ISSN NUMBER: 1938-7172
Issue 3.6

Michael A. Fiedler, PhD, CRNA

Contributing Editors:
Mary A. Golinski, PhD, CRNA
Gerard T. Hogan, Jr., DNSc., CRNA
Alfred E. Lupien, PhD, CRNA
Dennis Spence, PhD, CRNA
Steven R. Wooden, MS, CRNA

Guest Editor:
Penelope S. Benedik PhD, CRNA, RRT

Assistant Editor
Jessica Floyd, BS

A Publication of Lifelong Learning, LLC © Copyright 2009

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










This issue of Anesthesia Abstracts includes two abstracts and comments on Obstructive Sleep Apnea (OSA) at the request of a subscriber; one by Penelope S. Benedik, PhD, CRNA, RRT (respiratory physiologist) and another by Mary A. Golinski, PhD, CRNA.


Anesthesia Abstracts encourages your topic suggestions. While not every topic appears in the current literature, our contributing editors will certainly look for the topics you request. My thanks this month to Johnny J. Sacco, CRNA, who requested information about OSA and inpatient vs. outpatient treatment guidelines.


Michael A. Fiedler










Surgenor SD, Kramer RS, Olmstead EM, Ross CS, Sellke FW, Likosky DS, Marrin CAS, Helm RE, Leavitt BJ, Morton JR, Charlesworth DC, Clough RA, Hernandez F, Frumiento C, Benak A, DioData C


The association of perioperative red blood cell transfusions and decreased long-term survival after cardiac surgery

Anesth Analg 2009;108:1741-1746

Surgenor SD, Kramer RS, Olmstead EM, Ross CS, Sellke FW, Likosky DS, Marrin CAS, Helm RE, Leavitt BJ, Morton JR, Charlesworth DC, Clough RA, Hernandez F, Frumiento C, Benak A, DioData C



Purpose            The purpose of this study was to determine if there was an association between exposure to small amounts of allogenic Red Blood Cells and increased mortality in patients undergoing cardiac surgical procedures.

Background            There is growing evidence that exposure to even small amounts of blood products during surgery can affect morbidity and/or mortality. Past studies did not discriminate between those receiving large transfusions (3 or more units of RBCs) and those receiving small transfusions (2 units or less of RBCs). It is well documented that massive transfusion is a cause for peri- and post-operative mortality, but no study here to date has examined whether or not small transfusions play a role in postoperative mortality for cardiac surgery patients.  Past studies have implicated transfusion as a cause of reduced long-term survival in open heart surgery patients.

Methodology            This prospective, observational study included 15,512 patients having either Coronary Artery Bypass Grafting (CABG), Valve Replacement, or a combination of both. The study was completed by a consortium of eight medical centers in New England from 2001 to 2004.  Excluded from the study were those with prior open heart procedures, patients who received more than two units of RBCs, and those who were taken back to the operating room postoperatively for reexploration.  After exclusion, the data set included 9,079 patient procedures performed by 37 surgeons. Of these, 964 were off pump procedures.  Data were collected on patient demographics, type of procedure, and RBC utilization.  The outcome analysis measure was all-cause mortality over a 5 year period.  Kaplan-Meier survival curves were plotted according to RBC exposure.  The log-rank test and Cox Proportional Hazards modeling were used for univariate and multivariate tests of statistical significance.  There was also a propensity-adjusted analysis of the data.

Result            The number of subjects receiving one or two units of RBCs was 3,254 (35.8%), leaving 5,825 (64.2%) who did not receive blood products during their hospitalization. Of note, 63% of women received RBCs compared to only 27% of men in this study. Significant differences existed between the two groups.  The RBC group had a larger number of comorbidities, lower ejection fractions, and higher intraaortic balloon pump use. Those receiving RBCs were more likely to have undergone CABG/Valve procedure. Those having off pump procedures rarely required RBC transfusion. The overall crude hazard ratio for the RBC group vs. the no blood group was 1.94 (95% CI = 1.71-2.20; P< 0.001).

Conclusion            During the early postoperative period (surgery to 6 months), exposure to 1-2 units of RBCs led to a 67% increased hazard of death.  During the late phase (6 months to 5 years) there was no statistically significant difference in survival.



This was a very interesting finding. It is important for a number of reasons. We often do not consider transfusion of one or two units of RBCs significant. Studies are now proving that assumption to be false. Those who consider transfusion as unimportant must begin to question their beliefs. There are centers that routinely transfuse. The authors believe that if we treated every patient as we do those with religious objections to blood products, overall survival statistics would improve.

Patients in the “blood” group had a larger number of comorbid conditions than the “no blood” group.  Even with a risk adjusted methodology, it is difficult for me to imagine that conditions such as lower ejection fraction, longer pump run, diabetes, and peripheral vascular disease in and of themselves did not lead to a higher mortality rate.  Making the assertion that blood transfusion alone was the greatest determinant of short term survival is a bold assertion.

 The limitations included not knowing the actual cause of death, whether or not the blood was leukoreduced or not, and (most importantly in my opinion) did not consider whether other products (platelets, fresh frozen plasma, or cryoprecipitate) was used. There is a lesson for the CRNA here – transfusions can have deleterious outcomes.


Gerard T. Hogan, DNSc., CRNA




© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009


Correll D, Hepner D, Chang C, Tsen L, Hevelone N, Bader A

Preoperative electrocardiograms

Anesthesiology 2009;110:1217-22

Correll D, Hepner D, Chang C, Tsen L, Hevelone N, Bader A


Purpose            The hypothesis of this study was that significant abnormalities on preoperative electrocardiograms do not exist in the absence of specific risk factors. The study also evaluated age as an independent risk factor for significant electrocardiogram abnormalities.

