ISSN NUMBER: 1938-7172
Issue 7.2 VOLUME 7 | NUMBER 2

Michael A. Fiedler, PhD, CRNA

Contributing Editors:
Penelope S Benedik, PhD, CRNA, RRT
Mary A Golinski, PhD, CRNA
Gerard Hogan Jr., DNSc, CRNA
Alfred E Lupien, PhD, CRNA, FAAN
Lisa Osborne, PhD, CRNA
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

Observational study comparing non-invasive blood pressure measurement at the arm and ankle during caesarean section
Download Text File

Occurrence of rapid eye movement sleep deprivation after surgery under regional anesthesia
Download Text File

A randomized controlled comparison of epidural analgesia and combined spinal-epidural analgesia in a private practice setting: pain scores during first and second stages of labor and at delivery
Download Text File

Comparison of the effects of dexmedetomidine, ketamine, and placebo on emergence agitation after strabismus surgery in children
Download Text File

The hyperglycemic response to major noncardiac surgery and the added effect of steroid administration in patients with and without diabetes
Download Text File

The association between nitrous oxide and postoperative mortality and morbidity after noncardiac surgery
Download Text File

Equipment & Technology
Observational study comparing non-invasive blood pressure measurement at the arm and ankle during caesarean section

Anaesthesia 2013;68:461-466

Drake MJP, Hill JS


Purpose The purpose of this study was to assess the accuracy and reliability of Non-Invasive Blood Pressure (NIBP) measurements at the ankle vs. the arm during elective C-section with regional anesthesia.


Background Blood pressure often needs to be measured frequently during C-section with regional anesthesia, but movement in an awake patient can impede these measurements. Aortic compression by the gravid uterus can cause hypotension in the lower extremities that is not evident when BP is measured in the arms. As a result, there may be advantages to measuring BP at the calf rather than in the arms. Systolic BP measured at the distal end of the lower extremity is slightly higher than in the arm due to changes in transmission of the pulse wave over a longer distance. Mean BP, however, remains unchanged. As long as the arms and calf are at the same vertical height compared to the heart, BP measurements would not be altered by a hydrostatic gradient.


Methodology This prospective, observational study included women undergoing elective C-section of a single fetus of at least 37 weeks gestation with regional anesthesia. Blood pressure cuffs were placed on the left arm and left ankle. Appropriately sized cuffs were used based on the size of the extremity to which each cuff was applied. Simultaneous BP measurements were taken at one minute intervals from the injection of spinal anesthesia until five minutes after delivery of the fetus; and then every 2.5 minutes for another 20 minutes. Left table tilt was applied from injection of regional anesthesia until delivery. Only BP data obtained in a supine patient were included. BP data from both NIBP machines was excluded if either cuff failed to produce a BP measurement (due, for example, to extremity movement or the cuff being bumped).


Result Data from 64 women, and 1,986 pairs of BP measurements, were included in the analysis. Average patient age was 35 years. Average gestation was 38.6 weeks. Mean differences between ankle and arm measurements were:

  • 11 mm Hg Systolic
  • -0.5 mm Hg Mean
  • -4 mm Hg Diastolic

The range of differences between individual ankle and arm measurements (95% limits) were:

  • -20 to 43 mm Hg Systolic
  • -21 to 20 mm Hg Mean
  • -25 to 18 mm Hg Diastolic

BP measurements at opposite ends of the range of differences between ankle and arm were as much as 41 mm Hg different for MAP. Differences between ankle and arm BP was smaller after delivery.


Conclusion While average differences between arm and calf BP readings in women undergoing C-section with regional anesthesia were not great, the range of differences makes it impossible to know when the ankle BP is accurate and when it is not. 



Like you, I’m guessing, I’ve been known to occasionally put the NIBP cuff on a supine patient’s calf when the arm was unavailable or inconvenient for some reason. I’ve believed that as long as there was no vertical distance between the arm and the leg, the measured BPs would be the same. (Hydrostatic pressure changes measured BP by 7.8 mm Hg for each 10 cm (3 in) vertical distance between measurement sites.) Apparently I was wrong, at least when it comes to pregnant women. Where did this variability in BP come from; uterine pressure from the gravid uterus, effects of the regional anesthesia, some other cause? The results of this study, which has some authority by the shear number of data points compared, make me wonder about two things. First, I wonder what we’d find if we did this study in non-pregnant patients under general anesthesia. Would the variability in BP measured at the calf still be troublingly large? Second, I have to wonder if we should reexamine our threshold for placing an arterial line if placing the BP cuff on an arm is truly not an option. For years we’ve been using art lines less and less because of the accuracy and dependability of NIBP machines. But, now, we find that this accuracy may depend entirely on where we place the cuff – at least in women undergoing C-section.

