ANESTH ANALG 1990:71:419-22
419
Cerebrospinal Fluid Pressure in Patients With Brain Tumors: Impact of Fentanyl Versus Alfentanil During Nitrous OxideOxygen Anesthesia Richard Jung, MD, Nitin Shah, MD, Ruth Reinsel, P ~ D ,William Marx, Wayne Marshall, MD, Joseph Galicich, MD,and Robert Bedford, MD JUNG R, SHAH N, REINSEL R, MAW W, MARSHALL W, GALICICH J, BEDFORD R. Cerebrospinal fluid pressure in patients with brain tumors: impact of fentanyl versus alfentanil during nitrous oxide-oxygen anesthesia. Anesth Analg 1990;71:419-22.
The efects on the cerebrospinal fluid pressure (CSFP) of alfentanil and fentanyl were compared during nitrous oxide-oxygen (N,O-0,) anesthesia in 24 patients who had brain tumors. Monitored variables included CSFP (lumbar subarachnoid catheter), heart rate from electrocardiographic lead 11, mean radial arterial blood pressure, and arterial blood gas tensions. General anesthesia was induced with thiopental, 5 mglkg IV in divided doses, and maintained with 70% N,O in 0,;ventilation was held constant (Paco, = 37.4 i 1.6 mm Hg [mean SEMI). After baseline data were recorded, 7 6 subjects were randomly assigned to receive either 5 pglkg fentanyl as an intravenous bolus or 50 pglkg alfentanil as an intravenous bolus, followed by an infusion of alfentanil at 1 pg.kg-l.min-'. Monitored variables were continuously recorded for 15 min after opioid injection. A third group of 8 patients was studied subse-
Nitrous oxide-opioid anesthesia is commonly used for neurosurgical anesthesia. The role of the newer synthetic opioids, alfentanil and sufentanil, however, has recently come under scrutiny. Although these drugs are thought to impart greater hemodynamic stability than fentanyl (l,Z),a study in patients with
Received from the Department of Anesthesiology and Critical Care Medicine and the Neurosurgical Service, Memorial SloanKettering Cancer Center, and Cornell University Medical College, New York, New York; and the Department of Anesthesiology, The Pennsylvania State University College of Medicine, Hershey, Pennsylvania. Accepted for publication July 2, 1990. Address correspondence to Dr. Bedford, Department of Anesthesiology and Critical Care Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021. 01990 by the International Anesthesia Research Society
MD,
quently; they received only N,O-0, during a 15-min observation period and served as controls. Blood pressure was held constant with an intravenous infusion of 0.1% phenylephrine, as needed; noxious stimulation was carefully avoided. Cerebrospinal fluid pressure remained unchanged both in patients who received N,O-0, alone and in those who received fentanyl-N20-0,. By contrast, those who received alfentanil-N,O-0, had a gradual increase in CSFP, reaching 30% above baseline values after 10 min and stabilizing thereafter. Although the absolute increase in CSFP during normocarbic alfentanil-N,O anesthesia was relatively small (9.5 2 1.3 mm Hg to 13.0 1.3 mm Hg [mean -t SE], P < 0.05), the absence of a similar effect after fentanyl administration suggests that precautionary measures such as hyperventilation are advisable if alfentanil is used for potentiating normocarbic N,O-0, anesthesia in neurosurgical patients with intracranial mass lesions.
*
Key Words: ANESTHETICS, GAsEs-nitrous oxide. ANALGESICS, orroIus-alfentanil, fentanyl. ANESTHESIA, NEUROSURGICAL. BRAIN, CEREBROSPINAL FLUID PRESSURE.
brain tumors has suggested that significant increases in cerebrospinal fluid pressure (CSFP) may occur when alfentanil and sufentanil are administered to supplement nitrous oxide-oxygen (N,O-0,) anesthesia when intracranial compliance is compromised (3). By contrast, fentanyl was found not to be associated with adverse CSFP effects in this situation. One problem with the previous study was that alfentanil administration was associated with a significant reduction in blood pressure (BP), which made the changes in CSFP difficult to interpret. Therefore, the present study was designed to compare the impact of alfentanil and fentanyl on CSFP during N,O-0, anesthesia in patients with brain tumors while BP was held constant.
420
JUNG ET AL.
