ISSN NUMBER: 1938-7172
Issue 1.2

Michael A. Fiedler, PhD, CRNA

Contributing Editor:
Chuck Biddle, PhD, CRNA

Guest Editor:
Joseph F. Burkard, DNSc, CRNA

Assistant Editor
Jessica Floyd, BS

A Publication of Lifelong Learning, LLC © Copyright 2007

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

















Welcome! As the development of Anesthesia Abstracts continues, additional contributing editors will be providing content with each issue. This month we welcome the contribution of Dr. Joe Burkard. Dr. Burkard has extensive expertise in regional anesthesia and is a tireless clinician and educator.

We have added some links to websites and video related to the content in this issue. Look in the "Resource Links" area. Notes about this additional content are found at the end of an abstract & comment.

Please remember that we are in the process of fine tuning Anesthesia Abstracts. Your comments and suggestions are welcome in "Blog with the Editors" and through the "Contact Us" tab.

Michael A. Fiedler, PhD, CRNA


Benedetto WJ, Hess DR, Gettings E, Bigatello LMToon H, Hurford WE, Schmidt U



Urgent tracheal intubation in general hospital units: an observational study

J Clin Anesthesia 2007;19:20-24

Benedetto WJ, Hess DR, Gettings E, Bigatello LM

Toon H, Hurford WE, Schmidt U


Purpose            The primary purpose of this study was to measure the incidence of complications resulting from urgent or emergent intubations in general nursing units.

Background            Cardiovascular and respiratory emergencies often require endotracheal intubation in general nursing units. Such intubations are assumed to be associated with a higher incidence of complications and poorer patient outcomes than are intubations in the operating suite. Few studies have examined these outcomes. A notable exception is those intubations performed in the emergency department. The most common complications reported in the studies available were difficult intubation, esophageal intubation and gastric aspiration.

Induction drugs are not uncommonly used to facilitate intubations performed outside the operating suite, though usually at lower doses than used for induction of general anesthesia. Muscle relaxants are less commonly used for fear of a “can’t intubate – can’t ventilate” scenario in a location lacking the equipment and personnel normally available in the operating suite to handle such a problem. True rapid sequence intubation with muscle relaxant has been used in the emergency department with a reported reduction in the rate of complications.

Methodology            This prospective observational study included 150 patients who required urgent or emergent intubation in a general nursing unit (intensive care units excluded). At the study institution, intubations were performed by a physician anesthesia resident with a minimum of six months of anesthesia experience. Data collected included patient demographics, the reason for intubation, number of intubation attempts, drugs used to facilitate intubation, and complications.

Result            Mean age of those intubated was 67?16 years. The reasons for intubation included: a respiratory emergency (61%), cardiac arrest (24%), airway protection (9%), and a neurologic event (7%). Twenty-one percent were urgent and 79% were emergent. Most patients received either propofol or etomidate to facilitate intubation. Muscle relaxants were used in 5% of patients. Most patients were intubated on the first attempt (76%).

Complications of intubation occurred in 27% of patients. The rate of complications was not statistically significantly different between urgent (28%) and emergent (22%) intubations (P=0.64). Complications included: multiple attempts at intubation (24%), esophageal intubation (9%), regurgitation and aspiration (7%), airway trauma (7%), and dental trauma (3%).

There was no difference in survival rate for those who experienced a complication during intubation and those who did not (P=0.63). The survival to discharge was only 52% overall for those who were intubated on a nursing unit.

The anesthesia department in the study hospital changed their out of the operating suite intubation procedures as a result of this study. They appointed an anesthesiologist to train resident physicians to better respond to airway emergencies and assigned an attending anesthesiologist to supervise CA-1 residents during out of the operating suite airway management. When a supervising attending anesthesiologist is not available, senior residents respond to airway calls.

Conclusion            Urgent and emergent intubation in general nursing units by anesthesia personnel had a complication rate of 27%. Patients who require urgent or emergent intubation in a nursing unit frequently die before discharge.



I’m not sure why the authors collected information on the survival of patients who were intubated in this study. The study did not examine whether or not the complications recorded during intubation contributed to the low survival rate and there is no reason I can see to think they did. There was also no control group to compare the intubated group to in order to make a determination that intubation improved patient outcome.

This study is interesting to me for two reasons. First, in many institutions non-anesthesia personnel respond to urgent and emergent intubation requests. While we may view first year anesthesia residents as relatively inexperienced, I wonder what the complication rate is when non-anesthesia providers respond to intubation requests. While a 27% complication rate sounds high compared to my experiences, the conditions are often more difficult on the floor than in the operating room. As long as the esophageal intubations were recognized and corrected (the study was silent on this point) it is likely that the outcome of the intubations was acceptable in most cases.

The second reason I am interested in this study is because, while most clinicians are interested in improving patient care these folks actually did something about it. As a result of this study experienced anesthesia providers began responding to intubation requests, either as supervisors or to perform the intubation. Given the economics of an academic medical center this represents a not insignificant commitment and demonstrates a dedication to improving patient care. I would be interested in knowing what the complication rate was after the change.

Michael A. Fiedler, PhD, CRNA


What is the incidence of complications following urgent and emergent intubations outside the operating room in your institution? How does it compare with the incidence in this study? If you know, share it with us in Blog with the Editors.

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Goodman EJ, Ziegler EJ, Douglas AM

The PAXpress airway causes more pharyngeal irritation than the reusable laryngeal mask airway

AANA J 2007;75:123-125

Goodman EJ, Ziegler EJ, Douglas AM


Purpose            To compare the incidence of complications following the use of two different airway devices.

Background            Since their introduction, supraglottic airway devices have grown in popularity, largely replacing traditional face masks when an endotracheal tube is not used. Pharyngeal airways that do not enter the trachea are available in several configurations from a number of manufacturers. The one size fits all adults PAXpress (Vital Signs, Totowa, NJ) is a pharyngeal airway that is inserted on top of the esophageal opening. A large low pressure cuff is inflated with up to 60 mL of air to seal the pharynx above the glottic opening. Like the original laryngeal airways, the PAXpress is intended to be inserted blindly.

Methodology            This retrospective chart review examined the records of 15 patients in which the PAXpress airway had been used and 15 patients in which the reusable LMA classic had been used during general anesthesia. Both devices were lubricated with a non-local anesthetic containing “gel” and inserted according to manufacturer’s instructions. The airway device cuffs were inflated with the minimum volume of air needed for a seal when up to 20 cm H2O was produced during positive pressure ventilation. All LMAs used were size #4.

The incidences of difficulty placing the devices, blood on the devices, and patient reports of pharyngeal pain were compared. A verbal analogue pain scale was used to quantify pharyngeal pain and a student’s t test was used to compare these values.

Result            Insertion of the airway device was charted as “easy” for all LMAs but for only 20% of the PAXpress. Multiple insertion attempts were needed with one PAX express and no LMAs.

No blood was noted on any LMA. Blood was noted on 27% of PAXpress airways.

No patients in the LMA group reported pharyngeal soreness in the PACU but 33% of PAXpress patients did. The following day, 13% of LMA patients reported a sore throat which they graded as 1.5 and 3 in severity on a 10 point scale. Two PAXpress patients were lost to follow up. Of the 13 contacted, 11 (85%) reported a severe sore throat. Some reported difficulty swallowing food. Over half graded their sore throat ≥ 5 on a 10 point scale.

Conclusion            The PAXpress was associated with a higher incidence and severity of sore throat than the LMA classic.



My sense is that the airway device examined in this study has little if any relevance in clinical anesthesia practice. But this article is still worthwhile. The authors were faced with evaluating a new airway device for possible purchase and use in their practice. They did so using an organized quality assurance approach founded on patient outcomes (complication rates) and patient satisfaction. Bravo! We need to make more of these types of decisions based upon dispassionate objective evidence. As the authors demonstrated, the process doesn’t have to be rocket science to have value.

From a scientific standpoint this study has a number of methodologic and analytic problems. Nevertheless, using the information on the number of patients who reported a sore throat (not the severity of the sore throat they reported) I performed a quick ad hoc analysis (Fischer’s exact). It showed that statistically significantly more PAXpress patients than LMA patients reported a sore throat both in the PACU (P=0.042) and during the next day follow up phone call (P<0.001). The authors qualitative report of just how much PAXpress patients complained of their discomfort is a further basis for their clinical decision.

Simple, meaningful studies to evaluate uncomplicated clinical questions like what airway device to buy or which way of doing something produces a better patient outcome can be done in a formulaic manner with just a little expertise and advanced preparation. Planning ahead and gathering the data prospectively strengthens the results in a number of ways with little, if any, added effort. Leaders in anesthesia groups and hospital administration would be well advised to use little studies like this one more often in their decision making process.

