An 85-year-old Caucasian male presents with a heel ulcer after spending eight weeks in a rehabiltation home following hip replacement surgery. He presently spends much of his time in a lounge chair or in a wheelchair. He has a history of angina, arthritis, aortic valve replacement, prostate cancer with radiation treatment and Parkinson’s disease. He is taking carbidopa-levodopa (Sinemat, Merck), ropinirole (Requip, GlaxoSmithKline) and warfarin (Coumadin, Bristol-Myers Squibb). He is allergic to sulfa.
The patient initially underwent a workup and a debridement as well as an aerobic culture with sensitivities. He also received a Multipodus splint for offloading. The ulcer received a wet to dry dressing. Later the culture showed that the ulcer was infected with community acquired methicillin-resistant Staph aureus (MRSA), which was sensitive to clindamycin and Bactrim. After the normal debridement of the ulcer, the patient received clindamycin 300 TID.
Four days later, the patient presented to the office with a painful violaceous bruise-like area on the dorsum of his right foot with an area of central necrosis. The lesion did not blanch on diascopic compression, which indicated that the discoloration was external to the blood vessels. The discoloration went distal into the second and third toes, which were also painful to touch. There was no pruritus nor cellulitis. I incised the small blister and noted serous fluid. I obtained a culture and it came back as normal flora. I irrigated the affected area and utilized a wet to dry dressing. 
Since the patient was already on antibiotics, I felt that this was not an infectious process but I wanted to follow up with the patient in three days. The patient has an aide who denies that there were any spiders around and notes the patient uses all of his shoes year round. She also denied any trauma to the foot. She denied that he had any other area of rashes or new skin problems.
On the next visit, the area was more painful than before. There was more evidence of a “knot” on the dorsum of the foot and it was harder in that area. There was no warmth to the area.
In an effort to gain more information, I performed an ultrasound and found a large hematoma under this area. I determined that the skin discoloration was due to a spontaneous hematoma from an unknown trauma. I placed a needle in the area of the hematoma in an effort to drain the area but no fluid was present. 
At this time, I contacted the cardiologist and reviewed the INR, which was between 1.6 and 3.1 for the past nine months with the prothrombin time (PT) between 16 and 29.4. However, the INR reached a peak of 6.0 (PT 54.5) four days after the patient started on clindamycin and that was the exact time that the patient underwent evaluation for the swelling on the dorsum of the right foot. The cardiologist held the Coumadin for six days and the INR was back down to its normal range of 2 to 3.
The diagnosis is Coumadin induced hematoma with subclinical trauma. This patient demonstrates that podiatric dermatology is not in a bubble and one must use all of one’s medical knowledge to figure out what is going on. It was the use of the diagnostic ultrasound that showed the hematoma. It was the review of the INR from the cardiologist who provided the data to back up the diagnosis. 
Warfarin is the most commonly used oral anticoagulant and has established efficacy for the prevention of thromboembolic events in patients with chronic atrial fibrillation, prosthetic heart valves, venous thromboembolism and coronary artery disease. Warfarin exerts its effect by lowering the amount of active vitamin K available for the activation of clotting factors.1
A fat-soluble vitamin, vitamin K, is an essential component in the production of many of the coagulation proteins. An alteration in the synthesis of vitamin K-dependent coagulation factors is usually reflected by changes in the PT and INR.2
Many antibiotics alter the balance of gut flora, thereby enhancing the effect of warfarin. In the gut, bacteria would normally produce vitamin K but when less bacteria do not produce the vitamin K, the vitamin K dependent clotting factors (factors 2, 5, 9 and 10) are affected. Also note that older patients who may not be eating well and have poor nutritional habits may already have a deficiency of vitamin K.
An increased INR secondary to warfarin interactions with various antibacterial agents is a known phenomenon. An increased awareness of warfarin–antibiotic interaction and appropriate monitoring is essential to control the INR levels and prevent bleeding complications. 4 
Spider bite. Insect and spider bites often cause minor swelling, redness, pain and itching. These mild reactions are common and may last from a few hours to a few days. In this case, once I ruled in the hematoma, I ruled out a spider bite.
Insect bite. Insect bites may be grouped, popular or urticarial, and have a surrounding halo. Occasionally one can see a central depression. Some insect bites result in a hemorrhagic bullae. Bullae formation and tissue necrosis are more common with spider bites but may also be caused by insect bites. This patient had a bullae that was not hemorrhagic. This finding initially had me thinking this was due to an insect or spider bite.
Cellulitis. This is a common infection of the dermis and the subcutaneous tissues caused by a bacterial infection. It can arise without an obvious source for the bacterial infection. Staphylococci are the bacteria that most commonly cause cellulitis. Streptococci is the next most common cause.
