- Email: [email protected]
Comparison of four antiemetic regimens for the treatment of cisplatin-induced vomiting
In1 J Gynecol Obstet, 1993, 42: 19-24 International Federation of Gynecology
19 and Obstetrics
Comparison of four antiemetic regimens for the treatment cisplatin-induced vomiting R. Poka, Z. Hernadi,
of
B. Juhasz and L. Lamp6
Department of Obstetrics and Gynecology, University Medical School of Debrecen (Hungary) (Received November l&h, 1992) (Revised and accepted February 12th, 1993)
Abstract OBJECTIVE: To improve the treatment of cisplatin-induced acute emesis and vomiting, the aa’juvant effect of two different doses of metoclopramide and metoclopramide plus droperidol was compared to a defined standard antiemetic regiment. METHOD: One-hundred and twenty-one consecutive cisplatin-based chemotherapy treatments for ovarian carcinoma were randomly assigned to receive standard plus no aq’juvant (control, n = 24), high-dose metoclopramide (HDM, n = 33), very high dose metoclopramide (VHDM, n = 50) and very high dose metoclopramide plus droperidol ( VHDM + DP, n = 14) antiemetic treatment. The number of vomiting episodes up to 4 h after the treatment were recorded and compared in the four different antiemetic groups. The effect of the number of previous chemotherapy cycles, the strength of previous antiemetic treatments, age and stage of disease on the number of vomiting episodes was also assessed. RESULT.. The mean number of vomiting episodes during the first 4 h after the chemotherapy was 9, 6.4, 6.2 and 2.7 among patients receiving the standard HDM, antiemetic VHDM + DP VHDM and regimens, respectively. Analysis of variance of vomiting episodes confirmed significant differ(P < 0.001, ences between the groups F = 406.5, df = 120). The number of previous chemotherapy cycles and, to a lesser extent, the 0020-7293/93/$06.00 @ 1993 International Federation Printed and Published in Ireland
stage of disease, the strength of previous antiemetic regimens and the age showed significant correlation with the number of vomiting episodes (r = 0.443, -0.159, -0.14 and -0.009 respectively). CONCLUSION: Substantial improvement in the control of cisplating-induced vomiting can be achieved by high-dose metoclopramide and adding droperidol to a basic antiemetic regimen consisting prochlorperazine, dexamethasone and thiethylperazine.
Keywords: Ovarian cancer; Metoclopramide; Droperidol.
Cisplatin;
Introduction Cisplatin is one of the most emetogenic chemotherapeutic agents. Efficient control of cisplatin-induced nausea and vomiting is a major issue in the supportive care of ovarian carcinoma patients. Metoclopramide was shown to be effective for the control of chemotherapy-induced nausea and vomiting [6]. Positive dose-response relationship for the antiemetic effect of metoclopramide was established in a randomized crossover study [14]. Despite the established superiority of high-dose metoclopramide during the 1980s antiemetic trials often showed substantial improvement in its efficacy by adding other agents such as dexamethasone, lorazepam and Article
of Gynecology
and Obstetrics
20
Pdka et al.
chlorperazine [3,4,9,16]. A single dose combined antiemetic regimen, consisting of very high dose metoclopramide, lorazepam and dexamethasone, appeared to be equally effective as repeated low-dose regimens. A further advantage of the single dose administration is the lower cost due to less drug use and a shorter hospital stay [3]. A combination of high-dose steroid and droperidol with or without chlorpromazine was also shown to provide effective control of cisplatin-induced vomiting [ 1,111. Recent developments in pharmacological and physiological research determined the neurochemical mechanisms of nausea and vomiting. Accordingly, serotonin antagonist drugs such as ondansetron have been developed and proved superior to all other antiemetics both in terms of efficacy and side-effects [lo]. Unfortunately, for financial reasons, traditional combinations based on high-dose metoclopramide are still expected to remain in the first line for years. We therefore studied the effect of two different doses of metoclopramide and metoclopramide plus droperidol on the antiemetic efficacy of a defined standard regimen for the control of cisplatin-induced vomiting. Materials and methods During the study period (between 1st January and 31st June, 1990) 52 patients received a total of 121 chemotherapy treatments for ovarian carcinoma using cyclophosphamide/cisplatin (1000:50 or 750:75 mg/m2) or cyclophosphamide/doxorubicin/ cisplatin (500:50:50 mg/m2) combinations on a day-care basis. The mean age of the patients was 50.6 years (range 15-77) and by the end of the study, on average, they had received 3.8 treatment cycles (range l-10) of which an average of 2.3 were given during the study period. In an earlier study, we showed that the three different chemotherapy protocols have the same degree of emetogenic effect, therefore stratification for these was considered unnecessary [7]. The standard antiregimen included intramuscular emetic Int J Gynecol Obstet 42