ANESTH ANALG 1990;71:419-22
Table 1. Demographic Data for the Three Groups of Patients
Anesthetic Fen tanyl-N,O Alfentanil-N,O N,O alone
Age (Yr)
Primary tumors (n)
Secondary tumors (n)
Tumor diameter (cm)
Brain edema (%3+ scale)
54 2 5.06 42 2 5.51 48 2 5.99
5 5 5
3 3 3
3.06 2 0.32 4.00 2 0.32 3.09 2 0.18
1.75 2 0.25 1.50 2 0.29 1.13 2 0.40
N,O, nitrous oxide. Values for age, tumor diameter, and brain edema are mean
* SEM
Methods Twenty-four patients undergoing excision of supratentorial tumors were studied. The preoperative computed tomography scan of each patient's head was evaluated with regard to the following features: (a) widest diameter of tumor (cm); (b) lateral shift of midline structures (mm);and (c) amount of peritumor edema on a scale of 0-3 + , where 1+ = a small rim of edema, 2+ = a wide rim of edema, and 3+ = edema extending throughout the ipsilateral hemisphere. All patients received high-dose dexamethasone for a period of 3-4 wk before the time of surgery. All were alert and oriented preoperatively, and none had neurologic signs indicative of elevated CSFP before induction of general anesthesia. The protocol was approved by the local institutional review board, and informed consent was obtained from each patient on the evening before surgery. General anesthesia was induced with thiopental, 5 mglkg IV, in divided doses, and maintained with 70% N,O in 0,. Vecuronium, 0.1 mg/kg IV, was given to facilitate tracheal intubation; ventilation was controlled to maintain a constant end-tidal CO, tension in the 30-35 mm Hg range (Perkin-Elmer Mass Spectrometer). Radial arterial and lumbar subarachnoid catheters were placed to continuously monitor BP and CSFP, respectively. Pressure transducers were zeroed to the level of the midcranium. Heart rate (HR) was determined from electrocardiographic lead 11. During a period of stable hemodynamics without surgical stimulation, baseline HR, BP, and CSFP were recorded, and a blood sample for determination of arterial blood gas tension was collected anaerobically. Eight patients were randomized to receive fentanyl, 5 pg/kg, as an intravenous bolus; an additional eight received alfentanil, 50 pglkg, as an intravenous bolus, followed by an infusion of 1 pg.kg-'.min-'. To determine the effects on CSFP of N,O-0, anesthesia alone, a third group of 8 patients was studied subsequently; they received only 70% N 2 0 in 0, with no opioid supplementation. In all three groups, CSFP and cardiovascular parameters were recorded continuously for 15 min after place-
ment of lumbar subarachnoid catheters. Throughout the data collection period end-tidal CO, tension was held constant and external sensory stimuli were avoided. Mean arterial pressure was held constant with an intravenous infusion of 0.1% phenylephrine, as needed. Blood gas tension measurements were repeated at the end of the study period. After data acquisition was completed, each patient was hyperventilated to an end-tidal Pco, of 20-25 mm Hg and surgical incision was made. Data were subjected to statistical comparisons using analysis of variance for repeated measures and Student's t-test for paired and nonpaired data, as appropriate. To minimize the effect of minute-tominute variability, CSFP measurements obtained after the baseline values were averaged over three successive 5-min periods. These four averages were then subjected to a repeated measures analysis of variance to separate the main effects of drug and time, as well as their interaction (drug x time). To identify specifically where differences occurred, orthogonal contrasts were computed between the baseline values and those of each successive 5-min period. P < 0.05 was regarded as statistically significant.
Results The patient demographic data are shown in Table 1. There was no difference between the three groups with regard to patient age, origin of brain tumor, tumor size, or the amount of peritumor brain edema seen preoperatively on CT scan. Table 2 summarizes the baseline and final values for the measured variables during the 15-min observation period. As per the study protocol, there was no change in mean arterial pressure or Paco, in any group. There was no difference in baseline CSFP between the three patient groups. Nc, change in CSFP occurred after fentanyl administration. By contrast, alfentanil was associated with a small but statistically significant increase in CSFP by the end of the observation period. Patients receiving fentanyl and alfentanil required phenylephrine to maintain blood pres-
ALFENTANIL, FENTANYL, AND CSF PRESSURE
ANESTH ANALG 1990;71:419-22
Table 2. Measurements at Beginning and End of 15-min Observation Period CSFP (mm Hg) Alfentanil-N,O Fentanyl-N,O N,O alone MAP (mm Hg) Alfentanil-N,O Fen tanyl-N,O N,O alone HR (beatslmin) Alfentanil-N,O Fentanyl-N,O N,O alone Paco, (mm Hg) Alfentanil-N,O Fentanyl-N,O N,O alone
Baseline
15 Min
9.5 t 1.3 8.9 5 1.7 11.0 t 2.0
13.0 +- 2.0" 9.5 t 2.1 11.0 & 2.5
93.8 5 4.1 87.5 9.3 89.3 5 3.5
91.9 t 5.2 83.0 2 7.0 86.8 t 7.1
69.6 f 5.5 80.9 2 5.9 72.6 t 5.9
57.9 2 4.8" 56.2 f 3.3" 70.0 5 4.8
37.6 ? 1.6 37.3 5 1.3 35.6 t 0.9
38.2 5 1.4 36.5 t 1.2 34.0 t 1.0
*
CSFP, cerebrospinal fluid pressure; MAP, mean arterial blood pressure; HR, heart rate; Paco,, partial pressure of CO, (arterial); N,O, nitrous oxide. All values are mean -+ SE. "P < 0.05 versus baseline value.
