Patient-controlled analgesia in children and adolescents: A randomized, prospective comparison with intramuscular administration of morphine for postoperative analgesia Charles B. Berde, ME), PhD, B e a t e M. Lehn, DiptPsych, John D, Y e e , MD, Navil F. Sethna, MB,ChB, a n d Dennis Russo, PhD From the Departments of Anesthesia and Psychiatry,Children's Hospital and Harvard Medical School, and the Department of Behavioral and Pediatric Programming, NewMedico Associates, Inc., Boston, Massachusetts
A randomized, prospective trial of patient-controlled analgesia (PCA), that is, a method of analgesia administration involving a computer-driven pump activated by patients to receive small doses within defined limits, was performed in 82 children and adolescents after major orthopedic surgery to c o m p a r e (I) intramuscularly administered morphine, (2) PCA morphine, and (3) PCA morphine with a low-dose continuous morphine infusion (PCA-plus). Patients receiving PCA and PCA-plus had lower pain scores and greater satisfaction than patients receiving intramuscularly administered morphine. The three groups used equal amounts of morphine and most measures of recovery were identical in the groups. In particular, PCA and PCA-plus did not increase the incidence of opioid-related complications, and patients receiving PCA-plus were less sedated than patients receiving intramuscular therapy. We conclude that PCA and PCAplus are safe and effective methods of pain relief in children and adolescents after orthopedic surgery, are better a c c e p t e d than intramuscular injections, and do not increase perioperative morbidity. (J PEDIATR1991;118:460-6)
Severe postoperative pain in children and adolescents is most commonly managed by intramuscular injections of opioids such as morphine or meperidine. 1"3 This approach may result in fluctuating plasma drug levels and cycles of pain, comfort, and sedation. 4 In addition, the fear and distress caused by frequent injections may cause children to
Supported in part by a grant from Abbott Laboratories (Dr. Berde), and by contributions from the Christopher Coakley Memorial Fund (Dr. Berde), from the Karen Grunebaum Cancer Research Fund (Dr. Yee), and from a Student Research Award, Federal Republic of Germany (Dr. Lehn). Submitted for publication May 7, 1990; accepted Oct. 9, 1990. Reprint requests: Charles B. Berde, MD, PhD, Department of Anesthesia, Children's Hospital, 300 Longwood Ave., Boston, MA 02115. 9/25/26019 460
underreport their level of pain. 5 The problem of painful injections can be overcome by administering intermittent doses through an intravenous catheter; however, the problem of cyclic pain and comfort remains. Continuous intravenous opioid infusions have been used to provide a steadier ANOVA IM PCA PCA-plus VAS
Analysis of variance Intramuscularly administered Patient-contro'lled analgesia Patient-controlled analgesia plus continuous infusion Visual analog scale
plasma drug level,69 but patients vary widely (up to fivefold 1~ in their opioid clearances and effective analgesic levels; "average" infusion rates or plasma concentrations may lead to unacceptable levels of sedation or inadequate analgesia in certain patients.
