- Email: [email protected]
Prospective randomized, double‐blind comparative study of dexamethasone, ondansetron, and ondansetron plus dexamethasone as prophylactic antiemetic therapy in patients undergoing day‐case gynaecological surgery
British Journal of Anaesthesia 87 (4): 588±92 (2001)
Prospective randomized, double-blind comparative study of dexamethasone, ondansetron, and ondansetron plus dexamethasone as prophylactic antiemetic therapy in patients undergoing day-case gynaecological surgery R. Thomas* and N. Jones Department of Anaesthetics, North Hampshire Hospital, Aldermaston Road, Basingstoke, Hampshire RG24 9NA, UK *Corresponding author: Shackleton Department of Anaesthetics, Southampton General Hospital, Tremona Road, Southampton, Hampshire SO16 6YD, UK Dexamethasone alone and in combination with selective 5-hydroxytryptamine receptor antagonists is of bene®t in the prophylaxis of post-operative nausea and vomiting. In this study, the effectiveness of such a combination in comparison to either drug alone is investigated in day case gynaecological surgery. A total of 177 patients were randomized to three treatment groups: dexamethasone 8 mg, ondansetron 4 mg, and dexamethasone 8 mg plus ondansetron 4 mg. The only signi®cant difference between groups was seen in the ®rst 3 h when failure of prophylaxis was more frequent in patients who had received dexamethasone alone (P=0.0085; Fisher's exact probability test). Con®dence interval analysis indicates a modest treatment effect for the combination and the decision whether to perform a larger study depends upon whether such an effect is clinically relevant. Br J Anaesth 2001; 87: 588±92 Keywords: vomiting, antiemetics; vomiting, incidence; vomiting, nausea Accepted for publication: June 12, 2001
A recent meta-analysis examined the role of dexamethasone for the prevention of postoperative nausea and vomiting.1 It concluded that `it is very likely that the best prophylaxis of postoperative nausea and vomiting (PONV) currently available is by combining dexamethasone with a selective 5-hydroxytryptamine type 3 (5-HT3) receptor antagonist'. Such combinations are both safe and ef®cacious in paediatric,2 3 obstetric,4 breast,5 middle ear,6 and other7 surgery associated with a high risk of PONV. Despite work demonstrating the bene®ts of such combination therapy in major gynaecological surgery,8±11 little evidence exists regarding the use of an ondansetron and dexamethasone combination in a day-surgical gynaecological population.12 PONV occurs frequently after day case gynaecological surgery.13±15 Both 5-HT3 receptor antagonists14 16 and dexamethasone17 are superior to placebo in the prophylaxis of PONV in this setting. Our study has compared the ef®cacy of ondansetron plus dexamethasone with each antiemetic alone for the prevention of PONV in patients undergoing day case gynaecological surgery.
Patients and methods After obtaining approval from the local research ethical committee and written informed consent, we studied 177 ASA I or II pre-menopausal female patients, aged 19±53 yr, undergoing day case gynaecological surgery. Patients with gastrointestinal, liver, or renal diseases, those who had received any antiemetic medication within 24 h before surgery or complained of pre-operative nausea or vomiting, and those who were pregnant, lactating or with a body mass index of greater than 35 were excluded. Patients were allocated randomly to receive one of three treatment regimens: ondansetron 4 mg, dexamethasone 8 mg, or ondansetron 4 mg with dexamethasone 8 mg. A randomization list was prepared using a mechanical randomization device. The drugs were administered by slow i.v. injection immediately after induction of anaesthesia. Personnel not involved in the study prepared identical syringes containing the study drug(s). The drug(s) were diluted with normal saline to achieve a volume of 5 ml. The study drugs are known to be compatible when mixed.18
Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2001
Antiemetic therapy in day-case gynaecological surgery
A standardized anaesthetic technique was used. This was administered by one of the two investigators who were unaware of which treatment the patient had been given. Patients received no pre-anaesthetic medication. Anaesthesia was induced with 2.0±2.5 mg kg±1 propofol and 1 mg kg±1 fentanyl and maintained with 2.0±3.0% sevo¯urane (inspired concentration) and 66% nitrous oxide in oxygen. Ventilation was controlled mechanically via a laryngeal mask airway (LMA²) and adjusted to maintain end-tidal carbon dioxide concentration at 4.5±5.3 kPa throughout surgery. Neuromuscular relaxation was achieved with mivacurium 0.1±0.2 mg kg±1 and was not reversed. All patients received a 100 mg diclofenac suppository before surgery started. In addition, tramadol 100 mg i.v. was administered to those patients who underwent laparoscopic procedures.
Patients were transferred to the recovery area once adequate spontaneous respiration was established and the LMA was removed once the patient was awake. Whilst in recovery, rescue analgesia was provided by a further dose of tramadol 100 mg i.v. if required. If patients experienced PONV whilst in hospital, cyclizine 50 mg i.m. was administered as rescue antiemesis. On discharge home, patients were provided with cyclizine 50 mg tablets to be taken if needed. All patients were discharged with a 3-day supply of analgesics. Those who had undergone laparoscopic procedures were provided with diclofenac 50 mg and codeine phosphate 30 mg/paracetamol 500 mg tablets. Other patients were provided with dextropropoxyphene 32.5 mg/paracetamol 325 mg tablets. Patients were advised to take their analgesic medication regularly for the ®rst 24 h post-discharge.
