Hydroxyethyl starch impairs renal water reabsorption in patients with cardiac shock

2.1 Study population

A retrospective cohort was constructed of all patients with cardiac shock admitted to the Cardiology Department in Shanxi Bethune Hospital and in the affiliated hospital of Army Medical University NCO School between January 2010 and December 2020. All written informed consent was obtained from participants. Cardiac shock was primarily judged by their treating clinician, depending on factors (such as blood pressure $<$90/60 mmHg, left ventricular ejection fraction $\le$40% (LVEF), acute myocardial infarction or chronic heart failure).

234 patients were randomized into four groups (NaCl group, HES group, HES + DOP group, HES + NE group) according to the treatment. Patients with cardiac shock received fluid resuscitation firstly. When fluid resuscitation was not enough to alleviate shock, vasoactive agents were titrated to patients, restoring tissue perfusion. In detail, patients admitted in hospital firstly received fluid resuscitation (HES or NaCl). If hemodynamics of these patients was maintained well, these patients were assigned into HES group and NaCl group. If some of them are hemodynamically not stable, and fluid resuscitation was not enough to alleviate shock, vasoactive agents (dopamine or norepinephrine) were titrated to patients, restoring tissue perfusion. Those patients are assigned into HES + DOP group and HES + NE group. Seventy-two patients were excluded because of death in the first 24 hours (n = 10) or prior renal dysfunction (n = 62).

Patients in NaCl group received 0.9% sodium chloride. Patients in HES group received 6% HES with a molecular weight of 130 kD and molar substitution ratio of 0.4 (130/0.4) in 0.9% sodium chloride. Patients in HES + DOP group received dopamine and HES (130/0.4), and ones in HES + NE group received norepinephrine and HES (130/0.4).

2.2 Protocol

The protocol was initiated after patients admitted to hospital before the occurrence of cardiac shock, depicted as first phase (D0). Data collected included age, sex, blood pressure, heart rate, left ventricular ejection fraction, serum creatinine, blood urea nitrogen, urine specific gravity, urine volume, oxygen saturation serum, drug dosage, and so on. The anti-shock treatments were terminated when patients were hemodynamically stable (such as blood pressure $\ge$90/60 mmHg). The stage that three days after the hemodynamic stability of patients depicted as second phase (D3). The reason we collected data 3 days after anti-shock treatment is that almost no residue of HES, dopamine or norepinephrine in circulation might influence the evaluation of renal function. Data related to renal function were recorded at both phases.

2.3 Statistics

Quantitative variables are expressed as mean $\pm$ standard deviation, and qualitative variables are expressed as frequency (percentage) and analyzed using a chi-square test. Normally distributed data were analyzed by two-sided unpaired Student’s t-test. One-way ANOVA was used to compare means across multiple groups. Statistical analysis was performed using SPSS Statistics Version 18 (IBM Corp., Armonk, NY, USA), and a two-sided p $<$ 0.05 was considered statistically significant.

3.1 Patient population and baseline patient characteristics

In this study 234 patients were initially screened for eligibility, of which 72 patients were excluded because of death or renal dysfunction. 162 patients were randomized into four groups with different anti-shock treatments (Fig. 1).

Fig. 1.Patient population. Patients were randomized into four groups with different anti-shock treatments. Values are number.

Baseline patient characteristics were described in Table 1. The four groups had similar characteristics at baseline. After statistical analysis, we found that there were no differences among four groups in characteristics (such as gender, age, left ventricular ejection fraction, heart rate, blood pressure or oxygen saturation) (p $>$ 0.05). Coronary artery disease was the high-risk factor of cardiac shock, and the prevalence of this disease was almost the same as each other among four groups (p $>$ 0.05). The morbidity rates of both diabetes and hypertension were also statistically not different in four groups (p $>$ 0.05).

3.2 The effect of different anti-shock treatments on renal function

HES was administered to patients in three groups. Total HES use was 2144.83 $\pm$ 1464.85 mL in HES group, 2101.24 $\pm$ 1680.17 mL in HES + DOP group, and 2179.79 $\pm$ 1529.48 mL in HES + NE group. No difference of HES dosage was observed in groups (p $>$ 0.05). 0.9% NaCl solution (3518 $\pm$ 1223.52 mL) was provided to patients in NaCl group. While intravenous fluid therapy (HES solution) could not improve hemodynamics in patients with shock, patients in HES + DOP group received dopamine (552.19 $\pm$ 252.15 mg), and patients in HES + NE group received norepinephrine (21.52 $\pm$ 8.59 mg) (Table 2).

The serum creatinine (Scr) and blood urea nitrogen (BUN) of patients at D0 phase were similar in four groups (p $>$ 0.05). Different treatments including fluid resuscitation (NaCl, HES) and vasoactive agents (dopamine and norepinephrine) did not influence the Scr and BUN in patients at D3 phase (p $>$ 0.05). However, NaCl solution and dopamine + HES resulted in more urine volume at D3 phase than that at D0 phase (p $<$ 0.001). While NaCl solution and HES + dopamine did not influence urine specific gravity (p $>$ 0.05), HES and norepinephrine reduced the urine specific gravity at D3 phase in HES group (1.019 $\pm$ 0.006 vs 1.012 $\pm$ 0.005, p $<$ 0.001) and in HES + NE group (1.019 $\pm$ 0.006 vs 1.011 $\pm$ 0.004, p $<$0.001) (Table 3).

There was a high incidence of hypokalemia and hyponatremia in patients with cardiac shock. What’s more, patients receiving anti-shock treatments were more prone to suffering from hypokalemia (p $<$ 0.05) (Table 3).

