- Email: [email protected]
Retrobulbar Anesthesia Risk
Retrobulbar Anesthesia Risk Do Sharp Needles Really Perforate the Eye More Easily than Blunt Needles?
Stephen G. Waller, MD/ John Taboada, PhD,2 Patrick O'Connor, MIY Purpose: Conventional wisdom in ophthalmology is that the force required to perforate an eye during retrobulbar injection is noticeably greater with a specially designed blunt needle than with a standard hypodermic needle. A search of the literature showed no measurements of scleral perforation pressure with specific needle tips. The authors investigated this concept. Methods: A computerized search for eye perforations of the surgical records over a 5-year period at Wilford Hall United States Air Force Medical Center was conducted. A double-masked trial by experienced ophthalmologists, using preserved eye bank eyes and several commercially available needles, subjectively assessed the force required to perforate the globe. The authors designed and built a portable transducer system to objectively measure the perforation pressure of human globes with the needles. Measurements were done with preserved and unpreserved human cadaver eyes. Resuns: No globe penetrations or perforations were found in this consecutive series of over 4000 retrobulbar anesthesia procedures. A subjective difference between the hypodermic and blunt needles was detected. Objective measurements showed a significant difference between the hypodermic and blunt needles, and between types of blunt needles. The difference was present with both eye bank eyes and fresh cadaver eyes. Conclusion: Blunt-tipped needles do require greater force to penetrate the eye. The non cutting edge, blunt-tipped needles have higher scleral perforation pressures than those with cutting edges. Ophthalmology 1993;100:506-510
Over one million ophthalmic surgical procedures are performed in the United States annually. Minimizing the risk of systemic agents and prompt return to ambulatory status are desirable goals in outpatient surgery. Adequate
Originally received: June 15, 1992. Revision accepted: October 15, 1992. I Department of Ophthalmology, United States Air Force Medical Center, Lackland Air Force Base, San Antonio. 2 Ophthalmology Branch, United States Air Force School of Aerospace Medicine, Brooks Air Force Base, San Antonio. Dr. Waller is currently affiliated with the Massachusetts Eye and Ear Infirmary, Boston. The views expressed in this article are those of the authors and do not reflect the official policy of the Department of Defense or other Departments of the United States Government. Reprint requests to Stephen G. Waller, MD, Cornea Service, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114.
506
anesthesia and akinesia during ophthalmic surgery are often paramount to success. For these reasons, the great majority of eye surgeries are performed under local anesthesia, usually induced with a specially designed retrobulbar needle. Many retrobulbar and peribulbar blocks are administered by non ophthalmologists, possibly enhancing the risk of needle perforation during injection. I Other substances, such as corticosteroids, antibiotics, and alcohol also are injected into the orbit in selected circumstances. Retrobulbar and peri bulbar anesthesia have an important role in ophthalmology. The act of injecting substances into the orbit is associated with a number of rare but catastrophic complications. Globe perforation,2 optic nerve injection,3 brain stem anesthesia, 3 and vascular occlusive events4 all have been reported. The first retrobulbar anesthesia was used in an enucleation by Hermann Knapp 5 in 1884. The toxicity of the agent he used, cocaine, delayed acceptance of his tech-
Waller et al . Retrobulbar Needles nique. It was not until the work of Atkinson 6 in the 1930s and the development ·of the safer agent, procaine, that retrobulbar injections became common. To avoid scleral perforation, it is recommended that the beveled surface of the retrobulbar needle be aligned with and apposed to the globe. It should pass posteriorly along the orbital floor until it is behind the equator of the globe, then be directed into the muscle cone. Peribulbar injections are made after the needle passes parallel to the orbital floor or roof, carefully avoiding movement of the needle tip toward the globe. Despite these precautions, some iatrogenic scleral perforations do occur. The incidence of globe penetration or perforation during the administration of local orbital anesthesia can be estimated. Published studies from two retinal surgery practices show an approximate incidence of I in 1000. 7•8 Risk appears to be higher when the globe is longer, particularly if it is more than 26 mm on A-scan ultrasonic measurement. 2 Presumably, a disproportionate share of the eyes undergoing retinal surgery are in this category, and the true incidence may be somewhat less than 0.1 %. Other authors report no scleral perforations in 1600 peribulbar anesthetic injections,9 and one perforation in 12,000 cases of both retrobulbar and peribulbar anesthesia. lo There are three different needle tip designs in common retrobulbar and peribulbar use today. The standard hypodermic needle has a steep bevel and a cutting edge. The Atkinson (Visitec Co., Sarasota, FL) needle has a flat bevel and a cutting edge. The Straus needle (Visitec Co., Sarasota, FL) is a curved needle with an Atkinson-style tip. The Thornton (Alcon Surgical Inc., Ft. Worth, TX) needle has a flat bevel and a non-cutting edge. The use of flat-bevel ("blunt-tipped") needles, designed specifically for orbital injections, is thought to lessen the risk of inadvertent ocular penetration or perforation. II However, a number of these events occur despite the use of the blunt needles. Two recent reports of the clinical course of inadvertent globe perforations cited the use of blunt needles in 5 of 12 and 7 of 23 of the injuries. 1•12 There is also a possibility that a perforating blunt needle tip does more retinal damage than its sharper counterpart. 1 A recent article described the mechanics of various surgical needles used for passing suture efficiently through tissue. 13 However, there is little similar literature on needles used in injections. The scarcity of literature on penetration or perforation rates includes an absence of any comparison between rates using different needle tip shapes or caliber. We examined the force necessary to penetrate a human globe with the various sharp and blunt needles available on the market today. Are the needle tips different and can the difference be detected by experienced surgeons?
