ISSN NUMBER: 1938-7172
Issue 7.12

Michael A. Fiedler, PhD, CRNA

Contributing Editors:
Penelope S Benedik, PhD, CRNA, RRT
Mary A Golinski, PhD, CRNA
Alfred E Lupien, PhD, CRNA, FAAN
Dennis Spence, PhD, CRNA
Cassy Taylor, DNP, DMP, CRNA
Steven R Wooden, DNP, CRNA

Assistant Editor
Jessica Floyd, BS

A Publication of Lifelong Learning, LLC © Copyright 2013

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

High STOP-BANG score indicates a high probability of obstructive sleep apnea

Br J Anaesth 2012;108:768-75

Chung F, Subramanyam R, Sasaki E, Shapiro C, Sun Y


Purpose The purpose of this study was to determine if a high score on the STOP-BANG questionnaire was associated with a higher probability of more severe Obstructive Sleep Apnea (OSA).


Background OSA is associated with increased perioperative morbidity and mortality. In the United States, OSA affects between 2-26% of the population. Unfortunately, 82% of men and 92% of women with moderate to severe OSA have never been diagnosed. One instrument, the STOP-BANG questionnaire, has been found to be highly sensitive at detecting moderate to severe OSA in patients presenting for surgery who have never been diagnosed. A score of ≥3 indicates the patient is at high risk for OSA. Unfortunately, the false positive rate is fairly high with a cutoff score of ≥3. The investigators hypothesized that a higher score on the STOP-BANG would predict a higher probability of severe OSA.


Methodology Adults who presented for elective surgery were enrolled in the study. All patients completed the STOP-BANG questionnaire then either had a laboratory or portable polysomnography exam completed prior to surgery. Patients were diagnosed with OSA if the Apnea Hypopnea Index (AHI) was >5 and they reported fragmented sleep and daytime sleepiness. AHI scores of 5-15 indicated mild OSA, scores of 15-30 moderate OSA, and scores >30 severe OSA. Statistical analysis was appropriate and a P < 0.05 was considered significant.


Result Over a 2 year period 6,369 surgical patients were screened with the STOP-BANG at two hospitals in Toronto, Canada. Only 12% (N = 746) of these patients had complete STOP-BANG and polysomnography data. These 746 patients were included in the analysis. Forty-nine percent (49%) were male, with a median age of 60 years, neck circumference of 39 cm, and a BMI = 30 kg/m2. Overall, 68% had some degree of OSA. Most commonly, patients had a STOP-BANG score of either 3 (23%) or 4 (22%). Mild OSA was present in 30%, moderate OSA in 21%, and severe OSA in 18% of patients.


As the STOP-BANG score increased, the specificity increased and the sensitivity decreased. [See comment for definitions of sensitivity & specificity.] When the STOP-BANG score was ≥3 the specificity was only 28% for severe OSA. However, with a STOP-BANG score ≥5 the specificity for severe OSA increased to 74% and with a score of ≥7 the specificity was 96%. Thus, at higher scores (≥5) the STOP-BANG had a low false positive rate for identifying patients with severe OSA.


In contrast, the sensitivity in identifying a patient with severe OSA with a STOP-BANG score of ≥3 was 95%; with a score of ≥5 it was 56%, and with a score of ≥7 sensitivity was only 12%. This low sensitivity result for a cut score of ≥7 indicates the STOP-BANG had a high false negative rate. A previous investigation reported the specificity of a cut score ≥3 to be 37% for severe OSA. In that study the sensitivity for a cut score of ≥3 was 100%. Area under the receiver operator characteristic curve analysis indicated that the STOP-BANG questionnaire was best at discriminating those patients with severe OSA (all OSA = 0.65; severe OSA 0.71).


The odds of any degree of OSA (AHI >5) and of severe OSA (AHI > 30) increased as the STOP-BANG score increased (Table 1). For example, a patient with a score of ≥7 was almost 15 times more likely to have severe OSA compared to a patient with a STOP-BANG score <3 (P < 0.05). Likewise, the probability of the patient having any degree of OSA or severe OSA increased as the STOP-BANG score increased (Figure 1). 



Table 1. Odds Ratios of OSA based on STOP-BANG Score


Odds All OSA (AHI>5)

Odds Severe OSA (AHI>30)


3 (1.9-4.7)

3.6 (1.5-8.6)


3.2 (2.0-5)

5.3 (2.3-12.5)


4 (2.4-6.7)

10.4 (4.5-24.3)


4.5 (2.3-8.7)

11.6 (4.6-28.7)

7 or 8

7 (2.8-17.6)

14.9 (5.6-39.6)

Note. Odds ratios & (95% confidence intervals) of OSA for a given STOP-BANG score as compared to a score of <3. AHI = Apnea Hypopnea Index.


