ISSN NUMBER: 1938-7172
Issue 6.2

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 2012

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

Automatic notifications mediated by anesthesia information management systems reduce the frequency of prolonged gaps in blood pressure documentation

Combined spinal and epidural anaesthesia and maternal intrapartum temperature during vaginal delivery: a randomized clinical trial

A prospective survey of patient-controlled epidural analgesia with bupivacaine and clonidine after total hip replacement: A pre- and postchange comparison with bupivacaine and hydromorphone in 1,000 patients

Propofol wastage in anesthesia

Ketamine as an adjunct to fentanyl improves postoperative analgesia and hastens discharge in chidren following tonsillectomy—a prospective, double-blinded, randomized study

Obstructive sleep apnea and incidence of postoperative delirium after elective knee replacement in the nondemented elderly



Does your group or institution pay for CRNA Continuing Education? 

Anesthesia Abstracts and CRNA Webinars are available at group rates. Both offer AANA approved CE credit. Anesthesia Abstracts and CRNA Webinars are more convenient and less costly than traditional CE meetings. Plus, there is no expense for travel or hotel. You can earn CE anytime.

Call us for more information and to see how much you can save: 800.586.1425.


Automatic notifications mediated by anesthesia information management systems reduce the frequency of prolonged gaps in blood pressure documentation

Anesth Analg 2011;113:356–63

Ehrenfeld JM, Epstein RH, Bader S, Kherterpal S, Sandberg WS


Purpose The purpose of this study was to determine if notifying anesthesia providers in real-time that there had been a gap of ≥10 minutes between successive blood pressure recordings would reduce the incidence of cases with at least 1 blood pressure gap.


Background Standards for anesthetic monitoring require blood pressure evaluation at least every 5 minutes unless under extenuating circumstances. Anesthesia information management systems (AIMS) provide accurate records of physiologic parameters during an anesthetic. However, in the authors’ institution they found a higher incidence of at least one ≥10 minute gap in blood pressure recording with AIMS when compared to paper records. The investigators wanted to determine the incidence, possible causes, and evaluate possible solutions to the problem of gaps in intraoperative blood pressure monitoring.


Methodology This was a multi-center quality assurance project carried out at 3 large academic institutions. The investigators conducted a baseline evaluation of the fraction of cases at each institution with at least one ≥10 minute gap in intraoperative blood pressure monitoring. Records from all patients ≥16 years old who underwent general, regional, or MAC anesthesia whose cases were recorded in an AIMS system where included in the analysis. Cardiac bypass cases were excluded. Hospitals A and B served as intervention facilities and Hospital C as a control. At Hospital A a convenience sample of 500 paper records where reviewed to determine the incidence of gaps in blood pressure monitoring. These records were created at a time when AIMS system was unavailable.


After baseline data were collected Hospitals A and B implemented a notification system in which a pop-up message displayed on the AIMS screen if there was not a recorded blood pressure in the preceding 10 minutes. At Hospital A the anesthesia provider also received an alphanumeric pager notification. Additionally, at Hospital A there was a one time announcement at grand rounds on how the notification system would be used. At Hospital B there was no education or notification of the changes since these types of notifications on the AIMS screen were already in place for similar events. Hospital C served as a control facility with no notification system being implemented. Data was not collected during periods when new anesthesia personnel were arriving.


After an eight month intervention period, the AIMS databases at all three facilities were queried for the fraction of cases with at least one blood pressure gap. Data was compared with the baseline data collection period. A root cause analysis was conducted on a sample of 400 random charts from Hospitals A and B to determine possible causes of BP recording gaps. At Hospital B, after analysis of the results the interval for gap notification was decreased to 6 minutes because of lack of significant change in the incidence of blood pressure gaps ≥10 minutes. At Hospital B there was a quality assurance meeting at grand rounds about the notification and education on the importance of monitoring standards. After a 10 week follow-up period, data was reevaluated at all three facilities for the incidence of gaps of at least 10 minutes in blood pressure monitoring.


Result Review of the 500 paper records at Hospital A revealed not a single event of a 10-minute gap in blood pressure recordings. Prior to implementation of the notification, the incidence of blood pressure monitoring gaps for the 212,706 records reviewed at Hospitals A, B, and C were 1.5%, 2.7% and 7%, respectively.


