Effect of dobutamine on extravascular lung water index, ventilator function, and perfusion parameters in acute respiratory distress syndrome associated with septic shock

Abstract

Background: The role of dobutamine in the relief of pulmonary edema during septic shock-induced acute respiratory distress syndrome (ARDS) remains undetermined, due to a lack of controllable and quantitative clinical studies. Our objective was to assess the potential effects of dobutamine on extravascular lung water index (ELWI) in septic shock-induced ARDS, reflecting its importance in pulmonary edema. At the same time, ventilator function and perfusion parameters were evaluated. Methods: We designed a prospective, non-randomized, non-blinded, controlled study to compare the differences in PiCCO parameters after 6 h of constant dobutamine infusion (15 μg/kg/min), in the baseline parameters in 26 septic shock-related ARDS patients with cardiac index ≥ 2.5I/min/m² and hyperlactatemia. These patients (12 survivors/14 non-survivors) were monitored using the PiCCO catheter system within 48 h of onset of septic shock. The dynamic changes in ELWI, which is typically used for quantifying the extent of pulmonary edema, were evaluated, and the corresponding ventilator function and tissue perfusion parameters were also measured. Results: Decreasing ELWI (p = 0.0376) was accompanied by significantly decreased SVRI (p < 0.0001). Despite a significant increase in cardiac output (p < 0.0001), no differences were found in ITBI or GEDI. Moreover, the required dose of norepinephrine was decreased (p = 0.0389), and urine output was increased (p = 0.0358), accompanied by stabilized lactacidemia and MAP. Additionally, airway pressure was moderately improved. Conclusion: During the early stage of septic shock-induced ARDS, dobutamine treatment demonstrated a beneficial effect by relieving pulmonary edema in patients, without a negative elevation in preload or hemodynamics, which might account for the improvements in ventilator function and tissue hypoperfusion.

Keywords::

• acute respiratory distress syndrome
• dobutamine
• extravascular lung water
• pulmonary edema
• septic shock

Abbreviations
ACCP	=	American College of Chest Physicians
SCCM	=	Society of Critical Care Medicine
APACHE	=	Acute Physiology and Chronic Health Evaluation
SOFA	=	Sequential Organ Failure Assessment scores
SAPS	=	Simplified Acute Physiology Score
CHF	=	chronic heart failure
MAP	=	mean arterial pressure
CO	=	cardiac output
ITBI	=	intrathoracic blood volume index
GEDI	=	global end-diastolic volume index
ELWI	=	extravascular lung water index
PVPI	=	pulmonary vascular permeability index
HR	=	heart rate
SV	=	stroke volume
SVRI	=	systemic vascular resistance index
dPmx	=	index of left ventricular contractility
Crs	=	respiratory-system compliance

Introduction

Acute respiratory distress syndrome (ARDS), which is characterized by noncardiogenic pulmonary edema due to the accumulation of extravascular lung water (EVLW), remains the leading complication for morbidity and mortality in patients with severe sepsis and septic shock. Thus, reducing EVLW or balancing intravascular volume expansion against the negative effects of causing pulmonary edema might be critical for patients with ARDS and septic shock (Mitchell et al. 1992).

Patients with severe ARDS exhibit greater pulmonary edema (Chew et al. 2012). Indeed, both persistent hypoperfusion and pulmonary edema are strongly correlated with outcomes in septic shock-induced ARDS. Dobutamine is the first- choice inotrope in patients with low cardiac output or ongoing signs of persistent tissue hypoperfusion after initial adequate fluid resuscitation (Dellinger et al. 2013), owing to its predominantly vasodilatory properties. Previous experiments (Minnear et al. 1993, Goodman et al. 1984, Mason et al. 1982, McAuley et al. 2004, Berthiaume et al. 1987, 1988, Saumon et al. 1987), including recent data from an ex vivo human lung experiment (Sakuma et al. 1994), have demonstrated that the β2 receptor is crucial for recovery from pulmonary edema, based on a mechanism linked to water transport across alveolar type II cells in septic shock (Mutlu et al. 2004, Perkins et al. 2007). It has also been shown that patients who cleared EVLW early experienced more rapid resolution of lung injury, shorter duration of mechanical ventilation (Wyncoll and Evans 1999), and improved survival. Thus, it is critical and essential to focus on resolution of EVLW to control the formation of pulmonary edema and to avoid the aggravation of hypoperfusion during the early stage of ARDS in septic shock. Although dobutamine has been shown to be beneficial for persistent hypoperfusion, it remains unclear whether dobutamine might relieve pulmonary edema during sepsis-induced ARDS (Li et al. 2013).

