| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Pediatric Imaging |
1 From the Departments of Anesthesia (K.P.M.), Biostatistics (D.Z.), and Radiology (K.P.M., L.C., V.E.K., P.J.F., P.A.S., P.E.B.), Children’s Hospital, 300 Longwood Ave, Boston, MA 02115. From the 2003 RSNA scientific assembly. Received November 20, 2003; revision requested January 29, 2004; revision received February 17; accepted March 23. Address correspondence to K.P.M. (e-mail: keira.mason@tch.harvard.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: The institutional review board approved the review of medical records and determined informed consent to be unnecessary. All parents gave informed consent for patient sedation. Prior to MR imaging or CT, infants younger than 12 months were sedated with PO pentobarbital (4–8 mg per kilogram body weight) or IV pentobarbital (2–6 mg/kg), depending on the presence of an IV catheter or need for IV contrast medium. A computer database used to record sedation data was reviewed for data from January 1997 to September 2003. PO and IV sedation groups were compared for mean age, weight, dose, time to sedation, time to discharge, and duration of sedation with a two-sample Student t test. Multivariate analysis of covariance was used to determine whether differences in sedation time, time to discharge, and duration of sedation between groups were independent of age, weight, sex, American Society of Anesthesiologists physical status classification, dose, and type of procedure. Sedation effectiveness (outcome) was determined as the percentage of sedation failures in each group. Safety was determined separately for other adverse events as a total and for respiratory adverse events.
RESULTS: A total of 2164 infants received 2419 (1264 PO, 1155 IV) doses of pentobarbital for sedation. Weight and sex were comparable between groups. Time to sedation was significantly longer with PO than with IV pentobarbital (18 minutes ± 11 vs 7 minutes ± 7; P < .01), but time to discharge was similar, at approximately 108 minutes ± 35. Total adverse events rate during sedation was not significantly different (0.8% [PO] vs 1.3% [IV]), but incidence of abnormal oxygen saturation (5% decrease from baseline, >1 minute duration) differed significantly (0.2% [PO] vs 0.9% [IV]; P = .02). Sedation effectiveness was comparable (failure rate, 0.5% [PO] vs 0.3% [IV]; P = .76).
CONCLUSION: PO pentobarbital has comparable effectiveness and a lower rate of respiratory complications compared with IV pentobarbital in infants younger than 12 months; its use should be considered, regardless of presence of an IV catheter.
© RSNA, 2004
Index terms: Anesthesia • Computed tomography (CT), in infants and children • Magnetic resonance (MR), in infants and children • Radiography, in infants and children
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Since 1999, our radiology sedation protocol has specified the use of oral pentobarbital for infants younger than 12 months who will not require an intravenous contrast medium and who do not have an intravenous catheter in place. Infants who have an intravenous catheter in place or who require an intravenous contrast medium are sedated with pentobarbital administered via intravenous catheter. To our knowledge, the safety and effectiveness of pentobarbital delivered intravenously as opposed to orally have not been reported. Thus, the purpose of our study was to compare the effectiveness and safety of oral versus intravenous pentobarbital for MR imaging and CT in infants younger than 12 months.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sedation Database
In December 1993, our institution established a radiology sedation committee to establish and modify protocols and review sedation outcomes. The radiology sedation committee consists of an anesthesiologist, three radiologists, three registered nurses, two nurse practitioners, and an administrator with experience in data collection and computerized databases. Both the radiology and the hospital sedation committees must approve any new protocols or modifications prior to the institution of new practices or practice changes. In November 1997, the radiology department established a computerized database (FileMaker Pro, version 2.1; Claris, Cupertino, Calif) in which a standardized set of parameters and demographic data are recorded for each patient in whom sedation is performed. Parameters and outcomes include the type of study, the date and time of examination, medications administered (with doses in milligrams per kilogram body weight), the patient’s physical status according to the American Society of Anesthesiologists (ASA) classification system, adverse events, the time required to sedate, the time from sedation to discharge, and failure to achieve sedation. Other data recorded include the medical diagnosis and fasting status. Adverse events that occur in any of three time periods—during sedation, in the recovery room, and within 24 hours after discharge—are recorded. A radiology nurse contacts all families within 24 hours after sedation to document any untoward outcome. Radiology nurses collect and record these data on standardized forms that are later transferred to the computerized database by a single designated data coordinator who is specially trained in data input and retrieval. Every month, the radiology sedation committee holds quality assurance meetings at which patient demographic data, outcomes, and adverse events are reviewed. During these meetings, outcomes and comparisons between our existing drug regimens are reviewed, and changes to existing protocols, in addition to trials of new protocols, are discussed. Potential changes are presented to the hospital sedation committee for final approval prior to institution.
