"Pediatric Sedation Newsletter � September 2004"
Pediatric Sedation Newsletter – September 2004 Departments of Anesthesiology and Pediatrics, Children’s Hospital at Dartmouth, Dartmouth Hitchcock Medical Center, Lebanon, NH Editors: Joseph Cravero MD (email@example.com), George Blike MD firstname.lastname@example.org Website = http://an.hitchcock.org/PediSedation/ Circulation 3440 We hope everyone has had a great summer. To start this newsletter we would like to point you toward the content of the “1st International Multispecialty Conference on Pediatric Sedation” held on May 22, 2004 outside of Denver, Colorado. The conference was intended to evaluate the “state of the art” in pediatric sedation from the standpoint of several different specialties that provide this care for children. This meeting was sponsored by the Denver Children’s Hospital and the Pediatric Sedation Research Consortium. A formal report and analysis of the conference is still in preparation but we have provided the content of the primary lecture presentations in PowerPoint slides on the website www.PediatricSedationRC.org. During this conference we took questions from the audience and entertained a few of them in a structured panel discussion during lunch. Unfortunately there was not nearly enough time to review all of the questions that were submitted (literally hundreds). In the next several newsletters we will review some of the questions that were not addressed during the sedation conference and offer our own attempt at answers. Any comments on or additions to the answers that are provided are welcome and we will be happy to print them in the next newsletter. Questions from the 1st International Multidisciplinary Conference on Pediatric Sedation lunchtime question and answer period. 1. MRI – what types of pumps are used in the MRI scanner at this point? Answer: There are a number of MRI compatible pumps on the market. We are not supported by any company and we do not promote any one pump over another. Currently we use a unit made by Medex. We have found it useful and relatively reliable – it is kept at least 5 feet away from the core. We have also used Bard pumps and attached about 7 – 36 inch low volume extensions so that the pumps are at the door of the MRI scanner; and they seem to work OK in our hands. Having said this, there has been some work that shows the performance of non-MRI compatible pumps is less than ideal when used anywhere near the scanner. We would also include the fact that even the MRI compatible pumps can be “sucked” into the scanner if they are too close to the core. In this case MRI “compatible” means that the pumps work in the environment but it does not necessarily mean they are completely non-magnetic. We reviewed a case concerning this exact issue in our Pediatric Sedation Newsletter several months ago – please refer to http://an.hitchcock.org/pedisedation. 2. Can anyone discuss how to sedate the very hard to sedate patients i.e. patients with Autism or severe AD/HD? Are they best served by GA? Answer: A very difficult dilemma – many approaches would be acceptable. We begin our approach to these patients by having our child life specialist (or ourselves) determine the exact nature of the child’s problem and how it might impact on the induction/maintenance of sedation. Some of these children have had a lot of blood draws for medication levels etc. and they may become “habituated” to them – meaning they seem to have gotten beyond the stage of being horrified by the experience. On the other (more common) hand they are made extremely anxious and out of control simply at the thought of venipuncture. If the patient has a history of being “OK” with blood draws, we start an IV and use an IV technique appropriate for the procedure to be done. In our hands this includes a sedative hypnotic such as propofol for diagnostic procedures, with a short acting opiate added for painful procedures. (We are providing deep sedation/anesthesia). Other options, including midazolam with fentanyl, pentobarbital etc., have also been used with success. One caution would include the fact that (in our experience) paradoxical reactions to versed or pentobarbital are more common with this patient population than they are in the general population of children that we sedate. In these cases the usual sedation dose of medication results in a wildly “out of control” child rather than a docile sleepy one. We should recognize that there are a wide range of clinical entities and behaviors that are currently labeled “autistic”. Some of these patients have significant tactile aversion - so placing the IV or putting a mask on their face is really a huge problem. In these cases they may or may not cooperate with an oral med. We are a little careful about midazolam oral prep with this subgroup of patients because of the potential for a paradoxical response as mentioned above. In light of this we find that induction of sedation with an IM medication (ketamine +/- midazolam) has yielded the greatest success. This has to be done with awareness of all the possible side effects of this route and these meds, but that is the way we approach the most difficult of these tough cases. We would term this type of technique deep sedation or anesthesia – some would refer to it as dissociative sedation – a subject that has been reviewed in this newsletter on a number of occasions. In any case to (finally) answer the question, we do not opt for classic “GA” with inhaled agents as we do not find it offers real advantages over the techniques mentioned above; and inhaled induction of anesthesia is quite traumatic for many of these patients. 