Figure 1. Predicted Probability of OSA based on STOP-Bang Score

Figure 1

Note. The red line represents the predicted probability that a patient with a given STOP-BANG score will have OSA of any severity. The blue line represents the predicted probability that a patient will have severe OSA for a given STOP-BANG score. As the scores increased so did the probability of having any degree of OSA (AHI >5) or severe OSA (AHI >30). The lower probabilities for severe OSA as compared to any degree of OSA are probably due to the lower overall incidence of severe OSA in this surgical population.


Conclusion In this study, the odds of having any degree of OSA (AHI >5), especially severe OSA (AHI >30), increased as the STOP-BANG score increased. Providers should consider using the STOP-BANG to help stratify patients with unrecognized OSA who need further preoperative diagnosis and treatment. The STOP-BANG can also help identify patients who may need perioperative precautions implemented to prevent complications such as preparation for difficult intubation, using short-acting anesthetics, adequate neuromuscular blockade reversal, and use of continuous positive airway therapy (CPAP) and postoperative monitoring (e.g. SPO2, ETCO2).



The STOP-BANG is one of the simplest preoperative screening tools for OSA. When the instrument was initially developed, a cutoff score of ≥3 was established as indicating the patient was at risk for having OSA. However, it did not tell you about the severity of OSA, which is of clinical import to anesthesia providers. The results suggested that a patient with a higher score on the STOP-BANG had a higher probability of having severe OSA. Before we take the authors conclusion at face value I believe further clarification of their results is in order.


When examining a diagnostic test it is useful to evaluate the sensitivity, specificity, and area under the curve receiver operator characteristic results. Specificity refers to the ability of the test to correctly identify patients without a disease; in this case without severe OSA. So for a STOP-BANG score of ≥7 the specificity was 96%, which indicates that 4 out of 100 patients with this score would be incorrectly identified as having severe OSA (4% false positive rate). However, the sensitivity at a score ≥7 was 12%, suggesting that 88% of patients who had a score of 7 may not be identified as having severe OSA (88% false negative rate). On the surface these results seem confusing.


Sensitivity and Specificity results are relatively crude measures of an instrument’s ability to accurately identify a patient with a disease or condition.1 However, what one typically sees with tests like the STOP-BANG is that as the cutoff score increases, say from ≥3 to ≥7, there will be fewer false positives but more false negatives. In contrast if the cut score is low, then the test is more sensitive but less specific. With a lower score you may identify more patients who may have OSA who really do not have it. To determine the overall accuracy of the STOP-BANG, one needs to examine the area under the curve receiver operator characteristic results, this takes into account the sensitivity and specificity. For severe OSA the area under the curve was 0.71, which is moderately good. This suggests that a patient with severe OSA will have a 71% chance of having a higher STOP-BANG score as compared to a patient without severe OSA diagnosed by polysomnography. So these results support the author’s conclusion that a higher STOP-BANG score is associated with a higher likelihood of severe OSA. I think the instrument is good, but not perfect, in identifying patients with OSA, especially those with severe OSA.


How can we apply these results to practice? The authors of this study recommended using a score of ≥3 in patients with high OSA prevalence such as those having bariatric surgery. At this score the false negative rate is pretty low, so you could be confident in excluding severe OSA in a bariatric patient if their STOP-BANG score was <3. However, the authors recommend using a cut score of ≥5 in a general patient population to reduce the false positive rate. This would allow providers to identify those patients who may have moderate to severe OSA that need further workup and treatment, such as starting CPAP therapy, and implementation of perioperative precautions. To me this seems reasonable and may help reduce perioperative complications secondary to OSA. Readers are referred to the September 2011 single topic issue on OSA in Anesthesia Abstracts.

Dennis Spence, PhD, CRNA

The views expressed in this article are those of the author and do not reflect official policy or position of the Department of the Navy, the Department of Defense, the Uniformed Services University of the Health Sciences, or the United States Government.

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 12, December 30, 2013

Awake fiberoptic or awake video laryngoscopic tracheal intubation in patients with anticipated difficult airway management

Anesthesiology 2012;116:1210–6

Rosenstock CV, Thørgerson B, Afshari A, Christensen AL, Eriksen C, Gätke MR


Purpose The purpose of this study was to determine if intubation time was faster with the awake video laryngoscopy or awake flexible fiberoptic technique in patients with anticipated difficult intubations.


Background The safest technique for management of the anticipated difficult airway is an awake fiberoptic intubation. Unfortunately, it is difficult for some anesthesia providers to obtain and maintain the necessary psychomotor skills to be competent to perform an awake fiberoptic intubation. This lack of experience may lead anesthesia providers to choose to perform an asleep intubation on patients with an anticipated difficult intubation. However, closed claims analysis has documented severe complications such as brain damage and death secondary to failed intubation in patients with anticipated or known history of difficult intubation.