Volume 118 Number 3
Patient-controlled analgesia in children and adolescents
Patient-controlled analgesia is achieved by a new method of analgesia administration that consists of a computerdriven pump with a button that the patient may press to administer a small dose of analgesic drug. IH3 The machines are designed with several built-in safety features to prevent overdosing. The patient controls the frequency of dosing, but the machines are programmed to control the size of each dose, the minimum amount of time the patient must wait between each dose (lockout interval), and the total amount of drug that can be administered within a given period (typically for 3 to 4 hours). Patient-controlled analgesia has several theoretic advantages over other methods of analgesia delivery. 1418 This method allows the patient to titrate the amount of drug, balancing analgesia against sedation and other side effects. It may allow steadier plasma drug levels and thus less fluctuation in the level of analgesi'a, avoiding the cycles of pain. Intramuscular injections are avoided, reducing distress for the patient as well as potentially saving time for the nursing staff and improving the quality of nurse-patient interaction. By allowing the patient to retain control of the dosing, PCA may also have psychologic advantages over traditional methods of analgesia administration. Preliminary reports regarding PCA in children are encouraging, demonstrating effective analgesia for most children with safety and acceptable rates of opioid-associated side effects. 19"24 Most of the pediatric populations studied to date have not been randomly assigned to treatment and control groups, and drug dosing has generally not been standardized. Our prospective, randomized, controlled study compared intramuscular administration of morphine, PCA alone, and PCA plus a continuous low-dose infusion for analgesia after major orthopedic surgery. We sought to address several questions: (1) Can PCA and PCA-plus be used safely and effectively in children and adolescents aged 7 to 19 years? (2) Do IM morphine, PCA, and PCA-plus produce different degrees of pain relief postoperatively? (3) Do these methods differ in the rates of side effects; amount of drug used, or patient satisfaction? (4) Are psychologic modifiers such as anxiety and depression related to the eff• of PCA use? (5) Are there age- and development-related modifiers of PCA efficacy and side effects. Questions 1 to 3 are the subject of this report. Questions 4 and 5 will be the subject of a separate publication. METHODS Subjects. Patients were enrolled from February 1989 until July 1989 at Children's Hospital, Boston, Mass., according to a protocol approved by the committee on clinical investigation and with informed parental consent and patient assent. Enrollment was opened to patients between the ages of 7 and 19 years who had been scheduled for elective orthopedic surgery, with significant postoperative pain antic-
4 61
ipated and with a predicted postoperative hospital stay of at 1east 2 days. Orthopedic procedures were chosen to standardize the type of postoperative pain and to minimize the postoperative side effects related to the procedure (e.g., ileus or pulmonary complications after abdominal or thoracic surgery) in contradistinction to those side effects related to general anesthesia and postoperative administration of analgesia. Patients were interviewed during a routine preoperative visit at least 1 day before the day of surgery. Potential participants were screened for the ability to comprehend the concepts of PCA and visua ! analog rating scales. Patients were excluded if they had a history of allergy or adverse reaction to morphine, a history of chronic opioid use, or any major cardiac, respiratory, hepatic, renal, neurologic, or psychiatric disease. Patients undergoing posterior spine fusion or surgery for bone malignancies were excluded from the study, in the former case because of the altered anesthetic requirements posed by the "wakeup" test and in the latter case because of the associated psychologic burden associated with surgery for malignancy." One hundred nineteen patients were identified for the study. Twenty patients (parents) chose not to participate. Among them, 10 patients (parents) objected to the additional burden of participating in a study. Four patients stated medical reasons such as preference for another method of pain management or anticipation of only minimal postoperative pain. One patient chose not to participate because of concerns about the safety of the PCA machine. For five patients the reasons remained unknown to the investigators. Of the 99 patients who agreed to participate in the study, 17 did not complete the protocol for the following reasons: anesthetic or surgical factors that altered the patient's eligibility (10 patients); lack of available beds on nursing units participating in the study (five); and other reasons (two). A total of 82 subjects completed the protocol and were included in the data analysis. Preoperative procedure. During the preoperative visit, the subjects and parents were asked to complete the Center for Epidemiological Studies Depression Scale 25 and the Revised Children's Manifest Anxiety Scale. 26 The parents were asked to complete a questionnaire on the child's past experience with medical procedures. A 10 cm (0 to 10 continuous scale) VAS for the assessment of pain, nausea, anxiety, sedation, and satisfaction was explained to each subject. To ensure proper understanding, we asked the subjects to mark pain of different intensities on the VAS (e.g., pain after a mosquito bite, an injection, a fall from a bike). Group assignment. After consent was obtained, the subjects were assigned to one of three study groups by the use of a random numbers table. Of patients who completed the study, 23 were assigned to the control group (IM morphine), 32 were assigned to the PCA group (intermittent patient-
462
Berde et al.