Table 1 Patient characteristics. Values are numbers or means (SD) Group
Dexamethasone
Ondansetron
Combination
n Age (yr) Body mass index Past history of PONV or travel sickness Non-smokers Migraine sufferers Menstruation Duration of anaesthesia (min) Laparoscopic procedure Laparoscopic sterilization Procedure involving cervical dilation Rescue analgesia provided in recovery (tramadol)
60 34.9 (20±50) 25.1 (4.2) 35 40 20 4 30.5 (9.94) 33 13 53 3
59 36.2 (21±53) 23.8 (3.5) 34 41 18 9 29.7 (7.41) 33 18 52 4
58 36.0 (19±53) 24.1 (3.3) 33 32 18 8 30.2 (9.3) 38 14 55 6
Table 2 Incidence of nausea, retching, vomiting, use of rescue antiemetics and patients in whom prophylaxis failed. Values are numbers (proportions)
Group 0±3 h
Overall proportion of subjects classi®ed as `failures' Combination vs dexamethasone: 0.2 (95% CI: 0.06 to 0.33, 99% CI: 0.01 to 0.37) Combination vs ondansetron: 0.13 (95% CI: 0.01 to 0.27) Ondansetron vs dexamethasone: 0.063 (95% CI: ±0.09 to 0.22) Group 3±12 h
Overall proportion of subjects classi®ed as `failures' Combination vs dexamethasone: 0.11 (95% CI: ±0.06 to 0.27) Combination vs ondansetron: 0.064 (95% CI: ±0.10 to 0.22) Ondansetron vs dexamethasone: 0.045 (95% CI: ±0.12 to 0.21) Group 12±24 h
Overall proportion of subjects classi®ed as `failures'
589
Dexamethasone
Ondansetron
Combination
Nausea Retching Vomiting Rescue antiemetic
16 5 4 8 17/60 (0.283)
13 1 0 7 13/59 (0.22)
5 1 1 2 5/58 (0.086)
Nausea Retching Vomiting Rescue antiemetic
20 4 6 16 21/60 (0.35)
16 1 4 12 18/59 (0.305)
14 1 3 12 14/58 (0.241)
Nausea Retching Vomiting Rescue antiemetic
7 0 0 7 9/60 (0.15)
6 0 1 7 8/59 (0.135)
8 0 0 8 9/58 (0.155)
Thomas and Jones
Sample size was determined using the software package nQuery Advisor v 3.0 (Statistical Solutions Ltd, Cork, Ireland). The progress of the study was monitored by an independent third party who kept a record of numbers per treatment group and terminated the study when a minimum of 58 patients in each group was obtained. The statistical signi®cance of observed differences in the treatment outcomes was assessed using the recommended method (Wilson20) for comparing two proportions in the software package Con®dence Interval Analysis v. 2.0.0 (BMJ Books, 2000). To make allowances for multiple testing, all differences found to be signi®cant at the 5% level were re-tested at the 1% level (using 99% con®dence interval). Thence, the exact probability of any differences surviving such tests was obtained using Fisher's exact test (StatXact v. 3.0, Cytel Software Corporation, MA, USA).
Patients were interviewed on discharge from the day surgical unit (3 h post-surgery) by blinded nursing staff and by phone at 24 h post-surgery by one of the two blinded investigators. The patients were asked if retching or vomiting had occurred, whether rescue antiemetics had been used and if they had felt nauseated in three time periods: 0±3, 3±12, and 12±24 h post-surgery, with only two possible answers (yes/no). Nausea was de®ned as the unpleasant sensation associated with awareness of the urge to vomit; retching was de®ned as laboured, spasmodic, rhythmic contraction of the respiratory muscles without expulsion of gastric contents; vomiting was de®ned as the forceful expulsion of gastric contents from the mouth.19 Failure of prophylaxis was de®ned as any episode of nausea, retching, vomiting or use of rescue antiemetic. The nursing staff and investigators recorded details of adverse events using an open questioning technique. The patients were asked about analgesic consumption after leaving hospital and pain was scored at 3 and 24 h using a simple scoring system (1, no pain; 2, mild pain; 3, moderate pain; 4, severe pain). A sample-size of 174 patients (58 in each of three treatment groups) was required to detect a difference in response rates of 0.25 from a baseline prevalence of 0.5, with 80% power at a two-sided signi®cance level of 5%.
