ISSN NUMBER: 1938-7172
Issue 2.2

Michael A. Fiedler, PhD, CRNA

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

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

Assistant Editor
Jessica Floyd, BS

A Publication of Lifelong Learning, LLC © Copyright 2008

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

Table of Contents














In response to the suggestions of subscribers and would be subscribers, Anesthesia Abstracts now has Continuing Education Credit available for selected abstracts and comments. Once you have completed the CE quiz, your certificate is available immediately and, for AANA members, credits are automatically reported to the AANA (non-members file for recertification in the usual manner). When an issue offers CE credit in specialty areas, such as pharmacology, it will be noted on your certificate. Click on the green "CONTINUING EDUCATION" link on the left side of the web site for more information.

On a personal note I'd like to thank all of you who have chosen to be early subscribers to Anesthesia Abstracts. The first issue was published one year ago this month. Your support has enabled this publication to grow and develop. Your suggestions have inspired its improvement. With your continued subscriptions and comments we will work hard to make Anesthesia Abstracts an even better way to keep up with all the new anesthesia information -- quickly and economically.


Michael A. Fiedler


Levitan R.


Design rationale and intended use of a short optical stylet for routine fiberoptic augmentation of emergency laryngoscopy

Am J Emerg Med 2006;24:490-495

Levitan R.



Purpose            The purpose of this article was to evaluate the use of a short optical stylet to supplement routine and emergency intubation.

Background            Preoperative evaluation of the airway is important, but it does not always provide an adequate prediction of a difficult airway that may exist. Failed airway management continues to create significant morbidity and mortality. Studies show that three or more attempts to secure a difficult airway translates into a seven times greater risk of hypoxemia, regurgitation, and cardiac arrest. Several devices, often called “rescue” devices, have been developed to assist with the intubation of a difficult airway. Rescue devices are used infrequently and often require expertise. In order to gain expertise, providers need to use these devices during routine intubation but many devices create logistical problems with setup and use, often leading to a prolonged induction. A short optical stylet, called the Levitan FPS Stylet, is designed to augment direct laryngoscopy and to be used routinely without difficult setup or significant logistical issues.

Methodology            The device is a malleable stylet of optical fibers connected to an eyepiece and a removable light source. The stylet will accept a standard endotracheal tube. The stylet is designed to be positioned distal to the tip of the endotracheal tube (ETT) very much like a conventional stylet. A port allows blowing of oxygen through the tip of the stylet keeping the distal point clear of secretions. The device does require preparing the ETT for proper fit. This particular device is designed to be used as an adjunct to direct visualization with a conventional laryngoscope. The stylet is guided under the epiglottis where the eyepiece is then used to direct the ETT into the trachea. The tube is then slid off the stylet, the ETT cuff is inflated, and the tube secured.

Result            This device is said to be easier to use than other optical stylets because it is shorter. It costs considerably less than a flexible bronchoscope, and this device may be cleaned with standard cleaning solutions.

Conclusion            Many practitioners continue to rely on a seldom used emergency intubation device when an unexpected difficult airway arises. Lack of experience with emergency devices could create problems for the practitioner and patient in a difficult situation. It is suggested that the FPS Stylet be used during routine intubations in order to prepare the practitioner to deal more effectively when an unexpected difficult airway presents itself.



I have not used this particular device, or any other optical stylet, but I absolutely agree with the author that when we find ourselves in a difficult airway situation we often struggle to secure that airway with techniques and devices that we seldom use.

My practice covers small rural facilities and I often do not have control over the types of airway devices available. I think that many CRNAs who practice in multiple rural facilities, or those who do locum work can understand this situation. For many years, a dedicated flexible fiberoptic laryngoscope, or in many small institutions a shared fiberoptic bronchoscope was the only available rescue device. These devices required a bulky light source and the shared devices were not always immediately available. Most facilities have invested in different rescue devices over time, and each device takes a somewhat unique skill and understanding of how to best use it. When the patient’s airway becomes compromised and oxygen saturation is rapidly falling, it is not the time to learn or relearn how to use a rescue device.

The author suggests, and I strongly agree, that when choosing among the many airway rescue devices available today, that you choose one that is compact, dedicated to your use, and something that you can conveniently use on routine cases in order to maintain a skill level adequate to be used when the dreaded unexpected airway emergency arises.

Technology has advanced far enough to give us compact light and video sources attached to hardware that is intuitive and easy to use. Loss of airway continues to be one of the most significant morbidity and mortality problems in anesthesia today. There is no excuse for a facility and provider not to have a well established plan and readily available rescue devices for the management of such emergencies. I predict that in the near future, we will no longer be using the direct aluminum laryngoscopes we routinely use today and all of us have used throughout our careers.


Steven R. Wooden, MS, CRNA



© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Crabtree NA, Chandra DB, Weiss ID, Joo HW, Naik VN

Fibreoptic airway training: correlation of simulator performance and clinical skill

Can J Anesth 2008;55:100-104

Crabtree NA, Chandra DB, Weiss ID, Joo HW, Naik VN


Purpose            The primary purpose of this study was to determine whether or not the time needed to place a fiberoptic scope for intubation on low fidelity or high fidelity simulators predicted clinical fiberoptic skill. A secondary purpose was to compare the time needed to place a fiberoptic bronchoscope in a patient’s trachea with fiberoptic skill as scored on a checklist and a global rating scale.

Background            Fiberoptic intubation is an important airway management tool. Historically, this skill has been taught in the operating room. Production pressure and other factors are making it more difficult to learn fiberoptic intubation in the OR. Airway simulators have been developed in an effort to improve learning and supplant lost opportunities to practice fiberoptic intubation in the OR. High fidelity simulation in general has been shown to facilitate learning when used in conjunction with feedback, repetitive practice, and clinical correlation. Low fidelity and High fidelity fiberoptic airway management simulators have been developed. Both have been shown to facilitate learning of fiberoptic airway skills to a greater extent than classroom lecture alone. Like other technical skills, fiberoptic airway management can be assessed with a checklist, a global rating scale, and by the time required to complete the task. Whether or not the time required to complete fiberoptic airway management on a simulator accurately predicts the ability of an individual to perform the same task in the clinical area is unknown.

Methodology            This prospective, randomized, single-blind study included 30 registered respiratory therapists who had not independently completed two or more fiberoptic bronchoscopies. They were divided into two groups of 15 each. One group received hands on fiberoptic training on a low fidelity airway simulator. The low fidelity simulator was a non-anatomic box designed to allow practice with a fiberoptic bronchoscope. The other group received hands on fiberoptic training on a high fidelity airway simulator. The high fidelity simulator was a virtual reality bronchoscope (Accutouch® endoscopy simulator, Immersion Medical, Gaithersburg, MD). All subjects received standardized written and verbal instruction on use of the fiberoptic bronchoscope before simulation training. Simulation training time averaged about 30 minutes, but lasted no more than one hour. Informal guidance from an experienced fiberoptic bronchoscopist was continuously available during simulation training. After simulation training was complete, each subject’s fiberoptic bronchoscopy skill was tested twice. First, they were tested on the same type of simulator they used during training. Second, they were tested on consenting patients during general anesthesia and paralysis. Patients with a body mass index >25 kg/m2 and those with a difficult airway were excluded. Patients did not receive a drying agent. Fiberoptic performance in the OR was assessed by two attending anesthesiologists who were unaware of the training group the respiratory therapist was in. Performance was assessed with a checklist, a global rating scale, and by measuring the time required to perform fiberoptic bronchoscopy. (The checklist and global rating scale had both previously been validated as measurement tools.) If fiberoptic bronchoscopy took longer than three minutes to complete or the patients oxygen saturation fell below 94% the attempt was graded a failure.

Result            Twenty-eight subjects were analyzed. One subject in each group did not complete the study. The checklist and global rating scale scores recorded by the two anesthesiologists were in close agreement (r=0.9 and 0.85 respectively).

Six respiratory therapists in each group (43%) failed to place the fiberoptic bronchoscope in the patient’s trachea within three minutes and failed the clinical test. There was no correlation between the time to place the bronchoscope in either the low fidelity or high fidelity simulator and fiberoptic skills demonstrated clinically. There was a highly significant correlation between the time required for subjects to place the fiberoptic bronchoscope in the trachea of patients and both the checklist and global rating scale scores assigned by the anesthesiologists (P<0.001).

Conclusion            The time required to successfully place a fiberoptic bronchoscope on a low or high fidelity simulator does not correlate with clinical fiberoptic skill. The time required to successfully place a fiberoptic bronchoscope in the trachea of a patient correlated well with other measurements of clinical skill and is thus a usable proxy for assessing this skill.



