Article Text
PDF
Abstract
Methylphenidate and lisdexamfetamine are recommended as first-line pharmacological treatment options for adults with attention deficit hyperactivity disorder (ADHD). Formulations of methylphenidate can generally be classified into three groups according to their duration of action: one group lasts 12 hours, another group lasts 8 hours and the immediate-release group lasts 3–4 hours. Patients are usually able to substitute brands with one of the equivalent release profiles without significant problems. Lisdexamfetamine is a prodrug which has a slow onset and long duration (approximately 12 hours), ensuring minimal potential for abuse compared with its active metabolite dexamfetamine. Second-line treatments such as atomoxetine are also available for those who cannot tolerate or do not respond to methylphenidate or lisdexamfetamine. In the UK, ADHD has been previously managed largely in tertiary clinics, but many cases could be managed by appropriately trained clinicians in secondary or primary care (as already happens in some countries), with great benefit for patients and job satisfaction for clinicians.
- Drug-Related Side Effects and Adverse Reactions
- Health Care Quality, Access, and Evaluation
Statistics from Altmetric.com
Key learning points
Choice of medication for attention deficit hyperactivity disorder (ADHD) in adults should be tailored to the individual with regard to the intended duration of action and relevance of comorbidity.
Optimal dosage may vary unpredictably between individuals, requiring careful titration for best ADHD symptom control with minimum adverse effects.
Most people can be managed effectively after a few dose titration steps.
For those who do not benefit from the first medication that they try, other pharmacological treatment options are often effective.
We understand how to treat ADHD, but how to deliver such care at scale remains a challenge, particularly in the UK.
Introduction
Part 1 of this two-part series considered diagnostic criteria,1 non-pharmacological interventions and provided an overview of pharmacological management for adults with attention deficict hyperactivity disorder (ADHD). In this article, I consider medication in more detail.
First-line treatment: stimulants
The National Institute for Health and Care Excellence (NICE) guideline (NG87) and the more recent Australian clinical practice guideline recommend methylphenidate and lisdexamfetamine as first-line pharmacological treatment options for ADHD in adults.2–4
A network meta-analysis (51 studies, 8131 participants) published since the NICE guideline shows that stimulants are more effective than non-stimulants and are generally well tolerated5; also, amfetamines (including mixed amfetamine salts such as Adderall, which are not available in the UK) seem slightly more effective than methylphenidate.
A more recent network meta-analysis compared pharmacological, psychological and neurostimulatory interventions for ADHD in adults.6 The primary outcomes were the severity of ADHD core symptoms according to self-rated and clinician-rated scales at timepoints closest to 12 weeks, and acceptability based on all-cause discontinuation rate. The authors included 113 randomised controlled trials (RCTs) with a total of 14 887 participants who had a median age of 36 years. Only stimulants and atomoxetine had evidence of short-term improvement in ADHD core symptoms based on self-reported and clinician-reported ratings. Atomoxetine was less well tolerated than placebo. ADHD medications were not shown to improve other outcomes, such as quality of life, and there is a lack of RCT evidence on long-term effects. The efficacy of non-pharmacological interventions varied according to who was rating the symptoms.6 A commentary published alongside the network meta-analysis recognised some of its limitations.7 In particular, it noted that there can be difficulties in comparing pharmacological and psychological interventions for ADHD because trials often differ in blinding, treatment duration and intensity. In addition, non-pharmacological interventions may not improve ADHD symptoms but may improve other aspects of functioning which were not part of the network meta-analysis’ primary outcomes.7
However, efficacy varies by patient: in cross-over RCTs published between 1974 and 1999 (10 studies, 222 participants), approximately 41% of participants had equally good responses to amphetamines and methylphenidate, 28% had a better response to amphetamines, 16% had a better response to methylphenidate and the rest did not respond to either.8 Differences in overall effectiveness of stimulants may be small enough that acquisition cost or stock availability become more significant factors in choosing a drug.
See table 1 for a summary of some of the medications used to manage ADHD.
