Acetate-buffered crystalloid infusate versus infusion of 0.9% saline and hemodynamic stability in patients undergoing renal transplantation

Infusion therapy is one of the most frequently prescribed medications in hospitalized patients. The importance of research on fluid therapy has been unrecognized for a long time with normal saline being the most popular infusate [1‐4]. During recent years newer balanced infusates containing lactate or acetate became increasingly popular in many European countries whereas in the United States 0.9% saline is still the most widely used infusate in the perioperative and intensive care setting [4]. Currently used crystalloid solutions have a variable composition and may therefore influence acid-base status, intracellular and extracellular water content and plasma electrolyte compositions [5] and have a major impact on organ function and outcome [6]. Despite continuing evaluation no superiority of one particular type of fluid has been achieved so far [7‐10]. Nonetheless during recent years it has been shown that the theoretically more physiologically balanced buffered infusion solutions may have a respectable advantage in terms of patient morbidity. A study on healthy human volunteers found a balanced chloride-reduced infusion solution to be associated with better mean renal artery flow velocity, renal cortical tissue perfusion and urine output than infusion of 2 l of 0.9% saline [11]. In a further study including elderly surgery patients, gastric mucosal perfusion was reduced in patients receiving 0.9% saline compared to those receiving chloride-reduced infusion solutions [12]. These data, together with data from rodents with experimental sepsis, which showed significantly lower mean arterial pressure levels when receiving chloride-rich infusion solutions compared to lactated Ringer’s, could suggest an effect of the crystalloid fluid used on a patients’ hemodynamic situation [13].

In the present study we performed a sub-analysis using data from our previously performed prospective randomized controlled trial comparing an acetate-buffered balanced infusion solution compared to 0.9% saline in patients with end-stage renal disease receiving cadaveric renal transplantation [14]. Our aim was to investigate whether use of an acetate-buffered, chloride-reduced balanced infusion solution would result in A.) less need for catecholamine use than use of 0.9% saline and B.) in better hemodynamic stability of patients expressed by the mean arterial blood pressure.

A total of 150 patients were included in the study, 76 patients were randomized to normal saline and 74 patients to an acetate-buffered balanced crystalloid. No differences between the groups were found for the following parameters: age, sex, height, actual and dry weight, residual daily urine output, number of prior renal transplantations. For an overview on baseline characteristics see Table 1. The CONSORT flow chart is given in Fig. 1.

In the 0.9% saline group the mean volume of fluid received during surgery was 1691 ± 664 ml compared to 1798 ± 679 ml in the balanced acetate-based infusate group (p = 0.34). Noradrenaline for cardiocirculatory support during surgery was administered significantly more often in the normal saline group compared to patients receiving an acetate-buffered balanced crystalloid (30% versus 15%, p = 0.027). Patients receiving normal saline needed noradrenaline significantly earlier and the cumulative noradrenaline dose was significantly higher than patients receiving an acetate-based crystalloid solute (68 min ±45 versus 75 min ±60, p = 0.0055 and 0.000492 µg/kg body weight/min, ±0.002311 versus 0.000107 µg/kg/min, ±0.00039, p = 0.04, respectively). No difference in necessity for vasopressor or cumulative dose of vasopressors was seen between the groups (p = 0.47 and p = 0.08, respectively) (see Table 2; Fig. 2).

Mean minimum arterial blood pressure was significantly lower in patients randomized to 0.9% saline than in patients receiving the balanced infusion solution (57.2 [SD 8.7] versus 60.3 [SD 10.2] mm Hg; p = 0.024). Consequently, mean minimum systolic (77.7 [SD 17.1] versus 80.1 [SD 16.6] mm Hg; p = 0.0398) and mean minimum diastolic blood pressure (43.6 [SD 7.2] versus 45.7 [SD 8.6] mm Hg; p = 0.0485) were significantly lower in patients randomized to 0.9% saline. Patients randomized to 0.9% saline over time had a mean arterial blood pressure, which was 4 mm Hg lower than in patients receiving an acetate-buffered balanced solution during whole surgery (95% CI: 1–7 mm Hg; p = 0.009). Fig. 3 gives mean and systolic blood pressure levels during surgery for patients receiving either 0.9% saline or an acetate-buffered balanced infusion solution. In the Cox regression analysis it could be shown that patients receiving catecholamines had significantly lower blood pressure levels before administration of catecholamines than patients without catecholamines (see Fig. 4). There was no difference in heart rate (normal 73 beats per minute ±12.6 versus acetate-buffered solution 74 beats per minute ±12.2, p = 0.66) and central venous pressure (normal saline 9 ± 10 mmH₂O, acetate-buffered solution 8 ± 5 mmH₂O, p = 0.34) between the groups.

Peak chloride levels were significantly higher in patients randomized to the normal saline group (109 mmol/L [107 to 111] versus 107 mmol/L [105 to 109]), p = 0.001 as was fluctuation in serum chloride levels during surgery (4 mmol/L [3 to 6] versus 3 mmol/L [2 to 5]), p = 0.03. Serum sodium levels were not significantly different between the groups (4 mmol/L [3 to 5] versus 3 mmol/L [2 to 5]), p = 0.92. Acid base disturbances are described in more detail elsewhere [14].

These are the first data from a prospective, randomized, controlled study comparing the effects of 0.9% saline and an acetate-buffered balanced crystalloid solution on hemodynamics in the perioperative period. We showed that perioperative use of a balanced crystalloid as a maintenance fluid is associated with better hemodynamic stability of patients with end-stage renal disease receiving cadaveric renal transplantation. These findings are expressed by a significantly lower need for use of catecholamines for hemodynamic support and higher systolic and mean arterial blood pressure levels.

