1. Background
Liberal infusion of intravenous (IV) fluids, whether in the form of a crystalloid solution or colloid-crystalloid combination, is usually taken into consideration in major surgeries (1). Nonetheless, studies on the effects of IV fluid infusion are limited to complications after the surgery, in which the protocols have challenged the prescription of liberal IV fluids and led to the design and implementation of various studies on the superiority of prescribing a restrictive volume of fluids over liberal fluids (2, 3). An earlier review study suggested that high volumes of fluids are accompanied by an increased risk of postoperative complications such as respiratory, cardiac, and even renal failures (4). It seems that the restriction of fluids reduces cell damage processes such as oxidative stress, by modulating the level of cell oxygenation (5, 6).
The level of tissue oxygenation is directly affected by the cardiac output and the volume of circulating fluid, keeping in mind that the excessive increase of fluid in circulation will apply extra cardiac loads (7, 8). In general, the strategy of substituting the restrictive volume of fluids is acceptable for certain surgeries (9, 10). Monitoring the renal function while maintaining the cardiac output will help find the optimum strategy of intraoperative fluid therapy.
2. Objectives
This study was conducted to evaluate the effects of restrictive versus liberal fluid administration on the renal function of patients in laparoscopic gastric bypass surgery.
3. Methods
In a double-blinded randomized clinical trial, 72 adult candidates of laparoscopic bariatric surgery (Roux-en-Y Gastric Bypass (RYGB) or Sleeve gastrectomy (SG)) with a BMI of more than 40 were recruited. The exclusion criteria included a history of renal disease, a history of cardiopulmonary disease, and the use of diuretics. The participants were randomly assigned into two groups of restrictive or liberal fluid therapy. Block randomization was used to assign an equal number of patients to each group. Both patients and the outcome assessor were blinded to the assignments. The local ethics committee approved the study protocol, and informed consent was obtained from all patients. The clinical trial was registered under the code of IRCT20170109031852N3.
Standard monitoring was implemented. In the restrictive group, half of the fluid deficit up to 500 ml was infused at preinduction, and total maintenance fluid according to the 1-2-4 formula was infused hourly, without substitution of third space. Blood loss was replaced with crystalloids in a 3:1 ratio up to the allowable blood loss (ABL). In the liberal group, 5 mL/kg of crystalloids were infused for volume expansion before anesthesia induction. Half of the estimated volume deficit was delivered in the first hour and the rest in the second and third hours of the surgery according to the 1-2-4 formula. The third-space losses were replaced by 3 mL/kg and blood loss with three times of crystalloids up to the ABL. All patients in both groups received Ringer’s lactate and normal saline solutions sequentially with a 1:1 ratio.
The anesthesia protocol in both groups was identical with midazolam 2 mg, fentanyl 2 µg/kg, and lidocaine 1 mg/kg as premedication, and propofol 2 mg/kg and atracurium 0.5 mg/kg for induction. Anesthesia was maintained with isoflurane 0.8 - 1.5%, fentanyl, and atracurium with BIS values in the range of 40 - 60. Drops in blood pressure were treated with the least effective doses of ephedrine. High blood pressures were controlled with incremental doses of labetalol. Blood samples were taken one hour before and 24 hours after the surgery, and renal function indices, including BUN, creatinine, creatine kinase, and GFR, were evaluated. Urinary catheterization with single-use Nelaton was performed for patients before awakening at the end of surgery, and urine volume was measured.
The sample size was calculated based on the standard formula to estimate the difference in means between two independent populations, where α = 0.05, β = 0.2, and the s1 and s2 were considered to be 0.1 and 0.2, respectively, for creatinine values. The estimated required sample size in each group was 36 participants. The normality of quantitative variables was assessed with the Kolmogorov-Smirnov test. The Student t test was used to compare quantitative variables and the chi-square test to compare nominal variables between the two groups. A paired t test was used for within-group analyses of quantitative variables. All comparisons were two-tailed. The P-values of < 0.05 were considered statistically significant. Statistical analyses were performed with SPSS version 22.0 software (SPSS, Inc., Chicago, IL, USA).
4. Results
The demographic variables, type, and duration of surgery were comparable between the two groups (Table 1). Inter-group comparisons showed no significant difference in BUN, creatinine, creatine kinase, and GFR both before and after the surgery (P > 0.05). However, intragroup comparisons showed significant differences before and after the surgery in both groups (Table 2). BUN, creatinine, and GFR in the liberal fluid group showed significant differences before and after the surgery (P < 0.05). Similarly, BUN decreased after the surgery in patients who received restrictive fluid (P < 0.05). The volume of crystalloid infusion was 1824 ± 310 ml in the liberal group and 746 ± 127 ml in the restrictive group. The intraoperative urine volume was 305 ± 39 mL in the liberal fluid group and 116 ± 42 ml in the restrictive fluid group (P = 0.25).
