ISSN NUMBER: 1938-7172
Issue 1.1

Michael A. Fiedler, PhD, CRNA

Contributing Editor:
Chuck Biddle, PhD, 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





















Thank you for being an early adopter and welcome to the first issue of Anesthesia Abstracts. Our goal is to make it as quick and easy as possible for you to keep informed about developments that impact your practice of clinical anesthesia. We are interested in hearing what you want to know more about. Please share your thoughts and ideas about how we can make Anesthesia Abstracts more valuable to you in "Blog with the Editors" or through the "Contact Us" link.

Over the next few monthly issues you will see contributions from a growing number of CRNA contributing editors. Ultimately there will be a core group of about 10 frequent contributors whose abstracts and comments appear regularly throughout the year. There will also be regular contributions from a variety of Guest Editors. This will add a diversity of viewpoints and great expertise to the abstracts and comments you read.

Best Regards,

Michael A. Fiedler


Jones PM, Turkstra TP, Armstrong KP, Armstrong PM, Cherry RA, Hoogstra J, Harle CC


Effect of stylet angulation and endotracheal tube camber on time to intubation with the GlideScope

Can J Anesth 2007;54:21-27

Jones PM, Turkstra TP, Armstrong KP, Armstrong PM, Cherry RA, Hoogstra J, Harle CC



Purpose            The purpose of this study was to determine the better angulation of an endotracheal tube for use with the Glidescope videolaryngoscope, an acute 90? bend or a 60? bend that mimics the curvature of the Glidescope blade.

Background            The Glidescope videolaryngoscope has been shown by several studies to provide an improved view of the glottis in patients who are difficult to intubate using a standard laryngoscope. Despite an improved view, advancing the endotracheal tube (ETT) through the glottis is sometimes difficult. This difficulty may be the result of the angle at which the ETT approaches the trachea. Several ETT angulations have been recommended as alternatives to the 60? angle recommended by the Glidescope manufacturer. The best ETT angulation is unknown.

Methodology            This prospective study compared the time to intubation, success rate, ease of intubation, and frequency of use of external airway manipulation with two different ETT angulations during Glidescope laryngoscopy. Using a lubricated stylet, the 60? ETT was bent to an angle of 60? at a point 8 cm from the distal end of the ETT. The bend was gradual, mimicking the configuration of the Glidescope blade. The 90? ETT was bent to an acute 90? angle 8 cm from the distal end of the ETT.

The study was carried out in 196 elective surgery patients. Those with a known difficult airway were excluded from the study. After induction of general anesthesia and paralysis, laryngoscopy was performed by 53 different anesthesia residents and attending anesthesiologists who had previously completed at least three intubations with the Glidescope. The time to intubation was recorded by an observer who could not see the configuration of the ETT being used. Failure to intubate was defined as an intubation that took longer than 150 seconds. Ease of intubation was graded by the laryngoscopist on a 100 mm visual analogue scale (VAS) where 0 was easy and 100 was hard.

Result            Of the 196 patients included in the study, 14 had Mallampati scores of III and none had a score of IV. The 14 Mallampati III airways were evenly distributed between the two groups. The airways of 182 (93%) of the study patients were graded Mallampati I or II.

The average time to intubation was 54.4 (?28.2) seconds in the 60? ETT group and 47.1 (?21.2) seconds in the 90? ETT group (P=0.042). There were three failures in the 60? ETT group and one in the 90? ETT group. All four failures occurred in patients with a grade I view of the glottis. Intubation with the 90? ETT was graded as easier than the 60? ETT with average VAS values of 16.4 and 27.3 respectively. External airway manipulation was used in 18 of the 60? ETT patients and 5 of the 90? ETT patients (P=0.003).

Conclusion            The 90? ETT bend resulted in faster and easier intubation.



The authors had a good idea with this study. I suspect everyone who has seen the cords with a Glidescope has had trouble getting the tube to go there at least some of the time. They effectively asked the question, “if we bend the tube differently, will it go?” It is curious to me that they tested their 90? bend in patients with good airways. When 93% of the study patients had a Mallampati class I or II airway, I wonder if the study results are applicable to patients who are difficult to intubate. Those are the patients in whom I use the Glidescope. Let me be clear, though, I’m not saying the results are not applicable to patients with a difficult airway, I’m only expressing concern that they may not be.

Intubation was accomplished 17 seconds faster with the 90? ETT. It is difficult to judge how important this is clinically. Still, any progress towards securing an airway more quickly is a step in the right direction.

The authors made a common mistake when analyzing the visual analogue scale (VAS) scores for the ease of intubation. Although you can measure the distance between the baseline and the recorded score on the VAS with a ruler, studies have shown that VAS data is not linear. The VAS value tells you that a score is more or less than another score but it doesn’t tell you how much more or less. The difference between 10 and 20 is not the same as the difference between 50 and 60. As a result, the “average” VAS scores reported and statistically analyzed are meaningless. I do not accept the authors’ conclusion that the 90? ETT was statistically significantly easier to use based upon their analysis. There is, however, some strong circumstantial evidence that the 90? ETT was easier to intubate with. While there were only four intubation failures, three of them occurred in the 60? ETT group. A look at a plot of the “ease of intubation” VAS scores for both groups shows a clump of most scores near the “easy” end in the 90? ETT group while the scores extend all the way to the “hard” end of the VAS scale in the 60? ETT group. Lastly, airway manipulation was used far less often in the 90? ETT group than in the 60? ETT group (statistically significant with an appropriate test). Taken together this information does suggest to me that the laryngoscopists in this study thought it was easier to intubate patients with normal airways with the 90? ETT.

While this study was a good idea, I’m not sure we can solve the problem of difficult ETT advancement during Glidescope intubations by bending things. Just as the Magill forceps was invented to aid in nasoendotracheal tube placement, I wonder if we don’t need a new instrument to aid in the placement of the ETT during Glidescope laryngoscopy.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Xue F, Zhang G, Liu J, Li X, Sun H, Wang X, Li C, Liu K, Xu Y, Liu Y

A clinical assessment of the glidescope videolaryngoscope in nasotracheal intubation with general anesthesia

J Clin Anesth 2007;18:611-615

Xue F, Zhang G, Liu J, Li X, Sun H, Wang X, Li C, Liu K, Xu Y, Liu Y


Purpose            The purpose of this study was to evaluate the effectiveness of the Glidescope during nasotracheal intubation and to compare its effectiveness with a standard Macintosh laryngoscope in patients predicted to have a difficult airway.

Background            The Glidescope video laryngoscope (Saturn Biomedical Systems, Burnaby, British Columbia, Canada) is a reusable curved laryngoscope with a built in camera which displays the image on an 18 inch liquid crystal video screen on a stand next to the user. The laryngoscope blade is angled anteriorly to a much greater extent than the angle on a Macintosh laryngoscope blade. The camera is pointed anteriorly and covers a 50? to 60? field of view. The laryngoscope portion of the Glidescope is more bulky than a traditional laryngoscope blade.

Methodology            This prospective, experimental and observational study 1) compared the view obtained with the Glidescope and a standard Macintosh laryngoscope during nasotracheal intubation in patients with a predicted difficult airway and 2) compared the success or failure of nasotracheal intubation with the two laryngoscopes. It also made observations about the ease of use and problems encountered with both methods of laryngoscopy for nasotracheal intubation.

The study included 156 healthy adults scheduled for elective plastic or intraoral procedures who required general anesthesia and nasotracheal intubation. The Mallampati classification (III or IV) and thyromental distance (≤6 cm) were used to predict difficult airways. All airway assessments were performed by the same individual. All patients were premedicated with midazolam and scopolamine. Their nasal mucosa was prepared with local anesthetic and vasoconstrictor. Both the Glidescope laryngoscope and the nasal endotracheal tube (NETT) were lubricated. After induction of general anesthesia and paralysis the head was placed in the sniffing position.

Nasal intubation was first performed with the Glidescope. In patients predicted to have a difficult airway the NETT was then withdrawn until the tip was in the nasopharynx. Next laryngoscopy was performed with a #3 Macintosh laryngoscope. The laryngoscopic view produced by both methods was assessed using the Cormack and Lehane grading system (grade I through IV). After the second laryngoscopy the NETT was again placed in the trachea by the most expeditious means. Failure of the Glidescope assisted nasotracheal intubation occurred when one of the following criteria were met: 1) the NETT could not be aligned with the glottis in three attempts, 2) Glidescope intubation was abandoned when laryngeal views were grade III or IV, or 3) oxygen saturation decreased to 95% or less during a prolonged intubation attempt.

All laryngoscopies were performed by one of three clinicians who had never before used the Glidescope.

Statistical analysis was adequate for the data and study goals.

Result            All laryngeal views obtained with the Glidescope were grade I (113 cases, 72%) or II (43 cases, 28%). Nasotracheal intubation using the Glidescope was successful in one attempt in 98% of patients. In these patients, the average time to intubation was 52.7 ?12.3 seconds (range, 28-98 seconds). Nasotracheal intubation failed in three patients using the Glidescope. In two of these patients nasotracheal intubation was successful using a standard Macintosh blade and Magill forceps. In one case a fiberoptic bronchoscope was required. The greater anterior angulation of the Glidescope blade compared to the Macintosh blade makes using Magill forceps to assist in nasoendotracheal intubation quite difficult.

Of the 156 patients enrolled in the study, 36 met the criteria for potentially difficult intubation. All Glidescope laryngeal views in these patients were grade I or II. The corresponding Macintosh laryngeal views were grade I in 9 patients (25%), grade II in 6 patients (17%), grade III in 18 patients (50%),and grade IV in 3 patients (8%). The frequency of laryngeal views III or IV (difficult intubation) was significantly greater in the Macintosh group (58%) than the Glidescope group (0%) (P<0.05).

Conclusion            The Glidescope provides a better laryngeal view in patients predicted to have a difficult airway than does the Macintosh laryngoscope during nasotracheal intubation. Novice Glidescope users with traditional laryngoscopy skills can achieve improved visualization in a high percentage of patients and intubate the trachea in a reasonable period of time.


The best part of this article isn’t the research, it’s the volumes of “how to” comments about using the Glidescope to assist with nasal intubation! The authors provide paragraph after paragraph of information about how to manipulate the NETT and/or the airway to get the tube to go into the glottis. They provide tips on how to deal with a tube that hangs up on tracheal rings due to the acute angle at which it enters the trachea. They even identify the anatomic characteristics of the patients they could not intubate with the Glidescope. If you are not an expert in using the Glidescope for nasal intubation (as I am not) you need to read the full version of this article.

As to the research, there are several take home messages. In this study, novice Glidescope users were able to accomplish nasal intubations on the first try 98% of the time and in an average of 53 seconds. (The longest time was 98 seconds.) While the Glidescope is fairly easy to learn to use and often provides a better view of the larynx than a traditional laryngoscope in a difficult airway, it is not always better than a Macintosh blade for nasal intubations. This is probably due to the fact that one can use a Magill forceps with a Macintosh blade but it is exceedingly difficult to use a Magill forceps with the Glidescope.

The Glidescope is a great tool and was easy for me to learn to use. One could make a good case that every anesthesia department needs at least one. It makes some difficult intubations so much easier that it is tempting to think, “I can intubate anyone with this thing, it’s all I’ll ever need!” This study reminds us that this isn’t true. We still need skill with a wide range of methods to control the airway. The Glidescope is simply another good method to know.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Adamzik M, Groeben H, Rarahani R, Lehmann N, Peters J


Intravenous lidocaine after tracheal intubation mitigates bronchoconstriction in patients with asthma

Anesth Analg 2007;104:168-172

Adamzik M, Groeben H, Rarahani R, Lehmann N, Peters J



Purpose            The purpose of this study was to compare airway resistance after endotracheal intubation in patients with asthma given intravenous lidocaine versus placebo.

Background            Patients with asthma experience bronchoconstriction following endotracheal intubation more often than those with normal airways. In a study of closed insurance claims, 90% of patients who experienced bronchoconstriction died or had brain damage. The bronchoconstriction usually occurred at induction of general anesthesia. In some studies, intravenous (IV) lidocaine has been shown to attenuate the bronchoconstriction responses to a number of stimuli. In other studies, the effect was not observable. In those studies where the effects of lidocaine on bronchoconstriction were not seen it may have been hidden by the attendant bronchodilatory effects of induction or inhalation anesthetic agents.

