ISSN NUMBER: 1938-7172
Issue 2.3

Michael A. Fiedler, PhD, CRNA

Contributing Editors:
Mary A. Golinski, PhD, CRNA
Alfred E. Lupien, PhD, CRNA

Guest Editors:
Terri M. Cahoon, MSN, CRNA
Steven R. Wooden, MS, CRNA

Assistant Editor
Jessica Floyd, BS

A Publication of Lifelong Learning, LLC © Copyright 2008

New health information becomes available constantly. While we strive to provide accurate information, factual and typographical errors may occur. The authors, editors, publisher, and Lifelong Learning, LLC is/are not responsible for any errors or omissions in the information presented. We endeavor to provide accurate information helpful in your clinical practice. Remember, though, that there is a lot of information out there and we are only presenting some of it here. Also, the comments of contributors represent their personal views, colored by their knowledge, understanding, experience, and judgment which may differ from yours. Their comments are written without knowing details of the clinical situation in which you may apply the information. In the end, your clinical decisions should be based upon your best judgment for each specific patient situation. We do not accept responsibility for clinical decisions or outcomes.

Table of Contents













Enomoto Y, Asai T, Arai T, Kamishima K, Okuda Y


Pentax-AWS, a new videolaryngoscope, is more effective than the Macintosh laryngoscope for tracheal intubation in patients with restricted neck movements: a randomized comparative study

Br J Anaesth 2008;100:544-548

Enomoto Y, Asai T, Arai T, Kamishima K, Okuda Y



Purpose            The purpose of this study was to compare the view of the glottis and the success rate of endotracheal intubation with the Pentax-AWS and a Macintosh laryngoscope in patients whose head and neck were stabilized with manual in-line traction.

Background            Manual in-line stabilization of the neck during laryngoscopy makes endotracheal intubation more difficult. Neck stabilization prevents positioning the head and neck in “sniffing” position, which aligns the oral and pharyngeal axes. These limitations may also make fiberoptic laryngoscopy and laryngeal mask placement more difficult. Video laryngoscopes are rigid devices that can be inserted in the airway of a patient during neck stabilization. They are designed to provide an improved view of the glottis when airway anatomy or an inability to place the head and neck in “sniffing” position limits the view with traditional direct laryngoscopy. Several video laryngoscopes are clinically available. The Pentax-AWS (Tokyo, Japan) includes a battery operated handset with a liquid crystal (LCD) view screen and a disposable combination laryngoscope blade and endotracheal tube guide. Some studies have shown the Pentax-AWS provided a better view of the glottis than other videolaryngoscopes. One study has shown less cervical spine movement during intubation with the Pentax-AWS than direct laryngoscopy using a Macintosh blade. While other videolaryngoscopes often provide a clear view of the glottis, advancing the endotracheal tube (ETT) into the trachea can sometimes be difficult when using them. The tracheal tube guide built into the Pentax-AWS aims the ETT at the cross hairs on the LCD view screen without the use of a stylet.

Methodology            This prospective, randomized comparison study included healthy patients scheduled for elective surgery and general anesthesia. Patients with airway pathology or risk for aspiration of gastric contents were excluded. All airways were graded according to Mallampati classification before randomization. General anesthesia was induced with propofol and maintained with sevoflurane 2%. Neuromuscular blockade with vecuronium 7 mg to 10 mg preceded intubation attempts. If facemask ventilation was judged to be inadequate the patient was withdrawn from the study.

All patients had a laryngoscopy performed with both the Pentax-AWS and either a #3 or a #4 Macintosh laryngoscope. The use of a bougie was allowed during direct laryngoscopy with the Macintosh blade. No external manipulation or pressure on the neck was used during laryngoscopy with either device. The device used for the first laryngoscopy was randomized by the flip of a coin. A 7 mm internal diameter (ID) ETT was used in women and an 8 mm ID ETT was used in men. Laryngoscopic view was graded 1-4 according to Cormack and Lehane criteria (1=full view of the entire glottis, 4=neither the glottis nor the epiglottis is visible). First, laryngoscopy was performed with either the Pentax-AWS or the Macintosh laryngoscope and the laryngoscopic view graded. Intubation was not attempted during the first laryngoscopy. Next, laryngoscopy was performed with the other laryngoscope and the laryngoscopic view graded. During the second laryngoscopy a single attempt lasting no more than two minutes was made to intubate the trachea. If the first attempt at intubation failed the other laryngoscope was used for the second attempt. If the second attempt failed data collection was terminated and the airway was managed according to usual clinical practice.

Result            The study included 203 patients, 117 men and 86 women, average age 57 (range 18-86 years). Their Mallampati classifications were I=154, II=40, III=8, and IV=1. The first laryngoscopy was performed with the Macintosh laryngoscope in 99 patients and with the Pentax-AWS in 104 patients.

Glottic view during laryngoscopy was significantly better with the Pentax-AWS than with the Macintosh (P<0.001). Using the Pentax-AWS the glottic view was always grade 1. Using the Macintosh laryngoscope the glottic view was obscured in 11% of patients (grade 3 in 21 patients and grade 4 in 1 patient). All attempts at intubation with the Pentax-AWS were successful (99 patients). Intubation with the Macintosh laryngoscope was successful in 89% of patients (93 of 104 patients; 11 intubation failures) (P=0.001). The time until the glottis was seen was shorter when the Pentax-AWS was used but the total intubation time was comparable with either laryngoscope, when intubation was successful.

Conclusion            In patients whose head and neck were stabilized with manual in-line traction, the Pentax-AWS provided a better laryngoscopic view and a higher tracheal intubation success rate than direct laryngoscopy with a Macintosh laryngoscope.



I appreciate the design of this study because the laryngoscopic view was compared with both the Pentax-AWS and traditional direct laryngoscopy on the same patient. This, combined with the manual in-line stabilization of the neck, is a pretty “real world” comparison from my perspective. Using patients as their own controls allowed for a more rigorous statistical analysis which strengthened the study.

While the investigators mentioned it almost in passing, in my mind, the key difference between the Pentax Airway Scope and other videolaryngoscopes I’ve used is the guide for the endotracheal tube built in to the disposable portion of the device. With other videolaryngoscopes, once you can see the glottis the challenge is getting the ETT to the glottis (usually at quite an angle) and then getting it to go into the glottis once you get it there. A stylet is needed to get the tube to the glottis but makes it harder to advance the tube into the glottis. The ETT guide on the Pentax Airway Scope eliminates the need for a stylet and aims the tube at a crosshair on the video screen. If the cross hair is placed over the glottis that is where the tube will go when you advance it. Not having a stylet in it the tube is more flexible and may enter the trachea with less difficulty. I noticed in photographs accompanying this article that the authors used an anode tube which is quite flexible and may be even easier to advance into the trachea than a standard ETT.

While a flexible bronchoscope is often cited as the gold standard for difficult intubation it has many drawbacks. Despite what some would say, I’m convinced that few individuals are really skilled at fiberoptic intubation, which takes continued practice. Fiberoptic bronchoscopes are quite expensive so there aren’t a lot of them around. They are difficult and time consuming to clean and easily damaged. As a result, fiberoptic bronchoscopes tend to be less available even when a skilled operator is present. The videolaryngoscopes we are beginning to see come to the market now may largely replace the need for fiberoptic bronchoscopes for the vast majority of difficult intubations. They are less expensive, smaller, self contained, and include a disposable component so little, if any, cleaning is needed after use. Best of all, they appear to be much easier to use.

Michael Fiedler, PhD, CRNA

For photographs and more information on the Pentax-AWS, go to in the anesthesia section of the medGadget web site.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

Equipment & Technology

Wee AN, Sanderson, PM


Are melodic medical equipment alarms easily learned?

Anesth Analg 2008;106:501-508

Wee AN, Sanderson, PM



Purpose            The purpose of this investigation was to determine whether teaching mnemonic lyrics and rationale for alarm melodies would improve the abilities of critical care nurses to learn and differentiate among 16 medium and high priority alarms.