At Hospital A, there was a significant decrease in the incidence of at least one blood pressure gap ≥10 minutes after implementation of the notification system (P < 0.05). At Hospital B there was initially no significant decrease in the incidence of gaps in blood pressure monitoring post-notification. However, after staff education and a decrease in the notification time from 10 minutes to 6 minutes there was a significant decrease in the incidence in blood pressure gaps of ≥ 10 minutes pre and post-notification (2.7 vs. 1.5, P <0.0001). At Hospital C there was no significant difference in the incidence of blood pressure gaps at either time point.


The majority of blood pressure monitoring gaps at Hospital A occurred after the start of surgery (90%). Root cause analysis of 400 cases at Hospitals A and B revealed that 36% of the causes were indeterminate, while 19% occurred when the BP machine was disconnected for a position change, 19% when the patient was awake waiting for a PACU bed, 10% for blood pressure monitor failure, 4% for communication failure with the AIMS, 4% for the blood pressure being off while placing an arterial line, and 9% for other reasons.


Conclusion Blood pressure monitoring gaps of ≥10 minutes were common with AIMS records. The incidence of monitoring gaps was decreased, but not completely eliminated, with institution of real-time notification of the anesthesia provider. Anesthesia providers need to be aware of the potential for blood pressure gaps, and take steps to minimize their occurrence and document why they occurred.



There are multiple reasons for missing a blood pressure recording, such as stopping the cuff to place an arterial line, surgeon leaning on the cuff, etc. In the past, when a blood pressure was missed, some anesthesia providers may have “interpolated,” or recorded the average of the blood pressures before and after the “missing” blood pressure. With AIMS systems this is not possible. I am not saying this previous practice was right or wrong, but it was possibly what occurred. It also explains why the investigators did not find a single instance of a missing blood pressure on the paper records.


I found it interesting that Hospital C had a 7% incidence of blood pressure gaps as compared to less than 2.5% at the other two institutions. The authors speculate this was because the default timing for BP monitoring was 5 minutes at Hospital C, and closer to 3 minutes at the other two institutions. The authors also pointed out that at Hospitals A and C, the blood pressure cuff turned off if there was a failure to obtain a blood pressure (i.e., surgeon leaning on cuff), and that an obvious warning is not given when this occurred. This highlights the importance of knowing your equipment, and at the same time always being vigilant. If there is a gap, then it is important to document the reason why it occurred. This problem also supports the need for automated notifications when there is a gap in blood pressure monitoring.


This study demonstrated the inherent weakness of AIMS when it comes to anesthesia monitors. Since all vital signs are automatically recorded, anesthesia providers many times need to go in the electronic record and document why there was a missing blood pressure value. Additionally, some monitors do not automatically reset between cases to the default 3-5 minute interval. At my institution you have to manually set the cuff cycle at the start of every new case. On a busy day it can be easy to forget to do this.


If your facility is using an AIMS system, and plans to set up automated notifications, then it is important to educate the staff about the process and to continually track compliance. Without education and ongoing follow-up, you may find the notifications do little to change practice.

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 2012 Anesthesia Abstracts · Volume 6 Number 2, February 29, 2012

Obstetric Anesthesia
Combined spinal and epidural anaesthesia and maternal intrapartum temperature during vaginal delivery: a randomized clinical trial

Br J Anaesth 2011;107:762-8

de Orange FA, Passini R, Amorim MMR, Almeida T, Barros A


Purpose The purpose of this study was to compare the incidence of intrapartum fever in women who received combined spinal-epidural analgesia (CSE group) for labor to women who received only non-pharmacologic methods of analgesia (control group).


Background Epidural analgesia during labor has been reported to be associated with intrapartum maternal fever. The exact mechanism is unclear; however, epidural analgesia may contribute to changes in maternal thermoregulation. The problem with increased maternal temperature secondary to epidural analgesia is that this may result in unnecessary maternal and neonatal evaluation to eliminate the possibility of infection. CSE has become a popular technique for labor analgesia; however, the relationship between CSE use and maternal fever has not been evaluated. The investigators of this study wanted to determine if there was a relationship between CSE and increased maternal intrapartum temperature and fever.