controlled intravenous morphine doses), and 27 were assigned to the PCA-plus group (intermittent patient-controlled intravenous morphine doses plus continuous intravenous morphine background infusion). The subjects were informed of their group assignment postoperatively. Perioperative management. A standardized protocol for anesthesia and recovery room management was used for all subjects to diminish residual anesthetic effects as a source of variation in subsequent requirements for analgesia. Induction proceeded with either thiopental or a combination of halothane and nitrous oxide. A single intravenous dose of morphine sulfate (0.15 mg/kg) was given shortly after induction of anesthesia. No other opioid analgesics and no antiemetics or other fixed sedative-hypnotics were administered intraoperatively. Vecuronium was used for neuromuscular blockade. Anesthesia was maintained with a nitrous oxide-oxygen combination (65%:35%) and isoflurane (0.2% to 1.2%). In the recovery room, subjects received morphine sulfate, 0.05 mg/kg given intravenously every 10 minutes as needed until comfort was achieved. Subjects remained in the recovery room for 30 minutes after the last morphine dose. The PCA, PCA-plus, or I M morphine was initiated on subjects' arrival on the nursing units. The study was conducted on orthopedic and general surgery units familiar with the use of parenteral opioid analgesics; in-service training of the nursing staff on the use of PCA took place before the beginning of this study. Postoperative analgesia. Standardized protocols were used for the treatment of postoperative pain and for side effects of analgesia. The control group received morphine sulfate, 0.1 to 0.15 mg/kg IM every 3 hours as needed and increased to 0.18 mg/kg if necessary. No other opioid analgesics were administered during the study period. For the subjects in the PCA and PCA-plus groups, the Abbott Life Care PCA Infuser System (Abbott Laboratories, Chicago, Ill.) was employed. The system consists of a microprocessor-controlled infusion syringe connected to the patient's intravenous catheter through a Y tube with a oneway valve to prevent drug reflux. Morphine sulfate is provided in premade 30 ml vials containing 1 mg/ml. The patient activates the system by pressing a button similar to a nurse call button on a cord extending from the unit. The machine can be programmed in three modes: (1) PCA only, (2) continuous infusion, and (3) PCA plus continuous infusion. Only modes 1 and 3 were used for this study. The machine records several variables, including the cumulative opioid requirement summarized during the previous hour and the previous 12 hours; the duration and amount of each successful delivery of a dose of morphine; the number of demands made by the patient that did not result in an actual dose administration (button pressed during lockout in-
The Journal o f Pediatrics March 1991
terval); and the time and frequency of reaching the 4-hour limit. The machines for the subjects receiving PCA only were programmed for the following variables: amount of morphine for each dose--0.025 mg/kg; lockout time = 10 minutes (additional dose will not be given for 10 minutes even if demand button is pushed); 4-hour limit = 0.24 mg/ kg (maximum cumulative dose of morphine permitted for a 4-hour interval). The 4-hour PCA limit of 0.24 mg/kg was equivalent to the 3-hour IM morphine limit of 0.18 mg/kg (0.06 mg/kg per hour). The machines for the PCA-plus group were programmed as follows: amount of morphine for each dose = 0.018 mg/ kg; continuous background morphine dose = 0.015 mg/kg per hour; lockout time-- 10 minutes; and 4-hour limit -- 0.24 mg/kg. The continuous rate of 0.015 mg/kg per hour was chosen because it represents 30% to 50% of average hourly requirements of children receiving continuous infusions for postoperative pain or cancer pain. 7-9 Assessment of pain intensity and side effects. Patients and nurses were asked to record data for 48 hours after the IM morphine, PCA, or PCA-plus protocol was initiated. Patients in all three groups were asked to assess their pain, sedation, nausea, anxiety, and satisfaction every 2 hours postoperatively (except when asleep), utilizing a 10 cm (0 to 10 continuous scale) VAS for each variable. 27 Nurses were asked to assess pain, sedation, nausea, and anxiety at the same time, using a VAS. 28 Scores for each assessment were recorded on separate forms to diminish the influence of prior scores on the present assessment. In addition, nurses were asked to record the following variables: asleep or awake (every hour), respiratory rate (every hour), and pulse and blood pressure (every 4 hours). Vomiting, urinary retention (need for catheterization), and oral intake were recorded. Two nurses were asked to complete a global assessment of the experience for each patient--one on the night shift of postoperative day 0-1 and one on the day shift of postoperative day 1. Patients and parents were also asked to complete a global assessment at 48 hours or at the time of leaving the study. Antiemetic (perphenazine) and antipruritic (diphenhydramine) drugs were administered intravenously as needed for patients in all three groups. A rough scale of operative severity was developed as follows: three orthopedic surgeons were given a list of the 82 procedures performed. They were asked to rate the severity of each procedure in terms of the extensiveness of the procedure and the amount of postoperative pain expected. Study termination. Subjects were allowed to switch to orally administered opioids at any time after 8 AM on the second postoperative day if they were tolerating oral fluids and solids well. They ended the study at that time. If subjects were unable to take opioids orally by the second post-
Volume 118 Number 3
Patient-controlled analgesia in children and adolescents
463
Table. Demographics of the three study groups Study group
Age (yr) Gender Male Female Weight (kg) Preoperative depression Preoperative anxiety Physician-predicted postoperative pain Extensiveness of procedure Operating room time (hr) Recovery room time (hr) Morphine in recovery room (mg/kg) Length of hospital stay (days) Values (except for gender)are expressed as mean + SD.
