Secondary Logo

Journal Logo

Ambulatory Anesthesiology: Research Report

A Prospective, Randomized Comparison of the Effects of Inhaled Sevoflurane Anesthesia and Propofol/Remifentanil Intravenous Anesthesia on Salivary Excretion During Laryngeal Microsurgery

Kang, Jin Gu, MD*; Kim, Jin Kyoung, MD, PhD*; Jeong, Han-Sin, MD, PhD; Jung, Soo-Chan, MD; Ko, Moon Hee, MD; Park, Shin Hong, MD; Cho, Jae Keun, MD; Lee, Gil Joon, MD; Choi, Ji Won, MD*; Lee, Byung Dal, MD, PhD*

Section Editor(s): Glass, Peter S. A.

Author Information
doi: 10.1213/ane.0b013e3181730063
  • Free

Anesthesia for laryngeal microsurgery requires rapid control of anesthetic levels during laryngeal manipulation.1 The operation times for laryngeal microsurgery are relatively short, but the extensive manipulation of the laryngeal airway often produces hypertension, tachycardia, and arrhythmia, which are serious adverse cardiovascular effects.2 Therefore, to achieve immediate control of anesthetic levels during laryngeal microsurgery, anesthetics with rapid onset and short duration are desirable.1 In this respect, sevoflurane and propofol are widely used for the induction and maintenance of anesthesia during laryngeal microsurgery.

In addition, another important goal of anesthesia in laryngeal microsurgery is to provide a clear surgical field. For this purpose, an antisialagogue like glycopyrrolate may be helpful to reduce salivary excretion.2,3 However, anticholinergics have side effects,3 including tachycardia, an increased risk of pulmonary aspiration by relaxing the lower esophageal sphincter,4 increased respiratory dead-space due to relaxation of bronchial smooth muscles, increased airway resistance due to thickened secretions, and increased body temperature due to interference with sweating.5 For these reasons, the routine premedication of anticholinergics is generally not recommended and an anesthetic that produces less saliva is usually requested, especially for laryngeal microscopic surgery.

Early studies of propofol reported increased salivation.6,7 However, other reports have concluded that propofol reduces or causes no significant changes in salivary excretion.8–10 A rigorous comparison of the effects of sevoflurane and propofol on salivation has not yet been performed. Therefore, in this study, we prospectively compared the effects of sevoflurane and propofol/remifentanil on salivary excretion in patients who underwent laryngeal surgery.


This prospective study (SMC 2006-09-038-001) was approved by the IRB at Samsung Medical Center (a fully Association for the Accreditation of Human Research Protection Programs-accredited center). All patients provided written informed consent before being enrolled. A preliminary investigation of salivary excretion under sevoflurane and propofol/remifentanil anesthesia revealed that the difference between the two was around 0.1 mL/min, with a mean 0.21 mL/min for sevoflurane and 0.3 mL/min for propofol/remifentanil. Sample size analysis indicated that more than 19 patients were needed in each group for an 80% chance of detecting a difference of 0.09 mL/min (0.3–0.21) in the salivary flow rates of the anesthetics at a 0.05 significance level.


From October 2006 to April 2007, 40 patients scheduled to undergo laryngeal microsurgery by one ear–nose–throat surgeon (H.S.J.) under general anesthesia were enrolled in this study (Table 1). Exclusion criteria were age <20 yr or older than 80 yr; an ASA physical status of more than II; possible pregnancy; renal or liver disease; any other systemic disease, including diabetes; or use of any medication for 1 mo before surgery.

Table 1
Table 1:
Subject Characteristics

All patients fasted overnight before surgery. To exclude the potential influence of diurnal variations of salivation, the subject cases were performed uniformly at the time around midday. Subjects did not take any oral medication before surgery.

Salivary function was evaluated before anesthesia using a modified standard table for judging the severity of xerostomia and pharyngoxerosis.11 This table consisted of two categories, including subjective and objective symptoms. Items for subjective symptoms were (1) a parched feeling of the mouth, (2) stickiness in the mouth, (3) strong yearning to drink (thirst), (4) pain in the oral cavity, (5) taste abnormality, (6) speech impairment, (7) difficult mastication, and (8) difficult swallowing. Objective symptoms were determined using the following nine items; (1) dryness of mucosa in the oral cavity, (2) redness of mucosa in the oral cavity, (3) ulcer of the oral cavity, (4) coating of the tongue surface, (5) smoothing of the tongue surface, (6) wrinkles and creases on the tongue surface, (7) cracks on the tongue surface, (8) redness of mucosa in the oropharynx, and (9) dryness of mucosa in the oropharynx. Each item was scored semiquantitatively from score 0 (no complaints) to score 3 (severe discomfort or findings). Total scores (from 0 to 51) were calculated using the standard table. Scores more than 25 were considered to indicate the presence of salivary dysfunction, and patients with scores >25 were excluded to avoid bias.