Table 3. Mean Cerebrospinal Fluid Pressure Values at Baseline and During Three Successive 5-Min Intervals Mean cerebrospinal fluid pressure Anesthetic
f"
1-5 min"
6-10 min"
7-15 min"
Alfentanil-N,O 9.5 2 1.3 10.1 & 1.2 11.6 & 1.4' 12.7 t Fentanyl-N,O 8.9 k 1.7 8.1 t 2.4 7.8 t 2.3 6.7 k 2.8 N,Oalone 11.0 t 2.0 10.6 t 1.7 10.4 t 2.1 11.5 f 2.5 to, baseline; NzO, nitrous oxide. All values are mean 2 SE.
nIndicates minutes after baseline. bP < 0.05 versus baseline value.
sure, and there were significant reductions in HR in both of these groups by the end of the study. The control patients, given no opioid supplementation, had no change in CSFP; they required no blood pressure support, and mean arterial pressure and HR were unchanged during the observation period. In all three groups of patients, institution of hyperventilation at the end of the observation period promptly lowered CSFP below baseline values before surgical incision. Table 3 compares baseline CSFP values with subsequent 5-min averages in the three anesthetic groups studied. The drug x time interaction was significant (F6,42= 3.21, P = 0.011), indicating that the time-course of change in CSFP differed significantly among the three groups. In the first 5-min period, the effects of the drugs did not differ significantly; they diverged markedly, however, between the second and third 5-min intervals ( P < 0.03). Patients given alfentanil sustained a gradual increase
421
in CSFP that became statistically significant after 10 min and persisted for the remaining 5 min of the observation period. By contrast, CSFP was unchanged from baseline values among the patients given fentanyl and in those who received no opioid.
Discussion This study confirms that alfentanil may be associated with small increases in CSFP in normocapnic patients with brain tumors who are anesthetized with 70% N 2 0 in 0,. Although the exact mechanism for the increase in CSFP remains to be elucidated, these data are compatible with those showing an effect on alfentanil similar to that previously shown for sufentanil during N20-02 anesthesia in humans (3). Although sufentanil was not examined in the present study, it was associated with a prolonged increase in CSFP in a canine model of compromised intracranial compliance (4).This apparently was the result of an increase in cerebral blood flow, which occurred without a corresponding increase in cerebral metabolic activity (5). Alfentanil, however, does not have a similar cerebral vasodilating action in the dog (L.N. Milde, personal communication) and, to our knowledge, has not been associated with cerebral vasodilation in humans, either. Although the data in the present study suggest that alfentanil is associated with cerebrovascular effects that differ from those associated with fentanyl, it is conceivable that alfentanil might affect other aspects of intracranial compliance, such as the altering of cerebrospinal fluid production or reuptake. Preliminary work addressing this issue in a canine model, however, has found no significant effect of alfentanil on cerebrospinal fluid dynamics (A.A. Artru, personal communication). A possible criticism of this study is the difference in technique of opioid administration. Whereas fentanyl was given only as a bolus, alfentanil was given as an infusion after an initial bolus. The cerebral effects of fentanyl, as determined by the cortical electroencephalogram, last for 15 min (6). This is equal to the duration of our data collection period. By contrast, the effects of alfentanil on the cortical electroencephalogram last for only 5 min (6). It is because of this brief duration of action of alfentanil that a continuous infusion was required during our 15-min observation period. It is possible that the different administration techniques could have affected our results, particularly because the absolute increases in intracranial pressure were small. If this were the case, however, one might have anticipated an early change
422
JUNG ET AL.