administration of 12.5 mg prochlorperazine followed by 10 mg dexamethasone and 13 mg thiethylperazine in 500 ml preloading normal saline infusion. Apart from the trial drugs in the different groups, no further antiemetics were given within 4 h of the completion of cytostatic infusion. One hundred and twentyone consecutive patients were randomly assigned to receive the standard plus no adjuvant antiemetics (controls, n = 24), 1 mg/kg single dose intravenous metoclopramide (HDM, n = 33), 2 mg/kg single dose intravenous metoclopramide (VHDM, n = 50) and 2 mg/kg single dose intravenous metoclopramide plus 1 mg droperidol intramuscularly (VHDM + DP, n = 14) prior to the chemotherapy. The randomization was performed by the opaque envelope method and because of the relatively large number of treatments, no measures were taken to keep the size of the groups close to equal informed consent was obtained from the patients before each treatment. The number of vomiting episodes and sideeffects were recorded by the patients. Due to individual differences between further antiemetic requirements, the number of vomiting episodes during only the first four posttreatment hours were considered for analysis. Previous exposure to antiemetics during earlier treatments of the same chemotherapy course was quantitated by processing the means of an explanatory variable which had a value of 1, 2, 3 or 4 for each previous treatment when ‘standard’, HDM, VHDM or VHDM + DP regimens were administered, respectively. This method avoided the possible confounding effect of the number of previous cycles which, as it will be shown later, had a much stronger impact on the antiemetic efficacy. Spearman’s rank correlation analysis was performed to identify the most important factors that modify emetic response. The x2-test was used to assess similarity of the four antiemetic groups with regard to distribution of the most important confounding factors (i.e. number of previous cycles, stage of disease,
Aniiemetic trial for cisplatin-induced vomiting
Table 1.
Prognostic
factors
Antiemetic regime ra P
-0.23
+?Spearman’s rank correlation
and their correlation
with cisplatin-induced Previous antiemetic
Age
Stage
0.4433
-0.14
-0.009
-0.159
coefficient.
Results The type of antiemetic regimen, number of cycles, stage of previous chemotherapy of antiemetic disease, average strength regimens during the previous chemotherapy cycles and age were shown to have significant correlation with the number of vomiting episodes within 4 h of cisplatin based chemotherapy treatment (Table 1). In order to assess the similarity of the four antiemetic groups with regard to the identified influencing factors, x2-tests based on the number of vomiting episodes were performed for relating the number of previous chemotherapy cycles, stage of disease and the strengths of antiemetic regimens during previous chemotherapy cycles to the different antiemetic regimens (Table 2). Having made the assumption that the four antiemetic groups are similar with regard to
Table 2. Comparison of distribution of prognostic between the four different antiemetic groups.
b2 P
27 31.6 >0.2
vomiting.
No. cycles
previous exposure to antiemetics). The distribution of the number of vomiting episodes among different antiemetic groups was assessed by one-way analysis of variance and the P value of between group differences were generated by the F-test.
No. of cycles
21
Stage
27 26.3 >0.2
Previous emetic 21 54.2 0.001
factors
anti-
the two most important factors which influence emetic response, one-way analysis of variance was performed to compare the frequencies of vomiting in each group (Table 3). The overall mean of vomiting episodes was 6.4, and the means within the control, HDM, VHDM and VHDM-DP groups were 9, 6.4, 6.2 and 2.7, respectively. The improvement in the control of vomiting across the four groups was highly significant. Only one patient developed mild diarrhea following VHDM therapy and no dyskinetic side-effects were observed. No attempt was made to quantify the patients’ attitude to sedation. However, despite frequent accounts of sedation causing discomfort, prevention or reduction of nausea and vomiting was always appreciated. Discussion A combination of high-dose steroid and droperidol either with or without chlorpromazine has already been shown to provide effective control of cisplatin-induced vomiting [ 1,111.The key for determination of efficacy is the selection of a standardized patient population and the emetic stimuli. Besides the actual antiemetic regime there are several other factors which influence the patients’ response to antiemetic drugs. In the present study, the following patient characteristics were essential to take into consideration: type of antiemetic regime, number of previous chemotherapy cycles, stage of disease, average strength of antiemetic regime during the previous chemotherapy cycles and age. All these were shown to have a significant correlation Article
22
P6ka et al.
