Brain-Relaxing Effect of Different Diuretic Regimens in Supratentorial Tumor Surgery: A Comparative Study Guided by Optic Nerve Sheath Diameter.

BACKGROUND: Hyperosmolar therapy is a well-established method to approach brain relaxation during craniotomy. Mannitol is used with a wide range of dosing regimens, combination with loop diuretics exerts a synergistic effect resulting in both reduction of the dose and its complications. Ultrasound measurement of optic nerve sheath diameter (ONSD) gives reliable information about intracranial pressure (ICP) and avoids overdosing and complications of osmotherapy. AIMS AND
OBJECTIVES: In this study, we compare the ordinary dose of mannitol with the low dose combined with furosemide and detect the effect on ICP by ONSD. SETTING AND
DESIGN: This is a prospective, randomized, double-blind study involving 60 patients undergoing supratentorial brain tumor surgery.
MATERIALS AND METHODS: Sixty patients were enrolled in this study, divided into two equal groups: Group M received mannitol 1 g.kg-1: while Group F received mannitol 0.25 g.kg-1 and furosemide 0.5 mg.kg-1. Reduction in ONSD measurement was the primary objective, while brain-relaxation score (BRS), hemodynamic changes, urine output, serum lactate, and changes in serum electrolyte were the secondary objectives. STATISTICAL ANALYSIS: Data collected were analyzed using SPSS software, IBM, USA, version 22. P value was considered significant if <0.05.
RESULTS: ONSD and BRS showed no statistically significant difference between the studied groups. After diuresis, Group M showed significant reduction in heart rate and mean arterial blood pressure, serum sodium, potassium, and lactate (P = 0.02, P = 0.02, P = 0.001, P = 0.001, P = 0.001, P = 0.001 respectively), with increased urine output (UOP) and fluids replacement (P = 0.00, P = 0.01, respectively).
CONCLUSION: Compared to high dose, adding loop diuretics to low-dose mannitol during supratentorial brain tumor surgeries resulted in comparable BRSs with a lower incidence of hemodynamic and metabolic disturbances. Copyright:

INTRODUCTION

Brain-relaxation measures during brain tumor surgeries are mandatory as they improve the operating conditions, reduce retraction pressure, and improve patient's outcome.[1] Besides hyperventilation, osmotherapy considered the cornerstone in the management of cerebral edema. For decades, mannitol is the most commonly used agent.[2] However, it has several adverse effects including hypochloremic metabolic alkalosis associated with volume contraction and diuresis, electrolytes disturbance, heart, and renal dysfunction or nephrotoxicity. Although widely and longly using mannitol, there is no consensus about its safe, effective dose, and duration of administration.[3]

Mannitol and furosemide combination exerts synergistic action, as furosemide enhances the effect of mannitol on plasma osmolality, resulting in a greater reduction of brain water content, relaxing the brain, lowering intracranial pressure (ICP) effectively, and hence lowering mannitol dose and side effects.[45]

Many techniques can be applied to monitor brain relaxation and ICP during craniotomy. A catheter is inserted inside the ventricular system for the detection of ICP changes, but this is an invasive technique and carries the risk of infection and bleeding.[6] Furthermore, the brain relaxation score (BRS) is described in many studies for observation of brain state during craniotomy, but still is a subjective not conclusive method.[78] The use of optic nerve sheath diameter (ONSD) is a newly proposed technique for the diagnosis of increased ICP. Some studies in different parts of the world have observed that the changes in ONSD are strongly in consonance with computed tomography (CT) scan image findings suggestive of an increase in ICP.[91011]

In our trial, we compared the ordinary dose of mannitol with the lowest prescribed dose combined with furosemide and detected the effect on ICP depending on ONSD reduction.

MATERIALS AND METHODS

This prospective randomized blinded study was approved by our Institutional Review Board, and informed consent was obtained from all subjects participating in the trial. Before patient enrolment, the trial was registered in the Pan African Clinical Trial Registry (PACTR-202003536095702, date of registration: March 12, 2020). Sixty patients were enrolled in this study which adheres to the applicable CONSORT guidelines [Figure 1] from March to September 2020. Included patients were adults of both sex belonged to American Society of Anesthesiologists physical status classes I or II aging 18–65 years and had Glasgow Coma Scale score >13 scheduled for supratentorial brain tumor surgery. Exclusion criteria were patient's refusal, major cardiopulmonary disorders, hepatic or renal dysfunction, diagnosed optic nerve disease (neuritis, trauma, or pathology), and known allergy to used drugs. Random number generator with closed envelope technique randomized patients into two groups based on osmotherapy regimen: 30 for mannitol group (M Group) and 30 for mannitol–furosemide group (F Group).

