Comparison of normal saline and balanced crystalloid (plasmalyte) in patients undergoing elective craniotomy for supratentorial brain tumors: A randomized controlled trial
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.241347
Source of Support: None, Conflict of Interest: None
Keywords: Balanced crystalloid, metabolic profile, normal saline, serum osmolality
Normal saline and ringer lactate are the crystalloids commonly used in neurosurgical procedures and this influences the metabolic status, solute load to the kidney and brain edema. The effective osmolarity of ringer lactate is 278 mosm/L. Infusion of large volumes of ringer lactate has the potential to cause cerebral edema,, whereas normal saline results in hyperchloremic metabolic acidosis, which may have numerous clinical implications, including a theoretical risk of acute kidney injury.,,, It has been found to result in afferent arteriolar vasoconstriction, reduced glomerular filtration rate and delayed micturition following surgery.
The balanced crystalloid (BC), Plasmalyte is a relatively newer crystalloid, the composition of which is akin to that of plasma. It contains electrolytes in a similar concentration as that of plasma. The actual osmolality of BC is 295 mosm/kg and its pH is 7.4. Being more physiological in composition than normal saline, it may help circumvent the problems associated with an overzealous use of 0.9% saline. BC is expected to have a more favorable effect on the metabolic profile compared to normal saline. A thorough search of the available literature did not reveal any studies comparing the effects of these two crystalloids on brain relaxation and plasma osmolality.
Several biomarkers which facilitate early detection of acute kidney injury have been studied. Of these, NGAL (neutrophil gelatinase associated lipocalin) appears to be most widely used. It has been shown to predict the risk of acute kidney injury (AKI) as early as within 6 hours of a clinical insult. In contrast, the conventionally employed biomarkers like creatinine start to rise 24-72 hours after the injury has occurred. Besides, serum creatinine values are affected by several other factors. Thus, NGAL may be used as a surrogate marker of AKI, which occurs as a consequence of hyperchloremic metabolic acidosis associated with administration of large volumes of normal saline. Early detection of AKI may facilitate the institution of preventive or corrective measures to manage the renal injury and thereby reduce its burden on health care. Our study was designed with the aim of comparing normal saline and BC in terms of their effects on metabolic profile, intraoperative brain relaxation and the risk of acute kidney injury in the perioperative period.
After obtaining approval from the scientific advisory committee and the Institute Ethics Committee (IEC), the trial was registered with the Clinical Trials Registry India (CTRI/2017/06/008771). Assuming a level of significance of 5% and a power of 90%, in order to detect a mean difference of 0.03 in pH between the two study groups with a standard deviation of 0.03 in the saline group and 0.02 in the BC group, the sample size was calculated to be 22 in each group.
After obtaining a written informed consent, 44 patients, aged 18-60 years, belonging to the American Society of Anesthesiologists (ASA) class 1/2/3, posted for elective craniotomy for supratentorial tumor resection were included in the study [Figure 1]. Patients with preexisting renal failure (defined as serum creatinine >2), electrolyte imbalance, those who were already receiving mannitol, and those with a preoperative Glasgow Coma Score (GCS) < 13 were excluded from the study. Baseline complete hemogram and renal function tests (serum urea, serum creatinine) were noted for all the patients.
In the operation theatre, standard monitors (electrocardiogram [ECG], non-invasive blood pressure monitoring [NIBP], and peripheral capillary oxygen saturation [SpO2]) were attached and baseline parameters were noted. An 18 G intravenous cannula was inserted into one of the forearm veins. Patients were then randomized into two groups of 22 patients each (NS -normal saline, P -BC). The patients in the NS group received normal saline as the maintenance fluid during the intraoperative period. Patients in the P group received BC (Plasmalyte) as the maintenance fluid throughout the intraoperative period. Randomization was done using random allocation software. Allocation concealment was achieved using the Serially Numbered Opaque Sealed Envelope (SNOSE) technique. The operating neurosurgeon was blinded to the infused fluid by covering the label. Standard induction protocol (fentanyl 2 micrograms/kg, propofol 2 mg/kg and vecuronium 0.1 mg/kg) was followed in all the patients. Direct laryngoscopy and tracheal intubation were performed 3-5 minutes after administration of the neuromuscular blocker. After intubation and initiation of mechanical ventilation, the right subclavian vein was cannulated using a 7 Fr triple lumen central venous cannula. One of the radial arteries was cannulated using a 22 G cannula and invasive blood pressure was monitored.