Background            Electrocardiograms (EKG) for preoperative screening are a poor tool in determining occult coronary artery disease or postoperative outcome. Many abnormalities seen in these preoperative tests are not clinically significant but do have a cost and often delay needed surgery. The American Heart Association and the American College of Cardiology do not consider preoperative electrocardiograms on asymptomatic patients to be of any value regardless of patient age. They feel the risk of additional costs and workup is not worth the benefit. Only those patients with a history of coronary artery disease or at high risk for coronary artery disease by history may benefit.

Methodology            At the medical centers where the study took place, all patients over the age of 50 were given an EKG. Any abnormalities were considered significant if they were suggestive of ischemia or infarct as interpreted by a review committee. In addition, major Q wave changes, major ST depression or elevation, major T wave changes, Type II or Type III blocks, left bundle branch blocks, or atrial fibrillation were all considered significant. Any patients without significant finding were pooled and random samples of those patients were selected as the control group. The two groups were then compared for postoperative complications after non-cardiac surgery as evidenced by chart review, cardiac symptoms, cardiac enzymes, or the appearance of a new significant rhythm.

Result            Of the 1,149 EKGs evaluated, 540 (47%) had abnormalities. Significant abnormalities were found in 89 (7.7%) of the EKGs. When reviewing the risk factors associated with significant abnormalities, the factors in order of influence were high cholesterol, age over 65, history of severe valvular disease, myocardial infarction, angina, or congestive heart failure. Intervention in 25% of the patients with significant EKG abnormalities was needed prior to proceeding with the scheduled surgery. There was not a significant difference between the two groups in terms of postoperative cardiac complications.

Conclusion            This study revealed that age 65 years and older is an independent predictor of having a significantly abnormal EKG preoperatively. It also identified that a history of high cholesterol, myocardial infarction, angina, congestive heart failure, or severe valvular disease as being predictors of an abnormal EKG which might warrant further action prior to surgery. The Center for Medicare and Medicaid Services (CMS) no longer pays for preoperative EKGs based on age alone. This study determined that an age factor should remain a criteria for preoperative EKG evaluation.



As this country goes through another struggle to reform healthcare, it is inevitable that the cost benefit ratio of procedures will get a hard look. The amount of testing we do as providers has always been criticized as overdone. We are often accused of ordering numerous tests to protect ourselves from lawsuits or to increase facility revenue. I am not sure that this is particularly true, but I do think we have a tendency to order any test that might identify a problem no matter how unlikely it is because of course “there is always a chance” we might find a problem.

Insurance companies have stopped paying for tests that they deem “not medically necessary.” This study suggests that the decision by CMS not to pay for a preoperative EKG based solely on age may not be the right decision. As anesthesia providers it is important that we evaluate our testing practices and determine if they have a benefit and ask ourselves if the test can really have a positive impact on the care of the patient or are we doing it just because there is a protocol in place. Preoperative EKGs, chest x-rays, and some other test based on arbitrary conditions such as age may or may not be appropriate. The institution at which this study was conducted had a policy to do a preoperative EKG on every patient over the age of 50 even in the absence of other factors. I think this age criteria is pretty standard for both EKG and chest x-ray. In my experience, I have never seen a chest x-ray on an asymptomatic patient play a role in changing the surgical decision. On the other hand I have seen many EKGs delay surgery for what turns out to be no good reason. Yes, a few have helped change the cardiovascular care of the patient in question and perhaps prevented anesthesia provider distress during surgery, but I doubt the enormous overall cost is worth doing these test based solely on age.

I welcome studies like this that question the “sacred cow” and provide some real evidenced based rational for preoperative testing. I am curious if the institution in this study changed its policy on EKG testing from age 50 to age 65. One of the hardest things to do in an institution is to change a policy that has been in the books “forever.”


Steven R Wooden, MS CRNA

© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009

Gali B, Whalen F, Gay P, Olson E, Schroeder D, Plevak D, Morgenthaler T

Management plan to reduce risks in perioperative care of patients with presumed obstructive sleep apnea syndrome

J Clin Sleep Med 2007;3:582-588

Gali B, Whalen F, Gay P, Olson E, Schroeder D, Plevak D, Morgenthaler T


Purpose            The purpose of this study was to use a preoperative measurement tool aimed at identifying patients at risk of perioperative cardiorespiratory events, the events typically observed in those with diagnosed (and undiagnosed) obstructive sleep apnea syndrome. Information gained prior to surgery was used to closely assess patients while recovering in the post-anesthesia care unit (PACU), specifically for critical cardiorespiratory events. It was hypothesized that those with elevated risk factors noted prior to surgery and anesthesia were at greater risk for critical incidences of oxygen desaturation while in the PACU.

Background            Obstructive Sleep Apnea (OSA) is a common disorder but is often undiagnosed in patients presenting for surgery and anesthesia. Patients with OSA have an associated increase in perioperative morbidity and mortality and a higher incidence of prolonged length of postoperative hospital stay.  Several drugs routinely administered as part of an anesthetic (most significantly analgesics) decrease pharyngeal tone and depress ventilatory responses to hypoxia and hypercarbia.  This may exacerbate physiologic problems during recovery for OSA patients.

There are typical risk factors associated with OSA that should lead us to presume OSA is present.  These include a high BMI, increased neck circumference, snoring, craniofacial abnormalities affecting the airway, persistent daytime sleepiness, and other abnormalities on the airway examination.  The added risk of coexistent diseases has not been validated in the clinical setting; the sensitivity and specificity of these coexistent diseases is unknown (See notes).

American Academy of Sleep Medicine guidelines recommend monitoring OSA patients for the first 24 hours post-anesthetic and make mention that patient controlled analgesia may result in excessive respiratory depression. Identifying high risk patients prior to surgery and anesthesia, as well as determining how to best monitor these for post-operative events, are major issues.  Several prediction formulas can help to identify the high risk individual; they have good sensitivity, but often low specificity.  The Sleep Apnea Clinical Score (SACS-See notes) has a strong likelihood ratio and a rigorous probability score for OSA.  Since patients in the recovery room are often in pain and treated for their discomfort, patients with OSA (whether diagnosed or not), are at high risk for cardio-respiratory events.  Such events can clearly lead to poor outcomes.

