PAIN MANAGEMENT/BRIEF RESEARCH REPORT

Intravenous Morphine at 0.1 mg/kg Is Not Effective for Controlling Severe Acute Pain In the Majority of Patients Polly E. Bijur, PhD Mark K. Kenny, PhD E. John Gallagher, MD

From the Department of Emergency Medicine, Albert Einstein College of Medicine, Bronx, NY.

Study objective: The objective was to quantify the analgesic effect of a dose of intravenous morphine, 0.1 mg/kg, to emergency department (ED) patients presenting in acute, severe pain. Methods: This was a prospective convenience cohort of patients aged 21 to 65 years and presenting to an academic urban ED with acute, severe pain. Patients rated their pain intensity on a validated 11point verbal numeric rating scale ranging from 0, ‘‘no pain,’’ to 10, ‘‘worst possible pain,’’ immediately before they received 0.1 mg of intravenous morphine per kilogram of body weight and 30 minutes later. The main outcome was proportion of patients whose pain decreased by less than 50% during the 30-minute interval. Results: Of 119 patients who received intravenous morphine at 0.1 mg/kg, the average age was 42 years (SD=11 years), 55% were female patients, 65% were Hispanic, 28% were black, and 7% were classified as other. The median numeric rating scale pain score at baseline was 10 (interquartile range 9 to 10). Sixty-seven percent (95% confidence interval 58% to 76%) of the patients receiving intravenous morphine at 0.1 mg/kg reported a less than 50% decrease in pain. No patient required an opioid antagonist at any time during or after the study period. Conclusion: The data suggest that a 0.1 mg/kg dose of morphine may be too low to adequately control acute severe pain. [Ann Emerg Med. 2005;46:362-367.]

0196-0644/$-see front matter Copyright ª 2005 by the American College of Emergency Physicians. doi:10.1016/j.annemergmed.2005.03.010

INTRODUCTION Undertreatment of pain in the emergency department (ED) has received considerable attention in the past 15 years. Studies that have documented this trend have focused on the proportion of patients treated with any analgesics and opioid analgesics. Advances are being made in the provision of analgesics to ED patients, as documented by increasing proportions of patients treated.1,2 However, there is little evidence about what doses of analgesics are being administered in the ED and whether they are providing adequate analgesia. Morphine is the prototypic analgesic for treatment of severe pain in the ED. The question of how much morphine is needed to obtain satisfactory analgesia in acute severe pain has not received systematic attention in the emergency medicine literature. It is widely recognized that there is large interindividual variability in response to analgesics, thus making treatment with any single dose problematic. An early study found that, whereas 66% of patients who received a subcutaneous dose of 10 mg per 70 kg of morphine achieved moderate to complete relief, 34% of the patients reported less than moderate relief.3 362 Annals of Emergency Medicine

Dahlstrom et al4 showed that the effective analgesic concentration for morphine postoperatively ranged from 6 to 33 ng/mL. In a recent study of postoperative pain, Aubrun et al5 also found wide variability in the amount of morphine needed to achieve an adequate level of analgesia. A standard means of taking into account the heterogeneity of analgesic response in treatment is titration of dosage, with small increases of dose over short periods of time. Some emergency medicine texts recommend a range of doses (eg, 0.05 to 0.15 mg/kg)6; others provide a single dose (eg, 10 mg)7 or a single weight-based dose (0.1 mg/kg)8 with the proviso that the dose should be titrated to desired analgesic effect. There is little evidence about whether these recommendations are routinely followed in ED care. Although titration is a goal for optimal management of acute pain, a first step is to assess the recommended starting dose. Given the various recommendations for intravenous morphine, ranging from 0.05 mg/kg to 10 mg, we chose to assess the analgesic response to the recommended weight-based dose of 0.1 mg/kg. The purpose of this investigation was to quantify the proportion of patients in acute pain who had less than a 50% Volume 46, no. 4 : October 2005

Bijur, Kenny & Gallagher

Editor’s Capsule Summary What is already known on this topic Intravenous morphine is the criterion standard for the treatment of severe, acute pain. Clinicians often use suboptimal doses of morphine based on weight and overestimate its potential for causing adverse effects, specifically, respiratory depression. What question this study addressed This study examined the efficacy of intravenous morphine, given in a single dose of 0.1 mg/kg, to emergency department (ED) patients presenting with a wide variety of acute pain complaints. What this study adds to our knowledge Only one third of patients in this study achieved greater than 50% relief of pain after a 0.1 mg/kg bolus of intravenous morphine, and none had significant adverse effects requiring the administration of an opioid antagonist. How this might change clinical practice This study confirms the knowledge that an initial 0.1 mg/kg dose of intravenous morphine is inadequate to relieve pain for most ED patients and that titration of additional doses is necessary for many patients. reduction in pain intensity 30 minutes after intravenous administration of 0.1 mg/kg of morphine.