Some medications used to manage ADHD13
Combining stimulants with non-stimulants is not generally recommended. There is limited evidence for adding methylphenidate to those with a partial response to atomoxetine.9 A Canadian Health Technology Review, largely based on a guideline from Australia, suggests ‘If stimulants are not tolerated or ineffective, non-stimulants, such as atomoxetine or guanfacine, should be offered as a second-line treatment. A combination of stimulants and non-stimulants may be prescribed to increase the benefit of the treatment. Other drugs such as bupropion, clonidine, modafinil, reboxetine and venlafaxine could be offered as third-line treatments.’10
It may be more important to understand the pattern of the patient’s daily activities (eg, shiftwork) and priorities (eg, concentrating in afternoon meetings) to select a stimulant with an appropriate duration of action (perhaps with twice daily dosing) for the individual. The choice of drug should be tailored to the patient, but even with the best planning, there will be trial and error as pharmacokinetics and clinical response can vary considerably between individuals.
The clinical effect of an increase in the dose of a stimulant should be apparent within days, and doses can be increased weekly, although those with a highly variable day-to-day life or significant comorbidity may need longer intervals to assess. More rapid titration may be acceptable in straightforward cases where there is urgency to reach treatment goals (eg, when there is a major work deadline).
Some patients seem to be ‘fast metabolisers’ and need higher doses or need to take extended-release versions twice a day because they wear off faster. If the patient has a partial response at the maximum licensed dose without adverse effects, higher unlicensed doses might be justified. Some prescribers may prefer to try a different stimulant at licensed doses, rather than prescribe one which is only effective at higher unlicensed doses.
Using two different stimulants in the same patient is not recommended in any guidelines and has not been studied, but some patients may be treated with a short-acting formulation at the time when a long-acting formulation tends to wear off.
If stimulants are ‘not working’, consider whether comorbidity is not being addressed or treatment goals are inappropriate (eg, oppositional behaviour may not respond well to stimulants); whether the medication is being taken appropriately, including optimal timing (simply increasing the dosage may not be helpful if a longer duration is needed); and whether there are extraneous factors affecting response. Sometimes too much stimulant can cause adverse effects (eg, agitation) which resemble ADHD—that is, the dose-response curve can be lambda-shaped. If higher doses are not better than the first dose, it may be worth retitrating from a lower starting dose.
Methylphenidate
Methylphenidate preparations have different durations of action. They can be divided into three groups:
Immediate release - lasting approximately 4 hours.
Prolonged release - lasting approximately 8 hours.
‘Extended’ release - lasting approximately 12 hours.
Confusingly, the abbreviation ‘XL’ is applied to formulations of prolonged-release and extended-release methylphenidate. Patients can be started on any of the prolonged-release or immediate-release preparations. Patients stabilised on immediate-release products can change to equivalent doses of prolonged-release products, if appropriate.
Immediate-release methylphenidate is rapidly absorbed with clinical effects less than 30 min after ingestion. Food does not reduce bioavailability of methylphenidate, but a high-fat meal may delay its absorption. Methylphenidate is metabolised in the liver by carboxylesterase (CES1A1), causing a significant ‘first pass’ effect, lowering bioavailability to approximately 22%. The clinical effect lasts approximately 4 hours, but individuals vary considerably due to differences in metabolism.11 Approximately 4% of white and black (but not Asian) populations have a mutation in CES1A1 which dramatically reduces their ability to metabolise methylphenidate, making them very sensitive to low doses: in such cases, very low doses such as 5 mg immediate-release methylphenidate may be effective; however, toxicity may mean that lisdexamfetamine should be tried instead. The manufacturers state that proton pump inhibitors, such as omeprazole, do not affect the absorption of extended-release methylphenidate. However, methylphenidate may auto-hydrolyse at pH 8, which may be found in the small intestine, especially in those taking antacid medication.12 This might explain why anecdotally some patients believe antacids make their stimulants less effective.
Immediate-release methylphenidate
Monotherapy with immediate-release methylphenidate allows flexible dosing—for example, in shift workers (and may be cheaper than slow-release preparations). However, it requires the patient to be organised enough to take medication several times during the day, even if away from home or work. Immediate-release methylphenidate may also be used to extend the effects of methylphenidate when slow-release preparations wear off.
