Is ADHD Medication Safe for a Child's Heart? What the Studies Counted

In a US study of 1,200,438 children and young adults aged 2 to 24, there were 81 serious cardiovascular events, or 3.1 per 100,000 person-years. Current users of ADHD medication had an adjusted rate that wasn't statistically significantly different from that of nonusers (hazard ratio 0.75; 95% confidence interval 0.31 to 1.85).¹

The interval means the study couldn't rule out a rate up to 85% higher in current users, or one as much as 69% lower, and the events themselves were rare.¹

Serious heart events were rare in the cohort studies

The 81 events were 33 sudden cardiac deaths (1.3 per 100,000 person-years), 9 heart attacks (0.3), and 39 strokes (1.5). There was no evidence of increased risk for any of the three, or for methylphenidate specifically (hazard ratio 0.96; 95% CI 0.31 to 2.97). Children with congenital heart disease were included.¹

A second US study identified 241,417 children aged 3 to 17 who were new users of ADHD medication. The rate of validated sudden death or ventricular arrhythmia didn't differ significantly between users and nonusers (hazard ratio 1.60; 95% CI 0.19 to 13.60), and neither did death from any cause (0.76; 0.52 to 1.12).³

In a 2007 study of 55,383 young people with ADHD, no cardiac death occurred during 42,612 person-years of stimulant use. Stimulant use was associated with a 20% increase in the hazard of emergency department visits for cardiac causes or symptoms (adjusted hazard ratio 1.20; 95% CI 1.04 to 1.38).⁴

A Danish registry study used a broader outcome. Among 8,300 children with ADHD, there were 111 cardiovascular events (170 per 100,000 person-years), and stimulant use was associated with a higher hazard (adjusted hazard ratio 2.34; 95% CI 1.15 to 4.75). That figure is from the paper's Results section; its abstract prints a different one. The events were cardiovascular diagnoses of any kind, including arrhythmias (23%) and hypertension (8%).⁵

One study found an association with sudden unexplained death

Gould and colleagues used state mortality data from 1985 to 1996 to identify 564 sudden deaths at ages 7 through 19, and matched them to 564 young people who died as passengers in motor vehicle accidents. Stimulant use was found in 10 of the sudden unexplained deaths (1.8%) and in 2 of the comparison group (0.4%). The odds ratio was 7.4, with a 95% confidence interval of 1.4 to 74.9.²

The association was statistically significant, and it rests on 12 young people with stimulant exposure in total, and exposure was established from informant reports, medical examiner records, toxicology results, or death certificates. The authors wrote that "this finding should be considered in the context of other data about the risk and benefit of stimulants in medical treatment."²

The two US cohort studies published in 2011 did not find a statistically significant increase in sudden death among users, and their intervals leave room for a relative increase.^{1, 3}

Blood pressure and heart rate over 10 years

The MTA study randomized 579 children aged 7 to 9 to 14 months of behavior therapy, intensive medication treatment, the combination, or community management, then observed them for 10 years. No treatment effect on systolic or diastolic blood pressure could be detected, and there was no evidence that stimulant treatment increased the risk of prehypertension or hypertension. Stimulant medication did not increase the risk of tachycardia, although greater cumulative stimulant exposure was associated with higher heart rate at years 3 and 8.⁶

The ADDUCE study included 1,410 participants, 756 of them with ADHD and taking methylphenidate. After 24 months, pulse rate and systolic and diastolic blood pressure had increased by a statistically significant amount compared with participants with ADHD who weren't taking it. The authors described the changes as relatively small and as requiring regular monitoring.⁷

A 2011 review reported that stimulants as a class were associated with mean elevations of up to 5 mm Hg in blood pressure and up to 10 beats per minute in heart rate in healthy children and adolescents, without changes in ECG measurements.⁸ The class includes Adderall XR (an amphetamine product), Vyvanse (a prodrug of amphetamine), and Concerta (methylphenidate).^{9, 10, 11} I've written separately about how blood pressure and pulse are monitored.

Does a child need an ECG before starting ADHD medication?

The Adderall XR and Concerta labels tell the prescriber to assess for cardiac disease before treatment with a careful history, a family history of sudden death or ventricular arrhythmia, and a physical exam. Neither label's text mentions a routine ECG.^{9, 11} The atomoxetine (Strattera) label asks for the same history and exam, and says patients "should receive further cardiac evaluation if findings suggest such disease (e.g., electrocardiogram, echocardiogram)."¹²

Of 1,470 children with ADHD who had screening ECGs, 119 (8.1%) had an ECG read as abnormal, and cardiac disease was identified in 5 (0.3% of the cohort), a positive predictive value of 4.2%.¹³ In a second study, 372 ECGs were ordered for stimulant screening in the year after a 2008 American Heart Association statement. Twenty-four (6.4%) had abnormal findings, 18 patients were referred for further evaluation, and none had been found to have definitive disease at last follow-up.¹⁴

According to a 2010 review by Stiefel and Besag, the American Heart Association states that an ECG is reasonable to consider and not mandatory, and the American Academy of Pediatrics does not recommend routine ECGs before starting stimulant therapy.¹⁵

Limitations of the evidence

This evidence has limitations. First, the MTA's randomized phase lasted 14 months and the event studies are observational, so they cannot establish cause. Second, serious events were rare enough that the confidence intervals are wide, with upper limits of 1.85 in the first cohort and 13.60 in the second. Third, the 10-year follow-up began at ages 7 to 9, so these studies don't address heart disease in middle age. Nevertheless, the absolute rate in the 1.2 million cohort was 3.1 serious events per 100,000 person-years, among users and nonusers combined.^{1, 3, 6}

What to tell the prescriber before starting

The Adderall XR Medication Guide asks families to tell the prescriber, before treatment, about heart problems, heart disease, heart defects, or high blood pressure, and about a family history of sudden death or heart problems. The label names exertional chest pain and unexplained fainting as symptoms that suggest cardiac disease.⁹

During treatment, call 911 or go to an emergency department for chest pain, fainting, or shortness of breath. Other concerns go to the prescriber, who decides with the family whether anything about the medication should change.

Chest pain, fainting, or shortness of breath during treatment needs immediate medical assessment. Call 911 or go to the nearest emergency department for severe symptoms. Medication changes are made with the prescriber.

Related: Effects of ADHD medication on the heart · The long-term heart risk study · ADHD medication with high blood pressure · Heart palpitations · Heart attack and stroke · Adderall vs Ritalin · Raynaud's and cold hands