Figure 1

Consort flow diagram for the study

All patients were subjected to routine preoperative assessment including (history, clinical examination, Glasgow coma score assessment, electrocardiography (ECG), echocardiography, complete blood count, liver functions, renal functions, serum lactate, sodium, potassium, and coagulation profile). In the operative suite, patients were connected to monitor (General ElectricDatex B850, USA) for monitoring ECG, non-invasive blood pressure, and peripheral oxygen saturation (spO2). Premeditations included intravenous injection of pantoprazole (Zurcal 40 mg, AUG pharma, Spain), dexamethasone (Dexamethasone, Sigmatic, Egypt) 8 mg, and 3 mg of midazolam (Midathetic, Amoun pharmaceuticals, Egypt).

Anesthesia was induced using propofol 1–2 mg.kg−1 (Diprivan, Fresenius KABI, USA), fentanyl 1 μ.kg−1 (Fentanyl Hameln, Hameln pharmaceuticals, Germany), and atracurium 0.6 mg.kg−1 (Atrabesylate, Egypharm, Egypt). A proper-sized endotracheal tube was inserted and fixed in place after confirmation of correct positioning. Patients were ventilated using volume-controlled ventilation mode to keep end-tidal CO2 (EtCO2) between 30 and 32 mmHg. Anesthesia was maintained using sevoflurane 1%–2% in 40% oxygen–air gas mixture, a top-up dose of atracurium given on needs, and fentanyl infusion 1 μg.kg−1.h−1 and 1 g of paracetamol were infused as a part of the multimodal analgesia technique. Prior to the start of surgery, two large peripheral lines, suitable size urinary catheter, and central line 7.5 G triple lumen at the right subclavian vein were secured and Ringer acetate infusion at rate 5 ml.kg−1.h−1 was started.

ONSD was assessed by a skilled physician with ultrasound guidance (Mindray Z60, superficial probe, frequency 3.2–11.2 MHz). Patients’ heads were kept in the neutral position, closing eye with a sterilized eye cover, and the probe was applied perpendicularly over the eye globe in a transverse direction with the right orientation to see the clearest view of the eye globe and optic nerve behind. Using two-dimensional mode, ONSD measurement, 3 mm posterior to the posterior scleral margin at the area of maximal dimension, was recorded. The measurement was taken also from the other eye for confirmation and the average of both the readings reading was recorded.

At the start of skin incision for craniotomy, patients were categorized into two groups according to randomization: Group M received mannitol 1 g.kg−1 over 15 min, while Group F received mannitol 0.25 g.kg−1 over 15 min and furosemide 0.5 mg.kg−1 in 100 ml saline over 30 min. To ensure blinding, all infused drugs were covered and coded syringes were prepared by an independent anesthesiologist and in the M Group, the second syringe was filled with saline at the same volume and pump settings as per the study protocol. After the ending of diuretics infusion, and before the surgical opening of the dura, the second measurement of ONSD was taken with a cutoff value of 5.5 mm [Figure 2]. If ONSD >5.5 mm, patients received additional mannitol dose 0.25 g.kg−1 and reported as the second dose needed. Immediately, after the opening of the dura, the surgeon was asked to assess the BRS and reported as 1 = very good, 2 = good, 3 = bad, and 4 = very bad. Thirty minutes later, second BRS and third ONSD measurements were taken and reported.

Figure 2

Optic nerve sheath diameter before and after diuresis

Patients’ hemodynamic data including heart rate (HR), mean arterial blood pressure (MABP), oxygen saturation, and EtCO2 were recorded basally, postinduction, at the end of diuretics infusion, and 30 min later. Serum electrolytes (serum sodium and potassium) and serum lactate levels were measured 30 min after the end of diuretic infusion. At end of the surgery, urine output and the total volume of infused fluids were recorded.

In this trial, we hypothesized that adding loop diuretics to the lowest dose of mannitol will make a similar effect to the ordinary dose of mannitol in reducing ICP and brain relaxation based on ONSD, hence reducing the dose, side effects, and better hemodynamic tolerating during the procedure.

