Pediatric Sedation Newsletter – January 2001 Departments of Anesthesiology and Pediatrics, Dartmouth Hitchcock Medical Center, Lebanon, NH Editors: Joseph P. Cravero MD (firstname.lastname@example.org) , George T. Blike MD (email@example.com ) Website (http://an.hitchcock.org/PediSedation/) Circulation = 1790 A slight change for this month’s newsletter. We are foregoing literature reviews to update the readers on ongoing discussions concerning ketamine and to review a recent sedation related letter in Pediatrics. We will return with literature and book reviews next time. Thanks to all who have been writing us and updating us on sedation information in your fields! Housekeeping: Ketamine Continued…….: As mentioned in the December newsletter, the November Sedation Newsletter’s editorial “The Semantics of Ketamine” by Drs Green and Kraus elicited a lot of mail. In the interest of continuing a constructive dialogue on this subject (rather than print a confusing series of strongly worded opinion letters) we are awaiting an editorial on the subject that will address the semantics and use of ketamine sedation/analgesia (? anesthesia) from a slightly different point of view. We hope to bring that to you soon. We continue to receive many questions about the issue of ketamine administration and its politics – particularly concerning policy writing for use outside the operating room. For those who are looking for more ketamine information, excellent reviews of the subject exist and many are cited in the references for the ketamine editorial mentioned above (can be found at our website). For a nice review of the drug’s pharmacology and recent applications please see Kohrs R, Durieux ME. Ketamine: Teaching an old drug new tricks. Anesthesia and Analgesia Vol 87(5), Nov 1998, pp 1186-1193. You will find additional papers in the “Ketamine” section of the “literature reviews” partition on our website. The question of NPO status and ketamine also continues to be raised by our readers. Clinical experience tells us that aspiration is a rare event during sedation/analgesia – particularly when ketamine is involved. The need for appropriate NPO intervals remains an area of disagreement. A couple of old papers (we can not find anything very recent) that addressed (specifically) the issue of the laryngeal protection and ketamine anesthesia include: Carson IW, Moore J, Balmer P, et al. Laryngeal competence with ketamine and other drugs. Anesthesiology 38:128-33, 1973 and Yeung MI, Lin RS: Laryngeal reflexes in children under ketamine anaesthesia. Br J Anaesth 44:1089-1092, 1972. These papers confirm that tracheal soiling can occur when contrast is placed in the stomach of patients under ketamine sedation/anesthesia (airway protection is not complete). Similar findings occur with other drugs during general anesthesia however, and the clinical significance of these findings remains uncertain. The multitude of clinical studies that have looked at ketamine sedation in the emergency department have failed to reveal any cases of significant aspiration although rare case reports do exist. To determine current practice concerning NPO status and ketamine sedation we questioned experts from many disciplines about this subject at a sedation conference here at Dartmouth in September 2000. Anesthesiologists tended to be conservative, preferring to provide airway protection when ASA NPO guidelines have not been fulfilled. Our discussions with emergency medicine experts suggest that the decisions concerning airway protection and ketamine sedation in the emergency department are a bit more flexible. They are based on the urgency of a procedure, exact nature of the oral intake, and the timing of last PO intake and a given injury. Finally, our ongoing discussion concerning ketamine sedation/analgesia has yielded many (over twenty) anecdotal reports from anesthesiologists and dentists concerning episodes of laryngospasm during ketamine sedation. We presented one of these cases as a “Close Calls” case last month. Case reports of this problem are actually very rare in peer reviewed journals. Only one case in the emergency medicine literature can be found: Smith JA, Santer IJ. Respiratory arrest following intramuscular ketamine injection in a 4 year old child. Annals of Emergency Medicine. 22(3):613-5, 1993 Other ketamine related papers that address laryngospasm report wildly differing incidences of laryngospasm - ranging from 4 cases in 1,022 sedations with 4mg/kg IM (Green et al. Intramuscular ketamine for pediatric sedation in the emergency department safety profile in 1,022 cases. Annals of Emergency Medicine 31(6):688-97 1998) to 4 incidents in 10 patients given 6mg/kg IM (Zsigmond EK et. al. A new route, jet injection for anesthetic induction in children II, ketamine dose range finding studies. International Journal of Clinical Pharmacology and Therapeuics 34(2):84-8, 1996 Feb) Clearly there must be inconsistency in terms of what is considered “laryngospasm”. The incidence is relatively low and most likely varies with dose of the drug, the use of adjunctive anticholinergics, the route of administration, and the procedure that is being performed. While conclusions are difficult to draw from a review of the literature and the reports sent in to this newsletter, there is undoubtedly a need to define the incidence of laryngospasm with ketamine sedation/analgesia so that future recommendations concerning the use of this highly effective agent can be made with more certainty. We welcome your thoughts and comments on this matter…. Chloral Hydrate and Monitoring: In our October newsletter we reviewed a study by Dr. Charles Cote and coworkers involving a critical incident analysis of pediatric sedation cases (Coté CJ, Notterman DA, Karl HW, Adverse sedation events in pediatrics: a critical incident analysis of contributing factors. Pediatrics. 