In recent years video laryngoscopes such as the McGrath Series 5 have been found to be useful in difficult intubations. The McGrath scope can improve visualization of the vocal cords by one to two grades when compared to traditional direct laryngoscopy with a Macintosh laryngoscope blade. Recent case reports have described the successful use of the McGrath video laryngoscope during an awake intubation. However, awake intubation with the McGrath video laryngoscope has not been compared in a randomized controlled trial to the gold standard awake fiberoptic technique in patients with anticipated difficult intubations.


Methodology This was a prospective, randomized controlled trial comparing the McGrath video laryngoscope with awake fiberoptic intubation in 88 ASA I-III adult patients with known or suspected difficult airways scheduled for elective general anesthesia. Exclusion criteria included mouth opening less than 15 mm, poor dental status, contraindication to transtracheal injection, or surgeon request for nasal intubation. Awake intubation was performed by one of six anesthesia providers experienced with both airway techniques.


After randomization, patients were given glycopyrrolate 4-5 µg/kg and taken to the operating room. The airway was prepared with local anesthetic in a standard fashion. Patients were sedated with a remifentanil infusion between 0.1-0.15 µg/kg/min. If needed, a remifentanil bolus of 0.75 µg/kg or propofol 10-20 mg was administered. Patients in the McGrath group were intubated in the sniffing position. Positioning of patients in the fiberoptic group was at the discretion of the anesthesia provider. When the first technique failed after three attempts, the airway was secured with the alternative technique. Endotracheal tube placement was confirmed with capnography and bilateral breath sounds.


The primary outcome was time to tracheal intubation starting from the McGrath blade or fiberoptic passing behind the teeth until observation of a capnography curve by an independent observer. Secondary outcomes included number of intubation attempts, number of esophageal intubations, failure of technique, and Cormach-Lehane glottic view. The ease of the technique and patient discomfort with the procedure were evaluated by the anesthesia provider using a 0-100 mm visual analogue scale. Sample size and statistical analysis were appropriate. A P < 0.05 was considered significant. 


Result A total of 84 patients completed the study (McGrath group N = 41, Fiberoptic group N = 43). No significant differences were found between the two groups in demographics, airway examination, or sedation requirements. A total of 42 patients had a history of a previous difficult intubation. In five of these patients there was a history of difficult mask ventilation. The median time to intubation was 80 seconds (range: 33-424) in the fiberoptic group and 62 seconds (20-678) in the McGrath group (P = 0.17). Most patients in both groups where intubated on the first attempt (Fiberoptic group: 79% vs. McGrath group 71%, P = NS). One patient in the fiberoptic group and two patients in the McGrath group required three attempts at intubation (P = NS). In one patient, three failed flexible fiberoptic intubation attempts where followed by a successful McGrath intubation. Oxygen desaturation of <90% occurred in 21% of the fiberoptic group compared to 12% of the McGrath group. However, there was no difference in the duration of desaturation between the two groups. In approximately 50% of patients in both groups, the Cormach-Lehane vocal cord grade view was 1. The anesthesia provider rated the ease of the procedure very high in both groups. Patient discomfort was low and similar between the two groups.


Conclusion When intubation was performed by experienced anesthesia providers, no significant differences were found in time to awake intubation between flexible fiberoptic and McGrath intubation techniques in patients with anticipated difficult airways. Awake McGrath intubation may be a potential alternative to awake flexible fiberoptic intubation.



When I was in training I was always taught to err on the side of caution and to perform an awake fiberoptic intubation if I felt the patient was known or suspected to be a difficult intubation. However, as video laryngoscopy has come to the forefront as a primary or backup device for intubation in the difficult airway, many anesthesia trainees are graduating with little or no experience with awake fiberoptic intubation. In this study the investigators found that anesthesia providers with experience in both techniques could intubate most patients on the first attempt in less than 2 minutes. They concluded that both techniques were similar, and therefore an awake intubation may be a safe alternative to flexible fiberoptic intubation in the known or suspected difficult airway.  There is one major caveat to this conclusion; these results only apply to anesthesia providers with experience performing both techniques in an awake patient. Thus these results may not apply to less experienced anesthesia providers. Additionally, investigators excluded patients with poor mouth opening. In my experience video laryngoscopes are difficult or impossible to use in patients with small mouth openings. Therefore, I think anesthesia providers should take every opportunity to gain experience in performing awake fiberoptic intubations.


Being able to perform an awake intubation requires judgment to know when to perform the procedure, and the necessary psychomotor skills to drive the scope or perform a gentle awake intubation with a video laryngoscope. It also requires skill and patience to ensure adequate topicalization of the airway and appropriate sedation so the procedure can be performed efficiently and safely. It takes time and practice to become proficient in performance of an awake intubation, regardless of the technique. Anesthesia providers should seek out opportunities to perform the procedure with an experienced colleague and consider attending a difficult airway workshop.