Table 3.
ANOVA of vomiting episodes in four antiemetic groups.
Source of variation
df
Sums of squares
Mean squares
F
Between groups Within groups Total
3 117 120
43 651.5 4187.4 47 838.9
14 550.5 35.8
406.5
P < 0.001
with the number of vomiting episodes within 4 h of cisplatin-based chemotherapy treatment (Table 1). With respect to these factors, and to eliminate bias by anticipatory emesis, it has been suggested for clinical trials to include only those patients who had not received chemotherapy before [S]. In the present study a statistical weighing method was used to avoid this problem and to confirm statistically significant similarity of the distributions of cycle numbers in the compared groups. In order to assess the similarity of the four antiemetic groups with regard to the identified confounding factors, x*-tests, based on the number of vomiting episodes, were performed for relating the number of previous chemotherapy cycles, the stage of disease and the strength of antiemetic regimens during previous chemotherapy cycles to the different antiemetic regimens (Table 2). Considering several variables (number of vomiting episodes, duration, volume, nausea, sedation, etc.) that can be measured during an antiemetic trial, counting vomiting was found to be the most objective way of assessment [13]. A comparison between patients’ and observers’ assessment of antiemetic efficacy by recording the number of vomiting episodes, severity of vomiting, duration of nausea, severity of nausea and duration of vomiting showed that patients’ assessment on antiemetic studies are reliable [ 151. In the present study, 121 consecutive cisplatin-based chemotherapy treatments for ovarian carcinoma were randomly assigned to receive standard plus no adjuvant, high-dose metoclopramide, very high dose metoclopramide Int J Gynecol Obstet 42
and very high dose metoclopramide plus droperidol antiemetic treatment. The unequal number of patients in the treatment groups (control = 24, HDM = 33, VHDM = 50 and VHDM + DP = 14) could be explained by the opaque envelope method used. The randomization was not designed in a way which would ensure equal group numbers and the differences were due to chance. The number of vomiting episodes up to 4 h after treatment were recorded by the patients and these data were compared in the four different antiemetic groups. The effect of the number of previous chemotherapy cycles, the strength of previous antiemetic treatments, patients’ age and stage of disease on the number of vomiting episodes were also assessed. The mean number of vomiting episodes during the first four hours after the chemotherapy was 9, 6.4,6.2 and 2.7 among patients receiving the standard, HDM, VHDM and VHDM + DP antiemetic regimens, respectively. Analysis of variance of vomiting episodes confirmed significant differences between the groups (P < 0.001, F = 406.5, df = 120). The number of previous chemotherapy cycles and, to a lesser extent, the stage of disease, the strength of previous antiemetic regimens and the age showed significant correlation with the number of vomiting episodes rS = 0.443, -0.159, -0.14 and -0.009, respectively). These data suggest that substantial improvement in the control of cisplatin induced vomiting can be achieved by adding high-dose metoclopramide and droperidol to the basic antiemetic regime. In previous antiemetic studies, prochlorperazine and metoclopramide were shown to
Antiemetic
be equally effective in improving an antiemetic regime consisting dexamethasone and lorazepam, and no significant difference was found between the antiemetic efficacy of highdose metoclopramide and a combination of betamethasone plus dixyrazine [2,17]. In 36% of patients undergoing the first medium-dose cisplatin therapy, nausea was prevented, but in only 9% was vomiting absent. The ground for using anxiolytic drugs in the treatment of chemotherapy-induced nausea is that anxiety has a major impact on the patient’s susceptibility to nausea and vomiting [8]. The importance of that prospective study lies in the relation it shows between pretreatment level of anxiety and gastrointestinal side-effects of an emetogenic chemotherapy. One-way analysis of variance was performed to compare frequencies of vomiting in the different treatment groups (Table 3). The overall mean of vomiting episodes was 6.4, and the means within the control, HDM, VHDM and VHDM-DP groups were 9, 6.4, 6.2 and 2.7, respectively. Substantial improvement of antiemetic control was achieved by adding single high-dose metoclopramide and droperidol to the ‘standard’ regimen. Only one patient developed mild diarrhea following VHDM therapy and no dyskinetic side-effects were observed. Previous clinical trials have shown that dyskinetic side-effects are expected to develop at higher doses of metoclopramide [4,6,12]. No attempt was made to quantify the patients’ attitude to sedation. However, despite frequent account of sedation caused discomfort, prevention or reduction of nausea and vomiting was always better appreciated. Recent development has resulted in a shift of antiemetic research from dopamin antagonists to selective serotonin antagonists [lo]. However, for financial reasons, combinations of traditional antiemetics can still be expected to remain in the first line for years. A more cautious review of antiemetic trials including those with serotonin antagonists suggest that these new drugs show an efficacy at least equivalent to metoclopramide, therefore metoclopramide-based combinations should be studied further [ 121.
