- Email: [email protected]
ORAL NALBUPHINE FOR THE TREATMENT OF PAIN AFTER DENTAL EXTRACTIONS
Br. J. Anaesth. (1988), 61, 313-317
ORAL NALBUPHINE FOR THE TREATMENT OF PAIN AFTER DENTAL EXTRACTIONS B. KAY, R. G. LINDSAY, C. J. MASON AND T. E. J. HEALY
PATIENTS AND METHOD
The investigation was approved by the Hospital Ethics Committee. Ninety patients (ASA grade I or II) aged between 18 and 60 yr, requiring general anaesthesia for extraction of teeth and who had given written informed consent to participate in the study, were allocated randomly to one of three groups, to receive dihydrocodeine 30 mg, nalbuphine 30 mg, or nalbuphine 60 mg, B. KAY, M.B., CH.B., F.F.A.R.C.S. ; R. G. LINDSAY, M.B., CH.B., F.F.A.R.C.S.; C. J. MASON, CH.B., F.F.A.R.C.S.; T. E. J. HEALY,
M.SC, M.D., F.F.A.R.C.S.; Department of Anaesthesia, University Hospital of South Manchester, Manchester M20 8LR. Accepted for Publication: February 15, 1988. Correspondence to T.E. J.H.
SUMMARY A randomized, double-blind comparison of nalbuphine 30 mg or 60 mg by mouth and dihydrocodeine 30 mg by mouth was conducted in 75 patients with moderate to severe pain after surgery for dental extractions under general anaesthesia. A significant reduction in pain intensity followed each treatment and persisted throughout the 4-h observation period after nalbuphine, but only for 3 h after dihydrocodeine was given. Reduction in pain intensity was significantly greater 2, 3 and 4 h after the use of nalbuphine 60 mg than following dihydrocodeine 30 mg, and the mean total pain intensity difference was greater following nalbuphine 60 mg than following dihydrocodeine. Nalbuphine 60 mg effectively provided complete or good pain relief in more than 50 % of the patients and only three patients in this group required additional analgesia during the period of observation, compared with nine patients in each of the other groups. However, the patients who received nalbuphine 30 mg had a significantly higher mean pain intensity before treatment than those in the other groups. The side-effects encountered were those typical of opioid medication; there were no statistically significant differences between the groups.
for the initial treatment of postoperative pain. Before anaesthesia was induced the patients practised the use of a 10-cm horizontal linear analogue scale (LAS) ("None" to "Worst, unbearable"), for recording the degree of pain. Anaesthesia was induced in the unpremedicated patients with thiopentone—approximately 4 mg kg"1 i.v., followed by suxamethonium (1.5 mg kg"1). The trachea was intubated and anaesthesia maintained using nitrous oxide in oxygen
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
Nalbuphine is a partial agonist opioid that is approximately equipotent with, and has a similar duration of effect to, morphine 8-10 mg [1]. After i.m. injection it has been shown to be effective in the treatment of moderate to severe pain, and to produce a low incidence of side-effects [2]. It has been shown to have a limited respiratory depressant effect in man [3] and to have clinically unimportant cardiovascular and respiratory effects following doses of up to 200 mg given i.v. over 1 h by patient-controlled on-demand apparatus [4]. Nalbuphine has recently been formulated in tablets (30 mg) for administration by mouth. Dihydrocodeine is an opioid frequently given by mouth as tablets of 30 mg, for the treatment of moderate to severe pain including pain after dental extraction. It has a duration of effect (46 h) similar to that of morphine, and has similar side-effects [5]. The aim of the investigation was to compare the efficacy of dihydrocodeine 30 mg by mouth with that of nalbuphine 30 or 60 mg by mouth for the treatment of moderate pain following dental surgery under general anaesthesia.
