Diagnosis and management of nontraumatic unilateral diaphragmatic paralysis (complete or partial) in adults

INTRODUCTION — The diaphragm is the most important inspiratory muscle of the body and plays a critical role in ventilation and gas exchange [1-3]. The diaphragm also serves as a mechanical barrier between the abdominal and thoracic cavities and maintains the pressure gradient between the cavities.

Diaphragmatic paralysis can be unilateral or bilateral; unilateral paralysis is more common. The presentation, diagnostic approach to, and management of unilateral and bilateral paralysis are different. The causes and diagnostic evaluation of non-traumatic unilateral diaphragmatic paralysis in adults will be reviewed here. Evaluation and management of bilateral diaphragmatic paralysis, and respiratory muscle weakness due to other respiratory neuromuscular disorders, and surgical treatment of traumatic phrenic nerve injury are reviewed separately:

●(See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation".)

●(See "Diagnostic evaluation of adults with bilateral diaphragm paralysis".)

●(See "Treatment of bilateral diaphragmatic paralysis in adults".)

●(See "Surgical treatment of phrenic nerve injury".)

●(See "Disorders of ventilatory control".)

●(See "Diaphragmatic paralysis in the newborn".)

PATHOPHYSIOLOGY AND ETIOLOGY

Pathophysiology — The left and right sides of the diaphragm (figure 1) are innervated by the ipsilateral phrenic nerves, which derive from cervical nerve roots three, four, and five [4]. Each phrenic nerve divides into four trunks that innervate the anterolateral, posterolateral, sternal, and crural portions of the diaphragm. Interruption or injury to one of the phrenic nerves or branches results in unilateral complete or partial diaphragmatic paralysis, respectively [5-9].

Depending upon the degree of hemidiaphragm compromise, the contralateral normal hemidiaphragm and the accessory muscles of respiration compensate to perform the work of breathing. They do this by more intense contraction and progressive recruitment of other, less important accessory inspiratory muscles (eg, sternomastoid, trapezius, latissimus dorsi, and pectoralis minor and major). Patients may also recruit abdominal muscles to augment exhalation, which forces the respiratory system to a lower volume at end-exhalation (ie, below functional residual capacity).

Immediately after denervation of the hemidiaphragm, the normal contralateral hemidiaphragm is initially able to generate enough negative pleural pressure to promote adequate inspiration [10]. During inspiration, the unaffected diaphragm moves downward to draw air into the lungs via the trachea. However, air is also drawn across the chest from the affected side into the unaffected lung resulting in paradoxical (ie, upward) movement of the affected hemidiaphragm. This upward movement results in basal atelectasis and poor gas exchange in that portion of the ipsilateral lung and is used to facilitate the diagnosis. However, over time (approximately one year) the paralyzed hemidiaphragm becomes less compliant and less responsive to the pull of negative pleural pressure from the contralateral side. This reduces the amount of airflow from the paralyzed to the normal side. (See 'Sniff test' below.)

Normal diaphragm structure and function are described separately. (See "Eventration of the diaphragm in adults", section on 'Normal diaphragm anatomy and function'.)

Etiology — The causes of unilateral diaphragmatic paralysis are similar to that of bilateral involvement (table 1). However, the proportional likelihood of each of the causes is different and intrathoracic disease (or injury) of the phrenic nerve is added to the list of unilateral diaphragmatic paralysis etiologies.

In most cases, phrenic nerve injury or damage, most often due to cardiothoracic surgery, is the most likely reason for unilateral paralysis. Primary myopathies, systemic conditions, or central nervous system or neuromuscular junction diseases rarely cause isolated unilateral paralysis or weakness.

Cardiothoracic surgery — Phrenic nerve injury due to stretching or cooling during cardiac surgery is the most common etiology of unilateral diaphragmatic paralysis [11]. Up to 20 percent of cardiac bypass cases can be complicated by left hemidiaphragmatic paralysis [12], explaining why left-sided paralysis is more commonly encountered than right-sided involvement.

Unilateral paralysis has also been described following cardiac radiofrequency ablation [13].

Thoracic surgery involving the mediastinum including pacemaker placement can also directly injure or inflame the phrenic nerve to cause unilateral diaphragmatic paralysis or weakness [14].

Further details regarding neurologic injury of the phrenic nerve following cardiac surgery are found elsewhere. (See "Neurologic complications of cardiac surgery", section on 'Phrenic nerve'.)

Viral or infectious neuritis — Herpes zoster, poliomyelitis, zika virus, human immune deficiency virus, SARS-CoV-2 (COVID-19), Lyme disease, and other infections have been associated with unilateral diaphragmatic paralysis [15,16].

Cervical spine and neck disease — Unilateral phrenic nerve injury due to cervical spine (at the C3 to C5 level) and neck disease is an infrequent cause of unilateral diaphragmatic paralysis. These disorders include cervical spondylosis, cervical compressive tumors, blunt neck trauma, neck surgery, syringomyelia, and iatrogenic bronchial artery embolization [17-19]. (See "Evaluation and management of life-threatening hemoptysis", section on 'Efficacy and adverse effects'.)