Michael A. Fiedler, PhD, CRNA

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 2, February 28, 2013

Occurrence of rapid eye movement sleep deprivation after surgery under regional anesthesia

Anesth Analg 2013;116:939-43

Dette F, Cassel W, Urban F, Zoremba M, Koehler U, Wuff H, Graf J, Steinfeldt T


Purpose The purpose of this pilot study was to determine if rapid eye movement (REM) sleep deprivation occurred in patients after regional anesthesia for total knee arthroplasty.


Background Many patients experience disturbances in their sleep after surgery. Previous research has found surgery under general anesthesia is associated with significant reductions in REM and stage 3 slow-wake non-REM (NREM) sleep and a rebound in the density and duration over subsequent nights. During REM sleep in particular, muscle tone of the throat and neck, as well as the vast majority of all skeletal muscles, is almost completely attenuated, allowing the tongue, soft palate, and oropharynx to relax, and, in the case of obstructive sleep apnea, to impede the flow of air to a degree ranging from light snoring to complete collapse. This leads to hypoxemia and hypercarbia and is associated with increased rates of cardiac events secondary to activation of the sympathetic nervous system. Furthermore, reduced stage 3 NREM and REM can interfere with normal growth patterns, healing, and immune response. REM rebound has been found to be associated with worsening of sleep apnea symptoms, stroke, myocardial infarction, mental confusion, delirium, hemodynamic instability, and wound dehiscence.


Methodology This was a prospective, observational study that examined changes in sleep patterns in 26 patients who underwent total knee arthroplasty with spinal anesthesia plus three days of continuous peripheral femoral and sciatic nerve block. Patients did not receive opioids during the first three days after surgery. Oral opioids were started after removal of the nerve block catheters on postoperative day 3. Metamizole [Editor’s note: NSAID not available in the USA] was administered 4 g/day starting on the day of surgery. Normal patient care on the postoperative ward required no interventions during sleep. Polysomnography was performed on the night prior to surgery and then again on postoperative nights 1 and 5. A P < 0.025 was considered significant.


Result Twelve patients completed the study. Of these, 50% were male; their age range was 46-79 years old with an average age of 61 years; and their weight range was 75-138 kg. Patients received 0.16 mg/kg of morphine equivalents on the third postoperative night. Total sleep time averaged 330 minutes the night prior to surgery. Sleep decreased to 211 minutes on postoperative night 1 and increased to 300 minutes on postoperative night 5. REM sleep significantly decreased between the preoperative night and postoperative night 1 (- 8%, P = 0.02; Figure 1). Between postoperative nights 1 and 5, REM sleep increased significantly (+10%, P = 0.01; Figure 1). However, REM sleep was similar between the preoperative night and postoperative night 5. The median number of oxygen desaturations of >4% (oxygen desaturation index or ODI) decreased on postoperative night 1 then almost doubled on postoperative night 5 compared to the preoperative night. There was a positive correlation between the increase in REM sleep from postoperative nights 1 and 5 and pain scores at rest (r = 0.62, P = 0.01). This suggests that increased pain was associated with REM rebound.



Figure 1. Percent REM Sleep and Oxygen Desaturation Index 4%

Figure 1

Note: There was a significant decrease in percent REM sleep between the preoperative night and postoperative night 1 (P = 0.02), and significantly higher percent REM sleep on postoperative night 5 when compared to postoperative night 1 (P = 0.01).



Conclusion REM sleep significantly decreased then rebounded on postoperative day 5 after regional anesthesia for total knee arthroplasty. These results suggest REM sleep patterns after surgery with regional anesthesia are similar to those that occur after general anesthesia.



Anticipation of surgery is a stressful time for many patients and this may contribute to sleep disturbances in the days leading to surgery. Combined with the stress of surgery, anesthesia, postoperative pain, noisy postoperative wards, and a new environment, it is easy to see why patients may experience alterations in their sleep patterns. We do not often think about the quality of a patients sleep during the perioperative period. However, I would argue it is important to consider because it affects the overall patient experience. Additionally, in patients with known or suspected Obstructive Sleep Apnea disturbances in sleep patterns, specifically REM sleep may lead to exacerbations in obstructive events when REM rebound occurs. This could increase the risk of postoperative complications.


