Peribulbar Anesthesia Effect of Bicarbonate on Mixtures of Lidocaine, Bupiyacaine, and Hyaluronidase with or without Epinephrine ;
KENNETH ZAHL, MD, l ALAN JORDAN, MD/ JAMES McGROARTY, MD,2 BARBARA SORENSEN, MD,3 ALEXANDER W. GOTTA, MD3
Abstract: The pH-adjustment of local anesthetic solutions with sodium bicarbonate may shorten onset time and improve spread of neural blockade, The authors undertook a prospective, double-masked, randomized study to see if a pH-adjusted mixture of lidocaine, bupivacaine, and hyaluronidase had faster and more complete onset of neural blockade, when used for peribulbar anesthesia, Eighty patients were randomly assigned to four groups and received a peribulbar block with one of four mixtures: group 1 (L) = 2% lidocaine, group 2 (LPH) = 2% lidocaine with 0.06 meq/ml sodium bicarbonate, group 3 (LE) = 2% lidocaine with 1:100,000 epinephrine (commercially prepared), or group 4 (LEPH) = 2% lidocaine with 1 :100,000 epinephrine with 0,06 meq/ml sodium bicarbonate. To 5 ml of each of the preceding groups, 5 ml of 0.75% bupivacaine and 150 units of hyaluronidase was added, After each block, extraocular muscle movement was followed in each quadrant until akinesia developed. In the event of incomplete akinesia, blocks were supplemented at 20 minutes. The LPH group had the fastest onset to complete akinesia (7,0 ± 2.0 minutes, mean ± SEM) when compared with the onset time of all other groups (group 1 = 11,5 ± 1.9 minutes, group 4 = 13.1 ± 1.4 minutes, and group 3 = 16.0 ± 1.8 minutes, significance greater than 95% by analysis of variance) , Furthermore, when compared with group 3 by analysis of variance, group 4 had a faster onset time . The authors conclude that pH-adjustment of solutions with bicarbonate of either lidocaine/bupivacaine/hyaluronidase or commercially prepared lidocaine with epinephrine/bupivacaine/hyaluronidase decreases the onset time of peribulbar anesthesia. Ophthalmology 1991 ; 98:239-242
The pH-adjustment of local anesthetics with sodium bicarbonate may improve onset time and spread of neural Originally received: August 4, 1989, Revision accepted: July 2, 1990, , Department of Anesthesiology, Mt. Sinai Sc hool of Medicine, New York , Department of Ophthalmology, Long Island College Hospital, SUNY Health Science Center at Brooklyn. 3 Department of Anesthesiology, Long Island College Hospital, SUNY Health Science Center at Brooklyn, 2
Dr. Zahl is currently affiliated with St. Luke 's/Roosevelt Hospital Center, New York , NY, Presented in part at the annual meeting of the American Society of Anes· thesiologists, San Francisco, October 1988, Address reprint requests to Kenneth Zahl, MD, Director of Outpatient Anesthesiology, Roosevelt Hospital, 428 W 59th St, New York , NY 10019,
blockade. I ,2 A mixture oflidocaine, bupivacaine, and hyaluronidase is commonly used for retrobulbar anesthesia during intraocular surgery.5 The retrobulbar block can be associated with several serious complications. 4- 6 The peri bulbar approach has been advocated to prevent the complications of retrobulbar block.? Frequent clinical problems associated with the peri bulbar block are increased time for onset of anesthesia and akinesia of the eye, higher initial failure rate, and higher volume oflocal anesthetic used when compared with retrobulbar block. Because of incomplete anesthesia, further injections often are required. We undertook a prospective, double-masked, randomized study to see if a pH-adjusted mixture of lidocaine, bupivacaine, and hyaluronidase had faster and more complete onset of neural blockade when used for peri bulbar anesthesia.