IM MORPHINE
PCA
PCA-PLUS
p
13.65 _+ 3.31
14.28 --+ 3.42
13.41 + 2.82
0.56
17 (65.4%) 9 (34.6%) 53.52 _+ 17.44 14.41 + 7.81 47.39 _+ 10.51 4.70 _+ 1.08 4.77 _+ 1.27 2.34 + 1.17 1.69 _+ 0.59 0.078 _+ 0.067 5.04 _+ 1.80
15 (46.9%) 17 (53.1%) 54.95 _+ 14.16 15.38 _+ 11.95 46.00 -+ 9.02 4.79 _+ 1.66 4.73 + 1.66 2.36 + 1.36 1.55 _+ 0.73 0.046 _+ 0.055 5.41 --+ 2.61
15 (55.6%) 12 (44.4%) 53.44 _+ 19.50 9.85 + 8.47 45.19 + 9.50 4.17 + 1.17 4.36 -+ 1.26 1.82 _+ 0.79 1.50 - 0.79 0.039 _+ 0.067 4.67 + 1.98
0.4 0.93 0.10 0.73 0.19 0.51 0.15 0.63 0.07 0.56
operative day, they were either kept on their previous regimen or started on another regimen at the discretion of their physician. The study was terminated after 48 hours for all subjects. Data analysis. Data from the patient and nursing VAS ratings were analyzed in two separate ways. The mean score for each variable was calculated for each subject by 8-hour time blocks (e.g., postoperatively to 11 PM [variable length], 11 PM to 7 AM, 7 AM to 3 PM, 3 PM to 11 PM). These data were analyzed by repeated-measures ANOVA. In the second method, all the raw data (0 to 10 scale) for each variable were categorized as mild (
A randomized, prospective trial of patient-controlled analgesia (PCA), that is, a method of analgesia administration involving a computer-driven pump activated by patients to receive small doses within defined limits, was performed in 82 children and adolescents after major orthopedic surgery to c o m p a r e (I) intramuscularly administered morphine, (2) PCA morphine, and (3) PCA morphine with a low-dose continuous morphine infusion (PCA-plus). Patients receiving PCA and PCA-plus had lower pain scores and greater satisfaction than patients receiving intramuscularly administered morphine. The three groups used equal amounts of morphine and most measures of recovery were identical in the groups. In particular, PCA and PCA-plus did not increase the incidence of opioid-related complications, and patients receiving PCA-plus were less sedated than patients receiving intramuscular therapy. We conclude that PCA and PCAplus are safe and effective methods of pain relief in children and adolescents after orthopedic surgery, are better a c c e p t e d than intramuscular injections, and do not increase perioperative morbidity. (J PEDIATR1991;118:460-6)
Severe postoperative pain in children and adolescents is most commonly managed by intramuscular injections of opioids such as morphine or meperidine. 1"3 This approach may result in fluctuating plasma drug levels and cycles of pain, comfort, and sedation. 4 In addition, the fear and distress caused by frequent injections may cause children to
Supported in part by a grant from Abbott Laboratories (Dr. Berde), and by contributions from the Christopher Coakley Memorial Fund (Dr. Berde), from the Karen Grunebaum Cancer Research Fund (Dr. Yee), and from a Student Research Award, Federal Republic of Germany (Dr. Lehn). Submitted for publication May 7, 1990; accepted Oct. 9, 1990. Reprint requests: Charles B. Berde, MD, PhD, Department of Anesthesia, Children's Hospital, 300 Longwood Ave., Boston, MA 02115. 9/25/26019 460
underreport their level of pain. 5 The problem of painful injections can be overcome by administering intermittent doses through an intravenous catheter; however, the problem of cyclic pain and comfort remains. Continuous intravenous opioid infusions have been used to provide a steadier ANOVA IM PCA PCA-plus VAS
Analysis of variance Intramuscularly administered Patient-contro'lled analgesia Patient-controlled analgesia plus continuous infusion Visual analog scale
plasma drug level,69 but patients vary widely (up to fivefold 1~ in their opioid clearances and effective analgesic levels; "average" infusion rates or plasma concentrations may lead to unacceptable levels of sedation or inadequate analgesia in certain patients.