Immediate-release methylphenidate is available as 5 mg, 10 mg and 20 mg tablets. Usually, patients start on 5 mg two times per day and increase by 5–10 mg per day at weekly intervals up to a maximum licensed dose of 60 mg per day (but 100 mg is offered as a maximum unlicensed dose by the British National Formulary [BNF], which may be taken as three divided doses).13
Delayed-release methylphenidate
The Medicines and Healthcare products Regulatory Agency (MHRA) advises caution if switching between long-acting methylphenidate products.14 It recommends consulting with the patient to discuss the reasons for switching between products and the possible changes they may experience in symptom management and adverse effects, and what to do if these occur. The patient should be given clear instructions on using the new formulation and whether it should be taken with or without food.14
12-hour-release methylphenidate
Extended-release preparations lasting approximately 12 hours are popular because they can provide sustained benefit over the day. Bioavailability of extended-release formulations is approximately 15%–20% lower (18 mg is comparable to 15 mg immediate release) as some drug remains in the capsule and passes through the small intestine. The usual starting dose is 18 mg once a day. If this has minimal effect, it may be increased to 36 mg after a few days, although usually dose increases are at weekly intervals or longer. If 36 mg is not sufficient, the dose can be increased to 54 mg. If 36 mg is too much, 27 mg can be tried. Doses above 54 mg may be required and the maximum licensed dose is 72 mg. The BNF suggests an unlicensed dose up to 108 mg per day can be used for Concerta XL.13 When assessing the effects of higher doses, or when dealing with more complex cases (eg, severe comorbidities causing week to week variation in function), it might be preferred to wait longer (eg, 4 weeks) before judging the effect at each dose.
Brands include Affenid XL (currently cheapest in the UK), Atenza XL, Concerta XL (currently most expensive), Delmosart, Kixel XL, Matoride XL, Xaggitin XL and Xenidate XL. Most brands have had supply problems with some dosages. The brands are effectively interchangeable (but see MHRA comment above), although the extended-release mechanism is different for Xenidate. Most brands contain 22% ‘immediate’-release methylphenidate and the remaining 78% is contained in an inner compartment which allows osmotic ingress of water, causing a slower release with a second peak at 6–8 hours after ingestion. Delmostart and Xaggitin have 25% immediate release and 75% extended release. Xenidate tablets have a shell which dissolves to allow some methylphenidate to be absorbed as immediate release, and the rest is in sugar spheres/beads which dissolve slowly to release the remainder, with a similar biphasic 12 hours plasma level profile. Xenidate, unlike the others, is scored and allows the tablet to be split (but not crushed) without affecting absorption.
8-hour-release methylphenidate
In the UK, modified-release preparations lasting approximately 8 hours include Addepta XL, Equasym XL, Exattent XL, Focusim XL, Medikinet XL, Meflynate XL, Metyrol XL and Ritalin XL. These preparations may be useful for patients who want a shorter duration of action than the longer acting extended-release preparations.
Preparations range from 10 mg to 60 mg; the licensed maximum dose is the lower of 80 mg/day or 1 mg/kg (for those <80 kg), but the BNF suggests a maximum unlicensed dose of 100 mg.13 Starting doses are from 10 mg in adults, increasing by 10–20 mg every week or so. The capsules may be opened and sprinkled on food (but not crushed or chewed). Bioavailability of these intermediate modified release formulations is insignificantly (approximately 10%) lower than immediate-release formulations. They contain differing proportions of immediate-release methylphenidate, so there are modest differences in duration and clinical effect. Equasym XL, Exattent XL and Addepta XL have a higher proportion (70%) of prolonged-release beads, whereas the others have 50%. They are not as easily interchangeable as the group lasting 12 hours, although supply shortages may force patients to change to the nearest alternative.
Swapping from a 12-hour preparation to an 8-hour preparation can be problematic; patients may need a lower initial dose, followed by a ‘top up’, to achieve similar effects (eg, a patient who was on Delmosart 54 mg may find 50 mg Medikinet XL too high a dose in the morning, but it may also not last long enough, whereas 40 mg Medikinet XL in the morning and 10 mg immediate-release methylphenidate taken 6–7 hours after the Medikinet XL may be better). Alternatively, a different preparation can be dose-titrated from scratch.