Methodology            In this prospective, randomized, double-blind, placebo controlled study 30 adult patients with asthma scheduled for elective procedures requiring general anesthesia were intubated and then given either IV lidocaine or placebo. Of these 30 patients, 13 took inhaled bronchodilators daily, 5 took only antihistamines prn, 13 took inhaled steroids, and 1 oral steroids. Twenty patients, 10 in each group, were active cigarette smokers.

All patients were premedicated with midazolam and anesthesia was induced with 0.3 mg/kg etomidate, 5 ?g/kg fentanyl, and 0.6 mg/kg rocuronium. After endotracheal intubation, oxygen and 50% nitrous oxide was administered via standardized mechanical ventilation for the 15 minute study period. Baseline airway resistance data was collected after intubation because the ETT was necessary to make the measurements.

Blood lidocaine levels were drawn and airway resistance data collected at <1, 5, 10, and 15 minutes after intubation. After the 5 minute airway resistance data was collected, patients were given either lidocaine 2 mg/kg over five minutes followed by 3 mg/kg/hour or a placebo infusion. The investigator did not know whether the infusion was lidocaine or placebo.

Result            No patients in either group had wheezing on auscultation prior to induction of general anesthesia. Immediately after intubation, the average airway resistance across all patients in both groups was 23 ?12 cm H2O/sec/L. On average, airway resistance increased 38% over the study period (minute 1 to minute 15) in the placebo group and decreased 26% in the lidocaine group (P<0.01). Airway resistance continued to increase over time in most placebo patients while it almost uniformly decreased over time in lidocaine patients. The overall difference in airway resistance between placebo and lidocaine patients averaged approximately 50%.

Peak lidocaine concentrations averaged 3.3 ?2.2 ?g/mL with two patients achieving levels at or above the 5 ?g/mL toxic threshold in awake patients at both 5 minutes and 10 minutes, but not at 15 minutes.

Conclusion            The IV lidocaine loading dose and infusion used in this study statistically and clinically significantly attenuated the increase in airway resistance seen in asthma patients following endotracheal intubation.



I’ve seen lidocaine used for all sorts of things over the years, some well demonstrated by research and others not. Lidocaine has been shown to briefly attenuate the cough reflex to laryngoscopy and endotracheal intubation1, 2 which indicates that it addresses airway reactivity. This study looked at lidocaine’s effects on airway resistance in patients known to react to airway stimuli with increased resistance, those with asthma. The tightly controlled methodology in this study was impressive and the investigators were wise to use a lidocaine infusion. Most of a bolus dose of IV lidocaine is removed from the circulation by a single pass through the lungs.3

In my experience, bronchoconstriction is probably a more frequent, and certainly less deadly, problem during general anesthesia than reflected in the closed claims study cited in this article. Most commonly, bronchoconstriction is dealt with by anesthesia without any ensuing morbidity or mortality. I’ll bet, though, that if we had accurate airway resistance data on all patients we’d see that it went up more frequently than we know about without causing any sort of clinical problem. I suspect that inhalation agents and sufficiently deep anesthesia prevent increased airway resistance from being a clinical problem more times than we know. If I’m right the information in this study, while important, may have limited clinical applicability.

The investigators in this study avoided anesthetic agents that produced bronchodilation, undoubtedly so the effects of the lidocaine could be clearly seen. They achieved that goal at the cost of removing the study at least one step further from the most common clinical scenario. And I wonder if patients were not a bit light after 15 minutes of only 50% nitrous oxide despite the fentanyl bolus at induction. If they were, the bronchoconstriction in the placebo group may have been magnified a bit by the light anesthesia. While I understand that answering the question would necessitate including a much larger number of patients, what I think we really want to know is, “does lidocaine reduce airway resistance in asthma patients when inhalation agent and adequate anesthetic depth isn’t enough?” Nevertheless, this is a good study that does show us that an infusion of lidocaine is an effective way to reduce airway resistance in asthma patients after endotracheal intubation.

The authors didn’t say why they chose to use non-parametric statistics exclusively despite generating cardinal data. It may have had to do with the way the data was distributed. But, the fact that they were able show such significance with non-parametric tests only makes the results stronger.


Michael A. Fiedler, PhD, CRNA




1            Yukioka H, Yoshimoto N, Nishimura K, Fujimori M. Intravenous lidocaine as a suppressant of coughing during tracheal intubation. Anesth. Analg. 1995;64:1189-1192.

2            Yukioka H, Hayashi M, Terai T, Fujimori M. Intravenous lidocaine as a suppressant of coughing during tracheal intubation in elderly patients. Anesth. Analg. 1993;77:309-312.

3            Bokesch PM, Castaneda AR, Ziemer G, Wilson JM. The influence of a right-to-left cardiac shunt on lidocaine pharmacokinetics. Anesthesiology. 1987;67:739-744.

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007


Puskas JD, Williams WH, Mahoney EM, et al


Off pump versus conventional coronary artery bypass grafting:  Early and 1-year graft patency, cost, and quality of life outcomes  A randomized trial

JAMA 2004;291:1841-1849

Puskas JD, Williams WH, Mahoney EM, et al



Purpose            Outcomes for off pump coronary artery bypass (OPCAB) must be compared with those for coronary artery bypass surgery utilizing cardiopulmonary bypass (CABG/CPB) so that informed decision making can be made.

Background:            New techniques and technologies are constantly being introduced that offer the hope that patient outcomes may be improved with lower risks of complications and reduced cost. In the domain of cardiac surgery for vessel disease, OPCAB is the most recent major advance, yet its safety and efficacy compared to traditional CABG/CPB is still being debated.

Methodology            A randomized control trial of consecutive patients was performed with 100 patients in the OPCAB group and 100 patients in the CABG/CPB group. Patients were similar in both groups and had otherwise similar management in terms of their intraoperative and postoperative care. Patients in the study were excluded if during the course of their care it was determined that severe aortic disease or renal failure occurred. There were 189 patients surviving at one year, with 178 who were successfully followed-up at 1-year.

Result            The groups did not differ in the incidence of death, infarction, stroke, readmission rates or quality-of-life measures. The overall cost difference between the procedures favored the OPCAB group as being $1955.00 less expensive.

Conclusion            Little to no differences were observed in the major outcome measures when OPCAB was compared to CPB in this one center study.



Nurse anesthetists observe many innovations that are availed patients. Innovation is rarely an isolated phenomenon and consequences, good and bad, attend every new foray into medical and nursing intervention. “Starfish” and “Octopus” suction retractors manipulate the heart into bizarre presentations in the chest and can have profound hemodynamic consequences that can prove very challenging to us. The need to control the rate (to minimize movement and preserve supply/demand relationships) places great demand on the anesthesia provider. The need for constant vigilance, as opposed to the “safety net” of the traditional bypass procedure, is clearly in evidence with OPCAB. Therefore, studies that reveal outcome from new procedures, or modification of existing procedures are clearly an essential element of our contemporary focus on evidence-based care.

This study helped me to understand the validity and logic of OPCAB, and represents a good example of a randomized trial, the gold standard of modern clinical research. Such studies are the most demanding of all clinical research to perform. In applying the results to one’s practice, it should be noted that this study did not include emergency cardiac surgery, did not include patients with balloon assist devices, and tracked cases where the surgeons were experienced and ostensibly highly skillful. That there was virtually no differences between the groups in the major outcomes studied (except for the very small difference in cost) came as a surprise to me. Given the relative “newness” of OPCAB, this might change in the future, in much the same way as has been observed with significant decreases in length of stay and cost associated with many of the less invasive intraabdominal procedures that we provide anesthetic care for.


Chuck Biddle, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Equipment & Technology

Larson ER, Nuttall GA, Ogren BD, Severson DD, Wood SA, Torsher LC, Oliver WC, Shirk Marienau ME


A prospective study on anesthesia machine fault identification

Anesth Analg 2007;104:154-156

Larson ER, Nuttall GA, Ogren BD, Severson DD, Wood SA, Torsher LC, Oliver WC, Shirk Marienau ME


Purpose            The purpose of this study was 1) to determine if there had been any improvement in the ability of anesthesia clinicians to detect anesthesia machine faults compared to earlier studies and 2) to discover whether the ability to detect anesthesia machine faults correlated with the number of years in anesthesia practice.

Background            Previous studies have concluded that human error was a primary factor in 65% of anesthetic incidents and that failure to perform an anesthesia machine check was commonly involved. In a 1984 study of 190 anesthesia providers, an average of only 2.2 anesthesia machine faults were identified out of a possible 5.

Methodology            This prospective observational study examined the number of anesthesia machine faults detected by volunteers attending a national anesthesia meeting. Five faults were created on an Ohmeda Excel 210SE. The faults were: 1) a gas leak in the water trap on the CO2 absorber head, 2) an empty oxygen tank, 3) a stuck exhalation valve, 4) absence of backup battery power, and 5) a non-functional oxygen – nitrous oxide fail-safe. Ten minutes was allowed to check the machine.

Result            Of the 87 participants, 55% worked in an academic setting and 45% in private practice. On average, 3.1 of the 5 faults were discovered. Those with more than 7 years experience discovered fewer faults (2.3) than those with less experience. Just over 25% of all participants failed to detect both the gas leak and the empty oxygen cylinder. Three participants found none of the faults. Only 10 participants found all 5 faults.

Conclusion            The ability of the anesthesia providers represented in this study sample to detect significant, safety related anesthesia machine faults was unacceptably low but it may indicate an overall improvement since 1984.



It seems to be fairly well documented that failure to complete a machine check is often a factor in critical incidents that involve anesthesia equipment. Despite the development of the FDA checklist, I have to wonder how many anesthesia providers faithfully perform a complete anesthesia machine check at the start of each day and recheck critical tests, such as circuit pressure and suction, before the start of each case. I was stunned that 25% of participants in this study failed to find both a circuit leak and an empty oxygen tank. One can only hope that this does not generalize to 25% of all clinical anesthesia providers. I was dumbfounded that three participants (3.4%) failed to find any of the machine faults. My hat is off to those 10 who found all five faults (I’d like to think I would have been among them).

The authors correctly point out several limitations of this study. The first was that they did not provide the participants with an FDA checklist. It seems reasonable to assume that more faults would have been found if the participants had worked from the FDA checklist. It is easy and cheap to keep a copy of the checklist on every anesthesia machine. We should make sure every machine has one and use it. Next, it is entirely possible that participants talked with each other about the faults they found; one participant alerting the next about what to look for. Lastly, the fact that they were asked to find machine faults may have heightened the participants vigilance compared to routine daily practice. These last two limitations would tend to bias the study toward finding more faults than participants would otherwise find, so, if anything, the results may be better than the real world. This is another scary thought.

For a variety of reasons, I am not yet a proponent of retesting professionals throughout their careers. But, I must admit, these findings lead me to rethink that position. As much as we might all hate it, perhaps we should consider periodic evaluation of key aspects of our profession such as the anesthesia machine check. If such an evaluation was structured in a way that informed the anesthesia provider how well they had done without the information being used punitively perhaps it would be more palatable.

Finally, it is worth noting that this study was conducted by three physicians, a CRNA, and four nurse anesthesia graduate students. This kind of cooperative quality improvement research would be truly valuable for all our graduate students.


Michael A. Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Bailey PL, Glance LG, Eaton MP, Parshall B, McIntosh S



A survey of the use of ultrasound during central venous catheterization

Anesth Analg 2007;104:491-497

Bailey PL, Glance LG, Eaton MP, Parshall B, McIntosh S




Purpose            The purpose of this study was to assess the frequency of ultrasound use by anesthesiologists who frequently place central venous lines. The relationships between physician, hospital factors, and ultrasound use was also investigated.

Background            By one estimate, approximately five million central venous catheterizations (CVCs) are performed in the USA each year. Complications are estimated to occur 2% to 15% of the time during CVC placement. Real-time ultrasound guidance during CVC placement has been recommended by the Agency for Healthcare Research and Quality and the National Institute for Clinical Excellence (United Kingdom) as a method to reduce complications. Ultrasound guidance may also improve the success rate of CVC placement. Suitable ultrasound machines cost between $12,000 and $30,000.

Methodology            Invitations to participate in this survey were emailed to the >4000 members of the Society of Cardiovascular Anesthesiologists. The survey itself was internet based. The survey asked respondents to rate their use of ultrasound during CVC placement on a Likert scale (numeric response from 0 to 5, with never = 0 and always = 5). Demographic, practice setting, reasons for using / not using ultrasound, and related questions were also asked.

Result            The survey response rate was 35.3%. Respondents identified their practice settings as: university 48.6%, private practice 46.6%, veterans administration 3.8%, or other 0.7%. The right internal jugular vein was most commonly (94.3%) used for CVC placement.

