- Email: [email protected]
Double-Blind, Bilateral Pain Comparison with Simultaneous Injection of 2% Lidocaine versus Buffered 2% Lidocaine for Periocular Anesthesia
Double-Blind, Bilateral Pain Comparison with Simultaneous Injection of 2% Lidocaine versus Buffered 2% Lidocaine for Periocular Anesthesia Mark N. Welch, DO,1 Craig N. Czyz, DO, FACOS,2 Kevin Kalwerisky, MD,2,3 David E. E. Holck, MD, FACS,3 Lisa D. Mihora, MD3 Purpose: Determine if raising the pH of 2% lidocaine with epinephrine 1:100 000 to a physiologic level decreases pain perception during periocular, subcutaneous anesthesia. Design: Double-blind, prospective, randomized study. Simultaneous unilateral injections of buffered and unbuffered lidocaine solutions were given before surgery to patients having bilateral, periocular surgery. Participants: Fifty-four consecutive patients (27 male and 27 female; mean age, 68 years; standard deviation, 11 years). Intervention: Patients were given simultaneous injections of buffered and unbuffered 2% lidocaine with epinephrine 1:100 000. The needles were inserted simultaneously and the anesthesia was injected for a 20-second count for a total volume of 1.0 ml per injected side. Main Outcome Measures: After the simultaneous injections, the patients were asked to rate the pain on each side on a Likert-type visual analog scale of 0 to 10. Results: Sixty-five percent of patients preferred the buffered lidocaine with a scaled pain reduction of 0.9 (P ⫽ 0.0005). Additionally, for the patients who believed that the buffered solution was less painful, the mean decrease in scaled pain rating was 2, for a 51% reduction in pain level (P ⫽ 0.001). No gender differences were noted. Conclusions: Buffering 2% lidocaine with epinephrine 1:100 000 with sodium bicarbonate 8.4% offers a clinically and statistically significant reduction in pain experienced by two-thirds of patients receiving periocular subcutaneous anesthesia. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Ophthalmology 2012;119:2048 –2052 © 2012 by the American Academy of Ophthalmology.
Studies have indicated that the most feared and uncomfortable part of an outpatient surgical procedure for patients is the initial injection of local anesthetic.1 There are instances where patients decline procedures or request general anesthesia based on this fear or past experience. Anecdotal reports of pain on initial injection during outpatient surgical procedures may even influence patient referral patterns. Therefore, maximization of patient comfort during the initial anesthetic injection is important in an outpatient or office-based surgical practice. This is particularly true for procedures where oral or intravenous sedation is not given before local anesthetic injection. This practice is common in periocular surgery, such as aponeurotic ptosis repair, where minimal sedation and maximum patient cooperation is desired. Lidocaine is the commercially available local anesthetic used most often in the United States, and commercial 2% lidocaine hydrochloride with epinephrine 1:100 000 is utilized routinely in outpatient oculofacial procedures for subcutaneous anesthesia and hemostasis.2 Lidocaine can precipitate as an amino peptide in its natural base form; therefore, an acidic medium with a pH of 6.4 is used for stabilization and preservation. The addition of epinephrine requires further lowering of the pH to 4.5.3
2048
© 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
There have been multiple studies investigating the use of buffered lidocaine as a topical anesthetic, and none have reported any serious complications.4 –17 Both lidocaine with epinephrine 1:100 000 and sodium bicarbonate are approved by the Food and Drug Administration as injectables (006488 and A077394, respectively). There is currently no commercially available, Food and Drug Administrationapproved buffered lidocaine solution. Therefore, the mixture must be prepared and used on site. A PubMed and Medline review of the literature was conducted in June 2011 using the search terms: lidocaine, 2% lidocaine, bicarbonate, sodium bicarbonate, buffered lidocaine, alkalinized lidocaine, and pain reduction with local anesthesia. The results revealed that 35 studies have attempted to address this question in the past 16 years. However, these studies provide disparate results. Some studies demonstrate benefit to buffered lidocaine, whereas others reveal no difference. One manuscript concludes that unbuffered lidocaine actually may be better tolerated.4 Three studies specifically evaluate eyelid anesthesia, but with conflicting conclusions.5–7 A brief review of a cross-section of prior manuscripts illustrates the large number of variables involved. A study ISSN 0161-6420/12/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.05.029
Welch et al 䡠 Pain Level with 2% Lidocaine published in a general medicine journal evaluated buffered and unbuffered lidocaine use in consecutive emergency room patients requiring local anesthetic.8 The results showed no statistically significant difference in pain rating between the lidocaine solutions. However, the study design did not address multiple confounding variables, including injured skin with enhanced pain mediators, variable size and location of wounds, injection rate, and evaluation of only 1 lidocaine solution per patient. Although these results may be somewhat relevant for wound anesthesia in an emergency room setting, the lack of controls for the confounding variables does not allow for the results to be extrapolated to other patient populations. In another study conducted by Burns et al,9 patients received randomized, sequential subcutaneous injections of buffered and unbuffered lidocaine with epinephrine. The study reported that patients experienced less pain with the buffered lidocaine solution. As in the previous study, the authors did not control for injection location or rate. Although this study did use an identical patient control, injections were sequential, not simultaneous. There also exist some large, prospective studies evaluating the pain reduction merits of buffered lidocaine in specific surgical procedures.10,11 These studies investigate nonfacial procedures such as carpal tunnel decompression surgery and neonatal circumcision. As with the previously discussed studies, these larger studies all contain differing amounts of uncontrolled confounding variables and no direct comparison to periocular procedures. The disparate nature of the results found in the literature can be the result of the following variables: basic study design, needle and syringe size, patient demographics, injection site, rate of injection, volume of anesthetic injected, temperature of anesthetic, use of control injections, acidity
of solutions, condition of target tissue, superficial versus deep injections, stress level of the patient, and injecting only 1 solution per patient or using sequential injections in the same patient.12–15 These last 2 variables likely are the most confounding. Although each previously conducted study attempted to control for some of these variables, they were all designed such that only 1 solution was injected per patient or sequential injections of the experimental and control solutions both were placed in each patient. The method of 1 solution per patient is least sensitive because of the subjective and variable ways people perceive pain. Although sequential injections in the same patient provide a level of control, it is only slightly more sensitive than the 1 solution per patient method. The confounding nature of sequential injections was demonstrated by Bartfield and Orlinsky in prospective, blinded trials, which both concluded that the second injection is perceived by the patient as more painful, thus likely skewing results even in randomized studies if the method of sequential injection is used.16,17 As a result, this calls into question the conclusions of all previous studies that address this topic. The only way to control this effect is to give bilateral, simultaneous injections in the same patient, instead of sequential injections. This double-blind, prospective, randomized study was designed to control for most of the confounding variables found in other studies. In the authors’ opinion, injection method was the most important variable to control for based on the reviewed literature. Hence, this study used a simultaneous bilateral tissue entry and injection method with only 1 entry point per needle (Fig 1). This was done in an attempt to reduce further the subjective nature of pain reporting by providing the patient with the best real-time comparison of the anesthetic solutions under study.