Some of these factors we probably cannot do much about (e.g., anesthetic agent effects on sleep); however, we can improve the patient experience by tailoring multimodal analgesia regimens that provide effective postoperative analgesia while at the same time minimizing risks to OSA patients. As leaders in the perioperative arena we should try to influence policies that ensure patients get a good night of sleep (ensuring quite time and minimize nursing interventions during normal sleep hours). Next time you postop a patient, ask them how they slept.

Dennis Spence, PhD, CRNA

The ODI 4% is the number of oxygen desaturations of >4% per hour of sleep. An ODI 4% of 5 or more is suggestive of obstructive sleep apnea.


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 2, February 28, 2013

Obstetric Anesthesia
A randomized controlled comparison of epidural analgesia and combined spinal-epidural analgesia in a private practice setting: pain scores during first and second stages of labor and at delivery

Anesth Analg 2013;116:636-43

Gambling D, Berkowitz, J, Farrell TR, Pue A, Shay D


Purpose This study sought to compare pain scores during the first and second stages of labor in parturients randomized to receive either continuous lumbar epidural (CLE) or combined-spinal epidural (CSE) analgesia in a private practice maternity hospital. The secondary aims were to examine patient-controlled epidural analgesia (PCEA) use, number of epidural top-up doses, epidural catheter replacements, side effects, and labor outcomes.


Background For 40 years, CLE analgesia has been used to relieve pain during labor. Since the early 1990s combined-spinal epidural analgesia has become popular because it is associated with less motor block and more rapid onset of analgesia. Multiple studies have compared CLE vs. CSE, with some finding overall superior pain relief with CSE, whereas other studies have not. Many of the studies had design flaws and most had trainee initiated neuraxial analgesia. The authors hypothesized that CSE would provide superior analgesia throughout the course of labor when compared to CLE when initiated by experienced anesthesia providers.


Methodology This was a prospective, randomized controlled trial of ASA II-III parturients presenting for uncomplicated term labor and requesting labor analgesia. Exclusion criteria included gestational age < 37 weeks, malpresentation, previous cesarean delivery, multiple gestation, and body mass index >40 kg/m2. Parturients, nurses, and obstetricians where blinded to study group. Parturients were randomized to receive either CLE or CSE (via PCEA). In the CSE group, after loss of resistance with a 17-gauge Touhy needle the anesthesia provider injected 10 mL of 0.125% bupivacaine with fentanyl 2 µg/mL in divided doses through the epidural needle. No test dose was administered. The anesthesia provider then placed a 19-gauge Flextip Plus single orifice epidural catheter.


In the CSE group, after loss of resistance a 26-gauge pencil point needle was placed through the Touhy needle. After confirming cerebrospinal fluid, 2.5 mL of the same epidural solution used for CLE analgesia was injected (total 3.125 mg bupivacaine + fentanyl 5 µg). A PCEA infusion was then initiated with 0.125% bupivacaine with fentanyl 2 µg/mL at 6 mL/hr, PCEA bolus of 5 mL, lockout 5 minutes, and a one hour limit of 26 mL/hr. A P of <0.05 was considered significant for primary and key secondary outcomes, and for other secondary and tertiary outcomes a P < 0.005 was considered significant.


Result There were 398 parturients in the CLE group and 402 in the CSE group. No significant differences were found in patient characteristics, and labor and delivery characteristics. Neuraxial analgesia was initiated at approximately 3.7 cm dilation. The time from analgesia initiation to full cervical dilation was 4.6 ± 3 hours. The duration of the second stage of labor was approximately 74 ± 60 minutes. Spontaneous vaginal deliveries occurred in 73% of parturients, 9.7% required vacuum assistance, 2.5% required forceps, and 15% had a cesarean delivery. Of the cesarean deliveries, only 22% were for nonreassuring fetal status.


