An estimated 85.6 million American adults have at least one cardiovascular disease such as heart disease, stroke, heart failure, or hypertension (Mozaffarian et al., 2016). Each year cardiovascular diseases account for more than 800,000 deaths (i.e., they are the underlying cause listed on the death certificate), or 30 percent of all deaths in the United States (Mozaffarian et al., 2016).
Heart disease is the leading cause of mortality in the United States, accounting for more than 600,000 deaths per year (Kochanek et al., 2016). Within subcategories of heart disease, coronary heart disease (CHD) is by far the largest, with 364,000 deaths annually (Kochanek et al., 2016). CHD is a disease in which a waxy substance called plaque builds up inside the blood vessels supplying the heart (i.e., the coronary arteries). Over the course of years or decades, the plaque can harden or rupture, resulting in an inadequate supply of blood to the heart which may, in some instances, result in death of heart muscle (myocardial infarction).
Both coronary heart disease and stroke are associated with aging, with nearly 93 percent of CHD deaths and 94 percent of stroke deaths occurring in individuals 55 years and older (Kochanek et al., 2016). More
than one-third (about 36 percent) of CHD deaths occur in individuals ages 85 years and older, while 43 percent of stroke deaths occur in this age group (Kochanek et al., 2016).
Current (past-month) cannabis use is fairly low in the older populations most likely to experience cardiovascular diseases—in particular, about 2 percent past-month prevalence in those ages 50 years and older. In younger adults, by contrast, the prevalence of cannabis use has been estimated to be as high as 19.6 percent for past-month use among those ages 18 to 25 years (Azofeifa et al., 2016), but these rates decline dramatically with aging. In contrast, tobacco smoking—a known risk factor for heart disease and stroke—has a much higher prevalence among older adults: 18 percent in those ages 45 to 64 years and 8.5 percent in those ages 65 years and older who smoke (Jamal et al., 2015).
Cardiometabolic disorders result in a substantial economic burden on the United States. From 2011 to 2012 the estimated annual cost of cardiovascular diseases, including heart disease, stroke, hypertensive disease, and other circulatory conditions, was $316.6 billion ($207.3 billion for heart disease, $33.0 billion for stroke). The total estimated cost of diagnosed diabetes in 2012 was $245 billion (Mozaffarian et al., 2016).
The objective of the review of cannabis and cardiometabolic conditions was to assess the independent association of cannabis with these conditions in studies in which the association has been quantified. The justification for examining cannabis use in relation to cardiometabolic conditions is that these conditions are among the leading causes of death; are highly prevalent in the United States; account for high levels of medical care utilization and cost; and are caused, in significant part, by modifiable lifestyle risk factors, including diet, physical activity, and cigarette smoking. The high prevalence of these conditions means that a behavior that is associated with a small degree of increased risk for heart disease, stroke, or diabetes can be associated with a high level of attributable risk, that is, the number of cases of disease that result from that behavior. While the prevalence of cardiometabolic conditions is concentrated in the older-adult age groups which have low rates of cannabis use, it is expected that the expanding legalization of cannabis use will cause the rates of use to increase.
The discussion in this review is limited to acute myocardial infarction, stroke, metabolic dysregulation and metabolic syndrome, and diabetes. Sudden death and arrhythmias such as atrial fibrillation were other topics of interest for which no data were available to quantify the association with cannabis use. The 1999 Institute of Medicine (IOM) report Marijuana and Medicine: Assessing the Science Base (IOM, 1999) reviewed the cardiovascular system; however, no conclusions or recommendations related to cannabis use and cardiometabolic outcomes were included in that report.
The literature search conducted by the current committee did not identify any systematic reviews that were rated as “good” or “fair” for cannabis use and acute myocardial infarction, stroke, dyslipidemia or metabolic syndrome, or diabetes, so all of the available primary literature for these outcomes dating back to 1999 was reviewed and the 12 primary articles rated as “good” or “fair” by the committee have been included in this chapter.
ACUTE MYOCARDIAL INFARCTION
Each year, an estimated approximately 550,000 Americans have an incident (i.e., first-time) heart attack (acute myocardial infarction, or AMI) and about 200,000 have a recurrent attack (Mozaffarian et al., 2016). Of those who have a heart attack each year, about 116,000 die as a result of their coronary event (Mozaffarian et al., 2016). The committee responsible for the IOM report Marijuana and Medicine: Assessing the Science Base (1999) did not make any conclusions or recommendations regarding cannabis use and acute myocardial infarctions.
The acute cardiovascular effects of cannabis include increases in heart rate and supine blood pressure and postural hypotension (Beaconsfield et al., 1972; Benowitz and Jones, 1981). Smoking cannabis decreases exercise test duration on maximal exercise tests and increases the heart rate at submaximal levels of exercise (Renaud and Cormier, 1986). These acute effects provide a physiological basis for cardiac ischemia to develop in cannabis users. In fact, the time from exercise to the onset of angina pectoris is decreased by smoking one cannabis cigarette (Aronow and Cassidy, 1974). Tolerance develops to the acute effects of tetrahydrocannabinol (THC) over several days to a few weeks (Gorelick et al., 2013). Reported cardiovascular effects that may increase the risk of AMI include irregular heart rate (Khiabani et al., 2008) and impaired vascular endothelial function (demonstrated in rates from exposure to secondhand cannabis smoke) (Wang et al., 2016). Additionally, carbon dioxide production from smoked cannabis decreases the oxygen-carrying capacity of the blood and may contribute to the development of cardiac ischemia.