Figure 1. Diagram showing simultaneous injection of buffered and unbuffered lidocaine at a metronomic rate.
2049
Ophthalmology Volume 119, Number 10, October 2012
Figure 2. Diagram showing the Wong-Baker FACES Pain Rating Scale (visual analog scale shown to patients when asked for a pain rating). Reprinted from: Hockenberry MJ, Wilson D, Winkelstein ML. Wong’s Essentials of Pediatric Nursing, 7th ed. St. Louis: Mosby; 2005:1259. Copyright 2005 Mosby. Reprinted with permission.
Patients and Methods Institutional review board approval was obtained from Wilford Hall Medical Center, San Antonio, Texas. Written informed consent was obtained from each patient by the investigators, and all data were stored in a Health Insurance Portability and Accountability Actcompliant manner. Fifty-four patients (27 male and 27 female), with a mean age of 68 years (range, 42– 89 years; standard deviation, 11 years) were enrolled in this double-blind, prospective, randomized study. Inclusion criteria included patients undergoing bilateral, periocular surgery requiring subcutaneous local anesthesia. Exclusion criteria included age younger than 18 years, pregnancy, nursing, history of scarring, facial asymmetry, allergy to lidocaine or other amides, trauma, or previous surgery to the target site. Data obtained from each patient included gender, age, race, diagnosis, procedure performed, anesthetic solution used on each side, physician injector side, and the Likert-type Wong-Baker FACES Pain Rating visual analog scale ranking for each injection side. Two 1-ml tuberculin syringes with 26- or 30-gauge 1-inch needles were prepared as follows. The control syringe contained 0.9 ml 2% lidocaine with epinephrine 1:100 000 and 0.1 ml balanced salt solution, for a resulting pH of 4.5 to 5.0. The experimental syringe contained 0.9 ml 2% lidocaine with epinephrine 1:100 000 and 0.1 ml sodium bicarbonate 8.4% from a fresh bottle for a resulting pH of 7.0 to 7.5. The syringes were drawn up individually for each patient 30 minutes before use. All solutions were equilibrated at room temperature (70°–74° F) before injection. Thirty-gauge needles (outer diameter, 0.31 mm) were used for injection of eyelids and 26-gauge needles (outer diameter, 0.46 mm) were used for all other sites. Based on a randomization algorithm, each physician was assigned to a side and either the control or experimental syringe. The same 4 physicians (M.W., K.K., L.M., and D.H.) performed all injections and subsequent surgical procedures. No preoperative oral or intravenous sedation was given, nor was there pretreatment of the skin with any type of topical anesthetic or pain reduction technique before injection. The needles were inserted simultaneously (1 puncture per side), and the anesthesia was injected at a rate of 0.05 ml/second for 20 seconds, resulting in 1.0 ml total volume per side (Fig 1). Injection depth was bilaterally symmetric in the postorbicularis fascial plane for eyelids and the dermal layer for all other sites. After injection, the patients immediately were asked to rate their pain level experienced during injection on the Wong-Baker FACES Pain Rating visual analog scale for each side (Fig 2). Statistical analysis was performed with Prism 5 (GraphPad Software, Inc, La Jolla, CA) statistical software. A priori power calculation was not conducted because of the disparate results found in similar studies within the literature. The outcome measures consisting of ordinal data (pain) were analyzed using the
2050
Wilcoxon signed-rank test for intrapatient groups. This test was selected over the Mann–Whitney U test because the authors believed the sample data met the criteria for paired samples because of the extensive variable controls built into the protocol. Spearman’s correlation was used to test for adequate pairing. The Mann–Whitney U test was used for comparing interpatient groups. The Fisher exact test was used to compare all categorical data, such as patients who assigned 0 pain ratings to one solution but not the other, versus those who assigned a 0 rating for both. Statistical significance for all tests, except pairing correlation, were 2 tailed and performed at the 0.05 ␣ level.
