Postoperative Morphine Consumption in the Elderly Patient

This study was approved by the hospital's ethical committee (Comité de Protection des Personnes se Prêtant à la Recherche Biomédicale Pitié-Salpêtrière, Paris, France). Because data were recorded without any intervention and according to a protocol already used routinely in our PACU, 11authorization was given to waive informed consent.

All nurses in the PACU and in the wards have been trained to assess pain by use of unidimensional scales and to perform morphine titration (PACU) and subcutaneous administration (PACU and wards). They used the visual analog scale (VAS) (0 to 100, hand-held slide-rule type). 12When patients had difficulties in manipulating the VAS, nurses were allowed to use a numerical rating scale (from 0 to 100), 13because these two methods are equivalent. 14 

A strict protocol of intravenous morphine titration was implemented in the PACU after a previous study had determined the optimal regimen of morphine titration. 11This protocol defined the dose of intravenous boluses of morphine, the interval between boluses, the absence of limitation on the total dose, the VAS threshold required to administer morphine, and the criteria to stop titration. Immediately after the patients were tracheally extubated and awake, they were questioned as to the presence of pain in the PACU (at least every 15 min before the onset of morphine titration) and were asked to rate pain intensity on a scale (VAS). When the pain increased to more than 30, intravenous morphine was titrated every 5 min by 3-mg increments (2 mg in patients weighing 60 kg or less), and pain was assessed every 5 min until pain relief, defined as a VAS of 30 mm or less. When the patient was asleep, no attempt was made to wake him or her, and the patient was considered as having pain relief. When pain was too severe to obtain a VAS (patient refusal), it was scored 100. The delay to morphine titration was defined as time elapsed between the arrival in the PACU and onset of intravenous morphine titration. Clinical monitoring included respiratory rate measurements, pulse oximetry (Spo²), sedation according to the Ramsay score, 15arterial blood pressure, and heart rate. All patients received oxygen supplementation. Morphine titration was stopped if the patient had a respiratory rate lower than 12 breaths/min and/or an Spo²lower than 95% and/or experienced a serious adverse event related to morphine administration (allergy, vomiting, severe pruritus). In case of severe ventilatory depression (respiratory rate <10 breaths/min), naloxone (intravenous bolus of 0.04 mg) was administered until the respiratory rate was greater than 12 breaths/min, and this was defined as a severe morphine-related adverse effect. Severe postoperative pain was defined as an initial VAS of 70 or more, as reported previously. 16A high dose of intravenous morphine was defined as a dose of 0.15 mg/kg or more, as reported previously. 16 

A strict protocol of subcutaneous administration of morphine has been implemented in the PACU and in the wards. Subcutaneous morphine could be started 2 h after the end of intravenous morphine titration, either in the PACU or in the ward. Subcutaneous morphine was administered every 4 h, and the dose was adjusted according to the VAS and the patient's weight. The subcutaneous dose was 10 mg when VAS was 60 or greater (7.5 mg in patients weighing 60 kg or less) and 7.5 mg when VAS was lower than 60 (5 mg in patients weighing 60 kg or less). No morphine was administered when VAS was lower than 30. A high dose of subcutaneous morphine was defined as a dose of 0.12 mg/kg or more, because this threshold was the median dose administered in a previous study in our center that used the same protocol. 16 

During the data collection period, consecutive patients who fulfilled the criteria of (1) total hip replacement and (2) understanding of the unidimensional methods were included. Thus, patients with delirium or dementia or who were non-French speaking were not included in the study.