Due to a lack of controlled and dynamic monitoring, very few studies or direct evidence have assessed the development of pulmonary edema in patients with ARDS and septic shock. Recently, more studies have demonstrated that ELWI, a more sensitive and specific criterion for quantifying pulmonary edema, could be monitored with PiCCO devices (Michard et al. 2004, Galstian et al. 2011, Martin et al. 2005). In most cases, increased ELWI, as a hallmark of early ARDS, has been shown to be the best pulmonary-specific index of disease severity and outcome prediction (Sakka 2013, Aman et al. 2012). Indeed, patients with septic shock are usually equipped with a PiCCO catheter device, allowing for the easy access of dynamic ELWI changes (Neumann 1999). Using PiCCO parameters, we demonstrated here that dobutamine has a specific potential contribution to hemodynamically stable septic shock patients who have ongoing signs of tissue hypoperfusion and are at risk for pulmonary edema.

Materials and methods

This study was conducted in the medical and surgical adult intensive care units (ICUs) of 4 tertiary care hospitals in China and was approved by the Regional Ethical Review Board and registered with a database (ChiCTR-ONC-13003320, UTN: U1111-1145-9384). Informed consent was obtained from the next of kin of each patient.

Study subjects

Adult ICU patients were recruited into this prospective, non-randomized, open interventional study. The diagnosis of septic shock was defined according to the criteria of the ACCP/SCCM Consensus Conference (Bone et al. 1992). All of the patients were monitored using a transpulmonary thermodilution device (PiCCO, Pulsion Medical System, Munich, Germany). To be eligible, patients were required to have developed septic shock and secondary ARDS onset within the preceding 48 h. ARDS was defined according to the criteria recommended by the Berlin definition (2012). Mechanical ventilation was performed according to the protocol described in the ARDSNet study (No Authors Listed 2000). Sedation was provided with propofol, and analgesia with remifentanil.

All patients were required to have both a right femoral artery catheter and a right central venous catheter in place. The correct placement of the catheter for insertion was further confirmed by chest radiography. Dobutamine was administered to increase CO because of the persistence of signs of hypoperfusion (oliguria, mottled skin, central venous oxygen saturation [SvO₂] < 70% despite a hemoglobin > 8 g/dl). Achieving adequate intravascular volume, adequate mean arterial pressure (MAP > 65 mmHg), and ELW1 ≥ 8 were required (Saumon et al. 1987, Galstian et al. 2011). In addition, patients meeting the criteria for normal ITBV and significantly elevated SVR (SVR ≥ 2000) could be admitted within the precedent of dobutamine infusion.

We excluded patients according to the following criteria: pregnancy; age less than 18 years; absence of fluid resuscitation; unstable hemodynamic condition (change in vasoactive drug dosage or fluid administration within 1 h preceding the protocol); acute coronary syndrome within the last 72 h, and uncontrolled tachyarrhythmia (heart rate > 140 beats/min). In addition, patients who died within 48 h after the implementation of the PiCCO system, had hemodialysis, or were resistant to dobutamine were also excluded.

Study protocol

The baseline clinical characteristics and laboratory values of subjects were obtained upon admission (Table I). Thermodilution parameters and pulse contour parameters were obtained with the PiCCO monitor, based on triplicate central venous injections of 20 ml of iced (4°C) 0.9% saline solution, and were recorded as the average of the three measurements. The corresponding ventilator function and perfusion parameters were observed. Sedation drugs were kept constant during the 6-h period preceding the measurements. The patients were kept in the horizontal position.

Table I. Clinical features of the subjects with ARDS associated with septic shock.

Gender(M/F; n)^a	15/11
Age (years)^b	52.2 (22–68)
Weight (kg)^b	66 (49–81)
BMI (kg.cm^−2)^b	28.2 (25–36)
Height (cm)^b	165 (157–181)
Mortality (n,%)^a	11 (42)
MODS^b	7 (6–12)
APACHE II^b	27.3 (23–36)
LIS^b	2.8 (2.2–4.1)
SOFA^b	11.5 (8–15)
SAPS^b	59.1 (46–68)
Norepinephrine dose (ug/kg/min)^b	0.31 (0.21–0.35)
Serum lactate (mmol/L ^b	2.7 (2.4–3.3)
Urinary output (ml/kg/h)^b	0.33 (0.11–0.92)
Hemoglobin (g/dl)^b	9.2 (8.2–11.5)
Net fluid balance (liters)	7.8 (3.6–13.8)
Serum endotoxin (Eu/ml)^b	2.87 (0.21–12.76)
Underlying disorder, n (%)^a	9 post-op (35)
	8 pancreatitis (31)
	6 obstruction (23)
	3 trauma (11)
Infection source, n (%)^a	15 intra-abdominal (58)
	6 Catheter/bloodstream (23)
	4 urinary tract (15)
	1 skin and soft tissue (4)

^aData are expressed as number.

