The diaphragm, the most important muscle of ventilation, develops negative intrathoracic pressure to initiate ventilation. Innervated by cervical motor neurons C3-C5 via the phrenic nerves, these two nerves provide both sensory and motor function to the diaphragm.[1, 2] With contraction, the cone-shaped muscle of the diaphragm decreases intrapleural pressure during inspiration and thereby facilitates movement of air into the lungs. Diaphragmatic paralysis is an uncommon, yet underdiagnosed cause of dyspnea.
Diaphragmatic paralyses encompass a spectrum of diseases involving a single leaflet, known as unilateral diaphragmatic paralysis (UDP), and that involving both leaflets, known as bilateral diaphragmatic paralysis (BDP).
Although the diaphragm performs most of the work, normal ventilation also requires the simultaneous contraction of respiration accessory muscles (ie, scalene, parasternal portion of the internal and external intercostal muscles, sternocleidomastoid, trapezius). In bilateral diaphragmatic paralysis, accessory muscles assume some or all of the work of breathing by contracting more intensely. An increased effort in the struggle to breathe may fatigue the accessory muscles and lead to ventilatory failure.
Incidence is unknown.
Like diaphragm eventration, diaphragm paralysis is more common among males.
Depending on the etiology of the diaphragmatic paralysis, the prognosis of unilateral disease usually is excellent unless the patient has significant underlying pulmonary disease. Patients develop compensatory mechanisms, and patients with phrenic injuries may recover fully or partially. At times, patients may spontaneously recover from idiopathic disease. Patients who do not recover from unilateral diaphragmatic dysfunction generally lead relatively normal lives. In this group, dyspnea may develop with exertion, leading to increased ventilatory demands.
Bilateral diaphragmatic paralysis 
The prognosis depends on the nature of the underlying disease. Patient diaphragm function may recover if nerve injury is not permanent, while other patients may require long-term treatment as elaborated before. If recovery occurs, it usually takes considerable time, in excess of one year.
Unilateral diaphragmatic paralysis
The morbidity of the unilateral paralysis is mainly based on the underlying pulmonary functional status and the etiology of the paralysis. Because most cases of unilateral diaphragmatic paralysis are found incidentally during imaging studies, many patients have no symptoms. Diaphragmatic paralysis is more likely to affect the left hemidiaphragm. The patients with unilateral diaphrmatic paralysis that do have symptoms and decreased quality of life are those who have concurrent underlying lung diseases.
Bilateral diaphragmatic paralysis
Patients with bilateral diaphragmatic paralysis are usually symptomatic and, when symptoms are severe or in the presence of underlying lung pathology, may develop ventilatory failure without medical intervention.
Bilateral diaphragmatic paralysis is characterized by profound abnormalities of pulmonary and respiratory muscle function. Patients develop severe restrictive ventilatory impairment, and the vital capacity and total lung capacity frequently are below 50% of predicted for that patient. Lung capacity is reduced further when the patient assumes the supine position. Symptoms depend on whether the paralysis is unilateral or bilateral, how rapid the paralysis occurs, and the presence of underlying pulmonary disease.
Unilateral diaphragmatic paralysis is often discovered incidentally in patients undergoing chest radiography for some other reason. Patients usually are asymptomatic at rest but may experience dyspnea upon exertion and may have a decrease in exercise performance.
If the patient has an underlying lung disease, dyspnea may occur at rest.
Some patients may develop orthopnea, which is generally less intense than patients bilateral diaphragmatic paralysis.
Patients typically present with respiratory failure or dyspnea (should be distinguished from right-sided heart failure [fluid retention and dyspnea]) that worsens in the supine position. Tachypnea and rapid, shallow breathing occur when the patient adopts the recumbent position.
Patients also report anxiety, insomnia, morning headache, excessive daytime somnolence and fatigue, and poor sleep habits. In some patients, nonspecific GI symptoms such as heartburn, regurgitation, nausea, and epigastric pain can also develop.
In bilateral diaphragmatic paralysis, additional clues to diagnosis include dyspnea further exacerbated in a swimming pool or sleep that only occurs in the sitting position. The duration of symptoms prior to diagnosis can be quite long because of a lack of suspicion for the condition.
Physical examination findings depend on whether the paralysis is unilateral or bilateral. Generally, a breathing pattern of paradoxical abdominal wall retraction during inspiration occurs. The physician can evaluate the patient further by palpating under the costal margin and feeling for the descending hemidiaphragms during inspiration.