Results All 54 of the patients enrolled completed the study. The gender, race, and age distribution of the patients is summarized in Table 1. Injections were performed before surgery for the following bilateral surgical procedures: 4 direct brow lifts, 45 upper blepharoplasties, 3 lower blepharoplasties, and 3 external levator advancements. No precipitate was noted in any of the syringes at the time of injection. All procedures were accomplished in under 45 minutes. Additional anesthesia was not required, indicating that both solutions had similar efficacy within this surgical time frame. There were no complications resulting from the injections or the surgical procedures in the preoperative, intraoperative, or postoperative periods. Thirty-five (65%) of 54 patients reported that the buffered lidocaine was less painful than the unbuffered lidocaine on initial injection. Seven patients (13%) distinguished no difference, and 12 patients (22%) felt less pain with the unbuffered lidocaine. Eleven patients (20%) reported no pain with buffered lidocaine while reporting some amount of pain on the unbuffered side, whereas only 3 patients (5%) reported no pain on the unbuffered side with some amount on the buffered side. Further analyzing these data, 13 patients Table 1. Demographic Data of the Sample Population
White Black Hispanic Total Average age (SD) Age range SD ⫽ standard deviation.
Male
Female
22 3 2 27 68 (11) 42–89
23 3 1 27 68 (12) 43–87
Welch et al 䡠 Pain Level with 2% Lidocaine Table 2. Results of the Visual Analog Pain Scale (0 –10) Survey: Intergender and Overall Population Comparison of Buffered Versus Unbuffered Lidocaine Solutions Buffered Male Mean Median Delta P value
Unbuffered
Female
2.4 2
2.7 3
Male
Female
3.3 3
0.3 0.7653
3.7 3 0.4 0.6680
Buffered vs. Unbuffered: All Patients Mean Median SD Delta P value Spearman (rs) rs P value
2.6 2 2.1
3.5 3 2.2 0.9 0.0006 0.6092 0.0001
rs ⫽ Spearman rank correlation coefficient; SD ⫽ standard deviation. Intergender unpaired comparisons were analyzed using the Mann-Whitney U test, and overall population paired comparisons were analyzed with the Wilcoxon signed-rank test.
in total reported no pain on the buffered side (24%), 2 of whom also reported no pain on the unbuffered side (15% of the 24%). Of the 5 total patients who reported 0 pain on the unbuffered side (9%), 2 also reported no pain on the buffered side (40% of the 9%). This discrepancy between groups was statistically significant (P ⫽ 0.0136). Comparing the pain rating of buffered versus unbuffered lidocaine across the entire sample population, the patients experienced 0.9 less scaled pain units on the buffered side (P ⫽ 0.0006). Pairing of the samples was found to be strong (rs ⫽ 0.6092, P ⫽ 0.0001; Table 2). When comparing the 2 lidocaine preparations between genders, there was no statistically significant difference for buffered (P ⫽ 0.7653) or unbuffered (P ⫽ 0.668) pain ratings (Table 2). Anecdotally, male patients reported lower pain scale ratings on average versus females for both lidocaine solutions. Intragender comparisons resembled that of the full sample analysis, with patients reporting less pain on the buffered side (male, P ⫽ 0.0255; female, P ⫽ 0.0150). Pairing of the samples was found to be strong for both groups (male, rs ⫽ 0.5724, P ⫽ 0.0009; female, rs ⫽ 0.6285, P ⫽ 0.0002). These results are detailed in Table 3. When the patients were grouped into those who rated the buffered solution lower on the pain scale, there was a mean difference of 2 scaled pain units, which represents a 51% reduction in pain perception from the unbuffered side (P ⫽ 0.001). Those who rated the unbuffered lidocaine as less painful had a narrower mean difference between ratings on the pain scale (1 unit), which equated to a 38% reduction in pain perception (P ⫽ 0.001).
Discussion The findings of this study clearly indicate that buffered lidocaine solution causes the patient to experience less pain with periocular injections when no adjunctive topical, oral, or intravenous anesthesia or sedation is used. There was also no variation between genders in the results. The statistically significant difference in the number of patients who reported no pain for buffered versus unbuffered solutions
further supports the clinical relevance of the findings. Similarly, those who rated the buffered solution lower on the pain scale had a larger percentage decrease in pain than those who rated the unbuffered solution lower. Although the preference of two-thirds of patients for buffered lidocaine is significant, the amount of scale-rated pain reduction between the solutions may be called into question clinically. The statistical analysis illustrates that the difference is significant, and this was supported by clinical indications. The 0.9-unit difference in pain scale rating between solutions is likely an effect of the scale used. A scale with fewer value choices would likely yield a larger delta. This is supported by 2 findings within the data. The first is that more patients (20% vs. 5%) rated the buffered solution as zero pain on the scale, while assigning a nonzero value to the unbuffered solution. The second is that when the 65% of patients who preferred the buffered solution are extracted from the sample, they reported a 10-fold increase in pain reduction over that of the sample as a whole. Within the group, the pain reduction was 51%. A 50% to 100% reduction in perceived pain certainly is significant on both clinical and statistical levels. When used in the manner described, the safety profile of buffered lidocaine is excellent. Narváez et al7 recently compared short-term complications using buffered versus unbuffered lidocaine in blepharoplasty surgery and reported no difference in the complication rate. This confirms many earlier reports that buffered lidocaine has a similar safety profile as unbuffered lidocaine and offers a substantial reduction in the pain experienced with little additional risk to the patient. The authors have 3 recommendations to maintain the safety profile of this procedure. Firstly, verify that there are no dilution errors. In a syringe with buffered lidocaine, the sodium bicarbonate should consist of only 10% of the volume of the syringe. If the ratio were reversed mistakenly, there is the potential for causing an alkali burn of the skin and even necrosis. As with any drug requiring dilution or reconstitution, such as botulinum toxin type A, care must be taken to ensure proper mixing in a sterile fashion. Secondly, use a clear syringe and visually examine the mixture before Table 3. Results of the Visual Analog Pain Scale (0 –10) Survey: Intragender Comparison of Buffered Versus Unbuffered Lidocaine Solutions Male Buffered Mean Median SD Delta P value Spearman (rs) rs P value
Female
Unbuffered
2.4 2 1.7
3.3 3 2.1 0.9 0.0255 0.5724 0.0009
Buffered
Unbuffered
2.7 3 2.4
3.7 3 2.3 1.0 0.0150 0.6285 0.0002
rs ⫽ Spearman rank correlation coefficient; SD ⫽ standard deviation. Paired intragender comparisons were analyzed using the Wilcoxon signedrank test.