Background            Alarm safety has been a national priority of the Joint Commission on Accreditation of Healthcare Organizations since 2004.  That same year, the American Association of Nurse Anesthetists and American Society of Anesthesiologists adopted recommendations for the continuous use of variable-pitch alarms for capnography and pulse oximetry.  Subsequently, the International Electrotechnical Commission (IEC) published standard 60601-1-8 establishing specifications for audible and visible medical system alarms.  The standard promotes the use of melodic alarms to differentiate medium and high priority warnings for 8 conditions: general, oxygen, ventilation, cardiovascular, temperature, infusions, perfusion, and power failure.  Unfortunately, standard 60601-1-8 retained melodic constraints of an earlier International Organization for Standardization (ISO) standard for medium and high priority alarms (ISO 9703-2), and effectiveness of the new alarm specification was not tested before implementation.  Findings of independent researchers suggest that it is difficult for users to differentiate among melodic alarms even when rationale and mnemonic lyrics for the alarms are provided; however, each of these previous studies has limitations in terms of the population studied, participant training, or alarms tested.

Methodology            Twenty-two critical care nurses were randomized into two groups.  One group was introduced to each of the 16 alarms by an investigator who sang the alarm’s mnemonic lyrics (such as “In-fu-sion a-larm”) and explained how the lyric and alarm were related.  Comparison group participants were taught alarm labels (such as ‘high priority oxygen alarm’) without explanation of the associated lyric or rationale for development of each specific alarm.  During the first day of the investigation, both groups participated in familiarization exercises followed by learn-test cycles where individual participants could listen to each alarm as desired until they were ready for testing.  The test portion of each cycle concluded when a subject either attained perfect scores on two consecutive tests, or 45 minutes of testing had elapsed.

Six to 11 days after the first training day, the nurses participated in a second day of exercises.  Subjects completed a long-term memory test similar to the alarm recognition tests of the previous session.  Then the participants completed learn-test cycles similar to the first day’s cycles, concluding when two consecutive perfect test scores were achieved, 45 minutes had elapsed, or 8 learn-test cycles were completed.  Finally, participants were asked to identify alarms while completing a simultaneous “timeshared” task involving arithmetic calculations.


Result            No participant correctly identified 100% of the alarms on two consecutive tests on Day 1 and only 1 participant (4.5%) achieved consecutive perfect scores on the second day.  By the end of Day 2, the general alarm was identified correctly 89% of the time whereas specific parameter alarms were identified correctly only 37% to 82% of the times (57% average).  Participants with musical training correctly identified twice as many alarms compared to nurses without training (73% versus 38%).  Medium priority alarms were identified more accurately than high priority alarms (58% versus 53%).  Participants in the mnemonic group were unable to identify alarms more accurately, but were less likely to persistently confuse one alarm with another.  Accuracy decreased and time to react to alarms lengthened when subjects were completing the simultaneous timeshared task.

Conclusion            The requirement for consistency in underlying rhythmic structure for all alarms, as established in a previous ISO standard, made the new alarms difficult to differentiate.  Mapping alarm melodies to clinical conditions was no more effective than if the alarm melodies were assigned randomly.  Clinical alarms should vary musically in terms of pitch, timbre, and rhythm.  Future design changes should be explored by interdisciplinary teams of experts, and then be tested by clinicians before incorporation into international standards.



The tone of Wee and Sanderson’s report suggested that the investigators were somewhat pleased to report that the new standard would not be as helpful as the IEC had intended.  After playing the alarm melodies on a piano, I began understanding the authors’ apparent frustration.  Compared with the broad potential diversity of a musical palette, many of the alarms sounded too similar.  The longer patterns of high priority alarms increased complexity rather than focusing my attention.  Next, I studied the written musical notation, again hoping to identify an underlying logical structure to the alarm sequences.  After my one relatively short study session, undistracted and in a quiet room, I could not find either a musical or analytic “key” to help me unlock the alarm riddle.  It would be reasonable to conclude that correctly identifying alarms in the operating room would be even more difficult as we balance simultaneous attention demands, change task priorities, and manage frequent task interruptions.  Such conditions not only delay alarm recognition, as demonstrated by the investigators, but also delay our ability to resume previous activities after an interruption.1

Alternatively, alarm differentiation may improve in vivo.  Aside from a select few extraordinary situations, most of us will not find ourselves managing cases where all 8 conditions (16 alarms) would be active. Intuitively, the likelihood of correctly identifying an alarm should increase as the number of potential alarm options decrease.  Secondly, clinical context will cue us toward some alarms.  For example, managing an anesthetic where the connection between endotracheal tube and anesthesia circuit is obscured by surgical drapes heightens our sensitivity for ventilation alarms.  Ultimately, this investigation is another illustration of the challenges we face when translating research into practice.  

To place the study findings in an historical context, I suspect that 50 years ago nurses would have had similar difficulties differentiating among the graph paper tracings we now easily interpret as electrocardiogram rhythm strips.  Just as visual pattern recognition has become an essential element of contemporary practice, future generations of anesthesia providers will likely be expected to distinguish complex sonic patterns.  From my perspective, it is less important whether these 22 nurses were able to learn aural patterns after one day than it would be to know whether we as humans are physically capable of reliably differentiating among these patterns.  There is at least anecdotal evidence that we have this ability.  How many of us can, or know someone who can, differentiate among hundreds of popular songs after hearing only a few opening notes?

The broader question is how to prepare for a future practice that places a higher premium on sonic pattern recognition.  In the pioneering days of aviation, potential pilots were screened by judging their reactions to being spun on piano stools or being startled by pistols fired next to their ears.2  Eventually, the aviation community recognized that there were more effective ways to identify those few with the “right stuff.” What about anesthesia providers?  Currently countless hours of entry-level education are dedicated to studying topics such as physiology and pharmacology in order to understand how our patients’ body systems are structured and respond.  But how much time do we dedicate to developing our personal perceptive abilities?  Pattern recognition is an essential element of clinical practice. I’m not so courageous as to suggest that studying musical structure is as important as understanding anatomical structures or the structure of organic compounds, but perhaps an examination such as the Miller Analogies Test®, which emphasizes pattern recognition, can help us identify the best candidates to become our next generation of nurse anesthetists. Read in isolation, Wee and Sanderson’s investigation leads us to the conclusion that current international standards for melodic alarms are not ready for “prime time,” but perhaps we are not ready either.



Alfred E. Lupien, Ph.D., CRNA

1.  Wickens CD, McCarley JS. Applied Attention Theory. Boca Raton, FL: CRC Press; 2008.

2.  Beaty D. The Naked Pilot. Shrewsbury, England: Airlife; 1995.

Computer .wav files for each of the 16 IEC standard 60601-1-8 alarms can be found at the website for the University of Queensland’s Cognitive Engineering Research Group at the following URL: Accessed April 3, 2008.

More extensive discussion of each IEC alarm, its musical characteristics and mnemonics, as well as other alarm strategies can be found at the following URL: Accessed April 3, 2008.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

Avidan MS, Zhang L, Burnside BA, Finkel KJ, Searleman AC, Selvidge JA, Saager L, Turner MS, Rao S, Bottros M, Hantler C, Jacobsohn E, Evers AS


Anesthesia awareness and the Bispectral Index

N Eng J Med 2008;358:1097-1108

Avidan MS, Zhang L, Burnside BA, Finkel KJ, Searleman AC, Selvidge JA, Saager L, Turner MS, Rao S, Bottros M, Hantler C, Jacobsohn E, Evers AS



Purpose            The purpose of this study was to determine whether patients at high risk for awareness during general anesthesia had a lower incidence of awareness when anesthesia was administered by a BIS guided protocol compared to an End Tidal Anesthetic Gas guided protocol.

Background            Awareness is rare during general anesthesia. Awareness has been estimated to occur in about 1% of patients at high risk for recall. When end tidal anesthetic gas (ETAG) concentration is 0.7 MAC or greater the likelihood of awareness is small. A number of monitors are available that purport to monitor “depth of anesthesia” and their use is said to reduce the likelihood of awareness. Most commonly, they are totally or partially based on monitoring of processed EEG activity. The most widely used is the bispectral index (BIS, Aspect Medical Systems). BIS values between 60 and 40 are recommended by the manufacturer as indicating adequate depth of general anesthesia. BIS values below 60 are supposed to make awareness unlikely. The company suggests that BIS values below 40 represent excessive anesthetic depth and that lightening the anesthetic when the BIS is below 40 will result in a savings of anesthetic drugs. A previous study showed an absolute reduction in the risk of awareness of 0.74% when anesthesia was administered according to the BIS value compared to a control group. Other studies have shown no reduction in the incidence of awareness with BIS use. Despite this lack of consensus, the BIS has gained fairly widespread clinical use.