There have been numerous case reports suggesting that cannabis use is associated with the occurrence of AMI. The two primary studies that have quantified the risk of AMI associated with cannabis use and that were rated as good or fair are reviewed below.
Is There an Association Between Cannabis Use and Acute Myocardial Infarction?
The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis and AMI. Three descriptive review articles provided useful background: Sidney (2002), Thomas et al. (2014), and Franz and Frishman (2016).
A retrospective cohort study (Sidney, 2002; Sidney et al., 1997) assessed the risk of hospitalization for AMI associated with cannabis use in a cohort of 62,012 men and women ages 15 through 49 years who had, from mid-1979 through 1985, completed self-administered research questions on their cannabis, tobacco, and alcohol use. AMI was assessed by linkage to electronically maintained records of all overnight hospitalizations in Kaiser Permanente Northern California. Follow-up was conducted for up to 12 years. Current use of cannabis was reported by 22 percent and former use by 20 percent of the cohort. There were 209 incident AMIs, 173 in men and 36 in women. The relative risk associated with cannabis use was assessed by a Cox proportional hazards model with adjustments for age, race, education, body mass index (BMI), history of hypertension, smoking, and alcohol use. The relative risk for AMI in current users was 1.1 (95% confidence interval [CI] = 0.7–1.7) for men and 1.8 (95% CI = 0.5–6.3) for women; in former users it was 0.9 (95% CI = 0.6–1.5) for men and 1.0 (95% CI = 0.2–4.5) for women. Both current and former cannabis use were unassociated with an increased risk of AMI.
Study limitations included a reliance of self-report of cannabis use which may result in misclassification of this exposure; the lack of availability of longitudinal data on cannabis use; and the relatively young age (mean age 33 years), which meant that the AMIs occurred in a relatively young age range that is not representative of the older age range in which the vast majority of AMIs occur. Cannabis use was assessed at only one point in time.
A case crossover study design was used to examine the role of cannabis use as a possible trigger for myocardial infarction in 3,882 AMI patients in an inception cohort study identified between August 1989 and September 1996 from 64 community and tertiary medical centers in the United States that were part of the Determinants of Myocardial Onset Study (Mittleman et al., 2001). The mean ages of cannabis users and abstainers were 43.7 and 62.0 years, respectively, while 68 percent of cannabis users and 32 percent of abstainers were current tobacco smok-
ers. Nine patients (0.2 percent) interviewed soon after admission for AMI reported cannabis use during the hour preceding the symptoms of AMI. The risk for AMI associated with cannabis use during the hour preceding symptoms of AMI was 4.8 (95% CI = 2.9–9.5) as assessed by a case-crossover analysis. The exclusion of three of the nine patients who reported other triggering behaviors during the hour prior to the AMI (cocaine use and/or sexual intercourse) resulted in a relative risk of 3.2 (95% CI = 1.4–7.3).
The major limitations of this study were its small size and its reliance on self-report for cannabis use status, which meant that any misclassification could have had a significant effect on the results. While the case-crossover design controls for confounding by traditional risk factors for cardiovascular disease, it does not control for interaction of these factors, and one cannot determine whether cannabis acts as a trigger in low-risk individuals or those who are nonsmokers of tobacco.
Discussion of Findings
While there are a number of reports of an association between cannabis use and AMI, only the two studies described above quantify risk, with the Sidney (2002) study demonstrating no association with an increased or decreased risk of AMI and the Mittleman et al. (2001) study finding that cannabis use may act as a trigger for AMI. The limitations of these studies were described. More generally, with the Mittleman study as an exception, most reports of adverse cardiovascular effects of cannabis, including AMI, have been conducted in a relatively young age range, while major cardiovascular events are concentrated in older adults and the findings may not be generalizable to this age group. Other general limitations beyond those already mentioned in the description of the studies include the absence of the impact of the route of consumption (e.g, smoked, edible, etc.); dose, including accounting for the content of THC and other cannabinoids and potential additives or contaminants; and total lifetime duration/dose of cannabis use. Overall, the articles were judged to be of fair quality for assessing the risk of acute myocardial infarction associated with cannabis use.
The role of cannabis as a trigger of AMI is plausible, given its cardiostimulatory effects, which may cause ischemia in susceptible hearts. Carboxyhemoglobinemia from combustion of cannabis resulting in a decreased oxygen-carrying capacity of blood may also contribute to ischemia. Given the physiologic plausibility for a trigger effect, smoking cannabis may put individuals, particularly those at high risk for cardiovascular disease, at increased risk for AMI.
6-1(a) There is limited evidence of a statistical association between cannabis smoking and the triggering of acute myocardial infarction.
6-1(b) There is no evidence to support or refute a statistical association between chronic effects of cannabis use and the risk of acute myocardial infarction.
Stroke is the fifth leading cause of death in the United States, accounting for 133,000 deaths annually (Kochanek et al., 2016). A stroke is the death of a portion of brain tissue due to a disruption of the blood supply. Strokes may be ischemic (inadequate blood/oxygen supply) or hemorrhagic (bleeding into the brain) in origin. Each year, approximately 795,000 people experience a new or recurrent stroke. Approximately 610,000 of these are first stoke occurrences and 185,000 are recurrent stroke events (Mozaffarian et al., 2016). The committee responsible for the IOM report Marijuana and Medicine: Assessing the Science Base (1999) did not make any conclusions or recommendations regarding cannabis use and stroke.