Methodology            This was conducted as a clinical practice improvement initiative with the aim to identify patients at risk for OSA.  In the preadmission testing process, patients were assessed using Flemons Criteria (see notes), and a SACS was generated.  Patients with SACS scores > 15 were considered high-risk.  This initiative took place over a one year period.  Patients who consented to chart review for research purposes were eligible for participation; only those without a previous diagnosis of OSA were included.  On the day of surgery, the anesthesia provider was aware of the individual risk stratifications based upon the SACS.  The planned anesthetic regime was not altered for the purpose of the study.  After surgery, each enrollee was assessed in the PACU for respiratory events, which included:

§  Bradypnea

§  Apnea > 9 seconds

§  Desaturation <90% while on oxygen at 4 L nasal cannula

§  Inability to wean to a nasal cannula

§  Pain-sedation mismatch (high pain score with a high sedation score)

The recovery room nurses assessed the patients at 3 time points in their stay:  30, 60, and 90 minutes after extubation or PACU admission.  The clinical evaluation was continuous and the nurses recorded events as they occurred.  If a patient had repeated events at 2 or 3 time points, they were mandated to a monitored bed with pulse oximetry.  If patients had high SACS but no events in the PACU, no repeated events, or events that resolved before discharge, they were sent to a regular bed, however pulse oximetry was used.  Low risk-patients had routine post-operative care. A comparison group (the control group) was established consisting of 30 consecutive patients who were consider low-risk using the same scoring system.  Any PACU events in the control group were recorded and their data was compared with the high-risk group of patients.  Study patients were classified into 1 of 3 groups:  Group 1 patients were low risk for OSA, Group 2 was high risk but with no recurrent PACU events, and Group 3 were high-risk and had recurrent PACU events.

Result            Preoperative screening resulted in 1,923 patients with low SACS and 251 with a high SACS. Demographic data did not show statistically significant differences between any of the groups.

Unplanned ICU admissions were greater for those with high SACS (p<0.001) and therefore SACS was useful in identifying patients at high-risk of unplanned ICU admission. None of those with low SACS had recurrent PACU events.  Complete data was collected on 115 of 195 high risk patients (SACS > 15 without a diagnosis of OSA). The patients with high SACS were divided into groups: those without recurrent PACU events (Group 2) and those with recurrent events (Group 3).  The data showed that compared with low-risk patients, the patients at high-risk for OSA had higher BMIs (p<0.001), however there was no significant difference in BMIs between the high SACS without recurrent events and those with recurrent events.  The same was true for neck circumference in terms of significant differences noted between low-risk and high-risk groups, but not between Groups 2 and 3.  Also interesting to note, a significantly larger number of patients in the high-risk Group 2 received regional analgesia for post-operative pain management compared to high-risk Group 3.

The Oxygen Desaturation Index (ODI, see notes) differed significantly between the three groups. Group 1 had a 12% incidence of ODI >10, Group 2 had a 37% incidence of ODI >10, and Group 3 had a 57% incidence of ODI >10.  The greater incidence of the ODI >10 was statistically significant in Groups 2 and 3 compared with Group 1.

Conclusion            In this study, patients was a SACS score > 15 and with recurrent immediate postoperative respiratory events had the highest incidence of oxygen desaturation after surgery and anesthesia.  Those with an increased risk but no PACU events had the lowest incidence of desaturation following surgery and anesthesia.  These individuals were unlikely to require unplanned ICU admissions.



Great concerns remain regarding the ability to identify those with OSA.  While the more obvious clinical presentations may be known and evident at the time of a preoperative assessment, it is the ‘ordinary’ patient who presents to us who has no known risk factors, or fails to disclose all health information to us, that causes the concern.  For example, the patient who has a history of snoring (do we ask all?) but is thin and/or has a ‘normal’ BMI.   Or the individual who is hypertensive but not receiving clinical care and is undiagnosed.  Consider these individuals who are to receive a general anesthetic for a high-pain inducing high-risk procedure (adult tonsillectomy and adenoidectomy, or endoscopic sinus procedures with nasal packing!) where regional anesthesia isn’t feasible or local infiltration isn’t adequate, and there is a high probability of a poor outcome.  Typically we would discharge these individuals home (no known risks) but if we were better informed of their risk, their complete history, and/ or their potential for desaturation, we clearly could prevent untoward outcomes by prolonging their post anesthesia monitoring period.  The pre-operative assessment therefore is crucial.  Ask all patients about snoring.  Think twice about methods of post-operative pain management.  And consider consultations when suspicions exist.  We cannot let ourselves be fooled into thinking only those with the highest BMIs have OSA.  We must use our assessments to investigate further, probe further regarding abnormal findings, and be meticulous in identifying the appropriate analgesic methodology and length of stay post-surgical procedure.

Mary Golinski PhD, CRNA


Sensitivity is the proportion of true positives that are correctly identified by a diagnostic test. Specificity is the proportion of true negatives that are correctly identified by a diagnostic test. Sensitivity and specificity are one approach to quantifying the diagnostic ability of the test.  In real clinical practice, the ‘test result’ is often all that is known, so it behooves us to want to know how good a test is at predicting an abnormality.  What proportion of patients with abnormal test results, are truly abnormal?


Flemons developed a clinical prediction formula which requires a neck circumference measurement, a determination of a history of hypertension and other reported clinical symptoms.  There is a likelihood ratio that is generated.  Using the Flemons formula, a sleep apnea clinical score of 15 or greater has a likelihood ratio of 5.17 and a posttest probability of 81% that a patient has OSA.