2000; 105:805-814). Most of the readers of this newsletter have read this study or heard Dr. Cote speak about this work. For those who have not seen the study – the authors compiled a group of over one hundred cases of critical incidents that occurred during pediatric sedations that resulted in serious neurological injury or death in the patient. The data was collected from a federal reporting database and the US Pharmacopia over a 10 year period. After reviewing all the cases, this group of experts was able to agree on a cause for the critical incident in 95 cases. The authors went on analyze the cases in depth. They noted trends in the cases that lead to particularly bad outcomes, and they grouped the errors into categories (i.e. drug overdose, mixing of drugs etc) A brief analysis of the systems involved in the various cases was offered. In the December issue of Pediatrics, a letter to the editor was printed as a response to this study. I present the letter below because I think it is indicative the manner in which some providers view issue of sedation safety. In addition it seems germane to the discussion that took place concerning chloral hydrate in our last issue. We offer our analysis of this letter below: Oranges and Apples: Sedation and Analgesia To the Editor. Coté et al1 have performed a valuable service in their critical analysis of reported adverse sedation events. Some of the pitfalls inherent in this study are pointed out in the accompanying editorial. However, the editorial did not emphasize that once again the authors are combining analysis of patients given either analgesia and sedation they are studying apples and oranges, and evaluating fruit salad. The Coté study evaluated a selected sample, from a biased cohort, which lacked a denominator. In a larger, differently structured study,2 97.7% of children receiving proper management of sedation and analgesia in an emergency room setting had no adverse events. All of the events that did occur in the remaining 2.3% were transient, minor, and easily managed. Even in that study, sedation and analgesia are combined, and yet every physician, anesthesiologist and grandmother knows that sedation and analgesia are different. They are different in the goals sought, they may be different in the medications used, and different in the depth of neurologic alteration desired. They are also probably different in their complications. For years, indeed for generations, chloral hydrate has been used for sedation to provide sleep and thereby lessen anxiety. Chloral hydrate is not an analgesic. In situations where mild sedation, not analgesia, is desirable, chloral hydrate has been used widely and safely. For example, young children are often anxious about electroencephalograms (EEGs), and sleep is an important state of the brain in which to evaluate epileptic activity. Analgesic medications might even invalidate the EEG study. And yet we are no longer permitted to use chloral hydrate for sedation for EEGs without having a nurse (or physician) in attendance for monitoring. Why? Where are the data on the incidence of adverse events after appropriate doses of chloral hydrate for such sedation? Where are the "critical events analyses" where only chloral hydrate or other medications for conscious sedation are used? In Table 1 of his article, Coté cites 3 examples of such adverse events. He cites 1 child who received 6000 mg of chloral hydrate. Why was the child given such a large dose, and what was the adverse event? A second child was "flown up from Mexico for a cardiology procedure and received 60 mg/kg of chloral hydrate. Depression, bradycardia, and arrest followed..." The implication is that the chloral hydrate caused the event. It is unlikely that the child did not receive other medications for the procedure, whatever it was. What was the role of the digoxin? A third child had an adverse event of unstated type, and "the mother had given 2 prescriptions of chloral hydrate at home." What was dose? What was the event? What were its consequences? As a result of Coté's work and that of others, children who require EEGs often get inadequate studies, without sleep. They are, for example, often sent away from Johns Hopkins for studies because it is fiscally unsound for us to subsidize a nurse to monitor each child because the additional costs will not be reimbursed by insurance. Have the recommendations of the Academy resulted in safer studies? Where are the data that chloral hydrate properly prescribed for sleep is an unsafe drug? What is the magnitude of the risk to children? How does it compare with the risks of the procedures being performed at other sites? Does the quality of the studies performed elsewhere and their interpretation factor into the cost-effectiveness of the recommended Analgesia clearly may require more careful monitoring and more vigorous resuscitation than was done in the past. However, to lump sedation and analgesia together and develop a numerator of adverse events without any denominator is to lump together apples and oranges. We need far better data in both the numerators and in the denominators of the risks of sedation, and separately of analgesia. We need to understand the risk/benefit equation of each before we accept the proscriptions that have been forced on us and the families we serve. John M. Freeman Eileen P. G. Vining References: 1. Coté CJ, Notterman DA, Karl HW, Adverse sedation events in pediatrics: a critical incident analysis of contributing factors. Pediatrics. 