314
BRITISH JOURNAL OF ANAESTHESIA
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
with clinically appropriate concentrations of en- already withdrawn from the study, the final grade flurane. No analgesics were administered on the that had been assessed was recorded for this time. day of surgery, before the postoperative analgesic Other categorical data were assessed using the was given. When the patients requested pain relief Chi-squared test. after recovery from anaesthesia they were given oral medication randomized according to the RESULTS number of their entry into the study and rendered double-blind by the use of a double-dummy Fifteen of the 90 patients who entered the study technique. were withdrawn before recording of the effects of Before medication the patients indicated the the medication could start. Some of these vomited intensity of their pain (by LAS) and this assess- shortly after medication; others requested an ment was repeated 0.5, 1, 2, 3 and 4 h after injection because of severe pain. Full records were medication, or until another analgesic was given. obtained for 75 patients: 27 received dihydroAt each assessment after medication the assessor codeine 30 mg, 24 nalbuphine 30 mg and 24 recorded an opinion regarding pain relief, grading nalbuphine 60 mg. Demographic data are reit as 0, 1 (slight), 2 (moderate) 3 (good) or 4 corded in table I; the groups were similar in (complete). Any response to non-specific ques- composition. tions regarding side-effects or discomfort was Pain intensity values are illustrated in figure noted also. The intensity of side-effects was 1. graded as mild, moderate or severe, their duration All the treatments provided some reduction in noted, and any medication required to treat them pain intensity for 3 h. At the 4-h assessment was recorded. dihydrocodeine was no longer effective, whereas Alternative analgesic therapy was given im- both doses of nalbuphine were still providing a mediately on demand to any patient who re- significant reduction in pain intensity (P < 0.01; quested it and these patients were withdrawn paired t test). The reduction in effectiveness of dihydrocodeine towards the end of the obserfrom the study. vation period is confirmed by the number of Statistical analysis patients requiring additional analgesia. Only three Pain intensity scores (LAS) were treated as patients who received nalbuphine 60 mg required parametric data and analysed (including pire- an additional analgesic compared with nine in the treatment scores) by repeated measures analysis of group which received dihydrocodeine—eight of variance (using BMDP2V [6]). LAS values miss- these after 3 h. All but five of the remaining ing because of withdrawals from the study were patients in this group showed increasing pain replaced by values calculated by regressing the between the 3- and 4-h assessments. missing values on the existing pain intensity Nine patients who received nalbuphine 30 mg scores, and also on the pain relief scores using the also required additional analgesia, eight before 2 h. BMDPAM algorithm [6]. The degrees of freedom However, on those occasions when nalbuphine for the error terms in the analysis of variance were 30 mg was effective, it worked effectively throughreduced by the number of missing values involved out the observation period. The initial mean pain in the term. For the missing pain relief gradings intensity score for the nalbuphine 30 mg group resulting from withdrawal from the trial, an extra (61.5) was significantly higher than those of the category (withdrawn) was included, to allow the other groups (53 and 52.4, respectively) (P < use of a continuation odds model for regression analysis (GLIM algorithm) [7]. TABLE I. Demographic data (mean± SEAT) The Fisher-Tukey approach to avoid overAge Weight Height stating the statistical significance of a posteriori Group n (cm) Sex (yr) (kg) tests was used. Duration of effect was assessed by use of the paired J-test to compare the 4-h Dihydrocodeine 27 27.5 65.2 10F 168 assessments with pre-treatment values. Between30 mg + 19 ±2.1 17M ±2 group comparisons of the predicted peak effect of Nalbuphine 24 27.6 66.8 173 12F 30 mg +2 + 2.1 + 2.2 12M the analgesics were made by comparing pain relief 24 24 62.6 9F Nalbuphine 168 grades at the 2-h assessment, using the Mann60 mg + 3.2 15M ±1.3 ±2.4 Whitney U test. In the case of patients who had
315
ORAL NALBUPHINE AFTER DENTAL EXTRACTIONS 70T
60-
50-
E
r 4030H
£
20-
10-
0J 0.5
2 Time
(h)
FIG. 1. Mean (SEM) pain intensity scores (linear analogue scale) before (time 0) and after oral medication with dihydrocodeine 30 mg (A, n = 27), nalbuphine 30 mg ( # , n = 24) or nalbuphine 60 mg (O. « = 24). TABLE II. Mean pain intensity (LAS). *P < 0.05 analysis of variance (Fisher-Tukey approach) Assessment time Before treatment 30min 1h 2h 3h 4h
Mean total pain intensity difference (P = 0.028)
Nalbuphine Dihydrococeine 60 mg 30 mg 52.4 40.8 31.2 27.5 25.8 34.3 103.2
0.05, t test), indicating the recruitment into this group of several patients with more severe pain than the general level observed. Indeed, the patients in the nalbuphine 30 mg group who required additional analgesia had an initial mean pain intensity score of 78.5. It appears to be inappropriate to treat pain of such high intensity with oral medication, and most of these patients required a parenteral opioid.