Oxygen Desaturation Index (ODI) is defined as the number of desaturations recorded per hour. A desaturation was defined as a decrease in saturation of ≥ 4% for 10 seconds or more.  This ODI was calculated for each patient for 24 to 48 hours after PACU discharge.  An ODI >10 was indicative of a high frequency of oxygen desaturations.

This topic was suggested by Johnny J. Sacco, CRNA. 

© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009

Baumann, Studnicska D, Audibert G, Bondar A, Fuhrer Y, Carteaux J-P, Mertes PM


Refractory anaphylactic cardiac arrest after succinylcholine administration

Anesth Analg 2009;109:137-140

Baumann, Studnicska D, Audibert G, Bondar A, Fuhrer Y, Carteaux J-P, Mertes PM



Purpose            This report described two cases of anaphylaxis and cardiac arrest due to succinylcholine allergy.

Background            Allergic reactions during anesthesia are estimated to occur between 1 in 10,000 and 1 in 20,000 anesthetics. Up to 70% of allergic responses to anesthetic drugs have been associated with neuromuscular blockers, most often succinylcholine. Mortality has been reported to be as high as 10%. Anaphylaxis during anesthesia can be refractory to epinephrine and vasopressors despite prompt treatment.

Tryptase is a marker of mast cell activation and levels above 20 µg/mL are highly sensitive for the detection of anaphylaxis. IgE for the quaternary ammonium component of succinylcholine is both sensitive and specific.

Methodology            Case 1 involved a 74 year old, 79 kg woman scheduled for emergency appendectomy. Her history included hypertension controlled with diuretics. She had undergone two previous surgeries without complication. She had no history of allergies or atopy. General anesthesia was induced with propofol 2 mg/kg and succinylcholine 1 mg/kg. Endotracheal tube placement was verified. Four minutes later she developed bronchospasm, high airway pressure, ETCO2 decreased to 15 torr, oxygen saturation was not measurable, systolic BP was 40 torr, heart rate 40 bpm, and there was no carotid pulse. CPR was begun immediately. No external signs of anaphylaxis were visible, but an allergic reaction was suspected.

A total of 25 mg IV epinephrine was administered in escalating doses of 1, 2, and 5 mg at three minute intervals. Lactated Ringers 1,500 mL was administered over the 90 minute resuscitation. Multiple defibrillations were performed when ventricular fibrillation was present. Calcium chloride 2 gm, 250 mL 8.4% sodium bicarbonate, and two 1 mg doses of terlipressin (Editor’s note: this is a vasopressin analogue available in Europe) were also administered during the resuscitation.

A tryptase level and IgE specific for succinylcholine were drawn 70 minutes into the resuscitation. Both were grossly elevated confirming the presence of anaphylaxis due to succinylcholine.

Case 2 involved a 49 year old, 98 kg man scheduled for emergency appendectomy. His history included hypertension treated with diuretics and an alpha-2 blocker. He had no surgical history and no allergies. General anesthesia was induced with propofol 2 mg/kg and succinylcholine 1 mg/kg. Endotracheal tube placement was verified. A low ETCO2, erythema, and piloerection were seen instantly. Anaphylaxis was immediately suspected. His BP quickly decreased to 50/35 torr and oxygen saturation was unmeasurable. Pulseless electrical activity ensued and CPR was begun immediately. A total of 20 mg IV epinephrine was administered in escalating doses of 1, 2, and 5 mg. Normal saline 2,000 mL was administered during the resuscitation. Multiple defibrillations were performed. Calcium chloride 2 gm, 500 mL 8.4% sodium bicarbonate, amiodarone 300 mg, and two 1 mg doses of terlipressin were also administered during the resuscitation. At 75 minutes the patient was placed on extracorporeal membrane oxygenation (ECMO) which produced a systolic BP of 60 torr.

A tryptase level and IgE specific for succinylcholine were drawn 50 minutes into the resuscitation. Both were grossly elevated confirming the presence of anaphylaxis due to succinylcholine.

Result            Both patients developed pulmonary edema and cardiac activity was not restored. In the second patient, ECMO resulted in inadequate BP. He died 12 hours later of multiple organ failure, acidosis, rhabdomyolysis, and coagulopathy.

Conclusion            Anaphylaxis to succinylcholine can result in cardiac arrest and death despite immediate recognition and appropriate treatment. Epinephrine and vasopressin resistant anaphylaxis to succinylcholine indicates the need for alternate treatments.



Anaphylaxis is a rare occurrence during anesthesia but it does occur. Most anesthetists I have talked with are unaware that succinylcholine is one of the drugs most commonly associated with anaphylaxis during anesthesia (albeit a common cause of an uncommon event). Succinylcholine is an important drug in anesthesia. It plays a pivotal role in good anesthesia care, especially in emergency situations, and I’m not suggesting that we do away with it. Nonetheless, the fact that succinylcholine is a chief cause of the admittedly rare incidences of anaphylaxis that occur during anesthesia is one of the several reasons I don’t use succinylcholine unless there is a specific reason to use it rather than a nondepolarizer. Avoiding succinylcholine unless is it specifically required marginally reduces patient risk.

Perhaps the most troublesome aspect of these cases is not that succinylcholine resulted in anaphylaxis, but that death occurred despite early and aggressive treatment. At the very least, this information should encourage us to act decisively to identify and treat the problem when we see signs consistent with anaphylaxis within minutes after induction of anesthesia. It should also prompt us to call for assistance early on. (Of course, problems other than anaphylaxis can cause hypotension, bronchospasm, reduced ETCO2, and oxygen desaturation after induction.) Lastly, we should be open to additional treatments that may improve a patient’s chance of survival when they do experience anaphylaxis during anesthesia.


Michael Fiedler, PhD, CRNA



© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009

Ezaru CS, Mangione MP, Oravitz TM, Ibinson JW, Bjerke RJ


Eliminating arterial injury during central venous catheterization using manometry

Anesth Analg 2009;109:130-134

Ezaru CS, Mangione MP, Oravitz TM, Ibinson JW, Bjerke RJ



Purpose          The purpose of this study was to describe the incidence of arterial puncture and arterial injury when manometry was used to verify catheter placement during 9,348 central line insertions.