MATERIALS AND METHODS Study Design We used an observational prospective cohort design. The data were drawn from a series of 4 unpublished dose-finding randomized controlled trials in which all patients received a standard weight-based dose of morphine of 0.1 mg/kg intravenously. The patients in the control arms received the weight-based dose of morphine plus normal saline solution. These trials compared the change in pain in the control and experimental arms. Observation of change in pain of patients in the control arms over 30 minutes is the focus of the analyses summarized in this article. All patients signed informed consent. The study was approved by the institutional review board of Montefiore Medical Center. Setting The study took place in the academic medical center ED of the Montefiore Medical Center. The adult component of this urban ED receives 80,000 visits per year by patients who are predominantly black, Hispanic, and medically underserved. Selection of Participants In all 4 studies, trained research associates evaluated patients for inclusion after a preliminary physical examination was Volume 46, no. 4 : October 2005

Morphine and Severe Acute Pain performed. The research associates approached patients whose triage assessment indicated significant pain and whose attending physicians believed that opioids were warranted. Additionally, patients were eligible if they were between 21 and 65 years of age, spoke English or Spanish, or had acute pain with onset within the past 7 days. Exclusion criteria included previous use of methadone, use of other opioids or tramadol within the past 7 days, previous adverse reaction to morphine, chronic pain syndrome, altered mental status, pregnancy, use of monoamine oxidase inhibitors in the past 30 days, systolic blood pressure less than 100 mm Hg, or inability to provide informed consent. Data were collected between December 2, 2002, and December 15, 2003, and between March 15, 2004, and July 16, 2004. During this 16-month period, there was varying coverage in the ED by research staff. Throughout the period, we had coverage during the day shift Monday through Friday, 8 AM to 4 PM, and at least 16 hours on the weekends. An additional 8 to 16 hours per day were covered during evening and night shifts. Methods of Measurement Patients were asked by the research associates to rate their pain intensity at baseline and 30 minutes postbaseline. Peak analgesia from intravenous morphine is achieved within 5 minutes of administration in most patients.9,10 Clinically, 30 minutes seemed to be a reasonable time within which adequate analgesia should be achieved in patients with severe pain. Further, it is unlikely that an analgesic effect would be missed with this interval because the elimination half-life of morphine is 2 to 4 hours.9,10 Pain intensity was measured by an 11-point verbally administered numeric rating scale ranging from 0, ‘‘no pain,’’ to 10, ‘‘worst possible pain.’’ This scale has been shown to be valid and reliable compared to the visual analog scale.11 The primary measure of adequate analgesic response to morphine is percentage of reduction in pain intensity dichotomized into less than 50% versus greater than 50%. Although there are other measures, we chose a 50% or greater reduction in pain because this threshold has been used frequently in pain meta-analyses and has the appeal of quantitative simplicity and easy clinical interpretation. Patients’ age, sex, and pain location were obtained from the medical record or the patient. Ethnicity was self-reported. Additional administration of analgesics was ascertained from the records and consultation with the ED staff. Patients in this study were placed near the physicians’ and nurses’ station, directly in the line of sight of the staff and thus were under constant supervision. The research associates monitored the patients carefully as well. Vital signs were routinely measured by the nursing staff at 0, 15, and 30 minutes. For the purposes of the study, the research associates monitored the oxygen saturation, blood pressure, pulse rate, and respiratory rate at 0 and 30 minutes. Patients were reassessed by the clinical staff if systolic blood pressure was less than 100 mm Hg, pulse rate was less than 60 beats/min, or respiratory rate was less than 12 breaths/min to determine whether an opioid antagonist was Annals of Emergency Medicine 363

Bijur, Kenny & Gallagher

Morphine and Severe Acute Pain Table 1. Distribution of diagnoses. Diagnosis

No. (%)

Abdominal/pelvic Acid-peptic disease (gastritis, ulcer, dyspepsia) Appendicitis Biliary tract disease (biliary colic, cholecystitis, cholangitis) Diverticulitis Epididymitis Gastroenteritis Inflammatory bowel disease NSAP (nonspecific abdominal pain) Pancreatitis Pelvic inflammatory disease Perirectal abscess Small-bowel obstruction Spontaneous bacterial peritonitis Uterine myomata, degenerating Chest Ischemic chest pain Lung cancer Pneumonia Rib fractures Extremity Arthritis Avulsed digit Fracture/dislocation Flank/back Low back pain Pyelonephritis Renal colic HEENT Headache Otitis Total