Numerous reports have suggested that smoking cannabis increases the risk of stroke, including case series (Phillips et al., 2011) and studies describing cannabis-associated vascular changes that may be associated with stroke (Herning et al., 2001; Wolff et al., 2011, 2015). Several reports have indicated a close temporal relationship between cannabis smoking and stroke (Wolff et al., 2013). The cardiovascular effects of cannabis that have been proposed as a possible mechanism in the etiology of stroke include orthostatic hypotension with secondary impairment of the auto-regulation of cerebral blood flow, altered cerebral vasomotor function, supine hypertension and swings in blood pressure, cardioembolism with atrial fibrillation, other arrhythmias, vasculopathy, vasospasm, reversible cerebral vasoconstriction syndrome, and multifocal intracranial stenosis (Wolff et al., 2015).
Is There an Association Between Cannabis Use and Stroke?
The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis use and stroke.
A large reported study on the association of cannabis and stroke by Rumalla et al. (2016a) used the Nationwide Inpatient Sample, which provides admission data from a 20 percent sample of all U.S. hospitalizations, to examine the cross-sectional association between cannabis use and hospitalization for acute ischemic stroke (AIS) among patients ages 15 to 54 years during the time period 2004–2011. The primary International Classification of Diseases (ICD)-9-CM discharge code was used to identify AIS, and current cannabis use was identified using the ICD-9-CM code 340.30, which includes both cannabis dependence and nondependent cannabis abuse. Current cannabis use was identified in 11,320 of 478,649 AIS events (2.4 percent). Tobacco use prevalence was higher in current cannabis users than in nonusers (64.4 percent versus 31.5 percent) as was cocaine use (26.7 percent versus 3.1 percent). The odds ratio (OR) associated with current cannabis use and hospitalization for AIS was 1.17 (95% CI = 1.15–1.20) as calculated with multivariable logistic regression adjusted for age, gender, race, substance use, payer status, Charlson’s comorbidity index, and other comorbid risk factors. Analyses stratified on tobacco use status were not available. The limitations of this study include the cross-sectional design; the probable under-ascertainment of current cannabis use (2.4 percent is low for this age range); the absence of data on duration of tobacco use; and the absence of analyses that are stratified by tobacco and by cocaine use to determine the OR in non-tobacco use and non-cocaine users, given the high prevalence of these known risk factors for ischemic stroke.
In a case-control study conducted in a New Zealand hospital (Barber et al., 2013), 160 of 218 (73 percent) of ischemic stroke/transient ischemic attack (TIA) patients, ages 18 to 55 years, had urine drug screens between January 2009 and April 2012 (150 ischemic stroke, 10 TIA). Control urine samples were obtained from 160 patients matched for age, sex, and ethnicity. Twenty-five (15.6 percent) of the stroke/TIA patients and 13 (8.1 percent) of the control patients had positive cannabis drug screens. Eighty-eight percent of cannabis-positive patients were current tobacco smokers versus 28 percent of cannabis-negative patients. The OR associated with current cannabis use was 2.30 (95% CI = 1.08–5.08), but it was no longer statistically significant when an additional adjustment was made for tobacco use (1.59, 95% CI = 0.71–3.70).
In a cross-sectional analysis by Westover et al. (2007) of all ischemic (N = 998) and hemorrhagic strokes (N = 937) identified in 2003 by ICD-9 codes from an administrative database maintained by the state of Texas in young adults ages 18 to 44 years the ORs of cannabis and other illicit drugs being associated with ischemic and hemorrhagic stroke were estimated using a multivariable logistic regression adjusting for alcohol,
tobacco, amphetamines, cocaine, opioids, cardiovascular risk factors, and other medical conditions associated with increased risk of these outcomes. The prevalence of cannabis use, identified by ICD-9 codes, was approximately 1 percent. Cannabis was associated with an increased risk of ischemic stroke (OR, 1.76; 95% CI = 1.15–2.71) but was not associated with a risk of hemorrhagic stroke (OR, 1.36; 95% CI = 0.90–2.06). The prevalence rate of tobacco use was not reported, and analyses stratified by category of tobacco use were not performed.
A retrospective cohort study (Sidney, 2002; Sidney et al., 1997) assessed the risk of hospitalization for stroke associated with cannabis use in a cohort of 62,012 men and women of ages 15 to 49 years who had, from mid-1979 through 1985, completed self-administered research questions on cannabis, tobacco, and alcohol use. Stroke was assessed by linkage to electronically maintained records of all overnight hospitalizations in Kaiser Permanente Northern California. Follow-up was conducted for up to 12 years. Current use of cannabis was reported by 22 percent and former use by 20 percent of the cohort. There were 130 incident strokes, 68 in men and 62 in women. The relative risk associated with cannabis use was assessed by Cox proportional hazards model with adjustments for age, race, education, BMI, history of hypertension, smoking, and alcohol use. The relative risk for stroke in current users was 1.0 (95% CI = 0.5–1.9) for men and 0.7 (95% CI = 0.3–2.2) for women; in former users it was 0.8 (95% CI = 0.4–1.8) for men and 1.5 (95% CI = 0.7–3.5) for women. Both current cannabis use and former cannabis use were not associated with increased risk of stroke.
The study’s limitations included its reliance on self-report of cannabis use, which may result in misclassification of this exposure; the lack of availability of longitudinal data on cannabis use; and the relatively young age of subjects (mean age 33 years) so that the strokes occurred in a relatively young age range that is not representative of the older age range in which the vast majority of strokes occur. Cannabis use was assessed at only one point in time.