Michael A. Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Kikura M, Suzuki K, Itagaki T, Takada T, Sato S


Age and comorbidity as risk factors for vocal cord paralysis associated with tracheal intubation

Br J Anaesth. 2007;98:524-530

Kikura M, Suzuki K, Itagaki T, Takada T, Sato S



Purpose            The purpose of this study was to define the relationship between endotracheal intubation and vocal cord paralysis.

Background            Vocal cord paralysis is associated with endotracheal intubation and can result in difficulty phonating and in aspiration. The ASA closed claims analysis reports that a third of airway injury claims are laryngeal and most of these involve vocal cord paralysis, hematoma, or granuloma. Furthermore, temporary or permanent postoperative hoarseness is not uncommon even when vocal cord paralysis is absent. Patient specific risk factors may increase the likelihood of postoperative vocal cord paralysis but the relationship between multiple risk factors and vocal cord paralysis is not well defined.

The cause of post-intubation vocal cord paralysis is not known. It may be due to inflammation, recurrent laryngeal nerve compression, or reduced circulation to the recurrent laryngeal nerve.

Methodology            This observational cohort study included 31,241 consecutive patients who were intubated for general anesthesia and elective surgery. All patients were intubated with a Macintosh blade and a stylet. Those requiring other intubation methods or postoperative intubation were excluded. Adult female patients received a 7.0 or 7.5 mm endotracheal tube (ETT). Adult male patients received a 7.5 or 8.0 mm ETT. Children received an ETT appropriate for their age. The ETT cuff was inflated to a “no leak” pressure below 20 torr and the pressure was checked periodically during the case. All ETTs were lubricated with 2% lidocaine gel. Muscle relaxation was used for intubation in all patients. Patients were evaluated for vocal cord paralysis on the day of surgery or the first postoperative day. All patients with persistent dysphonia were evaluated by an ENT surgeon. Vocal cord paralysis was attributed to endotracheal intubation by exclusion.

Result            In all, 24 patients (0.08%) had vocal cord paralysis attributed to endotracheal intubation within 30 days of surgery. Intubation was accomplished in a single attempt in each of these patients. The left vocal cord was paralyzed twice as often as the right vocal cord. In each case patients were found to be dysphonic during the postoperative visit (same day or first postoperative day). Normal vocal cord function returned in all patients in an average of 68 days (range 41 – 97 days). No vocal cord paralysis occurred in patients less than 20 years old.

The duration of intubation was most strongly associated with vocal cord paralysis. Patients intubated for between 3 and 6 hours were twice as likely to experience vocal cord paralysis as those intubated for <3 hours. Those intubated for >9 hours were 30 times more likely to have vocal cord paralysis (P<0.0001).

Patients greater than 50 years old were over 3 times more likely to experience vocal cord paralysis than patients younger than 50 years. Patients with diabetes or hypertension were 2 times more likely to have vocal cord paralysis than those without diabetes or hypertension.

These associations are consistent with a cause that involves reduced arterial circulation to the recurrent laryngeal nerve.

Conclusion            The risk of vocal cord paralysis increases noticeably with the duration of intubation. The risk is over three times higher in patients over 50 years old and about twice as great in patients with diabetes or hypertension.



Vocal cord paralysis is a rare but important postoperative complication. The proximate cause is not precisely known and it would be a mistake to simply blame it on endotracheal intubation. Some patients who are not intubated experience postoperative vocal cord paralysis. Nevertheless, vocal cord paralysis is associated with endotracheal intubation and there are factors related to intubation that increase the likelihood that vocal cord paralysis will occur. In this study, all patients who developed vocal cord paralysis were hoarse within a day after surgery. But some degree of hoarseness is a complaint of up to half of patients and few of them go on to develop vocal cord paralysis. This rather large study helps us better understand which patients are at increased risk of postoperative vocal cord paralysis. Understanding which patients are at highest risk helps raise our index of suspicion and provides guidance about who to watch most closely. Most patients who are hoarse recover their normal voice quickly. But for those few in whom it is a forerunner of vocal cord paralysis, early detection is essential. With treatment, all patients in this study recovered vocal cord function within a few months.

Hindsight is great, of course, but I really wish these investigators would have collected a little bit more information. Even in an observational study, having information about oxygen carrying capacity, the magnitude and duration of any hypotensive episodes, and actual ETT cuff pressures throughout the case might have been enough to suggest the cause of vocal cord paralysis.

Lastly, the fact that the left vocal cord was paralyzed twice as often as the right makes me wonder if something about how we commonly secure the ETT plays a role in vocal cord paralysis. When we have a better idea what the cause of vocal cord paralysis is perhaps we can prevent the complication by changing the way we secure the ETT.


Michael Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Shippey B, Ray D, McKeown D


Case series: The McGrath videolaryngoscope – an initial clinical evaluation

Can J Anesth 2007;54:307

Shippey B, Ray D, McKeown D



Purpose            The purpose of this report was to evaluate the McGrath videolaryngoscope.

Background            Several video laryngoscopy devices are now available and gaining wider use. They may offer advantages over direct laryngoscopy, most notably the ability to gain a better view of the glottis in patients with a difficult airway. Experienced and novice laryngoscopists often view video laryngoscopy as quicker and easier than direct laryngoscopy when practicing difficult intubations on a model.

The McGrath videolaryngoscope is self-contained. It is powered by a single double A battery and has a small color tilt and swivel LCD display attached to the laryngoscope handle. The length of the curved laryngoscope blade is adjustable and part of the blade is disposable.

Methodology            This two phase evaluation included 150 adult patients who required general anesthesia and endotracheal intubation. Patients were neither included nor excluded based on their airway evaluation. Laryngoscopy was performed by one of three experienced anesthesiologists who had used an early version of the McGrath videolaryngoscope in an airway manikin. None of them had used the McGrath videolaryngoscope in humans.

In phase I the anesthesiologists refined their laryngoscopy technique in 75 patients. Refinement included gaining experience with the McGrath videolaryngoscope and evaluating the need for a stylet or bougie for endotracheal tube (ETT) placement. In phase II the device was evaluated in 75 additional patients. A stylet was routinely used for ETT placement. The laryngoscope blade was extended to the middle of its adjustable range. The number of attempts at laryngoscopy, the time to obtain an optimal view, the view on the display during laryngoscopy (grades I through IV), and the total intubation time were recorded.

Result            Tracheal intubation was successful using the McGrath vidoelaryngoscope in 147 of 150 patients. Eight required two laryngoscopy attempts. A grade I view was obtained in 143 patients (95%) and a grade II view in six patients (4%). (Equipment failure prevented any view in one patient.) Several attempts to place the ETT without a stylet were unsuccessful.

In the 75 phase II patients, the median time required to obtain the best glottic view was 6.3 seconds (range 2 – 26.3 seconds). The median total time for intubation was 24.7 seconds (range 11.4 – 286 seconds). Two patients required more than three minutes to intubate with the McGrath videolaryngoscope. One had a fused cervical spine and was graded a Mallampati 3. The other had undergone oral surgery on the floor of the mouth. Ninety-six percent of intubations were accomplished in less than one minute.

Intubation was subjectively rated as “easy” in 79% of patients, “slightly difficult” in 17%, and “moderately difficult” in 4%. Despite a good view of the glottis and use of a stylet, there was difficulty placing the ETT into the glottis in 14 patients (19%) and difficulty advancing the ETT into the trachea in 9 patients (12%). Since pharyngeal tissue is not displaced anteriorly to the extent that it would be during a direct laryngoscopy, the ETT approaches the glottic opening and trachea at a much greater angle when using a video laryngoscope. Advancing the ETT into the trachea is typically not as easy as it is during direct laryngoscopy.

Eighteen of the 150 patients (12%) were potentially difficult intubations. Using the McGrath videolaryngoscope, a grade I view was obtained in 16 of these patients and a grade II view in the other 2 patients. Intubation using the McGrath videolaryngoscope failed in 3 of the 150 patients and they were intubated under direct vision with a standard laryngoscope.

Conclusion            The McGrath videolaryngoscope provides excellent laryngoscopic views in normal airways and in some patients with predictors of difficult intubation. Advancing the ETT into the trachea may, at times, be difficult but routine use of a stylet reduces the difficulty.



Death or brain injury due to airway problems is not common but remains an important quality improvement area in anesthesia practice. Failed intubation occurs in up to 0.3% of surgical patients. While a fiber optic flexible bronchoscope can often be used to successfully intubate these patients, fiber optic scopes are expensive, easily damaged, hard to clean, and require specialized training to use proficiently. The recent development of a number of less complex devices to aid in difficult intubation is heartening. The McGrath videolaryngoscope is interesting because it does not require any wires, an AC power supply, or a separate stand to support the view screen. It may also be quicker and easier to clean.