Baseline pre-neuraxial analgesia pain scores were similar between the groups (CLE: 7.5 vs. CSE: 7.4, P =NS). The time to complete analgesia was 11 minutes faster in CSE group (P < 0.001; Figure 1). Pain scores one hour after initiation of neuraxial analgesia were significantly lower in the CSE group (CLE: 0.7 vs. CSE: 0.3, P < 0.001). While typical pain scores in the CSE group were significantly lower; the difference was not clinically significant (CLE: 1.9 vs. CSE: 1.4). The percent of patients with no pain after initiation of neuraxial analgesia during stage one was significantly lower in the CSE group (CLE: 31% vs. 42%, P = 0.004). Typical pain scores in the second stage of labor were similar (CLE: 1.9 vs. CSE: 1.7, P = NS). The percent of patients with no pain during the second stage of labor was similar (CLE: 47% vs. 48%, P = NS). Patients in the CSE group required significantly less anesthesia provider administered top-up boluses when compared to the CLE group (CLE: 27% vs. 16%, P = 0.001). In the CLE group, 5% of parturients required more than 1 top-up bolus compared to only 1.2% in the CSE group (P = NS). Eight catheters were replaced in the CLE group and 5 in the CSE group. There were no differences in neonatal outcomes between groups.



Figure 1. Time to Complete Analgesia (Minutes)

Figure 1

Note: In the CLE group after loss of resistance, 10 mL in divided doses of 0.125% bupivacaine + 2 µg/mL fentanyl was administered through the needle. In the CSE group 2.5 mL of the same solution was injected into the intrathecal space (3.125 mg bupivacaine + fentanyl 5 µg). No test dose was administered.



Conclusion In a private maternity hospital, CSE analgesia provided a faster onset of analgesia and better overall pain relief during the first stage of labor when compared to CLE. Parturients who received CSE analgesia required significantly less anesthesia-provider administered top-up boluses.



Students often ask me why I prefer to do CSEs over CLEs for parturients in labor. My response is that I feel they get a faster onset of analgesia, have less hypotension (especially when no test dose is given), and overall I find that CSE provides better analgesia. I also find using CSE with PCEA that I have to provide less top-up boluses, which when working on a busy labor deck equates to less workload. In this study, the time to complete analgesia was 11 minutes faster in the CSE group; this occurred despite the fact that the CLE group received 10 mL of local directly through the epidural needle and before catheter placement. Typically, one would wait 5-10 minutes after the test dose to bolus a CLE catheter. Therefore, I think if they had waited to bolus the CLE catheter that the time to complete analgesia would have been even longer; maybe 20 minutes. Certainly one could argue in this study that the difference in pain scores was not clinically significant in the first stage of labor. However, approximately 10% fewer parturients in the CSE group required an anesthesia provider administered top-up bolus. Additionally, the proportion of parturients with no pain during the first stage of labor was almost 10% higher in the CSE group. Therefore, I believe this very well-designed study supports my clinical practice and argument for why I prefer CSE over CLE analgesia.


One of the controversial things about this study was the practice of injecting 10 mL of the local anesthetic solution through the epidural needle after obtaining loss of resistance (no test dose was given). The authors argue this is a common accepted practice at their institution, and they reported no significant adverse outcomes secondary to this practice. If you think about the dosage administered, 12.5 mg bupivacaine, is probably the average amount that one would administer for a cesarean delivery. While I do not inject directly through the epidural needle, I do not think this practice is outside the standard of care or a terribly risky practice. If the entire local were injected intrathecally you probably would detect it rather early.


Overall, this study demonstrated that neuraxial analgesia (CLE and CSE) provide excellent labor analgesia. The epidural failure rates were extremely low (< 2%) and consistent with what most obstetric anesthesia providers experience. Some anesthesia providers argue that you never truly know if a CSE catheter is working in the first few hours after placement, and therefore CSEs are not as reliable if a crash cesarean delivery is required when compared to a traditional epidural. This study was not designed to examine differences in epidural failure rates for cesarean delivery. However, I think the results provide some evidence demonstrating that CSE does not result in a higher failure rate when compared to CLE at the time of cesarean delivery. In fact, the number of top-up boluses in the CSE group was significantly lower. Therefore, I agree with the authors when they say in a private practice setting that concerns about higher CSE catheter failure rate at the time of cesarean delivery are unfounded.

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 2, February 28, 2013

Pediatric Anesthesia
Comparison of the effects of dexmedetomidine, ketamine, and placebo on emergence agitation after strabismus surgery in children

Can J Anesth 2013;60:385-392

Chen JY, Jia JE, Qin MJ, LI WX


Purpose The purpose of this study was to compare the effects of dexmedetomidine and ketamine on emergence agitation in children undergoing strabismus surgery. A secondary aim was to examine the effects of dexmedetomidine and ketamine on postoperative vomiting compared to placebo.