Definition of Terms
Since 1997, we have established well-defined descriptions of all patient demographic parameters and adverse events. By clearly defining these terms, we have ensured, as best we can, that all descriptors in the computerized database are consistent. Adverse events recorded in the database are as follows (12): (a) failed sedation, defined as inadequate sedation subsequent to delivery of the maximum allowable dose per the sedation protocol or as inability to complete the planned procedure, secondary to unacceptable motion artifact; (b) paradoxical reaction, defined as sustained irritability or combativeness of more than 30 minutes’ duration that occurs after administration of the sedative; (c) prolonged sedation, defined as the inability to meet discharge criteria 3 hours after administration of the sedative or as failure to return to baseline mental and behavioral status within 24 hours after sedation; (d) abnormal oxygen saturation, defined as a sustained (>1-minute) decrease in oxygen saturation of more than 5% from baseline despite delivery of 6 L/min oxygen via face mask, head repositioning, suctioning, and physical stimulation; (e) need for resuscitation, defined as any decline in respiratory rate and oxygen saturation that requires resuscitative efforts (positive pressure ventilation, cardiopulmonary resuscitation, or the use of medications to reverse sedation); (f) cardiovascular complication, defined as a sustained (>5-minute) decrease of more than 20% in mean arterial pressure with or without a decrease in heart rate below the lower limit of the normal range for the patient’s age; (g) unplanned admission, defined as unexpected admission to the hospital overnight as the result of an adverse event directly related to sedation; (h) gastrointestinal side effect, defined as vomiting, aspiration, or diarrhea that occurs within 24 hours after administration of sedation; (i) allergic reaction, defined as an unexplained rash or allergic symptoms that develop within 24 hours after sedation. The time to sedation is defined as minutes from initial administration of sedative to achievement of adequate sedation of the patient. The time to discharge is defined as minutes from the initial administration of sedative to the time at which the patient meets the criteria for discharge from the recovery room.
Sedation Protocol
In 1997, the radiology sedation committee established a standard protocol for the administration of oral and intravenous pentobarbital. All sedatives are administered by credentialed radiology nurses under the supervision of a radiologist, who orders the medication. Credentialing, for both radiologists and nurses, requires successful completion of annual basic life support and biannual advanced pediatric life support courses, in addition to an annual written hospital sedation examination.
Prior to scheduling the patient for an imaging study, a radiology nurse screens the patient to confirm that nurse-administered sedation is medically appropriate. After reviewing the available medical history, the nurse contacts the parent to confirm the medical history, complete a review of systems, and establish whether any additional measures are necessary (eg, specialist consultation, laboratory study, electrocardiography, echocardiography, or sleep study). The radiology sedation committee has established a set of medical contraindications for nurse-administered sedation (Figure). Any patients with these medical conditions are referred to the anesthesiology department. In the event that there is uncertainty regarding the medical appropriateness of nurse-administered sedation, the nurse consults with the supervising radiologist, who then requests a consultation with the anesthesiology department. A staff anesthesiologist then assesses the patient to make a final determination about whether the patient should be scheduled for nurse-administered sedation or for anesthesia. In either case, parental informed consent is obtained prior to administration of the sedative or anesthetic.
|
For oral administration, pentobarbital (50 mg/mL, Nembutal; Abbott) is mixed with cherry syrup (Humco, Texarkana, Tex) in a 3:1 ratio. The initial dose is 4 mg/kg. If needed, supplemental doses of 2–4 mg/kg may be administered every 30 minutes, to a maximum total dose of 8 mg/kg. Oral pentobarbital is usually administered via a 3-mL syringe while the parent holds the infant. For intravenous administration, pentobarbital is diluted in normal saline and administered in aliquots of 1–2 mg/kg at 1–2-minute intervals. The total intravenous dose generally ranges from 2 to 6 mg/kg, but patients who are currently receiving barbiturates for medical therapy (eg, for seizure disorders) may receive as much as 9 mg/kg intravenous pentobarbital.