This evaluation is a nice first step for those who have not had a chance to use a McGrath videolaryngoscope in clinical practice. It simultaneously produces some encouraging and some puzzling information. As with similar devices, the view of the glottis was uniformly quite good, despite some patients with assessment finding typically associated with a poor view of the glottis under direct laryngoscopy. Especially encouraging was that in all 18 potentially difficult intubations the McGrath videolaryngoscope view was quite good and intubation accomplished without trouble. The difficulty maneuvering the ETT into the glottis is probably unavoidable in patients with a difficult airway since their glottis tends to be more anterior. This is often what makes direct laryngoscopy difficult in the first place. What is puzzling is why the authors were unable to intubate three patients who did not have signs predictive of difficult intubation.

It will be helpful to see some studies comparing the percent success, the view, and total time needed for intubation with devices like the McGrath videolaryngoscope and traditional direct laryngoscopy.


Michael Fiedler, PhD, CRNA



Click on “Resource Links” for links to YouTube video of the McGrath laryngoscope in action.

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Equipment & Technology

Rees LM, Sheraton TE, Modestini C, Wilkes AR, Hall JE


Assessing the efficacy of HME filters at preventing contamination of breathing systems

Anaesthesia 2007;62:67-71

Rees LM, Sheraton TE, Modestini C, Wilkes AR, Hall JE


Purpose            The purpose of this study was to evaluate the effectiveness of anesthesia circuit bacterial / viral filters using ATP bioluminescence.

Background            Hospital acquired infection results in morbidity and mortality, and adds to the cost of health care. Anesthesia equipment may play a role in spreading infection. Anesthesia breathing circuit filters are intended to reduce the likelihood that infectious agents will be spread from one patient to another. Currently available circuit filters have a wide range of filtration performance.

Adenosine Triphosphate (ATP) bioluminescence is used in the food industry to monitor the cleanliness of surfaces that come into contact with food. ATP bioluminescence has been shown to compare favorably with microbiological culturing but it requires less training to use and produces nearly real time results.

Methodology            In this prospective, observational study, 235 used breathing circuit filters of four types were collected from a single suite of operating rooms. The Biotrace Clean-Trace® (Biotrace International Inc., Bothell, WA) was used to evaluate the patient side and anesthesia machine side contamination of each breathing circuit filter. Contaminates were collected with a disposable swab containing an agent that releases ATP from microbial cells and other organic materials. A reagent released by the?Clean-Trace® device reacts with ATP on the disposable swab and produces light. Light intensity is measured by the hand held Clean-Trace® device. The light produced by the reaction is proportional to the amount of ATP released from the microbial / organic contamination, and thus is proportional to the amount of contaminates present on the swab.

Result            Overall, breathing circuit filters were far more contaminated on the patient side than on the anesthesia machine side, though this was not always the case. The anesthesia machine side was contaminated in 9% of filters. There was no correlation between the amount of machine side filter contamination and the length of the anesthetic. There was a low, but very real, correlation between the patient side filter contamination and the length of the anesthetic.

Conclusion            The anesthesia circuit bacterial / viral filters tested removed most, but not all, contaminates introduced into the breathing circuit during patient use. Overall, they were highly effective.



It is reassuring to know that breathing circuit filters do a pretty good job of keeping contaminates out of the anesthesia machine. That, however, is not what caught my attention in this article. What caught my interest was this nifty little handheld device which apparently is able to quantify surface contamination with secretions, blood, bacteria, and other organic materials almost immediately.

We all know that anesthesia equipment can become contaminated with patient blood, secretions, and other organic materials. We understand the importance of cleaning anesthesia equipment to prevent disease transmission between patients or from patients to staff. We also know that our equipment isn’t always as clean as it looks. What we have a hard time knowing is how well our equipment has been cleaned, how often it needs to be cleaned, and how clean it is right now. In the past we have only been able to sample the cleanliness of anesthesia equipment. Equipment was cultured, cultures sent to the lab, some time later we’d get the results. Because this culture procedure was slow and expensive we don’t use it very often.

Now, let me be clear, reading this article was the first I’d heard of the Clean-Trace®. All I know about it as I write this, I’ve learned from the article and the manufacturer’s web site. But from what I see, the Clean-Trace® may allow real time monitoring of anesthesia equipment contamination. Such a device would be useful to evaluate the effectiveness of established cleaning protocols and to spot check the performance of individuals responsible for cleaning. If my institution’s infection control department doesn’t know about this device already, they soon will.

Michael Fiedler, PhD, CRNA

The Clean-Trace®  is available from Biotrace International. Their web site is

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007


Chung F, Ward B, Ho J, Yuan H, Kayumov L, Shapiro C


Preoperative identification of sleep apnea risk in elective surgical patients, using the Berlin questionnaire

J Clin Anesth 2007;19:130-134

Chung F, Ward B, Ho J, Yuan H, Kayumov L, Shapiro C



Purpose            The purpose of this study was to screen elective surgical patients for Obstructive Sleep Apnea (OSA) and describe the percentage at high risk for OSA.

Background            OSA occurs in approximately 12% of men and 5% of women between 30 and 60 years old. It has been estimated that over 80% of both men and women with OSA have not been diagnosed.

OSA is associated with morbidity and mortality. Unexplained cardiopulmonary arrest following general anesthesia and surgery has been associated with sleep apnea and opioid administration. This increased risk may be associated with an elevated sensitivity to opioids. Joint replacement patients are twice as likely to experience serious postoperative respiratory or cardiac complications if they have sleep apnea. One study reported that 6.7% of orthopedic patients had OSA. Sleep apnea has also been associated with difficult intubation.

The American Society of Anesthesiologists has published guidelines recommending that anesthesiologists and surgeons develop a preoperative assessment pathway for patients with clinical findings suggestive of OSA. The gold standard for diagnosis is the overnight sleep study but it is not feasible as a screening tool. The most widely used OSA screening tool is the Berlin questionnaire.

Methodology            This combined prospective screening and retrospective chart review screened 305 surgical patients for OSA. Patients were scheduled for general, orthopedic, urologic, plastic, ophthalmologic, or neurosurgical procedures. Patients completed the Berlin questionnaire and demographic data were collected. Patients were segregated on their risk for OSA as either High or Low. The primary care physician of patients categorized as High risk was asked to refer the patient for polysomnography (overnight sleep test). Postoperatively, patient charts were reviewed for perioperative complications.

Result            Berlin questionnaire screening identified 73 patients (24%) at high risk for OSA. Nine of these patients had previously been diagnosed with OSA. Six patients underwent polysomnography as requested. Of these six, four were diagnosed with OSA. Thus, 13 of 305 patients (4.2%) were known to have OSA.

Of the 73 patients identified as being at High risk, 18% were diagnosed with OSA, 3% were tested but did not have OSA, and the OSA status of 79% of the High risk patients was unknown. Those identified as being at High risk of OSA by the Berlin questionnaire were older, had higher BMIs, and were more likely to have hypertension. High risk patients experienced perioperative complications at the same rate as those at Low risk of OSA.

Conclusion            In this study, 24% of elective surgical patients screened were identified as being at High risk for OSA. Of 305 elective surgical patients screened, 4.2% were confirmed to have OSA. The actual incidence may have been higher as 79% of those identified as being at High risk for OSA did not receive diagnostic follow-up and none of the Low risk patients had follow-up.



Obstructive Sleep Apnea (OSA) is associated with a body mass index (kg/m2) >35, tonsils that touch or nearly touch each other in the midline of the pharynx, frequent loud snoring, observed pauses in breathing during sleep, and frequent daytime sleepiness or falling asleep easily during the day despite adequate sleep. These and other factors have been gathered into the Berlin questionnaire in an attempt to predict an individual’s risk of OSA.

While this study is limited in a number of ways, it serves to highlight an important area in which preventable postoperative complications may currently be all but invisible. OSA may contribute to postoperative morbidity and mortality that is chalked up to “causes unknown.” There are procedures which experts widely agree should not be done on an outpatient basis in patients with OSA. Likewise, there are procedures for which inpatients with OSA would benefit from specific anesthetic planning and enhanced postoperative monitoring. Of course, the problem is that the vast majority (>80%) of patients with OSA are undiagnosed. The Berlin questionnaire has fairly good predictive power for identifying patients at risk for OSA and it is simple to administer and score. The Berlin questionnaire is short enough and simple enough that it could be integrated into the pre-anesthetic evaluation process. This would be especially easy to accomplish when initial pre-anesthetic data is gathered over the phone by an RN.

In the end, though, we need to better understand which surgical procedures are associated with increased risk of morbidity and mortality in patients with OSA. We can then focus our efforts on identifying OSA in patients scheduled for those procedures and develop an appropriate anesthetic plan for patients at risk for OSA. Until that time, we can become more aware of the factors associated with OSA and develop an anesthetic plan that provides for OSA when we suspect a patient may be at risk.


