Predictive factors of early morphine requirements in the post-anaesthesia care unit (PACU)

S. Dahmani, H. Dupont, J. Mantz, J. M. Desmonts and H. Keita*

Department of Anaesthesia and Intensive Care, Hospital Bichat, 46 rue Henri Huchard, F-75018 Paris, France*Corresponding author

Accepted for publication: April 23, 2001


    Abstract
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Use of morphine by titration in the post-anaesthesia care unit (PACU) is often the first step in postoperative pain management. This approach provides rapid analgesia but shows a wide inter-individual variability in morphine requirements and may prolong patient stay in the PACU. The aim of this study was to identify the patient characteristics, surgical, anaesthetic, and postoperative factors predictive of early morphine requirements. The study included 149 patients undergoing various non-cardiac surgical procedures under general anaesthesia. In the multiple regression analysis of nine variables, only ethnicity (Caucasian), emergency surgery, major surgery, surgery exceeding 100 min, and pain score on arrival in PACU were predictive factors of morphine requirements. This observational study identifies for the first time independent predictive factors of morphine requirements in the early postoperative period. Future studies are warranted to evaluate the impact of intervention on these factors and any resulting improvement in postoperative pain treatment.

Br J Anaesth 2001; 87: 385–9

Keywords: pain, postoperative; analgesics opioid, morphine; audit, post-anaesthesia care unit


    Introduction
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
The management of acute pain has undoubtedly improved over the last few years. There is now increased interest in it, and effective pain relief after surgery is an essential element of good postoperative management.

In the post-anaesthesia care unit (PACU), the use of titrated doses of morphine is often the first step in postoperative pain treatment. Morphine by titration requires initial assessment of the pain but provides rapid analgesia and the ability to adapt the dose to individual requirements.1 2 Although in common use, there are several limitations of this approach. Intravenous (i.v.) administration of boluses of morphine needs regular pain and sedation assessment. Such management is time consuming for nurses, prolongs patient stay,3 and contributes to the increased cost of PACU admission. There is also wide inter-individual variability in opioid requirements.4 5 Contrary to the later postoperative period (24 h plus) where correlation between physical characteristics, psychological factors, and opioid requirements has been reported,68 there has been little evaluation of the factors that contribute to variations in morphine requirement in the early postoperative period.

The aim of this study was to identify the patient characteristics, surgical, anaesthetic, and postoperative factors that contribute to variability in early postoperative morphine requirements.


    Methods
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Study site
This prospective study was carried out during a 2-week period in a university hospital. This hospital has 402 surgical beds and admits on average 12 600 surgical patients each year. All surgical procedures except cardiac surgery (i.e. orthopaedic, general, urological, gynaecological, and ear, nose and throat) were included in the study.

Patients
All in-patients (including for emergency procedures) undergoing surgery under general anaesthesia, were included in the study. Patients underwent routine pre-operative assessment and were informed about postoperative pain management. Patients having any one of the following criteria were excluded: regional block, epidural or spinal anaesthesia or analgesia, inability to read or speak French, the likelihood of delayed postoperative tracheal extubation in the PACU, ASA physical status over 3, or incomplete records.

Study design
We performed an observational study and patients were included consecutively as they presented. During the study period, the anaesthetic and postoperative pain management did not vary from routine practice; hence, ethics committee approval and informed consent were not needed. We prospectively collected the physical characteristics, surgical, anaesthetic, and postoperative data of each patient. A data sheet was kept postoperatively (from day 1) for each patient by the PACU resident. Operations were grouped as major, intermediate, or minor procedures according to the expected level of postoperative pain as proposed by Gould and the French Society of Anaesthesiology (Table 1).9 10


View this table:
[in this window]
[in a new window]
 
Table 1 Types of operation classified as major, intermediate, and minor procedures according to the expected level of postoperative pain9 10
 
In the PACU after surgery, patients complaining of pain were given i.v. bolus doses of morphine that were repeated until the patient felt comfortable. Pain intensity level was assessed using a verbal score (VS) (Fig. 1). Morphine boluses were given by the nurse using the standardized procedure currently used in our hospital (Fig. 1). The procedure sets no limit to the total dose that can be given in the absence of excessive sedation or hypotension.



