ISSN NUMBER: 1938-7172
Issue 2.7 VOLUME 2 | NUMBER 7

Michael A. Fiedler, PhD, CRNA

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

Guest Editor:
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

Download Text File

Download Text File

Download Text File

Download Text File

Download Text File

Download Text File

Download Text File

Download Text File

Download Text File

Download Text File



Krebs MJ, Sakai T


Retropharyngeal dissection during nasotratheal intubation: a rare complication and its management

J Clin Anesth 208;20:218-221

Krebs MJ, Sakai T



Purpose            The purpose of this report was to describe a case in which the tip of an ETT tube tunneled submucosally to the level of the posterior oropharynx during nasoendotracheal intubation.

Background            Nasotracheal intubation is commonly used for ENT surgery as well as other purposes. Airway trauma is more likely during nasoendotracheal intubation than during oral intubation due to the smaller space through which the endotracheal tube (ETT) passes and at least partially blind insertion of the ETT. Complications reported following nasoendotracheal intubation include bleeding, bacteremia, sinusitis, dislodgment of the adenoids, intracranial placement of the ETT, pharyngoesophageal perforation, and retropharyngeal dissection. If a nasal ETT that has tunneled beneath the pharyngeal mucosa is advanced too forcefully or ventilation is attempted through the ETT, additional trauma, bleeding, and airway obstruction are likely. Retropharyngeal abscesses are often life threatening.

Methodology            A healthy, 95 kg, 54 year old women presented for mandibular surgery. A general nasoendotracheal anesthetic was planned. After nasal preparation with oxymetazoline nasal spray for vasoconstriction general anesthesia was induced. After introducing Surgilube into the right nostril a 7 mm ID nasal RAE ETT was inserted 7 cm without resistance. Intubation was to be completed with direct laryngoscopy but the tip of the ETT was not visible when laryngoscopy was performed. There was, however, a bulge in the right posterior oropharynx. Palpation of the bulge revealed the tip of the ETT beneath the pharyngeal mucosa.

Result            The nasal ETT was removed and an oral ETT was placed. An ENT surgeon was consulted to examine the damage produced by the retropharyngeal penetration of the ETT. Subsequently, the surgical procedure was performed and the patient was admitted postoperatively. She received follow up studies, broad spectrum antibiotics, and was discharged home on postoperative day two without sequella.

Conclusion            Retropharyngeal dissection of the ETT is a known, though unusual, complication of nasoendotracheal intubation. Early recognition of the complication, prompt assessment, and proper treatment are needed to avoid potentially life threatening complications.



This complication may seem a little far fetched, or like something that would only happen in the hands of an oafish, ham-handed anesthetist, but I did it once late in my training. Imagine my surprise when I did a laryngoscopy intending to grab the tip of the tube with the Magill forceps only to find a big bump in the throat where the tube should be. (Or more precisely, where the tube was … just beneath the pharyngeal mucosa!) Retropharyngeal placement of a nasal ETT can happen even when it doesn’t feel like you are advancing the ETT against too much resistance. This case report is not really extraordinary, but it serves to remind us of the possibility of this complication. Fortunately, with proper follow up examination and treatment it rarely becomes a problem.


Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Equipment & Technology

Holbrook SP, Quinn A


An unusual explanation for low oxygen saturation

Br J Anaesth. 2008;101:350-353

Holbrook SP, Quinn A


Purpose            This report described a case of persistently low pulse oximetry readings in an apparently healthy adult despite supplemental oxygen.

Background            Pulse oximetry has been widely used in anesthesia for over two decades and is an essential monitor. Pulse oximetry determines oxygen saturation by measuring hemoglobin’s absorption of two different wavelengths of light as blood circulates past a probe. Under some conditions pulse oximetry underestimates or overestimates the true saturation of hemoglobin with oxygen. Arterial Blood Gas (ABG) machines use a more accurate technology to determine the percent concentration of oxyhemoglobin as well as several other species of hemoglobins (e.g. hemoglobin bound to carbon monoxide).

Adult hemoglobin is composed of two pairs of two different polypeptide molecules bound to a heme molecule. Hundreds of hemoglobin variants have been described. Since variant hemoglobins absorb light differently than normal hemoglobin, some of them may alter not only oxygen carrying capacity but the accuracy of oxygen saturation measurements produced by a pulse oximeter even if they are physiologically benign.

Methodology            A 62 year old man presented for elective surgery. He was a former smoker and experienced seasonal “asthma” but was otherwise healthy. He had no apparent cardiovascular or respiratory symptoms, showed no cyanosis, and had a normal chest x-ray and lab. He had had multiple previous anesthetics without complications.

In the OR, his pulse oximeter oxygen saturation (SpO2) was 88% on room air. Supplemental oxygen resulted in an increase in oxygen saturation to 93%. His previous anesthetic records noted the same problem but surgery proceeded uneventfully and follow up was recommended. Prior to induction, an arterial line was inserted. An ABG was drawn and surgery proceeded with general anesthesia. Periodic ABGs were drawn throughout the case at various inspired oxygen concentrations.

Result            While the patient’s room air pulse oximeter oxygen saturation was only 88%, an ABG drawn on room air reported 98% oxygen saturation and a PaO2 of about 88 torr.

Further ABGs reported:

FIO2 30% oxygen              SpO2 90%            SaO2 99.5%            PaO2 157 torr

FIO2 60% oxygen              SpO2 90%            SaO2 99.6%            PaO2 411 torr

FIO2 94% oxygen              SpO2 92%            SaO2 99.7%            PaO2 482 torr

The operation was completed without complications. Postoperatively, the patient was diagnosed with an uncommon hemoglobinopathy affecting 7% of his circulating hemoglobin.

Conclusion            This case reported a man with a benign hemoglobin variant that consistently resulted in a pulse oximeter oxygen saturation value about 10% lower than his actual hemoglobin oxygen saturation.



Pulse oximeters are marvelous devices and we rely upon them every day. The technology has been so well developed that pulse oximeters are quite accurate and dependable. We are aware that there are a few conditions when the pulse oximeter will report a falsely high oxygen saturation. For example, heavy cigarette smoking or smoke inhalation produces carboxyhemoglobin (COHb, hemoglobin bound to carbon monoxide). Patients with elevated carboxyhemoglobin have falsely elevated pulse oximeter oxygen saturation readings because pulse oximeters can’t tell the difference between oxyhemoglobin and carboxyhemoglobin. We are, I think, less suspicious that a pulse oximeter might be giving us a falsely low saturation reading. Nevertheless, there are a number of conditions, such as sickle cell crisis or variant hemoglobins, when the pulse oximeter will give us a falsely low reading. This case report served to remind me of the limitations of my beloved pulse oximeter. While we don’t do nearly as many ABGs as we did 20 years ago, ABGs are still useful and are usually a low risk procedure. Whenever we have a significant question about the accuracy of pulse oximetry readings during general anesthesia we should probably consider checking an ABG.


Michael Fiedler, PhD, CRNA

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008


Sessler, D



Temperature Monitoring and Perioperative Thermoregulation

Anesthesiology 2008;109:318-38

Sessler, D




Purpose            The purpose of this article was to review the impact of anesthesia on temperature regulation.

Background            Normal human thermoregulation can overcome many of the typical environmental temperature changes we are exposed to. However, we find most unwarmed surgical patients to be hypothermic indicating that anesthesia negatively impacts the normal thermoregulatory mechanisms. Understanding the aspects of temperature monitoring and the impact of anesthesia on thermoregulation is important.