Methods Globe perforation occurs at a low rate among patients undergoing routine ophthalmic surgery preceded by retrobulbar or peribulbar anesthesia. To establish the mag-
nitude of this problem in our multi-specialty, residencytraining hospital practice, we searched the inpatient record files at Wilford Hall United States Air Force (USAF) Medical Center for iatrogenic scleral perforations during local anesthesia, retrospectively evaluating all ophthalmology records coded as scleral, ocular, eye, or globe penetration or perforation over a 5-year period. We included surgical cases from all subspecialities in ophthalmology. To test the hypothesis that commercially available, blunt-tipped retrobulbar needles are less likely to penetrate the globe than standard hypodermic needles, we designed a three-part experiment. The first part involved the use of preserved eye bank eyes. Ten board-certified ophthalmologists passed sharp and blunt (Atkinson tip) needles and judged which needle was easier and which was more difficult. The needles used were the 23-gauge X 38-mm (Ilh-inch) hypodermic and the Atkinson 23-gauge X 38 mm. The needle hubs were masked, and close inspection of the needle tips was not permitted, so that neither the surgeon nor an observer who recorded the results knew which needle type was being used. The sequence of needle use also was randomly varied, and each surgeon was only permitted one pass for each needle. The globes were secured with gauze and pins to a flat wooden board. Needle passes were made perpendicular to the sclera near the equator of the globe. To control for variations in scleral thickness on the same eye, each surgeon made passes in the same general area of the globe. Each ophthalmologist used three or more globes, depending on the availability of eye bank eyes. No eye or needle was used more than three times. In the second part of our experiment, we obtained objective data. An appropriate transducer (Fig 1) was custom designed and constructed for the purpose of transferring pressure data from the needle tip to an electronic recorder. The transducer is a double-tipped, water-filled syringe. Force is mechanically transmitted along the long axis of the needle to the distal end of the transducer. There a water-filled cavity is used to conduct the pressure to an electromechanical coupling with the low pressure (umbilical vein) channel of a Spacelabs (Redmond, WA) Model 906 22A portable electronic recorder, normally used to continuously monitor vital signs on neonatal patients at Wilford Hall USAF Medical Center. The arrangement was calibrated by holding it at the level of the eye. The transducer was attached to the recorder through a Medex Sure-Cal manual, pre-calibrated, 100-mmHg, square-wave generator (Medex, Hilliard, OH), which was used to verify calibration of the electronic recorder. The transducer apparatus was first used with ten preserved eye bank eyes. The globes were again fixed by wrapping them in gauze and pinning the gauze to a platform. The needle passes were made obliquely to simulate the angle of an inadvertent penetration. Scleral perforation pressure values were measured from strip chart recordings of the needle pass (Fig 2). Each needle and each globe were used twice in succession. Because the results were objectively measured, no attempt to mask the needle hub or to vary the sequence of needle use was made. In addition to the above needles, 27- and 18-gauge hypodermic
507
Ophthalmology
Volume 100, Number 4, April 1993 of the original pass. We believe that this effect did not change our results.
Results
Figure 1. Photograph of the custom-made transducer used by the authors for measuring the scleral perforation pressures of various retrobulbar needles. The needle is attached to the tip of the syringe and the waterfilled tube leading from the syringe transmits pressure information to an adjacent portable electronic recorder.
and 25-gauge X 34 mm (1 %-inch) curved needle with an Atkinson tip (Straus design) needles were tested. In the final phase of the study, unpreserved cadaver globes were tested in the orbit with the same portable pressure-monitoring system. The needle was passed in the frontal plane near the lateral canthus and perpendicular to the globe under direct visualization. Orbital passes were avoided to allow measurement of globe perforation only, unobscured by resistance from orbital tissue. Scleral perforation pressures were measured on ten undisturbed eyes from five cadavers, each within 24 hours of death. Since the hypodermic needle passes seemed to produce no detectable hypotony or fluid leak from the sclera, the 30gauge needle and a 25-gauge X Sis-inch hypodermic needle were used first. Each globe was then used twice with a varied sequence of four other needles: the Atkinson and Straus needles, and 38-mm Thornton needles of 23 and 25 gauge. No individual needle was used more than three times, but each needle type was used 20 times. Minimal leakage did occur during this part of the experiment, but second passes of the same needle measured within 10%
Figure 2. Typical recording of scleral perforation pass with the retrobulbar needle. Notice the gradual rise in pressure as the needle tip attempts to perforate the sclera. When the pressure is sufficient, the needle passes through the sclera with a precipitous drop in pressure. The peak pressure is the value interpreted as scleral preforation pressure.