53.7 44.7 36.8 40.6 41.8 43.7 56.7
Differences between treatments 1.3 3.9 5.6
13.1* 16.0* 9.4*
Because of the difference in pain intensity before medication, and the increased number of early withdrawals compared with the other groups, the nalbuphine 30 mg group was omitted from the between group comparisons of pain intensity and total pain intensity differences. Mean pain intensities (table II) were significantly (P < 0.05) lower in the nalbuphine-treated patients than in the dihydrocodeine-treated
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
g
BRITISH JOURNAL OF ANAESTHESIA
316
TABLE III. Percentage of patients obtaining complete or good pain relief
Group
30
60
120
180
240
Mean total pain relief score
Nalbuphine 30 mg Nalbuphine 60 mg Dihydrocodeine 30 mg
21 29 25 50 18 30
33 62 22
29 50 19
33 29 19
13.1 14.7 12.7
Assessment time (min)
TABLE IV. Side-effects reported, n= Total number of patientsin group
Nalbuphine 30 mg (n = 24)
Nalbuphine 60 mg (n = 24)
Dihydrocodeine 30 mg (n = 27)
3 Nausea 2 Headache 1 Drowsiness 2 Light-headed 2 Nausea 1 Drowsiness 2 Light-headed 1 Itchy nose | Light-headed | I Nausea ) 1 Rash
patients at 2, 3 and 4 h after medication. The total decrease in pain intensity scores was also greater in the nalbuphine group (P < 0.05). The percentage of patients in whom pain relief was assessed as complete or good is indicated in table III, together with the total pain relief scores that were recorded. Only in the nalbuphine 60 mg group did at least 50 % of the patients achieve these degrees of pain relief. Two hours after medication, nalbuphine 60 mg provided significantly better pain relief than dihydrocodeine (P < 0.01, Mann-Whitney U test). The number and type of side-effects encountered are listed in table IV and were typically those caused by opioids. There was no statistical difference between the groups. One patient in each of the nalbuphine groups was treated (with an antiemetic) for side-effects before additional analgesics were administered. DISCUSSION
The comparison made in the present study was primarily that between nalbuphine 60 mg and dihydrocodeine 30 mg and this showed statistical
(1 severe, 2 mild) (moderate) (mild) (mild) (moderate, 1 vomited) (mild) (mild) (mild) (moderate)
(mild)
differences between the groups in favour of nalbuphine before withdrawals seriously affected the analysis. Most early withdrawals occurred in the group of patients who received nalbuphine 30 mg, but this observation may be related less to the ineffectiveness of this dose than to the fact that this group contained more patients with severe pain than the other groups, a quirk of random allocation. For these reasons this group was omitted from the analysis of variance used to compare the other groups but, as can be seen in figure 1 (mean pain intensity scores (including substituted values)) and table III (the pain relief obtained), nalbuphine 30 mg was at least as good an analgesic as dihydrocodeine 30 mg and, like the larger dose of nalbuphine, was still effective after 4 h, whereas dihydrocodeine was not. The statistical assessment of duration of effect was made by comparing pain intensity before treatment with that at the end of the observation period using Student's paired t test. This method of analysis avoids the theoretical problems of assessing probability after multiple testing. For the same reason, between-group comparisons of pain relief were confined to the anticipated time of
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
Side-effect
Group
ORAL NALBUPHINE AFTER DENTAL EXTRACTIONS