Results Patient characteristics and surgical procedures were similar in each group (Table 1). Failure of prophylaxis during the ®rst 3 h (0±3 h) after surgery was recorded in 22, 28.3, and 8.6% of patients who had received ondansetron, dexamethasone, and ondansetron/dexamethasone, respectively. The incidences during the next 9 h (3±12) were 30.5, 35, and 24.1%, respectively. The incidences during the following 12 h (12±24) were 13.5, 15, and 15.5%, respectively. The overall incidences for the 24 h post surgery were 42.4, 48.3, and 34.5%, respectively. Thus, only in the ®rst 3 h was failure of prophylaxis seen signi®cantly less often in patients who had received a combination of ondansetron and dexamethasone than in those who had received dexamethasone alone: exact probability P=0.0085. For ondansetron, the difference was signi®cant at the 5% level, but not by our stricter 1% criterion (see Table 2). No patient required hospital admission for PONV. The most frequently reported adverse events were fatigue, headache, and dizziness, but there were
Table 3 Incidence of adverse effects. Values are numbers Group
Dexamethasone
Ondansetron
Combination
Fatigue Dizziness Headache Pruritis Flushing Cough Diarrhoea Transient chest pain Heartburn
1 1 1 1 0 0 0 2 1
4 2 2 1 0 0 1 0 0
2 3 1 0 3 1 0 0 0
Table 4 Incidence and severity of pain and use of opiate analgesia after hospital discharge. Values are numbers
Group at 3 h
Group at 24 h
Use of opiate analgesia after hospital discharge in ®rst 24 hrs (either codeine phosphate or dextropropoxyphene )
Dexamethasone
Ondansetron
Combination
No pain Mild pain Moderate pain Severe pain
25 24 11 0
30 21 7 1
28 22 8 0
No pain Mild pain Moderate pain Severe pain
21 25 14 0
22 25 10 1
21 26 11 0
47
43
46
590
Antiemetic therapy in day-case gynaecological surgery
no differences between groups (Table 3). There were no differences with respect to analgesic consumption and the incidence and severity of pain (Table 4).
Discussion The aetiology of PONV after gynaecological surgery performed under general anaesthesia remains elusive, but is probably multifactorial. Age, gender, smoking habits, past history of previous PONV and/or motion sickness, surgical procedure, length of procedure, and use of postoperative opioids are of importance in predicting PONV.21±23 In our study, these factors were well balanced between groups and, therefore, differences between groups may be attributed to differences in the antiemetic drugs administered. We decided to include a heterogeneous group of operations in our study to increase the ability to generalize our ®ndings. Although theoretically, the baseline incidences of and mechanisms for PONV may be different for different operations, evidence for this is lacking in the group of operations we included.23 We did speci®cally exclude the procedures of dilation and curettage and termination of pregnancy, as the incidence of PONV is lower.23 As PONV is recognized to be a common complication of gynaecological surgery performed under general anaesthesia,24 we did not believe it to be ethical to include a placebo arm in our study. We based our choice of a baseline prevalence for PONV of 0.5 on the ®ndings of other studies8 10 13 24 and on local audit data. The high incidence of PONV in this population may justify the use of prophylactic antiemetics. Using antiemetics in this way, however, does raise cost-effectiveness issues.25 At the drug dosages used, we found that ondansetron and dexamethasone performed similarly, despite an approximate 7-fold price differential. Our selection of drug dosages was based on previous work that demonstrated that these doses were effective.26 27 Our study demonstrated a signi®cant difference between the ondansetron/dexamethasone combination and either drug alone in the 0±3 h period. Although this difference failed to maintain signi®cance overall (0±24 h), or during the 3±12 or 12±24 h periods, the 95% con®dence intervals show a trend suggesting that the combination has some bene®t over the individual drugs. This is probably a result of underpowering of our study for the size of the actual difference observed. In designing this study, we had to choose an arbitrary reduction in PONV that we believed clinically important. We based this upon a reduction of 50% in the incidence of PONV in the combination group when compared with either of the single treatment groups. We believed this to be a reasonable and clinically important effect that would lead to anaesthetists using the ondansetron/dexamethasone combination in preference to either drug alone.
Our failure to demonstrate signi®cance overall (0±24 h) may be because of other reasons. It could be argued that our base line prevalence for PONV of 0.5 was too high, in the absence of a placebo group. This would result in an underestimation of sample size. The fact that the dexamethasone treatment group demonstrated a PONV rate of 0.48 would argue against this, however. Additionally, we may have reduced the likelihood of a signi®cant result by our decision to include a heterogeneous group of operations. This decision has been discussed above. It may be that a smaller than 50% reduction in PONV as a result of using the combination therapy in preference to single agent therapy is still clinically valuable. This is in the context of a day case gynaecological population as a whole or for certain high risk patients within it. We estimate that 1565 patients per group would be required in an identical study of the same power investigating a 10% difference between treatment groups. Other work has suggested that dexamethasone alone28 or in combination,1 12 may be particularly ef®cacious in preventing late or delayed PONV. Our results do not support this, but this may be because of problems associated with our study design as discussed above.
