposed in our trial (i.e.g, alternating regimen of 1000mL HES and 1000mL RL, respectively).

The results of the BaSES trial neither call for complete suspension of HES from the market, nor support that HES is needed for volume resuscitation today. It remains unclear which colloid to choose when a more efficient restoration of circulating blood volume with less interstitial fluid accumulation could be beneficial for the patient, or if colloids should not be used at all.

Funding

BaSES (ClinicalTrials.gov Identifier identifier NCT00273728) was funded by the University Hospital Basel, Switzerland. HES was provided by Fresenius Kabi, Switzerland.

Conflicts of interest/Competing interests

Martin Siegemund was supported by Fresenius with speaker honoraria and the HES supply used in the trial. The other authors report no conflicts of interest.

Ethics

The study was conducted in Switzerland. The study protocol was approved by the local ethical committee (Ethics Committee reference number 2004/04, approval date 15.11.2004, study title: BaSES_{(130; 0.4)} Trial - Basel Starch Evaluation in Sepsis Trial) and by Swissmedic (reference number 2005 DR 3123, approval date 09.05.2005; study title: BaSES_{(130; 0.4)} Trial - Basel Starch Evaluation in Sepsis Trial) and conducted in accordance with Directive 2001/20/EC, as well as Good Clinical Practices (I.C.H. version 4 of May 1, 1996 and Decision of November 24, 2006) and the declaration of Helsinki. Patients were included from April 18, 2005, to August 23, 2007. No Institutional Review Board (IRB) was involved.

Consent to participate

All patients included in this trial gave written consent for participation.

Availability of data and material

Prof Martin Siegemund owns the data and may provide access to all data in the study. He takes responsibility for the integrity of the data and the accuracy of the data analysis.

Authors' contributions (alphabetical)

Study concept, planning, design and conduction: Siegemund

Acquisition of data: Siegemund, Steiner, Pargger, Marsch, Balestra

Analysis and interpretation of data: Hollinger, Siegemund, von Felten

Drafting of the manuscript: Hollinger, von Felten

Critical revision of the manuscript for important intellectual content: all declared authors

Statistical analysis: von Felten

Obtained funding: Siegemund

Administrative, technical, or material support: Siegemund

Study supervision: Siegemund

All authors have read and approved the final manuscript.

Patient / public involvement

Patients or the public were not involved in the design of our research, and will not be involved in the conduct, or reporting, or dissemination of our research.

Acknowledgements

The authors are particularly grateful to the nurses of the medical and surgical intensive care unit of the University Hospital Basel who supported and coordinated organizational aspects of the study.