239
OPHTHALMOLOGY
•
FEBRUARY 1991
METHODS After approval by the Long Island College Hospital Institutional Review Board, 80 patients gave written informed consent for this protocol. All patients were scheduled for elective intraocular surgery with regional anesthesia and monitored anesthesia care. The only exclusion criteria was prior vitreo-retinal surgery. Eighty patients were randomly assigned to 4 groups of 20: group I (L) = 2% lidocaine (Elkins-Sinn, Cherry Hill, NJ), group 2 (LPH) = 2% lidocaine plus sodium bicarbonate 0.06 meq/ ml of L (achieved by the addition of 1.8 meq of bicarbonate to 30 ml of anesthetic), group 3 (LE) = 2% lidocaine with I: 100,000 epinephrine (premixed Elkins-Sinn), group 4 (LEPH) = 2% lidocaine with I: 100,000 epinephrine plus sodium bicarbonate 0.06 meq/ml. To avoid precipitation of the solution, 5 ml of 0.75% bupivacaine (Astra, Westborough, MA) and 150 units of hyaluronidase (Wyeth Labs, Philadelphia, PA) was added just before the block (final concentrations, lidocaine 1%, bupivacaine 0.375%, and hyaluronidase 15 u/ml). All lidocaine-containing solutions were prepared by an operating room nurse or by one of the authors (BS) and used within 3 hours. An intravenous infusion was started and all patients received 50 to 100 mcg of fentanyl and 0.5 mg of droperi dol 5 to 10 minutes intravenously before the peribulbar block. No further intravenous medications were given until complete akinesia was obtained. All blocks were performed by one of the investigators (KZ), who was unaware of the mixture selected for use. One injection was at the 6 o'clock position, just above the inferior orbital rim and the second injection was made at the superior orbital rim just below the supratrochlear notch. A 25gauge, 1.5-inch needle, with the bevel facing the globe, was inserted past the equator of the globe through the corresponding lid. Approximately 4 to 5 ml was injected in a fan-like manner below the eye and another 4 ml was given above the eye (for a total of8-9 ml). Both injections were given outside of the muscle cone. After the initial injections, each patient had light digital massage of the eyelids with gauze for 2 minutes. Extraocular muscle movement was evaluated in each quadrant at I-minute intervals for 20 minutes. A block was considered satisfactory when akinesia occurred (defined as movement of less than I mm in each quadrant). Supplemental injections (with the same mixture) were given at 20 minutes in the case of residual movement. If there was inferior and/or lateral movement, the inferior injection was repeated with I to 2 ml of injectate. The superior injection was repeated for superior and/or medial movement with I to 2 ml of anesthetic. In these cases, the actual time when complete akinesia occurred from the time of the initial injection was used for statistical analysis. Surgery was never initiated until complete akinesia was present. Adequate analgesia was determined by lack of response to incision of the conjunctiva and by lack of any patient complaint of pain during surgery. During the course of the study, sample vials represen240
•
VOLUME 98
•
NUMBER 2
tative of the lots of local anesthesia used were selected. The solutions were combined so as to simulate each of the four resulting mixtures and the final pH was measured by a Beckman model 4500 pH meter (Beckman Instruments, Inc, Fullerton, CA). Differences between onset times of akinesia between groups were evaluated by an analysis of variance with statistical significance set at greater than 95%. The need for supplemental injections was compared among groups by Fisher's exact test. A P value ofless than 0.05 was used to reject the null hypothesis.
RESULTS The mean pH values (±SEM) and volume of solution used are presented in Table I. No statistical difference was found in terms of initial volume of local anesthesia used. The mean time (from the first two injections) to akinesia in all quadrants (±SEM) is presented in Figure 1. Statistical analysis by analysis of variance shows that the LPH group had a faster onset of complete akinesia (7.0 ± 2.0 minutes) when compared with any of the other three groups. Of the epinephrine-containing solutions studied, group 4 had a statistically faster onset of extraocular muscle akinesia (13.1 ± 1.4 minutes) than did group 3 (16.0 ± 1.8 minutes). The number of patients in each group requiring supplemental injections is presented in Table 2. Although group 3 had the highest need for supplemental injections, this was not statistically significant by a Fisher's exact analysis. All patients had adequate anesthesia for the start of surgery and required no additional supplementation during surgery.
DISCUSSION Most local anesthetics are supplied as acidic salts to avoid precipitation during shipping and storage. It has been shown that the cation form of the local anesthetic is the active form of the drug. 8 Alkalinization of a local anesthetic solution with sodium bicarbonate should increase the amount of drug available in the noncation form . Although the cation is the active form of the local anesthetic, it has been postulated that the noncation penetrates through soft tissues and the nerve sheath faster. 9 Thus, increased nerve penetrance decreases onset time. The ability of pH-adjusted local anesthetic solutions to decrease the onset time for neural blockade has been questioned by some investigators. It has been shown that pH-adjusted lidocaine 2 and bupivacaine (unpublished data; Douglas et al) decreased onset time in two separate studies on epidural anesthesia. Hilgier lO showed that a pH-adjusted solution of 0.5% bupivacaine with epinephrine had a faster onset than did the control group in a study on brachial plexus anesthesia. Bedder et ai, 11 in contrast, failed to show any benefit of alkalinized bupivacaine for brachial plexus anesthesia. These different results may be secondary to differences in study designs, in
ZAHL et al
•
PERI BULBAR ANESTHESIA
Table 1. Resulting pH Values and Initial Volume Injected for the Four Solutions Used in the Study*
pH Volume (ml) L
=
L
LPH
LE
LEPH
5.43 ± 0.21 8.6 ± 0.11
6.97 ± 0.12 8.4 ± 0.12
3.84 ± 0.47 8.5 ± 0.11
6.57 ± 0.18 8.6 ± 0.13
lidocaine/bupivacaine; LPH
+ bupivacaine.
=
pH adjusted "L"
+ bupivacaine; LE
=
lidocaine with epinephrine/bupivacaine; LEPH
=
pH adjusted "LE"
* All solutions were prepared as described in the Methods section. The resulting mean pH values (± standard error of the mean) for the four final solutions as measured on the Beckman (Fullerton, tAl 4500 pH meter are shown. The initial volumes (± standard error of the mean) in milliliters used for the block are shown in the second row. There was no difference found by analysis of variance in terms of initial volume used. Table 2. Number of Patients Requiring a Supplementallnjection* L
LPH
LE
LEPH
4/20
2/20
7/20
4/20
* Numbers of patients in each group who required supplemental injections for incomplete akinesia at 20 minutes. Although the LE group had the highest need for supplemental injections, this was not statistically significant.