View larger version (23K):
[in this window]
[in a new window]
 
Fig 1 Standard procedure for morphine titration in PACU. Pain assessment with the verbal score (VS): ask the patient ‘Which of the following words best describes the pain you are experiencing at the moment’: none=0, mild pain=1, moderate pain=2, severe pain=3. Sedation score (SS): look at the patient and decide which of the following apply: awake=0, dozing intermittently=1, mostly sleeping=2, only awakens when aroused=3.

 
For patients who did not require morphine titration because of low pain intensity (VS <=1), postoperative analgesia was achieved with i.v. propacetamol. Patients were not discharged from the recovery room until they were comfortable with a VS <=1.

Statistical analysis
Data were categorized as shown in Table 2. In the analysis, all factors have been considered as binary variables: age (<=65 yr=1, >65 yr=2); sex (female=1, male=2); ethnicity (African and Asian people=1, Caucasian=2); type of surgery (major=1, intermediate and minor=2); context of surgery (scheduled=1, emergency=2); duration of surgery (<=100 min=1, >100 min =2); surgeon grade (senior=1, junior=2); type of intra-operative opioid (none or alfentanil=1, fentanyl or sufentanil=2); total dose of intra-operative opioid (low=1, high=2); intra-operative administration of non-opioid analgesic (none=1, propacetamol, nefopam or ketoprofen=2); time between last injection of opioid and arrival in the PACU (low=1, high=2); sedation score on arrival in the PACU (0 or 1=1, 2 or 3=2); and pain score on arrival in the PACU (0 or 1=1, 2 or 3=2). For each variable, the total corresponding morphine dose (mg) is expressed as mean (SD) with 95% confidence intervals. For all variables, an appropriate test has been performed to ensure the data is normally distributed. For small subgroups (below 30 patients each) we used a graphical technique. The total dose of morphine required for pain relief was compared between groups using analysis of variance (ANOVA).


View this table:
[in this window]
[in a new window]
 
Table 2 Data categorization used for statistical analysis. *For the duration of surgery and the dose of intra-operative opioid, the categorization is determined by the mean value of each variable. Patients received any one of the three opioids. §For the time between last injection of opioid and arrival in PACU, the categorization into low or high is based on the duration of analgesic effect of each opioid
 
To identify predictive factors of morphine consumption, variables that were found significant using ANOVA were included in a stepwise, backward, multivariate analysis model. The multiple regression analysis (MRA) model used was: Y=B+B1X1+B2X2+B3X3+ ... BnXn, where Y is the dependent variable (total morphine dose titrated) and X1,2,3... the independent variables.

A P value of <0.05 was considered statistically significant. All statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) version 9.0 for Windows 95/98.


    Results
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
Patient characteristics
During the study period, 205 patients underwent a surgical procedure under general anaesthesia and 149 patients fulfilled the criteria for inclusion in the analysis. Fifty-six patients were subsequently excluded from the study because of: delayed extubation (13 patients); ASA physical status over 3 (eight patients); inability to read or speak French (15 patients); and incomplete data records (20 patients). The characteristics of the study population are shown in Table 3.


View this table:
[in this window]
[in a new window]
 
Table 3 Morphine requirements for pain relief in the PACU
 
Univariate analysis
Factors for which a significant difference in the total dose of morphine was found were: sex, ethnicity, type, context and duration of surgery, type and amount of intra-operative opioids, time between last injection of opioid and arrival in the PACU, and finally, postoperative pain on arrival in the PACU. In contrast, there was no significant difference for age, surgeon grade, intra-operative administration of non-opioid analgesics such as propacetamol, nefopam or keto profen, and sedation score at the arrival in the PACU (Table 3).