Ultrasound was “never” or “almost never” used during CVC placement by 67% of respondents (never = 37.4%). Ultrasound was “always” or “almost always” used during CVC placement by 15% of respondents (always = 7.7%). When ultrasound was used, it was used to visualize the central vein in real time during the procedure in only 26% of cases. Often, ultrasound was used to visualize anatomy prior to performing the CVC placement (31%) or during secondary attempts (40%) after a failed CVC placement without ultrasound. When ultrasound was not used it was considered unnecessary by 46% of respondents while 18% stated it was unavailable.

In institutions where ultrasound equipment was always available respondents were about 19 times more likely to use it in some capacity during CVC placement (P<0.001). Ultrasound use was about twice as common in academic hospitals (P<0.001) and over three times more common in VA hospitals (P<0.005) than in private practice settings. Anesthesiologists who placed over 100 CVCs per year tended to be less likely to use ultrasound than those who placed 20 or fewer CVCs per year (P not significant).

Conclusion            Use of ultrasound for CVC placement is limited, with 15% or fewer cardiovascular anesthesiologists using it for real time visualization during CVC placement. Ultrasound use was much more common where the equipment was always available.



There is an old joke about NASA spending hundreds of thousands of dollars on the development of a ball point pen that would write reliably in zero gravity and the Russians sending their cosmonauts up with 5 cent pencils. There are a number of things that can be learned from this story, but, in the end, there are some times when you need that zero gravity pen.

Several authoritative sources recommend ultrasound use as a means to reduce complications during CVC placement. Recommendations for ultrasound use flowed out of studies of overall complication rates. This study contrasts those recommendations for ultrasound use with a low actual rate of use in clinical practice.

The first question that comes to my mind is this, is there a pencil that would reduce the risks associated with CVC placement or do we really need a zero gravity pen (ultrasound)? The study authors point out that some of the complications prevented with ultrasound use (expensive) could also be prevented by measuring intravascular pressure with a simple manometer before inserting a large gauge introducer (cheap).

Next, I wonder why ultrasound use reduces the overall level of complications during CVC placement. Is it because there is some inherent improvement in safety during each and every procedure for which ultrasound is used? Or is there some subgroup of CVC placements at much higher risk for complications in which ultrasound use reduces the risk, and, in so doing, reduces the average risk shown in large studies? My gut reaction is that ultrasound can be quite useful for at least some CVC placements. Like other manual tasks, there are probably some clinicians whose skills are such that ultrasound won’t make much difference in their complication rate and others for whom ultrasound will greatly reduce their complication rate.

While not stated in their formal conclusions, in the discussion section the investigators assert, “…  use of (ultrasound) during CVC … remains limited. Availability of equipment and practitioners’ perceptions of the utility of (ultrasound) contribute to this.” This conclusion is not supported by their study. While respondents use of ultrasound and the availability of ultrasound did correlate to a degree, the study was not constructed to determine a cause and effect. We don’t know whether ultrasound was used more simply because it was available, or whether it was available because practitioners wanted to use it and thus insisted that it be available.

So, what are we to do? Should we be using ultrasound for all CVC placements? There are recommendations to that effect but doing so is clearly not the standard of care right now. There is convincing evidence that using ultrasound reduces the overall incidence of CVC placement complications. That is enough to make a good argument for using it. On the other hand, I’ve not read anything to convince me that ultrasound must be used for every line placement to achieve the reduction in complication rate or that ultrasound is the most effective and efficient way to reduce the complication rate. I really want to know more.


Michael A. Fiedler, PhD, CRNA





© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

O?Sullivan CT, Dexter F, Lubarsky DA, Vigoda MM

Evidence-based management assessment of return on investment from anesthesia information management systems

AANA J 2007;75:43-48

O’Sullivan CT, Dexter F, Lubarsky DA, Vigoda MM


Purpose            The purpose of this review was to describe the pros and cons of anesthesia information management systems in terms of return on investment and societal benefit.

Background            Healthcare costs in the USA are passing $1.7 trillion annually. In an era of reimbursement based upon Diagnostic Related Groups, reducing costs and capturing all reimbursable charges is necessary to preserve revenue levels. One of the methods talked about to reduce the cost of care and increase revenue capture is the implementation of an Anesthesia Information Management System (AIMS). For an AIMS to be financially advantageous, it must result in some combination of reduced overall cost and/or increased revenue that results in a positive return on investment.

Methodology            The authors of this conceptual analysis reviewed the scientific literature for information about AIMS and return on investment, productivity, revenue, savings, or resource utilization. MEDLINE, the Cumulative Index to Nursing & Allied Health Literature, HealthSTAR, International Pharmaceutical Abstracts, Biological Abstracts, and Current Contents were searched. The authors extracted, reported, and commented on the information contained in these studies.

Result            Ten articles were identified. Four areas were identified for which there was evidence that AIMS might result in a positive return on investment:

  • Reducing anesthesia drug costs
  • Reducing staffing costs by changes in scheduling
  • Increased capture of anesthesia charges
  • Increased hospital reimbursement due to improved billing coding

Anesthesia drug costs have been shown to be reduced by as much as 50% as a result of changes resulting from the information made available by AIMS. Drug cost reduction was achieved primarily through a feedback system whereby clinicians were provided monthly reports of their drug use that fell “outside protocol” compared with aggregate data about the drug use of their peers.

AIMS provide highly detailed information about the income generating activities of all providers by time and date allowing detailed analysis of staffing requirements. Using this information, staffing costs were reduced by over 10% in 9 of 11 facilities.

Capturing anesthesia charges by filling out paper billing sheets during an anesthetic which are then transcribed by billing clerks and processed by computer has long been known to be an inefficient and error prone method process. Capturing charges with an AIMS increases the number of legitimate charges captured and reduces errors. It also speeds the claims process resulting in faster payment from their party payers.

Improved hospital reimbursement is possible if the pre-anesthetic evaluation results in the inclusion of all appropriate disease classifications. In one study, when an AIMS was used, at least one ICD-9 code was added in 12% of patient charts resulting in an increase in hospital reimbursement by 1.5%.

AIMS may also reduce costs and improve care by “reminding” clinicians to perform certain tasks, such as administration of preoperative antibiotics. Analysis of AIMS data can alert anesthesia managers to the actual incidence of key procedures, such as antibiotic administration, resulting in departmental or individual notification of a need for improvement.

While the cost of an AIMS is most often born by the institution (hospital), the anesthesia group is more likely to realize the gain. Owing to differences in payment sources and practice patterns, gains resulting from AIMS use for both institutions and anesthesia groups are highly variable, with some gaining little and others achieving noticeable reductions in cost and/or increases in income.

Conclusion   ?Anesthesia Information Management Systems may offer financial and quality of care benefits to institutions, anesthesia groups, and society. The magnitude of these benefits depends upon a variety of factors and will not be achievable in all cases.


Computer Generated Anesthesia Records, the heart of any Anesthesia Information Management System, have been available since 1985. In 1986 they were predicted to be used almost uniformly “within the next few years.” I have long been surprised and disappointed by how slowly AIMSs have been adopted by clinical anesthesia. Judging by my experiences with three different systems, it is apparently quite difficult to develop and implement a system that works well, is compatible with a hospital’s overall information system, and is judged to be affordable.

In my view, AIMSs offer great advantages to anesthesia from both financial and patient care perspectives. I suggest that there are two major barriers to realizing the benefits of AIMSs we should work to overcome. First, I wonder if we don’t get so focused on getting today’s cases done that we fail to look at the long term advantages of an AIMSs. Second, just like we need anesthesia providers “cross trained” in business and education to manage anesthesia groups and teach at the university, we need anesthesia providers “cross trained” in computer and information technology to develop and implement AIMS. It will take visionary leadership and CRNAs willing to pursue degrees in addition to their anesthesia education to remove these roadblocks.

Comprehensive and accurate data is necessary to bill and collect for all legitimate services, manage an anesthesia group, and to monitor and continuously improve the quality of anesthesia care delivered. Without an AIMS of some type we lack much of this data and what we have is error prone. Computer technology has developed to the point where high quality, dependable AIMSs are conceptually possible. Refining and effectively using such systems is likely to improve all aspects of anesthetic practice.


Michael A. Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007


Br?uer A, English MJM, Steinmetz N, Lorenz N, Perl T, Weyland W, Quintel M

Efficacy of forced-air warming systems with full body blankets

Can J Anesth 2007;54:34-41

Bräuer A, English MJM, Steinmetz N, Lorenz N, Perl T, Weyland W, Quintel M


Purpose            The purpose of this study was to ascertain the 1) overall efficiency of heat transfer and 2) the amount of heat transferred by forced air warming blankets at four different skin surface temperatures.

Background            In the surgical setting unintended hypothermia is associated with increased bleeding, prolonged mechanical ventilation, longer hospital stays, a higher incidence of wound infection, cardiovascular stress, and patient discomfort.

Methodology            This prospective observational study examined 11 different forced air warming blankets made by 7 different manufacturers. The warming blankets were applied to a mechanical model designed specifically to study the heat transfer of warming devices. The model had been previously validated and shown to accurately predict heat transfer to human volunteers. Heating units were set at their maximum. The warming blanket was covered with two cotton sheets. The temperature and flow of air exiting the warming hose and entering the blanket was measured by several sensors simultaneously and averaged. Sixteen pairs of heat sensors on the model measured the temperature of the model’s “skin surface” and the air temperature 1 cm above the “skin surface” (essentially the temperature of the air exiting the warming blanket). Each experiment was repeated three times with a new blanket and the average value of the three sets of observations reported. From this information heat transfer in Watts (W) was derived.

Result            The temperature of the air exiting the warming hoses ranged from 41.5?C to 47.6?C. Warming hose air flow ranged from 9.4 L•sec-1 to 26.2 L•sec-1. Heat flow (the “amount” of heat exiting the warming hose per unit time) was thus 249W to 623W. The absolute amount of heat entering the model “body” ranged from 30.7W to 77.3W at a “skin” temperature of 32?C. The heat exchange coefficient (used to describe the efficiency of heat transfer from warming blanket to body) ranged from 13.4 W•m-2•°C-1 to 32.2 W•m-2•°C-1. As the temperature gradient between the skin and the warming blanket became smaller, the rate of heat transfer from the blanket to the body decreased. At skin temperatures of 36?C and lower all blankets transferred some heat to the body.

Conclusion            The efficiency of heat transfer between forced air warming blankets and the body varied considerably, not only from manufacturer to manufacturer but also between different models of forced air warming blankets made by the same manufacturer.



While we often think of postoperative hypothermia as a nuisance, it does contribute to morbidity and even mortality. Hypothermia retards the elimination of anesthetic and adjuvant drugs, impairs coagulation, increases blood loss, activates the sympathetic nervous system, increases the incidence of adverse cardiac events (e.g. angina, MI, ventricular arrhythmias), increases the incidence of wound infections, and is associated with longer hospital stays.

Although perhaps not widely known by clinicians, there has been evidence since the early 1990’s that the efficiency of heat transfer by forced air warming blankets varies by brand. The transfer of heat energy from the warming blanket to the patient depends upon more than simply how hot the air is blowing out of the warming unit. The velocity of airflow, the design of the warming blanket, and various patient factors are also important.

One might be tempted (as I was) to dismiss this study because it used a set of copper tubes with water flowing inside them to simulate a patient’s head, trunk, arms, and legs. But the authors cite studies demonstrating the validity of the test model. Previous studies using the model accurately predicted the temperature changes seen when the studies were repeated in humans. While a degree in physics might be helpful to dig into all the details of the study methodology, the overview is that the authors used the model to determine two factors. The first factor described was the overall efficiency of the warming blankets expressed as the Watts of energy transferred per square meter of skin surface per degree centigrade difference between blanket temperature and skin temperature (W•m-2•°C-1). The second factor reported was how much heat energy was transferred to the simulated patient by each warming blanket at several skin surface temperatures (32?C, 34?C, 36?C, and 38?C). Efficiency describes the percent of the heat leaving the warming unit that gets into the patient while heat transfer describes the absolute amount of heat energy that gets into the patient.

The most efficient warming blanket used a middle of the range warming hose air temperature. The unit that used the coolest warming hose air temperature was the 3rd (of 11) most efficient at transferring heat to the patient and was 1st in the actual amount of heat energy transferred at skin temperatures of 32?C, 34?C, and 36?C. The unit that used the hottest warming hose air temperature was the 8th most efficient at transferring heat to the patient. Despite its inefficiency, however, it ranked 3rd, 3rd, 2nd, and 1st respectively at the actual amount of heat energy transferred at skin temperatures of 32?C, 34?C, 36?C, and 38?C. The most efficient unit transferred over twice as much heat energy to the patient as the least efficient unit.