References
591
1 Henzi I, Walder B, Tramer MR. Dexamethasone for the prevention of postoperative nausea and vomiting: a quantitative systematic review. Anesth Analg 2000; 90: 186±94 2 Fujii Y, Tanaka H, Toyooka H. Granisetron and dexamethasone provide more improved prevention of postoperative emesis than granisetron alone in children. Can J Anaesth 1996; 43: 1229±32 3 Fujii Y, Saitoh Y, Tanaka H, Toyooka H. Prophylactic therapy with combined granisetron and dexamethasone for the prevention of post-operative vomiting in children. Eur J Anaesthesiol 1999; 16: 376±9 4 Fujii Y, Saitoh Y, Tanaka H, Toyooka H. Granisetron/ dexamethasone combination for reducing nausea and vomiting during and after spinal anesthesia for cesarean section. Anesth Analg 1999; 88: 1346±50 5 Fujii Y, Tanaka H, Toyooka H. Prophylactic antiemetic therapy with granisetron-dexamethasone combination in women undergoing breast surgery. Acta Anaesthesiol Scand 1998; 42: 1038±42 6 Fujii Y, Tanaka H, Toyooka H. Prophylactic antiemetic therapy with granisetron-dexamethasone in patients undergoing middle ear surgery. Br J Anaesth 1998; 81: 754±6 7 Janknegt R, Pinckaers JW, Rohof MH, et al. Double-blind comparative study of droperidol, granisetron and granisetron plus dexamethasone as prophylactic anti-emetic therapy in patients undergoing abdominal, gynaecological, breast or otolaryngological surgery. Anaesthesia 1999; 54: 1059±68 8 Fujii Y, Tanaka H, Toyooka H. The effects of dexamethasone on antiemetics in female patients undergoing gynecological surgery. Anesth Analg 1997; 85: 913±7 9 Fujii Y, Tanaka H, Toyooka H. Granisetron-dexamethasone combination reduces postoperative nausea and vomiting. Can J Anaesth 1995; 42: 387±90 10 McKenzie R, Tantisira B, Karambelkar DJ, Riley TJ, Abdelhady H. Comparison of ondansetron with ondansetron plus
Thomas and Jones
11
12 13
14
15
16
17 18
dexamethasone in the prevention of postoperative nausea and vomiting. Anesth Analg 1994; 79: 961±4 Lopez-Olaondo L, Carrascosa F, Pueyo FJ, Monedero P, Busto N, Saez A. Combination of ondansetron and dexamethasone in the prophylaxis of postoperative nausea and vomiting. Br J Anaesth 1996; 76: 835±40 Rajeeva V, Bhardwaj N, Batra YK, Dhaliwal LK. Comparison of ondansetron and dexamethasone in prevention of PONV in diagnostic laparoscopy. Can J Anaesth 1999; 46: 40±4 Cholwill JM, Wright W, Hobbs GJ, Curran J. Comparison of ondansetron and cyclizine for prevention of nausea and vomiting after day-case gynaecological laparoscopy. Br J Anaesth 1999; 83: 611±4 McKenzie R, Kovac A, O'Connor T, et al. Comparison of ondansetron versus placebo to prevent postoperative nausea and vomiting in women undergoing ambulatory gynecological surgery. Anesthesiology 1993; 78: 21±8 Malins AF, Field JM, Nestling PM, Cooper GM. Nausea and vomiting after gynaecological laparoscopy: comparison of premedication with oral ondansetron, metoclopramide and placebo. Br J Anaesth 1994; 72: 231±3 Paxton LD, McKay AC, Mirakhur RK. Prevention of nausea and vomiting after day case gynaecological laparoscopy. A comparison of ondansetron, droperidol, metoclopramide and placebo. Anaesthesia 1995; 50: 403±6 Wang JJ, Ho ST, Liu YH, et al. Prophylactic antiemetic effect of dexamethasone in women undergoing ambulatory laparoscopic surgery. Br J Anaesth 2000; 84: 459±62 Trissel LA. Handbook of Injectable Drugs, 11th edn. Special
592
19 20 21 22 23 24 25 26 27 28
Publishing Department of the American Society of HealthSystem Pharmacists; 2001 Watcha MF, White PF. Postoperative nausea and vomiting. Its etiology, treatment, and prevention. Anesthesiology 1992; 77: 162±84 Wilson EB. Probable inference, the law of succession, and statistical inference. J Am Stat Assoc 1927; 22: 209±12 Toner CC, Broomhead CJ, Littlejohn IH, et al. Prediction of postoperative nausea and vomiting using a logistic regression model. Br J Anaesth 1996; 76: 347±51 Apfel CC, Greim CA, Haubitz I, et al. The discriminating power of a risk score for postoperative vomiting in adults undergoing various types of surgery. Acta Anaesthesiol Scand 1998; 42: 502±9 Sinclair DR, Chung F, Mezei G. Can postoperative nausea and vomiting be predicted? Anesthesiology 1999; 91: 109±18 Koivuranta M, Laara E, Snare L, Alahuhta S. A survey of postoperative nausea and vomiting. Anaesthesia 1997; 52: 443±9 Watcha MF. The cost-effective management of postoperative nausea and vomiting. Anesthesiology 2000; 92: 931±3 Pearman MH. Single dose intravenous ondansetron in the prevention of postoperative nausea and vomiting. Anaesthesia 1994; 49 (Suppl): 11±15 Liu K, Hsu CC, Chia YY. The effect of dose of dexamethasone for antiemesis after major gynecological surgery. Anesth Analg 1999; 89: 1316±8 Rothenberg DM, McCarthy RJ, Peng CC, Normoyle DA. Nausea and vomiting after dexamethasone versus droperidol following outpatient laparoscopy with a propofol-based general anesthetic. Acta Anaesthesiol Scand 1998; 42: 637±42