Patients were randomized to the sevoflurane or propofol/remifentanil groups by using a random permuted block method.


In the sevoflurane group, anesthesia was induced with propofol (Pofol inj., JEIL Pharm. CO, Ltd., Korea) 1.5 mg/kg and rocuronium bromide (Esmeron®, Organon Korea, Korea) 1 mg/kg. After mask ventilation with sevoflurane 8% in oxygen for 3 min, the trachea was intubated. Anesthesia was maintained with sevoflurane and oxygen at a 1:1 ratio. No narcotic drugs were used in this group. Sevoflurane was titrated to maintain the mean arterial blood pressure at 80–100 mm Hg.

In the propofol/remifentanil group, anesthesia was induced and maintained with propofol (Fresofol 2% Inj. Fresenius Kabi, Germany) and remifentanil hydrochloride (Ultiva™, Pharmacia and Upjohn NV/SA, Belgium) via a target-controlled infusion pump (Orchestra® pump, Fresenius Kabi Korea, Korea); patients’ lungs were ventilated using an air/oxygen mix (1:1). Rocuronium (1 mg/kg) was injected IV to facilitate endotracheal intubation. The target effect-site concentrations for propofol was 4–6 μg/mL and for remifentanil 3–5 ng/mL during induction; 3–5 μg/mL and 2–4 ng/mL for maintenance, respectively. Propofol and remifentanil target effect-site concentrations were titrated to maintain a mean arterial blood pressure of 80–100 mm Hg. In both groups 2% lidocaine 40 mg was given IV to reduce propofol’s injection pain before injection or infusion.

Patients had standard ASA monitoring applied, including electrocardiography, noninvasive arterial blood pressure, pulse oximetry, and end-tidal CO2. Arterial blood pressures were recorded every 3 min.

The settings for infusion target-controlled infusion pumps and vaporizers were identical in all subjects, although only one of the anesthetic techniques was actually used as determined by the earlier randomization. The IV lines from the pump to the patient were also hidden from the surgeon so that he was blind to which drug the patient received. The surgeon (H.S.J.) was blinded to the randomization, and at all times patients were also unaware of the anesthetic technique used.


Before induction, with cooperation of the subject, the oral secretions in the mouth were collected by suctioning into a collection bottle for 5 min. We also measured the amylase concentration of the collected secretion. The results showed the amylase concentration was more than 500,000 U/L, suggesting the collected secretions arose mainly from the salivary glands. Preanesthetic baseline salivary flow rates were determined by dividing volumes by time (mL/min). An analysis of results from the first 10 subjects showed that preanesthetic baseline salivary flow rates were below 0.05 mL/min. After that, we did not measure the baseline salivary flow rates in the subsequent subjects. On the basis of this finding, we considered that the study anesthetics were the sole contributors to salivary flow rate after induction.

Total oral and pharyngeal secretions after induction were determined by collecting them with frequent suction until the start of awakening from anesthesia. Salivary flow rates during anesthesia were defined as total amounts collected divided by total time (mL/min). Collected saliva was immediately sent to our central laboratory for determinations of osmolality, and total protein, amylase, and chloride concentrations. Salivary total protein and chloride secretion rates were calculated by multiplying the salivary flow rate by saliva total protein and chloride concentrations, respectively.

In addition, the surgeon recorded the number of suctioning episodes required to clear the surgical field of secretions before beginning the actual operative procedure and any contribution of blood loss to the suctioned secretions. The two groups were then compared in terms of numbers of suctions. To determine the amounts of residual accumulated secretions in the oral cavity and pharynx at the end of surgery, secretions were collected separately and volumes and associated times were recorded (through d = 2.0 mm Laryngeal suction tip 8291.22, Richard Wolf GmbH, Germany).

Data in the tables and figures are presented as mean values and standard deviations. The two groups were compared using a nonparametric comparison method (nonparametric Mann–Whitney test) because data were not normally distributed. Statistical significance was accepted for two-tailed P values of <0.05.


In the sevoflurane group, the end-tidal concentration of sevoflurane for anesthesia maintenance was mean 2.8% ± 0.92%. In the propofol/remifentanil group, the actual doses of propofol and remifentanil for maintenance were 3.4 ± 0.55 μg/mL and 3.2 ± 0.41 ng/mL, respectively.

The salivary flow rates were determined by dividing total amounts of secretions collected by the durations of anesthesia (in minutes). When compared with preanesthetic baseline levels (<0.05 mL/min), salivary flow rates increased substantially in both study arms after induction. This increase was significantly more profound in the propofol/remifentanil group (0.53 ± 0.39 mL/min) than in the sevoflurane group (0.28 ± 0.15 mL/min) (P < 0.001).

