Ex) Article Title, Author, Keywords
Ex) Article Title, Author, Keywords
R Clin Pharm 2023; 1(2): 115-126
Published online December 31, 2023 https://doi.org/10.59931/rcp.23.030
Copyright © Asian Conference On Clinical Pharmacy.
Hanh T.H. Nguyen1 , Linh K. Duong1 , Tien D Vu2 , Thu T.M. Hoang2 , Duong V Bach3 , Hao T.L Hoang3 , Huong T.L. Nguyen1
Correspondence to:Huong T.L. Nguyen
E-mail Huongntl@hup.edu.vn
ORCID
https://orcid.org/0000-0002-8051-5017
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: Chemotherapy-induced nausea and vomiting (CINV) is considered as the most distressing side effect of chemotherapy (CT). Overuse of prophylaxis leads to waste of resources, whereas underuse results in poor CINV prevention. Therefore, it is important to explore the pattern of CINV prophylaxis and CINV occurrence to improve the quality of medication use.
Methods: This prospective, observational study was conducted in two oncology-specialized hospitals in North Vietnam in 2019 to investigate the CINV prophylaxis regimens and CINV events among cancer patients receiving intravenous high-, moderate-, or lowemetic-risk CT (HEC, MEC, or LEC, respectively) after the administration of the first cycle of CT. The Multinational Association of Supportive Care in Cancer Antiemesis Tool was used to assess nausea and vomiting.
Results: The study included 294 patients (HEC: 224, MEC: 61, and LEC: 9). They were prescribed a similar pattern of prophylaxis for both acute and delayed emesis regardless of their emetic risks. Patient-specific factors, including sex, age, balance disorder, and comorbidity, but not the emetic risk of CT, were associated with the choice of prophylactic regimens. The incidence rate of CINV was 47.3% in all patients. Multivariate analysis revealed that HEC was associated with an increased risk of acute CINV compared with MEC (odds ratio, 2.34; 95% confidence interval, 1.03–5.32), whereas female sex and younger age were associated with a higher risk of delayed CINV.
Conclusion: CINV management is challenging and may involve over- and underuse of prophylaxis. A more effective approach to prophylaxis in the context of limited resources needs to be developed to improve patient outcomes.
KeywordsChemotherapy-induced nausea and vomiting; Prophylaxis; Patient care
Chemotherapy-induced nausea and vomiting (CINV) which are considered the most distressing side effects related to chemotherapy [1], would lead to deteriorate patients’ clinical outcomes (e.g., dehydration, electrolyte imbalances, weight loss, and malnourishment), and negatively impact their quality of life and daily activities [2,3]. These events would also result in additional supportive care, outpatient visits, or hospitalization, which might incur an economic burden [4]. Moreover, inadequate control of CINV could lead to patients’ delay or refusal of possible life-saving chemotherapy (CT) [5]. Although antiemetic medications have reduced the incidence of vomiting and nausea substantially, approximately 30% to 60% of patients still experienced acute or delayed nausea after chemotherapy [2]. This demonstrates a great challenge for healthcare professionals to prevent and treat CINV with available antiemetic agents in clinical practice settings, especially in limited-resource countries. Therefore, proper antiemetic use is claimed as one of the five key opportunities to improve care and reduce costs in cancer by the American Society of Clinical Oncology (ASCO) [6].
To foster the appropriate use of antiemetic medications in cancer patients treated with CT, the evidence-based guidelines for CINV prevention and treatment have been developed and updated by various cancer societies, e.g., ASCO, National Comprehensive Cancer Network (NCCN), Multinational Association of Supportive Care in Cancer and the European Society of Medical Oncology (MASCC and ESMO) [7-9]. As per their consensus, the prophylaxis of CINV with medications such as serotonin receptor antagonists (5-HT3RA), dexamethasone, neurokinin-1 receptor antagonists (NK1RA), and olanzapine, should be the primary goal and be implemented specifically for those patients treated with CT classified as highly emetogenic chemotherapies (HEC) [7-10]. Adherence to the guidelines demonstrated a strong correlation with the improvement of CINV control in PEER or INSPIRE studies with significant reductions in CINV incidences [11,12].
However, the use of antiemetic drugs for CINV was still an issue. In high-income countries, where mostly antiemetic medications are available, overuse was reported in 24.1% of patients of various CINV risk groups [13], and in up to 70% of patients receiving low-emetogenic chemotherapies (LEC) [6]. No prophylaxis or a lack of prophylactic duration were also reported as the main reasons for non-adherence to the clinical guidelines in these countries [11,12]. In resource-limited countries, it’s more challenging to follow the antiemetic guidelines, as it is presumed that “socio-economic status” is one of the rationales for non-adherence [14]. A study in Jordan showed that there were no NK1RA in the CINV prophylaxis therapy, and the prevalence of CINV was as high as more than 70% in the included patients [15]. In a study based on medication data in China, there was also a lack of NK1RA and an overutilization of the other antiemetic agents [16]. Therefore, it’s essential to reveal prophylactic problems to improve the practice of CINV prophylaxis and patient outcomes. In Vietnam, a resource-limited country, antiemetic medications were not available in full range, and information about CINV prophylaxis and its effectiveness was limited. Therefore, we conducted this study to investigate the patterns of CINV prophylaxis in Vietnam and patient-reported CINV events to explore the potential gap in practice. With that, we expected to propose measurements to use the limited healthcare resources optimally and to improve patient outcomes in the future.
This prospective, observational study was conducted at two oncology-specialized hospitals in North Vietnam from March 2019 to June 2019. Patients were eligible if they had a confirmed cancer diagnosis, were aged ≥18 years, and were scheduled to take the first cycle of a single-day, intravenous chemotherapy. The emetic risk of chemotherapy was classified as HEC, moderate emetogenic chemotherapy (MEC), and LEC according to the NCCN guidelines for anti-emesis version 2019.1 [17]. Patients who concurrently took (an) oral cytotoxic agent(s), used to be treated with chemotherapy, experienced nausea or vomiting in 24 hours before the administration of the chemotherapy, had brain metastases or bowel obstruction, or whose medical records were not accessed, were excluded.
Data collection consisted of medical record extraction and patient interviews. A form was created to retrieve data from medical records, which included patient demographic and clinical characteristics, the prescribed chemotherapy and medications for CINV prophylaxis, and other co-medications. The demographic and clinical characteristics consisted of age, gender, body weight and height, type of cancer, comorbidities, and the status of concurrent radiotherapy. The chemotherapy and all prescribed medications were recorded for their names and active ingredients, route of administration, dose, and duration. A questionnaire about risk factors for CINV, including history of balance disorder, motion sickness, nausea and vomiting related to pregnancy (if applicable), anxiety about chemotherapy, and frequency of alcohol drinking, was used to interview patients. The Multinational Association of Supportive Care in Cancer Antiemesis Tool (MAT), a validated tool to measure CINV, was used to assess patients’ experiences of the occurrence and severity of nausea and vomiting [18]. Permission to use MAT was granted by the Multinational Association of Supportive Care in Cancer (MASCC).
There was a pre-established process for data collection. Every day during the study period, the clinical pharmacists screened a list of patients admitted to the hospitals for those with the first cycle of chemotherapy. Their medical records were reviewed to select eligible patients for the study and to collect the data on the form. The selected patients were interviewed to obtain their risk factors for CINV and were introduced to MAT before or during their administration of chemotherapy. The time of the initiation of the chemotherapy administration was noted to identify the acute and delayed phases of CINV. The patients were approached twice, 24 hours and 120 hours after the chemotherapy, for their responses to MAT in the acute and delayed phases, respectively. The process is summarized in Fig. 1.
The MAT used to evaluate patients’ experiences of nausea and vomiting consists of eight items. The first four are for nausea and vomiting for the acute phase, which is 24 hours following chemotherapy, and the last four items are for the delay phase, the period after that until 120 hours. In both phases, patients were asked whether they experienced nausea (items 3 and 7) and vomiting (items 1 and 5). They were asked about the severity of nausea (items 4 and 8) on a scale from 0 (no nausea) to 10 (as much as possible) and the number of vomiting episodes (items 2 and 6) [18]. In this study, the events of CINV were defined as severe if patients reported from four episodes of vomiting, or from level six of nausea in the questions about CINV severity in MAT.
As for the 5-HT3RA, the upper limit of an agent’s daily dose recommended for the acute phase by NCCN [17], which were similar to those by ASCO, MASCC and ESMO [7-10], and was considered one recommended dose (RD). One RD of the locally available 5-HT3RA was 0.25 mg of intravenous (IV) palonosetron, 24 mg of oral ondansetron, 16 mg of IV ondansetron, or 1 mg of IV granisetron. If the 5-HT3RA daily dose of a patient was higher than one RD, it would be classified as a high dose of 5-HT3RA.
As the NCCN guidelines recommended that in prophylactic regimens with a NK1RA, the daily doses of dexamethasone are 12 mg and 8 mg for the acute and delayed phases, respectively, and higher glucocorticoid (GC) doses should be used when a NK1RA agent is not given concomitantly [17]. Therefore, the daily dose of GC was considered to be high if it was higher than 12 mg of dexamethasone for the acute phase and more than 8 mg of dexamethasone for the delayed phase. The GC included in the chemotherapy regimen was also calculated for the CINV prophylaxis.
The medications used on the day of CT were recorded for the acute phase (day 1), and those on the following days were for the delayed phase (from day 2 to day 4). Patients were considered to be covered by 5-HT3RA in the delayed phase if they were prescribed IV palonosetron on day 1 due to its long half-life [12].