Brachial plexus injury — In a minority of patients with neuralgic amyotrophy, an inflammatory disorder of the brachial plexus, the phrenic nerve is affected, causing unilateral diaphragmatic paralysis [20]. A related condition, hereditary brachial plexopathy (MIM 162100) can also be associated with phrenic nerve injury. Diaphragmatic paralysis complicates about 5 percent of congenital brachial plexus palsies. Nerve blocks of the brachial plexus can also be associated with unilateral diaphragm paralysis [21]. (See "Brachial plexus syndromes", section on 'Neuralgic amyotrophy' and "Brachial plexus syndromes", section on 'Hereditary brachial plexopathy' and "Diaphragmatic paralysis in the newborn", section on 'Birth injury'.)

Intrathoracic disease — Intrathoracic disease or trauma such as pneumonia, tumors, fibrosis (eg, sclerosing mediastinitis), or mediastinal chest tubes can compress, inflame, or injure the phrenic nerve locally to cause unilateral paralysis [22]. For example, about 5 percent of lung cancers involve the phrenic nerve [12]. (See "Clinical manifestations of lung cancer", section on 'Intrathoracic clinical manifestations' and "Mediastinal granuloma and fibrosing mediastinitis", section on 'Clinical presentation'.)

Miscellaneous and idiopathic — Rare case reports of unilateral diaphragmatic disease due to cholecystectomy, or paravertebral nerve blocks have been reported [23,24].

Up to 20 percent cases of unilateral diaphragmatic paralysis are idiopathic [12].

CLINICAL MANIFESTATIONS — Most patients with unilateral diaphragmatic paralysis present with an elevated hemidiaphragm as an incidental finding on chest radiography. (See 'Incidental finding on chest radiography (ie, asymptomatic)' below and 'Chest radiograph' below.)

Symptoms are more likely to occur in patients with comorbidities (eg, obesity) or underlying acute or chronic cardiopulmonary disease (eg, asthma, pneumonia, chronic obstructive pulmonary disease [COPD], lung fibrosis); in such patients, the increase in respiratory load required for the work of breathing may fatigue the already overburdened accessory muscles, leading to symptoms and signs. (See 'Less common presentations' below.)

Details regarding clinical manifestations of patients with respiratory muscle weakness from neuromuscular disease are provided separately. (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Clinical manifestations'.)

Incidental finding on chest radiography (ie, asymptomatic) — Most patients with unilateral diaphragmatic paralysis are asymptomatic and present with an elevated hemidiaphragm on chest imaging (chest radiography or computed tomography) obtained for another reason (image 1 and image 2).

Less common presentations — Less commonly, patients may have the following [7,9,25-27]:

●Dyspnea – Some patients present with exertional dyspnea and decreased exercise performance since inefficient or lack of hemidiaphragm contraction contributes to the exercise capacity limitation in these patients [28]. Progressive dyspnea is unusual unless associated with a progressive neuromuscular disorder. Patients may also complain of dyspnea at rest or dyspnea when in the lateral recumbent position with the paralyzed side down.

●Orthopnea – Orthopnea is an unusual symptom and is less intense compared with bilateral diaphragm paralysis. (See "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Clinical manifestations'.)

●Sleep disturbance – Some patients have symptoms of sleep disruption (eg, fatigue, daytime somnolence, snoring). This is thought to be due to apneas and hypopneas during rapid eye movement sleep and can also result in nocturnal hypoventilation (ie, hypercapnia). (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Nocturnal hypoventilation'.)

●Respiratory failure – Respiratory failure, typically with hypercapnia and hypoxemia, is a rare presentation of unilateral diaphragm paralysis. Respiratory failure is more likely to occur when chronic lung disease is progressive (chronic respiratory failure) or an acute illness develops (eg, acute respiratory failure due to an asthma or COPD exacerbation, or pneumonia).

Ventilatory failure may be reversible if the lung injury is reversible. However, in patients with progressive lung or neuromuscular disease, the symptoms may require long-term ventilator support. (See "Noninvasive ventilation in adults with chronic respiratory failure from neuromuscular and chest wall diseases: Patient selection and alternative modes of ventilatory support".)

Physical examination and laboratory findings — Physical examination findings are generally nonspecific but may reveal dullness to percussion and reduced air entry on the affected side.

Generally, laboratory tests are nonspecific, unless paralysis is associated with a systemic neuromuscular disorder, endocrine disorder (eg, hypothyroidism), or connective tissue disease (eg, systemic lupus erythematosus). However, these findings are rare in unilateral disease since these disorders are more likely to cause bilateral rather than unilateral diaphragmatic paralysis.

Details on the clinical manifestations of bilateral diaphragmatic paralysis and neuromuscular disease are provided separately. (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Clinical manifestations' and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Clinical manifestations'.)

DIAGNOSTIC EVALUATION

General approach — In patients with suspected unilateral hemidiaphragmatic paralysis, we obtain an upright, inspiratory, posteroanterior chest radiograph, if not already performed, to look for an elevated hemidiaphragm (algorithm 1). (See 'Chest radiograph' below.)

●Regardless of symptoms, if unilateral elevation of the hemidiaphragm is present on chest imaging, we proceed with a fluoroscopic (or ultrasound) sniff test, chest computed tomography, spirometry (in the supine and sitting positions) and respiratory muscle strength testing. (See 'Chest computed tomography' below and 'Sniff test' below and 'Pulmonary function testing' below.)