Sample size and statistical analysis

A pilot study was conducted in 10 patients scheduled for brain tumor surgery where the reading of the ONSD was 5.5 ± 0.5 mm after induction of anesthesia. The authors hypothesized that the combination of low-dose mannitol and diuretic can return the ONSD to the normal value (≤5 mm) and used G * power software version 3.1.9.4 (Franz Faul, Kiel University, Germany) to calculate the required sample size to assure a study power of 90% with an α error of 0.05. The total sample size for the two study groups was 52 patients. Sixty patients were recruited to compensate for dropouts. Data collected during the study were tabulated and analyzed using SPSS software, IBM, USA, version 22. Data were tested for normality of distribution and significant differences between the study groups were assessed using independent sample t-test, Mann–Whitney test, Chi-square test, or Kruskal–Wallis test as appropriate.

RESULTS

In this study, 61 patients were assessed for eligibility criteria; one patient was excluded due to severe hemodynamic instability after anesthesia induction [Figure 1]. Patients’ baseline data and perioperative characteristics are presented in Table 1. No statistical differences were found between the two groups regarding age, gender, or BMI, while the total volume of fluids’ replacement and UOP was statistically higher in M Group than in the F Group (3.1 ± 0.4 L for M Group, 2.4 ± 0.3 L for F group, P = 0.00 and 4.9 ± 0.9 L for M Group, F 4.2 ± 1.1 L for F Group, P = 0.01, respectively).

Table 1

Perioperative characteristics in the two studied groups

	Group M (n=30)	Group F (n=30)	P
Age (years)	41±16	39±16	0.74
Gender (male/female)	13/17	12/18	0.50
BMI (kg/m2)	25.5±2.7	26±3.6	0.54
Intraoperative UOP (L)	3.1±0.4	2.4±0.3	0.001*
Intraoperative infused fluids (L)	4.9±0.9	4.2±1.1	0.01*
Second dose of mannitol, n (%)	6 (20)	4 (14)	0.73

Data are presented as mean±SD and absolute n (%). *P value significant if<0.05. BMI=Body mass index, UOP=Urine output, SD=Standard deviation

Hemodynamic data of the included patients are demonstrated in Figure 3 with a statistically significant difference in HR reading immediately after diuresis (72.03 ± 5.35 for M Group, F 76.90 ± 6.13 for F Group, P 0.02) and MABP immediately after diuresis (79.23 ± 6.10 for M Group, 85.76 ± 6.10 for F Group, P 0.001).

Figure 3

Hemodynamic data of the included patients

Table 2 shows that the BRS and ONSD of the included patients are with no statistically significant difference among the studied group.

Table 2

Brain-relaxation score and optic nerve sheath diameter in the two groups

	Group M (n=30)	Group F (n=30)	P
Basal ONSD (mm)	6.8±0.32	6.7±0.38	0.30
ONSD immediately after diuresis (mm)	5.4±0.12	5.3±0.12	0.15
ONSD 30 min after diuresis (mm)	5.3±0.13	5.2±0.10	0.06
BRS1 very good/good	26/4	25/5	0.72
BRS2 very good/good	30/0	30/0	1

Data are presented as mean±SD or absolute numbers. ONSD=Optic nerve sheath diameter, BRS=Brain-relaxation score, SD=Standard deviation

The biochemical profile of the included patients, before and after diuresis, is displayed in Table 3. A statistically significant change in the serum electrolytes and serum lactate level, between M and F groups, is noticed (131 ± 2, 137 ± 3, P 0.001 for serum sodium, respectively), (3.5 ± 0.32, 3.9 ± 0.3, P = 0.001 for serum potassium, respectively), and (3.15 ± 0.43, 1.9 ± 0.29, P = 0.001 serum lactate, respectively).

Table 3

Biochemical profile in the two study groups before and after diuresis

	Before diuresis			After diuresis
	Group M	Group F	P	Group M	Group F	P
Serum sodium (mmol/L)	141±3	140±4	0.88	131±2	137±3	0.001*
Serum potassium (mmol/L)	4.4±0.37	4.3±0.41	0.39	3.5±0.32	3.9±0.3	0.001*
Serum lactate (mmol/L)	1.19±0.28	1.15±0.21	0.54	3.15±0.43	1.9±0.29	0.001*

*P significant if<0.05

DISCUSSION

In this clinical trial, we compared the efficacy of adding furosemide to the lowest mannitol dose as a brain-relaxation method in contrast to the ordinary mannitol dose. Sixty patients with supratentorial brain surgery were enrolled in the study, and ONSD measurement was adapted as a primary outcome objective. We found that adding furosemide to the lowest mannitol dose as a brain-relaxation method has the same effect as an ordinary dose of mannitol guided by the reduction in ONSD and BRS. Moreover, the used combination has better hemodynamic stability, less urine output, and lower replacement fluids volume. In addition, electrolyte disturbance and tissue perfusion parameters were significantly better in the combination group.