Patient demographic and clinical characteristics, risk factors were summarized using descriptive statistics. These factors, the pattern of prophylaxis, and the incidence of CINV were compared among groups of patients treated with HEC, MEC, and LEC using the Pearson Chi-square test or Fisher exact test for categorical variables and one way ANOVA for the continuous variables. The associations between demographics, clinical characteristics, risk factors, and the pattern of CINV prophylaxis were analyzed using χ2-test. The variable of emetic risks of CT but not cancer type was employed in the analysis because they are correlated, and the risk of CT was the primary criterion for CINV prophylactic regimen selection. The associations of the demographics and clinical characteristics, risk factors, the patterns of CINV prophylaxis with the occurrence of events in the acute and delayed phases were also examined using χ2-test. Then the factors that had the
The study included 294 patients in three groups: HEC, MEC, and LEC. The groups were significantly different in patients’ age, the frequent alcohol usage, the concomitant use of radiation therapy (RT), and the type of cancer (Table 1). The mean age was 54.4 years in the whole studied population, however, more patients under 60 years old were in the HEC group compared with MEC and LEC groups, 70.1% vs. 59.0% and 22.2% respectively (
Table 1 Patient demographics and clinical characteristics
Patient characteristics\N | Total (N=294) | HEC (N=224) | MEC (N=61) | LEC (N=9) | |
---|---|---|---|---|---|
Age (year), mean (SD) | 54.4 (11.1) | 53.4 (10.7) | 56.6 (11.4) | 64.2 (13.7) | 0.003 |
Age <60 years old, n (%) | 195 (66.3) | 157 (70.1) | 36 (59.0) | 2 (22.2) | 0.005 |
BMI, n (%) | 0.13 | ||||
<18.5 | 41 (13.9) | 29 (13.0) | 8 (13.1) | 4 (44.4) | |
18.5≤BMI<25 | 226 (76.9) | 174 (77.7) | 48 (78.7) | 4 (44.4) | |
≥25 | 27 (9.2) | 21 (9.4) | 5 (8.2) | 1 (11.1) | |
Female gender, n (%) | 171 (58.2) | 128 (57.1) | 39 (63.9) | 4 (44.4) | 0.44 |
Having comorbidity, n (%) | 100 (34.0) | 74 (33.0) | 23 (37.7) | 3 (33.3) | 0.79 |
Concurrent RT, n (%) | 86 (29.3) | 74 (33.0) | 12 (19.7) | 0 | 0.02 |
Nervousness before CT, n (%) | 15 (5.1) | 12 (5.4) | 3 (4.9) | 0 | 1.00 |
History of balance disorder, n (%) | 36 (12.2) | 31 (13.8) | 5 (8.2) | 0 | 0.26 |
History of carsick, n (%) | 103 (35.0) | 76 (33.9) | 26 (42.6) | 1 (11.1) | 0.14 |
History of nausea/vomit related to pregnancy, n (%) | 58 (19.7) | 42 (18.8) | 14 (23.0) | 2 (22.2) | 0.75 |
Frequent alcohol usage, n (%) | 64 (21.8) | 48 (21.4) | 11 (18.0) | 5 (55.6) | 0.04 |
Type of cancer, n (%) | <0.0001 | ||||
Breast cancer | 88 (30.0) | 61 (27.2) | 25 (41.0) | 2 (22.2) | |
Digestive tract cancer* | 75 (25.5) | 53 (23.7) | 22 (36.1) | 0 | |
Lung cancer | 53 (18.0) | 44 (19.6) | 8 (13.1) | 1 (11.1) | |
Gynecologic cancer | 45 (15.3) | 41 (18.3) | 2 (3.3) | 2 (22.2) | |
Non-Hodgkin lymphoma | 17 (5.8) | 13 (5.8) | 4 (6.5) | 0 | |
Others | 16 (5.4) | 12 (5.4) | 0 | 4 (44.5) |
*Including tongue, nasopharyngeal, esophagus, gastric, colorectal cancer.
BMI=body mass index, CT=chemotherapy, HEC=high emetic risk chemotherapy, LEC=low emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, RT=radiotherapy, SD=standard deviation.
In the acute phase, a combination therapy of GC and 5-HT3RA was prescribed for all patients, regardless of the potential emetogenicity of the chemotherapy they used (Table 2). The proportions of patients treated with high doses of GC and 5-HT3RA were not significantly different among the groups (
Table 2 Pattern of prophylaxis for chemotherapy induced-nausea and vomiting
Pattern of CINV prophylaxis\N | HEC (N=224) | MEC (N=61) | LEC (N=9) | |
---|---|---|---|---|
Acute phase - day 1 | ||||
Prophylaxis therapy | ||||
GC+5HT3 RA | 224 (100.0) | 61 (100.0) | 9 (100.0) | - |
High dose of GC | 99 (44.2) | 24 (39.3) | 3 (33.3) | 0.67 |
High dose of 5-HT3 RA | 151 (67.4) | 34 (55.7) | 5 (55.6) | 0.20 |
Delayed phase - day 2 | ||||
Prophylaxis therapy | 0.37 | |||
No use | 48 (21.4) | 14 (23.0) | 4 (44.4) | |
GC only | 23 (10.3) | 11 (18.0) | 1 (11.1) | |
5-HT3 RA only | 89 (39.7) | 19 (31.1) | 3 (33.3) | |
GC+5-HT3 RA | 64 (28.6) | 17 (27.9) | 1 (11.1) | |
High dose of GC | 11 (4.9) | 3 (4.9) | 0 (0) | 1.00 |
High dose of 5-HT3 RA | 29 (13.0) | 12 (19.7) | 1 (11.1) | 0.40 |
Delayed phase - day 3 | ||||
Prophylaxis therapy | 0.10 | |||
No use | 124 (55.4) | 31 (50.8) | 8 (88.9) | |
GC only | 13 (5.8) | 6 (9.8) | 0 (0) | |
5-HT3 RA only | 70 (31.2) | 15 (24.6) | 0 (0) | |
GC+5-HT3 RA | 17 (7.6) | 9 (14.7) | 1 (11.1) | |
High dose of GC | 10 (4.5) | 1 (1.6) | 0 (0) | 0.62 |
High dose of 5-HT3 RA | 8 (3.6) | 7 (11.5) | 0 (0) | 0.06 |
Delayed phase - day 4 | ||||
Prophylaxis therapy | 0.36 | |||
No use | 143 (63.8) | 45 (73.8) | 9 (100) | |
GC only | 9 (4.0) | 2 (3.3) | 0 (0) | |
5-HT3 RA only | 66 (29.5) | 13 (21.3) | 0 (0) | |
GC+5-HT3 RA | 6 (2.7) | 1 (1.6) | 0 (0) | |
High dose of GC | 8 (3.6) | 0 (0) | 0 (0) | 0.38 |
High dose of 5-HT3 RA | 3 (1.3) | 5 (8.2) | 0 (0) | 0.04 |
HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists.
In terms of prophylaxis for delayed nausea and vomiting, there was no statistically significant difference in the usage of anti-emesis medications for the patients in the groups from day 2 to day 4. The percentages of patients without prophylaxis increased from day 2 to day 4 (21.4%, 55.4% and 63.8% of the patients with HEC, respectively, and 23.0%, 50.8% and 73.8% of those with MEC, respectively). The proportion of the patients in HEC group treated with the combination of GC and 5-HT3RA was 28.6% on day 2 and decreased to 2.7% on day 4. Whereas 49.1% and 34.4% of the patients in MEC, were prophylactic with either GC or 5-HT3RA on day 2 and 3 respectively. The proportions of patients with high dose GC were under 5% across the groups of all the days.
The body mass index (BMI) equal or greater than 25 was the only factor related to the choice of high dose GC in the acute phase with OR=0.3 (0.10–0.87) compared to BMI less than 18.5 (Table 3). Fewer female than male patients were prescribed 5-HT3RA at the high dose in the acute phase (OR=0.42, 95% CI=0.02–0.87). However, the female patients were 2.47 times more likely to be treated enough or more days for the delayed prophylaxis than their male counterparts. They also tended to be treated with more drugs than men during the delayed phase, but were not significantly different. The patients having history of balance disorder were less likely to be treated enough or more days for the delayed prophylaxis (OR=0.4, 95% CI=0.17–0.91), while those with comorbidity were indicated two medications less frequent than the patients without comorbidity (OR=0.39, 95% CI=0.19–0.80).
Table 3 Multi-variable analysis for factors associated with the pattern of CINV prophylaxis
Factors | Acute phase | Delayed phase | |||
---|---|---|---|---|---|
High dose of GC* | High dose of 5-HT3 RA* | More or enough days of delayed prophylaxis* | Number of drugs selection# | ||
2 drugs (5-HT3 RA and GC) vs. no use | 1 drug (5-HT3 RA or GC) vs. no use | ||||
Emetic risk of CT | - | - | - | - | |
HEC | 0.64 (0.35-1.17) | ||||
MEC | 1 | ||||
LEC | NE | ||||
Female gender (yes vs. no) | - | 0.42 (0.20-0.87) | 2.47 (1.29-4.75) | 1.38 (0.64-2.98) | 1.92 (0.96-3.85) |
Age <60 years old (yes vs. no) | - | 0.57 (0.32-0.99) | - | - | - |
Concurrent RT (yes vs. no) | - | 1.69 (0.92-3.10) | 1.03 (0.57-1.85) | - | - |
Nervousness before CT (yes vs. no) | - | - | - | - | - |
History of balance disorder (yes vs. no) | - | - | 0.40 (0.17-0.91) | - | - |
History of carsick (yes vs. no) | - | 0.92 (0.51-1.64) | 1.25 (0.69-2.26) | 1.10 (0.47-2.59) | 1.35 (0.64-2.86) |
History of nausea/vomit related to pregnancy (yes vs. no) | - | - | 1.23 (0.63-2.40) | - | - |
Frequent alcohol usage (yes vs. no) | - | 1.24 (0.53-2.93) | - | - | - |
Having comorbidity (yes vs. no) | - | - | - | 0.39 (0.19-0.80) | 0.78 (0.42-1.43) |
BMI | - | - | - | - | |
<18.5 | 1 | ||||
18.5≤BMI<25 | 0.65 (0.33-1.27) | ||||
≥25 | 0.30 (0.10-0.87) |
Values are presented as odds ratio (95% confidence interval).
*Multi-variable logistic regression model including variables with
CINV=chemotherapy induced nausea and vomiting, OR=odds ratio, CI=confident interval, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists, CT=chemotherapy, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, NE=not estimated due to no event in the group, RT=radiotherapy, BMI=body mass index.
