LEFT brachial plexus injury and musculoskeletal pain have been reported after left internal mammary (LIMA) harvesting for left anterior descending coronary artery bypass grafting. 1Prolonged elevation and retraction of the left hemithorax with the Favoloro sternal retractor during LIMA harvesting has been implicated as a cause of left upper extremity peripheral neuropathy from brachial plexus stretch injury and postoperative musculoskeletal pain from rib fracture, intercostal muscle disruption, and costotransverse and costosternal cartilage separations. 2We report a case of factitious pseudoangina from prolonged postoperative musculoskeletal pain treated as angina pectoris with nitrates and anticoagulants after LIMA harvesting for left anterior descending coronary artery bypass grafting (table 1).
Case Report
A 52-yr-old man with a 70% stenosis of the left main coronary artery underwent a single coronary artery bypass graft with a LIMA graft to the proximal left anterior descending coronary artery in 1988 at age 42. The coronary artery bypass graft was performed during balanced narcotic and inhalational anesthesia through a median sternotomy incision with the patient in the supine position with both arms adducted. Invasive vascular monitoring included a right radial artery catheter and a central venous catheter inserted via the right internal jugular vein. The LIMA was harvested over 25 min with exposure provided by a Favaloro retractor. The LIMA-to-left anterior descending anastomosis was conducted during 25 min of complete cardiopulmonary bypass and circulatory arrest. The immediate postoperative course was uneventful, and the patient was discharged on the sixth postoperative day. The patient was readmitted for a 24-h hospital stay 2 days after discharge for tube thoracostomy drainage of a spontaneous left pneumothorax.
One week after the second hospital discharge, the patient first noted left anterior chest wall pain during movement of the left arm with radiation to the left side of the neck, the left shoulder, and left-sided scapula. The pain was dysesthetic, nonlancinating, and pressing, and was initiated by exercise of the left arm, particularly lifting and carrying light items, such as books or a briefcase. The pain was relieved only by resting the left arm. Exercise electrocardiography showed no evidence of cardiac ischemia or myocardial infarction. Cardiac angiography showed no significant additional coronary artery disease or LIMA graft stenosis or occlusion. Cardiac ejection fraction and myocardial performance were assessed as normal, and oral analgesics for prolonged postoperative incisional pain were prescribed.
Between 1988 and 1997, the patient consulted innumerable specialists, including cardiovascular surgeons, physiatrists, cardiologists, and rheumatologists, and underwent five additional coronary arteriograms to assess “recurrent angina.” The patient was treated for presumed costochondritis with oral analgesics, nonsteroidal antiinflammatory agents, muscle relaxants, and trigger-point injections in the back. The patient also was prescribed long-term aspirin prophylaxis for coronary artery disease and lovastatin for hypercholesterolemia, and encouraged to take sublingual nitroglycerin for exertional “anginal”, i.e. , left chest wall and upper arm pain and muscle spasm. The patient continued to exercise despite attacks of left-sided chest wall pain, participated in several marathon runs over the years, but gave up golf and weight-lifting because of severe pain and spasms in the left pectoralis major muscle precipitated by movement and weight-bearing exercise of the left upper extremity.
The patient was ultimately referred to the Louisiana State University Medical Center Multidisciplinary Pain Mastery Center for evaluation in November 1997, and, at physical examination, was found to have a 5-cm circumferential area of dysesthesia, muscular irritability, and spasm overlying the left-sided first and second parasternal intercostal spaces. The remainder of the physical examination, particularly the neuromuscular examination of the left upper extremity, was within normal limits. A magnetic resonance imaging study of the thorax in January 1998 showed no evidence of costochondritis, costochondral, or costotransverse junction separation or dislocation; rib fracture, callus, or pseudoarthrosis; sternal or xiphisternal abscess; pseudoarthrosis; or dehiscence. The mediastinum, hemidiaphragms, lungs, and hemithoraces were consistent with poststernotomy and otherwise normal. The hemidiaphragms were symmetrically shaped and positioned bilaterally, and there was no radiographic evidence of phrenic neuropathy. Phrenic nerve conduction studies were not indicated.
The patient received an intramuscular injection of 50 U botulinum A toxin suspended in 10 ml bupivacaine, 0.25%, into the dysesthetic trigger point described in the left pectoralis major muscle in January 1998, with dramatic reduction of symptoms. Supplemental therapy for breakthrough pain was initiated at the same time with 75 mg oral venlafaxine and 500 mg oral methocarbamol, both taken at bedtime. The patient remains on a regimen of semiannual intramuscular trigger-point injections with 50 U botulinum toxin supplemented orally with night-time venlafaxine, a selective norepinephrine-reuptake-inhibiting antidepressant with analgesic properties, and methocarbamol, a skeletal muscle relaxant. The patient continues to participate in jogging marathons and has restarted a weekly golf game.
Discussion
Brachial plexus and musculoskeletal injuries of the chest wall and upper extremities have been reported after cardiac surgery, with incidences ranging from 2% to 38%. 2Brachial plexus injuries have been associated with the median sternotomy technique, paramedian sternotomy, forced or prolonged sternal retraction with the Favoloro retractor, penetrating trauma from first-rib fractures, hyperabducted arm positioning, and needle trauma during insertion of internal jugular catheters. 3Musculoskeletal injuries have also been associated with rib fractures from forced sternal retraction, costochondral cartilage separations, and costotransverse rib disarticulations. 1–3In a prospective investigation of 162 patients undergoing median sternotomy for cardiac surgery, Roy et al. 1reported a significantly greater incidence (39%) of musculoskeletal complaints and neurologic dysfunction in patients undergoing internal mammary artery grafts for aortocoronary bypasses than in patients (17%) undergoing valvular and other cardiac operations without internal mammary artery graft harvesting.