Data Collection
The radiology sedation committee members (K.P.M., D.Z., L.C., V.E.K., P.J.F., P.A.S., P.E.B.) reviewed the computer database for all instances of oral or intravenous pentobarbital administration between January 1, 1997, and September 30, 2003, for sedation in infants younger than 12 months. For quality control, we routinely perform a double review of sedation records. First, the data coordinator (P.J.F.) continually reviews the patient charts and data forms for logical consistency and completeness. The resultant evaluation is sent to the study coordinators (K.P.M., P.E.B., L.C.) for their review and correction. After agreement, the patient data are entered into the database by the data coordinator. Second, the database is reviewed independently by two members of the radiology sedation committee (V.E.K., P.A.S.), and any corrections or concerns are brought to the attention of the study coordinators. We believe that this process works quite well. Review by the data coordinator is important for consistency and for keeping study evaluations up to date. For reasons of medical expertise, we also think it important to have the study coordinator and sedation committee members review the data to ensure that the results are not compromised by flawed data and that the radiologic imaging studies are monitored for patient safety.
All demographic data and outcome variables were surveyed with particular attention to adverse events. Power analysis was conducted by using software (nQuery Advisor, version 5.0; Statistical Solutions, Saugus, Mass) to determine sufficient sample size for the oral and intravenous pentobarbital groups, since the expected number of adverse events and failures was small. A minimum sample size of 1000 patients in each of the two groups would provide 80% power (
= .05) to detect a difference of 0.5% or more in sedation failure and respiratory complications between the groups.
Statistical Analysis
Oral and intravenous sedation groups were compared with regard to mean age, weight, dose, time to sedation, time to discharge, and duration of sedation by using the two-sample Student t test. The proportion of males and females and the type of procedure (MR imaging or CT) were assessed with the Fisher exact test, and the Pearson 
2 test was used to compare ASA classification levels between the two groups. Continuous variables were correlated with the Pearson product moment correlation coefficient (r). Multivariate analysis of covariance was applied to determine whether differences in sedation time, time to discharge, and duration of sedation between oral and intravenous pentobarbital groups were independent of age, weight, sex, ASA classification level, dose, and type of procedure, and the F test was used to assess the significance of each covariate (13). There was an inherent bias due to differences in the distribution of ASA classification levels between the two groups in our study; we removed this bias by adjusting not only for ASA classification level but also for age, weight, sex, dose, and type of procedure in the multivariate analyses. Analysis of covariance was used to evaluate group differences in time to sedation, time to discharge, and duration of sedation; logistic regression was used to evaluate differences in adverse events and sedation failures. Specifically, multiple logistic regression was used to identify which variables, including method of pentobarbital administration, age, weight, sex, ASA classification level, dose, and type of procedure, were associated with decreased oxygen saturation and with sedation failure; the odds ratio and 95% confidence interval were used to measure the adjusted odds of these outcomes on the basis of the final stepwise model (14). Confidence intervals were determined for the incidence of abnormal oxygen saturation levels and for sedation failure by using the normal approximation method (15). Statistical analysis was performed by using software (SPSS, version 11.5; SPSS, Chicago, Ill). A two-tailed P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our comparison of adverse effects of sedation with oral versus intravenous pentobarbital in patients of the same age and similar characteristics revealed a higher incidence of decreased oxygen saturation with intravenous administration than with oral administration (0.9% vs 0.2%). This difference remained significant after controlling for the imbalance between the study groups with regard to the distribution of ASA classification levels. This result was contrary to our assumption that careful intravenous titration of the sedative, compared with oral bolus dosing, would result in fewer respiratory complications. Also surprising was our finding that the failure rate between the sedation techniques was comparable (0.3%–0.5%). Predictably, the time to sedation was longer with oral pentobarbital. However, the mean difference of 11 minutes between oral and intravenous pentobarbital was lower than we expected. Despite careful titration of intravenous pentobarbital to achieve the desired effect, the overall duration of sedation and the time required to meet discharge criteria did not differ between the two groups.
To our knowledge, this is the first reported comparison of two different methods of administration of the same sedative in a study population of similar age and demographics. Sedation in infants can be particularly challenging because children of this age are more susceptible to airway obstruction and hypoventilation for multiple reasons. Airway obstruction is more likely in infants than in adults because the infant tongue is larger relative to the total size of the mouth (20). The infant pharynx is at risk of collapse from the negative pressure that can occur as the tongue falls backward during inspiration (21). Neonates are susceptible to airway obstruction during head flexion because of their soft necks, compressible airways, and large tongues (22). If possible, imaging is attempted without sedation in infants by using techniques such as sleep deprivation or feeding prior to examination to increase the likelihood of the infant’s falling asleep during the imaging study.