4.1 HES is cautiously used clinically with regard to adverse effect on renal function

HES solution as a less-expensive synthetic colloid to albumin is widely used in clinical practice, aiming at restoring intravascular volume [10, 11]. The toxicity of HES results from the cumulative tissue uptake. HES can be detected in plasma, urine and tissues (such as liver, muscle, spleen and skin) on the first day after infusion, associated with impaired coagulation, renal injury and severe persistent pruritus [12]. Endothelial and epithelial cells in proximal kidney tubule have been reported to participate in the ingestion of HES [12]. In-vitro studies have confirmed that HES impairs the endothelial barrier but not epithelial barrier, and side effects of HES are correlated to reduced cell-matrix or cell–cell adhesion [11]. Several HES products have been developed in recent years, differing from each other in molecular weight, degree of substitution, and hydroxyethylation ratio [11, 13]. Lower molecular weight of HES seems to lead to increased tissue uptake [12].

In this study we found that NaCl solution did not affect urine specific gravity, and HES reduced urine specific gravity. Therefore, we thought that HES impairs renal water reabsorption of kidney tubules. When water reabsorption is impaired, it is reasonable to speculate that the urine volume will be increased. In our study, we found reduced urine specific gravity and no increased urine volume in HES group and HES + NE group. I think there are some reasons: (1) when plenty of HES solution is transfused into circulation, it will lead to increased plasma colloid osmotic pressure and decreased glomerular effective filtration pressure. The consequence is that more water is preserved in circulatory system, and less water is filtered and eliminated to the outside of the body. Although we found HES impaired water reabsorption and decreased urine specific gravity, the effect of decreased glomerular effective filtration pressure and reduced urine specific gravity compensated each other, leading to no change in urine volume of HES group. (2) Norepinephrine activates $\alpha$1 adrenergic receptor, producing decreased renal blood flow in HES + NE group. When renal blood flow decreases, glomerular effective filtration pressure will further reduce, inducing less urine volume. So we did not find increased urine volume in HES group and HES + NE group.

The normal range of urine specific gravity is 1.015 to 1.025. There are also some experts considering the value of 1.010 to 1.025 as normal range in healthy persons. In this trail, we found urine specific gravity decreased to 1.012 $\pm$ 0.005 in HES group and to 1.011 $\pm$ 0.004 in HES + NE group after anti-shock treatments, which is still in normal range. However, there is statistically significant difference after treatments. Technically, it is reasonable to conclude that HES impairs renal water reabsorption. If we prolong the study period, maybe we can obtain the decreased value of urine specific gravity below the lower limit.

Interestingly, hypokalemia and hyponatremia are common in patients before and after anti-shock treatments. It is reasonable to speculate that reduced urine specific gravity and increased urine volume may worsen electrolyte imbalance.

A survey revealed that 66% of physicians believed HES solution improves survival rate of patients [12], and HES accounted for 43% of colloids to improve abnormal perfusion and vital signs [10]. However, the use of HES in critical care has been shown to be associated with increased risks of kidney failure, renal replacement therapy and mortality [10]. Some organizations (e.g., European Society of Critical Care Medicine and European Medicines Agency) recommended the avoidance of HES in patients with renal dysfunction or sepsis or burn injuries or critical ill [9], whereas the American Thoracic Society Guidelines advise cautious use of HES in critically ill patients [10]. A lot of effort has been given to balance the safety and efficacy of HES. The latest balanced HES 130/0.4 was developed in 2001, but until now clinical evidence is not adequate to conclude that HES 130/0.4 is safer than other HES products [10]. Generally HES is still administered to critically ill patients despite of the increased risk of kidney dysfunction and mortality [14], and a consensus has been reached world widely that renal function must be monitored in HES-treated patients [8, 15].

4.2 Dopamine plays a protective role on renal function in HES-treated patients

When fluid resuscitation is not able to restore tissue perfusion and arterial pressure ($<$80–85 mmHg) [16], vasoactive agents should be administered immediately in patients with cardiac shock [3]. By stimulating receptors in the heart and vessels, vasoactive agents produce direct inotropic effect and peripheral vascular effect, increasing cardiac output and systemic vascular resistance [3, 17], which, in turn, preserves visceral perfusion [16].

Different dose of dopamine activates dopaminergic receptors, $\beta$1 adrenergic receptors and $\alpha$1 adrenergic receptors, resulting in natriuretic effects and increased cardiac output and systemic vascular resistance [3]. Activated dopaminergic receptors lead to increased renal blood and creatinine clearance [18]. The mechanism of HES-related renal toxicity is that luminal epithelial cells in the proximal tubule can take up HES molecules by pinocytosis, impairing renal function [12, 18]. Our study elucidated the protective role of dopamine on renal water absorption in HES-treated patients. We thought the underlying mechanism is that: activated dopamine receptors increase renal blood flow and then inhibit the process of epithelial cells absorbing HES, preserving the function of proximal tubule and reducing the toxicity of HES on kidney. Therefore, we recommend joint use of dopamine and HES in patients with cardiac shock, to decrease the risk of renal failure.

Norepinephrine is regarded as the first-line vasopressor for shock. Norepinephrine is approximately 100 times the potency of dopamine [3]. However, excessively activated-$\alpha$1 adrenergic receptor produces decreased renal blood flow, predisposing to kidney injury [19]. Considering the risk of precipitating acute renal failure, clinicians are hesitant to choose norepinephrine [20]. Our study found the combined utilization of HES and norepinephrine led to decreased urine specific gravity, suggesting injured renal water absorption. While HES alone has a toxic effect on renal function, we postulate that the joint use of norepinephrine may aggravate renal ischemia, inducing increased risk of renal injury. Therefore, we recommend norepinephrine not be jointly used with HES during anti-shock treatment.