During the study, the anesthesiologists could decide to administer propacetamol and/or nonsteroidal antiinflammatory drugs (NSAIDs), but these prescriptions were also included in the protocol routinely used in our center. Vials containing 2 g of propacetamol (Pro-Dafalgan®, Laboratoires UPSA, Rueil-Malmaison, France), yielding 1 g of paracetamol, were administered intravenously within 15 min. The first administration was performed at the beginning of skin closure in the operating room, as described previously. 17The dose regimen was 2 g of propacetamol every 6 h. Ketoprofen (Profenid®, Laboratoire Aventis, Paris, France) was used because it is the sole intravenous NSAID available in France. The first administration was also performed at the beginning of skin closure. The dose regimen was 100 mg of ketoprofen every 8 h, as described previously. 18 

Patients were divided into two groups according to age: young and elderly (≥70 yr old), as described previously. 10The primary endpoint was the dose of intravenous morphine during titration in the PACU. The secondary endpoint was the dose of subcutaneous morphine administered over the first 24 h. The following adverse effects were noted over the first 24 h: ventilatory depression, defined as a respiratory rate below 10 breaths/min and/or the need for naloxone administration; nausea and vomiting; pruritus; retention of urine requiring drainage in the PACU; bradycardia; bronchospasm; and cutaneous rash or other allergic events. Sedation was defined as a Ramsay score 15greater than 3 but was not considered an adverse effect. However, the percentage of sedated patients during morphine titration was also recorded.

The decision to leave the PACU was made by the anesthesiologist after the nurse had checked that the patient fulfilled the Aldrete criteria 19and did not suffer from emesis, severe pain, or major postoperative bleeding. Like the nurses, the anesthesiologists were not aware of the data collecting period.

The main characteristics of the patients were recorded, including the preoperative use of analgesics. The analgesic drugs used by the patient were classified according to the World Health Organization classification: level 1 (paracetamol), level 2 (dextropropoxyphen, codeine, tramadol), and level 3 (major opioids). 20Moreover, the use of antiinflammatory drugs (NSAIDs or corticosteroids) was also recorded. The main characteristics of surgery were recorded, including indication for surgery, type of surgery, surgical approach, experience of the surgeon (less than or more than 5 yr), postoperative drainage, and early (less than 24 h) postoperative mobilization.

Data are expressed as means ± SD or medians and their 95% CIs in nongaussian variables (time delay and duration). The main endpoint being the dose of intravenous morphine during titration, we assumed that this dose was 0.15 ± 0.10 mg/kg in young patients, as reported previously, 10and that elderly patients represent 30% of patients undergoing total hip replacement in our center. We estimated that at least 310 patients would be needed for the study to be able to detect a 25% difference with 90% certainty (1−β) and a two-sided 5% significance level (α) (NQuery 3.0, Statistical Solutions Ltd., Cork, Ireland). Therefore, we decided to include up to 340 patients, excluding those who were removed from the study (a maximum of 8% of patients in this class was expected). Student t  test and repeated-measures ANOVA were used for continuous gaussian variables. The Mann–Whitney U test was used to compare two medians. The chi-square test or Fisher's exact method were used for categorical variables. Correlation between two variables was performed by use of the least-squares method. For multivariate analysis, a forward step-by-step logistic regression was used, and odds ratios with their 95% CIs were calculated. All comparisons were two-tailed, and a value of P < 0.05 was needed to rule out the null hypothesis. Statistical analysis was performed with a computer and NCSS 6.0 software (Statistical Solutions Ltd.).

Three hundred forty patients were included in the study over an 18-month period. Eleven patients were excluded because some data were lacking. Thus, 329 patients were analyzed. Mean age was 62 ± 13 yr, 153 patients (46%) were male, and 176 (54%) were female. Two hundred twenty-four patients (68%) were young, and 105 (32%) were age 70 yr or older. The main characteristics of the two groups of patients are summarized in table 1. Before surgery, elderly patients took corticosteroids more often than young patients (table 1), but the proportion of patients receiving antiinflammatory drugs (NSAIDS or corticosteroids) was not significantly different between young and elderly patients (22 vs . 28%, P = NS). The main characteristics of surgery and anesthesia in the two groups were comparable. Only the indication for surgery differed slightly between groups (table 2).