This study had several limitations due to the retrospective nature of its design. First, because the patients were not randomly assigned to sedation treatment groups, there were certain imbalances between the two groups. For example, significantly more patients with ASA classification level 1 received oral pentobarbital than intravenous pentobarbital (62.5% vs 46.5%). These kinds of imbalances are typical of nonrandomized retrospective studies, although statistical procedures should be used to eliminate the potential biases that may be caused by different composition of study groups. To remove the potential bias due to differences in the distribution of ASA classification level between our study groups, we applied analysis of covariance and logistic regression methods and treated ASA classification level as a covariate when comparing the two groups with respect to sedation and discharge times, as well as when evaluating differences in respiratory complications and sedation failures. This strategy, in essence, was used to adjust for the inequalities between the oral and intravenous treatment groups.
Another limitation of this study was the lack of double blinding or masking. Blinding is used to avoid the risk of personal bias in comparing treatments. With double blinding, neither the investigator nor the patient knows to which treatment group the patient has been assigned. Often, in clinical settings, bias occurs because of preconceptions of the investigator and/or supporting staff that might influence the reporting of outcomes or adverse events. When individuals involved with a study have knowledge of the treatments given to patients and other relevant information, then their personal subjective biases may interfere (consciously or unconsciously) with the reporting and evaluation of the data. It is extremely difficult in practice to assess and control for the biases that result from a lack of blinding.
The third study limitation is the lack of randomization. Since this study was not prospective, the patients were not randomly assigned to the oral and intravenous pentobarbital treatment groups. Randomization is an effective way to prevent selection bias in the assignment of patients to treatment groups. Ideally, a prospective randomized double-blind study would have been superior to control for potential bias, not only to generate comparable groups of patients who have similar characteristics but also to allow the most objective assessment for clinical evaluation of the two methods of administration under investigation.
On the basis of our findings, we recommend that pentobarbital be administered with the oral method rather than the intravenous method for sedation in infants younger than 1 year, regardless of the presence of an intravenous catheter. In the event that the child does not have an intravenous catheter but requires an intravenous contrast medium, venous access should be attained prior to administration of oral pentobarbital, because pentobarbital has no analgesic properties and because attempts to achieve venous access after the administration of oral pentobarbital could jeopardize the success of sedation. In conclusion, our findings indicate that oral pentobarbital, compared with intravenous pentobarbital, is safer and equally effective in infants younger than 12 months. For this reason, oral and not intravenous sedation should be used for routine imaging of infants in the radiology department.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, K.P.M., D.Z., P.E.B.; study concepts, all authors; study design, K.P.M., D.Z., P.E.B.; literature research, K.P.M.; clinical studies, K.P.M., L.C., V.E.K., P.A.S.; data acquisition, L.C., V.E.K., P.J.F., P.A.S., P.E.B.; data analysis/interpretation, D.Z., K.P.M., P.E.B.; statistical analysis, D.Z., K.P.M.; manuscript preparation, all authors; manuscript definition of intellectual content, K.P.M., D.Z., P.E.B.; manuscript editing, all authors; manuscript revision/review and final version approval, K.P.M., D.Z., P.E.B.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. E. Zgleszewski, D. Zurakowski, P. J. Fontaine, M. D'Angelo, and K. P. Mason Is Propofol a Safe Alternative to Pentobarbital for Sedation during Pediatric Diagnostic CT? Radiology, May 1, 2008; 247(2): 528 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Pershad, J. Wan, and D. L. Anghelescu Comparison of Propofol With Pentobarbital/Midazolam/Fentanyl Sedation for Magnetic Resonance Imaging of the Brain in Children Pediatrics, September 1, 2007; 120(3): e629 - e636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Dalal, D. Murray, T. Cox, J. McAllister, and R. Snider Sedation and Anesthesia Protocols Used for Magnetic Resonance Imaging Studies in Infants: Provider and Pharmacologic Considerations Anesth. Analg., October 1, 2006; 103(4): 863 - 868. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. P. Mason, S. E. Zgleszewski, J. L. Dearden, R. S. Dumont, M. A. Pirich, C. D. Stark, P. D'Angelo, S. MacPherson, P. J. Fontaine, L. Connor, et al. Dexmedetomidine for pediatric sedation for computed tomography imaging studies. Anesth. Analg., July 1, 2006; 103(1): 57 - 62. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Pershad and B. Gilmore Successful Implementation of a Radiology Sedation Service Staffed Exclusively by Pediatric Emergency Physicians Pediatrics, March 1, 2006; 117(3): e413 - e422. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