Michael Fiedler, PhD, CRNA


Gross JB, Bachenberg KL, Benumof JL, et al. Practice guidelines for the perioperative management of patients with obstructive sleep apnea: a report by the American Society of Anesthesiologists Task Force on Perioperative Management of patients with obstructive sleep apnea. Anesthesiology 2006; 104: 1081-93.

Netzer NC, Stoohs RA, Netzer CM, Clark K, Strohl KP. Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Annals of internal medicine 1999; 131: 485-91



© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Greif R, Akca O, Horn EP, Kurz A, Sessler D

 Supplemental perioperative oxygen to reduce the incidence of surgical wound infection

New Engl J Med 2000;342:161-167

Greif R, Akca O, Horn EP, Kurz A, Sessler D


Purpose            To determine if supplemental administration of oxygen in the immediate perioperative period would decrease the rate of postoperative wound infection in patients undergoing colorectal surgery.

Background            Surgical wound infections occur with surprising frequency in the general population; in the case of colorectal surgery, they are reported in up to 27% of patients. Associated increase costs, increased length of stay and even mortality make the issue an important one to study.  It was hypothesized that increasing the fiO2 in the immediate perioperative period would reduce surgical wound infections in this population.

Methodology            The authors enrolled 500 patients aged 18 to 80 years who underwent open colorectal resection.  All received a similar general anesthetic and follow-up was standardized. Intraoperatively the patients were randomized to receive an fiO2 of 30% oxygen (in nitrogen) or 80% oxygen (in nitrogen), concentrations that were continued for 2 hours into the postoperative phase.  Preoperatively, standard regimens of prophylactic antibiotics were administered equivalently in both arms of the study.  Both groups received a brief period of 100% oxygen just prior to extubation.  Core temperature was maintained at 36 degrees C in both groups and postoperative pain control was standardized as well.  Wounds were evaluated postoperatively by physicians unaware of the oxygen treatment arms of the study using valid and reliable standardized wound infection rating scales.

Result            Although the study planned to include 1000 patients, an interim analysis was performed once 500 patients were enrolled, and the study was discontinued because the incidence of wound infection between the two groups was so great.  The overall incidence of wound infection was 8%.  In patients who received 80% oxygen the rate of infection was 5.2% (CI 2.4 – 8.0) while in the group that received 30% oxygen the infection rate was 11.2% (7.3-15.1), differences that were significant at P=0.01.

Conclusion            The authors concluded that supplemental, 80% oxygen administered during and in the first 2 hours postoperatively halved incidence of surgical wound infection.


The more we look, the more we see.  Decisions that we make intraoperatively, even ostensibly simple ones like choice of opiate, amount of fluid administered, use of a warming device, or in this case, selection of inspired oxygen tension, may have dramatic, long term consequences.  Oxygen is clearly not without adverse effects.  In some, it may depress respiratory drive.  In high enough concentrations for a long enough period, it produces pulmonary toxicity.  Oxygen may facilitate the development of atelectasis.  Yet oxygen is life-sustaining and it can be demonstrated in laboratory preparations that the bactericidal activity of neutrophils is mediated by oxidative killing.  Furthermore, surgical tissue trauma perturbs local and regional blood flow in such a way that driving gradients may be taxed.  Doesn’t it make sense then, that giving some “Os” in the perioperative period might help mitigate against wound infections?  This study seems a compelling example.

            But let’s consider another study (Pryor et al.  Surgical site infection and the routine use of perioperative hyperoxia.  JAMA.  2004;291:79-87).  These researchers basically repeated the study in 165 patients undergoing major abdominal surgery (a more diverse group of procedures).  Using randomized assignment to either a 35% or 80% arm of the study they employed equally standardized and rigorous methodology and found an overall infection rate of 18%, more importantly they revealed a 25% rate in the 80% oxygen group and 11.3% rate of infection in the 35% arm (P=0.02). Virtually the exact opposite of the Greif (above) et al paper!!  What are we to do (as we not uncommonly find) with results of studies that appear to conflict?

            On closer examination it is clear that there are major differences between the studies.  For example, a much more diverse surgical procedure population in the Pryor paper.  More transfusions (30% vs 4%), greater mortality (1.3% vs 0.6%) and longer length of stay (12.4 days vs 7.4 days) all in the Greif population.  What stands out to me, in this day and age of evidence-based practice, is the need to carefully weigh all the decisions that we make, every one, in the context of the best available information in the context of the particular patient that is presented to us.  Although ostensibly conflicting, I find studies like this clinically revealing as they help to further illuminate landscape that is often in the shadows.  What I gain from these studies, and what I will carry into the operating room and pass along to my students, is the need to carefully weigh each decision that we make.  High tissue oxygen partial pressures may benefit some, but may prove harmful to others.  Our task is to decide, as we must with all decisions that we make, what will benefit a particular patient in a particular circumstance.

Chuck Biddle PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007


White H, Black RJ, Jones M, Mar Fan GC


Randomized comparison of two anti-emetic strategies in high-risk patients undergoing day-case gynaecological surgery

Br J Anaesth 2007;98:470-476

White H, Black RJ, Jones M, Mar Fan GC



Purpose            The purpose of this study was to compare the incidence and severity of PONV for 24 hours in patients who received an inhalation anesthetic with dolasetron and those who received a propofol-based total intravenous anesthetic (TIVA). A secondary purpose was to identify factors associated with PONV in a high risk group.

Background            Postoperative nausea and vomiting (PONV) is an especially important cause of morbidity in outpatient surgery patients where it can result in delayed discharge and unplanned admission. PONV is also associated with wound dehiscence and aspiration of gastric contents. PONV is often cited as the complication most troubling to patients and is the most common reason patients are unsatisfied with their surgical experience. Post discharge nausea and vomiting (PDNV) is often not reported but may occur in up to 50% of patients. No single best strategy for PONV prophylaxis is agreed upon. Universal prophylaxis is not recommended but may be appropriate in patient groups at high risk for PONV.

Methodology            This randomized, single blind study included 126 women scheduled for elective gynecologic surgery on an outpatient basis. All subjects were estimated to have a > 40% risk of PONV. Patients were randomized to receive either a sevoflurane anesthetic with dolasetron PONV prophylaxis (SD group) or a propofol TIVA anesthetic (TIVA group). The SD group received propofol for induction of anesthesia. General anesthesia was then maintained with oxygen, air, and sevoflurane. Dolasetron 12.5 mg was given before the end of the case. The TIVA group received no other PONV prophylaxis. All patients received one liter of crystalloid IV fluid beginning in preoperative holding. All patients received midazolam and fentanyl. Neuromuscular block and reversal was used in 30% to 40% of patients in both groups. Reversal of neuromuscular block was accomplished with neostigmine 2.5 mg when necessary.

 In recovery, nurses were not told whether the patient was in the SD or the TIVA group but the chart was available to them. The highest pain and PONV scores in each half hour period were recorded. A complete response to PONV prophylaxis was defined as no nausea, vomiting or medication for nausea for 24 hours after surgery.

Result            Ultimately, 67 SD patients and 55 TIVA patients were included in the analysis. The number of patients in each group who received morphine in recovery, who were paralyzed during the case, and who received neostigmine for antagonism of neuromuscular block were not statistically significantly different between groups. Over 90% of patients in both groups received midazolam and fentanyl and the average doses were the same in each study group. The duration of surgery averaged 16.5 minutes in the SD group and 22 minutes in the TIVA group (P=0.046).

Overall, 28% of patients experienced PONV prior to discharge but there was no significant difference in PONV scores or use of rescue antiemetics between groups prior to discharge. PDNV, from discharge to 24 hours after surgery, was more common in the TIVA group (P=0.019). Both post discharge nausea (P=0.004) and vomiting (P=0.03) were more common in the TIVA group. Of those patients who experienced PONV, 28% experienced PDNV only and 39% experienced both pre- and post discharge nausea and vomiting.

Multivariate analysis identified the type of anesthetic (P=0.022), greater patient weight (P=0.005), and longer duration of anesthesia (P<0.0001) to be associated with a greater risk of PONV. The odds of complete response to PONV prophylaxis decreased by 32% for each additional 10 kg of patient body weight. Patients in the sevoflurane dolasetron group were 2.7 times more likely than TIVA patients not to experience PONV during the 24 hours after surgery.

Conclusion            Propofol TIVA is significantly less effective than antiemetic prophylaxis for prevention of post discharge nausea and vomiting. Propofol TIVA and dolasetron antiemetic prophylaxis are equally effective at preventing nausea and vomiting prior to discharge.