Result            Core temperature can be 2-4°C warmer than the extremities. Temperature monitoring is important to detect not only hypothermia but hyperthermia caused by excess warming, infection, transfusion reaction, or malignant hyperthermia.

Monitoring sites: Practical core temperature monitoring sites include distal esophageal, tympanic, nasopharyngeal, and pulmonary artery. Each of these sites requires proper sensor placement which is not always easy to acquire or maintain. Core temperature can be reasonably approximated by “near-core” monitoring. These sites include oral, axillary, and bladder. Each of these sites has circumstantial limitation so the practitioner should use these sites selectively. As an example, bladder temperature is highly influenced by urine flow and oral temperature is influenced by ambient gas flow. Skin temperature is lower than core temperature and different skin sites vary depending on environmental conditions. Forehead temperature is fairly consistent, and is typically 2°C less than core temperature. Surprisingly, even intense vasodilatation and vasoconstriction does not alter the variance between forehead skin and core temperature significantly.

Newer infrared devices that are supposed to detect near core temperature by scanning the forehead and temporal artery are not accurate enough for clinical anesthesia use.

Normal thermoregulation: The body tightly regulates core body temperature. The dominant controller of body temperature is the hypothalamus. It receives input from skin, neuraxis, and deep tissues. Nearly 80% of the input is derived from core structures. Core temperatures normally range between 36.5°C and 37.5°C. Values below 36°C or above 38°C usually indicate a loss of thermoregulatory defenses. Nonshivering thermogenesis is the primary defense against cooling in infants, but relatively unimportant in adults. Shivering is a late, but important mechanism to elevate core body temperature in adult humans, while sweating is the most important mechanism in decreasing core temperature.

Hyperthermia: Passive hyperthermia is caused by overheating. This type of hyperthermia is more common in infants and children. It can be treated by removing the heat source or insulation. In contrast, malignant hyperthermia results from increased metabolism and although the thermoregulatory system remains intact it can not respond properly because of intense vasoconstriction caused by catecholamine levels that are often 20 times higher than normal.

Fever is a body defense mechanism that raises the thermoregulatory target temperature in response to immune system activation. General anesthesia and opioids inhibit this immune fever response. Post surgical fever is often seen in response to inflammation, but may be caused by other more serious causes. Treatment of hyperthermia should be specific to the cause. Active cooling should be used with caution. It often fails to reduce the core temperature and can worsen the situation by triggering thermoregulatory defenses.

General Anesthesia and Thermoregulation: All general anesthetic agents impair the thermoregulatory system. General anesthesia causes the warm response threshold to be increased slightly and the cold response threshold to be decreased dramatically. The impact that drugs have on cold response varies depending on the drug and its dose. Propofol and alfentanil decrease vasoconstriction and shivering in a linear fashion while inhalation agents decrease the same cold responses non-linearly. This means that propofol and alfentanil have a greater impact on cold response at lower doses while inhalation agents have a greater impact at higher doses. Midazolam has minimal influence on thermoregulation.

Response in Infants and Elderly: Because infants have a larger surface area relative to body weight, they are at greater risk of hypothermia than an adult. In addition, inhalation anesthetics inhibit non-shivering thermogenesis which is important to infant thermoregulation. People greater than 60 years old have a vasoconstriction response that initiates 1°C lower than younger adults which predisposes them to hypothermia under similar anesthetic conditions.

Neuraxial Anesthesia and Thermoregulation: Neuraxial anesthesia reduces central thermoregulation slightly and prevents the patient’s ability to sense the existence of hypothermia. Spinal and epidural anesthesia impact central thermoregulation for unknown reasons, but do so proportionally to the number of spinal segments blocked. In addition, they decrease the intensity of shivering and vasoconstriction while reducing the gain that shivering has on hypothermic response. When propofol is given for sedation in addition to using neuraxial anesthesia, the thermoregulation impact is compounded. Exaggerated shivering is often seen after neuraxial anesthesia in obstetric patients. Studies have been unable to identify a reason for this response specific to obstetric patients. Also, prolonged epidural anesthesia has been associated with hyperthermia in obstetric patients as well as postoperative non-pregnant patients. No convincing mechanism for this has been identified. It could be caused by passive hyperthermia, increased metabolic rate, sweating inhibition, infection, or inflammation. It is interesting to note that high dose steroids significantly reduced elevated temperature in obstetric patients suggesting that inflammation plays a role in hyperthermia for these patients.

Conclusion            Core temperature monitoring is the best indicator of body heat distribution. Body temperature should be monitored by core monitoring or “near-core” monitoring in patients undergoing general anesthesia exceeding 30 minutes and neuraxial anesthesia procedures lasting more than one hour. Patients susceptible to hypothermia should be actively warmed as necessary to maintain normothermia.



I enjoyed this review of thermoregulation. Even though the article went into painfully detailed and redundant descriptions of the technical aspects of temperature monitoring and the biological aspects of temperature control, I think there was enough clinical information to make reading the entire text worthwhile. However, there were some essential issues missing from this review. I would have liked to have seen more information on appropriate methods for clinical temperature monitoring, a discussion on methods of warming patients, a better discussion on the prevention of hypothermia, and perhaps the impact that hypothermia has on patient recovery and infection.

The AANA has addressed temperature monitoring in its practice recommendations. The current recommendation pertaining to temperature monitoring in an AANA document suggests that the anesthesia provider should “monitor body temperature continuously on all pediatric patients receiving general anesthesia and when indicated, on all other patients.”1  This recommendation provides for situational flexibility, but I would agree with the authors opinion that all patients receiving a general anesthetic greater than 30 minutes in duration should have temperature monitoring as well as all patients receiving neuraxial anesthesia in cases longer than one hour.

I continue to fight the misconception by my nursing staff that warmed blankets provide core temperature protection for patients. I agree they feel good and they warm the skin which triggers the temperature receptors to reduce shivering, but what they fail to understand is that cotton blankets provide minimal increase in core body temperature. Patients in the recovery room with core temperature hypothermia need aggressive warming with active warming devices and frequent temperature monitoring. Shivering promotes increased oxygen consumption and should be treated by raising the core body temperature instead of stimulating the skin receptors.

During surgery we promote heat loss by having patients minimally covered in a cold environment, wash them with cold fluids which we allow to evaporate, increase heat loss through the incision, infuse cold fluids, and reduce their metabolic and thermogenic response with anesthetic agents. Then in the recovery room we give patients opioids which reduce the shivering response and their ability to indicate that they feel cold, we throw warm cotton blankets on them, and take one or two skin temperature readings without understanding the inaccuracies of some of those readings. It is well documented that hypothermic patients are more susceptible to postoperative infection and recover poorly from anesthesia and surgery. I think it is about time that we recognize that maintaining normothermia is almost as important as maintaining normal oxygen saturation.


Steven R Wooden, MS, CRNA



© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Rajagopalan S, Mascha E, Na J, Sessler D



The effects of mild perioperative hypothermia on blood loss and transfusion requirement

Anesthesiology 2008;108:71-7

Rajagopalan S, Mascha E, Na J, Sessler D




Purpose            This was a meta-analysis to evaluate the effect that hypothermia had on surgical blood loss and transfusion needs.

Background            Previous studies have indicated that even mild hypothermia can cause serious problems such as wound infections, prolonged recovery from anesthesia, increased myocardial oxygen demand, and patient discomfort. It has also been suggested that hypothermia impairs platelet function by inhibiting thromboxane A2 release and other enzymes which are important to coagulation. The initial review of previous studies concerning hypothermia and its impact on blood loss found conflicting conclusions. Some studies even suggest that blood loss is reduced by mild hypothermia. This article presents a systematic review and meta-analysis of studies involving the evaluation of mild hypothermia and blood loss.