CINV happened in 20.8% and 46.4% of the patients in the acute and delayed phase respectively, and 47.3% experienced CINV in either phase (Table 4). The incidence of nausea was higher than of vomiting in both phases, and the delayed events happened more frequently than the acute ones in all categories. When comparing amongst groups of emetic risk, the higher risk groups had the higher incidences of the events with significant difference in some types of events such as: all CINV (
Table 4 CINV incidences and the comparison among emetogenic risk groups
CINV\N | Total (n=294) | HEC (n=224) | MEC (n=61) | LEC (n=9) | |
---|---|---|---|---|---|
All CINV | 139 (47.3) | 116 (51.8) | 22 (36.1) | 1 (11.1) | 0.01 |
All acute CINV | 61 (20.8) | 53 (23.7) | 8 (13.1) | 0 | 0.06 |
Acute vomiting | 29 (9.9) | 25 (11.2) | 4 (6.6) | 0 | 0.34 |
Acute nausea | 60 (20.4) | 52 (23.2) | 8 (13.1) | 0 | 0.07 |
Severe acute CINV | 31 (10.5) | 27 (12.1) | 4 (6.6) | 0 | 0.27 |
Severe acute vomiting | 18 (6.1) | 16 (7.1) | 2 (3.3) | 0 | 0.65 |
Severe acute nausea | 28 (9.5) | 25 (11.2) | 3 (4.9) | 0 | 0.21 |
All delayed CINV# | 131 (46.4) | 108 (50.0) | 22 (37.9) | 1 (12.5) | 0.04 |
Delayed vomiting | 72 (25.5) | 62 (28.7) | 10 (17.2) | 0 | 0.05 |
Delayed nausea | 125 (44.3) | 103 (47.7) | 21 (36.2) | 1 (12.5) | 0.06 |
Severe delayed CINV# | 69 (23.5) | 61 (28.2) | 8 (13.8) | 0 | 0.02 |
Severe delayed vomiting | 35 (12.4) | 30 (13.9) | 5 (8.6) | 0 | 0.31 |
Severe delayed nausea | 57 (20.2) | 50 (23.2) | 7 (12.1) | 0 | 0.06 |
#There were 282, 216, 61 and 8 patients in total, HEC, MEC and LEC group respectively because some patients were lost to contact in the delayed phase.
CINV=chemotherapy induced nausea and vomiting, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy.
The multiple logistic regression analysis showed that patients treated with HEC was the only factor related to the 2.34 times higher risk of developing acute CINV than with MEC (OR=2.34 95% CI=1.03–5.32). Whereas HEC also tended to have more delayed CINV but non-significantly after controlling other factors. Female gender was associated with higher risk of acute and delayed CINV but only statistically significant in two models of delayed CINV (Table 5). For the risk of delayed CINV, in the model 1, besides gender, the occurrence of acute CINV increased the risk of delayed events by 10.26 times (95% CI=4.45–23.63). In the model 2, which excluded the patients with acute CINV, younger age was an additional factor associated with the increased risk of the event by 2.08 times (95% CI=1.07–4.07) after controlling for the others.
Table 5 Multivariate logistic regression analysis for acute and delayed CINV
Factors\N | Acute CINV* (N=294) | Delayed CINV | ||
---|---|---|---|---|
Model 1*,# (N=282) | Model 2 | |||
Emetic risk of CT | ||||
HEC | 2.34 (1.03-5.32) | 1.47 (0.74-2.90) | 1.87 (0.88-3.96) | |
MEC | 1 | 1 | 1 | |
LEC | NE | 0.44 (0.05-4.27) | 0.45 (0.04-4.83) | |
Female gender (yes vs. no) | 1.95 (0.902-4.21) | 3.22 (1.41-7.33) | 3.38 (1.38-8.28) | |
Age <60 years old (yes vs. no) | - | 1.76 (0.96-3.22) | 2.08 (1.07-4.07) | |
Concurrent RT (yes vs. no) | - | - | - | |
Nervousness before CT (yes vs. no) | - | - | - | |
History of balance disorder (yes vs. no) | - | - | - | |
History of carsick (yes vs. no) | 1.57 (0.80-3.10) | 1.18 (0.61-2.29) | 1.19 (0.58-2.44) | |
History of nausea/vomiting related to pregnancy (yes vs. no) | 1.34 (0.65-2.76) | 1.36 (0.64-2.91) | 1.97 (0.86-4.54) | |
Frequent alcohol usage (yes vs. no) | - | 1.56 (0.65-3.73) | 1.32 (0.50-3.52) | |
Having comorbidity (yes vs. no) | - | - | - | |
BMI (18.5≤ BMI<25, ≥25 vs. <18) | - | - | - | |
High dose of acute GC (yes vs. no) | - | - | - | |
High dose of 5-HT3RAs in the acute phase (yes vs. no) | - | - | - | |
Days of delayed CINV prophylaxis | NT | - | - | |
Therapy selection | NT | - | - | |
No used | 1 | - | ||
1 drug | 1.74 (0.78-3.88) | 1.41 (0.60-3.33) | ||
2 drugs | 1.17 (0.56-2.44) | 0.85 (0.38-1.88) | ||
Occurrence of acute CINV | NT | 10.26 (4.45-23.63) | NT |
Values are presented as odds ratio (95% confidence interval).
*Multi-variable logistic regression model including variables with
CINV=chemotherapy induced nausea and vomiting, OR=odds ratio, CI=confident interval, CT=chemotherapy, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, NE=not estimated due to no event in the group. RT=radiotherapy, BMI=body mass index, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists, NT=not tested in univariate and multivariate analysis.
Although there were international and local emetic prophylaxis guidelines for patients receiving chemotherapy, we could not evaluate the adherence to these because of the lack of preventive medications (NK1RA) and the non-updated local guidelines [19]. However, this study revealed that the prophylactic patterns had the potential for both overuse and underuse. Underuse, i.e., lack of medication or insufficient doses according to the recommendations of the clinical guidelines, such as the ASCO, NCCN, and MASCC and ESMO guidelines for CINV prevention and treatment, may result in suboptimal patient outcomes, specifically nausea and vomiting, which occurred in approximately half of the included patients in the current study. Overuse, or the use of a medication not recommended by the guidelines mentioned above or at a higher dose than the recommended one, may lead to adverse effects and a waste of resources that could be allocated to better choices.
The patients in the HEC group were supposed to have more aggressive treatment. However, the prophylaxis patterns of the group did not differ from the other two groups MEC and LEC (Table 2), and the emetic risks of CT were not associated with physicians’ decisions about the anti-emesis drug selection, doses and durations of treatment (Table 3). Even patients in the HEC group seemed to have more other risk factors for nausea and vomiting such as concurrently treated with RT, younger age, and without the habit of consuming alcohol frequently [20,21].These results demonstrated a potential for improper CINV prophylaxis in the study.
Due to the unavailability of NK1RA in Vietnam, the prophylaxis for patients treated with HEC might be more likely to be insufficient. Olanzapine, which could be a choice according to the international guidelines, was not used in the study. A higher dose of GC, which was recommended in case of no NK1RA [9,17], was prescribed in 44.2% of the patients in the acute phase and in less than 5% in the delayed phase. The lack of prophylaxis for the delayed phase could also be seen in patients with no prevention medication in HEC group (21.4% on day 2, 55.4% on day 3 and 63.8% on day 4) and MEC groups (23.0% and 50.8% on day 2 and 3, respectively), or low proportion of patients in HEC having combination therapy.
Olanzapine was not utilized, which could be due to its weak evidence and the fear of its side effects such as QT prolongation (especially when combined with a 5-HT3RA). Moreover, olanzapine was not yet in the local guideline for CINV prophylaxis and it did not have the indication in the label. As a result, it was not covered by insurance, which could be a barrier to the usage. However, olanzapine did not cost much for several days and could be an acceptable solution for several specific patients receiving HEC. Ideally, the local guideline should be updated with taking olanzapine into consideration.
Another resolution for the lack of NK1RA was a higher dose of GC. However, HEC was not shown to relate with the choice of high dose GC in this study. The median dose of GC in the patients with high dose was still low (16 mg, IQR: 16–20 mg) for the acute phase, comparing with the recommendation of 20 mg in the local or ESMO guideline in the absence of NK1RA [9,19]. Therefore, high dose GC in this study did not show relation to the patients’ outcomes. This practice still gave a room to improve the clinical outcomes by using a higher dose of GC for patients with HEC.
In contrast, there was still a potential overuse of prophylactic medications. It was seen in the use of combination therapy in 100% of patients with LEC on day 1, and in 27.9% and 14.7% of patients with MEC on days 2 and 3, respectively, where a single drug of GC or a 5-HT3RA is recommended. The use of a high dose of 5-HT3RAs for the acute phase in 55.6% to 67.4% of patients in all groups was also a source of wasting. The high doses of 5-HT3RAs in this study were almost double those in the international guidelines. Although these guidelines are similar for 5-HT3RA dose recommendations [9,17], the local guidelines do not provide guidance on the doses, which could be a difficulty for physicians [19]. The prices for one-day treatment of these medications are usually 5–10 times higher than those of GC or olanzapine, according to the prices on the website of Drug Administration of Vietnam [22]. A high dose of 5-HT3RA also did not relate to the decrease in CINV occurrence in this study. Therefore, to save resources, the selection or dose of 5-HT3RA should be considered carefully.
In the current study, gender was a factor related to many aspects of CINV prophylaxis therapy, which could be understandable. Female patients, who are associated with a higher risk of CINV occurrence [20,21], were more likely to be treated with more drugs or for a longer duration in the delayed phase than their male counterparts. However, they were less likely to be prescribed high doses of 5-HT3RA, which could be explained by the fact that physicians may be more careful with the adverse effects of 5-HT3RA such as QT prolongation and torsade point, which are more likely to happen in women. In contrast, the potential emetogenicity of CT, which is usually the basis for choosing the therapy in the evidence guidelines, was not shown to be associated with the CINV prophylaxis patterns. Besides, patients with other risk factors for CINV, including younger age and a history of balance disorder [20,21], were chosen for less intensive therapy in the current study. These findings supported the aforementioned results that no difference in the prophylactic patterns among the LEC, MEC, and HEC groups was seen.