^bData are expressed as medians and their interquartile ranges.

PiCCO measurement was performed before admission, to capture a high ELWI, conforming to the criteria for pulmonary edema. After baseline measurements, an infusion of dobutamine was initiated at 5 μg/kg/min and was increased by 5 μg/kg/min at every 20-min interval, reaching a level up to 15 μg/kg/min, and continuous infusion with 15 μg/kg/min was administered for 6 h. During infusion, additional noradrenaline support (beyond baseline needs) to maintain a MAP of 65 mmHg was mandated. No new vasopressors or inotropes were administered after starting the protocol. During the study period, all participants had the same environmental control system and followed the same control protocol. As a safeguard, in case of any transient adverse effects such as life-threatening hypotension, tachycardia > 150 bpm, acute atrial fibrillation, or ST changes on the cardiac monitor, the study was stopped and treatment was begun immediately. A diagram of the study protocol is shown in Figure 1. After completing the study, dobutamine was tapered over 20 min and was then discontinued. All subjects were continuously monitored with continuous electrocardiographic monitoring and pulse oximetry, according to the standard protocols of the ICU. The patients’ subsequent intensive care and hospital course were directed by normal clinical practice.

Figure 1. Diagram of study protocol.

Data analysis

Date at baseline (before dobutamine but after fluid resuscitation) and at the end of the dobutamine infusion for all of the subjects were pooled as the mean ± SD and were analyzed. Fisher's exact test was used for proportional analysis, and the independent-samples t-tests with Levene's correction for unequal group variances were used for measurements (Table II). Statistical analyses were performed with SAS/STAT software, version 9.4. All of the significance tests used were two-tailed, and statistical significance was defined as p < 0.05.

Table II. Measurements of PiCCO, ventilator function, and perfusion parameters.

	Before	After	P value
Hemodynamic and thermodilution parameters
HR (beats/min)	98.2 ± 11.7	102.7 ± 13.3	0.2011
MAP (mmHg)	81.6 ± 11.2	86.2 ± 14.7	0.1584
CO (l/min)	3.2 ± 0.6	5.8 ± 0.6	< 0.0001
SV (ml)	48.2 ± 12.6	73.2 ± 15.4	< 0.0001
GEDI (ml/m^2)	823 ± 163	881 ± 206	0.2656
ITBI (ml/m^2)	928 ± 123	966 ± 48	0.1485
ELWI (ml/kg)	11.23 ± 2.6	9.11 ± 1.8	0.0376
SVRI (DSm^2/cm^5)	2685 ± 416	1532 ± 309	< 0.0001
dPmx	1524 ± 221	1046 ± 284	< 0.0001
Perfusion parameters
Norepinephrine dose (mcg/kg/min)	0.26 ± 0.12	0.20 ± 0.08	0.0389
Urine output (ml)	65.7 ± 38.2	90.1 ± 43.2	0.0358
Mixed venous oxygen saturation (%)	73 ± 9	79 ± 8	0.2098
Arterial lactate (mmol/l)	2.92 ± 0.51	2.74 ± 0.44	0.1791
Oxygen delivery (ml/min/m^2)	512 ± 215	676 ± 220	0.0090
Oxygen consumption (ml/min/m^2)	133 ± 34	153 ± 57	0.1307
Ventilator function
PaO2/FiO2 (mmHg)	124.7 ± 22.5	129.2 ± 31.9	0.5145
Respiratory-system compliance (ml/cm of water)	28.3 ± 3.1	30.1 ± 3.9	0.0714
Positive end-expiratory pressure (cm of water)	12.5 ± 1.6	11.3 ± 2.3	0.0337
Minute volume (l/min)	12.3 ± 0.5	12.6 ± 0.3	0.0115
Respiratory rate (breaths/min)	25.1 ± 1.7	24.9 ± 0.9	0.5983
Peak inspiratory pressure (cm of water)	37.0 ± 2.6	35.4 ± 2.3	0.0227
Plateau pressure (cm of water)	31.8 ± 3.4	28.8 ± 3.9	0.0047

Data are expressed as the mean ± SD or median (first–third quartiles).