Methodology            A search was conducted for published, randomized, controlled trials that compared the outcomes of normothermic patients as compared to patients with mild hypothermia. Mild hypothermia was defined as a core temperature between 34-36°C. Studies were included that were published between 1966 and 2006. Information that was evaluated included physical status, type of surgery, anesthesia technique, operating room temperature, core temperature at the end of surgery, blood loss, and transfusion requirements. Blood loss in various studies was reported as either a mean or median. For consistency, all information was converted to a mean and standard deviation. The calculated mean blood loss was then incorporated into a formula that yielded a ratio of normothermic to hypothermic blood loss where a ratio of 1.0 indicated no difference in blood loss. The true amount of blood transfused was not considered, but instead a relative risk analysis was performed comparing transfusion risks in hypothermic and normothermic patients. A relative risk ratio of less than 1.0 indicated that the risk of transfusion was less for normothermic patients as opposed to the risk or transfusion for hypothermic patients. A quality score system was developed with points given to studies that used blood loss as the primary outcome, those who objectively measured blood loss, those that specified a randomization method, and those that used specific “intent to treat” methodology. Finally, the difference between mean temperatures and blood loss was assessed.

Result            Approximately 1800 articles were reviewed. Articles excluded were those that reviewed temperature reductions that were less than 34°C and studies that were smaller than 15 subjects. There were 18 studies that met criteria but only 14 were selected because the blood loss data of four studies were either inconsistent or missing. Of those studies selected, some reported intraoperative or postoperative blood loss only, while others included total blood loss. The 14 studies included 1,219 patients and 985 transfusions.

There was only a 0.85°C median difference between normothermia and hypothermia in selected studies. The resulting analysis indicated that normothermic patients had significantly less blood loss than hypothermic patients, and transfusion was significantly less frequent in the normothermia group.

Conclusion            This meta-analysis evaluated 14 studies with contradictory conclusions. The combined results of all 14 studies showed that blood loss was 16% greater in moderately hypothermic patients than in normothermic patients. In addition, hypothermia increased the risk of blood transfusion by 22%.



The authors of this study were quick to note that meta-analyses have inherent methodological problems. These types of studies rely on the data of various other studies with a variety of methods, outcomes, and reliabilities. This meta-analysis reviewed 1800 studies and found only 14 that met its criteria. The 14 studies selected were so different from each other in respect to data collection and outcome that complicated formulas had to be used to develop analyzable information. As a clinical practitioner with a limited educational background in statistics, I often find myself somewhat mystified by the methods used in developing this type of information and question if the outcome was the results of statistical wizardry or if the outcome was truly accurate. For that reason, I usually take the results of a meta-analysis with a grain of salt, but I also understand that sometimes a meta-analysis is the only way to bring small studies together in order to paint a larger picture. Those larger pictures can help develop a better focus for randomized studies that can put homogenous data together for clinically significant results.

Regardless of how I might feel about a meta-analysis, I think this particular study has some merit. It brings to light the possibility that even a small drop in core temperature (less than 0.85°C) can significantly increase blood loss and increase the risk of requiring a transfusion. I believe that temperature monitoring and efforts to maintain normothermia are critical to improving surgical patient outcome. This study supports that conclusion.


Steven R Wooden, MS, CRNA



© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Obstetric Anesthesia

Svanstr?m MC, Biber B, Hanes M, Johansson G, N?slund U, B?lfors EM


Signs of myocardial ischaemia after injection of oxytocin: a randomized double-blind comparison of oxytocin and methylergometrine during caesarean section

Br J Anaesth 2008;100:683-689

Svanström MC, Biber B, Hanes M, Johansson G, Näslund U, Bålfors EM



Purpose            The purpose of this study was to determine whether ECG changes and complaints of chest pain during cesarean section with regional anesthesia were associated with oxytocin administration.

Background            ECG changes consistent with myocardial ischemia have been reported in healthy women during cesarean section with regional anesthesia. These ECG changes have often been associated with hypotension, tachycardia, and subjective complaints such as chest pain, headache, and dyspnea.

Intravenous (IV) oxytocin produces hypotension and tachycardia and is often being administered when ECG changes are noted. Oxytocin receptors are present in the myocardium, vessel walls, the CNS, lactation glands, and the myometrium. Oxytocin causes direct smooth muscle relaxation in many blood vessels resulting in decreased systemic vascular resistance and hypotension. In coronary arteries, however, oxytocin produces vasoconstriction. Heart rate increases occur due to direct effects on the myocardium and secondary to hypotension. Previous studies have observed normal ventricular wall motion during complaints of chest pain during cesarean section, suggesting that the pain was not related to myocardial ischemia.

Methodology            This randomized, double-blind study included healthy term pregnant women scheduled for elective cesarean section under spinal anesthesia. Women were randomized into an oxytocin group and a methylergotamine group. An additional group of healthy, non-pregnant female volunteers received oxytocin alone and served as controls.

All women undergoing cesarean section had baseline vital signs measured and received 500 mL dextran 70 IV. A spinal anesthetic was induced with 12.5 mg to 13.75 mg hyperbaric bupivacaine in the sitting position. Two liters nasal cannula oxygen was administered during the cesarean section. Ephedrine 5 mg IV was administered as needed to maintain a systolic blood pressure (BP) above 95 torr. Left uterine displacement was maintained at all times when supine. After the umbilical cord was clamped, women received either 10 IU oxytocin or 0.2 mg ergotamine IV. Ten non-pregnant controls received 10 IU oxytocin in the same manner. Arterial BP was measured with a radial arterial line in all subjects.

Subjects were monitored for 15 minutes after oxytocin or ergotamine administration. ST segment depression was measured at a point 60 ms past the “J-point.” Horizontal or down sloping ST depression greater than 0.1 mV from baseline was considered abnormal. A cardiologist blinded to subject group membership interpreted all ECGs.

Result            Fifty women were the subjects of this study; 20 each undergoing cesarean section and receiving either oxytocin or ergotamine and an additional 10 non-pregnant controls who received only oxytocin. The sensory anesthesia level was T-4 or higher in 19 of 20 women in each cesarean section group. The remaining women in each group had a sensory level of T-6.

Oxytocin administration produced hypotension and tachycardia. Within less than a minute after 10 IU oxytocin was administered IV there was a significant decrease in BP and an increase in heart rate (HR). HR increased by an average of 28 bpm (P<0.05 vs ergotamine). Systolic, mean, and diastolic BP fell by an average of 47, 33, and 27 torr respectively (P<0.05 vs ergotamine). ST depression occurred in 11 of 20 women and T-wave changes in 7 of 20 women.

Ergotamine administration produced an increase in BP. Within about a minute after 0.2 mg ergotamine was administered IV there was a significant increase in BP. There was no change in HR. HR increased by an average of 4 bpm. Systolic, mean, and diastolic BP increased by an average of 15, 11, and 9 torr respectively. Less severe ST depression occurred in 6 of 20 women and T-wave changes in 3 of 20 women.

