Perioperative administration of buffered versus non-buffered crystalloid intravenous fluid to improve outcomes following adult surgical procedures: a Cochrane systematic review

Background

Buffered intravenous fluid preparations contain substrates to maintain acid-base status. The objective of this systematic review was to compare the effects of buffered and non-buffered fluids administered during the perioperative period on clinical and biochemical outcomes.

Methods

We searched MEDLINE, EMBASE, CINAHL and the Cochrane Library until May 2017 and included all randomised controlled trials that evaluated buffered versus non-buffered fluids, whether crystalloid or colloid, administered to surgical patients. We assessed the selected studies for risk of bias and graded the level of evidence in accordance with Cochrane recommendations.

Results

We identified 19 publications of 18 randomised controlled trials, totalling 1096 participants. Mean difference (MD) in postoperative pH was 0.05 units lower immediately following surgery in the non-buffered group (12 studies of 720 participants; 95% confidence interval (CI) 0.04 to 0.07; I² = 61%). This difference did not persist on postoperative day 1. Serum chloride concentration was higher in the non-buffered group at the end of surgery (10 trials of 530 participants; MD 6.77 mmol/L, 95% CI 3.38 to 10.17). This effect persisted until postoperative day 1 (5 trials of 258 participants; MD 8.48 mmol/L, 95% CI 1.08 to 15.88). Quality of this evidence was moderate. We identified variable protocols for fluid administration and total volumes of fluid administered to patients intraoperatively. Outcome data was variably reported at disparate time points and with heterogeneous patient groups. Consequently, the effect size and overall confidence interval was reduced, despite the relatively low inherent risk of bias. There was insufficient evidence on the effect of fluid composition on mortality and organ dysfunction. Confidence intervals of this outcome were wide and the quality of evidence was low (3 trials of 276 participants for mortality; odds ratio (OR) 1.85, 95% CI 0.37 to 9.33; I² = 0%).

Conclusions

Small effect sizes for biochemical outcomes and lack of correlated clinical follow-up data mean that robust conclusions on major morbidity and mortality associated with buffered versus non-buffered perioperative fluid choices are still lacking. Buffered fluid may have biochemical benefits, including a significant reduction in postoperative hyperchloraemia and metabolic acidosis.

Electronic supplementary material

The online version of this article (10.1186/s13741-018-0108-5) contains supplementary material, which is available to authorized users.

Perioperative buffered versus non-buffered fluid administration for surgery in adults

BACKGROUND

Perioperative fluid therapy influences clinical outcomes following major surgery. Fluid preparations may be based on a simple non-buffered salt solution, such as normal saline, or may be modified with bicarbonate or bicarbonate precursor buffers, such as maleate, gluconate, lactate or acetate, to better reflect the human physiological state. These latter fluids have theoretical advantages over normal saline in preventing hyperchloraemic acidosis. A number of clinical studies have now compared fluid preparations with and without a buffer to achieve a balanced electrolyte solution for perioperative fluid resuscitation.

OBJECTIVES

To review the safety and efficacy of perioperative administration of buffered versus non-buffered fluids for plasma volume expansion or maintenance in adult patients undergoing surgery.

SEARCH METHODS

We electronically searched the Cochrane Central Register of Controlled Trials (CENTRAL) (2011, Issue 4), MEDLINE (1966 to May 2011), EMBASE (1980 to May 2011), and CINAHL (1982 to May 2011). We handsearched conference abstracts and where possible, contacted leaders in the field.

SELECTION CRITERIA

We only included randomized trials of buffered versus non-buffered intravenous fluids for perioperative fluid resuscitation. The trials with other forms of comparisons such as crystalloids versus colloids and colloids versus different colloids were excluded. We also excluded trials using hypertonic fluids and dextrose-based fluids.

DATA COLLECTION AND ANALYSIS

Two authors independently extracted data and assessed the methodological quality of clinical trials. We resolved any disagreements by discussion. We contacted the trial authors to provide additional information where appropriate. We presented pooled estimates of the dichotomous outcomes as odds ratios (OR) and on continuous outcomes as mean differences, with 95% confidence intervals (CI). We analysed data on Review Manager 5.1 using fixed-effect models, and when heterogeneity was high (I² > 40%) random-effect models were used.