Patients reveal dullness to percussion and absent breath sounds over the lower chest on the involved side. Excursion on the involved hemithorax is decreased when compared with the healthy side.
Chest examination reveals limitation of diaphragmatic excursions and bilateral lower chest dullness with absent breath sounds. Patients are tachypneic and use accessory respiration muscles. The diagnostic finding is a paradoxical inward movement of the abdomen with inspiration.
The most common diagnosed cause is a malignant (ie, metastatic lung cancer) lesion leading to nerve compression (approximately 30% of patients).
If malignancy is not the cause, many times the etiology cannot be determined.
Other causes in the differential include blunt cervical trauma, surgical trauma (mainly thoracic), herpes zoster, cervical spondylosis, and supraclavicular brachial plexus block (which can be largely avoided with the use of ultrasound.) Upper cervical radiculopathies as a cause of Hemidiaphragmatic paralysis have also been reported.
The most common causes are secondary to motor neuron disease, including amyotrophic lateral sclerosis and postpolio syndrome.
Other causes include thoracic trauma, cardiac surgery, multiple sclerosis, myopathies, muscular dystrophy (acid maltase deficiency), Guillain-Barré syndrome, and Parsonage-Turner syndrome (neuropathy of brachial plexus). See Complications.
Phrenic nerve injury commonly occurs from cold cardioplegia or mechanical stretching during open-heart surgery.
Diaphragmatic dysfunction often occurs postoperatively in patients undergoing cardiac surgery. This has been attributed to pleurotomy in order to harvest internal mammary artery (IMA) grafts, which results in greater chest wall and parenchymal trauma, greater pain, and impairment of cough and deep breathing. In addition, IMA dissection may reduce blood supply to ipsilateral intercostal muscles and may cause mechanical injury to the phrenic nerve.
In the past, studies have confirmed phrenic nerve injury from cold-induced injury during myocardial protection, although in current practice most centers use warm cardioplegia.
The consequences of post–cardiac surgery diaphragm dysfunction vary from asymptomatic radiographic abnormalities to severe pulmonary dysfunction requiring prolonged mechanical ventilation and increased morbidity and mortality.
In one study, the incidence of diaphragmatic dysfunction was 11% (5 of 44 patients). But only one patient had phrenic nerve palsy.
Most patients with post–cardiac surgery diaphragmatic dysfunction improve with conservative measures such as chest physiotherapy, prevention and treatment of pneumonia, treatment of underlying chronic obstructive pulmonary disease (if present), and overall care. Rarely, diaphragmatic plication may also be required in such patients.
Arterial blood gas analysis may demonstrate hypoxemia in persons with bilateral diaphragmatic paralysis. Hypoxemia develops from atelectasis and ventilation-perfusion mismatching. Progressive hypercapnia also develops with disease progression.
The hallmark of patients with diaphragmatic paralysis is hypercapnia and a respiratory acidosis. The hypoxemia is a consequence of the hypoventilation. This is a key distinction and P(A-a) O2 gradients may be normal if there is no underlying parenchymal lung disease.
This study reveals elevated hemidiaphragms, small lung volumes, and atelectasis.
In contrast to bilateral disease, physicians can usually diagnose unilateral paralysis with only radiographic studies. (see the image below).
Acute unilateral left diaphragmatic paralysis in a patient with moderately severe chronic obstructive pulmonary disease. The patient previously was as....
Bilateral diaphragmatic paralysis can be subtler to recognize with radiographic studies alone. See image below.
Radiograph of a patient with bilateral diaphragmatic paralysis displaying low lung volumes.
Because accessory muscle contraction may create the appearance of diaphragmatic movement, this study may mislead the physician when diagnosing bilateral diaphragmatic paralysis (see the image below).
Fluoroscopy of elevated left hemidiaphragm in a patient with unilateral diaphragmatic paralysis. The diaphragm moves paradoxically upward during inspi....
In fluoroscopic sniff testing, paradoxical elevation of the paralyzed diaphragm is observed with inspiration and confirms diaphragmatic paralysis (see the image below). However, the sniff test is not very specific; 6% of normal persons exhibit paradoxical motion on fluoroscopy. Due to compensatory respiratory strategies, apparently normal decent of diaphragms may also be seen with sniff test in bilateral diaphragmatic paralysis.