REFERENCES 1. Beaver WT, Feiser GA. A comparison of the analgesic effect of i.m. nalbuphine with morphine in patients with post-operative pain. Journal of Pharmacology and Experimental Therapeutics 1978; 204: 487-496. 2. Fahmy NR. Agonist/antagonist opioid analgesics. In: Estafanous FG, ed. Opioids in Anesthesia. London: Butterworth; 1984; 27. 3. Romagnoli A, Keats AS. Ceiling effect for respiratory depression by nalbuphine. Clinical Pharmacology and Therapeutics 1980; 27: 478-185. 4. Kay B, Krishnan A. On-demand nalbuphine for postoperative pain relief. Acta Anaesthesiologica Belgica 1986; 37: 33-37. 5. Wood M, Wood AJ, eds. In: Drugs and Anaesthesia. London: Williams and Wilkins, 1982; 171. 6. Dixon WJ. BMDP Statistical Software. Berkeley: University of California Press; 1979. 7. Iver R. Continuation-Odds Models of Ordinal Variable Regression. GLIM Newsletter 10. Oxford: Numerical Algorithms Group, 1984.
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
maximum effect (2 h). From previous experience it was not anticipated that the maximum effect of oral dihydrocodeine after surgery would appear earlier, and this biased this assessment in favour of nalbuphine. In the other analyses the FisherTukey approach was used to maintain at least a minimum statistical significance level when a large number of comparisons are made. The results of the investigation indicate that nalbuphine by mouth reduces pain after dental surgery. However, whilst a dose of 30 mg is as effective as, and acts for longer than dihydrocodeine 30 mg, nalbuphine 60 mg is significantly more effective than dihydrocodeine, and was the only medication to produce good or complete pain relief in more than 50% of the patients who received it.
317
ORAL NALBUPHINE FOR THE TREATMENT OF PAIN AFTER DENTAL EXTRACTIONS B. KAY, R. G. LINDSAY, C. J. MASON AND T. E. J. HEALY
PATIENTS AND METHOD
The investigation was approved by the Hospital Ethics Committee. Ninety patients (ASA grade I or II) aged between 18 and 60 yr, requiring general anaesthesia for extraction of teeth and who had given written informed consent to participate in the study, were allocated randomly to one of three groups, to receive dihydrocodeine 30 mg, nalbuphine 30 mg, or nalbuphine 60 mg, B. KAY, M.B., CH.B., F.F.A.R.C.S. ; R. G. LINDSAY, M.B., CH.B., F.F.A.R.C.S.; C. J. MASON, CH.B., F.F.A.R.C.S.; T. E. J. HEALY,
M.SC, M.D., F.F.A.R.C.S.; Department of Anaesthesia, University Hospital of South Manchester, Manchester M20 8LR. Accepted for Publication: February 15, 1988. Correspondence to T.E. J.H.
SUMMARY A randomized, double-blind comparison of nalbuphine 30 mg or 60 mg by mouth and dihydrocodeine 30 mg by mouth was conducted in 75 patients with moderate to severe pain after surgery for dental extractions under general anaesthesia. A significant reduction in pain intensity followed each treatment and persisted throughout the 4-h observation period after nalbuphine, but only for 3 h after dihydrocodeine was given. Reduction in pain intensity was significantly greater 2, 3 and 4 h after the use of nalbuphine 60 mg than following dihydrocodeine 30 mg, and the mean total pain intensity difference was greater following nalbuphine 60 mg than following dihydrocodeine. Nalbuphine 60 mg effectively provided complete or good pain relief in more than 50 % of the patients and only three patients in this group required additional analgesia during the period of observation, compared with nine patients in each of the other groups. However, the patients who received nalbuphine 30 mg had a significantly higher mean pain intensity before treatment than those in the other groups. The side-effects encountered were those typical of opioid medication; there were no statistically significant differences between the groups.