MAIN RESULTS

We identified 14 publications reporting 13 trials or comparisons with a total of 706 participants. For many of the outcomes reported, there was significant clinical and statistical heterogeneity. The primary outcome of mortality at any time was reported in only three studies with a total of 267 patients. The mortality rate was 2.9% for the buffered fluids group and 1.5% for the non-buffered fluids group but this difference was not statistically significant. The Peto OR was 1.85 (95% CI 0.37 to 9.33, P = 0.45, I(2)= 0%). Organ dysfunction was only presented for renal impairment. There was no difference in renal insufficiency leading to renal replacement therapy between the buffered and non-buffered groups (OR 0.61, 95% CI 0.23 to 1.63, P = 0.32, I(2) = 0%). Markers of organ system failure as assessed by urine output, creatinine and its variables (for renal function), PaC0(2) (respiratory function) and postoperative nausea and vomiting (gastro-intestinal function) showed a statistically significant difference only in PaC0(2) levels. The mean difference was 1.18 with lower PaC0(2) levels in the non-buffered fluid group (95% CI 0.09 to 2.28, P = 0.03, I(2) = 0%) compared to the buffered fluid group.There was no difference in intraoperative blood loss nor the volumes of intraoperative red cell or fresh frozen plasma transfused between groups. There was an increase in platelet transfusion in the non-buffered group which was statistically significant after analysing the transformed data (log transformation because the data were highly skewed).A number of metabolic differences were noted. There was a difference in postoperative pH of 0.06 units, lower in the non-buffered fluid group (95% CI 0.04 to 0.08, P < 0.00001, I(2) = 74%). However, this difference was not maintained on postoperative day one. The non-buffered fluid group also had significantly greater base deficit, serum sodium and chloride levels.There was no difference demonstrated in length of hospital stay and no data were reported on cost or quality of life.

AUTHORS' CONCLUSIONS

The administration of buffered fluids to adult patients during surgery is equally safe and effective as the administration of non-buffered saline-based fluids. The use of buffered fluids is associated with less metabolic derangement, in particular hyperchloraemia and metabolic acidosis. Larger studies are needed to assess robust outcomes such as mortality.

Perioperative buffered versus non-buffered fluid administration for surgery in adults

BACKGROUND

Perioperative fluid therapy influences clinical outcomes following major surgery. Fluid preparations may be based on a simple non-buffered salt solution, such as normal saline, or may be modified with bicarbonate or bicarbonate precursor buffers, such as maleate, gluconate, lactate or acetate, to better reflect the human physiological state. These latter fluids have theoretical advantages over normal saline in preventing hyperchloraemic acidosis. A number of clinical studies have now compared fluid preparations with and without a buffer to achieve a balanced electrolyte solution for perioperative fluid resuscitation.

OBJECTIVES

To review the safety and efficacy of perioperative administration of buffered versus non-buffered fluids for plasma volume expansion or maintenance in adult patients undergoing surgery.

SEARCH METHODS

We electronically searched the Cochrane Central Register of Controlled Trials (CENTRAL) (2011, Issue 4), MEDLINE (1966 to May 2011), EMBASE (1980 to May 2011), and CINAHL (1982 to May 2011). We handsearched conference abstracts and where possible, contacted leaders in the field.

SELECTION CRITERIA

We only included randomized trials of buffered versus non-buffered intravenous fluids for perioperative fluid resuscitation. The trials with other forms of comparisons such as crystalloids versus colloids and colloids versus different colloids were excluded. We also excluded trials using hypertonic fluids and dextrose-based fluids.

DATA COLLECTION AND ANALYSIS

Two authors independently extracted data and assessed the methodological quality of clinical trials. We resolved any disagreements by discussion. We contacted the trial authors to provide additional information where appropriate. We presented pooled estimates of the dichotomous outcomes as odds ratios (OR) and on continuous outcomes as mean differences, with 95% confidence intervals (CI). We analysed data on Review Manager 5.1 using fixed-effect models, and when heterogeneity was high (I² > 40%) random-effect models were used.