It is good to see attention paid to the nausea and vomiting experienced by outpatients after they have been discharged home. To few studies consider post discharge nausea and vomiting in their search for effective ways to reduce PONV. While propofol is important in our efforts to reduce the incidence of PONV, it is critical to remember, as this study so persuasively points out, that propofol alone is not enough to extinguish PONV for 24 hours. The emetic influence of inhalation anesthetic agents is relatively short lived, about 2 hours.1 Likewise, the anti-emetic influence of propofol is also relatively short lived; in this study only lasting the few hours prior to discharge. A significant number of patients experienced PONV only after they had been discharged. Our patients will be happier with us if we plan for 24 hour PONV prophylaxis. A propofol anesthetic is a good start in high risk patients but it will take more than propofol to prevent PONV for a full 24 hours postoperatively.


Michael A. Fiedler, PhD, CRNA


1. Apfel CC, Kranke P, Katz MH, et al. Volatile anaesthetics may be the main cause of early but not delayed postoperative vomiting: a randomized controlled trial of factorial design. Br J Anaesth 2002; 88: 659-68.




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007


Hurley RW, Cohen SP, Williams KA, Rowlingson AJ, Wu CL


The Analgesia effects of perioperative gabapentin on postoperative pain: a meta-analysis

Reg Anesth Pain Med 2006;31:237-247

Hurley RW, Cohen SP, Williams KA, Rowlingson AJ, Wu CL



Purpose            The purpose of this study was to perform a meta-analysis of the available randomized controlled trials that were conducted to evaluate the analgesic effect of perioperative gabapentin.

Background            Numerous neurophysiological or neurochemical mechanisms are involved in postoperative pain including peripheral nerve and central sensitization.  These mechanisms can contribute to the patient’s enhanced pain sensitivity, represent a corollary of neuropathic pain in the acute setting, and may be responsive to the treatments used in the palliation of neuropathic pain.  The versatility of gabapentin in treating a wide array of pain conditions and its favorable side effects have generated interest in its use as a perioperative analgesic.

Methodology            The National Library of Medicine’s PubMed database along with abstracts from the Society for Neuroscience and the American Society of Anesthesiologists were searched with terms related to gabapentin and postoperative pain.  Inclusion and exclusion criteria were established.  Demographic and clinical data were extracted and analyzed.  The primary outcome measure was pain score at 20 - 24 hours after surgery.  A random effects model was used, and all statistical analyses were performed with RevMan 4.2.7.

Result            The search resulted in 85 abstracts with 12 articles meeting all inclusion criteria.  There were 449 subjects who received gabapentin and an equal number who received placebo.  Perioperative administration of oral gabapentin was associated with significant decrease in postoperative pain at both 20 - 24 and 0 - 4 hours after surgery. (Weighted Mean Difference WMD = -0.74 95% CI, -1.03 to -0.45 and -1.57 95% CI, -2.14 to -0.99, respectively) There was also no difference in the incidence of nausea, vomiting, or dizziness between groups, but the oral gabapentin group had a significantly higher incidence of sedation.

Conclusion            Compared with placebo, the perioperative administration of gabapentin produced significantly better postoperative analgesia.  Postoperative analgesia as measured by pain scores and decreased opioid consumption was improved in the immediate postoperative period and up to 24 hours after a wide range of surgical operations commonly associated with significant postoperative pain.  A single study in this systematic review examined the dose dependence of the effect of gabapentin on postoperative pain.  These authors concluded that 600 mg administered before surgery provided the best balance between pain relief and unwanted effects.  The results of this meta-analysis indicate that oral gabapentin administered within 4 hours before surgery significantly reduced postoperative pain.



Preemptive analgesia and postoperative pain relief continues to be a cornerstone of perioperative anesthesia practice.  The authors mention common secondary outcome measures which include perioperative pain, patient satisfaction, nausea and vomiting, and postoperative sedation.  A single, one time 600 mg dose of gabapentin given preoperatively to reduce postoperative pain up to 24 hours seems to be a useful adjunct in the management of postoperative pain.  In my anesthesia practice, I use a multi-model approach combining regional anesthesia, total intravenous medications when appropriate, and other adjuncts that decrease pain through mechanisms other than opioids such as non-steroidal anti-inflammatory agents, alpha-2 agents and now gabapentin.  I believe this approach decreases peripheral nerve and central sensitization leading to decreased pain and higher patient satisfaction.


Joseph F. Burkard, DNSc, CRNA


© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

White PF, Sacan O, Tufanogullari B, Eng M, Nuangchamnong N, Ogunnaike B


Effect of short-term postoperative celecoxib administration on patient outcome after outpatient laparoscopic surgery

Can J Anesth 2007;54:342-348

White PF, Sacan O, Tufanogullari B, Eng M, Nuangchamnong N, Ogunnaike B



Purpose            The purpose of this study was to test the hypothesis that postoperative oral celecoxib (Celebrex) 400 mg/day for 4 days would result in improved quality of recovery and earlier return to activities of daily living after laparoscopic surgery.

Background            Effective pain relief has been reported to aid in the recovery and increase the satisfaction of outpatients. Ineffective acute pain relief has been suggested to facilitate the development of chronic pain, even following minor surgical procedures. Nonsteroidal anti-inflammatory drugs (NSAIDs) are used as part of a multimodal approach to postoperative pain relief. Ketorolac has been shown to be opioid sparing after laparoscopic surgery and to be associated with earlier discharge following anorectal surgery.

Cyclooxygenase-2 (COX-2) selective NSAIDs provide effective analgesia after outpatient surgery and are associated with earlier bowel function and return to normal daily activities. Some other COX-2 NSAIDs have been withdrawn from clinical use due to safety concerns. Their use has become controversial since the withdrawal of two COX-2 NSAIDs due to concerns about cardiovascular complications. Celecoxib has not yet been associated with cardiovascular complications, especially during short term use.

Methodology            This randomized, double-blind, placebo-controlled study included 80 ASA class I-III outpatients scheduled for laparoscopic surgery. Patients were divided into two groups. The control group received a placebo. The Treatment group received celecoxib 200 mg twice a day for 4 days postoperatively. The first dose of placebo or celecoxib was administered 10-20 minutes after patients arrived in recovery. Patients were sent home with identical capsules containing either placebo or celecoxib to take over the following three days.

All patients were premedicated with midazolam. General anesthesia was induced with propofol and fentanyl. Rocuronium was used for muscle relaxation. Anesthesia was maintained with desflurane in air and oxygen. After induction, patients received droperidol and dexamethasone. Surgical wounds were infiltrated with bupivacaine at the end of the case and muscle relaxant antagonized with neostigmine if needed.

Patients were assessed for pain, nausea, quality of recovery, and patient satisfaction with pain management. Pain and nausea was assessed at 30, 60, 120, and 240 minutes and then daily. Postoperative analgesic use was recorded both in recovery and at home. Times to normal fluid and solid food consumption, first bowel movement, and return to activities of daily living were recorded.

Result            Patients in the celecoxib group received an average of 84?45 ?g fentanyl in recovery compared to 127?58 ?g in control patients (P=0.03). Pain scores in recovery were similar overall. Average pain scores 24 and 48 hours postoperatively were lower in the celecoxib group (P=0.028 and 0.01 respectively). Fewer patients in the celecoxib group required rescue pain medications postoperatively as well.

Patient satisfaction with pain management and quality of recovery scores were both greater in the celecoxib group on all three postoperative days. Bowel function returned a day earlier in the celecoxib group (P<0.05). Normal activities of daily living resumed on day 4?2 in the celecoxib group and day 6?3 in the placebo group (P<0.05). No study patients experienced wound or cardiovascular complications within 30 days following discharge.

Conclusion            Following laparoscopic surgery, celecoxib 200 mg twice a day for 4 days resulted in better pain relief and faster return to activities of daily living without complications associated with other COX-2 NSAIDs.



Opioids are great analgesics, but opioids cannot relieve any and all pain and they do have side effects. The most common side effects (e.g. nausea, constipation) are generally much less troublesome than the pain experienced without opioids. But if pain relief can be achieved while limiting or eliminating opioid use these side effects can be reduced as well and patients feel better.

If you are like me, you grew up thinking of NSAIDs like aspirin as rather impotent; not worth much in the area of surgical pain relief. Part of the reason for this is that NSAIDs do not relieve pain, they prevent pain formation. While opioids bind with receptors and reduce pain perception, NSAIDs interrupt a biologic process (they inhibit cyclooxygenase), preventing inflammation and other responses to tissue injury that cause pain. NSAIDs reduce the sensitivity of pain receptors (nociceptors) to a stimulus like touch that would not normally be painful but can be painful at the edge of a fresh wound (allodynia). They also reduce the increased sensitivity to a stimulus that would normally be painful but has become more so as a result of unrelieved pain (hyperalgesia). For these reasons, NSAIDs are widely agreed to be the preferred foundation for acute pain management.