Background In pediatric patients, emergence agitation has a reported incidence of 10% - 80% after sevoflurane anesthesia. Risk factors include pain, preschool age, ENT or ophthalmic procedures, and rapid emergence secondary to the administration of sevoflurane or desflurane. Some studies suggest sedation and analgesia administration during emergence may reduce the incidence of emergence agitation. Dexmedetomidine and ketamine both have analgesic and sedative properties, and, depending on the dose, may reduce the incidence of emergence agitation. The investigators in this study hypothesized that the incidence of emergence agitation would be significantly less with dexmedetomidine due to its sedative, anxiolytic, and analgesic properties.


Methodology This was a prospective, randomized, double-blind, placebo-controlled study comparing dexmedetomidine to ketamine on the incidence of emergence agitation in 84 pediatric patients, aged 2 years to 7 years, undergoing strabismus surgery. Patients were randomized to receive either dexmedetomidine 1 µg/kg, ketamine 1 mg/kg, or saline after induction of anesthesia with oxygen and sevoflurane. A laryngeal mask airway (LMA) was placed by an anesthesia provider blinded to group assignment. A continuous infusion of dexmedetomidine (1 µg/kg/hr), ketamine (1 mg/kg/hr), or saline was administered during the surgery. Anesthesia was maintained with sevoflurane 3% to 4% with an oxygen/air mixture to maintain SaO2 >95%. Mechanical ventilation was titrated to maintain an end tidal CO2 of 40-45 mm Hg. Fentanyl 1 µg/kg was administered if the patient’s heart rate was ≥20% of baseline during surgery. If patients received fentanyl, they were excluded from the analysis. All patients received topical anesthesia with 0.4% oxybuprocaine drops. At the end of surgery, sevoflurane was turned off and the study drugs were discontinued. The patients were ventilated with 100% oxygen and transferred to the PACU. In the PACU the LMA was removed while subjects were still under deep anesthesia. Patients were discharged from the PACU when they had an Aldrete score of 9 or 10.


A blinded research nurse evaluated the incidence of emergence agitation in the PACU. Emergence agitation scores were evaluated with the Pediatric Anesthesia Emergence Delirium Scale (PAED) (PAED: range 0-20; <10 absence of emergence agitation, ≥10 = emergence agitation, ≥ 15 = severe agitation). The total number of postoperative vomiting episodes were recorded in the PACU and for 24 hours on the ward . Other outcomes included pain scores, incidence of oculocardiac reflex requiring atropine; time to LMA removal, resumption of mental orientation, and time to discharge from the PACU in minutes. Statistical and power analyses were appropriate. A P < 0.05 was considered significant.


Result There were 78 patients who completed the study; 6 were excluded because they received intraoperative fentanyl (saline group n = 4, dexmedetomidine n = 1, ketamine n = 1). The average age was approximately 4 ± 1, weight 17.5 ± 4 kg, and surgical duration 34.5 ± 7.5 minute (P = NS). Time to LMA removal after arrival in the PACU was 1.4 minutes (P = NS). The incidence of the oculocardiac reflex was significantly higher in the saline group (33%), compared to the dexmedetomidine (4%) or ketamine group (0%) (P = 0.006). Time to return of mental orientation and discharge from the recovery room was approximately 10 minutes longer in the dexmedetomidine and ketamine groups than the saline group (P < 0.001; Figure 1).



Figure 1. Time to Discharge from PACU

Figure 1



Dexmedetomidine and ketamine both significantly decreased peak median emergence agitation scores compared to the saline group (dexmedetomidine = 3.5, ketamine = 4, saline = 9, P <0.017). The incidence of emergence agitation (PAED ≥10) and severe agitation (PAED ≥15) were significantly greater in the saline group compared to the dexmedetomidine group, but similar to the ketamine group (P < 0.017; Figure 2). Pain scores were similar in all three groups in the PACU. On the ward, median pain scores were 50% higher in the saline group compared to the dexmedetomidine or ketamine groups (saline = 4, dexmedetomidine = 2, ketamine 2, P < 0.017). The incidence of postoperative vomiting was significantly lower in the dexmedetomidine group (15%) compared to the ketamine (44%) or saline (46%) groups (P = 0.02). 



