| JOURNAL HOME | CME HOME | THIS MONTH | PAST ISSUES | ETOC | COLLECTIONS |
| AUTHORS | REVIEWERS | EDITORIAL BOARD | FEEDBACK | RSS | HELP |
| ||||||||||||||
| ||||||||||||||
|
|
|
|
||||||||||||
|
||||||||||||||
Departments of
*Anesthesiology and Critical Care Medicine and
Pediatrics, The Johns Hopkins Hospital, Baltimore, Maryland;
Department of Pediatrics, Children’s Hospital of Los Angeles, Los Angeles, California; and the
§Pediatric Pain Service, The Johns Hopkins Hospital, Baltimore, Maryland
Address correspondence and reprint requests to Constance L. Monitto, MD, Blalock 943, Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Hospital, 600 North Wolfe St., Baltimore, MD 21287. Address e-mail to cmonitto{at}welchlink .welch.jhu.edu.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
3/10 (objective pain scale) or 2/5 (objective or self-report pain scale) in more than 80% of all occasions of PNCA use. PNCA usage was associated with an 8% incidence of pruritus and a 15% incidence of vomiting on the first day of treatment. Nine children studied received naloxone, four (1.7%) for treatment of PNCA-related apnea or desaturation. All had improvement in their symptoms after naloxone administration.
Implications: Parent-/nurse-controled analgesia provided effective pain relief in most children <6 yr of age experiencing nonsurgical or postoperative pain. The observed incidence of vomiting and pruritis was similar to that seen in older patients treated with patient-controlled analgesia. However, significant respiratory depression, although uncommon, did occur, thus reinforcing the need for close patient monitoring.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Modifying this paradigm, over the past several years we have used a standard PCA pump and demand technology to treat pain in young and developmentally delayed patients with small-dose continuous opioid infusions, supplemented by small, but potentially more frequent, bolus doses of opioid, which can be administered by a nurse or parent to treat acute exacerbations of pain. We call this technique parent-/nurse-controlled analgesia (PNCA). Although PNCA has become common practice in our institution, the technique is controversial because of concerns regarding the risk of overdosage and the potential for respiratory depression if someone other than the patient is allowed to initiate bolus doses (8). In addition, the use of demand dosing technology in young children has been limited by a lack of evidence demonstrating its efficacy and safety. We have therefore undertaken a prospective, observational study to determine patient demographics, safety, efficacy, and the incidence of complications in patients for whom PNCA was provided at The Children’s Center of The Johns Hopkins Hospital.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
After the Pediatric Pain Service had been consulted by the patient’s primary physician, and opioid analgesia was deemed to be the optimal analgesic for treating the patient’s pain, PNCA was initiated using our standard protocols (9). A PCA pump (CADD-PrizmTM PCS Model 6100, SIMS Deltec, Inc., St. Paul, MN) programmed to deliver a basal infusion and bolus doses of morphine, fentanyl, or hydromorphone (Table 1) was attached to a continuously infusing IV catheter. The IV infusion tubing contained a one-way, back-check valve to prevent backflow and inadvertent dosing of opioid by gravity.
|
Demographic data collected on each patient included patient age, weight, sex, diagnosis, surgical site/type, coexisting medical conditions, and patient location. Choice of opioid, total dose received (µg/day) and the number of boluses administered were recorded over the 5 days of study on a daily basis. To allow comparisons between patients when they did not receive PNCA for full a 24-hr interval, usage was averaged to a µg · kg-1 · hr-1 dose (total daily dose received/[patient weight x number of hours on PNCA]) for each study day. To compare daily opioid usage patterns, opioid consumption was subsequently converted to "morphine equivalents" using a ratio of 1:40:5 for morphine:fentanyl:hydromorphone. When reported in morphine equivalents, standard bolus doses (i.e., 20 µg/kg morphine, 0.5 µg/kg fentanyl, or 4 µg/kg hydromorphone) and hourly basal infusions are all equivalent, irrespective of opioid choice. Use of additional drugs, such as nonnarcotic analgesics, benzodiazepines, antihistamines, antiemetics, and opioid antagonists, was recorded on a daily basis from nursing medication flow sheets.
Pain scores were measured by the patient’s nurse or by the patient, where appropriate, using an objective 6-point scale (pain rating: 0 = none, 1 = mild, 2 = discomforting, 3 = distressing, 4 = horrible, and 5 = excruciating), an objective 11-point (0–10) scale (10), or a subjective Wong-Baker FACES Pain Rating Scale (0–5) (11).