The drugs we group into the category “NSAIDs” are actually quite chemically diverse. This study doesn’t tell us that celecoxib is a better foundation for postoperative pain than other NSAIDs. What it does tell us is remarkable. On average, patients who received celecoxib as the foundation of their pain management after laparoscopic surgery reported less pain and returned to normal daily activities two days sooner (day 4 vs. day 6) than patients who received only opioids. It would seem that my grandmother was right again; an ounce of prevention is worth a pound of cure.


Michael Fiedler, PhD, CRNA



One set of postoperative pain management guidelines that include NSAIDs has been developed by the American Society of Anesthesiologists Task Force on Acute Pain Management. The guidelines were published in Anesthesiology 2004;100:1573 and are available online at

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007


Montero-Huerta P, Hess DR, Head A

 Inhaled nitric oxide for treatment of sickle cell stroke

Anesthesiology 2006;105:619-621

Montero-Huerta P, Hess DR, Head A


Purpose            A report of the successful management of a nonhemorrhagic stroke that occurred in the immediate postoperative period in a 13 year old African-American boy with sickle cell disease.

Background            Risks and complications associated with sickle cell disease present major challenges to anesthesia providers.  In this case a child with a long-standing history of sickle cell disease and vasooclusive crisis underwent a routine general anesthetic to facilitate placement of a central venous access port for subsequent, and ongoing exchange transfusion.  Despite an uneventful intraoperative course, the child went on to have major neurological sequelae (seizures, paralysis and unresponsiveness) in the immediate postoperative period.  Diagnostic studies revealed a life threatening right-sided cerebrovascular accident affecting the right middle cerebral artery and severe bilateral internal carotid artery stenosis.

Methodology  and Result            After receiving emergency institutional approval and parental consent, inhaled nitric oxide (INO) was administered to the patient (80 parts per million) and was continued for 48 hours.  The patient was neurologically intact at the 3 hour epoch (exchange transfusion still had not occurred).  INO therapy was continued for 48 hours with the patient demonstrating near complete neurological recovery.

Conclusion            While the authors note that recovery from sickle cell stroke can and does occur in the absence of INO therapy, the physicians involved (including a consultant hematologist) suggested that the stroke occurring in this child would not have resolved spontaneously.  They further cautioned readers that this case report does not demonstrate proof of the intervention’s efficacy.



Reading this case report was a hand wringer.  Those of us who take care of children and adults with sickle cell disease and who have seen vasooclusive crisis, know that these patients are at great risk for negative outcome and pose major challenges to us during their care.  The miraculous recovery of this patient was temporally related to initiation of the INO, and though “proof” cannot be demonstrated except in carefully executed prospective trials, the attribution of efficacy is tantalizingly close at hand.

Nitric oxide (NO) is truly a remarkable molecule.  Crowned Science magazine’s “molecule of the year” in 1992, NO research was published in literally thousands of scientific articles, and in 1987 it was revealed that what we knew of as endothelial-derived relaxing factor was in fact, NO!  Up to that point (which led to the awarding of the Nobel Prize to 3 investigators in 1998) NO was largely considered an environmental pollutant emitted from automotive exhaust and industrial smoke stacks.

We now appreciate that endogenous NO possesses an impressive physiological resume, one ranging from its role as a major biological signaling molecule, to its being a primary determinant of blood vessel tone, to its service in preventing platelet adhesion.  It is an ubiquitous biological molecule that plays essential functionality in memory formation, sleep cycling and mood, and is strategically enhanced by drugs that treat erectile dysfunction in males. 

The above report is not the first publication in the anesthesia literature of INO’s efficacy in sickling occlusive phenomenon.  Atz and Wessel previously described two cases of young children who were successfully resuscitated from acute chest syndrome as a result of their underlying sickle cell disease. The children would likely have died if INO had not been employed.1

Stay tuned for more developments on this fascinating, chemically simple, yet dynamic molecule.  NO may one day find itself as a commonly administered agent when vascular tone (especially in the pulmonary circulation) must be modified.


Chuck Biddle PhD, CRNA


1.  Atz A, Wessel D.  Inhaled nitric oxide in sickle cell disease with acute chest syndrome.  Anesthesiology.  1997;87:988-990.

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

David JS, Ferreti C, Amour J, Vivien B, Eve O, Petit P, Riou B, Gueugniaud PY



Effects of bupivacaine, levobupivacaine and ropivacaine on myocardial relaxation

Can J Anesth 2007;54:208-217

David JS, Ferreti C, Amour J, Vivien B, Eve O, Petit P, Riou B, Gueugniaud PY




Purpose            The purpose of this study was to compare the cardiotoxic effects of three long acting local anesthetics on rat papillary muscle.

Background            Impaired diastolic function is a major cause of heart failure. A high percentage of patients with heart failure have normal systolic function but abnormal diastolic function. Diastolic dysfunction contributes to hemodynamic instability in operative patients. Diastolic dysfunction is associated with advanced age, hypertension, sepsis, diabetes, myocardial ischemia, and, perhaps, halogenated inhalation agents. In patients with poor myocardial performance or who are at risk for hemodynamic instability, regional blocks are often used to limit or eliminate the need for general anesthesia.

Bupivacaine is known to possess myocardial specific toxicity. Ropivacaine and levobupivacaine were developed as less cardiotoxic alternatives. But levobupivacaine has been shown to concentrate in myocardial tissue to a level 2 to 3 times that present in arterial blood. Acidosis, beta blockers, and calcium channel blockers can displace these long acting local anesthetics from plasma proteins, increasing the free fraction of circulating local anesthetic.

Methodology            This prospective, comparative, laboratory study used 83 left ventricular papillary muscles from adult Wistar rats. Papillary muscle was suspended in an oxygenated, pH adjusted, temperature controlled nutritive bath. The inotropic (cardiac contraction) and lusitropic (cardiac relaxation) effects of bupivacaine, levobupivacaine, and ropivacaine on cardiac papillary muscle were tested. Sequentially higher concentrations of each local anesthetic were tested as the experiment progressed. For each local anesthetic, the effects on papillary muscle were tested under both isometric and isotonic conditions.

Result            Sufficiently high concentrations of each local anesthetic decreased the rate at which papillary muscle relaxed after a contraction. Levobupivacaine and bupivacaine significantly increased the time until papillary muscle relaxed half way back to its baseline state after a contraction, while ropivacaine did not. Levobupivacaine had the greatest negative effect on the relaxation of cardiac muscle. In fact, when bupivacaine and ropivacaine concentrations were increased enough to reduce the rate of relaxation by approximately 40%, levobupivacaine reduced the rate of papillary muscle relaxation by 80%.

Sufficiently high concentrations of each local anesthetic reduced the contractility of cardiac muscle as well. Again, levobupivacaine had the greatest myocardial depressant effect. Increasing the available calcium had a positive inotropic effect in the presence of ropivacaine but this positive inotropic effect was attenuated in the bupivacaine and levobupivacaine groups. Reducing the rate of muscle contractions also improved contractility except in the levobupivacaine group.

Conclusion            Ropivacaine, bupivacaine, and levobupivacaine each impaired both papillary muscle contraction and relaxation. The negative inotropic and lusitropic effects were noticeably worse with levobupivacaine. In patients with diastolic dysfunction, toxic levels of these long acting local anesthetics may further impair diastolic function and produce heart failure, with levobupivacaine having the most pronounced negative diastolic effects.



In the late 1970’s there had been a rash of deaths attributed to the relatively new long acting local anesthetic, bupivacaine. After several publications1 on the problem and testimony before the US Food and Drug Administration (FDA), some limitations were placed on the maximum concentration of bupivacaine used clinically, notably in obstetric anesthesia. As it turned out, bupivacaine is more toxic in pregnant women than in non-pregnant individuals. Despite these problems, bupivacaine went on to be heavily used for a variety of local anesthetic applications in anesthesia practice. As with so many other “old” drugs, we have gained skill in “adjusting” to the low margin of safety of bupivacaine and routinely produce good patient outcomes.

Levobupivacaine and ropivacaine were introduced as safer alternatives to bupivacaine with similar pharmacodynamic profiles. Levobupivacaine is the S stereoisomer of bupivacaine. We know that with some drugs more of the toxic effects are found in one stereoisomer than in the other so this approach seemed to make sense. Ropivacaine is a purified S isomer of an analogue of mepivacaine and bupivacaine. There is good experimental and clinical evidence to show that ropivacaine has a wider margin of safety than bupivacaine.2-6

Interestingly, this wider margin of safety apparently does not extend to levobupivacaine. The graphs is the original article are quite compelling. With each of the long acting local anesthetics, as the drug concentration is increased, the heart muscle ultimately fails. In each case cardiac depression begins gradually. In each case there is a point at which the graph makes a sharp downward turn; a point where a small increase in the local anesthetic concentration results in a very large increase in cardiac depression. But in the case of levobupivacaine, that sharp downward turn usually occurred earlier, and at a lower local anesthetic concentration.