Rumalla et al. (2016b) used the Nationwide Inpatient Sample, which provides admission data from a 20 percent sample of all U.S. hospitalizations, to examine the cross-sectional association between cannabis use and hospitalization for aneurysmal subarachnoid hemorrhage (SAH) among patients ages 15 to 54 years during the time period 2004–2011. The primary ICD-9-CM discharge code was used to identify SAH, and current cannabis use was identified using the ICD-9-CM code 340.30, which includes both cannabis dependence and nondependent cannabis abuse. Current cannabis use was identified in 2,104 of the 94,052 (2.2 percent) SAH events. Tobacco use prevalence was higher in current cannabis users than in nonusers (59.3 percent versus 25.4 percent). The OR associated
with current cannabis use was 1.18 (95% CI = 1.12–1.24) according to a multivariate logistic regression adjusted for age, gender, race, substance use, primary payer status, Charlson’s comorbidity index, and other SAH risk factors. The limitations of this study include its cross-sectional design, the probable under-ascertainment of current cannabis use (2.2 percent is low for this age range), the absence of data on duration of cannabis use, and the absence of analyses that are performed stratified by tobacco to determine the OR in non-tobacco use, given the high prevalence of this known risk factor for ischemic stroke.
Discussion of Findings
The studies by Rumalla et al. (2016a,b) and Westover et al. (2007) were cross-sectional studies using administrative data consisting of ICD-9 codes. Cross-sectional studies do not allow one to assess temporality between exposure and outcome. The miscoding of strokes does occur, although the reliability is probably reasonable for epidemiological studies. The classification of exposure status using ICD-9 is particularly concerning, given the likelihood that the percentage of cannabis users appears to be low compared to population norms in each of these studies, most notably the Westover et al. (2007) study.
With the exception of the Sidney (2002) study, none of the studies have data on the temporal relation between the cannabis or tobacco use and the stroke. A general problem was the analytic treatment of tobacco use. Given the much longer duration and frequency of tobacco smoking than of cannabis smoking for most people and the very common co-use of both substances, it is not appropriate to assume that an adjustment for tobacco use in a multivariable model will provide an accurate assessment of the risk associated with cannabis use. Additional analytic approaches, when feasible, may include testing the interaction between cannabis and tobacco use and performing stratified analyses to test the association of cannabis use with clinical endpoints in nonusers of tobacco. Other general limitations beyond those already mentioned in the description of the studies include the absence of the impact of the route of consumption (e.g., smoked, edible, etc.); the absence of information on dose, including accounting for the content of THC and other cannabinoids and potential additives or contaminants; and the lack of information on the total lifetime duration/dose of cannabis use.
All the articles were judged to be of fair quality for assessing the risk of stroke associated with cannabis use. With the exception of Sidney (2002) and Barber et al. (2013), all showed an increased risk of stroke associated with cannabis use but had significant limitations. For ischemic stroke, two of the studies indicated an increased risk while one showed a
nonsignificant finding in the direction of increased risk. For subarachnoid hemorrhage, the single study found an increased risk. For the combined hemorrhagic stroke endpoint assessed by Westover et al. (2007), the study showed no association of cannabis use with the risk of this endpoint.
CONCLUSION 6-2 There is limited evidence of a statistical association between cannabis use and ischemic stroke or subarachnoid hemorrhage.
METABOLIC DYSREGULATION, METABOLIC SYNDROME, PREDIABETES, AND DIABETES MELLITUS
Ranked as the seventh-leading cause of death in the United States, diabetes accounts for more than 76,000 deaths annually (Kochanek et al., 2016). An estimated 29 million adults in the United States have diabetes (CDC, 2014a), which is a group of conditions characterized by high blood glucose (sugar) levels due to the inability to metabolize glucose effectively. The number of new (incident) cases of diabetes diagnosed annually is more than 1.4 million (CDC, 2015). Similar to the case with cardiovascular diseases, the prevalence of diabetes increases with age, from 4.4 percent among those ages 20 to 44 years, to 16.2 percent at ages 45 to 64 years, and 25.9 percent at ages 65 years and older (CDC, 2014a). A major risk factor for the development of the most common type of diabetes (type 2) is obesity, which results in resistance to the effect of the glucose regulating hormone, insulin. An epidemic of obesity has resulted in the prevalence of obesity increasing from 22.9 percent in 1988–1994 to 34.9 percent in 2011–2012 (Flegal et al., 2002; Ogden et al., 2014), contributing to a near tripling of the prevalence of diabetes since 1990 to its current level of 9.3 percent (CDC, 2014b). The committee responsible for the IOM report Marijuana and Medicine: Assessing the Science Base (IOM, 1999) did not make any conclusions or recommendations regarding cannabis use and metabolic dysregulation, metabolic syndrome, prediabetes, or diabetes mellitus.
Obesity, most notably central adiposity, is the dominant risk factor for the development of type 2 diabetes (Klil-Drori et al., 2016). Stimulation of the endogenous cannabinoid receptor system (the CB1 receptor and, to a lesser extent, the CB2 receptor) by Δ9-THC, the major psychoactive component of cannabis, and by endogenous cannabinoids increases appetite and promotes adipogenesis, the production of body fat (Di Marzo et al., 2011). This physiological pathway suggests that cannabinoids such as Δ9-THC may promote weight gain, which would increase the risk of an individual developing diabetes.
As noted earlier, the approximately 30-year epidemic of increasing
obesity rates in the United States has been associated with increasing rates of diabetes. A number of studies have examined the association of cannabis use with BMI and obesity. Counterintuitively, the majority of the reviewed studies showed that cannabis was associated with a lower BMI or a lower prevalence of obesity, or both (Hayatbakhsh et al., 2010; Le Strat and Le Foll, 2011; Smit and Crespo, 2001; Warren et al., 2005), or to have no association with BMI or obesity (Rodondi et al., 2006).