Volume 118 Number 3
Patient-controlled analgesia in children and adolescents
Patient-controlled analgesia is achieved by a new method of analgesia administration that consists of a computerdriven pump with a button that the patient may press to administer a small dose of analgesic drug. IH3 The machines are designed with several built-in safety features to prevent overdosing. The patient controls the frequency of dosing, but the machines are programmed to control the size of each dose, the minimum amount of time the patient must wait between each dose (lockout interval), and the total amount of drug that can be administered within a given period (typically for 3 to 4 hours). Patient-controlled analgesia has several theoretic advantages over other methods of analgesia delivery. 1418 This method allows the patient to titrate the amount of drug, balancing analgesia against sedation and other side effects. It may allow steadier plasma drug levels and thus less fluctuation in the level of analgesi'a, avoiding the cycles of pain. Intramuscular injections are avoided, reducing distress for the patient as well as potentially saving time for the nursing staff and improving the quality of nurse-patient interaction. By allowing the patient to retain control of the dosing, PCA may also have psychologic advantages over traditional methods of analgesia administration. Preliminary reports regarding PCA in children are encouraging, demonstrating effective analgesia for most children with safety and acceptable rates of opioid-associated side effects. 19"24 Most of the pediatric populations studied to date have not been randomly assigned to treatment and control groups, and drug dosing has generally not been standardized. Our prospective, randomized, controlled study compared intramuscular administration of morphine, PCA alone, and PCA plus a continuous low-dose infusion for analgesia after major orthopedic surgery. We sought to address several questions: (1) Can PCA and PCA-plus be used safely and effectively in children and adolescents aged 7 to 19 years? (2) Do IM morphine, PCA, and PCA-plus produce different degrees of pain relief postoperatively? (3) Do these methods differ in the rates of side effects; amount of drug used, or patient satisfaction? (4) Are psychologic modifiers such as anxiety and depression related to the eff• of PCA use? (5) Are there age- and development-related modifiers of PCA efficacy and side effects. Questions 1 to 3 are the subject of this report. Questions 4 and 5 will be the subject of a separate publication. METHODS Subjects. Patients were enrolled from February 1989 until July 1989 at Children's Hospital, Boston, Mass., according to a protocol approved by the committee on clinical investigation and with informed parental consent and patient assent. Enrollment was opened to patients between the ages of 7 and 19 years who had been scheduled for elective orthopedic surgery, with significant postoperative pain antic-
4 61
ipated and with a predicted postoperative hospital stay of at 1east 2 days. Orthopedic procedures were chosen to standardize the type of postoperative pain and to minimize the postoperative side effects related to the procedure (e.g., ileus or pulmonary complications after abdominal or thoracic surgery) in contradistinction to those side effects related to general anesthesia and postoperative administration of analgesia. Patients were interviewed during a routine preoperative visit at least 1 day before the day of surgery. Potential participants were screened for the ability to comprehend the concepts of PCA and visua ! analog rating scales. Patients were excluded if they had a history of allergy or adverse reaction to morphine, a history of chronic opioid use, or any major cardiac, respiratory, hepatic, renal, neurologic, or psychiatric disease. Patients undergoing posterior spine fusion or surgery for bone malignancies were excluded from the study, in the former case because of the altered anesthetic requirements posed by the "wakeup" test and in the latter case because of the associated psychologic burden associated with surgery for malignancy." One hundred nineteen patients were identified for the study. Twenty patients (parents) chose not to participate. Among them, 10 patients (parents) objected to the additional burden of participating in a study. Four patients stated medical reasons such as preference for another method of pain management or anticipation of only minimal postoperative pain. One patient chose not to participate because of concerns about the safety of the PCA machine. For five patients the reasons remained unknown to the investigators. Of the 99 patients who agreed to participate in the study, 17 did not complete the protocol for the following reasons: anesthetic or surgical factors that altered the patient's eligibility (10 patients); lack of available beds on nursing units participating in the study (five); and other reasons (two). A total of 82 subjects completed the protocol and were included in the data analysis. Preoperative procedure. During the preoperative visit, the subjects and parents were asked to complete the Center for Epidemiological Studies Depression Scale 25 and the Revised Children's Manifest Anxiety Scale. 26 The parents were asked to complete a questionnaire on the child's past experience with medical procedures. A 10 cm (0 to 10 continuous scale) VAS for the assessment of pain, nausea, anxiety, sedation, and satisfaction was explained to each subject. To ensure proper understanding, we asked the subjects to mark pain of different intensities on the VAS (e.g., pain after a mosquito bite, an injection, a fall from a bike). Group assignment. After consent was obtained, the subjects were assigned to one of three study groups by the use of a random numbers table. Of patients who completed the study, 23 were assigned to the control group (IM morphine), 32 were assigned to the PCA group (intermittent patient-
462
Berde et al.