The incidences of different types of CINV showed the higher proportions in the higher emetic risk groups with significant differences (for all CINV:
Beside the emetic risk of CT, from multivariable logistic regression analysis, we found female gender and younger age were associated with both acute and delayed events. These findings were consistent with previous studies [20,23,24]. Recently, the prediction models of CINV risk have been developed taking into account patients’ factors for appropriate choices of antiemetics [25,26].
About the incidence of CINV, it was 47.3% in the current study, meaning that 52.7% patients did not experience any CINV. This rate was lower than that reported in a study conducted in 6 Asia Pacific countries in 2014 of which the complete response rate was 69% [27]. However, it was similar to the findings of a study in 8 European countries in 2010 which showed that the proportion of no CINV in the first cycle were 50.7% and 59.9% in the groups of non-adherence and adherence to anti emesis guidelines respectively [11]. At that time NK1RA and olanzapine had not yet been recommended in the ESMO guideline [10].
When looking at nausea and vomiting separately, the incidences of nausea were higher in both acute and delayed phases in the current study, 20.4% vs. 9.9% and 42.5% vs. 24.5% respectively, which showed that nausea was more difficult to control than vomiting. These patterns were similar to those in the recent studies in Korea and Jordan. In the Korean study, the rates of nausea in the acute and delayed phases were much higher than ours (46% and 83% respectively), while those of acute and delayed vomiting were quite similar (8.3% and 23%, respectively) [28]. However, the patients in that study received HEC and MEC only. In Jordan where NK1RA and olanzapine were not prescribed in the prophylaxis like the practice in ours, nausea and vomiting were at higher rates, about 60% and 42% respectively in acute phase, and similarly in the delayed phase [15].
Although this current study revealed the situation of CINV prophylaxis and patients’ outcomes, it had some limitations to be acknowledged. First, it only recorded the prescribed preventive medications for the patients to use at the hospitals and to take home after their CT administration. During the delayed phase at home, whether they took the medication and whether their medication were changed due to the occurrence of CINV were not studied. Therefore, it might interfere the effects of the prophylaxis regimens on CINV occurrence and severity. However, the description of the patterns of CINV prophylaxis still reflected the reality of the current practice, which was the main objective of this study. Second, the disparity in the number of patients among groups of CINV risks (HEC, MEC, and LEC) and their characteristics could affect the comparisons among the groups. This limitation is due to the nature of the cross-sectional study design used in this study, which is not allowed to allocate or match patients among groups. Third, this study did not consider physicians’ prescribing behaviors, which could also contribute to the pattern of CINV prophylaxis, especially the overuse in LEC and MEC groups and the underuse in HEC. Future research, including physician interviews about their prescribing behaviors, is needed to explore underlying reasons.
In conclusion, the pattern of CINV prophylaxis in this study showed a potential over- and under-use of anti-emesis medication. Emetic risk of CT has not yet been a key factor to consider a pattern of CINV prophylaxis, and CT with higher emetic risk was associated with higher possibility of developing CINV. While the resource was limited, more effort should be made to allocate the current resources effectively by taking into account the aspects recommended in the reputed guidelines such as emetic risk of CT and doses of medications. Ideally, an up-to-date local guideline should be established based on local resources, the available data and the international evidence-based guidelines to standardize the practice.
In conclusion, the pattern of CINV prophylaxis in this study showed a potential over- and under-use of anti-emesis medication. Emetic risk of CT has not yet been a key factor to consider a pattern of CINV prophylaxis, and CT with higher emetic risk was associated with higher possibility of developing CINV. While the resource was limited, more effort should be made to allocate the current resources effectively by taking into account the aspects recommended in the reputed guidelines such as emetic risk of CT and doses of medications. Ideally, an up-to-date local guideline should be established based on local resources, the available data and the international evidence-based guidelines to standardize the practice.
No potential conflict of interest relevant to this article was reported.
R Clin Pharm 2023; 1(2): 115-126
Published online December 31, 2023 https://doi.org/10.59931/rcp.23.030
Copyright © Asian Conference On Clinical Pharmacy.
Hanh T.H. Nguyen1 , Linh K. Duong1 , Tien D Vu2 , Thu T.M. Hoang2 , Duong V Bach3 , Hao T.L Hoang3 , Huong T.L. Nguyen1
1Department of Clinical Pharmacy, Hanoi University of Pharmacy, Hanoi, Vietnam
2Department of Pharmacy, Vietnam National Cancer Hospital, Hanoi, Vietnam
3Department of Pharmacy, Hanoi Oncology Hospital, Hanoi, Vietnam
Correspondence to:Huong T.L. Nguyen
E-mail Huongntl@hup.edu.vn
ORCID
https://orcid.org/0000-0002-8051-5017
This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: Chemotherapy-induced nausea and vomiting (CINV) is considered as the most distressing side effect of chemotherapy (CT). Overuse of prophylaxis leads to waste of resources, whereas underuse results in poor CINV prevention. Therefore, it is important to explore the pattern of CINV prophylaxis and CINV occurrence to improve the quality of medication use.
Methods: This prospective, observational study was conducted in two oncology-specialized hospitals in North Vietnam in 2019 to investigate the CINV prophylaxis regimens and CINV events among cancer patients receiving intravenous high-, moderate-, or lowemetic-risk CT (HEC, MEC, or LEC, respectively) after the administration of the first cycle of CT. The Multinational Association of Supportive Care in Cancer Antiemesis Tool was used to assess nausea and vomiting.
Results: The study included 294 patients (HEC: 224, MEC: 61, and LEC: 9). They were prescribed a similar pattern of prophylaxis for both acute and delayed emesis regardless of their emetic risks. Patient-specific factors, including sex, age, balance disorder, and comorbidity, but not the emetic risk of CT, were associated with the choice of prophylactic regimens. The incidence rate of CINV was 47.3% in all patients. Multivariate analysis revealed that HEC was associated with an increased risk of acute CINV compared with MEC (odds ratio, 2.34; 95% confidence interval, 1.03–5.32), whereas female sex and younger age were associated with a higher risk of delayed CINV.
Conclusion: CINV management is challenging and may involve over- and underuse of prophylaxis. A more effective approach to prophylaxis in the context of limited resources needs to be developed to improve patient outcomes.
Keywords: Chemotherapy-induced nausea and vomiting, Prophylaxis, Patient care
Chemotherapy-induced nausea and vomiting (CINV) which are considered the most distressing side effects related to chemotherapy [1], would lead to deteriorate patients’ clinical outcomes (e.g., dehydration, electrolyte imbalances, weight loss, and malnourishment), and negatively impact their quality of life and daily activities [2,3]. These events would also result in additional supportive care, outpatient visits, or hospitalization, which might incur an economic burden [4]. Moreover, inadequate control of CINV could lead to patients’ delay or refusal of possible life-saving chemotherapy (CT) [5]. Although antiemetic medications have reduced the incidence of vomiting and nausea substantially, approximately 30% to 60% of patients still experienced acute or delayed nausea after chemotherapy [2]. This demonstrates a great challenge for healthcare professionals to prevent and treat CINV with available antiemetic agents in clinical practice settings, especially in limited-resource countries. Therefore, proper antiemetic use is claimed as one of the five key opportunities to improve care and reduce costs in cancer by the American Society of Clinical Oncology (ASCO) [6].
To foster the appropriate use of antiemetic medications in cancer patients treated with CT, the evidence-based guidelines for CINV prevention and treatment have been developed and updated by various cancer societies, e.g., ASCO, National Comprehensive Cancer Network (NCCN), Multinational Association of Supportive Care in Cancer and the European Society of Medical Oncology (MASCC and ESMO) [7-9]. As per their consensus, the prophylaxis of CINV with medications such as serotonin receptor antagonists (5-HT3RA), dexamethasone, neurokinin-1 receptor antagonists (NK1RA), and olanzapine, should be the primary goal and be implemented specifically for those patients treated with CT classified as highly emetogenic chemotherapies (HEC) [7-10]. Adherence to the guidelines demonstrated a strong correlation with the improvement of CINV control in PEER or INSPIRE studies with significant reductions in CINV incidences [11,12].
However, the use of antiemetic drugs for CINV was still an issue. In high-income countries, where mostly antiemetic medications are available, overuse was reported in 24.1% of patients of various CINV risk groups [13], and in up to 70% of patients receiving low-emetogenic chemotherapies (LEC) [6]. No prophylaxis or a lack of prophylactic duration were also reported as the main reasons for non-adherence to the clinical guidelines in these countries [11,12]. In resource-limited countries, it’s more challenging to follow the antiemetic guidelines, as it is presumed that “socio-economic status” is one of the rationales for non-adherence [14]. A study in Jordan showed that there were no NK1RA in the CINV prophylaxis therapy, and the prevalence of CINV was as high as more than 70% in the included patients [15]. In a study based on medication data in China, there was also a lack of NK1RA and an overutilization of the other antiemetic agents [16]. Therefore, it’s essential to reveal prophylactic problems to improve the practice of CINV prophylaxis and patient outcomes. In Vietnam, a resource-limited country, antiemetic medications were not available in full range, and information about CINV prophylaxis and its effectiveness was limited. Therefore, we conducted this study to investigate the patterns of CINV prophylaxis in Vietnam and patient-reported CINV events to explore the potential gap in practice. With that, we expected to propose measurements to use the limited healthcare resources optimally and to improve patient outcomes in the future.
This prospective, observational study was conducted at two oncology-specialized hospitals in North Vietnam from March 2019 to June 2019. Patients were eligible if they had a confirmed cancer diagnosis, were aged ≥18 years, and were scheduled to take the first cycle of a single-day, intravenous chemotherapy. The emetic risk of chemotherapy was classified as HEC, moderate emetogenic chemotherapy (MEC), and LEC according to the NCCN guidelines for anti-emesis version 2019.1 [17]. Patients who concurrently took (an) oral cytotoxic agent(s), used to be treated with chemotherapy, experienced nausea or vomiting in 24 hours before the administration of the chemotherapy, had brain metastases or bowel obstruction, or whose medical records were not accessed, were excluded.