2051
Ophthalmology Volume 119, Number 10, October 2012 injection. The solution should be clear without any cloudy precipitate. Do not inject any solution that appears cloudy because this may indicate that the anesthetic has precipitated out of solution because of time elapsed or higher-thanrequired concentration of sodium bicarbonate. Third, the authors recommend that the mixture be made the same day as surgery, because the sodium bicarbonate has a shelf life of 1 day after the bottle is opened, and potentially a few days longer if the mixture is refrigerated.18 The sterility of the mixed solution is no less than unbuffered lidocaine based on previous study results that reported no increase in infections for buffered solution.7 Although no studies have examined the infection rate of buffered lidocaine as the primary outcome measure, there are studies detailing the infection rate of botulinum toxin type A, which also must be mixed at time of use. These studies found no increased incidence of infection caused by performing mixing and even long-term storage and reuse.19 The cost and time required to buffer lidocaine injections are minimal, as are the potential risks, and should be considered to improve patient comfort. Because this study controls many of the confounding variables present in previous studies, we believe it definitively establishes that buffered lidocaine reduces initial injection pain in periocular subcutaneous anesthesia without adjunctive topical, oral, or intravenous anesthesia or sedation. When used appropriately, buffered lidocaine minimizes the pain associated with injection of the eyelids and offers a safety profile equal to that of unbuffered lidocaine. A follow-up study using a validated Likert-type pain scale with fewer data points likely would show a larger effect and is worth investigating. Ultimately, surgeons should not underestimate the effect of pain reduction, in any amount, on a patient’s satisfaction with their care.
References 1. Arndt KA, Burton C, Noe JM. Minimizing the pain of local anesthesia. Plast Reconstr Surg 1983;72:676 –9. 2. Hyun BH, Stevenson AJ, Hanau CA. Fundamentals of bone marrow examination. Hematol Oncol Clin North Am 1998;8: 651– 63. 3. Milner QJ, Guard BC, Allen JG. Alkalinization of amide local anaesthetics by addition of 1% sodium bicarbonate solution. Eur J Anaesthesiol 2000;17:38 – 42. 4. Lugo-Janer G, Padial M, Sánchez JL. Less painful alternatives for local anesthesia. J Dermatol Surg Oncol 1993;19:237– 40.
5. da Fonseca NL Jr, Lucci LM, Badessa MP, Rehder JR. Comparison of two modified lidocaine solutions for local anesthesia in blepharoplasty [in Portuguese]. Arq Bras Oftalmol 2009;72:211– 4. 6. Eccarius SG, Gordon ME, Parelman JJ. Bicarbonate-buffered lidocaine-epinephrine-hyaluronidase for eyelid anesthesia. Ophthalmology 1990;97:1499 –501. 7. Narváez J, Wessels I, Bacon G, et al. Prospective randomized evaluation of short-term complications when using buffered or unbuffered lidocaine 1% with epinephrine for blepharoplasty surgery. Ophthal Plast Reconstr Surg 2010;26:33–5. 8. Karcioglu O, Topacoglu H, Ayrik C, et al. Prilocaine versus plain or buffered lidocaine for local anesthesia in laceration repair: randomized double-blind comparison. Croat Med J 2003;44:716 –20. 9. Burns CA, Ferris G, Feng C, et al. Decreasing the pain of local anesthesia: a prospective, double-blind comparison of buffered, premixed 1% lidocaine with epinephrine versus 1% lidocaine freshly mixed with epinephrine. J Am Acad Dermatol 2006;54:128 –31. 10. Newton CW, Mulnix N, Baer L, Bovee T. Plain and buffered lidocaine for neonatal circumcision. Obstet Gynecol 1999;93:350–2. 11. Watts A, Hooper G. Pain scores for patients undergoing open carpal tunnel decompression under local anesthetic with plain or buffered lidocaine [letter]. J Hand Surg Br 2005;30:235; author reply 235. 12. Younis I, Bhutiani RP. Taking the ‘ouch’ out— effect of buffering commercial Xylocaine on infiltration and procedure pain—a prospective, randomized, double-blind, controlled trial. Ann R Coll Surg Engl 2004;86:213–7. 13. Scarfone RJ, Jasani M, Gracely EJ. Pain of local anesthetics: rate of administration and buffering. Ann Emerg Med 1998;31:36–40. 14. Palmon SC, Lloyd AT, Kirsch JR. The effect of needle gauge and lidocaine pH on pain during intradermal injection. Anesth Analg 1998;86:379 – 81. 15. Ruegg TA, Curran CR, Lamb T. Use of buffered lidocaine in bone marrow biopsies: a randomized, controlled trial. Oncol Nurs Forum 2009;36:52– 60. 16. Bartfield JM, Pauze D, Raccio-Robak N. The effect of order on pain of local anesthetic infiltration. Acad Emerg Med 1998;5:105–7. 17. Orlinsky M, Hudson C, Chan L, Deslauriers R. Pain comparison of unbuffered versus buffered lidocaine in local wound infiltration. J Emerg Med 1992;10:411–5. 18. Larson PO, Ragi G, Swandby M, et al. Stability of buffered lidocaine and epinephrine used for local anesthesia. J Dermatolog Surg Oncol 1991;17:411– 4. 19. Czyz CN, Hill RH, Foster JA. Cosmetic use of botulinum toxins and tissue fillers. In: Tse DT, ed. Color Atlas of Oculoplastic Surgery, 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2011:432.