Fluoroscopy of elevated left hemidiaphragm in a patient with unilateral diaphragmatic paralysis. The diaphragm does not move during expiration. For co....
Computerized tomography may be indicated in certain patients to evaluate for potential causes of diaphragmatic paralysis that are due to mediastinal pathology and malignancy.
MRI may be indicated in certain patients to determine the presence of pathologic conditions involving the spinal column or nerve roots that are causing diaphragmatic paralysis.
Dynamic MRI has been used by some institutions to evaluate diaphragmatic disorders.
M-mode ultrasonography is a relatively simple and accurate test for diagnosing paralysis of the diaphragm in the adult population and it can be performed at the bedside.
The paralyzed side shows no active caudal movement of the diaphragm with inspiration and abnormal paradoxical movement (ie, cranial movement on inspiration), particularly with the sniff test.
Patients can be scanned in the anterior axillary line with a curved linear transducer probe angled cranially at a 90° angle to the diaphragm. In this view, the liver is used as a window on the right, while the spleen is used on the left.
B-mode ultrasonography of diaphragm thickness in the zone of apposition of the diaphragm to the rib cage can also provide a sensitive and specific noninvasive assessment of diaphragmatic paralysis. Less than 20% thickening of the diaphragm muscle during inspiration is diagnostic of diaphragmatic paralysis.[3, 13]
Ultrasonography can also be used to serially monitor patients with diaphragmatic paralysis for recovery.
Measuring the vital capacity in the upright and supine positions is the most important part of the pulmonary function test.
Diaphragmatic paralysis reduces the measured compliance of the lungs and a restrictive pattern can develop.
Normally, vital capacity in recumbency decreases by 10%. In contrast, patients with bilateral diaphragmatic paralysis show a 50% decrease in vital capacity when they are supine. This decrease is from cephalad displacement of abdominal contents.
Pulmonary function test results, however, are not always consistent and do not always correlate with the severity of dyspnea from diaphragmatic paralysis.
Patients with diaphragmatic dysfunction and paralysis have a decrease in maximal inspiratory pressures (PI max). These patients cannot generate high negative inspiratory pressures. Therefore, the Pl max in these patients is less negative than -60 cm water.
Important to note is that decreased maximal pressures are the hallmark of bilateral diaphragmatic paralysis. The decrease may not be as easy to detect in those with unilateral diaphragm paralysis.
Electromyography may reveal a neuropathic versus myopathic pattern, depending on the etiology. This can be accomplished by stimulation of the phrenic nerve at the neck. Phrenic nerve stimulation can be done with electrical (surface or needle electrodes) and magnet stimulation. Operator expertise is an important factor in testing.
Technical issues with electromyography include proper electrode placement to avoid “cross-talk” from adjacent muscles and variable results due to variable subcutaneous fat among individuals.
This is the criterion standard for diagnosis.
The transdiaphragmatic pressure is measured by placing a thin-walled balloon transnasally at the lower end of the esophagus, allowing reflection of the changes in pleural pressure. Then, a second balloon manometer is placed in the stomach to measure changes in intra-abdominal pressure. Gastric pressure should become more positive during inspiration. Esophageal pressure should become more negative during inspiration, demonstrating an increase in gradient during normal inspiration. In cases in which the sniff test is negative and clinical suspicion for diaphragmatic paralysis is still high, transdiaphragmatic pressure should be considered.
Normal transdiaphragmatic pressure is approximately 148 cm water in men and 122 cm water in women. Unilateral diaphragmatic paralysis is associated with a maximal transdiaphragmatic pressure of greater than 70 cm water, and thus does not significantly effect transdiaphragmatic pressure generation during normal ventilatory behaviors, but can compromise higher-force, nonventilatory, behaviors like coughing or sneezing. Bilateral diaphragmatic paralysis, however, can impair normal ventilatory behaviors as it is associated with a maximal transdiaphragmatic pressure of less than 40 cm water.[15, 16, 17]
Consult with an expert to perform the test and interpret the results. This measurement can help differentiate diaphragmatic paralysis from other causes of respiratory failure.
Most patients with unilateral diaphragmatic paralysis are asymptomatic and do not require treatment. If the underlying causes are discovered, they can be treated. Even when the etiology is not known, many times paralysis resolves on its own, albeit slowly over a period of months to more than a year. In a select group of patients with unilateral diaphragmatic paralysis who have severe dyspnea upon excursion, surgical treatment has been shown to be beneficial.