Methodology This was a prospective, randomized, clinical trial of 70 parturients undergoing vaginal delivery to examine the incidence of intrapartum fever. Inclusion criteria included singleton pregnancy, full-term fetus, and cervical dilation of 3-6 cm. High risk pregnancies and women with a preexisting fever or who were on antibiotics were excluded.


Parturients were randomized to receive either Combined Spinal Epidural analgsia or no neuraxial or intravenous analgesia. The CSE group received 2.5 mg of 0.5% hyperbaric bupivacaine with sufentanil 5 µg, followed 30 minutes later with a 5 mL bolus of bupivacaine 0.05% with sufentanil 0.2 µg/mL, then intermittent boluses every 30 minutes as needed. Women in the control group received no neuraxial or intravenous analgesia.


All women were monitored hourly and their temperature, blood pressure, heart rate and respiratory rate were recorded. Temperature was measured in the axilla on all patients. The primary outcome was the incidence of maternal fever between the two groups, which was defined as temperature ≥38°C. Secondary outcomes included maternal and neonatal temperature, cesarean delivery rate, instrumental delivery rate, use of oxytocin, duration of the first and second stage of labor, maternal or neonatal infection, need for maternal or neonatal antibiotic therapy, Apgar scores and umbilical cord blood pH. Sample size calculations and statistical analysis were appropriate. A P < 0.05 was considered significant.


Result A total of 68 subjects completed the study (N = 34 each group). There were no significant differences in demographics between groups. The average age was 22 with 69% being primaparas at an average 39.2 weeks gestation. The instrumental delivery rate was 11.4% in the CSE group and 0% in the control group (P = NS). The cesarean delivery rate was 11.4% in the CSE group and 20.6% in the control group (P = NS). The duration of the first stage of labor was 85 minutes longer in the control group (P = 0.01).


Subjects in the CSE group had a significantly higher incidence of fever (14.3% vs. 0%, P = 0.027). In the CSE group the temperature was significantly higher starting the first hour after CSE placement and continuing until the sixth hour (P < 0.05; Figure 1). There were no cases of chorioamnionitis or signs of maternal infection in either group. No parturient required antibiotic therapy. Apgar scores and cord gases were similar in both groups, and no neonate born to a mother with fever developed neonatal sepsis or required antibiotic therapy.



Figure 1. Comparison of Temperature 

Figure 1

Note. Median (10-90th percentiles).



Conclusion The use of CSE was associated with a significant increase in maternal temperature and fever. However, this increase was not associated with complications in the mother or neonate.



This is the second study recently to evaluate the relationship between neuraxial analgesia and maternal fever / neonatal outcomes. In a retrospective study, Greenwell et al (1) reported that women who received epidural analgesia and had a temperature >38.3ºC (101ºF) had a 2 to 6 fold increase in neonatal adverse events. In contrast, this randomized clinical trial found no adverse events in any neonate. However, it must be pointed out that the current study sample size was too small to evaluate differences in neonatal outcomes in women who received CSE. A much larger, multi-center, clinical trial would be needed to truly determine if neuraxial analgesia is associated with serious neonatal adverse outcomes.


Overall, this was a well-designed clinical trial. It is the first study to examine maternal fever in association with CSE rather than epidural analgesia for labor. The mechanism is most likely similar, and some authors suggest neuraxial analgesia is associated with an inflammatory response with subsequent release of cytokines which contribute to maternal fever. (1) It may be that with CSE the onset of temperature elevation is earlier than with epidural analgesia. The authors of this study report previous investigations have found delayed onset of temperature elevation and fever in women undergoing epidural analgesia for labor.


I would have liked to have seen an epidural analgesia group included in this study because it would have helped determine how soon the temperature elevation occurs with these two techniques (CSE vs. epidural). Also I believe a limitation of this study is that temperatures were taken in the axilla as opposed to orally. Nonetheless, I found this to be a good start investigating this phenomenon … but only a start. At this point we are beginning to accumulate data that neuraxial analgesia during labor is somehow associated with maternal fever. We do not yet know if it is the cause of fever and we don’t know if the increase in maternal temperature is harmful in any way.