Throughout PNCA use, patients were monitored for vital signs and blood oxygen saturation (via pulse oximetry). Continuous pulse oximetry was performed for all patients during the first 24 h of PNCA use and whenever basal or bolus doses were increased. Supplemental oxygen was provided for oxygen saturations below 95%. Because of concerns about the potential for respiratory depression, as part of our standard management protocol, naloxone was ordered on an as-needed basis for all patients receiving PNCA. Complications (including vomiting, pruritus, and respiratory depression) were recorded from the nursing data record.
Data collection was terminated on discontinuation of PNCA or on completion of 5 days of PNCA therapy. Data are reported as mean ± SD for normally distributed data; skewed data are presented as median (25th–75th percentiles). Data were analyzed by using descriptive statistics, 
2 analysis, analysis of variance, and Wilcoxon ranked sum for nonparametric data where appropriate. Significance was defined as P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
|
|
|
3/10 or 2/5 were recorded in 81%–95% of patients throughout the study period. Efficacy of pain control was not influenced by surgical site, patient age, sex, or the presence of developmental delay.
|
2 yr of age had significantly more vomiting on Day 1 than patients <2 yr of age (26% vs 4%, P < 0.05). The incidence of vomiting decreased with each day of PNCA use, and there was no association with age on Days 2–5. Pruritus occurred during 8% (20 of 240) of PNCA usages on Day 1, peaked at 14% (31 of 228) on Day 2, and declined to 7% (5 of 74) by Day 5.
On the first day of PNCA therapy, in 25% (54/217) of the instances during which PNCA was administered to unintubated patients, supplemental oxygen was provided to maintain oxygen saturations 95%. During the study, nine patients (4%) received naloxone on 10 occasions (Table 5). One additional patient whose trachea was intubated and whose ventilation was mechanically controlled accidentally received a 10-fold drug overdose as a result of a medication concentration error. This patient did not require naloxone, and her apnea resolved after transient discontinuation of the PNCA and correction of the concentration error.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
3/10 or 2/5, in more than 80% of children on all five days of study. These results are comparable with those seen in adolescents receiving bolus-only PCA or PCA plus a basal infusion (2). Direct comparison to PCA is limited, however, by the fact that older patients generally self-rate pain using a visual analog scale, whereas almost 90% of our patients had their pain assessed by nurses using observational pain scales. Prior studies have shown that in older patients objective measurements of pain may correlate with, but generally underrate, pain when compared with direct patient rating (2,12). Hence pain may have been under-diagnosed in our young patients. Conversely, because pediatric observational pain scales assess surrogate measures of pain, such as crying, which may be caused by other factors, such as hunger or a desire to be held, incorrect presumptions (and overtreatment) of pain may have occurred as well. Although we do not know the relative contribution of each of these biases to our assessment of pain, we have tried to err on the side of overestimating pain in evaluating the efficacy of PNCA. We used maximum, as opposed to average, daily pain scores and used an objective pain score 3/10 as our criteria for effective analgesia. Even given these constraints, PNCA appears to provide superior analgesia to that reported when prn dosing of analgesics in young children is assessed by objective measures (5).
Although PNCA was safe in the large majority of patients studied, some patients did experience opioid-induced side effects, most notably respiratory compromise. On Day 1, in 25% of the instances during which PNCA was administered to unintubated patients, supplemental oxygen was required to maintain oxygen saturation 95%. Given our complex patient population (which included patients admitted to the PICU, neonates and young infants, former preterm infants, children with chronic lung disease, and those with developmental delay), as well as the diverse nature of procedures performed, we could not clearly differentiate those patients who required supplemental oxygen solely to manage hypoxemia caused by PNCA from those who would be expected to require oxygen to treat hypoxemia caused by baseline respiratory disease or postoperative respiratory compromise. However, using the criterion that oxygen saturation be maintained at >94%, our frequent use of supplemental oxygen was not unexpected. In a small study, Tyler et al. (13) previously demonstrated that preoperatively or postoperatively, while receiving opioid analgesics, more than 70% of children breathing room air will spend some time with an oxygen saturation <90%. Many of these children would receive supplemental oxygen in our standard management protocol.
We therefore used the discrete event of naloxone administration to identify patients who may have developed significant respiratory compromise primarily as a result of PNCA-based opioid administration. Naloxone was administered to nine patients, four of whom (1.7%, confidence interval: 0.5%–4.2%) received the opioid antagonist to treat PNCA-related apnea or episodes of desaturation. Infants less than three months of age were not at increased risk of respiratory compromise, except on the basis of postoperative apnea; however, because only four patients less than one month of age received PNCA, one of whom remained intubated after his episode of postoperative apnea, we cannot be sure that PNCA is without added risk to these extremely young patients. No specific risk factor was statistically associated with naloxone administration, but individual clinical characteristics may have predisposed some of our patients to excessive sedation or respiratory compromise, including coexisting disease and the concomitant use of sedative drugs (Table 4). Although not a statistically significant result in our series, the association between respiratory depression and concurrent administration of opioids and sedatives has previously been reported in surveys of PCA-related complications in adults (14,15).