MAIN RESULTS

We identified 14 publications reporting 13 trials or comparisons with a total of 706 participants. For many of the outcomes reported, there was significant clinical and statistical heterogeneity. The primary outcome of mortality at any time was reported in only three studies with a total of 267 patients. The mortality rate was 2.9% for the buffered fluids group and 1.5% for the non-buffered fluids group but this difference was not statistically significant. The Peto OR was 1.85 (95% CI 0.37 to 9.33, P = 0.45, I(2)= 0%). Organ dysfunction was only presented for renal impairment. There was no difference in renal insufficiency leading to renal replacement therapy between the buffered and non-buffered groups (OR 0.61, 95% CI 0.23 to 1.63, P = 0.32, I(2) = 0%). Markers of organ system failure as assessed by urine output, creatinine and its variables (for renal function), PaC0(2) (respiratory function) and postoperative nausea and vomiting (gastro-intestinal function) showed a statistically significant difference only in PaC0(2) levels. The mean difference was 1.18 with lower PaC0(2) levels in the non-buffered fluid group (95% CI 0.09 to 2.28, P = 0.03, I(2) = 0%) compared to the buffered fluid group.There was no difference in intraoperative blood loss nor the volumes of intraoperative red cell or fresh frozen plasma transfused between groups. There was an increase in platelet transfusion in the non-buffered group which was statistically significant after analysing the transformed data (log transformation because the data were highly skewed).A number of metabolic differences were noted. There was a difference in postoperative pH of 0.06 units, lower in the non-buffered fluid group (95% CI 0.04 to 0.08, P < 0.00001, I(2) = 74%). However, this difference was not maintained on postoperative day one. The non-buffered fluid group also had significantly greater base deficit, serum sodium and chloride levels.There was no difference demonstrated in length of hospital stay and no data were reported on cost or quality of life.

AUTHORS' CONCLUSIONS

The administration of buffered fluids to adult patients during surgery is equally safe and effective as the administration of non-buffered saline-based fluids. The use of buffered fluids is associated with less metabolic derangement, in particular hyperchloraemia and metabolic acidosis. Larger studies are needed to assess robust outcomes such as mortality.

The Role of Endogenous Kallikrein Inhibition in Perioperative Transfusion and Adverse Outcome in Cardiac Surgical Patients

OBJECTIVE

The goal of this study was to explore the relationship among endogenous plasma kallikrein inhibition (KI), perioperative bleeding, and adverse outcomes in cardiac surgery.

DESIGN

A prospective, observational study.

SETTING

University teaching hospitals.

PARTICIPANTS

Cardiac surgical patients.

INTERVENTIONS

Endogenous plasma KI levels (%) and kallikrein-like activity (KKA) were measured preoperatively, 30 minutes into cardiopulmonary bypass, and at the end of surgery. Patients were divided into quartiles of preoperative KI. Data including risk factors, blood loss, transfusion requirements, and postoperative outcomes were collected.

MEASUREMENTS AND MAIN RESULTS

Preoperative endogenous KI ranged from 40% to 175%, where 100% represents the activity of pooled healthy volunteer plasma. The quartiles of KI levels were as follows: quartile 1, KI = 40% to 83.8% (n = 40); quartile 2, KI = 84% to 101.5% (n = 40); quartile 3, KI = 102% to 120% (n = 42); and quartile 4, KI = 121% to 175% (n = 38). The hematocrits on admission to the intensive care unit were as follows: quartile 1, 28% +/- 4%; quartile 2, 26% +/- 4%; quartile 3, 26% +/- 4%; and quartile 4, 24% +/- 4% (p = 0.009). Blood product use was similar among quartiles in the operating room. Quartiles 3 and 4 received more blood (p = 0.003) and platelet (p = 0.04) transfusions than quartiles 1 and 2 in the first 24 hours after surgery. More patients in quartile 4 were ventilated for more than 24 hours after surgery (p < 0.05). Hospital length of stay was longest in quartile 4 (p = 0.002).