Collected secretions were submitted for composition analysis, and these showed high concentrations (>500,000 U/L) of amylase, which suggested that collected secretions were due mainly to salivary excretion (Table 2). The mean total protein concentration was slightly higher in the sevoflurane group, whereas the protein secretion rate was significantly higher in the propofol/remifentanil group, because salivary excretions were much greater in the propofol/remifentanil group. The mean chloride concentration was also higher in this group, although osmolality was similar in the two groups.

Table 2
Table 2:
Comparison of Saliva Compositions in the Sevoflurane and Propofol/Remifentanil Groups

To evaluate the impact of salivary excretions on surgery, we counted the number of suction episodes required to clear the surgical field before starting the surgical procedures. In the sevoflurane group, the mean number of suction episodes before the procedures was 2.1 ± 1.5, whereas in the propofol/remifentanil group this was 5.0 ± 2.3 (P < 0.001). Thus, significantly more frequent suction episodes were needed in the propofol/remifentanil group.

Mean residual volume of accumulated secretions after the main procedures was significantly higher in the propofol/remifentanil group (2.13 ± 0.56 mL) than in the sevoflurane group (0.45 ± 0.32 mL) (P < 0.001). To remove secretions from the oral cavity and oropharynx, more time was needed in the propofol/remifentanil group because secretions were more profuse and tenacious (propofol/remifentanil 69.8 ± 15.6 s vs sevoflurane 18.3 ± 12.4 s, P < 0.001) (Fig. 1).

Figure 1.
Figure 1.:
Amounts of residual accumulated secretions in the oral cavity and pharynx and the times to clear them by suction.


We evaluated the impact of anesthetics on salivary excretion with respect to securing an optimal surgical field during laryngeal surgery. We found salivary excretion is more rapid during propofol/remifentanil anesthesia than after a pure volatile anesthetic with sevoflurane. Several reports have compared sevoflurane inhaled anesthesia with propofol IV anesthesia during cystoscopy,12 minor gynecologic surgery,13 elective surgery,14 and in an outpatient setting.15,16 Primary outcome measures in these studies were anesthetic induction time, postoperative incidence of nausea and vomiting, recovery profiles, costs, and patient satisfaction.17

However, studies on the effects of sevoflurane and propofol/remifentanil on salivary excretion are limited and contradictory. One study concluded that propofol compared with methohexitol anesthesia does not affect mucus secretion or clearance from tracheal mucosa of healthy dogs,10 and another study found that propofol plus ketamine reduced salivary flow versus midazolam plus ketamine.9 In yet another study, a direct comparison between propofol and inhaled anesthesia (isoflurane) showed marked short-term hyposalivation in both groups.8 However, propofol anesthesia was also reported to induce increased salivation.6,7

In the propofol/remifentanil group, anesthesia was induced and maintained with propofol and remifentanil. From our data, both groups in this study showed similar arterial blood pressure values (slightly higher in the propofol/remifentanil group) during anesthesia, although direct measurement of catecholamine levels was not performed. Therefore, the direct effect of remifentanil via vagal or parasympathetic stimulation on salivation was thought to be minimal in the propofol/remifentanil group. In the sevoflurane group, we also used propofol for induction. Although we cannot precisely determine or evaluate the effect of propofol on the sevoflurane group from this study, the effect of propofol during induction was thought to be only temporary, considering the short onset and recovery time. Therefore, we think that the effect on salivation of propofol in the sevoflurane group was negligible.

In laryngeal surgery, using a microscope, reducing the amount of secretions is critical for obtaining a satisfactory view, as is the case for oral and dental surgery. Our results showed a definite increase of salivary flow in the propofol/remifentanil group, which unfavorably affects the fine microscopic manipulations during laryngeal surgery. The effects of anesthetics on salivation may last throughout anesthesia. In particular, before the main procedure, more than five suction episodes were needed in the propofol/remifentanil group. We were unable to measure the number of suction episodes during surgery because the bleeding during surgery hindered counting the exact suction episodes for only salivary secretion.

Composition analysis results also supported that the salivary excretion was greater in the propofol/remifentanil group. The mean total protein concentration was slightly higher in the sevoflurane group, whereas the protein secretion rate was significantly higher in the propofol/remifentanil group because salivary excretions were much greater in this group. Chloride in saliva also reflects salivary gland production.18 Therefore, the higher concentration and secretion rate of chloride in the saliva of the propofol/remifentanil group implies that salivary gland production was more stimulated in that group.

