Secondary Logo

Journal Logo

Original Article

Effects of remifentanil and fentanyl on intraocular pressure during the maintenance and recovery of anaesthesia in patients undergoing non-ophthalmic surgery

Sator-Katzenschlager, S. M.*†; Oehmke, M. J.*; Deusch, E.*; Dolezal, S.; Heinze, G.; Wedrich, A.

Author Information
European Journal of Anaesthesiology: February 2004 - Volume 21 - Issue 2 - p 95-100


There is increasing awareness of the potential side-effects of drugs used in anaesthesia on intraocular pressure (IOP) [1-3]. An increase in IOP is undesirable in patients with glaucoma and may be detrimental in patients with penetrating open-eye injury. Accordingly, the perioperative management must prevent IOP increases. Regional anaesthesia for eye surgery, the standard anaesthesia technique for most ophthalmological patients, has been associated with ischaemic complications such as central retinal vascular occlusion, optic atrophy and ischaemic optic neuropathy [4]. Especially, patients with compromised ocular blood flow are at risk [5]. In these patients, general anaesthesia appears to be indicated. Most anaesthetics for general anaesthesia are suggested not to induce increasing effects on IOP [1,6]. Opioid analgesics are an integral part of the practice of general anaesthesia. They are used to reduce vegetative responses and blunt haemodynamic reactions, when surgical stimuli are applied. Unfortunately, opioids increase the risk of postoperative nausea and vomiting (PONV), and may thereby indirectly increase IOP. However, novel opioids such as remifentanil provide potent intraoperative analgesia, smooth recovery and a rapid return of reflexes with a minimal risk of restlessness [7,8]. According to its pharmacokinetic properties, remifentanil allows for a high-dose opioid-based anaesthetic technique, whereas with fentanyl only small doses are used to prevent postoperative respiratory depression. Thus, IOP responses throughout anaesthesia and surgery may be different during remifentanil and fentanyl as a result of the different anaesthetic technique. The present study was performed to evaluate the effect of remifentanil on IOP in comparison with the well-established opioid fentanyl during and after general anaesthesia in patients undergoing elective, non-ophthalmic surgery.


The Ethics Committee of the University of Vienna, Vienna General Hospital, approved the study. Thirty-two patients (16-60 yr, ASA I-II) scheduled for elective non-ophthalmic surgery (Table 1) were enrolled in this randomized, prospective, double-blind study after giving written, informed consent. Patients with pre-existing ophthalmic diseases, known allergies or adverse reactions to any of the anaesthetic agents used were excluded. Patients were not liable to receive any catecholamine during the study.

Table 1
Table 1:
Opioid treatment and type of surgery.

Before surgery, patients were allocated randomly to one of the two groups: Group R (remifentanil, n = 16) or Group F (fentanyl, n = 16). All patients were anaesthetized by the same anaesthesiologist. Premedication consisted of midazolam 7.5 mg (Dormicum®; Hoffmann-La Roche AG, Basel, Switzerland) orally 1 h before surgery. Anaesthesia was induced in Group R with remifentanil 1 μg kg−1 (Ultiva®; GlaxoWellcome Operations, Greenford, UK) injected over 30 s, propofol 2 mg kg−1 (Diprivan®; Zeneca Österreich, Vienna, Austria) and vecuronium 0.1 mg kg−1 (Norcuron®; Organon Teknika, Eppelheim, Germany). In Group F, anaesthesia was induced with fentanyl 2 μg kg−1 (Fentanyl-Janssen®; Janssen Pharmaceutica, Beerse, Belgium), propofol 2 mg kg−1 and vecuronium 0.1 mg kg−1. The trachea was intubated and lung ventilation was adjusted to maintain end-tidal carbon dioxide tension (ETCO2) at 4.7-5.4 kPa. All patients' lungs were ventilated with 50% air in oxygen; nitrous oxide was not administered. Ventilation was volume-controlled (Cicero®; Dräger, Lübeck, Germany). For the maintenance of anaesthesia, all patients received a continuous infusion of propofol 4-8 mg kg−1 h−1. Supplemental intermittent bolus doses of vecuronium 0.03 mg kg−1 were administered as clinically indicated by monitoring the patient's train-of-four ratio. A continuous infusion of remifentanil 0.25-0.5 μg kg−1 min−1 was started after endotracheal intubation in the remifentanil group. Patients in the fentanyl group received bolus doses of fentanyl 2-5 μg kg−1 as clinically indicated. The application of propofol, remifentanil infusion and fentanyl, respectively, was discontinued on completion of skin closure.