Simulation is assumed by some to be at least as good as, if not superior to, the traditional model of clinical education. This is a seductive thought because, unlike real patients and real clinical situations, simulators can be controlled, they are available at any time, and you can do it over and over until you get it right. But I’m a skeptic for now. There is little evidence that currently available simulation results in demonstrated clinical skill in anesthesia. (And simulation is very expensive.) The investigators note that a systematic review of simulation showed that “high fidelity” simulation facilitated learning when combined with repeated practice, feedback, and integration with clinical education. I’m left wondering if people wouldn’t learn just as well if they had repeated practice, feedback and integration with clinical education without simulation. That said, a few institutions have clearly put simulation to good use, at least in teaching and evaluating things that are difficult to teach or evaluate in the OR in the first place (e.g. crisis management of rare events). It is with this skepticism that I approach the study we look at here.

This study stands out because of its high quality methodology and analysis. The statistical analysis was conservative and appropriate for the type of data produced. The checklist and global rating scale used to measure the skill level of the subjects had been validated. Two investigators evaluated each subject’s fiberoptic skill level and their evaluations had a very high level of agreement, providing further evidence that the skill evaluation was a valid measurement. I am confident I can believe what this study has to teach. One thing it teaches me is that, at least where learning fiberoptic bronchoscopy is concerned, how long it takes to perform a task on a simulator doesn’t say anything about how well I can perform the task in a real patient. This should give pause to those who want clinicians to “demonstrate their skills” on a simulator as part of a certification or credentialing process. The other thing this study teaches is that the time it took to place the bronchoscope in the trachea of a real patient did correlate with other tests of bronchoscopy skill. The faster subjects placed the bronchoscope in a patient the higher their score on other assessments of fiberoptic skill. That is helpful because time then becomes an easy way to test skill level with a bronchoscope.

The study did have a couple important limitations; the previous experience of subjects learning fiberoptic bronchoscopy and the time they had to learn it. The investigators stated that respiratory therapists were excluded from the study if they had “… independently completed two or more fiberoptic bronchoscopes (sic).” Thus, if they had independently completed only one bronchoscopy they were eligible for the study and, presumably, had some prior training in fiberoptic bronchoscopy. Prior training and experience with bronchoscopy could bias the study. The other limitation was the brief duration of training in bronchoscopy that subjects had prior to their skills being tested. It is quite possible that more training would have resulted in much greater skill in fiberoptic bronchoscopy and, thus, markedly different study results. As a result, this study reinforces my concern that simulation may not be as useful as some would like it to be. But … even I will have to admit that the reason it doesn’t show a benefit to simulation may have to do with the aforementioned bias and/or the very limited simulation training the study subjects received.


Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Equipment & Technology

Sandberg W, Sandberg E, Seim A, Anupama S, Ehrenfeld J, Spring S, Walsh J


Real-time checking of electronic anesthesia records for documentation errors and automatically text messaging clinicians improves quality of documentation

Anesth Analg 2008;106:192-201

Sandberg W, Sandberg E, Seim A, Anupama S, Ehrenfeld J, Spring S, Walsh J


Purpose            The purpose was to assess the efficacy and outcomes of a newly developed custom software system which was built to be interactive with a currently used anesthesia information management system (also known as the electronic anesthesia record). The custom software alerts anesthesia providers via a text page alert, when clinically significant pieces of information are not entered into the anesthesia record within 15 minutes of the start of the anesthetist-patient relationship. Anesthesia information management systems (AIMS) function as a databank of patient demographics and clinical information during the anesthetic. Efficacy and outcomes of this text page alert system were defined as the answers to the following questions:  1) Does a one-time alert become sufficient for changing an anesthetist’s documentation performance?  2) How quickly does performance on documentation tasks change once the reminder is activated?  And 3) How does an alerting system such as this act to improve performance?

Background            Complete and thorough documentation is unequivocally necessary for an anesthesia record. Comprehensive documentation facilitates communication regarding the anesthetic itself as well as facilitates communication with subsequent providers who are downstream in the healthcare process. Just as important, comprehensive documentation in the operating room is how we, as providers, demonstrate appropriate intra-operative diligence and care. Reimbursement is becoming increasingly complex and often is based on provider compliance with mandated acts; those that are deemed to improve outcomes of care and therefore elevate the quality of care received. The omission of documented acts leads to reductions in revenue or payment for services; it truly questions the care and diligence of the provider. AIMS can identify missing data and logical inconsistencies in documentation that would, in addition to identifying problematic situations, reduce revenue and payment for services if left uncorrected.

Methodology            Software was originally developed by the investigators for the purpose of scanning the anesthesia record for documentation errors; errors that would prevent billing. This occurred after the anesthetic was completed and prior to charges being sent. This original software was modified to scan the anesthesia record in real time to discover missing information. For the purpose of this study, the system identified that nothing was entered into the allergy section of the anesthesia record. The missing allergy information would interact with the AIMS and generate an alphanumeric text page through the hospital paging system. The provider was alerted in real time via text page, when the allergy data was missing.

The start of anesthesia care was operationally defined in this study as the start of the anesthesia-patient relationship. It was the point when a case was begun and data recording was initiated. The relationship start time was typically in the preoperative holding area and not meant to be defined as the induction start time. Allergy documentation was considered incomplete if no information was entered within 15 minutes of the start of the relationship which coincided with the start of the establishment of the record. After 15 minutes, a one time prompt was sent via pager to the clinician. The 15 minute interval was chosen because it was felt the anesthetist should complete documentation of the preanesthesia events, which included allergy comments, before induction. This data piece was selected for study because it should be ascertained and recorded at the start of the relationship and before induction, which makes it safer if care should be handed off.

Pages were sent on weekdays between 7 am and 6 pm. To measure effectiveness of this one time page text alert system, 6 months of anesthesia records were reviewed prior to the start of this study. This established a baseline rate of missing allergy information. It determined how often allergy data was lacking relative to the numbers of cases done each day. (Note that the original software system, which alerted providers about documentation errors and potential billing problems, had been fully implemented for one full year before the experiment described here was implemented. Providers were accustomed to receiving pages regarding problems with documentation related to billing). After this text page alert system was initiated regarding allergy data omissions, data was collected for nine months. The data gathered included: the percentage of time an alert occurred relative to the number of cases done in each day, and, after alerts stopped (end of data collection), the percentage and number of times allergy information was omitted without alerts, relative to the number of cases done each day.

Result            Baseline data gathered before the start of this study (six months prior), determined the fraction of charts without an allergy comment in each 5 day sample to be greater than 30%. Once the researchers starting sending the one-time alert pages reminding clinicians whose charts lacked allergy documentation, a significant improvement in performance for entering allergy data was immediately seen. Within 10 days of activating the pager alert system, only 8.2% of charts lacked a comment in the allergy field. This demonstrated a nearly fourfold reduction (relative to the numbers of cases done on each day) of the documentation error rate, and was achieved within days of the intervention. After a period of six months, omissions in the allergy data field crept back up, but performance remained better than the baseline measured before the trial. In addition to reminding providers during the first 15 minutes to correct the missing allergy information data field, the system seemed to have taught people to add allergy data early on, to prevent the text page. Interesting also to note, nurse anesthetists performed less than 20% of all cases with the majority being done by residents and/or attending anesthesiologists. Nurse anesthetists corrected a significantly larger fraction of their records than did the residents, and in turn the residents corrected more records than the attending anesthesiologists.

Conclusion            This study indicated that one time messages sent via a text page alert to clinicians about particular documentation errors (allergy data field not completed) improved performance of this particular task. Improvement occurred within days of beginning the intervention and persisted for just under two months after suspension of the alert system. While compliance wavered after the two months of suspension of the alert system, the ongoing compliance remained greater than the baseline rate. The process was employed in 50 operating room suites, during specific times of the day, and for the typical Monday through Friday work week. All providers participated irrespective of their role; for example, anesthesiology residents, nurse anesthetists, and MD anesthesiologists. The clinicians, having worked with an electronic anesthesia record, were used to page alerts regarding their documentation and billing requirements, but other than that had not received further instruction. The process resulted in improved compliance without extensive training.



Electronic anesthesia information systems are becoming increasingly more commonplace and certainly more complex. The electronic medical record now plays an integral part in creating and promoting a culture of safety. This complexity of technical systems does add to our need for an expanded knowledge base and is most likely synergistic with increased vigilance and improvement in care delivery. Never taking our focus from the individual patient as a human being, real time data input into an information system coupled with real time analysis and alerts to potential mistakes is an innovative way to promote the culture of safety. While documentation is critical for more than compliance and reimbursement, being “reminded” to document may also play a role in reminding us to perform an intervention or task in general. Everything we have to do to begin an anesthetic or treat and stabilize a critical patient, or even cautiously, skillfully, and methodically deliver care to low acuity patients, can challenge the most prudent provider. Technology of this type doesn’t replace the human brain, it is simply a tool to increase the safety of anesthesia delivery.