Dennis Spence, PhD, CRNA

1. Pediatrics. 2012;129:e447-54. Epub 2012 Jan 30.

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 2012 Anesthesia Abstracts · Volume 6 Number 2, February 29, 2012

A prospective survey of patient-controlled epidural analgesia with bupivacaine and clonidine after total hip replacement: A pre- and postchange comparison with bupivacaine and hydromorphone in 1,000 patients

Anesth Analg 2011;113:1213-17

Liu S, Bae J, Bieltz M, Wukovits B, Ma Y


Purpose The purpose of this study was to evaluate the difference in efficacy and side effects between epidural hydromorphone and epidural clonidine.


Background This prospective survey examined the use of epidural patient controlled analgesia (PCEA) after total hip replacement. A solution that is frequently used for this type of pain management contains bupivacaine and hydromorphone. This solution has been reported to cause a significant amount of nausea and pruritus (15-30%). The source of the side effects is almost certainly hydromorphone. This study replaced the hydromorphone with clonidine in half of the patients and hypothesized that the use of clonidine would provide similar pain relief with fewer side effects.


Methodology The study reviewed two groups. Each group contained 500 patients that were undergoing total hip replacement or hip resurfacing. The hydromorphone group received postoperative epidural analgesia with bupivacaine 0.06% and hydromorphone 10 µg/mL running at 4 mL/hr with a PCEA dose of 4 mL/10 min prn for pain. The clonidine group received postoperative epidural analgesia with bupivacaine 0.06% and clonidine 1 µg/mL running at 4mL/hr with a PCEA dose of 4 mL/10min prn for pain. In both groups, the epidural was discontinued on the first postoperative day and patients were placed on Meloxicam 7.5 to 15 mg qd. Both groups had orders for nalbuphine 5mg IV prn itching and ondansetron 4mg IV prn nausea.


Each patient was evaluated for pain using a verbal pain score and side effects using the following definitions for three postoperative days. The primary variables compared between groups were pain, nausea, pruritus, and hypotension. Nausea was defined as the use of ondansetron while receiving PCEA. Pruritus was defined as the use of nalbuphine while receiving PCEA. Hypotension was defined as a systolic blood pressure <90 mm Hg. Variables were analyzed using a Student’s t test with unequal variance and a Fisher 2-tailed exact test. The study used a P<0.01 to determine significance.


Result Patient characteristics were similar between the two groups. The clonidine group had a lower verbal pain score at rest (2.3 vs 3.7, P=0.001) and a lower incidence of pruritus (1% vs 10%, P=0.01) than the hydromorphone group. The hydromorphone group had a lower incidence of hypotension (11% vs 20%, P=0.001). Rates of nausea were not significantly different between groups. Secondary findings revealed no incidence of sedation or respiratory depression in either group. A majority of physician and non-physician staff caring for the patients considered the clonidine PCEA to be a “worse” choice than the hydormorphone PCEA.


Conclusion Even though the clonidine group had statistically significant improvement over the hydromorphone group with respect to the verbal pain score and pruritus, the clinical appearance was similar. Both groups had relatively low verbal pain scores and a low incidences of pruritus and the hypotension associated with clonidine grup was easily managed.



I thought this was a well done study in a reasonably controlled situation, with the exception of the lack of randomizaiton. The parameters were simple, well controlled, and clinically relevant. Although I was interested in the outcome of this study in order to determine which drug combination might be superior, my main interest was in finding a drug that could replace preservative free morphine for postoperative pain management. The shortage of preservative free morphine and fentanyl has forced all of us to pursue alternatives. Preservative free morphine has been an extremely valuable drug when given by the intrathecal or epidural route and used to supplement pain management for a number of surgical procedures.  Now that preservative free morphine is almost impossible to obtain, I have started to use hydromorphone for both routes. My initial impression of hydromorphone has been positive, but my experience is still limited.  I continue to look for alternatives, and I feel clonidine may play a role in epidural pain management. This study has confirmed that assumption, and I will consider using it in the future.


We are being forced to consider what alternative drugs may be effective when our drugs of choice are no longer available.  I don’t think this drug shortage situation will be improving in the near future so I am always looking for alternatives for all of my current drug plans. 