Background   More than five million central lines are placed in the USA each year. Central line complications occur in between 3% and 25% of patients. Arterial puncture during central line placement has been reported to occur in 2% to 4.5% of central line placements with insertion of a large bore cannula, introducer, or dilator into an artery in 0.1% to 0.5% of cases. Arterial placement of a large bore device intended for a central vein results in arterial injury and a substantial increase in the risk of stroke or death.

During central line insertion, venous cannulation is typically differentiated from arterial cannulation by assessing the color and flow of blood through a catheter or needle 18 gauge or smaller. The color and flow of blood through a small gauge catheter or needle is an unreliable indicator. Ultrasound visualization of vascular structures during cannulation has been recommended to reduce the risk of introducing large diameter cannula into an artery but ultrasound is infrequently used even when available. A survey of cardiovascular anesthesiologists reported that 67% never or almost never used ultrasound. The respondents reported that it was unnecessary (46%), unavailable (18%), or took too long (16%). The infrequent use of ultrasound for identification of vascular structures during central line placement suggests the need for a more “user-friendly and efficient alternative.”

Manometry is a reliable, quick, and easy technique to verify venous placement during central line insertion. The only equipment needed that is not included in the central line kit is a 20 inch IV extension tube.

Methodology This retrospective, descriptive study included 9,348 central venous catheterizations with manometry over a 15 year period. During the central line procedure, the internal jugular, subclavian, or femoral vein was cannulated with an 18 gauge catheter. At that point, a 20 inch IV extension tube was attached to the catheter. The tubing was held vertically above the patient and blood aspirated with a syringe on the end of the extension tubing. The syringe was then removed and the blood column observed. A descending column of blood that moved with respiration or mechanical ventilation indicated placement within a vein. An ascending column of blood indicated arterial placement. A static column suggested catheter kinking or impingement upon a vessel wall, which required repositioning. A guidewire was inserted to facilitate placement of a larger catheter or introducer only after evidence of venous placement via manometry. The manometry procedure added one minute or less to central line placement.

Result Ninty-eight percent of central lines were placed in the internal jugular vein, while 1% each were placed in the subclavian or femoral veins. Similarly, 98% of the lines placed were 7 French or larger. Attending anesthesiologists placed 67% of the lines and trainees 33%.

Arterial puncture by an 18 gauge or smaller needle was tracked in a subset of 511 central line insertions. Arterial punctures occurred in 5% of central line placements (28 patients). No complications arose. In 4 of the 28 arterial placements arterial puncture was identified on the basis of manometry alone; not by blood flow or color.

Arterial injury was defined as the introduction of a catheter or dilator larger than 18 gauge into an artery. No arterial injuries occurred in any of the 9,348 central line insertions.

The reported incidence of arterial injury during central line insertion predicted between 9 and 47 arterial injuries in 9,348 central line insertions.

Conclusion    Manometry is a quick, simple, effective technique for identifying venous cannulation during central line placement. Using manometry can prevent arterial injury and subsequent complications from inadvertent arterial cannulation or dilation.



Ultrasound has been recommended for visualization of vascular structures during central line placement in an attempt to reduce complications. I would not argue that, properly used, ultrasound is ineffective for this purpose. But the incidence of complications during insertion (notwithstanding their potentially grave outcome) is relatively low, ultrasound equipment is expensive, and training is needed to use ultrasound properly. Thus, the question becomes, how does the reduction in risk when ultrasound is used compare to less expensive, easier ways of reducing the risk of complications during central line insertion?

Using the manometer method described in this study to verify venous placement before thrusting an up to 10 French cannula or dilator into the neck is certainly cheaper and easier than using ultrasound. It is the method I was taught in training and it has served me well over the years. I can attest to the clarity of the manometry test in almost all circumstances.

One might wonder if, given the rare incidence of arterial injury during central line placement, the reported incidence of 0 arterial injuries out of 9,348 central lines is really any different than the reported nationwide overall incidence of between 0.1% and 0.5%. The answer is most likely, “yes.” The more conservative comparison is against a nationwide incidence of 0.1%. Fisher’s Exact test produces a P < 0.0002 comparing 0.1% against an incidence of 0 of 9,348. Certainly the strength of the evidence provided by a post hoc statistical test of a retrospective study compared to extrapolated nationwide overall data is limited. Nevertheless, the difference is highly statistically significant and the comparison data in the retrospective study was collected over a prolonged period of time.

Whether or not the manometry test is “the” best way to avoid arterial cannulation and whether or not ultrasound is used, manometry only takes a minute, it has virtually no risk, and it is insanely cheap. Why not use it? If your results are as good as those reported by the study authors you’ll be glad you did.

Michael Fiedler, PhD, CRNA



© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009


Kok-Yuen H, Jones L, Tong G


The effect of cultural background on the usage of complementary and alternative medicine for chronic pain management

Pain Physician 2009;12:685-688

Kok-Yuen H, Jones L, Tong G


Purpose            This study evaluated the racial background of patients who used complementary or alternative medicine (CAM). It was hypothesized that there was a difference between the attitudes of Caucasians and non-Caucasians related to the use of CAM in treating chronic pain.

Background            The cost of treating chronic pain in the United States exceeds $100 billion annually. There are many treatments not commonly used in conventional medicine or taught in medical school.

Methodology            The study group was adults who were diagnosed with a chronic pain condition. Race, age, income, educational level, type of CAM used, the reason for using CAM, and the effectiveness of both conventional and CAM interventions were all recorded. The researchers selected 92 patients for interview from the Duke University Pain Clinic.