Mary A. Golinski, PhD, CRNA


Not all may be familiar with the statistical process control, or SPC methodology used. The process involves X-bar charts, which is a common way to chart or graph variables while testing for systematic performance changes. SPC is often used to monitor performance over time. The improvement/non-improvement in allergy documentation before and after the text page system was deployed was monitored over time, including also when the page system was stopped. When analyzing the data, the researchers plotted the fraction of charts lacking allergy comments as a function of calendar date. The SPC methods were developed for use with sampling techniques; where only a fraction of the work output is tested. This makes sense as the system was used only on week days and during a specific daytime work shift, not a 24/7 paging and data gathering process. Using this method, the data gathered is grouped into samples of a reasonable size before the analysis is performed. These researchers created uniform samples of five sequential days, Monday through Friday.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Gandsas A, Parekh M, Bleech MM, Tong DA


Robotic telepresence: Profit analysis in reducing length of stay after laparoscopic gastric bypass

J Am Coll Surg 2007;205:72-7

Gandsas A, Parekh M, Bleech MM, Tong DA



Purpose            To evaluate the financial implications of using robotic telepresence in combination with standard postoperative rounds to reduce inpatient length of stay following laparoscopic gastric bypass surgery.

Background            Health care costs in the United States are projected to increase to $4 trillion in 2015, representing 20% of the gross domestic product. The combination of low reimbursement rates for health-care services and continuously increasing health-care costs create financial pressures as institutional operating margins decrease. Decreasing the hospitalized patient’s length of stay is one attractive potential solution to reduce hospital expenses and create additional capacity for growth.

Robotic telepresence was achieved using a human-sized Intouch Health RP7 mobile incorporating 2-way video communication via high resolution cameras, a directional microphone, and a 15-inch video screen. The robot was navigated throughout the hospital using a Windows-based computer via a wireless network accessible from either the facility’s Ethernet or a virtual private network. All data transmitted through the system were encrypted.

Methodology            Retrospective analysis was conducted of records from 376 patients receiving a Roux-en-Y gastric bypass for obesity. All surgical procedures were completed by the same surgeon. The first 284 patients received traditional post-operative care whereas the subsequent 92 were followed post-operatively by the same schedule of physical post-operative rounds augmented by up to 4 robotic rounds: one on the evening of surgery and up to three visits on the first post-operative day. Standardized criteria were used to determine readiness for discharge. Length of stay data were combined with hospital accounting information to determine cost savings.

Result            The mean length of stay for patients receiving supplemental robotic rounds was 1.26 days, with 77% of patients discharged on the first postoperative day and no patient remaining hospitalized beyond 3 days. In contrast, the mean length of stay for patients receiving only traditional rounds was 2.33 days, with 77% of patients discharged on the second post-operative day. Six percent of the traditionally managed patients were discharged after the 3rd post-operative day. Based on a $3,800 average profit for each new hospital admission, and assuming a 76% reoccupancy rate for vacated beds, implementing the robotic telepresence strategy resulted in an annualized benefit of $358,180.

Conclusion            Robotic telepresence reduced costs associated with patient boarding and increased hospital capacity. Monitoring patients more frequently through telepresence can help physicians anticipate events potentially prolonging hospital stays, improve pain and nausea management, and provide opportunities for encouragement of patient activities such as ambulation and incentive spirometry.



I was attracted to this article because of its uniqueness: a novel application of advanced communication technology in surgical care. After reading the article, I still think that using a robot to conduct patient rounds is “cool” and apparently I’m not alone in thinking this way. Ellison et al reported high patient satisfaction with robotic follow-up after urological surgery.1   Nurses appreciated the increased opportunity for collaboration with physicians after interactions with an RP6.2  But aside from the novelty factor, there doesn’t seem to be much science associated with this report. This study’s design is a classic illustration of the idiom that association does not equal causation. No evidence was provided to support the premise that the decreased length of hospital stay should be attributed to robotic patient rounds. Because the patients were assigned to groups chronologically, with the traditional group receiving surgery first; shorter hospitalizations in the group where physician rounds were supplemented with robotic telepresence may have been the result of the increased experience of the surgeon and entire health care team. And without a human comparison group, readers are left to wonder whether similar, or even better, outcomes could have been achieved by employing an advanced practice nurse to conduct the robot’s patient rounds.

I also have concerns regarding author impartiality. The Journal of the American College of Surgeons is praiseworthy because it consistently requires authors to disclose potential conflicts of interest. According to the disclaimer published with the manuscript, there are no author competing interests; however, the primary author is listed on the Intouch Health website as an “Applications and Clinical Advisor.”  Competing interests are not uncommon. Many advances in health care can be attributed to research sponsored by manufacturers, or conducted by investigators with close or frequent ties to manufacturers; however, alerting consumers is essential when the potential for investigator bias exists. Most every CRNA has attended corporately-sponsored lectures, and when speaker affiliations are openly disclosed, we listeners have been forewarned to approach the information presented with our “eyes wide open.”  In the case of this article, readers were assured that no such caution was necessary.

Moving beyond the methodological concerns, I am still intrigued by the notion of robots scurrying around our hospitals. I can easily imagine a scenario where all the passageways of a facility are digitally mapped so that an RP7 can be instructed to navigate to a certain room and the robot will be standing by to examine a patient as soon as the surgeon “breaks scrub”, without the surgeon ever leaving the operating room. Or dare I ask … could TEFRA conditions for medical direction of anesthesia care be met through robotic telepresence?


Alfred E. Lupien, PhD, CRNA


1.  Ellision LM, Pinto PA, Kim F et al. Telerounding and patient satisfaction after surgery. J Am Coll Surg. 2004;199:523-30.

2.  Petelin JB, Nelson ME, Goodman J. Deployment and early experience with remote-presence patient care in a community hospital. Surg Endosc. 2007;21:53-6.

Additional information about Intouch Health and remote telepresence can be found at the following URL:

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008


Comfere T, Sprung J, Case K, Dye P, Johnson J, Hall B, Schroder D, Hanson A, Marienau M, Warner D


Predictors of mortality following symptomatic pulmonary embolism in patients undergoing noncardiac surgery

Can J Anesth 2007;54:634-641

Comfere T, Sprung J, Case K, Dye P, Johnson J, Hall B, Schroder D, Hanson A, Marienau M, Warner D


Purpose            The perioperative period is associated with a higher incidence of venous thromboembolism than the incidence in the general population. Limited knowledge is available about mortality rates in those who suffer a perioperative pulmonary embolism (PE). The purpose of this study was twofold; to determine mortality rates after perioperative PE and to determine if any relationships existed between demographic factors, comorbidities, or surgical procedures and mortality.

Background            In the general population, the incidence of venous thromboembolism is one to two per 1,000 people. PE remains a significant cause of perioperative morbidity and mortality despite thromboembolism prophylaxis. Why is the incidence of PE high in the surgical population?  Current evidence points to a hypercoagulable state that may occur during major surgery and extend into the postoperative period. What is the cause of the hypercoagulable state? It is theorized that physiologic stress plays a key role in the etiology. General risk factors for PE include:  malignancy, prolonged immobility, age, and cardiovascular disease. Surgical risk factors include: the type and duration of surgery and the type and use of prophylactic measures. There is a possibility that if perioperative risk factors associated with mortality from PE can be identified, more effective treatments could be developed.

Methodology            A computerized search was completed for patients coded with the diagnosis of PE between January 1, 1998 and December 31, 2001. This information was merged with the anesthesia database to identify the subset of patients diagnosed with PE within 30 days after surgery. Only those undergoing non-cardiac surgery with general or neuraxial anesthesia were included. The Clinical Pathology Autopsy Registry was also reviewed for deaths related to PE during the same period. PE was confirmed by:  clinical diagnosis coupled with presentation typical for PE, a clinical notation referring to the event as a PE, the initiation of therapy for PE, confirmatory diagnosis by ventilation/perfusion lung scan, pulmonary angiography, computerized tomography/spiral scan, lower extremity Doppler, and/or autopsy. Excluded from the study were those who were coded with a diagnosis of PE but did not receive any treatment and those who had negative diagnostic imaging. The primary objectives were to determine the 30 day mortality following perioperative PE, and to determine the risk factors associated with 30 day mortality.

