 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Acute Ischemia of the Upper Limb Fifteen Years after Anterior Dislocation of the Glenohumeral Joint and a Modified Bristow Procedure. A Case Report* , CHICAGO, ILLINOISInvestigation performed at the Department of Orthopaedic Surgery and the Division of Vascular Surgery, Department of Surgery, Northwestern University Medical School, Chicago
In 1958, Helfet described a stabilization procedure for recurrent anterior dislocations of the shoulder. He named it after his mentor, Bristow, who had taught it to him almost twenty years earlier. The procedure consisted of detaching the tip of the coracoid process from the scapula just distal to the insertion of the pectoralis minor muscle. This piece of coracoid, with the tendons of the short head of the biceps and coracobrachialis attached, was then passed through the subscapularis muscle and placed in contact with a roughened surface of the scapula. It was held in place by suturing the biceps and coracobrachialis tendons to the subscapularis tendon. In 1964, Mead and Sweeney described a modification of the procedure whereby the coracoid is secured to the anterior part of the glenoid rim with a screw. Since then, many modifications have been made concerning the splitting of the subscapularis muscle or tendon; however, all of these modifications have involved the use of a screw to secure the coracoid process to the anterior part of the glenoid rim12. Vascular complications have been rare after the modified Bristow procedure for recurrent anterior dislocations of the glenohumeral joint1,3,4,7,11,15. A review of the literature revealed four previous reports of pseudoaneurysm of the artery after the modified procedure1,3,4,7. We report an unusual case in which ischemia of the distal part of the upper extremity developed secondary to emboli that originated from a pseudoaneurysm of the axillary artery. The pseudoaneurysm was adjacent to a screw that had been inserted fifteen years earlier during a modified Bristow procedure for recurrent anterior dislocations of the shoulder.
A fifty-six-year-old man, a general surgeon, was seen in the emergency department of our institution because of acute pain, numbness, tingling, and coolness in the right forearm and hand that had begun four hours earlier. The pain, which was severe, began in the elbow and traveled distally to the hand and fingers. There was tingling in the entire hand, but it was more intense in the ring and small fingers. The hand and forearm were paler and cooler than the proximal portion of the extremity. Sixteen years previously, the patient had sustained a traumatic anterior dislocation of the right glenohumeral joint in a skiing accident. The upper limb had been without a pulse for approximately forty-five minutes until the glenohumeral joint had been reduced. No vascular studies had been performed, as the findings on the clinical vascular examination had returned to normal. He had also had a traction injury of the right brachial plexus, which had resulted in neurapraxia of all three of its trunks. Clinically, he had had no sensation or motor control of the extremity after the initial dislocation. During the next eighteen months, he slowly regained sensation and motor control. The only residual deficits were mild paresthesias of the entire little finger and the ulnar half of the ring finger. He had full motor strength of the entire extremity and was able to return to work as a general surgeon. As motor function returned and use of the extremity increased, the shoulder began to redislocate with minimum provocation. One dislocation occurred while the patient was swimming, and two dislocations occurred while he was externally rotating and abducting the extremity during sleep. Twenty months after the initial dislocation, the patient had a modified Bristow procedure in an attempt to prevent additional dislocations. One centimeter of the distal coracoid was removed with the attached tendons of the coracobrachialis and the short head of the biceps. The subscapularis muscle was detached from its insertion, and the anterior aspect of the joint capsule was incised to expose the glenoid. The labrum and the capsule were found to be detached from the anterior portion of the glenoid, demonstrating a classic Bankart lesion. A one and one-quarter-inch (thirty-one and three-quarter-millimeter) compression screw was used to attach the coracoid to the anteroinferior portion of the glenoid. The subscapularis muscle and the anterior aspect of the capsule were advanced and oversewn to tighten the soft tissues, as done in a Putti-Platt procedure. Adequate stability was noted intraoperatively. The patient tolerated the procedure well and did not have any additional dislocations of the shoulder. He continued to have paresthesias in the ring and small fingers, which remained stable compared with the preoperative condition. He was able to return to work as a general surgeon. Approximately eight months before being seen in our emergency department, the patient had had transient discoloration of the tips of the right index and long fingers. This had been attributed to the beta-blocker that he was taking to prevent hypertension. The