Footnotes and Financial Disclosures Originally received: June 15, 2011. Final revision: May 9, 2012. Accepted: May 9, 2012. Available online: July 6, 2012.
Presented at: American Society of Ophthalmic Plastic and Reconstructive Surgery Annual Meeting, October 2009, San Francisco, California. Manuscript no. 2011-881.
1
Department of Ophthalmology, Dwight David Eisenhower Army Medical Center, Augusta, Georgia.
2
Division of Ophthalmology, Section Oculofacial Plastic and Reconstructive Surgery, Ohio University/OhioHealth Doctors Hospital, Columbus, Ohio.
3
Department of Ophthalmology, Wilford Hall Medical Center, San Antonio, Texas.
2052
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Correspondence: Mark N. Welch, DO, Department of Ophthalmology, Dwight David Eisenhower Army Medical Center, 300 Hospital Road, Augusta, Georgia 30905. E-mail: [email protected].
Studies have indicated that the most feared and uncomfortable part of an outpatient surgical procedure for patients is the initial injection of local anesthetic.1 There are instances where patients decline procedures or request general anesthesia based on this fear or past experience. Anecdotal reports of pain on initial injection during outpatient surgical procedures may even influence patient referral patterns. Therefore, maximization of patient comfort during the initial anesthetic injection is important in an outpatient or office-based surgical practice. This is particularly true for procedures where oral or intravenous sedation is not given before local anesthetic injection. This practice is common in periocular surgery, such as aponeurotic ptosis repair, where minimal sedation and maximum patient cooperation is desired. Lidocaine is the commercially available local anesthetic used most often in the United States, and commercial 2% lidocaine hydrochloride with epinephrine 1:100 000 is utilized routinely in outpatient oculofacial procedures for subcutaneous anesthesia and hemostasis.2 Lidocaine can precipitate as an amino peptide in its natural base form; therefore, an acidic medium with a pH of 6.4 is used for stabilization and preservation. The addition of epinephrine requires further lowering of the pH to 4.5.3
2048
© 2012 by the American Academy of Ophthalmology Published by Elsevier Inc.
There have been multiple studies investigating the use of buffered lidocaine as a topical anesthetic, and none have reported any serious complications.4 –17 Both lidocaine with epinephrine 1:100 000 and sodium bicarbonate are approved by the Food and Drug Administration as injectables (006488 and A077394, respectively). There is currently no commercially available, Food and Drug Administrationapproved buffered lidocaine solution. Therefore, the mixture must be prepared and used on site. A PubMed and Medline review of the literature was conducted in June 2011 using the search terms: lidocaine, 2% lidocaine, bicarbonate, sodium bicarbonate, buffered lidocaine, alkalinized lidocaine, and pain reduction with local anesthesia. The results revealed that 35 studies have attempted to address this question in the past 16 years. However, these studies provide disparate results. Some studies demonstrate benefit to buffered lidocaine, whereas others reveal no difference. One manuscript concludes that unbuffered lidocaine actually may be better tolerated.4 Three studies specifically evaluate eyelid anesthesia, but with conflicting conclusions.5–7 A brief review of a cross-section of prior manuscripts illustrates the large number of variables involved. A study ISSN 0161-6420/12/$–see front matter http://dx.doi.org/10.1016/j.ophtha.2012.05.029
Welch et al 䡠 Pain Level with 2% Lidocaine published in a general medicine journal evaluated buffered and unbuffered lidocaine use in consecutive emergency room patients requiring local anesthetic.8 The results showed no statistically significant difference in pain rating between the lidocaine solutions. However, the study design did not address multiple confounding variables, including injured skin with enhanced pain mediators, variable size and location of wounds, injection rate, and evaluation of only 1 lidocaine solution per patient. Although these results may be somewhat relevant for wound anesthesia in an emergency room setting, the lack of controls for the confounding variables does not allow for the results to be extrapolated to other patient populations. In another study conducted by Burns et al,9 patients received randomized, sequential subcutaneous injections of buffered and unbuffered lidocaine with epinephrine. The study reported that patients experienced less pain with the buffered lidocaine solution. As in the previous study, the authors did not control for injection location or rate. Although this study did use an identical patient control, injections were sequential, not simultaneous. There also exist some large, prospective studies evaluating the pain reduction merits of buffered lidocaine in specific surgical procedures.10,11 These studies investigate nonfacial procedures such as carpal tunnel decompression surgery and neonatal circumcision. As with the previously discussed studies, these larger studies all contain differing amounts of uncontrolled confounding variables and no direct comparison to periocular procedures. The disparate nature of the results found in the literature can be the result of the following variables: basic study design, needle and syringe size, patient demographics, injection site, rate of injection, volume of anesthetic injected, temperature of anesthetic, use of control injections, acidity
of solutions, condition of target tissue, superficial versus deep injections, stress level of the patient, and injecting only 1 solution per patient or using sequential injections in the same patient.12–15 These last 2 variables likely are the most confounding. Although each previously conducted study attempted to control for some of these variables, they were all designed such that only 1 solution was injected per patient or sequential injections of the experimental and control solutions both were placed in each patient. The method of 1 solution per patient is least sensitive because of the subjective and variable ways people perceive pain. Although sequential injections in the same patient provide a level of control, it is only slightly more sensitive than the 1 solution per patient method. The confounding nature of sequential injections was demonstrated by Bartfield and Orlinsky in prospective, blinded trials, which both concluded that the second injection is perceived by the patient as more painful, thus likely skewing results even in randomized studies if the method of sequential injection is used.16,17 As a result, this calls into question the conclusions of all previous studies that address this topic. The only way to control this effect is to give bilateral, simultaneous injections in the same patient, instead of sequential injections. This double-blind, prospective, randomized study was designed to control for most of the confounding variables found in other studies. In the authors’ opinion, injection method was the most important variable to control for based on the reviewed literature. Hence, this study used a simultaneous bilateral tissue entry and injection method with only 1 entry point per needle (Fig 1). This was done in an attempt to reduce further the subjective nature of pain reporting by providing the patient with the best real-time comparison of the anesthetic solutions under study.