Mannitol, as a brain dehydrating drug, is commonly used during craniotomy. It acts by inducing osmotic diuresis and hence reducing brain swelling and ICP, allowing better surgical conditions.[12] Currently, there is no universal consensus or guidelines for describing an effective and safe dose of mannitol. Many studies demonstrated a better brain relaxation scores when a higher mannitol doses were used, however, this was associated with a significant extracellular fluid shifts with subsequent volume overload, kidney damage and excessive diuresis. This will affect hemodynamic HR, MABP, and systemic perfusion indicated by serum lactate and excessive urine output and electrolyte disturbances. With fluid transfusion during operation and water movement across three fluid compartments, dilutional hyponatremia and fluid balance changes occur. These rapid and consecutive changes may be harmful, especially in patients with cardiac or renal dysfunctions. Changes in electrolytes, especially potassium levels, can affect cardiac rhythm and contractility. In addition, rapid fluctuations in sodium levels have harmful central effects. Rapid intravascular volume decrease leads to the disturbed systemic circulation, affection of HR, hypotension, poor peripheral perfusion, anaerobic metabolism, acidosis, and increased lactate levels.[5812]

Seo et al.[3] did a comparative study among 124 patients with supratentorial brain surgery for four rising doses of mannitol (0.25, 0.50, 0.75, and 1.0 g.kg−1). The study found that brain relaxation determined by BRS was better in 1.0, 0.75 g.kg−1 groups versus 0.25 g.kg−1 group (67.7% and 64.5% vs. 32.2%, P = 0.011 and 0.022, respectively). Interestingly, the study related the higher mannitol doses to an increased osmotic gap and electrolyte disturbance, especially hyponatremia. Quentin et al.[8] enrolled 80 patients undergoing supratentorial craniotomy for tumor resection and found that a 1.4 g.kg−1 dose of mannitol resulted in greater brain relaxation than did a 0.7-g.kg−1 dose.

Similar to our hypothesis, Akcil et al.[13] used furosemide mannitol combinations in different doses compared with mannitol 0.5 g.kg−1 alone. They concluded that administration of furosemide with low or high doses of mannitol may cause a reduction in sodium and chloride levels as well as a rise in the lactate level. Moreover, it may cause high urine output and negative intraoperative fluid balance with adequate brain relaxation. These results do not match our findings regarding electrolytes and fluid balance and perfusion indicators. This may be due to raised mannitol dose in combination. Akcil et al. used 0.5 and 1.0 g.kg−1 mannitol, while our study used 0.25 g.kg−1 mannitol. Mannitol–furosemide combination has a synergistic effect; doubling mannitol dose may have excessive and rapid diuresis effect resulting in more urine output, electrolyte disturbance, and negative fluid balance.

The intraoperative use of ONSD measurement is a sensitive indicator for raised ICP and its reduction after osmotherapy.[14] Furthermore, this technique is considered simple, noninvasive, reliable, and available to be done inside the operating room unlike another imaging process as CT and MRI. In a meta-analysis by Lee et al. including five studies, ocular ultrasound and ONSD measurement showed equal sensitivity (0.91 vs. 0.90; P = 0.35) and higher specificity (0.82 vs. 0.58; P = 0.01) compared to those using brain CT.[15]

For many trials, brain-relaxation assessment depends on the BRS score which is a subjective method. It may expose the patient to a useless higher mannitol dose with its drawbacks. Using a ventricular catheter carries a high risk of infection, bleeding, displacement, and accuracy affection.[16] In their studies, Dimitriou et al. and Poblete et al. accounted for the infection rate of a ventricular catheter as 9.2% and 7.3%, respectively,[1718] while the bleeding rate was 21.6%, 31.9%, and 44.3% according to Miller and Tummala Sussman et al., and Gardner et al., respectively.[192021]

Our study has some limitations; we did not monitor the patient for used regimen regarding ONSD, hemodynamics, electrolytes, and serum lactate more than 30 min post diuretics end. Longer time monitoring may reflect more information, benefits, or drawbacks, also the used regimen was effective; further studies may be needed to evaluate the effects of reducing the dose of furosemide while having a reasonable brain dehydrating effect.

CONCLUSION

Combined furosemide with the lowest dose of mannitol has the same effect in brain relaxation as an ordinary dose of mannitol monitored by ONSD measurement with better hemodynamic stability and fewer electrolyte disturbances. Further studies may be needed to evaluate the effect of reducing furosemide in this combination regimen.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.