Table 1.Demographic and Baseline Variables in Two Groupsa
| Variable | Fluid Regimen | P Value | |
|---|---|---|---|
| Liberal (N = 36) | Restrictive (N = 36) | ||
| Age (years) | 38 ± 10 | 38 ± 9 | 0. 92 |
| Male Sex, No. (%) | 13 (36.1) | 12 (33.3) | 0.97 |
| Height (cm) | 182 ± 24 | 175 ± 15 | 0.28 |
| BMI (kg/m2) | 40.1 ± 3.3 | 40.9 ± 2.3 | 0.29 |
| Type of surgery | 0.87 | ||
| Roux-en-Y Gastric Bypass | 24 (66.6) | 22 (61.1) | |
| Sleeve gastrectomy | 12 (33.3) | 14 (38.8) | |
| Duration of surgery (min) | 114 (28) | 108 (19) | 0.46 |
aData are presented as mean (SD).
Table 2.Renal Function Indices in Two Groups Before and After Surgery
| Variable/Time | Fluid Regimen | P Valuea | |
|---|---|---|---|
| Liberal (N = 36) | Restrictive (N = 36) | ||
| BUN | |||
| Before surgery | 28.3 ± 9.8 | 27.6 ± 9.5 | 0.07 |
| After surgery | 24.8 ± 9.4 | 24.3 ± 6.2 | 0.8 |
| P valueb | 0.01 | 0.045 | |
| Creatinine | |||
| Before surgery | 0.95 ± 0.24 | 0.91 ± 0.17 | 0. 24 |
| After surgery | 0.88 ± 0.16 | 0.9 ± 0.12 | 0.29 |
| P value | 0.001 | 0.88 | |
| Creatine kinase | |||
| Before surgery | 185 ± 108 | 186 ± 107 | 0. 98 |
| After surgery | 165 ± 86 | 191 ± 114 | 0.47 |
| P value | 0.18 | 0.39 | |
| GFR | |||
| Before surgery | 179 ± 64 | 179 ± 39 | 0. 54 |
| After surgery | 192 ± 63 | 178 ± 43 | 0.23 |
| P value | 0.01 | 0.58 | |
aP value: intergroup comparisons.
bP value: intragroup comparisons before and after surgery.
5. Discussion
The two groups were comparable in terms of age, gender, height, and BMI, and these confounding factors were unlikely to affect the results. In inter-group comparisons, there was no significant difference in BUN, creatinine, creatine kinase, and GFR between patients receiving restrictive or liberal fluid before and after the surgery. In addition, only BUN decreased after the surgery in both groups, which could be a dilutional effect, and lower indices of BUN and creatinine and a higher GFR index in the liberal fluid therapy group than in the restrictive group could be due to the difference in the volume loading of patients. Therefore, it seems that the two methods of liberal and restrictive fluid therapy have comparable effects on traditional kidney function indices in laparoscopic bariatric surgery, keeping in mind that this study could not investigate the clinical significance of the modest observed statistical differences.
The safety of limited volume treatment fluid has been challenging for physicians over the last years. Their main concern is the potential of hypovolemia and organ dysfunction after the surgery, such that acute renal injury is the main concern. Systematic reviews suggest that the liberal fluid regimen may result in extra-vascular losses and end-organ damages, namely kidney injuries. Thus, the restrictive fluid regimen, together with individualized goal-directed colloid administration to maintain a maximal stroke volume, could be the optimal strategy (3). However, there are limited studies so far on the comparison of the effects of liberal and restrictive fluid on renal function indices in laparoscopic gastric bypass surgery. An earlier study reported no difference in the creatinine level of patients suffering from early respiratory and renal failures during major abdominal surgeries between the two groups of limited (5 mL/kg/h) and free (10 mL/kg/h) volumes of fluid therapy (7). In a more recent study, the rate of acute kidney injury (AKI) was higher after the surgery in patients undergoing major abdominal surgeries who had received limited volumes of fluid than in those who were on a more liberal fluid regimen. This inconsistency might be due to the prescription method of the treatment fluid and the surgical method applied to patients in the two studies (11). Neutral findings in renal complications after abdominal surgeries between patients receiving liberal and restrictive regimens have been reported in meta-analysis and clinical trials, as well (10, 12-14).
Renal blood flow, GFR, and urine output will decrease in laparoscopic surgeries due to increased intra-abdominal pressure. Therefore, the monitoring of urine output for the adequacy of hydration could be misleading. Obesity makes this index more challenging. Serial serum creatinine measurements are most often used to detect acute kidney injury. However, there is a delay between the onset of kidney injury and the rise in serum creatinine levels. Several novel markers for the early detection of AKI have been introduced. Of those, cystatin C, β-trace protein, and β-2 microglobulin are used to detect filtration-based renal dysfunction (15-17). As a limitation of this study, the early markers of AKI were not available, and we did not measure serum and plasma osmolality as a marker of hydration status.
Finally, it seems that the two regimens of liberal and restrictive fluid therapy have comparable effects on traditional renal function indices, including BUN, creatinine, creatine kinase, and GFR in laparoscopic bariatric surgery. The clinical significance of observed differences in the outcome should be investigated in further studies. Avoidance of excessive crystalloids and goal-directed fluid therapy may decrease the cardiac load and reasonably reduce the damage to end-organs, such as the renal system (18-20). Further studies with the early biomarkers of AKI are warranted.