Methodology            This prospective, randomized, single-blind study included patients at high risk for awareness during general anesthesia 18 years old and older. Awareness risk factors were divided into major criteria and minor criteria. “High risk” was defined as having at least one major risk factor or two minor risk factors. Consenting patients were divided into a BIS group and an ETAG group. In the BIS group the anesthesia provider endeavored to keep the BIS index between 60 and 40. In the ETAG group the anesthesia provider endeavored to keep the end tidal concentration of isoflurane, sevoflurane, or desflurane between 0.7 and 1.3 MAC. The use of nitrous oxide was allowed and was taken into consideration when calculating MAC. A BIS Quatro sensor was applied to the forehead of all patients. While the BIS was recorded in all patients, BIS index values were not visible to the anesthesia provider while caring for patients in the ETAG group. End tidal anesthetic gas concentrations were visible during all cases.

Awareness during general anesthesia was assessed on three occasions postoperatively with the Brice questionnaire: within 24 hours, between 24 and 72 hours, and 30 days postoperatively. After data collection was complete, three experts blinded to the group each patient was in reviewed patient responses to the questionnaire and classified each patient as having “definite” anesthesia awareness, “possible” awareness, or “no” awareness. Agreement by two of the three experts was required to classify a patient. Graphic records of BIS and ETAG (as % MAC) values were recorded during each case and examined for sustained periods of 30 seconds or longer during which BIS values were greater than 60 or ETAG values were below 0.7 MAC.

Result            A total of 2000 patients were randomized and 1941 were included in the analysis. Thirty-three patients from the BIS group and 26 patients from the ETAG group were excluded for technical reasons, canceled surgery, or use of an anesthetic technique other than general. Data was available from all three interviews for 90.4% of patients.

Awareness was “definite” in 4 patients, 2 in the BIS group and 2 in the ETAG group. The overall incidence of definite awareness was 0.21%. Awareness was “possible” in 5 patients, 4 in the BIS group and 1 in the ETAG group. The overall incidence of definite or possible awareness was 0.46% (0.62% in the BIS group and 0.31% in the ETAG group). There were no significant differences between patients with and without awareness in regards to patient characteristics or anesthetic drugs administered.

Definite awareness was experienced by 2 BIS patients. In one patient the BIS value was below 40 and end tidal anesthetic concentration was at least 0.7 MAC throughout the case. In the other patient, the BIS was above 60 for about 5 minutes out of the approximate 51 minutes during which awareness was documented. The end tidal anesthetic concentration was 0.7 MAC or less (as low as 0.1 MAC) during the period of awareness. Possible awareness was experienced by 4 BIS patients. In 3 of these patients BIS values were 40 or less throughout the case. MAC values varied widely during the periods of awareness. In the 4th possible awareness patient, awareness was present for the vast majority of the case. Both BIS and end tidal anesthetic values varied widely throughout.

Definite awareness was experienced by 2 ETAG patients. In both patients the BIS values (recorded but unseen during the case) were below 60 throughout the case. One of the patients had BIS values well below 40 for most of the case. End tidal anesthetic concentration was less than 0.7 MAC during part of the period of awareness in one patient and at 0.7 MAC in the other. Possible awareness was experienced by 1 ETAG patient. BIS values were 60 or less and, except for a period of about 5 minutes of 0.6 MAC, ETAG concentration was 0.7 MAC throughout the period of awareness.

The mean (±sd) time-averaged end tidal anesthetic concentration during maintenance of anesthesia was 0.81±0.25 MAC in the BIS group and 0.82±0.23 MAC in the ETAG group (P=0.10).

BIS values were sustained above 60 in 55% of patients who did not experience awareness. Likewise, end tidal anesthetic concentrations were less than 0.7 MAC in 75% of patients who did not experience awareness.

Conclusion            Compared to administering general anesthesia based upon an end tidal anesthetic gas protocol, a protocol based upon BIS values neither reduced the incidence of awareness nor the amount of potent inhalation agent administered. Awareness during general anesthesia cannot be predicted in all patients using the BIS protocol in this study. Reliance on BIS values may give a false sense of security.



This report in the New England Journal of Medicine is perhaps the most prestigious study to date on the topic of consciousness monitoring and awareness. While the study was generally well executed and deserves our attention, several areas warrant a comment.

There are a couple things I wish had been done differently. First, while BIS values were not visible to the anesthesia provider during a case being conducted by the End Tidal Anesthetic Gas (ETAG) protocol, End Tidal Anesthetic concentrations were visible to the anesthesia provider during a case being conducted by the BIS protocol. This would allow the anesthetist to keep the BIS below 60 and the ETAG concentration above 0.7 MAC during BIS cases and may have biased the study toward deeper anesthesia in the BIS group than if the ETAG concentration had not been visible. Next, a one-tailed statistical test was used to compare the incidence of awareness between groups. A one-tailed test is rarely used and assumes that the investigators know that a value can only move in one direction. In this case the test assumes the incidence of awareness can only go down in the BIS group. I don’t believe there is enough information to make this assumption. The one-tailed test would more easily find a significant reduction in the incidence of awareness when using the BIS. Both these drawbacks make it more likely that the study would find that the incidence of awareness was lower when the BIS was used, but that was not the case.

Other aspects of the study are noteworthy for how carefully they were handled. The investigators assessed patients for awareness at three different time periods postoperatively. The three time periods used have previously been shown to give the greatest chance of finding awareness if it occurred. Studies that are serious about detecting awareness use this same method. And the results of the awareness assessments were reviewed by three experts. Only when there was agreement could a result be classified as definite, possible, or no awareness.

I’m grateful that manufacturers are spending research and development money trying to develop a consciousness monitor. Having said that, I’m not convinced that currently available devices, such as the BIS, reliably measure the level of consciousness or depth of anesthesia, let alone the likelihood of awareness. In this study, there were four incidences of definite awareness during general anesthesia. In three of them the BIS numbers were well below 60, and often below 40, during awareness. In the fourth, the BIS number was somewhat above 60 for perhaps 10 minutes out of a more than hour long period of awareness. If one includes cases of both definite and possible awareness than the BIS numbers were below 60 in 6 of 9 cases of awareness.

Awareness is hard to study in part because it occurs infrequently. Despite being rare, when awareness occurs it can be traumatic for patients. Time and effort spent to make awareness even more rare serves our patients well. Awareness occurs in part due to anesthetic mistakes, in part because there are times when preserving life and brain function come at the cost of awareness, and in part for reasons we don’t understand. Studies, such as this one, that I’ve read to date fail to convince me that consciousness monitors can help me prevent awareness. For now, I think our efforts to prevent awareness are better spent in assuring adequate depth of anesthesia with a potent inhalation agent and the use of amnestic drugs.

Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008


Young B



intraoperative detection of methemoglobinemia in a patient given benzocaine spray to relieve discomfort from a nasogastric tube: a case report

AANA J 2008;76:99-102

Young B




Purpose            This case report concerns the identification of methemoglobinemia in a patient who was using benzocaine spray to relieve nasogastric tube discomfort, and the impact of the methemoglobinemia on intraoperative anesthetic management.

Background            Methemoglobinemia is a rare cause of oxygen desaturation. It can be caused by intrinsic congenital deficiencies of enzymes that normally convert methemoglobin to hemoglobin, or from chemical and mechanical alterations associated with drugs or cardiopulmonary bypass. Although rare, it should be included in the differential diagnosis of patients with low saturation not attributed to other causes.

Methodology            This case involved a 27 year old man who had traumatic injuries associated with a fall from a ladder. He underwent exploratory laparotomy to repair a suspected bladder rupture. Prior to the surgical procedure he was hospitalized and had a nasogastric tube in place. He was allowed to use a benzocaine spray PRN to relieve the discomfort of the tube. On the evening of surgery his blood pressure was 130/98, pulse 100, hemoglobin 13.7, skin was gray, and the patient had low oxygen saturation (90%) while on 100% oxygen by non-rebreather. During induction, the endotracheal tube was confirmed to be in the trachea, there were bilateral breath sounds, and the patient’s vital signs were stable with the exception of a persistently low oxygen saturation of 91%. The surgeon noted that his “blood was brown.” Blood gases revealed a pH of 7.31, PCO2 of 40 mmHg, PO2 of 533 mmHg, and a calculated saturation of 100%. Because the blood gas results were inconsistent with the “brown blood” and low pulse oximeter value, a spectrophotometer analysis was done. Methemoglobin levels were significantly elevated. The patient was treated with IV methylene blue and within 30 minutes his oxygen saturation was 99% and his heart rate was 80/min. The blood in the surgical field appeared normal and the patient’s recovery was uneventful.