Results

Subject characteristics

From July 2013 to June 2014, 26 patients (15 male, 11 female) met the criteria for septic shock-induced ARDS and were eligible to be enrolled in the study. The median time between diagnosis of septic shock-induced ARDS and enrollment was 8.2 (4–15) h. Seventy-five percent of the patients were admitted to the ICU from the emergency unit. The overall ICU mortality was 42% (11/26). Death was caused by multiple organ dysfunction syndrome and/or irreversible septic shock. Baseline characteristics and laboratory values of the subjects are summarized in Table I. Treatment was generally well tolerated in all of the patients, and no clinically significant adverse effects occurred during the course of the constant dobutamine administration (except that, in 6 patients, there was a trend toward higher heart rates during the initial period of dobutamine administration, but the rates then rapidly returned to normal levels).

Hemodynamic and thermodilution parameter variables

Hemodynamic and thermodilution parameters measured with the PiCCO system are listed in Table II and Figure 3. ELWI decreased at the final assessment (after 6 h of constant dobutamine administration), compared with baseline (11.2 ± 2.6 vs 9.1 ± 1.8, respectively, p = 0.0376). This decrease was accompanied by a significant decrease in SVRI (2685 ± 416 vs 1532 ± 309, respectively, p < 0.0001), suggesting beneficial effects in the relief of pulmonary edema and afterload during septic shock-induced ARDS. Moreover, cardiac output was significantly increased (3.2 ± 0.6 vs 5.8 ± 0.6, respectively, p < 0.0001), which was paralleled by a statistically significant increase in oxygen delivery (512 ± 215 vs 676 ± 220, respectively, p = 0.0090). However, the intrathoracic blood volume index (ITBI) and global end-diastolic volume index (GEDI) remained unchanged (p = 0.1485, p = 0.2656, respectively), instead of showing the expected elevation that, in theory, would be caused by an intravascular volume increase. There was no difference in pulmonary blood volume (PBV, calculated as the difference between ITBV and GEDV, data not shown). Despite an initial increase in heart rate in some patients, heart rates returned to baseline and stabilized shortly thereafter (p = 0.2011). Additionally, no differences were observed in MAP. Indeed, MAP decreased in only 2 cases, but it rapidly increased and stabilized quickly after administering additional norepinephrine, accompanied by decreased doses of norepinephrine (0.26 ± 0.12 vs 0.20 ± 0.08, respectively, p = 0.0389).

Figure 2. Classification of pulmonary edema as measured using the PiCCO system. The results of categorization by the PiCCO system of pulmonary edema patients in whom a precise diagnosis of pulmonary edema could not be made.

Figure 3. Changes in ELWI, SVRI, Co, ITBI, PaO2/FiO2, and Pplat in patients. Data in the final 6 h corresponding to baseline values. P values were calculated by comparing the differences between two groups (Baseline- Final). The symbols * and ** represent significant differences (P < 0.05, P < 0.001). Data are expressed as medians, interquartile ranges, and 5th/95th percentiles.*.

Ventilator function and perfusion parameter variables

The ventilator function and tissue perfusion parameters were measured, and they are listed in Table II and Figure 3. Although there were no differences in mixed venous oxygen saturation, arterial lactate, or oxygen consumption, both urine output and oxygen delivery were increased (p = 0.0358, p = 0.0090, respectively). Accordingly, there was a decrease in the dose of norepinephrine required (p = 0.0389), indicating an improvement in tissue perfusion.

Ventilator function variables were also assessed, and they are described in Table II and Figure 3. Indeed, the PaO₂/FiO₂ ratio, mixed venous oxygen saturation (SvO₂), respiratory-system compliance (Crs), and minute volume (MV) remained unchanged, and there was a moderate reduction in airway pressure indicators (positive end-expiratory pressure [PEEP], peak inspiratory pressure [PIP], and plateau pressure [Pplat]), indicating an improvement in ventilator function (p = 0.0337, p = 0.0227, and p = 0.0047, respectively) (Table II, Figure 3).

