Fracture blisters are a relatively uncommon complication of high-energy fractures, with an incidence of 2.9%.1 In the lower extremity, fracture blisters almost always occur distal to the knee.1 Histologically, the blisters represent an injury to the dermoepidermal junction.2 On physical examination, there are tense blood- and/or clear fluid–filled bullae overlying markedly swollen and edematous soft tissue,1 resembling a second-degree burn.3 Infection may develop after fracture blisters,1 and this is perhaps the most dreaded complication of total knee arthroplasty (TKA). The patient provided written informed consent for print and electronic publication of this case report.
Case Report
A 71-year-old man with end-stage osteoarthritis of the right knee underwent an elective TKA with cemented components (Legion PS; Smith & Nephew). His medical history included venous insufficiency, type 2 diabetes mellitus, chronic obstructive sleep apnea, hypertension, morbid obesity (body mass index, 50), and a previous uneventful left TKA. Tourniquet time was 78 minutes and estimated blood loss was 100 mL. An intra-articular drain was used and was removed on the first postoperative day. After wound closure, a soft splint bandage consisting of 2 to 3 layers of cotton and bias wrap was applied. Deep vein thrombosis (DVT) prophylaxis with enoxaparin 40 mg once daily was started on the first postoperative day.
Upon removal of the surgical dressings on the second postoperative day, the anterior leg was found to have a combination of tense clear fluid– and blood-filled blisters on markedly swollen and erythematous skin. The incision was minimally involved (Figure A). There was diffuse 2+ pitting edema with hyperesthesia in the affected skin distal to the knee. Prior to these findings, the patient had complained of increasing pain in his operative leg, but there was no escalation in analgesic requirements. There was no evidence of compartment syndrome on serial examinations. An ultrasound of the lower extremity was negative for DVT. Plain films did not show iatrogenic fractures. There was no intraoperative vascular injury, and the foot pulses remained unchanged between the time the patient was in the preoperative holding unit, the postanesthesia care unit, and the orthopedic ward. The operative leg was treated with elevation and loosely applied Kerlix roll gauze (Kendall, Covidien), but active blister formation continued for another 2 days. A 10-day prophylactic course of trimethoprim/sulfamethoxazole was initiated on the third postoperative day after the blisters started to rupture. The patient was allowed to bear weight as tolerated, but his physical therapy (PT) course was limited by pain and fear “of losing his leg.” He declined several PT sessions and was hesitant to use continuous passive motion. The patient was discharged to a short-term rehabilitation facility with weekly outpatient follow-up. On the second postoperative week, his fluid-filled blisters completely reepithelialized, but the blood-filled blisters required an additional week for reepithelialization (Figure B). While the patient’s knee was stiff because of limited PT participation, it was not until the second postoperative week when most of the fracture blisters had healed that he was able to resume an intensive knee exercise program, avoiding the need for manipulation under anesthesia.
Discussion
Giordano and colleagues2 identified 2 types of fracture blisters: clear fluid– and blood-filled. While both types involved disruption of the dermoepidermal junction, greater disruption and complete absence of dermal epithelial cells was observed in the hemorrhagic type. Clinical follow-up of the patients in the study by Giordano and colleagues2 showed that the mean time for reepithelialization was 12 days for fluid-filled blisters and 16 days for blood-filled blisters. These findings are similar to what we observed in our case report. In particular, the fluid-filled blisters healed in 2 weeks, whereas the blood-filled blisters required 3 weeks to heal.
The etiology of the fracture blisters in this patient is likely multifactorial and related to age, obesity, venous insufficiency, and diabetes mellitus. Farage and colleagues4 described a series of progressive degenerative changes in the aging skin, including vascular atrophy and degradation of dermal connective tissue, leading to compromised skin competence. The integrity of the dermis can be further reduced in patients with diabetes through glycosylation of collagen fibrils.5 Compared with age-matched normal controls, patients with insulin-dependent diabetes have a reduced threshold to suction-induced blister formation.6 Obesity is another potential contributing factor, with multiple studies showing significantly impaired venous flow in obese patients.7,8 Taken together, soft-tissue swelling after surgery in the setting of chronic venous insufficiency and compromised skin due to advanced age and diabetes may lead to markedly elevated interstitial pressure. One mechanism to relieve such abnormally high pressure is the formation of fracture blisters.1