Clearly, simply being hot is not enough to make a forced air warming blanket system work well. Transferring twice as much heat energy does not require being twice as hot.

The bottom line demonstrated by this and previous studies is that the ability to warm a patient is not simply a matter of how hot the air is the blows into the blanket or the manufacturers name. The effectiveness of forced air warming blankets varies between manufacturers and between products of the same manufacturer. The best way to choose warming blankets to most effectively warm our patients is to look at the data in articles such as this one.


Michael A. Fiedler, PhD, CRNA


Previous worthwhile studies of the efficiency of forced air warming blankets:

Camus Y, Delva E, Just B, Lienhart A. Leg warming minimizes core hypothermia during abdominal surgery. Anesth Analg. 1993;77:995-999.

Just B, Trévien V, Delva E, Lienhart A. Prevention of intraoperative hypothermia by preoperative skin-surface warming. Anesthesiology. 1993;79:214-218.

Giesbrecht GG, Ducharme MB, McGuire JP. Comparison of forced-air patient warming systems for perioperative use. Anesthesiology. 1994;80:671-679.



© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Hemingway CJ, Malhotra S, Almeida M, Azadian B, Yentis SM


The effect of alcohol swabs and filter straws on reducing contamination of glass ampoules used for neuroaxial injections

Anaesthesia 2007;62:286-288

Hemingway CJ, Malhotra S, Almeida M, Azadian B, Yentis SM



Purpose            The purpose of this study was to define the incidence of contamination of the drug inside a glass ampoule when it was opened in the usual manner. Usual ampoule opening procedure was contrasted with wiping the ampoule with an alcohol swab before opening, aspirating the drug with a filter needle, or both.

Background            Contamination of drugs with pathogens may lead to patient infection and harm. This is especially so when the drugs are injected into the epidural or subarachnoid spaces. Previous research has separately demonstrated “considerable” contamination of the outer surface of unopened fentanyl ampoules and that wiping the outside of an ampoule prior to opening resulted in a reduction in bacterial contamination of the drug within.

Methodology            This prospective, experimental, clinical and laboratory study was divided into two parts. In the first part 100 glass ampoules were opened in the course of normal clinical use without any special precautions. While the clinicians routinely washed their hands after patient contact, hands were not washed immediately prior to opening the ampoules. Gloves were not worn during ampoule opening. The upper third of 50 ampoules was wiped with an isopropyl alcohol swab and air dried. The other 50 ampoules were opened without being wiped with alcohol. After aspirating the drug within for clinical use, the inside of the ampoules was swabbed and the swab used to colonize a blood agar plate which was incubated for 4 days at 37??C.

In the second part, the exterior surface 100 unopened glass ampoules containing normal saline was contaminated with a pathogen in the laboratory. Wearing clean gloves to protect the researcher, the ampoules were opened. Half were first wiped with alcohol and half were not. The saline inside was then aspirated with a filter needle in half the cases and with a regular hypodermic needle in the other half. This resulted in four groups combining alcohol wiping or not with a filter needle or not. The aspirated saline was then deposited onto a blood agar plate and incubated for two days at 37 ?C.

Result            In the first part of the study 18% (95% CI: 10-31%) of the ampoules that were not wiped with alcohol before being opened grew organisms. None (95% CI: 0-9%) of the ampoules wiped with alcohol grew organisms (P = 0.004).

In the second part of the study the number of bacterial colonies grown were counted. A large number of ampoules opened without first being wiped with an alcohol pad grew multiple bacterial colonies, often greater than 25 colonies. Wiping the ampoule with an alcohol pad prior to opening reduced the subsequent growth of bacterial colonies to the greatest extent. Using a filter needle helped further reduce the number of colonies grown but the effect was small.

Conclusion            The exterior of drug ampoules is often contaminated with bacterial pathogens. These pathogens contaminate the drug inside when the ampoule is opened without any special precautions. Wiping the top third of the ampoule with an alcohol pad and allowing the alcohol to dry before opening the ampoule significantly reduces the incidence and magnitude of bacterial contamination of the drug inside. Aspirating the drug with a filter needle reduces the contamination of the drug, but to a lesser degree.



I can’t tell you how surprised I was to read this study. I never dreamed that opening a glass ampoule could routinely result in the bacterial contamination of the drug inside. I’m impressed that someone asked this question and that they did such a good job finding the answer. But it is the second half of their study that is the most helpful clinically. This study shows convincingly that simply wiping the neck and top part of the ampoule with an alcohol swab and allowing the alcohol to dry prior to breaking the top off dramatically reduced the incidence and magnitude of bacterial contamination. It is always a plus when simple steps are also highly effective.

While the clinical consequences of this bacterial contamination are unknown, it seems reasonable to believe it would pose a threat to some patients. The investigators’ focus was on the risk to patients receiving neuraxial anesthesia. Given the potentially devastating consequences of injecting bacteria in, for example, the subarachnoid space, their concern is warranted. Based on this information, I will have to change my practice and wipe ampoules with alcohol before breaking them for spinal and epidural use. We should give some thought to wiping ampoules with alcohol prior to opening for all other uses as well.

One caution is necessary as a final thought. Ethyl alcohol is used in concentrations of 50% to 100% as a neurolytic. While I’m not familiar with the neurolytic potential of isopropyl (rubbing) alcohol (generally 50% to 70%) I would take care not to allow it to contaminate the drug I’m drawing up for neuraxial injection.


Michael A. Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Bhananker SM, Posner KL, Cheney FW, Caplan RA, Lee LA, Domino KB


Injury and liability associated with monitored anesthesia care

Anesthesiology 2006;104:228-234

Bhananker SM, Posner KL, Cheney FW, Caplan RA, Lee LA, Domino KB



Purpose            The purpose of this closed claims analysis was to compare claims involving Monitored Anesthesia Care (MAC) to claims involving general and regional anesthesia.

Background            More and more cases are done with MAC anesthesia. Cardiorespiratory depression and death during MAC cases has been associated with administration of multiple drugs, medication errors, inadequate monitoring, and delayed or inadequate resuscitation.

Methodology            This retrospective, closed claims analysis included information from 1,952 claims filed with 35 insurance companies between 1990 and 2002. Obstetric, pain management, and dental claims were not included in the analysis.

Result            There were 121 MAC related claims, 1,519 general anesthesia claims, and 312 regional anesthesia claims during the inclusion period. MAC claims included a higher proportion of patients ≥ 70 years old and those assigned an ASA physical status of III, IV, or V. MAC claims involved plastic or reconstructive procedures of the head and neck more often (19%) than did claims for general anesthesia (5%) or regional anesthesia (1%).

Injury severity was similar between MAC and general anesthesia. Claims involved death or permanent brain injury in 41% of MAC cases and 37% of general cases. The mechanism of injury involved inadequate oxygenation / ventilation in 18% of MAC claims but only 2% of general and regional claims.

Fires ignited by electrocautery were far more common in MAC claims (17%) than in general (1%) or regional (0%) claims.

Inadequate anesthesia or patient movement was the mechanism of injury in 11% of MAC claims but only 3% of general claims and 2% of regional claims. Eighty five percent of MAC claims for inadequate anesthesia / patient movement involved eye surgery or sedation for eye blocks.

Conclusion            The two most common causes of injury during MAC were death or brain injury due to respiratory depression and burns associated with supplemental oxygen and electrocautery.



Clearly, MAC cases can and do result in anesthesia related morbidity and even mortality. One of the most important uses for the type of information in this closed claims analysis is to see where the problems are and ask oneself, “what can I do to reduce the risk of these problems in my patients?”

This closed claims analysis puts the risk of MAC into perspective. While MAC related claims made up only 6% of overall anesthesia claims, the severity of injury in MAC claims was almost identical to that of general anesthesia. Furthermore, airway problems resulting in inadequate oxygenation and ventilation were nine times more common in MAC claims than in general anesthesia claims. The fact that inadequate anesthesia and patient movement was the mechanism of injury almost four times more often in MAC claims than in general anesthesia claims suggests to me that some cases were being attempted under MAC that really needed a regional or general anesthetic.

From my perspective, these are the take home messages.

1)    Airway problems may result in proportionately more harm during MAC than during general anesthesia. This may be even more true in the elderly, those with high ASA physical status classifications, and in patients where the airway is in the surgeon’s field (head and neck procedures). Perhaps we should be using less propofol for sedation (a potent respiratory depressant) and more dexmedetomidine or a mixture of propofol / ketamine. Despite our expertise with sedation, doing so would increase the margin of safety for respiratory depression.

2)    Perhaps we should reevaluate our guidelines for sedation for eye blocks and eye surgery. It may be that general anesthesia would be safer for some of these patients. (The closed claims analysis does not, unfortunately, provide us with any guidance as to which eye patients might benefit from general anesthesia. We’ll have to figure that out ourselves.)


3)    The problem of burns during MAC is too complex to propose a solution based on the information presented here but we should at least become aware that it is a problem. With awareness we can be more cautious, observe conditions that might contribute to the risk of fires during MAC, and devise procedures to reduce this risk.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Obstetric Anesthesia

Siddik-Sayyid SM, Taha SK, Aouad MT, Daaboul DG, Deeb PG, El Khatib MF, Baraka AS



The effect of injection of two vs 10 mL saline on the subsequent spread and quality of epidural analgesia in parturients

J Clin Anesth 2006;18:575-579

Siddik-Sayyid SM, Taha SK, Aouad MT, Daaboul DG, Deeb PG, El Khatib MF, Baraka AS




Purpose            The primary purpose of this study was to determine whether the spread of local anesthetic block or extent of labor pain relief was different in parturients who received 2 mL versus 10 mL of saline through the Touhy needle prior to epidural catheter insertion and dosing. A secondary purpose was to compare the ease of catheter insertion and the incidence of paresthesia or blood return through the catheter.

Background            Preservative free normal saline is often injected to identify entry into the epidural space by loss of resistance. Previous research has shown that the volume of saline affects characteristics of the subsequent block. Some studies have shown that injecting 10 mL of saline into the epidural space before injecting local anesthetic results in a greater spread of block. In the term parturient, distended epidural veins and increased intraabdominal pressure result in a reduction in the volume of the epidural space. Injecting fluid volume through the Touhy needle before inserting an epidural catheter decreases the incidence of both venous entry and paresthesia. The optimal volume needed to facilitate atraumatic insertion of an epidural catheter is not known in the parturient.

Methodology            This prospective, randomized, double-blind study divided 105 healthy term pregnant women with cervical dilation of 5 cm or less into two groups. Group 2 mL received an epidural injection of 2 mL of preservative free normal saline through the Touhy needle with loss of resistance. Group 10 mL received an epidural injection of 10 mL of saline at loss of resistance.

All parturients received 500 mL to 1000 mL of lactated ringers and all were receiving oxytocin for augmentation of labor. A 20 gauge, closed end, multiple side hole catheter was inserted 4 cm into the epidural space through a 19 gauge Touhy needle at the L3-4 interspace in all parturients while in the sitting position. Patients did not know which group they were in and all evaluations of block height, etc were performed by an independent observer blinded to maternal group assignment.

After epidural saline injection and catheter placement, all women received 10 mL of 0.1% bupivacaine with 2 ?g/mL fentanyl. Assessment of block spread, pain relief, etc was performed 25 minutes after local anesthetic injection. Block spread was assessed with an alcohol swab and pinprick. Visual analog scale (VAS) pain scores were reported at the peak of a uterine contraction.

Result            Block height assessed with cold (alcohol swab) was T-10 (T-6 to L-1) in the 2 mL group and T-9 (T-4 to L-1) in the 10 mL group (P=0.049). Block height assessed by pinprick was T-11 (T-6 to L-1) in the 2 mL group and T-11 (T-5 to L-1) in the 10 mL group (P=0.099).

The spread of block (total number of dermatomes in both directions) assessed with both cold and pinprick was statistically significantly greater in the 10 mL group (P<0.001 and P=0.001 respectively). The average difference in spread was two dermatomes or less.

Pain relief was almost identical in the two groups with a median VAS score of 1 on both groups 25 minutes after injection of the local anesthetic / fentanyl solution. While there was no statistically significant difference between the groups in regard to one sided blocks, there was a noticeable difference between groups. Just over 13% of parturients had a one sided block in the 2?mL group compared to 5.7% in the 10 mL group.