L
L PH
LE
LE PH
Fig 1. Onset time to complete akinesia of all extraocular muscles (mean ± SEM). Patients received a peribulbar injection with one of four solutions. L = 2% lidocaine; LPH = pH-adjusted lidocaine; LE = 2% lidocaine with epinephrine; LEPH = pH-adjusted lidocaine with epinephrine. All of the solutions were mixed with an equal volume of 0.75% bupivacaine and had hyaluronidase added (see Methods section). The LPH group had the fastest onset time (level of significance greater than 95% when analyzed by analysis of variance).
that Hilgier used an epinephrine-containing solution that was more acidic. 12 It may be that part of the controversy on the efficacy of pH-adjusted local anesthetics could be attributed to confusion from work on carbonated lidocaine. Sukhani and Winnie l3 have reported improved brachial plexus blockade with carbonated lidocaine, but Knade et al (unpublished data) report conflicting results and no advantage in using carbonated lidocaine for brachial plexus block. Carbonated salts oflocal anesthetics release carbon dioxide, which is believed to lower intracellular pH and promote intraneural local anesthetic diffusion. 13 The pH of the carbonate salt is generally equivalent to the hydrochloride salt. 13 Difficulties in needle placement for brachial block may well account for different results among investigators. The authors propose that the peribulbar block model may be more useful for studies using pH-adjusted local anesthetics. In contrast to a retrobulbar block, the injection is given within the orbit but does not penetrate the muscle cone. Thus, the onset of neural blockade
should depend mainly on diffusion. Needle placement seems to be easier and less variable, in our experience, when compared with that for a brachial plexus block. Other variables should be considered before interpreting the results. We gave each patient 8 to 9 ml of local anesthetic. For practical purposes, it is hard to give precisely the same exact volume of drug to each patient. This is due to differences in the volume of the orbit and degree of proptosis during the injection. Although there may have been 11.1 % difference in injectate, there was no statistical difference among the groups. Furthermore, since the investigator was unaware of group assignment this should not have influenced the results. We attempted to adjust the pH of the local anesthetic closer to seven for the group 4 patients. However, bupivacaine cannot be alkalinized as easy as lidocaine. Because of difficulties with precipitation of the local anesthetic after combination of the bupivacaine and lidocaine, we were unable to use more than 0.6 meq of bicarbonate per 10 ml of the 2% lidocaine with epinephrine. The consequence was that group 4 patients (the LEPH/B mixture) did not have the same pH as the group 2 patients (the LPH/B mixture). Similar problems with precipitation of bupivacaine have been reported elsewhere (unpublished data; Peterfreund et al). The role of hyaluronidase also should be addressed. As it is our customary practice to use hyaluronidase for peribulbar and retrobulbar anesthesia, we chose to continue to do so for this study. Hyaluronidase may not be effective in reducing the onset time of epidural I 5 or other regional blocks. 16 However, its value in ophthalmic regional anesthesia has been demonstrated in reducing the onset time of motor block in retrobulbar anesthesia in two studies. Nicoll et al 17 reported quicker onset of motor blockade when hyaluronidase was added to a 3:2 mixture of bupivacaine and lidocaine. Mindel 18 has reported a more 241
OPHTHALMOLOGY
•
FEBRUARY 1991
rapid onset of retrobulbar block with hyaluronidase and mepivacaine. Hyaluronidase depolymerizes hyaluronic acid and theoretically promotes the spread of local anesthetics. It may be that pH-adjustment of the lidocaine/ bupivacaine solution improves the activity of hyaluronidase. This may account for differences in our studies. To prevent oxidation of the epinephrine most commercially available local anesthetic solutions containing epinephrine are supplied at a lower pH than their counterparts without epinephrine. In our case, the premixed lidocaine with epinephrine solution used was supplied at a mean pH of approximately 3.5. Many ophthalmologists prefer to add epinephrine to counteract the vasodilating effects of the local anesthetics and prolong postoperative analgesia with lidocaine. Other studies with freshly added epinephrine may be indicated. Our initial results with pHadjusted lidocaine/bupivacaine/hyaluronidase, with or without freshly added epinephrine (unpublished data), indicate that epinephrine does not affect the onset time of peri bulbar anesthesia. The most common block performed for intraocular surgery is the retrobulbar block. With experience, the practitioner can obtain reliable anesthesia in 5 to 10 minutes with a single injection of 2 to 5 ml oflocal anesthetic. Although not common, complications from this block fall into two categories. They may be life-threatening such as a respiratory or cardiac arrest, or blindness or decreased visual acuity may result from direct trauma to the globe, optic nerve, or retina. Retrobulbar hemorrhage results in increased intraocular pressure, which can cause retinal ischemia and lead to blindness. The peribulbar block has been offered as an alternative to the retrobulbar block. 