Figure 2. Incidence of Emergence Agitation

Figure 2


Conclusion Dexmedetomidine and ketamine both appear to decrease emergence agitation after strabismus surgery in pediatric patients. Dexmedetomidine also decreased the incidence of postoperative vomiting compared to saline.



Emergence agitation has a reported incidence ranging from 2% to 80%. For anesthesia providers that take care of a lot of pediatric patients emergence agitation is almost an expected part of the normal emergence process.1 Emergence agitation commonly presents as agitation associated with a confused state and lack of recognition of the surrounding environment. Patients may kick and thrash around, may not make eye contact, and generally appear inconsolable. Anesthesia providers should rule out hypoxemia, hypotension, upper airway obstruction, hypoglycemia, and pain before confirming the presence of emergence agitation.1 The Pediatric Anesthesia Emergence Delirium Scale (PAED) is a useful tool anesthesia providers can use for scoring the severity of emergence agitation.


Numerous medications have been studied and in this study the investigators compared dexmedetomidine to ketamine and placebo. They found dexmedetomidine resulted in a 35% lower incidence of emergence agitation when compared to placebo. The incidence in the ketamine group was 24% lower when compared to the saline group; however this difference was not statistically significant. I suspect the reason for this latter finding is the study may have been underpowered. Unfortunately, both dexmedetomidine and ketamine resulted in an almost 10 minute longer PACU stay. Anesthesia providers should consider how these medications may reduce or prevent emergence agitation, but also consider what effects they may have on discharge times.


The one issue I had with this study was that the investigators chose not to pretreat the children for postoperative nausea and vomiting. Strabismus repair is associated with a high incidence of nausea and vomiting. Therefore, I think anesthesia providers should consider administration of prophylactic antiemetic therapy in pediatric patients undergoing strabismus repair.

Dennis Spence, PhD, CRNA

1. Dahmani S, Mantz J, Veyckemans F. Case Scenario: Severe Emergence Agitation after Myringotomy in a 3-yr-old Child. Anesthesiology 2012;117:399-406.

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 2, February 28, 2013

The hyperglycemic response to major noncardiac surgery and the added effect of steroid administration in patients with and without diabetes

Anesth Analg 2013;116:1116-22

Abdelmalak B, Bonilla A, Yang D, Chowdary A, Gottlieb A


Purpose The purpose of this study was to test the hypothesis that preoperative administration of 8 mg dexamethasone would cause a greater increase in intraoperative hyperglycemia in diabetic patients compared to non-diabetics. The study also described when the greatest rise in blood glucose occurred during major non-cardiac surgical procedures.


Background Elevations in blood glucose occur during major surgical procedures irrespective of whether or not diabetes exists. The term “diabetes of injury” describes the syndrome that occurs when physiologic stress, like that caused by surgery, induces insulin resistance, glucose intolerance, and hyperglycemia. High plasma glucose concentrations are physiologically insulting; creating an inflammatory response and impairing the immune response.


The use of intraoperative steroids is becoming more and more commonplace for reasons such as preventing PONV, minimizing tissue inflammation and edema, and even to reduce “post- operative fatigue.” Steroids do create an insulin resistant and hyperglycemic state; what remains unknown and somewhat controversial about the use of steroids in those with or without diabetes, is the extent to which they affect glucose concentrations during the perioperative period.


Methodology Patients who participated in the original clinical DeLiT trial (Dexamethasone, Light Anesthesia, and tight Glucose Control) were enrolled. Within the DeLiT trial a “conventional glucose group” was identified. Patients in the conventional glucose group were stratified based on the presence or absence of a diagnosis of type 1 or type 2 diabetes. The conventional glucose group received insulin when blood glucose concentrations exceeded 215 mg/dL and adjusted to maintain a target of 180-200 mg/dL. The conventional glucose group was the study population used in this trial. These patients were randomized to receive dexamethasone or placebo 1-2 hours prior to incision.


