A Comparison of Equivolume, Equiosmolar Solutions of Hypertonic Saline and Mannitol for Brain Relaxation in Patients Undergoing Elective Intracranial Tumor Surgery: A Randomized Clinical Trial Pavel Dostal, MD, PhD,* Vlasta Dostalova, MD, PhD,* Jitka Schreiberova, MD, PhD,* Tomas Tyll, MD, PhD,w Jirina Habalova, MD, PhD,z Vladimir Cerny, PhD, FCCM,* Svatopluk Rˇehak, MD, PhD,z and Tomas Cˇesak, MD, PhDz

Background: Hyperosmolar solutions have been used in neurosurgery to modify brain bulk and prevent neurological deterioration. The purpose of the study was to compare the effects of equivolume, equiosmolar solutions of mannitol and hypertonic saline (HTS) on brain relaxation and postoperative complications in patients undergoing elective intracranial tumor surgery. Methods: In this prospective, randomized study, patients with American Society of Anesthesiologists physical status I to III scheduled to undergo a craniotomy for intracranial tumors were enrolled. Patients received a 3.75 mL/kg intravenous infusion of either 3.2% HTS (group HTS, n = 36) or 20% mannitol (group M, n = 38). The surgeon assessed the condition of the brain using a 4-point scale after opening the dura. Recorded measures included duration of surgery, blood loss, urine output, volume and type of infused fluids, hemodynamic variables, electrolytes, glucose, creatinine, predefined postoperative complications, and length of intensive care unit and hospital stays. Results: Brain relaxation conditions in group HTS (score 1/2/3/ 4, n = 10/17/2/7) were better than those in group M (score 1/2/ 3/4, n = 3/18/3/14, P = 0.0281). Patients in group M had higher urine output, received more crystalloids during surgery, and displayed lower central venous pressure and lower natremia at the end of surgery than did patients in group HTS. No significant differences in postoperative complications or lengths of intensive care unit and hospital stays were observed between the groups.

Received for publication March 25, 2014; accepted June 12, 2014. From the Departments of *Anesthesia and Intensive Care; zNeurosurgery, Faculty of Medicine Hradec Kralove, Charles University in Prague, University Hospital Hradec Kralove, Hradec Kralove; and wDepartment of Anesthesia and Intensive Care, 1st Faculty of Medicine Prague, Charles University in Prague, Military University Hospital, Prague, Czech Republic. Supported by the program PRVOUK (Research Area Development Programs at Charles University) P37/02. The authors have no conflicts of interest to disclose. Reprints: Pavel Dostal, MD, PhD, Department of Anesthesia and Intensive Care, University Hospital Hradec Kralove, Sokolska 581, 500 05, Hradec Kralove, Czech Republic (e-mail: [email protected]). Copyright r 2014 by Lippincott Williams & Wilkins

J Neurosurg Anesthesiol

Volume 27, Number 1, January 2015

Conclusions: Our results suggest that HTS provides better brain relaxation than mannitol during elective intracranial tumor surgery. Key Words: mannitol, osmotherapy, hypertonic saline, brain tumor, neurosurgery, brain relaxation (J Neurosurg Anesthesiol 2015;27:51–56)


uring a craniotomy, tumor bulk and vasogenic edema can generate clinical situations in which adequate intracranial volume management becomes a key factor to facilitate surgical removal of the tumor.1 Hyperosmolar solutions have been used during neurosurgical procedures to improve operating conditions and prevent transdural herniation and pertinent neurological deterioration.1–3 The effectiveness of a hyperosmolar solute depends on its “reflection coefficient” (RC), which determines relative impermeability to the blood-brain barrier (BBB) of the solute, wherein 1 indicates an impermeable solute and 0 an ideally permeable solute. Because the RC of sodium is 1 and that of mannitol is 0.9, hypertonic saline (HTS) may have advantages over mannitol.4–6 The use of a hyperosmolar solution with a higher RC is theoretically more effective in reducing increased intracranial pressure and carries a lower risk of the so-called rebound phenomenon after lowering the blood concentration of a hyperosmolar solute.4,5,7 Mannitol is recommended as a first-choice hyperosmotic agent for treatment of increased intracranial pressure.8,9 However, a number of prospective clinical trials comparing the effects of mannitol and HTS on intracranial pressure have suggested that HTS is at least as effective if not superior to mannitol for treating intracranial hypertension.10–12 HTS and mannitol have recently been studied in patients without intracranial hypertension. Clinical studies comparing the use of mannitol and HTS in patients undergoing elective neurosurgical procedure used either different osmolar loads of mannitol and HTS,2,3 enrolled patients undergoing urgent neurosurgical procedures |


J Neurosurg Anesthesiol

Dostal et al

including patients with subarachnoid hemorrhage (SAH),13 or used nonequivolemic, nonequiosmolar hypertonic solutions in patients with supratentorial brain tumors.14 The aim of our study was to compare the effects of equivolemic, equiosmolar solutions of mannitol and HTS on intraoperative brain relaxation as the primary endpoint and early postoperative complications as secondary endpoints in patients undergoing craniotomy for elective intracranial tumor surgery.