In adults, the incidence of PCA-associated respiratory complications has been reported to be between 0.2% and 0.7% (14–16). In a large series of adolescent patients, the incidence of hypoventilation related to PCA use was approximately 1% (17). Our incidence of respiratory events in young patients might be expected to be somewhat higher than that observed in older patients receiving PCA, given our use of a basal infusion (14,15,18), as well as the loss of the "safety mechanism" of only the patient triggering the pump (8,15,16). These limitations are especially notable in our infant who received a fentanyl cassette with a 10-fold concentration error. In terms of comparing PNCA with other therapeutic modalities, we do not know whether PNCA causes more or fewer respiratory complications than does routine prn dosing of opioids in children because there are little data available for comparison (19). However, in a small series of children receiving continuous IV infusions of morphine to manage pain after cardiac surgery, 9% of patients received naloxone to treat hypoventilation (7). Thus, this finding suggests that the respiratory compromise seen with PNCA may be similar to that seen when titration of a continuous opioid infusion is the primary method of managing postoperative pain.
Because the risk of respiratory compromise from PNCA exists but could not be predicted a priori in individual patients, we do not advocate restricting access to a select, "low-risk" subpopulation of patients. This approach would have denied treatment to many children who tolerated PNCA well, without necessarily eliminating the incidence of serious complications. Particular attention, however, must be paid to each patient’s coexisting medical problems, as well as the use of additional sedative medications, both of which may act to decrease the safety margin of the technique. In addition, we advocate that PNCA only be used in settings where adequate resources are available to minimize the risk of serious complications, and to intervene rapidly and effectively if they occur. In our large teaching hospital, this includes a Pediatric Pain Service with 24-h, 7-days a week on-call expert availability; extensive patient monitoring and nursing protocols; and nursing, parent (20), and physician education (9).
Although the incidence of PCA-related respiratory compromise has been found to be small, previous studies have shown that many patients who receive IV PCA experience other opioid-related side effects. For example, nausea and vomiting have been reported in 30%–50% of children receiving PCA (21,22), and pruritus has been reported in up to 20% of patients (22). We observed a similar incidence of these complications in our patients receiving PNCA. Although we did not find an association between the incidence of these complications and the choice of opioid administered, age may have affected the incidence of vomiting, because less vomiting was reported in children less than two years of age on the first postoperative day. Whether this smaller incidence was due to a resistance to opioid-induced or postanesthetic vomiting or to underreporting (occasional vomiting or "spitting" is frequently viewed as expected in infants) of this side effect in this preverbal population is unknown. Given the frequent occurrence of these common "nuisance" complications, however, if PNCA is instituted, treatment paradigms should be in place to treat these side effects.
There are a number of limitations in this study. Because this was strictly an observational study, and because PNCA has become a standard treatment for moderate-to-severe pain in children in our institution, we did not randomize patients to PNCA, bolus-only PNCA, or prn nurse-administered analgesics. Hence, we can only compare our safety and efficacy profiles with available historical controls. Nor can we separate the relative contribution of the two components of PNCA, the basal infusion and the demand doses, to either the incidence of complications or pain relief. PCA with basal opioid infusions in older patients is controversial, in part because pain management may not be improved (23), and the addition of an infusion has been associated with an increased frequency of desaturation (3,24). However, in children, the addition of a basal infusion has also been shown to provide improved analgesia or sleep patterns (2,3). We continue to use basal infusions because our clinical impression is that they are beneficial in our youngest patients who would otherwise be totally dependent on others to provide analgesia. We use basal infusion rates that are 30%–50% of those administered to patients who receive continuous opioid infusions as their primary method of analgesia (7,25). However, these rates are higher than those generally recommended for patients receiving PCA (3). Whether altering or eliminating the basal infusion would affect analgesia, improve safety, change opioid consumption, or impact on the use of the demand component of PNCA, are all important questions that require further study.
In addition, because we do not routinely document who administers each PNCA bolus, we cannot say to what degree parents participated in their child’s pain management, and whether there were any differences in adequacy of pain control or incidence of side effects between children receiving primarily nurse-controlled or parent-controlled analgesia. We did find, however, that postoperative patients who were tracheally extubated at the conclusion of surgery received comparable opioid dosing irrespective of operative site, age, and postoperative location. This similarity suggests that parents and nurses throughout the hospital used PNCA in a consistent fashion to medicate these children. Two subgroups of patients did demonstrate different PNCA usage patterns: children who received PNCA for control of medical pain remained on PNCA longer than surgical patients, possibly because once these patients experienced pain it lasted longer. In addition, on Day 1, endotracheally intubated patients in the PICU received more opioid than unintubated patients. It is unclear whether this difference was related to increased illness and more severe pain in these patients or if PNCA was used to help sedate them.