Cellulitis usually begins as a small area of pain and redness on the skin. This area spreads to surrounding tissues, resulting in the typical signs of inflammation: redness, swelling, warmth and pain. One can also develop fever and/or swollen lymph nodes in the area of the infection. This patient has a negative culture and a lack of warmth to the area. 
Trauma. Trauma can cause various types of bleeding into tissues. It is the primary cause for hematomas. Occasionally after an injury, blood collects and pools under the skin, resulting in a hematoma and giving the skin a spongy, rubbery, lumpy feel. A regular bruise is more spread out and may not feel like a firm lump. When there is trauma, an X-ray is normally required to aid in the diagnosis.
Drug induced purpura. These are primarily found on the feet and ankles (dependent areas).
Drug reaction. Drug reactions can mimic a wide range of dermatoses. The morphologies are myriad and include morbilliform, urticarial, papulosquamous, pustular and bullous. Medications can also cause pruritus and dysesthesia without an obvious eruption. One should consider the possibility of a drug-induced reaction in any patient who is taking medications and who suddenly develops a symmetric cutaneous eruption.
In this patient, the rash was only on the top of the foot and there was no pruritus. Severe dermatologic reaction as a result of drug side effects usually manifests in multiple blisters whereas this patient just had a single blister.
In general, the quicker one applies ice after the injury, the less bleeding will result. If possible, elevate the bruised limb, which may lead to less swelling. Resting the limb will also help to prevent further injury.
If the area is still painful after 48 hours, apply gentle heat with warm towels, a hot water bottle or a heating pad. Apply heat for 20 minutes at a time to promote absorption and repair. Since heat causes swelling and increases tissue fluid, which may impair function, one may follow hot compresses with cold applications to minimize the secondary effects of heat. Pressure in the form of an elastic adhesive bandage may be helpful to reduce hemorrhage and swelling.
In a patient on anticoagulants, conservative management always involves temporary discontinuation of the Coumadin and close monitoring of the PT and INR. Prompt surgical decompression and hemostasis is indicated for patients with large progressing hematomas or for patients with nerve compression. 5
When there is poor resorption of the hematoma, the body will progressively sequester the hematoma. The body slowly eliminates the hematoma and replaces it with a calcium deposit. This is called traumatic myositis ossificans, which most commonly occurs in the skeletal muscle of the arms and legs. 6
The ulcer of the heel eventually closed and continued to stay closed with the use of the Multipodus splint. The hematoma slowly resolved with the use of warm wet compresses. If I was aware of the diagnosis earlier, I would have tried to evacuate the hematoma or apply the warm compresses earlier during the course of treatment.
Patients taking warfarin are susceptible to numerous drug interactions. The majority of interactions result in an increased risk of hemorrhage, and most, but not all, of these are accompanied by an elevated international normalized ratio.
One must see these patients every other day when trying to determine the cause of the dermatological issue. One may see skin changes that help to give clues as to the cause of the podiatric problem. It is also valuable to consult with other physicians who are involved with the care of the patient.
Dr. Morse is the President of the American Society of Podiatric Dermatology. He is a Fellow of the American College of Foot and Ankle Surgeons, and the American College of Foot and Ankle Orthopedics and Medicine. Dr. Morse is board certified in foot surgery.
For further reading, see “Expert Insights On Diagnosing Pigmented Skin Lesions” in the April 2005 issue of Podiatry Today.
1. Holbrook AM, et al. Systematic overview of warfarin and its drug and food interactions. Arch Intern Med 2005;165:1095-1106.
2. Hoffman R. Hematology: Basic Principles and Practice, fourth edition. Churchill Livingstone, New York, 2005.
4. Davydov L. Warfarin and amoxicillin/clavulanate drug interaction. Annals Pharmacotherapy 37(3): 367-370, 2003.
5. Sakakibara Y. Lower extremity hematoma as a complication of warfarinization in patients with artificial heart valves. Japanese Heart Journal 40(2):239-245, 1999.
6. Person DA, Pattekar MA. Myosistis Ossificans www.emedicine.com/  PED/topic1538.htm, 2006.
7. Coller BS, Schneiderman PI. Clinical Evaluation of hemorrhagic disorders: the bleeding history and differential diagnosis of purpura. In Hoffman R (ed.): Hematology: Basic Principles and Practice. Chapter 112, fourth edition, Churchill Livingston, 2005.
8. Kennedy M, Krusinski P, Dermatologic manifestations of hematologic disease. http://emedicine.medscape.com/article/1096183-overview , 2006.
9. Wendling P. Differential diagnosis of purpura delineated. Int Med News 38(20):33, 2005.