Steven R Wooden, DNP, CRNA

© Copyright 2012 Anesthesia Abstracts · Volume 6 Number 2, February 29, 2012

Propofol wastage in anesthesia

Anesth Analg 2012;X: published ahead of print

Mankes RF


Purpose The goal of this brief study was to quantify and reduce the volume of propofol wastage in a small surgical site.


Background Two issues were key in this work. First, the fact that drug shortages are rampant and reductions in wastage could potentially alleviate some shortage situations. Second, propofol is environmentally hazardous because wasted drug is non-biodegradable, accumulates in fat, is harmful to aquatic life, and can only be fully destroyed by incineration at > 1,000°C for at least 2 seconds.


Methodology Drug waste bins in 8 operating rooms were collected over 8 consecutive months. The weight of propofol discarded in the bins, including vials and syringes, was recorded as propofol waste. Self-recorded waste data from the PYXIS dispensing system was also collected. Both of these data sets were combined to calculate total propofol waste.


After data collection, the facility removed all 50- and 100 mL vials of propofol so that only 20 mL vials were available. Estimated propofol waste was reassessed by reviewing 153 PYXIS records along with random checks of discard bins and new data were collected.


Result Propofol represented 45% of all wasted drug at this facility and was wasted at a rate of 29.2 mL/day/bin. Overall, 32% of over 70,000 mL of dispensed propofol was wasted. After changing to only 20 mL propofol bottles, propofol waste was reduced to 2.8 mL/day/bin.


Conclusion Drug waste can be reduced by giving providers feedback about usage and by using system measures to make wasting medication more difficult.



On the clinical side, this brief report really hit home for me. Last fall in Texas, we experienced a massive shortage of propofol. The pharmacies at all three local hospitals were unwilling to aseptically prepare 200 mg doses of propofol from the 50 mL and 100 mL bottles. When the 20 mL bottles ran out, the pharmacies stocked 50 mL or 100 mL bottles in the anesthesia drug tray and required that we waste all unadministered drug from that drug tray. Anesthesia providers were not even allowed to aseptically draw up propofol in a sterile syringe with date, time, initial, and expiration date for use within the appropriate time frame. Surely these actions (or really lack of action) contributed to the shortage nor was it a good example of collaboration among health care professionals.


On the environmental side, Mankes has provided detailed information on the “ecotoxicity” of several drugs used in anesthesia service, including propofol. The PBT (persistence, bioaccumulation, and toxicity) for propofol is 6 (range 0 to 9, with 9 the worst). If propofol is discarded into a sewage system and ultimately enters a water table, it is very toxic to aquatic organisms and may cause long term effects in an aquatic environment. Propofol does not appear to biodegrade in water or soil nor does it degrade in anaerobic conditions. It can only be destroyed effectively by incineration at > 1,000°C for at least 2 seconds. Therefore all propofol should be wasted into the red waste containers that are destined for the incinerator and not disposed of down a sink.

Penelope S Benedik, PhD, CRNA, RRT

NOTE: I have appended two interesting articles on the effects of volatile anesthetics on greenhouse gas production for readers who are interested in these issues.


Anesth Analg. 2011;112:213-7 Special article: general anesthetic gases and the global environment.

Anesth Analg. 2010;111:92-8. Global warming potential of inhaled anesthetics: application to clinical use.

© Copyright 2012 Anesthesia Abstracts · Volume 6 Number 2, February 29, 2012

Ketamine as an adjunct to fentanyl improves postoperative analgesia and hastens discharge in chidren following tonsillectomy—a prospective, double-blinded, randomized study

Paediatr Anaesth 2011;21:1009-14

Elshammaa N, Chidambaran V, Housny W, Thomas J, Zhang X, Michael R


Purpose The purpose of this study was two-fold:  first, to determine if the combination of ketamine and fentanyl was efficacious in alleviating pain postoperatively in children who had outpatient tonsillectomy surgeries and second, to determine if ketamine was conducive to a safe and speedy recovery in the same population.