dose was decreased; however, the discoloration resolved before the decrease could have had an effect, and it did not recur until the episode that led to his admission to our emergency department. In the emergency department, the heart rate was eighty beats per minute and regular. With use of a pneumatic cuff and a stethoscope, the blood pressure in the left upper limb was found to be 140/80 millimeters of mercury (18.66/10.67 kilopascals). With use of a pneumatic cuff and a Doppler probe, the systolic pressures of the right brachial and radial arteries were measured as 140 and 120 millimeters of mercury (18.66 and 16.00 kilopascals, respectively). The ulnar pulse was not audible with use of the Doppler probe. There was a variable audible pulse over the right palmar arch. The hand and forearm were white on both the volar and the dorsal surface and were cooler than the proximal portion of the extremity. The brachial pulse was palpable proximal to the elbow. The radial and ulnar pulses were not palpable at the wrist. The patient had a full range of motion of the upper extremity, including the shoulder, elbow, wrist, and hand, and no tenderness to palpation. The glenohumeral joint was stable. There was no pulsation or bruit around the clavicle or the shoulder with any position of the extremity. The muscles of the hand and the forearm were weak, which the patient attributed to the severe pain. There was decreased sensation to touch over the little finger and the ulnar side of the ring finger and diffuse paresthesias over the entire forearm and hand. Angiography performed from a right common femoral artery approach showed that the aortic arch was normal. The angiogram revealed minimum irregularity in the axillary artery proximal to the posterior circumflex humeral artery. The radiologist did not associate the irregularity with the symptoms. A screw, which appeared to be attached to the inferior aspect of the glenoid labrum, was adjacent to the irregularity. The branches of the axillary artery filled appropriately around the shoulder, as did the profunda brachii and the ulnar collateral arteries. In the distal portion of the brachial artery, an inverted Y-shaped thromboembolus occluded the bifurcation of the artery. The radial and ulnar arteries filled from the collaterals around the elbow. The predominant supply to the palmar arch and the hand was from the radial artery. There appeared to be an occlusion of the distal portion of the ulnar artery. The digital arteries did not fill because of the upstream occlusions. A decision was made to proceed with urokinase therapy. As there is no standard dose for this therapy8, we relied on the experience of the interventional radiologists to select the dose. The catheter was placed with its tip just proximal to the embolus of the brachial artery, and the infusion of urokinase was started at 4000 units per minute for four hours; this was followed by an infusion of 2000 units per minute. To prevent propagation of the embolus, heparin was administered intravenously. Throughout the urokinase therapy, the dose of heparin was adjusted, at rates of 800 to 1300 units per hour, in an attempt to maintain the partial thromboplastin time between 1.5 and 2.0 times the control. Prothrombin time, partial thromboplastin time, and levels of fibrin split products and fibrinogen were monitored during the urokinase therapy. The prothrombin time and the partial thromboplastin time were used to monitor the status of the systemic coagulation, and the levels of fibrin split products and fibrinogen were used to monitor the degree of systemic fibrinolysis caused by the urokinase. At no time during the urokinase or heparin therapy did the degree of systemic fibrinolysis increase. The patient was also monitored with chemistry and hematological studies. He was kept in the operative intensive-care unit for monitoring of the vital signs and the vascular status of the limb and for detection of any evidence of complications from the anticoagulation and thrombolytic therapies. His condition remained stable throughout the course of treatment. The patient was monitored with physical examination to palpate the distal pulses and with use of a Doppler probe to listen for the distal tones. Doppler pulsation of the ulnar artery and the palmar arch returned after approximately twelve hours of urokinase therapy. The patient noted a decrease in the pain in the forearm but continued to have paresthesias of the hand. A second angiogram was made after seventeen hours of urokinase therapy. The irregularity of the axillary artery remained unchanged. The embolus of the brachial artery had been partially lysed and remained only in the ulnar interosseous trunk (Fig. 1-A). The flow to the palmar arch and the digital arteries had improved. The catheter was advanced to bathe preferentially the embolus of the ulnar artery. The urokinase therapy was decreased at this time to 1500 units per minute.
The patient began to note a decrease in the paresthesias of the hand. No pulse was palpable in the ulnar artery, but it remained audible with the Doppler probe. A third angiogram was made after forty-one hours of urokinase therapy (Fig. 1-B). The axillary artery remained unchanged.