Intraocular pressure was determined in both eyes using a hand-held applanation tonometer (Perkins MK2®; Clement Clarke International Ltd, Harlow, UK) [9,10]. Patients were positioned supine for IOP measurements and in the lithotomy position during abdominal hysterectomy and prostatectomy, respectively. An ophthalmologist blinded to the anaesthetic technique performed all IOP measurements. IOP was measured before induction of anaesthesia (baseline), after induction (immediately before intubation), 1 min after intubation, 5 min after intubation, 2 min after skin incision, every 15 min during the maintenance, upon completion of skin closure and 2 min after extubation. A final measurement of IOP was performed in the recovery room, 30 min after the end of anaesthesia. Continuous monitoring consisted of electrocardiography (ECG), heart rate (HR), pulse oximetry (SPO2) and ETCO2 measurements. Noninvasive blood pressure measurement (mean arterial pressure, MAP) and HR were documented at the above-mentioned time points. Nasopharyngeal body temperature was maintained at 36-37°C with an upper or lower body forced-air cover (Bair-Hugger®; Augustine Medical, Eden Prairie, MN, USA). Additionally, intravenous fluids were warmed to 37°C. An anaesthesiologist in charge of the recovery room (and not participating in the study) was responsible for post-operative monitoring. Postoperative pain was treated intravenously by using a patient-controlled analgesia pump with bolus doses of pirinitramide 3 mg (Dipidolor®; Janssen Pharmaceutica) no more than every 10 min. PONV was treated intravenously with metoclopramide 0.15 mg kg−1 (Paspertin®; Solvay Pharmaceuticals GmbH, Hannover, Germany). If this was not effective within 10 min, a dose of ondansetron 8 mg (Zofran®; GlaxoWellcome Operations) was administered intravenously.


For sample size planning, a Type I error rate of 0.05 and a power of 0.80 were proposed. This resulted in a sample size of 16 patients per group to detect a group difference of 3 mmHg, if a standard deviation of 3 mmHg in IOP was assumed. Randomization of patients was achieved using a computer-generated table. Variables were reported as mean ± standard deviation, except where indicated. Baseline values of variables were compared by using unpaired t-tests. ANOVA for repeated measurements was used to assess differences of outcome values to baseline values within groups and to compare these differences between groups. For the between-group comparison, differences to baseline were first averaged across time and then compared between groups [9]. This test is denoted by the global test of group differences. Only if the global test was significant, tests at specific time points were performed. For adjusting P values for multiple comparisons, the Bonferroni-Holm procedure was applied. P < 0.05 was considered as statistically significant. SAS System 8.1® (SAS Institute, Inc., Cary, NC, USA) was used for statistical analysis.


All 32 patients included were evaluated. Patients in both groups were comparable with respect to age, body weight, height and gender (Table 2). No significant difference in the type and duration of anaesthesia could be observed. The dose of propofol was 5.7 ± 1.7 mg kg−1 h−1 in the remifentanil group and 6.1 ± 1.6 mg kg−1 h−1 in the fentanyl group. The total volumes of infusion were comparable (Table 3).

Table 2
Table 2:
Patient characteristics.
Table 3
Table 3:
Anaesthetic characteristics.