controlled intravenous morphine doses), and 27 were assigned to the PCA-plus group (intermittent patient-controlled intravenous morphine doses plus continuous intravenous morphine background infusion). The subjects were informed of their group assignment postoperatively. Perioperative management. A standardized protocol for anesthesia and recovery room management was used for all subjects to diminish residual anesthetic effects as a source of variation in subsequent requirements for analgesia. Induction proceeded with either thiopental or a combination of halothane and nitrous oxide. A single intravenous dose of morphine sulfate (0.15 mg/kg) was given shortly after induction of anesthesia. No other opioid analgesics and no antiemetics or other fixed sedative-hypnotics were administered intraoperatively. Vecuronium was used for neuromuscular blockade. Anesthesia was maintained with a nitrous oxide-oxygen combination (65%:35%) and isoflurane (0.2% to 1.2%). In the recovery room, subjects received morphine sulfate, 0.05 mg/kg given intravenously every 10 minutes as needed until comfort was achieved. Subjects remained in the recovery room for 30 minutes after the last morphine dose. The PCA, PCA-plus, or I M morphine was initiated on subjects' arrival on the nursing units. The study was conducted on orthopedic and general surgery units familiar with the use of parenteral opioid analgesics; in-service training of the nursing staff on the use of PCA took place before the beginning of this study. Postoperative analgesia. Standardized protocols were used for the treatment of postoperative pain and for side effects of analgesia. The control group received morphine sulfate, 0.1 to 0.15 mg/kg IM every 3 hours as needed and increased to 0.18 mg/kg if necessary. No other opioid analgesics were administered during the study period. For the subjects in the PCA and PCA-plus groups, the Abbott Life Care PCA Infuser System (Abbott Laboratories, Chicago, Ill.) was employed. The system consists of a microprocessor-controlled infusion syringe connected to the patient's intravenous catheter through a Y tube with a oneway valve to prevent drug reflux. Morphine sulfate is provided in premade 30 ml vials containing 1 mg/ml. The patient activates the system by pressing a button similar to a nurse call button on a cord extending from the unit. The machine can be programmed in three modes: (1) PCA only, (2) continuous infusion, and (3) PCA plus continuous infusion. Only modes 1 and 3 were used for this study. The machine records several variables, including the cumulative opioid requirement summarized during the previous hour and the previous 12 hours; the duration and amount of each successful delivery of a dose of morphine; the number of demands made by the patient that did not result in an actual dose administration (button pressed during lockout in-
The Journal o f Pediatrics March 1991
terval); and the time and frequency of reaching the 4-hour limit. The machines for the subjects receiving PCA only were programmed for the following variables: amount of morphine for each dose--0.025 mg/kg; lockout time = 10 minutes (additional dose will not be given for 10 minutes even if demand button is pushed); 4-hour limit = 0.24 mg/ kg (maximum cumulative dose of morphine permitted for a 4-hour interval). The 4-hour PCA limit of 0.24 mg/kg was equivalent to the 3-hour IM morphine limit of 0.18 mg/kg (0.06 mg/kg per hour). The machines for the PCA-plus group were programmed as follows: amount of morphine for each dose = 0.018 mg/ kg; continuous background morphine dose = 0.015 mg/kg per hour; lockout time-- 10 minutes; and 4-hour limit -- 0.24 mg/kg. The continuous rate of 0.015 mg/kg