ANESTH ANALG 1990;71:419-22
in CSFP after a single bolus of fentanyl, and no such change was observed. It is important to emphasize that none of the patients in the present study had signs of increased intracranial pressure preoperatively and in none was intracraniai pressure measured before induction of anesthesia. Therefore, further study is required to elucidate fully the effects of alfentanil in patients or animals with severely reduced intracranial compliance. It might be anticipated, however, that greater increases in CSFP would occur if alfentanil were administered in the presence of severely reduced intracranial compliance. The background anesthetic, N,O, also requires discussion. Several studies have shown that N20, given alone or in combination with other anesthetics, may increase CSFP, especially during the first 5-10 min of administration (7-10). Once N 2 0 uptake is nearly complete, no further increase in CSFP would be expected to occur. This was observed in our control group, which had been receiving N,O for approximately 20-30 min before CSFP measurement. Likewise, the patients who received an opioid had been receiving N 2 0 for 20-30 min before our obtaining baseline measurements. At that time, mass spectrometry indicated that end-tidal and inspired N 2 0 concentrations were virtually identical. For these reasons, we think it probable that the changes observed after alfentanil were related to the effect of the opioid and not to some other confounding factor. Although N,O may have potentiated the cerebrovascular effects of alfentanil, we believe that the role of N 2 0 was most likely to increase cerebral blood volume so that a small additional effect of alfentanil on either intracranial blood volume or cerebrospinal fluid volume would result in an accentuated CSFP increase relative to what might have been observed if N,O had not been present. The role of arterial CO, tension in this study also should be addressed. We deliberately chose to maintain normocarbia so that if an adverse effect on CSFP occurred, it could be identified easily. As has been shown during isoflurane-N,O anesthesia (ll), we found that hyperventilation during alfentanil-N,O resulted in a prompt reduction of CSFP below baseline values. This observation suggests that alfentanil may be a mild cerebral vasodilator in humans, with an effect similar to that of isoflurane. In this regard, it is not surprising that Young et al. recently observed that both isoflurane and sufentanil had virtually identical effects on cerebrovascular dynamics during N20-0, anesthesia (12). Given these findings, it
could be concluded that alfentanil is an appropriate agent to use in well-compensated patients with brain tumors, as long as hyperventilation is an integral part of the anesthetic technique. The results of this study suggest that alfentanil may be associated with an increase in CSFP in normocarbic neurosurgical patients with intracranial mass lesions. The absence of a similar effect after fentanyl administration suggests that greater caution should be exercised if alfentanil is used for potentiation of N,O-0, anesthesia. Further study is required to examine the effects of alfentanil in patients or in animals with severely reduced intracranial compliance. Pending the results of such studies, however, we suggest that hyperventilation should be employed whenever alfentanil N,O-0, anesthesia is used in patients with potentially compromised intracranial compliance.
References 1. McKay RD, Varner PD, Hendricks PL, Adams ML, Harsh GR. The evaluation of sufentanil N,O-0, vs fentanyl N,O-0, anesthesia for craniotomy (abstr). Anesth Analg 1984;63:A175. 2. Shupack RC, Harp JR. Comparison between high-dose sufentanil-oxygen and fentanyl-oxygen for neuroanaesthesia. Br J Anaesth 1985;57:37%31. 3. Marx W, Shah N, Long C, et al. Sufentanil, alfentanil and fentanyl: impact on cerebrospinal fluid pressure in patients with brain tumors. Anesthesiology 1988;69:A627. 4. Milde LN, Milde JH. Cerebral effects of sufentanil in dogs with reduced intracranial compliance (abstract). Anesth Analg 1989; 68:S196. 5. Milde LN, Milde JH. Effects of sufentanil on cerebral circulation and metabolism in dogs. Anesth Analg 1990;70:13%46. 6. Scott JC, Ponganus KV, Stanski DR. EEG quantification of narcotic effect: the comparative pharmacodynamics of fentanyl and alfentanil. Anesthesiology 1985;62:23441. 7. Henriksen HT, Jorgensen PB. The effect of nitrous oxide on intracranial pressure in patients with intracranial disorders. Br J Anaesth 1973;45:486-92. 8. Sakabe T, Kuramoto T, Inoue S, Takeshita H. Cerebral effects of nitrous oxide in the dog. Anesthesiology 1978;48:195-200. 9. Todd MM. The effects of Paco, on the cerebrovascular response to nitrous oxide in the halothane-anesthetized rabbit. Anesth Analg 1987;66:1090-5. 10. Samra SK, Deutsch G, Arens JF. The effect of nitrous oxide on global and regional cerebral blood flow in humans (abstract). Anesthesiology 1988;69:A536.
11. Adams RW, Cucchiara RF, Gronert GA, Messick JM, Michenfelder JD. Isoflurane and cerebrospinal fluid pressure in neurosurgical patients. Anesthesiology 1981;54:97-9. 12. Young WL, Prohovnik I, Correll JW, et al. A comparison of the cerebral hemodynamic effects of sufentanil and isoflurane in humans undergoing carotid endarterectomy. Anesthesiology 1989;71:863-9.