Lisdexamfetamine and dexamfetamine
Lisdexamfetamine is a prodrug that has no effect until it has been hydrolysed (in red cells) into dexamfetamine and lysine. This prolongs the duration of action to approximately 12 hours or more, facilitating once daily dosing. It is also less likely to be abused compared with dexamfetamine.15 Lisdexamfetamine is absorbed well and not affected by food. Contraindications, cautions and adverse effects are similar to methylphenidate. Lisdexamfetamine mesylate 30 mg contains the equivalent of 8.9 mg of dexamfetamine, but the different pharmacodynamics preclude simple substitution.
Preparation strengths range from 20 mg (for children) to a maximum recommended daily dose of 70 mg (or 50 mg in severe renal impairment).13 Capsules can be opened, and the contents dissolved in water. When a particular strength of capsule is out of stock, the equivalent dose can be made up from another capsule—for example, dissolve the contents of a 70 mg capsule in 70 mL of water and take 50 mL for a 50 mg dose. The manufacturer advises to discard any unused solution, but lisdexamfetamine appears not to degrade in water at room temperature for days or weeks. This has not been subject to clinical studies but has been widely practised by many patients during periods of supply shortages and highlighted in patient forums. One study shows there is very little degradation in mildly acid or alkaline solution in vitro.16
The normal starting dose for adults is 30 mg daily, increased after 1 week, usually to 50 mg daily (or to 40 mg if there is already moderate benefit at 30 mg or a smaller increment is preferred for other reasons). If 50 mg daily is insufficient, 70 mg can be tried (or 60 mg if appropriate). If there is a partial benefit at 70 mg without significant adverse effects, unlicensed higher doses may be justified, especially if an alternative treatment is not appropriate. Adverse effects at 30 mg may suggest the optimal dose may be as low as 20 mg. Some patients find the duration of action considerably less than 12 hours and may prefer twice daily dosing. Conversely, some find the duration too long and it may affect sleep.
Dexamfetamine (eg, from lisdexamfetamine) is excreted in urine, but the rate varies with tubular pH: in acidic urine dexamfetamine ionises, less is reabsorbed and elimination is increased.17 Anecdotally, patients have reported exploiting this by taking 1 g of vitamin C at 17:00–18:00 hours, to acidify urine and enhance excretion of dexamfetamine before bedtime, to prevent it disturbing sleep (if it does). It may also be useful after inadvertent double dosing. Dose reductions should be considered in those with renal impairment.
The NICE clinical guideline (NG87) recommends consideration of dexamfetamine for adults whose ADHD symptoms are responding to lisdexamfetamine but who cannot tolerate the longer effect profile2; conversely, it might be used as a ‘top up’ if lisdexamfetamine wears off early (but immediate-release methylphenidate could also be used if suitable), although no guidelines recommend this.
In the UK, dexamfetamine is licensed for use in adults for narcolepsy, not ADHD. It is licensed in children aged 6–17 years with refractory ADHD under specialist supervision.17 The BNF mentions off-label use for refractory ADHD in adults starting at 5 mg two times per day increased at weekly intervals up to a maximum of 60 mg per day.13 This is also the maximum for narcolepsy and is equivalent to approximately 180 mg of lisdexamfetamine, which would be considered a very high dose for ADHD. Dexamfetamine prescribing may be discouraged because of greater concerns about abuse or diversion.
Adverse effects
Nausea, loss of appetite, dyspepsia, headaches, palpitations and muscle cramps are commonly described as adverse effects of medication for ADHD but may occur independently. Comparing the timing of the symptom with the timing of exposure to medication or asking if it occurs on days when medication is not taken can be helpful to ascertain likely causation.
Appetite suppression is common with stimulants and may be a concern in those with eating disorders. Weight loss can be avoided by eating meals before taking the dose or after it has worn off or by scheduling snacks during the day even if the person is not hungry. If there is a history of an eating disorder, this may need addressing to reduce the risk of relapse.