3. Is there a policy concerning chloral hydrate sedation for full term infants less than one month of age and former preemies who are less than 60 weeks PCA? Are they required to stay overnight? Answer: Wonderful question that we have put to the readership of our newsletter in the past. I think the practice varies significantly from institution to institution, but we would strongly suggest that you keep these patients overnight for observation just as they would be kept after GA. We do not know of studies specific to sedation in this age group, but logic would say that a long acting sedative could potentially have the same depressant effect as the lingering effects of anesthetics and it seems only prudent to observe them. The exact age cut offs that are used will also vary from institution to institution and we may individualize this dictum with regard to former preemies depending on the extent of their pre-maturity (i.e. 26wker vs. 35 wker) but our general rule is to admit and observe. Fortunately we find that many of the tests and procedures, such as MRI scans, that require sedation in other age groups can be accomplished without sedation in these patients when experienced techs are handling the children using some sleep deprivation, feeding just before the study, and swaddling the child for the study. Having said this we are asked to sedate these children and we do keep them for observation when we have to sedate. Question from a reader: We recently had a reader question us about the practice of sedating children without supplemental oxygen. This is done by some practitioners to allow faster detection of hypoventilation or apnea. We are interested to know if anyone routinely uses this practice – we would love to hear your personal experience and any evidence for improved safety associated with this practice. Thanks. ____________________________________________________________________ Literature Reviews: We would like to start by presenting a letter to the editor that appeared in Anesthesia and Analgesia Vol 98 pp 1814, 2004. Dexmedetomidine in Pediatrics: Controlled Studies Needed Steve Serlin, MD Valley Anesthesiology Consultants, Clinical Associate Professor Anesthesiology, University of Arizona, Phoenix, Arizona To the Editor: Clonidine, an alpha-2 agonist, is widely used for pediatric pain and neuropsychobehavioral therapy (1). Although not approved for pediatric use by the Federal Drug Administration, dexmedetomidine, 8 to 10 times more specific than clonidine, has been administered to 48 pediatric patients for perioperative, pain, or behavioral management at Phoenix Children’s Hospital with the approval of the Pharmacy and Therapeutics Committee. All patients received an initial dose of 0.5 µg/kg over 15 min, followed by an infusion of 0.25–1.25 mg · kg–1 · h–1. Duration of administration was 12 to 144 h. Outside the operating room all patients were monitored in an intensive care unit. The patient’s ages ranged from 10 mo to 19 yr. Diagnoses included congenital muscular dystrophy, spinal muscular atrophy, myasthenia gravis, Rhett syndrome, opioid-augmented respiratory failure, recurrent pancreatitis, pectus excavatum, Henoch-Schonlein syndrome, familial polyposis, severe neurobehavioral disorders, congenital heart disease, and 14 cases of idiopathic scoliosis. Dexmedetomidine was effective for all selected indications. Patients with pulmonary compromise from muscular or neuromuscular disorders undergoing a thoracotomy or spinal fusions maintained spontaneous ventilation postoperatively, avoiding mechanical ventilation. Patients with severe neurobehavioral disorders remained calm and responsive postoperatively. A patient with recurrent pancreatitis pain refractory to all analgesics, including ketamine and epidural catheter placement, demonstrated dramatic analgesia. A ventilator dependent patient receiving infusions of opioids and benzodiazepines was extubated within 40 h after initiating dexmedetomidine. Dexmedetomidine also helped to differentiate severe inflammatory abdominal pain from a noninflammatory etiology. Although all patients were monitored in an intensive care unit, dexmedetomidine was safe. Hypotension was the most significant side effect and only occurred with inadequate fluid replacement unless desired intraoperatively. We hope this letter will encourage controlled studies of dexmedetomidine in academic centers so its pediatric surgical and medical applications can be better delineated. Reference 1. Nishina K, Mikawa K, Shiga M, Obara H. Clonidine in pediatric anesthesia. Paediatr Anesth 1999; 9: 187–202. Commentary: We thought we would include this letter for two reasons: 1) it provides some interesting insight into how one institution is using this medication in ICU children; 2) the author advocates for what we really need – prospective, randomized studies of the effectiveness and usefulness of this drug. There is still relatively little data in the collected world literature about the use of dexmedetomidine in children (see reference list below) however we believe that our conversations with people across the country indicate that its use is growing rapidly. In our own institution we have used the drug in our Pediatric ICU on 4 patients with outstanding results. Unfortunately we must once again object to Dr. Serlin’s use of the term “safe” with respect to dexmedetomidine as a conclusion from an observational report of 48 total patients. The main drawback to the use of this medication has been associated hypotension and bradycardia, noted most commonly in adults. As children are less apt to have hemodynamic swings with sympathectomy (data on spinal anesthesia supports this) it is possible that these complications may be much less common in children. There is no doubt that dexmedetomidine has potential for use as a sedative in the ICU setting and even for procedural sedation, however we absolutely need to go beyond anecdotal reports of its usefulness to controlled studies that look carefully at it effectiveness and safety in a number of different venues. 