Result            The most common pain condition seen in this study was back pain, and CAM was used by 81% of the patients interviewed. There was no difference between ethnic groups in the use of CAM. The use of CAM by Caucasians and African Americans was 80% while Hispanic and Native Americans were recorded at 100%. Reasons given for the use of CAM were, control over treatment (25%), conventional methods did not work (23%), believed that CAM was safer (12.3%), CAM was more effective than conventional treatment (12.3%).

Conventional treatment was effective for 56% of the patients while CAM was reported to be effective for 63% of the patients. Of those patients using opioids alone for pain management, 52.9% of them were satisfied with their treatment, while those prescribed pain treatments other than opioids reported a satisfaction rate of 56.2%. The average out of pocket cost of conventional prescription pain medications for these patients was $114 while the average cost of CAM was $62.

Conclusion            This study indicated that not only was CAM therapy more effective for these patients but it was also less expensive. Larger surveys conducted using National Center for Health Statistics information showed that 36% of 3,000 patients in 2002 reported using CAM. This particular study was much smaller and did not include any Asian patients and had representation from a limited number of minority patients (27.1%). The use of opioids did not improve patient satisfaction, and there was no difference between races and the use of CAM.



This was a pretty simple study. I was surprised by some of its findings. Although the study had limitations in supporting the hypothesis, small case numbers and limited number of non-Caucasians, I was more intrigued by the secondary findings. The study suggested, at least with this study population, that these chronic pain patients were more satisfied with non-conventional treatment than they are with conventional treatment. Perhaps even more surprising (or not) was that patients on opioids are the least satisfied of all. Perhaps this tells us a couple of things. First of all I believe that conventional health care providers are doing a poor job treating chronic pain. We do not use multimodal therapy enough. We tend to focus on a single conventional therapy and this leaves patients to look for alternatives if they are not satisfied with their treatment. I try to incorporate as many methods of treatments for pain patients as I feel are appropriate. If they ask me about non-conventional therapy I do tell them to discuss this with their physician, but I do not discourage its use unless there is some obvious downside for the patient. I also believe in using physical therapy, chiropractic, psychology, nursing care, pharmacy consultation, and anything else that might help provide the patient with physical or psychological support.


Steven R Wooden, MS CRNA

© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009

Quality Improvement

Rothwell M, Pearson D, Wright K, Barlow D



Bacterial contamination of PCA and epidural infusion devices

Anaesthesia 2009;64:751-753

Rothwell M, Pearson D, Wright K, Barlow D




Purpose            The purpose of this study was to audit the cleanliness of reusable pumps used for patient controlled analgesia and epidural infusions at one institution.

Background            Computer keyboards in a surgical intensive care unit have been shown to be colonized with pathogenic bacteria but little infection control literature exists for other reusable items in patient care areas. Pumps for Patient Controlled Analgesia (PCA) and epidural infusions are typically in use seven days a week and are moved from ward to ward as needed. Often, they are moved from patient to patient without much thought and may not even be regularly cleaned. At the authors’ institution, pumps were typically used by 10 different patients each month. Cross-contamination from patient to patient was possible across several different clinical areas. In the authors’ institution these pumps were cleaned with soapy water and allowed to dry after every use. They appeared clean to visual inspection.

Methodology            For a one month period all pumps returned to the operating room for assignment to another patient were cultured for bacterial contamination. Those cleaning the pumps and the nurses on the wards were not informed that the pumps were being cultured before their next use. Pumps were cultured from two locations, the keypad used by staff and the demand dosing handset used by patients. The entire keypad was swabbed. Handsets were swabbed on the bolus button and the handset that contained the button.

Following the baseline cultures an additional cleaning step was added to the pump cleaning procedure. Cleaning with soapy water after use was continued but pumps were also cleaned with isopropyl alcohol immediately before being used on the next patient. After the alcohol had dried pumps were cultured in the same manner as before in a second round of data collection.

Result            Direct observation verified that pumps were being cleaned with soapy water as procedure dictated. During the initial one month data collection period 13 unique pumps were returned to the OR an average of 4.3 times each resulting in a total of 56 cultures. Positive bacterial cultures were grown from 26 keypads and 25 handsets; 45.5% of surfaces cultured were positive overall.

In the post alcohol cleaning tests 50 pumps were cultured. Positive bacterial cultures were grown from 2 keypads and 3 handsets; 5% of surfaces cultured were positive (P<0.001 compared to baseline cultures).

Conclusion            Pumps used for PCA and epidural infusions may be contaminated with potentially pathogenic bacteria even after routine cleaning. Secondary cleaning with isopropyl alcohol immediately before pumps are assigned to another patient significantly reduced bacterial contamination.



Preventing the transmission of infectious agents from one patient to another patient or healthcare provider is more challenging than we often realize. I suspect this will become more true as “healthcare reform” forces greater and greater productivity. Greater productivity often translates into doing things faster. Economic pressures invariably move the responsibility for “cleaning” (cleaning is such an oversimplification where preventing the transmission of potentially deadly infection is concerned) down to the lowest paid, least educated individual. I do not intend the description, “least educated” to be disparaging. I simply believe that, all things being equal, someone earning minimum wage is least likely to comprehend the gravity of cleaning, disinfection, and sterilization. As a result, they may not be as motivated to perform cleaning duties as obsessively as someone who fully understands how dangerous cross contamination can be. At the very least, individuals in the OR responsible for cleaning and disinfection should be trained and supervised by a professional registered nurse well versed in infection control and zealous for the prevention of cross contamination.

This study shows that even simple devices that we don’t think about as sources for cross contamination can be highly contaminated even after routine cleaning. The bacteria cultured from the pumps in this study were almost exclusively flora normally found on human skin; not usually considered very pathogenic. I don’t think that is the message. I think the message is that the cleaning measures recommended by the manufacturer were insufficient. As a result, any pathologic bacteria that may find its way onto the device could potentially infect the next patient or staff member. As healthcare reform progresses, and production pressures increase, I believe we are going to have to become much more aware of how we use and clean equipment if we are to prevent an increase in significant nosocomial infections.