Result            The patient’s problem with methemoglobin was likely the results of a liberal use of benzocaine. Hemoglobin can only transport oxygen when the iron on the hemoglobin is in the reduced state (Fe2+). Methemoglobin, or Fe3+, can not transport oxygen in spite of adequate oxygen tension in the blood. Small amounts of methemoglobin normally exist in the blood. Patients with congenital enzyme deficiencies that inhibit the conversion of Fe3+ to Fe2+ have lifelong cyanosis. Benzocaine acts as an indirect oxidant converting hemoglobin from Fe2+ to Fe3+. Benzocaine does have a short half life and is usually cleared within 24-72 hrs, but excessive use can contribute to methemoglobin with clinically significant consequences which could potentially include death. Contributing factors can include preexisting cardiovascular and respiratory problems as well as hereditary lack of the ability to reduce methemoglobin. Inflamed tissue in the area of benzocaine use can also contribute to an increased systemic absorption of the drug.

If methemoglobinemia is suspected, supportive treatment should be initiated and immediate action should be taken to restore the oxygen carrying capacity of the blood. Treatment includes the administration of methylene blue, 1-2 mg/kg. Adequate blood glucose is necessary for methylene blue to be effective, so dextrose should be administered as well. Diagnosis of chemically induced methemoglobinemia is definitive if this treatment results in oxygen saturation returning to normal. A poor response to treatment should arouse suspicions of enzyme deficiency which might require treatment with transfusion and/or dialysis.

Conclusion            There are several agents used by anesthesia providers that can cause methemoglobinemia. They include prilocaine, lidocaine, benzocaine, metoclorpramide, nitric oxide, nitroglycerin, and nitroprusside. The actions of cardiopulmonary bypass can also be a cause of the problem. Methemoglobinemia should be part of the differential diagnosis especially when the clinical picture is inconsistent with the finding of a persistently low pulse oximeter reading. The offending agent should be removed, treatment initiated, and adequate oxygen delivery established.



I find this case study very interesting because I believe few health care providers are aware of the potential impact benzocaine use can have on the oxygen carrying capacity of the blood. Liberal use of benzocaine is common for patients with nasogastric tube irritation, used prior to upper endoscopy procedures, and other situations of oral and nasal irritation. This is just the kind of situation that promotes absorption of benzocaine and increases the potential for methemoglobinemia.

This case also reveals the multitude of issues surrounding surgical patients that make it difficult to diagnose a potentially life threatening situation. A patient with trauma who might have undiagnosed pulmonary problems, a mechanical airway in place that could be obstructed or displaced, or other issues unrelated to the obvious make such a diagnosis very difficult indeed. It is possible and often likely that benzocaine is used without the knowledge of the anesthesia provider and without an understanding of its potential problems. Sometimes when you least expect it, the most unusual problem will present itself and only those who constantly challenge themselves, read about unusual but real clinical situations, and stay on top of the most current developments in anesthesia care are most likely to readily recognize and resolve the problem. Keep reading!


Steven R. Wooden, MS, CRNA



© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008


Heid F, M?ller N, Piepho T, B?res M, Giesa M, Drees P, R?melin A, Werner C


Postoperative analgesic efficacy of peripheral levobupivacaine and ropivacaine: a prospective, randomized double-blind trial in patients after total knee arthroplasty

Anesth Analg 2008;106:1559-1561

Heid F, Müller N, Piepho T, Bäres M, Giesa M, Drees P, Rümelin A, Werner C



Purpose            The purpose of this study was to compare the analgesic efficacy of patient controlled femoral nerve analgesia with either levobupivacaine 0.125% or ropivacaine 0.2% for three days following total knee replacement.

Background            Analgesia following total knee replacement is important not only for patient comfort, but also to allow the joint mobility crucial to a successful surgical outcome. To avoid local anesthetic toxicity, knowledge of the potency of local anesthetics used for specific blocks is needed.

Methodology            This prospective, randomized, double-blind study included patients undergoing total knee replacement. Patients were randomized to receive patient controlled regional analgesia (PCRA) of the femoral nerve postoperatively with either levobupivacaine or ropivacaine. Preoperatively, the femoral nerve of all patients was identified with nerve stimulation at 0.4 mA or less using an insulated needle. A total of 35 mL of levobupivacaine 0.3125% or ropivacaine 0.5% was injected incrementally. Next, a femoral nerve catheter was inserted 3-4 cm. Following the placement of a femoral nerve catheter, a single shot sciatic nerve block was performed by the posterior approach and 25 mL of the same local anesthetic was injected. A total of 60 mL of local anesthetic solution was used for the initial blocks.

All patients received a general anesthetic with propofol 2 mg/kg and sufentanil 10 µg. The airway was managed with either an LMA or an ETT. If an ETT was used atracurium 0.5 mg/kg was administered prior to intubation. Anesthesia was maintained with sevoflurane 1.0 to 1.5% (end tidal concentration) in oxygen and air. Supplemental sufentanil or atracurium was administered if required. Before emergence, acetaminophen was administered to all patients. No local anesthetic was administered during the surgical procedure.

Upon arrival in the post anesthesia care unit (PACU) the femoral nerve catheter was connected to a patient controlled regional anesthesia pump set as follows: continuous infusion 5 mL/h, bolus dose 5 mL, lockout 30 minutes. The pump contained either levobupivacaine 0.125% or ropivacaine 0.2% consistent with the local anesthetic used for the initial blocks. Data was collected upon admission to the PACU (T0), and at 24h (T1), 48h (T2), and 72h (T3) postoperatively. An opioid was available as a rescue medication.

Result            The total volume of local anesthetic used for the initial block and PCRA, and the number of bolus doses was similar between groups (P=0.98). There was no significant difference in pain scores, either at rest or with joint movement, at any of the time points (PACU admission to 72 hours postoperatively). All pain scores were below 25 on a 0-100 numeric rating scale during rest. All pain scores were between 22 and 31 during joint movement. One patient in the levobupivacaine group was prevented from walking by motor block. All ropivacaine patients walked. Patient satisfaction was high and not different between groups.

Conclusion            Levobupivacaine 0.125% and ropivacaine 0.2% provided equally effective femoral nerve analgesia following total knee replacement.



This study would have benefited from a little better focus. In one place the investigators’ stated purpose was to compare “postoperative analgesic efficacy,” in another they identify the study’s “primary end point” as “postoperative local anesthetic consumption” (but fail to specify whether this consumption was in mL or mg), and in yet another they talk about the importance of understanding the potency of the respective local anesthetics in order to use the lowest dose and avoid local anesthetic toxicity.

Both local anesthetic techniques appear to have provided exceptional postoperative analgesia. Pain scores were low, even with joint mobility, and patient satisfaction was high. While the mg dose of ropivacaine used was greater, the difference was almost exactly proportional to the difference in concentration of the two local anesthetics. This suggests the investigators did a good job selecting concentrations of the two local anesthetics there were equally potent. This is a good thing, because the study was not designed to find equal potent concentrations of levobupivacaine and ropivacaine. It was also not designed to find the lowest dose of local anesthetic that could be used so as to minimize the risk of local anesthetic toxicity. While the investigators compared the analgesic potency of the two different local anesthetic concentrations, analgesic potency and toxicity are not the same thing. A significant drawback of this study is that no information about plasma concentrations or signs of local anesthetic toxicity were reported.

On a mg per mg basis, levobupivacaine is less toxic than bupivacaine by several measures. There are, however, some measures of cardiac performance for which levobupivacaine has similar or even greater toxicity than bupivacaine (LVEDP, dP/dt, and fiber shortening). Ropivacaine is less toxic than bupivacaine and levobupivacaine by all measures. Ropivacaine has a higher margin of safety between the dose that results in CNS toxicity and the dose that results in death. Ropivacaine also has a significantly shorter elimination half life than levobupivacaine (1.9 hours vs. 3.3 hours) reducing the chance of systemic absorption toxicity during a long term infusion. In this study, an average 68% greater mg dose of ropivacaine than levobupivacaine was used. For the initial femoral and sciatic nerve blocks an average of 2.3 mg/kg levobupivacaine and 4 mg/kg ropivacaine was used. We don’t really know if the greater dose of ropivacaine was still less toxic than the smaller dose of levobupivacaine. We also don’t know if a lower dose of one or both local anesthetics would have produced equally satisfactory pain relief. All we know for sure is that these two specific doses of two different local anesthetics resulted in great pain relief after total knee replacement.

