Thrombocytopenic purpura with depot medroxyprogesterone acetate and subsequent use of contraceptive implant: making decisions in the absence of evidence.

A 19-year-old woman presented to her general practitioner with a purpuric rash 3 weeks after receiving her first 150 mg intramuscular injection of depot medroxyprogesterone acetate (DMPA) at a family planning clinic. She had no medical or family history of note and was taking no other medication. Following investigation by the haematology department she was diagnosed with thrombocytopenic purpura. Her platelet count at diagnosis was 41×1109/l (local laboratory normal range: 100–1501×1109/l). Other haematological and biochemical indices were normal. Three weeks prior to the administration of DMPA the patient had a full blood count taken before a minor surgical procedure under general anaesthesia, and on that occasion her platelet count was normal. Two months after receiving DMPA the patient’s platelet count remained low (in the range 2–41×1109/l). She returned to the family planning clinic to discuss her contraceptive options. Since DMPA may have been a causal factor in the development of her thrombocytopenia there was concern that other progestogens might have a similar effect. However, other progestogens are reversible: DMPA is not. Despite an extensive literature search, no evidence was identified to support an association between other hormonal contraceptives and thrombocytopenic purpura. No information was identified to guide clinicians and patients on appropriate contraceptive use with thrombocytopenic purpura. This patient had used combined oral contraception in the past but expressed concern that she might not take it reliably, and this had been a factor in choosing DMPA. She did not wish to use an intrauterine device or system. After counselling the patient opted for a subdermal progestogen-only implant (Implanon®). Following discussion about the potential risks, and with the agreement of her haematologist, the implant was inserted. This was 14 weeks after the first injection of DMPA. The patient was not at risk of pregnancy at this time. Her platelet count at insertion was 81×1109/l. Her platelet count was assessed weekly. The implant was to be removed if her platelet count fell. The insertion procedure was uneventful. She was advised to contact the family planning clinic if she developed swelling or excessive tenderness at the site of insertion. Her platelet count returned to normal 5 months after her first