Insomnia from stimulants may be helped by taking the dose earlier or taking ascorbic acid as mentioned above.
Patients are often anxious while starting stimulants. This could be an adjustment reaction, a nocebo effect, a symptom of ADHD, comorbid anxiety or a direct effect of medication. The timing of the anxiety is key to managing it. If the anxiety correlates with the level of stimulant, the stimulant is more likely to be a direct cause and warrant reducing dose or changing drug. If the anxiety is mainly when the drug is wearing off (‘crashing’), then a ‘top-up’ dose or a longer-acting preparation may help. If the anxiety is at unrelated times and persists, then anxiolytic treatment may be helpful.
Serotonin syndrome is a risk highlighted in the Summary of Product Characteristics (SPC) for different formulations of lisdexamfetamine and methylphenidate, particularly if combined with an antidepressant. However, it is important to keep this risk in perspective. One pharmacovigilance review of nearly 9000 reports of serotonin syndrome, collected in a global database between 2004 and 2022, reported only 194 cases from methylphenidate and 77 from amfetamines, compared with 4548 for selective serotonin reuptake inhibitors, 3159 for opioids, 1258 for serotonin noradrenaline reuptake inhibitors and even the antibiotic linezolid had 475 reports.18 The reporting odds ratio was highest in patients aged over 65 years, suggesting caution may be more relevant in older patients. There is no published data on the risk of serotonin syndrome with atomoxetine.
Stimulants may increase the risk of cardiovascular disease, but the main ‘disease’ found in studies is hypertension, which is a recognised adverse effect of stimulants and atomoxetine.19 A meta-analysis of 10 clinical trials reported that ADHD medications significantly increased resting heart rate by 5.7 beats/min, and systolic blood pressure (BP) by 2 mm Hg20; but a meta-analysis of studies covering nearly four million people showed no increase in myocardial infarction or cerebrovascular disease.21 It may be better to treat the hypertension to allow stimulants to be used.22 One review states: ‘Concerns regarding QT effects and attendant sudden cardiac death risks deter clinicians from initiating much-needed ADHD medications in patients with heart disease.23 This overly cautious approach is potentially depriving low-risk individuals of significant benefits associated with timely ADHD drug treatment.’
ADHD medications’ adverse effects may attract concern, yet other medication in psychiatry has risks. Do we overemphasise ADHD in comparison? One study of 80 commonly prescribed drugs in psychiatry shows that rates of adverse effects from ADHD medications compare favourably with many antipsychotics and mood stabilisers, although long-term studies are lacking.24 This study also suggested that methylphenidate was associated with fewer adverse events than other ADHD medication.
Monitoring
With stimulants, monitoring of blood pressure (BP), pulse and weight is recommended after dose changes and every 6 months when the dose is stable. If the BP is consistently in the hypertensive range or the resting pulse is over 120 bpm, consider reducing, pausing, or changing medication, or seeking cardiology advice. Sometimes the patient has underlying hypertension which needs treatment before stimulants can be resumed.
Other considerations
In the short term, some patients need higher doses for demanding work and lower doses at other times. Some patients use ‘top up’ doses to get started on a difficult task. The prescriber should discuss the circumstances when this is appropriate. Some patients prefer to avoid stimulants while exercising because of tachycardia.
Premenstrual exacerbation of ADHD is important. Oestrogen enhances dopamine activity.25 Allopregnanolone, a metabolite of natural progesterone, can bind gamma-aminobutyric acid receptors with anxiolytic effects, which rapidly diminish prior to menstruation or post partum, thereby exacerbating premenstrual low mood.26 Suppressing the menstrual cycle with hormonal contraceptives may not help because they contain synthetic progestogens that cannot be metabolised into allopregnanolone. They may cause depression, especially in those with ADHD.27 In addition, there may be an increase in risk-taking behaviour mid-cycle.28 Premenstrual increase in stimulant dose may be helpful.29
Annual review
The NICE guideline (NG87) recommends annual specialist review, with the purpose ‘to assess their need for continued treatment.’2 This question should not be reduced to a yes/no answer without addressing other issues which may have arisen, such as altering the dose or managing comorbidities.