Prospective randomized, double-blind comparative study of dexamethasone, ondansetron, and ondansetron plus dexamethasone as prophylactic antiemetic therapy in patients undergoing day-case gynaecological surgery R. Thomas* and N. Jones Department of Anaesthetics, North Hampshire Hospital, Aldermaston Road, Basingstoke, Hampshire RG24 9NA, UK *Corresponding author: Shackleton Department of Anaesthetics, Southampton General Hospital, Tremona Road, Southampton, Hampshire SO16 6YD, UK Dexamethasone alone and in combination with selective 5-hydroxytryptamine receptor antagonists is of bene®t in the prophylaxis of post-operative nausea and vomiting. In this study, the effectiveness of such a combination in comparison to either drug alone is investigated in day case gynaecological surgery. A total of 177 patients were randomized to three treatment groups: dexamethasone 8 mg, ondansetron 4 mg, and dexamethasone 8 mg plus ondansetron 4 mg. The only signi®cant difference between groups was seen in the ®rst 3 h when failure of prophylaxis was more frequent in patients who had received dexamethasone alone (P=0.0085; Fisher's exact probability test). Con®dence interval analysis indicates a modest treatment effect for the combination and the decision whether to perform a larger study depends upon whether such an effect is clinically relevant. Br J Anaesth 2001; 87: 588±92 Keywords: vomiting, antiemetics; vomiting, incidence; vomiting, nausea Accepted for publication: June 12, 2001
A recent meta-analysis examined the role of dexamethasone for the prevention of postoperative nausea and vomiting.1 It concluded that `it is very likely that the best prophylaxis of postoperative nausea and vomiting (PONV) currently available is by combining dexamethasone with a selective 5-hydroxytryptamine type 3 (5-HT3) receptor antagonist'. Such combinations are both safe and ef®cacious in paediatric,2 3 obstetric,4 breast,5 middle ear,6 and other7 surgery associated with a high risk of PONV. Despite work demonstrating the bene®ts of such combination therapy in major gynaecological surgery,8±11 little evidence exists regarding the use of an ondansetron and dexamethasone combination in a day-surgical gynaecological population.12 PONV occurs frequently after day case gynaecological surgery.13±15 Both 5-HT3 receptor antagonists14 16 and dexamethasone17 are superior to placebo in the prophylaxis of PONV in this setting. Our study has compared the ef®cacy of ondansetron plus dexamethasone with each antiemetic alone for the prevention of PONV in patients undergoing day case gynaecological surgery.
Patients and methods After obtaining approval from the local research ethical committee and written informed consent, we studied 177 ASA I or II pre-menopausal female patients, aged 19±53 yr, undergoing day case gynaecological surgery. Patients with gastrointestinal, liver, or renal diseases, those who had received any antiemetic medication within 24 h before surgery or complained of pre-operative nausea or vomiting, and those who were pregnant, lactating or with a body mass index of greater than 35 were excluded. Patients were allocated randomly to receive one of three treatment regimens: ondansetron 4 mg, dexamethasone 8 mg, or ondansetron 4 mg with dexamethasone 8 mg. A randomization list was prepared using a mechanical randomization device. The drugs were administered by slow i.v. injection immediately after induction of anaesthesia. Personnel not involved in the study prepared identical syringes containing the study drug(s). The drug(s) were diluted with normal saline to achieve a volume of 5 ml. The study drugs are known to be compatible when mixed.18
Ó The Board of Management and Trustees of the British Journal of Anaesthesia 2001
Antiemetic therapy in day-case gynaecological surgery
A standardized anaesthetic technique was used. This was administered by one of the two investigators who were unaware of which treatment the patient had been given. Patients received no pre-anaesthetic medication. Anaesthesia was induced with 2.0±2.5 mg kg±1 propofol and 1 mg kg±1 fentanyl and maintained with 2.0±3.0% sevo¯urane (inspired concentration) and 66% nitrous oxide in oxygen. Ventilation was controlled mechanically via a laryngeal mask airway (LMA²) and adjusted to maintain end-tidal carbon dioxide concentration at 4.5±5.3 kPa throughout surgery. Neuromuscular relaxation was achieved with mivacurium 0.1±0.2 mg kg±1 and was not reversed. All patients received a 100 mg diclofenac suppository before surgery started. In addition, tramadol 100 mg i.v. was administered to those patients who underwent laparoscopic procedures.
Patients were transferred to the recovery area once adequate spontaneous respiration was established and the LMA was removed once the patient was awake. Whilst in recovery, rescue analgesia was provided by a further dose of tramadol 100 mg i.v. if required. If patients experienced PONV whilst in hospital, cyclizine 50 mg i.m. was administered as rescue antiemesis. On discharge home, patients were provided with cyclizine 50 mg tablets to be taken if needed. All patients were discharged with a 3-day supply of analgesics. Those who had undergone laparoscopic procedures were provided with diclofenac 50 mg and codeine phosphate 30 mg/paracetamol 500 mg tablets. Other patients were provided with dextropropoxyphene 32.5 mg/paracetamol 325 mg tablets. Patients were advised to take their analgesic medication regularly for the ®rst 24 h post-discharge.