Because of the significance of diabetes as a highly prevalent disease, as a risk factor for cardiovascular diseases, and as a significant economic burden in our society, the question of whether cannabis use is associated with increased risk of diabetes is important. Included in this review are the assessments of three studies of cannabis use and metabolic dysregulation/metabolic syndrome, one study of cannabis use and prediabetes, and three studies of cannabis use and diabetes.
Is There an Association Between Cannabis Use and Metabolic Dysregulation, Metabolic Syndrome, Prediabetes, or Diabetes Mellitus?
The committee did not identify a good- or fair-quality systematic review that reported on the association between cannabis and metabolic dysregulation, metabolic syndrome, prediabetes, or diabetes mellitus. A review by Sidney (2016), published after the cutoff date for literature considered in this report, informed the discussion regarding the studies described in this section.
Metabolic Dysregulation and Metabolic Syndrome Three cross-sectional studies were conducted using data from the National Health and Nutrition Examination Survey (NHANES) to examine the associations between cannabis use and glucose, insulin, and insulin resistance (Penner et al., 2013); cannabis use and the metabolic syndrome (Vidot et al., 2016); and cannabis use and tobacco cigarette smoking with metabolic syndrome (Yankey et al., 2016).
The study by Penner et al. (2013) included 4,657 NHANES participants from three exams conducted from 2005 to 2010 who were categorized as current, former, or never users of cannabis. The fasting mean glucose levels were not found to be significantly different in current users than in never users according to multivariable analyses that adjusted for age, sex, race/ethnicity, income, marital status, tobacco use, alcohol use,
BMI, and physical activity. Hemoglobin A1c did not vary by cannabis use status, while fasting insulin and homeostasis models of insulin resistance (HOMA-IR) were about 12 percent lower in current cannabis users than in never users. A study by Vidot et al. (2016) of 8,478 NHANES participants from three exams conducted from 2005 to 2010 found that the odds of metabolic syndrome were lower in current users than in never users, with an OR of 0.69 (95% CI = 0.47–1.00) according to a multivariable analysis that adjusted for age, sex, race/ethnicity, poverty-to-income ratio, tobacco smoking, and exam cycle year. Yankey et al. (2016) studied the association between cannabis and cigarette smoking with the prevalence of metabolic syndrome, using data from 3,051 2011–2012 NHANES participants. Compared with findings from respondents who reported never having used cannabis, regular use of cannabis (defined as smoking cannabis or hashish at least once per month for more than 1 year) was associated with reduced odds for metabolic syndrome (OR, 0.23; 95% CI = 0.06–0.90). The multivariable analysis controlled for age, education, family-income-to-poverty ratio, sex, medical insurance, marital status, tobacco smoking, physical activity, other drug use, and rehabilitation.
Prediabetes Bancks et al. (2015) examined the association of self-reported cannabis use with both the prevalence and the incidence of prediabetes in the Coronary Artery Risk Development in Young Adults (CARDIA) study. A cross-sectional analysis for diabetes was conducted in 3,024 participants at the Year 25 exam. Cannabis use was assessed by self-administered questions. Prediabetes was defined according to American Diabetes Association criteria and was present in 45 percent of participants. Relative to never use, the current use of cannabis was associated with an OR for prediabetes of 1.65 (95% CI = 1.15–2.38), and lifetime cannabis use of at least 100 times was associated with an OR of 1.49 (95% CI = 1.06–2.11). The multivariable analysis adjusted for age, sex, race, tobacco smoking, alcohol use, education, field center, systolic blood pressure, C-reactive protein (CRP), physical activity, and the use of other illicit drugs. The CARDIA longitudinal analysis examined the association of self-reported cannabis use at the Year 7 follow-up exam to incident prediabetes (51 percent of participants) at the four subsequent follow-up examinations, with an average of 13.8 years of follow-up. The adjusted hazard ratio (HR) for prediabetes associated with lifetime use of at least 100 times relative to never use of cannabis was 1.39 (95% CI = 1.13–1.71).
Diabetes Bancks et al. (2015) also examined the association of self-reported cannabis use and diabetes in both cross-sectional and longitudinal analyses conducted in the CARDIA study. The study population was the same for the cross-sectional analysis, and the adjustment variables
were the same as described for the prediabetes analysis. Diabetes was present in 11.8 percent of Year 25 exam participants. The ORs for diabetes were 1.18 (95% CI = 0.67–2.10) for current use and 1.42 (95% CI = 0.85–2.38) for lifetime use of at least 100 times relative to never use of cannabis. The longitudinal analysis examined the association between Year 7 exam and self-reported cannabis use to incident diabetes (11.1 percent of participants) at the four subsequent follow-up examinations (years 10, 15, 20, and 25). Relative to never use, the HR associated with diabetes for lifetime use of at least 100 times was 1.10 (95% CI = 0.74–1.64), adjusted for the same variables as the longitudinal analysis of prediabetes.
Two cross-sectional studies were conducted using data from the NHANES to examine the association of cannabis use with diabetes (Alshaarawy and Anthony, 2015; Rajavashisth et al., 2012). The first study (Rajavashisth et al., 2012) used interviewer-administered data regarding cannabis use and diabetes collected from 10,896 adults, ages 20 to 29 years, during NHANES III, conducted from 1988 to 1994. Relative to nonusers, the OR for diabetes associated with current and past cannabis use was 0.36 (95% CI = 0.24–0.55), adjusted for race/ethnicity, physical activity, alcohol use, alcohol × cannabis use interaction, BMI, total cholesterol, triglyceride, CRP, and hypertension.