The following risk factors were considered in the data collection:  age, gender, ASA physical status, body mass index, and tobacco use status. The co-morbidities that were considered included: a history of hypertension, cardiac disease, peripheral vascular disease, stroke or immobility, chronic renal insufficiency, diabetes mellitus of all types, pulmonary disease, deep venous thrombosis or PE, current malignancy, use of estrogen replacement therapy, and use of various antithrombotic prophylactic strategies preoperatively or perioperatively. Identified also were surgery or anesthesia related factors including: type of surgery, length of surgery, time of the diagnosis of PE in relation to surgery, and intra-operative use of blood products. Additionally, if a central line or pacemaker was inserted at any point for surgery or during three months before the surgery, this was documented. Factors related to the presentation and treatment of the PE were noted for all, and included the need for admission to the ICU, the initiation of mechanical ventilation after the PE, the placement of an inferior vena cava filter during the same hospitalization, and the need for vasopressor support. The cause of death was determined for all patients who died by the 30th day following PE. Death was considered PE-related if the PE was confirmed via autopsy and if PE was listed as one of the diagnoses on the hospital medical summary.

Result            A total of 365 patients met the criteria for perioperative PE in the time period analyzed. Review of the medical records identified 158 patients who met the inclusion criteria for perioperative PE. Patients were diagnosed with perioperative PE in the following manner:  CT/MRI/SPIRAL scan-68%, clinical impression-14%, ventilation/perfusion scan 10%, lower extremity Doppler 3%, autopsy 3%, and angiography 2%. The incidence of mortality at 30 days was 25.3% or 40/158 patients. PE was either the primary or a significant contributory factor to 30 day mortality. Survival following a diagnosis of perioperative PE was 63%. Among patients who survived a minimum of 30 days following PE, the estimated survival at 30 days plus one year, equaled 83%.

Univariate statistical analysis revealed that a higher ASA physical status classification; the presence of peripheral vascular disease; long surgery time; the transfusion of red blood cells, fresh frozen plasma or platelets; and central vein catheterization within the three months preceding the PE were significant predictors of increased 30-day mortality (p < 0.05). Statistically significant factors related to the severity of PE presentation included: the need for vasopressor or inotropic support, ICU admission, and the requirement for ventilatory support. A logistic regression analysis was done, and after eliminating non-significant variables, 30-day mortality was found to be associated with vasopressor or inotropic support (P< 0.001), diabetes mellitus (P= 0.004), older age (P= 0.003 and longer duration of surgery (P= 0.012).

It was assumed that a clinically significant PE would be consistently noted in the medical record of the patient who experienced the PE, however, subclinical PEs were not necessarily diagnosed in the absence of a consistent surveillance strategy. Missing data regarding the numbers of subclinical PEs may have produced a bias towards increased mortality. Additionally, surgical patients may have been lost in the follow up process, so not all PE cases that occurred in the perioperative period were captured. Some patients may have suffered a fatal PE within 30 days post-operatively without the event being documented in the medical record.

Conclusion            In this study, mortality after perioperative PE fell within the lower range of mortality following PE in the general population. This may be due to the fact that prophylactic measures were taken in the hospital to prevent PE, and compliance in the general population with preventative measures may not have been as strong. Strong predictors of 30 day mortality included hemodynamic instability at the onset of the PE that required vasopressors. Almost all patients who required major vasopressor support following their PE died. This most likely reflected the severity of the emboli. Additionally, prolonged surgery may have favored thromboembolism during the surgical period. It is possible that patients who experienced a perioperative PE may clearly benefit from more aggressive interventions such as pulmonary embolectomy or even extracorporeal membrane oxygenation (ECMO). Treatment of these patients warrants further investigation given the high mortality rate.



While pulmonary embolectomy and ECMO may be considered for those experiencing severe PE, these treatments are simply not an option if the facility does not have the resources to provide them. Those at high risk for PE, including the elderly with a history of diabetes and those having lengthy surgical procedures, are often seen in small community hospitals. Is it feasible for small facilities to offer these procedures?  Should patients at high risk have their surgical procedures done only at the most resource rich institutions?  The answer to these questions will require a more definitive picture of the etiology and pathophysiology of PE. With a more definitive understanding of what is occurring in individuals who suffer a perioperative PE, interventional strategies including new or different pharmacotherapy or even less “resourceful rich” techniques may be discovered.


Mary A. Golinski, PhD, CRNA


The “need for vasopressor support” was operationally defined as arterial hypotension requiring bolus epinephrine, continuous infusions of dopamine, dobutamine, epinephrine, norepinephrine and/or vasopressin to treat the hypotension.


Statistical analyses involved the univariate association between each of the candidate risk factors and 30-day mortality. This was evaluated using the two-sample t test for continuous variables and Fisher’s exact test for categorical variables. Patient’s age, body mass index, procedure duration, and the number of days from surgery to PE were analyzed as continuous variables; all other variables were analyzed as categorical variables. An exploratory multivariable logistic regression analysis was performed to identify those variables demonstrating the strongest independent associations with 30-day mortality. This analysis appropriately included those variables expected to be related to survival such as age, ASA ranking, peripheral vascular disease, diabetes, type of surgery, length of surgery, transfusion occurrence, vasopressor and ventilatory support. 

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Obstetric Anesthesia

Cohen SE Andes LC, Carvalho B


Assessment of knowledge regarding cardiopulmonary resuscitation of pregnant women

Int J Obstet Anesth 2008;17:20-25

Cohen SE Andes LC, Carvalho B


Purpose            The purpose of this study was to assess the understanding of pregnancy specific ACLS protocols amongst anesthesiologists, obstetricians, and emergency physicians (both residents in training and specialty certified).

Background            Cardiac arrest is rare in pregnant women but may occur due to complications of pregnancy (e.g. amniotic fluid embolus) or factors unrelated to pregnancy (e.g. trauma or preexisting cardiovascular pathology). Most health professionals have not had first hand experience with resuscitation of a pregnant patient. Normal physiologic changes due to pregnancy require important changes to Advanced Cardiac Life Support (ACLS). Standard ACLS algorithms and training courses do not include modifications to ACLS protocols necessary in pregnant women.

Methodology            A 12 question descriptive survey was developed by a group of five subspecialty-trained obstetric anesthesiologists. The survey covered four areas of knowledge key to the successful resuscitation of a pregnant woman. Those four areas were: 1) the need for left uterine displacement, 2) ACLS algorithms, 3) physiologic changes accompanying pregnancy, and 4) a recommendation to consider immediate cesarean section in arrested women > 20 weeks pregnant after 5 minutes of unsuccessful ACLS. The survey was not validated prior to use. The survey was administered to 75 physicians in a single academic medical center in the presence of an investigator. The medical center logged between 5,000 and 6,000 deliveries a year. All physicians surveyed provided care for pregnant patients. Participants were surveyed in the hospital, during the day, while performing their usual clinical duties. No reference materials were allowed during completion of the survey.

Result            One of the 12 questions was excluded from analysis because obstetricians and anesthesiologists failed to agree on the correct answer. One of 75 surveys was excluded from analysis because it was not completed in the presence of the investigator. Of those who completed the survey, 43% were anesthesiologists, 37% were obstetricians, and 20% were emergency physicians. Sixty-two percent were residents and 38% were attending physicians. Only about 50% of participants had completed an ACLS course in the previous two years.

Average scores on the 11 item multiple choice exam were 76% (±15%) for anesthesia, 63% (±16%) for obstetrics, and 72% (±13%) for emergency. Residents as a group scored an average 70% (±17%) and attending physicians 74% (±13%). Those who participated in an ACLS course in the previous two years did not score significantly differently than those who had not participated in an ACLS course in five years or longer. Only 15% of all participants scored above 85%. Up to 43% of specialty physicians caring for pregnant patients failed to demonstrate adequate knowledge of left uterine displacement or the role of cesarean delivery during resuscitation upon which to base clinical decisions.

Conclusion            The anesthesiologists, obstetricians, and emergency physicians surveyed lacked knowledge required for the successful resuscitation of pregnant women. The authors recommended that pregnancy specific resuscitation information be added to ACLS training for these specialties.



Successful resuscitation of a pregnant woman requires several important changes to widely accepted ACLS protocols. This simple study described the level of understanding of those changes among physicians who cared for pregnant women at a large academic medical center. From a research point of view, the study was lacking. The survey wasn’t validated, the sample was neither random nor representative of the larger population, and the “test” questions had some problems from an educational perspective. But what the study lacked in methodologic rigor, it made up in simplicity and in raising awareness. And the results weren’t pretty. Most physicians fell short of what any well educated person would consider a passing score. (And I have no reason to believe that nurse anesthetists would have fared any better.)