Table 1 Patient characteristics. Values are numbers or means (SD) Group
Dexamethasone
Ondansetron
Combination
n Age (yr) Body mass index Past history of PONV or travel sickness Non-smokers Migraine sufferers Menstruation Duration of anaesthesia (min) Laparoscopic procedure Laparoscopic sterilization Procedure involving cervical dilation Rescue analgesia provided in recovery (tramadol)
60 34.9 (20±50) 25.1 (4.2) 35 40 20 4 30.5 (9.94) 33 13 53 3
59 36.2 (21±53) 23.8 (3.5) 34 41 18 9 29.7 (7.41) 33 18 52 4
58 36.0 (19±53) 24.1 (3.3) 33 32 18 8 30.2 (9.3) 38 14 55 6
Table 2 Incidence of nausea, retching, vomiting, use of rescue antiemetics and patients in whom prophylaxis failed. Values are numbers (proportions)
Group 0±3 h
Overall proportion of subjects classi®ed as `failures' Combination vs dexamethasone: 0.2 (95% CI: 0.06 to 0.33, 99% CI: 0.01 to 0.37) Combination vs ondansetron: 0.13 (95% CI: 0.01 to 0.27) Ondansetron vs dexamethasone: 0.063 (95% CI: ±0.09 to 0.22) Group 3±12 h
Overall proportion of subjects classi®ed as `failures' Combination vs dexamethasone: 0.11 (95% CI: ±0.06 to 0.27) Combination vs ondansetron: 0.064 (95% CI: ±0.10 to 0.22) Ondansetron vs dexamethasone: 0.045 (95% CI: ±0.12 to 0.21) Group 12±24 h
Overall proportion of subjects classi®ed as `failures'
589
Dexamethasone
Ondansetron
Combination
Nausea Retching Vomiting Rescue antiemetic
16 5 4 8 17/60 (0.283)
13 1 0 7 13/59 (0.22)
5 1 1 2 5/58 (0.086)
Nausea Retching Vomiting Rescue antiemetic
20 4 6 16 21/60 (0.35)
16 1 4 12 18/59 (0.305)
14 1 3 12 14/58 (0.241)
Nausea Retching Vomiting Rescue antiemetic
7 0 0 7 9/60 (0.15)
6 0 1 7 8/59 (0.135)
8 0 0 8 9/58 (0.155)
Thomas and Jones
Sample size was determined using the software package nQuery Advisor v 3.0 (Statistical Solutions Ltd, Cork, Ireland). The progress of the study was monitored by an independent third party who kept a record of numbers per treatment group and terminated the study when a minimum of 58 patients in each group was obtained. The statistical signi®cance of observed differences in the treatment outcomes was assessed using the recommended method (Wilson20) for comparing two proportions in the software package Con®dence Interval Analysis v. 2.0.0 (BMJ Books, 2000). To make allowances for multiple testing, all differences found to be signi®cant at the 5% level were re-tested at the 1% level (using 99% con®dence interval). Thence, the exact probability of any differences surviving such tests was obtained using Fisher's exact test (StatXact v. 3.0, Cytel Software Corporation, MA, USA).
Patients were interviewed on discharge from the day surgical unit (3 h post-surgery) by blinded nursing staff and by phone at 24 h post-surgery by one of the two blinded investigators. The patients were asked if retching or vomiting had occurred, whether rescue antiemetics had been used and if they had felt nauseated in three time periods: 0±3, 3±12, and 12±24 h post-surgery, with only two possible answers (yes/no). Nausea was de®ned as the unpleasant sensation associated with awareness of the urge to vomit; retching was de®ned as laboured, spasmodic, rhythmic contraction of the respiratory muscles without expulsion of gastric contents; vomiting was de®ned as the forceful expulsion of gastric contents from the mouth.19 Failure of prophylaxis was de®ned as any episode of nausea, retching, vomiting or use of rescue antiemetic. The nursing staff and investigators recorded details of adverse events using an open questioning technique. The patients were asked about analgesic consumption after leaving hospital and pain was scored at 3 and 24 h using a simple scoring system (1, no pain; 2, mild pain; 3, moderate pain; 4, severe pain). A sample-size of 174 patients (58 in each of three treatment groups) was required to detect a difference in response rates of 0.25 from a baseline prevalence of 0.5, with 80% power at a two-sided signi®cance level of 5%.