CONCLUSION

Contrary to expectation, higher endogenous KI levels were associated with more blood product transfusion, longer postoperative mechanical ventilation, and hospital length of stay. These findings raise questions as to the role of KI in postoperative outcomes.

Low-dose endotoxin elicits variability in the inflammatory response in healthy volunteers

Endotoxin has been implicated as a cause of sepsis, inflammation and organ dysfunction after surgery. Patients differ in their response to endotoxin, and this may account for differences in outcome. The traditional human model of endotoxin challenge (2-4 ng/kg) is not associated with significant inter-individual variability in systemic inflammation and may not be suitable for studying variability in the inflammatory response. We examined whether low-dose regimens of endotoxin cause significant variability in inflammation. Volunteers (n = 30) were randomised in a double-blinded ('double-dummy') study to one of 6 dosing regimens: saline (placebo) or Escherichia coli O:113 endotoxin as a 4 ng/kg bolus (positive control), 0.25 ng/kg (bolus), 0.25 ng/kg (30 min infusion), 0.75 ng/kg (bolus) or 0.75 ng/kg (30 min infusion). Temperature, white cell count, platelet count, C-reactive protein and cytokine changes from baseline were measured. In contrast to subjects receiving placebo, those randomised to 4 ng/kg endotoxin exhibited significant systemic inflammation during the 10-h observation period. The four low-dose regimens elicited variability in most markers of inflammation. We conclude that low-dose endotoxin elicits inter-individual variability in inflammation and could be used to test factors that may affect the human response to endotoxin.

A randomized, double-blind comparison of lactated Ringer's solution and 0.9% NaCl during renal transplantation

Normal saline (NS; 0.9% NaCl) is administered during kidney transplantation to avoid the risk of hyperkalemia associated with potassium-containing fluids. Recent evidence suggests that NS may be associated with adverse effects that are not seen with balanced-salt fluids, e.g., lactated Ringer's solution (LR). We hypothesized that NS is detrimental to renal function in kidney transplant recipients. Adults undergoing kidney transplantation were enrolled in a prospective, randomized, double-blind clinical trial of NS versus LR for intraoperative IV fluid therapy. The primary outcome measure was creatinine concentration on postoperative Day 3. The study was terminated for safety reasons after interim analysis of data from 51 patients. Forty-eight patients underwent living donor kidney transplants, and three patients underwent cadaveric donor transplants. Twenty-six patients received NS, and 25 patients received LR. There was no difference between groups in the primary outcome measure. Five (19%) patients in the NS group versus zero (0%) patients in the LR group had potassium concentrations >6 mEq/L and were treated for hyperkalemia (P = 0.05). Eight (31%) patients in the NS group versus zero (0%) patients in the LR group were treated for metabolic acidosis (P = 0.004). NS did not adversely affect renal function. LR was associated with less hyperkalemia and acidosis compared with NS. LR may be a safe choice for IV fluid therapy in patients undergoing kidney transplantation.

Endotoxaemia during left ventricular assist device insertion: relationship between risk factors and outcome

BACKGROUND

Endotoxaemia, caused by splanchnic ischaemia during surgery, is believed to trigger systemic inflammation and cause postoperative organ dysfunction. A relationship between the plasma concentration of endotoxin during surgery and known risk factors for postoperative morbidity and mortality (e.g. age, abnormal gastric tonometric variables) and adverse outcome after surgery has not been demonstrated.

METHODS

In a prospective study, the plasma concentration of endotoxin was measured in 12 patients undergoing implantation of a left ventricular assist device. Automated air gastric tonometry was performed in all patients. The relationship between plasma endotoxin concentration, risk factors, and postoperative outcome was explored.

RESULTS

Carbon dioxide gap increased from 0.7 (0.3) to 3.6 (1.6) kPa at the end of surgery. Endotoxin was detected in one of 12 patients at baseline and in nine of 12 patients at the end of surgery (P=0.003). A high plasma concentration of endotoxin at the end of surgery was associated with a higher carbon dioxide gap (r=0.59, P<0.05), and a higher postoperative multiple organ dysfunction score (r=0.7, P=0.01).