The group surveyed was probably not representative of anesthesiologists, obstetricians, or emergency physicians as a whole. This limitation was compounded by the lack of a random sample in the survey participants. But I would argue that, if anything, a more representative sample would probably have performed even worse. The physicians surveyed cared for pregnant patients. On average, over 13 babies a day were delivered at the institution so they cared for a lot of pregnant women. This would not be true for all physicians in a more representative sample. They all worked at an academic institution and some were in residency training. Education should have been active and ongoing at such an institution. It seems reasonable to believe that those working in non-academic settings where few pregnant women are cared for might be even less knowledgeable about the resuscitation of pregnant women.

So what is this additional critical knowledge needed to successfully resuscitate a pregnant woman? As reported by the authors of the survey it is this.

Left Uterine Displacement of at least 15° - the gravid uterus can obstruct the abdominal aorta and/or inferior vena cava further limiting the limited cardiac output provided by chest compressions.

Perform Chest Compressions 1-2 cm Higher on the Sternum in the Term Pregnant Woman – this allows for the proper depth of chest compressions.

Consider Immediate Cesarean Section at >20 Weeks Gestation if Response is Inadequate to 5 Minutes of ACLS – left uterine displacement may be inadequate to restore circulation in some women. There are numerous reports of successful maternal resuscitation immediately after the delivery of the fetus.

Use Full Doses of ACLS Drugs – Limiting vasopressors in an effort to avoid uterine artery constriction is misguided. Without restoration of maternal circulation the fetus will surely die.

This survey effectively raises awareness of the need for additional education of those most likely to be involved in resuscitation of pregnant women. It should really be a simple matter to add this crucial information to ACLS training for the appropriate clinicians. (We already have different “flavors” of ACLS for different provider types.) The recommendations and rationales are well supported by scientific evidence and clinical experience. Whether or not the American Heart Association makes adjustments in ACLS training, I hope all nurse anesthesia programs will make sure this information is included in their curricula.


Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Kodali B-S, Chandrasekhar S, Bulich LN, Topulos GP, Datta S


Airway changes during labor and delivery

Anesthesiology 2008;108:357-362

Kodali B-S, Chandrasekhar S, Bulich LN, Topulos GP, Datta S


Purpose            The purpose of this study was to look for airway changes between early and late in the labor and delivery process. The study was intended to be exploratory, the foundation for future more detailed investigations of the airway in laboring women.

Background            Failed intubation is much more common in pregnant than in nonpregnant individuals. The relative risk of a difficult intubation in pregnant women with a Mallampati class III airway has been shown to be 7.58 times greater than pregnant women with a class I airway. For those with a class IV airway the relative risk is 11.3 times greater than those with a class I airway. Airway problems during general anesthesia for cesarean section are the number one cause of anesthesia related maternal morbidity and mortality. At least one study has also identified airway obstruction or hypoventilation during emergence and extubation as a cause of maternal death.

Airway soft tissue changes and edema are associated with pregnancy and may contribute to difficult airway management. Some studies have shown that airway edema increases during pregnancy and increases the Mallampati score. Anecdotal reports suggest the airway of pregnant women may also change during the labor and delivery process.

The upper airway has two components, an oral component and a pharyngeal component. Only the oral component is assessed with the Mallampati score. Acoustic reflectometry can be used to assess both components of the airway. Acoustic reflectometry has been shown to correlate well with computed tomography measurements of the upper airway. An airway volume less than about 40 mL has been associated with difficulty viewing the glottis in nonpregnant subjects. And acoustic reflectometry has been shown to predict difficult mask ventilation.

Methodology            This prospective, single-blind study included healthy women who underwent labor and spontaneous vaginal delivery. Women who ultimately had an operative delivery were excluded. Two different methods of assessing the airway of laboring women were employed. First, using the Mallampati scoring system and second, by using acoustic reflectometry.

For the airway classification section of the study, healthy women admitted in early active labor (2-3 cm cervical dilation) were recruited. Women with a class IV airway were excluded. The oral airway of women was photographed in a standard manner at admission, 20 minutes after delivery of the placenta, and 36-48 hours postpartum. Photographs were taken with women in the sitting position, with their mouths wide open, and without phonation. Each photograph was taken from the same angle and from a distance of 10 inches from the uvula or soft palate. Photographs were assigned a numeric code. Each photograph was later assigned a Mallampati classification by a single senior anesthesiologist.

In the second section of the study acoustic reflectometry was used to describe the length and cross sectional area of the oral and pharyngeal portions of the upper airway. Healthy women admitted in early active labor (2-3 cm cervical dilation) were recruited. Airway measurements were made while in early active labor and again 20 minutes after spontaneous vaginal delivery.

Statistical analysis simply used descriptive statistics and Spearman correlation analysis of airway assessments with pregnant weight, patient height, duration of labor, and the volume of intravenous fluids administered during labor and delivery.

Result            The Mallampati classification section of the study included 61 women. (Seventy women were recruited but nine delivered by cesarean section and were not included in the analysis.) Overall Mallampati class changed significantly between early active labor and post-delivery assessments (P<0.001). In 23 of the 61 women (38%) the Mallampati class increased from early labor to post-delivery. (In no cases did the Mallampati class decrease across these two measurements.) In 20 women (33%) the Mallampati class increased by one grade. In 3 women (5%) the Mallampati class increased by two grades. There were 30 women with Mallampati class III or IV at the post-delivery measurement but only 17 in the early labor measurement (an increase of 76%)(P<0.001). Immediately post-delivery, half of the women in the study had class III or IV airways. While no women with a Mallampati class IV airway during early labor were included in the study, 8 women had a class IV airway after delivery. Most women whose Mallampati class increased during labor reverted to their early labor airway classification by 48 hours postpartum. There was no correlation between airway changes and weight, height, duration of labor, or IV fluid administration.

The acoustic spectrometry section of the study included 21 women. (Twenty-eight women were recruited but seven delivered by cesarean section and were not included in the analysis.) Twenty of the 21 women received labor epidural analgesia. Both oral (P<0.05) and pharyngeal (P<0.001) volume decreased significantly between early active labor and post-delivery. There was no correlation between the decreases in airway volumes and weight, height, duration of labor, or IV fluid administration.

Conclusion            There was a significant increase in the Mallampati score and a decrease in total upper airway volume between early labor and immediate post vaginal delivery. An airway exam should be performed immediately prior to endotracheal intubation as it may have changed since the start of labor and delivery.



I have always been pretty quick to perform a pre-anesthetic assessment in labor and delivery if I had any ideas I might be called upon to provide anesthesia for a cesarean section later on. I was especially interested in looking at their airway so if I had to do a stat general cesarean section I’d “know” what to expect during intubation. I never dreamed that the airway could change significantly between that early, “just in case,” assessment and when I actually induced general anesthesia. This article convinces me that it can.

Obstetric anesthetists have long regarded regional anesthesia as safer than general anesthesia for cesarean section. This article gives us even more reason to hold to that belief. Here is the way the authors of this article have laid out the increased risk. 1) Airway problems during cesarean section and general anesthesia are the number one cause of maternal morbidity and mortality. 2) In this study, 50% of women had a Mallampati class III or IV airway after delivery. 3) The risk of a difficult intubation in those women (half the pregnant women in the study) was between 7 and 11 times greater than a pregnant woman with a class I airway.

We don’t know from this study precisely when the airway gets worse during labor and delivery. It may occur only very late in the process, after the woman has pushed for an hour or more. We also don’t know if the airway changes this study found are applicable to pregnant women in general or, for some reason, were peculiar to the women in this one study. But this is powerful information and, at least until further studies give us an idea how widespread the decline in airway class is during labor and delivery, it should give us pause. And, it should give us a reason to reexamine the airway immediately before the induction of general anesthesia in a pregnant or recently postpartum woman.


Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008


Ryu J-H, Kang M-H, Park K-S, Do S-H


Effects of magnesium sulphate on intraoperative anaestehtic requirements and postoperative analgesia in gynaecology patients receiving total intravenous anaesthesia

Br J Anaesth 2008:100:397-403

Ryu J-H, Kang M-H, Park K-S, Do S-H


Purpose            The purpose of this study was to determine the effects of Magnesium on the amount of propofol required for general anesthesia and the use of intravenous patient controlled analgesia (IVPCA) postoperatively.

