Diaphragmatic dysfunction and respiratory illness (Review, NEJM)

Diaphragmatic dysfunction can result from nerve damage, primary muscle problems, or problems with the muscle’s interaction with the chest wall. The true incidence of diaphragmatic paralysis is unknown, since many patients are asymptomatic. Treatment for diaphragmatic dysfunction usually consists of watchful waiting, addressing underlying causes, with mechanical ventilation if respiratory failure develops.

Causes of Diaphragmatic Weakness or Paralysis

Diaphragmatic paralysis is likely most often idiopathic and unilateral. When a cause for diaphragmatic paralysis can be identified, it may be due to:

• Trauma or surgery causing cervical cord or phrenic nerve damage (high C-spine injuries involving C3-C5, phrenic nerve injury during cardiac surgery);
• Mechanical ventilation;
• COPD and other diseases that cause lung hyperinflation;
• Myopathies and neuropathies (myasthenia gravis; critical illness neuro/myopathy; amyotrophic lateral sclerosis, poliomyelitis, with a 35 year delay until diaphragmatic weakness);
• Inflammatory disorders (e.g., sepsis);
• Mediastinal masses.

 

Symptoms of Unilateral & Bilateral Diaphragmatic Paralysis

Unilateral diaphragmatic paralysis or weakness rarely causes symptomatic dyspnea at rest, but may result in dyspnea on exertion or the patient’s voluntary restriction of activity. It can sometimes cause dyspnea when lying on one’s back (supine). Often, unilateral diaphragmatic paralysis is detected incidentally on a chest X-ray obtained for other purposes.

Bilateral diaphragmatic paralysis frequently causes dyspnea at rest, with exertion, when supine (necessitating sleeping in a recliner), bending over, or when swimming with water above waist level. Sleep disorders are also common in these patients, and symptoms thereof (fatigue, somnolence, awakening during sleep) may be the first presentation of bilateral diaphragmatic paralysis. Recurrent pneumonias (possibly due to basilar atelectasis) and recurrent respiratory failure are also possible.

Rather than causing problematic dyspnea or respiratory insufficiency on its own, diaphragmatic paralysis likely acts more often as a “co-conspirator” that reduces respiratory reserve and the threshold for respiratory failure. Many people with diaphragmatic paralysis are well-compensated when at rest and not acutely ill, but an acute illness such as pneumonia or an exacerbation of heart or lung disease may increase respiratory demand beyond the ability of the paralyzed diaphragm, with resulting severe dyspnea or respiratory failure. Diaphragmatic dysfunction also likely makes it more difficult to escape from dependence on mechanical ventilation.

Diagnosis of Diaphragmatic Paralysis

A number of tests can help identify diaphragmatic dysfunction. Tests for diaphragmatic paralysis include:

Chest X-rays in diaphragm paralysis may show elevated hemidiaphragms and basal subsegmental atelectasis; insensitive in detecting bilateral paralysis as films may often be interpreted as “poor effort” or “low lung volumes.” Chest X-ray is 90% sensitive for unilateral paralysis but only 44% specific (high false positive rate).

Fluoroscopy of the diaphragm (“sniff test”): the patient sniffs energetically during fluoroscopy; descent of the diaphragm is the normal response. People with unilateral diaphragmatic paralysis have a paradoxical upward movement of the weak hemidiaphragm, which gets “sucked up” by the negative pleural pressure created by the working hemidiaphragm. The sniff test is not considered to be accurate in diagnosing bilateral diaphragmatic paralysis, with a ~6% false positive rate and a higher false negative rate.

Pulmonary function tests show restriction, which may be moderate to severe (30-50% predicted total lung capacity) in bilateral diaphragmatic paralysis. The restriction worsens when supine, evidenced by a drop in vital capacity of 30 to 50% in bilateral diaphragm paralysis. This test is sensitive and has a high negative predictive value: if there is no reduction in FVC when supine, there is probably no significant diaphragmatic paralysis.