508
1 ')
To ascertain the frequency of globe injuries from anesthesia injections at our large military teaching hospital, a computer search of the inpatient record files was made. We examined over 4000 charts and found no instances of scleral perforation during retrobulbar or peri bulbar injection. The injections were all performed by ophthalmologists, with a majority by residents at all stages of training. A high percentage of the injections were done with the Atkinson blunt-tipped needle, and the retrobulbar technique was used. While the possibility of occult perforation exists, there were no admissions or surgeries to repair iatrogenic detachments from injections during a prior admission. The incidence of high myopia, posterior staphyloma, and multiple needle passes during the administration of local anesthesia was not determined. The initial experimental part of our study was subjective. Ten board-certified ophthalmologists tested which of two needles, both of the same caliber and length but of different tip design, penetrated a preserved eye bank eye with less force. Of 32 trials, the sharp needle was chosen 17 of 24 times, with eight passes judged identical. This difference is statistically significant at less than 0.05 by chi-square testing. None of the surgeons had a perfect or zero score, nor had enough needle passes to acquire a learning effect. The peak amplitude for each needle pass was measured in millimeters of mercury to obtain the objective results for the second and third parts of the experiment. Figure 2 shows the gradual increase in pressure until globe penetration occurs, when the pressure drops promptly. The peak pressure was considered the scleral perforation pressure. Statistical evaluation of the objective data involved the calculation of means of the different data groups, and computerized analysis to evaluate the null hypothesis of no difference between any of the needle types examined. The F probability from a one-way analysis of variance
Waller et al . Retrobulbar Needles was less than 0.0001. Using the Tukey HSD paired comparison methods, separate sets of analyses were conducted for preserved and fresh cadaver eyes. The means and standard deviations for each needle type are shown in Table 1. The striking differences between the hypodermic data and the results from the three blunt-tipped needles are significant at 0.0001. In addition, the Atkinson and Straus needle results are different from the Thornton needle data at 0.05. The Atkinson and Straus needles were not significantly different from each other. In addition, our data allowed assessment of the effect of needle caliber on scleral perforation pressure. These data also are shown in Table 1. Differences between different caliber needles of identical tip design were seen with the Thornton needles, and this was significant at less than 0.001. The hypodermic and Straus needles did not show statistically significant differences between the various needle calibers tested. Although the variances for the Thornton needles were somewhat larger, the coefficient of variation was similar among the needle types; we therefore assumed homogeneity of variance.
Discussion The incidence of inadvertent scleral perforation during retrobulbar or peribulbar injection is known only from two reports of retinal surgery practices. We undertook a retrospective search of inpatient records at our facility, and found no perforations in more than 4000 consecutive cases of routine ophthalmologic surgery over a 5-year period. Although perforations are catastrophic, they are fortunately rare in our multi-specialty, residency-training practice situation. In conclusion, a sharp hypodermic needle tip should perforate human tissue, including the eye, more easily than a tip with a blunt design. Likewise, a small caliber needle should pass through a tissue barrier with less force
Table 1. Scleral Perforation Pressures Scleral Perforation Pressure (mmHg)
Mean ± SD Preserved Eyes 3D-gauge hypodermic 27-gauge hypodermic 18-gauge hypodermiC 23-gauge Atkinson Fresh Cadaver Eyes 3D-gauge hypodermic 2S-gauge hypodermic 23-gauge Atkinson 2S-gauge Straus 23-gauge Straus 2S-gauge Thornton 23-gauge Thornton SD
=
standard deviation.
2± 2± 3± 12 ±
O.S 0.5 1.3 2.6
1± 1 2±1 10 ± 1 12 ± 3.8 12 ± 1
29 ± 4.5
3S ± S
than its larger caliber counterpart. Even though there is popularity for blunt-tipped, moderate-gauge needles for local orbital anesthesia, we know of no published experimental support for these hypotheses. In a double-masked test with qualified surgeons, we found the Atkinson-style bltint-bevel tip was perceived as requiring greater force to perforate the sclera of preserved globes than a hypodermic needle of similar caliber. The trained hands of the surgeons were the transducer. The sUbjective data were obtained to lend clinical significance to this study. Our objective data would be less meaningful without this subjective "clinical" confirmation. Using this information as a pilot study, we designed an objective transducer, which enabled us to accurately measure scleral perforation pressure. Instead of the perpendicular passes made in the initial part of the study, the study needles on the syringe transducer made oblique scleral perforations, much like an errant retrobulbar injection pass. Again, the Atkinson blunt-tipped needle required more force to penetrate the globe than a hypodermic needle. The perforation pressures measured for the sharp needles, even as large as 18 gauge, were very low, and would usually be undetectable to the anesthetist. In our experience, pressures of 1 to 2 mmHg were very subtle, 5 to 10 mmHg were easily felt, and 25 to 40 mmHg would often bow the needle markedly before penetrating. These findings support our sUbjective data from the first part of the experiment. Resistance from orbital tissue might add to the perforation pressure and obscure these perceptible differences. The final part of the