8 Because the needle does not go behind the eye into the muscle cone, in theory, there is a lower chance of ocular trauma, retrobulbar hemorrhage, or central spread of the anesthetic causing respiratory difficulties. Two or more injections are occasionally needed to obtain satisfactory anesthesia for surgery (Davis DB and Mandel M, personal communication). The extra volume used (often 2 to 4 times greater than used for retrobulbar block) may cause positive vitreous pressure and make surgery difficult. More injections may lead to an increased complication rate and defeat the conceptual advantages of peribulbar anesthesia. In this study on peribulbar anesthesia, we have demonstrated that pH-adjustment of solutions of lidocaine/ bupivacaine/hyaluronidase, with or without epinephrine, improved the onset time and quality of the motor block of the extraocular muscles. Due to the quicker onset time, more blocks were successful in the pH-adjusted groups within 20 minutes. The 4.5-minute reduction in onset time with the LPH solution and the 2.9-minute reduction with the LEPH solution do not appear to be large. However, by using a clinically relevant and arbitrary cutoff of 20 minutes for complete ocular akinesia, any reduction in onset of the block will make the peribulbar block approach the typical onset time of retrobulbar block. Many of the failed blocks in our study had gross motor movement, usually in two or more quadrants. It has been our experience that these blocks will still fail even after 40 minutes of waiting. The pH-adjustment of the local an" 242
•
VOLUME 98
•
NUMBER 2
esthetic is very simple and not time-consuming. The only disadvantage is that precipitation may occur if too much bicarbonate is added, especially with bupivacaine or if the solution is not used expeditiously. Decreased pain on injection with pH-adjusted local anesthetic for intravenous catheter insertion has been reported. 19 We believe that this is applicable to the peribulbar block, and that this is yet another argument in favor of pH-adjustment of the local anesthetic. In conclusion, pH-adjustment of lidocaine/bupivacaine/hyaluronidase mixtures with or without epinephrine using sodium bicarbonate shortens the onset time of peribulbar anesthesia. The pH-adjustment oflocal anesthetic solutions is an attractive method for improving the efficacy of peribulbar anesthesia.
REFERENCES 1. Galindo A. pH-adjusted local anesthetics: clinical experience. Reg Anaesth 1983; 8: 35-40. 2. DiFazio CA, Carron H, Grosslight KR, et al. Comparison of pH-adjusted lidocaine solutions for epidural anesthesia. Anesth Analg 1986; 65: 760-4. 3. Chin GN, Almquist HT. Bupivacaine and lidocaine retrobulbar anesthesia. A double-blind clinical study. Ophthalmology 1983; 90: 36972. 4. Javitt JC, Addiego R, Friedberg HL, et al. Brain stem anesthesia after retrobulbar block. Ophthalmology 1987; 94: 718-24. 5. Ahn JC, Stanley JA. Subarachnoid injection as a complication of retrobulbar anesthesia. Am J Ophthalmol1987; 103: 225-30. 6. Nicoll JMV, Acharya PA, Ahlen K, et al. Central nervous system complications after 6000 retrobulbar blocks. Anaesth Analg 1987; 66: 1298-302. 7. Davis DB II, Mandel MR. Posterior peribulbar anesthesia: an altemative to retrobulbar anesthesia. J Cataract Refract Surg 1986; 12: 182-4. 8. Ritchie JM, Ritchie B, Greengard P. The active structure of local anesthetics. J Pharmacol Exp Ther 1965; 150: 152-9. 9. Ritchie JM, Ritchie B, Greengard P. The effect of the nerve sheath on the action of local anesthetics. J Pharmacol Exp Ther 1965; 150: 160-4. 10. Hilgier M. Alkalinization of bupivacaine for brachial plexus block. Reg Anaesth 1985; 10: 59-61. 11. Bedder MD, Kozody R, Craig DB. Comparison of bupivacaine and alkalinized bupivacaine in brachial plexus anesthesia. Anaesth Analg 1988; 67: 48-52. 12. Pamass SM, Ivankovich AD. Alkalinized bupivacaine in brachial plexus blocks [Letter]. Anaesth Analg 1988; 67: 1017-18. 13. Sukhani R, Winnie AP. Clinical pharmacokinetics of carbonated local anesthetics I: subclavian perivascular brachial block model. Anaesth Analg 1987; 66: 739-45. 14. Sukhani R, Winnie AP. Clinical pharmacokinetics of carbonated local anesthetics II: interscalene brachial block model. Anaesth Analg 1987; 66: 1245-50. 15. Scott DB. Hyaluronidase in epidural analgesia. Br J Anaesth 1956; 28: 187-93. 16. Eckenhoff JE, Kirby CK. The use of hyaluronidase in regional nerve blocks. Anesthesiology 1951; 12: 27-32. 17. Nicoll JMV, Treuren B, Acharya PA, et al. Retrobulbar anesthesia: the role of hyaluronidase. Anaesth Analg 1986; 65: 1324-8. 18. Mindel JS. Value of hyaluronidase in ocular surgical akinesia. Am J Ophthalmol 1978; 85: 643-6. 19. McKay W, Morris R, Mushlin P. Sodium bicarbonate attenuates pain on skin infiltration with lidocaine, with or without epinephrine. Anaesth Analg 1987; 66: 572-4.