419
Cerebrospinal Fluid Pressure in Patients With Brain Tumors: Impact of Fentanyl Versus Alfentanil During Nitrous OxideOxygen Anesthesia Richard Jung, MD, Nitin Shah, MD, Ruth Reinsel, P ~ D ,William Marx, Wayne Marshall, MD, Joseph Galicich, MD,and Robert Bedford, MD JUNG R, SHAH N, REINSEL R, MAW W, MARSHALL W, GALICICH J, BEDFORD R. Cerebrospinal fluid pressure in patients with brain tumors: impact of fentanyl versus alfentanil during nitrous oxide-oxygen anesthesia. Anesth Analg 1990;71:419-22.
The efects on the cerebrospinal fluid pressure (CSFP) of alfentanil and fentanyl were compared during nitrous oxide-oxygen (N,O-0,) anesthesia in 24 patients who had brain tumors. Monitored variables included CSFP (lumbar subarachnoid catheter), heart rate from electrocardiographic lead 11, mean radial arterial blood pressure, and arterial blood gas tensions. General anesthesia was induced with thiopental, 5 mglkg IV in divided doses, and maintained with 70% N,O in 0,;ventilation was held constant (Paco, = 37.4 i 1.6 mm Hg [mean SEMI). After baseline data were recorded, 7 6 subjects were randomly assigned to receive either 5 pglkg fentanyl as an intravenous bolus or 50 pglkg alfentanil as an intravenous bolus, followed by an infusion of alfentanil at 1 pg.kg-l.min-'. Monitored variables were continuously recorded for 15 min after opioid injection. A third group of 8 patients was studied subse-
Nitrous oxide-opioid anesthesia is commonly used for neurosurgical anesthesia. The role of the newer synthetic opioids, alfentanil and sufentanil, however, has recently come under scrutiny. Although these drugs are thought to impart greater hemodynamic stability than fentanyl (l,Z),a study in patients with
Received from the Department of Anesthesiology and Critical Care Medicine and the Neurosurgical Service, Memorial SloanKettering Cancer Center, and Cornell University Medical College, New York, New York; and the Department of Anesthesiology, The Pennsylvania State University College of Medicine, Hershey, Pennsylvania. Accepted for publication July 2, 1990. Address correspondence to Dr. Bedford, Department of Anesthesiology and Critical Care Medicine, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021. 01990 by the International Anesthesia Research Society
MD,
quently; they received only N,O-0, during a 15-min observation period and served as controls. Blood pressure was held constant with an intravenous infusion of 0.1% phenylephrine, as needed; noxious stimulation was carefully avoided. Cerebrospinal fluid pressure remained unchanged both in patients who received N,O-0, alone and in those who received fentanyl-N20-0,. By contrast, those who received alfentanil-N,O-0, had a gradual increase in CSFP, reaching 30% above baseline values after 10 min and stabilizing thereafter. Although the absolute increase in CSFP during normocarbic alfentanil-N,O anesthesia was relatively small (9.5 2 1.3 mm Hg to 13.0 1.3 mm Hg [mean -t SE], P < 0.05), the absence of a similar effect after fentanyl administration suggests that precautionary measures such as hyperventilation are advisable if alfentanil is used for potentiating normocarbic N,O-0, anesthesia in neurosurgical patients with intracranial mass lesions.
*
Key Words: ANESTHETICS, GAsEs-nitrous oxide. ANALGESICS, orroIus-alfentanil, fentanyl. ANESTHESIA, NEUROSURGICAL. BRAIN, CEREBROSPINAL FLUID PRESSURE.
brain tumors has suggested that significant increases in cerebrospinal fluid pressure (CSFP) may occur when alfentanil and sufentanil are administered to supplement nitrous oxide-oxygen (N,O-0,) anesthesia when intracranial compliance is compromised (3). By contrast, fentanyl was found not to be associated with adverse CSFP effects in this situation. One problem with the previous study was that alfentanil administration was associated with a significant reduction in blood pressure (BP), which made the changes in CSFP difficult to interpret. Therefore, the present study was designed to compare the impact of alfentanil and fentanyl on CSFP during N,O-0, anesthesia in patients with brain tumors while BP was held constant.
420
JUNG ET AL.