The expectation that patients will remain stable following initial dose titration may not be appropriate, especially in young adults: in a cohort of 558 children followed up for up to 16 years (to a mean age of 25 years), 64% showed a fluctuating course of remission and recurrence.30
In the medium or long term, the impact of ADHD symptoms on each patient may be affected by changes in lifestyle, occupation, relationships, etc. For example, someone who starts medication during a demanding occupation requiring self-direction and self-motivation, living alone, on a low income or in poor accommodation, may need less medication once they have the external structure and support of an enjoyable, interesting, well-paid job with a regular routine, supportive colleagues and contented home life with a healthy lifestyle and understanding partner. Some patients may need dose adjustment if their weight changes substantially. There may be some patients who benefit from psychotherapy enough to reduce their medication. Also, other psychotropic medication, such as antidepressants, may be suitable for discontinuation once the ADHD is well controlled.
There are also many who cease medication, sometimes due to life circumstances or challenges in accessing services.31 A support worker from primary or secondary care could help them re-engage. Others may cease medication because they feel it is not sufficiently effective or causes adverse effects, but some patients discontinue medication because they dislike taking medication, find it inconvenient, have poor adherence or other reasons such as cost.32 Perhaps some of these might have carried on if their original dose titration had been more effective or regular review had addressed their concerns.
If the original dose titration is effective, it is assumed that most patients do not need subsequent dose adjustments. Some services may be contracted to provide a limited number of consultations for initial dose titration, which, in some cases, could lead to inadequate titration, suboptimal response and a higher risk of defaulting from treatment.
Review may rescue some patients from disengagement due to suboptimal dosing. Review may also allow retitration for patients whose medication has since become unobtainable.
Access to specialist review may not be easy, just like access to initial assessment. Annual review may not always be commissioned or attended. Patients initiated on medication in the private sector may not feel they can afford to pay for private annual review. A questionnaire, automated or virtual review may identify those who would benefit from a more detailed review, helping to target best use of resources. Specialist support to quickly answer queries (eg, by email) from generalists may also be helpful.
Tolerance to stimulants
Tolerance to stimulants is rare (<3%) at least after the initial dose titration period.33 If there are repeated requests for increasing doses, this might suggest non-medical use, or that treatment goals may not reflect what ADHD medication can achieve. Drug ‘holidays’ are not necessary to avoid the risk of tolerance (but may be helpful to assess or mitigate adverse effects, or to establish continuing need for treatment). If tolerance is confirmed, it may be worth stopping all stimulants for 7 days and retitrating.
Withdrawal from stimulants
There are many SPCs for different stimulants mention the possibility of withdrawal symptoms. In practice, many patients experience periods without medication but do not suffer withdrawal symptoms. Discontinuation may unmask symptoms of ADHD, but small trials of discontinuing therapeutic doses of methylphenidate and lisdexamfetamine have not found withdrawal symptoms.34 35
Non-stimulants
Atomoxetine and guanfacine are the main non-stimulants in use. Some antidepressants have limited evidence of benefit in ADHD36: these are not recommended as alternatives to ADHD medication but may be useful to treat comorbid anxiety or depression.
Atomoxetine
Atomoxetine selectively blocks the noradrenaline transporter in the prefrontal cortex (PFC), as do stimulants. However, it has various other effects: the mode of action is less well understood, and effects can take weeks to accumulate. Usually, a response is apparent after 1–4 weeks but in some cases, it can take 8 weeks. Atomoxetine tends to be either effective or not, more like a bimodal distribution, with intermediate efficacy being less common.37 It has the advantage of ‘24 hours’ symptom control, unlike stimulants.
It is well absorbed but is subject to first pass metabolism in the liver, by cytochrome P450 2D6 (CYP2D6), which has considerable polymorphic variation between individuals (resulting in up to 8–10 fold difference in plasma levels).38 Patients with liver disease may need a lower dose. Exposure to atomoxetine is greatly increased by concurrent use of potent CYP2D6 inhibitors such as bupropion, paroxetine, fluoxetine, terbinafine or quinidine.