KENNETH ZAHL, MD, l ALAN JORDAN, MD/ JAMES McGROARTY, MD,2 BARBARA SORENSEN, MD,3 ALEXANDER W. GOTTA, MD3
Abstract: The pH-adjustment of local anesthetic solutions with sodium bicarbonate may shorten onset time and improve spread of neural blockade, The authors undertook a prospective, double-masked, randomized study to see if a pH-adjusted mixture of lidocaine, bupivacaine, and hyaluronidase had faster and more complete onset of neural blockade, when used for peribulbar anesthesia, Eighty patients were randomly assigned to four groups and received a peribulbar block with one of four mixtures: group 1 (L) = 2% lidocaine, group 2 (LPH) = 2% lidocaine with 0.06 meq/ml sodium bicarbonate, group 3 (LE) = 2% lidocaine with 1:100,000 epinephrine (commercially prepared), or group 4 (LEPH) = 2% lidocaine with 1 :100,000 epinephrine with 0,06 meq/ml sodium bicarbonate. To 5 ml of each of the preceding groups, 5 ml of 0.75% bupivacaine and 150 units of hyaluronidase was added, After each block, extraocular muscle movement was followed in each quadrant until akinesia developed. In the event of incomplete akinesia, blocks were supplemented at 20 minutes. The LPH group had the fastest onset to complete akinesia (7,0 ± 2.0 minutes, mean ± SEM) when compared with the onset time of all other groups (group 1 = 11,5 ± 1.9 minutes, group 4 = 13.1 ± 1.4 minutes, and group 3 = 16.0 ± 1.8 minutes, significance greater than 95% by analysis of variance) , Furthermore, when compared with group 3 by analysis of variance, group 4 had a faster onset time . The authors conclude that pH-adjustment of solutions with bicarbonate of either lidocaine/bupivacaine/hyaluronidase or commercially prepared lidocaine with epinephrine/bupivacaine/hyaluronidase decreases the onset time of peribulbar anesthesia. Ophthalmology 1991 ; 98:239-242
The pH-adjustment of local anesthetics with sodium bicarbonate may improve onset time and spread of neural Originally received: August 4, 1989, Revision accepted: July 2, 1990, , Department of Anesthesiology, Mt. Sinai Sc hool of Medicine, New York , Department of Ophthalmology, Long Island College Hospital, SUNY Health Science Center at Brooklyn. 3 Department of Anesthesiology, Long Island College Hospital, SUNY Health Science Center at Brooklyn, 2
Dr. Zahl is currently affiliated with St. Luke 's/Roosevelt Hospital Center, New York , NY, Presented in part at the annual meeting of the American Society of Anes· thesiologists, San Francisco, October 1988, Address reprint requests to Kenneth Zahl, MD, Director of Outpatient Anesthesiology, Roosevelt Hospital, 428 W 59th St, New York , NY 10019,
blockade. I ,2 A mixture oflidocaine, bupivacaine, and hyaluronidase is commonly used for retrobulbar anesthesia during intraocular surgery.5 The retrobulbar block can be associated with several serious complications. 4- 6 The peri bulbar approach has been advocated to prevent the complications of retrobulbar block.? Frequent clinical problems associated with the peri bulbar block are increased time for onset of anesthesia and akinesia of the eye, higher initial failure rate, and higher volume oflocal anesthetic used when compared with retrobulbar block. Because of incomplete anesthesia, further injections often are required. We undertook a prospective, double-masked, randomized study to see if a pH-adjusted mixture of lidocaine, bupivacaine, and hyaluronidase had faster and more complete onset of neural blockade when used for peri bulbar anesthesia.