experiment involved in situ testing of fresh cadaver globes, a situation which mimics actual preoperative injections in most respects. The flat-bevel, non-cutting Thornton needle was found to have a much higher perforation pressure than the Atkinson needle, which was again higher than the hypodermic needles. Needle caliber also became a factor in this part of the study. Both the Straus and Thornton needles were tested in the 23- and 25-gauge version with unpreserved sclera, whereas the hypodermic needle was tested in three gauges with the preserved eyes. The larger caliber Thornton needles were found to have significantly higher scleral perforation pressures than their smaller counterparts, but this difference could not be demonstrated with the other needles. A larger number of trials or a wider range of needle calibers in the hypodermic and Atkinson groups might have shown a detectable difference. Each eye was used for 12 needle passes, 2 from each needle. Originally, we believed that several 23-gauge perforations of the sclera would allow leakage, loss of intraocular pressure, and difficulty with subsequent passes due to poor scleral support. This was not the case. When the sequence of needle use was varied, the scleral perforation pressure values did not vary more than 10%. The intraocular pressure was not objectively measured, but it did not fatigue palpably from leaks through the needle holes. It would be possible to equilibrate intraocular pressure with a manometer between needle passes, but this would require a large-bore perforating catheter, further compro-
509
Ophthalmology
Volume 100, Number 4, April 1993
mising globe integrity. We believe that the reproducibility ofthese data supports its validity. Mean perforation pressures and the standard deviations leave little question about the answer to our title question. There are clear subjective and objective differences. Hypodermic needles perforate the sclera easily and effortlessly, with an often imperceptibly small pressure. The Atkinson-style needles require greater force to pass through the sclera, and thus are less likely to inadvertently enter the eye. However, the Thornton needles are so resistant to perforation pressure that the needle would often bow and markedly displace the eye before perforating the sclera on a perpendicular pass. This finding was noted on each of the cadaver eyes tested. These data make a strong case for the use of blunttipped design on all peribulbar and retrobulbar needles. Needles that pass near the globe, unless passed under direct visualization, should be blunt-tipped to minimize the risk of inadvertent perforation. Our data also support the use of needles with the Thornton tip over the Atkinson tip in retrobulbar or peribulbar anesthesia. Using an oblique angle, we were unable to pass the Thornton needle through a fresh cadaver globe in the orbit with any amount of manual force. The globe was merely deflected within the orbit without perforation. Because Thornton needles of 23 gauge require more force to perforate the globe than 25-gauge needles of the same tip design, the authors also recommend the larger caliber needle in situations where globe perforation is of greater than normal risk.
Summary The authors tested the hypothesis that the Atkinson and Thornton needles, with relatively blunt-tipped designs, would make scleral perforation more difficult than with a standard hypodermic needle. In subjective doublemasked measurement on preserved eye bank eyes, this hypothesis is true. Objective recording of perforation pressure on preserved eye bank eyes, and on fresh, in situ cadaver globes confirmed the original study. The Thornton-tipped needles required much greater force before perforating sclera than did the Atkinson-tipped needle.
510
The authors recommend the use of a large caliber, blunttipped needle for the delivery of peribulbar or retrobulbar anesthetics, especially when the globe is larger than 26 mm or lengthened by a scleral buckle.
References 1. Grizzard WS, Kirk NM, Pavan PR, et al. Perforating ocular injuries caused by anesthesia personnel. Ophthalmology 1991 ;98: 10 11-6. 2. Duker JS, Belmont JB, Benson WE, et al. Inadvertent globe perforation during retrobulbar and peribulbar anesthesia: patient characteristics, surgical management, and visual outcome. Ophthalmology 1991 ;98:519-26. 3. Hamilton RC, Gimbel HV, Javitt Je. The prevention of complications of regional anesthesia for ophthalmology. Ophthalmol Clin North Am 1990;3:111-25. 4. Cowley M, Campochiaro PA, Newman SA, Fogle JA. Retinal vascular occlusion without retrobulbar or optic nerve sheath hemorrhage after retrobulbar injection of lidocaine. Ophthalmic Surg 1988;19:859-61. 5. Knapp H. On cocaine and its use in ophthalmic and general surgery. Arch Ophthalmol 1884;13:402-48. 6. Atkinson WS. Retrobulbar injection of anesthetic within the muscle cone (cone injection). Arch OphthalmoI1936;16: 494-503. 7. Cibis PA. General discussion: opening remarks. In: Schepens CL, Regan CDJ, eds. Controversial Aspects in the Management of Retinal Detachment. Boston: Little, Brown, 1965;222. 8. Ramsay RC, Knobloch WHo Ocular perforation following retrobulbar anesthesia for retinal detachment surgery. Am J Ophthalmol 1978;86:61-4. 9. Davis DB II, Mandel MR. Posterior peribulbar anesthesia: an alternative to retrobulbar anesthesia. J Cataract Refract Surg 1986; 12: 182-4. 10. Hamilton RC, Gimbel HV, Strunin L. Regional anesthesia for 12,000 cataract extraction and intraocular lens implantation procedures. Can J Anaesth 1988;35:615-23. II. Lichter PR. Avoiding complications from local anesthesia [editorial]. Ophthalmology 1988;95:565-6. 12. Hay A, Flynn HW Jr, Hoffman 11, Rivera AH. Needle penetration of the globe during retrobulbar and peri bulbar injections. Ophthalmology 1991 ;98: 1017-24. 13. Mcaung WL, Thacker JG, Edlich RF, et al. Biomechanical performance of ophthalmic surgical needles. Ophthalmology 1992;99:232-7.