Data collection consisted of medical record extraction and patient interviews. A form was created to retrieve data from medical records, which included patient demographic and clinical characteristics, the prescribed chemotherapy and medications for CINV prophylaxis, and other co-medications. The demographic and clinical characteristics consisted of age, gender, body weight and height, type of cancer, comorbidities, and the status of concurrent radiotherapy. The chemotherapy and all prescribed medications were recorded for their names and active ingredients, route of administration, dose, and duration. A questionnaire about risk factors for CINV, including history of balance disorder, motion sickness, nausea and vomiting related to pregnancy (if applicable), anxiety about chemotherapy, and frequency of alcohol drinking, was used to interview patients. The Multinational Association of Supportive Care in Cancer Antiemesis Tool (MAT), a validated tool to measure CINV, was used to assess patients’ experiences of the occurrence and severity of nausea and vomiting [18]. Permission to use MAT was granted by the Multinational Association of Supportive Care in Cancer (MASCC).
There was a pre-established process for data collection. Every day during the study period, the clinical pharmacists screened a list of patients admitted to the hospitals for those with the first cycle of chemotherapy. Their medical records were reviewed to select eligible patients for the study and to collect the data on the form. The selected patients were interviewed to obtain their risk factors for CINV and were introduced to MAT before or during their administration of chemotherapy. The time of the initiation of the chemotherapy administration was noted to identify the acute and delayed phases of CINV. The patients were approached twice, 24 hours and 120 hours after the chemotherapy, for their responses to MAT in the acute and delayed phases, respectively. The process is summarized in Fig. 1.
The MAT used to evaluate patients’ experiences of nausea and vomiting consists of eight items. The first four are for nausea and vomiting for the acute phase, which is 24 hours following chemotherapy, and the last four items are for the delay phase, the period after that until 120 hours. In both phases, patients were asked whether they experienced nausea (items 3 and 7) and vomiting (items 1 and 5). They were asked about the severity of nausea (items 4 and 8) on a scale from 0 (no nausea) to 10 (as much as possible) and the number of vomiting episodes (items 2 and 6) [18]. In this study, the events of CINV were defined as severe if patients reported from four episodes of vomiting, or from level six of nausea in the questions about CINV severity in MAT.
As for the 5-HT3RA, the upper limit of an agent’s daily dose recommended for the acute phase by NCCN [17], which were similar to those by ASCO, MASCC and ESMO [7-10], and was considered one recommended dose (RD). One RD of the locally available 5-HT3RA was 0.25 mg of intravenous (IV) palonosetron, 24 mg of oral ondansetron, 16 mg of IV ondansetron, or 1 mg of IV granisetron. If the 5-HT3RA daily dose of a patient was higher than one RD, it would be classified as a high dose of 5-HT3RA.
As the NCCN guidelines recommended that in prophylactic regimens with a NK1RA, the daily doses of dexamethasone are 12 mg and 8 mg for the acute and delayed phases, respectively, and higher glucocorticoid (GC) doses should be used when a NK1RA agent is not given concomitantly [17]. Therefore, the daily dose of GC was considered to be high if it was higher than 12 mg of dexamethasone for the acute phase and more than 8 mg of dexamethasone for the delayed phase. The GC included in the chemotherapy regimen was also calculated for the CINV prophylaxis.
The medications used on the day of CT were recorded for the acute phase (day 1), and those on the following days were for the delayed phase (from day 2 to day 4). Patients were considered to be covered by 5-HT3RA in the delayed phase if they were prescribed IV palonosetron on day 1 due to its long half-life [12].
Patient demographic and clinical characteristics, risk factors were summarized using descriptive statistics. These factors, the pattern of prophylaxis, and the incidence of CINV were compared among groups of patients treated with HEC, MEC, and LEC using the Pearson Chi-square test or Fisher exact test for categorical variables and one way ANOVA for the continuous variables. The associations between demographics, clinical characteristics, risk factors, and the pattern of CINV prophylaxis were analyzed using χ2-test. The variable of emetic risks of CT but not cancer type was employed in the analysis because they are correlated, and the risk of CT was the primary criterion for CINV prophylactic regimen selection. The associations of the demographics and clinical characteristics, risk factors, the patterns of CINV prophylaxis with the occurrence of events in the acute and delayed phases were also examined using χ2-test. Then the factors that had the
The study included 294 patients in three groups: HEC, MEC, and LEC. The groups were significantly different in patients’ age, the frequent alcohol usage, the concomitant use of radiation therapy (RT), and the type of cancer (Table 1). The mean age was 54.4 years in the whole studied population, however, more patients under 60 years old were in the HEC group compared with MEC and LEC groups, 70.1% vs. 59.0% and 22.2% respectively (
Table 1 . Patient demographics and clinical characteristics.
Patient characteristics\N | Total (N=294) | HEC (N=224) | MEC (N=61) | LEC (N=9) | |
---|---|---|---|---|---|
Age (year), mean (SD) | 54.4 (11.1) | 53.4 (10.7) | 56.6 (11.4) | 64.2 (13.7) | 0.003 |
Age <60 years old, n (%) | 195 (66.3) | 157 (70.1) | 36 (59.0) | 2 (22.2) | 0.005 |
BMI, n (%) | 0.13 | ||||
<18.5 | 41 (13.9) | 29 (13.0) | 8 (13.1) | 4 (44.4) | |
18.5≤BMI<25 | 226 (76.9) | 174 (77.7) | 48 (78.7) | 4 (44.4) | |
≥25 | 27 (9.2) | 21 (9.4) | 5 (8.2) | 1 (11.1) | |
Female gender, n (%) | 171 (58.2) | 128 (57.1) | 39 (63.9) | 4 (44.4) | 0.44 |
Having comorbidity, n (%) | 100 (34.0) | 74 (33.0) | 23 (37.7) | 3 (33.3) | 0.79 |
Concurrent RT, n (%) | 86 (29.3) | 74 (33.0) | 12 (19.7) | 0 | 0.02 |
Nervousness before CT, n (%) | 15 (5.1) | 12 (5.4) | 3 (4.9) | 0 | 1.00 |
History of balance disorder, n (%) | 36 (12.2) | 31 (13.8) | 5 (8.2) | 0 | 0.26 |
History of carsick, n (%) | 103 (35.0) | 76 (33.9) | 26 (42.6) | 1 (11.1) | 0.14 |
History of nausea/vomit related to pregnancy, n (%) | 58 (19.7) | 42 (18.8) | 14 (23.0) | 2 (22.2) | 0.75 |
Frequent alcohol usage, n (%) | 64 (21.8) | 48 (21.4) | 11 (18.0) | 5 (55.6) | 0.04 |
Type of cancer, n (%) | <0.0001 | ||||
Breast cancer | 88 (30.0) | 61 (27.2) | 25 (41.0) | 2 (22.2) | |
Digestive tract cancer* | 75 (25.5) | 53 (23.7) | 22 (36.1) | 0 | |
Lung cancer | 53 (18.0) | 44 (19.6) | 8 (13.1) | 1 (11.1) | |
Gynecologic cancer | 45 (15.3) | 41 (18.3) | 2 (3.3) | 2 (22.2) | |
Non-Hodgkin lymphoma | 17 (5.8) | 13 (5.8) | 4 (6.5) | 0 | |
Others | 16 (5.4) | 12 (5.4) | 0 | 4 (44.5) |
*Including tongue, nasopharyngeal, esophagus, gastric, colorectal cancer..
BMI=body mass index, CT=chemotherapy, HEC=high emetic risk chemotherapy, LEC=low emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, RT=radiotherapy, SD=standard deviation..
In the acute phase, a combination therapy of GC and 5-HT3RA was prescribed for all patients, regardless of the potential emetogenicity of the chemotherapy they used (Table 2). The proportions of patients treated with high doses of GC and 5-HT3RA were not significantly different among the groups (
Table 2 . Pattern of prophylaxis for chemotherapy induced-nausea and vomiting.
Pattern of CINV prophylaxis\N | HEC (N=224) | MEC (N=61) | LEC (N=9) | |
---|---|---|---|---|
Acute phase - day 1 | ||||
Prophylaxis therapy | ||||
GC+5HT3 RA | 224 (100.0) | 61 (100.0) | 9 (100.0) | - |
High dose of GC | 99 (44.2) | 24 (39.3) | 3 (33.3) | 0.67 |
High dose of 5-HT3 RA | 151 (67.4) | 34 (55.7) | 5 (55.6) | 0.20 |
Delayed phase - day 2 | ||||
Prophylaxis therapy | 0.37 | |||
No use | 48 (21.4) | 14 (23.0) | 4 (44.4) | |
GC only | 23 (10.3) | 11 (18.0) | 1 (11.1) | |
5-HT3 RA only | 89 (39.7) | 19 (31.1) | 3 (33.3) | |
GC+5-HT3 RA | 64 (28.6) | 17 (27.9) | 1 (11.1) | |
High dose of GC | 11 (4.9) | 3 (4.9) | 0 (0) | 1.00 |
High dose of 5-HT3 RA | 29 (13.0) | 12 (19.7) | 1 (11.1) | 0.40 |
Delayed phase - day 3 | ||||
Prophylaxis therapy | 0.10 | |||
No use | 124 (55.4) | 31 (50.8) | 8 (88.9) | |
GC only | 13 (5.8) | 6 (9.8) | 0 (0) | |
5-HT3 RA only | 70 (31.2) | 15 (24.6) | 0 (0) | |
GC+5-HT3 RA | 17 (7.6) | 9 (14.7) | 1 (11.1) | |
High dose of GC | 10 (4.5) | 1 (1.6) | 0 (0) | 0.62 |
High dose of 5-HT3 RA | 8 (3.6) | 7 (11.5) | 0 (0) | 0.06 |
Delayed phase - day 4 | ||||
Prophylaxis therapy | 0.36 | |||
No use | 143 (63.8) | 45 (73.8) | 9 (100) | |
GC only | 9 (4.0) | 2 (3.3) | 0 (0) | |
5-HT3 RA only | 66 (29.5) | 13 (21.3) | 0 (0) | |
GC+5-HT3 RA | 6 (2.7) | 1 (1.6) | 0 (0) | |
High dose of GC | 8 (3.6) | 0 (0) | 0 (0) | 0.38 |
High dose of 5-HT3 RA | 3 (1.3) | 5 (8.2) | 0 (0) | 0.04 |
HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists..