Figure 1. Diagram showing simultaneous injection of buffered and unbuffered lidocaine at a metronomic rate.
2049
Ophthalmology Volume 119, Number 10, October 2012
Figure 2. Diagram showing the Wong-Baker FACES Pain Rating Scale (visual analog scale shown to patients when asked for a pain rating). Reprinted from: Hockenberry MJ, Wilson D, Winkelstein ML. Wong’s Essentials of Pediatric Nursing, 7th ed. St. Louis: Mosby; 2005:1259. Copyright 2005 Mosby. Reprinted with permission.
Patients and Methods Institutional review board approval was obtained from Wilford Hall Medical Center, San Antonio, Texas. Written informed consent was obtained from each patient by the investigators, and all data were stored in a Health Insurance Portability and Accountability Actcompliant manner. Fifty-four patients (27 male and 27 female), with a mean age of 68 years (range, 42– 89 years; standard deviation, 11 years) were enrolled in this double-blind, prospective, randomized study. Inclusion criteria included patients undergoing bilateral, periocular surgery requiring subcutaneous local anesthesia. Exclusion criteria included age younger than 18 years, pregnancy, nursing, history of scarring, facial asymmetry, allergy to lidocaine or other amides, trauma, or previous surgery to the target site. Data obtained from each patient included gender, age, race, diagnosis, procedure performed, anesthetic solution used on each side, physician injector side, and the Likert-type Wong-Baker FACES Pain Rating visual analog scale ranking for each injection side. Two 1-ml tuberculin syringes with 26- or 30-gauge 1-inch needles were prepared as follows. The control syringe contained 0.9 ml 2% lidocaine with epinephrine 1:100 000 and 0.1 ml balanced salt solution, for a resulting pH of 4.5 to 5.0. The experimental syringe contained 0.9 ml 2% lidocaine with epinephrine 1:100 000 and 0.1 ml sodium bicarbonate 8.4% from a fresh bottle for a resulting pH of 7.0 to 7.5. The syringes were drawn up individually for each patient 30 minutes before use. All solutions were equilibrated at room temperature (70°–74° F) before injection. Thirty-gauge needles (outer diameter, 0.31 mm) were used for injection of eyelids and 26-gauge needles (outer diameter, 0.46 mm) were used for all other sites. Based on a randomization algorithm, each physician was assigned to a side and either the control or experimental syringe. The same 4 physicians (M.W., K.K., L.M., and D.H.) performed all injections and subsequent surgical procedures. No preoperative oral or intravenous sedation was given, nor was there pretreatment of the skin with any type of topical anesthetic or pain reduction technique before injection. The needles were inserted simultaneously (1 puncture per side), and the anesthesia was injected at a rate of 0.05 ml/second for 20 seconds, resulting in 1.0 ml total volume per side (Fig 1). Injection depth was bilaterally symmetric in the postorbicularis fascial plane for eyelids and the dermal layer for all other sites. After injection, the patients immediately were asked to rate their pain level experienced during injection on the Wong-Baker FACES Pain Rating visual analog scale for each side (Fig 2). Statistical analysis was performed with Prism 5 (GraphPad Software, Inc, La Jolla, CA) statistical software. A priori power calculation was not conducted because of the disparate results found in similar studies within the literature. The outcome measures consisting of ordinal data (pain) were analyzed using the
2050
Wilcoxon signed-rank test for intrapatient groups. This test was selected over the Mann–Whitney U test because the authors believed the sample data met the criteria for paired samples because of the extensive variable controls built into the protocol. Spearman’s correlation was used to test for adequate pairing. The Mann–Whitney U test was used for comparing interpatient groups. The Fisher exact test was used to compare all categorical data, such as patients who assigned 0 pain ratings to one solution but not the other, versus those who assigned a 0 rating for both. Statistical significance for all tests, except pairing correlation, were 2 tailed and performed at the 0.05 ␣ level.
Results All 54 of the patients enrolled completed the study. The gender, race, and age distribution of the patients is summarized in Table 1. Injections were performed before surgery for the following bilateral surgical procedures: 4 direct brow lifts, 45 upper blepharoplasties, 3 lower blepharoplasties, and 3 external levator advancements. No precipitate was noted in any of the syringes at the time of injection. All procedures were accomplished in under 45 minutes. Additional anesthesia was not required, indicating that both solutions had similar efficacy within this surgical time frame. There were no complications resulting from the injections or the surgical procedures in the preoperative, intraoperative, or postoperative periods. Thirty-five (65%) of 54 patients reported that the buffered lidocaine was less painful than the unbuffered lidocaine on initial injection. Seven patients (13%) distinguished no difference, and 12 patients (22%) felt less pain with the unbuffered lidocaine. Eleven patients (20%) reported no pain with buffered lidocaine while reporting some amount of pain on the unbuffered side, whereas only 3 patients (5%) reported no pain on the unbuffered side with some amount on the buffered side. Further analyzing these data, 13 patients Table 1. Demographic Data of the Sample Population
White Black Hispanic Total Average age (SD) Age range SD ⫽ standard deviation.