Background            Pain is associated with delayed recovery. N-methyl-D-aspartate (NMDA) receptor blockers contribute to reducing post-surgical pain through prevention of central sensitization. Magnesium (Mg) is a chief intracellular cation involved in numerous ion channel and enzymatic processes. It is also an NMDA receptor blocker. Some studies have shown that perioperative Mg administration reduces anesthetic requirements and reduces postoperative pain. For example, Mg has been shown to reduce requirements for remifentanil, mivacurium, vecuronium, or propofol requirements. Other studies have failed to demonstrate a significant reduction in anesthetic requirement and/or postoperative pain. Mg, in excess, can result in vasodilation (and thus hypotension), skeletal muscle weakness similar to that produced by nondepolarizing neuromuscular blocking drugs, and, at very high concentrations, cardiac arrest. Mg potentiates neuromuscular blockade.

Methodology            This prospective, double-blind study included ASA I and II patients aged 30 years to 65 years scheduled for total abdominal hysterectomy. Patients with a history of opioid or analgesic abuse, Mg administration, or those taking calcium channel blockers were excluded.

The Magnesium group received 50 mg/kg magnesium sulphate (MgSO4) IV immediately before induction of general anesthesia. They also received a continuous infusion of 15 mg/kg/hour MgSO4 during surgery. The control group received equal volumes of normal saline during the same periods. General anesthesia was induced and maintained with propofol, remifentanil, and rocuronium. Nitrous oxide was not used. Neuromuscular block was monitored at the wrist with a peripheral nerve stimulator. Propofol effect site concentrations were regulated to maintain a Bispectral Index of 40 to 50 during the case. Additional doses of 0.15 mg/kg rocuronium were given only when the train-of-four was 2 or more. At the end of the procedure, residual neuromuscular block was antagonized with pyridostigmine 0.3 mg/kg and glycopyrrolate 0.01 mg/kg. Ramosetron 0.5 mg and dexamethasone 5 mg were given to prevent postoperative nausea and vomiting (Editors Note: ramosetron is a 5-HT3 receptor blocker). Postoperatively, patients received IVPCA with morphine and ketorolac (Toradol).

Result            At the end of surgery, serum Mg concentrations were about 15.4% higher than the upper limit of the normal range in Magnesium group patients and within the normal range in Control group patients. Mean arterial blood pressure (MAP) was lower in the Magnesium group than the Control group throughout most of the procedure. MAP was never lower than 60 torr.

The average cumulative doses of propofol and remifentanil used during the procedure were similar in the Magnesium and Control groups. Cumulative IVPCA use was lower in the Magnesium than the Control group at 24 hours (P=0.026) and 48 hours (P=0.005) postoperatively.

Patients in the Magnesium group received less rocuronium than those in the Control group, an average of 0.35 (sd 0.07) mg/kg/h compared to 0.45 (sd 0.09) mg/kg/h in the Control group (P<0.001). On average, the duration of surgery was 6 minutes longer in the Magnesium group. PONV was less common in Magnesium patients (40%) than Controls (76%) (P=0.01).

Conclusion            A preoperative loading dose of MgSO4 followed by an intraoperative infusion until the end of surgery did not reduce the use of propofol or remifentanil intraoperatively. It did reduce the use of rocuronium intraoperatively and of IVPCA during the first 48 hours postoperatively.



I was more impressed by this study before I really started to think about it.

There are two aspects of the study that concern me and they are both, unfortunately, altogether too common in biomedical studies. The first is the analysis of visual analogue data, used in this study to rate pain, as if the VAS was a linear scale. It is fine to have patients report their pain on a VAS, but patients don’t do so linearly. That is, a VAS score of 20 doesn’t represent twice as much pain as a VAS score of 10. And a VAS score of 40 isn’t half as much pain as a VAS of 80. The scores tend to bunch up near the top and bottom of the scale so that there isn’t much difference between the numbers there. There is a proper way to analyze such data to know if the differences between scores are significant but too often, as in this study, it is not done properly. When the differences are quite large it is likely that they were still real, though perhaps inflated. But when, as in this study, the differences are less than 10 mm near the bottom of a 100 mm scale I doubt the difference would be even statistically significant if properly analyzed. I’d not judge them to be clinically significant in either case.

The second concern is the use of the Bispectral Index (BIS) as a research tool. The BIS is arguably a marginally useful clinical monitor at a qualitative level. No work that I’m aware of has validated the BIS, or any other currently available “level of consciousness” monitor as a quantitative research tool. I am continually bewildered that the BIS is used as a research tool with little, if any, critical thought.

One finding that seems well founded (and was properly analyzed) was the reduction in IVPCA use over the first 48 hours postoperatively. The investigators reported this in a way that makes it hard for the reader to know what the actual differences were but it appears from a figure in the article that the magnesium group used about 33% less of the morphine / ketorolac IVPCA solution over that time. This is all the more interesting because the magnesium infusion was discontinued at the end of surgery (48 hours earlier) and excess magnesium is rapidly and completely eliminated by the kidneys. Alas, they did not measure the magnesium level at 48 hours to verify that levels had declined into the normal range. Nevertheless, it would appear that the NMDA blockade during surgery may well have been responsible for a persistent reduction in pain perception. This is all the more interesting since magnesium is a relatively weak NMDA receptor blocker, a point the investigators did not include in their background information.

The finding that PONV was considerably less frequent in the group the received magnesium is intriguing but we must be careful not to give it too much weight. This study was designed to look for differences in the areas of intraoperative anesthetic requirement and postoperative analgesia. It was not designed to look for differences in PONV. As such, there may well have been invisible reasons why the PONV rate was lower in the magnesium group. As an example, risk factors for PONV were not tracked (or at least, not reported). It could have been that the control group contained a high percentage of patients with multiple PONV risk factors while the magnesium group did not. There is always a danger of being mislead when a study draws or implies conclusions that are outside what it was designed to investigate. Such conclusions should be viewed with caution. If the administration of magnesium did produce a reduction PONV, I have no idea what the mechanism may have been.

I do see a danger in the administration of magnesium in an effort to reduce postoperative pain that the investigators didn’t discuss. As noted, magnesium in excess can produce significant weakness of skeletal muscle, much the same as nondepolarizing muscle relaxants. Magnesium also potentiates nondepolarizers. Before the availability of the shorter acting nondepolarizing relaxants we use now, morbidity and mortality from residual neuromuscular block were constant threats. Residual neuromuscular block is a much less frequent cause of anesthetic complications now. I understand that with proper monitoring of neuromuscular blockade there is no reason for magnesium administration to once again increase the incidence of morbidity and mortality from residual paralysis. But I have to wonder if the analgesic benefits of magnesium administration outweigh the neuromuscular risks, especially when there are alternative analgesic strategies available.


Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Pediatric Anesthesia

Lardner DRR, Cox RG, Ewen A, Dickinson D


Comparison of laryngeal mask airway (LMA) Proseal and the LMA-Classic in ventilated children receiving neuromuscular blockade

Can J Anesth 2008;55:29-35

Lardner DRR, Cox RG, Ewen A, Dickinson D



Purpose            The purpose of this study was to compare the #2 LMA Proseal and the LMA Classic in pediatric patients during positive pressure ventilation with neuromuscular blockade.

Background            The Laryngeal Mask Airway (LMA) is commonly used in pediatric patients. The LMA-Classic (LMA-C) is associated with gastric insufflation and leaks during positive pressure ventilation. In adults, the LMA-Proseal (LMA-P) is associated with a tighter seal allowing for positive pressure ventilation with less of a leak then the LMA-C. Children, however, have somewhat different airway anatomy and lung mechanics compared to adults. Furthermore, neuromuscular blockade may increase the likelihood of airway leaks during positive pressure ventilation with an LMA. Airway leaks have increased in some adults being ventilated through an LMA when neuromuscular blockade was added. A size #2 LMA-P is now available. It differs in design, compared to adult sizes of the LMA-P, in that it lacks the dorsal cuff extension present on the adult sizes. It is this extension that has been thought to improve the seal of the LMA-P and allow positive pressure ventilation with less of a leak than when using the LMA-C.

Methodology            This prospective, randomized, single-blind study included a convenience sample of ASA physical status I or II patients weighing between 10 kg and 20 kg. All patients were undergoing elective day surgery requiring general anesthesia. Exclusion criteria included: risk for aspiration, asthma or other causes of reduced respiratory compliance, upper respiratory infection, and known or suspected difficult airway. All LMAs were new and inserted according to manufacturers instructions. Anesthesia was induced with sevoflurane and nitrous oxide using a pediatric circle system with carbon dioxide absorber. After induction, an IV was started and 0.2 mg/kg to 0.25 mg/kg mivacurium was administered. Anesthesia was maintained with isoflurane 1.8% to 2.5 %. Nitrous oxide was discontinued after induction. Neuromuscular blockade was confirmed with a peripheral nerve stimulator. Next, an LMA-P or an LMA-C was inserted according to the randomization procedure. The LMA cuff was inflated to 60 cm H2O. Head and neck position were standardized and maintained throughout the study period. Patients were mechanically ventilated with a Datex ADU ventilator in pressure mode. The fresh gas flow was 150 mL/kg/min. and inspiratory pressure was ≤ 20 cm H2O.