If nothing else, this study shows us a way to produce great pain relief in total knee patients. I’d feel better, though, if we knew more about how close these doses were to toxic levels, especially in the elderly population who most commonly undergo knee replacement and who have lower levels of plasma proteins to which local anesthetics are bound.


Michael Fiedler, PhD, CRNA


Local anesthetic pharmacology and toxicity information primarily from Hadzic A. Textbook of regional anesthesia and acute pain management. New York: McGraw Hill; 2007.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

Pediatric Anesthesia

Rait RA, Voepel-Lewis T, Burke C, Kostrzewa A, Lewis I


Incidence and risk factors for perioperative adverse respiratory events in children who are obese

Anesthesiology 2008;108:375-380

Rait RA, Voepel-Lewis T, Burke C, Kostrzewa A, Lewis I



Purpose            This study evaluated the risk factors and incidence of perioperative morbidity associated with childhood obesity.

Background            Childhood obesity has increased dramatically over the last three decades. It is estimated that 15.5% of children and adolescents are obese. The definition of obesity is well established for adults, but no such criterion is universally accepted for children. There have been recent attempts to correlate body mass index (BMI) data for children to accepted adult norms with some success. Obesity is associated with hypertension, asthma, sleep apnea, and eventually an increased risk of heart disease in adulthood. Studies of adult obesity indicate an increase risk of aspiration and postoperative respiratory complications. It is unknown whether or not these same risks occur with obese children. The goals of this study were to determine the prevalence of childhood obesity, examine co-morbidities, measure perioperative respiratory events, and identify risk factors.

Methodology            This prospective, non-randomized, and non-blinded study included children 2-18 years old undergoing elective surgery, with the exception of cardiac surgery. Obese children were compared to a random sample of normal and overweight children. Preoperative height and weight were used to determine BMI (BMI = kg/m2). Combined international data adjusting for sex and age was used to categorize obese children. Adverse events were recorded prospectively at critical times including securing the airway, anesthesia maintenance, emergence, and recovery. The anesthesia providers were also asked if they had to modify their anesthesia plan because of the patient’s weight. Aspiration, nausea and vomiting, and any unplanned admissions were also noted. A previous study indicated that the incidence of adverse respiratory events in children was 24.2% which lead the researchers to determine the minimum sample size for this study to be 1,181 normal weight and 294 obese children.

Result            Data was collected from 2,025 children over a twenty month period. Of those, 14.5% were classified as obese, 17.3% overweight, and 68.1% normal. Overweight and normal groups were combined because they responded in a similar manner. The obese children were significantly older than the normal children as a group and had more coexisting conditions including asthma (28% vs 17%), hypertension (2.4% vs 1.1%), sleep apnea (14% vs 4.4%), gastric reflux (16% vs 9%), and type II diabetes (3% vs 0.1%). Obese children presented with more difficult airways on initial assessment, but actual management of the airways was not significantly different between the two groups. Anesthesia providers caring for obese patients reported changing their anesthesia plan 12% of the time because their patient was obese, but changing the anesthesia plan did not decrease the incidence of adverse respiratory events when compared to plans that were not changed once confronted with an obese patient.

Obese children had a higher incidence of adverse respiratory events when compared to the normal population. Significant complications included difficulty with mask ventilation (9% vs 2%), airway obstruction (19% vs 11%), bronchospasm (6% vs 2%), and oxygen desaturation (17% vs 9%). When comparing other complications, there were no significant differences between the obese and normal group with the exception of nausea and vomiting at home. The normal group had more nausea and vomiting than the obese group (17% vs 5%). In addition, seven children vomited during induction, only one of which was classified as obese.

Conclusion            Obese children were more likely to present with preexisting medical conditions than their normal weight counterparts. Childhood obesity and a history of obstructive sleep apnea were independent predictors of perioperative adverse respiratory events which included difficult mask ventilation, airway obstruction, bronchospasm, and oxygen desaturation. Other studies have shown obese children to have increased risk of intraoperative oxygen desaturation, unexpected hospitalization, difficult mask ventilation, difficult laryngoscopy, and postoperative airway obstruction.



The term “a priori” was used throughout this study. A priori truths are conceptual truths, and that in itself describes this study very well. The study is full of concepts, observations, and assumptions, but very little scientifically based evidence. However, this does not negate the importance of many of the observations stated in this study. It is well recognized that childhood obesity is an epidemic and those of us who anesthetize obese children on a regular basis have observed the same issues found in this study. Multiple coexisting medical conditions, difficult airways, difficult dosing of anesthetic agents, positioning issues, and other significant findings make the anesthetic management of obese children a bit challenging to say the least. The most significant scientific evidence presented in the study was the fact that obese children who also have obstructive sleep apnea are at higher risk for perioperative respiratory complications.

Of interest was the finding of increased nausea and vomiting in the non-obese group. I would not have expected this, and will pay attention to see if this occurs in my practice. With the exception of the nausea and vomiting, I am not sure that this study revealed any surprises, and it certainly did not suggest any solutions, but it did remind us of the problems we face and the importance of further studies in the area of anesthesia management of obese children. I personally will be looking forward to additional research into childhood obesity which will hopefully identify anesthesia management techniques as well.


Steven R. Wooden, MS, CRNA



© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

Flick RP, Wilder RT, Pieper SF, Vankoeverden K, Ellison KM, Marienau MES, Hanson AC, Schroeder DR, Sprung J


Risk factors for laryngospasm in children during general anesthesia

Pediatric Anesthesia 2008;18:289-296

Flick RP, Wilder RT, Pieper SF, Vankoeverden K, Ellison KM, Marienau MES, Hanson AC, Schroeder DR, Sprung J



Purpose            The purpose of this study was to describe the association between patient, demographic, and surgical factors in 130 cases of documented perioperative laryngospasm.

Background            Adverse respiratory events are common in pediatric anesthesia. Laryngospasm is one of the most common of these adverse respiratory events. The incidence of laryngospasm has been described as 14% in children < 6 years old and 3.6% in children > 6 years old. Previously identified risks for laryngospasm in children include upper respiratory infection, the type of airway device used, and a higher ASA physical status classification. History of a recent upper respiratory infection coupled with use of an LMA during surgery has been shown to more than double the risk of a laryngospasm. Use of an endotracheal tube in children with a recent upper respiratory infection increased the risk of laryngospasm 11 times. A robust method of identifying risk factors is multivariate analysis but only one study has included sufficient numbers of patients to generate information about risk factors for pediatric laryngospasm.

Methodology            This retrospective, case-control, descriptive study examined the records of pediatric patients included in the “Anesthesia Performance Improvement Database” at a large teaching medical center who’s records were coded for “laryngospasm” and contained documentation of the laryngospasm event. Consent was required for entry into this database so it may not have included all patients. Laryngospasm patients were compared to patients who had the same type of anesthesia (all general), type of surgery in the same calendar year, and age ±3 months.

Several groups of factors were examined for association with laryngospasm. Patient factors included gender, ASA physical status, elective vs. emergency surgery, medications being taken, and inpatient vs. outpatient status. Co-morbidities included congenital heart disease, recent upper respiratory infection, asthma, other lung disease, airway abnormalities, neurologic disease, a history of gastroesophageal reflux, and the presence of a variety of syndromes. Surgical and anesthesia factors included inhalation vs. intravenous induction; halothane, sevoflurane, or isoflurane for maintenance of anesthesia; use of nitrous oxide, and use of muscle relaxants. Airway management was with an endotracheal tube, an LMA, or a traditional face mask. The timing of laryngospasm was categorized as induction, maintenance, or emergence. Odds ratios (OR) with 95% confidence intervals were used to describe the magnitude of the risk of laryngospasm for factors that were positively correlated with laryngospasm.