Carlsen S, Perner A, East Danish Septic Shock Cohort I. Initial fluid resuscitation of patients with septic shock in the intensive care unit. Acta anaesthesiologica Scandinavica 2011;55(4):394-400. doi: 10.1111/j.1399-6576.2011.02399.x
Rochwerg B, Alhazzani W, Gibson A, et al. Fluid type and the use of renal replacement therapy in sepsis: a systematic review and network meta-analysis. Intensive care medicine 2015 doi: 10.1007/s00134-015-3794-1
Micek ST, McEvoy C, McKenzie M, et al. Fluid balance and cardiac function in septic shock as predictors of hospital mortality. Critical care 2013;17(5):R246. doi: 10.1186/cc13072
Lafrance JP, Miller DR. Acute kidney injury associates with increased long-term mortality. J Am Soc Nephrol 2010;21(2):345-52. doi: 10.1681/ASN.2009060636
Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a multinational, multicenter study. Jama 2005;294(7):813-8. doi: 10.1001/jama.294.7.813
Doyle JF, Forni LG. Acute kidney injury: short-term and long-term effects. Critical care 2016;20(1):188. doi: 10.1186/s13054-016-1353-y
Leone M. Septic shock resuscitation: assembling the puzzle*. Critical care medicine 2014;42(10):2294-5. doi: 10.1097/CCM.0000000000000536
Malbrain M, Van Regenmortel N, Saugel B, et al. Principles of fluid management and stewardship in septic shock: it is time to consider the four D's and the four phases of fluid therapy. Ann Intensive Care 2018;8(1):66. doi: 10.1186/s13613-018-0402-x
Haase NR. Hydroxyethyl starch in sepsis. Dan Med J 2014;61(1):B4764.
Zarbock A, Buhre W. Hydroxyethyl Starch in the Perioperative Period: Friend, Foe, or Still an Unsolved Issue? Anesthesia and analgesia 2022;134(4):683-85. doi: 10.1213/ANE.0000000000005903 [published Online First: 2022/03/18]
Toyoda D, Shinoda S, Kotake Y. Pros and cons of tetrastarch solution for critically ill patients. J Intensive Care 2014;2(1):23. doi: 10.1186/2052-0492-2-23
Levy MM, Fink MP, Marshall JC, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Critical care medicine 2003;31(4):1250-6. doi: 10.1097/01.CCM.0000050454.01978.3B
Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). Jama 2016;315(8):801-10. doi: 10.1001/jama.2016.0287
Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive care medicine 1996;22(7):707-10.
Muller RB, Haase N, Lange T, et al. Acute kidney injury with hydroxyethyl starch 130/0.42 in severe sepsis. Acta anaesthesiologica Scandinavica 2015;59(3):329-36. doi: 10.1111/aas.12453
Perner A, Haase N, Guttormsen AB, et al. Hydroxyethyl starch 130/0.42 versus Ringer's acetate in severe sepsis. N Engl J Med 2012;367(2):124-34. doi: 10.1056/NEJMoa1204242 [published Online First: 2012/06/29]
Lehmann EL. Nonparametrics: Statistical Methods Based on Ranks. 2006:76-81.
R Core Team (2022). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.
Hollinger A, Gayat E, Feliot E, et al. Gender and survival of critically ill patients: results from the FROG-ICU study. Ann Intensive Care 2019;9(1):43. doi: 10.1186/s13613-019-0514-y
Zettersten E, Jaderling G, Bell M, et al. Sex and gender aspects on intensive care. A cohort study. Journal of critical care 2020;55:22-27. doi: 10.1016/j.jcrc.2019.09.023
Vezzani A, Mergoni M, Orlandi P, et al. Gender differences in case mix and outcome of critically ill patients. Gend Med 2011;8(1):32-9. doi: 10.1016/j.genm.2010.11.011
Sunden-Cullberg J, Nilsson A, Inghammar M. Sex-based differences in ED management of critically ill patients with sepsis: a nationwide cohort study. Intensive care medicine 2020;46(4):727-36. doi: 10.1007/s00134-019-05910-9
Mathieu C, Leone M. Gender and sepsis: first step of personalized medicine? Minerva anestesiologica 2018;84(4):434-36. doi: 10.23736/S0375-9393.18.12625-3
Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J Med 2008;358(2):125-39. doi: 10.1056/NEJMoa070716
Schortgen F1 LJ, Bruneel F, Cattaneo I, Hemery F, Lemaire F, Brochard L. Effects of hydroxyethylstarch and gelatin on renal function in severe sepsis: a multicentre randomised study. Lancet 2001;357(9260):911-6.
Myburgh JA, Finfer S, Bellomo R, et al. Hydroxyethyl starch or saline for fluid resuscitation in intensive care. N Engl J Med 2012;367(20):1901-11. doi: 10.1056/NEJMoa1209759