Stephen G. Waller, MD/ John Taboada, PhD,2 Patrick O'Connor, MIY Purpose: Conventional wisdom in ophthalmology is that the force required to perforate an eye during retrobulbar injection is noticeably greater with a specially designed blunt needle than with a standard hypodermic needle. A search of the literature showed no measurements of scleral perforation pressure with specific needle tips. The authors investigated this concept. Methods: A computerized search for eye perforations of the surgical records over a 5-year period at Wilford Hall United States Air Force Medical Center was conducted. A double-masked trial by experienced ophthalmologists, using preserved eye bank eyes and several commercially available needles, subjectively assessed the force required to perforate the globe. The authors designed and built a portable transducer system to objectively measure the perforation pressure of human globes with the needles. Measurements were done with preserved and unpreserved human cadaver eyes. Resuns: No globe penetrations or perforations were found in this consecutive series of over 4000 retrobulbar anesthesia procedures. A subjective difference between the hypodermic and blunt needles was detected. Objective measurements showed a significant difference between the hypodermic and blunt needles, and between types of blunt needles. The difference was present with both eye bank eyes and fresh cadaver eyes. Conclusion: Blunt-tipped needles do require greater force to penetrate the eye. The non cutting edge, blunt-tipped needles have higher scleral perforation pressures than those with cutting edges. Ophthalmology 1993;100:506-510
Over one million ophthalmic surgical procedures are performed in the United States annually. Minimizing the risk of systemic agents and prompt return to ambulatory status are desirable goals in outpatient surgery. Adequate
Originally received: June 15, 1992. Revision accepted: October 15, 1992. I Department of Ophthalmology, United States Air Force Medical Center, Lackland Air Force Base, San Antonio. 2 Ophthalmology Branch, United States Air Force School of Aerospace Medicine, Brooks Air Force Base, San Antonio. Dr. Waller is currently affiliated with the Massachusetts Eye and Ear Infirmary, Boston. The views expressed in this article are those of the authors and do not reflect the official policy of the Department of Defense or other Departments of the United States Government. Reprint requests to Stephen G. Waller, MD, Cornea Service, Massachusetts Eye and Ear Infirmary, 243 Charles St, Boston, MA 02114.
506
anesthesia and akinesia during ophthalmic surgery are often paramount to success. For these reasons, the great majority of eye surgeries are performed under local anesthesia, usually induced with a specially designed retrobulbar needle. Many retrobulbar and peribulbar blocks are administered by non ophthalmologists, possibly enhancing the risk of needle perforation during injection. I Other substances, such as corticosteroids, antibiotics, and alcohol also are injected into the orbit in selected circumstances. Retrobulbar and peri bulbar anesthesia have an important role in ophthalmology. The act of injecting substances into the orbit is associated with a number of rare but catastrophic complications. Globe perforation,2 optic nerve injection,3 brain stem anesthesia, 3 and vascular occlusive events4 all have been reported. The first retrobulbar anesthesia was used in an enucleation by Hermann Knapp 5 in 1884. The toxicity of the agent he used, cocaine, delayed acceptance of his tech-
Waller et al . Retrobulbar Needles nique. It was not until the work of Atkinson 6 in the 1930s and the development ·of the safer agent, procaine, that retrobulbar injections became common. To avoid scleral perforation, it is recommended that the beveled surface of the retrobulbar needle be aligned with and apposed to the globe. It should pass posteriorly along the orbital floor until it is behind the equator of the globe, then be directed into the muscle cone. Peribulbar injections are made after the needle passes parallel to the orbital floor or roof, carefully avoiding movement of the needle tip toward the globe. Despite these precautions, some iatrogenic scleral perforations do occur. The incidence of globe penetration or perforation during the administration of local orbital anesthesia can be estimated. Published studies from two retinal surgery practices show an approximate incidence of I in 1000. 7•8 Risk appears to be higher when the globe is longer, particularly if it is more than 26 mm on A-scan ultrasonic measurement. 2 Presumably, a disproportionate share of the eyes undergoing retinal surgery are in this category, and the true incidence may be somewhat less than 0.1 %. Other authors report no scleral perforations in 1600 peribulbar anesthetic injections,9 and one perforation in 12,000 cases of both retrobulbar and peribulbar anesthesia. lo There are three different needle tip designs in common retrobulbar and peribulbar use today. The standard hypodermic needle has a steep bevel and a cutting edge. The Atkinson (Visitec Co., Sarasota, FL) needle has a flat bevel and a cutting edge. The Straus needle (Visitec Co., Sarasota, FL) is a curved needle with an Atkinson-style tip. The Thornton (Alcon Surgical Inc., Ft. Worth, TX) needle has a flat bevel and a non-cutting edge. The use of flat-bevel ("blunt-tipped") needles, designed specifically for orbital injections, is thought to lessen the risk of inadvertent ocular penetration or perforation. II However, a number of these events occur despite the use of the blunt needles. Two recent reports of the clinical course of inadvertent globe perforations cited the use of blunt needles in 5 of 12 and 7 of 23 of the injuries. 1•12 There is also a possibility that a perforating blunt needle tip does more retinal damage than its sharper counterpart. 1 A recent article described the mechanics of various surgical needles used for passing suture efficiently through tissue. 13 However, there is little similar literature on needles used in injections. The scarcity of literature on penetration or perforation rates includes an absence of any comparison between rates using different needle tip shapes or caliber. We examined the force necessary to penetrate a human globe with the various sharp and blunt needles available on the market today. Are the needle tips different and can the difference be detected by experienced surgeons?