●If paralysis is confirmed and the etiology clear, we do not necessarily proceed with additional testing (eg, post cardiac surgery unilateral diaphragmatic paralysis). If the etiology is unclear, we typically perform additional imaging (eg, chest and neck computed tomography [CT]) to facilitate the exclusion of conditions that can cause an elevated hemidiaphragm on imaging and to identify a potential underlying etiology for the paralysis. (See 'Follow-up testing' below.)

History, physical, and laboratory evaluation — Unilateral diaphragmatic paralysis should be suspected in individuals who present with an elevated hemidiaphragm on chest imaging. (See 'Clinical manifestations' above.)

When suspected, we inquire about dyspnea during previously tolerated activities, sleep disturbance, or positional dyspnea. We also inquire about any history of childhood poliomyelitis or other serious viral illnesses (eg, Lyme disease), diabetes, neurological illnesses, remote history of trauma, previous cardiothoracic surgery (especially if the paralysis is left-sided), chest, neck, or cervical spine injury, manipulation, or irradiation. We inquire about upper extremity weakness, known chest or neck pathology, and shoulder pain or paresthesia of the arm. We also ask about symptoms that might suggest lung cancer (eg, weight loss, cough, hemoptysis). (See "Clinical manifestations of lung cancer".)

Examination may reveal dullness to percussion and reduced air entry into the affected lung base to support the diagnosis [29]. The examination may also suggest a potential cause such as upper extremity weakness, wasting, and paraesthesia, to support brachial plexus or cervical disease, lymphadenopathy to suggest possible underlying tumor, or scars consistent with prior surgery.

We generally obtain a complete blood count to assess whether anemia is playing a role in dyspnea. Testing for etiologies that may underlie diaphragmatic paralysis are discussed below. (See 'Follow-up testing' below.)

Imaging — Imaging involves chest radiography and in suspected cases of diaphragmatic paralysis, chest computed tomography and sniff testing.

Chest radiograph — When suspected clinically, and chest radiography is not already performed, we obtain a frontal, inspiratory, upright and lateral, chest radiograph to look for an elevated hemidiaphragm.

A chest radiograph that shows an elevated hemidiaphragm is sensitive, but not specific for the diagnosis of unilateral diaphragm paralysis (image 1 and image 2) [30], the differential of which is discussed below. (See 'Differential diagnosis' below.)

Normally, the dome of the right hemidiaphragm projects anteriorly over the 5th or 6th rib and posteriorly over the 10th rib, and the dome of the left hemidiaphragm is up to one interspace lower than the right. In patients with unilateral disease, the dome of the affected diaphragm is elevated and smooth in contour. The costophrenic and costovertebral sulci are deep and narrow on the affected side. Small lung volumes and plate-like atelectasis may be seen on the affected side.

Chest computed tomography — In patients with an elevated hemidiaphragm on chest radiography, we generally perform chest computed tomography (CT).

●The main purpose of chest CT is to help confidently exclude other etiologies associated with an elevated hemidiaphragm (eg, subpleural effusion or mass) and conditions that mimic respiratory muscle weakness (restriction due to interstitial lung disease). The value to CT in this regard is discussed separately. (See 'Differential diagnosis' below.)

●Chest CT is also useful in patients diagnosed with unilateral diaphragmatic paralysis to look for causative etiologies (eg, mediastinal or other tumor affecting the phrenic nerve (table 1)). (See 'Follow-up testing' below.)

●Chest CT also allows assessment of diaphragm thickness, which is most reliably measured at the level of the diaphragmatic crus (image 2) [31]. Unilateral diaphragmatic paralysis typically leads to atrophy of the ipsilateral crus, a finding which is seldom seen in other causes of diaphragmatic elevation. However, thinning of the diaphragmatic crus on the affected side may not be reliably seen in early paralysis and exact parameters of normal diaphragm thickness are poorly defined. In our opinion, values below 1.5 mm and an obvious disparity in thickness between one side and the other support the diagnosis.

In a retrospective study of 72 patients who had undergone diaphragmatic fluoroscopic sniff testing, thinning of the crus to ≤2.5 mm at the L1 vertebral level on CT identified paralysis of the hemidiaphragm with a sensitivity of 100 percent and a specificity of 88 percent on the right and with a sensitivity of 100 percent and a specificity of 77 percent on the left [31]. Patients with diaphragmatic paralysis who had follow-up CT showed progressive crus atrophy, suggesting that patients with recent paralysis may not demonstrate crus atrophy.

Sniff test — To test whether the observed hemidiaphragm elevation on imaging is due to diaphragmatic paralysis (complete or partial), we perform sniff testing. We prefer a fluoroscopic approach, although ultrasonography is an alternative or can be used in equivocal cases:

●Fluoroscopic sniff testing – In a fluoroscopic "sniff" test, diaphragm movement is observed fluoroscopically while the patient sniffs forcefully in the standing position [32]. A positive test is one that shows paradoxical elevation or no movement of the paralyzed hemidiaphragm with inspiration compared with the rapid descent of the normal hemidiaphragm.

The sniff test is positive in over 90 percent of patients with unilateral diaphragm paralysis [33]. We use a positive test in conjunction with spirometry and respiratory muscle strength testing to make a clinical diagnosis of unilateral diaphragm paralysis. (See 'Diagnosis' below.)

If the sniff test shows normal descent of the hemidiaphragm, we evaluate the patient for other etiologies associated with unilateral elevation of the diaphragm. (See 'Unilateral elevation of the hemidiaphragm' below.)