Bouchard J, Soroko SB, Chertow GM, et al. Fluid accumulation, survival and recovery of kidney function in critically ill patients with acute kidney injury. Kidney Int 2009;76(4):422-7. doi: 10.1038/ki.2009.159 [published Online First: 2009/05/14]
Payen D, de Pont AC, Sakr Y, et al. A positive fluid balance is associated with a worse outcome in patients with acute renal failure. Critical care 2008;12(3):R74. doi: 10.1186/cc6916 [published Online First: 2008/06/06]
Fulop T, Pathak MB, Schmidt DW, et al. Volume-related weight gain and subsequent mortality in acute renal failure patients treated with continuous renal replacement therapy. ASAIO J 2010;56(4):333-7. doi: 10.1097/MAT.0b013e3181de35e4 [published Online First: 2010/06/19]
Grams ME, Estrella MM, Coresh J, et al. Fluid balance, diuretic use, and mortality in acute kidney injury. Clin J Am Soc Nephrol 2011;6(5):966-73. doi: 10.2215/CJN.08781010 [published Online First: 2011/03/12]
Murugan R, Kellum JA. Fluid balance and outcome in acute kidney injury: is fluid really the best medicine? Critical care medicine 2012;40(6):1970-2. doi: 10.1097/CCM.0b013e31824e1a1f [published Online First: 2012/05/23]
Bechir M, Puhan MA, Neff SB, et al. Early fluid resuscitation with hyperoncotic hydroxyethyl starch 200/0.5 (10%) in severe burn injury. Critical care 2010;14(3):R123. doi: 10.1186/cc9086
Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Intensive care medicine 2017;43(3):304-77. doi: 10.1007/s00134-017-4683-6
Hahn RG, Lyons G. The half-life of infusion fluids: An educational review. European journal of anaesthesiology 2016;33(7):475-82. doi: 10.1097/EJA.0000000000000436 [published Online First: 2016/04/09]
Mahmood A, Gosling P, Barclay R, et al. Splanchnic microcirculation protection by hydroxyethyl starches during abdominal aortic aneurysm surgery. European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery 2009;37(3):319-25. doi: 10.1016/j.ejvs.2008.11.003
Rittoo D, Gosling P, Simms MH, et al. The effects of hydroxyethyl starch compared with gelofusine on activated endothelium and the systemic inflammatory response following aortic aneurysm repair. European journal of vascular and endovascular surgery : the official journal of the European Society for Vascular Surgery 2005;30(5):520-4. doi: 10.1016/j.ejvs.2005.04.046
Chen QJ, Yang ZY, Wang CY, et al. Hydroxyethyl starch resuscitation downregulate pro-inflammatory cytokines in the early phase of severe acute pancreatitis: A retrospective study. Exp Ther Med 2016;12(5):3213-20. doi: 10.3892/etm.2016.3744
Sossdorf M, Marx S, Schaarschmidt B, et al. HES 130/0.4 impairs haemostasis and stimulates pro-inflammatory blood platelet function. Critical care 2009;13(6):R208. doi: 10.1186/cc8223
Haase N, Perner A, Hennings LI, et al. Hydroxyethyl starch 130/0.38-0.45 versus crystalloid or albumin in patients with sepsis: systematic review with meta-analysis and trial sequential analysis. BMJ 2013;346:f839. doi: 10.1136/bmj.f839
Roberts I, Shakur H, Bellomo R, et al. Hydroxyethyl starch solutions and patient harm. Lancet 2018;391(10122):736. doi: 10.1016/S0140-6736(18)30255-1
Caironi P, Tognoni G, Masson S, et al. Albumin replacement in patients with severe sepsis or septic shock. N Engl J Med 2014;370(15):1412-21. doi: 10.1056/NEJMoa1305727
Lewis SR, Pritchard MW, Evans DJ, et al. Colloids versus crystalloids for fluid resuscitation in critically ill people. The Cochrane database of systematic reviews 2018;8:CD000567. doi: 10.1002/14651858.CD000567.pub7

No competing interests reported.

BaSESSupplementOctober29th2023.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

Hydroxyethyl starch and Ringer’s lactate for volume resuscitation in sepsis and septic shock – The BaSES Trial

Status:

Version 1

Abstract

Figures

BACKGROUND AND RATIONALE

METHODS

Trial design

Institutional Review Boards

Eligibility criteria

Study groups and follow-up

Statistical methods

RESULTS

Primary endpoint

Secondary endpoints

Hospital LOS and mortality

Markers of kidney function

Inflammatory and metabolic markers, SOFA score

DISCUSSION

CONCLUSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1