Dennis Spence, PhD, CRNA

The views expressed in this article are those of the author and do not reflect official policy or position of the Department of the Navy, the Department of Defense, the Uniformed Services University of the Health Sciences, or the United States Government.

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 12, December 30, 2013

Obstetric Anesthesia
Mallampati class changes during pregnancy, labour, and after delivery: can these be predicted?

Br J Anaesth 2010;104:67-70

Boutonnet M, Faitot V, Salomon L, Keita H


Purpose The purpose of this study was to describe changes in the airway exam (Mallampati class) before, during and after labor, and to identify any predictive factors in worsening Mallampati classification.

Background Mallampati classification is an easy and reliable measure that is useful in predicting potential difficult airways. Several recent investigations suggest the Mallampati class increases during the course of labor; however no study has described the changes that occur between the 8th month of pregnancy and 48 h after delivery.

Methodology A descriptive, repeated measures design was used to evaluate changes in the Mallampati classification at the 8th month of pregnancy (T1), epidural placement (T2), 20 minutes after delivery (T3), and 48 hours after delivery (T4). Mallampati classification was evaluated with parturients in the semi-sitting position with the head in a neutral position, the mouth wide open and no phonation. At T1 all airway evaluations were evaluated by a single investigator and during the last three evaluations during labor by a second investigator. Potential predictive factors of Mallampati classification change included weight gain between T1 and T2, duration of 1st and 2nd stages of labor, and total intravenous fluid administered during labor. Mallampati classes 3 and 4 were compared at each time point with the McNemar test. Logistic regression was used to identify predictors of changes in Mallampati classification between the time points. Data are presented as the mean (SD) or n (%). A P < 0.05 was considered significant.

Result A total of 87 out of 90 parturients completed the study. The mean age was 31.2 years (5.2). The majority of the subjects were Caucasian females (81.6%) with a pre-pregnancy weight of 64.2 kg (13.2 ) and a BMI of 23.9 (6.3). The full breakdown of BMI was as follows: < 20: 17.2%, 20-25: 49.4%, 25-30: 24.1%, and >30 9.2%). Table 1 lists obstetrical data.


Table 1. Obstetrical Data (n = 87)


Mean (SD) or n (%)

Gestational Age (weeks)

39.8 (1.2)

Gestational age at T1 (weeks)

33 (2.2)

Gestational weight gain (kg)

12.3 (5.8)



Diabetes Mellitus



5 (5.7%)

10 (11.5%)

3 (3.4%)

Cervical dilation at T2 (cm)

3.9 (1.6)

Fluids administered (ml)

1864.9 (709)

Duration of 1st stage (min)

409.3 (197.2)

Duration of 2nd stage (min)

17.8 (15)

Delivery type





58 (66.7%)

20 (23%)

9 (10.3%)



A little over a third of the parturients Mallampati classification did not change between the 8th month of pregnancy and 48 hours postpartum (n = 32, 36.8%). Increases in Mallampati classification were noted in a majority of parturients between both T1 and T2 and between T2 and T3; whereas there was a decrease between T3 and T4. The number of parturients with a Mallampati 3 and 4 airway increased significantly between T1 and T2 (10.3% vs. 36.8%, P < 0.0001), between T2 and T3 (36.8% vs. 51.7%, P = 0.0005), between T3 and T4 (51.7% vs. 20.7%, P < 0.0001), and between T4 and T1 (20.7% vs. 10.3%, P = 0.0062) (Figure 1). None of the postulated factors were significant predictors of changes in Mallampati classification.



Figure 1. Comparison of Mallampati Classification at Different Time Points

Figure 1

Note. MP = Mallampati classification. T1: 8 months of pregnancy. T2: prior to epidural placement. T3: 20 minutes after delivery. T4: 48 hours after delivery. Significant increases were found in the incidence of MP3 or MP4 airways between T1 and T2 and T2 and T3 (P <0.001), respectively. The incidence of MP3 and MP4 airways decreased between T3 and T4 (P = 0.0062).


Conclusion This study shows that the Mallampati classification worsens over the course of labor, and does not fully reverse by 48 hours after delivery. These results suggest that anesthesia providers need to continually reevaluate the airway during the course of labor.



A thorough airway evaluation is an essential component to the preoperative evaluation of any patient, especially a parturient. This was an important study because the findings support that changes do occur with the airway (i.e., Mallampati classification) during the course of labor. The implications for this is that anesthesia providers may want to consider reevaluating the airway periodically during the course of labor. This is especially important in parturients who are at risk for a cesarean section or who may require general anesthesia. Anesthesia providers should also consider these findings when caring for a patient presenting for a postpartum tubal ligation the day after delivery because the airway exam may still not have normalized.