Methods Globe perforation occurs at a low rate among patients undergoing routine ophthalmic surgery preceded by retrobulbar or peribulbar anesthesia. To establish the mag-
nitude of this problem in our multi-specialty, residencytraining hospital practice, we searched the inpatient record files at Wilford Hall United States Air Force (USAF) Medical Center for iatrogenic scleral perforations during local anesthesia, retrospectively evaluating all ophthalmology records coded as scleral, ocular, eye, or globe penetration or perforation over a 5-year period. We included surgical cases from all subspecialities in ophthalmology. To test the hypothesis that commercially available, blunt-tipped retrobulbar needles are less likely to penetrate the globe than standard hypodermic needles, we designed a three-part experiment. The first part involved the use of preserved eye bank eyes. Ten board-certified ophthalmologists passed sharp and blunt (Atkinson tip) needles and judged which needle was easier and which was more difficult. The needles used were the 23-gauge X 38-mm (Ilh-inch) hypodermic and the Atkinson 23-gauge X 38 mm. The needle hubs were masked, and close inspection of the needle tips was not permitted, so that neither the surgeon nor an observer who recorded the results knew which needle type was being used. The sequence of needle use also was randomly varied, and each surgeon was only permitted one pass for each needle. The globes were secured with gauze and pins to a flat wooden board. Needle passes were made perpendicular to the sclera near the equator of the globe. To control for variations in scleral thickness on the same eye, each surgeon made passes in the same general area of the globe. Each ophthalmologist used three or more globes, depending on the availability of eye bank eyes. No eye or needle was used more than three times. In the second part of our experiment, we obtained objective data. An appropriate transducer (Fig 1) was custom designed and constructed for the purpose of transferring pressure data from the needle tip to an electronic recorder. The transducer is a double-tipped, water-filled syringe. Force is mechanically transmitted along the long axis of the needle to the distal end of the transducer. There a water-filled cavity is used to conduct the pressure to an electromechanical coupling with the low pressure (umbilical vein) channel of a Spacelabs (Redmond, WA) Model 906 22A portable electronic recorder, normally used to continuously monitor vital signs on neonatal patients at Wilford Hall USAF Medical Center. The arrangement was calibrated by holding it at the level of the eye. The transducer was attached to the recorder through a Medex Sure-Cal manual, pre-calibrated, 100-mmHg, square-wave generator (Medex, Hilliard, OH), which was used to verify calibration of the electronic recorder. The transducer apparatus was first used with ten preserved eye bank eyes. The globes were again fixed by wrapping them in gauze and pinning the gauze to a platform. The needle passes were made obliquely to simulate the angle of an inadvertent penetration. Scleral perforation pressure values were measured from strip chart recordings of the needle pass (Fig 2). Each needle and each globe were used twice in succession. Because the results were objectively measured, no attempt to mask the needle hub or to vary the sequence of needle use was made. In addition to the above needles, 27- and 18-gauge hypodermic
507
Ophthalmology
Volume 100, Number 4, April 1993 of the original pass. We believe that this effect did not change our results.
Results
Figure 1. Photograph of the custom-made transducer used by the authors for measuring the scleral perforation pressures of various retrobulbar needles. The needle is attached to the tip of the syringe and the waterfilled tube leading from the syringe transmits pressure information to an adjacent portable electronic recorder.
and 25-gauge X 34 mm (1 %-inch) curved needle with an Atkinson tip (Straus design) needles were tested. In the final phase of the study, unpreserved cadaver globes were tested in the orbit with the same portable pressure-monitoring system. The needle was passed in the frontal plane near the lateral canthus and perpendicular to the globe under direct visualization. Orbital passes were avoided to allow measurement of globe perforation only, unobscured by resistance from orbital tissue. Scleral perforation pressures were measured on ten undisturbed eyes from five cadavers, each within 24 hours of death. Since the hypodermic needle passes seemed to produce no detectable hypotony or fluid leak from the sclera, the 30gauge needle and a 25-gauge X Sis-inch hypodermic needle were used first. Each globe was then used twice with a varied sequence of four other needles: the Atkinson and Straus needles, and 38-mm Thornton needles of 23 and 25 gauge. No individual needle was used more than three times, but each needle type was used 20 times. Minimal leakage did occur during this part of the experiment, but second passes of the same needle measured within 10%
Figure 2. Typical recording of scleral perforation pass with the retrobulbar needle. Notice the gradual rise in pressure as the needle tip attempts to perforate the sclera. When the pressure is sufficient, the needle passes through the sclera with a precipitous drop in pressure. The peak pressure is the value interpreted as scleral preforation pressure.