Result            One hundred thirty cases of laryngospasm were identified in the database over a 10 year period. Of those, 46% (60 cases) occurred during induction, 44% (57 cases) during emergence, and 10% (13 cases) during maintenance of anesthesia. Fifty-seven percent of laryngospasms occurred in children ≤3 years old and 68% of laryngospasms occurred in children ≤5 years old.

Recent upper airway infection (OR = 2.03, P=0.022) and airway abnormalities (OR 3.35, P=0.030) were significant risk factors for laryngospasm. When these risk factors were adjusted for, the type of airway management used during maintenance of anesthesia was a significant risk factor for laryngospasm (P=0.013). The greatest risk of laryngospasm was in those who’s airway was managed with an LMA (OR 8.8, 95% CI 1.3 – 59.8) compared to the traditional facemask group.

Postoperative complications also differed between laryngospasm and non-laryngospasm patients. One or more complications were experienced by 7.7% (10 of 130) laryngospasm patients vs. only 1.2% (3 of 250) control non-laryngospasm patients (P=0.008). Complications in the laryngospasm group included unplanned admission, unplanned postoperative ventilation, aspiration, and unplanned ICU admission. In the non-laryngospasm group only unplanned ICU admission occurred.

Conclusion            Upper respiratory infection and managing the airway with an LMA were risk factors for laryngospasm. Children who experienced a laryngospasm had a higher incidence of postoperative complications than those who did not experience a laryngospasm.



Only 130 cases of laryngospasm over a ten year period? This is one factor of this study that diverts my attention from the evidence that it presents. I suspect that the sample might be flawed due to the required documentation on the chart as well as inclusion in the database. Do practitioners routinely chart the occurrence of an uncomplicated laryngospasm or simply treat it and proceed with the case? In comparison, a prospective study examining respiratory adverse events in pediatric patients had 128 out of 318 patients experience a laryngospasm as verified by an observer.1 This percentage of 40% is significantly higher than the 3.6 – 14% reported in the literature. However, the reports in the literature are mostly recorded retrospectively. Again, what is the accuracy of self-reporting and discovery? If the Flick study does not capture the true sample, then the findings are weakened. Based on my personal experience, I find the very low incidence of only 13 laryngospasms per year suspect or unrealistic.

The second factor that is distracting to me is the correlation of LMA use with laryngospasm. Given the time period of this study, 1996-2006, the use of the LMA may have been a relatively new technique being employed. This aspect was not discussed in this article, but it may have been a factor that should be investigated. The finding that use of an LMA increases risks when compared to traditional facemask supports previous studies. I would be interested in the specifics of the LMA technique, like removal when the patient is deep or awake. Homer and associates found that the incidence of laryngospasm when the LMA was removed with the patient awake (6.1%) was not significantly different from mask technique (4.4%).1

The third aspect of this study that causes some suspicion of bias is the fact that patients who experience laryngospasm have a higher incidence of postoperative complications. Is the fact that they had a complication the reason that the laryngospasm was documented? Or, if a patient had an uncomplicated laryngospasm, would it likely be omitted from the record and, thus, not included in this retrospective study?

Upper respiratory infections (URI) have been associated with adverse respiratory events including laryngospasm both retrospectively and prospectively in the literature. Given that children average six URIs per year2, we should consider the risk of laryngospasm and postoperative complications and employ measures to prevent adverse events including postponing the anesthetic.


Terri M. Cahoon, MSN, CRNA

1. Homer JR, Elwood T, Peterson D, Rampersad S. Risk factors for adverse events in children with colds emerging from anesthesia: a logistic regression. Pediatric Anesthesia 2007; 17:154-161.

2. Monto AS, Ullman BM. Acute respiratory illness in an American community. The Tecumseh study. JAMA 1974; 227:164-169.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008


Cammu G, De Kam PJ, Demeyer I, Decoopman M, Peeters PAM, Smeets JMW, Foubert L


Safety and tolerability of single intravenous doses of sugammadex administered simultaneously with rocuronium or vecuronium in healthy volunteers

Br J Anaesth 2008;100:373-379

Cammu G, De Kam PJ, Demeyer I, Decoopman M, Peeters PAM, Smeets JMW, Foubert L


Purpose            The primary purpose of this study was to serve as a pilot study of the safety and tolerability of sugammadex administered simultaneously with rocuronium or vecuronium in both anesthetized and non-anesthetized patients. A secondary purpose was to describe the plasma pharmacokinetic profiles of sugammadex compared to those of rocuronium and vecuronium.

Background            Sugammadex is different than current drugs used to antagonize nondepolarizing neuromuscular blockade. Rather than increasing acetylcholine at the neuromuscular junction it encapsulates steroidal relaxants; specifically rocuronium and vecuronium. Sugammadex is not metabolized, but excreted renally. It has not been associated with cardiovascular side effects and the co-administration of an antimuscarinic is unnecessary. Nevertheless, part of the drug investigation and approval process involves testing for prolongation of the QT interval. QT prolongation predisposes the heart to arrhythmias, including torsade de pointes, and may result in ventricular fibrillation and sudden death. To eliminate the confounding effects of anesthetics testing must also be done in non-anesthetized volunteers. Prior to such a study, a smaller, pilot study, is conducted.

Methodology            This was a phase I, non-blinded, descriptive study. All subjects were healthy volunteers aged 18 years to 45 years with a body mass index of 18 to 30 kg/m2. Exclusion criteria included: pregnant or breast-feeding women, women who might be pregnant, those with a history of difficult intubation or an assessment suggesting difficult intubation, and those with a history of alcohol or drug abuse.

There were three study groups. Group I subjects received sugammadex 32 mg/kg, 20 mg/kg, or 16 mg/kg (two subjects each) along with rocuronium 1.2 mg/kg. Group II subjects received the same sugammadex doses along with vecuronium 0.1 mg/kg. Neuromuscular block was monitored at the ulnar nerve throughout the study with an acceleromyograph (TOF-Watch SX, Organon Ireland Ltd, Dublin, Ireland). A T4 to T1 ratio (TOF ratio) of <0.9 was defined as evidence of neuromuscular blockade. Before sugammadex administration all Group I and II subjects were anesthetized with remifentanil and propofol. A laryngeal mask airway (LMA) was inserted and subjects were ventilated with air and oxygen. Sugammadex and the nondepolarizing muscle relaxant were administered simultaneously over four minutes via two different IVs so the drugs would not mix in the IV line. All anesthetics were continued for a minimum of two hours to allow time for the muscle relaxant to wear off. To further avoid awakening a partially paralyzed subject, at the end of the two hour anesthetic emergence was not begun until the TOF ratio was >0.9 for 10 minutes or more.

The study procedure was begun with Group III subjects only after information about the level of neuromuscular blockade was derived from groups I and II. Group III subjects were not anesthetized. Prior to administration of the study drugs, group III subjects underwent a baseline assessment of neuromuscular function. Subjects were assessed in the following areas: ability to smile, swallow, and speak; 5 second sustained head lift; hand grip strength; and sustained tongue depressor test. Group III subjects then received sugammadex 32 mg/kg along with either 1.2 mg/kg rocuronium or 0.1 mg/kg vecuronium. Neuromuscular function was assessed regularly thereafter.

Blood was drawn from subjects in all three groups to measure plasma concentrations of each drug administered at six points throughout the study.

Subjects of groups I, II, and III were monitored for eight hours after study drug administration.

Result            Sixteen subjects were enrolled and completed the study; 10 men and 6 women. They ranged in age from 18 years to 43 years. Six each were in groups I and II and four were in group III. There was no evidence of neuromuscular blockade in any subject during the study period. Twelve lead ECGs were normal in all subjects. The plasma concentrations of rocuronium and vecuronium decreased more rapidly than the plasma concentration of sugammadex. Adverse events reported by study subjects included nausea (5 subjects) and moderate injection site irritation (13 subjects). Aspartate aminotranferase (AST) and gamma-glutamyltransferase were mildly increased in nine subjects each. The increase was clinically significant in only one subject.

Conclusion            A single intravenous dose of up to 32 mg/kg of sugammadex along with rocuronium or vecuronium was associated with no neuromuscular block and only minimal side effects in healthy volunteers. No changes in the QT interval were observed.



I have been following early reports of sugammadex with interest. These early reports make sugammadex sound incredible. It can completely antagonize rocuronium or vecuronium before they even start to cause paralysis. That means I can use rocuronium to intubate even if the case will only last 10 minutes! It has no side effects even if you give way too much of it (see the June 2007 issue of Anesthesia Abstracts). Wow!