Maintenance fluid was given at the rate of 2 ml/kg/hour. Mannitol 1gm/kg was administered to all patients before opening of dura. Blood transfusion was carried out in accordance with the ASA practice guidelines. Baseline arterial sample for blood gas analysis was obtained for all patients. Thereafter, blood gas analysis was done once every two hours. In addition, venous samples were collected in ethylene diamine tetra acetic acid (EDTA) tube before induction and once in three hours thereafter, for the measurement of plasma osmolality. Osmolality was measured using the Advanced ® 3250 Single-Sample osmometer (Advanced Instruments INC., Massachusetts, USA).
A venous sample was collected from all the study participants for the measurement of serum NGAL, 3 hours after the conclusion of the surgery. NGAL, was estimated using an enzyme linked immunosorbent assay (ELISA) kit (human lipocalin-2/NGAL ELISA, Biovendor Research and Diagnostic Products). Instructions given in the operator's manual were followed. A cut off of 150 ng/ml was selected to have a high sensitivity, and a cut off of 400 ng/ml was selected to have a high specificity for the diagnosis of acute kidney injury, in accordance with previous studies.,
The blood loss, blood transfused (if any), volume of crystalloids given, volume of colloids given and urine output were also noted during the intra-operative period. On postoperative day 1, complete hemogram and renal function tests were repeated. Renal function test was also repeated on the 5th postoperative day. In addition, the patients were monitored for clinical signs and symptoms of renal failure, till their discharge from the hospital.
The primary outcome measure was the metabolic profile (primarily pH, base excess and chloride levels) in the study participants. It was postulated that BC would maintain a more favorable metabolic profile compared to normal saline. Plasma osmolality was the other primary outcome measure. The brain relaxation score and serum NGAL value were the secondary outcome measures. The brain relaxation score was assessed by the operating surgeon, who was blinded to the fluid infused on a 4-point scale (1-perfectly relaxed, 2 – satisfactorily relaxed, 3 – firm brain, 4 – bulging brain) and the same was documented.
Statistical Package for the Social Sciences [SPSS] version 19.0 (Armonk, NY: IBM Corp) was used for carrying out the statistical analysis. Comparison of age and weight between the two groups was done using unpaired student t test. Categorical variables like gender and ASA category were expressed as proportions (percentages). The difference in metabolic parameters (pH, pCO2, HCO3-, lactate, anion gap, base excess) between the two groups was analyzed using the two-way repeated measures analysis of variance (ANOVA). The difference in electrolytes (sodium, potassium, calcium, chloride), hematocrit, plasma osmolality and renal function parameters between the two groups was also analyzed using the two-way repeated measures ANOVA. Serum NGAL was analyzed using the unpaired t test. Brain relaxation scores between the two groups was analyzed using the Mann Whitney U test. The differences between the two groups with respect to duration of surgery, blood loss, blood transfused, volume of crystalloids given and urine output were analyzed using the unpaired Student t test. A probability value (p value) of less than 0.05 was considered significant.
A total of 52 patients were assessed for eligibility. Of these, 44 patients were randomized into two groups of 22 each. Patients in the group NS received normal saline as the maintenance fluid intra-operatively, and patients in the group P received BC.
The two groups were comparable in terms of demographic parameters [Table 1]. The duration of surgery was comparable between the two groups (P value = 0.51).
The metabolic profile was more favorable in patients who received BC as the maintenance fluid. There was a statistically significant difference between the two groups in terms of pH [Table 2]. We found that the pH was significantly lower in patients who received normal saline compared to those who received BC. However, the pH values were within the normal physiological range in both the groups. There was a trend towards acidosis in patients who received normal saline.
The two groups were not significantly different in terms of partial pressure of carbon dioxide. Similarly, the two groups appeared to be comparable with reference to bicarbonate levels, with a declining trend observed in both the groups. The absolute values were lower in those who received normal saline. We also noted a significant inter-group variability in the anion gap values. The anion gap values were higher in those who received normal saline, compared to those who received BC [Table 2].