Discussion

Consistent with previous animal studies (Wu et al. 2009), our data provided the evidence that dobutamine treatment diminishes ELWI. The differences in ELWI values in our study were small, however, with an early decrease associated with more rapid improvements in ARDS, reduced durations of ventilation, and improved survival (Ware and Matthay 2001). Therefore, the augmentation of resolved ELWI with dobutamine administration during the early stage of ARDS associated with septic shock seems promising and could be important for clinical strategies to prevent pulmonary edema. This finding was in contrast to some previous reports (Pittet et al. 1994) that found elevated alveolar fluid clearance (AFC) in an animal experiment on sepsis onset and no improvement in peripheral perfusion following general concentrations of dobutamine intervention (5 μg/kg/min) (Hernandez et al. 2013). The reasons for these differences could be as follows:

1. In this clinical trial, patients with ARDS caused by severe septic shock had more severe alveolar epithelial cell damage than that observed in animal experiments, because the self-cleaning protection of the alveolar epithelium, which increases during light infection, could result in temporarily elevated AFC.

2. Patients received a bolus of dobutamine, followed by continuous infusion at a constant rate (15 μg/kg/min), in contrast to the low concentrations used in animal experiments.

Early resolution of edema is critical for recovery from ARDS due to the development of pulmonary edema, which can impair gas exchange, causing refractory hypoxemia, and an early diagnosis of the pathological accumulation of EVLW (Figure 2)during resuscitation could allow for earlier intervention times and considerable changes in the therapeutic plan (Pino-Sanchez et al. 2009). In this study, we chose 8 ml/kg (Michard et al. 2012) as a cut-off for normal EVLW, not the higher threshold (EVLW > 14 ml/kg) (Eisenberg et al. 1987)used in previous studies. This lower threshold (EVLW > 8 ml/kg) represented the relatively early stage of permeability pulmonary edema in ARDS patients, and it could be utilized to monitor early dobutamine intervention, whereas the higher threshold might represent the aggravate period and a missed opportunity for interventions (LeTourneau et al. 2012). In addition, only in this situation, without the interference of diuretics and nitroglycerin preparation, the effect of dobutamine could be objectively evaluated, theoretically.

At study entry, we noted significantly increased SVRI, which might be connected to the pathophysiology of persistent shock status and the use of norepinephrine during septic shock. Increasing SVRI—a negative effect caused by norepinephrine—can aggravate hypoxia, leading to reduced alveolar fluid clearance, even if CO and MAP are normal after initial adequate fluid resuscitation. Because the signs of persistent tissue hypoxia/hypoperfusion (hyperlactacidemia) were present at enrollment, the high SVRI must be integrated into treatment considerations. We also demonstrated that there was a trend toward higher sensitivity to dobutamine, as shown by the significant decrease in SVRI, indicating that parallel mechanisms acted through dobutamine as a pulmonary vasodilator to reduce pulmonary pressures, thereby reducing one of the driving forces of edema formation (Garcia-Delgado et al. 2001). Additionally, as a pulmonary vasodilator, dobutamine inhibited hypoxic pulmonary vasoconstriction, thereby increasing the shunt and overcoming the beneficial effects on oxygenation caused by the reduction in alveolar edema. Although dobutamine infusion increased CO because of its predominantly inotropic properties, which could theoretically increase the preload and pulmonary perfusion volume and result in elevated ITBV or GEDV, there were no actual changes in ITBV or GEDV, making large changes in pulmonary hemodynamics less likely and avoiding the risk of hydrostatic pulmonary edema.

In this study, we found that pulmonary blood volume (PBV), a value calculated by the difference in value between ITBV and GEDV, was also unaffected by the increase in CO caused by dobutamine infusion. Chew et al. (Chew et al. 2012)showed aggravated permeability of pulmonary edema due to a higher pulmonary vascular permeability index (PVPI), which is the ratio of high EVLW to normal PVB. In this study, although CO increased, PVPI decreased according to decreased EVLW and stable PVB, resulting in the resolution of permeability pulmonary edema. Based on further analysis of dobutamine as a pulmonary vasodilator, if it influenced pulmonary perfusion, it would have increased, rather than decreased ELWI. One possibility to tie these observations together is the role of water transportation, instead of vasodilation, in parallel with dobutamine infusion to relieve the abnormal accumulation of fluid in the extravascular compartments of the lung, in accordance with a component of the molecular mechanisms that potentially link to active water transport across the epithelial barrier and that might be responsible for decreasing ELWI after dobutamine treatment (Wu et al. 2009).