In non-pregnant control women, 10 IU oxytocin IV resulted in a significant decrease in BP and an increase in HR in about 1.5 minutes. HR increased by an average of 52 bpm (P<0.05 vs ergotamine). Systolic, mean, and diastolic BP fell by an average of 35, 30, and 28 torr respectively (P<0.05 vs ergotamine). ST depression occurred in 5 of 10 women and T-wave changes in 6 of 10 women. At the same time, these 10 women reported headache (n=9), flushing (n=8), palpitations (n=6), chest pressure (n=3), and dyspnea (n=1).

Conclusion            Hemodynamic and ECG changes seen during cesarean section with regional anesthesia were due to the administration of oxytocin. These changes were associated with subjective complaints typical of myocardial ischemia.



Obstetric anesthetists have long known that women often complained of chest pressure / pain / discomfort, nausea, dyspnea, and the like during cesarean section with regional anesthesia. These complaints often occurred shortly after delivery. The complaints were usually brief and only occasionally severe. There didn’t seem to be any morbidity associated with them. Then, about 1990, the literature began to buzz with reports that these women were exhibiting ECG changes characteristic of myocardial ischemia at the same time they were complaining of chest pain. (1, 2) A few years later several studies of ventricular wall motion showed no abnormal motion associated with myocardial ischemia and anesthesia was left scratching its head. The complaints were apparently not related to myocardial ischemia and women didn’t seem to suffer any harm. The problem got pushed to the back burner. Subclinical air emboli? Noxious visceral stimuli from exteriorization of the uterus?

I’ve long been discontented with the “we don’t know what it is but it doesn’t kill anybody” answer to the question, what causes women to complain of chest pain during cesarean section? Now comes this study that shows pretty convincingly that the complaints of chest pain were strongly associated with the administration of oxytocin. (While administered as an IV bolus in the study, it is reasonable to assume that a rapid IV infusion would produce the same result.) But more interestingly, and not really addressed by the study, this finding rekindles the question of whether or not the chest pain is cardiac in origin. Previous findings of normal ventricular wall motion not withstanding, this study paints a pretty convincing picture. These women are hyperdynamic to begin with (high myocardial oxygen demand) and then they get oxytocin in large doses to rapidly put a stop to uterine bleeding. If the surgeon isn’t happy with the uterine contraction they get more oxytocin. As a result their heart rates increase even further. Much of the time they have some level of hypotension for a while, occasionally severe hypotension. And all this with a drug that is known to constrict coronary arteries. Even healthy people will eventually get chest pain if you raise their oxygen demand high enough and lower they oxygen supply low enough. So why don’t they have bad outcomes? Probably because they are usually young and healthy, and the duration of their myocardial ischemia is almost always short; too short to be able to see ventricular wall motion abnormalities. In hindsight, really, we should have been more suspicious of this.

This is a highly important study that, for me, closes a loop that has been left open too long. The lesson it offers us, I believe, is one of margin of safety. The women that complain of chest pain during a cesarean section under regional anesthesia while the pitocin is running are probably being pushed right up to the edge of myocardial ischemia … but only for a minute or two. During those minutes their cardiovascular margin of safety is gone. If the ischemia gets worse or lasts longer or something else happens they will likely suffer harm. I want the widest margin of safety for my patients that I can get and all this makes me nervous. Our task now is to continue gaining more understanding of this phenomenon and to make an anesthetic plan that will prevent maternal myocardial ischemia and restore the margin of safety. We can begin by being careful to administer “just enough” pitocin to get the uterus to contract rather than running the pit wide open and then cutting it back after the uterus begins to look like a rock.


Michael Fiedler, PhD, CRNA


1.            Palmer CM, Norris MC, Giudici MC et al. Incidence of electrocardiographic changes during cesarean delivery under regional anesthesia. Anesth Analg. 1990;70:36-43.

2.            Zakowski MI, Ramanathan S, Baratta JB et al. Electrocardiographic changes during cesarean section: a cause for concern? Anesth Analg. 1993;76:162-7.


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Pediatric Anesthesia

Bachiller P, McDonough J, Feldman J


Do new anesthesia ventilators deliver small tidal volumes accurately during volume-controlled ventilation

Anesth Analg 2008;106:1392-1400

Bachiller P, McDonough J, Feldman J



Purpose            The purpose of this study was to evaluate the accuracy of tidal volume delivery during volume controlled ventilation via four different anesthesia ventilators, including older and newer ventilator models. The research question asked was whether newer ventilators used by anesthesia providers were capable of delivering accurate tidal volumes to pediatric patients when using the volume controlled mode.

Background            Mechanically ventilating the pediatric patient during an anesthetic has posed challenges. This remains true even when anesthetizing the healthiest of children. If the pediatric patient has lung pathology, the problem can be worsened. Small changes in tidal volume during mechanical ventilation can result in barotrauma and volutrauma when lungs are exposed to high pressure or volume.

Throughout the years, various ways to safely and effectively ventilate the pediatric patient have been tried. Most recently, pressure controlled ventilation has been the preferred method of ventilating the pediatric patient. During pressure controlled ventilation, the tidal volume delivered to the patient mostly depends upon lung compliance. Delivered tidal volume is independent of circuit compliance and fresh gas flow. The ventilator is able to deliver its maximum volume, and if uncuffed endotracheal tubes are used, the result is not a reduction of set volume unless the leak is inappropriately large. When the maximum pressure to which the patient’s lungs are exposed is set by the provider, the potential for barotrauma is substantially reduced.

A limitation of pressure controlled ventilation is the fluctuation of tidal volume that occurs when lung compliance changes. Exhaled volume must be carefully monitored so that fluctuations in tidal volume are detected. Newer anesthesia ventilators on the market have been designed to accurately deliver set tidal volumes to the airway. Their features include compensation for the compliance of the breathing system, and for changes in fresh gas flow, so that the volume set is delivered to the patient. These features may enhance the use of volume controlled ventilators for pediatric patients, … if we can be confident set tidal volume is actually what is delivered.

Methodology            This study was performed in the laboratory using simulated test lungs. The goal of the research was to determine the accuracy of tidal volume delivery by anesthesia ventilators using varied breathing circuit and lung compliance. Four different ventilators were connected to a test lung using a standard circle anesthesia circuit. Test lungs simulated both infant and adult lungs over a wide range of airway compliance. Airway resistance was set to 20 cm H2O for all measurements. The infant test lung was set to tidal volumes of 100 and 200 mL. The adult test lung was set to tidal volumes of 500 mL. The breathing circuit used was expandable and tested at different lengths. Three different anesthesia ventilators were studied, one had two different software configurations. Thus, four ventilators were studied: two with, and two without circuit compliance compensation.

Anesthesia ventilators without breathing circuit compliance compensation:  Smartvent 7900 (Datex-Ohmeda) and Avance (GE Healthcare).

Anesthesia ventilators with circuit compliance compensation: Aisys (GE Healthcare) and Apollo (Dräeger Medical).

The volume set to be delivered by the ventilator was compared with the volume actually delivered to the airway. Circuits were either extended or contracted. Tidal volumes were recorded as displayed, delivered, and displayed volume minus delivered volume. Three different protocols were used.

  • Protocol 1. Respiratory rate set at 20 breaths per minute with an IE ratio of 1:2. Fresh gas flow set at 2 L/min of air plus 0.2 L/min of oxygen. Measurements of volume were done at three different settings after allowing the ventilator to equilibrate. This protocol mimicked a typical scenario using pressure controlled ventilation with the Ohmeda Smart vent 7900. The goal was to assess the variation in delivered volume as lung compliance changed.
  • Protocol 2 was designed to determine the degree to which different anesthesia ventilators delivered set tidal volume during volume controlled ventilation. Both contracted and extended breathing circuits were used. Tidal volumes were measured at varying ventilator settings and lung compliance was measured and recorded. The displayed inspiratory airway volume was compared to delivered volume.
  • Protocol 3 was performed in the same way as protocol 2 except breathing circuit compliance was changed by switching the circuit configuration. The data were compared with delivered tidal volume measurements obtained in protocol 2. Only ventilators with compliance compensation were tested in protocol 3.