Electrolytes and hematocrit
We found a significant difference in chloride levels between the two groups (p value = 0.03) [Table 3]. The chloride levels were higher in those who received normal saline. There was no significant difference with reference to sodium, potassium, calcium and hematocrit between the two groups.
The change in osmolality over time was not significantly different between the two groups (P value = 0.16). The baseline plasma osmolality (measured before induction) was similar in both the groups. The plasma osmolality values were higher in those who received normal saline [Table 4]. This may be attributed to the higher sodium and chloride content of normal saline. However, the values were within the normal physiological range in both the groups.
Brain relaxation score
Brain relaxation was assessed on a 4-point scale by the operating surgeon who was blinded to the crystalloid used. There was no significant difference between the two groups in terms of brain relaxation scores [Table 5].
Renal parameters and serum NGAL
The two groups were homogeneous in terms of baseline serum urea and creatinine. However, we found a significant difference between the two groups with respect to postoperative urea and creatinine levels [Table 6]. Although the values were well within the reference range in both the groups, the serum urea on the 5th postoperative day (mean ± standard deviation) in the normal saline group was 28.5 ± 15.9 and it was 24.82 ± 8.28 in the BC group. We observed a similar trend in the serum creatinine values. While the mean creatinine value in the normal saline group on the 5th postoperative was 0.98 ± 0.22, the value in the BC group was 0.74 ± 0.18 (P value <0.001). However, the serum creatinine values in both the groups were within the physiological range. All the patients were monitored for clinical signs and symptoms of renal failure till discharge. None of them was found to develop clinically significant AKI. Serum NGAL was measured 3 hours after completion of the surgery. The NGAL value in those who received normal saline (mean ± standard deviation) was 249.13 ± 91.0 ng/ml. The serum NGAL value in those who received BC was 71.13 ± 28.7 ng/ml (P value <0.001) [Table 6]. The mean NGAL in those who received normal saline was well above the cut-off value predictive of acute kidney injury derived from prior studies (150 ng/ml). However, the subsequent clinical follow up did not reveal AKI in any of these patients.
We found a significant difference in blood loss between the two groups (p value 0.01). The blood loss was found to be higher in patients who received normal saline intra-operatively [Table 7].
Our study was designed as a prospective, randomized controlled trial to compare the effects of the two crystalloids: normal saline and BC, on the metabolic status of patients scheduled to undergo elective craniotomy for supratentorial tumor resection. The primary objective was to compare the metabolic status (pH, chloride, base excess) among the two groups of patients. We also compared the brain relaxation scores and the risk of acute kidney injury in the postoperative period between the two groups by measuring serum parameters.
Acid base status
The acid base status was more favorable in patients who received BC. We found a significant difference between the two groups in terms of pH (p = 0.004). The mean pH was significantly lower in the normal saline group compared to the BC group at all-time points during the intraoperative period, even though the mean values in both the groups were maintained within the normal physiological range. Hadimioglu et al., compared the effects of normal saline, ringer lactate and BC on acid base status after kidney transplant. They found a statistically significant decrease in pH, in patients who received normal saline intra-operatively (7.44 ± 0.50 vs 7.36 ± 0.05) as against those who received BC. In a similar study conducted by Kim et al., on renal transplant recipients, a statistically significant reduction in pH was found in those who received normal saline, compared to those who received BC. Similarly, comparing these two crystalloids for the initial resuscitation of trauma patients, Young et al., noted a more favorable acid base status in those who received BC. They found that the pH at the end of 24 hours in patients who received normal saline in the initial phase of resuscitation was 7.37 ± 0.07, and it was 7.41 ± 0.06 in those who received BC. The findings of our study was similar to theirs.