508
1 ')
To ascertain the frequency of globe injuries from anesthesia injections at our large military teaching hospital, a computer search of the inpatient record files was made. We examined over 4000 charts and found no instances of scleral perforation during retrobulbar or peri bulbar injection. The injections were all performed by ophthalmologists, with a majority by residents at all stages of training. A high percentage of the injections were done with the Atkinson blunt-tipped needle, and the retrobulbar technique was used. While the possibility of occult perforation exists, there were no admissions or surgeries to repair iatrogenic detachments from injections during a prior admission. The incidence of high myopia, posterior staphyloma, and multiple needle passes during the administration of local anesthesia was not determined. The initial experimental part of our study was subjective. Ten board-certified ophthalmologists tested which of two needles, both of the same caliber and length but of different tip design, penetrated a preserved eye bank eye with less force. Of 32 trials, the sharp needle was chosen 17 of 24 times, with eight passes judged identical. This difference is statistically significant at less than 0.05 by chi-square testing. None of the surgeons had a perfect or zero score, nor had enough needle passes to acquire a learning effect. The peak amplitude for each needle pass was measured in millimeters of mercury to obtain the objective results for the second and third parts of the experiment. Figure 2 shows the gradual increase in pressure until globe penetration occurs, when the pressure drops promptly. The peak pressure was considered the scleral perforation pressure. Statistical evaluation of the objective data involved the calculation of means of the different data groups, and computerized analysis to evaluate the null hypothesis of no difference between any of the needle types examined. The F probability from a one-way analysis of variance
Waller et al . Retrobulbar Needles was less than 0.0001. Using the Tukey HSD paired comparison methods, separate sets of analyses were conducted for preserved and fresh cadaver eyes. The means and standard deviations for each needle type are shown in Table 1. The striking differences between the hypodermic data and the results from the three blunt-tipped needles are significant at 0.0001. In addition, the Atkinson and Straus needle results are different from the Thornton needle data at 0.05. The Atkinson and Straus needles were not significantly different from each other. In addition, our data allowed assessment of the effect of needle caliber on scleral perforation pressure. These data also are shown in Table 1. Differences between different caliber needles of identical tip design were seen with the Thornton needles, and this was significant at less than 0.001. The hypodermic and Straus needles did not show statistically significant differences between the various needle calibers tested. Although the variances for the Thornton needles were somewhat larger, the coefficient of variation was similar among the needle types; we therefore assumed homogeneity of variance.
Discussion The incidence of inadvertent scleral perforation during retrobulbar or peribulbar injection is known only from two reports of retinal surgery practices. We undertook a retrospective search of inpatient records at our facility, and found no perforations in more than 4000 consecutive cases of routine ophthalmologic surgery over a 5-year period. Although perforations are catastrophic, they are fortunately rare in our multi-specialty, residency-training practice situation. In conclusion, a sharp hypodermic needle tip should perforate human tissue, including the eye, more easily than a tip with a blunt design. Likewise, a small caliber needle should pass through a tissue barrier with less force
Table 1. Scleral Perforation Pressures Scleral Perforation Pressure (mmHg)
Mean ± SD Preserved Eyes 3D-gauge hypodermic 27-gauge hypodermic 18-gauge hypodermiC 23-gauge Atkinson Fresh Cadaver Eyes 3D-gauge hypodermic 2S-gauge hypodermic 23-gauge Atkinson 2S-gauge Straus 23-gauge Straus 2S-gauge Thornton 23-gauge Thornton SD
=
standard deviation.
2± 2± 3± 12 ±
O.S 0.5 1.3 2.6
1± 1 2±1 10 ± 1 12 ± 3.8 12 ± 1
29 ± 4.5
3S ± S
than its larger caliber counterpart. Even though there is popularity for blunt-tipped, moderate-gauge needles for local orbital anesthesia, we know of no published experimental support for these hypotheses. In a double-masked test with qualified surgeons, we found the Atkinson-style bltint-bevel tip was perceived as requiring greater force to perforate the sclera of preserved globes than a hypodermic needle of similar caliber. The trained hands of the surgeons were the transducer. The sUbjective data were obtained to lend clinical significance to this study. Our objective data would be less meaningful without this subjective "clinical" confirmation. Using this information as a pilot study, we designed an objective transducer, which enabled us to accurately measure scleral perforation pressure. Instead of the perpendicular passes made in the initial part of the study, the study needles on the syringe transducer made oblique scleral perforations, much like an errant retrobulbar injection pass. Again, the Atkinson blunt-tipped needle required more force to penetrate the globe than a hypodermic needle. The perforation pressures measured for the sharp needles, even as large as 18 gauge, were very low, and would usually be undetectable to the anesthetist. In our experience, pressures of 1 to 2 mmHg were very subtle, 5 to 10 mmHg were easily felt, and 25 to 40 mmHg would often bow the needle markedly before penetrating. These findings support our sUbjective data from the first part of the experiment. Resistance from orbital tissue might add to the perforation pressure and obscure these perceptible differences. The final part of the