trial for cisplatin-induced
vomiting
23
References 1
2
3
4
5 6
7
8
9
10
11
12
13
14
B&e P, Gero Gy, Cseh I, Gati I: Management of CAP (cyclophosphamid, adriamycin, cisplatin) chemotherapy induced emesis. Magyar Noorv L 48: 193, 1985. Carr BI, Doroshow JH, Morgan RJ, Leong LA, Margolin KA, Akman SA, Hill LR: Combination antiemetic therapy based on high doses of either prochlorperazine or metoclopramide give complete protection against platinum-induced emesis (Abstract). Proc Ann Meet Am Sot Clin Oncol 8: 325, 1989. Clark RA, Gralla RJ, Kris MG, Tyson LB: Exploring very high doses of metoclopramide (4-6 mg/kg): preservation of efficacy and safety with only a single dose in combination antiemetic regimen (Abstract). Proc Annu Meet Am Sot Clin Oncol 8: 330, 1989. Craig JB, Powell BL: Review: The management of nausea and vomiting in clinical oncology. Am J Med Sci 293: 34, 1987. Gralla R: Approaches to management of nausea and vomiting in the clinical setting. Clinician 6: 26, 1988. Gralla RJ, Itri LM, Pisko SE et al.: Antiemetic efficacy of high-dose metoclopramide: randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting. N Engl J Med 305: 905, 1981. Hernadi Z, Juhasz B, Poka R, LampC L: Randomised trial comparing combinations of cyclophosphamide and cisplatin with or without doxorubicin or 4’-epi-doxorubicin in the treatment of advanced ovarian cancer. Int J Gynecol Obstet 27: 199, 1988. Jacobsen PS, Andrykowski MA, Redd WH et al.: Nonpharmacologic factors in the development of posttreatment nausea with adjuvant chemotherapy for breast cancer. Cancer 61: 379, 1987. Kris MG, Gralla RJ, Clark RA et al.: Consecutive dosefinding trials adding lorazepam to the combination of metoclopramide plus dexamethasone: improved subjective effectiveness over the combination of diphenhydramine plus metoclopramide plus dexamethasone. Cancer Treat Rep 69: 1257, 1988. Marty M: Ondansetron in the prophylaxis of acute cisplatin-induced nausea and vomiting. Eur J Cancer Clin Oncol 25: 541, 1989. Mason BA, Dambra J, Grossman B, Catalan0 RB: Effective control of cisplatin-induced nausea using high-dose steroid and droperidol. Cancer Treat Rep 66: 243, 1982. Merrifield KR, Chaffee BJ: Recent advances in the management of nausea and vomiting caused by antineoplastic agents. Clin Phann 8: 187, 1989. Olver IN, Simon RM, Aisner J: Antiemetic studies: a methodological discussion. Cancer Treat Rep 70: 555, 1986. Onsrud M, Moxnes A, Sollien A, Grande T, Solesvik 0: High dose versus low dose metoclopramide in the prevention of cisplatin-induced emesis. A randomised crossover study in patients with ovarian carcinoma. Cancer 61: 2429, 1988.
Article
24
Pdka et al.
15 Smith RA, Matthews JP, Bishop JF, Olver IN: The role of patient assessment in antiemetic trials (Abstract). Proc Ann Meet Am Sot Clin Oncol 8: 325, 1989. 16 Smyth JF: The problem of emesis induced by cancer chemotherapy. Clinician 6: 2, 1988. 17 Sorbe B, Hall& C: Antiemetic treatment of chemotherapy-induced nausea in ovarian carcinoma patients. Gynecol Oncol 34: 141, 1989.