Results Patient characteristics and surgical procedures were similar in each group (Table 1). Failure of prophylaxis during the ®rst 3 h (0±3 h) after surgery was recorded in 22, 28.3, and 8.6% of patients who had received ondansetron, dexamethasone, and ondansetron/dexamethasone, respectively. The incidences during the next 9 h (3±12) were 30.5, 35, and 24.1%, respectively. The incidences during the following 12 h (12±24) were 13.5, 15, and 15.5%, respectively. The overall incidences for the 24 h post surgery were 42.4, 48.3, and 34.5%, respectively. Thus, only in the ®rst 3 h was failure of prophylaxis seen signi®cantly less often in patients who had received a combination of ondansetron and dexamethasone than in those who had received dexamethasone alone: exact probability P=0.0085. For ondansetron, the difference was signi®cant at the 5% level, but not by our stricter 1% criterion (see Table 2). No patient required hospital admission for PONV. The most frequently reported adverse events were fatigue, headache, and dizziness, but there were
Table 3 Incidence of adverse effects. Values are numbers Group
Dexamethasone
Ondansetron
Combination
Fatigue Dizziness Headache Pruritis Flushing Cough Diarrhoea Transient chest pain Heartburn
1 1 1 1 0 0 0 2 1
4 2 2 1 0 0 1 0 0
2 3 1 0 3 1 0 0 0
Table 4 Incidence and severity of pain and use of opiate analgesia after hospital discharge. Values are numbers
Group at 3 h
Group at 24 h
Use of opiate analgesia after hospital discharge in ®rst 24 hrs (either codeine phosphate or dextropropoxyphene )
Dexamethasone
Ondansetron
Combination
No pain Mild pain Moderate pain Severe pain
25 24 11 0
30 21 7 1
28 22 8 0
No pain Mild pain Moderate pain Severe pain
21 25 14 0
22 25 10 1
21 26 11 0
47
43
46
590
Antiemetic therapy in day-case gynaecological surgery
no differences between groups (Table 3). There were no differences with respect to analgesic consumption and the incidence and severity of pain (Table 4).
Discussion The aetiology of PONV after gynaecological surgery performed under general anaesthesia remains elusive, but is probably multifactorial. Age, gender, smoking habits, past history of previous PONV and/or motion sickness, surgical procedure, length of procedure, and use of postoperative opioids are of importance in predicting PONV.21±23 In our study, these factors were well balanced between groups and, therefore, differences between groups may be attributed to differences in the antiemetic drugs administered. We decided to include a heterogeneous group of operations in our study to increase the ability to generalize our ®ndings. Although theoretically, the baseline incidences of and mechanisms for PONV may be different for different operations, evidence for this is lacking in the group of operations we included.23 We did speci®cally exclude the procedures of dilation and curettage and termination of pregnancy, as the incidence of PONV is lower.23 As PONV is recognized to be a common complication of gynaecological surgery performed under general anaesthesia,24 we did not believe it to be ethical to include a placebo arm in our study. We based our choice of a baseline prevalence for PONV of 0.5 on the ®ndings of other studies8 10 13 24 and on local audit data. The high incidence of PONV in this population may justify the use of prophylactic antiemetics. Using antiemetics in this way, however, does raise cost-effectiveness issues.25 At the drug dosages used, we found that ondansetron and dexamethasone performed similarly, despite an approximate 7-fold price differential. Our selection of drug dosages was based on previous work that demonstrated that these doses were effective.26 27 Our study demonstrated a signi®cant difference between the ondansetron/dexamethasone combination and either drug alone in the 0±3 h period. Although this difference failed to maintain signi®cance overall (0±24 h), or during the 3±12 or 12±24 h periods, the 95% con®dence intervals show a trend suggesting that the combination has some bene®t over the individual drugs. This is probably a result of underpowering of our study for the size of the actual difference observed. In designing this study, we had to choose an arbitrary reduction in PONV that we believed clinically important. We based this upon a reduction of 50% in the incidence of PONV in the combination group when compared with either of the single treatment groups. We believed this to be a reasonable and clinically important effect that would lead to anaesthetists using the ondansetron/dexamethasone combination in preference to either drug alone.
Our failure to demonstrate signi®cance overall (0±24 h) may be because of other reasons. It could be argued that our base line prevalence for PONV of 0.5 was too high, in the absence of a placebo group. This would result in an underestimation of sample size. The fact that the dexamethasone treatment group demonstrated a PONV rate of 0.48 would argue against this, however. Additionally, we may have reduced the likelihood of a signi®cant result by our decision to include a heterogeneous group of operations. This decision has been discussed above. It may be that a smaller than 50% reduction in PONV as a result of using the combination therapy in preference to single agent therapy is still clinically valuable. This is in the context of a day case gynaecological population as a whole or for certain high risk patients within it. We estimate that 1565 patients per group would be required in an identical study of the same power investigating a 10% difference between treatment groups. Other work has suggested that dexamethasone alone28 or in combination,1 12 may be particularly ef®cacious in preventing late or delayed PONV. Our results do not support this, but this may be because of problems associated with our study design as discussed above.