Motor block was minimal in both groups. There was no difference in the ease of catheter insertion, frequency of paresthesia, blood in the catheter or change in maternal blood pressure between groups.

Conclusion            Block height was no different in term pregnant women who received 2 mL versus 10 mL of epidural saline before injection of 10 mL of 0.1% bupivacaine with 2 ?g/mL fentanyl. The total number of dermatomes across which the epidural block spread was statistically significantly greater in the 10 mL saline group but of little if any clinical significance. The ease of epidural catheter insertion, paresthesia, and blood in the epidural catheter was no different between groups.



I used the loss of resistance to air technique to identify the epidural space for many years. I favored it chiefly because air is compressible so I could see the plunger move while checking for loss of resistance. I had seen others puncture the dura using loss of resistance to saline with a glass syringe only to discover afterward that their plunger was stuck making it impossible to detect the loss of resistance when it occurred. A few years ago, however, it became clear that using saline was associated with a lower incidence of dural punctures, unblocked segments (“hot spots”), and inability to thread the epidural catheter.1 Plastic syringes designed for loss of resistance have also become available removing concerns of plungers sticking. I now teach and use loss of resistance to saline.

Once we decided that finding loss of resistance using saline was superior to air, the next question became, how much saline should be injected once the epidural space was entered? Presumably, not enough and we would miss the benefits of the technique. Too much and our precious local anesthetic would be diluted and fail to do its job. Many clinicians used about 2 mL but the old adage, “If a little bit is good, a lot more must be better” was hanging out there.

This study set out to answer the question, “how much saline should we inject?” in the context of labor epidural analgesia. I’m impressed with the tightly controlled methodology of this study. They also did a good job maximizing the effects of pain, including only women receiving oxytocin augmentation of labor and measuring pain scores only at the peak of contractions. This was important so that if 10 mL of saline diluted the analgesic effects of the bupivacaine / fentanyl solution the reduction in pain relief would be easy to detect. Nothing else matters if the pain relief isn’t there.

I would have expected the larger volume of saline to increase the block spread and dilute the concentration of the local anesthetic resulting in reduced analgesia. To my surprise, the study found that injecting 10 mL of saline didn’t change the block that much. Those changes in block height and spread that were statistically significant were of little or no clinical significance. Pain relief was virtually identical. While it was not statistically significant, I noted with interest that over twice as many women in the 2 mL group had one sided blocks compared to the 10 mL group. A larger study looking at this factor in specific might provide some solid rational for choosing the larger volume of saline. As it is now, what I take away from this study is that any volume of saline between 2 mL and 10 mL is better than using air.


Michael A. Fiedler, PhD, CRNA


1            Evron S, Sessler D, Sadan O, et al. Identification of the epidural space: loss of resistance with air, lidocaine, or the combination of air and lidocaine. Anesth. Analg. 2004;99:245-250.


© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Rofaeel A, Lilker S, Fallah S, Goldszmidt E, Carvalho J


Intrathecal plain vs hyperbaric bupivacaine for labour analgesia: efficacy and side effects

Can J Anesth 2007;54:15-20

Rofaeel A, Lilker S, Fallah S, Goldszmidt E, Carvalho J


Purpose            The purpose of this study was to determine whether subarachnoid block with a plain bupivacaine / fentanyl solution produced greater labor pain relief than a hyperbaric bupivacaine / fentanyl solution. A secondary purpose was to contrast the side effects associated with both solutions.

Background            A combined spinal epidural (CSE) produces effective labor analgesia more quickly than a pure epidural technique. Initial block characteristics are largely controlled by the absolute dose and baricity of the solution injected into the cerebrospinal fluid. Hyperbaric solutions produce a more predictable block and may be associated with less hypotension, nausea, and itching. The limited cephalad spread of hyperbaric solutions may reduce analgesia efficacy compared to clinically isobaric solutions. But CSF density has been shown to be lower in term pregnant women (1.0003 g/mL) than in non-pregnant individuals and solutions often considered clinically isobaric are slightly hypobaric in term pregnant women. The clinical effects of a slightly hypobaric solution are most pronounced when injected into a sitting patient, common when performing a CSE in pregnant women. Conversely, hyperbaric solutions move down and away from nerve roots carrying the sensation of labor pain in sitting patients, potentially reducing the effectiveness of labor analgesia.

Methodology            This prospective, single blind study included 62 term pregnant women in active labor. All women had ≥5 cm cervical dilation and a verbal pain score of >5 on a 0 – 10 scale. Women were randomly assigned to receive 2.5 mg hyperbaric or plain bupivacaine in 1 mL solution as part of their CSE labor analgesia. Hyperbaric solutions contained 2.75% dextrose. In both groups 15 ?g fentanyl was added to the bupivacaine solution. The CSE was performed at the L3-4 or L4-5 interspace with patients in the sitting position. The epidural catheter was not dosed during the first 10 minutes after the subarachnoid injection of bupivacaine and fentanyl. This included the test dose. After 10 minutes, patients reporting a pain score of >3 were offered additional analgesia via the epidural catheter (5-10 mL of 0.125% bupivacaine). Patients with a pain score of 3 or less who requested more relief also received an epidural dose. Patient controlled epidural analgesia was begun at 30 minutes, after data collection for the study had ended.

Fetal bradycardia was defined as a fetal heart rate (FHR) < 100 beats per minute for > 60 seconds. Maternal hypotension was defined as a 20% or greater decrease in systolic pressure.

Result            Before CSE analgesia, median pain scores were identical in the plain and hyperbaric bupivacaine groups. Five minutes after the subarachnoid injection, no women in the plain bupivacaine group had a pain score of > 3 while 11 women in the hyperbaric bupivacaine group (37%) had a pain score > 3 (P<0.001). Likewise, 10 minutes after the subarachnoid injection, no women in the plain bupivacaine group had a pain score of > 3 while six women in the hyperbaric bupivacaine group (20%) had a pain score > 3 (P=0.024). At 20 minutes and beyond pain scores were not statistically significantly different.

Block level was higher in the plain bupivacaine group at 5, 10, 20, and 30 minutes. Motor block was greater in the hyperbaric bupivacaine group.

Fetal bradycardia occurred in 33% of patients who received plain bupivacaine and 10% of patients who received hyperbaric bupivacaine (P=0.03). Itching was equally more common in the plain than the hyperbaric bupivacaine group. Maternal hypotension was more common in the plain bupivacaine group (23% vs 10%) but was not statistically significant. There was no difference between groups in regards to nausea.

Conclusion            During CSE labor analgesia, plain bupivacaine with fentanyl produced analgesia more quickly, higher sensory block levels, and less motor block. Plain bupivacaine was also associated with a higher incidence of fetal bradycardia and maternal itching.



Epidural analgesia has long been the gold standard for labor pain relief but subarachnoid blocks do have a faster onset of action. Combined spinal and epidural (CSE) techniques for labor analgesia try to get the best of both techniques while avoiding Post Dural Puncture Headache with a very small gauge spinal needle. One of the limitations of the CSE technique for labor is that the subarachnoid dose is generally hyperbaric. When administered in the lumbar area it moves toward sacral nerve roots in the sitting patient. But the pain from the first stage of labor is carried by nerve roots from T-10 to L-1 or L-2, above the level at which the subarachnoid dose is injected. This may limit the analgesia provided by the spinal dose.

Here, the investigators compared the analgesia of spinal bupivacaine with and without dextrose. In an attempt to make the bupivacaine with dextrose “less hyperbaric” they reduced the concentration of dextrose from the commercially prepared 8.25% to 2.75%. The plain bupivacaine was commercially prepared 0.25% bupivacaine.

Clearly, the plain bupivacaine solution produced better pain relief in more women faster than hyperbaric bupivacaine. But plain bupivacaine was also associated with a noticeably higher incidence of “sustained fetal bradycardia” than the hyperbaric solution. The authors cite studies showing that fetal bradycardia in this context is most likely due to rapid pain relief and transient uterine hypertonus due to lower circulating catecholamine levels. They also cite studies showing that this bradycardia is not associated with “adverse neonatal outcomes.” Yet we know that uterine hypertonus can reduce fetal oxygenation and result in harm in some circumstances so it seems fair to say, at the least, that the margin of safety is reduced if fetal bradycardia is occurring due to uterine hypertonus.

So here are my questions, and in asking them I will admit I am not yet a fan of CSE techniques for labor analgesia. While traditional labor epidural techniques do have a slower onset of analgesia than CSE, it isn’t that much slower. Furthermore, pain relief often begins in about five minutes giving the woman the relief of knowing that things are getting better. How important is the marginally faster onset of pain relief offered by CSE? (No, I am not a woman and I have never experienced labor pain. I am not minimizing this pain experience. I’m asking the question in good faith, not rhetorically.) Does the faster pain relief outweigh the incidence of fetal bradycardia seen with CSE? Even if it is not associated with neonatal harm, the high incidence of false positives may sometimes mask a real problem which is then discovered and treated later than it otherwise would have been.


Michael A. Fiedler, PhD, CRNA

How important is the faster onset of analgesia with CSE versus traditional epidural only labor analgesia? How often do you see fetal bradycardia during CSE analgesia? Is it accompanied by increased uterine tone? Tell us at “Blog with the Editors.”

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007


Chung F, Kayumov L, Sinclair DR, Edward R, Moller HJ, Shapiro CM



What is the driving performance of ambulatory surgical patients after general anesthesia?

Anesthesiology 2005;103:951-956

Chung F, Kayumov L, Sinclair DR, Edward R, Moller HJ, Shapiro CM




Purpose            To assess the ability to drive an automobile following general anesthesia in patients undergoing anesthesia and surgery.

Background            We traditionally inform patients who undergo general anesthesia not to drive their car or operate potentially dangerous equipment.   In the current practice climate this recommendation invites a critical look because it has not been studied in patients actually undergoing anesthesia and surgery.

Methodology            This was a prospective study involving 20 patients who underwent general anesthesia for knee arthroscopy with midazolam, propofol, fentanyl, and nitrous oxide in combination with either 1 MAC desflurane or sevoflurane. A standardized postoperative analgesic/antiemetic protocol was followed. Driving ability was assessed preoperatively, then at 2 and 24 hours postoperatively in a standardized driving simulator that measured a number of performance demands that occur during routine driving.  Additionally a number of other physiological variables (e.g., fatigue/sleep/alertness) were recorded.  Twenty subjects matched for the usual demographics of age, gender and baseline health represented the control group; this groups’ driving performance was assessed once between the hours of 10am and 12pm.

Result            Patients, compared to controls, had a greater number of lapses and generally lower attention acuity in the preoperative period.  Significant decrement in performance (impaired driving skill, microsleep, and longer reaction time) was noted at 2 hours postoperatively in those undergoing general anesthesia.  At 24 hours the performance in both groups was equivalent.

Conclusion            Subjects undergoing general anesthesia were safe to drive at 24 hours.  Preoperatively, likely do to distractions of the upcoming surgical procedure, and at 2 hours postoperatively, likely do to residual drug effects, those undergoing general anesthesia demonstrated impairment (worse at 2 hours than preoperatively).



Studies like this get to a fundamentally important patient safety concern in a practical way.  We know that anesthetic drugs have residual effects on the central/autonomic nervous system.  What was surprising to me was to learn that research designed to determine driving readiness following anesthesia and surgery had never been performed.  I suspect that like myself, many of you have simply advised patients, “don’t drive for 24 hours,” with a degree of certainty.  Although there are at least 2 studies of driving ability following general anesthesia, neither involved patients undergoing surgery.  So the applicability of those studies to the real world is not certain.

While this study is helpful to us in advising our patients, I could not help but notice that the patients were all relatively healthy and relatively young. Also given that the intraoperative fentanyl was limited to 1.5 mcg/kg and the inhalational agents were limited to those with low solubility, we must be very careful in extending these findings too generously. This study also helps to remind me that what we do during the management of our patients has long term consequences that may not, at first blush, be obvious.  For example, I wonder about those lapses in ability they found prior to anesthesia--how many of our ambulatory patients are driving to the hospital impaired?


Chuck Biddle, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007


Nitahara K, Sugi Y, Higa K, Shono S, Hamada T

Neuromuscular effects of sevoflurane in myasthenia gravis patients

Br J Anaesth 2007;98:337-341

Nitahara K, Sugi Y, Higa K, Shono S, Hamada T


Purpose            The purpose of this study was to describe the effect of 1.7% and 3.4% sevoflurane on neuromuscular function in myasthenia gravis patients who had not received any muscle relaxants. A secondary goal was to discover whether or not it was possible to predict the neuromuscular depression produced by sevoflurane in myasthenia gravis patients by observing the Train-of-Four (TOF) ratio before sevoflurane administration.