239
OPHTHALMOLOGY
•
FEBRUARY 1991
METHODS After approval by the Long Island College Hospital Institutional Review Board, 80 patients gave written informed consent for this protocol. All patients were scheduled for elective intraocular surgery with regional anesthesia and monitored anesthesia care. The only exclusion criteria was prior vitreo-retinal surgery. Eighty patients were randomly assigned to 4 groups of 20: group I (L) = 2% lidocaine (Elkins-Sinn, Cherry Hill, NJ), group 2 (LPH) = 2% lidocaine plus sodium bicarbonate 0.06 meq/ ml of L (achieved by the addition of 1.8 meq of bicarbonate to 30 ml of anesthetic), group 3 (LE) = 2% lidocaine with I: 100,000 epinephrine (premixed Elkins-Sinn), group 4 (LEPH) = 2% lidocaine with I: 100,000 epinephrine plus sodium bicarbonate 0.06 meq/ml. To avoid precipitation of the solution, 5 ml of 0.75% bupivacaine (Astra, Westborough, MA) and 150 units of hyaluronidase (Wyeth Labs, Philadelphia, PA) was added just before the block (final concentrations, lidocaine 1%, bupivacaine 0.375%, and hyaluronidase 15 u/ml). All lidocaine-containing solutions were prepared by an operating room nurse or by one of the authors (BS) and used within 3 hours. An intravenous infusion was started and all patients received 50 to 100 mcg of fentanyl and 0.5 mg of droperi dol 5 to 10 minutes intravenously before the peribulbar block. No further intravenous medications were given until complete akinesia was obtained. All blocks were performed by one of the investigators (KZ), who was unaware of the mixture selected for use. One injection was at the 6 o'clock position, just above the inferior orbital rim and the second injection was made at the superior orbital rim just below the supratrochlear notch. A 25gauge, 1.5-inch needle, with the bevel facing the globe, was inserted past the equator of the globe through the corresponding lid. Approximately 4 to 5 ml was injected in a fan-like manner below the eye and another 4 ml was given above the eye (for a total of8-9 ml). Both injections were given outside of the muscle cone. After the initial injections, each patient had light digital massage of the eyelids with gauze for 2 minutes. Extraocular muscle movement was evaluated in each quadrant at I-minute intervals for 20 minutes. A block was considered satisfactory when akinesia occurred (defined as movement of less than I mm in each quadrant). Supplemental injections (with the same mixture) were given at 20 minutes in the case of residual movement. If there was inferior and/or lateral movement, the inferior injection was repeated with I to 2 ml of injectate. The superior injection was repeated for superior and/or medial movement with I to 2 ml of anesthetic. In these cases, the actual time when complete akinesia occurred from the time of the initial injection was used for statistical analysis. Surgery was never initiated until complete akinesia was present. Adequate analgesia was determined by lack of response to incision of the conjunctiva and by lack of any patient complaint of pain during surgery. During the course of the study, sample vials represen240
•
VOLUME 98
•
NUMBER 2
tative of the lots of local anesthesia used were selected. The solutions were combined so as to simulate each of the four resulting mixtures and the final pH was measured by a Beckman model 4500 pH meter (Beckman Instruments, Inc, Fullerton, CA). Differences between onset times of akinesia between groups were evaluated by an analysis of variance with statistical significance set at greater than 95%. The need for supplemental injections was compared among groups by Fisher's exact test. A P value ofless than 0.05 was used to reject the null hypothesis.
RESULTS The mean pH values (±SEM) and volume of solution used are presented in Table I. No statistical difference was found in terms of initial volume of local anesthesia used. The mean time (from the first two injections) to akinesia in all quadrants (±SEM) is presented in Figure 1. Statistical analysis by analysis of variance shows that the LPH group had a faster onset of complete akinesia (7.0 ± 2.0 minutes) when compared with any of the other three groups. Of the epinephrine-containing solutions studied, group 4 had a statistically faster onset of extraocular muscle akinesia (13.1 ± 1.4 minutes) than did group 3 (16.0 ± 1.8 minutes). The number of patients in each group requiring supplemental injections is presented in Table 2. Although group 3 had the highest need for supplemental injections, this was not statistically significant by a Fisher's exact analysis. All patients had adequate anesthesia for the start of surgery and required no additional supplementation during surgery.
DISCUSSION Most local anesthetics are supplied as acidic salts to avoid precipitation during shipping and storage. It has been shown that the cation form of the local anesthetic is the active form of the drug. 8 Alkalinization of a local anesthetic solution with sodium bicarbonate should increase the amount of drug available in the noncation form . Although the cation is the active form of the local anesthetic, it has been postulated that the noncation penetrates through soft tissues and the nerve sheath faster. 9 Thus, increased nerve penetrance decreases onset time. The ability of pH-adjusted local anesthetic solutions to decrease the onset time for neural blockade has been questioned by some investigators. It has been shown that pH-adjusted lidocaine 2 and bupivacaine (unpublished data; Douglas et al) decreased onset time in two separate studies on epidural anesthesia. Hilgier lO showed that a pH-adjusted solution of 0.5% bupivacaine with epinephrine had a faster onset than did the control group in a study on brachial plexus anesthesia. Bedder et ai, 11 in contrast, failed to show any benefit of alkalinized bupivacaine for brachial plexus anesthesia. These different results may be secondary to differences in study designs, in
ZAHL et al
•
PERI BULBAR ANESTHESIA
Table 1. Resulting pH Values and Initial Volume Injected for the Four Solutions Used in the Study*
pH Volume (ml) L
=
L
LPH
LE
LEPH
5.43 ± 0.21 8.6 ± 0.11
6.97 ± 0.12 8.4 ± 0.12
3.84 ± 0.47 8.5 ± 0.11
6.57 ± 0.18 8.6 ± 0.13
lidocaine/bupivacaine; LPH
+ bupivacaine.
=
pH adjusted "L"
+ bupivacaine; LE
=
lidocaine with epinephrine/bupivacaine; LEPH
=
pH adjusted "LE"
* All solutions were prepared as described in the Methods section. The resulting mean pH values (± standard error of the mean) for the four final solutions as measured on the Beckman (Fullerton, tAl 4500 pH meter are shown. The initial volumes (± standard error of the mean) in milliliters used for the block are shown in the second row. There was no difference found by analysis of variance in terms of initial volume used. Table 2. Number of Patients Requiring a Supplementallnjection* L
LPH
LE
LEPH
4/20
2/20
7/20
4/20
* Numbers of patients in each group who required supplemental injections for incomplete akinesia at 20 minutes. Although the LE group had the highest need for supplemental injections, this was not statistically significant.