We did not find a statistically significant difference with reference to base excess between the two groups (p = 0.17), unlike prior studies. Young et al., conducted a study comparing normal saline and BC in the initial resuscitation of trauma patients. They found a significant improvement in base excess over 24 hours in patients who were resuscitated with BC as against those who received normal saline for the same (7.5 ± 4.7 vs 4.4 ± 3.9 mmol/L). In their study on living donor kidney transplant recipients, Kim et al., found that the base excess was significantly reduced in those who received normal saline. Studies which found a statistically significant difference in base excess between the two groups were conducted in the setting of trauma or critical care; on the other hand, the patients in our study belonged to the ASA categories 1 and 2 and were posted for elective neurosurgical procedures. Therefore, their preoperative metabolic status was well preserved, unlike in the previous scenarios. Also, in our study, base excess was measured during the intra-operative period, which lasted for about 9 hours in both the groups, as opposed to previous studies which studied change in base excess over 24 hours. These factors might be responsible for the findings of our study.
Normal saline causes hyperchloremic metabolic acidosis due to its supraphysiological chloride content. We found a statistically significant difference between the two groups with respect to changes in chloride levels over time (p = 0.03). This may be explained by the fact that hyperchloremia is an expected consequence of administration of large volumes of normal saline. In our study, the total volume of fluid administered was comparable between the two groups. Yunos et al., found that a chloride restrictive intravenous fluid strategy resulted in a significant reduction in the incidence of acute kidney injury in critically ill patients. In their study on major trauma patients, Jin Young Lee et al., found that hyperchloremia (defined in their study as serum chloride >110 mEq/l) was positively associated with a 30-day mortality. In our study, the acidosis which developed in patients receiving normal saline was purely metabolic. There was no significant difference with respect to partial pressure of carbon dioxide between the two groups. This, in turn, may be attributed to the hyperchloremia in the normal saline group. In their study comparing normal saline and BC for the resuscitation of trauma patients, Young et al., found that the chloride level 24 hours after admission was higher in those who received normal saline compared to those who received BC (111 ± 8 vs 104 ± 4 mEq/l). Similarly, Shaw et al., comparing 0.9% saline and BC in patients undergoing open abdominal surgeries, found a statistically significant increase (P value <0.001) in the incidence of renal failure mandating dialysis in patients who received normal saline. Even though this may not be accounted for by hyperchloremia alone, it was postulated that the incidence of acid base disorders and electrolyte disturbances was higher in those who received normal saline.
We noted a statistically higher values in anion gap over time (P value <0.001) in those who received normal saline as the maintenance fluid. This could be attributed to the hyperchloremic metabolic acidosis which occurred as a consequence of infusion of large volumes of normal saline. The anion gap is maintained within the normal range (12-18 mmol/l) in hyperchloremic metabolic acidosis. In other words, the acidosis due to hyperchloremia may be classified as normal anion gap metabolic acidosis. Although there was a statistically significant difference between the two groups, the mean values were within the normal reference range in both the groups.
Maintenance of plasma osmolality in the normal or a mildly hyperosmolar range is a pre-requisite for fluid management in neurosurgical patients. Plasma osmolality was measured at predetermined time points throughout the surgery (baseline, 3 hours, 6 hours, 9 hours). Both the groups were comparable in terms of baseline osmolality. We also did not find any significant difference between the two groups with respect to change in osmolality over time (p = 0.16). However, the mean plasma osmolality values were higher in those who received normal saline than in those who received BC. The mean values at 3rd, 6th and at 9 hours post induction were 305.2 ± 11.5, 307.5 ± 12.4 and 310.4 ± 17.2 mosm/kg respectively, in patients who received normal saline. The mean values at corresponding time points in those who received BC were 293.3 ± 11.2, 295.2 ± 8.0 and 293.5 ± 3.2 mosm/kg respectively. These changes are due to the higher sodium and chloride content of normal saline, compared to BC. BC, being an iso-osmolar fluid, maintained plasma osmolality within the physiological range. We, however, could not find any study comparing the osmolality of these two solutions.
Brain relaxation score
The changes in plasma osmolality may have a significant impact on the intraoperative brain relaxation. The brain relaxation score was comparable between the two groups in our study (p = 0.65). Although BC is relatively hypoosmolar compared to normal saline (295 mosm/kg vs 308 mosmkg), it did not appear to have any adverse impact upon brain relaxation in our study. We could not find any prior study comparing these two fluids in terms of their effect on cerebral relaxation.