per hour was chosen because it represents 30% to 50% of average hourly requirements of children receiving continuous infusions for postoperative pain or cancer pain. 7-9 Assessment of pain intensity and side effects. Patients and nurses were asked to record data for 48 hours after the IM morphine, PCA, or PCA-plus protocol was initiated. Patients in all three groups were asked to assess their pain, sedation, nausea, anxiety, and satisfaction every 2 hours postoperatively (except when asleep), utilizing a 10 cm (0 to 10 continuous scale) VAS for each variable. 27 Nurses were asked to assess pain, sedation, nausea, and anxiety at the same time, using a VAS. 28 Scores for each assessment were recorded on separate forms to diminish the influence of prior scores on the present assessment. In addition, nurses were asked to record the following variables: asleep or awake (every hour), respiratory rate (every hour), and pulse and blood pressure (every 4 hours). Vomiting, urinary retention (need for catheterization), and oral intake were recorded. Two nurses were asked to complete a global assessment of the experience for each patient--one on the night shift of postoperative day 0-1 and one on the day shift of postoperative day 1. Patients and parents were also asked to complete a global assessment at 48 hours or at the time of leaving the study. Antiemetic (perphenazine) and antipruritic (diphenhydramine) drugs were administered intravenously as needed for patients in all three groups. A rough scale of operative severity was developed as follows: three orthopedic surgeons were given a list of the 82 procedures performed. They were asked to rate the severity of each procedure in terms of the extensiveness of the procedure and the amount of postoperative pain expected. Study termination. Subjects were allowed to switch to orally administered opioids at any time after 8 AM on the second postoperative day if they were tolerating oral fluids and solids well. They ended the study at that time. If subjects were unable to take opioids orally by the second post-
Volume 118 Number 3
Patient-controlled analgesia in children and adolescents
463
Table. Demographics of the three study groups Study group
Age (yr) Gender Male Female Weight (kg) Preoperative depression Preoperative anxiety Physician-predicted postoperative pain Extensiveness of procedure Operating room time (hr) Recovery room time (hr) Morphine in recovery room (mg/kg) Length of hospital stay (days) Values (except for gender)are expressed as mean + SD.
IM MORPHINE
PCA
PCA-PLUS
p
13.65 _+ 3.31
14.28 --+ 3.42
13.41 + 2.82
0.56
17 (65.4%) 9 (34.6%) 53.52 _+ 17.44 14.41 + 7.81 47.39 _+ 10.51 4.70 _+ 1.08 4.77 _+ 1.27 2.34 + 1.17 1.69 _+ 0.59 0.078 _+ 0.067 5.04 _+ 1.80
15 (46.9%) 17 (53.1%) 54.95 _+ 14.16 15.38 _+ 11.95 46.00 -+ 9.02 4.79 _+ 1.66 4.73 + 1.66 2.36 + 1.36 1.55 _+ 0.73 0.046 _+ 0.055 5.41 --+ 2.61
15 (55.6%) 12 (44.4%) 53.44 _+ 19.50 9.85 + 8.47 45.19 + 9.50 4.17 + 1.17 4.36 -+ 1.26 1.82 _+ 0.79 1.50 - 0.79 0.039 _+ 0.067 4.67 + 1.98
0.4 0.93 0.10 0.73 0.19 0.51 0.15 0.63 0.07 0.56
operative day, they were either kept on their previous regimen or started on another regimen at the discretion of their physician. The study was terminated after 48 hours for all subjects. Data analysis. Data from the patient and nursing VAS ratings were analyzed in two separate ways. The mean score for each variable was calculated for each subject by 8-hour time blocks (e.g., postoperatively to 11 PM [variable length], 11 PM to 7 AM, 7 AM to 3 PM, 3 PM to 11 PM). These data were analyzed by repeated-measures ANOVA. In the second method, all the raw data (0 to 10 scale) for each variable were categorized as mild (