Stabilization from surgical plication of the paralyzed diaphragm provides good results in selected patients. Following plication, the paralyzed diaphragm does not paradoxically move cephalad into the thorax during inspiration and, therefore, improves ventilation to the affected site. Furthermore, the procedure also favors the healthy diaphragm, which now performs less work.
In a select group of patients, diaphragmatic plication decreased breathlessness, improved vital capacity by 10-20%, and improved PaO2 by 10%. In one surgical series, the mean forced tidal volume improved dramatically from 216 mL to 415 mL after plication and it was possible to discontinue mechanical ventilation within 2-12 days of plication. Functional and physiologic results of diaphragm plication have been shown to endure over long-term follow-up. In another study, 41 patients underwent plication of the hemidiaphragm. Patients were followed up for at least 48 months. Mean forced vital capacity, forced expiratory volume at 1 second, functional residual capacity, and total lung capacity all improved by 17%, 21%, 20%, and 20% (P < .005), respectively, at 48 months. These mean values had remained constant when compared with the 6-month follow-up.
Plication of the diaphragm can be performed using a number of techniques through a thoracotomy, video-assisted thoracoscopic surgery (VATS), or laparoscopy. The VATS approach can have similar results as the thoracotomy series, with fewer complications.
A common relative contraindication to plication is morbid obesity, as surgical plication is technically more difficult in these patients. This group of patients should be evaluated for bariatric surgery and may be able to avoid plication with improvement of pulmonary function after significant weight loss. Patients with certain neuromuscular disorders (ie, amyotrophic lateral sclerosis and muscular dystrophy) should be approached with caution as plication provides only modest benefit with more complications.
The treatment of bilateral diaphragmatic paralysis mainly depends on the etiology and severity of the paralysis. Invasive ventilation was historically the main treatment for patients who developed respiratory failure as a result of bilateral diaphragmatic paralysis. Later, a subset of these patients who did not have intrinsic lung pathology became candidates for noninvasive ventilation.
Currently phrenic pacing is increasingly being employed in patients with central respiratory paralysis and upper cervical spinal cord injury (lesions above C3) to wean them off the ventilators. These patients ideally should not have any intrinsic lung disease. Electrodes can be implanted intrathoracically via thoracotomy and, more recently, with VATS. Alternatively, electrodes can be placed intramuscularly via a laparoscopic approach. In this approach, intramuscular electrodes are placed near the entrance points of the phrenic nerves using motor-point mapping techniques.
Diaphragm pacing allows patients to speak again and use their olfaction system. It reduces the occurrence of respiratory infections, provides more natural breathing, and avoids dependency on a mechanical ventilator. The phrenic nerve should be tested with a phrenic nerve conduction study before planning for diaphragmatic pacing. Deconditioning and atrophy of the diaphragm prior to pacing is the main limiting factor in weaning patients off the ventilators.
Negative-pressure systems may induce obstruction of the upper airway, particularly if the upper airway dilators are weak and unable to counteract the negative pressure generated by the ventilator. Therefore, sleep studies are required for patients who are being considered for negative-pressure ventilation. Consideration of positive-pressure ventilation lessens the need for screening sleep studies.
Most patients with mild-to-moderate diaphragmatic weakness maintain daytime gas exchange but worsen during sleep. Sleep studies and ventilatory-assist device treatments may identify this condition. Nighttime noninvasive ventilation could be used in this group of patients.
Patients in whom nasal or oral positive-pressure ventilation is unsuccessful may need other forms of noninvasive ventilation (eg, negative-pressure cuirass, pulmonary wrap, rocking bed, positive-pressure pneumobelt).
Tracheostomy with positive-pressure intermittent or permanent ventilation is reserved for patients who are not candidates for less invasive methods or in whom less invasive methods fail.
Nerve reconstruction techniques
In a select group of patients, nerve surgery may be used to restore function to the paralyzed hemidiaphragm.[24, 25] Neurolysis, nerve grafting, and neurotization have demonstrated promise in returning function to unilateral phrenic nerve injury that occurred as a result of anesthetic procedures and operative and nonoperative trauma to the neck. With microscopic neurolysis, fibrous tissue from the compressed portion of the phrenic nerve is removed.
Inspiratory muscle strength and endurance training
Daily inspiratory muscle strength and endurance training can lead to increased nondiaphragmatic inspiratory muscle recruitment and help those with mild symptoms from diaphragmatic paralysis.