Results            For changes in delivered volume during pressure controlled ventilation, Protocol 1, each circuit configuration (extended or contracted) resulted in a decrease in tidal volume measured at the airway proportionate to a decrease in lung compliance. In other words, if lung compliance was reduced by 50%, this demonstrated a similar decrease in delivered tidal volume. If lung compliance was consistent, volume delivered was independent of the circuit configurations.

Protocol 2 determined the accuracy of volume delivered to the airway during volume controlled ventilation with a test lung compliance of 0.0025 L/cm H2O and a set tidal volume of 100 mL. The Smartvent was designed to deliver the set tidal volume to the breathing circuit. The measured volume delivered to the circuit was slightly larger than the set tidal volume. The measured volume delivered to the airway was lower than the set tidal volume. When the circuit was lengthened, the delivered tidal volume decreased even more.

Changing lung compliance affected the tidal volume delivered to the airway to the greatest degree in ventilators without breathing circuit compliance compensation. With lung compliance set at what is typical of a neonate, the tidal volume actually delivered to the airway ranged from 45.6% to 100.3% of set tidal volume, depending on the ventilator used. The Apollo and Aisys that use breathing circuit compliance compensation delivered 95.5% - 106.2% of their set tidal volumes. The Smartvent 7900 and Avance delivered tidal volumes ranging from 45.6% to 109.3% of set volumes. Peak inspiratory pressures ranged from 21 to 53 cm H2O, and depended on lung compliance and tidal volume delivered, not on the actual ventilator used. The GE Aisys and Dräeger Apollo, both with breathing circuit compliance compensation software, had an error rate of <9% over all conditions tested, as long as the pre-use circuit compliance test was performed.

Protocol 3 results showed that without compliance compensation an extended or contracted breathing circuit had a major effect on the tidal volume delivered when lung compliance and tidal volume were low. Ventilators with compliance compensation software avoided this problem when a breathing circuit compliance check was done after each change in circuit configuration.

Conclusion            There are two major ways that ventilators attempt to determine accuracy of set tidal volume to the actual airway of the patient:  by measuring the volume delivered to the circuit using an inspiratory flow sensor and controlling the ventilator by using the sensor, and by using a software algorithm set to have the machine compensate for the compliance of the breathing circuit. The Ohmeda Smartvent 7900 and GE Avance (without compliance software) use the first way. The actual volumes delivered by the Smartvent and Avance differed from set tidal volumes by up to 55%. This was true for lung compliances and tidal volumes tested similar to those in neonates and smaller infants. An exhaled volume sensor located at the expiratory valve was shown to over estimate the volume that the patient received. If accurate tidal volume information is wanted using these ventilators, an adult or pediatric airway flow sensor should be used. By using a flow sensor, ventilator settings can be adjusted guided by the tidal volumes delivered to the airway during volume controlled or pressure controlled ventilation.

The GE Aisys and the Dräeger Apollo use breathing circuit compliance compensation technology to deliver tidal volumes accurately. Both of these ventilators delivered the set tidal volumes to the airway with a great degree of accuracy (+5%), both when tested with low lung compliances and low tidal volumes. During volume controlled ventilation, the compliance of the breathing circuit lead to fluctuations in tidal volume delivery in the absence of compliance compensation technology.


Comment            Over the past decade, anesthesia experts have been provided the utmost of sophisticated technology to use, some invented by providers, to offer the safest of care to our patients. It will take a concerted effort, however, to spend time truly learning the intricacies of the technology; it is only as good as our understanding of it, and our willingness to use it. This research, while presented in a complex manner, is a wonderful example of the scientific technologic advancement of anesthesia. We have become smarter regarding a ‘task’ we perform with every patient ventilation. Our patients hopefully will experience ‘more’ positive outcomes. I applaud the authors for their determination to dive in to the details of the technology component of ventilation, albeit based on human physiology, in their efforts to disseminate information about providing high quality and safe care.



Mary A. Golinski, PhD, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Farion K, Splinter K, Newhook K, Gaboury I, Splinter W


The effect of vapocoolant spray on pain due to intravenous cannulation in children:  a randomized controlled trial

CMAJ 2008;179:31-6

Farion K, Splinter K, Newhook K, Gaboury I, Splinter W



Purpose            The purpose of the research was to determine if vapocoolant spray was an acceptable alternative to topical anesthetics, when used for anesthetizing the skin of a pediatric patient who must have intravenous cannulation.

Background            Intravenous cannulation causes pain and additional stress and anxiety in young children who are already acutely ill; often times they simply cannot tolerate the discomfort. When children must have intravenous cannulation because of an urgent situation, irrespective of ‘where’ in the hospital the child is physically located, inducing anesthesia, attempting to sedate them, or numbing the skin with topical cream, isn’t always a feasible option. Several of the methods often used to alleviate discomfort, stress, and anxiety take a great deal of time. This prolonged time is often impractical, especially in non-elective situations. A new medication has been developed which, due to it’s composition and chemical makeup, decreases the time between administration and cannulation. Vapocoolant spray provides transient anesthesia via evaporation-induced skin cooling. This skin cooling decreases pain, specifically during intravenous catheter placement. Pain Ease is a new product and the researchers hypothesized that it would be efficacious in minimizing the discomfort of intravenous cannulation and practical to use in the pediatric patient.

Methodology            The study was conducted as a double-blind, randomized controlled trial at Children’s Hospital of Eastern Ontario. Inclusion criteria involved children 6-12 years of age who required ‘urgent’ intravenous cannulation. Exclusion criteria involved those who needed emergency vascular access, those that had any contraindication to vapocoolant spray, and those who had a history of peripheral vascular disease. A child life specialist, as part of the ‘team’ had to be present when the child was approached to participate. Patients were randomly assigned to one of two groups. Group 1 patients received placebo spray and group 2 patients received Pain Ease vapocoolant spray. Baseline demographic data were gathered. Additionally, children were asked to rate their anxiety regarding the IV placement by indicating their level of anxiety on a color visual analogue scale. Other information that was recorded at baseline included the location of the IV placement, the size of the cannula, the nurse’s experience, and the assessment of the child life specialist as well as any distraction activity chosen by the child. Each application of either placebo or coolant spray, was administered at room temperature, sprayed at a distance of 8-18 cm and for 4-10 seconds until the skin blanched. Those who received the normal saline placebo spray received their treatment in a similar manner. The primary outcome measure was the child’s self-reported pain during intravenous cannulation. The secondary outcome measures included the success rate on the first attempt for the IV placement and assessments of the child’s parents as well as from the child life specialist. Lastly, the nurses rated the ease of cannulation.

Result            A total of 80 children met the inclusion criteria and were enrolled in the study; 40 children in each group. Demographic data were not different between the two groups except for IV cannula size. Those in the treatment group received more 24 gauge IV catheters compared to the placebo group (98% vs. 80%). (All other patients in both groups received 22 gauge catheters.) This was adjusted for statistically. Pain scores differed significantly between the two groups; those in the treatment group had a mean score of 37 mm on the VAS versus 56 mm with placebo (p<0.01). The significance remained when the adjustment for needle size was made. Successful IV starts on the first attempt occurred more often in the treatment group (p = 0.03). Three sets of observers; the child life specialist, the parents, and the nurses; all gave ratings of the child’s comfort either favoring the use of the vapocoolant spray, or noting similar responses by the child comparing the two treatments. The nurses rated the ease of the IV start  as ‘easy’ or ‘very easy’ more often when the treatment spray was used (p = 0.02). There were no adverse treatment outcomes noted for the entire study.