The VAS was not significantly different in the two groups during and at the end of morphine titration (fig. 1). The total dose of intravenous morphine administered during titration was lower in elderly patients, but the body weight was also significantly lower in this group (table 3). When the dose was normalized for body weight, no significant difference was observed between groups (table 3). No significant differences in morphine-related adverse effects were observed between the two groups (table 3). No patient required naloxone administration. The number of sedated patients and the number of patients requiring termination of morphine titration were not significantly different between groups. Elderly patients exhibited significantly longer delay in morphine titration and duration of stay in the PACU than young patients (table 3). The use of NSAIDS in the postoperative period was more frequent in young patients (table 3). Even when excluding patients receiving NSAIDS, we did not observe a significant difference between young (n = 118) and elderly (n = 82) patients in the dose of intravenous morphine administered during titration (0.15 ± 0.12 vs . 0.14 ± 0.09 mg/kg, P = NS), in contrast to the dose of subcutaneous morphine (0.19 ± 0.16 vs . 0.13 ± 0.11 mg/kg, P < 0.001) administered later.

When patients were grouped in decades of age, no significant differences in the postoperative dose of intravenous morphine during titration were observed (fig. 2). In contrast, elderly patients required a significantly lower subcutaneous dose of morphine in the postoperative period (fig. 2).

We looked for variables associated with the requirement of a high dose of intravenous morphine. By univariate analysis, patients requiring a high dose of intravenous morphine (n = 131, 40%) experienced severe pain more frequently (VAS of 70 or more, 82 vs . 35%, P < 0.001) and had a higher initial VAS (81 ± 16 vs . 58 ± 22, P < 0.001) and a lower weight (69 ± 14 vs . 72 ± 14 kg, P = 0.02). The mean age was lower (60 ± 14 vs . 63 ± 13 yr, P < 0.05), but the proportion of elderly patients (28 vs . 38%, P = NS) was not significantly different. However, by multivariate analysis, only severe pain was significantly associated with a high dose of intravenous morphine (table 4).

We looked for variables associated with the requirement of a high dose of subcutaneous morphine. By univariate analysis, patients requiring a high dose of subcutaneous morphine (n = 167, 51%) experienced severe pain more frequently (VAS 70 or greater, 65 vs . 42%, P < 0.001) and had a higher initial VAS (71 ± 15 vs . 63 ± 23, P = 0.002), a lower age (59 ± 14 vs . 65 ± 12 yr, P < 0.001), and a longer duration of surgery (144 ± 62 vs . 128 ± 47 min, P = 0.007). The proportion of patients less than 60 yr old was higher (48 vs . 31%, P = 0.002), as was the proportion of patients receiving no NSAIDs (67 vs . 55%, P = 0.04) and no minor analgesic drugs (28 vs . 17%, P = 0.02). However, by multivariate analysis, only severe pain, age less than 60 yr, and no NSAIDs were significantly associated with a high dose of subcutaneous morphine (table 4).

The dose of intravenous morphine titration was not significantly correlated to age (R = 0.06, P = NS). In contrast, the dose of subcutaneous morphine titration was significantly correlated with age (R = 0.15, P < 0.009).

The main finding of our study is that the dose of intravenous morphine during postoperative titration (i.e ., acute administration) was not significantly modified in elderly patients, in contrast to the dose of subcutaneous morphine administered over the first 24 postoperative hours (i.e ., subacute administration).

Elderly patients have been noted to be more susceptible to the effects of opioid analgesic than young patients, 2,21because aging affects some phases of pharmacokinetics, including distribution, metabolism, and elimination. 22–24For any given weight-adjusted dose of an opioid, elderly patients have been shown to have higher blood levels and more pronounced pharmacologic effects than young patients. 6,8In elderly patients, the morphine volume of distribution is approximately half that of young patients, probably because of a decreased cardiac output. 25Plasma clearance of morphine is also lower in the elderly. 26Moreover, renal failure is more frequent in elderly patients and can induce an accumulation of morphine metabolites (morphine-3-glucuronide and morphine-6-glucuronide), which explains its prolonged effect in the elderly patients. 27 