In children, dosing is adjusted for weight, whereas for adults (>70 kg) the licence recommends 40 mg daily for the first week, then increasing as tolerated to usual maintenance 80–100 mg daily. Safety of higher doses has not been established, although the BNF suggests an unlicensed maximum of 120 mg per day.13 Nausea is a common adverse effect and may be tolerated by building slowly from a lower dose, such as 18 mg. Atomoxetine capsules should not be opened as the contents can be irritant. They are available in a range of strengths from 10 mg to 100 mg, but all preparations contain gelatine making them unsuitable for vegetarians.
Adverse effects are like those of stimulants. Atomoxetine increases BP and heart rate and is ‘contraindicated’ in cardiovascular disease. Monitoring of BP and pulse is recommended after each dose adjustment and every 6 months. Somnolence with atomoxetine may be mitigated by moving the timing of the dose to the evening. Annual review by a specialist is recommended.
Guanfacine
Guanfacine binds to the alpha-2A noradrenaline receptor (α2A-AR). Historically, the α2A-AR was thought to be presynaptic, but in the human PFC, it is mostly postsynaptic39: hence guanfacine may work by enhancing propagation of nerve impulses in the PFC.40 It also binds the alpha-2 receptor on microglia, potentially affecting neuroplasticity.
Guanfacine is licensed for initiation in children aged 6–17 years, and those benefitting at age 16–18 years may continue it as adults (off-label, usually after tertiary specialist advice). Trial evidence of benefit in adults was sparse until recently.41
Unlike stimulants, guanfacine may lower BP and pulse rate. Commonly, it can cause sedation and somnolence which may settle after 2–3 weeks’ use. Other adverse events include headache, fatigue, appetite suppression, abdominal pain and weight gain. The SPC recommends weekly monitoring for adverse effects during titration, and every 6 months when the dose is stable.
Guanfacine should not be stopped suddenly due to a risk of rebound hypertension. Daily dosage should be reduced no faster than 1 mg every 3 days. Guanfacine has the potential for significant cytochrome p450 interactions (CYP3A4 and CYP3A5 inhibitors/inducers such as clarithromycin, verapamil, itraconazole, grapefruit, carbamazepine). Guanfacine does not interact with methylphenidate or lisdexamfetamine.
Other drugs
Clonidine may, like guanfacine, bind to the postsynaptic α2A-AR in the PFC, but also binds to presynaptic α2A-AR causing a sympatholytic effect in the locus coeruleus (reducing BP, causing sedation) and potentially in the amygdala (reducing arousal, which may be useful for anxiety). Presynaptic α2A-ARs are also expressed on neurons releasing other neurotransmitters in different areas of the brain and cord, making the pharmacology complex.42 Clonidine (extended release 100–200 μg) has a licence for children with ADHD in the USA but not in the UK, where immediate-release 25 μg and 100 μg preparations are available for migraine prevention or hypertension. Clonidine has been shown to help adolescents with hyperactive symptoms as add-on therapy, but studies of ADHD in adults are still lacking43 44; the benefit of clonidine in children seems small and less than with guanfacine.5
Drugs such as modafinil and bupropion are occasionally used but are not recommended as other drugs are more effective. There is only weak evidence that they may help.5 45
The existing pharmacopoeia for ADHD seems effective enough that new drugs are unlikely to be significantly better, but some may have a specific role. For example, viloxazine is a noradrenaline reuptake inhibitor historically marketed as an antidepressant46: it was recently approved in the USA for ADHD and can help those who have not responded to atomoxetine.47 Centanafadine has a different mode of action (inhibitor of noradrenaline, dopamine and serotonin reuptake transporters) and seems to be effective and well tolerated in early trials, but its place in clinical practice is not yet established.48
Newer mechanisms of drug delivery (transdermal methylphenidate, which avoids hepatic first pass effect, or delayed onset slow release, which can be effective at the time of waking up) are available in the USA and may become more widespread. Dexmethylphenidate, the active enantiomer of methylphenidate and serdexmethylphenidate, a prodrug metabolised into dexmethylphenidate, are available in some countries.