Figure 1 shows the course of IOP during the study. After induction of anaesthesia, a significant decrease in IOP in the remifentanil group from 13.6 ± 2.6 to 7.1 ± 3.1 mmHg (P < 0.0001) and in the fentanyl group from 13.7 ± 2.2 to 9.7 ± 3.4 mmHg (P < 0.0001) was observed. Intraocular pressure remained significantly lower than baseline values during the whole maintenance of anaesthesia in both groups (P < 0.0001). After extubation, IOP increased but remained under the baseline IOP values in both groups (remifentanil 11.1 ± 2.9 mmHg; fentanyl 11.4 ± 2.3 mmHg). Thirty minutes after the end of anaesthesia, IOP returned to baseline values in both groups (remifentanil 13.9 ± 2.8 mmHg, P = 0.28; fentanyl 13.6 ± 2.3 mmHg, P = 0.59). At baseline as well as throughout all measurements, there were no significant differences in IOP between both groups (P = 0.7327, global test) (Fig. 1). The different intraoperative positioning of the patients - supine for patients undergoing mastectomy or a lithotomy position during abdominal hysterectomy and prostatectomy - showed no statistically significant difference in IOP (P = 0.4618, ANOVA, global test) in a subgroup analysis.

Figure 1
Figure 1:
Intraocular pressure (IOP) in patients undergoing unstressed non-ophthalmic surgery after the application of remifentanil (♦,n = 16) and fentanyl (□, n = 16). IOP measurements were taken before the induction of anaesthesia as baseline, after induction, 1 min after intubation, 5 min after intubation, 2 min after skin incision, during the maintenance of anaesthesia, upon completion of skin closure, after extubation and in the recovery room 30 min after the end of the anaesthesia. Data are the mean and standard deviation. Level of significance in comparison with IOP baseline values: *P < 0.05.

In both groups, ETCO2 and SPO2 remained stable throughout the study. After induction of anaesthesia, MAP decreased significantly (P < 0.0001), remaining significantly lower than the baseline values during the whole period of the maintenance of anaesthesia (P < 0.0001). After extubation, MAP increased but remained below the baseline values (P = 0.0124) and returned to baseline values 30 min after the end of the anaesthesia in both groups (P = 0.77). No significant differences in MAP could be observed at any time between both groups (P = 0.1295). Heart rates were significantly lower than baseline throughout maintenance (P < 0.0001). No significant differences in HR between the two groups were recorded at any time (P = 0.8601) (Table 4).

Table 4
Table 4:
Haemodynamic values (HR and MAP) before the induction of anaesthesia (baseline), after anaesthesia induction (before intubation), 1 min after intubation, 5 min after intubation, 2 min after skin incision, every 15 min during the maintenance (averaged), upon completion of skin closure, 2 min after extubation and 30 min after the end of anaesthesia in the remifentanil and fentanyl group.

In the postoperative period, four patients in the remifentanil group and three in the fentanyl group complaining of PONV were treated with metoclopramide, and one patient in Group R was given additionally ondansetron intravenously. A comparable number of patients in each group were treated postoperatively for pain with pirinitramide: in the remifentanil group, five patients received 3 mg, eight patients received 6 mg and two patients received 9 mg; in the fentanyl group, five patients received 3 mg, seven patients received 6 mg and one patient received 9 mg.


It has been shown that remifentanil and fentanyl have comparable decreasing effects on IOP in patients undergoing anaesthesia for non-ophthalmic surgery. No significant difference in IOP was observed between the groups. Comparable incidences of PONV and pain treatment were found in both groups.

Until now, no studies have compared the effect of remifentanil and fentanyl on IOP in humans during the maintenance and recovery of anaesthesia. Remifentanil is a new ultrashort-acting opioid structurally related to fentanyl that has been approved for clinical use during the induction and maintenance of anaesthesia. Remifentanil is rapidly and extensively metabolized by non-specific esterases in blood and tissues. Thus, the context-sensitive half-time of remifentanil is short, regardless of the duration of infusion or the age of the patient [7,8]. Accordingly, it is a suitable analgesic drug for both short- and long-lasting surgical interventions and - most importantly - its metabolism is independent of liver and renal function. Therefore, especially elderly patients can benefit from its use.