ANESTH ANALG 1990;71:419-22
Table 1. Demographic Data for the Three Groups of Patients
Anesthetic Fen tanyl-N,O Alfentanil-N,O N,O alone
Age (Yr)
Primary tumors (n)
Secondary tumors (n)
Tumor diameter (cm)
Brain edema (%3+ scale)
54 2 5.06 42 2 5.51 48 2 5.99
5 5 5
3 3 3
3.06 2 0.32 4.00 2 0.32 3.09 2 0.18
1.75 2 0.25 1.50 2 0.29 1.13 2 0.40
N,O, nitrous oxide. Values for age, tumor diameter, and brain edema are mean
* SEM
Methods Twenty-four patients undergoing excision of supratentorial tumors were studied. The preoperative computed tomography scan of each patient's head was evaluated with regard to the following features: (a) widest diameter of tumor (cm); (b) lateral shift of midline structures (mm);and (c) amount of peritumor edema on a scale of 0-3 + , where 1+ = a small rim of edema, 2+ = a wide rim of edema, and 3+ = edema extending throughout the ipsilateral hemisphere. All patients received high-dose dexamethasone for a period of 3-4 wk before the time of surgery. All were alert and oriented preoperatively, and none had neurologic signs indicative of elevated CSFP before induction of general anesthesia. The protocol was approved by the local institutional review board, and informed consent was obtained from each patient on the evening before surgery. General anesthesia was induced with thiopental, 5 mglkg IV, in divided doses, and maintained with 70% N,O in 0,. Vecuronium, 0.1 mg/kg IV, was given to facilitate tracheal intubation; ventilation was controlled to maintain a constant end-tidal CO, tension in the 30-35 mm Hg range (Perkin-Elmer Mass Spectrometer). Radial arterial and lumbar subarachnoid catheters were placed to continuously monitor BP and CSFP, respectively. Pressure transducers were zeroed to the level of the midcranium. Heart rate (HR) was determined from electrocardiographic lead 11. During a period of stable hemodynamics without surgical stimulation, baseline HR, BP, and CSFP were recorded, and a blood sample for determination of arterial blood gas tension was collected anaerobically. Eight patients were randomized to receive fentanyl, 5 pg/kg, as an intravenous bolus; an additional eight received alfentanil, 50 pglkg, as an intravenous bolus, followed by an infusion of 1 pg.kg-'.min-'. To determine the effects on CSFP of N,O-0, anesthesia alone, a third group of 8 patients was studied subsequently; they received only 70% N 2 0 in 0, with no opioid supplementation. In all three groups, CSFP and cardiovascular parameters were recorded continuously for 15 min after place-
ment of lumbar subarachnoid catheters. Throughout the data collection period end-tidal CO, tension was held constant and external sensory stimuli were avoided. Mean arterial pressure was held constant with an intravenous infusion of 0.1% phenylephrine, as needed. Blood gas tension measurements were repeated at the end of the study period. After data acquisition was completed, each patient was hyperventilated to an end-tidal Pco, of 20-25 mm Hg and surgical incision was made. Data were subjected to statistical comparisons using analysis of variance for repeated measures and Student's t-test for paired and nonpaired data, as appropriate. To minimize the effect of minute-tominute variability, CSFP measurements obtained after the baseline values were averaged over three successive 5-min periods. These four averages were then subjected to a repeated measures analysis of variance to separate the main effects of drug and time, as well as their interaction (drug x time). To identify specifically where differences occurred, orthogonal contrasts were computed between the baseline values and those of each successive 5-min period. P < 0.05 was regarded as statistically significant.
Results The patient demographic data are shown in Table 1. There was no difference between the three groups with regard to patient age, origin of brain tumor, tumor size, or the amount of peritumor brain edema seen preoperatively on CT scan. Table 2 summarizes the baseline and final values for the measured variables during the 15-min observation period. As per the study protocol, there was no change in mean arterial pressure or Paco, in any group. There was no difference in baseline CSFP between the three patient groups. Nc, change in CSFP occurred after fentanyl administration. By contrast, alfentanil was associated with a small but statistically significant increase in CSFP by the end of the observation period. Patients receiving fentanyl and alfentanil required phenylephrine to maintain blood pres-
ALFENTANIL, FENTANYL, AND CSF PRESSURE
ANESTH ANALG 1990;71:419-22
Table 2. Measurements at Beginning and End of 15-min Observation Period CSFP (mm Hg) Alfentanil-N,O Fentanyl-N,O N,O alone MAP (mm Hg) Alfentanil-N,O Fen tanyl-N,O N,O alone HR (beatslmin) Alfentanil-N,O Fentanyl-N,O N,O alone Paco, (mm Hg) Alfentanil-N,O Fentanyl-N,O N,O alone
Baseline
15 Min
9.5 t 1.3 8.9 5 1.7 11.0 t 2.0
13.0 +- 2.0" 9.5 t 2.1 11.0 & 2.5
93.8 5 4.1 87.5 9.3 89.3 5 3.5
91.9 t 5.2 83.0 2 7.0 86.8 t 7.1
69.6 f 5.5 80.9 2 5.9 72.6 t 5.9
57.9 2 4.8" 56.2 f 3.3" 70.0 5 4.8
37.6 ? 1.6 37.3 5 1.3 35.6 t 0.9
38.2 5 1.4 36.5 t 1.2 34.0 t 1.0
*
CSFP, cerebrospinal fluid pressure; MAP, mean arterial blood pressure; HR, heart rate; Paco,, partial pressure of CO, (arterial); N,O, nitrous oxide. All values are mean -+ SE. "P < 0.05 versus baseline value.