Male
Female
22 3 2 27 68 (11) 42–89
23 3 1 27 68 (12) 43–87
Welch et al 䡠 Pain Level with 2% Lidocaine Table 2. Results of the Visual Analog Pain Scale (0 –10) Survey: Intergender and Overall Population Comparison of Buffered Versus Unbuffered Lidocaine Solutions Buffered Male Mean Median Delta P value
Unbuffered
Female
2.4 2
2.7 3
Male
Female
3.3 3
0.3 0.7653
3.7 3 0.4 0.6680
Buffered vs. Unbuffered: All Patients Mean Median SD Delta P value Spearman (rs) rs P value
2.6 2 2.1
3.5 3 2.2 0.9 0.0006 0.6092 0.0001
rs ⫽ Spearman rank correlation coefficient; SD ⫽ standard deviation. Intergender unpaired comparisons were analyzed using the Mann-Whitney U test, and overall population paired comparisons were analyzed with the Wilcoxon signed-rank test.
in total reported no pain on the buffered side (24%), 2 of whom also reported no pain on the unbuffered side (15% of the 24%). Of the 5 total patients who reported 0 pain on the unbuffered side (9%), 2 also reported no pain on the buffered side (40% of the 9%). This discrepancy between groups was statistically significant (P ⫽ 0.0136). Comparing the pain rating of buffered versus unbuffered lidocaine across the entire sample population, the patients experienced 0.9 less scaled pain units on the buffered side (P ⫽ 0.0006). Pairing of the samples was found to be strong (rs ⫽ 0.6092, P ⫽ 0.0001; Table 2). When comparing the 2 lidocaine preparations between genders, there was no statistically significant difference for buffered (P ⫽ 0.7653) or unbuffered (P ⫽ 0.668) pain ratings (Table 2). Anecdotally, male patients reported lower pain scale ratings on average versus females for both lidocaine solutions. Intragender comparisons resembled that of the full sample analysis, with patients reporting less pain on the buffered side (male, P ⫽ 0.0255; female, P ⫽ 0.0150). Pairing of the samples was found to be strong for both groups (male, rs ⫽ 0.5724, P ⫽ 0.0009; female, rs ⫽ 0.6285, P ⫽ 0.0002). These results are detailed in Table 3. When the patients were grouped into those who rated the buffered solution lower on the pain scale, there was a mean difference of 2 scaled pain units, which represents a 51% reduction in pain perception from the unbuffered side (P ⫽ 0.001). Those who rated the unbuffered lidocaine as less painful had a narrower mean difference between ratings on the pain scale (1 unit), which equated to a 38% reduction in pain perception (P ⫽ 0.001).
Discussion The findings of this study clearly indicate that buffered lidocaine solution causes the patient to experience less pain with periocular injections when no adjunctive topical, oral, or intravenous anesthesia or sedation is used. There was also no variation between genders in the results. The statistically significant difference in the number of patients who reported no pain for buffered versus unbuffered solutions
further supports the clinical relevance of the findings. Similarly, those who rated the buffered solution lower on the pain scale had a larger percentage decrease in pain than those who rated the unbuffered solution lower. Although the preference of two-thirds of patients for buffered lidocaine is significant, the amount of scale-rated pain reduction between the solutions may be called into question clinically. The statistical analysis illustrates that the difference is significant, and this was supported by clinical indications. The 0.9-unit difference in pain scale rating between solutions is likely an effect of the scale used. A scale with fewer value choices would likely yield a larger delta. This is supported by 2 findings within the data. The first is that more patients (20% vs. 5%) rated the buffered solution as zero pain on the scale, while assigning a nonzero value to the unbuffered solution. The second is that when the 65% of patients who preferred the buffered solution are extracted from the sample, they reported a 10-fold increase in pain reduction over that of the sample as a whole. Within the group, the pain reduction was 51%. A 50% to 100% reduction in perceived pain certainly is significant on both clinical and statistical levels. When used in the manner described, the safety profile of buffered lidocaine is excellent. Narváez et al7 recently compared short-term complications using buffered versus unbuffered lidocaine in blepharoplasty surgery and reported no difference in the complication rate. This confirms many earlier reports that buffered lidocaine has a similar safety profile as unbuffered lidocaine and offers a substantial reduction in the pain experienced with little additional risk to the patient. The authors have 3 recommendations to maintain the safety profile of this procedure. Firstly, verify that there are no dilution errors. In a syringe with buffered lidocaine, the sodium bicarbonate should consist of only 10% of the volume of the syringe. If the ratio were reversed mistakenly, there is the potential for causing an alkali burn of the skin and even necrosis. As with any drug requiring dilution or reconstitution, such as botulinum toxin type A, care must be taken to ensure proper mixing in a sterile fashion. Secondly, use a clear syringe and visually examine the mixture before Table 3. Results of the Visual Analog Pain Scale (0 –10) Survey: Intragender Comparison of Buffered Versus Unbuffered Lidocaine Solutions Male Buffered Mean Median SD Delta P value Spearman (rs) rs P value
Female
Unbuffered
2.4 2 1.7
3.3 3 2.1 0.9 0.0255 0.5724 0.0009
Buffered
Unbuffered
2.7 3 2.4
3.7 3 2.3 1.0 0.0150 0.6285 0.0002
rs ⫽ Spearman rank correlation coefficient; SD ⫽ standard deviation. Paired intragender comparisons were analyzed using the Wilcoxon signedrank test.