In terms of prophylaxis for delayed nausea and vomiting, there was no statistically significant difference in the usage of anti-emesis medications for the patients in the groups from day 2 to day 4. The percentages of patients without prophylaxis increased from day 2 to day 4 (21.4%, 55.4% and 63.8% of the patients with HEC, respectively, and 23.0%, 50.8% and 73.8% of those with MEC, respectively). The proportion of the patients in HEC group treated with the combination of GC and 5-HT3RA was 28.6% on day 2 and decreased to 2.7% on day 4. Whereas 49.1% and 34.4% of the patients in MEC, were prophylactic with either GC or 5-HT3RA on day 2 and 3 respectively. The proportions of patients with high dose GC were under 5% across the groups of all the days.
The body mass index (BMI) equal or greater than 25 was the only factor related to the choice of high dose GC in the acute phase with OR=0.3 (0.10–0.87) compared to BMI less than 18.5 (Table 3). Fewer female than male patients were prescribed 5-HT3RA at the high dose in the acute phase (OR=0.42, 95% CI=0.02–0.87). However, the female patients were 2.47 times more likely to be treated enough or more days for the delayed prophylaxis than their male counterparts. They also tended to be treated with more drugs than men during the delayed phase, but were not significantly different. The patients having history of balance disorder were less likely to be treated enough or more days for the delayed prophylaxis (OR=0.4, 95% CI=0.17–0.91), while those with comorbidity were indicated two medications less frequent than the patients without comorbidity (OR=0.39, 95% CI=0.19–0.80).
Table 3 . Multi-variable analysis for factors associated with the pattern of CINV prophylaxis.
Factors | Acute phase | Delayed phase | |||
---|---|---|---|---|---|
High dose of GC* | High dose of 5-HT3 RA* | More or enough days of delayed prophylaxis* | Number of drugs selection# | ||
2 drugs (5-HT3 RA and GC) vs. no use | 1 drug (5-HT3 RA or GC) vs. no use | ||||
Emetic risk of CT | - | - | - | - | |
HEC | 0.64 (0.35-1.17) | ||||
MEC | 1 | ||||
LEC | NE | ||||
Female gender (yes vs. no) | - | 0.42 (0.20-0.87) | 2.47 (1.29-4.75) | 1.38 (0.64-2.98) | 1.92 (0.96-3.85) |
Age <60 years old (yes vs. no) | - | 0.57 (0.32-0.99) | - | - | - |
Concurrent RT (yes vs. no) | - | 1.69 (0.92-3.10) | 1.03 (0.57-1.85) | - | - |
Nervousness before CT (yes vs. no) | - | - | - | - | - |
History of balance disorder (yes vs. no) | - | - | 0.40 (0.17-0.91) | - | - |
History of carsick (yes vs. no) | - | 0.92 (0.51-1.64) | 1.25 (0.69-2.26) | 1.10 (0.47-2.59) | 1.35 (0.64-2.86) |
History of nausea/vomit related to pregnancy (yes vs. no) | - | - | 1.23 (0.63-2.40) | - | - |
Frequent alcohol usage (yes vs. no) | - | 1.24 (0.53-2.93) | - | - | - |
Having comorbidity (yes vs. no) | - | - | - | 0.39 (0.19-0.80) | 0.78 (0.42-1.43) |
BMI | - | - | - | - | |
<18.5 | 1 | ||||
18.5≤BMI<25 | 0.65 (0.33-1.27) | ||||
≥25 | 0.30 (0.10-0.87) |
Values are presented as odds ratio (95% confidence interval)..
*Multi-variable logistic regression model including variables with
CINV=chemotherapy induced nausea and vomiting, OR=odds ratio, CI=confident interval, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists, CT=chemotherapy, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, NE=not estimated due to no event in the group, RT=radiotherapy, BMI=body mass index..
CINV happened in 20.8% and 46.4% of the patients in the acute and delayed phase respectively, and 47.3% experienced CINV in either phase (Table 4). The incidence of nausea was higher than of vomiting in both phases, and the delayed events happened more frequently than the acute ones in all categories. When comparing amongst groups of emetic risk, the higher risk groups had the higher incidences of the events with significant difference in some types of events such as: all CINV (
Table 4 . CINV incidences and the comparison among emetogenic risk groups.
CINV\N | Total (n=294) | HEC (n=224) | MEC (n=61) | LEC (n=9) | |
---|---|---|---|---|---|
All CINV | 139 (47.3) | 116 (51.8) | 22 (36.1) | 1 (11.1) | 0.01 |
All acute CINV | 61 (20.8) | 53 (23.7) | 8 (13.1) | 0 | 0.06 |
Acute vomiting | 29 (9.9) | 25 (11.2) | 4 (6.6) | 0 | 0.34 |
Acute nausea | 60 (20.4) | 52 (23.2) | 8 (13.1) | 0 | 0.07 |
Severe acute CINV | 31 (10.5) | 27 (12.1) | 4 (6.6) | 0 | 0.27 |
Severe acute vomiting | 18 (6.1) | 16 (7.1) | 2 (3.3) | 0 | 0.65 |
Severe acute nausea | 28 (9.5) | 25 (11.2) | 3 (4.9) | 0 | 0.21 |
All delayed CINV# | 131 (46.4) | 108 (50.0) | 22 (37.9) | 1 (12.5) | 0.04 |
Delayed vomiting | 72 (25.5) | 62 (28.7) | 10 (17.2) | 0 | 0.05 |
Delayed nausea | 125 (44.3) | 103 (47.7) | 21 (36.2) | 1 (12.5) | 0.06 |
Severe delayed CINV# | 69 (23.5) | 61 (28.2) | 8 (13.8) | 0 | 0.02 |
Severe delayed vomiting | 35 (12.4) | 30 (13.9) | 5 (8.6) | 0 | 0.31 |
Severe delayed nausea | 57 (20.2) | 50 (23.2) | 7 (12.1) | 0 | 0.06 |
#There were 282, 216, 61 and 8 patients in total, HEC, MEC and LEC group respectively because some patients were lost to contact in the delayed phase..
CINV=chemotherapy induced nausea and vomiting, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy..
The multiple logistic regression analysis showed that patients treated with HEC was the only factor related to the 2.34 times higher risk of developing acute CINV than with MEC (OR=2.34 95% CI=1.03–5.32). Whereas HEC also tended to have more delayed CINV but non-significantly after controlling other factors. Female gender was associated with higher risk of acute and delayed CINV but only statistically significant in two models of delayed CINV (Table 5). For the risk of delayed CINV, in the model 1, besides gender, the occurrence of acute CINV increased the risk of delayed events by 10.26 times (95% CI=4.45–23.63). In the model 2, which excluded the patients with acute CINV, younger age was an additional factor associated with the increased risk of the event by 2.08 times (95% CI=1.07–4.07) after controlling for the others.
Table 5 . Multivariate logistic regression analysis for acute and delayed CINV.
Factors\N | Acute CINV* (N=294) | Delayed CINV | ||
---|---|---|---|---|
Model 1*,# (N=282) | Model 2 | |||
Emetic risk of CT | ||||
HEC | 2.34 (1.03-5.32) | 1.47 (0.74-2.90) | 1.87 (0.88-3.96) | |
MEC | 1 | 1 | 1 | |
LEC | NE | 0.44 (0.05-4.27) | 0.45 (0.04-4.83) | |
Female gender (yes vs. no) | 1.95 (0.902-4.21) | 3.22 (1.41-7.33) | 3.38 (1.38-8.28) | |
Age <60 years old (yes vs. no) | - | 1.76 (0.96-3.22) | 2.08 (1.07-4.07) | |
Concurrent RT (yes vs. no) | - | - | - | |
Nervousness before CT (yes vs. no) | - | - | - | |
History of balance disorder (yes vs. no) | - | - | - | |
History of carsick (yes vs. no) | 1.57 (0.80-3.10) | 1.18 (0.61-2.29) | 1.19 (0.58-2.44) | |
History of nausea/vomiting related to pregnancy (yes vs. no) | 1.34 (0.65-2.76) | 1.36 (0.64-2.91) | 1.97 (0.86-4.54) | |
Frequent alcohol usage (yes vs. no) | - | 1.56 (0.65-3.73) | 1.32 (0.50-3.52) | |
Having comorbidity (yes vs. no) | - | - | - | |
BMI (18.5≤ BMI<25, ≥25 vs. <18) | - | - | - | |
High dose of acute GC (yes vs. no) | - | - | - | |
High dose of 5-HT3RAs in the acute phase (yes vs. no) | - | - | - | |
Days of delayed CINV prophylaxis | NT | - | - | |
Therapy selection | NT | - | - | |
No used | 1 | - | ||
1 drug | 1.74 (0.78-3.88) | 1.41 (0.60-3.33) | ||
2 drugs | 1.17 (0.56-2.44) | 0.85 (0.38-1.88) | ||
Occurrence of acute CINV | NT | 10.26 (4.45-23.63) | NT |
Values are presented as odds ratio (95% confidence interval)..
*Multi-variable logistic regression model including variables with
CINV=chemotherapy induced nausea and vomiting, OR=odds ratio, CI=confident interval, CT=chemotherapy, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, NE=not estimated due to no event in the group. RT=radiotherapy, BMI=body mass index, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists, NT=not tested in univariate and multivariate analysis..
Although there were international and local emetic prophylaxis guidelines for patients receiving chemotherapy, we could not evaluate the adherence to these because of the lack of preventive medications (NK1RA) and the non-updated local guidelines [19]. However, this study revealed that the prophylactic patterns had the potential for both overuse and underuse. Underuse, i.e., lack of medication or insufficient doses according to the recommendations of the clinical guidelines, such as the ASCO, NCCN, and MASCC and ESMO guidelines for CINV prevention and treatment, may result in suboptimal patient outcomes, specifically nausea and vomiting, which occurred in approximately half of the included patients in the current study. Overuse, or the use of a medication not recommended by the guidelines mentioned above or at a higher dose than the recommended one, may lead to adverse effects and a waste of resources that could be allocated to better choices.