We also examined hemodynamic and hypoperfusion parameters in these patients. Interestingly, MAP was not decreased and it remained more stable during treatment, as evidenced by the reduced dose of norepinephrine and stabilization of lactacidemia, suggesting that there were no negative or aggravated signs in hypotension or hemodynamic parameters (Singh and Pai 2014). Despite the lack of differences in SvO₂, arterial lactate, and oxygen delivery, the improvement in urine output and in the doses of norepinephrine indicate beneficial effects on tissue hypoperfusion. In addition, both the stable ITBV and the decrease in norepinephrine demonstrated improved fluid responsiveness (Donati et al. 2008), although no obvious improvement in PaO₂/FiO₂, Crs, or SvO₂ were noted, which could have been due to three reasons: the infusion time was short, the sample size was small, and the recovery from ventilator function and tissue hypoperfusion might have been a late response related to the critical pathophysiology of septic shock. In addition, this finding indicates that monitoring of the risk of pulmonary edema via bedside ELWI measurements is more sensitive than PaO₂/FiO₂. One possible explanation is that the delay in oxygenation response observed reflected the time necessary for regeneration and repair of the damaged alveolar–capillary barrier.

Furthermore, some corresponding improvements in ventilator function (PEEP, PIP, and Pplat) were also noted, consistent with the possibility that the presence of cardiovascular response and ELWI reserve were critical for recovery from hypoxia/hypoperfusion and lung dysfunction in septic shock, thus providing clinicians with additional information on ventilator function and with the hypotension parameters that are concerned with promising outcomes in parallel with dobutamine administration.

Therefore, it seems that the reversion of pulmonary edema, cardiovascular response, ventilator function, and tissue hypoperfusion might potentially be accelerated by more rapid decreased in ELWI and SVRI, resulting in promising outcomes parallel to dobutamine administration in ARDS associated with septic shock. The challenges with dobutamine by PiCCO monitoring include understanding the multitude of variables that are measured and integrating the clinically relevant and adequately validated variables with the appropriate therapeutic interventions. These results are important for at least three reasons. First, they support the clinical observation with quantitative measurements that intravenous dobutamine can potentially reduce ELWI with normal PBV, thereby reducing the formation of edema to prevent ARDS development. This finding confirmed the potential effects suggested by previous studies (Pittet et al. 1994) that showed that the stimulation of fluid clearance from the distal airspaces of the lung was mediated by beta-adrenergic active water transport across the epithelial barrier. Second, this study might also have demonstrated that there is no clear proof that reduced MAP is accompanied by stable hemodynamics, despite a significant decrease in SVRI. Third, despite a lack of evidence for increases in PaO₂/FiO₂ and Crs, the data in this study suggested beneficial effects on ventilator function and tissue perfusion, suggesting promising effects of dobutamine administration. Based on these results, our observations provided important integrated insights regarding the cardiovascular and volume mechanisms underlying the effects of dobutamine infusion on pulmonary edema in early ARDS onset, associated with septic shock.

Our study showed a correlation between the statistical reductions in the amount of ELWI and dobutamine treatment. This study showed improved tissue perfusion, without the negative effects of increased hydrostatic pulmonary edema and unstable hemodynamics, suggesting both prognostic and therapeutic utility in the early clinical stage of ARDS associated with septic shock. We only sought to evaluate the incremental value of adding constant infusion of dobutamine to preliminary treatment, which would be a logical next step in future research. Further studies are needed to confirm these findings in an appropriately powered, randomized, and controlled trial.

Limitations

There were several significant limitations to this study. First, the very small number of patients in this study represented its main limitation and the primary source of possible error. Second, we could not ensure that the same volume status was present in the entire patient population that was enrolled because volume status could have been affected by time-dependent compensatory fluid shifts. These factors, and unaccounted compensatory fluid changes in perfusion, might have influenced the results. Finally, because the guidelines did not specify the dose for dobutamine, a relatively large dose (15 μg/kg/min) was adopted in this study. Thus, we were unable to select the optimal dose due to a lack of comparisons between different concentrations.

Author contributions

M.Z. contributed to the study concept, research design, and data analysis and wrote the manuscript; J.D., M.D., W.W., C.G., Y.W., R.T., F.X., and Z.R., participated in performing the study and the acquisition of data; G.S. contributed to the research design.

Trial registration: ChiCTR-ONC-13003320, UTN: U1111-1145-9384

Funding statement

This research article and the work were supported by the National Natural Science Foundation of China (No. 81201488 to M.Z., No. 81370170 to G.S.). The funders played no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Declaration of interest

The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.