for the initial treatment of postoperative pain. Before anaesthesia was induced the patients practised the use of a 10-cm horizontal linear analogue scale (LAS) ("None" to "Worst, unbearable"), for recording the degree of pain. Anaesthesia was induced in the unpremedicated patients with thiopentone—approximately 4 mg kg"1 i.v., followed by suxamethonium (1.5 mg kg"1). The trachea was intubated and anaesthesia maintained using nitrous oxide in oxygen
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
Nalbuphine is a partial agonist opioid that is approximately equipotent with, and has a similar duration of effect to, morphine 8-10 mg [1]. After i.m. injection it has been shown to be effective in the treatment of moderate to severe pain, and to produce a low incidence of side-effects [2]. It has been shown to have a limited respiratory depressant effect in man [3] and to have clinically unimportant cardiovascular and respiratory effects following doses of up to 200 mg given i.v. over 1 h by patient-controlled on-demand apparatus [4]. Nalbuphine has recently been formulated in tablets (30 mg) for administration by mouth. Dihydrocodeine is an opioid frequently given by mouth as tablets of 30 mg, for the treatment of moderate to severe pain including pain after dental extraction. It has a duration of effect (46 h) similar to that of morphine, and has similar side-effects [5]. The aim of the investigation was to compare the efficacy of dihydrocodeine 30 mg by mouth with that of nalbuphine 30 or 60 mg by mouth for the treatment of moderate pain following dental surgery under general anaesthesia.
314
BRITISH JOURNAL OF ANAESTHESIA
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
with clinically appropriate concentrations of en- already withdrawn from the study, the final grade flurane. No analgesics were administered on the that had been assessed was recorded for this time. day of surgery, before the postoperative analgesic Other categorical data were assessed using the was given. When the patients requested pain relief Chi-squared test. after recovery from anaesthesia they were given oral medication randomized according to the RESULTS number of their entry into the study and rendered double-blind by the use of a double-dummy Fifteen of the 90 patients who entered the study technique. were withdrawn before recording of the effects of Before medication the patients indicated the the medication could start. Some of these vomited intensity of their pain (by LAS) and this assess- shortly after medication; others requested an ment was repeated 0.5, 1, 2, 3 and 4 h after injection because of severe pain. Full records were medication, or until another analgesic was given. obtained for 75 patients: 27 received dihydroAt each assessment after medication the assessor codeine 30 mg, 24 nalbuphine 30 mg and 24 recorded an opinion regarding pain relief, grading nalbuphine 60 mg. Demographic data are reit as 0, 1 (slight), 2 (moderate) 3 (good) or 4 corded in table I; the groups were similar in (complete). Any response to non-specific ques- composition. tions regarding side-effects or discomfort was Pain intensity values are illustrated in figure noted also. The intensity of side-effects was 1. graded as mild, moderate or severe, their duration All the treatments provided some reduction in noted, and any medication required to treat them pain intensity for 3 h. At the 4-h assessment was recorded. dihydrocodeine was no longer effective, whereas Alternative analgesic therapy was given im- both doses of nalbuphine were still providing a mediately on demand to any patient who re- significant reduction in pain intensity (P < 0.01; quested it and these patients were withdrawn paired t test). The reduction in effectiveness of dihydrocodeine towards the end of the obserfrom the study. vation period is confirmed by the number of Statistical analysis patients requiring additional analgesia. Only three Pain intensity scores (LAS) were treated as patients who received nalbuphine 60 mg required parametric data and analysed (including pire- an additional analgesic compared with nine