2051
Ophthalmology Volume 119, Number 10, October 2012 injection. The solution should be clear without any cloudy precipitate. Do not inject any solution that appears cloudy because this may indicate that the anesthetic has precipitated out of solution because of time elapsed or higher-thanrequired concentration of sodium bicarbonate. Third, the authors recommend that the mixture be made the same day as surgery, because the sodium bicarbonate has a shelf life of 1 day after the bottle is opened, and potentially a few days longer if the mixture is refrigerated.18 The sterility of the mixed solution is no less than unbuffered lidocaine based on previous study results that reported no increase in infections for buffered solution.7 Although no studies have examined the infection rate of buffered lidocaine as the primary outcome measure, there are studies detailing the infection rate of botulinum toxin type A, which also must be mixed at time of use. These studies found no increased incidence of infection caused by performing mixing and even long-term storage and reuse.19 The cost and time required to buffer lidocaine injections are minimal, as are the potential risks, and should be considered to improve patient comfort. Because this study controls many of the confounding variables present in previous studies, we believe it definitively establishes that buffered lidocaine reduces initial injection pain in periocular subcutaneous anesthesia without adjunctive topical, oral, or intravenous anesthesia or sedation. When used appropriately, buffered lidocaine minimizes the pain associated with injection of the eyelids and offers a safety profile equal to that of unbuffered lidocaine. A follow-up study using a validated Likert-type pain scale with fewer data points likely would show a larger effect and is worth investigating. Ultimately, surgeons should not underestimate the effect of pain reduction, in any amount, on a patient’s satisfaction with their care.
References 1. Arndt KA, Burton C, Noe JM. Minimizing the pain of local anesthesia. Plast Reconstr Surg 1983;72:676 –9. 2. Hyun BH, Stevenson AJ, Hanau CA. Fundamentals of bone marrow examination. Hematol Oncol Clin North Am 1998;8: 651– 63. 3. Milner QJ, Guard BC, Allen JG. Alkalinization of amide local anaesthetics by addition of 1% sodium bicarbonate solution. Eur J Anaesthesiol 2000;17:38 – 42. 4. Lugo-Janer G, Padial M, Sánchez JL. Less painful alternatives for local anesthesia. J Dermatol Surg Oncol 1993;19:237– 40.
5. da Fonseca NL Jr, Lucci LM, Badessa MP, Rehder JR. Comparison of two modified lidocaine solutions for local anesthesia in blepharoplasty [in Portuguese]. Arq Bras Oftalmol 2009;72:211– 4. 6. Eccarius SG, Gordon ME, Parelman JJ. Bicarbonate-buffered lidocaine-epinephrine-hyaluronidase for eyelid anesthesia. Ophthalmology 1990;97:1499 –501. 7. Narváez J, Wessels I, Bacon G, et al. Prospective randomized evaluation of short-term complications when using buffered or unbuffered lidocaine 1% with epinephrine for blepharoplasty surgery. Ophthal Plast Reconstr Surg 2010;26:33–5. 8. Karcioglu O, Topacoglu H, Ayrik C, et al. Prilocaine versus plain or buffered lidocaine for local anesthesia in laceration repair: randomized double-blind comparison. Croat Med J 2003;44:716 –20. 9. Burns CA, Ferris G, Feng C, et al. Decreasing the pain of local anesthesia: a prospective, double-blind comparison of buffered, premixed 1% lidocaine with epinephrine versus 1% lidocaine freshly mixed with epinephrine. J Am Acad Dermatol 2006;54:128 –31. 10. Newton CW, Mulnix N, Baer L, Bovee T. Plain and buffered lidocaine for neonatal circumcision. Obstet Gynecol 1999;93:350–2. 11. Watts A, Hooper G. Pain scores for patients undergoing open carpal tunnel decompression under local anesthetic with plain or buffered lidocaine [letter]. J Hand Surg Br 2005;30:235; author reply 235. 12. Younis I, Bhutiani RP. Taking the ‘ouch’ out— effect of buffering commercial Xylocaine on infiltration and procedure pain—a prospective, randomized, double-blind, controlled trial. Ann R Coll Surg Engl 2004;86:213–7. 13. Scarfone RJ, Jasani M, Gracely EJ. Pain of local anesthetics: rate of administration and buffering. Ann Emerg Med 1998;31:36–40. 14. Palmon SC, Lloyd AT, Kirsch JR. The effect of needle gauge and lidocaine pH on pain during intradermal injection. Anesth Analg 1998;86:379 – 81. 15. Ruegg TA, Curran CR, Lamb T. Use of buffered lidocaine in bone marrow biopsies: a randomized, controlled trial. Oncol Nurs Forum 2009;36:52– 60. 16. Bartfield JM, Pauze D, Raccio-Robak N. The effect of order on pain of local anesthetic infiltration. Acad Emerg Med 1998;5:105–7. 17. Orlinsky M, Hudson C, Chan L, Deslauriers R. Pain comparison of unbuffered versus buffered lidocaine in local wound infiltration. J Emerg Med 1992;10:411–5. 18. Larson PO, Ragi G, Swandby M, et al. Stability of buffered lidocaine and epinephrine used for local anesthesia. J Dermatolog Surg Oncol 1991;17:411– 4. 19. Czyz CN, Hill RH, Foster JA. Cosmetic use of botulinum toxins and tissue fillers. In: Tse DT, ed. Color Atlas of Oculoplastic Surgery, 2nd ed. Philadelphia: Lippincott Williams & Wilkins; 2011:432.
Footnotes and Financial Disclosures Originally received: June 15, 2011. Final revision: May 9, 2012. Accepted: May 9, 2012. Available online: July 6, 2012.
Presented at: American Society of Ophthalmic Plastic and Reconstructive Surgery Annual Meeting, October 2009, San Francisco, California. Manuscript no. 2011-881.
1
Department of Ophthalmology, Dwight David Eisenhower Army Medical Center, Augusta, Georgia.
2
Division of Ophthalmology, Section Oculofacial Plastic and Reconstructive Surgery, Ohio University/OhioHealth Doctors Hospital, Columbus, Ohio.
3
Department of Ophthalmology, Wilford Hall Medical Center, San Antonio, Texas.
2052
Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article. Correspondence: Mark N. Welch, DO, Department of Ophthalmology, Dwight David Eisenhower Army Medical Center, 300 Hospital Road, Augusta, Georgia 30905. E-mail: [email protected].