1. Tobias, J. D., J. W. Berkenbosch, et al. (2003). "Additional experience with dexmedetomidine in pediatric patients. [Review] [21 refs]." Southern Medical Journal 96(9): 871-5. 2. Tobias, J. D. and J. W. Berkenbosch (2004). "Sedation during mechanical ventilation in infants and children: dexmedetomidine versus midazolam." Southern Medical Journal 97(5): 451-5. 3. Tobias, J. D. and J. W. Berkenbosch (2002). "Initial experience with dexmedetomidine in paediatric-aged patients." Paediatric Anaesthesia 12(2): 171-5. 4. Berkenbosch, J. W. and J. D. Tobias (2003). "Development of bradycardia during sedation with dexmedetomidine in an infant concurrently receiving digoxin." Pediatric Critical Care Medicine 4(2): 203-5. Literature Continued…… Shobha Malviya, Terri Voepel-Lewis, Alan Tait et. al. Pentobarbital vs. chloral hydrate for sedation of children undergoing MRI: efficacy and recovery characteristics. Paediatric Anesthesia 14:589-595. Abstract Excerpted: The authors point out that Chloral hydrate (CH) sedation for MRI scans has been associated with significant failure rates, adverse events and delayed recovery. Pentobarbital (PB) has a reported success rate of 98%. This study was intended to compare the efficacy, adverse events, and recovery characteristics of CH vs. PB in children undergoing MRI scans. Methods: Children were randomly assigned to receive either PB in incremental doses up to 5 mg/kg or oral CH 75mg/kg prior to their MRI. Sedation was augmented with 0.05mg/kg of midazolam as necessary. Adverse effects (hypoxemia, failed sedation, paradoxical reactions, and behavior changes), the return of baseline activity, and parental satisfaction were documented. The quality of the MRI scans was evaluated by a radiologist blinded to the sedation technique. Results: PB had an earlier onset of sedation (P = 0.001), higher sedation scores (deeper sedation) (P = 0.01), and less need for supplemental midazolam. Severe hypoxemia occurred in two children (6%) in the PB group. Fourteen per cent of the PB group experienced a paradoxical reaction, 9% sedation failure, and 11% major motion artifact, compared with 0% (P = 0.05) and 2% (P = NS) in the CH group. CH and PB were both associated with a high incidence of motor imbalance and agitation. Children who received PB had a slower return to baseline activity (P = 0.04). The authors conclude that PB facilitated quicker sedation and less requirement for supplemental O2 but produced a higher incidence of paradoxical reaction and prolonged recovery with a similar failure rate when compared to CH. Commentary: We are always impressed with the thorough nature of the studies this group produces. Specifically, the attention to the quality of the MRI scan with the two drugs is an important (and not always reported) piece of data in studies of sedation in the radiology setting. Unfortunately we are only told that 70% of the scans were available for review (seems a little odd) and that 15% of the PB scans and 2% of the CH scans could not be read because of movement. This is important data but hard to evaluate. We could not find the distribution of the available scans or the reasons some were not available – were unreadable scans in the CH group simply discarded? The authors followed up patients 24-48 hours after the study. Once again this kind of data is often missing in sedation studies and is particularly important when long acting medications have been used. It is interesting to note that patients who received IV PB took longer to return to baseline than the oral CH patients in this particular study paradigm. In fact 66% of patients who received PB did not return to baseline activity for more than 8 hours. We found it somewhat unusual that paradoxical reactions were only found in the PB group since we have definitely seen this reaction in children receiving CH in our institution. The findings of inadequate sedation frequency are at odds with other studies (see Mason et. al. below) which have shown a much higher success rate with PB than is reported in this paper. Clearly this could be due to technique used by the sedation nurses who titrated in PB with relatively small doses of 2mg/kg and a maximum of 5mg/kg. Comfort and experience with this medication could clearly change the outcomes – we are not aware of the relative use of these two drugs in the study institution. In addition, midazolam boluses delivered at the discretion of the radiology nurses are not standard procedure in many hospitals and may change the outcomes of this study. We could not easily discern how many of the children who had significant desaturation with PB had received midazolam IV doses (or not). It is hard to interpret some of the data in light of this fact. We would also be cautious in applying the satisfaction data collected in this study which showed a “somewhat to very” satisfied rate of 76% with PB and 88% with CH. Since satisfaction is heavily dependent on previous experience, without knowing what kind of background these patients had with respect to sedation it is hard to know what to make of this comparison. Comparison satisfaction data in a cross over study of these two groups (for instance) would be much more helpful. Mason KP. Sanborn P. Zurakowski D. Superiority of pentobarbital versus chloral hydrate for sedation in infants during imaging. Radiology. 