The patients in the HEC group were supposed to have more aggressive treatment. However, the prophylaxis patterns of the group did not differ from the other two groups MEC and LEC (Table 2), and the emetic risks of CT were not associated with physicians’ decisions about the anti-emesis drug selection, doses and durations of treatment (Table 3). Even patients in the HEC group seemed to have more other risk factors for nausea and vomiting such as concurrently treated with RT, younger age, and without the habit of consuming alcohol frequently [20,21].These results demonstrated a potential for improper CINV prophylaxis in the study.
Due to the unavailability of NK1RA in Vietnam, the prophylaxis for patients treated with HEC might be more likely to be insufficient. Olanzapine, which could be a choice according to the international guidelines, was not used in the study. A higher dose of GC, which was recommended in case of no NK1RA [9,17], was prescribed in 44.2% of the patients in the acute phase and in less than 5% in the delayed phase. The lack of prophylaxis for the delayed phase could also be seen in patients with no prevention medication in HEC group (21.4% on day 2, 55.4% on day 3 and 63.8% on day 4) and MEC groups (23.0% and 50.8% on day 2 and 3, respectively), or low proportion of patients in HEC having combination therapy.
Olanzapine was not utilized, which could be due to its weak evidence and the fear of its side effects such as QT prolongation (especially when combined with a 5-HT3RA). Moreover, olanzapine was not yet in the local guideline for CINV prophylaxis and it did not have the indication in the label. As a result, it was not covered by insurance, which could be a barrier to the usage. However, olanzapine did not cost much for several days and could be an acceptable solution for several specific patients receiving HEC. Ideally, the local guideline should be updated with taking olanzapine into consideration.
Another resolution for the lack of NK1RA was a higher dose of GC. However, HEC was not shown to relate with the choice of high dose GC in this study. The median dose of GC in the patients with high dose was still low (16 mg, IQR: 16–20 mg) for the acute phase, comparing with the recommendation of 20 mg in the local or ESMO guideline in the absence of NK1RA [9,19]. Therefore, high dose GC in this study did not show relation to the patients’ outcomes. This practice still gave a room to improve the clinical outcomes by using a higher dose of GC for patients with HEC.
In contrast, there was still a potential overuse of prophylactic medications. It was seen in the use of combination therapy in 100% of patients with LEC on day 1, and in 27.9% and 14.7% of patients with MEC on days 2 and 3, respectively, where a single drug of GC or a 5-HT3RA is recommended. The use of a high dose of 5-HT3RAs for the acute phase in 55.6% to 67.4% of patients in all groups was also a source of wasting. The high doses of 5-HT3RAs in this study were almost double those in the international guidelines. Although these guidelines are similar for 5-HT3RA dose recommendations [9,17], the local guidelines do not provide guidance on the doses, which could be a difficulty for physicians [19]. The prices for one-day treatment of these medications are usually 5–10 times higher than those of GC or olanzapine, according to the prices on the website of Drug Administration of Vietnam [22]. A high dose of 5-HT3RA also did not relate to the decrease in CINV occurrence in this study. Therefore, to save resources, the selection or dose of 5-HT3RA should be considered carefully.
In the current study, gender was a factor related to many aspects of CINV prophylaxis therapy, which could be understandable. Female patients, who are associated with a higher risk of CINV occurrence [20,21], were more likely to be treated with more drugs or for a longer duration in the delayed phase than their male counterparts. However, they were less likely to be prescribed high doses of 5-HT3RA, which could be explained by the fact that physicians may be more careful with the adverse effects of 5-HT3RA such as QT prolongation and torsade point, which are more likely to happen in women. In contrast, the potential emetogenicity of CT, which is usually the basis for choosing the therapy in the evidence guidelines, was not shown to be associated with the CINV prophylaxis patterns. Besides, patients with other risk factors for CINV, including younger age and a history of balance disorder [20,21], were chosen for less intensive therapy in the current study. These findings supported the aforementioned results that no difference in the prophylactic patterns among the LEC, MEC, and HEC groups was seen.
The incidences of different types of CINV showed the higher proportions in the higher emetic risk groups with significant differences (for all CINV:
Beside the emetic risk of CT, from multivariable logistic regression analysis, we found female gender and younger age were associated with both acute and delayed events. These findings were consistent with previous studies [20,23,24]. Recently, the prediction models of CINV risk have been developed taking into account patients’ factors for appropriate choices of antiemetics [25,26].
About the incidence of CINV, it was 47.3% in the current study, meaning that 52.7% patients did not experience any CINV. This rate was lower than that reported in a study conducted in 6 Asia Pacific countries in 2014 of which the complete response rate was 69% [27]. However, it was similar to the findings of a study in 8 European countries in 2010 which showed that the proportion of no CINV in the first cycle were 50.7% and 59.9% in the groups of non-adherence and adherence to anti emesis guidelines respectively [11]. At that time NK1RA and olanzapine had not yet been recommended in the ESMO guideline [10].
When looking at nausea and vomiting separately, the incidences of nausea were higher in both acute and delayed phases in the current study, 20.4% vs. 9.9% and 42.5% vs. 24.5% respectively, which showed that nausea was more difficult to control than vomiting. These patterns were similar to those in the recent studies in Korea and Jordan. In the Korean study, the rates of nausea in the acute and delayed phases were much higher than ours (46% and 83% respectively), while those of acute and delayed vomiting were quite similar (8.3% and 23%, respectively) [28]. However, the patients in that study received HEC and MEC only. In Jordan where NK1RA and olanzapine were not prescribed in the prophylaxis like the practice in ours, nausea and vomiting were at higher rates, about 60% and 42% respectively in acute phase, and similarly in the delayed phase [15].
Although this current study revealed the situation of CINV prophylaxis and patients’ outcomes, it had some limitations to be acknowledged. First, it only recorded the prescribed preventive medications for the patients to use at the hospitals and to take home after their CT administration. During the delayed phase at home, whether they took the medication and whether their medication were changed due to the occurrence of CINV were not studied. Therefore, it might interfere the effects of the prophylaxis regimens on CINV occurrence and severity. However, the description of the patterns of CINV prophylaxis still reflected the reality of the current practice, which was the main objective of this study. Second, the disparity in the number of patients among groups of CINV risks (HEC, MEC, and LEC) and their characteristics could affect the comparisons among the groups. This limitation is due to the nature of the cross-sectional study design used in this study, which is not allowed to allocate or match patients among groups. Third, this study did not consider physicians’ prescribing behaviors, which could also contribute to the pattern of CINV prophylaxis, especially the overuse in LEC and MEC groups and the underuse in HEC. Future research, including physician interviews about their prescribing behaviors, is needed to explore underlying reasons.
In conclusion, the pattern of CINV prophylaxis in this study showed a potential over- and under-use of anti-emesis medication. Emetic risk of CT has not yet been a key factor to consider a pattern of CINV prophylaxis, and CT with higher emetic risk was associated with higher possibility of developing CINV. While the resource was limited, more effort should be made to allocate the current resources effectively by taking into account the aspects recommended in the reputed guidelines such as emetic risk of CT and doses of medications. Ideally, an up-to-date local guideline should be established based on local resources, the available data and the international evidence-based guidelines to standardize the practice.
In conclusion, the pattern of CINV prophylaxis in this study showed a potential over- and under-use of anti-emesis medication. Emetic risk of CT has not yet been a key factor to consider a pattern of CINV prophylaxis, and CT with higher emetic risk was associated with higher possibility of developing CINV. While the resource was limited, more effort should be made to allocate the current resources effectively by taking into account the aspects recommended in the reputed guidelines such as emetic risk of CT and doses of medications. Ideally, an up-to-date local guideline should be established based on local resources, the available data and the international evidence-based guidelines to standardize the practice.
No potential conflict of interest relevant to this article was reported.
Table 1 Patient demographics and clinical characteristics
Patient characteristics\N | Total (N=294) | HEC (N=224) | MEC (N=61) | LEC (N=9) | |
---|---|---|---|---|---|
Age (year), mean (SD) | 54.4 (11.1) | 53.4 (10.7) | 56.6 (11.4) | 64.2 (13.7) | 0.003 |
Age <60 years old, n (%) | 195 (66.3) | 157 (70.1) | 36 (59.0) | 2 (22.2) | 0.005 |
BMI, n (%) | 0.13 | ||||
<18.5 | 41 (13.9) | 29 (13.0) | 8 (13.1) | 4 (44.4) | |
18.5≤BMI<25 | 226 (76.9) | 174 (77.7) | 48 (78.7) | 4 (44.4) | |
≥25 | 27 (9.2) | 21 (9.4) | 5 (8.2) | 1 (11.1) | |
Female gender, n (%) | 171 (58.2) | 128 (57.1) | 39 (63.9) | 4 (44.4) | 0.44 |
Having comorbidity, n (%) | 100 (34.0) | 74 (33.0) | 23 (37.7) | 3 (33.3) | 0.79 |
Concurrent RT, n (%) | 86 (29.3) | 74 (33.0) | 12 (19.7) | 0 | 0.02 |
Nervousness before CT, n (%) | 15 (5.1) | 12 (5.4) | 3 (4.9) | 0 | 1.00 |
History of balance disorder, n (%) | 36 (12.2) | 31 (13.8) | 5 (8.2) | 0 | 0.26 |
History of carsick, n (%) | 103 (35.0) | 76 (33.9) | 26 (42.6) | 1 (11.1) | 0.14 |
History of nausea/vomit related to pregnancy, n (%) | 58 (19.7) | 42 (18.8) | 14 (23.0) | 2 (22.2) | 0.75 |
Frequent alcohol usage, n (%) | 64 (21.8) | 48 (21.4) | 11 (18.0) | 5 (55.6) | 0.04 |
Type of cancer, n (%) | <0.0001 | ||||
Breast cancer | 88 (30.0) | 61 (27.2) | 25 (41.0) | 2 (22.2) | |
Digestive tract cancer* | 75 (25.5) | 53 (23.7) | 22 (36.1) | 0 | |
Lung cancer | 53 (18.0) | 44 (19.6) | 8 (13.1) | 1 (11.1) | |
Gynecologic cancer | 45 (15.3) | 41 (18.3) | 2 (3.3) | 2 (22.2) | |
Non-Hodgkin lymphoma | 17 (5.8) | 13 (5.8) | 4 (6.5) | 0 | |
Others | 16 (5.4) | 12 (5.4) | 0 | 4 (44.5) |
*Including tongue, nasopharyngeal, esophagus, gastric, colorectal cancer.