Multiple regression
Independent predictive factors of morphine requirements were: context of surgery (emergency), duration (>100 min), pain score on arrival in the PACU (2 or 3), ethnicity (Caucasian), and type of surgery (major) with a R2=0.38. The equation for the regression model was: Y=2.3 (emergency surgery)+1.9 (duration of surgery >100 min)+1.6 (pain score on arrival in PACU at 2 or 3)+1.6 (Caucasian ethnicity)+0.9 (major surgery)–1.36. Sex, type, and amount of intra-operative opioid, and time between last injection of opioid and arrival in PACU were not found to be independent predictive factors of morphine requirements (Table 4).


View this table:
[in this window]
[in a new window]
 
Table 4 Independent predictive factors of morphine requirements in the PACU from the multiple regression analysis
 

    Discussion
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
This prospective, observational study suggests that ethnicity, type, context and duration of surgery, and initial postoperative pain score are predictive factors of morphine requirements in the PACU after general anaesthesia.

However, some methodological limitations have to be considered. First, this study included only a restricted sample of 149 patients. This may have decreased the power of the study, particularly for the negative findings, which were significant in the univariate analysis but not in the multiple regression analysis. Second, there is a possibility of confounders among the factors identified as significant in the univariate analysis. However, multiple regression analysis looks at the independency of these factors thereby ruling out those considered as possible confounders. Third, the value of R2 in the multiple regression was 0.38. This means that 38% of the variability in the morphine requirements was explained by variability in the factors analysed. That leaves 62% of the variability that is not explained by any of the measured variables. Factors not considered in this study have to be taken into account. For example, psychological factors may be responsible for some of the inter-patient variability.1113 In our study, an increased anxiety level could probably explain the finding that emergency surgery is a predictive factor of morphine requirements. Patients operated on as an emergency are more anxious because they receive less information and do not benefit from any anxiolytic pre-medication. Because of these methodological considerations, this study can only be considered as preliminary work.

The positive findings in this study are not exactly the same as those reported for the later postoperative period (24 h plus). Ethnicity appears as a predictive factor of morphine consumption. Caucasian people required more morphine than Africans and Asians. It could be argued that this result is related to a higher incidence of Caucasian people in our study. However, the expression of pain and analgesic requirements has been shown to differ in various ethnic groups.1417 In addition, Uhl and colleagues18 have demonstrated that a major source of variability in pain sensation and response to opioids is genetic. Alternatively, our finding could be explained by a higher socioeconomic status and level of formal education in Caucasian people in our institution.19 It cannot be excluded that the predictive value of ethnicity could be related to other variables such as alcohol intake with hepatic enzyme induction, as increased requirements for intra-operative opioid have been reported in chronic alcoholics.20

As with previous observations,21 22 we found a significant influence of surgical factors on morphine requirements. An original finding was that surgery exceeding 100 min duration was associated with increased morphine requirements. Pain score on arrival in PACU (moderate or severe) was also a predictive factor of morphine requirements. This is in agreement with previous work that reported an increase in postoperative opioid requirements with increased postoperative pain levels.23 It could be argued that this result was to be expected as the procedure for morphine titration was based on pain scores. Nevertheless, the sedation score also used in the procedure did not appear as a predictive factor of morphine requirements.

In contrast, some variables previously reported to predict postoperative opioid requirements had no significant influence on morphine consumption in the PACU in our study. Increasing age has been found to decrease opioid requirements in numerous studies.68 24 However, we failed to show a significant influence of age on early morphine requirements. This result could be related to the procedure used because it took into account the age of the patients (boluses of 3 mg under 65 yr and 2 mg over 65 yr, Fig. 1). Nor was our procedure weight related because it has been demonstrated that morphine requirements are better correlated to the age than the weight of the patient.7 Procedures of morphine titration only based on age are used in other institutions.25 Sex also did not appear as a predictive factor of morphine consumption in the PACU. This result is consistent with a clinical study in volunteers that has failed to find an effect of sex on morphine-induced behavioural and physiological responses.26

For the later postoperative period, previous results are conflicting. Macintyre and Jarvis7 noted that male gender was a predictive factor of postoperative PCA morphine requirements although in their study, age was the best predictive factor. In contrast, other authors have reported a greater need for pain relief in women.27