Background Opioids are typically administered in the post-anesthesia care unit (PACU) to control pain following tonsillectomy. Their use, however, is limited by an extensive adverse effect profile which includes drowsiness and respiratory depression. The adverse effect profile is especially problematic in this high risk group. Children who require tonsillectomy are also often diagnosed with sleep apnea or they have exhibited sleep apnea-like symptoms. Lethargy, drowsiness, and respiratory depression can be detrimental during the immediate recovery period. Many different classes of medications, such as non-steroidal anti inflammatory agents, have been studied as potential alternatives for opioids in order to avoid the adverse effects. Most have met with unimpressive levels of success and acceptance.


Ketamine, which non-competitively binds to the N-methyl-D-aspartate receptor, produces both preemptive analgesia and analgesia. Given in varying doses, it appears to prevent pain before one has an exposure to a painful stimulus, such as surgical interventions. Most notably, ketamine does not suppress respirations. Fentanyl is often chosen for analgesia in the PACU for outpatients due to its short duration of action. While numerous studies have been conducted in the adult population assessing the effects of ketamine when used with fentanyl and other opioids, this study focused specifically on post-tonsillectomy pediatric patient. Would the addition of ketamine facilitate a reduction in fentanyl dose, produce effective analgesia, and allow safe and rapid recovery?


Methodology This was a prospective, randomized, double-blinded study. A total of 60 children, aged 2-7 years, ASA I and II, and scheduled for elective tonsillectomy with or without adenoidectomy were enrolled. Randomization assigned the participants to one of four groups as follows:


  • Group F1 Fentanyl 1 µg/kg
  • Group F2 Fentanyl 2 µg/kg
  • Group Ketamine Ketamine 0.5 mg/kg
  • Group Fentanyl-Ketamine Fentanyl 1 µg/kg plus Ketamine 0.5 mg/kg


No child received premedication. Study medications were administered intravenously prior to incision. Induction and maintenance consisted of an inhalation technique with sevoflurane in nitrous oxide and oxygen. Neuromuscular blocking agents were used at the discretion of the provider. All children received dexamethasone and ondansetron and were extubated awake upon case conclusion. Pain was measured by a blinded observer at 4 time points:  arrival to the PACU and at 30, 60, and 90 minutes following PACU arrival. Supplemental analgesia was administered as needed. The need for supplemental analgesia in the PACU was determined by a FLACC score > 5. Those who required supplements were excluded from subsequent measurements.


Result There were no significant demographic differences noted between the groups considering age, weight, and gender. There were no differences between groups in time to awaken / emergence, anesthesia time, incidence of PONV, or emergence agitation/delirium. Only total surgical time and study group a patient was assigned to were predictors of pain upon arrival in the PACU (P = 0.02). Pain scores increased with surgical time.


After adjusting for surgical time, pain scores for study groups Ketamine and Fentanyl-Ketamine were significantly lower at PACU arrival time compared to the F1 group. Supplemental analgesia was required in 5 of 15 children in each of the fentanyl only groups; 2 children in the Ketamine group, and 1 child in the Fentanyl - Ketamine group. Groups Fentanyl-Ketamine and F2 had the shortest stays in the PACU and therefore quicker discharge times.


Conclusion Children receiving ketamine with or without fentanyl prior to surgical incision for tonsillectomy procedures had statistically significant lower pain scores upon arrival to the PACU compared with those who received only fentanyl at doses of 1-2 µg/kg. In addition, those who received ketamine required less supplemental analgesia in the PACU although this was not statistically significant. In this study, ketamine was found to be a valuable adjunct to fentanyl analgesia with limited adverse effects and a resultant shorter PACU length of stay.



While this study did not demonstrate that pain scores in the PACU remained lower for those who received ketamine, I found this data to be clinically significant and absolutely helpful for enhancing quality care in the immediate post-operative period. Typically, tonsillectomy procedures are not lengthy even when technically challenging to the surgeon, yet those who do require the surgery often have the associated diagnosis of sleep apnea and the immediate post-operative period is a critical time. (It is noted in the literature that if a child does have sleep apnea preceding tonsillectomy, it takes approximately 3 weeks post surgery for all symptoms to disappear). These short procedures are: 

  • extremely painful during recovery
  • involve the airway
  • have a potential for post-operative bleeding which may compromise the airway further
  • are performed on pediatric patients who often have associated high risk clinical diagnoses

They indeed place these individuals at greater risk and present challenges for the anesthesia provider. Identifying the most appropriate technique to minimize pain and respiratory depression is always imperative. This is especially true with this surgical population. The multimodal approach described in this study may indeed be safer and more effective than common practice.