The study was performed in patients without any eye disease to exclude an influence on IOP by concurrent eye pathology. Anaesthesia was identical except for the studied opioids. Remifentanil and fentanyl were used in clinical usual dosages. Nitrous oxide was not used in order to study only the effects of the pure analgesics. IOP was measured with a hand-held Perkins applanation tonometer, which is an easy-to-use tool enabling the investigator, regardless of the patient's position, to measure precisely the IOP according to the Goldmann applanation method [10,11].

A stable or reduced IOP is necessary in patients with ocular hypertension, glaucoma and open eye injuries, and in patients with retinal eye diseases undergoing cataract surgery. Particular attention must be paid to the induction and emergence of anaesthesia since both pharmacological stressors, such as succinylcholine, and mechanical stressors, have an amplifying effect on IOP. A predictably rapid emergence from anaesthesia and sustained alertness, which are highly desirable characteristics for any anaesthetic technique, could be dangerous in patients with eye diseases, because any acute rise in intra-abdominal or intrathoracic pressure, not infrequently observed during extubation, might increase IOP. During anaesthesia, IOP is influenced by factors such as blood pressure, ETCO2, PEEP ventilation, PaO2, catecholamine concentrations as well as by drugs. Most anaesthetic and hypnotic agents, including volatile anaesthetics, barbiturates, opioids, neuroleptics and benzodiazepines, decrease IOP in proportion to the depth of anaesthesia [1-3]. A decrease in IOP after induction with propofol is well known [12,13], and remifentanil has already been shown to prevent an increase in IOP after succinylcholine and tracheal intubation [14,15].

A potential shortcoming of the present study might be the question of equivalent levels of anaesthesia in both groups. The remifentanil group had a slightly lower propofol consumption during anaesthesia (5.7 ± 1.7 mg kg−1 h−1) when compared with the fentanyl group (6.1 ± 1.6 mg kg−1 h−1), which is in accordance with clinical experience. The clinical signs for sufficient anaesthesia and analgesia, such as blood pressure, HR and ETCO2, revealed no differences between the two anaesthetic groups; however, they are weak indicators of the depth of anaesthesia. The aim was to compare the IOP during two common clinical intravenous anaesthetic techniques using either remifentanil or fentanyl.

Until now, the mechanisms of IOP reduction by remifentanil are unknown. Muscle rigidity by remifentanil is an unfavourable drug effect. This effect was not observed despite a bolus remifentanil being given over 30 s. The mechanisms of IOP reductions may be due to the attenuated muscle rigidity induced by coadministration of a hypnotic agent or neuromuscular blocking drug that may have a relaxing effect on the extraocular muscles. Other mechanisms of IOP reduction may be caused by vitreous volume reduction brought about by increased trabecular outflow or lowered aqueous humour production [16].

Anaesthesia for eye surgery may be complicated by intraoperative dysrhythmias and postoperative vomiting. These may be mediated via a variety of vagal reflexes [2]. In the present study, incidences of PONV were comparable in the remifentanil and fentanyl groups. Minkowitz described similar results [17]. However, Bekker and colleagues found a trend toward a more frequent incidence of PONV after remifentanil infusion [18]. The low incidence of PONV in the present study may reflect an intrinsic antiemetic property of propofol [19].

In conclusion, the results show that general anaesthesia with remifentanil as an analgesic decreases IOP in patients undergoing non-ophthalmic surgery similarly to fentanyl. The incidence of PONV as well as the amount of pirinitramide necessary for pain treatment were comparable in both groups. Therefore, when general anaesthesia is indicated, remifentanil-propofol anaesthesia might be suitable in patients with glaucoma undergoing cataract surgery and in patients with open globe injuries. Further studies should examine the different anaesthetic regimens in these clinical settings.


