Answers

Short answers

1 Respiratory muscle weakness causing nocturnal hypoventilation and cor pulmonale. Overnight oximetry showed precipitous desaturation associated with reflex tachycardia on lying flat, suggestive of respiratory muscle weakness. This is supported by the history of orthopnoea and previous episode of type 2 respiratory failure.

2 First line investigations are: lung function tests (spirometry, lung volumes, and gas transfer); lying, sitting, and standing spirometry; maximal inspiratory mouth pressures and sniff nasal inspiratory pressures. Other investigations include measurement of transdiaphragmatic pressure and phrenic nerve magnetic stimulation if available.

3 Possible causes are: neural abnormalities (Guillain-Barré syndrome, trauma, mediastinal malignancies, herpes zoster, motor neurone disease, vasculitis, hereditary sensorimotor polyneuropathy, critical illness polyneuropathy, alcoholism, diabetes); disorders of the neuromuscular junction (toxins, drugs, myasthenia gravis); and abnormalities of muscle tissue (muscular dystrophies, myopathies, acid maltase deficiency, hypothyroidism, systemic lupus erythematosis, mechanical ventilation induced diaphragmatic dysfunction).^{1 2}

Long answers
1 Likely reason
Overnight oximetry showed repeated precipitous oxygen desaturations and reflex tachycardia (figure) when the patient was lying flat. Lung function tests showed a restrictive spirometry defect, which worsened when supine (sitting forced expiratory volume in one second 1.95, sitting forced vital capacity 2.09, ratio 93%; supine forced expiratory volume in one second 1.43, supine forced vital capacity 1.57, ratio 91%). In addition, maximal inspiratory pressure was greatly reduced at 31 cm H₂O (normal 80-120) and sniff nasal inspiratory pressure was 51% of the predicted value, results that suggest respiratory muscle weakness. More specialised investigations of respiratory muscle function are not currently available at our institution. The patient had a raised body mass index (31.6) and the baseline overnight oximetry trace also showed frequent smaller desaturations (between 3-5%) associated with short rises in heart rate, raising the possibility of coexistent obstructive sleep apnoea. However, this was not investigated further.

View larger version (12K): [in this window] [in a new window] [PowerPoint Slide for Teaching]   Overnight oximetry trace. Red trace: oxygen saturation; blue trace: heart rate

 
Respiratory muscle weakness was further suggested by the initial presentation with type 2 respiratory failure. Respiratory failure is broadly divided into type 1 respiratory failure, which is characterised by severe hypoxia (arterial oxygen tension <8.0 kPa) with low or normal carbon dioxide, and type 2 respiratory failure, which is defined as the combination of hypoxia and hypercapnia. Hypercapnia is primarily caused by hypoventilation, as a result of reduced central respiratory drive, underlying neuromuscular disease, obesity, or an abnormality of the thoracic wall.^{1 3} Ventilation-perfusion mismatch can also cause hypercapnia, particularly when associated with an increase in alveolar dead space, as seen in some patients with chronic obstructive pulmonary disease.³ However, in healthy young people with no predisposing comorbidities, abnormal gas exchange associated with pneumonia would be expected to be characterised by hypoxia, without hypercapnia. Our patient had received high flow supplemental oxygen during interhospital transfer to our unit. This could have contributed to the observed hypercapnia by reducing respiratory drive and increasing ventilation-perfusion mismatch. Weakness of the inspiratory muscles causes an imbalance between muscle load and capacity, which may precipitate type 2 respiratory failure. Expiratory muscle weakness reduces cough and predisposes towards infection and lung atelectasis. The clinical features in this case suggest weakness of both inspiratory and expiratory muscle groups. Our patient’s bicarbonate measurement was not raised, as would be expected in chronic type 2 respiratory failure. It is therefore likely that our patient had coexistent metabolic acidosis, probably related to lactic acidosis caused by a combination of tissue hypoxia and concurrent pneumonia.