57 (48) 3 7 8 4 1 2 6 15 3 1 1 4 1 1 8 (7) 1 1 5 1 13 (11) 5 1 7 38 (32) 3 6 29 3 (2) 2 1 119 (100)

Table 2. Characteristics of patients with less than 50% pain reduction versus 50% or greater pain reduction in 30 minutes. Patients With Patients With \50% Decrease R50% Decrease in Pain N=80 in Pain N=39 Age, y, meanGSD Sex, % (No.) Male Female Race/ethnicity, % (No.) Hispanic Black Other Location of pain, % (No.) Abdominal/pelvic Flank/back Extremity Chest HEENT Baseline pain (NRS), medianGinterquartile range Nausea, % (No.) None Some Vomited, % (No.) No Yes Reassessed for abnormal vital signs, % (No.)* No Yes

42.3G11.7

40G10.3

43% (34) 57% (46)

51% (20) 49% (19)

66% (53) 29% (23) 5% (4)

64% (25) 26% (10) 10% (4)

49% 31% 10% 8% 2%

46% 33% 12% 5% 3%

(39) (25) (8) (6) (2)

(18) (13) (5) (2) (1)

10G1

10G1

76% (61) 24% (19)

82% (32) 18% (7)

97 (77) 2 (3)

100 (39) 0

89% (71) 11% (9)

92% (36) 8% (3)

NRS, Numeric rating scale. *Blood pressure \100 mg Hg, pulse rate \60 beats/min, respiratory rate %12 breaths/min, or SO2 decreased O5%. No patients received an opioid antagonist.

HEENT, Head, eyes, ears, nose, and throat.

needed. If oxygen saturation dropped by more than 5%, the patient was reassessed, and oxygen was administered at a fraction of inspired oxygen that returned the oxygen saturation to its baseline level. Patients whose presenting oxygen saturation was 95% or less were given oxygen on presentation. The protocol specified a dose of 0.1 mg/kg morphine to be given intravenously during 1 to 2 minutes. Because of rounding up or down, some patients received weight-based doses that varied slightly from the specified dose. Before data analysis, we decided to accept weight-based doses ranging from 0.09 to 0.11 mg/kg as meeting the protocol criterion for a 0.1 mg/kg standardized dose.

Because this was an analysis of previously collected data, we calculated the precision of the estimated proportion of patients classified as having less than a 50% decrease in pain from a sample of 119 patients. If the proportion of patients with inadequate analgesia were 50%, a cohort of at least 119 patients would provide us with 95% CIs with reasonable precision (ie, within 10% of the point estimate). SPSS version 12.2 (SPSS, Inc.) was used for the descriptive analyses. Stata version 8.2 (Stata Corporation, College Station, TX) was used to calculate the exact binomial CIs.

RESULTS Data Collection and Processing Data were double entered by 2 individuals into a database by SPSS Data Entry 4.0 (SPSS, Inc., Chicago, IL). Primary Data Analysis Descriptive statistics are reported as means with 95% confidence intervals (CIs), medians with interquartile ranges, and proportions with exact binomial 95% CIs. 364 Annals of Emergency Medicine

Four hundred twenty-eight patients were assessed for eligibility and asked to participate in the 4 trials. Three hundred sixty were enrolled in the studies, 225 in the experimental arms of the trials, and 135 in the control arms, the focus of these analyses. Major reasons for noneligibility included the following: opioids had been used in the past week (25%), patient did not want to receive morphine (12.5%), pain started more than 1 week before the ED visit (12.5%), opioids were given Volume 46, no. 4 : October 2005

Bijur, Kenny & Gallagher

Morphine and Severe Acute Pain

Figure. Distribution of 30-minute pain score by baseline pain score.* *Shaded area indicates number and percentage of patients whose pain scores decreased \50%.