Mary A Golinski, PhD, CRNA

NOTE: The FLACC Pain Scoring system (Face, Legs, Activity, Cry, Consolability) was designed for children ages 2 months to 7 years. Similar to the APGAR scoring system, the FLACC assigns a score of 0, 1, or 2 to observable characteristics in each of the 5 areas assessed. Higher scores correlate with greater pain perception. The maximum score is 10.

© Copyright 2012 Anesthesia Abstracts · Volume 6 Number 2, February 29, 2012

Respiration and Ventilation
Obstructive sleep apnea and incidence of postoperative delirium after elective knee replacement in the nondemented elderly

Anesthesiology 2012;116:788-96

Flink BJ, Rivelli SK, Cox EA, White WD, Falcone G, Vail TP, Young CC, Bolognesi ML, Krystal AD, Trzepaca PT, Moon RE Kwatra MM


Purpose The goal of this study was to examine the association between preexisting obstructive sleep apnea (OSA) and postoperative delirium in elderly nondemented patients. The investigators  looked for “molecular markers of postoperative delirium” in a group of elderly patients who underwent elective total knee replacement under both general and regional anesthesia.


Background Postoperative delirium is relatively common in the elderly and has multiple known etiologies. Dementia is not delirium; dementia is a significant and persistent inability to perform simple cognitive tasks. Delirium is a disturbance that develops over a short time and tends to fluctuate throughout the day. To diagnose delirium, one must find its three key characteristics: 1) disturbance of consciousness, 2) change in cognition, and 3) no identifiable physiologic cause. Known causes of delirium include:

  • D: drugs, too many
  • E: electrolyte abnormalities
  • L: (lack of ) drugs, pain
  • I: infection
  • R: reduced sensory input
  • I: intracranial problems
  • U: urinary retention or fecal impaction
  • M: myocardial infarction

When these possibilities are ruled out, postoperative delirium; probably better termed postoperative cognitive decline (POCD) may be the cause of symptoms. Older patients, sicker patients, and those exposed to more extensive surgeries will more commonly develop POCD.


Postoperative delirium is associated with increased morbidity and mortality. The incidence of postoperative delirium after major surgery in an elderly patient with no history of delirium ranges from 5% to 15%. There are limited therapies for postoperative delirium because there is limited understanding of the biology of delirium in elderly patients after surgery.


Methodology This was a prospective cohort study of 106 patients aged 65 years or older without pre-existing delirium, alcohol dependence, psychiatric, or neurologic disorders who underwent elective TKA. Subjects were initially screened by medical record review, interview about activities of daily living, administration of the mini-mental state exam, CAGE questionnaire for alcohol abuse, the Geriatric Depression Scale, the Confusion Assessment Method, and the Delirium Rating Scale Revised-98. These tools were administered again on postoperative day (POD) 2 and 3. Charts were also reviewed for evidence of changes in cognition on POD 2 and 3. Sixteen patients with obstructive sleep apne were identified by either polysomnography or medical record review. Logistic regression was used to examine clinical characteristics which were associated with the presence or absence of postoperative delirium.


Result A total of 106 patients completed the study. There were no significant differences in demographics or type of anesthesia between patients with and without postoperative delirium. The incidence of postoperative delirium in these older patients was 25.5% (27 patients). The postoperative delirium seen in patients was of mild severity. The majority of patients (90%) exhibited delirium on POD 2. There were no differences in delirium based on type of anesthesia, sex, BMI, or preoperative oxygen saturation.


Two preoperative conditions were statistically associated with postoperative delirium, Obstructive Sleep Apnea (P=0.01) and a lower average preoperative hemoglobin (12.8 vs. 13.4, P = 0.05). In those patients with a history of OSA, the incidence of postoperative delirium was 53%, compared to 21% in those without an OSA history. OSA patients were significantly more obese (BMI 36 compared to 30, P=0.005), had more medical comorbidities (4 versus 3, p=0.007), and were slightly younger (70 vs. 74 years, p=0.01). After multivariate analysis, OSA was identified as a risk factor for postoperative delirium with an odds ratio of 4.3 (95% confidence interval 1.2 to 15.8).