L
L PH
LE
LE PH
Fig 1. Onset time to complete akinesia of all extraocular muscles (mean ± SEM). Patients received a peribulbar injection with one of four solutions. L = 2% lidocaine; LPH = pH-adjusted lidocaine; LE = 2% lidocaine with epinephrine; LEPH = pH-adjusted lidocaine with epinephrine. All of the solutions were mixed with an equal volume of 0.75% bupivacaine and had hyaluronidase added (see Methods section). The LPH group had the fastest onset time (level of significance greater than 95% when analyzed by analysis of variance).
that Hilgier used an epinephrine-containing solution that was more acidic. 12 It may be that part of the controversy on the efficacy of pH-adjusted local anesthetics could be attributed to confusion from work on carbonated lidocaine. Sukhani and Winnie l3 have reported improved brachial plexus blockade with carbonated lidocaine, but Knade et al (unpublished data) report conflicting results and no advantage in using carbonated lidocaine for brachial plexus block. Carbonated salts oflocal anesthetics release carbon dioxide, which is believed to lower intracellular pH and promote intraneural local anesthetic diffusion. 13 The pH of the carbonate salt is generally equivalent to the hydrochloride salt. 13 Difficulties in needle placement for brachial block may well account for different results among investigators. The authors propose that the peribulbar block model may be more useful for studies using pH-adjusted local anesthetics. In contrast to a retrobulbar block, the injection is given within the orbit but does not penetrate the muscle cone. Thus, the onset of neural blockade
should depend mainly on diffusion. Needle placement seems to be easier and less variable, in our experience, when compared with that for a brachial plexus block. Other variables should be considered before interpreting the results. We gave each patient 8 to 9 ml of local anesthetic. For practical purposes, it is hard to give precisely the same exact volume of drug to each patient. This is due to differences in the volume of the orbit and degree of proptosis during the injection. Although there may have been 11.1 % difference in injectate, there was no statistical difference among the groups. Furthermore, since the investigator was unaware of group assignment this should not have influenced the results. We attempted to adjust the pH of the local anesthetic closer to seven for the group 4 patients. However, bupivacaine cannot be alkalinized as easy as lidocaine. Because of difficulties with precipitation of the local anesthetic after combination of the bupivacaine and lidocaine, we were unable to use more than 0.6 meq of bicarbonate per 10 ml of the 2% lidocaine with epinephrine. The consequence was that group 4 patients (the LEPH/B mixture) did not have the same pH as the group 2 patients (the LPH/B mixture). Similar problems with precipitation of bupivacaine have been reported elsewhere (unpublished data; Peterfreund et al). The role of hyaluronidase also should be addressed. As it is our customary practice to use hyaluronidase for peribulbar and retrobulbar anesthesia, we chose to continue to do so for this study. Hyaluronidase may not be effective in reducing the onset time of epidural I 5 or other regional blocks. 16 However, its value in ophthalmic regional anesthesia has been demonstrated in reducing the onset time of motor block in retrobulbar anesthesia in two studies. Nicoll et al 17 reported quicker onset of motor blockade when hyaluronidase was added to a 3:2 mixture of bupivacaine and lidocaine. Mindel 18 has reported a more 241
OPHTHALMOLOGY
•
FEBRUARY 1991
rapid onset of retrobulbar block with hyaluronidase and mepivacaine. Hyaluronidase depolymerizes hyaluronic acid and theoretically promotes the spread of local anesthetics. It may be that pH-adjustment of the lidocaine/ bupivacaine solution improves the activity of hyaluronidase. This may account for differences in our studies. To prevent oxidation of the epinephrine most commercially available local anesthetic solutions containing epinephrine are supplied at a lower pH than their counterparts without epinephrine. In our case, the premixed lidocaine with epinephrine solution used was supplied at a mean pH of approximately 3.5. Many ophthalmologists prefer to add epinephrine to counteract the vasodilating effects of the local anesthetics and prolong postoperative analgesia with lidocaine. Other studies with freshly added epinephrine may be indicated. Our initial results with pHadjusted lidocaine/bupivacaine/hyaluronidase, with or without freshly added epinephrine (unpublished data), indicate that epinephrine does not affect the onset time of peri bulbar anesthesia. The most common block performed for intraocular surgery is the retrobulbar block. With experience, the practitioner can obtain reliable anesthesia in 5 to 10 minutes with a single injection of 2 to 5 ml oflocal anesthetic. Although not common, complications from this block fall into two categories. They may be life-threatening such as a respiratory or cardiac arrest, or blindness or decreased visual acuity may result from direct trauma to the globe, optic nerve, or retina. Retrobulbar hemorrhage results in increased intraocular pressure, which can cause retinal ischemia and lead to blindness. The peribulbar block has been offered as an alternative to the retrobulbar block. 