The investigators failed to identify any predictors of changes in the airway exam in their study. Factors such as prolonged 2nd stage labor and large amounts of crystalliods have been postulated to be associated with worsening airway exam. The problem with this study is that the investigators did not present any of the statistical results (beta coefficients and/or odds ratio) for the logistic regression analysis so it is difficult to interpret their results. Additionally, no power analysis was presented, so I suspect the study may have been underpowered to find any significant predictors in a logistic regression analysis.

It is important to point out that the mean BMI was only 23.9 so it is difficult to determine from their data  if obese parturients have a worsening airway exam over the course of labor. Though given obese parturients may be at increased risk for a difficult airway, one could speculate that a worsening of the Mallampati score in this population could be more problematic. This could be even more of an issue in a severe preeclamptic where airway edema has been reported to be an issue.

Dennis Spence, PhD, CRNA

The views expressed in this article are those of the author and do not reflect official policy or position of the Department of the Navy, the Department of Defense, the Uniformed Services University of the Health Sciences, or the United States Government.

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 12, December 30, 2013

Single-dose Etomidate is not associated with increased mortality in ICU patients with sepsis: Analysis of a Large Electronic ICU Database

Crit Care Med. 2013;41:774-783

McPhee L, Badawi O, Gilles F, Lerwick P, Riker, R et al.


Purpose The purpose of this retrospective study was to assess whether or not a single dose of etomidate used for intubation in a large cohort of patients with sepsis was correlated with in-hospital mortality or other clinically important outcomes of care.


Background Etomidate, an hypnotic agent without analgesic properties, is selected frequently to facilitate intubation in the critically ill. Its unique properties minimize untoward hemodynamic effects such as increases in heart rate and decreases in blood pressure. These are typically the reasons why it is selected for use in the high acuity population. As with any drug, it is associated with certain adverse responses. Due to its inhibitory effect on adrenal function (inhibiting the 11-B-hydroxylase enzyme, which converts 11-deoxycortisol to cortisol) which has been reported to last up to 72 hours, much research has been conducted to determine its appropriateness in patients with sepsis or septic shock. The clinical consequences of its adrenal inhibitory effect remain unclear. Two pivotal clinical trials reported higher mortality rates in people diagnosed with sepsis that were intubated with etomidate though a cause and effect relationship has not been established.


Methodology This study was conducted as a retrospective review of an existing dataset (the Philips eICU Research Institute eRI). The dataset contained information taken from the medical records of critically ill adult patients who were hospitalized in an Intensive Care Unit across the United States. Inclusion criteria were as follows:

  1. Adults (>18 years) admitted to the ICU between 2008-2010
  2. Intubated only in the ICU
  3. Comprehensive medication documentation available, including:
    • medical history and physical examination findings
    • demographic profiles
    • other diagnoses (multiple listed when applicable)
    • laboratory data
    • vital signs
  4. Diagnosis of sepsis (see notes)
  5. No evidence of receiving etomidate for a previous intubation


Data was collected from each patient’s clinical ICU flow sheet including the average mean arterial pressure, the lowest systolic blood pressure, and the average heart rate during 24 hours before and after intubation. Vasopressor use was also recorded and defined as the number of days following the intubation that the patient received epinephrine, norepinephrine, phenylephrine, dopamine, dobutamine, and/or vasopressin. The specific medications used for the intubation and whether or not corticoids were used was also recorded.


The primary endpoint for data analysis was in-hospital mortality in septic patients. The variable of interest was not the intubation, but the use of etomidate compared with no etomidate for the intubation. The secondary endpoints used for data analysis were Intensive Care Unit (ICU) mortality, ICU and hospital length-of-stay, days of mechanical ventilation, and days of vasopressor/inotrope use in those with sepsis, severe sepsis and septic shock. The clinical characteristics of those who received etomidate for intubation were compared with those who did not. Records and corresponding data were identified and assessed until the pre-calculated sample size needed for appropriate power was obtained.


Result A total of 2,014 patient records that met inclusion criteria were identified. Of these, 1,102 received etomidate for intubation. Patients who received etomidate were slightly older, had lower pre-intubation blood pressure, and were more likely to have received steroids before intubation (P < 0.05). In the regression model used to assess in-hospital mortality; advanced age (P<0.001), higher APACHE IV score (P < 0.001), African American race (P < 0.05), steroid administration (before or after intubation)(P < 0.001), and the presence of another comorbid diagnoses had a higher risk of death.