Conclusion            Compared to placebo, Pain Ease vapocoolant spray showed a reduction in pain experienced by pediatric patients during IV cannulation. While prior studies have not demonstrated this level of efficacy, this study demonstrated an alternative to topical anesthetic creams that have been used in the past. One noted limitation of the study was related to the discomfort experienced when administering the vapocoolant spray. Since the IV was inserted within seconds of the spray, it was difficult to assess discomfort from the spray versus the IV itself.



One of the greatest limitations of the topical local anesthetic creams used to minimize discomfort for IV insertion, is the fact that the creams have to be applied for a significant period of time prior to the cannulation. There appears to be great potential regarding this vapocoolant spray, not only related to its immediate onset, but also because it is nontoxic and ozone friendly. And very importantly, it is reported to be less expensive than the topical creams. With the focus always on patient safety, there are times that having an IV a priori and NOT performing inhalation inductions on pediatric patients, is most appropriate. If this can be done efficiently, safely, in a timely manner, and not cause distress to the child, it appears to lend itself to enhancing quality and safety in anesthesia for this particular patient population.


Mary A. Golinski, PhD, CRNA


Note: Topical spray of this type should not exceed 10 seconds of spray or administration time, as severe local hypothermia may cause the death of local skin cells.

Editor’s Note: Pain Ease spray (1,1,1,3,3-Pentafluoropropane and 1,1,1,2-Tetrafluoroethane) is available from:

Gebauer Company

4444 East 153rd Street

Cleveland, OH 44128


Phone: (216) 581-3030

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008


Puhringer F, Rex C, Sielenkamper A, Claudius C, Larsen P, Prins M, Eikermann M, Dhuenl-Brady K



Reversal of Profound high dose rocuronium induced neuromuscular blockade by sugammadex at two different time points

Anesthesiology 2008;109:188-97

Puhringer F, Rex C, Sielenkamper A, Claudius C, Larsen P, Prins M, Eikermann M, Dhuenl-Brady K


Purpose            The purpose of this study was to determine the effectiveness and safety of sugammadex in reversing high doses of rocuronium.

Background            Sugammadex is a modified γ-cyclodextrin agent designed to bind and reverse rocuronium induced neuromuscular blockade. A previous study indicated that sugammadex fully reversed within two minutes, a dose of rocuronium (0.6mg/kg) given three minutes prior to the sugammadex. However, it did not evaluate the dose response relationship of sugammadex. This study was designed to evaluate the effectiveness and safety of sugammadex used to reverse high doses of rocuronium.

It has been recommended that 1.0-1.2 mg/kg of rocuronium could be used as an alternative to succinylcholine for rapid sequence intubations. Without an effective reversal agent, doses this high might produce unacceptably prolonged paralysis.

Methodology            This was a multicenter, randomized, blinded, placebo controlled, parallel, dose finding, phase II trial. The study planned for 176 patients who were older than 18 years and were classified as physical status I-III. The surgical procedures chosen were expected to last at least 120 minutes in the supine position. Patients were excluded that had difficult airways, a family history of malignant hyperthermia, neuromuscular disorders; those who were pregnant, breast feeding, using contraceptives, or were receiving drugs that might interfere with neuromuscular blockade.

Anesthesia was induced with an opioid and propofol. Neuromuscular function of the adductor pollicis was monitored using “train-of-four” response (TOF). After obtaining a baseline TOF, the patient received a randomized rocuronium dose of either 1.0 or 1.2 mg/kg. Each patient was randomized to receive a dose of sugammadex or a placebo at either 3 or 15 minutes. The dose of sugammadex was also randomized at 2, 4, 8, 12, or 16 mg/kg. The patients TOF response was monitored to full recovery. Full recovery was determined to be a TOF ratio of 0.9 for at least 30 minutes after the administration of sugammadex. All patients were monitored for evidence of residual or reoccurrence of blockade for an additional 60 minutes. A safety assessment was performed by a physician at 10-24 hours and again at 7 days after the procedure. The assessing physician was blinded to the drugs administered. The assessment included evaluation of physical changes, intraoperative and post operative electrocardiogram assessment, and any adverse events temporally associated with the administration of sugammadex or placebo.

The primary variable was time from administration of sugammadex or placebo to full recovery of TOF and the secondary variable was time from administration of sugammadex or placebo to a TOF ratio of 0.7. Additional variables included the occurrence of residual blockade (final TOF less than 0.9) or reoccurrence of blockade (decrease in TOF to less than 0.8 after full recovery).

Sample size and expected recovery times were based on previous sugammadex studies. Analysis of each dose group and time group was performed for the primary and secondary variable. The safety results were included as descriptive data.


Efficacy: A total of 173 patients were included in the study after exclusions. There were no significant differences in patient characteristics between the randomized groups in relationship to age, weight, height, sex or physical status. All doses of sugammadex showed significant reductions in time to recovery from neuromuscular blockade along with a clear dose response relationship compared to placebo. The 12 and 16 mg/kg doses of sugammadex given 3 minutes after rocuronium resulted in full recovery at 1.9 minutes or less compared to 111 minutes or higher for the placebo. The time to recovery following 4, 8, 12, and 16 mg/kg sugammadex doses were consistent for both 1.0 and 1.2 mg/kg rocuronium.

The 2 mg/kg dose of sugammadex showed more rapid recovery when administered 15 minutes after rocuronium compared to 3 minutes after rocuronium and more rapid recovery after the 1.0 mg/kg dose of rocuronium compared to the 1.2 mg/kg dose. For doses of sugammadex at 8 mg/kg or above, when administered 3 minutes after rocuronium there was a 95% confidence that reversal would occur within 4 minutes for both rocuronium doses, and within 3.2 minutes for reversal at the 15 minute time point.

Sugammadex 8 mg/kg had a median reversal time of 3.6 minutes or less for rocuronium 1.0 mg/kg and 1.2mg/kg while sugammadex 16 mg/kg had a median reversal time of 1.3 and 1.6 minutes for rocuronium 1.0 mg/kg and 1.2 mg/kg respectively. A predictable dose dependent time to recovery curve was established for sugammadex doses at 4 mg/kg and greater.

Safety: There was no evidence of residual or reoccurrence of blockade. Adverse events (AE) were reported in 62% of patients receiving sugammadex and 81% of patients receiving placebo. The majority of AEs after the administration of sugammadex were signs of insufficient anesthetic depth and not related directly to sugammadex. The most common drug related AE was nausea and vomiting (12 patients). There were 9 patients who had prolonged QTc intervals which occurred after the administration of sugammadex. Only one of those cases was considered to be possibly related to sugammadex (4 mg/kg). There were 2 cases of hypertension reported after the administration of sugammadex.

Conclusion            The study concluded that rapid and dose dependent reversal of high dose rocuronium was possible with sugammadex. The speed of recovery was dose dependent and predictable. With the higher doses of rocuronium administered in this study, a sugammadex dose of 8 mg/kg or higher was required to get consistent recovery times. This study suggested that it was possible for rocuronium 1.2 mg/kg, along with early reversal using sugammadex 16 mg/kg, to potentially replace succinylcholine 1 mg/kg for onset and recovery from neuromuscular blockade. There was no evidence of residual or reoccurrence of blockade. Side effects from sugammadex were generally mild. A prolongation of the QTc interval was possible but probably rarely significant.