Table 3. Mean Cerebrospinal Fluid Pressure Values at Baseline and During Three Successive 5-Min Intervals Mean cerebrospinal fluid pressure Anesthetic
f"
1-5 min"
6-10 min"
7-15 min"
Alfentanil-N,O 9.5 2 1.3 10.1 & 1.2 11.6 & 1.4' 12.7 t Fentanyl-N,O 8.9 k 1.7 8.1 t 2.4 7.8 t 2.3 6.7 k 2.8 N,Oalone 11.0 t 2.0 10.6 t 1.7 10.4 t 2.1 11.5 f 2.5 to, baseline; NzO, nitrous oxide. All values are mean 2 SE.
nIndicates minutes after baseline. bP < 0.05 versus baseline value.
sure, and there were significant reductions in HR in both of these groups by the end of the study. The control patients, given no opioid supplementation, had no change in CSFP; they required no blood pressure support, and mean arterial pressure and HR were unchanged during the observation period. In all three groups of patients, institution of hyperventilation at the end of the observation period promptly lowered CSFP below baseline values before surgical incision. Table 3 compares baseline CSFP values with subsequent 5-min averages in the three anesthetic groups studied. The drug x time interaction was significant (F6,42= 3.21, P = 0.011), indicating that the time-course of change in CSFP differed significantly among the three groups. In the first 5-min period, the effects of the drugs did not differ significantly; they diverged markedly, however, between the second and third 5-min intervals ( P < 0.03). Patients given alfentanil sustained a gradual increase
421
in CSFP that became statistically significant after 10 min and persisted for the remaining 5 min of the observation period. By contrast, CSFP was unchanged from baseline values among the patients given fentanyl and in those who received no opioid.
Discussion This study confirms that alfentanil may be associated with small increases in CSFP in normocapnic patients with brain tumors who are anesthetized with 70% N 2 0 in 0,. Although the exact mechanism for the increase in CSFP remains to be elucidated, these data are compatible with those showing an effect on alfentanil similar to that previously shown for sufentanil during N20-02 anesthesia in humans (3). Although sufentanil was not examined in the present study, it was associated with a prolonged increase in CSFP in a canine model of compromised intracranial compliance (4).This apparently was the result of an increase in cerebral blood flow, which occurred without a corresponding increase in cerebral metabolic activity (5). Alfentanil, however, does not have a similar cerebral vasodilating action in the dog (L.N. Milde, personal communication) and, to our knowledge, has not been associated with cerebral vasodilation in humans, either. Although the data in the present study suggest that alfentanil is associated with cerebrovascular effects that differ from those associated with fentanyl, it is conceivable that alfentanil might affect other aspects of intracranial compliance, such as the altering of cerebrospinal fluid production or reuptake. Preliminary work addressing this issue in a canine model, however, has found no significant effect of alfentanil on cerebrospinal fluid dynamics (A.A. Artru, personal communication). A possible criticism of this study is the difference in technique of opioid administration. Whereas fentanyl was given only as a bolus, alfentanil was given as an infusion after an initial bolus. The cerebral effects of fentanyl, as determined by the cortical electroencephalogram, last for 15 min (6). This is equal to the duration of our data collection period. By contrast, the effects of alfentanil on the cortical electroencephalogram last for only 5 min (6). It is because of this brief duration of action of alfentanil that a continuous infusion was required during our 15-min observation period. It is possible that the different administration techniques could have affected our results, particularly because the absolute increases in intracranial pressure were small. If this were the case, however, one might have anticipated an early change
422
JUNG ET AL.