CONCLUSIONS

The finding of an association between high intraoperative plasma concentrations of endotoxin, abnormal gastric tonometric variables and adverse outcome supports the view that endotoxaemia is caused by gut hypoperfusion during surgery and is associated with postoperative organ dysfunction.

Abnormal gastric tonometric variables and vasoconstrictor use after left ventricular assist device insertion

BACKGROUND

Abnormal gastric tonometric variables, a surrogate for splanchnic ischemia, occur in approximately 50% of patients at the end of routine cardiac operations and are associated with postoperative morbidity. We sought to determine whether gastric tonometric variables deteriorate after left ventricular assist device insertion and to explore the association between abnormal gastric tonometric variables and vasoconstrictor use.

METHODS

Nineteen patients who had insertion of a left ventricular assist device were enrolled in a prospective, observational study. Automated air tonometry was used to determine the difference between gastric and arterial partial pressure of carbon dioxide (CO2 gap) at five time points perioperatively.

RESULTS

Compared with baseline, systemic blood flow was significantly increased at the end of operation (1.9 +/- 0.6 versus 2.9 +/- 0.7 L x min(-1) x m(-2), p < 0.0001). Tonometric variables, which were normal at baseline, became abnormal in 90% of patients (baseline CO2 gap 4 +/- 2 mm Hg versus end of operation CO2 gap 24 +/- 15 mm Hg, p < 0.0001). Elevated CO2 gaps correlated with larger doses of norepinephrine (r = 0.69, p = 0.001) and vasopressin (r = 0.88, p < 0.0001). Abnormal gastric tonometric variables at the end of operation correlated with postoperative intensive care unit length of stay (r = 0.70, p = 0.0009) and multiple organ dysfunction score (r = 0.64, p = 0.0033).

CONCLUSIONS

Despite a significant increase in systemic blood flow after left ventricular assist device implantation, abnormal gastric tonometric variables developed and were associated with larger vasoconstrictor dose. These data provide evidence that gastric ischemia can develop independently of changes in systemic blood flow and support the potential role of vasoconstrictors as a cause of splanchnic ischemia.

Stroke after cardiac surgery: a retrospective analysis of the effect of aprotinin dosing regimens

BACKGROUND

Half-dose aprotinin (HDA) appears to be equivalent to full-dose aprotinin (FDA) in its ability to prevent bleeding. However, data regarding the potential effect of aprotinin use and dosage on other outcomes such as the occurrence of perioperative stroke are limited. It has been postulated that the higher level of kallikrein inhibition obtained with FDA dosing may be required for end-organ protection. Therefore, we performed a retrospective study in cardiac surgery patients at high risk for developing stroke to determine the relative effects of FDA and HDA regimens on the incidence of postoperative stroke.

METHODS

Records of 1,524 patients undergoing cardiac surgery over a 15-month period were reviewed. Patients at high risk for stroke were selected if they met all of the following predefined criteria: age greater than 70 years, history of hypertension, history of diabetes mellitus, history of stroke or transient ischemic attack, and presence of aortic atheroma. A validated preoperative stroke risk index was calculated for each patient. Postoperative stroke required confirmation by computed tomography or magnetic resonance imaging. Patients were divided into three groups according to whether they were administered no aprotinin, HAD, or FDA.

RESULTS

A total of 149 patients fulfilled the criteria for being at high risk for stroke. Stroke risk index was very similar (p = 0.56) in the three groups: those who received no aprotinin and served as a control group (124 +/- 15, n = 56), those who were given HDA (123 +/- 12, n = 67), and those who received FDA (122 +/- 11 n = 26). Preoperative and intraoperative characteristics were also similar between the three study groups. Overall, the incidence of stroke was 16% (24/149). The incidence of stroke differed (p < 0.05) among the three groups: no aprotinin 16% (9/56), HDA 22% (15/67), and FDA 0% (0/26).

CONCLUSIONS

In this retrospective study of cardiac surgery patients at high risk for postoperative stroke, the administration of FDA but not HDA was associated with a lower incidence of stroke.