230(2):537-42, 2004 Feb Abstract: The authors sought to compare the effectiveness and safety of oral pentobarbital and oral chloral hydrate for sedation in infants younger than 1 year during magnetic resonance (MR) imaging and computed tomography (CT). Methods: A computerized database was used to collect information about all cases in which sedation was used. Outcomes of all infants who received oral pentobarbital or oral chloral hydrate for sedation between 1997 and 2002 were reviewed. Two study groups were compared for sedation and discharge times by using Student t test and for adverse events by using Fisher exact test and multiple logistic regression analysis. Results: Infants (n = 1,316) received an oral medication for sedation. Mean doses were 50 mg/kg chloral hydrate and 4 mg/kg pentobarbital. Student t test demonstrated no difference in mean time to sedation and in time to discharge between groups. Overall adverse event rate during sedation was lower with pentobarbital (0.5%) than with chloral hydrate (2.7%) (P <.001). There were fewer episodes of oxygen desaturation with pentobarbital (0.2%) than with chloral hydrate (1.6%) (P <.01). Both medications were equally effective in providing successful sedation. The authors conclude that although oral pentobarbital and oral chloral hydrate are equally effective, the incidence of adverse events with pentobarbital was significantly reduced. Commentary: A very different type of study than the Malviya et al. paper above - using a database to record and to evaluate outcomes based on medication PB vs. CH in a different (specifically) infant age cohort using only oral medications of both types. The authors should be complimented for very strictly defining their criteria for failed sedation, paradoxical reaction, prolonged sedation, abnormal oxygen saturation, cardiovascular complication, and unplanned admission at the beginning of this paper – if only we could all agree on how to define these terms…..The other obvious strength of this study is the relatively large numbers involved – we seldom see sedation studies with more than one thousand patients, and on this basis alone we see advantages to this type of investigation. Unfortunately in this case there were 1,024 patients in the PB group and 374 in the CH group – a more even distribution might be desirable. Both oral chloral hydrate and oral pentobarbital had high rates of success (>98%) but this is not too surprising since almost all studies looking at sedation using oral chloral hydrate indicate that there is a high success rate in the infant age group – and by extension it is not surprising that if oral pentobarbital worked at all, it would be similarly successful in this cohort. The incidence of oxygen desaturation was low in both groups compared to other sedation studies (1.6% for CH and 0.2% for PB) – one patient required PPV in the CH group. (Is this another argument for requiring monitoring in CH sedated patients?) As with the preceding paper, the authors took the time to provide a 24 hour follow up for late adverse event rates. In this case the rate was low for both drugs in all of the important outcome categories including prolonged sedation, irritability etc. Once again it must be pointed out that the infant age group is probably not as susceptible to some of these problems as older patients, in our experience, and therefore a direct comparison with the previous study is not appropriate. Difficulties with this type of research must be appreciated – including the fact that all of the Chloral Hydrate data came from 1997-2000 and all of the PB data came from 2000-2002 when the group obviously changed technique. As with any non-randomized study one has to assume that no bias has been introduced by change in experience or other environmental alterations over this time period in order to accept this as a fair comparison. In addition, from the description in this paper, this is a highly trained and skilled group of nurse sedation providers at a very large children’s hospital – how easily the results would be reproduced at other institutions is not certain. What is clear from the study is that oral pentobarbital works well in this age group. Is the dose used (avg 4mg/kg oral) exactly equivalent to the CH dose (50mg/kg oral)? We can only point out that the lack of airway intervention required for these patients would indicate that the depth of sedation produced by PB is not excessive – hard to know how exact this comparison is however. Finally we would point out that this study has extensive detail as to methodology and various outcomes – a full review of this paper is highly recommended for anyone who uses these meds. No case this time – please keep sending any interesting clinical scenarios to us and we will be happy to review and discuss them. As always your comments are welcome.