I am excited about the prospects of sugammadex. It is being extensively studied and has so far been found to be effective and safe in reversing rocuronium at various doses. What I am particularly interested in is the possibility of replacing succinylcholine with the combination of rocuronium and sugammadex. It is apparent that more study is needed to determine if this combination is truly comparable to succinylcholine time of onset and recovery, but if it is true we will hopefully be able to replace succinylcholine with a combination of drugs that would have fewer potential side effects.

In my practice, I anesthetize a number of patients for relatively short procedures. Once mivicurium was taken off the market, I have struggled with the use of neuromuscular blocking agents. Mivicurium was far from perfect, but it fit a niche that met my needs quite well. No other drug really fills that void. Rocuronium is an excellent drug and having an agent that will predictably reverse it without significant side effects will be a substantial benefit to many practices.

Overall I found this study to be very well done. As far as I am concerned, it addressed the most important points which were effectiveness, dose response, and safety. I found it very interesting that the most frequent adverse effect found was that when rocuronium was reversed the provider recognized his patient was inadequately anesthetized. I have always told my students that we must not replace adequate anesthesia with paralysis.


Steven R Wooden, MS, CRNA


© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Cesanek P, Schwann N, Wilson E, Urffer S, Maksimik C, Nabhan S, et al

The effect of beta-blocker dosing strategy on regulation of perioperative heart rate and clinical outcomes in patients undergoing vascular surgery:  a randomized comparison


Ann Vasc Surg 2008 in press

Cesanek P, Schwann N, Wilson E, Urffer S, Maksimik C, Nabhan S, et al



Purpose            The optimal course of therapy for beta-blockers administered during the peri-operative period (operationally defined as dosing for a target heart rate) remains ill-defined and somewhat controversial. The purpose of this study was to assess whether titrating beta-blockade to heart rate (HR) achieved clinical outcomes with minimal complications compared with a fixed dosing regimen.

Background            The incidence of peri-operative cardiovascular morbidity and mortality remains high for those undergoing vascular surgery who are at high risk of cardiovascular complications. It is thought that the surge of catecholamines causes hypertension and tachycardia that can be detrimental. Beta-blockade attenuates the ß agonist effects of catecholamines and current data supports their use in the peri-operative period to prevent cardiovascular complications in those at greatest risk. Studies to date however, are inconclusive in answering the questions: what should the target heart rate be, what is the optimal dosing strategy, and what are appropriate dosing intervals? According to American College of Cardiology / American heart Association published guidelines, a better definition of ‘target heart rate’ is needed. Data supports maintaining a heart rate <80 bpm during the anesthetic. The relationship between maintaining a normal sinus rhythm and favorable intra- and post-operative outcomes is positive. Evidence is lacking to suggest the optimal strategy for titrating beta-blockade to an optimal heart rate.

Methodology            This study was a randomized comparison of two treatment regimens for peri-operative beta-blockade in high risk patients undergoing vascular surgery procedures. The Fixed Dose Group received a fixed-dosing schedule of beta-blockade throughout the peri-operative period; dosing and dosing intervals were chosen by either the patient’s primary care provider or vascular surgeon. The Titrated Dose Group received beta-blockade titrated to their HR. A clearly delineated beta-blocker dosing regimen was followed based upon the patient’s HR at a specific time during the intra- and post-operative period. In the Fixed Dose Group, there were no uniform beta-blocker dosing orders and the administration of the drug was at the discretion of the treating physician. If the patient was randomized to be in the Titrated Dose Group they received up to 2 doses of IV metoprolol at 5 minute increments just prior to the start of anesthesia, according to the following:

  • HR >80 bpm, 5 mg IV metoprolol
  • HR 60-80 bpm, 2.5mg of IV metoprolol
  • HR <60 bpm or systolic blood pressure < 100mmHg, dose held

The same dosing regimen was repeated at time zero (arrival to PACU) and at 6, 12, 18, and 24 hours postoperatively.

The study was stopped 24 hours postoperatively. All patients were monitored throughout their hospital stay. Follow up evaluations were made by phone 30 days post discharge to inquire about any adverse events. The primary end points noted during the peri-operative period included:  death, MI, and CHF. Heart rate end points included the mean HR at each dosing time period, the mean HR for the entire dosing period, the mean HR change from baseline, and the percentage of HR measurements greater than 80 and 100 bpm. Primary safety end points assessed included any serious adverse events, such as death, related to beta-blocker administration, hypotension requiring vasopressor support, and bradycardia requiring a pacemaker. A sample size was calculated, with a power of .90, to detect a mean HR difference of 10 bpm, with an assumption that the mean HR in the fixed dose group would be 80 bpm. Data analysis was performed using unpaired t-tests for comparison of heart rates, and chi-square analysis to compare nominal data and safety outcomes. A P<0.05 was considered statistically significant.

Result            Sixty-four patients were enrolled in the study; 33 in the Fixed Dose Group and 31 in the Titrated Dose Group. Demographic data was comparable between the groups. The following vascular procedures were performed:

  • Peripheral bypass 40.6%
  • Aortic endovascular 26.6%
  • Abdominal aortic aneurysm repair 25%
  • Lower extremity limb amputation 4.7%
  • AV fistula creation 3.1%

There were no significant differences in the total amount of metoprolol received between the two groups. No deaths occurred during the hospitalization or the 30 day follow up period. Mean heart rates were not significantly different between groups at any time except at admission to the PACU where they were 70 bpm in the Fixed Dose Group vs. 58 bpm in the Titrated Dose Group, P=0.012). Heart rates >80 bpm occurred more frequently in the fixed dose group at all time points and for the entire study time period (P<0.001). Heart rates >100 bpm did not occur in the Titrated Dose Group. Heart rates > 100 bpm did occur 6% of the time in the fixed dose group (P=0.005). The Titrated Dose Group showed a significant reduction in total heart rate change, compared with baseline (P=0.034). There were no significant beta-blocker associated adverse events noted in either group.

Conclusion            Perioperative beta-blockade titrated to HR was associated with better control of HR without a greater amount of drug-related complications. Perioperative beta-blockade requires careful identification of patients who will benefit and a dosing regimen that permits titration of therapy based on the patient’s heart rate.



Controversy remains surrounding the administration of perioperative beta-blockers for the prevention of cardiovascular related morbidity and mortality. One key limitation to the study was that it was not statistically powered to evaluate clinical outcomes. Additionally, the physicians caring for those enrolled were not blinded and this may have affected the way care was delivered. It was possible that those randomized to the fixed dose regimen may have been more aggressively treated than they would otherwise have been.

While this study is not without its limitations, it is the type of study we need to fill the gaps in our knowledge base. It appears that our largest gaps are related to dosing, parameters to use for dosing strategy, and whether or not those at intermediate risk may benefit from beta-blocker therapy. There is clearly more work to be done regarding dosing strategy; this study will contribute to the evidence we have to date.


American college of Cardiology / American Heart Association guidelines for perioperative ß-blockade are available on the American College of Cardiology web site at:

Go to “Quality and Science” and look for “Clinical Statements.” Under Clinical Statements look for “Beta-Blockers/Adrenergic Beta-Antagonist.” In that category look for “Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery: ACC/AHA 2007 Guidelines on.” Both a full text article and an executive summary of the most up-to-date guidelines are available. In the full text document look under section 7.2 “Perioperative Medical Therapy” for perioperative ß-blockade recommendations.