MATERIALS AND METHODS Ethical approval for this study (Ethical Committee no. 200814 S14P) was granted by the Ethical Committee of the University Hospital Hradec Kralove, Hradec Kralove, Czech Republic (Chairperson Jiri Vorel, MD) on April 3, 2008. After providing written informed consent, adult patients scheduled for an intracranial brain tumor craniotomy were considered for inclusion in this prospective, randomized study. All subjects were recruited between April 5, 2008 and October 31, 2012. The inclusion criteria were: age 18 to 70 years, elective intracranial tumor surgery with indication for perioperative osmotherapy, American Society of Anesthesiologists physical status I to III, and preoperative natremia of 135 to 145 mmol/L. Exclusion criteria were: history or presence of congestive heart failure (New York Heart Association class III to IV), history or presence of renal failure, presence of preoperative disturbance of water or sodium metabolism (diabetes insipidus, cerebral salt wasting syndrome, or syndrome of inappropriate antidiuretic hormone secretion), preoperative Glasgow Coma Scale score r13, preoperative need for hemodynamic support, preoperative presence of obstructive hydrocephalus, treatment with cyclosporine within the last month, or a neurosurgical procedure within the last 3 months. After randomization (computer-generated random list of patients in sealed envelopes) patients were assigned to receive 3.75 mL/kg body weight of either 3.2% HTS (group HTS) or 20% mannitol (group M) solution administered intravenously over 30 minutes using an infusion pump at the time of skin incision. Both solutions had the same osmolarity (1099 mosm/L) and were infused by means of a central venous catheter. The 3.2% HTS solution was prepared by the hospital pharmacy. The volume infused was equivalent to a dose of 0.75 g/kg body weight mannitol. The surgeon was blinded to the type of solution used. All patients were administered midazolam preoperatively (patients less than 60 y of age 7.5 mg, older patients 3.75 mg) orally. Preoperative pharmacotherapy, including preoperative corticosteroids, was continued. General anesthesia was induced with propofol or thiopentone, along with an opiate and muscle relaxant as determined by the attending anesthesiologist. Tracheal intubation was facilitated with atracurium or cisatracurium. Anesthesia was maintained with isoflurane (0.5 to 1 minimum alveolar concentration) in oxygen and

52 |

Volume 27, Number 1, January 2015

air along with sufentanil boluses. Muscle relaxants were used as needed to maintain a single twitch on a train-offour stimulation. A radial artery catheter, central venous catheter, urinary bladder catheter, and esophageal temperature probe were inserted in all patients. Dexamethasone was administered intravenously to patients, at the surgeon’s discretion before the skin incision. Lumbar or ventricular drains were not inserted during the surgery. An esophageal temperature of 36 to 37.51C and end-tidal carbon dioxide tension (ETCO2) of 35 to 38 mm Hg were maintained. Intravenous fluids were managed according to our standard practice: baseline fluid intake of 2 mL/kg/h was provided with 0.9% NaCl. The decision to give fluid boluses, colloids, or blood transfusion was at the discretion of the attending anesthesiologist, although the protocol recommended administering a fluid bolus if hypotension due to hypovolemia was suspected. We targeted our interventions to maintain mean arterial pressure (MAP) in the range of 75 to 85 mm Hg. Hypotension was defined as MAP < 75 mm Hg lasting >5 minutes. Norepinephrine was indicated if hypotension persisted for >10 minutes, the dose of norepinephrine was adjusted to maintain MAP above 75 mm Hg. Brain relaxation was assessed by the attending neurosurgeon immediately after opening the dura on a scale of 1 to 4 (1 = perfectly relaxed, 2 = satisfactorily relaxed, 3 = firm brain, 4 = bulging brain),13,15 and this scale was used to evaluate dural tension at incision. A second dose of osmotic agent and short-term hyperventilation to reach an ETCO2 of 28 to 30 mm Hg was allowed at the neurosurgeon’s request if needed to facilitate surgical exposure or in the case of significant cerebral edema (brain relaxation score 3 to 4). At the end of surgery, we recorded the neurosurgeon’s assessment of the quality of osmotherapy on a subjective scale (3 = very good effect of osmotherapy, 2 = acceptable effect of osmotherapy, 1 = insufficient effect of osmotherapy). This scale was used to assess the effect on brain bulk throughout surgery. Recorded variables included age, sex, type, location, and size of brain tumor, presence and size of midline shift, head position during surgery, preoperative corticosteroid use, dose of HTS used during surgery, hemodynamic variables (MAP and heart rate), temperature, urine output, perioperative fluid balance, blood loss, and laboratory values (creatinine, electrolytes, and osmolarity), duration of surgery, duration of postoperative ventilation, length of intensive care unit (ICU) and hospital stay, number of surgical interventions, wound infections, and outcome. A difference of 1 point in brain relaxation score between the groups was considered clinically significant for the power analysis.13 A power analysis based on an a error of 0.05 and a b error of 0.2 was performed using G*Power 3.0.9 (Franz Faul, University Kiel, Germany). The sample size needed for the Wilcoxon-Mann-Whitney (2 groups) test (expected mean difference of 1.0, SD in both groups of 1.2) with the minimal asymptotic relative r

