This site is intended for health professionals only

At the heart of general practice since 1960

Read the latest issue online

CAMHS won't see you now

What you need to know about eating disorders

Psychiatrist Dr Adrienne Key answers GP Dr Mandy Fry‘s questions on hypokalaemia, bone loss, male patients and using the Mental Health Act

Psychiatrist Dr Adrienne Key answers GP Dr Mandy Fry‘s questions on hypokalaemia, bone loss, male patients and using the Mental Health Act

1 Individuals with eating disorders often seem to be hypokalaemic, presumably as a result of protracted vomiting. How often should we be checking their potassium levels? Are there any other biochemical abnormalities we should be concerned about?

41230721Hypokalaemia is one of the common electrolyte and fluid disturbances that occur in an eating disorder – the others include dehydration, hyopchloraemia and alkalosis.

Hypokalaemia is caused by persistent purging behaviour, vomiting or laxative misuse but also sometimes diuretic abuse. It causes weakness in all muscle and, most worryingly, cardiac arrhythmias, which may lead to death in severe cases. Renal function can also be affected. Metabolic alkalosis can augment potassium depletion.

Diuretic abuse, particularly thiazide and loop diuretics, can produced marked sodium as well as potassium depletion.

GPs should be concerned as soon as the potassium is below the normal range since purging rarely stops without intervention, so the patient continues to be vulnerable to an escalating severity of hypokalaemia. Potassium supplements are frequently sufficient but should be closely monitored.

Low sodium and magnesium levels occur less commonly but both have potentially serious consequences. If severe, low sodium may cause central nervous system disturbances. Low magnesium results in muscle weakness, cardiac arrhythmias and mood changes and is also associated with other abnormalities such as hypocalcaemia and hypokalaemia. Consider the presence of low magnesium levels in the face of refractory hypokalaemia.

2 Should we be undertaking ECGs in practice to look for arrhythmias secondary to biochemical disturbance?

Yes, all patients with an eating disorder should have an ECG.

Biochemical disturbances lead to arrhythmias, particularly hypokalaemia in patients with normal-weight bulimia but more frequent changes are due to starvation. These include prolongation of the QT interval, bradycardia and T-wave inversion. Any ECG change increases the level of risk to the patient of a cardiac event and should lead to an urgent referral to an eating disorder clinic.

Be aware of the risk of drugs that prolong the QT interval on the ECG – for example, antipsychotics, tricyclic antidepressants, macrolide antibiotics, and some antihistamines.

In patients with anorexia nervosa at risk of cardiac complications, prescribing drugs with side-effects that may compromise cardiac function should be avoided.

3 Anorexia nervosa is a risk factor for osteoporosis so should we screen these individuals with bone densitometry in later life? If they had a prolonged period of amenorrhoea, should we try to prevent development of osteoporosis by giving those who are amennorrhoeic oestrogen supplements in the form of the combined pill or HRT?

Yes, young women with anorexia nervosa are at increased risk of bone fractures later in life and screening should be considered in those with a history of ammenorrhea and anorexia nervosa. In addition, DXA bone scans can be useful as a motivational tool in those patients with current anorexia and ammenorrhea by helping them accept treatment. Resumption of normal menses is the best protection against development or worsening of osteoporosis.

The most recent NICE guidance on eating disorders also states that people with eating disorders and osteoporosis or related bone disorders should be advised to avoid physical activities that significantly increase the likelihood of falls.

Research has been contradictory as to whether exogenous oestrogens offer long-term protection to those with persistent amenorrhoea, but probably leans towards their use now, particularly if recovery is unlikely.

4 Is there any correlation between eating disorders and other forms of deliberate self-harm?

No. There is a small subgroup of patients with eating disorders, drug, alcohol and self-harm tendencies, which form a more difficult-to-treat group, but self-harm in people with eating disorders is no more common than in patients with other mental health diagnoses.

5 The incidence of eating disorders appears to be increasing in men. Why is this? Are the theories about its aetiology the same for men and women?

Eating disorders are increasing, particularly atypical and partial syndrome cases. For men, it is assumed this is because of the increase in scrutiny of the male body image, but it is also because we are better at identifying eating disorders in men and it is more socially acceptable for them to ask for help.

However, despite changing attitudes and understanding, eating disorders are still considered primarily a female issue, which complicates both diagnosis and management in men.

Many men drop out of treatment and we have still not identified the reasons for this.

6 One of the theories about eating disorders in girls is that it is an attempt to delay sexualisation. Does this have implications for group therapy? Should it be single sex?

The theories about anorexia being due to attempts at delaying sexualisation are not now considered to be so relevant. Anorexia nervosa is caused by an interplay of biological, genetic, psychological and social factors. Most groups are mixed sex but there are certainly arguments for allowing male single-sex groups as qualitative research has demonstrated men experience difficulties in dealing with what society considers to be a ‘women's disease' (see question 5).

7 What long-term health implications are there for individuals with anorexia – apart from the well-established risk of osteoporosis – particularly those with pre-pubertal disease? Does a delay in menarche increase their risk of breast cancer for example, or does the fact that it is effectively a hypo-oestrogenic state prevent this?

We have not established a link between anorexia nervosa and breast cancer.

The prognosis for children and adolescents with anorexia nervosa is variable. Some, particularly those with a rapid and early onset, will make a full recovery from a first episode. This is most likely where early physical and psychosocial development has been healthy and where there is an identified precipitating negative life event such as bereavement.