Int J Gynecol Obstet 42
Address for reprints:
R. Polka Southend Health Care Trust Rochford Hospital Dalys Road, Rochford Essex SS4 65RZ, UK
19 and Obstetrics
Comparison of four antiemetic regimens for the treatment cisplatin-induced vomiting R. Poka, Z. Hernadi,
of
B. Juhasz and L. Lamp6
Department of Obstetrics and Gynecology, University Medical School of Debrecen (Hungary) (Received November l&h, 1992) (Revised and accepted February 12th, 1993)
Abstract OBJECTIVE: To improve the treatment of cisplatin-induced acute emesis and vomiting, the aa’juvant effect of two different doses of metoclopramide and metoclopramide plus droperidol was compared to a defined standard antiemetic regiment. METHOD: One-hundred and twenty-one consecutive cisplatin-based chemotherapy treatments for ovarian carcinoma were randomly assigned to receive standard plus no aq’juvant (control, n = 24), high-dose metoclopramide (HDM, n = 33), very high dose metoclopramide (VHDM, n = 50) and very high dose metoclopramide plus droperidol ( VHDM + DP, n = 14) antiemetic treatment. The number of vomiting episodes up to 4 h after the treatment were recorded and compared in the four different antiemetic groups. The effect of the number of previous chemotherapy cycles, the strength of previous antiemetic treatments, age and stage of disease on the number of vomiting episodes was also assessed. RESULT.. The mean number of vomiting episodes during the first 4 h after the chemotherapy was 9, 6.4, 6.2 and 2.7 among patients receiving the standard HDM, antiemetic VHDM + DP VHDM and regimens, respectively. Analysis of variance of vomiting episodes confirmed significant differ(P < 0.001, ences between the groups F = 406.5, df = 120). The number of previous chemotherapy cycles and, to a lesser extent, the 0020-7293/93/$06.00 @ 1993 International Federation Printed and Published in Ireland
stage of disease, the strength of previous antiemetic regimens and the age showed significant correlation with the number of vomiting episodes (r = 0.443, -0.159, -0.14 and -0.009 respectively). CONCLUSION: Substantial improvement in the control of cisplating-induced vomiting can be achieved by high-dose metoclopramide and adding droperidol to a basic antiemetic regimen consisting prochlorperazine, dexamethasone and thiethylperazine.
Keywords: Ovarian cancer; Metoclopramide; Droperidol.
Cisplatin;
Introduction Cisplatin is one of the most emetogenic chemotherapeutic agents. Efficient control of cisplatin-induced nausea and vomiting is a major issue in the supportive care of ovarian carcinoma patients. Metoclopramide was shown to be effective for the control of chemotherapy-induced nausea and vomiting [6]. Positive dose-response relationship for the antiemetic effect of metoclopramide was established in a randomized crossover study [14]. Despite the established superiority of high-dose metoclopramide during the 1980s antiemetic trials often showed substantial improvement in its efficacy by adding other agents such as dexamethasone, lorazepam and Article
of Gynecology
and Obstetrics
20
Pdka et al.
chlorperazine [3,4,9,16]. A single dose combined antiemetic regimen, consisting of very high dose metoclopramide, lorazepam and dexamethasone, appeared to be equally effective as repeated low-dose regimens. A further advantage of the single dose administration is the lower cost due to less drug use and a shorter hospital stay [3]. A combination of high-dose steroid and droperidol with or without chlorpromazine was also shown to provide effective control of cisplatin-induced vomiting [ 1,111. Recent developments in pharmacological and physiological research determined the neurochemical mechanisms of nausea and vomiting. Accordingly, serotonin antagonist drugs such as ondansetron have been developed and proved superior to all other antiemetics both in terms of efficacy and side-effects [lo]. Unfortunately, for financial reasons, traditional combinations based on high-dose metoclopramide are still expected to remain in the first line for years. We therefore studied the effect of two different doses of metoclopramide and metoclopramide plus droperidol on the antiemetic efficacy of a defined standard regimen for the control of cisplatin-induced vomiting. Materials and methods During the study period (between 1st January and 31st June, 1990) 52 patients received a total of 121 chemotherapy treatments for ovarian carcinoma using cyclophosphamide/cisplatin (1000:50 or 750:75 mg/m2) or cyclophosphamide/doxorubicin/ cisplatin (500:50:50 mg/m2) combinations on a day-care basis. The mean age of the patients was 50.6 years (range 15-77) and by the end of the study, on average, they had received 3.8 treatment cycles (range l-10) of which an average of 2.3 were given during the study period. In an earlier study, we showed that the three different chemotherapy protocols have the same degree of emetogenic effect, therefore stratification for these was considered unnecessary [7]. The standard antiregimen included intramuscular emetic Int J Gynecol Obstet 42