before patient was enrolled (11%), the patient refused, no reason given (9%), and not age eligible (8%). Of 135 patients meeting inclusion criteria, 119 patients received between 0.09 and 0.11 mg/kg intravenous morphine. The remaining 16 subjects received morphine doses outside the prespecified range and were not included. The average age was 42 years (SD=11 years), 55% were female patients, 65% were Hispanic, 28% were black, and 7% were classified as other. The median pain score at baseline was 10 (interquartile range 9 to 10). Oxygen was administered to 3 patients whose presenting PO2 was 95% or less. The diagnoses of the patients are shown in Table 1. One patient was reassessed by the medical staff because the respiratory rate was 12 breaths/min, 1 patient’s oxygen saturation dropped more than 5%, 2 patients had a systolic blood pressure less than 100 mg Hg, and 8 patients had a pulse rate less than 60 beats/min (range 52 to 60 beats/min). None of the patients required administration of an opioid antagonist at any time during the 30-minute study period or for 2 hours thereafter. Patients who had less than 50% and those with 50% or greater reduction in pain at 30 minutes had similar distributions of age, sex, race or ethnicity, location of pain, and initial pain severity (Table 2). As would be expected, the patients classified Volume 46, no. 4 : October 2005

as having a 50% or greater analgesic response had a substantially greater mean decrease in pain than those with less than 50% analgesia: –6.6 versus –1.9 (95% CI for difference 4.7 [4.1 to 5.3]). None of the patients in the less than 50% pain relief group and 1 patient in the 50% or greater pain relief group received additional morphine between baseline and 30 minutes. Distributions of nausea, vomiting, and reassessment for abnormal vital signs were similar. The figure shows the distribution of pain intensity at 30 minutes by baseline pain intensity. Eighty patients (67%; 95% CI 58% to 76%) reported a numeric rating scale score at 30 minutes that indicated less than 50% reduction in pain from baseline. Nineteen patients (16%) reported no change or an increase in pain in the 30-minute interval.

LIMITATIONS Our sample is almost entirely composed of poor, inner-city Hispanic and black patients. Given the many cultural influences on perception and expression of pain, it is quite possible that the prevalence of poor response to morphine in other settings and other populations may differ from what we have reported. Because a convenience sample was recruited when trained research associates were present, the findings might have differed if consecutive patients had been enrolled. The patients in this Annals of Emergency Medicine 365

Bijur, Kenny & Gallagher

Morphine and Severe Acute Pain study had heterogeneous locations of pain, having in common only high pain intensity and need for opioid analgesics. However, this reflects the true variety of clinical emergency practice. Some of the pain, though severe, was episodic and fluctuating. Thus, assessment of it at 30 minutes using a single numeric rating scale reading may not provide an accurate overall reflection of the degree of pain relief experienced during the preceding half hour. There is no a priori reason, however, to postulate that this methodology of sampling produces bias because pain that is changing over time seems about as likely to worsen as it does to improve throughout a constant interval.

DISCUSSION In the late 1980s and early 1990s, the emergency medicine community began to focus on the problem of oligoanalgesia in the ED. Early studies showed that less than half of patients who came to the ED for a variety of causes of acute pain (burns, extremity fractures, sickle cell crisis) received any analgesic medication.12,13 As the proportion of patients receiving treatment for pain increases, a next logical step is to assess the adequacy of pain relief provided by the type and dose of analgesics that emergency physicians elect to administer. This study assessed an initial dose of morphine that is consistent with starting doses recommended in standard texts.6-8 We found that about two thirds of the patients in acute severe pain (median numeric rating scale score 10) receive less than 50% pain relief from a dose of 0.1 mg/kg of morphine within the first 30 minutes after administration. This dose clearly provides inadequate analgesia, and it may well be higher than what is routinely administered in many EDs. In an early report, Wilson and Pendleton13 found that the mean dose of intravenous morphine given to 21 patients (11% of the total number of patients) who received morphine was 5.5 mg. Todd et al14 reported that 61% of white patients and 43% of Hispanic patients received a low dose (less than a 10 mg of morphine or equivalent). Karpman et al15 similarly found that 67% of white and 50% of Hispanic patients received low doses of analgesics. Thus, it may be that the dose examined in our study, with its inadequate provision of analgesia, is higher than what is given in practice. Concern about adverse effects of opioids is a primary barrier to using them in adequate doses. Patients in this study were under constant observation and routinely had their vital signs monitored. It is significant that at the specified dose of 0.1 mg/kg, there were no patients who needed an opioid antagonist and few who had abnormal vital signs requiring clinical reassessment. The existence of interindividual variability in response to opioids has long been recognized. Thus, it should not be surprising that a fixed dose of morphine does not produce adequate analgesia in all patients. The accepted management of pain in postoperative patients is titration to effect, reflecting a recognition of the heterogeneity of analgesic response. This is also the recommended approach in ED medicine, but it is 366 Annals of Emergency Medicine