BMI=body mass index, CT=chemotherapy, HEC=high emetic risk chemotherapy, LEC=low emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, RT=radiotherapy, SD=standard deviation.
Table 2 Pattern of prophylaxis for chemotherapy induced-nausea and vomiting
Pattern of CINV prophylaxis\N | HEC (N=224) | MEC (N=61) | LEC (N=9) | |
---|---|---|---|---|
Acute phase - day 1 | ||||
Prophylaxis therapy | ||||
GC+5HT3 RA | 224 (100.0) | 61 (100.0) | 9 (100.0) | - |
High dose of GC | 99 (44.2) | 24 (39.3) | 3 (33.3) | 0.67 |
High dose of 5-HT3 RA | 151 (67.4) | 34 (55.7) | 5 (55.6) | 0.20 |
Delayed phase - day 2 | ||||
Prophylaxis therapy | 0.37 | |||
No use | 48 (21.4) | 14 (23.0) | 4 (44.4) | |
GC only | 23 (10.3) | 11 (18.0) | 1 (11.1) | |
5-HT3 RA only | 89 (39.7) | 19 (31.1) | 3 (33.3) | |
GC+5-HT3 RA | 64 (28.6) | 17 (27.9) | 1 (11.1) | |
High dose of GC | 11 (4.9) | 3 (4.9) | 0 (0) | 1.00 |
High dose of 5-HT3 RA | 29 (13.0) | 12 (19.7) | 1 (11.1) | 0.40 |
Delayed phase - day 3 | ||||
Prophylaxis therapy | 0.10 | |||
No use | 124 (55.4) | 31 (50.8) | 8 (88.9) | |
GC only | 13 (5.8) | 6 (9.8) | 0 (0) | |
5-HT3 RA only | 70 (31.2) | 15 (24.6) | 0 (0) | |
GC+5-HT3 RA | 17 (7.6) | 9 (14.7) | 1 (11.1) | |
High dose of GC | 10 (4.5) | 1 (1.6) | 0 (0) | 0.62 |
High dose of 5-HT3 RA | 8 (3.6) | 7 (11.5) | 0 (0) | 0.06 |
Delayed phase - day 4 | ||||
Prophylaxis therapy | 0.36 | |||
No use | 143 (63.8) | 45 (73.8) | 9 (100) | |
GC only | 9 (4.0) | 2 (3.3) | 0 (0) | |
5-HT3 RA only | 66 (29.5) | 13 (21.3) | 0 (0) | |
GC+5-HT3 RA | 6 (2.7) | 1 (1.6) | 0 (0) | |
High dose of GC | 8 (3.6) | 0 (0) | 0 (0) | 0.38 |
High dose of 5-HT3 RA | 3 (1.3) | 5 (8.2) | 0 (0) | 0.04 |
HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists.
Table 3 Multi-variable analysis for factors associated with the pattern of CINV prophylaxis
Factors | Acute phase | Delayed phase | |||
---|---|---|---|---|---|
High dose of GC* | High dose of 5-HT3 RA* | More or enough days of delayed prophylaxis* | Number of drugs selection# | ||
2 drugs (5-HT3 RA and GC) vs. no use | 1 drug (5-HT3 RA or GC) vs. no use | ||||
Emetic risk of CT | - | - | - | - | |
HEC | 0.64 (0.35-1.17) | ||||
MEC | 1 | ||||
LEC | NE | ||||
Female gender (yes vs. no) | - | 0.42 (0.20-0.87) | 2.47 (1.29-4.75) | 1.38 (0.64-2.98) | 1.92 (0.96-3.85) |
Age <60 years old (yes vs. no) | - | 0.57 (0.32-0.99) | - | - | - |
Concurrent RT (yes vs. no) | - | 1.69 (0.92-3.10) | 1.03 (0.57-1.85) | - | - |
Nervousness before CT (yes vs. no) | - | - | - | - | - |
History of balance disorder (yes vs. no) | - | - | 0.40 (0.17-0.91) | - | - |
History of carsick (yes vs. no) | - | 0.92 (0.51-1.64) | 1.25 (0.69-2.26) | 1.10 (0.47-2.59) | 1.35 (0.64-2.86) |
History of nausea/vomit related to pregnancy (yes vs. no) | - | - | 1.23 (0.63-2.40) | - | - |
Frequent alcohol usage (yes vs. no) | - | 1.24 (0.53-2.93) | - | - | - |
Having comorbidity (yes vs. no) | - | - | - | 0.39 (0.19-0.80) | 0.78 (0.42-1.43) |
BMI | - | - | - | - | |
<18.5 | 1 | ||||
18.5≤BMI<25 | 0.65 (0.33-1.27) | ||||
≥25 | 0.30 (0.10-0.87) |
Values are presented as odds ratio (95% confidence interval).
*Multi-variable logistic regression model including variables with
CINV=chemotherapy induced nausea and vomiting, OR=odds ratio, CI=confident interval, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists, CT=chemotherapy, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, NE=not estimated due to no event in the group, RT=radiotherapy, BMI=body mass index.
Table 4 CINV incidences and the comparison among emetogenic risk groups
CINV\N | Total (n=294) | HEC (n=224) | MEC (n=61) | LEC (n=9) | |
---|---|---|---|---|---|
All CINV | 139 (47.3) | 116 (51.8) | 22 (36.1) | 1 (11.1) | 0.01 |
All acute CINV | 61 (20.8) | 53 (23.7) | 8 (13.1) | 0 | 0.06 |
Acute vomiting | 29 (9.9) | 25 (11.2) | 4 (6.6) | 0 | 0.34 |
Acute nausea | 60 (20.4) | 52 (23.2) | 8 (13.1) | 0 | 0.07 |
Severe acute CINV | 31 (10.5) | 27 (12.1) | 4 (6.6) | 0 | 0.27 |
Severe acute vomiting | 18 (6.1) | 16 (7.1) | 2 (3.3) | 0 | 0.65 |
Severe acute nausea | 28 (9.5) | 25 (11.2) | 3 (4.9) | 0 | 0.21 |
All delayed CINV# | 131 (46.4) | 108 (50.0) | 22 (37.9) | 1 (12.5) | 0.04 |
Delayed vomiting | 72 (25.5) | 62 (28.7) | 10 (17.2) | 0 | 0.05 |
Delayed nausea | 125 (44.3) | 103 (47.7) | 21 (36.2) | 1 (12.5) | 0.06 |
Severe delayed CINV# | 69 (23.5) | 61 (28.2) | 8 (13.8) | 0 | 0.02 |
Severe delayed vomiting | 35 (12.4) | 30 (13.9) | 5 (8.6) | 0 | 0.31 |
Severe delayed nausea | 57 (20.2) | 50 (23.2) | 7 (12.1) | 0 | 0.06 |
#There were 282, 216, 61 and 8 patients in total, HEC, MEC and LEC group respectively because some patients were lost to contact in the delayed phase.
CINV=chemotherapy induced nausea and vomiting, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy.
Table 5 Multivariate logistic regression analysis for acute and delayed CINV
Factors\N | Acute CINV* (N=294) | Delayed CINV | ||
---|---|---|---|---|
Model 1*,# (N=282) | Model 2 | |||
Emetic risk of CT | ||||
HEC | 2.34 (1.03-5.32) | 1.47 (0.74-2.90) | 1.87 (0.88-3.96) | |
MEC | 1 | 1 | 1 | |
LEC | NE | 0.44 (0.05-4.27) | 0.45 (0.04-4.83) | |
Female gender (yes vs. no) | 1.95 (0.902-4.21) | 3.22 (1.41-7.33) | 3.38 (1.38-8.28) | |
Age <60 years old (yes vs. no) | - | 1.76 (0.96-3.22) | 2.08 (1.07-4.07) | |
Concurrent RT (yes vs. no) | - | - | - | |
Nervousness before CT (yes vs. no) | - | - | - | |
History of balance disorder (yes vs. no) | - | - | - | |
History of carsick (yes vs. no) | 1.57 (0.80-3.10) | 1.18 (0.61-2.29) | 1.19 (0.58-2.44) | |
History of nausea/vomiting related to pregnancy (yes vs. no) | 1.34 (0.65-2.76) | 1.36 (0.64-2.91) | 1.97 (0.86-4.54) | |
Frequent alcohol usage (yes vs. no) | - | 1.56 (0.65-3.73) | 1.32 (0.50-3.52) | |
Having comorbidity (yes vs. no) | - | - | - | |
BMI (18.5≤ BMI<25, ≥25 vs. <18) | - | - | - | |
High dose of acute GC (yes vs. no) | - | - | - | |
High dose of 5-HT3RAs in the acute phase (yes vs. no) | - | - | - | |
Days of delayed CINV prophylaxis | NT | - | - | |
Therapy selection | NT | - | - | |
No used | 1 | - | ||
1 drug | 1.74 (0.78-3.88) | 1.41 (0.60-3.33) | ||
2 drugs | 1.17 (0.56-2.44) | 0.85 (0.38-1.88) | ||
Occurrence of acute CINV | NT | 10.26 (4.45-23.63) | NT |
Values are presented as odds ratio (95% confidence interval).
*Multi-variable logistic regression model including variables with
CINV=chemotherapy induced nausea and vomiting, OR=odds ratio, CI=confident interval, CT=chemotherapy, HEC=high emetic risk chemotherapy, MEC=moderate emetic risk chemotherapy, LEC=low emetic risk chemotherapy, NE=not estimated due to no event in the group. RT=radiotherapy, BMI=body mass index, GC=glucocorticoids, 5-HT3 RA=5-HT3 receptor antagonists, NT=not tested in univariate and multivariate analysis.