Table 5.
Sensitivity analyses of the effect of ondansetron on the latency to the onset of emesis induced by cisplatin (5 or 10 mg kg−1)
| Effect estimate (proportion of latency change) | 95% CI | Number of comparisons | Overall effect (Z tests) | Heterogeneity (χ2) | |
|---|---|---|---|---|---|
| Cisplatin dose | |||||
| 10 mg kg−1 | 0.88 | 0.49 to 1.27 | 15 | P < 0.001*** | P < 0.001*** |
| 5 mg kg−1 | 0.60 | −0.97 to 2.16 | 3 | P = 0.45 | P = 0.29 |
| Ondansetron dose | |||||
| 0.01–0.02 mg kg−1 | 0.36 | 0.00 to 0.72 | 3 | P = 0.050 | P = 0.008** |
| 0.03–0.06 mg kg−1 | 0.85 | 0.30 to 1.40 | 3 | P = 0.002** | P < 0.001*** |
| 0.1 mg kg−1 | 0.49 | 0.31 to 0.67 | 5 | P < 0.001*** | P = 0.140 |
| 1 mg kg−1 | 1.96 | 1.66 to 2.25 | 7 | P < 0.001*** | P = 0.27 |
| Mode of delivery of ondansetron | |||||
| i.v. | 0.60 | 0.35 to 0.85 | 9 | P < 0.001*** | P < 0.001*** |
| p.o. | 0.34 | 0.15 to 0.53 | 2 | P < 0.001*** | N/A |
| i.p. | 1.54 | 0.67 to 2.41 | 5 | P < 0.001*** | P < 0.001*** |
| s.c. | 0.36 | −0.94 to 1.65 | 2 | P = 0.59 | P = 0.31 |
| Treatment time | |||||
| With cisplatin | 0.62 | 0.34 to 0.68 | 6 | P < 0.001*** | P = 0.010* |
| 20–30 min before | 1.11 | 0.25 to 1.96 | 5 | P = 0.010** | P < 0.001*** |
| 1 h before | 0.34 | 0.15 to 0.53 | 2 | P = 0.010** | N/A |
| 30 min before and 45 min after | 0.94 | −0.01 to 1.89 | 2 | P = 0.050* | P < 0.001*** |
| Twice daily | 0.36 | −0.94 to 1.65 | 2 | P = 0.590 | P = 0.310 |
| Three times daily | 5.83 | −3.02 to 14.68 | 1 | P = 0.020* | N/A |
| Mode of administration of cisplatin | |||||
| i.v. | 0.60 | 0.35 to 0.85 | 9 | P < 0.001*** | P < 0.001*** |
| i.p. | 1.16 | 0.35 to 1.96 | 9 | P = 0.005 | P < 0.001*** |
| Animal origin | |||||
| UK | 0.76 | 0.44 to 1.07 | 5 | P < 0.001*** | P = 0.19 |
| USA | 0.86 | 0.39 to 1.33 | 13 | P < 0.001*** | P < 0.001*** |
| Quality score | |||||
| Less that 5/9 | 0.94 | −0.01 to 1.89 | 2 | P = 0.050 | P < 0.001*** |
| 5 to 7/9 | 0.63 | 0.32 to 0.94 | 11 | P < 0.001*** | P < 0.001*** |
| 7.5 to 9/9 | 1.42 | 0.31 to 2.53 | 5 | P = 0.01** | P < 0.001*** |
The effect estimate was computed as the weighted mean difference (WMD) and represents the impact of the treatment on the latency expressed as a proportion of the latency in the control group. An effect estimate <0 indicates that the latency was shorter in the control group than in the treatment group, 0 indicates that the treatment had no effect on the latency and an effect estimate of 1 indicates that the treatment increased the latency by 100%. The variables examined were the dose and mode of administration of cisplatin, the dose of ondansetron, the regimen of ondansetron administration (mode of delivery and timing relative to cisplatin administration), the animal origin (country animals were bred) and the quality score assigned to the study where comparisons were extracted