In the second study, Alshaarawy and Anthony (2015) examined the association of cannabis use with diabetes in eight different replication samples and in a meta-analysis. The samples were obtained from four NHANES surveys (2005–2006, 2007–2008, 2009–2010, 2011–2012) and from a survey performed for the National Survey on Drug Use and Health (NSDUH) during the same time periods. A composite indicator of diabetes from the NHANES data combined interview reports of diabetes, current use of insulin and/or oral hypoglycemic medication, and lab-derived glycohemoglobin. Self-report of cannabis was assessed from the NSDUH surveys. Compared to nonusers, the adjusted odds ratios (aORs) for diabetes associated with current cannabis use ranged from 0.4 to 0.9, with a meta-analytic OR summary of 0.7 (95% CI = 0.6–0.8). Meta-analytic summary analyses performed within cigarette smoking strata found aORs were 0.8 (95% CI = 0.5–1.2) in respondents who reported never having smoked cigarettes and 0.8 (95% CI = 0.6–1.0) in current smokers.
Discussion of Findings
Overall, the articles reviewed by the committee were judged to be of good to fair quality for assessing the risk of metabolic dysregulation, metabolic syndrome, prediabetes, or diabetes mellitus associated with cannabis use. In their review of the evidence, the committee found that cannabis use had either an inverse association or no association with
BMI, an inverse association with metabolic dysregulation and metabolic syndrome, and an inverse association or no association with diabetes mellitus. The only study showing an increased risk was the prediabetes portion of the CARDIA study analysis.
As noted earlier, these are counterintuitive findings because THC tends to stimulate appetite, promote fat deposition, and promote adipogenesis. Potential explanations include the following:
- Cross-sectional studies do not allow one to assess temporality between exposure and outcome. With the exception of the longitudinal findings reported in the CARDIA study, all of the reported findings were from cross-sectional analyses.
- Dose estimates of cannabis exposure were generally imprecise and lacking information on cannabis strength, dose, frequency of use, and duration of use, although this may be because the cumulative dose for most cannabis users is not high enough to affect fat and glucose-insulin metabolism. Statistical confounders may exist in these studies which are not adequately controlled for by standard multivariable modeling. For example, in general, high levels of cannabis use are strongly associated with younger age, which is inversely associated with the incidence and prevalence of diabetes. They are also associated with tobacco cigarette smoking, a known risk factor for diabetes (Willi et al., 2007). Cannabis use was associated with increases in physical activity in the CARDIA study (Bancks et al., 2015) and in one of the NHANES studies (Rajavashisth et al., 2012). Physical activity is protective against obesity and diabetes.
- Reverse causality might result in a chronic illness such as diabetes leading to the cessation of potentially unhealthy habits, including cannabis use. This might help to explain why cannabis use is associated with prediabetes but not with diabetes.
6-3(a) There is limited evidence of a statistical association between cannabis use and decreased risk of metabolic syndrome and diabetes.
6-3(b) There is limited evidence of a statistical association between cannabis use and increased risk of prediabetes.
The major gaps and opportunities relate to the paucity of longitudinal data for all of the cardiometabolic disorders and to the lack of data on the impact of cannabis use on risk in the older-adults age groups in which the majority of cardiovascular endpoints (e.g., acute myocardial infarction, stroke) occur. To address research gaps the committee suggests the following:
- Establishing a population cohort(s) in which cannabis use is regularly evaluated with standardized questionnaires accounting for the type of preparation, THC/other cannabinoid strength, the amount smoked or consumed, assessment of frequency and duration of use, and other cardiovascular disease (CVD) risk data, and in which researchers collect medical record and toxicology data or other biological marker data for cannabis use on incident CVD events.
- The cohort(s) need to be large enough that the association of cannabis with CVD events in the presence of potential statistical confounding variables (e.g., tobacco use, physical activity) can be validly assessed.
- Promote the collection of cannabis use data in electronic health records.
An additional suggestion is that basic research needs to be carried out to better understand the mechanisms for the role of cannabis as a possible trigger of AMI.
This chapter summarizes the good and fair cardiometabolic literature published since 1999. The committee found limited evidence of an association between acute cannabis use—but not chronic cannabis use—and AMI risk. The committee also determined that there is limited evidence of an association between cannabis use and an increased risk of ischemic stroke or subarachnoid hemorrhage and also prediabetes and an association between cannabis and a decreased risk of metabolic dysregulation, metabolic syndrome, and diabetes. The limitations of the reviewed studies include a lack of information on different routes of cannabis administration (e.g, smoked, edible, etc.), a lack of adequate dose information, insufficient information on potential additives or contaminants, and inadequate data on total lifetime duration/dose of cannabis use. The committee has formed a number of research conclusions related to these health endpoints; however, it is important that each of these conclusions be
interpreted within the context of the limitations discussed in the Discussion of Findings sections. Box 6-1 contains a summary of the conclusions for this chapter.
Alshaarawy, O., and J. C. Anthony. 2015. Cannabis smoking and diabetes mellitus: Results from meta-analysis with eight independent replication samples. Epidemiology 26(4):597–600.
Aronow, W. S., and J. Cassidy. 1974. Effect of marihuana and placebo-marihuana smoking on angina pectoris. New England Journal of Medicine 291(2):65–67.
Azofeifa, A., M. E. Mattson, G. Schauer, T. McAfee, A. Grant, and R. Lyerla. 2016. National estimates of marijuana use and related indicators—National Survey on Drug Use and Health, United States, 2002–2014. Morbidity and Mortality Weekly Report Surveillance Summaries 65(11):1–25.
