Division of Psychiatry, University of Western Australia, Crawley, Western Australia, 6009, Australia.
School of Medical and Health Sciences, Edith Cowan University, Joondalup, Western Australia, 6027, Australia.
BMC Pediatr. 2020 Nov 30;20(1):539. doi: 10.1186/s12887-020-02431-z.
Cardiovascular anomalies are the largest group of congenital anomalies and the major cause of death in young children, with various data linking rising atrial septal defect incidence (ASDI) with prenatal cannabis exposure. Objectives / Hypotheses. Is cannabis associated with ASDI in USA? Is this relationship causal?
Geospatiotemporal cohort study, 1991-2016. Census populations of adults, babies, congenital anomalies, income and ethnicity. Drug exposure data on cigarettes, alcohol abuse, past month cannabis use, analgesia abuse and cocaine taken from National Survey of Drug Use and Health (78.9% response rate). Cannabinoid concentrations from Drug Enforcement Agency. Inverse probability weighted (ipw) regressions. Analysis conducted in R.
ASDI rose nationally three-fold from 27.4 to 82.8 / 10,000 births 1991-2014 during a period when tobacco and alcohol abuse were falling but cannabis was rising. States including Nevada, Kentucky, Mississippi and Tennessee had steeply rising epidemics (Time: Status β-estimate = 10.72 (95%C.I. 8.39-13.05), P < 2.0 × 10 ). ASDI was positively related to exposure to cannabis and most cannabinoids. Drug exposure data was near-complete from 2006 thus restricting spatial modelling from 2006 to 2014, N = 282. In geospatial regression models cannabis: alcohol abuse term was significant (β-estimate = 19.44 (9.11, 29.77), P = 2.2 × 10 ); no ethnic or income factors survived model reduction. Cannabis legalization was associated with a higher ASDI (Time: Status β-estimate = 0.03 (0.01, 0.05), P = 1.1 × 10 Weighted panel regression interactive terms including cannabis significant (from β-estimate = 1418, (1080.6, 1755.4), P = 7.3 × 10 Robust generalized linear models utilizing inverse probability weighting interactive terms including cannabis appear (from β-estimate = 78.88, (64.38, 93.38), P = 1.1 × 10 Marginal structural models with machine-aided SuperLearning association of ASDI with high v. low cannabis exposure R.R. = 1.32 (1.28, 1.36). Model e-values mostly > 1.5.
ASDI is associated with cannabis use, frequency, intensity and legalization in a spatiotemporally significant manner, robust to socioeconomicodemographic adjustment and fulfilled causal criteria, consistent with multiple biological mechanisms and similar reports from Hawaii, Colorado, Canada and Australia. Not only are these results of concern in themselves, but they further imply that our list of the congenital teratology of cannabis is as yet incomplete, and highlight in particular cardiovascular toxicology of prenatal cannabinoid and drug exposure.
心血管异常是先天性异常中最大的一组，也是幼儿死亡的主要原因，各种数据表明，随着大麻暴露导致的房间隔缺损发生率（ASDI）上升。目的/假设。大麻是否与美国的 ASD 有关？这种关系是否有因果关系？
1991-2016 年的地理时空队列研究。成人、婴儿、先天畸形、收入和种族的普查人群。药物暴露数据包括香烟、酗酒、过去一个月大麻使用、滥用镇痛药和可卡因，来源于国家药物使用和健康调查（78.9%的应答率）。大麻中的大麻素浓度来源于毒品执法局。采用逆概率加权（ipw）回归进行分析。分析在 R 中进行。
1991-2014 年，在烟草和酒精滥用下降但大麻使用上升期间，美国的 ASD 发病率全国上升了三倍，从 27.4/10000 升至 82.8/10000。内华达州、肯塔基州、密西西比州和田纳西州等州的发病率急剧上升（时间：状态β估计值=10.72（95%CI 8.39-13.05），P<2.0×10）。ASDI 与大麻暴露和大多数大麻素呈正相关。2006 年以后，药物暴露数据几乎完整，因此从 2006 年到 2014 年限制了空间模型，N=282。在地理空间回归模型中，大麻与酒精滥用的关系显著（β估计值=19.44（9.11，29.77），P=2.2×10）；没有种族或收入因素在模型简化后仍然存在。大麻合法化与 ASDI 升高有关（时间：状态β估计值=0.03（0.01，0.05），P=1.1×10）加权面板回归交互项包括大麻显著（β估计值=1418（1080.6，1755.4），P=7.3×10）稳健广义线性模型利用包括大麻的逆概率加权交互项呈阳性（β估计值=78.88（64.38，93.38），P=1.1×10）具有机器辅助超学习的边际结构模型，ASDI 与高 v. 低大麻暴露的关联 R.R. =1.32（1.28，1.36）。模型 e 值大多大于 1.5。
ASDI 与大麻使用、频率、强度和合法化呈时空显著相关，经社会经济人口调整后仍具有稳健性，并满足因果关系标准，与多种生物学机制一致，与来自夏威夷、科罗拉多、加拿大和澳大利亚的类似报告一致。这些结果本身不仅令人担忧，而且进一步表明，我们对大麻先天性畸形的清单还不完整，特别是强调了产前大麻素和药物暴露的心血管毒理学。