The rate of in-hospital mortality was no different between patients who received etomidate for intubation and those who did not (37% vs. 38%, P = 0.77), There was no statistically significant trend between etomidate use and increased mortality. There was no association between etomidate use and any secondary outcome of interest. A separate regression model created to assess in-hospital mortality only in patients with septic shock (N= 650); showed no statistically significant relationship between etomidate use for intubation and in-hospital mortality.


Conclusion In this retrospective cohort study, a single dose of etomidate used for intubation in ICU patients with a diagnosis of sepsis, severe sepsis, or septic shock was not associated with hospital mortality, increased vasopressor or inotrope use, longer duration of mechanical ventilation, or ICU or hospital length-of-stay.



I noted that the researchers did a rigorous job developing the methodology for this study, and analyzing the data knowing a limitation of a retrospective analysis. I very much appreciated the diligence used when defining inclusion criteria; if the patient was NOT intubated in the ICU or a medication interface was absent for example. Those records were not used in the analysis. Retrospective reviews are only as good as the information available. No assumptions were made about missing data and this contributed to the robustness of the study.


Guidelines for the care of septic patients continue to be updated; care is clearly becoming more evidence based. Etomidate does cause cortisol suppression for a short period of time. What is the clinical impact of this fact? This study showed no difference in outcomes when etomidate was used to facilitate intubation in the septic ICU patient. Unless a specific contraindication is noted, I continue to use etomidate in select individuals but I advocate talking with the ICU care team and communicate my plans. Effective communication is a key to safe patient care and our ICU colleagues can provide us with information about patients response to our care.

Mary A Golinski, PhD, CRNA

Septic patients were defined in any one of the following three ways:

  1. Suspected or confirmed infection plus two or more of the following criteria within 24 hours of intubation:  T > 38 C or < 36 C, heart rate > 90 bpm, respiratory rate > 20 breaths per minute or PaCO2 < 32 mm Hg, or WBC > 12,000 or < 4000
  2. A diagnosis of sepsis, severe sepsis, or septic shock within 24 hours after the initiation of mechanical ventilation in the medical record.
  3. ICU admission diagnosis of sepsis

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 12, December 30, 2013

Regional Anesthesia
Comparative effectiveness of regional versus general anesthesia for hip fracture surgery in adults

Anesthesiology 2012;117:72-92

Neuman MD, Silber JH, Elkassabany NM, Ludwig JM, Fleisher LA


Purpose The purpose of this study was to compare mortality rates as well as pulmonary and cardiac complication rates in patients who received regional anesthesia vs. general anesthesia for hip fracture surgery.


Background As the population ages worldwide it is anticipated that the number of elderly patients who experience hip fractures will increase. These elderly patients experience high rates of morbidity, mortality, and disability after hip fracture repair. Approximately 5% die during their hospitalization and 10% die within 30 days due to pulmonary or cardiovascular complications. Unfortunately, there are few interventions available to reduce hip fracture surgery mortality.


The type of anesthesia may influence the outcomes. Some previous investigations have found benefit with regional (i.e., spinal, epidural, or peripheral nerve blockade) versus general anesthesia for hip fracture repair. However, other studies have not. Additionally, it is not known if there are differences in outcomes based on the type of hip fracture; for example, femoral neck vs. intertrochanteric fractures. The authors of this study hypothesized that use of regional anesthesia for hip fracture repair would reduce the probability of inpatient mortality, and major pulmonary and cardiovascular complications when compared to general anesthesia. The authors also hypothesized that these outcomes would vary by type of fracture.


Methodology This was a retrospective study examining adult patients >50 years old or older who underwent hip fracture repair at hospitals in New York State between 2007 and 2008. The investigators used the New York State Inpatient Database to identify records of patients having hip fracture surgery. The investigators recorded data on anesthesia type (general or regional), patient and hospital characteristics, as well as pulmonary and cardiovascular complication rates from the database.


The primary outcome was in-hospital mortality; secondary outcomes included major pulmonary and cardiovascular complications. Complications included: pneumonia, empyema, aspiration, respiratory failure, acute myocardial infarction, congestive heart failure, and cardiac arrest. The investigators used logistic regression techniques to compare their primary and secondary outcomes while controlling for potential patient and hospital confounding variables that might influence the outcomes. A P value < 0.05 was considered significant.


Result A total of 18,158 patients were included in the sample. Of these patients, 29% (5,254) received regional anesthesia and 71% (12,904) received general anesthesia. An unknown number of patients may also have received both general and regional anesthesia. The average age of patients was 83.5 years with a majority of patients being Caucasian (83%) and female (74%). Patients in the regional anesthesia group had a slightly lower rate of pathologic fractures (1.9% vs. 2.7%). In the regional group, 24% had a history of chronic obstructive pulmonary disease compared to 20% in the general group (P < 0.0001). Likewise, 21% in the regional group had a history of dementia compared to 19% in the general group (P = 0.005).