References
591
1 Henzi I, Walder B, Tramer MR. Dexamethasone for the prevention of postoperative nausea and vomiting: a quantitative systematic review. Anesth Analg 2000; 90: 186±94 2 Fujii Y, Tanaka H, Toyooka H. Granisetron and dexamethasone provide more improved prevention of postoperative emesis than granisetron alone in children. Can J Anaesth 1996; 43: 1229±32 3 Fujii Y, Saitoh Y, Tanaka H, Toyooka H. Prophylactic therapy with combined granisetron and dexamethasone for the prevention of post-operative vomiting in children. Eur J Anaesthesiol 1999; 16: 376±9 4 Fujii Y, Saitoh Y, Tanaka H, Toyooka H. Granisetron/ dexamethasone combination for reducing nausea and vomiting during and after spinal anesthesia for cesarean section. Anesth Analg 1999; 88: 1346±50 5 Fujii Y, Tanaka H, Toyooka H. Prophylactic antiemetic therapy with granisetron-dexamethasone combination in women undergoing breast surgery. Acta Anaesthesiol Scand 1998; 42: 1038±42 6 Fujii Y, Tanaka H, Toyooka H. Prophylactic antiemetic therapy with granisetron-dexamethasone in patients undergoing middle ear surgery. Br J Anaesth 1998; 81: 754±6 7 Janknegt R, Pinckaers JW, Rohof MH, et al. Double-blind comparative study of droperidol, granisetron and granisetron plus dexamethasone as prophylactic anti-emetic therapy in patients undergoing abdominal, gynaecological, breast or otolaryngological surgery. Anaesthesia 1999; 54: 1059±68 8 Fujii Y, Tanaka H, Toyooka H. The effects of dexamethasone on antiemetics in female patients undergoing gynecological surgery. Anesth Analg 1997; 85: 913±7 9 Fujii Y, Tanaka H, Toyooka H. Granisetron-dexamethasone combination reduces postoperative nausea and vomiting. Can J Anaesth 1995; 42: 387±90 10 McKenzie R, Tantisira B, Karambelkar DJ, Riley TJ, Abdelhady H. Comparison of ondansetron with ondansetron plus
Thomas and Jones
11
12 13
14
15
16
17 18
dexamethasone in the prevention of postoperative nausea and vomiting. Anesth Analg 1994; 79: 961±4 Lopez-Olaondo L, Carrascosa F, Pueyo FJ, Monedero P, Busto N, Saez A. Combination of ondansetron and dexamethasone in the prophylaxis of postoperative nausea and vomiting. Br J Anaesth 1996; 76: 835±40 Rajeeva V, Bhardwaj N, Batra YK, Dhaliwal LK. Comparison of ondansetron and dexamethasone in prevention of PONV in diagnostic laparoscopy. Can J Anaesth 1999; 46: 40±4 Cholwill JM, Wright W, Hobbs GJ, Curran J. Comparison of ondansetron and cyclizine for prevention of nausea and vomiting after day-case gynaecological laparoscopy. Br J Anaesth 1999; 83: 611±4 McKenzie R, Kovac A, O'Connor T, et al. Comparison of ondansetron versus placebo to prevent postoperative nausea and vomiting in women undergoing ambulatory gynecological surgery. Anesthesiology 1993; 78: 21±8 Malins AF, Field JM, Nestling PM, Cooper GM. Nausea and vomiting after gynaecological laparoscopy: comparison of premedication with oral ondansetron, metoclopramide and placebo. Br J Anaesth 1994; 72: 231±3 Paxton LD, McKay AC, Mirakhur RK. Prevention of nausea and vomiting after day case gynaecological laparoscopy. A comparison of ondansetron, droperidol, metoclopramide and placebo. Anaesthesia 1995; 50: 403±6 Wang JJ, Ho ST, Liu YH, et al. Prophylactic antiemetic effect of dexamethasone in women undergoing ambulatory laparoscopic surgery. Br J Anaesth 2000; 84: 459±62 Trissel LA. Handbook of Injectable Drugs, 11th edn. Special
592
19 20 21 22 23 24 25 26 27 28
Publishing Department of the American Society of HealthSystem Pharmacists; 2001 Watcha MF, White PF. Postoperative nausea and vomiting. Its etiology, treatment, and prevention. Anesthesiology 1992; 77: 162±84 Wilson EB. Probable inference, the law of succession, and statistical inference. J Am Stat Assoc 1927; 22: 209±12 Toner CC, Broomhead CJ, Littlejohn IH, et al. Prediction of postoperative nausea and vomiting using a logistic regression model. Br J Anaesth 1996; 76: 347±51 Apfel CC, Greim CA, Haubitz I, et al. The discriminating power of a risk score for postoperative vomiting in adults undergoing various types of surgery. Acta Anaesthesiol Scand 1998; 42: 502±9 Sinclair DR, Chung F, Mezei G. Can postoperative nausea and vomiting be predicted? Anesthesiology 1999; 91: 109±18 Koivuranta M, Laara E, Snare L, Alahuhta S. A survey of postoperative nausea and vomiting. Anaesthesia 1997; 52: 443±9 Watcha MF. The cost-effective management of postoperative nausea and vomiting. Anesthesiology 2000; 92: 931±3 Pearman MH. Single dose intravenous ondansetron in the prevention of postoperative nausea and vomiting. Anaesthesia 1994; 49 (Suppl): 11±15 Liu K, Hsu CC, Chia YY. The effect of dose of dexamethasone for antiemesis after major gynecological surgery. Anesth Analg 1999; 89: 1316±8 Rothenberg DM, McCarthy RJ, Peng CC, Normoyle DA. Nausea and vomiting after dexamethasone versus droperidol following outpatient laparoscopy with a propofol-based general anesthetic. Acta Anaesthesiol Scand 1998; 42: 637±42