Background            It has long been known that potent inhalation agents produce some skeletal muscle relaxation during general anesthesia and that they potentiate the neuromuscular blocking effects of nondepolarizing muscle relaxants. The T4 to T1 ratio (T4/T1) compares the magnitude of the first twitch in a train-of-four with the magnitude of the fourth twitch. The smaller the fourth twitch compared to the first twitch, the smaller the numerator (T4 magnitude divided by the T1 magnitude) and the smaller the T4/T1 ratio. If the ratio is multiplied by 100 it essentially represents what percent T4 is of T1. The smaller the T4/T1 ratio the greater the inhibition of neuromuscular function; the greater the paralysis.

Previous research has shown that 1.9 MAC halothane reduced the T4/T1 to 0.72 in myasthenia gravis patients. Similarly, 1.9 MAC isoflurane reduced the T4/T1 to 0.59. Sevoflurane has been shown in some studies to potentiate nondepolarizing muscle relaxants to a greater extent than halothane or isoflurane.

The potential for potent inhalation agent neuromuscular depression in myasthenia gravis patients is not reliably predictable from clinical signs, anti-acetylcholine receptor antibody titers, or maintenance doses of anticholinesterase medications.

Methodology            This prospective, descriptive study compared the neuromuscular effects of sevoflurane in two subgroups of myasthenia gravis patients with a healthy control group. It also compared the pre-sevoflurane T4/T1 ratio with subsequent sevoflurane induced neuromuscular depression in myasthenia gravis patients.

The study included 16 myasthenia gravis patients and 12 healthy patients, all of whom were scheduled for “elective minor surgery.” In myasthenic patients, maintenance anticholinesterase and steroid drugs were continued until the day of surgery. Some, but not all, myasthenic patients received pyridostigmine on the day of surgery. General anesthesia was established with propofol, fentanyl, midazolam, and nitrous oxide. No muscle relaxants were administered to any patient in the study. End-tidal CO2 was kept between approximately 35 torr and 40 torr. Likewise, tympanic temperature was maintained at 37?1 ?C. Neuromuscular function was monitored via electromyography at the adductor pollicis muscle. Baseline T4/T1 was measured 15 minutes after induction of anesthesia and before sevoflurane administration. Measurements of T4/T1 during sevoflurane administration occurred following 30 minute stable end-tidal concentrations of 1.7% and 3.4%.

Statistical analysis was appropriate to the data and study goals.

Result            Myasthenic patients were classified into two subgroups according to their baseline T4/T1 ratio. Of the 16 patients with myasthenia gravis, 6 had a baseline T4/T1 of less then 0.9 and were classified as the “fade” subgroup. Their average T4/T1 was 0.81. The remaining 10 myasthenic patients had a baseline T4/T1 of 0.9 or greater and were classified as the “non-fade” subgroup. Their average T4/T1 was 0.97. The baseline T4/T1 in the healthy control group averaged 0.99.

Administration of 1.7% sevoflurane reduced the T4/T1 in the myasthenic fade group to 0.64 but produced almost no neuromuscular depression in the myasthenic non-fade group and the healthy control group (0.94 and 0.98 respectively). Administration of 3.4% sevoflurane decreased the T4/T1 ratio as follows: myasthenic fade group 0.43, myasthenic non-fade group 0.71, and control group 0.70. These reductions in neuromuscular function were statistically significantly different than their respective baseline values for each group (P<0.01). In addition, the T4/T1 ratios in the myasthenic fade group were all statistically significantly different than the corresponding values in both the non-fade and control groups (P<0.01).

Upon recovery from general anesthesia the T4/T1 ratio recovered to near baseline values in all patients.

Conclusion            Sevoflurane administration produced concentration dependent depression of neuromuscular function in all patients. At 1.7 %, sevoflurane depression of neuromuscular function was of little clinical significance in the myasthenic non-fade group and the healthy control group. Elimination of sevoflurane resulted in a return of baseline neuromuscular function.

Myasthenic patients whose baseline T4/T1 was less than 0.9 experienced a greater depression of neuromuscular function during sevoflurane anesthesia than myasthenic patients whose T4/T1 was 0.9 or greater.




Throughout my practice I have tried to maximize the muscle relaxation derived from potent inhalation agents administered to myasthenic patients in an effort to avoid using nondepolarizing relaxants. Performing endotracheal intubation during deep inhalation anesthesia is especially attractive in myasthenic patients so this article is particularly relevant to their care. It is helpful to know that we can achieve, on average, a 29% to 57% reduction in neuromuscular function in myasthenic patients at only 3.4% sevoflurane (non-fade and fade subgroups respectively). I can easily imagine an anesthetic plan that would allow for greater than 3.4% sevoflurane in many patients, and, thus, a further depression in neuromuscular function.

While I understood that there was a fair amount of variability in the susceptibility of myasthenics to the neuromuscular depressant effect of potent inhalation agents, I did not understand that this susceptibility could be so conveniently classified into “highly susceptible” (the fade subgroup) and only moderately susceptible (the non-fade subgroup). This knowledge has clinical significance. The non-fade subgroup responded quite similarly to healthy control patients at 1.7% sevoflurane. Using relatively low concentrations of sevoflurane in these patients preserves a wide margin of safety in regards to neuromuscular function. The fade subgroup experienced a disproportionately greater depression of neuromuscular function than the non-fade subgroup or healthy controls, yet their T4/T1 ratio recovered to baseline when sevoflurane was eliminated. I can thus use high concentrations of sevoflurane as a substitute for nondepolarizing muscle relaxants in many situations with the assurance that there will be little, if any, residual neuromuscular effects once the agent is eliminated. I will also know that if nondepolarizing relaxants are used in these more susceptible patients residual paralysis is much more likely than it is in myasthenics whose baseline T4/T1 is 0.9 or greater.

What this study didn’t tell me that I really wanted to know was how long it took for the T4/T1 to return to baseline once the sevoflurane was discontinued. It is helpful to know that postoperatively the T4/T1 ratio returns to baseline. It would be more helpful to have an idea how long that takes. Intuitively it would seem as though recovery of T4/T1 would occur simultaneously with sevoflurane elimination. Clinically, I wish the authors would have shown me evidence that this was the case … or, if not, tell me how long it did take.

There is one confounder in this study that we should be mindful of. Some, but not all, of the myasthenic patients in both the fade and non-fade groups received pyridostigmine on the day of surgery. This is the type of “problem” that is unavoidable in human clinical research. Not to have given pyridostigmine to patients who needed it on the day of surgery would have placed them at risk and been unethical. Patients in the fade group who received pyridostigmine may have had an even lower T4/T1 ratio if they had not received it. Patients in the non-fade subgroup who received pyridostigmine may have had a more “normal” T4/T1 ratio because they had it. We simply need to understand that this fact reduces the clarity of the study results to an unknown degree.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Habib AS, Reuveni J, Taguchi A, White WD, Gan TJ


A comparison of ondansetron with promethazine for treating postoperative nausea and vomiting in patients who received prophylaxis with ondansetron: a retrospective database analysis

Anesth Analg 2007;104:548-551

Habib AS, Reuveni J, Taguchi A, White WD, Gan TJ



Purpose            The purpose of this study was to compare the effectiveness of ondansetron or promethazine as rescue treatment of Post Operative Nausea and Vomiting (PONV) following a prophylactic dose of ondansetron. A secondary purpose was to compare different doses of promethazine.

Background            Changing to a different class of antiemetic drug when PONV prophylaxis fails is recommended but there is little data to support this recommendation. In particular, studies show that when ondansetron prophylaxis fails, a repeat dose of ondansetron is unlikely to relieve PONV.

Methodology            This was a retrospective analysis of information collected in a PONV quality improvement database. Adult patients who received isoflurane or sevoflurane general anesthesia for 30-240 minutes, with or without nitrous oxide, and who received ondansetron 4 mg for PONV prophylaxis were included in the analysis.

The quality and completeness of data entry was not controlled. No attempt was made to insure that the two groups were comparable in regards to PONV risk or other variables.

Result            A total of 18,209 patient records were analyzed. Almost a third (31%) received rescue antiemetics in recovery within four hours of a prophylactic ondansetron dose; either ondansetron 4 mg (72%) or 6.25 mg, 12.5 mg, or 25 mg promethazine (17%). Rescue PONV treatment was effective in 50% of ondansetron patients and 68% of promethazine patients (P<0.0001). Promethazine 6.25 mg was just as effective as 12.5 mg or 25 mg (P=0.3).

Conclusion            When ondansetron prophylaxis failed, promethazine was a more effective PONV rescue drug than ondansetron. Promethazine 6.25 mg was just as effective as larger doses.



The strength of the evidence offered by reports such as this one is limited. It is basically one step above “expert opinion.” Despite this caveat, such reports do have value on several levels. Participating in an institutional quality improvement process, such as the one that generated this report, can engage staff, get their heads out of the routine, and help them to critically examine patient care and outcomes.

I’ve seen enough evidence to believe that when prophylaxis with a serotonin receptor antagonist fails, rescue with the same drug is unlikely to succeed. This confirms my belief with a fairly large number of patients. I was surprised, however, by two of the findings. First, I was surprised that when ondansetron didn’t work prophylactically it worked as a PONV rescue drug 50% of the time. That is much higher than I would have predicted. But when ondansetron PONV prophylaxis fails I’ll still change to another drug. Second, I was surprised that 6.25 mg of promethazine was just as effective as larger doses. Given the side effects of promethazine I’ll be less likely to use more than 6.25 mg for PONV rescue from here on.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Gurbet A, Basagan-Mogol E, Turker G, Ugun F, Kaya FN, Ozcan B



Intraoperative infusion of dexmedetomidine reduces perioperative analgesic requirements

Can J Anesth 2006;53:646-652

Gurbet A, Basagan-Mogol E, Turker G, Ugun F, Kaya FN, Ozcan B


Purpose            The purpose of this study was to assess the effects of an intraoperative dexmedetomidine infusion on postoperative analgesia.

Background            Dexmedetomidine is a highly selective alpha-2 agonist that induces sedation, some analgesia, and reduces MAC with little if any respiratory depression. Direct analgesia appears to peak at about 0.5 ?g/kg in humans. Sedation is dose dependent, progressing to general anesthesia at high doses.

Methodology            This prospective, randomized, double-blind study included 50 women undergoing total abdominal hysterectomy. Subjects were divided into two groups. Group “D” received an infusion of 1 ?g/kg dexmedetomidine over 30 minutes before induction of general anesthesia followed by 0.5 ?g/kg/hour until the end of surgery. Group “P” received an equal volume infusion of normal saline over the same period.

General anesthesia was induced with 3 – 5 mg/kg of sodium pentothal, 3 ?g/kg fentanyl, and vecuronium 0.1 mg/kg. Anesthesia was maintained with 0.5% to 2% sevoflurane and 60% nitrous oxide. Additional fentanyl and vecuronium was administered during the case in response to vital signs and muscle relaxation. Postoperatively, all women received morphine by Patient Controlled Analgesia (PCA) for 48 hours. There was no basal rate; only boluses of morphine were administered.

Result            Intraoperative sevoflurane concentrations were similar in both groups. Sedation scores, pain scores, and time to first demand for pain medicine were similar in both groups. Group D received less fentanyl intraoperatively (P<0.05) and less morphine during the first 48 hours postoperatively (P<0.01). Group D used an average of 12 mg morphine in the PACU while group P used an average of 19.5 mg morphine. During the 46 hours after discharge from the PACU, group D patients used an average of 28.6 mg morphine and group P patients used an average of 65.8 mg morphine (group D used 56% less morphine). PONV treatment was required in 6 group D patients and 15 group P patients.

Conclusion            An intraoperative infusion of dexmedetomidine reduces PCA morphine demand for at least 48 hours following abdominal hysterectomy. Morphine related side effects are fewer in patients who have had a dexmedetomidine infusion intraoperatively.