●Ultrasonographic sniff testing – A sniff test using M-mode ultrasonography can demonstrate the same paradoxical elevation or no movement during the sniff maneuver [33-36]. In addition to imaging diaphragmatic excursion, diaphragm thickness can also be assessed on ultrasonography.

Some experts prefer ultrasonography since it can be readily performed at the bedside (unlike fluoroscopy) and can be used when fluoroscopy is not feasible (eg, the office, pulmonary function testing (PFT) laboratory, hospital department, and intensive care unit). For the same reason it can be useful during periods of acute illness, when transfer to the fluoroscopy suite in not feasible. Ultrasonography can also be used in the supine position; the latter can exaggerate the inability to move the diaphragm during inspiration, thereby enhancing the ability to detect paralysis or weakness.

However, the small field of view and obscuration of the diaphragm by intervening lung or bowel air can make assessment of diaphragm function and thickness difficult. Furthermore, this technique requires operator expertise.

●Other imaging modality – Diaphragmatic motion can also be assessed with dynamic magnetic resonance (MR) imaging [34,35]. This technique, however, is not widely used because of limited availability and high cost.

Some patients who have difficulty with the sniff maneuver (eg, neuromuscular disease) or patients with unclear results from the sniff test may require more definitive diaphragmatic testing (eg, electromyography [EMG], transdiaphragmatic pressure [Pdi], or phrenic nerve conduction studies [PNCS]). These studies are discussed separately. (See 'Specific diaphragm tests' below and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Measurement of transdiaphragmatic pressure' and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Diaphragmatic electromyography'.)

Pulmonary function testing — In patients with suspected unilateral diaphragmatic paralysis, we perform spirometry (in the supine and sitting positions), lung volumes, and diffusing capacity for carbon monoxide (DLCO). We do not routinely obtain maximum voluntary ventilation (MVV), unless needed for cardiopulmonary exercise testing. The degree of pulmonary dysfunction generally correlates with severity of the respiratory muscle weakness. We also obtain tests of respiratory muscle strength, and peripheral oxygen saturation, both in the supine and sitting positions.

Spirometry (supine and sitting), volumes, diffusion — When unilateral diaphragm paralysis is present, a pattern of restriction is typically seen, provided no other physiologic or abnormal lung or pleural pathology is present. A reduced forced vital capacity (FVC), vital capacity (VC), and total lung capacity (TLC) with normal or elevated forced expiratory volume in one second (FEV₁)/FVC ratio (greater than 70 percent of predicted) and DLCO is usually seen. (See "Overview of pulmonary function testing in adults" and "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Pulmonary function testing'.)

The FVC and VC in the upright sitting position can be decreased to values in the range of 70 to 80 percent of predicted, a less pronounced reduction than that seen with bilateral disease (see "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Supine and sitting spirometry, volumes, diffusion'). The FVC typically decreases further by greater than 10 percent in the supine position (normal is 10 percent or less). The effect of the supine position on FVC and VC is more pronounced with paralysis of the right than the left hemidiaphragm, possibly due to the larger pressure exerted by the liver during inspiration when lying flat.

Tests of respiratory muscle strength — In patients with suspected unilateral diaphragm paralysis, we perform maximal inspiratory pressure (MIP or PImax) and maximal expiratory pressure (MEP or PEmax). We do not routinely perform sniff nasal pressure (SNIP), since it provides no real additional information other than that provided by MIP; however, other experts use this test as a supplement to MIP [37].

The MIP is usually decreased to values of 60 percent of predicted (normal values are more negative than -60 cm H₂O) [38]. Rarely, the MIP is normal because of preserved strength of the accessory muscles.

The MEP is usually normal but may be slightly reduced.

An elevated MEP to MIP ratio (MEP/MIP) >1.5 is supportive of the diagnosis. In one study, a MEP/MIP of 1.5 had a sensitivity of 87 percent and a specificity of 45 percent for detecting unilateral diaphragmatic paralysis [39].

Further details regarding performance and interpretation of respiratory muscle strength tests are provided separately. (See "Tests of respiratory muscle strength".)

Oxygenation — We measure oxygen saturation in the sitting and supine position. Oxygenation is typically normal but may occasionally be reduced in the supine position (eg, with the paralyzed side down). While not sensitive or specific, a positive result may help support the diagnosis.

We do not routinely perform arterial blood gas (ABG) analysis since it is unusual for patients with unilateral diaphragm paralysis to develop an elevated arterial partial pressure of carbon dioxide (PaCO₂) in the absence of accompanying lung disease, or nocturnal hypoventilation during sleep [40]. However, we do obtain an ABG when hypoventilation is suspected (eg, those with an elevated bicarbonate or patients with symptoms suggestive of sleep disruption). (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Inadequate ventilation'.)

Patients with equivocal findings — In most patients the diagnosis of unilateral diaphragmatic paralysis can be made, using an approach that involves chest radiography, PFTs, the sniff test and respiratory muscle strength testing [41,42]. However, in patients with equivocal findings in whom a diagnosis of unilateral diaphragmatic paralysis cannot be confidently made, we perform diaphragm testing, which is targeted at measuring diaphragm function and innervation.