Logistic regression found that patients who received regional anesthesia for hip fracture surgery had a 29% lower adjusted odds of mortality compared to those who received general anesthesia (P = 0.014). Pulmonary complications were 24% lower in the regional anesthesia group compared to the general anesthesia group (P <0.0001). No difference was found in the incidence of cardiovascular complications between the two groups. Of those patients who had intertrochanteric fracture repair, regional anesthesia was associated with a 42% reduction in mortality (P = 0.013), and 37% reduction in any pulmonary complication compared to patients who received general anesthesia as the primary anesthetic (P = 0.001). No differences in mortality or cardiac and pulmonary complications were found between the groups in patients who had femoral neck fracture repair.


Conclusion Regional anesthesia was associated with a lower risk of mortality and pulmonary complications in patients who had hip fracture repair, especially in those with intertrochanteric fractures.



Patients who present with hip fractures tend to be elderly females with multiple comorbidities. Additionally, these patients can be in a significant amount of pain, and this may make it difficult to position them for the procedure while they are still awake. These factors combined make it challenging to administer anesthesia to these patients. Furthermore, anesthesia providers may feel pressured to administer general anesthesia rather than regional anesthesia (i.e., spinal, epidural, or peripheral nerve block) because of the concern that a regional technique may delay the start of the surgery.


It is important to point out this was a retrospective study, so the strength of the evidence is weaker. However, I believe this was a well designed study that used advanced statistical analysis to control for many confounding variables. Therefore, I think the results of this study do provide good evidence supporting improved outcomes with regional anesthesia in elderly patients who require hip fracture repair for intertrochanteric fractures. Also, it is impossible to determine which specific type of regional technique was used (i.e., spinal, epidural, or peripheral nerve block). I suspect the most common technique was a neuraxial technique such as spinal anesthesia. Furthermore, an unknown number of patients in the general anesthesia group may have received a regional technique as well, such as a peripheral nerve block for postoperative pain. According to the authors, this might bias the results and thus underestimate the true effect regional anesthesia had on morbidity and mortality after hip fracture repair.


The greatest benefit found was on pulmonary complications and overall mortality. Overall mortality was most likely decreased secondary to reduced pulmonary complications. Given the known effects of general anesthesia on postoperative lung function, especially in elderly patients with COPD, it seems prudent to consider administering a regional anesthetic for intertrochanteric hip fracture repair.


Anesthesia providers should share these results with their orthopedic and anesthesia colleagues and talk about how regional anesthesia may reduce morbidity and mortality in hip fracture patients. I think careful titration of opioids and sedatives to assist in positioning for the regional technique and procedure would make the procedure more tolerable for the patient. It would also be important to know if there are any contraindications to placing a regional anesthetic (i.e., aortic stenosis, coagulopathy), and how long the surgeon may take to do the repair, as these factors may be the rate limiting step.

Dennis Spence, PhD, CRNA

The views expressed in this article are those of the author and do not reflect official policy or position of the Department of the Navy, the Department of Defense, the Uniformed Services University of the Health Sciences, or the United States Government.

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 12, December 30, 2013

A comparison of epidural analgesia and traditional pain management effects on survival and cancer recurrence after colectomy: a population-based study

Anesthesiology 2012;116:797-806

Cummings KC, Xu F, Cummings LC, Cooper GS


Purpose The purpose of this study was to compare survival and cancer recurrence rates in colectomy patients who did or did not receive perioperative epidural analgesia.


Background Colon cancer is the third leading cause of death in the United States. Recurrence of cancer after colectomy may be affected by immunosuppressive factors such as surgical stress, blood transfusion, anesthetic agents, and opioids. Several previous research studies have reported reduced cancer recurrence and improved survival rates with the use of epidural analgesia. However, results after cancer surgery are mixed. The main hypothesized benefit of epidural analgesia on reducing mortality and cancer recurrence is in the reduction of anesthetic agent exposure and opioid requirements. 


Methodology This was a population-based cohort study using the Surveillance, Epidemiology, and End Results Medicare database to compare cancer recurrence and survival rates after colorectal surgery in patients who had received perioperative epidurals or those who did not. The study included patients aged 66 years or older diagnosed with nonmetastatic colorectal cancer between 1996 and 2005. Patients were excluded if they had stage IV cancer, developed a subsequent second malignancy, a prior diagnosis of cancer, renal disease, history of familial adenomatous polyposis or inflammatory bowel disease, or underwent a laparoscopic procedure. Survival was measured until death or 8 months, whichever came first. Cancer recurrence was measured up to 4 years after the cancer diagnosis or death in patients who survived at least 12 months after their surgery. Medicare databases were examined to determine whether or not the patient received an epidural, the type of surgical procedure, patient comorbidities, and perioperative complications. Perioperative complications examined did not include cardiopulmonary complications, i.e., myocardial infarction or pulmonary embolism.