But wait, history urges that I restrain my exuberance. When sufentanil was released the early studies misjudged how potent it was and we had recovery rooms full of ventilated patients for a while. When midazolam was released the early studies misjudged its potency and duration. For anesthesia providers this resulted in some patients who became apneic in the holding area. For non-anesthesia providers less adept at managing the airway this resulted in some deaths. And then there was Raplon (rapacuronium), the muscle relaxant that came and went in the blink of an eye. It made it through all the early studies with only a hint of pulmonary problems (in hindsight), was released into clinical practice, and was quickly removed from practice by the manufacturer. A number of patients developed significant bronchospasm after Raplon administration; some of them died.

I am not suggesting that sugammadex will be a disappointment nor that it is dangerous. I hope it is as good as it looks. But we must always remember that, while an important start, drug development and FDA approval studies do have limitations. Once a drug is released for clinical practice it is subject to a much larger number and range of patients. It is used in an almost infinite variety of clinical situations. As we begin using sugammadex lessons from previous “new drug” releases suggest that we are the final clinician / investigators. I think we will do our patients the most service if we approach sugammadex cautiously and learn everything we can from “real world” use. Experience has shown that there is much more to know about a new drug than what is known when it is released.

Sugammadex may fundamentally change our concept of muscle relaxant reversal and, with it, change our practice. It is up to us to use it safely.


Michael Fiedler, PhD, CRNA

For other abstracts and comments about sugammadex see the following issues of Anesthesia Abstracts:

• Volume 1 Number 3, June 30, 2007. Reversal of rocuronium-induced (1.2 mg/kg) profound neuromuscular block by accidental high dose of sugammadex (40 mg/kg).

• Volume 1 Number 5, August 31, 2007. Sugammadex reversal of rocuronium-induced neuromuscular blockade:  a comparison with neostigmine-glycopyrrolate and edrophonium-atropine.

• Volume 1 Number 7, November 30, 2007. A randomized, dose-finding, phase II study of the selective relaxant binding drug, sugammadex, capable of safely reversing profound rocuronium-induced neuromuscular block.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

Canbay O, Celebi N, Sahin A, Celiker V, Ozgen S, Aypar U


Ketamine gargle for attenuating postoperative sore throat

Br J Anaesth 2008;100:490-493

Canbay O, Celebi N, Sahin A, Celiker V, Ozgen S, Aypar U


Purpose            The purpose of this study was to compare gargling with ketamine or placebo for the prevention of postoperative sore throat.

Background            The incidence of postoperative sore throat has been reported to be as high as 65% following general anesthesia. A variety of non-pharmacologic interventions have been shown to reduce the incidence or severity of sore throat including smaller endotracheal tubes (ETT), lubricating the ETT with a water soluble lubricant, minimizing ETT cuff pressure, intubating after complete paralysis, atraumatic intubation, and atraumatic airway suctioning. Azulene sulphonate (Azunol) gargle and beclomethasone inhalation have also been used to reduce postoperative sore throat.

N-methyl-d-aspartate (NMDA) receptors are found in peripheral, as well as central, nerves. Peripheral NMDA receptor blockade produces antinociceptive and anti-inflammatory effects.

Methodology            This prospective, randomized, placebo controlled study included healthy patients undergoing elective septorhinoplasty with general anesthesia. All surgeries were performed by the same surgeon. Patients with a history of sore throat, asthma, an airway assessed to be Mallampati class III or greater, those taking NSAIDs, and those who required more than one attempt at intubation were excluded from the study.

Five minutes before induction of general anesthesia the control group gargled for 30 seconds with 30 mL saline solution. The ketamine group gargled with 40 mg ketamine mixed in 30 mL of saline solution. The anesthesia provider did not know which group the patient was in. Patients might have been aware of their group assignment due to the taste of the saline and ketamine solutions. Anesthesia was induced with fentanyl 2 µg/kg and propofol 2 mg/kg. Patients were intubated with a soft seal cuffed polyvinyl chloride ETT after full paralysis with vecuronium. Men received an 8 – 9 mm ETT and women a 7 – 8 mm ETT. ETT cuff pressure was maintained between 18 and 22 cm H2O as measured with a handheld pressure gauge. Anesthesia was maintained with remifentanil and sevoflurane in oxygen and air. Oropharyngeal suctioning was performed under direct vision.

Postoperatively patients were assessed for sore throat at admission to the post-anesthesia care unit (PACU), and 2, 4, and 24 hours later. Sore throat was graded from 0 to 3 with 0 being no sore throat, 1 complaining of sore throat only if asked, 2 complaining of sore throat without being asked, and 3 severe sore throat with a change in voice.

Result            Three patients were consented but omitted from analysis for technical reasons. Thus, there were 23 control patients and 20 ketamine patients. Age, weight, smoking habits, duration of surgery and anesthesia, and total remifentanil administered were similar between the control and ketamine groups. The incidence of postoperative sore throat was higher in the control group then the ketamine group at admission to the PACU, 2 hours, and 24 hours (P<0.01). Severe sore throat (grade 3) was more common in the control group than in ketamine patients at 4 hours and 24 hours (P<0.05). No patient who gargled with ketamine reported a severe sore throat. Some control patients reported severe sore throat at the 2 hour, 4 hour, and 24 hour assessments. No side effects were observed.

Conclusion            The incidence and severity of postoperative sore throat was lower in patients who gargled with ketamine than in those who gargled with saline solution.



Sore throat can be a significant cause of postoperative distress in some patients. In my experience patients usually accept a brief, mild sore throat without too much concern. Some surgeries and anesthetic techniques, however, are associated with a significant incidence of severe sore throat that patients are likely to find bothersome. These are the patients in which taking the extra time for an intervention like a ketamine gargle might be worth considering. The investigators provide some evidence that ketamine may work peripherally, producing both analgesia and reducing inflammation. Furthermore ketamine is highly lipid soluble so absorption through mucous membranes is believable. The investigators didn’t provide any support for the dose of ketamine used, the duration of the gargle, or the onset of a peripheral action after transmucosal absorption. But while they left that ground open, they controlled for confounding factors reasonably well and were appropriately conservative with their statistical analysis. Because of that I’m convinced the differences in sore throat they found were probably real even if I don’t know exactly how the analgesia was produced.

Some anesthetists believe that dysphoria is inevitable when ketamine is administered. Neither the scientific literature nor my own experience with ketamine support this but there is no doubt that the way ketamine was used 20 years ago commonly resulted in dysphoria. Benzodiazepine premedication effectively prevents dysphoria from doses of ketamine even larger than the 40 mgs these patients gargled with and most patients receive midazolam before going back to the OR. Furthermore, it is unlikely that patients swallowed much of the ketamine they gargled with. Even if they did, only 20% of ketamine administered orally is absorbed systemically. If patients swallowed 10% of the ketamine they gargled with (4 mg) and 20% of that was absorbed (0.8 mg) the absorbed dose ends up being so small that dysphoria seems incredibly unlikely. To top it off, general anesthesia was induced soon after patients gargled so they had the entire duration of their anesthetic for any possible dysphoria to wear off. I certainly hope that no one would decide against this treatment because of concerns about dysphoria, especially in patients who received midazolam.


Michael Fiedler, PhD, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

McCutchen T, Gerancher JC


Early intralipid therapy may have prevented bupivacaine-associated cardiac arrest

Reg Anesth Pain Med 2008;33:178-180

McCutchen T, Gerancher JC



Purpose            The purpose of this manuscript was to report a case of local anesthetic toxicity treated, in part, with intravenous Intralipid.

Background            Regional anesthesia is commonly used for acute pain management. Local anesthetic toxicity may result in cardiac arrest and difficult resuscitation. Bupivacaine specifically impedes the carnitine-dependent transport of lipids into the mitochondria which may at least partially account for its selective cardiac toxicity. Intravenous (IV) lipid emulsion has increased the survival rates of test animals given a cardiotoxic dose of bupivacaine. There are now at least two case reports of successful resuscitation of human patients with Intralipid after bupivacaine-induced cardiac arrest. In two cases Intralipid was administered after 10 or more minutes without spontaneous circulation and circulation promptly returned.

