The risk of adverse cardiovascular events from varenicline balanced against the benefits in mortality from smoking cessation.
The risk of adverse cardiovascular events from varenicline balanced against the benefits in mortality from smoking cessation.
复制标题
伐尼克兰引起的不良心血管事件的风险与戒烟带来的死亡率的益处相平衡。
DOI：
10.1093/ntr/nts028
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Goldstein,AdamO
中科院分区：
文献类型：
--
作者：
Kistler,ChristineE;Goldstein,AdamO
从伐尼克兰的不良心血管事件的风险合理的长期时间范围，以框预期戒烟的好处。虽然我们可以选择谈论在他们整个剩余的预期寿命中可能产生的益处，但我们选择了10年的时间范围，因为这似乎与临床最相关，并允许处方者谈论未来十年的益处。让我们假设一组妇女接受伐尼克兰的处方，另一组没有接受药物支持（安慰剂）。首先，我们想知道这两组女性发生心血管事件的初始风险。来自Fragrance研究的数据（Wilson等人，1998年）显示，对于一个平均50岁的女性吸烟者，在其他健康状况良好的情况下，一年中不良心血管事件的风险为0.4%。因此，在没有接受药物治疗的一组中，1000名妇女中有4名可能在一年内发生心血管事件。如果伐尼克兰使不良心血管事件的几率增加1.72，则伐尼克兰组发生心血管事件的平均绝对风险将增加0.69%或平均绝对风险增加0.29%。这意味着在服用伐伦克林的1000名妇女中，大约有7人在第一年会发生不良心血管事件。鉴于Singh研究中OR为1.72左右的95% CI，伐尼克兰组的绝对风险范围可能低至0.44%（OR为1.09）或高达1.08%（OR为2.71）;绝对风险增加0.04%-0.68%（分别为0.44%-0.4%和1.08%-0.4%）。所需的伤害数量是绝对风险增加0.29%的倒数。因此，每345名服用伐尼克兰的女性中，大约有一名女性在一年内发生心血管事件，在这种情况下，这种增加在未来几年不会增加，因为我们假设药物的使用仅在第1年。然而，我们不应该在这里结束我们的讨论，因为即使最初的危害风险增加，戒烟的可能益处改善了许多结果，其中最具体的是总体死亡率。根据科克伦关于戒烟结果的综述，（Cahill，Stead，&兰开斯特，2007）80/1，000在安慰剂组中，1,000人中有217人（8.0%）在1年结束时自发成功戒烟，而服用伐尼克兰的1,000人中有217人（21.7%），或137名额外的吸烟者在1年内因药物而戒烟。那些戒烟的人明显降低了死亡的风险，这种好处会随着时间的推移而增加。如果我们在我们选择的10年时间范围内观察死亡率的好处，我们需要计算两组中全因死亡率的差异。在30-55岁的女性吸烟者队列中，发现目前吸烟者每10，000人年有85.6例死亡（3，602例死亡/420，761人年）。戒烟5-10年的人的全因死亡风险比是目前吸烟者的三分之二（0.67）（Kenfield等人，2008年）。为了数学上的简单起见，我们假设在第1年戒烟后，在剩下的9年里，两组的吸烟状况变化是相等的，因此可以忽略不计。因此，安慰剂组中有920名女性继续吸烟10年（9200人-年），其中79人预计死亡（9200人-年乘以0.00856例死亡/人-年）。如果这80名妇女没有戒烟，预计其中6.8人死亡（800人-年乘以0.00856例死亡/人-年），但这一数字因戒烟而减少了三分之二（0.67），10年内死亡人数为4.6人。
The risk of adverse cardiovascular events from varenicline reasonable long-term time horizon in order to frame the expected smoking cessation benefits. While we could choose to talk about the benefits that might accrue over their entire remaining life expectancy, we have chosen a 10-year horizon because this seems the most clinically relevant and allows a prescriber to talk about benefits over the next decade of life. Let us suppose that one group of women receives a prescription for varenicline and the other group receives no pharmacologic support (placebo). First, we would want to know the initial risk of cardiovascular events in both groups of women. Data from the Framingham study (Wilson et al., 1998) show that for an average 50-year-female smoker, in otherwise good health, the risk of an adverse cardiovascular event in a year is 0.4%. Thus, in the group receiving no pharmacotherapy, 4 of the 1, 000 women will likely have a cardiovascular event within the year. If varenicline increases the odds of an adverse cardiovascular event by 1.72, the varenicline group will experience an average absolute risk of cardiovascular events of 0.69% or an average absolute risk increase of 0.29%. This means that roughly 7 of the 1, 000 women taking varenicline will have an adverse cardiovascular event in this first year. Given the 95% CI around the 1.72 OR in the Singh study, the range of the absolute risk in the varenicline group may be as low as 0.44%(for an OR of 1.09) or as high as 1.08%(for an OR of 2.71); a 0.04%–0.68% absolute risk increase (0.44% minus 0.4% and 1.08% minus 0.4%, respectively). The number needed to harm is the inverse of the absolute risk increase of 0.29%. Thus, approximately one additional woman has a cardiovascular event over a year for every 345 women who take varenicline, an increase that does not increase in future years in this scenario as we assume the use of the medication is in Y ear 1 only.Yet we should not end our discussion here, for even if there is an increased risk of harm initially, the possible benefit of smoking cessation improves a number of outcomes, the most concrete of which is overall mortality. According to the Cochrane review on outcomes of smoking cessation,(Cahill, Stead, & Lancaster, 2007) 80 of 1,000 (8.0%) in the placebo group will spontaneously and successfully quit at the end of 1 year, compared with 217 of 1,000 (21.7%) in those who take varenicline, or 137 additional smokers quit in 1 year as a result of the medication. Those that quit smoking clearly reduce their risk of mortality, a benefit that increases over time. If we look at the mortality benefit over our chosen 10-year horizon, we need to calculate the difference in all-cause mortality in our two groups. Among a cohort of women smokers aged 30–55, current smokers were found to have 85.6 deaths per 10,000 person-years (3, 602 deaths/420,761 person-years). The all-cause mortality hazard ratio of those who quit smoking for 5–10 years was two-thirds (0.67) that of current smokers (Kenfield et al., 2008). For the sake of mathematical simplicity, we will assume that after the effort to quit smoking in Y ear 1, smoking status changes will be equal in both groups over the remaining 9 years and are thus negligible. The placebo group, therefore, has 920 women who continue to smoke for 10 years (9, 200 person-years), of whom 79 are expected to die (9, 200 person-years multiplied by 0.00856 deaths/person-years). If the 80 women had not quit smoking, 6.8 of them are expected to die (800 person-years multiplied by 0.00856 deaths/person-years), but this number is reduced by smoking cessation by two-thirds (0.67) to 4.6 deaths over 10 years …