experiment involved in situ testing of fresh cadaver globes, a situation which mimics actual preoperative injections in most respects. The flat-bevel, non-cutting Thornton needle was found to have a much higher perforation pressure than the Atkinson needle, which was again higher than the hypodermic needles. Needle caliber also became a factor in this part of the study. Both the Straus and Thornton needles were tested in the 23- and 25-gauge version with unpreserved sclera, whereas the hypodermic needle was tested in three gauges with the preserved eyes. The larger caliber Thornton needles were found to have significantly higher scleral perforation pressures than their smaller counterparts, but this difference could not be demonstrated with the other needles. A larger number of trials or a wider range of needle calibers in the hypodermic and Atkinson groups might have shown a detectable difference. Each eye was used for 12 needle passes, 2 from each needle. Originally, we believed that several 23-gauge perforations of the sclera would allow leakage, loss of intraocular pressure, and difficulty with subsequent passes due to poor scleral support. This was not the case. When the sequence of needle use was varied, the scleral perforation pressure values did not vary more than 10%. The intraocular pressure was not objectively measured, but it did not fatigue palpably from leaks through the needle holes. It would be possible to equilibrate intraocular pressure with a manometer between needle passes, but this would require a large-bore perforating catheter, further compro-
509
Ophthalmology
Volume 100, Number 4, April 1993
mising globe integrity. We believe that the reproducibility ofthese data supports its validity. Mean perforation pressures and the standard deviations leave little question about the answer to our title question. There are clear subjective and objective differences. Hypodermic needles perforate the sclera easily and effortlessly, with an often imperceptibly small pressure. The Atkinson-style needles require greater force to pass through the sclera, and thus are less likely to inadvertently enter the eye. However, the Thornton needles are so resistant to perforation pressure that the needle would often bow and markedly displace the eye before perforating the sclera on a perpendicular pass. This finding was noted on each of the cadaver eyes tested. These data make a strong case for the use of blunttipped design on all peribulbar and retrobulbar needles. Needles that pass near the globe, unless passed under direct visualization, should be blunt-tipped to minimize the risk of inadvertent perforation. Our data also support the use of needles with the Thornton tip over the Atkinson tip in retrobulbar or peribulbar anesthesia. Using an oblique angle, we were unable to pass the Thornton needle through a fresh cadaver globe in the orbit with any amount of manual force. The globe was merely deflected within the orbit without perforation. Because Thornton needles of 23 gauge require more force to perforate the globe than 25-gauge needles of the same tip design, the authors also recommend the larger caliber needle in situations where globe perforation is of greater than normal risk.
Summary The authors tested the hypothesis that the Atkinson and Thornton needles, with relatively blunt-tipped designs, would make scleral perforation more difficult than with a standard hypodermic needle. In subjective doublemasked measurement on preserved eye bank eyes, this hypothesis is true. Objective recording of perforation pressure on preserved eye bank eyes, and on fresh, in situ cadaver globes confirmed the original study. The Thornton-tipped needles required much greater force before perforating sclera than did the Atkinson-tipped needle.
510
The authors recommend the use of a large caliber, blunttipped needle for the delivery of peribulbar or retrobulbar anesthetics, especially when the globe is larger than 26 mm or lengthened by a scleral buckle.
References 1. Grizzard WS, Kirk NM, Pavan PR, et al. Perforating ocular injuries caused by anesthesia personnel. Ophthalmology 1991 ;98: 10 11-6. 2. Duker JS, Belmont JB, Benson WE, et al. Inadvertent globe perforation during retrobulbar and peribulbar anesthesia: patient characteristics, surgical management, and visual outcome. Ophthalmology 1991 ;98:519-26. 3. Hamilton RC, Gimbel HV, Javitt Je. The prevention of complications of regional anesthesia for ophthalmology. Ophthalmol Clin North Am 1990;3:111-25. 4. Cowley M, Campochiaro PA, Newman SA, Fogle JA. Retinal vascular occlusion without retrobulbar or optic nerve sheath hemorrhage after retrobulbar injection of lidocaine. Ophthalmic Surg 1988;19:859-61. 5. Knapp H. On cocaine and its use in ophthalmic and general surgery. Arch Ophthalmol 1884;13:402-48. 6. Atkinson WS. Retrobulbar injection of anesthetic within the muscle cone (cone injection). Arch OphthalmoI1936;16: 494-503. 7. Cibis PA. General discussion: opening remarks. In: Schepens CL, Regan CDJ, eds. Controversial Aspects in the Management of Retinal Detachment. Boston: Little, Brown, 1965;222. 8. Ramsay RC, Knobloch WHo Ocular perforation following retrobulbar anesthesia for retinal detachment surgery. Am J Ophthalmol 1978;86:61-4. 9. Davis DB II, Mandel MR. Posterior peribulbar anesthesia: an alternative to retrobulbar anesthesia. J Cataract Refract Surg 1986; 12: 182-4. 10. Hamilton RC, Gimbel HV, Strunin L. Regional anesthesia for 12,000 cataract extraction and intraocular lens implantation procedures. Can J Anaesth 1988;35:615-23. II. Lichter PR. Avoiding complications from local anesthesia [editorial]. Ophthalmology 1988;95:565-6. 12. Hay A, Flynn HW Jr, Hoffman 11, Rivera AH. Needle penetration of the globe during retrobulbar and peri bulbar injections. Ophthalmology 1991 ;98: 1017-24. 13. Mcaung WL, Thacker JG, Edlich RF, et al. Biomechanical performance of ophthalmic surgical needles. Ophthalmology 1992;99:232-7.