Why is this patient short of breath?
A 52-year-old man presented with a 1-year history of persistent, progressive dyspnea on exertion. As a result of orthopnea, he usually slept sitting upright. He also complained of persistent cough, which occasionally produced white plugs of mucus.
Dyspnea and cough were worse following exposure to cold air and fumes, ingestion of a large meal, and laughter. He noted swelling of both feet and ankles.
The patient’s medical history was significant for diabetes mellitus, for which he received insulin, and for coronary artery disease, for which he underwent a 4-vessel coronary artery bypass graft surgery 1 year before presentation. The patient was not a current smoker but had smoked in the past. He reported no other systemic symptoms.
On physical examination, the patient was afebrile with normal vital signs. There was no adenopathy. The cardiac examination was unremarkable, but the lung examination revealed crackles in bilateral posterior lung fields from the bases to approximately three quarters of the way to the apices of the lungs. There was no wheezing.
The patient’s laboratory workup was remarkable for pulmonary function studies that revealed a forced vital capacity (FVC) of 1.93 L (44% of predicted), a forced expiratory volume in 1 second ([FEV.sub.1]) of 1.35 L (41% of predicted), and an [FEV.sub.1]:FVC ratio of 70%. These results indicate significant restrictive as well as obstructive ventilatory defects. There was no improvement following bronchodilator administration. The total lung capacity was 3.2 L (53% of predicted). The residual volume was 1.5 L (55% of predicted).
An arterial blood gas sample revealed a pH of 7.44, a [PCO.sub.2] 51 mm Hg, a [PO.sub.2] of 56 mm Hg, and a bicarbonate of 32 mEq/L. The carbon monoxide-diffusing capacity (DLCO) was 38% of predicted. The posteroanterior and lateral chest radiographs taken on presentation are shown below (Figures 1A and 1B).
Making the diagnosis
The chest radiographs revealed small lung volumes. Both hemidiaphragms were elevated, and discoid atelectasis was present at the lung bases. There were no pleural effusions or pulmonary edema. A chest CT scan showed bibasilar discoid atelectasis and elevated hemidiaphragms but no other significant pathology (Figure 2).
The patient’s clinical presentation and the pulmonary function tests show restrictive and obstructive ventilatory defects and a diminished DLCO. The concomitant arterial hypoxemia is associated with a broad differential diagnosis that includes congestive heart failure, interstitial lung diseases, and bronchiolitis obliterans with organizing pneumonia. However, the important clues that pointed toward the correct diagnosis included the proximity of the symptoms to the cardiac surgery, the profound degree of orthopnea, and the clear parenchyma by plain radiography and chest CT scan.
A diagnosis of diaphragmatic weakness, which may have been secondary to phrenic neuropathy following open-heart surgery, was considered. The pulmonary function tests were then repeated in the sitting and supine positions; FVC was significantly reduced in the supine position, compared with the upright position. The diagnosis was confirmed by phrenic nerve conduction studies and by the demonstration of paradoxic diaphragmatic movement by fluoroscopic examination.
The patient was treated with nocturnal bilevel positive airway pressure (BIPAP). Over 1 year, the patient’s exercise tolerance and or thopnea gradually improved, along with radiographic improvement and resolution of basilar discoid atelectasis.
Phrenic nerve injury during cardiac surgery leading to diaphragmatic paralysis is a well-recognized entity. The incidence ranges from 10% to 70%, according to a 1993 report.  However, with better recognition of the problem and new surgical measures to minimize injury, the incidence has recently fallen to between 2% and 17%. 
The consequences of phrenic nerve injury range from an asymptomatic course to significant morbidity and, in some cases, mortality from complications such as atelectasis, pneumonia, pleural effusion, and prolonged mechanical ventilation.
* Relevant anatomy: After arising from the anterior horn cells of nerve roots C3 to C6, the phrenic nerve enters the thoracic cavity. In its course, the nerve passes medially and posteriorly to the internal mammary artery (IMA) before passing along the lateral aspect of the pericardial sac. The phrenic nerve receives its blood supply from the pericardiophrenic artery, a branch of the IMA.
* Mechanisms of injury: The mechanisms of injury to the phrenic nerve during cardiac surgery include topical hypothermia and direct injury to the nerve during dissection of the IMA. Topical hypothermia is a technique of “cooling” the heart to temperatures lower than 4[degrees]C (39.2[degrees]F) by bathing the heart in a cold solution This cold temperature may induce a demyelination injury to the phrenic nerve, which results in protracted recovery over a variable period of time or in no recovery. 
* Presentation and diagnosis: The clinical presentation of phrenic nerve injury following open heart surgery may vary from an asymptomatic radiologic abnormality to severe symptomatology, as in our case, leading to extreme physical limitations and hypoxemia. The diagnosis relies on a high index of clinical suspicion of this complication of cardiac surgery and on the relationship of the symptoms to the surgery. Chest radiographs may show only the presence of elevated hemidiaphragms.
Ultrasonography is a noninvasive diagnostic modality that may also be useful for imaging the hemidiaphragms. In a prospective study of 16 patients, assessment of phrenic nerve function and fluoroscopy were found to be equally effective.  Diaphragmatic ultrasound has also been studied in the preoperative staging of non-small-cell lung cancer.  Phrenic nerve conduction studies definitively confirm the diagnosis. 
* Treatment: The treatment of phrenic nerve injury is supportive care. In extreme cases, mechanical ventilation may be required, but simple observation may be adequate in milder cases. BiPAP, as was employed in this case, provides ventilatory support to symptomatic patients. Intraoperative monitoring of phrenic nerve function has been suggested to reduce the incidence of injury. 
Dr Bhadriraju is a clinical fellow, and Dr Fowler is professor of medicine and chairman of pulmonary and critical care medicine, at the Medical College of Virginia, Virginia Commonwealth University in Richmond.
Dr Chiles is professor and section chief of thoracic imaging at Wake Forest University School of Medicine in WinstonSalem, North Carolina.
(1.) DeVita MA, Robinson LR, Rehder J, et al. Incidence and natural history of phrenic neuropathy occurring during open heart surgery. Chest. 1993;103:850-856.
(2.) Tripp HF, Bolton JW. Phrenic nerve injury following cardiac surgery: a review. J Card Surg. 1993;13:218-223.
(3.) Balaji S, Kunovsky P, Sullivan I. Ultrasound in the diagnosis of diaphragmatic paralysis after operation for congenital heart disease. Br Heart J. 1990;64:20-22.
(4.) Houston JG, Fleet M, McMillian N, Cowan MD. Ultrasonic assessment of hemidiaphragmatic movement: an indirect method of evaluating mediastinal invasion in non-small cell lung cancer. Br J Radiol. 1995;68:695-699.
(5.) Mazzoni M, Solinas C, Sisillo E, et al. Intraoperative phrenic nerve monitoring in cardiac surgery. Chest. 1996;109:1455-1460.
COPYRIGHT 2001 Cliggott Publishing Co.
COPYRIGHT 2001 Gale Group