Michael Fiedler, PhD, CRNA



© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009

Respiration & Ventilation

Bolden N, Smith C, Auckley D


Avoiding adverse outcomes in patients with obstructive sleep apnea (osa): development and implemenation of a perioperative OSA protocol

J Clin Anesth 2009;21:286-293

Bolden N, Smith C, Auckley D



Purpose            This article uses 3 case reports describing the perianesthesia care of sleep apnea patients to support and explain the need for an “OSA protocol” in a large medical center.

Background            Obstructive Sleep Apnea (OSA) is a potentially lethal condition pervasive in our overweight and obese society, with a prevalence of up to 14% in American men and 7% in American women. OSA is clinically defined by multiple episodes of obstructive or mixed apnea during sleep, loud intermittent or continuous snoring, and excessive daytime sleepiness. OSA occurs along a continuum, progressing from snoring and CNS arousal without apnea, to OSA syndrome (>5 apneic episodes an hour - quantified as the apnea-hypopnea index or AHI, desaturation, and daytime sleepiness) to awake respiratory and right ventricular failure (Pickwickian syndrome). Along with obesity and/or abnormal upper airway anatomy, OSA is associated with hypertension, cardiovascular remodeling, stroke/transient ischemic attacks, accumulation of proinflammatory cytokines, and glucose intolerance/insulin resistance. Sleep apnea is an independent and major risk factor for stroke; in fact, 60 to 80% of stroke and TIA patients have untreated and undiagnosed sleep apnea.

Key to perianesthesia care is the acute awareness that OSA patients have altered CNS responses to chemical stimuli (O2 and CO2). During sleep, two events occur that impact respiration and ventilation: loss of muscle tone in the pharyngeal musculature and altered control of respiration.

Higher CO2 levels and lower O2 levels are necessary to elicit CNS arousal and stimulate breathing. In the context of sleep, arousal is an appropriate response to acute airway obstruction and hypoxemia. However, chronic arousal phenomena contribute to chronic sleep deprivation and neuropsychologic symptoms of OSA. Untreated OSA tends to progress as chemoreceptors “reset” their sensitivity so that arousal does not occur as frequently. This initiates a vicious cycle in which hypoxia and hypercarbia become more and more profound before arousal occurs. Extreme hypoxemia and hypercarbia also have cardiac sequelae; O2 saturation below 85% may be associated with brief periods of asystole or sinus tachycardia, while O2 saturation less than 75% is associated with asystole > 3 seconds or ventricular tachycardia.

Anesthesia care for OSA revolves around airway management and the cardiorespiratory alterations that profoundly worsen the effects of anesthesia drugs. All of our “routinely” administered drugs decrease pharyngeal tone, depress ventilatory responses to hypoxia and hypercarbia, and inhibit arousal responses, thereby exacerbating the pre-existing depressed CNS response pattern of the OSA patient.

Methodology            Case 1: (occurred when no OSA protocol was in place). A 45-year old male, BMI 41 kg/m2, with untreated OSA underwent general endotracheal anesthesia for 2.5 hours and received 16 mg morphine during repair of a tibial fracture. He arrived in PACU with SpO2 80%. Oxygen was applied and after 2.5 hours the patient was delivered to the ward where he received additional IV and PO opioids. At 2:30 am the night nurse noted brief periods of apnea. At 4:00 am, the patient was found in arrest and resuscitation attempted. The patient was delivered to the organ transplant team 12 hours later.

Case 2: (occurred after protocol implementation, but protocol was not followed). A 38 year-old female, BMI 64 kg/m2, with “possible OSA” underwent uterine artery embolization in the radiology suite with radiologist-driven sedation. She received midazolam 4 mg, fentanyl 275 mcg, and ketorolac 30 mg during the 2.5 hour long procedure. Despite reported pain during the procedure of 8/10, the patient desaturated to 89% which required increased O2 flow and tactile stimulation. After a 1 hour recovery, the patient was sent to the ward with a hydromorphone PCA (basal rate 1 mg/hr, bolus 0.2 mg, lockout of 10 minutes). Eighteen hours later, she was found apneic and unarousable. By this time, she had received 20 mg hydromorphone. She was successfully resuscitated and discharged 8 days later.

Case 3: (occurred with protocol in place and properly followed). A 37 year-old female, BMI 63 kg/m2, with asthma, hypertension, GERD, and treated OSA underwent gastric bypass. She brought her CPAP machine to hospital and the perioperative team had reserved a designated OSA bed with continuous pulse oximeter monitoring for post-operative care. After an uneventful PACU, she was placed on morphine PCA (basal rate 1 mg/hr, bolus 1.0 mg, lockout of 6 minutes) and transferred to OSA bed on the ward. She used her own CPAP device during sleep and despite this had 8 episodes of desaturation < 90% over the next 3 days. When these episodes occurred, the pulse oximeter alarmed and the nursing staff awakened the patient for deep breathing. She was discharged home with an otherwise uneventful recovery.

Result            Recognition of the lack of a policy-driven approach for the high-risk patient population of OSA led to the creation of a protocol for management of anesthesia care for sleep apnea (Appendix A in the paper). The goal of the protocol was to reduce adverse outcomes in this population. Protocol development included some of the key features of the ASA Practice Guidelines for Management of OSA in 2006.

Conclusion            Significant principles for perioperative management of OSA (hereafter referred to as the “Bolden protocol”) include:

1) Documented OSA patients who will require postoperative IV narcotics should be admitted and observed for up to or beyond 24 hours. Those who require PO narcotics should be observed for up to 23 hours.

2) Patients with presumed OSA (no formal testing), but all clinical signs and symptoms support the diagnosis should also be monitored as noted in 1.