Mary A. Golinski, PhD, CRNA



Editors Note: Four previous abstracts and comments on perioperative ß-blocker therapy have been published in previous issues of Anesthesia Abstracts. I draw the reader’s attention to another larger study that, like this one, supports titrating perioperative ß-blockade to the patient’s heart rate in order to achieve the benefits while limiting the adverse effects of perioperative ß-blockade.

The effect of metoprolol on perioperative outcome in coronary patients undergoing nonvascular abdominal surgery. (J Clin Anesth 2008;20:284-289) Anesthesia Abstracts, July 2008, page 21.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008

Respiration & Ventilation

Klopfenstein CE, Schiffer E, Pastor CM, et. al.


Laparoscopic colon surgery : unreliability of end-tidal CO2 monitoring

Acta Anaesthesiol Scand 2008;52:700-707

Klopfenstein CE, Schiffer E, Pastor CM, et. al.


Purpose            The purpose of this study was to describe the PaCO2–PetCO2 difference over time in healthy patients undergoing laparoscopic surgery in the Trendellenburg position to ascertain whether or not end tidal CO2 monitoring accurately reflected intraoperative PaCO2.

Background            Laparoscopic techniques are being applied to more and more surgical procedures. Some laparoscopic procedures commonly require anesthesia and surgery lasting several hours. Laparoscopy and pneumoperitoneum, Trendellenburg position, and general anesthesia are associated with respiratory and hemodynamic alterations. Previous studies have shown that the gradient between PaCO2 and PetCO2 may increase over time. A variety of factors negatively affect the relationship between end tidal CO2 and actual arterial CO2. Maintenance of general anesthesia with isoflurane, for example, has been associated with greater end tidal to arterial CO2 gradients than maintenance with propofol. Trendellenburg position results in decreased lung volumes and increased dead space during general anesthesia.

Methodology            This observational study included 40 healthy, ASA physical status I or II, non-smoking patients, free of cardiovascular disease, who were 65 years old or less undergoing laparoscopic colon surgery.

All patients received 7.5 mg midazolam PO and 10 mL/kg lactated Ringers IV preoperatively. General anesthesia was induced with 2 µg/kg fentanyl and 4-5 mg/kg thiopental. Intubation was facilitated with 1 mg/kg succinylcholine and patients were paralyzed intraoperatively with rocuronium. Anesthesia was maintained with desflurane in oxygen and air with an FIO2 of 40% and additional fentanyl as needed to maintain hemodynamic stability. A forced air warming blanket was applied to all patients. Pressure-controlled ventilation was used in all patients with an inspiratory pressure limit of 30 cm H2O. Respiratory rate was adjusted to control minute ventilation and produce a PetCO2 of 30 to 41 torr throughout the case. Arterial blood gasses were measured at 30 minute intervals.

Pneumoperitoneum was produced by insufflation with CO2 to 12 – 14 torr. Twenty degrees Trendellenburg position was maintained throughout the case.

Result            Data from 25 women and 13 men were analyzed. Two patients were excluded from analysis because they were converted to an open procedure. Over six hours, an increase in minute ventilation from about 6 L/min to almost 12 L/min was required to maintain a stable end tidal CO2.

Overall, end tidal CO2 underestimated actual arterial CO2. Thirty-one percent (31%) of all end tidal CO2 measurements were more than 7.5 torr lower than actual arterial CO2. Only between 29% and 77% of the variability in actual arterial CO2 was explainable by changes in end tidal CO2 (R2 was between 0.29 and 0.77). The correlation between end tidal CO2 and arterial CO2 was both inconsistent and became less reliable over time. After 4.5 hours there was no correlation between end tidal CO2 and arterial CO2.

Conclusion            End tidal CO2 does not accurately represent actual arterial CO2 during prolonged laparoscopic surgery in healthy patients in moderate Trendellenburg position. The disconnect between arterial CO2 and end tidal CO2 became greater the longer the duration of the case. This may have been due to variably increased dead space ventilation due to ventilation-perfusion mismatch.



End tidal CO2 monitoring is highly valuable during general anesthesia. Not only does it provide evidence that the endotracheal tube is in the trachea, it allows us to customize ventilation to the patient and current conditions. The old rule of setting the ventilator to “a tidal volume of 10 mL/kg at a rate of 10” produces too much or too little ventilation far more often than it produces a PaCO2 within the normal range. End tidal CO2 monitoring allows us to appropriately reduce ventilation when a patient is 36.0°C and thus producing less CO2. It allows us to increase ventilation when CO2 is being absorbed from the abdomen during a lap chole just enough to maintain a normal arterial CO2 level. We have grown used to end tidal CO2 values being about 3 to 5 torr below actual arterial CO2 values and we have grown used to that relationship being quite reliable. We have grown to depend upon it. But we grew to expect this level of accuracy in patients in a level supine position undergoing open abdominal or extremity procedures. Now we are doing procedures that are quite different than the procedures we did when we learned to trust end tidal CO2 monitoring.

A number of factors can increase dead space ventilation (ventilation without perfusion) during general anesthesia. Areas of the lung that aren’t being perfused receive no CO2 from the metabolically active tissues of the body. When these non-perfused areas are ventilated, the exhaled gas from them contains no CO2 and that CO2 free gas dilutes the exhaled CO2 coming from all other areas of the lung. Dead space ventilation rapidly reduces end tidal CO2 values and increases the gradient between end tidal CO2 and actual arterial CO2.

I first began thinking about this problem when I started doing a lot of laparoscopic robot prostatectomies. These cases take between two and three hours during which the surgeon has a magnified stereoscopic view and manipulates instruments by remote control. During that entire time the patient has a CO2 pneumoperitoneum and the OR table is in all the Trendellenburg position it can produce, almost standing the patient on their head. I had developed an anesthetic technique that used very little opioid while still providing good postoperative analgesia. Because of the magnified view, the surgeon was very sensitive to even small movements of the diaphragm due to patient respiratory efforts. Near the end of these cases, while sewing the anastomosis between the urethra and the bladder the surgeon would often complain that there was movement in his field due to respiratory effort. I thought I had prevented this possibility with relative over ventilation and the end tidal CO2 was always in the low 30s. Then I ran across and read an article reporting markedly increased end tidal to arterial CO2 gradients in elderly patients undergoing laparoscopy in extreme Trendellenburg position. (1) That would mean that my end tidal CO2 value of 30 may have correlated to an arterial CO2 of 40 torr or greater. As a result, I began increasing ventilation to achieve an end tidal CO2 of 26 or 27 during the anastomosis. This change eliminated almost all problems with patient respiratory efforts and the surgeon was able to work in a motionless field.

The two articles I have read about the reduced accuracy of end tidal CO2 values during laparoscopic procedures in the Trendellenburg position and my own experience convince me that this is a real phenomenon. We need to better understand the limitations of end tidal CO2 monitoring and the circumstances in which it gives us “bad numbers.” This study begins that process.


Michael Fiedler, PhD, CRNA

1.            Takahata O, Kunisawa T, Nagashima M et al. Effect of age on pulmonary gas exchange during laparoscopy in the Trendellenburg lithotomy position. Acta Anaesthesiol Scand 2007;51:687-92.

© Copyright 2008 Anesthesia Abstracts · Volume 2 Number 7, August 31, 2008