Surprisingly, the type of anaesthetic, the amount of opioid used intra-operatively, and the time of the last injection of opioid, although significantly related to morphine consumption in the univariate analysis, were not independent predictive factors in the multiple regression analysis. This discrepancy could be related to the fact that these variables are probably dependent on the type of surgery performed and could be considered as confounding factors. An important finding of this study is the lack of significant association between intra-operative administration of propacetamol, ketoprofen or nefopam, and morphine requirements. These analgesics were usually administered between 30 and 60 min before the end of surgery in anticipation of postoperative pain. Although these results do not support the use of non-opioid analgesics, further work is required because of the lack of power of our study.

How can the identification of these predictive factors improve postoperative pain management? Few of these variables are accessible to medical intervention. However, reduction of pain intensity on arrival in the PACU can be considered a reasonable goal. This can be achieved by the use of regional analgesia when possible.2830 The concept of pre-emptive analgesia has been also suggested.31 Although pre-emptive analgesia is an attractive working hypothesis, the results appear conflicting in the literature.3234 Other approaches such as intra-operative titration of morphine could be investigated further. This technique decreases postoperative pain scores, morphine requirements, and patient length of stay in the PACU.25

In conclusion, this study opens up areas for further investigation of pain management in the early postoperative period. It is of importance because this period has clinical and economic implications.


    References
 Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 References
 
1 Upton RN, Semple TJ, Macintyre PE. Pharmacokinetic optimisation of opioid treatment in acute pain therapy. Clin Pharmacokinet 1997; 33: 225–44[ISI][Medline]

2 Zacharias M, Pfeifer MV, Herbison P. Comparison of two methods of intravenous administration of morphine for postoperative pain relief. Anaesth Intensive Care 1990; 18: 205–9[ISI][Medline]

3 Seago JA, Weitz S, Walczak S. Factors influencing stay in the postanesthesia care unit: a prospective analysis. J Clin Anesth 1998; 10: 579–87[ISI][Medline]

4 Gourlay GK, Willis RJ, Lamberty J. A double-blind comparison of the efficacy of methadone and morphine in post-operative pain control. Anesthesiology 1986; 64: 322–7[ISI][Medline]

5 Gourlay GK, Kowalski SR, Plummer JL, et al. Fentanyl blood concentration-analgesic response relationship in the treatment of postoperative pain. Anesth Analg 1988; 67: 329–37[Abstract]

6 Burns JW, Hodsman NBA, McLintock TTC, et al. The influenceof patient characteristics on the postoperative analgesia. Anaesthesia 1989; 44: 2–6[ISI][Medline]

7 Macintyre PE, Jarvis DA. Age is the best predictor of morphine requirements. Pain 1995; 64: 357–64[ISI]

8 Berkowitz BA, Ngai SH, Yang JC, et al. The disposition of morphine in surgical patients. Clin Pharmacol Ther 1975; 17: 629–35[ISI][Medline]

9 Gould TH, Crosby DL, Harmer M, et al. Policy for controlling pain after surgery. Effect of sequential changes in management. BMJ 1992; 305: 517–22[ISI][Medline]

10 Société Française d’Anesthésie et de Réanimation. Conférence de consensus sur la prise en charge de la douleur postopératoire chez l’adulte et l’enfant. Ann Fr Anesth Réanim 1998; 17: 445–61[ISI][Medline]

11 Scott LE, Clum GA, Peoples JB. Preoperative predictors of postoperative pain. Pain 1983; 15: 283–93[ISI][Medline]

12 Perry F, Parker RK, White PF, Clifford PA. Role of psychological factors in postoperative pain control and recovery with patient-controlled analgesia. Clin J Pain 1994; 10: 57–63[ISI][Medline]

13 Preble LM, Guveyan JA, Sinatra RS. Patient characteristics influencing postoperative pain management. In: Sinatra RS, Hord AH, Ginsberg B, Preble LM, eds. Acute Pain: Mechanisms and Management. St Louis, MO: Moseby Year Book, 1992; 140–4