Of these three long acting local anesthetics, ropivacaine appears to have the greatest margin of safety. Levobupivacaine, initially suggested to have safety advantages over bupivacaine, may be less safe than bupivacaine by the measures used in this study. As we know, animal studies don’t always apply directly to humans. And the dynamics of local anesthetic administration does play a role in the development of local anesthetic toxicity. Overall, though, it appears that ropivacaine has the widest margin of safety of the long acting local anesthetics.


Michael Fiedler, PhD, CRNA





1.  Albright GA. Cardiac arrest following regional anesthesia with etidocaine or bupivacaine. Anesthesiology 1979; 51: 285-7

2.  Scott DB, Lee A, Fagan D, Bowler GM, Bloomfield P, Lundh R. Acute toxicity of ropivacaine compared with that of bupivacaine. Anesthesia & Analgesia 1989; 69: 563-9

3.  Knudsen K, Beckman Suurkula M, Blomberg S, Sjovall J, Edvardsson N. Central nervous and cardiovascular effects of i.v. infusions of ropivacaine, bupivacaine and placebo in volunteers. British Journal of Anaesthesia 1997; 78: 507-14

4.  Moller RA, Covino BG. Effect of progesterone on the cardiac electrophysiologic alterations produced by ropivacaine and bupivacaine. Anesthesiology 1992; 77: 735-41

5.  Ohmura S, Kawada M, Ohta T, Yamamoto K, Kobayashi T. Systemic Toxicity and Resuscitation in Bupivacaine-, Levobupivacaine-, or Ropivacaine-Infused Rats. Anesth Analg 2001; 93: 743-8.

6.  Hasselstrom LJ, Mogensen T. Toxic reaction of bupivacaine at low plasma concentration. Anesthesiology 1984; 61: 99-100

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Apiliogullari S, Keles B, Apiliogullari B, Balasar M, Yilmaz H, Duman A


Comparison of diphenhydramine and lidocaine for prevention of pain after injection of propofol: a couble-blind, placebo-controlled, randomized study

Eur J Anaesthesiol 2007;24:235-238

Apiliogullari S, Keles B, Apiliogullari B, Balasar M, Yilmaz H, Duman A


Purpose            The purpose of this study was to compare diphenhydramine (Benedryl) and lidocaine for preventing pain during intravenous (IV) induction with propofol.

Background            Propofol produces pain during IV injection in up to 90% of patients. A number of methods of reducing or eliminating this pain have been tried with mixed results. The most effective method to date appears to be the injection of IV lidocaine with a venous tourniquet in place before tourniquet release and propofol injection. Adding lidocaine to propofol negatively affects the stability of the propofol solution and is not recommended.

Propofol has a strong irritant effect on the vascular intima. It may cause pain through activation of the kallikrein-kinin system, bradykinin release, and increased vascular permeability.

Diphenhydramine has a local anesthetic effect and 1% diphenhydramine (10 mg/mL) has been shown to be as effective a local anesthetic as 1% lidocaine. Diphenhydramine 10 mg has previously been shown to reduce propofol injection pain.

Methodology            This randomized, double-blind, placebo-controlled study included 180 patients divided into three groups. No patient was premedicated. Each had a 20 gauge IV placed in an antecubital vein. Group I received normal saline placebo. Group II received 40 mg lidocaine. Group III received 20 mg diphenhydramine. A proximal venous tourniquet was applied and the study drug was injected over five seconds. One minute after injection the tourniquet was released and the patient received 25% of the calculated induction dose of propofol over five seconds. Patients then rated their pain as none, mild, or severe. Induction was then completed with the balance of the calculated propofol dose.

Appropriate statistical tests were used in a more rigorous manner than commonly seen.

Result            The study doses of both lidocaine and diphenhydramine reduced the pain of propofol injection to a similar extent. Propofol injection pain was both clinically and statistically significantly decreased compared to placebo. In the placebo group, 42% of patients experienced pain (11 mild, 14 severe). Only 2 patients in the lidocaine group (3.3%) and 3 patients in the diphenhydramine group (5%) experienced any pain and only one of those five patients experienced severe pain.

Conclusion            Intravenous diphenhydramine prevents propofol injection pain as effectively as lidocaine.



It is interesting to note the “other” uses for so many of the drugs we use for their “correct” purposes. Meperidine (Demerol), for example, was developed in a search for a synthetic atropine. While it did increase heart rate a bit, meperidine turned out to be a far better opioid analgesic than vagolytic.

I’m glad to see these authors point out in their literature review that adding lidocaine to propofol results in undesirable physiochemical changes. A few authors have strongly recommended against mixing lidocaine with propofol but the practice seems to be quite common, despite limited effectiveness at reducing propofol injection pain. I was surprised the investigators chose to use an antecubital vein as the injection site for their study. Their extremely low incidence of pain with propofol injection after lidocaine or diphenhydramine pretreatment would be much more impressive if the injection had been in the hand or even the forearm.

I don’t expect we’ll all be changing to diphenhydramine to prevent propofol injection pain right away but I have tried it a time or two. So far I can’t say if I’ve seen improvement over my regular methods to reduce propofol injection pain. I did have one patient who was slow to awaken in recovery but I don’t know that diphenhydramine was the cause At the very least it’s nice to know one more way to skin the cat.


Michael A. Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Foxall G, McCahon R, Lamb J, Hardman JG, Bedforth NM


Levobupivacaine-induced seizures and cardiovascular collapse treated with intralipid®

Anaesthesia 2007;62:516-518

Foxall G, McCahon R, Lamb J, Hardman JG, Bedforth NM



Purpose            The purpose of this report was to describe a case of systemic levobupivacaine local anesthetic toxicity treated with Intralipid®.

Background            Levobupivacaine is the S enantiomer of bupivacaine. It was developed in the search for a long acting local anesthetic with less cardiovascular toxicity than bupivacaine. Nevertheless, animal studies suggest that levobupivacaine cardiac arrest is more difficult to treat than cardiac arrest caused by ropivacaine.

Local anesthetic induced cardiovascular collapse has been treated successfully with lipid administration in animals. Intralipid® 20% has been reported in the resuscitation of two patients with cardiac arrest due to local anesthetic toxicity. Lipid infusion has also been shown to increase the dose of bupivacaine required to produce asystole in rats.

Inadvertent intravascular injection has been reported following presumed lumbar plexus block even when no blood could be aspirated before or during intermittent injection. Failure to aspirate blood during lumbar plexus block is known not to preclude intravascular injection.

Methodology            This case report involved a 75 year old, ASA III, 85 kg woman with a femoral neck fracture. Her history included COPD, shortness of breath with minimal exertion, and pathologic q waves in several V leads. Her preoperative BP was 130/70. She was given midazolam 1 mg and alfentanil 0.5 mg prior to multiple unsuccessful attempts at a subarachnoid block. A lumbar plexus block was subsequently performed with 20 mL of 0.5% levobupivacaine. Aspiration for blood was negative before injection and during incremental injection of the full dose.

Result            Upon completion of the block the patient became unresponsive and seized. She was positioned supine and given 100% oxygen by face mask. Her first BP was 60/40 and her QRS complexes were widening and becoming progressively smaller. No radial pulse was palpable. Heart rate was unchanged from baseline at 110 bpm. Her skin appeared mottled. As time passed her ECG changes continued to progress with QRS complexes becoming wider and their voltage (amplitude) becoming lower. Prior to endotracheal intubation the patient was given metaraminol 0.5 mg, propofol 80 mg, and succinylcholine 100 mg. Within four minutes of the levobupivacaine injection, 100 mL of 20% Intralipid® was rapidly infused. Next, an arterial line was placed. Blood pressure by arterial line was 90/60. The pH on an ABG drawn at that time was 6.97. After the Intralipid® was infused, the QRS complexes rapidly returned to normal. Inhalation anesthesia was begun and her hip fracture repaired. Two hours later, her 12-lead ECG in recovery was unchanged from preoperative baseline and she recovered uneventfully.

Conclusion            Attempted lumbar plexus block with levobupivacaine resulted in an intravascular injection, seizure, ECG changed, and cardiovascular collapse. Administering 100 mL of 20% Intralipid® prior to cardiac arrest facilitated a successful resuscitation.



Information about using lipids to treat systemic local anesthetic toxicity has been emerging for several years now. Given the severity of systemic toxicity seen with some local anesthetics and the cumulation of bupivacaine and levobupivacaine in myocardial tissue, this information is good news indeed. This report was especially encouraging because: 1) it was a clinical case rather than an animal study, 2) the patient involved was elderly and already had cardiac disease (approaching a “worst case” scenario), 3) it showed the value of administering Intralipid® before cardiac arrest, 4) the result was fairly profound, and 5) the outcome was good.