The airway pressure at which the LMA began to leak was determined by auscultating the neck as airway pressure was rising. The pressure at which a leak was audible over the thyroid cartilage was recorded.

The volume of the leak during positive pressure ventilation was measured by comparing the inspiratory and expiratory tidal volumes during 20 consecutive positive pressure breaths delivered at 20 cm H2O. Capnometry was discontinued during this measurement to eliminate it as a source of gas loss from the circuit.

Result            Fifty-one patients were enrolled and randomized. One patient experienced inadequate ventilation and was withdrawn from the study. The time needed to insert the LMAs was almost identical in the LMA-C and LMA-P groups. The mean (standard error of the mean) airway pressure at which a leak began was 16.5 (1.0) cm H2O in the LMA-C group and 23.7 (1.9) cm H2O in the LMA-P group (P=0.009). The mean percent of tidal volume lost to leaks during 20 positive pressure breaths was 21.2% in the LMA-C group and 13.3% in the LMA-P group (P not significant). Gastric insufflation occurred in 46.1% of the LMA-C group and 8.0% of the LMA-P group during the study period (P=0.006). (At the end of the case, gastric distension was present in six LMA-C patients and four LMA-P patients.)

Conclusion            During positive pressure ventilation with neuromuscular blockade, the #2 LMA-Proseal was associated with a higher circuit pressure before a leak occurred and a lower incidence of gastric insufflation than the #2 LMA-Classic in 10 kg to 20 kg patients.



Using LMAs in pediatric patients is another one of those areas that demonstrates that pediatric patients are not just smaller versions of the adult patient.  I must admit that I prefer not to utilize an LMA Classic in patients in the lower weight range of the sample due to the challenge in consistently achieving appropriate position and ventilation.  However, the results of this nicely designed study are persuasive to try the LMA Proseal instead of the LMA Classic.  Based on these findings, it seems that the LMAP may have better construction to address those anatomic differences found in the younger pediatric patients.  Although I typically allow the patient to return to spontaneous ventilation after the LMA is placed, the ability to generate positive pressure ventilation with decreased risk of gastric insufflation, as shown with the LMAP, is attractive as a back up plan.  Even with a sample size of only 51 children, this evidence, if nothing else, encourages me to revisit the use of LMAs in more pediatric patients to explore the purported advantages of the LMAP over the LMAC.


Terri M. Cahoon, MSN, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Regional Anesthesia

Fischer R


epidural conduction device fractures and complications of retained fragments

AANA J 2008;76:37-40

Fischer R



Purpose            The purpose of this article was to review the issues concerning fragmentation of epidural catheters during or after patient placement.

Background            The Food and Drug Administration (FDA) is responsible for monitoring adverse events associated with medical devices. They do this by reviewing mandatory reports submitted by medical manufacturers, importers, and health care facilities. In addition, voluntary reports may be submitted by device distributors and individuals including patients, providers, and administrative personnel. These reports have identified incidences of epidural catheter fracture which had the potential to create serious patient complications associated with fragment migration or inflammatory response. This article reviews three cases in an attempt to educate providers to the potential problems associated with fragmented epidural catheters, how to deal with the problem if it does occur, and strategies to help prevent the problem from occurring.

Methodology            The following cases occurred between 2001 and 2007.

Case 1. Following an epidurolysis, the anesthesia provider could not remove the catheter. Under fluoroscopy there was no evidence of catheter kinks, knots, or other reason for the inability to remove the catheter. The provider, in consultation with a neurosurgeon decided to “yank” on the catheter which resulted in the catheter breaking. It was decided not to remove the fragment.

Case 2. During the advancement of an epidural catheter into the caudal space the provider met unexpected resistance. The practitioner pulled the catheter back with force at which time the catheter broke into two pieces. The fragment was not removed.

Case 3. During an epidural pain procedure using a catheter, the provider felt resistance during removal of the catheter through a Touhy needle. The tip of the catheter was missing when the needle and catheter was removed. There was radiographic evidence that 3 to 4 inches of catheter was left in the patient’s epidural space. The fragment was not removed.

Result            Contributing factors to catheter fragmentation include shearing of a catheter by withdrawing it back through the introducer needle, nicking the catheter on a flawed needle during insertion, advancing the introducer needle over an already placed catheter, and removing a catheter using excessive force. Insertion and withdrawal forces play a role in catheter fragmentation. It is suggested that a catheter be removed with the patient in the same position as it was inserted. Removing a catheter with the patient in a sitting position requires 2.5 times more force than when the patient is in the lateral position. Rarely, a catheter can loop, knot, or kink. This is the result of anatomical deflection and obstacles such as nerve roots, blood vessels, fascia, and boney processes. Excess catheter threading increases the risk of these rare events. Mechanical damage to the catheter by using a hemostat to remove it increases the risk of breakage by lowering the catheters tensile strength.

Generally a fragmented catheter is benign when inserted under sterile conditions, but concerns about having a foreign body left in the patient remain. Those concerns include the possibility of a fragment migrating into another area of the body and causing problems, interference with radiographic exams, and questions about the normally inert material remaining that way for extended periods of time.

Generally, fragmented catheters should be left in place. Surgical intervention to remove them has inherent risks. If the catheter is left in place, the provider has an ethical obligation to inform the patient of the retention related risks.

There are several strategies to avoid fragmentation if the provider is confronted with a situation where the catheter is difficult to remove. They include patient repositioning into a lateral decubitus position, injecting saline to determine if the catheter is kinked or knotted, avoid using mechanical devices such as hemostats, provide sedation to the patient or wait to allow a tense patient to relax, or the use of radiographic imaging in an attempt to determine what the problem may be. Surgical removal should be reserved for patients with neurological symptoms.

Conclusion            Prevention of catheter fragmentation can best be accomplished by using proper insertion and removal techniques. In the event that a catheter fragmentation does occur, it should be reported to the FDA so that sufficient information can be collected to determine the reasons for this happening, and help provide suggestions on how to avoid fragmentation.



I found this article somewhat informative, but lacking in evidence of the risks associated with retention of catheter fragments. It would have been helpful to have follow-up information on the three cases, and additional information on known complications associated with catheter fragment retention.

It does make it clear that fragmentation can be associated with poor technique. I continue to be amazed at providers who pull a misplaced catheter back through the Touhy needle rather than using the proper technique of removing the needle and catheter in tandem, and then removing the catheter from the needle. In those cases, the catheter should only be used for another attempt after it is inspected for integrity because any nick in the catheter can prevent proper function and increase the risk of fragmentation. Anesthesia is very much like flying an airplane. When you cut corners and take risks you might be able to get way with it most of the time, but when you least expect it something will occur that will ruin your day. Avoiding shortcuts and following established practice standards (and flying rules) will minimize the risks.

Like flying, even in the best of hands and under the best circumstances, complications can still occur. Sometimes these problems are beyond the control of the provider. Product flaws can lead to product failure and unknown patient issues can create unexpected problems and catheter fragmentation. Regardless of the minimal risk associated with leaving a fragmented catheter in a patient, doing so creates patient anxiety and raises questions about provider competency, but surgical removal of a fragment invites the possibility of surgical complications and escalating problems. The best suggestion is to always use proper technique and avoid situations that might produce a fragmented catheter.


Steven R. Wooden, MS, CRNA

Demiraran Y, Yucel I, Erdogmus B, Subcutaneous Effusion Resulting from an Epidural Catheter Fragment. Br J Anaesth 2006; 96: 508-9 

Fragneto RY, The broken epidural catheter: an anesthesiologist's dilemma. J Clin Anesth. 2007;19:243-244

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Esmaoglu A, Kin A, Mizrak A, Turk Y, Boyaci A


Addition of cisatracurium to lidocaine for intravenous regional anesthesia

J Clin Anesth 2006;18:194-197

Esmaoglu A, Kin A, Mizrak A, Turk Y, Boyaci A



Purpose            The purpose of the study was to determine if adding a small dose of the non-depolarizing muscle relaxant cisatracurium to lidocaine would improve the quality of a intravenous regional anesthetic block.  The improved quality of the intravenous regional block was operationally defined by demonstrating a quick onset, delayed sensory regression time, and decreased postoperative pain scores reported by patients.