in the treatment scores) by repeated measures analysis of group which received dihydrocodeine—eight of variance (using BMDP2V [6]). LAS values miss- these after 3 h. All but five of the remaining ing because of withdrawals from the study were patients in this group showed increasing pain replaced by values calculated by regressing the between the 3- and 4-h assessments. missing values on the existing pain intensity Nine patients who received nalbuphine 30 mg scores, and also on the pain relief scores using the also required additional analgesia, eight before 2 h. BMDPAM algorithm [6]. The degrees of freedom However, on those occasions when nalbuphine for the error terms in the analysis of variance were 30 mg was effective, it worked effectively throughreduced by the number of missing values involved out the observation period. The initial mean pain in the term. For the missing pain relief gradings intensity score for the nalbuphine 30 mg group resulting from withdrawal from the trial, an extra (61.5) was significantly higher than those of the category (withdrawn) was included, to allow the other groups (53 and 52.4, respectively) (P < use of a continuation odds model for regression analysis (GLIM algorithm) [7]. TABLE I. Demographic data (mean± SEAT) The Fisher-Tukey approach to avoid overAge Weight Height stating the statistical significance of a posteriori Group n (cm) Sex (yr) (kg) tests was used. Duration of effect was assessed by use of the paired J-test to compare the 4-h Dihydrocodeine 27 27.5 65.2 10F 168 assessments with pre-treatment values. Between30 mg + 19 ±2.1 17M ±2 group comparisons of the predicted peak effect of Nalbuphine 24 27.6 66.8 173 12F 30 mg +2 + 2.1 + 2.2 12M the analgesics were made by comparing pain relief 24 24 62.6 9F Nalbuphine 168 grades at the 2-h assessment, using the Mann60 mg + 3.2 15M ±1.3 ±2.4 Whitney U test. In the case of patients who had
315
ORAL NALBUPHINE AFTER DENTAL EXTRACTIONS 70T
60-
50-
E
r 4030H
£
20-
10-
0J 0.5
2 Time
(h)
FIG. 1. Mean (SEM) pain intensity scores (linear analogue scale) before (time 0) and after oral medication with dihydrocodeine 30 mg (A, n = 27), nalbuphine 30 mg ( # , n = 24) or nalbuphine 60 mg (O. « = 24). TABLE II. Mean pain intensity (LAS). *P < 0.05 analysis of variance (Fisher-Tukey approach) Assessment time Before treatment 30min 1h 2h 3h 4h
Mean total pain intensity difference (P = 0.028)
Nalbuphine Dihydrococeine 60 mg 30 mg 52.4 40.8 31.2 27.5 25.8 34.3 103.2
0.05, t test), indicating the recruitment into this group of several patients with more severe pain than the general level observed. Indeed, the patients in the nalbuphine 30 mg group who required additional analgesia had an initial mean pain intensity score of 78.5. It appears to be inappropriate to treat pain of such high intensity with oral medication, and most of these patients required a parenteral opioid.
53.7 44.7 36.8 40.6 41.8 43.7 56.7
Differences between treatments 1.3 3.9 5.6
13.1* 16.0* 9.4*
Because of the difference in pain intensity before medication, and the increased number of early withdrawals compared with the other groups, the nalbuphine 30 mg group was omitted from the between group comparisons of pain intensity and total pain intensity differences. Mean pain intensities (table II) were significantly (P < 0.05) lower in the nalbuphine-treated patients than in the dihydrocodeine-treated
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
g
BRITISH JOURNAL OF ANAESTHESIA
316
TABLE III. Percentage of patients obtaining complete or good pain relief
Group
30
60
120
180
240
Mean total pain relief score
Nalbuphine 30 mg Nalbuphine 60 mg Dihydrocodeine 30 mg
21 29 25 50 18 30
33 62 22
29 50 19
33 29 19
13.1 14.7 12.7
Assessment time (min)
TABLE IV. Side-effects reported, n= Total number of patientsin group
Nalbuphine 30 mg (n = 24)
Nalbuphine 60 mg (n = 24)
Dihydrocodeine 30 mg (n = 27)
3 Nausea 2 Headache 1 Drowsiness 2 Light-headed 2 Nausea 1 Drowsiness 2 Light-headed 1 Itchy nose | Light-headed | I Nausea ) 1 Rash
