COPYRIGHT © 2006 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED  Iatrogenic Symptomatic Chest Wall Hematoma After Shoulder Arthroplasty A REPORT OF T WO CASES BY EKAVIT KEYURAPAN, MD, SAMUEL J. HU, MD, MICHAEL B. STREIFF, MD, LAURA M. FAYAD, MD, AND EDWARD G. MCFARLAND, MD Investigation performed at the Division of Sports Medicine and Shoulder Surgery, the Department of Orthopaedic Surgery, the Department of Medicine and Pathology, and the Department of Radiology and Radiological Science, The Johns Hopkins University, Baltimore, Maryland otal shoulder arthroplasty is considered to be a safe procedure with a low complication rate. Cofield et al. 1 reviewed the literature and reported that the overall complication rate for this procedure was 10.4% (123 of 1183). Associated complications that have been reported include in- fection, nerve damage, arterial injury, fracture, instability, and long-term loosening of the implant 2-5 . The rate of bleeding or hematoma formation after total shoulder arthroplasty has been reported as 0.17% (two of 1183) 1 , 2.78% (one of thirty- six) 1,6 , and 3.3% (five of 152) 7 . Most reviews of total shoulder arthroplasty have not listed bleeding as a complication of the procedure 8-12 . The rate of bleeding after total hip and knee replace- ment has been reported as 1.3% (eighteen of 1414 wounds) 13 , 2.4% (six of 248 primary total hip arthroplasties) 14 , and 5.8% (fourteen of 243 revision total hip arthroplasties) 14 . Patients in whom a hematoma develops after total hip or knee replace- ment can experience higher-than-normal pain levels, are at increased risk for infection, and have poorer rehabilitation progress than those without a hematoma 13,15-17 . Hematoma for- mation has been reported to be more common in patients who receive prophylactic or therapeutic anticoagulation ther- apy in the perioperative period 7 than in those who do not. The impact of anticoagulation therapy on hematoma formation after total shoulder arthroplasty has not been addressed in the literature. We report the cases of two patients who, as a result of anticoagulation therapy, had development of a substantial chest wall hematoma after total shoulder replacement. These two cases represent 1.3% of the 158 shoulder arthroplasties that were performed by the senior author (E.G.M.) during the study period (from 1998 to 2005) and two of the six patients in the cohort who received preoperative anticoagulation therapy. The patients were informed that data concerning their cases would be submitted for publication. Case Reports ASE 1. One year before the time of presentation, a seventy-two-year-old man with at least a ten-year his- tory of painful bilateral degenerative joint disease of the shoulder underwent an uncomplicated and successful right total shoulder replacement with cemented humeral and glen- oid components. At the time of presentation to our practice, he requested a left total shoulder arthroplasty because an ad- ditional year of nonoperative treatment had failed to relieve the pain. The medical history included deep venous throm- bosis, chronic atrial fibrillation, coronary artery disease, recurrent cerebrovascular ischemic events, congestive heart failure, hypertension, mitral and aortic regurgitation, achala- sia, sick sinus syndrome, gout, chronic obstructive pulmo- nary disease, chronic renal insufficiency, and osteoporosis. His daily medications included warfarin (5.0 mg daily), digoxin, omeprazole, spironolactone, metoprolol, metolazone, allopu- rinol, bumetanide, potassium, and home oxygen. He had no history of a bleeding disorder and was receiving chronic war- farin therapy because of atrial fibrillation and previous deep venous thrombosis. He was admitted to the hospital four days before surgery to reverse the warfarin and to provide bridging anticoagula- tion with unfractionated heparin. Eight hours before surgery, the heparin infusion was stopped. Immediately before surgery, the prothrombin time and activated partial thromboplastin time were normal. The patient underwent an uncomplicated left total shoulder arthroplasty with cemented humeral and glenoid components through a standard deltopectoral ap- proach. The procedure took 177 minutes, and two Hemovac drains were placed deep in the wound. The patient had limited shoulder motion and a large amount of scarring necessitating an extensive surgical exposure (incision length, 4 in [10.2 cm]). The estimated blood loss was 1000 mL, and the replace- ment fluids included 3100 mL of crystalloid and two units of packed red blood cells. Immediately postoperatively, the he- moglobin level was 98 g/L, the prothrombin time was 12.3 seconds, the international normalized ratio was 1.0, and the activated partial thromboplastin time was 29.2 seconds. Eight hours after surgery, the patient was started on a continuous infusion of unfractionated heparin at a rate of 800 T C