Diaphragmatic testing is generally more specific for the diagnosis of diaphragm paralysis. It is generally invasive, has associated risks (eg, pneumothorax, lung perforation), is technically difficult to perform, and requires specialized expertise that is not universally available. Thus, a certain amount of clinical judgement is generally required when deciding to proceed with such testing (eg, unexplained symptoms, patient desire for a diagnosis, differentiating the contribution of paralysis to symptoms, patients with suspected bilateral involvement). (See 'Specific diaphragm tests' below and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Measurement of transdiaphragmatic pressure' and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Diaphragmatic electromyography'.)

Specific diaphragm tests — Diaphragm electromyography (EMG), measurement of transdiaphragmatic pressures (Pdi), or performance of phrenic nerve conduction studies (PNCS) are not necessary for routine evaluation of unilateral diaphragm weakness, unless the diagnosis is in doubt or surgical repair is planned. (See 'Severe symptoms' below.)

Diaphragmatic testing is described separately. (See "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Measurement of transdiaphragmatic pressure' and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Diaphragmatic electromyography'.)

Others — While investigational, measurement of the lung sound intensity (LSI) in both the right and left hemithorax using acoustic respiratory analysis has been proposed as a simple noninvasive method to help confirm the diagnosis of unilateral diaphragmatic paralysis [43]. It has also been suggested that LSI is useful for monitoring changes over time or after therapy. (See 'Treatment and prognosis' below.)

DIFFERENTIAL DIAGNOSIS — The differential diagnosis of unilateral diaphragmatic paralysis includes disorders known to give the appearance of an elevated hemidiaphragm on chest radiography and other diseases that are in the differential for patients who present with respiratory muscle weakness.

Unilateral elevation of the hemidiaphragm — The differential diagnosis of an elevated hemidiaphragm on chest radiography includes the following:

●A subpulmonic pleural effusion or mass. (See "Pleural fluid analysis in adults with a pleural effusion".)

●Diaphragmatic eventration. (See "Eventration of the diaphragm in adults".)

●Diaphragmatic hernia (rare in adults). (See "Congenital diaphragmatic hernia in the neonate".)

●Unilateral subdiaphragmatic processes (eg, abscess, splenomegaly, ascites, hepatomegaly). (See "Pyogenic liver abscess", section on 'Imaging'.)

●Lobar atelectasis or previous lobectomy. (See "Atelectasis: Types and pathogenesis in adults" and "Radiologic patterns of lobar atelectasis" and "Overview of pulmonary resection".)

This differential can be narrowed significantly using existing clinical findings and diagnostic tests.

●Chest imaging – For example, most of these etiologies can be readily identified on chest computed tomography (CT). Useful clues on chest imaging include the following:

•Subpulmonic effusions or masses may be associated with an irregular (rather than smooth) contour of the diaphragm, and diaphragm elevation may be eccentric (rather than central) and associated with blunting of the costophrenic angles.

•Diaphragmatic hernias are either anterior (Morgagni) or posterior (Bochdalek) which can usually be best appreciated on the lateral chest radiograph or on CT.

•Differentiating partial diaphragm eventration from unilateral diaphragm paralysis is typically clear radiologically by the demonstration of partial involvement of the diaphragm (suggests partial eventration) versus complete involvement (suggests unilateral paralysis).

However, distinguishing paralysis from severe diaphragm eventration (ie, eventration that involves the entire hemidiaphragm) is more challenging. Clues such as a smooth contour and lack of paradoxical movement during inspiration on sniff testing may favor eventration over paralysis.

If the distinction remains unclear, diaphragmatic testing can help distinguish these two entities [38]. For example, phrenic nerve conduction should be intact in cases of eventration, but not in cases of unilateral diaphragm paralysis due to phrenic nerve injury. However, we do not typically perform invasive diaphragmatic testing to distinguish these entities since the treatment is similar, rendering the distinction irrelevant. (See "Eventration of the diaphragm in adults" and 'Specific diaphragm tests' above.)

●Other diagnostic tests – In addition, with the exception of diaphragmatic eventration, the conditions in the differential are not associated with pulmonary function test (PFT) abnormalities, evidence of respiratory muscle weakness, or abnormal sniff testing.

●Clinical clues – Clinical findings such as fever, abdominal pain, and leukocytosis may support subdiaphragmatic abscess. Further imaging may be required to fully exclude such conditions (eg, abdominal imaging for liver or splenic pathology).

Disorders that mimic respiratory muscle weakness — The differential diagnosis that is considered in patients with respiratory muscle weakness should also be considered in patients with suspected unilateral diaphragmatic paralysis. This includes restrictive lung disorders (for patients with restriction on PFTs), poor effort (for patients with reduced respiratory muscle strength testing), heart failure (for patients who present with orthopnea), sleep apnea (for those with symptoms of sleep-disordered breathing), and disorders that cause hypercapnia (if hypercapnia is present). Distinguishing these disorders is discussed in detail separately. (See "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Differential diagnosis'.)

DIAGNOSIS — The diagnosis of unilateral diaphragm paralysis is usually made by a constellation of typical findings including all of the following:

●An elevated hemidiaphragm on upright, frontal and lateral chest radiograph. (See 'Chest radiograph' above.)

●Paradoxical, limited, or no movement of the affected hemidiaphragm on sniff testing using fluoroscopy or ultrasonography. (See 'Sniff test' above.)

●Evidence of reduced forced vital capacity and/or vital capacity (on spirometry) and inspiratory muscle weakness (on respiratory muscle strength testing). (See 'Spirometry (supine and sitting), volumes, diffusion' above and 'Tests of respiratory muscle strength' above.)