Maximal static inspiratory pressure (MIP) and sniff nasal inspiratory pressure are reduced to ~60% predicted in people with unilateral diaphragmatic paralysis and to ~30% predicted in bilateral diaphragmatic paralysis. However, these tests are effort-dependent and less reproducible than lung volumes; with a high false positive rate for respiratory weakness.

Ultrasound can be extremely useful in measuring diaphragmatic function. The point of contact between the diaphragm and the rib cage should be viewed. The diaphragm should thicken with inspiration, indicating shortening; if the diaphragm does not thicken, paralysis is present. As a fast, inexpensive and noninvasive test, ultrasound offers many advantages and can also be used serially to assess recovery of a paralyzed diaphragm. A 2011 study among 88 mechanically ventilated patients suggested that diaphragmatic weakness on ultrasound could help predict extubation failure and inability to wean from mechanical ventilation.

Electromyography of the diaphragm is technically difficult and its results can therefore be hard to interpret or rely upon. It can potentially help differentiate between a myopathy and neuropathy, if one of these is strongly believed to be the cause of diaphragmatic dysfunction.

 

Physical Examination Findings in Diaphragmatic Dysfunction

Generally speaking, physical findings are more likely in people with bilateral diaphragmatic paralysis. Some, all, or none of these physical examination findings may be present in people with diaphragmatic dysfunction.

Abdominal paradox, with the abdomen moving inward as the thorax expands during inspiration; this is the “classic” sign of diaphragmatic dysfunction, caused by the accessory muscles creating negative pleural pressure that “sucks up” the flaccid diaphragm into the chest during inspiration. Abdominal paradoxical breathing is almost exclusively found in people with bilateral diaphragmatic paralysis; if present in unilateral paralysis, it means the respiratory muscles in general are weak.

Other possible physical examination findings in diaphragmatic paralysis include:

• Tachypnea
• Use of accessory muscles during quiet breathing; detectable by contraction of the sternocleidomastoid muscles during neck palpation
• Decreased diaphragmatic excursion (percussing the lower rib cage at end inspiration and end expiration; the change in resonance should span at least 3 – 5 cm [video])

Treatments for Diaphragmatic Paralysis

Most people with diaphragmatic paralysis do not require treatment, other than watchful waiting, potentially with serial examinations. Many or most people with diaphragmatic paralysis have other likely contributing causes for dyspnea (obesity and deconditioning, lung and heart disease, etc.), making a determination of the contribution of diaphragmatic paralysis to dyspnea extremely difficult.

A common-sense approach to treatment of diaphragmatic paralysis can include:

• Remove/treat any obvious contributing factors (hypokalemia, hypophosphatemia, high-dose steroids, neurotoxic drugs, neuromuscular blockers).
• Nocturnal noninvasive ventilation for people with an awake pCO2 of 45+ mm Hg; nocturnal hypoxemia (SaO2 < 88% of >5 consecutive min); or progressive neuromuscular disease and a maximal static inspiratory pressure (MIP) < 60 cm H2O or forced vital capacity (FVC) < 50% predicted.
• Treat sleep-disordered breathing, if present, with continuous positive airway pressure (CPAP) or nocturnal noninvasive ventilation.

Surgical plication of the hemidiaphragm involves “tightening” the loose, paralyzed hemidiaphragm by oversewing its center. This therapy improved lung function and dyspnea in retrospective, uncontrolled trials.  It is of no use in bilateral diaphragmatic paralysis, and is relatively contraindicated in progressive neuromuscular disease and in severely obese people. Because unilateral paralysis is usually either minimally symptomatic or improves with time, plication should be considered only after a long period of watchful waiting.

Phrenic pacing is only appropriate for ventilator dependent patients, mainly quadriplegics with cervical spine injuries at C3-C5 or above (“high quads”).

McCool FD, Tzelepis GE. Dysfunction of the Diaphragm. N Engl J Med 2012;366:932-942.