administration of 12.5 mg prochlorperazine followed by 10 mg dexamethasone and 13 mg thiethylperazine in 500 ml preloading normal saline infusion. Apart from the trial drugs in the different groups, no further antiemetics were given within 4 h of the completion of cytostatic infusion. One hundred and twentyone consecutive patients were randomly assigned to receive the standard plus no adjuvant antiemetics (controls, n = 24), 1 mg/kg single dose intravenous metoclopramide (HDM, n = 33), 2 mg/kg single dose intravenous metoclopramide (VHDM, n = 50) and 2 mg/kg single dose intravenous metoclopramide plus 1 mg droperidol intramuscularly (VHDM + DP, n = 14) prior to the chemotherapy. The randomization was performed by the opaque envelope method and because of the relatively large number of treatments, no measures were taken to keep the size of the groups close to equal informed consent was obtained from the patients before each treatment. The number of vomiting episodes and sideeffects were recorded by the patients. Due to individual differences between further antiemetic requirements, the number of vomiting episodes during only the first four posttreatment hours were considered for analysis. Previous exposure to antiemetics during earlier treatments of the same chemotherapy course was quantitated by processing the means of an explanatory variable which had a value of 1, 2, 3 or 4 for each previous treatment when ‘standard’, HDM, VHDM or VHDM + DP regimens were administered, respectively. This method avoided the possible confounding effect of the number of previous cycles which, as it will be shown later, had a much stronger impact on the antiemetic efficacy. Spearman’s rank correlation analysis was performed to identify the most important factors that modify emetic response. The x2-test was used to assess similarity of the four antiemetic groups with regard to distribution of the most important confounding factors (i.e. number of previous cycles, stage of disease,
Aniiemetic trial for cisplatin-induced vomiting
Table 1.
Prognostic
factors
Antiemetic regime ra P
-0.23
+?Spearman’s rank correlation
and their correlation
with cisplatin-induced Previous antiemetic
Age
Stage
0.4433
-0.14
-0.009
-0.159
coefficient.
Results The type of antiemetic regimen, number of cycles, stage of previous chemotherapy of antiemetic disease, average strength regimens during the previous chemotherapy cycles and age were shown to have significant correlation with the number of vomiting episodes within 4 h of cisplatin based chemotherapy treatment (Table 1). In order to assess the similarity of the four antiemetic groups with regard to the identified influencing factors, x2-tests based on the number of vomiting episodes were performed for relating the number of previous chemotherapy cycles, stage of disease and the strengths of antiemetic regimens during previous chemotherapy cycles to the different antiemetic regimens (Table 2). Having made the assumption that the four antiemetic groups are similar with regard to
Table 2. Comparison of distribution of prognostic between the four different antiemetic groups.
b2 P
27 31.6 >0.2
vomiting.
No. cycles
previous exposure to antiemetics). The distribution of the number of vomiting episodes among different antiemetic groups was assessed by one-way analysis of variance and the P value of between group differences were generated by the F-test.
No. of cycles
21
Stage
27 26.3 >0.2
Previous emetic 21 54.2 0.001
factors
anti-
the two most important factors which influence emetic response, one-way analysis of variance was performed to compare the frequencies of vomiting in each group (Table 3). The overall mean of vomiting episodes was 6.4, and the means within the control, HDM, VHDM and VHDM-DP groups were 9, 6.4, 6.2 and 2.7, respectively. The improvement in the control of vomiting across the four groups was highly significant. Only one patient developed mild diarrhea following VHDM therapy and no dyskinetic side-effects were observed. No attempt was made to quantify the patients’ attitude to sedation. However, despite frequent accounts of sedation causing discomfort, prevention or reduction of nausea and vomiting was always appreciated. Discussion A combination of high-dose steroid and droperidol either with or without chlorpromazine has already been shown to provide effective control of cisplatin-induced vomiting [ 1,111.The key for determination of efficacy is the selection of a standardized patient population and the emetic stimuli. Besides the actual antiemetic regime there are several other factors which influence the patients’ response to antiemetic drugs. In the present study, the following patient characteristics were essential to take into consideration: type of antiemetic regime, number of previous chemotherapy cycles, stage of disease, average strength of antiemetic regime during the previous chemotherapy cycles and age. All these were shown to have a significant correlation Article
22
P6ka et al.