●Reasonable exclusion of other etiologies that can cause an elevated hemidiaphragm or mimic respiratory muscle weakness. (See 'Differential diagnosis' above.)

In most cases there is sufficient information from the clinical history, chest imaging, sniff testing, pulmonary function tests, and respiratory muscle strength testing that a diagnosis can be made without further testing.

Further testing may be needed when the diagnosis is in doubt (eg, specific diaphragm testing) or when other etiologies of an elevated hemidiaphragm or conditions that mimic respiratory neuromuscular weakness need to be confidently excluded (eg, echocardiography for heart failure, computed tomography of the abdomen for subdiaphragmatic pathology). (See 'Unilateral elevation of the hemidiaphragm' above and 'Disorders that mimic respiratory muscle weakness' above and 'Specific diaphragm tests' above and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Specific diaphragmatic tests'.)

FOLLOW-UP TESTING — We adopt the following approach to follow-up testing:

●In many patients, the etiology of unilateral diaphragmatic paralysis is clear (eg, recent cardiac surgery) and no further testing is needed.

●For patients with unilateral diaphragm paralysis in whom the etiology is unclear, we perform further testing. (See 'Investigating an underlying etiology' below.)

●For patients with symptoms suggestive of sleep-disordered breathing, we also perform sleep testing to look for sleep apnea as a potential complication of unilateral diaphragmatic paralysis. (See 'Polysomnography' below.)

Investigating an underlying etiology — There are several etiologies associated with unilateral diaphragmatic paralysis or weakness (table 1). (See 'Etiology' above.)

For patients with a diagnosis of unilateral diaphragmatic paralysis in whom the underlying etiology is unclear, we obtain neck and chest computerized tomography (CT; if not already performed), preferably with contrast, to evaluate the course of the ipsilateral phrenic nerve and exclude involvement by a tumor or other disease process.

For patients in whom other local pathology is suspected (eg, cervical or brachial plexus disease), we obtain additional imaging (CT or magnetic resonance imaging of the cervical spine or shoulder). (See "Clinical features and diagnosis of cervical radiculopathy", section on 'Imaging studies'.)

We also obtain thyroid tests and a connective tissue panel, though arguably such tests are often not necessary since such disorders are more likely to be associated with bilateral than unilateral paralysis. (See "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Etiology'.)

Polysomnography — Unilateral diaphragmatic paralysis can be complicated by sleep-disordered breathing. While some experts routinely perform sleep testing in all patents with unilateral diaphragm paralysis, we first inquire about symptoms that are consistent with sleep disruption (eg, daytime sleepiness, fatigue) and assess the severity of respiratory impairment.

If symptoms suggest sleep-disordered breathing and/or respiratory impairment is severe (eg, FVC or MIP <50 percent of predicted) (table 2), we perform overnight in-laboratory polysomnography. We do not perform home sleep apnea testing (HSAT) since HSAT has a low sensitivity for the detection of central events, which have an increased likelihood of being present in this population. (See "Evaluation of sleep-disordered breathing in adult patients with neuromuscular and chest wall disorders", section on 'Diagnostic approach' and "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Diagnostic tests'.)

Overnight pulse oximetry may be used as a screening test for nocturnal oxygen desaturation but is generally not adequate screening for sleep-disordered breathing in these patients, as it underestimates the degree of hypoventilation and sleep disturbance.

TREATMENT AND PROGNOSIS — In most patients with nontraumatic unilateral diaphragmatic paralysis, we suggest periodic clinical observation and treatment of the underlying disease, if feasible (algorithm 2). This approach is based upon the rationale that most patients are asymptomatic or have mild symptoms and have an excellent prognosis, unless the paralysis is caused by an inherently fatal illness (eg, metastatic lung cancer) [36]. In addition, partial or complete resolution is expected in many patients, generally over a period of one year or longer, particularly patients with reversible conditions (eg, phrenic nerve injury due to cardiac surgery or trauma, neuralgic amyotrophy). Surgery, typically ipsilateral diaphragm plication, is reserved for those with severe symptoms (in the absence of underlying disease) who do not improve after 6 to 12 months.

The management of non-trauma-related unilateral hemidiaphragm paralysis is discussed in this section. The approach to surgical repair of the phrenic nerve due to trauma is discussed separately. (See "Surgical treatment of phrenic nerve injury".)

Treat underlying cause — For patients with an underlying reversible cause, we suggest targeting therapy to that pathology. For example, patients with cervical compression due to tumors may respond to debulking or decompression; patients with viral neuropathy may respond to targeted antiviral therapy with or without glucocorticoids. (See "Cervical spinal column injuries in adults: Evaluation and initial management".)

Patients with traumatic transsection of the phrenic nerve may be suitable for early surgical repair of the phrenic nerve with or without diaphragmatic plication. Rarely, the phrenic nerve can be repaired after injury by transposition or interposition of a nerve graft during the later phase of trauma recovery (ie, 6 to 12 months) [44]. Further details are provided separately. (See "Surgical treatment of phrenic nerve injury", section on 'Phrenic nerve reconstruction'.)

General therapies and rehabilitation — We treat patients with routine vaccinations, counsel against smoking or vaping, and administer oxygen, if indicated. (See "Standard immunizations for nonpregnant adults" and "Overview of smoking cessation management in adults" and "Long-term supplemental oxygen therapy".)