Not having been there to experience the event, and, thus, undoubtedly lacking some information, I am reluctant to criticize the care described in this case report. I must say, though, that I find the decision to give propofol prior to intubation bewildering. The cardiovascular depressant effects of propofol would seem to contraindicate it in an elderly, debilitated patient experiencing cardiovascular collapse following systemic local anesthetic toxicity. While one does well to try to prevent awareness during intubation, this patient had not regained consciousness since collapsing. She was hypotensive, which makes recall less likely, and she was presumably post ictal at best following a seizure only moments earlier. Recall seems improbable under these circumstances. Perhaps the authors we hoping that the lipid in propofol would slow the progression of the local anesthetic toxicity and gave the metaraminol first to try to offset the cardiovascular effects of the propofol. If that was the case, we will never know if propofol helped or not because Intralipid® was infused moments later.

It appears certain that this patient benefited from the Intralipid® infusion. I wonder if we are approaching a time when Intralipid® should be available whenever major local anesthetic blocks are performed.


Michael Fiedler, PhD, CRNA



For more information about the use of 20% Intralipid in the treatment of systemic local anesthetic toxicity, including a proposed treatment protocol, go to

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Regional Anesthesia

Brull R, McCartney CJL, Chan VWS, El-Beheiry H



Neurological complications after regional anesthesia: contemporary estimates of risk

Anesth Analg 2007;104:965-974

Brull R, McCartney CJL, Chan VWS, El-Beheiry H




Purpose            The purpose of this article was to attempt to estimate the incidence of complications following central neuraxial blockade and peripheral nerve blockade using an aggregate of previously published studies.

Background            Regional anesthesia has been associated with lower morbidity and mortality, better postoperative analgesia, and lower cost than general anesthesia. Patients are, however, often concerned that regional anesthesia may result in nerve injury. Many of the studies reporting the incidence of neurologic injury following regional anesthesia are quite old and may reflect regional anesthetic techniques that are no longer used. Patient risk factors for neurologic injury also change over time.

Methodology            English language studies published between 1995 and 2005 were included. No pediatric studies were included. Transient neurologic symptoms (TNS) associated with subarachnoid blocks and epidural hematoma were not included. Permanent nerve injury was defined as a neurologic deficit that lasted over one year. The Medline search located 235 studies; 32 studies met inclusion criteria and were aggregated in this report. All included studies were cohort or case-control studies.

Result            In almost all cases, there was considerable variability in the reported incidences of specific complications between studies. The single largest study in the analysis, which included over 1.2 million subarachnoid blocks and 450,000 epidural blocks, reported the overall incidence of “severe” neurologic complications to be 0.004%. When epidurals placed in pregnant women were excluded, the incidence of severe neurologic complications was 0.028%, seven times higher.

Overall, the analysis suggests that there is a higher risk of radiculopathy or peripheral neuropathy with subarachnoid block (0.0378%) than with epidural block (0.0219%). Rates of permanent neurologic injury ranged from 0% to as high as 0.042% following subarachnoid block and 0.076% following epidural block. The incidence of paraplegia was exceedingly rare; 0.0006% (0.06 per 10,000) for subarachnoid blocks and 0.0009% (0.09 per 10,000) for epidural blocks.

Most peripheral nerve blocks studied were of the upper extremity, specifically brachial plexus blocks. Interscalene blocks were associated with the highest incidence of transient nerve injury at 2.84%. Axillary brachial plexus blocks were associated with a lower incidence of transient nerve injury, 1.48%. Supraclavicular blocks were associated with the lowest incidence of transient nerve injury, 0.03%. Permanent peripheral nerve injury was exceedingly rare following a brachial plexus block.

Conclusion            When epidural hematomas are excluded, permanent nerve injury is exceedingly rare in adult patients. The rates of temporary radiculopathy and peripheral neuropathy are less than 0.04% following central neuraxial blockade and less than 3% following peripheral nerve blockade.



I believe I can tell my patients that permanent nerve injury is exceedingly rare following regional anesthesia. In the studies included in this analysis only 11 of 2,577,425 patients developed paraplegia following subarachnoid or epidural block. But I’d really like a figure that included epidural hematomas. In deciding to omit studies looking specifically at epidural hematomas the authors reasoned that studies specific to this complication were published elsewhere. Yet, my clinical practice is best served with an overall complication rate; not an overall rate with the risk of epidural hematomas left out. I don’t get it. Nevertheless, there is good information here suggesting that the risk of a major complication following regional anesthesia is quite low. It is also reassuring, though not surprising, that the incidence of severe neurologic complications following epidurals in pregnant women is markedly lower than in the general population.

While this analysis may be among the best information we have available we should be cautious about quoting these numbers as exact complication rates for two reasons. First, many of the studies upon which it was based collected data through voluntary reporting and may not be an accurate estimate of the incidence of some complications. Second, the confidence intervals for these average complication rates were often quite wide. A confidence interval is a way of expressing how sure we are that the average number reported is accurate. A 95% confidence interval really says, “we’re 95% sure the real average number is between these two numbers.” When the confidence intervals are very wide, as they were here – sometimes differing by a factor or 10 or more between the low and high end, we are really saying we’re not so sure what the number really is.

In last month’s issue of Anesthesia Abstracts we summarized and commented on an article (cited at the end of this comment) that examined the frequency with which the risks of regional anesthesia were discussed with patients. Based upon this study of the risks of regional anesthesia I’m wondering if we shouldn’t rethink how we discuss the risks of nerve injury with our patients.


Michael A. Fiedler, PhD, CRNA




The reader may want to revisit a previous Abstract & Comment from Volume 1 Number 1 of Anesthesia Abstracts (March 2007) which is topically related to this one. The article reviewed is titled Disclosure of risks associated with regional anesthesia: a survey of academic regional anesthesiologists. The complete citation for the article is Reg Anesth Pain Med 2007;32:7-11.

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007

Ababou A, Marzouk N, Mosadiq A, Sbihi A


 The effects of arm position on onset and duration of axillary brachial plexus block

Anesth Analg 2007;104:490-491

Ababou A, Marzouk N, Mosadiq A, Sbihi A


Purpose            The purpose of this study was to compare the effects of two different arm positions during the set up of axillary brachial plexus block on the onset and duration of the block.

Background            Previously, most studies of brachial plexus block have sought improvements in block success, speed of onset, or duration through the number of nerves identified or the intensity of nerve stimulation accepted before injecting local anesthetic. The stated assumption has been that leaving the arm abducted after local anesthetic injection would reduce the proximal spread of local anesthetic due to pressure on the neurovascular sheath by the humeral head.

Methodology            This prospective, randomized, double-blind study included 64 patients scheduled for axillary brachial plexus block. The block was performed with the arm abducted 90? at the shoulder and the arm flexed 90? at the elbow with the hand above the patient’s head. A 22 gauge 5 cm needle was used with a nerve stimulator to locate the median, radial, and musculocutaneous nerves. Ten mL of local anesthetic (2% lidocaine and 0.5% bupivacaine mixed half and half) was injected when the desired motor response was observed at a 0.5 mA stimulating current. An additional 5 mL was deposited subcutaneously. Immediately after completion of local anesthetic injection, half the patients left their arm in the position used to place the block (the abduction group). The other half had their arms positioned straight down along the side of their body (the adduction group). These positions were maintained until the block had completely set up. A successful block was defined as complete sensory and motor block of all nerves in the brachial plexus. If any one nerve retained sensory or motor function the block was a failure for the purposes of the study.

Result            The block success rate was 100% in the abduction group and 90.6% in the adduction group (P=0.078). The time required for complete sensory and motor block of each individual nerve was shorter in the abduction group than the adduction group (P <0.001). Similarly, the time required for complete set up of the block was shorter in the abduction group (17?5 minutes) than the adduction group (22?10) (P<0.05).

The duration of sensory block was over two hours longer in the abduction group: 325?53 minutes in the abduction group versus 189?67 minutes in the adduction group (P<0.001).

Conclusion            After performing an axillary brachial plexus block, leaving the arm abducted until the block was completely set up increased both the speed of onset and the duration of the block.



This fairly simple study challenges the conventional wisdom about how best to achieve proximal local anesthetic spread when performing an axillary block. Using a fairly conservative methodological and statistical approach, they clearly show clinically significant improvements in both the onset and duration of the block by simply leaving the arm abducted until the block sets up. Perhaps even more impressive, these results were achieved with only a 30 mL injection. I would love to see this study repeated with a single injection into the neurovascular sheath and with a higher volume of local anesthetic. If leaving the arm abducted after injection facilitates the proximal spread of local anesthetic to the degree that this study seems to indicate perhaps we can reduce the number of injections (at three different nerves in this study) and still achieve their high success rate, faster onset, and longer duration.


Michael Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 2, May 31, 2007