2000; 105:805-814 2. Pena BMG, Krauss B Adverse events of procedural sedation and analgesia in a pediatric emergency department. Ann Emerg Med. 1999; 34:483-491 Commentary: There are plenty of points to be made here. I will approach this in a paragraph by paragraph manner: 1. Can we reliably differentiate sedation and analgesia? Often in performing pediatric sedation for infants and young children we have to sedate a child to the point of obtundation - just to achieve movement control. In these cases the line between where a child is sedated enough to tolerate pain, or when he/she is just sedated enough to hold still - is impossible to draw. Does the fact that a child is sedated without analgesia mean he/she is not at any risk for airway compromise, hypoxia, or aspiration? We reject the “fruit salad” argument. 2. The authors relate Dr. Cote’s study to one by Pena et. al.. While it is true we do not know the “n” for Cote’s study, it is clear that millions of cases were performed over the time period that the cases were collected and only a small fraction were reported. We all recognize (as does Dr. Cote) that sedation critical events are relatively rare - just not rare enough. We need to study a huge number of cases to be able to analyze where the problem areas are. The Pena study looked roughly 1,000 sedations and analyzed the incident rate. As we have pointed out in our previous newsletters, critical incidents in sedation should be no more common than 1 in several thousand to one in 10’s of thousands. While the Pena study is certainly noteworthy, the fact that no one had brain injury or death in a cohort of one thousand sedations does not prove it is safe! In fact the 2.3% that had some type of complication may well indicate that if enough sedations were evaluated a critical incident might well have occurred. The authors of this letter are mixing apples and oranges when it comes to reviewing the literature. We reject the “Grandmother” argument. 3. The authors ask why they should be required to have any monitoring of patients undergoing chloral hydrate sedation. Answer: 1: Response to any of the sedation agents is variable – eventually you will have a patient (perhaps already neurologically impaired) that will have respiratory depression with an “appropriate” dose of chloral hydrate. 2: Knowledge of human performance would indicate that a medication error will be made in their lab at some point and that error will put a child in severe danger if there is no monitoring. Yes, you need monitors even if you have never had a sedation problem with chloral hydrate in your institution – just as you should wear your seatbelt even if you have never experienced major trauma from a motor vehicle accident. What price can are we willing to put on the life of the child who is overdosed? Consider that chloral hydrate in low doses will fail a considerable percentage of the time (1). Providers will want to offer repeated dosing or larger doses (thereby mandating monitoring). Critical incident reporting such as that in Dr. Cote’s study reminds us that accidents will happen with chloral hydrate. Analysis of critical incidents is exactly where we need to start to improve safety for pediatric sedation. These incidents tell us where the problems lie when facing a low frequency but high cost problem such as pediatric sedation errors – these cannot be easily studied on a day to day basis. In fact this is exactly the type of analysis that improved safety in the aviation and nuclear power industries by orders of magnitude while medicine has remained in the relative dark ages of safety improvement (see the recent Institute of Medicine Report on line) Moving backward on monitoring criteria because you have never personally experience a critical incident (removing the need for monitoring) is poorly reasoned and flies in the face of all recent advances in creating medical safety. We do not think the “cost effectiveness argument” will stand up. 4. The last three paragraphs ask some great questions but result in some unfortunate conclusions. Better studies are definitely required in order to understand what constitutes true safety in pediatric sedation practice. We need better numerators and denominators for sedation errors – no one would argue with this point. We should try to determine which types of sedation carry the greatest risk to children. All of this will require very detailed analysis. What the authors appear to fail to recognize is that Dr. Cote’s study never intends to report frequency rate of critical incidents given certain types of sedation – rather it is compiling a number of critical problems in pediatric sedation with the intent of pointing out where significant errors have occurred and could continue to occur. The conclusion Freeman and Vining draw out of this perceived lack of definitive data, (that sedation for non-painful EEG’s is completely different than other types of sedation and therefore does not require monitoring) represents a misinterpretation of the current literature (in particular Dr. Cote’s study) If widely adopted their ideas would likely end up causing a new series of sedation critical events. 1. Malviya S. Et al. Sedation and general anesthesia in children undergoing MRI and CT: adverse events and outcomes. British Journal of Anaesthesia 84 (6):743-8, 2000 Jun Close Call and Critical Incident Reports We invite those receiving the newsletter to submit cases to this "Close Calls" section aimed at the key safety issues associated with providing pediatric procedural sedation. The cases have been "sanitized" such that the date, patient identifiers, institutional identifiers, provider information, etc. will not violate the privacy of the patient or care providers. Case #4: A 4yo child was having a diagnostic lumbar puncture performed by the oncology service. A senior anesthesia resident used an existing IV to induce sedation with propofol. Standard monitors were placed as the child became somnolent including EKG, NIBP, and SpO2. The child’s oxygen saturation (initially 98%) began to decrease and the child became apneic. Because of the offsite location and the use of IV sedation, the anesthesia