All patients received standardized general anesthesia for major non-cardiac surgical procedures. Baseline blood glucose concentrations were noted. Intraoperative blood glucose concentrations were measured hourly when no intervention was needed to keep within range, or every 30 minutes if insulin was administered. Two primary outcome variables were measured comparing those with or without a diagnosis of diabetes:

  1. Hyperglycemic surgical stress response (maximum glucose change from preoperative to intraoperative)
  2. Association between dexamethasone and glucose response


Result A total of 185 patients were in the conventional glucose group; 49 had a diagnosis of diabetes and 136 did not. A total of 19 patients with a diagnosis of diabetes (39%) and 8 patients without a diagnosis of diabetes (6%) were given insulin intraoperatively to keep glucose values within the specified target range. Demographic data was similar between those randomized to receive dexamethasone but were not similar between the diabetic and non-diabetic patients. Those with diabetes were more likely to be older, have a higher ASA physical status classification, a greater BMI, use insulin, and have higher baseline glucose values.


Key findings were as follows:

overall, the mean glucose change in all patients was not significant (unadjusted, diabetic and non- diabetic) (range of increase = 45mg/dL to 72 mg/dL) 

adjusted for age, ASA status, BMI, surgery type, dexamethasone use, and insulin use, the mean glucose increase was greater in non-diabetic than diabetic patients; 29 mg/dL mean change, range 13-46 mg/dL (P < 0.001)

dexamethasone 8 mg increased the mean maximal glucose increase in the non-diabetic patients: (86 + 41mg/dL) (P = 0.0012)

there was no significant steroid induced hyperglycemic effect in diabetic patients (63 mg/dL Decadron vs 63 mg/dL placebo)

Timing of the hyperglycemic response:

~ for all patients, mean glucose increased only slightly from baseline to incision; steroid use did not appear to influence this

~ glucose increased substantially from incision to surgery midpoint in non-diabetic patients; this was true whether they received steroids or not 

~ glucose remained elevated from surgery midpoint to emergence in non-diabetic patients who received steroids


Conclusion Blood glucose concentrations increased substantially from preoperative baseline values to the intraoperative period. The largest increase was observed in non-diabetic patients with a fasting glucose concentration >110 mg/dL. This substantial increase in glucose concentration occurred between incision to the midpoint of surgery, specifically in non-diabetic patients who received 8 mg of dexamethasone.



Very interesting findings and not what was expected at all. The increase in glucose concentrations noted during surgery in non-diabetic patients, specifically in those whose preoperative glucose values were >110 mg/dL, called to question whether there are significant numbers of undiagnosed people. The Cleveland Clinic states that approximately 21% of their non-diabetic general surgical patients present with glucose concentrations > 110 mg/dL and have a significant hyperglycemic response to surgical stress. It truly begs the question of whether or not we are paying close enough attention to those who are not clinically diagnosed with diabetes but whose baseline glucose values are higher than what we would call within an acceptable range. And should we be assessing their postoperative outcomes with a keen eye? Additionally, I agree with the authors that maybe we should consider algorithms for intraoperative glucose control based on both absolute values as well as a context sensitive component that considers the time elapsed since incision. And while I find the steroid results a bit perplexing, it also makes me reconsider withholding dexamethasone from diabetic patients as the benefits, at least in this study, appear to outweigh any problems.

Mary A Golinski, PhD, CRNA

To access the DeLiT trial- BMC Anesthesiol 2010;10:11

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 2, February 28, 2013

The association between nitrous oxide and postoperative mortality and morbidity after noncardiac surgery

Anesth Analg 2013;116:1026-33

Turan A, Mascha EJ, You J, Kurz A, Shiba A, Saager L, Sessler DI


Purpose The purpose of this study was to discover any relationship between nitrous oxide use during general anesthesia and 30 day postoperative mortality from any cause. A second goal was to assess the occurrence of complications or death during the inpatient stay.


Background Nitrous oxide (N2O) has been included in an estimated one billion anesthetics over 150 years of clinical use. N2O is known to be emetogenic, though less so than potent inhalation agents. Administering N2O continuously for several days produces toxicity, likely due to inhibition of methionine synthase. Among other effects, methionine synthase inhibition results in elevated plasma homocysteine and, thus, vascular endothelial damage. Myocardial ischemia may be exacerbated, at least in high risk patients. In the original ENIGMA trial (Evaluation of N2O In Gas Mixture for Anesthesia), N2O was associated with a trend towards increased cardiovascular morbidity, though it was statistically insignificant. Nevertheless, strong and convincing evidence that N2O causes cardiovascular injury as used in clinical anesthesia has not yet been produced.