Table 3.
ANOVA of vomiting episodes in four antiemetic groups.
Source of variation
df
Sums of squares
Mean squares
F
Between groups Within groups Total
3 117 120
43 651.5 4187.4 47 838.9
14 550.5 35.8
406.5
P < 0.001
with the number of vomiting episodes within 4 h of cisplatin-based chemotherapy treatment (Table 1). With respect to these factors, and to eliminate bias by anticipatory emesis, it has been suggested for clinical trials to include only those patients who had not received chemotherapy before [S]. In the present study a statistical weighing method was used to avoid this problem and to confirm statistically significant similarity of the distributions of cycle numbers in the compared groups. In order to assess the similarity of the four antiemetic groups with regard to the identified confounding factors, x*-tests, based on the number of vomiting episodes, were performed for relating the number of previous chemotherapy cycles, the stage of disease and the strength of antiemetic regimens during previous chemotherapy cycles to the different antiemetic regimens (Table 2). Considering several variables (number of vomiting episodes, duration, volume, nausea, sedation, etc.) that can be measured during an antiemetic trial, counting vomiting was found to be the most objective way of assessment [13]. A comparison between patients’ and observers’ assessment of antiemetic efficacy by recording the number of vomiting episodes, severity of vomiting, duration of nausea, severity of nausea and duration of vomiting showed that patients’ assessment on antiemetic studies are reliable [ 151. In the present study, 121 consecutive cisplatin-based chemotherapy treatments for ovarian carcinoma were randomly assigned to receive standard plus no adjuvant, high-dose metoclopramide, very high dose metoclopramide Int J Gynecol Obstet 42
and very high dose metoclopramide plus droperidol antiemetic treatment. The unequal number of patients in the treatment groups (control = 24, HDM = 33, VHDM = 50 and VHDM + DP = 14) could be explained by the opaque envelope method used. The randomization was not designed in a way which would ensure equal group numbers and the differences were due to chance. The number of vomiting episodes up to 4 h after treatment were recorded by the patients and these data were compared in the four different antiemetic groups. The effect of the number of previous chemotherapy cycles, the strength of previous antiemetic treatments, patients’ age and stage of disease on the number of vomiting episodes were also assessed. The mean number of vomiting episodes during the first four hours after the chemotherapy was 9, 6.4,6.2 and 2.7 among patients receiving the standard, HDM, VHDM and VHDM + DP antiemetic regimens, respectively. Analysis of variance of vomiting episodes confirmed significant differences between the groups (P < 0.001, F = 406.5, df = 120). The number of previous chemotherapy cycles and, to a lesser extent, the stage of disease, the strength of previous antiemetic regimens and the age showed significant correlation with the number of vomiting episodes rS = 0.443, -0.159, -0.14 and -0.009, respectively). These data suggest that substantial improvement in the control of cisplatin induced vomiting can be achieved by adding high-dose metoclopramide and droperidol to the basic antiemetic regime. In previous antiemetic studies, prochlorperazine and metoclopramide were shown to
Antiemetic
be equally effective in improving an antiemetic regime consisting dexamethasone and lorazepam, and no significant difference was found between the antiemetic efficacy of highdose metoclopramide and a combination of betamethasone plus dixyrazine [2,17]. In 36% of patients undergoing the first medium-dose cisplatin therapy, nausea was prevented, but in only 9% was vomiting absent. The ground for using anxiolytic drugs in the treatment of chemotherapy-induced nausea is that anxiety has a major impact on the patient’s susceptibility to nausea and vomiting [8]. The importance of that prospective study lies in the relation it shows between pretreatment level of anxiety and gastrointestinal side-effects of an emetogenic chemotherapy. One-way analysis of variance was performed to compare frequencies of vomiting in the different treatment groups (Table 3). The overall mean of vomiting episodes was 6.4, and the means within the control, HDM, VHDM and VHDM-DP groups were 9, 6.4, 6.2 and 2.7, respectively. Substantial improvement of antiemetic control was achieved by adding single high-dose metoclopramide and droperidol to the ‘standard’ regimen. Only one patient developed mild diarrhea following VHDM therapy and no dyskinetic side-effects were observed. Previous clinical trials have shown that dyskinetic side-effects are expected to develop at higher doses of metoclopramide [4,6,12]. No attempt was made to quantify the patients’ attitude to sedation. However, despite frequent account of sedation caused discomfort, prevention or reduction of nausea and vomiting was always better appreciated. Recent development has resulted in a shift of antiemetic research from dopamin antagonists to selective serotonin antagonists [lo]. However, for financial reasons, combinations of traditional antiemetics can still be expected to remain in the first line for years. A more cautious review of antiemetic trials including those with serotonin antagonists suggest that these new drugs show an efficacy at least equivalent to metoclopramide, therefore metoclopramide-based combinations should be studied further [ 121.
