Running on Empty

Running on empty
Running on - running on empty
Running on - running blind
— 1977 Lyrics by Jackson Browne

There are some days, some patients and even some ECGs when the treating team is running on empty, when the brain stem is making the diagnostic and therapeutic decisions and the brain is just humming along. 


This healthy 52 year old patient came to the Emergency Department three weeks ago complaining of palpitations. Examination of the cardiovascular system was normal. An ECG and blood tests were done, he spent 5 hours in the Short Stay Unit and was then discharged on metoprolol. An appointment for Cardiology outpatient was requested, and he is scheduled to be seen next week. He has returned because the palpitations have recurred. He is no longer taking metoprolol.

Figure 1

+ Figure 1 is the ECG taken on arrival. What are the major findings on this ECG, and what are the main management issues in this case ?

The main features of this ECG are:

  • Atrial fibrillation is present with a ventricular rate of about 100 beats per minute.
  • Apart from the atrial fibrillation the ECG is within normal limits.
  • The ECG recorded 3 weeks before was the same as this ECG

When a patient is first found to be in atrial fibrillation the immediate issues are:

  • Does the ventricular rate need slowing? The observed resting ventricular rate of 100 beats per minute is slower than the expected ventricular rate (of between 120-150 beats per minute) if atrioventricular node conduction was normal. This slower (than expected) ventricular rate could be due to the tachycardia-bradycardia (sick sinus) syndrome, or to the effect of drugs, or both. In this case there is no immediate need to reduce the ventricular rate.
  • Will the patient need cardioversion in the Emergency Department? This may be needed in persons with pre-excitation (Wolf-parkinson White) syndrome who have developed atrial fibrillation or if the ventricular rate is very fast, but this does not apply to this case.
  • Is there any evidence of underlying heart disease that could cause or predispose to the development of atrial fibrillation? In this case the cardiovascular system was normal.
  • Are there other (non cardiac) conditions that can cause atrial fibrillation? One important condition is hyperthyroidism, which is ruled out (or in) by blood (thyroid function) tests. When we review the blood tests taken three weeks ago we find the following results:
    • TSH : < 0.01 mIU/L (Normal 0.4-4.0 mIU/L)
    • Free thyroxine: 24.0 pmol/L (Normal 10-25 pmol/L)
    • These results are consistent with hyperthyroidism. The results were overlooked on the first presentation.
    • Repeat investigations are:
    • TSH : < 0.01 mIU/L (Normal 0.4-4.0 mIU/L)
    • Free thyroxine: 22.9 pmol/L (Normal 10-25 pmol/L)
    • Free T3: 10.2 pmol/L (Normal 4.0-8.0 pmol/L)
    • TSH receptor antibody: 2.8 (Elevated) Present in 80% of patients with Graves' disease.
    • Thyroglobulin antibody: 765 (Elevated) Immunoassay for thyroid microsomal/ peroxidase and thyroglobulin antibodies.
    • Positive thyroid microsomal antibodies or thyroglobulin antibodies occur in virtually all patients with Hashimoto disease, but positive results also occur in patients with other forms of autoimmune thyroid disease.
    • Thyroid gland scan: There is uniform, markedly increased tracer uptake throughout the thyroid gland

Final diagnosis: Graves Disease

The incidence of thyrotoxicosis in unselected patients with the first presentation of atrial fibrillation is less than 1 percent ([See "Thyroid Disease and the ECG" in Nuts & Bolts of the ECG]1). If a doctor working in a busy Emergency Department sees 20 patients per year with a first presentation of atrial fibrillation, then it may take up to five years before they see a patient with atrial fibrillation due to undiagnosed hyperthyroidism.

In this case thyroid function tests were ordered (Department of Brain Stem Investigations) but the results were not followed up.

A few things to ponder:

  • Is atrial fibrillation due to thyrotoxicosis less likely to be paroxysmal (and more likely to be persistent) than (so called) lone atrial fibrillation ?
  • Is the risk of systemic emboli greater in atrial fibrillation due to thyrotoxicosis?

Key Points

  • The incidence of thyrotoxicosis in unselected patients with the first presentation of atrial fibrillation is less than 1 percent
  • If you order a test check the result before discharging the patient