2014 Lippincott Williams & Wilkins

J Neurosurg Anesthesiol

Volume 27, Number 1, January 2015

efficiency setting was calculated. This calculation produced a sample size of 56 subjects (28 subjects in each treatment group). Sample size was increased to at least 35 patients per treatment group to compensate for potential dropouts and possible inaccuracy of predictions used for the power analysis. Data are presented as mean ± SD. Brain relaxation scores, subjective assessment scores, blood loss, and ICU and hospital days are presented as medians with interquartile ranges on the basis of the results of the Kolmogorov-Smirnov test. Differences between the groups were analyzed using a w2 test with Yates correction for continuity (demographic variables). The Mann-Whitney U test was used for brain relaxation and subjective assessment scales, and the Wilcoxon-Mann-Whitney odds (WMWodds) with 95% confidence intervals (CI) were calculated. The Mann-Whitney U test was also used when the sample distribution was not normal (blood loss, days of stay, creatinine, and selected glucose and osmolarity levels). An unpaired t test was used to compare other results between the groups. P < 0.05 was considered significant. The statistical analysis was performed using MedCalc 7.6.0. (MedCalc Software, Ostend, Belgium). WMWodds were calculated using Microsoft Excel 2007 (Microsoft, Redmond, WA).

Solutions of Hypertonic Saline and Mannitol

TABLE 1. Patients’ Characteristics Group HTS (n = 36) Age (mean ± SD) 52.1 ± 13.1 Sex (M/F) 16/20 Weight (kg) 80.1 ± 13.2 ASA physical status II/III (n) 30/6 Intracranial pathology (n [%]) Glioblastoma 13 (36.1) Meningioma 10 (27.7) Metastases of extracranial 5 (13.9) tumors Other brain tumor 8 (22.2) Localization of pathology (n [%]) Frontal 13 (36.1) Frontal/parietal and frontal/ 3 (8.3) temporal Parietal 3 (8.3) Temporal 9 (25.0) Infratentorial 3 (8.3) Base of skull 4 (11.1) Other site 1 (2.8) Tumor, largest diameter (mm) 47 ± 14 (mean ± SD) Midline shift (n [%]) 10 (27.8) Midline shift (mm) 7.7 ± 3.2 (mean ± SD) Preoperative corticosteroid 30 (83.3) use (n [%])

Group M (n = 38)


53.5 ± 13.0 14/24 78.4 ± 14.7 28/10

0.6366 0.7067 0.6097 0.4683

14 (36.8) 10 (26.3) 5 (13.2)

0.8579 0.6983 0.8011

9 (23.7)


15 (35.5) 6 (15.8)

0.8503 0.5280

6 (15.8) 6 (15.8) 0 (0) 3 (7.9) 2 (5.3) 49 ± 13

0.5280 0.4873 0.2216 0.3614 0.9645 0.5119

14 (36.8) 8.4 ± 3.5

0.5591 0.6446

32 (84.2)


HTS indicates hypertonic saline; M, mannitol.

RESULTS A total of 479 patients undergoing scheduled craniotomy for an intracranial brain tumor were considered. After excluding patients with contraindications, those not willing to participate, those involved in other studies, and those not recruited for logistic reasons, a total of 74 neurosurgical patients were recruited. Demographic data of all recruited patients are summarized in Table 1. No significant differences were observed between the 2 groups regarding age, sex, body weight, ASA physical status, type of brain tumor, brain tumor localization, or preoperative use of corticosteroids. All patients received dexamethasone intravenously before the skin incision (Table 1). The dose of HTS did not differ between the groups (Table 2). Brain relaxation scores and subjective assessment scores were significantly better in group HTS compared with those in group M (P = 0.0281; WMWodds, 1.84; 95% CI, 1.09-3.42 and P = 0.0154, WMWodds, 1.97; 95% CI, 1.17-3.70). Compared with HTS, mannitol caused higher urine output (1395 ± 825 vs. 656 ± 496 mL, P < 0.0001), and, although patients in the M group received more crystalloids during surgery (2971 ± 1390 vs. 2093 ± 1190, P = 0.0048) and had higher total fluid intake, including colloids and blood products (3391 ± 1823 vs. 2304 ± 1429, P = 0.0058), they had lower central venous pressure at the end of the surgery (8.1 ± 4.3 vs. 10.4 ± 4.8, P = 0.0398) (Table 2). We observed a nonsignificant trend toward longer length of surgery in the M group (295.9 ± 123.8 vs. 248.5 ± 79.1, P = 0.0553). No significant differences were observed in the volume of blood loss, infused colloids, fluid balance during surgery, r