Neurologic dysfunction from brachial plexus and peripheral nerve injuries after cardiac surgery often present with neuropathic pain and dermatomal sensorimotor disturbances that require diagnostic confirmation with nerve conduction studies and electromyography. 2Musculoskeletal complaints after cardiac surgery were described by Roy et al. 1as pain at rest and exercise and limitation of motion in the arms, shoulders, chest, and back. Postoperative musculoskeletal complaints were difficult to describe clinically and confirm radiographically in the absence of old fractures, cartilaginous separations, and rib disarticulations. 1Using either chest wall computed tomography or exploratory resternotomy, Shafir et al. 3identified inadvertent paramedian sternotomy as a cause of postoperative musculoskeletal pain, chronic thoracic fistulas, sternal osteomyelitis, and sternal dehiscence in 11 of 55 patients undergoing cardiac surgery. Shafir et al. 3recommended that when a painful paramedian sternotomy is diagnosed radiographically or surgically at resternotomy, sternectomy and myocutaneous flap wound closures be performed rather than simple sternal reclosures.
We reported a case of chronic myofascial pain in the left upper thorax and arm repeatedly misdiagnosed as angina pectoris and coronary artery spasm despite normal coronary angiography and left ventricular wall motion and ejection fraction (table 1). Although the patient requested sternal reexploration to assess faulty sternotomy, magnetic resonance imaging of the chest wall, mediastinum, and thorax was initially recommended to rule out sternal fistula, sternal osteomyelitis, occult sternal dehiscence, costochondral cartilage separations, old rib fractures or pseudoarthroses, costotransverse rib disarticulations, and pleuropericardial defects or hernias. Chest magnetic resonance imaging was selected over chest radionuclide scanning, a better imaging technique to detect chronic infection and deep abscesses, for its superior soft tissue and bone density delineation and because the erythrocyte sedimentation rate, differential count, and serum protein electrophoresis did not suggest chronic infection from osteomyelitis or fistula. In addition, chest magnetic resonance imaging ruled out chronic atelectasis and pleurisy from phrenic neuropathy as an unusual cause of postoperative chest pain. 4Although common after open heart surgery, with incidences ranging from 10% to 73%, phrenic neuropathy from topical cardioplegia is usually short-lived and painless and causes elevation and abnormal motion of the hemidiaphragms and persistent ipsilateral atelectasis. 4
Physical examination of the patient was consistent with a tight and painful myofascial band in the pectoralis major muscle, with irritation and fasciculation during percussion and radiation to the uppermost clavicular portion of the muscle and left upper arm. 5The pectoralis major, pectoralis minor, and intercostal muscles are all common sites of myofascial pain syndromes, especially after chest wall trauma and surgery, and have unique pain referral patterns, as initially mapped by Travell and Rinzler. 6
Conditions that may mimic myofascial pain syndromes of the anterior chest wall and upper arm may include angina pectoris, as in this case, diffuse myopathies, fibromyalgia, polymyalgia rheumatica, dermatomyositis, postherpetic neuralgia, connective tissue disorders, hypothyroidism, early Parkinsonism, multiple myeloma, osteoarthritis, bursitis, and tendonitis. 5Inflammatory and malignant diseases should be ruled out using the appropriate laboratory tests. Postherpetic neuralgia causes a characteristic neuropathic pain condition and usually is heralded by dermatomal shingles.
Conventional therapies for myofascial pain syndromes include physical therapy, regular stretching and exercises, vapocoolant spray and stretch techniques, dry needling and injections of trigger points with local anesthetics and antiinflammatory agents, psychotherapy, biofeedback, and pharmacotherapy. 5Conventional pharmacologic agents effective in myofascial pain syndromes include the tricyclic and selective norepinephrine-reuptake-inhibiting antidepressants, such as venlafaxine, nonsteroidal antiinflammatory drugs, and skeletal muscle relaxants, such as methocarbamol. 5Most recently, botulinum A and F toxin injections, which are approved as orphan drugs for managing severe pain and dystonia in cerebral palsy, 7thalamic stroke, 8spastic torticollis, 9and spinal cord infarcts and trauma, 10have been recommended for the management of chronic myofascial pain syndromes. 11
Myofascial pain syndromes may mimic angina pectoris after median sternotomy for cardiac surgery. Brachial plexus and musculoskeletal injuries may occur even more commonly after median sternotomy for cardiac surgery and have been correlated with harvesting of the ipsilateral internal mammary artery, usually the left, but not with intraoperative arm positioning. 1Precise diagnostic imaging techniques may be necessary to assess chronic chest wall pain after median sternotomy for cardiac or anterior mediastinal surgery and to eliminate faulty sternotomy techniques, cartilaginous injuries, and rib fractures or disarticulations. Nerve conduction studies may be necessary to assess sensorimotor dysfunction in brachial plexus injuries. Although usually reserved for painful spasticity disorders, botulinum A toxin injections may offer a new, safe, and effective technique to manage myofascial pain syndromes.