2 Further investigations
Which of the several tests of respiratory muscle function should be used depends on the clinical picture.^{1 2 4} First line investigations include maximal inspiratory pressure and sniff nasal inspiratory pressure. Maximum inspiratory pressure is the highest mouth pressure obtained during a maximum inspiratory effort that is sustained for one second. Sniff nasal inspiratory pressure measures peak inspiratory pressure in one nostril which is occluded by a nasal plug.⁵ It is easier to measure sniff nasal inspiratory pressure than it is to measure maximum inspiratory pressure, although nasal congestion may affect measurements. For both tests, poor technique, submaximal effort, and coexistent lung disease may result in falsely low readings.⁶ Diagnostic accuracy can be improved by using a combination of tests in conjunction with standard lung function measurements.⁷ Further invasive refinements include the direct placement of oesophageal and gastric transducers to obtain more direct measurements of transdiaphragmatic pressure during maximal manoeuvres (such as sniff and twitch).^{8 9} Non-volitional tests such as phrenic nerve magnetic stimulation are more reliable but less widely available.¹⁰

3 Causes of respiratory muscle weakness
Respiratory muscle weakness has several causes, which may be subdivided into neural abnormalities, disorders of the neuromuscular junction, and abnormalities of muscle tissue (box). In view of the chronic nature of this patient’s illness, and the absence of focal neurological signs or symptoms, the principle differential diagnoses included myasthenia gravis, critical illness neuropathy, and acid maltase deficiency.

Causes of diaphragmatic muscle weakness1 2    Neural abnormalities Upper motor neurone: Cerebral infarction or haemorrhage Lower motor neurone: Guillain-Barré syndrome Trauma Mediastinal malignancies Herpes zoster Motor neurone disease Vasculitis Hereditary sensorimotor polyneuropathy Critical illness polyneuropathy Alcoholism Diabetes    Abnormalities of the neuromuscular junction Toxins (such as botulinum) Drugs Myasthenia gravis    Abnormalities of muscle function Muscular dystrophies Myopathies Acid maltase deficiency Hypothyroidism Systemic lupus erythematosis ("shrinking lung syndrome") Mechanical ventilation induced diaphragmatic dysfunction

Echocardiography had shown right ventricular dilation associated with increased pulmonary artery pressures and no evidence of left ventricular dysfunction or valvular heart disease. A subsequent bubble contrast echo found no evidence of intracardiac shunting. Pulmonary arterial hypertension is a progressive condition characterised by raised pulmonary arterial pressures leading to right ventricular failure.¹¹ In 2003 the classification of pulmonary hypertension was revised (table).¹² Patients with raised pulmonary artery pressures need a careful and systematic assessment to identify those who may benefit from disease modifying treatment. Currently in the United Kingdom and Ireland, such treatment is initiated in national specialist pulmonary vascular units.¹³

View this table: [in this window] [in a new window]   Classification of pulmonary hypertension12

 
In our case, further normal results included thyroid function tests and negative autoimmune and vasculitis serology, anticholinesterase antibodies, and HIV testing.

Therefore, the above investigations support a diagnosis of respiratory muscle weakness causing nocturnal hypoventilation complicated by cor pulmonale. Serum concentrations of the enzyme acid maltase (-glucosidase without acarbose 1.4 µmol/g/h (3-20), -glucosidase with acarbose 0.84 µmol/g/h (2-10)) were reduced, and creatinine kinase concentrations were mildly raised, in keeping with acid maltase deficiency. We did not perform skeletal muscle biopsies because sufficient diagnostic information had been obtained. A subsequent review of the patient’s history showed that she had experienced reduced mobility since childhood and did not participate in sports at school. There was no family history of similar symptoms.

Acid maltase deficiency, also termed glycogen storage disease type II or Pompe’s disease, is an autosomal recessive condition. It is characterised by deficiency of the enzyme acid -glucosidase caused by mutation in the acid -glucosidase gene on chromosome 17.¹⁴ This leads to abnormal accumulation of intracellular lysosomal glycogen in most cells of the body, although the effects are most noticeable in muscle. The course of the disease is variable, with onset of symptoms occurring at any age up to the sixth decade. The classic infantile form is most severe, and affected neonates have almost no -glucosidase activity. Such patients rarely survive beyond the first year of life. Patients with milder later onset disease have greater -glucosidase activity. These patients usually present with proximal myopathy and eventually become wheelchair dependant and require ventilatory support.¹⁵ Treatment has traditionally been supportive, but enzyme replacement therapy (Myozyme) has recently been approved for selected patients.¹⁶

Patient outcome
Our patient was treated with non-invasive ventilation, which resulted in increased exercise tolerance, reduced orthopnoea, and a marked improvement in gas exchange. Right heart failure was initially treated with a combination of diuretics and anticoagulation. Repeat echocardiography six weeks later showed normalisation of pulmonary artery pressures and resolution of right ventricular dysfunction. Therefore, we did not undertake a more invasive assessment of right ventricular haemodynamics. The patient is currently undergoing genetic counselling and is being assessed for enzyme replacement therapy.

Cite this as: BMJ 2009;339:b3093