Methodology            An 82 year old woman was scheduled for a total knee replacement. Her maintenance medications were celecoxib, levothyroxine, valsartan, rosuvastatin, and aspirin. She was sedated with a total of 2 mg midazolam and 100 µg fentanyl in multiple doses but remained awake and responded verbally. A femoral block catheter was placed and dosed with 30 mL of 0.5% ropivacaine with 5 µg/mL epinephrine. Next, the patient was turned laterally and a sciatic block was performed. The sciatic nerve block was dosed with 30 mL bupivacaine with 5 µg/mL epinephrine. Both blocks were dosed 5 mL at a time after negative aspiration for blood. There was no change in heart rate while dosing either block.

Result            Less than one minute after completion of the sciatic nerve block the patient had a tonic-clonic seizure. Midazolam 3 mg was administered and the seizure ended 30 seconds later. Oxygen was administered via an ambu bag though the patient was breathing spontaneously. A minute later the patient had a second seizure that ended before treatment could be administered. At this point the patient was unresponsive. ECG showed ventricular tachycardia (V-tach). Heart rate (HR) was 200 bpm. Central pulses were palpable. Oxygen saturation was 100% and blood pressure (BP) was 114/64. Now, three minutes after the second seizure ended, amiodarone 150 mg and 100 mL 20% Intralipid was given IV. The ECG continued to show V-tach. Femoral pulses were now less intense. A synchronized countershock with 120 J resulted in conversion to normal sinus rhythm. A second dose of Intralipid 400 mL was infused over the subsequent 15 minutes. The patient’s HR, BP, and ECG were normal and remained so. Over the next two hours her level of consciousness progressed from obtunded to alert and oriented. Follow up echocardiogram and serial troponin levels were normal. Serum local anesthetic levels were never measured.

Conclusion            Clinical signs suggested that despite negative aspiration at least some of the bupivacaine was injected intravenously resulting in bupivacaine local anesthetic toxicity. In this case, while cardiac arrhythmia ensued, it did not progress to cardiac arrest. It is unknown whether the cardiotoxicity was aborted due to the amiodarone, Intralipid, cardioversion, or some combination of the three. However, Intralipid was used as part of a successful treatment, preventing the progression of bupivacaine cardiotoxicity to cardiac arrest.



Normally, when I read a report such as this one I’d be highly skeptical of any claims that a single treatment prevented a complication when a number of treatments were used. In this case it is a little hard to say that Intralipid prevented cardiac arrest when an antiarrhythmic and cardioversion were also used. Furthermore, trying to “prove” that cardiac arrest was prevented is difficult because we don’t really know for sure that cardiac arrest would have happened if Intralipid had not been given. This case is a little different, though, and I’m willing to be less skeptical of the author’s belief that Intralipid may have prevented cardiac arrest. The reasons I’m more accepting of their conclusion are as follows. 1) The case looks like bupivacaine cardiotoxicity following an inadvertent IV injection. 2) CNS toxicity was clearly present and bupivacaine cardiotoxicity occurs either at the same time as, or earlier than, CNS toxicity. 3) We have a lot of historical information indicating that once bupivacaine cardiotoxicity begins, cardiac arrest is common and difficult to resuscitate. And lastly, 4) we have animal studies and human clinical case reports of near miraculous recovery from bupivacaine induced cardiac arrest immediately following Intralipid administration. Given all this I believe there is a reasonable chance that Intralipid did play a key role in preventing cardiac arrest in this case.

It should be noted that Intralipid may be useful for treating toxicity following local anesthetics other than bupivacaine. Reports of Intralipid use in cases of levobupivacaine and ropivacaine toxicity have also been published.

I look forward to further information becoming available about Intralipid as a rescue from local anesthetic toxicity. It would not surprise me if Intralipid soon became a “must have” emergency drug in any area where regional blocks are being performed.


Michael Fiedler, PhD, CRNA


For more information about Intralipid as a rescue treatment for local anesthetic toxicity see the link to the Lipid Rescue web site in the “Resource Links” section of the Anesthesia Abstracts web site.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008

Regional Anesthesia

Whitty R, Moore M, Macarthur A


Identification of the lumbar interspinous spaces: palpation versus ultrasound

Anesth Analg 2008;106:538-540

Whitty R, Moore M, Macarthur A



Purpose            The purpose of this study was to compare the vertebral interspace identified by palpation prior to an epidural or subarachnoid block with the interspace identified using ultrasound.

Background            Neuraxial anesthesia is associated with a small incidence of transient and permanent traumatic spinal cord injury. Studies have shown that palpation and anatomical landmarks are an inaccurate method of identifying a specific intervertebral space. Clinicians tend to misidentify the space at which they place a spinal or epidural needle as being one or two spaces lower than it actually is. (That is, the space is identified as L2-3 when it is really T12-L1 or L1-2.) Compared to x-rays, clinicians correctly identified a selected lumbar interspace only 30% of the time. Other studies have shown that after minimal training with ultrasound an anesthesiologist could correctly identify the L3-4 interspace 76% of the time. Some anatomic variations, such as extra vertebrae or fused vertebrae may result in incorrect interspace identification with either palpation or ultrasound.

Methodology            This prospective study included postpartum women who had received either spinal or epidural anesthesia for a cesarean section. All blocks were performed in the sitting position. The day after cesarean section, patients’ backs were examined and the spinal or epidural puncture site was located. Ultrasound was then used to identify the interspace at which the needle mark appeared by starting at the sacrum and counting up to the puncture site. The anesthesiologist who performed the ultrasound was unaware of the interspace that had been recorded in the patient’s anesthesia record. The interspace identified with ultrasound was compared to the interspace identified by palpation and recorded in the patient’s chart.

Result            One hundred twenty-one women were included in the analysis. Epidural anesthesia was used in 68% of women and subarachnoid block anesthesia in the remaining 32%. The vertebral interspace identified by palpation at the time of the block and the interspace identified by ultrasound agreed in 55% of the cases. Ultrasound identified the puncture site to be at least one interspace higher than charted in 32% of cases. In 12% of cases ultrasound identified the puncture site to be lower than the charted level. No demographic variables (age, weight, etc.) were associated with the difference between the interspace identified by palpation and by ultrasound. On average, ultrasound required about a minute and a half to perform.

Conclusion            Identifying a lumbar spinous interspace by ultrasound agreed with palpation a little over half the time. When there was disagreement, the clinician most commonly placed the spinal or epidural needle one or two spaces higher than they thought they were placing it.



Ultrasound seems to be the latest trend to sweep over regional anesthesia. There is probably an advantage to using ultrasound during the performance of some blocks for at least some practitioners. I must acknowledge from the start, however, that I may have a significant bias in reviewing this study. I’ve placed a large number of spinal and epidural blocks over the years and I have a hard time believing that I’ve correctly identified the space I placed the needle only 30% of the time. (Of course, I’ve never checked my space identification against spine x-rays so I don’t really know for sure.) I suspect that anesthesia providers that pay close attention to their regional techniques get the space right far more frequently than this study suggests. But even if the interspace is misidentified most of the time, the only reason to introduce an additional expensive piece of equipment into the process is if the outcome can be improved by doing so. I understand that direct needle trauma to the spinal cord is possible if I mistakenly insert the needle too high or if an anatomic variation results in an exceptionally low conus medullaris, but I’ve never seen this complication. I’ve never talked to another anesthesia provider who has seen it. I don’t believe I’ve ever read an incidence for this complication but it seems as though it must be exceptionally low.

Can the incidence of spinal cord trauma be meaningfully reduced with the aid of ultrasound that correctly identifies the space about 76% of the time? The answer may depend upon how good the landmarks are. This study did not record how discernable the landmarks were. Knowing that would add important context to the results. I have long recommended to my students that they don’t depend upon a line drawn between the posterior superior iliac spines to identify L4. I suggest that once they have identified the space they count interspaces down to the sacrum to make sure they are at the correct space. We don’t know, but it seems unlikely that this was done in this study. If we always verified our interspace identification by counting down I suspect that we would get it right a high percentage of the time. Of course, counting down to the sacrum doesn’t work when you can’t find the landmarks. As far as spinals and epidurals are concerned, the times I can’t feel the landmarks are probably when ultrasound will be most helpful.

Anesthesia doesn’t get better simply because we throw more technology at it. Accurate, dependable technology, skillfully applied to a real problem in a way that improves outcomes, can improve anesthesia.


Michael Fiedler, PhD, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 3, April 30, 2008