Previous studies have found a significant correlation between the dose of opioid required in the postoperative period and the patient's age. 28–30Nevertheless, these studies assessed the variables influencing morphine requirements over a large time period (the initial 24 h). In our previous study, 10we observed no significant difference in the dose of intravenous morphine during titration between young and elderly patients. However, we could draw no definite conclusions, because these two populations differed in many respects (type of surgery undertaken, duration of anesthesia, and surgery) that were potential and important biases for this comparison. In the present study, we included only patients undergoing total hip replacement, generating approximately the same surgical trauma and postoperative pain complaints. As expected, the two populations were not completely comparable in American Society of Anesthesiologists status, delay for tracheal extubation, and weight (tables 1 and 2). However, age was not a significant variable associated with the requirement of a high dose of intravenous morphine when assessed by multivariate analysis (tables 3 and 4). Moreover, these differences between young and elderly patients are thought to decrease the dose of morphine and not potentially to mask a difference.

There is an important discrepancy between our present and past studies 10and the well-accepted concept that opioids are more potent in the elderly. We suggest that the influence of pharmacokinetic variables may not be so important during intravenous morphine titration. Nevertheless, because the dose of intravenous morphine was carefully adapted to pain, we cannot completely rule out the hypothesis that differences in pain intensity (or pain sensitivity to morphine) occur between young and elderly patients. Indeed, although the prevalence of painful conditions in the elderly patient is greater than it is in the young, 31pain complaints are less frequent in the elderly. 2Some studies have shown that pain tolerance is slightly increased in elderly patients, 32whereas others did not. 33Our study may lack sufficient power to detect a significant difference between groups. However, the small difference observed (−8%) (table 3) should not be considered clinically significant. Our results are strengthened by the significant decrease (−36%) in the dose of subcutaneous morphine we observed in elderly patients, which is consistent with the 41% decrease reported previously over the first postoperative day. 34After remifentanil anesthesia, Minko-witz et al . 35also observed that total morphine requirements in the 24-h postoperative period was almost halved in patients older than 65 yr. Thus, in contrast to intravenous morphine titration, the dose and the rate of morphine administration after titration should be adjusted according to age, as demonstrated by numerous studies. 26,28,30,34,35 

We did not observe a significant decrease in the dose of sufentanil required during surgery. This result contrasts with the 50% decrease in opioid requirements reported previously during the intraoperative period. 36,37Several methodologic issues preclude any valid comparison between the two groups in our study: (1) sufentanil was usually administered as fixed boluses of 10 μg, decreasing the probability of observing a difference over a short period, and (2) opioid requirements were not validated by a precise technique such as electroencephalographic recording.

Some remarks must be included to assess the limitations of our study. First, nurses were not blinded to patient age and thus might have behaved differently in the two groups. Despite this, pain relief was similar in the two groups (fig. 1) because of the strict protocol implemented in our PACU. Moreover, in our previous study, 10we observed that protocol violations occurred rarely, were minor in most cases, and did not occur more frequently in elderly patients. Second, polypharmacy occurs very often in the elderly population. In addition to age-related changes in either the pharmacokinetics or the pharmacodynamics of the opioids, further studies are necessary to assess interactions, accumulation, and toxicity risks between analgesics and other drugs used before prescription of the pain treatment. Third, the reliability of the pain measurement tools might be affected by age. 38,39We limited this problem by excluding severely cognitively impaired patients, but we cannot rule out the hypothesis that a subtle difference occurs. Fourth, further studies are necessary in the very elderly population (greater than 90 yr old), which was not included in our study, primarily because total hip replacement is rarely performed at these advanced ages.

In conclusion, we confirm that the dose of intravenous morphine during postoperative titration is not significantly modified in elderly patients, in contrast to the dose of morphine administered subcutaneously over a more prolonged period.

The authors thank the nurses of the postanesthesia care unit (Department of Anesthesiology, Centre Hospitalier Universitaire [CHU] Pitié-Salpêtrière, Paris) for their work on this study and David Baker, D.M., F.R.C.A. (Department of Anesthesiology, CHU Necker-Enfants Malades, Paris), for reviewing the manuscript.