Background            Intravenous regional anesthesia, also known as the Bier Block, has consistently demonstrated safety and efficacy as an anesthetic technique for those requiring hand surgeries.  As with any anesthetic technique, it is not without some shortcomings.  While the benefits include (but certainly are not limited to) a bloodless field which is optimal for the surgeon, concerns remain regarding local anesthetic toxicity, a delayed onset of action, a reported degree of poor muscle relaxation, and once the tourniquet is deflated, a rapid onset of pain.  The characteristics considered ideal in an intravenous regional anesthetic technique include a rapid onset, reduced or minimal intraoperative pain, and a long duration of analgesia following the procedure.  Previous attempts to improve perioperative analgesia and postoperative analgesia have not been totally successful, nor have attempts to add other drugs with hopes to shorten the onset time of both the sensory and motor block.

Methodology            This was a prospective, randomized, double-blinded study approved by the institutional review board.  Patients were scheduled to undergo either carpal tunnel release or tenolysis procedures.  The control group received 3 mg/kg of lidocaine, diluted with saline to a total volume of 40 mL and the experimental group received 0.01mg/kg of cisatracurium plus 3 mg/kg lidocaine diluted with saline to a total volume of 40 mL.  The operative extremity was exsanguinated in a standardized fashion for all subjects and a double tourniquet was positioned on the upper operative arm also in standardized fashion for all subjects.  Sensory blocks were assessed at 30 second intervals after administration of the solutions by an observer who was blinded as to which group the subjects were in.  Sites that were used for sensory testing (done with a 25 guage short beveled needle) included the thenar eminence (innervated by the median nerve), hypothenar eminence (innervated by the ulnar nerve), and the first web space (innervated by the radial nerve).  Sensory regression was assessed at these same sites after tourniquet deflation at 30 second intervals.  Motor blockade was documented when each patient could not induce any movement of their fingers after solution administration.  If the subjects experienced intraoperative breakthrough pain, fentanyl was used for an analgesic.  Upon completion of the surgery, quality of anesthesia was assessed using a numeric scale:  4 = excellent (no complaints of pain voiced by the patient), 3 = good (minor complaints, no fentanyl required), 2 = moderate (complaints from the patients; fentanyl required), and 1 = unsuccessful (this was a failed block and general anesthesia was required).  In an appropriate manner upon culmination of the surgery, tourniquet deflation was performed and sensory and motor block regression times were noted.  A blinded observer assessed the patient’s pain just after deflation, 15 minutes, 30 minutes, one and two hours after, using the visual analog scale.  Both intra-operative and post-operative fentanyl requirements were documented. 

Result            There were no significant differences in the demographic make up of the two groups, including duration of surgery and tourniquet time.  Sensory block onset time averaged 2.47 (±1.2) min. in the cisatracurium group and 4.27 (±2) min. in the control group (P<0.05). Statistically significant results were also seen in motor block onset time - shorter in the cisatracurium group, motor block regression time - longer in the cisatracurium group, and quality of anesthesia - better in the cisatracurium group. Intraoperative, postoperative, and total fentanyl requirements were significantly greater in the control group.

Conclusion            This study demonstrated that adding a small dose of cisatracurium (less than 1 mg in patients under 100 Kg) to lidocaine for an intravenous regional block shortened the sensory and motor onset time, improved the quality of the anesthetic, and controlled post-operative pain to an acceptable degree for the patients.  The hypothesis that the researchers gave regarding how cisatracurium achieved an analgesic property postoperatively is worth addressing and is as follows. Cisatracurium besilate undergoes pH and temperature-dependent chemical (Hoffman) degradation.  Hoffman degradation therefore is inhibited in the ischemic limb (induced by the use of the tourniquet) because of the metabolic acidosis.  The post-operative analgesia may be explained by the late regression of the block (ischemia blocks nerve conduction and motor end plate function) after tourniquet deflation when cisatracurium was used.



The thought of adding a non-depolarizing neuromuscular blocking agent to a local anesthetic for intravenous regional anesthesia, especially understanding what occurs with tourniquet deflation, appears scary at best.  However, the authors of this research, after citing studies whereby the addition of other non-depolarizing neuromuscular blocking agents caused serious side effects, did postulate that the manner in which cisatracurium undergoes degradation may be the reason for the absence of noted side effects.  I found it interesting that previous studies cited that a neuromuscular blocking agent offered greater post-operative pain relief over narcotics when the narcotics were added to the local anesthetic.  Understanding the mechanism of action of opioids and what appeared to be a lack of analgesic efficacy post tourniquet deflation, remains perplexing.  Further studies are indeed warranted to understand this phenomenon.


Mary A. Golinski, PhD, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008

Eltaki K, Abdulla H, Sinatra RS


A severe inflammatory cutaneous reaction after continuous epidural analgesia

Anesth Analg 2008;106:517-519

Eltaki K, Abdulla H, Sinatra RS



Purpose            The purpose of this case report was to describe a rare, delayed, cutaneous inflammatory reaction associated with the path of an epidural catheter on the back of a post-surgical patient.

Background            Epidural catheters are commonly used for anesthesia and for postoperative analgesia. Skin reactions to epidural catheters, antiseptic solutions, and adhesives are rare but known risks associated with epidural catheter placement. They may occur despite the use of hypoallergenic materials. Perifix epidural catheters are made of polyamide nylon, a material considered to be inert. A previous case report from the 1990s described six inflammatory cutaneous reactions associated with epidural catheter placement over the span of a few weeks. Pressure sores along the path of epidural catheters have also been reported.

Methodology            A 59 year old man was admitted for an abdominal perineal resection. He had no allergies to drugs. Prior to the procedure a 19 gauge nylon epidural catheter (Perifix™; B Braun Medical Inc., Bethlehem, PA; product code CESK REF 555963) was placed for postoperative pain management. The site was prepped with 10% providone-iodine solution and the excess dried off prior to epidural insertion. After catheter placement, residual providone-iodine solution was washed off with saline solution. The catheter was secured with a Tegaderm dressing.

Postoperatively the epidural site was checked regularly and no signs of tenderness, erythema, or infection were detected. The epidural was removed approximately 72 hours after insertion. At that time no erythema or tenderness was detected.

Result            Twenty-four hours after removing the epidural catheter (approximately 4 days after insertion) a 5 mm wide, nonulcerated, red, slightly raised rash was evident on the patient’s back along the path where the epidural catheter had been. Interestingly, the path of the epidural catheter outside the area which had been prepped with providone-iodine solution was unaffected. The skin surrounding the lesion, inside and outside the providone-iodine prepped area, was also unaffected. The rash was slightly tender to touch but otherwise unremarkable. Topical hydrocortisone cream was ordered. The plastic surgery service was consulted and believed the lesion to be a pressure sore. Topical bacitracin was ordered. A day later the redness had spread to encompass almost the entire area under the bacitracin and xeroform dressing and the initial area of redness underlying the former position of the epidural catheter had grown larger and intensified. The dermatology service was consulted and believed the lesion to be a contact dermatitis. Bacitracin, they thought, had made the initial lesion worse. Bacitracin was discontinued and topical hydrocortisone was ordered. The lesion cleared over the subsequent three weeks. The epidural catheter manufacturer was contacted and had received no other reports of similar complications associated with the use of Perifix catheters.

Conclusion            The cause of the contact dermatitis underlying the epidural catheter within the area prepped with providone-iodine solution is uncertain. Nevertheless, due in part to superior prevention of catheter-related infections, the authors reported changing their prep solution to chlorhexidine.



We will never know for sure what caused the rash underlying the path the epidural catheter had taken before it was removed a day earlier. I have a hunch the cause of the rash was not as simple as the combination of the epidural catheter and some residual betadine under the catheter path. My hunch is based in part on the fact that there is good documentation that the rash didn’t appear until about a day after the catheter was removed. That said, this case report reminds us of two other complications to watch for in patients who have had an epidural catheter; dermatitis and pressure sores. It also illustrates the importance of open lines of communication to anesthesia after we are done caring for a patient.

I must take a moment to complement the authors for avoiding the introduction of bias into their report. The anesthesia group appeared to have a bias that providone-iodine was likely to cause skin irritations. (To be clear, I am not being critical, simply making an observation.) I say this because they reported routinely washing the skin clean of providone-iodine after placing epidural catheters in an effort to prevent skin reactions. This single case also apparently prompted them to change to an alternate skin disinfectant. Despite this possible bias against providone-iodine  they were careful not to blame the contact dermatitis on providone-iodine, saying, “Although the offending agent is still unclear, …”   If we could all be this careful to avoid biased thinking we would, I believe, make better clinical decisions.


Michael Fiedler, PhD, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 2, March 31, 2008