Finally, the conversion ratio of 1:40:5 morphine: fentanyl:hydromorphone used to calculate morphine equivalents was chosen solely to allow comparison of usage patterns (i.e., the hourly combination of the basal infusion and additional bolus doses) among our three different PNCA drugs, and may not reflect equipotent doses of these opioids. IV PCA opioid potency studies comparing morphine and hydromorphone confirm that the 5:1 ratio used here to convert hydromorphone to morphine equivalents is an equianalgesic/equipotent ratio (26), but there are no analogous studies comparing either morphine or hydromorphone to fentanyl. When administered as a single bolus, fentanyl is approximately 50–100 times more potent than morphine (27). This ratio is higher than our conversion factor of 40:1, suggesting that more fentanyl may be required by patients receiving PNCA than might be expected based on its potency alone. However, this study was not designed to determine an equipotency ratio between IV PNCA fentanyl and morphine, and the difference between our conversion ratio and published potency ratios may be related to the pharmacokinetics of fentanyl, or may be skewed by our initial choice of basal and bolus opioid doses.
In conclusion, when used in accordance with our standard management protocols, PNCA produced effective analgesia in most children under six years of age who were experiencing moderate-to-severe nonsurgical or postoperative pain as assessed by objective and, when appropriate, self-report pain measures. Although this approach was safe in the large majority of patients, some patients did experience opioid-induced side effects, most notably a 1.7% incidence of apnea and episodes of desaturation requiring treatment with naloxone. These findings reinforce the need for treatment paradigms and close patient monitoring to minimize risk and allow for effective intervention whenever PNCA is used.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. D. Yates Parent or Nurse-Controlled Analgesia for Pain Control in Developmentally Delayed Children AAP Grand Rounds, March 1, 2009; 21(3): 32 - 32. [Full Text] [PDF]
|
|
|
|
 |
|
 T. Voepel-Lewis, A. Marinkovic, A. Kostrzewa, A. R. Tait, and S. Malviya The Prevalence of and Risk Factors for Adverse Events in Children Receiving Patient-Controlled Analgesia by Proxy or Patient-Controlled Analgesia After Surgery Anesth. Analg., July 1, 2008; 107(1): 70 - 75. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. El-Tahtawy, H. Kokki, and B. E. Reidenberg Population pharmacokinetics of oxycodone in children 6 months to 7 years old. J. Clin. Pharmacol., April 1, 2006; 46(4): 433 - 442. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. L. Anghelescu, L. L. Burgoyne, L. L. Oakes, and D. A. Wallace The Safety of Patient-Controlled Analgesia by Proxy in Pediatric Oncology Patients Anesth. Analg., December 1, 2005; 101(6): 1623 - 1627. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. G. Maxwell, S. C. Kaufmann, S. Bitzer, E. V. Jackson Jr, J. McGready, S. Kost-Byerly, L. Kozlowski, S. K. Rothman, and M. Yaster The Effects of a Small-Dose Naloxone Infusion on Opioid-Induced Side Effects and Analgesia in Children and Adolescents Treated with Intravenous Patient-Controlled Analgesia: A Double-Blind, Prospective, Randomized, Controlled Study Anesth. Analg., April 1, 2005; 100(4): 953 - 958. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. C. Walker and D. S. Wagner Treatment of Pain in Pediatric Patients Journal of Pharmacy Practice, August 1, 2003; 16(4): 261 - 275. [Abstract] [PDF]
|
|
|
|
|
|
P. K. Birmingham, M. Wheeler, S. Suresh, R. M. Dsida, B. R. Rae, J. Obrecht, V. A. Andreoni, S. C. Hall, and C. J. Cote Patient-Controlled Epidural Analgesia in Children: Can They Do It? Anesth. Analg., March 1, 2003; 96(3): 686 - 691. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. B. Berde and N. F. Sethna Analgesics for the Treatment of Pain in Children N. Engl. J. Med., October 3, 2002; 347(14): 1094 - 1103. [Full Text] [PDF]
|
|
|
|
|
|
T. Mancuso Parent/Nurse Controlled Analgesia Is Safe and Effective AAP Grand Rounds, January 1, 2001; 5(1): 3 - 4. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
), or >3 of 10 or 2 of 5 (
) on Days 1–5 of PNCA use. PNCA = parent-/nurse-controlled analgesia.