ANESTH ANALG 1990;71:419-22
in CSFP after a single bolus of fentanyl, and no such change was observed. It is important to emphasize that none of the patients in the present study had signs of increased intracranial pressure preoperatively and in none was intracraniai pressure measured before induction of anesthesia. Therefore, further study is required to elucidate fully the effects of alfentanil in patients or animals with severely reduced intracranial compliance. It might be anticipated, however, that greater increases in CSFP would occur if alfentanil were administered in the presence of severely reduced intracranial compliance. The background anesthetic, N,O, also requires discussion. Several studies have shown that N20, given alone or in combination with other anesthetics, may increase CSFP, especially during the first 5-10 min of administration (7-10). Once N 2 0 uptake is nearly complete, no further increase in CSFP would be expected to occur. This was observed in our control group, which had been receiving N,O for approximately 20-30 min before CSFP measurement. Likewise, the patients who received an opioid had been receiving N 2 0 for 20-30 min before our obtaining baseline measurements. At that time, mass spectrometry indicated that end-tidal and inspired N 2 0 concentrations were virtually identical. For these reasons, we think it probable that the changes observed after alfentanil were related to the effect of the opioid and not to some other confounding factor. Although N,O may have potentiated the cerebrovascular effects of alfentanil, we believe that the role of N 2 0 was most likely to increase cerebral blood volume so that a small additional effect of alfentanil on either intracranial blood volume or cerebrospinal fluid volume would result in an accentuated CSFP increase relative to what might have been observed if N,O had not been present. The role of arterial CO, tension in this study also should be addressed. We deliberately chose to maintain normocarbia so that if an adverse effect on CSFP occurred, it could be identified easily. As has been shown during isoflurane-N,O anesthesia (ll), we found that hyperventilation during alfentanil-N,O resulted in a prompt reduction of CSFP below baseline values. This observation suggests that alfentanil may be a mild cerebral vasodilator in humans, with an effect similar to that of isoflurane. In this regard, it is not surprising that Young et al. recently observed that both isoflurane and sufentanil had virtually identical effects on cerebrovascular dynamics during N20-0, anesthesia (12). Given these findings, it
could be concluded that alfentanil is an appropriate agent to use in well-compensated patients with brain tumors, as long as hyperventilation is an integral part of the anesthetic technique. The results of this study suggest that alfentanil may be associated with an increase in CSFP in normocarbic neurosurgical patients with intracranial mass lesions. The absence of a similar effect after fentanyl administration suggests that greater caution should be exercised if alfentanil is used for potentiation of N,O-0, anesthesia. Further study is required to examine the effects of alfentanil in patients or in animals with severely reduced intracranial compliance. Pending the results of such studies, however, we suggest that hyperventilation should be employed whenever alfentanil N,O-0, anesthesia is used in patients with potentially compromised intracranial compliance.
References 1. McKay RD, Varner PD, Hendricks PL, Adams ML, Harsh GR. The evaluation of sufentanil N,O-0, vs fentanyl N,O-0, anesthesia for craniotomy (abstr). Anesth Analg 1984;63:A175. 2. Shupack RC, Harp JR. Comparison between high-dose sufentanil-oxygen and fentanyl-oxygen for neuroanaesthesia. Br J Anaesth 1985;57:37%31. 3. Marx W, Shah N, Long C, et al. Sufentanil, alfentanil and fentanyl: impact on cerebrospinal fluid pressure in patients with brain tumors. Anesthesiology 1988;69:A627. 4. Milde LN, Milde JH. Cerebral effects of sufentanil in dogs with reduced intracranial compliance (abstract). Anesth Analg 1989; 68:S196. 5. Milde LN, Milde JH. Effects of sufentanil on cerebral circulation and metabolism in dogs. Anesth Analg 1990;70:13%46. 6. Scott JC, Ponganus KV, Stanski DR. EEG quantification of narcotic effect: the comparative pharmacodynamics of fentanyl and alfentanil. Anesthesiology 1985;62:23441. 7. Henriksen HT, Jorgensen PB. The effect of nitrous oxide on intracranial pressure in patients with intracranial disorders. Br J Anaesth 1973;45:486-92. 8. Sakabe T, Kuramoto T, Inoue S, Takeshita H. Cerebral effects of nitrous oxide in the dog. Anesthesiology 1978;48:195-200. 9. Todd MM. The effects of Paco, on the cerebrovascular response to nitrous oxide in the halothane-anesthetized rabbit. Anesth Analg 1987;66:1090-5. 10. Samra SK, Deutsch G, Arens JF. The effect of nitrous oxide on global and regional cerebral blood flow in humans (abstract). Anesthesiology 1988;69:A536.
11. Adams RW, Cucchiara RF, Gronert GA, Messick JM, Michenfelder JD. Isoflurane and cerebrospinal fluid pressure in neurosurgical patients. Anesthesiology 1981;54:97-9. 12. Young WL, Prohovnik I, Correll JW, et al. A comparison of the cerebral hemodynamic effects of sufentanil and isoflurane in humans undergoing carotid endarterectomy. Anesthesiology 1989;71:863-9.