This study has a limitation, which should be considered. It is possible that objective functions of the salivary gland might have been different in the two study groups. However, the risk of such a discrepancy between the two groups is minimal. The preanesthetic baseline level of secretion rates in some patients (n = 10) was uniformly <0.05 mL/min; therefore, the anesthetics were thought to mainly induce salivary excretion after induction. Presurgical patients likely experienced some sort of anxiety and their preanesthetic baseline salivary flow rates should not therefore be generalized as actual resting salivary flow rates in a larger or general population. In this study, the baseline salivary flow was used only as controls for comparison.

We conclude that salivary excretion is greater under propofol/remifentanil anesthesia than under sevoflurane anesthesia, and that this may unfavorably affect the surgical field during laryngeal surgery.


1. Morgan GE, Mikhail MS, Murray MJ, Larson CP. Clinical anesthesiology. 3rd ed. New York: McGraw-Hill Company, 2002:773
2. Donlon JV, Doyle DJ, Feldman MA. Anesthesia for eye, ear, nose, and throat surgery. In: Miller RD, ed. Miller’s anesthesia. 6th ed. Philadelphia: Elsevier Churchill Livingstone, 2005;2542–43
3. Hata TM, Moyers JR. Preoperative evaluation and management. In: Barash PG, Cullen BF, Stoelting RK, eds. Clinical anesthesia. 6th ed. Philadelphia: Lippincott Williams & Wilkins, 2006;495–7
4. Pick N, McDonald A, Bennett N, Litsche M, Dietsche L, Legerwood R, Spurgas R, LaForce FM. Pulmonary aspiration in a long-term care setting: clinical and laboratory observations and an analysis of risk factors. J Am Geriatr Soc 1996;44:763–8
5. Manchikanti L, Roush JR. Effect of preanesthetic glycopyrrolate and cimetidine on gastric fluid pH and volume in outpatients. Anesth Analg 1984;63:40–6
6. De Grood PM, Coenen LG, van Egmond J, Booij LH, Crul JF. Propofol emulsion for induction and maintenance of anaesthesia. A combined technique of general and regional anaesthesia. Acta Anaesthesiol Scand 1987;31:219–23
7. Schaer H. [Propofol infusion for the maintenance of short-term anesthesia.] Anaesthesist 1988;37:187–92
8. Lahteenmaki M, Salo M, Tenovuo J. Mucosal host defence response to hysterectomy assessed by saliva analyses: a comparison of propofol and isoflurane anaesthesia. Anaesthesia 1998;53:1067–73
9. Morse Z, Sano K, Kanri T. Decreased intraoral secretions during sedation-analgesia with propofol-ketamine and midazolam-ketamine combinations. J Anesth 2001;15:197–200
10. Padda GS, Kishioka C, Rubin BK. Propofol and methohexital have no significant effect on mucus secretion or clearance in the anesthetized dog. Crit Care Med 2001;29:1045–8
11. Sugano S, Takeyama I, Ogino S, Kenmochi M, Kaneko T. Effectiveness of formula ophiopogoins in the treatment of xerostomia and pharyngoxerosis. Acta Otolaryngol Suppl 1996;522:124–9
12. Thwaites A, Edmends S, Smith I. Inhalation induction with sevoflurane: a double-blind comparison with propofol. Br J Anaesth 1997;78:356–61
13. Nelskyla K, Korttila K, Yli-Hankala A. Comparison of sevoflurane-nitrous oxide and propofol-alfentanil-nitrous oxide anaesthesia for minor gynaecological surgery. Br J Anaesth 1999;83:576–9
14. Visser K, Hassink EA, Bonsel GJ, Moen J, Kalkman CJ. Randomized controlled trial of total intravenous anesthesia with propofol versus inhalation anesthesia with isoflurane-nitrous oxide: postoperative nausea with vomiting and economic analysis. Anesthesiology 2001;95:616–26
15. Fredman B, Nathanson MH, Smith I, Wang J, Klein K, White PF. Sevoflurane for outpatient anesthesia: a comparison with propofol. Anesth Analg 1995;81:823–8
16. Tang J, Chen L, White PF, Watcha MF, Wender RH, Naruse R, Kariger R, Sloninsky A. Recovery profile, costs, and patient satisfaction with propofol and sevoflurane for fast-track office-based anesthesia. Anesthesiology 1999;91:253–61
17. Joo HS, Perks WJ. Sevoflurane versus propofol for anesthetic induction: a meta-analysis. Anesth Analg 2000;91:213–9
18. Walsh NP, Montague JC, Callow N, Rowlands AV. Saliva flow rate, total protein concentration and osmolality as potential markers of whole body hydration status during progressive acute dehydration in humans. Arch Oral Biol 2004;49:149–54
© 2008 International Anethesia Research Society