provider was prepared to use a pediatric self-inflating AMBU bag. Despite attempted bag mask ventilation by this experienced airway manager, the child desaturated to approximately 50% and the attending was called STAT to help out. The attending was close by and noticed that the “pop- off” valve was not closed. He also noted that the round mask provided with the bag did not provide a very tight seal for ventilation. Subsequently the child was easily ventilated and the oxygen saturation normalized. The child resumed spontaneous ventilation 1minute later and the procedure was completed under propofol sedation with supplemental O2 by external mask. Comments: This is a good example of a “use” error related to equipment. This case involved a senior resident in anesthesiology using equipment found outside of the OR setting that they were not familiar with (i.e., an AMBU bag with a “pop-off” valve (range 15-50cm H2O). Interviews with respiratory therapy experts revealed that initial responders frequently attempted bag-mask ventilation with the “pop-off” valve open during respiratory arrest situations throughout the hospital. The adult AMBU bag used at the reporting institution does not have such a valve. The intent of such a valve is to provide “safety” from excess pressure developing when providing positive pressure ventilation. The hazard of accidental pneumothorax is real, but unlikely with a maximum pressure possible of 50cm H2O. During 1 of 2 mock codes using a pediatric simulator (part of an ongoing study at Dartmouth), a senior pediatric resident committed the same “use” error and was unable to ventilate the simulated toddler under difficult conditions. Another observation has been that the round mask attached to the AMBU bag included with pediatric code carts is difficult to use. In fact anesthesia personnel and respiratory therapists have tended to switch to a triangular mask with an air-pillow. These experts tend to use the round mask only for the preterm infants. Interestingly, the round mask is actually more expensive than a triangular mask! Key issues: 1) The more complex the equipment the harder it is to use. Use errors represent a hazard and certainly can harm patients. This is becoming a more common form of system failure as medical technology and devices become more complex. The FDA is interested in having manufacturers pay more attention to the “usability” of medical devices. They have issued Guidelines for Industry-Medical Device Use—Safety: Incorporating Human Factors Engineering into Risk Management as of July, 2000. Note--on February 14, 2001 from 1-3 pm EDT, the FDA is sponsoring a live, interactive satellite teleconference titled “Integrating Human Factors into Medical Device Design and Development”. For more information on this program, please see http://www.fda.gov/cdrh/Useerror/teleconf101100.html pediatric AMBU bag round mask triangular masks “pop-off” valve 2) Experts (in this case an anesthesiologist) using unfamiliar equipment can perform poorly. Unfortunately a great deal of equipment in a hospital is presumed to be “standard” when it is not. Equipment for positive pressure ventilation is a good example. There are at least 10 types of circuits available for delivering oxygen and positive pressure ventilation. We know that the the learning curve associated with aquiring the skill set to use a tool or technique optimally can vary. The learning curve for bag-mask ventilation requires ten’s of hours for first year anesthesia residents managing patients under anesthesia. In addition the notion that you must “use it or loose it” to maintain skills is true. This is in part why one study after another is showing clinical volumes to relate to outcome in many care settings, especially complex care settings like the OR (e.g., providers doing less than 100 cardiac bypass cases per year have higher mortalities). While the learning curve for remembering to close this “pop-off” valve is steep (i.e., learned rapidly), multiple seemingly trivial items that a clinician must “remember” when stressed (i.e., a patient is apneic and desaturating) is a “set-up” for failure. 3) Feedback is important. When I try to understand why this problem is rare in the OR, I concluded that it may be related to feedback. The standard anesthesia circuit used includes a rubber bag that does not self- inflate. It requires an oxygen source and a seal for it to inflate. No facemask seal, no inflation. If the “pop-off” valve is closed and the patient is over inflating-the bag over-inflates. What you see is what you have. A self inflating bag will look the same even if you forget to hook up oxygen and/or if you are over inflating the patient and need to open the “pop-off” valve. 4) “One size fits all” leads to a crummy fit for most. The round mask provided with the AMBU-bag in this case is best suited for neonates, but is not the first choice for older children. Triangular masks with an air- pillow provide a better anatomic fit for older children. The choice to go with the round mask was to have a single piece of equipment that worked on a larger range of children. This case caused us to look at our equipment for codes and the errors commited during simulations. We have made the following changes and will be using mock code performance to attempt to validate that the changes improve safety. 1) All pediatric AMBU bags we are using with a variable “pop-off” valve will have the valve closed such that the maximum pressure that can be delivered is 50cm H2O. The user will give less if they squeeze less hard. 2) Trangular masks in three sizes with be provided with all Pediatric AMBU-bags along with 3 sizes of oral airways. 3) Training videos and simulation will be used to teach initial responders how difficult it is to ventilate a child with the “pop-off” valve open. Please send us your thoughts and concerns.