8 Because the needle does not go behind the eye into the muscle cone, in theory, there is a lower chance of ocular trauma, retrobulbar hemorrhage, or central spread of the anesthetic causing respiratory difficulties. Two or more injections are occasionally needed to obtain satisfactory anesthesia for surgery (Davis DB and Mandel M, personal communication). The extra volume used (often 2 to 4 times greater than used for retrobulbar block) may cause positive vitreous pressure and make surgery difficult. More injections may lead to an increased complication rate and defeat the conceptual advantages of peribulbar anesthesia. In this study on peribulbar anesthesia, we have demonstrated that pH-adjustment of solutions of lidocaine/ bupivacaine/hyaluronidase, with or without epinephrine, improved the onset time and quality of the motor block of the extraocular muscles. Due to the quicker onset time, more blocks were successful in the pH-adjusted groups within 20 minutes. The 4.5-minute reduction in onset time with the LPH solution and the 2.9-minute reduction with the LEPH solution do not appear to be large. However, by using a clinically relevant and arbitrary cutoff of 20 minutes for complete ocular akinesia, any reduction in onset of the block will make the peribulbar block approach the typical onset time of retrobulbar block. Many of the failed blocks in our study had gross motor movement, usually in two or more quadrants. It has been our experience that these blocks will still fail even after 40 minutes of waiting. The pH-adjustment of the local an" 242
•
VOLUME 98
•
NUMBER 2
esthetic is very simple and not time-consuming. The only disadvantage is that precipitation may occur if too much bicarbonate is added, especially with bupivacaine or if the solution is not used expeditiously. Decreased pain on injection with pH-adjusted local anesthetic for intravenous catheter insertion has been reported. 19 We believe that this is applicable to the peribulbar block, and that this is yet another argument in favor of pH-adjustment of the local anesthetic. In conclusion, pH-adjustment of lidocaine/bupivacaine/hyaluronidase mixtures with or without epinephrine using sodium bicarbonate shortens the onset time of peribulbar anesthesia. The pH-adjustment oflocal anesthetic solutions is an attractive method for improving the efficacy of peribulbar anesthesia.
REFERENCES 1. Galindo A. pH-adjusted local anesthetics: clinical experience. Reg Anaesth 1983; 8: 35-40. 2. DiFazio CA, Carron H, Grosslight KR, et al. Comparison of pH-adjusted lidocaine solutions for epidural anesthesia. Anesth Analg 1986; 65: 760-4. 3. Chin GN, Almquist HT. Bupivacaine and lidocaine retrobulbar anesthesia. A double-blind clinical study. Ophthalmology 1983; 90: 36972. 4. Javitt JC, Addiego R, Friedberg HL, et al. Brain stem anesthesia after retrobulbar block. Ophthalmology 1987; 94: 718-24. 5. Ahn JC, Stanley JA. Subarachnoid injection as a complication of retrobulbar anesthesia. Am J Ophthalmol1987; 103: 225-30. 6. Nicoll JMV, Acharya PA, Ahlen K, et al. Central nervous system complications after 6000 retrobulbar blocks. Anaesth Analg 1987; 66: 1298-302. 7. Davis DB II, Mandel MR. Posterior peribulbar anesthesia: an altemative to retrobulbar anesthesia. J Cataract Refract Surg 1986; 12: 182-4. 8. Ritchie JM, Ritchie B, Greengard P. The active structure of local anesthetics. J Pharmacol Exp Ther 1965; 150: 152-9. 9. Ritchie JM, Ritchie B, Greengard P. The effect of the nerve sheath on the action of local anesthetics. J Pharmacol Exp Ther 1965; 150: 160-4. 10. Hilgier M. Alkalinization of bupivacaine for brachial plexus block. Reg Anaesth 1985; 10: 59-61. 11. Bedder MD, Kozody R, Craig DB. Comparison of bupivacaine and alkalinized bupivacaine in brachial plexus anesthesia. Anaesth Analg 1988; 67: 48-52. 12. Pamass SM, Ivankovich AD. Alkalinized bupivacaine in brachial plexus blocks [Letter]. Anaesth Analg 1988; 67: 1017-18. 13. Sukhani R, Winnie AP. Clinical pharmacokinetics of carbonated local anesthetics I: subclavian perivascular brachial block model. Anaesth Analg 1987; 66: 739-45. 14. Sukhani R, Winnie AP. Clinical pharmacokinetics of carbonated local anesthetics II: interscalene brachial block model. Anaesth Analg 1987; 66: 1245-50. 15. Scott DB. Hyaluronidase in epidural analgesia. Br J Anaesth 1956; 28: 187-93. 16. Eckenhoff JE, Kirby CK. The use of hyaluronidase in regional nerve blocks. Anesthesiology 1951; 12: 27-32. 17. Nicoll JMV, Treuren B, Acharya PA, et al. Retrobulbar anesthesia: the role of hyaluronidase. Anaesth Analg 1986; 65: 1324-8. 18. Mindel JS. Value of hyaluronidase in ocular surgical akinesia. Am J Ophthalmol 1978; 85: 643-6. 19. McKay W, Morris R, Mushlin P. Sodium bicarbonate attenuates pain on skin infiltration with lidocaine, with or without epinephrine. Anaesth Analg 1987; 66: 572-4.