1. Jantzen JP. Anaesthesia and intraocular pressure. Anaesthesist 1988; 37: 458-469.
2. Artru AA. Trabecular outflow facility and formation rate of aqueous humor during propofol, nitrous oxide, and halothane anesthesia in rabbits. Anesth Analg 1993; 77: 564-569.
3. Donlon JV Jr. Anesthesia for eye, ear, nose and throat surgery. In: Miller RD, ed. Anesthesia, 5th edn. New York, USA: Churchill Livingstone, 2000: 2173-2198.
4. Coupland SG, Deschenes MC, Hamilton RC. Impairment of ocular blood flow during regional orbital anesthesia. Can J Ophthalmol 2001; 36: 140-144.
5. Findl O, Dallinger S, Menapace R, et al. Effects of peribulbar anesthesia on ocular blood flow in patients undergoing cataract surgery. Am J Ophthalmol 1999; 127: 645-649.
6. Sator-Katzenschlager SM, Oehmke MJ, Kontaratos M, Wedrich A, Heinze G, Weinstabl C. Effect of different doses of cisatracurium on intraocular pressure in sedated patients. Eur J Anaesthesiol 2002; 19: 823-828.
7. Lauwers M, Camu F, Vanlersberghe C. Remifentanil, an esterase metabolised opioid: What advantages does it offer in analgesia and anaesthesia? CNS Drugs 1997; 8: 189-198.
8. Glass PS, Gan TJ, Howell S. A review of the pharmacokinetics and pharmacodynamics of remifentanil. Anesth Analg 1999; 89 (Suppl): 7S-14.
9. SAS Institute, Inc. SAS/STAT User's Guide, Version 8. Cary, NC, USA: SAS Institute, Inc., 1999: 1562.
10. Perkins ES. Hand-held applanation tonometer. Br J Ophthalmol 1965; 49: 591-593.
11. Dunn JS, Brubaker RF. Perkins applanation tonometer. Clinical and laboratory evaluation. Arch Ophthalmol 1973; 89: 149-151.
12. Sator S, Wildling E, Schabernig C, Akramian J, Zulus E, Winkler M. Desflurane maintains intraocular pressure at an equivalent level to isoflurane and propofol during unstressed non-ophthalmic surgery. Br J Anaesth 1998; 80: 243-244.
13. Schäfer R, Klett J, Auffarth G, et al. Intraocular pressure more reduced during anesthesia with propofol than with sevoflurane: both combined with remifentanil. Acta Anaesthesiol Scand 2002; 46: 703-706.
14. Alexander R, Hill R, Lipham WJ, Weatherwax KJ, el-Moalem HE. Remifentanil prevents an increase in intraocular pressure after succinylcholine and tracheal intubation. Br J Anaesth 1998; 81: 606-607.
15. Ng HP, Chen FG, Yeong SM, Wong E, Chew P. Effect of remifentanil compared with fentanyl on intraocular pressure after succinylcholine and tracheal intubation. Br J Anaesth 2000; 85: 785-787.
16. Cunningham AJ, Barry P. Intraocular pressure - physiology and implications for anaesthetic management. Can Anaesth Soc J 1986; 33: 195-208.
17. Minkowitz HS. Postoperative pain management in patients undergoing major surgery after remifentanil vs. fentanyl anesthesia. Multicentre Investigator Group. Can J Anesth 2000; 47: 522-528.
18. Bekker AY, Berklayd P, Osborn I, Bloom M, Yarmush J, Turndorf H. The recovery of cognitive function after remifentanil-nitrous oxide anesthesia is faster than after an isoflurane-nitrous oxide-fentanyl combination in elderly patients. Anesth Analg 2000; 91: 117-122.
19. Tramèr M, Moore A, McQuay H. Propofol anaesthesia and postoperative nausea and vomiting: quantitative systematic review of randomized controlled studies. Br J Anaesth 1997; 78: 247-255.

ANALGESICS, OPIOID, fentanyl, remifentanil; OCULAR PHYSIOLOGY, intraocular pressure; SURGICAL PROCEDURES, OPERATIVE, intraoperative period, postoperative period

© 2004 European Academy of Anaesthesiology