patients at 2, 3 and 4 h after medication. The total decrease in pain intensity scores was also greater in the nalbuphine group (P < 0.05). The percentage of patients in whom pain relief was assessed as complete or good is indicated in table III, together with the total pain relief scores that were recorded. Only in the nalbuphine 60 mg group did at least 50 % of the patients achieve these degrees of pain relief. Two hours after medication, nalbuphine 60 mg provided significantly better pain relief than dihydrocodeine (P < 0.01, Mann-Whitney U test). The number and type of side-effects encountered are listed in table IV and were typically those caused by opioids. There was no statistical difference between the groups. One patient in each of the nalbuphine groups was treated (with an antiemetic) for side-effects before additional analgesics were administered. DISCUSSION
The comparison made in the present study was primarily that between nalbuphine 60 mg and dihydrocodeine 30 mg and this showed statistical
(1 severe, 2 mild) (moderate) (mild) (mild) (moderate, 1 vomited) (mild) (mild) (mild) (moderate)
(mild)
differences between the groups in favour of nalbuphine before withdrawals seriously affected the analysis. Most early withdrawals occurred in the group of patients who received nalbuphine 30 mg, but this observation may be related less to the ineffectiveness of this dose than to the fact that this group contained more patients with severe pain than the other groups, a quirk of random allocation. For these reasons this group was omitted from the analysis of variance used to compare the other groups but, as can be seen in figure 1 (mean pain intensity scores (including substituted values)) and table III (the pain relief obtained), nalbuphine 30 mg was at least as good an analgesic as dihydrocodeine 30 mg and, like the larger dose of nalbuphine, was still effective after 4 h, whereas dihydrocodeine was not. The statistical assessment of duration of effect was made by comparing pain intensity before treatment with that at the end of the observation period using Student's paired t test. This method of analysis avoids the theoretical problems of assessing probability after multiple testing. For the same reason, between-group comparisons of pain relief were confined to the anticipated time of
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
Side-effect
Group
ORAL NALBUPHINE AFTER DENTAL EXTRACTIONS
REFERENCES 1. Beaver WT, Feiser GA. A comparison of the analgesic effect of i.m. nalbuphine with morphine in patients with post-operative pain. Journal of Pharmacology and Experimental Therapeutics 1978; 204: 487-496. 2. Fahmy NR. Agonist/antagonist opioid analgesics. In: Estafanous FG, ed. Opioids in Anesthesia. London: Butterworth; 1984; 27. 3. Romagnoli A, Keats AS. Ceiling effect for respiratory depression by nalbuphine. Clinical Pharmacology and Therapeutics 1980; 27: 478-185. 4. Kay B, Krishnan A. On-demand nalbuphine for postoperative pain relief. Acta Anaesthesiologica Belgica 1986; 37: 33-37. 5. Wood M, Wood AJ, eds. In: Drugs and Anaesthesia. London: Williams and Wilkins, 1982; 171. 6. Dixon WJ. BMDP Statistical Software. Berkeley: University of California Press; 1979. 7. Iver R. Continuation-Odds Models of Ordinal Variable Regression. GLIM Newsletter 10. Oxford: Numerical Algorithms Group, 1984.
Downloaded from http://bja.oxfordjournals.org/ at Pennsylvania State University on May 24, 2015
maximum effect (2 h). From previous experience it was not anticipated that the maximum effect of oral dihydrocodeine after surgery would appear earlier, and this biased this assessment in favour of nalbuphine. In the other analyses the FisherTukey approach was used to maintain at least a minimum statistical significance level when a large number of comparisons are made. The results of the investigation indicate that nalbuphine by mouth reduces pain after dental surgery. However, whilst a dose of 30 mg is as effective as, and acts for longer than dihydrocodeine 30 mg, nalbuphine 60 mg is significantly more effective than dihydrocodeine, and was the only medication to produce good or complete pain relief in more than 50% of the patients who received it.
317