A few days ago a colleague told me about a patient in which he’d used dexmedetomidine intraoperatively and how little pain they had postoperatively. I was familiar with dexmedetomidine but had not used it. Dexmedetomidine provides some analgesia but the analgesia it produces has a fairly low ceiling effect. Furthermore, it has a terminal elimination half life of only two hours. I couldn’t account for long lasting postoperative analgesia with the limited analgesia provided by dexmedetomidine and its short duration of action made effective postoperative analgesia seem even less likely. In the course of learning more about the postoperative analgesic effects of dexmedetomidine I ran across this article. There is a lot we have yet to learn about pain, but one thing that is clear now is that the amount of pain produced by a stimulus can be increased or decreased by factors in addition to the magnitude of the painful stimulus itself. Some drugs are capable of reducing the amount of pain that is formed by a noxious stimulus much in the same way robinul reduces the production of oral secretions. The general term for this effect is “preemptive analgesia” and it involves pain relief of greater magnitude and / or longer duration than can be accounted for by the direct analgesic effect of the drug. It appears that dexmedetomidine may have a notable preemptive analgesic effect. As we learn more about pain we can steadily reduce the pain our patients experience. And now I know a little more than I did before a colleague told me the story of his patient.


Michael A. Fiedler, PhD, CRNA

This abstract and comment topic suggested by Chris Womack ,CRNA.

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Lee L, Wang JJ, Yang YL, Chen A, Lai HY

Midazolam vs ondansetron for preventing postoperative nausea and vomiting: a randomized controlled trial

Anaesthesia 2007;62:18-22

Lee L, Wang JJ, Yang YL, Chen A, Lai HY


Purpose            The purpose of this study was to compare the antiemetic effects of a prophylactic dose of midazolam to ondansetron in patients undergoing sevoflurane induction and maintenance of general anesthesia.

Background            Postoperative nausea and vomiting (PONV) is common after inhalation inductions, reportedly occurring in 56% of individuals who undergo sevoflurane inhalation induction and maintenance of anesthesia. Some believe potent inhalation agents to be the greatest risk factor for PONV in a typical general anesthetic. Inhalation induction may increase the risk even more, perhaps increasing the overall PONV rate to 60%. Midazolam has been shown to have antiemetic properties.

Methodology            This prospective, randomized study included 90 non-smoking, healthy, female patients scheduled for hysteroscopy or ureteroscopy planned to last 1 to 2 hours. After preoxygenation, women underwent an inhalation induction with 8% sevoflurane and laryngeal mask insertion. Spontaneous respirations were maintained throughout induction and maintenance of anesthesia. Fresh gas flow during maintenance consisted of 1 L/min oxygen and 1 L/min air. Fentanyl up to 2 ?g/Kg was administered as needed. Thirty minutes before the end of the case patients received either 2 mg midazolam or 4 mg ondansetron. Each received 30 mg ketorolac. Nausea and vomiting assessments were made by a trained observer who was unaware of whether the patient had received midazolam or ondansetron.

Result            While PONV risk factors were not controlled, the average number of PONV risk factors was almost identical between groups. Demographic factors were comparable. Time to awakening, sedation scores, and time to readiness for discharge was no different between groups.

There were no significant differences in the incidence of nausea, vomiting, or nausea and vomiting between groups at any time during the first 24 hours postoperatively. Furthermore, the predicted incidence of PONV in the midazolam group was 55% but the actual incidence was only 30% (P=0.018). The predicted incidence of PONV in the ondansetron group was 53% but the actual incidence was only 27% (P=0.017).

Conclusion            In patients undergoing sevoflurane induction and maintenance of anesthesia, midazolam 2 mg near the end of the case was equally effective at preventing PONV as ondansetron 4 mg without prolonging recovery time.



It has been known that anxiety plays a role in PONV and I have taught my students to be aware of this fact and try to reduce preoperative anxiety as part of their overall strategy to minimize PONV. I’m guessing, though, that such a large portion of adult surgical patients already get midazolam preoperatively that there is little value in talking about “adding midazolam” for PONV prophylaxis.

This study captures my curiosity in a number of ways. Intuitively, it doesn’t seem like anxiety would be that big a contributor to PONV postoperatively unless pain was a problem. In this study, postoperative pain scores were low in both groups and almost identical. Was midazolam reducing PONV by reducing anxiety or does it have some more direct antiemetic effect?

Midazolam has a shorter elimination half life (1-4 hours) than does ondansetron (3-6 hours). And while midazolam patients received rescue antiemetics slightly more frequently (23%) than did ondansetron patients (18%) during the first 24 hours postoperatively, the difference was no where near statistically significant (P=0.597) so any perceived difference is unlikely to be real. Even if midazolam has some direct antiemetic effect why didn’t midazolam patients need more rescue antiemetics as time wore on? Why did the antiemetic effect of midazolam last so long?

Lastly, I had accepted that midazolam had some small antiemetic effect. But, I would have laughed at you if you had told me the prophylactic antiemetic effect of midazolam was equal to that of ondansetron under any circumstances. Yet this study presents some pretty convincing evidence that it does, at least in this circumstance. So, perhaps we have another prophylactic antiemetic to consider?


Michael A. Fiedler, PhD, CRNA

© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Policy, Process, & Economics

Kheterpal S, Gupta R, Blum JM, Tremper KK, O?Reilly M, Kazanjian PE


Electronic reminders improve procedure documentation compliance and professional fee reimbursement

Anesth Analg 2007;104:592-597

Kheterpal S, Gupta R, Blum JM, Tremper KK, O’Reilly M, Kazanjian PE



Purpose            The purpose of this study was to compare the rate at which arterial line placement was documented on the anesthesia record with and without electronic reminders to document the procedure.

Background            Commercial payors often require specific documentation as a condition of payment for anesthesia services. Payment of professional fees is thus inexorably linked to documentation specified by the payor. Missing documentation results in lost reimbursement and documentation is often omitted during high acuity patient care, high volume patient care, and when multiple clinicians are involved in the anesthesia care of a single patient. Evaluations of electronic reminders and documentation compliance in non-anesthesia settings have produced mixed results. Limited information regarding automated recordkeeping in anesthesia suggests that it may contribute to reducing the cost of care and/or increased reimbursement. In the year preceding the study 80% of arterial line insertions were documented in compliance with requirements for payment. Reimbursement for arterial line insertion averaged $83 at the study institution.

Methodology            Data on arterial line insertion and documentation was collected in patients scheduled for elective procedures over a two month period. An Anesthesia Information System (AIS) already in use at the institution was used to collect the data. The anesthesia record was kept using the AIS which included a user modifiable template for documenting the arterial line insertion note.

A system was developed by which the AIS would correctly identify when an arterial line was in use during a case. The system was validated prior to use in this study. Clinicians were divided into two groups without their knowledge or prior notification. A reminder group received reminders to complete an arterial line insertion note, during the case and for up to two days following, if an arterial line procedure note was not present in the anesthesia record of a patient who had an arterial line. Reminders were sent by alphanumeric pager and email. No reminders were shown on the anesthesia record display during the case. The control group received no reminders. No departmental communication addressed the need for improved documentation of arterial line insertion prior to the study period.

Statistical methods were not disclosed. The study was not designed to address issues of patient safety or legal liability in relation to documentation.

Result            During the two month study period, 670 arterial lines were placed in surgical patients. Of these, 267 (40%) were documented before surgical incision and before any reminders were sent. This left 403 arterial lines undocumented when the case began; 201 in the reminder group and 202 in the control group. The reminder group subsequently documented 88% of the previously undocumented arterial line insertions. The control group documented 75% (P<0.001). In total, the reminder group documented 93% of all arterial line insertions during the study period.

Conclusion            Electronic reminders resulted in statistically and financially significant increases in documentation of arterial line placement by anesthesia personnel.



There is good reason to believe that automated record keeping systems can help improve the completeness, accuracy, and legibility of computerized anesthesia records compared to hand written records. To help improve completeness the interface must be properly designed. A good user interface improves the completeness of documentation while reducing the time needed to write the procedure note.

With the information provided in this study, it appears that about 310 arterial lines were inserted each month at the study institution. At their baseline documentation rate of 80% they were not documenting 62 arterial lines a month. This resulted in $5,164 / month of lost revenue or $61,752 a year. (My calculations from data in the report.) Clinicians that got reminders ultimately documented 93% of arterial line insertions; an increase of 13 points over the previous 12 month baseline rate. This reflects documentation of 65% of all previously undocumented arterial lines and a $40,00 + yearly increase in revenue. Given that they already had an Anesthesia Information System in place, the authors calculated that it cost about $1,100 to realize this gain.

There is one aspect of this study that concerns me and raises questions about the validity of the reported results. Some arterial lines were documented before induction of general anesthesia and before any reminders were sent. Those procedure notes were written by physician residents and CRNAs who were already assigned to either the reminder group or the control group. If the tendency of the groups to document their arterial line placement was equal one would expect a fairly equal percent of arterial lines to be documented by the two groups before any reminders were sent. Yet, of the 267 arterial line notes written before reminders were sent, 60% were written by clinicians in the reminder group and 40% by clinicians in the control group. My quick ad hoc analysis (Chi-squared with Yates correction) shows the difference in documentation rate between these two groups to be pretty significant (P =0.017). This makes me wonder if there wasn’t some unidentified characteristic of the reminder group that predisposed them to a higher documentation rate. If so, it would bias the study in the direction of the reported results.

The baseline rate of arterial line documentation prior to reminders was 80%. The overall rate of arterial line documentation in the reminder group was 93% during the study period. Given that there were no departmental initiatives aimed at improving the documentation rate, if the results of the study are valid this is encouraging information. Applying similar reminders to other aspects of anesthesia charting linked to reimbursement might well capture a lot of lost revenue. I may be paid by the hour but this is where the money comes from that the anesthesia group pays me with.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Regional Anesthesia

Brull R, McCartney CJL, Chan VWS, Liguori GA, Hargett MJ, Xu D, Abbas S, El-Beheiry H

 Disclosure of risks associated with regional anesthesia: a survey of academic regional anesthesiologists

Reg Anesth Pain Med 2007;32:7-11

Brull R, McCartney CJL, Chan VWS, Liguori GA, Hargett MJ, Xu D, Abbas S, El-Beheiry H




Purpose            The purpose of this study was to identify the risks associated with regional anesthesia that were discussed with patients by academic anesthesiologists specializing in regional anesthesia. The percent of anesthesiologists who discussed each risk factor and magnitude of the risk (incidence of the complication) was also reported.

Background            Studies of what patients want to know about risk prior to anesthesia have produced a variety of results without agreement. A large proportion of patients appear to want information only about the main risks while a few want “full” disclosure. The authors believe that “anesthesiologists have a duty to accurately disclose the significant risks of the proposed anesthetic … as well as those that happen rarely but are severe in nature …” Yet, the most severe complications tend to occur quite infrequently, making it difficult to measure their incidence precisely. Many available studies of regional anesthesia risk suffer from limitations and may not be valid measures of the actual incidence of risk.

Methodology            This prospective survey was sent to 23 regional anesthesia fellowship programs in the USA and Canada with instructions to distribute the survey to all “regional anesthesiologists” and regional anesthesia fellows. Respondents picked from a list of complications associated with each type of regional anesthesia included in the survey. Next they picked from six different ranges of the incidence of the complication from “greater than 1 in 10” to a maximum of “less than 1 in 1,000,000.”

Result            Twelve of the 23 fellowship programs responded, three Canadian and nine from the USA. In total, 79 surveys were returned. Only nine of the surveys were completed by fellows.

In general, the most common and less severe risks were discussed frequently. Less common but more severe complications (paralysis, death) were discussed less frequently. Prior to epidural anesthesia, for example, local discomfort, infection, and headache were discussed by 82%, 73%, and 91% of anesthesiologists respectively while paralysis and death were discussed by 43% and 29% of respondents. For various peripheral nerve blocks infection was discussed by 62% to 74% of survey respondents depending upon the block. Permanent nerve damage was discussed by 45% to 77% of respondents depending upon the block. Cardiac arrest was discussed by 11% to 18% of respondents.

Seventy four percent of respondents said they discussed risks primarily to assist patients in making an informed choice while 26% said discussion of risk was primarily for legal reasons.

Conclusion            Relatively few academic regional anesthesiologists discuss severe risks of regional anesthesia with patients.



This study tries to provide some perspective on what the regional anesthesia “experts” think is important to discuss with patients prior to performing a block. The authors seem to be arguing for more universal discussion of almost all potential complications and for providing patients with more accurate data on the incidence of these complications. They write, “candid disclosure and accurate quantification of risks associated with regional anesthesia are imperative …” While the study had some important limitations I think there are larger issues than the study results and those limitations.

Accept, for the sake of discussion, that all or almost all risks of a particular regional anesthetic should be discussed with patients so they can make a decision about whether or not they are willing to accept the risk. An individual, at the least, would need information about the overall incidence of a complication. A “more informed” decision would be based upon a context specific incidence of the complication. The problem here is that many severe complications of regional anesthesia occur quite infrequently. Accurately determining the risk of these rare complications would require not only large numbers of patients but also meticulous methodology to exclude confounding variables. The bottom line is that we just don’t know with much certainty what the real incidence is of some compilations.