Bancks, M. P., M. J. Pletcher, S. G. Kertesz, S. Sidney, J. S. Rana, and P. J. Schreiner. 2015. Marijuana use and risk of prediabetes and diabetes by middle adulthood: The Coronary Artery Risk Development in Young Adults (CARDIA) study. Diabetologia 58(12):2736–2744.
Barber, P. A., H. M. Pridmore, V. Krishnamurthy, S. Roberts, D. A. Spriggs, K. N. Carter, and N. E. Anderson. 2013. Cannabis, ischemic stroke, and transient ischemic attack: A case-control study. Stroke 44(8):2327–2329.
Beaconsfield, P., J. Ginsburg, and R. Rainsbury. 1972. Marihuana smoking. Cardiovascular effects in man and possible mechanisms. New England Journal of Medicine 287(5):209–212.
Benowitz, N. L., and R. T. Jones. 1981. Cardiovascular and metabolic considerations in prolonged cannabinoid administration in man. Journal of Clinical Pharmacology 21(8–9 Suppl):214S–223S.
CDC (Centers for Disease Control and Prevention). 2014a. National diabetes statistics report: Estimates of diabetes and its burden in the United States, 2014. Atlanta, GA: U.S. Department of Health and Human Services.
CDC. 2014b. Division of Diabetes Translation: National Diabetes Surveillance System. Long-term trends in diabetes. https://www.cdc.gov/diabetes/statistics/slides/long_term_trends.pdf. (accessed October 25, 2016).
CDC. 2015. Annual number (in thousands) of new cases of diagnosed diabetes among adults aged 18–79 years, United States, 1980–2014. http://www.cdc.gov/diabetes/statistics/incidence/fig1.htm (accessed October 25, 2016).
Di Marzo, V., F. Piscitelli, and R. Mechoulam. 2011. Cannabinoids and endocannabinoids in metabolic disorders with focus on diabetes. Handbook of Experimental Pharmacology 203:75–104.
Flegal, K. M., M. D. Carroll, C. L. Ogden, and C. L. Johnson. 2002. Prevalence and trends in obesity among U.S. adults, 1999–2000. JAMA 288(14):1723–1727.
Franz, C. A., and W. H. Frishman. 2016. Marijuana use and cardiovascular disease. Cardiology in Review 24(4):158–162.
Gorelick, D. A., R. S. Goodwin, E. Schwilke, D. M. Schwope, W. D. Darwin, D. L. Kelly, R. P. McMahon, F. Liu, C. Ortemann-Renon, D. Bonnet, and M. A. Huestis. 2013. Tolerance to effects of high-dose oral D9-tetrahydrocannabinol and plasma cannabinoid concentrations in male daily cannabis smokers. Journal of Analytical Toxicology 37(1):11–16.
Hayatbakhsh, M. R., M. J. O’Callaghan, A. A. Mamun, G. M. Williams, A. Clavarino, and J. M. Najman. 2010. Cannabis use and obesity and young adults. American Journal of Drug and Alcohol Abuse 36(6):350–356.
Herning, R. I., W. E. Better, K. Tate, and J. L. Cadet. 2001. Marijuana abusers are at increased risk for stroke. Preliminary evidence from cerebrovascular perfusion data. Annals of the New York Academy of Sciences 939:413–415.
IOM (Institute of Medicine). 1999. Marijuana and medicine: Assessing the science base. Washington, DC: National Academy Press.
Jamal, A., D. M. Homa, E. O’Connor, S. D. Babb, R. S. Caraballo, T. Singh, S. S. Hu, and B. A. King. 2015. Current Cigarette Smoking Among Adults—United States, 2005–2014. Morbidity and Mortality Weekly Report 64(44):1233–1240.
Khiabani, H. Z., J. Mørland, and J. G. Bramness. 2008. Frequency and irregularity of heart rate in drivers suspected of driving under the influence of cannabis. European Journal of Internal Medicine 19:608–612.
Klil-Drori, A. J., L. Azoulay, and M. N. Pollak. 2016. Cancer, obesity, diabetes, and antidiabetic drugs: Is the fog clearing? Nature Reviews: Clinical Oncology August. doi:10.1038/ nrclinonc.2016.120.
Kochanek, K. D., S. L. Murphy, J. Q. Xu, and B. Tejada-Vera. 2016. Deaths: Final data for 2014. National Vital Statistics Reports 65(4):1–121. Hyattsville, MD: National Center for Health Statistics.
Le Strat, Y., and B. Le Foll. 2011. Obesity and cannabis use: Results from 2 representative national surveys. American Journal of Epidemiology 174(8):929–933.
Mittleman, M. A., R. A. Lewis, M. Maclure, J. B. Sherwood, and J. E. Muller. 2001. Triggering myocardial infarction by marijuana. Circulation 103(23):2805–2809.
Mozaffarian, D., E. J. Benjamin, A. S. Go, D. K. Arnett, M. J. Blaha, M. Cushman, S. R. Das, S. de Ferranti, J. P. Després, H. J. Fullerton, V. J. Howard, M. D. Huffman, C. R. Isasi, M. C. Jiménez, S. E. Judd, B. M. Kissela, J. H. Lichtman, L. D. Lisabeth, S. Liu, R. H. Mackey, D. J. Magid, D. K. McGuire, E. R. Mohler, 3rd, C. S. Moy, P. Muntner, M. E. Mussolino, K. Nasir, R. W. Neumar, G. Nichol, L. Palaniappan, D. K. Pandey, M. J. Reeves, C. J. Rodriguez, W. Rosamond, P. D. Sorlie, J. Stein, A. Towfighi, T. N. Turan, S. S. Virani, D. Woo, R. W. Yeh, M. B. Turner; American Heart Association Statistics Committee, and Stroke Statistics Subcommittee. 2016. Heart disease and stroke statistics—2016 update: A report from the American Heart Association. Circulation 133(4):e38–e360.