Methodology This was a retrospective analysis of existing data sets covering patients who had noncardiac surgery with general anesthesia between 2005 and 2009 at a major medical center. Patients who received N2O were compared to similar patients, undergoing the same surgical procedure, who did not receive N2O. Patients who had emergency surgery or who were classified as ASA 5 were excluded. In addition to tracking postoperative mortality for 30 days, data on the following inpatient complications were also collected:


  • neurologic
  • cardiac
  • pulmonary / respiratory
  • infection
  • urinary & hemorrhagic complications
  • wound disruption, peripheral vascular complications
  • death


Result At total of 10,747 patients who received N2O were matched with an equal number who did not. Overall, the N2O group had slightly lower ASA physical status scores and slightly fewer comorbidities. Nitrous patients also tended to receive lower concentrations of potent inhalation agents and oxygen (46%) than non-N2O patients (55%).


Nitrous oxide use was associated with a 17% decrease in the odds of major complications or death before hospital discharge (odds ratio 0.83, P<0.001). This was largely due to a 40% reduction in respiratory complications. N2O was similarly associated with a decreased odds of mortality during the first 30 days after hospital discharge (odds ratio 0.67, P=0.02). Neither in hospital nor post-discharge death differed by gender or the year of surgery for patients who had received N2O. Cardiac complications and intraoperative hemodynamics requiring intervention were no different between N2O and non-N2O groups. N2O was associated with between a 17% and 26% reduction in potent inhalation agent use.


Conclusion Use of N2O during general anesthesia for noncardiac procedures was associated with a decrease in 30 day mortality compared to those who did not receive N2O. It was also associated with a significant decrease in the incidence of respiratory complications. These results do not support the omission of N2O from general anesthesia over concerns for cardiovascular complications or death.



Years ago some anesthesia providers stopped using nitrous oxide all together. Others still use it heavily today. Like all drugs we use, there are pros and cons to using N2O. Some are related to the anesthetist’s level of concern about the drug in general; others to the patient population being served. Nitrous is a weak to moderate, though not unimportant, emetogenic. This alone has caused some to significantly limit their use of the drug. But a larger concern currently in the literature is related to methionine synthase inhibition and increased levels of homocysteine. 


It is well accepted that elevated plasma homocysteine damages the vascular endothelium and may disrupt coronary arterial plaques. Homocysteine also contributes to development of atherosclerosis and thrombus formation. Since we know that N2O inactivates methionine synthase resulting in an increase in homocysteine it might seem like a no brainer that N2O is bad for the cardiovascular system. But this “fact” is far from established at this time. First, the methionine synthase inactivated by N2O is replaced by the body in less than a week. Thus, any increase in homocysteine due to N2O administration should be short lived. Second, good quality scientific studies of the effects of N2O on subsequent complications and death have not yet shown that N2O is “bad for you.” The study that probably started the concern over N2O, the original ENIGMA trial, had important methodological problems, leaving it’s conclusions in question. A secondary analysis of the ENGIMA trial showed no difference in postoperative death rates between N2O and no-N2O patients.1 And the study being reviewed here also shows no increase in complications or death with N2O use. Hopefully, the ENIGMA-II trial, underway now, will better answer the N2O question.


Interestingly, one of the things we have learned from the recent investigations of N2O is that it may reduce respiratory complications. While speculative, one thought is that using nitrous necessarily limits the FIO2 resulting in less atelectasis. We also know that N2O is an NMDA receptor antagonist; like magnesium, methadone, and ketamine. NMDA receptors are present in the lungs and may mediate inflammation and oxygen radical induced damage. Thus, NMDA receptor blockade from N2O may be protective to the lungs. Again, speculation; but what is not speculation is that respiratory complications are reduced when N2O is used.


Fortunately, there are lots of ways to conduct a proper anesthetic. One can choose to omit N2O and still provide a perfectly good anesthetic. But, from the evidence I’ve seen so far, there is no reason we should eliminate the use of N2O for all patients over concerns that it will cause cardiovascular disease.

Michael A. Fiedler, PhD, CRNA

1. Leslie K, Myles PS, Chan MTV, Forbes A, Paech MJ, Peyton P, Silbert BS, Williamson E. Nitrous oxide and long-term morbidity and mortality in the ENIGMA trial. Anesth Analg 2011;112:387–93.


For more information about the patient circumstances in which N2O may be contraindicated see the April 2012 issue of Anesthesia Abstracts.

© Copyright 2013 Anesthesia Abstracts · Volume 7 Number 2, February 28, 2013