2014 Lippincott Williams & Wilkins

number of hypotensive episodes, or use of norepinephrine (Table 2). Compared with mannitol, HTS caused a higher serum sodium concentration at the time of admission to the ICU. No significant differences were observed in ICU stay, number of predefined postoperative complications, or hospital stay between the 2 groups (Table 3). Three patients in each group needed postoperative ventilation. Two patients in group M were postoperatively ventilated on the basis of the neurosurgeon’s decision, and 1 patient in group M and 3 patients in group HTS needed reintubation in the ICU. Postoperative bleeding was present in 4 patients in group HTS, and 3 required reoperation. Postoperative bleeding was present in 3 patients in group M, and 1 required reoperation due to bleeding. Two patients in group HTS and 3 patients in group M developed wound infections. Other infectious complications in the groups were respiratory tract infections and 1 urinary tract infection in group HTS.

DISCUSSION We demonstrated that HTS provided more satisfactory brain relaxation than did mannitol; mannitol had a more prominent diuretic effect, and patients in the M group received more fluids during surgery and had lower central venous pressure at the end of surgery. ICU postoperative course and ICU and hospital stays were |


J Neurosurg Anesthesiol

Dostal et al

Volume 27, Number 1, January 2015

TABLE 2. Comparison of the Osmotherapy Dose, Length of Surgery, Brain Relaxation Score, Subjective Assessment of the Effect of Osmotherapy, Fluid Intake, Fluid Balance, Urine Output, Episodes of Hypotension, Use of Norepinephrine, and Central Venous Pressure at the Beginning and End of Surgery Group HTS (n = 36)

Group M (n = 38)

300.1 ± 49.4 302.9 ± 52.2 1 (2.8) 248.5 ± 79.1 10/17/2/7 2 (1.0, 2.5) 23/10/3 3 (2, 3) 656 ± 496 425 (225, 600) 8 (22.2) 2 (5.6) 1 (2.8) 2093 ± 1190 0 (0, 0) 7 (19.4) 2304 ± 1429 1207 ± 1167

294.0 ± 55.2 322.8 ± 99.4 5 (13.2) 295.9 ± 123.8 3/18/3/14 2 (1.0, 4.0) 13/16/9 2 (2, 3) 1395 ± 825 300 (250, 800) 16 (42.1) 3 (7.9) 2 (5.3) 2971 ± 1390 0 (0, 500) 11 (28.9) 3391 ± 1823 1565 ± 1497

8.1 ± 4.7 10.4 ± 4.8

7.5 ± 4.5 8.1 ± 4.3

First dose (mL) Total dose (mL) No. patients with second dose of osmotic agent (n [%]) Length of surgery (min) Brain relaxation score (1/2/3/4) Median grade (quartile range) Subjective assessment of the effect of intervention (3/2/1) Median grade (quartile range) Urine output (mL) Blood loss (mL) Episode of hypotension (n [%]) Hypotension lasting >5 min (n [%]) Use of norepinephrine (n [%]) Crystalloids (mL) Colloids (mL) No. patients with colloids (n [%]) Total fluid intake (mL) Fluid balance (mL) Central venous pressure (mm Hg) Beginning of surgery End of surgery

P 0.6182 0.2890 0.2263 0.0553 0.0281 0.0154 < 0.0001 0.6458 0.1142 0.9448 0.9645 0.0048 0.4330 0.4956 0.0058 0.2573 0.5823 0.0398

Four-point scale score of brain relaxation: 1—perfectly relaxed, 2—satisfactorily relaxed, 3—firm (leveled) brain, 4—bulging brain; 3-point scale subjective assessment score of neurosurgeon satisfaction with the effect of osmotherapy: 3—very good, 2—satisfactory, 1—unsatisfactory; episode of hypotension—mean arterial pressure

A comparison of equivolume, equiosmolar solutions of hypertonic saline and mannitol for brain relaxation in patients undergoing elective intracranial tumor surgery: a randomized clinical trial.

Hyperosmolar solutions have been used in neurosurgery to modify brain bulk and prevent neurological deterioration. The purpose of the study was to com...
124KB Sizes 0 Downloads 3 Views