The strength of the current study is explained by the study collective: all patients had end-stage renal disease and received cadaveric renal transplantation. This collective of patients is especially vulnerable to effects due to infusion solutions since they lack the capacity of the kidneys to rapidly compensate for electrolyte and acid-base derangements to a variable degree. Theoretically, this makes even slight effects due to the infusion solution obvious, which would not be seen in patients with normal kidney function or only after administration of far larger volumes of fluid.

We found a significantly lower need for catecholamines, a longer time to catecholamines and a lower cumulative catecholamine dose in the acetate-buffered group. The significantly better hemodynamic stability in patients receiving a balanced acetate-based infusate compared to normal saline might be related to A) acetate itself and B) hyperchloremia induced by normal saline.

So far no study has compared the hemodynamic effects of normal saline to an acetate-buffered balanced crystalloid; however, there is some older evidence that the use of sodium acetate may have an influence on the cardiovascular system. In 1978 Liang and Lowenstein infused acetate and pyruvate in to anesthetized dogs to measure their impact on circulation [15]. They found that increased acetate levels were associated with a significant increase in cardiac output [15]. Even though myocardial oxygen consumption increased during acetate infusion, the decrease in myocardial oxygen uptake and the increase in coronary sinus blood oxygen saturation suggest that an active coronary vasodilation which was not a result of the increased cardiac work takes place [15]. Acetate infusion also increased blood flow to the gastrointestinal tract, kidneys, intercostal muscles and diaphragm [15]. In another study by Conahan et al. on resuscitation fluid composition and myocardial performance during burn shock in guinea pigs, the authors were able to show that treatment with Ringer’s acetate significantly improved cardiac output and contractility when compared to normal saline and Ringer’s lactate [16]. In the same study Ringer’s acetate was found to have better blood pressure stabilizing effects compared to Ringer’s lactate and normal saline [16]. Concerning blood pressure stabilization three studies on sodium acetate found that such an infusion may have a positive effect on cardiac output but it lowers peripheral vascular resistance [17‐19], a finding that we cannot confirm; however, it has to be born in mind that data on acetate and acetate-buffered crystalloid solutes and hemodynamics remain scarce and are mostly experimental. There is a lack of controlled randomized trials in humans that compare currently available crystalloid solutes with respect to hemodynamics and further research certainly should be encouraged.

In 2004 the study group around John Kellum showed in a model of experimental sepsis that infusion of chloride-rich crystalloids led to a decrease in arterial blood pressure in rodents compared to infusion of lactated Ringers [13]. In another cross-over study on 12 healthy volunteers Chowdhury et al. found that a colloid, embedded in a balanced crystalloid solution, led to increased renal cortical tissue perfusion while the same concentration of colloid, embedded in 0.9% saline, did not result in improved renal perfusion [20]. A similar trial of this group in which 2 l of 0.9% saline or a chloride-reduced, balanced crystalloid was administered to healthy volunteers showed that administration of 0.9% saline even resulted in a decline in renal blood flow velocity and renal cortical tissue perfusion [11]. Given this evidence, together with other recent studies on this subject [2, 21, 22], it is plausible that hyperchloremia induced by infusion of chloride-rich solutions, is the direct trigger for unfavorable hemodynamic effects, as seen in our current study. Only recently, large scale studies and meta-analysis showed that use of chloride-rich infusion solutions might be associated with adverse outcome [2, 23, 24]; however, a Cochrane review from 2012 still came to the conclusion that although balanced crystalloids are associated with less occurrence of hyperchloremia and concurrent metabolic acidosis, use of conventional solutions (i. e. 0.9% saline) can be considered safe in the perioperative period [25]. Additionally, a recently published randomized controlled trial comparing 0.9% saline to an acetate-buffered crystalloid solute found no differences in terms of acute kidney failure and renal replacement therapy [26]. Given the current data together with previously published studies in the field, the need for prospective randomized controlled trials is obvious. While studies comparing the currently available crystalloid solutes with mortality as the primary endpoint will be hard to perform due to the large patient numbers needed and probable issues with the financing of such a study, it seems applicable to perform studies focusing on patient hemodynamics or renal function.

Our study has several limitations. The major drawback of this study is the lack of invasive hemodynamic monitoring such as esophagus Doppler, pulmonary artery catheter or thermodilution techniques. Additionally, blood pressure was monitored non-invasively, which was performed in order to preserve blood vessels in patients potentially in need of dialysis shunts in the future. Additionally, it has to be borne in mind that the original study was designed to investigate incidence rates of perioperative hyperkalemia, other electrolyte and acid-base disorders and not hemodynamics. Therefore, hemodynamic management followed hospital standard operating protocols (SOP but not a hemodynamic protocol specially designed for the present study. Additionally, the study was not blinded, so investigator bias is possible but not very likely as hemodynamics were not an outcome in the first study and hospital SOP’s were followed. Nonetheless the results of the present study should be interpreted with some caution. We see our results as a first indiction that the choice of crystalloid may impact hemodynamic stability, but certainly further research needs to be done before an overall conclusion can be reached.

In conclusion, the results of the prospective, randomized, controlled trial suggest that use of an acetate-buffered, balanced infusion solution results in less occurrence of arterial hypotension and reduced need for use of catecholamines and cumulative catecholamine dose for hemodynamic support in the perioperative period. The more favorable hemodynamic outcome of patients receiving an acetate-buffered crystalloid solute may be attributed to increased cardiac output related to acetate as well as lower susceptibility for hyperchloremia and concurrent metabolic acidosis when compared to normal saline. Future randomized, controlled trials are certainly desirable to clarify the effects of current crystalloid infusates on hemodynamics in the perioperative field and the critically ill.