Ogden, C. L., M. D. Carroll, B. K. Kit, and K. M. Flegal. 2014. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA 311(8):806–814.
Penner, E. A., H. Buettner, and M. A. Mittleman. 2013. The impact of marijuana use on glucose, insulin, and insulin resistance among U.S. adults. American Journal of Medicine 126(7):583–589.
Phillips, M. C., J. M. Leyden, W. K. Chong, T. Kleinig, P. Czapran, A. Lee, S. A. Koblar, J. Jannes. 2011. Ischaemic stroke among young people aged 15 to 50 years in Adelaide, South Australia. Medical Journal of Australia 195(10):610–614.
Rajavashisth, T. B., M. Shaheen, K. C. Norris, D. Pan, S. K. Sinha, J. Ortega, and T. C. Friedman. 2012. Decreased prevalence of diabetes in marijuana users: Cross-sectional data from the National Health and Nutrition Examination Survey (NHANES) III. BMJ Open 2:e000494.
Renaud, A. M., and Y. Cormier. 1986. Acute effects of marihuana smoking on maximal exercise performance. Medicine and Science in Sports and Exercise 18(6):685–689.
Rodondi, N., M. J. Pletcher, K. Liu, S. B. Hulley, S. Sidney, and Coronary Artery Risk Development in Young Adults (CARDIA) Study. 2006. Marijuana use, diet, body mass index, and cardiovascular risk factors (from the CARDIA study). American Journal of Cardiology 98(4):478–484.
Rumalla, K., A. Y. Reddy, and M. K. Mittal. 2016a. Recreational marijuana use and acute ischemic stroke: A population-based analysis of hospitalized patients in the United States. Journal of Neurological Sciences 364:191–196.
Rumalla, K., A. Y. Reddy, and M. K. Mittal. 2016b. Association of recreational marijuana use with aneurysmal subarachnoid hemorrhage. Journal of Stroke and Cerebrovascular Disease 25(2):452–460.
Sidney, S. 2002. Cardiovascular consequences of marijuana use. Journal of Clinical Pharmacology 42(11 Suppl):64S–70S.
Sidney, S. 2016. Marijuana use and type 2 diabetes mellitus: A review. Current Diabetes Reports 16(11):117.
Sidney, S., J. E. Beck, I. S. Tekawa, C. P. Quesenberry, and G. D. Friedman. 1997. Marijuana use and mortality. American Journal of Public Health 87(4):585–590.
Smit, E., and C. J. Crespo. 2001. Dietary intake and nutritional status of U.S. adult marijuana users: Results from the Third National Health and Nutrition Examination Survey. Public Health Nutrition 4(3):781–786.
Thomas, G., R. A. Kloner, and S. Rezkalla. 2014. Adverse cardiovascular, cerebrovascular, and peripheral vascular effects of marijuana inhalation: What cardiologists need to know. American Journal of Cardiology 113(1):187–190.
Vidot, D. C., G. Prado, W. M. Hlaing, H. J. Florez, K. L. Arheart, and S. E. Messiah. 2016. Metabolic syndrome among marijuana users in the United States: An analysis of National Health and Nutrition Examination survey data. American Journal of Medicine 129(2):173–179.
Wang, X., R. Derakhshandeh, J. Liu, S. Narayan, P. Nabavizadeh, S. Le, O. M. Danforth, K. Pinnamaneni, H. J. Rodriguez, E. Luu, R. E. Sievers, S. F. Schick, S. A. Glantz, and M. L. Springer. 2016. One minute of marijuana secondhand smoke exposure substantially impairs vascular endothelial function. Journal of the American Heart Association 5(8):e003858.
Warren, M., K. Frost-Pineda, and M. Gold. 2005. Body mass index and marijuana use. Journal of Addictive Diseases 24(3):95–100.
Westover, A. N., S. McBride, and R. W. Haley. 2007. Stroke in young adults who abuse amphetamines or cocaine: A population-based study of hospitalized patients. Archives of General Psychiatry 64(4):495–502.
Willi, C., P. Bodenmann, W. A. Ghali, P. D. Faris, and J. Cornuz. 2007. Active smoking and the risk of type 2 diabetes: A systematic review and meta-analysis. JAMA 298(22):2654–2664.
Wolff, V., V. Lauer, O. Rouyer, F. Sellal, N. Meyer, J. S. Raul, C. Sabourdy, F. Boujan, C. Jahn, R. Beaujeux, and C. Marescaux. 2011. Cannabis use, ischemic stroke, and multifocal intracranial vasoconstriction: A prospective study in 48 consecutive young patients. Stroke 42(6):1778–1780.
Wolff, V., J. P. Armspach, V. Lauer, O. Rouyer, M. Bataillard, C. Marescaux, and B. Gény. 2013. Cannabis-related stroke: Myth or reality? Stroke 44(2):558–563.
Wolff, V., J. P. Armspach, V. Lauer, O. Rouyer, A. Ducros, C. Marescaux, and B. Gény. 2015. Ischaemic strokes with reversible vasoconstriction and without thunderclap headache: A variant of the reversible cerebral vasoconstriction syndrome? Cerebrovascular Disease 39(1):31–38.
Yankey, B. N., S. Strasser, and I. S. Okosun. 2016. A cross-sectional analysis of the association between marijuana and cigarette smoking with metabolic syndrome among adults in the United States. Diabetes and Metabolic Syndrome 10(2 Suppl 1):S89–S95.
This page intentionally left blank.