unclear whether these recommendations are being followed. Our clinical experience suggests that titration is not routine practice. Several strategies have been proposed to address variability in response to opioids to improve postoperative management of pain. Aggressive titration protocols, continuous infusion pumps, and patient-controlled analgesia devices have been adopted by anesthesiologists to manage postoperative pain. Changing from morphine to a second opioid is another strategy that has been successfully used to overcome poor response to morphine for treatment of chronic pain. The practices that are used in postoperative settings do not easily generalize to the ED. Barriers to adequate pain management in the ED include crowding, staff shortages, lack of continuity of care, inadequate training in pain management, and concern about serious adverse effects. Nevertheless, the knowledge gained about pain management from other disciplines may point in a direction that is worth following if our specialty is to improve management of pain in the ED. In conclusion, our results suggest that a 0.1 mg/kg dose of morphine may be too low to adequately control acute severe pain in the majority of patients. If our findings are replicated in other populations, we may need to reevaluate the way in which we use morphine in emergency practice. Supervising editors: Knox H. Todd, MD, MPH; Michael L. Callaham, MD Author contributions: PEB, MKK, and EJG contributed to the concept, design, and writing. MKK was responsible for editing. PEB and EJG incorporated revisions. PEB conducted the analysis. PEB takes responsibility for the paper as a whole. Funding and support: This study was supported in part by a grant from the Agency for Healthcare Research and Quality 1 R01 HS13924. Publication dates: Received for publication December 15, 2004. Revisions received February 7, 2005, and February 15, 2005. Accepted for publication March 10, 2005. Reprints not available from the authors. Address for correspondence: Polly E. Bijur, PhD, Albert Einstein College of Medicine, Rose F. Kennedy Center, 1410 Pelham Parkway South, Bronx, NY 10461; 718-430-4217, fax 718-430-8821; E-mail [email protected]. REFERENCES 1. Brown JC, Klein EJ, Lewis CW, et al. Emergency department analgesia for fracture pain. Ann Emerg Med. 2003;42:197-205. 2. Todd KH. Emergency medicine and pain: a topography of influence. Ann Emerg Med. 2004;43:504-506. 3. Lasagna L, Beecher H. The optimal dose of morphine. JAMA. 1954; 156:230-234. 4. Dahlstrom B, Tamsen A, Paalzow L, et al. Patient-controlled analgesic therapy, part IV: pharmacokinetics and analgesic plasma concentrations of morphine. Clin Pharmacokinet. 1982;7: 266-279.

Volume 46, no. 4 : October 2005

Bijur, Kenny & Gallagher 5. Aubrun F, Langeron O, Quesnel C, et al. Relationships between measurement of pain using visual analog score and morphine requirements during postoperative intravenous morphine titration. Anesthesiology. 2003;98:1415-1421. 6. Zimmer G. Acute pain management. In: Tintinalli J, Kelen G, Stapczynski J, eds. Emergency Medicine: A Comprehensive Study Guide. New York, NY: McGraw-Hill; 2004:257-264. 7. Paris P, Yealy D. Pain management. In: Marx JA, ed. Rosen’s Emergency Medicine: Concepts and Clinical Practice. St. Louis, MO: Mosby; 2002:2555-2577. 8. Ungar JR, Brandes D, Reinoehl BM, et al. Pain management. In: Schwartz GR, ed. Principles and Practice of Emergency Medicine. 4th ed. Baltimore, MD: Williams & Wilkens; 1999. 9. Berkowitz BA, Ngai SH, Yang JC, et al. The disposition of morphine in surgical patients. Clin Pharmacol Ther. 1975;17:629-635.

Volume 46, no. 4 : October 2005

Morphine and Severe Acute Pain 10. Stanski DR, Greenblatt DJ, Lowenstein E. Kinetics of intravenous and intramuscular morphine. Clin Pharmacol Ther. 1978;24:52-59. 11. Bijur PE, Latimer CT, Gallagher EJ. Validation of a verbally administered numerical rating scale of acute pain for use in the emergency department. Acad Emerg Med. 2003;10:390-392. 12. Selbst SM, Clark M. Analgesic use in the emergency department. Ann Emerg Med. 1990;19:1010-1013. 13. Wilson JE, Pendleton JM. Oligoanalgesia in the emergency department. Am J Emerg Med. 1989;7:620-623. 14. Todd KH, Samaroo N, Hoffman JR. Ethnicity as a risk factor for inadequate emergency department analgesia. JAMA. 1993;269: 1537-1539. 15. Karpman RR, Del Mar N, Bay C. Analgesia for emergency centers’ orthopaedic patients: does an ethnic bias exist? Clin Orthop. 1997; 334:270-275.

Annals of Emergency Medicine 367