But the long-term health risks are significantly worse in the prepubertal disease if recovery is delayed or not achieved. These include risk of kidney, heart and liver problems, short stature, delay or absence of puberty and infertility. The psychological risks remain depression, anxiety and suicide, and a generally diminished quality of life.

8 I find the Mental Health Act confusing when it comes to the physical consequences of mental illness. How might it be applied in severe cases of anorexia? How does the presence of significant mental illness affect the rights of young people under the age of 16 to be competent to determine their own treatment options?

The Mental Health Act explicitly states that when treating anorexia nervosa, the medical treatment is nutritional rehabilitation. The physical consequences of starvation place the patient at risk of dying and nutrition reverses this.

NICE guidance states that ‘when a young person with anorexia nervosa refuses treatment that is deemed essential, consideration should be given to the use of the Mental Health Act or the right of those with parental responsibility to override the young person's refusal'.

Force-feeding is possible under section 3 and in an emergency under section 2 – but NICE guidance clearly states an eating disorder specialist should be involved in the care of these patients.

This applies equally to patients under 16. The Mental Health Act should always be given due consideration if the patient's life is at risk because of the presence of anorexia nervosa and the consequent starvation.

9 Psychological therapies clearly have an important role to play in the treatment of eating disorders but what about medication? Do SSRIs have a role? What about symptomatic treatments such as metoclopramide or domperidone to increase gastric emptying and thus help with feelings of satiety and nausea when eating?

SSRIs are useful in the treatment of bulimia nervosa and NICE guidelines recommend 60mg fluoxetine with self-help material as first-line treatment. Note this is higher than the effective dose in people with depression.

There is little other evidence yet to firmly support psychotropic or other medication in the treatment of eating disorders but comorbid conditions occur frequently such as OCD and depression, so these must be addressed.

As mentioned in question 2, caution should be applied when prescribing any medication that prolongs the QT interval due to the ECG abnormalities that commonly occur.

Gastric emptying is a problem for those in recovery but patients should be reassured that normal eating will restore this and that it should be a short-term problem.

10 How can we try to identify individuals with bulimia who retain a normal BMI. What are the risk factors for this disease? Are there any indicators that should make us more alert to its possibility so we directly inquire about it?

Patients with bulimia are more likely to ask for help than those with anorexia, but it still remains a disorder often shrouded in shame for the sufferer.

Frequent problems that may indicate the presence of bulimia include:

• depression, particularly depression not responding to treatment

• irregular or absent menstrual periods

• unexplained infertility with multi-follicular ovaries

• low potassium

• borderline anaemia.

Asking directly about eating and weight fluctuations, while explaining how common the problem is and how help is at hand, may offer the patient the chance to open up.

11 The age of onset of anorexia seems to be ever decreasing. As one of the theories about its aetiology appears to centre on media pressure to be thin, how do you prevent the Government's current focus on childhood obesity resulting in a paradoxical increase in the incidence of eating disorders?

The anti-obesity messages are difficult and may lead to an increase in disordered eating and so eating disorders. Morbid obesity is frequently a result of disordered eating, making it impossible and pointless to separate the two states.

Messages should be about healthy eating, activity and a positive body image without promoting thin body ideals and strict diets.

Most research is now demonstrating that an increase in dieting behaviour in childhood and preoccupation with dietary restraint leads to increasing obesity and disordered eating in adulthood, so we need to review how ‘health' messages are distributed and implemented.

12 Individuals with eating disorders often seem to relapse in later life at subsequent times of stress. How can we minimise the likelihood of this occurring? What are the true cure rates?

Bulimia nervosa has a better prognosis than anorexia nervosa. Recommended specific CBT success rates as high as 60-80% have been achieved. Anorexia remains difficult to treat with a third of sufferers making a complete recovery, a third a partial one and a third remaining ill or in some cases dying. Improvement in prognosis is achieved through rapid referral and treatment, so at times of relapse patients should be given access to support as soon as possible.

Dr Adrienne Key is consultant psychiatrist and lead consultant for eating disorders at the Priory Hospital, Roehampton

Competing interests: None declared

Dr Key's colleague at the The Priory Group, Dr Alex Yellowlees, has made a video where he discusses how GPs can spot the signs of an eating disorder.

points eating What I will do now What I will do now

Dr Mandy Fry considers the responses to her questions

It's interesting to note that persistent hyokalaemia may be caused by an unrecognised low magnesium level so I'll check this in those who aren't responding to potassium supplementation, which I will now prescribe more routinely for these patients.
I'd also never considered using an ECG as part of a risk assessment to identify those who need an urgent referral to the specialist service.
Or indeed the use of bone densitometry as a motivational tool for someone with amenorrhoea to seek further treatment.
It would also be interesting to discover how many men are seen by our local service and whether they offer them single-sex groups.
The cure rates for anorexia remain disappointing but those for bulimia appear much more encouraging. So I'll try and identify those patients more quickly, by asking some more direct questions around weight and eating patterns, particularly in those with refractory depression.
I will then consider the role of high-dose fluoxetine for those individuals.
The increasing incidence of these conditions, particularly in partial and atypical forms, means that as GPs we need to be more vigilant about identifying cases, as it is clear that rapid treatment improves cure rates and decreases the likelihood of long-term health complications.

Dr Mandy Fry is a GP in Cirencester, Gloucestershire, and senior primary care lecturer at Oxford Brookes University

Eating disorders

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say