3) Procedures that will require NO postop narcotics may be scheduled on an outpatient basis. These patients should be monitored for at least 3 hours postoperatively with an oxygen saturation >94% for at least 2 hours (or saturation equal to preop values if lower than 94%). Patients who cannot meet these criteria should be admitted for observation to an OSA-designated bed.

4) Patients with mild OSA (AHI < 16) who require only oral narcotics may be discharged home after being monitored for 4 hours in the PACU.

5) All tonsillectomy with adenoidectomy patients < 3 years of age should be done at hospital and generally monitored overnight. Some of these patients may be discharged earlier after evaluation. Pediatric patients with documented OSA should stay overnight in an OSA-designated bed. Pediatric patients with mild OSA, craniofacial abnormalities, or suggestive but not documented OSA may be scheduled at the discretion of the ENT surgeon.

6) OSA patients who have undergone uvulopalatopharyngoplasty (UP3) cannot be considered to be “cured” of OSA unless documented by a post-operative sleep study. (UP3 is effective in only 50% of patients.)

7) Pharmacotherapy: “Narcotics and sedative agents should be avoided whenever possible in OSA patients . . . non-narcotic agents (acetaminophen, NSAIDs), regional anesthesia techniques, and neuraxial opioids are excellent choices.” If PCA is used, consider eliminating the basal infusion and using solely bolus dosing.

8) Patients who use CPAP or Bilevel pressure support at home should bring these devices with them to hospital. Patients who forget their machines should use one supplied by Respiratory Therapy.



By using case studies, the authors are effective in explaining why OSA patients require extra-ordinary perianesthesia care. Further analysis of these cases might be useful. In Case 1, the high dose of a long-acting narcotic, morphine 16 mg, could have been avoided by the use of peripheral nerve blockade (femoral and/or sciatic n.), regional anesthesia, or simply shorter-acting agents. In addition, any patient who arrives in the PACU with a saturation of 80% clearly did not undergo an adequate pre-extubation evaluation. Masking the hypoventilation that causes hypoxemia by giving the patient a high FIO2 oxygen mask worsens the problem. Desaturation is a warning that overall ventilation is inadequate. (Inadequate neuromuscular reversal may also contribute to this.) If a patient cannot respond appropriately to command (“take a deep breath”), they may require reintubation rather than a non-rebreather mask.

In Case 2, the patient had a clear presumptive diagnosis of moderate to severe OSA, however the radiologist did not elicit this history. When a long-acting narcotic with a basal rate on a PCA pump was then prescribed post-procedure, it should not have been a surprise that profound respiratory depression developed.

In Case 3, the Bolden protocol successfully kept adverse events to a minimum. The fact that the PCA was set with a basal rate (against the Bolden recommendations) may have contributed to the episodes of apnea despite the use of the CPAP device. It also demonstrated that the distending pressure of CPAP may need to be higher in the setting of pharmacologically-induced airway obstruction.

Other parts of the Bolden protocol need updating: it is not true that neuraxial opioids are an “excellent” choice for OSA patients. In fact, epidural analgesia with opioids confers the same risk of respiratory depression as IV or IM opioids. Recall that the first sign of respiratory depression due to opioids is somnolence, followed by bradypnea. Since OSA patients may present as somnolent independent of drug therapy, this means that evidence of respiratory depression will be delayed, as bradypnea is a much later sign.

The ASA has recently published Practice Guidelines for the prevention of respiratory depression during the use of neuraxial opioids.1 The Task Force reported that in frequency of respiratory depression there is no difference between single shot neuraxial versus parenteral (IV, IM, or PCA) opioids and no difference between extended release epidural morphine and immediate release morphine. The Task Force also reported no difference in the frequency of respiratory depression, ventilatory response to carbon dioxide, somnolence, or sedation when single-injection morphine was compared with single-injection fentanyl or sufentanil, administered by either an epidural or an intrathecal route. Single doses of epidural or intrathecal, or continuous epidural morphine or hydromorphone confer a greater risk of respiratory depression than single doses of epidural or intrathecal, or continuous epidural fentanyl or sufentanil. In addition, lower doses of shorter-acting agents confer the least amount of risk for respiratory depression. Many authors recommend that local anesthetics be the sole agents used in neuraxial anesthesia for OSA patients because of the high risk of respiratory depression.

The current monitoring guidelines1 recommend that after single dose neuraxial injections of hydrophilic opioids, monitoring should occur for a minimum of 24 hours, beginning at once per hour for the first 12 hours followed by every 2 hours for the second 12 hours after injection. This protocol exceeds standard ward nursing care and will require a specific program of training and implementation for the post-operative care team.

It may be that the Bolden protocol reaches not quite far enough to protect OSA patients from harm, however, it is a giant step toward improving patient safety. When coupled with the recommendations delineated in the 2009 Practice Guidelines noted above, these combined principles for management of OSA patients may reduce morbidity and mortality.


Penelope S. Benedik PhD, CRNA, RRT



1) Practice Guidelines for the Prevention, Detection, and Management of Respiratory Depression Associated with Neuraxial Opioid Administration. Anesthesiology 2009;110:218-230. (available free at,_Detection,


Other References of Interest

Grigg-Damberger M. Why a polysomnogram should become part of the diagnostic evaluation of stroke and transient ischemic attack. J Clin Neurophysiology 2006;23:21-38.

Practice Guidelines for the Perioperative Management of Patient with Obstructive Sleep Apnea: A report by the American Society of Anesthesiologists Task Force on Perioperative Management of Patients with Obstructive Sleep Apnea. Anesthesiology 2006;104:1081-93.

Bolden N, Smith C, Auckley D, Makarski J, Avula R. Perioperative Complications During Use of an Obstructive Sleep Apnea Protocol Following Surgery and Anesthesia. Anesth Analg 2007;105:1869-1870.


This topic was suggested by Johnny J. Sacco, CRNA.

© Copyright 2009 Anesthesia Abstracts · Volume 3 Number 6, June 26, 2009