Conclusion Obstructive sleep apnea may be a risk factor for the development of postoperative delirium in patients over 65 years undergoing elective total knee replacement.




Comment: Dr. Benedik

A molecular marker is a fragment of DNA that is associated with a certain location within the genome. The authors did not “explore the molecular markers of postoperative delirium” in the results or concluding remarks after introducing it as the purpose of the study which was disappointing. Possibly the manuscript was written as an incidental finding for a larger biological research project. Maybe this oversight is due to the fact that the primary author was a medical student; clearly neither the coauthors nor the editors, edited the introduction of the paper.


What was identified is that Obstructive Sleep Apnea patients may be more likely to develop postoperative delirium than non-OSA patients. Is there a physiologic basis for this? OSA is associated with chronic, nighttime hypoxemia and hypercapnia that is worsened by the effects of anesthesia and postoperative analgesia. OSA is highly associated with neuropsychological sequelae such as more frequent motor vehicle accidents (Odds Ratio 6.3), more on the job injury (Odds Ratio 2 to 3), self-assessed concentration problems and a decreased quality of life. Anemia and low postoperative hemoglobin levels are associated with postoperative delirium. Transient brain deficits in oxygen may also contribute to postoperative delirium (reference below).  Combining a predisposition to subtle abnormalities in the brain with chronic hypoxia and hypercarbia due to both low arterial oxygen content and poor microvascular flow, might you get a nudge into postoperative delirium? It seems plausible, especially if you add postoperative analgesia to the mix. Opioids are known to increase apneic events and suppress the arousal response to hypoxia and hypercarbia.


Unfortunately, much important data that could help us interpret these findings were not reported. This critical data includes the type and dose of intraoperative sedation and intravenous or central axial opioids: drugs that are well-known to adversely affect OSA by causing more profound hypoxemia and hypercapnia. Clearly these data are available on the patient’s medical records, but the authors did not pursue this part of the analysis. Data were not collected or reported on postoperative pulmonary function (i.e. pulse oximetry or arterial blood gases).  Data were not collected on the use of oxygen therapy in these postoperative patients. Many opportunities for a more informative analysis were lost.


Lastly, I must point out that the identified OSA patients did not appear to have been placed on continuous pulse oximetry postoperatively, which is a deviation from the ASA’s 2006 OSA practice guidelines.


Penelope S. Benedik, PhD, CRNA, RRT


Cerebral oxygen desaturation predicts cognitive decline and longer hospital stay after cardiac surgery. Ann Thorac Surg 2009;87:36 – 44.




Comment: Dr. Spence

Obstructive sleep apnea is a common comorbidity which creates many challenges for anesthesia providers during the perioperative period (see single topic issue on OSA in Anesthesia Abstracts, September 2011). However, for nurses on the postoperative wards I would imagine delirium, or in this case postoperative delirium, would create more of a challenge after surgery. It might also be concerning for family members.


In this study the investigators found an association between Obstructive Sleep Apnea and mild postoperative delirium after TKA. It is important to point out that these findings are only preliminary. The finding of an association does not mean causality. The investigators were very clear in pointing out these results were part of a larger study for which the primary aim was to evaluate molecular markers of postoperative delirium. So the study was not powered or designed to really assess causality. A lot of other unmeasured factors such as medications administered perioperatively could influence these results. So a larger randomized trial is needed.


Nonetheless, the investigators’ findings are important because they highlight some issues and areas for future research. The investigators did report they saw a trend in more postoperative delirium in patients who were on home CPAP who did not receive it while in the hospital. I think this highlights an important postoperative management point. If a patient is on home CPAP they should continue that therapy while in the hospital. If they didn’t bring their CPAP machine with them, then it should be ordered and a respiratory therapy technician should set one up, assuming it does not interfere with the surgical recovery. Doing so will help reduce more serious perioperative complications such as hypoxemia, but may also help reduce postoperative delirium. Although future studies will need to confirm the latter.


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.

Penelope S Benedik, PhD, CRNA, RRT Dennis Spence, PhD, CRNA

© Copyright 2012 Anesthesia Abstracts · Volume 6 Number 2, February 29, 2012