During the initial observation period, we also refer patients with mild or moderate symptoms for rehabilitation and diaphragmatic strength training. However, the impact of this on clinical outcomes is unclear. (See "Pulmonary rehabilitation", section on 'Ventilatory muscle training' and "Respiratory muscle training and resting in COPD".)

Asymptomatic or mild symptoms — Most patients with unilateral diaphragmatic paralysis are asymptomatic or have mild symptoms and require no treatment. The prognosis is good in this setting, and the paralysis is of little clinical relevance in the absence of new or underlying pulmonary disease [8,9,45]. Most patients are expected to improve over 6 to 12 months unless the etiology is progressive (eg, widespread cancer).

Clinical observation — We follow patients clinically every six months for the first year and periodically after that. At each visit we obtain pulmonary function tests and tests of respiratory muscle strength until they plateau or normalize.

Most patients, even those with idiopathic disease, recover partially or completely over that period of time (eg, patients with phrenic nerve damage following cardiac surgery, neuralgic amyotrophy, viral neuropathy). However, the expected time for recovery is unpredictable. Nonetheless, even those who do not recover diaphragm function can still generally lead a normal life, although breathlessness may occur during situations of increased ventilatory demands (eg, acute intercurrent illness or exercise). (See 'Intermittent symptoms with acute reversible illness' below.)

Severe symptoms — It is rare for patients with unilateral diaphragmatic paralysis to have severe symptoms and when present, they should prompt evaluation for etiologies other than diaphragmatic paralysis.

However, some patients may be severely impaired enough to warrant consideration of therapy, which is typically surgical plication (eg, patients unable to wean from mechanical ventilation). It is prudent to wait for at least 6 to 12 months before proceeding with surgical therapy since some patients may improve spontaneously during this time period.

Surgical plication — Surgical plication of the affected hemidiaphragm is successful in carefully selected patients with severe symptoms considered to be due to unilateral diaphragm paralysis [46-54]. Studies demonstrate an improvement in several parameters including lung and respiratory muscle function, exercise endurance, blood gas exchange, and possibly dyspnea [52,55-57].

Before subjecting patients to invasive surgery, preoperative evaluation should include the performance of specific diaphragm testing to definitively confirm unilateral paralysis. Preoperative evaluation, patient selection, and risk involved are otherwise similar to that of thoracic procedures in general. (See "Overview of minimally invasive thoracic surgery" and "Anesthesia for video-assisted thoracoscopic surgery (VATS) for pulmonary resection" and "Evaluation of perioperative pulmonary risk".)

The procedure involves creating folds in the diaphragm and suturing them in place to reduce paradoxical mobility of the paralyzed hemidiaphragm during inspiration (figure 2) [47,56-59]. After plication, the paralyzed hemidiaphragm no longer moves paradoxically upward into the hemithorax during inspiration and abdominal ribcage synchrony improves bilaterally, allowing the adjacent lung segments on the affected side to expand appropriately and improve gas exchange [57,59]. Furthermore, the healthy hemidiaphragm performs less work, rendering it less susceptible to fatigue and decreases the need to recruit accessory muscles of ventilation [57].

Postoperatively, patients are routinely managed in the intensive care unit since many patients require temporary mechanical ventilation. Extubation is high risk as there is a change in thoracic and abdominal compliance with the surgical incision and reduction of contents into the abdominal cavity. (See "Overview of the management of postoperative pulmonary complications" and "Strategies to reduce postoperative pulmonary complications in adults" and "Extubation management in the adult intensive care unit".)

Other — We do not generally perform diaphragmatic pacing in patients with unilateral disease. Competing input from the pacing device and the natural variable impulse of the normally functioning hemidiaphragm is uncomfortable and leads to dyspnea as well as increased work of breathing. (See "Pacing the diaphragm: Patient selection, evaluation, implantation, and complications".)

Intermittent symptoms with acute reversible illness

Transient ventilatory support — Occasionally, patients with unilateral diaphragm paralysis will develop acute respiratory insufficiency due to, for example, an intercurrent respiratory infection, exacerbation of asthma or chronic obstructive pulmonary disease, or general anesthesia. These patients may require temporary ventilatory support with noninvasive ventilation (NIV) or invasive mechanical ventilation. Patients are typically able to be weaned from ventilatory support once the intercurrent illness is resolved. (See "Noninvasive ventilation in adults with acute respiratory failure: Benefits and contraindications" and "Respiratory muscle weakness due to neuromuscular disease: Management" and "Respiratory muscle weakness due to neuromuscular disease: Management", section on 'Acute ventilatory support'.)

Sleep-disordered breathing — Patients with sleep-disordered breathing may warrant therapy with NIV tailored to the specific type of sleep disturbance. (See "Central sleep apnea: Treatment".)

SUMMARY AND RECOMMENDATIONS

●Etiology – Unilateral diaphragmatic paralysis is more common than bilateral paralysis. Causes of unilateral diaphragm paralysis in adults (table 1) include phrenic nerve injury due to stretching or cooling during cardiac surgery, viral infections (eg, Herpes zoster, poliomyelitis), neck or cervical spine disease (eg, cervical spondylosis, cervical compressive tumors, blunt neck trauma or surgery), brachial plexopathies, and intrathoracic disease (eg, lung cancer). Up to 20 percent of cases are idiopathic. (See 'Pathophysiology and etiology' above.)