Renal parameters and the risk of acute kidney injury
Several studies have postulated that the hyperchloremic metabolic acidosis which occurs as a consequence of infusions of large volumes of normal saline can compromise renal function by causing splanchnic vasoconstriction and reduction in glomerular filtration rate. Serum creatinine, which is probably the most widely used index of renal function at present, starts to rise 24 to 72 hours after a clinical insult. On the contrary, urine and serum levels of NGAL, which is a bio-marker of acute kidney injury, start peaking as early as within 6 hours of a clinical injury. In our study, serum samples for the measurement of NGAL were collected 3 hours after the completion of surgery. We found a very significant difference in the values between the two groups. The mean serum NGAL was significantly elevated in patients who received normal saline (249 ± 91.03 vs 71.13 ± 28.69 ng/ml, P value <0.001). Shapiro et al., found serum NGAL values >150 ng/ml to be 96% sensitive and 51% specific for acute kidney injury. However, Di Somma et al., found that NGAL had the highest predictive value for in-hospital mortality when 400 ng/ml was considered as the cut-off value.
We also observed a significant difference in the postoperative creatinine values between the two groups. The mean serum creatinine, measured on the 5th postoperative day, were 0.98 ± 0.22 and 0.74 ± 0.18 mg/dl, respectively, in the normal saline and in the BC group. However, the two groups were comparable in terms of baseline creatinine values. It may be worthwhile to note that even though the serum creatinine values were comparatively higher in those who received normal saline, the values in both the groups of patients were well within the reference range. None of the patients developed clinically significant AKI perioperatively. We observed a similar trend with respect to serum urea. While the mean serum urea values on the 5th post-operative day were higher in those who received normal saline, the values were within the normal physiological range in both the groups. The patients were followed up clinically till their discharge. None of them developed renal failure mandating hemodialysis.
NGAL serves to detect sub-clinical AKI. Therefore, it may be useful in the institution of preventive or salvage measures to manage AKI. The widespread use of this marker at present may be limited by the lack of a consensus on a cut-off value with an acceptable sensitivity and specificity, given that AKI has a multifactorial aetiology, whereas NGAL is predominantly elevated in renal tubular injury.
To our knowledge, no study comparing the effects of normal saline and BC on the metabolic status and the brain relaxation score of neurosurgical patients has been undertaken previously. Our study may be a small, but significant step, in this direction.
Advantages and limitations
Although there are several studies comparing the effects of normal saline and BC on the acid base status of patients in the setting of trauma and organ transplant, there is a paucity of literature comparing these two intravenous fluids with respect to their effects on neurosurgical patients. Despite several studies calling into question the practice of using normal saline indiscriminately, it continues to be used across a wide range of clinical scenarios. Being mildly hyperosmolar to plasma, it is commonly used in the neurosurgical patient. However, studies have shown that hyperchloremic metabolic acidosis which occurs as a consequence of infusion of large volumes of normal saline, poses a risk for acute kidney injury.,,, Our study aimed to compare these two fluids in terms of their effect on acid base status as well as intra- operative brain relaxation. Our study proves that BC is a promising alternative to normal saline in these patients, since it circumvents the problems associated with overt administration of normal saline, while maintaining the plasma osmolality as well as brain relaxation.
However, our study has several limitations. Serum NGAL was found to be significantly elevated in patients who received normal saline (p value <0.001). However, none of these patients developed clinically significant AKI. Due to financial constraints, we could measure serum NGAL only once. Serial measurements might have helped us to decide whether or not this finding was of any clinical significance. Urinary NGAL is more specific for detection of AKI compared to serum NGAL, because the latter may have extra-renal sources. Combining serum NGAL with other available biomarkers of renal injury like cystatin C, might have been more reliable in predicting the risk of AKI.
We conclude that the newer BC may be a wiser choice for fluid management in neurosurgical patients. The BC maintained the metabolic status more favorably than normal saline in our study. The postoperative renal parameters were also better maintained in those who received BC. There was no significant difference between the two groups with respect to brain relaxation score. Serum NGAL was found to be very significantly elevated in those who received normal saline, indicating that there may be a risk of AKI in these patients. However, meticulous clinical follow up did not reveal any adverse renal outcome in any of these patients. Further studies may be necessary to ascertain the clinical significance of this finding.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]