Arterial Doppler studies were performed at the patient's bedside with use of a pneumatic cuff and a Doppler probe to measure the pressures in the arteries of the upper extremity. The right brachial, radial, and ulnar systolic pressures were 148, 162, and 158 millimeters of mercury (19.73, 21.59, and 21.06 kilopascals), respectively. These values were comparable with those on the left side. It was still difficult to palpate the pulse of the ulnar artery at the wrist. The audible tone heard with the Doppler probe remained strong. A transesophageal echocardiogram was performed at the patient's bedside to search for a cardiac source of the embolus. This study did not reveal an intracardiac thrombus. The source of the embolus was still not known. A fourth angiogram, made after forty-eight hours of urokinase therapy, demonstrated further resolution of the thrombus of the ulnar interosseous trunk; however, some residual clot remained (Fig. 1-C). The distal circulation continued to improve. A pseudoaneurysm was visible in the axillary artery just proximal to the posterior humeral circumflex artery, the location of the irregularity that had been noted earlier (Fig. 1-D). There was also almost complete lysis of the thromboembolus of the ulnar interosseous trunk. Only a small residual amount of thrombus was adherent to the walls of the vessels involved in the original embolus. The flow to the palmar arch through both the radial and the ulnar artery had improved. The only occlusions distally were in the arteries of the ulnar side of the little finger, the radial side of the long finger, and the radial side of the index finger. The urokinase therapy was discontinued at this time, as there was adequate lysis of the original brachial embolus and adequate distal circulation on radiographic and physical examination. The sheath in the right common femoral artery was removed. The heparin therapy was discontinued for two hours to allow the site of the catheterization to coagulate and was then restarted.
An excision of the pseudoaneurysm of the axillary artery and a vein patch angioplasty was done through an incision into the axillary fold. The trunks of the brachial plexus were dissected away from the axillary artery. The saccular pseudoaneurysm was visible on the axillary artery just proximal to the orifice of the posterior circumflex humeral artery, extending posterolaterally from the vessel. The anterior and posterior aspects were dissected; however, heavy scar tissue prevented full circumferential dissection. The screw was found to be located in scar tissue, attached to a free-floating piece of bone immediately superior and slightly lateral to the pseudoaneurysm. It was clearly detached from the glenoid labrum (Fig. 1-E). The screw was removed with a screwdriver and a clamp, without difficulty or damage to the axillary artery or the pseudoaneurysm.
The pseudoaneurysm was then excised from the axillary artery, leaving an approximately 0.5 by 0.5-centimeter-diameter defect in the axillary wall posterolaterally. A segment of the left greater saphenous vein was removed from the left groin, and a patch was fashioned and was sewn into place over the defect. The brachial, radial, and ulnar pulses were palpable postoperatively. In the recovery room, a pneumatic cuff and a Doppler probe were used to measure the systolic pressures in the brachial, radial, and ulnar arteries; the pressures were 160, 150, and 150 millimeters of mercury (21.33, 19.90, and 19.90 kilopascals), respectively. Postoperatively, the patient was monitored in the operative intensive-care unit for one night. He began receiving one aspirin tablet a day, and the antihypertension medication was restarted. The findings on clinical examination returned to the baseline that had been present before the embolic event. Sensation over the little finger and the ulnar side of the ring finger remained decreased. Both the radial and the ulnar pulses were palpable. On the first postoperative day, repeat blood-flow studies were performed with use of a pneumatic cuff and a Doppler probe. A duplex scanner was also used to visualize and record the nature and amount of the arterial flow in the brachial, radial, and ulnar arteries. The pressures and natures of all arterial wave forms were within normal limits. The patient was discharged to home on the second postoperative day, while continuing to take one aspirin tablet a day as well as antihypertension and pain medications orally. One year postoperatively, there had been no additional embolic episodes, the distal pulses continued to be palpable, the ulnar paresthesias had not worsened, and the glenohumeral joint was stable. The patient continued to take one aspirin tablet a day and to work as a general surgeon.
The pseudoaneurysm of the axillary artery in our patient had one of two possible etiologies: the vascular insult that was incurred with the initial dislocation of the shoulder, or the migration of the screw that occurred years later. We believe that the pseudoaneurysm was caused by direct irritation or laceration of the axillary artery by the screw as it withdrew from its original position. This conclusion is based on the physical proximity of the screw to the pseudoaneurysm and on the delay in presentation. To our knowledge, only four cases of pseudoaneurysm of the axillary artery after a modified Bristow procedure have been reported previously1,3,4,7. Our patient was initially seen much later after the operation than were those who were described in the earlier reports, and he was the only one to have had embolic phenomena. The ischemia of the fingertips that occurred eight months before presentation was probably an earlier manifestation of the same problem. The pseudoaneurysm was not visible on the initial angiograms because it was filled with a thrombus. The urokinase caused lysis of the thrombus, which allowed the pseudoaneurysm to fill at the time of the most recent angiography. Emboli may migrate into the small arteries and arterioles, which are too small to be reached with the catheters used in embolectomy. The presence of these emboli made thrombolytic therapy useful for this patient. Urokinase can allow the palmar and digital vessels to reopen, thereby optimizing return to normal perfusion2,8,10,13,14. This case report demonstrates that serious complications, such as ischemia of the limb, may occur long after the operative procedure. The embolic events in our patient were limb-threatening, emphasizing the need for surgeons to consider carefully the use of hardware in procedures for anterior stabilization of the shoulder.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Introduction
Case Report
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |