Long Term Use of ADHD Medication and Cardiovascular Outcomes

 

 

I have several posts on this blog about prescribing ADHD medications with a goal of minimizing adverse psychiatric and medical side effects.  Like all medical treatments, close follow-up and monitoring is required to assure efficacy while reducing the risk of adverse effects.  To a trained physician it does not take much effort other than being rigorous in examinations and discussions with patients. In the area of ADHD, there is the frequent assumption that patients are young, healthy, and can probably tolerate medications better than older populations. With the increasing diagnosis of adult ADHD, all the comorbidities need to be carefully addressed and a recommendation of no treatment also needs to be considered.

Who have I advised against treatment? Older adults with obvious cardiovascular problems that are inadequately treated or controlled who may or may not have ADHD.  I do not really care if you are 60 years old and I think you really have ADHD, I am not going to start treatment if your blood pressure is not in good control or if you have other unstable conditions like angina, congestive heart failure, cardiomyopathy, or arrhythmias.

The commonest reason for not treating people was hypertension, measured by me in the office.  In some cases, there was an abnormal ECG showing a previously unknown arrhythmia. It can be difficult to tell a patient that you will not treat them because of a medical condition – but that is just the way it is.  Even if treatment is started – blood pressure monitoring needs to occur at every visit.  In some cases, I recommend that the patient purchase a home blood pressure monitor and send me the results.  Referring the patient to their primary care physician or cardiologist is useful to let that physician know that their patient wants stimulant treatment and provide feedback on what your assessment of their cardiac status was.  It is common for physicians prescribing adequate does of antihypertensives to not know that their patient is still hypertensive. It is always clear that the decision to prescribe stimulants is made by me and does not depend on the opinion of another physician.

White coat hypertension (WCH) is not an exception.  WCH is the idea that people get hypertensive related to the stress of being in physician’s office.  Conventional wisdom was that resolved when the patient left the office and therefore this was a being condition. The problem with that assessment is that it depends on knowing that the blood pressure did normalize away from the office.  That lead to a more modern definition that required ambulatory blood pressure measurements away from the office and subsequent more detailed definitions. As an example, the European Society of Hypertension recommends the following:  subjects with office systolic/diastolic blood pressure readings of ≥140/90 mm Hg and a 24-hour blood pressure <130/80 mm Hg.

In the most recent review, the authors do an excellent job pointing out some of the flaws in the early research that led to no significant differences between subjects with WCH and controls.  The control subjects often had cardiovascular disease or were treated with antihypertensives.  They also make the distinction between white coat effect (WCE) and white coat hypertension (WCH). WCE is defined as “an alerting reaction working through reflex activation of the sympathetic nervous system.”  A standard research technique to assess stress effects on blood pressure is to ask subjects to do mental arithmetic and it generally leads to a blood pressure effect like what the authors describe in this paper of 20 mm systolic or 10 mm diastolic.  The authors provide guidance on differentiating the various combinations of white coat hypertension and hypertension as well as providing guidance for future research.  In the office for the purpose of prescribing stimulants the key question is whether there is a white coat effect, white coat hypertension, and whether it occurs in the context of treated or untreated hypertension. Short of ambulatory blood pressure measurements other sources can provide some additional guidance.  Access to the electronic health record can show long term trends.  If indicated - I would not hesitate to suggest that the patient consult with their primary care physicians or hypertension specialist for ambulatory BP measurement.         

Studies have shown that patients who continue to exhibit a reactive blood pressure problem at home have similar cardiovascular risks to hypertensive individuals.  On the flip side, I have assessed many distressed patients in inpatient settings who were normotensive.  Based on this experience, I do not dismiss elevated blood pressure readings in the office especially if I am going to prescribe a medication that may elevate blood pressure.  

That brings me to the paper that led me to write this post (1).  This is a nested case control study of registry data in Sweden that looked at 10,388 cases of ADHD and 51,672 matched controls (aged 6-64 years old).  Exclusion criteria included pre-existing cardiovascular disease, previous use of ADHD medication, and emigration or death before baseline (defined as day of first ADHD medication or diagnosis – whichever came first).  This study design basically looks at the defined illness (in this case ADHD) and then matches the selected cases to controls from the same cohort – in this case up to 5 controls without known cardiovascular disease.  The exposure in this case was ADHD medications including study period, including methylphenidate, amphetamine] dexamphetamine lisdexamfetamine, atomoxetine, and guanfacine.  The last two medications are nonstimulants and guanfacine has also been used as an antihypertensive medication.  The cardiovascular outcomes included: Ischemic heart disease, cerebrovascular disease, hypertension, heart failure, arrhythmias, thromboembolic disease, and arterial disease.  The statistics of interest were adjusted odds rations comparing cases to controls.  The authors also did a brief literature review in both the introduction and discussion sections of the existing literature in this area and what can be described as mixed results.

Their main finding was that only two cardiovascular conditions – arterial disease and hypertension were significantly associated with stimulant medication use but not with atomoxetine or lisdexamfetamine use.  Risk also increased at a level of 1.5 DDD (defined daily doses) of stimulant medication.  Those specific doses except for guanfacine can be found at this link. 

The authors do a good job of interpreting the limitations of their data including the possibilities of under detection of the true rate of cardiovascular disease at baseline, the possibility of mediation nonadherence and underestimating the effects of medication exposure, and confounding by severity could be an issue through the effect for more severe ADHD on lifestyle factors important in the genesis of cardiovascular disease (CVD). Finally, since the study eliminated subjects with existing CVD – stimulant exposure was not measured at all in that population. The authors advise very cautious treatment and monitoring of those individuals.

All things considered, this was a good approach to studying the effects of ADHD medication exposure and the development of cardiovascular disease on a significant sample.  It was a convenience sample from a pre-existing registry.  The authors point out that some treatment groups were very small and advocated for a similar study with a larger N.   Just looking at the trends in their tables, there is clearly significant cardiovascular disease in both the test and control subjects.  The odds ratios for medication exposure were low when they were significant.  Few medical variables were controlled for (obesity, Type 2 diabetes mellitus, dyslipidemia, and sleep disorders) and of those 3 out of 4 are more common in patient with ADHD (3).  More subtle forms of effects from ADHD like whether there are affective changes (typically irritability and anger) leading to hypertension or a white coat effect are unknown currently.

That leads me back to the need for close monitoring for cardiovascular risk factors and conditions before any medication is considered. The group with pre-existing cardiovascular disease is at highest risk and they have not been studied. My speculation is that even using a large health plan database those numbers (patients with cardiovascular disease started on ADHD medication) will be small.  Any real world clinical scenario where this is being considered should be approached as cautiously as possible and monitored the same way.   

 

George Dawson, MD, DFAPA

Photo Credit:  Sgt. 1st Class Shane Klestinski, Public domain, via Wikimedia Commons.  For full details click on the photo to see Wikimedia Commons page.  I chose this photo because several adults that I diagnosed with ADHD told me that they had adapted to work in warehouse management and logistics - in many cases that involved driving fork lifts. 

References:

  1:  Zhang L, Li L, Andell P, Garcia-Argibay M, Quinn PD, D'Onofrio BM, Brikell I, Kuja-Halkola R, Lichtenstein P, Johnell K, Larsson H, Chang Z. Attention-Deficit/Hyperactivity Disorder Medications and Long-Term Risk of Cardiovascular Diseases. JAMA Psychiatry. 2024 Feb 1;81(2):178-187. doi: 10.1001/jamapsychiatry.2023.4294. PMID: 37991787; PMCID: PMC10851097.

2:  Franklin SS, Thijs L, Hansen TW, O'Brien E, Staessen JA. White-coat hypertension: new insights from recent studies. Hypertension. 2013 Dec;62(6):982-7. doi: 10.1161/HYPERTENSIONAHA.113.01275. Epub 2013 Sep 16. PMID: 24041952.

3:  Chen Q, Hartman CA, Haavik J, Harro J, Klungsøyr K, Hegvik TA, Wanders R, Ottosen C, Dalsgaard S, Faraone SV, Larsson H. Common psychiatric and metabolic comorbidity of adult attention-deficit/hyperactivity disorder: A population-based cross-sectional study. PLoS One. 2018 Sep 26;13(9):e0204516. doi: 10.1371/journal.pone.0204516. PMID: 30256837; PMCID: PMC6157884.

4:  Fuemmeler BF, Østbye T, Yang C, McClernon FJ, Kollins SH. Association between attention-deficit/hyperactivity disorder symptoms and obesity and hypertension in early adulthood: a population-based study. Int J Obes (Lond). 2011 Jun;35(6):852-62. doi: 10.1038/ijo.2010.214. Epub 2010 Oct 26. PMID: 20975727; PMCID: PMC3391591.