The primary outcome was survival time in patients with and without the use of epidural anesthesia/analgesia techniques. The secondary outcome was colon cancer recurrence rates. Statistical analysis was appropriate. A P < 0.05 was considered significant.


Result The study included 42,151 patients. Of those, 23% had epidurals at the time of their colon cancer resection. The two groups had often minor, but statistically significant differences in several patient characteristics. Patients who received epidurals were slightly younger (77.8 vs. 78.1 years, P < 0.001), had lower comorbidity scores, where more likely male (44.5% vs. 42%, P <0.001), were more often married (53% vs. 49%), and more likely lived in the Midwest (36% vs. 18%). Patients who received epidurals were more likely to have rectosigmoid or rectal cancers (26% vs. 19%, P < 0.001). The need for a blood transfusion was similar between the two groups (9% vs. 9.4%, P = NS).


The 5-year colon cancer survival rate was 5 percentage points higher in the epidural group compared to the traditional pain management group, 61% vs. 56%. The median survival rate was 7.24 years in the epidural group and 6.09 in the traditional pain management group. After controlling for multiple patient characteristics, epidural use was found to be associated with improved survival (adjusted hazard ratio = 0.91, P <0.001). Patient characteristics that were controlled for included age, gender, race, marital status, education level, median income, year of diagnosis, Charlson comorbidity score, tumor grade, need for blood transfusion, cancer site, and surgical complications. Several other factors were identified as significant predictors of earlier mortality:

  • older age
  • African American race
  • higher comorbidity score
  • unmarried
  • higher stage cancer
  • distal cancer site
  • diagnosis before 2005


The need for a blood transfusion was strongly associated with reduced survival (adjusted hazard rate = 1.34, P <0.001).


At four years, the cancer recurrence rate was 14.3% in the epidural group and 13.8% in the traditional pain management group. After adjusting for patient characteristics, there was no significant difference in the odds of cancer recurrence between the two groups. However, need for a blood transfusion was a significant predictor of cancer recurrence (odds ratio = 1.14, P = 0.01).


Conclusion Epidural use was associated with improved survival, but not reduced cancer recurrence rates, when compared to traditional pain management techniques. It is not known if the association between epidural use and improved survival was causative. Blood transfusions had a clearly negative effect both on cancer recurrence and survival.



This is one of the largest studies examining colon cancer survival and cancer recurrence rates with or without the use of an epidural. Overall, I found it to be a well-designed study, with some interesting findings. However, it is important to point out that it was a retrospective study, and thus causation cannot be determined. Because data was collected from databases, we do not know specifics about how the epidurals were used perioperatively, and if the patients experienced any cardiopulmonary complications. There may also be other unmeasured factors that may have contributed to the findings. More importantly, there were a number of differences between the groups that make me view the results cautiously; although, the investigators did statistically control for many of these differences. Additionally, the effect epidural use had on improving survival was small (hazard ratio = 0.91). Since the investigators were unable to record nonsurgical complications (e.g., myocardial infarction), we do not know what effect nonsurgical complications might have had on their results.


Not surprisingly, the use of an epidural had no effect on the rate of colon cancer recurrence. However, need for a blood transfusion had a negative effect on survival and cancer recurrence. This is consistent with what other investigators have found with allogenic blood transfusions during cancer surgery. The need for a blood transfusion could be a reflection of more advanced disease, sicker patients, and worse perioperative outcomes.


Despite the limitations of these findings, I am still a big proponent of epidural analgesia after open abdominal surgery for colon cancer, especially when a long midline incision is going to be used. However, the risks and benefits of epidural use should be weighed for each patient, and the plan discussed with the surgeon. If you wanted to provide some evidence supporting a long term benefit of epidural use on survival, I might consider using this data, though I would do it cautiously.

Dennis Spence, PhD, CRNA

Hazard Ratio is used to describe whether or not a treatment, in this case epidural use, reduces or increases survival. In this study the hazard ratio was the odds of a patient surviving longer when an epidural was used, however it does not convey any information about how much longer the patient will survive. For example, if the hazard ratio was 2 this implies that twice as many patients had the event, in this case death. In contrast, if the hazard ratio were 0.5 then half as many patients in the epidural group died at any point in time when compared to the traditional pain management group. When examining the hazard ratio, it is important to also look at the median survival rate of the event of interest.


For more information on blood transfusion in anesthesia see the August 2010 issue of Anesthesia Abstracts.

The views expressed in this article are those of the author and do not reflect official policy or position of the Department of the Navy, the Department of Defense, the Uniformed Services University of the Health Sciences, or the United States Government.

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 12, December 30, 2013