‘Shared care’
The NICE guideline anticipates that treatment will be initiated by a ‘specialist’, then continued in primary care (known as ‘shared care’), once the patient is ‘stable’.2 But what does ‘stable’ mean? ‘Shared care’ in the UK is at the discretion of the general practitioner (GP); the GP can decide what is ‘stable’, yet it is often the specialist who tells the patient they are ‘stable’ or ‘suitable for shared care’ and this can cause tension between the specialist or patient wanting the convenience of prescribing in primary care, and the GP wanting to be more confident that the patient will not be changing dose, that monitoring has not shown any problems, any adverse effects have been addressed, and that all appropriate psychoeducation has been arranged. As mentioned above, the concept of ‘stable’ may be further complicated by fluctuating symptom levels in relation to external or other factors. It is important that GPs have timely access to specialist advice to allay concerns that the patient’s condition may not be as ‘stable’ as the GP would like.
‘Shared care’ is outside the GP contract, often unfunded, and GPs may perceive it as work ‘dumped’ from secondary care. Patients may feel deserted too, especially if after being discharged, they have remaining questions or concerns which cannot be addressed by the GP. GPs may feel they lack the resources or sufficient expertise to treat ADHD, or that there is limited access to the specialist with whom care is ‘shared’. However, these patients are already using GP and health services resources, without necessarily benefiting; treating ADHD may help executive function to deal with a range of problems and optimise use of health services. It can be immensely satisfying to help such patients recognise and treat their ADHD and to help them unlock new ways to engage in healthier behaviours; those with ADHD have the potential for substantial improvement, which may benefit the patient, the efficient use of health services for comorbidities, and the GP’s job satisfaction.
Primary care clinicians are ‘well placed’ (with sufficient training, resources and access to specialist advice) to manage patients with ADHD, particularly those without other serious conditions requiring specialist services (eg, substance misuse, bipolar disorder, eating disorders, suicidal ideation, complex trauma). Unfortunately, this is also true of many other conditions, and primary care cannot take on all of them.
Particularly relevant to ADHD is that primary care clinicians may be better at taking a pragmatic holistic view of the patient within their family and community, including being able to consider common somatic comorbidities such as obesity, chronic fatigue, functional bowel or neurological symptoms, musculoskeletal problems, allergic or autoimmune disease and so on (which mental health services cannot address). Primary care teams are used to managing polypharmacy and repeat prescriptions, and to being approachable to patients who may struggle to access distant services.
There are an increasing number of GPs with an extended role (GPwER) in ADHD, and the Royal College of General Practitioners in the UK has recently acknowledged this with a credential profile.49 If the contractual framework for funding NHS services could accommodate appropriate innovative service delivery models (eg, a GPwER serving a Primary Care Network, with collaboration and supervision from secondary care), that might enable many more patients to benefit from treatment. In the long term, assuming adequate resource provision, GPs might manage ADHD as they currently manage many other conditions which used to be the remit of specialists, that is, managing straightforward cases in primary care and referring the more complex cases for advice and guidance or specialist-led care. The Canadian guidelines expect that GPs manage ADHD, and the new French guidelines recommend ‘any doctor’ trained in ADHD can diagnose and treat it.50 51
Conclusion
ADHD is a serious condition with high morbidity but responds well to treatment, which can be tailored to the patient’s medical and psychosocial needs. Stimulants are effective first-line medications and can be initiated and titrated over weeks to months, usually requiring only a few titration steps, but some patients may need higher, twice daily or ‘top-up’ doses, or non-stimulant medication. Although there are many nuances, most cases will benefit from standard dose titration and could be managed in primary care or general psychiatry by a clinician with appropriate training, resources and professional support.
Supplemental material
Acknowledgments
I would like to thank Anna Wright for valuable comments from a patient perspective, and Dr Sally Cubbin and Dr Ulrich Müller-Sedgwick, and DTB’s peer-reviewers for expert comments and encouragement.
References
Footnotes
Competing interests None declared. Refer to the online supplementary files to view the ICMJE form(s).
Provenance and peer review Commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.