The overall incidence of an epidural hematoma in patients who have had an epidural anesthetic is probably about 1 in 190,000.1 But this overall incidence is not agreed upon and the 95% confidence interval is between 1 in 96,949 and 1 in 406,242. That is, based upon the information in the study, we are 95% confident that the true incidence is somewhere between those two numbers. There is quite a bit of uncertainty in that confidence interval.

Context specific risk factors are even harder to know. Patient health status, their medications, and provider factors, such as technique and skill level, all modify the overall risk of any given complication. The risk of an epidural hematoma following an atraumatic subarachnoid block in a healthy patient has been reported to be 1 in 320,000.2 If the subarachnoid block was traumatic (e.g. multiple needle passes) the incidence increases 11 times to 1 in 29,000. And if the patient was on standard heparin thromboprophylaxis and the block was traumatic the incidence of epidural hematoma is increased 112 times to 1 in 2,900. Do we know in advance how much tissue trauma is likely to be caused by the block procedure? Are we to approach the discussion of risk in the same manner in each of these circumstances?

Now let us assume that we know with certainty the incidence and magnitude of the risks associated with every regional anesthetic. In my experience, anesthesia providers approach the discussion of risk prior to regional anesthesia differently than they do prior to general anesthesia. Frequently, what the anesthesia provider chooses to discuss is not based in the available scientific information. For example, a 1990 closed claims analysis reported that 61% of nerve injuries associated with anesthesia occurred during general anesthesia.3 A 1999 closed claims analysis reported that over 80% of ulnar and over 70% of brachial plexus nerve injuries occurred during general anesthesia.4 Yet, my observation is that we emphasize discussion of nerve injury prior to regional anesthesia but rarely discuss it prior to general anesthesia. If, as the authors contend, the presentation of risk information influences a patient’s choice between a regional and a general anesthetic, then using different standards for what risks to discuss prior to regional and general anesthesia would likely bias the patient’s choice.

Lastly, some believe that all available information about risk must be communicated to a patient before they are capable of making an appropriate decision about whether or not to consent to a procedure. This is a standard that we do not hold to in any other area of society that I am aware of and one that is at odds with the generally held principle of personal autonomy. As a patient, is it not up to me how much I want to know before consenting to a procedure? I am aware that an incredibly small percentage of airline passengers die each year. The likelihood of death is probably related to factors specific to the airline, the pilots, the weather en route, and the model of the jet I’ll be riding in. Yet, when I book an airline flight the agent does not provide me with any information about the risk of death overall or related to factors specific to my flight. I don’t want it either. I’ve accepted the small risk to get where I’m going fast. One cannot argue that commercial aviation and health care are not comparable in this regard. In both cases the outcome of interest is death, the risk is very small and the actual number is not accurately known, and the activity (anesthesia or flying) is normally elective. Clarke has written an especially insightful essay on this aspect of informed consent that is worthwhile reading.5

This study showed considerably variability in the percent of anesthesiologists that routinely discussed a number of specific risks of regional anesthesia. I suspect there is also variability in when a single anesthesiologist discusses a specific risk with a patient. Some patients simply don’t want the information. Some risks are almost infinitesimally small. Some patients are at much greater or lesser risk for the same complication due to factors specific to their care. Some risks are not known with any agreed upon certainty. With all this uncertainty, it is no surprise that clinical practices regarding risk discussion are neither more complete nor uniform.

Please don’t misunderstand me. I am not arguing against discussing risks with patients. I am simply pointing out that this is an area in which there is much we don’t know. It is an area in which we have not yet devoted concentrated effort to determine what best serves the patient. It is an area that requires considerable professional judgment and in which there remains room for thoughtful anesthesia providers to take a variety of clinical approaches.


Michael A. Fiedler, PhD, CRNA


1            Wulf H. Epidural anaesthesia and spinal haematoma. Can. J. Anaesth. 1996;43:1260-1271.

2            Horlocker TT, Wedel DJ, Benzon H, et al. Regional anesthesia in the anticoagulated patient: defining the risks (the second ASRA Consensus Conference on Neuraxial Anesthesia and Anticoagulation). Reg. Anesth. Pain Med. 2003;28:172-197.

3            Kroll DA, Caplan RA, Posner K, Ward RJ, Cheney FW. Nerve injury associated with anesthesia. Anesthesiology. 1990;73:202-207.

4            Cheney FW, Domino KB, Caplan RA, Posner KL. Nerve injury associated with anesthesia: a closed claims analysis. Anesthesiology. 1999;90:1062-1069.

5            Clarke S. Informed consent without bureaucracy. J Clin Neurosci. 2003;10:35-36.


© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007

Beach ML, Sites BD, Gallagher JD

Use of a nerve stimulator does not improve the efficacy of ultrasound-guided supraclavicular nerve blocks

J Clin Anesth 2006;18:580-584

Beach ML, Sites BD, Gallagher JD


Purpose            The purpose of this study was to 1) define how often ultrasound showed the block needle touching a trunk of the brachial plexus without nerve stimulation resulting in a motor response and 2) establish whether or not a motor response to nerve stimulation predicted a block that was adequate for surgery.

Background            Electrical stimulation is commonly used to ensure the success of nerve blocks. The authors have been augmenting their block procedures with ultrasound guidance since 2002. When ultrasound has shown the block needle in contact with a nerve, a motor response to nerve stimulation has sometimes been absent. Paresthesia or a motor response to nerve stimulation may not be as specific an indicator of proximity to a nerve as has historically been thought. Tissue compression may result in these responses without direct needle to nerve contact. It is not uncommon for multiple needle passes to be used in search of an “adequate” motor response which is commonly held to predict an optimal block success rate. Multiple needle passes are associated with patient dissatisfaction and nerve injury.

Methodology            This retrospective database and chart review included 94 patients who received a supraclavicular brachial plexus block as their surgical anesthetic. A successful block was defined as one that provided adequate anesthesia for the surgical procedure in the absence of induction of general anesthesia. Blocks were performed using ultrasound imaging (12 MHz) and nerve stimulation (2 Hz, 0.5 mA, 0.1 msec duration) with a 22?gauge B. Braun Stimuplex block needle and nerve stimulator. Bupivacaine 0.5% with 5 ?g/mL epinephrine was injected when the tip of the block needle was observed to be in contact with a nerve trunk. Contact of the tip of the block needle with a trunk of the brachial plexus was defined by the presence of all three of the following criteria: 1) at least two trunks of the brachial plexus were visualized on ultrasound, 2)?the full length of the block needle was visible on the ultrasound, and 3) displacement of the nerve trunk was visible on ultrasound as the needle encountered the nerve. If the local anesthetic was observed by ultrasound to be spreading away from the nerve trunks, the needle was repositioned until local anesthetic was seen spreading around the nerve trunks.

Patients were categorized into two groups. The No Twitch Group (NT) included patients in whom the needle was observed to be in contact with a brachial plexus nerve trunk but no motor response was observed to nerve stimulation. The Twitch Group (T) included patients in whom the needle was observed to be in contact with a brachial plexus nerve trunk and a motor response was observed to nerve stimulation.

Result            While 94 blocks were performed, in only 74 (79%) was the definition for needle contact with a nerve trunk met. The T group included 64 patients and the NT group 10 patients. Of the 91 patients for whom general anesthesia was not part of the anesthetic plan before the block was placed, the brachial plexus block provided adequate surgical anesthesia 92% of the time. Overall, 90% of patients in group NT and 92% in group T did not require general anesthesia.

General anesthesia was planned for three patients in group T due to patient preference. One patient in group T received general anesthesia for an unknown reason. Four patients in group T received general anesthesia as a result of an inadequate block.

Ninety percent of group T patients had complete sensory block and 80% had complete motor block of all four named peripheral nerves. All patients in group NT had complete motor and sensory block.

Of the 20 patients in whom ultrasound visualization was inadequate, 18 had a motor response to nerve stimulation (as did group T patients). Only 1 out of 18 required general anesthesia resulting in a 94.4% success rate. Only two patients with inadequate ultrasound visualization did not have a motor response to nerve stimulation (as did group NT patients). Neither of them required general anesthesia.

No demographic variables (age, gender, weight, local anesthetic volume, number of block attempts) correlated with an absence of motor response to nerve stimulation (NT).

The time needed to perform the block was 10 minutes (sd = 4.3 min) in group T and 8 minutes (sd=4.2) in group NT (P=0.20). Paresthesias were elicited in three group T patients but none of the group NT patients (P=0.48).

Conclusion            In this series of 94 blocks, ultrasound showed the block needle touching a trunk of the brachial plexus in 74 patients (78.7%). Of those 74, nerve stimulation did not yield a motor response in 10 (13.5%). The authors concluded that the rate of successful blocks was unchanged whether or not a motor response to nerve stimulation was attained. Lack of a motor response in blocks performed with the assistance of ultrasound visualization does not appear to predict block failure.



Using ultrasound as an aid in nerve localization during regional block procedures has been in the literature for over a year now but my impression is that few are using it seriously. We all know that some just seem to have a gift for performing blocks, others are so-so, while a few appear hopeless. I wonder if ultrasound isn’t the tool that can bump everyone up a notch on the regional skills hierarchy. The gifted won’t need to bother with it. Those with so-so regional skills may be elevated to gifted status with proper application of ultrasound techniques. And those who have appeared hopeless in the past may be elevated to at least so-so status. I am watching the application of ultrasound visualization in regional anesthesia with anticipation. I included this study, despite its flaws, because it is among the first that tries to show the latent value of ultrasound in regional anesthesia. How can it not help to see where the local anesthetic is spreading during injection rather than waiting 20 minutes to see if it spread where we wanted it to?

Too often, when we are all taught the same thing (often from the same textbook) it becomes fact and we never think to question it. These authors are to be commended for remembering to question. Sadly, though, the study overall disappoints. While it does have some things to teach us, it suffers from a number of fairly obvious methodological problems. These problems are compounded by the sometimes disorganized way in which the results were reported and related to each other.

The first question was clearly addressed. In 10 of the 74 cases (13.5%) where ultrasound showed the block needle touching a trunk of the brachial plexus there was no motor response to nerve stimulation. Especially interesting, all these patients had a complete motor and sensory block. Only one of them (10%) required a general anesthetic for a failed block. While the number of patients with failed blocks was not large enough to compare statistically, this 10% failure rate is similar to the 8% of patients who did have a twitch and yet had a failed block.

The study wasn’t designed to answer the second question. That question asked, “is a motor response to nerve stimulation needed to predict a block that is adequate for surgery?” The presence or lack of a motor response to nerve stimulation was really a secondary finding in all groups of this study. This is so because the primary criterion was the ability to visualize the tip of the block needle in contact with a trunk of the brachial plexus. In the groups compared statistically, local anesthetic was injected only after ultrasound showed the tip of the block needle in contact with a nerve trunk. To investigate whether or not a motor response was predictive of an adequate block, the motor response would have to have been the primary end point, rather than the ultrasound appearance.

There was a group of 20 patients in whom ultrasound did not show contact with a nerve trunk. Two in this group did not have a motor response and 18 did. (Only one of these patients had a failed block.) But this group was not included in the statistical analysis. The analysis that was done was biased by the inclusion of three patients in whom general anesthesia was part of the overall anesthetic plan, and thus could not be used as evidence of a failed block. Another patient was included that received a general anesthetic for unknown reasons. All four of these patients were in group T (needle touching a trunk and a twitch to nerve stimulation) resulting in a bias against group T in regards to whether or not a motor response was predictive of block success.

The authors concluded that a twitch in response to nerve stimulation had “limited” predictive value for block success compared to ultrasound visualization. This conclusion was not demonstrated by the statistical analysis. The data would probably support the statement that the twitch response was no better a predictor of block success than ultrasound visualization.

The authors conclude that “… block success was not different when no twitch was obtained …” While the analysis does not support this conclusion the data look like it is probably true.

The authors conclude that “… the inability to elicit a twitch does not … predict block failure.” Again, the analysis does not support this conclusion but when I look at the raw data the conclusion is believable.

I wouldn’t hang my hat on the results of this study or its conclusions. Nevertheless, it does suggest some important latent benefits to using ultrasound for at least some regional blocks. It encourages me to seek more information on the uses of ultrasound in regional anesthesia.


Michael A. Fiedler, PhD, CRNA




© Copyright 2007 Anesthesia Abstracts · Volume 1 Number 1, March 30, 2007