14 Faucett J, Gordon N, Levine J. Differences in postoperative pain severity among four ethnic groups. J Pain Symptom Manage 1994; 9: 383–9[ISI][Medline]

15 Lipton J, Marbach JJ. Ethnicity and the pain experience. Soc Sci Med 1984; 19: 1279–98[ISI][Medline]

16 Zborowski M. Cultural components in response to pain. J Soc Issues 1952; 8: 15–35

17 Greenwald HP. Interethnic differences in pain perception. Pain 1991; 44: 157–63[ISI][Medline]

18 Uhl GR, Sora I, Wang Z. The m opiate receptor as a candidate gene for pain: polymorphisms, variations in expression, nociception, and opiate responses. Proc Natl Acad Sci USA 1999; 96: 7752–5[Abstract/Free Full Text]

19 Weisenberg M, Caspi Z. Cultural and educational influences of pain of childbirth. J Pain Symptom Manage 1989; 4: 13–19[Medline]

20 Lemmens HJ, Bovill JG, Hennis PJ, Gladines MP. Alcohol consumption alters the pharmacodynamics of alfentanil. Anesthesiology 1989; 71: 669–74[ISI][Medline]

21 Benedetti C, Bonic JJ, Bellucci G. Pathophysiology and therapy of postoperative pain: a review. In: Benedetti C, Chapman CR, Moricca G, eds. Advances in Pain Research and Therapy. New York: Raven Press, 1984; 373–7

22 Bonica JJ. Postoperative pain. In: Bonica JJ, eds. The Management of Pain. Philadelphia, 1990; 461–80

23 Voulgari A, Lykouras L, Papanikolaou M, et al. Influence of psychological and clinical factors on postoperative pain and narcotic consumption. Psychother Psychosom 1991; 55: 191–6[ISI][Medline]

24 Owen JA, Sitar DS, Berger L, et al. Age-related morphine kinetics. Clin Pharmacol Ther 1983; 34: 364–8[ISI][Medline]

25 Pico L, Hernot S, Negre I, et al. Peroperative titration of morphine improves immediate postoperative analgesia after total hip arthroplasty. Can J Anaesth 2000; 47: 309–14[Abstract/Free Full Text]

26 Dennis W, Coalson MD, Debra J, Janiszewski BA, James P. Does gender modulate the behavioral and physiological effects of intravenous morphine in healthy volunteers? Anesthesiology 1999; 91: A213

27 Verbrugge LM. Gender and health: an update on hypotheses and evidence. J Health Soc Behav 1985; 26: 156–82[ISI][Medline]

28 Boylan JF, Katz J, Kavanagh BP, et al. Epidural bupivacaine-morphine analgesia versus patient-controlled analgesia following abdominal aortic surgery. Anesthesiology 1998; 89: 585–93[ISI][Medline]

29 Brown DL. Peripheral nerve blocks: regional anesthesia for the new millenium (editorial). Reg Anesth Pain Med 1998; 23: 237–40[ISI]

30 Capdevila X, Barthelet Y, Bilboulet P, et al. Effects of perioperative analgesic technique on the surgical outcome and duration of rehabilitation after major knee surgery. Anesthesiology 1999; 91: 8–15[ISI][Medline]

31 Kehlet H. Controlling acute pain-role of pre-emptive analgesia, peripheral treatment, balanced analgesia, and effects on outcome. In: Max M ed. Pain 1999 – An Updated Review. Seattle, WA: IASP Press, 1999; 459–62

32 Coderre TJ, Katz J, Vaccarino AL, et al. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993; 52: 259–85[ISI][Medline]

33 Dahl JB, Kehlet H. The value of pre-emptive analgesia in the treatment of postoperative pain. Br J Anaesth 1993; 70: 434–9[ISI][Medline]

34 Kissin I. Preemptive analgesia, why it is not always obvious. Anesthesiology 1996; 84: 1015–9[ISI][Medline]