trial for cisplatin-induced
vomiting
23
References 1
2
3
4
5 6
7
8
9
10
11
12
13
14
B&e P, Gero Gy, Cseh I, Gati I: Management of CAP (cyclophosphamid, adriamycin, cisplatin) chemotherapy induced emesis. Magyar Noorv L 48: 193, 1985. Carr BI, Doroshow JH, Morgan RJ, Leong LA, Margolin KA, Akman SA, Hill LR: Combination antiemetic therapy based on high doses of either prochlorperazine or metoclopramide give complete protection against platinum-induced emesis (Abstract). Proc Ann Meet Am Sot Clin Oncol 8: 325, 1989. Clark RA, Gralla RJ, Kris MG, Tyson LB: Exploring very high doses of metoclopramide (4-6 mg/kg): preservation of efficacy and safety with only a single dose in combination antiemetic regimen (Abstract). Proc Annu Meet Am Sot Clin Oncol 8: 330, 1989. Craig JB, Powell BL: Review: The management of nausea and vomiting in clinical oncology. Am J Med Sci 293: 34, 1987. Gralla R: Approaches to management of nausea and vomiting in the clinical setting. Clinician 6: 26, 1988. Gralla RJ, Itri LM, Pisko SE et al.: Antiemetic efficacy of high-dose metoclopramide: randomized trials with placebo and prochlorperazine in patients with chemotherapy-induced nausea and vomiting. N Engl J Med 305: 905, 1981. Hernadi Z, Juhasz B, Poka R, LampC L: Randomised trial comparing combinations of cyclophosphamide and cisplatin with or without doxorubicin or 4’-epi-doxorubicin in the treatment of advanced ovarian cancer. Int J Gynecol Obstet 27: 199, 1988. Jacobsen PS, Andrykowski MA, Redd WH et al.: Nonpharmacologic factors in the development of posttreatment nausea with adjuvant chemotherapy for breast cancer. Cancer 61: 379, 1987. Kris MG, Gralla RJ, Clark RA et al.: Consecutive dosefinding trials adding lorazepam to the combination of metoclopramide plus dexamethasone: improved subjective effectiveness over the combination of diphenhydramine plus metoclopramide plus dexamethasone. Cancer Treat Rep 69: 1257, 1988. Marty M: Ondansetron in the prophylaxis of acute cisplatin-induced nausea and vomiting. Eur J Cancer Clin Oncol 25: 541, 1989. Mason BA, Dambra J, Grossman B, Catalan0 RB: Effective control of cisplatin-induced nausea using high-dose steroid and droperidol. Cancer Treat Rep 66: 243, 1982. Merrifield KR, Chaffee BJ: Recent advances in the management of nausea and vomiting caused by antineoplastic agents. Clin Phann 8: 187, 1989. Olver IN, Simon RM, Aisner J: Antiemetic studies: a methodological discussion. Cancer Treat Rep 70: 555, 1986. Onsrud M, Moxnes A, Sollien A, Grande T, Solesvik 0: High dose versus low dose metoclopramide in the prevention of cisplatin-induced emesis. A randomised crossover study in patients with ovarian carcinoma. Cancer 61: 2429, 1988.
Article
24
Pdka et al.
15 Smith RA, Matthews JP, Bishop JF, Olver IN: The role of patient assessment in antiemetic trials (Abstract). Proc Ann Meet Am Sot Clin Oncol 8: 325, 1989. 16 Smyth JF: The problem of emesis induced by cancer chemotherapy. Clinician 6: 2, 1988. 17 Sorbe B, Hall& C: Antiemetic treatment of chemotherapy-induced nausea in ovarian carcinoma patients. Gynecol Oncol 34: 141, 1989.
Int J Gynecol Obstet 42
Address for reprints:
R. Polka Southend Health Care Trust Rochford Hospital Dalys Road, Rochford Essex SS4 65RZ, UK