●Clinical features – Most patients with unilateral diaphragmatic paralysis are asymptomatic and present with an elevated hemidiaphragm as an incidental finding on chest radiography. Symptoms such as dyspnea on exertion, positional dyspnea, orthopnea, and ventilatory failure are more likely to occur in patients with comorbidities (eg, obesity or underlying acute or chronic cardiopulmonary disease). Rarely, patients present with symptoms of sleep-disordered breathing. (See 'Clinical manifestations' above and "Clinical presentation and diagnosis of obstructive sleep apnea in adults", section on 'Clinical features'.)

●Evaluation – In patients with suspected unilateral diaphragmatic paralysis, we obtain a frontal upright inspiratory chest radiograph (if not already performed) to look for an elevated hemidiaphragm (algorithm 1). If present, we proceed with a chest computed tomography (CT), fluoroscopic (or ultrasound) sniff test, spirometry (in the supine and sitting positions), and respiratory muscle strength testing. (See 'Diagnostic evaluation' above.)

●Diagnosis – The diagnosis of unilateral diaphragm paralysis or weakness is usually made when all of the following are present:

•An elevated hemidiaphragm on upright, frontal and lateral chest radiograph. (See 'Chest radiograph' above.)

•Paradoxical, limited, or no movement of the affected hemidiaphragm on sniff testing. (See 'Sniff test' above.)

•Evidence of reduced forced vital capacity and/or vital capacity (eg, 70 to 80 percent predicted, decrease in FVC >10 percent in the supine position) and reduced maximal inspiratory pressure (eg, 60 percent predicted) on pulmonary function and respiratory muscle strength testing. (See 'Spirometry (supine and sitting), volumes, diffusion' above and 'Tests of respiratory muscle strength' above and "Tests of respiratory muscle strength" and "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Respiratory muscle strength testing'.)

•Reasonable exclusion of other etiologies that can cause an elevated hemidiaphragm on imaging (eg, subpulmonic pleural effusion or other process, diaphragm eventration or hernia, lobar atelectasis, or lobectomy) or mimic respiratory muscle weakness (eg, restrictive lung disorders, poor effort, heart failure, disorders associated with hypercapnia). Most of these conditions are distinguished by the evaluation and standard diagnostic testing. (See 'Differential diagnosis' above and "Respiratory muscle weakness due to neuromuscular disease: Clinical manifestations and evaluation", section on 'Differential diagnosis'.)

Further diagnostic testing may be needed when the diagnosis is in doubt (eg, specific diaphragm testing) or when other etiologies of an elevated hemidiaphragm or conditions that mimic respiratory neuromuscular weakness are suspected or need to be excluded. (See 'Specific diaphragm tests' above and "Diagnostic evaluation of adults with bilateral diaphragm paralysis", section on 'Specific diaphragmatic tests'.)

●Follow-up – In some cases, the etiology of diaphragmatic paralysis is obvious (eg, recent cardiac surgery). However, if the etiology is unclear, we perform additional testing, such as neck CT (and chest CT if not already performed), magnetic resonance imaging of the cervical spine, or shoulder imaging. When sleep-disordered breathing is suspected on the basis of symptoms or the degree of respiratory impairment, we obtain in-laboratory polysomnography (table 2). (See 'Follow-up testing' above and 'Investigating an underlying etiology' above.)

●Treatment and prognosis – The following is a reasonable approach to treatment (algorithm 2):

•Asymptomatic or mild symptoms – For most patients with unilateral diaphragm paralysis, we suggest periodic clinical observation over 12 months (Grade 2C). In most cases this provides time for the underlying etiology to be treated or spontaneously resolve. We also treat patients with routine vaccinations, counsel against smoking or vaping, and administer oxygen, if indicated. This approach is predicated on the rationale that most patients are asymptomatic or have mild symptoms, spontaneously improve, and have an excellent prognosis in the absence of progressive lung disease or other fatal disorder. (See 'Treat underlying cause' above and 'General therapies and rehabilitation' above and 'Asymptomatic or mild symptoms' above.)

•Severe symptoms – In the rare patient who has significant symptoms that are persistent or unresolving beyond 6 to 12 months, we suggest ipsilateral surgical diaphragm plication (Grade 2C). Observational data suggest that plication improves pulmonary function, gas exchange, and may improve symptoms. (See 'Treatment and prognosis' above and 'Severe symptoms' above and 'Surgical plication' above.)

•Sleep-disordered breathing or acute respiratory failure – If warranted, we treat patients with evidence of sleep-disordered breathing with noninvasive ventilation (NIV), and if indicated, administer short-term NIV or mechanical ventilatory support during intermittent episodes of acute respiratory failure. This is discussed separately. (See 'Intermittent symptoms with acute reversible illness' above and 'Sleep-disordered breathing' above and "Central sleep apnea: Treatment" and "Noninvasive ventilation in adults with acute respiratory failure: Benefits and contraindications" and "Respiratory muscle weakness due to neuromuscular disease: Management", section on 'Acute ventilatory support'.)

•Acute transection of the phrenic nerve – Patients with traumatic transsection of the phrenic nerve may be suitable for early surgical repair of the phrenic nerve with or without diaphragmatic plication. Further details are provided separately. (See "Surgical treatment of phrenic nerve injury", section on 'Phrenic nerve reconstruction'.)