Discussion

The aetiology of thrombocytopenia in hospitalised patients is heterogeneous and commonly includes infections, thrombotic thrombocytopenic purpura (TTP), disseminated intravascular coagulation, heparin and drugs such as antibiotics.5 Thrombocytopenia is also a common clinical manifestation in patients with SLE.6 Zhan et al reported that SLE-associated thrombocytopenia had a median platelet count of 66×10⁹/L with a range of 29–95×10⁹/L.7 Accordingly, the causes of acute reactive thrombocytopenia were highly suspected other than SLE itself. In our case the underlying causes were TTP, heparin or drug reactions.

TTP is a rare and potentially lethal thrombotic microangiopathy typically characterised by thrombocytopenia, microangiopathic haemolytic anaemia, neurological abnormalities and a severe deficiency or reduction of ADAMTS13 activity (less than 10%).8 Despite its rare occurrence, the association of TTP and SLE has been described in the literature.9 Our patient had a normal baseline platelet count on admission and remained stable during the first 2 days of treatment, but acute thrombocytopenia occurred unexpectedly on day 4 with a minimum platelet count of 1×10⁹/L. Compared with the corresponding results on admission, the platelet count and D-dimer level decreased more significantly within 24 hours after the onset of acute reactive thrombocytopenia (table 1). However, there were no significant changes in the red blood cell count, haemoglobin, lactic dehydrogenase and indirect bilirubin levels. The clinical symptoms were dominated by drowsiness and petechiae at the injection sites. In addition, the results of anti-ADAMTS13 IgG and ultrasonography did not support the diagnosis of TTP. Therefore, TTP-induced thrombocytopenia was ruled out first.

The cause of thrombocytopenia was then assessed according to the type of medication or duration of medication, and heparin- or drug-induced reactive thrombocytopenia was speculated to be the most likely cause. As enoxaparin and sulperazon both have the side effects of thrombocytopenia, we recommended the clinicians withdraw them immediately. The patient was treated with intravenous infusions of immunoglobulin 25 g daily for 6 days and the platelet count gradually returned to the normal range.

Enoxaparin is a low-molecular weight heparin that has a lower incidence of heparin-induced thrombocytopenia (HIT).10 Currently, HIT is not a plausible diagnosis for our patient due to the following reasons: (a) the onset of thrombocytopenia began only 4 days after starting enoxaparin, but HIT is a moderate thrombocytopenia that starts at least 5–10 days after heparin administration, too soon to be explained by heparin-induced immunisation; (b) HIT is not characterised by such severe thrombocytopenia and the platelet count is rarely less than 20×10⁹/L,11 but our patient’s platelet count fell to 1×10⁹/L; and (c) according to the widely used 4Ts scoring system, the total score of our patient was only 2, which had a very high negative predictive value of 97–99% for HIT.12 Therefore, HIT was largely excluded.

Thus, sulperazon was the most likely cause. Typically, drug-induced immune thrombocytopenia (DITP) occurs in one to several weeks following drug exposure.13 DITP, which refers only to immune-mediated platelet destruction rather than myelosuppression, is characterised by severe thrombocytopenia and is often misdiagnosed as idiopathic thrombocytopenic purpura. Its incidence is unknown as well as it is not fully recognised by clinicians.14 Antibodies only react with platelets in the presence of the sensitising drug, antibody drug/platelet interactions may result in severe thrombocytopenia.15 In the present case, the reactive thrombocytopenia showed a temporal relationship with sulperazon administration. When sulperazon was discontinued for 6 days, the platelet levels were reversed. However, the acute thrombocytopenia of our patient initially occurred within 29 hours of exposure to sulperazon, implying that drug-induced antibodies may have been present in vivo before exposure. Once the sensitising drug was withdrawn, the binding affinities of these antibodies were too low to cause severe thrombocytopenia. Further inquiry about the patient's medication history revealed that he had used ceftriaxone 8 days before admission. Evidence already exists that ceftriaxone can induce immune-mediated thrombocytopenia.13 Since both belong to the third-generation cephalosporins, cross-reactivity has also been described between cefoperazone and ceftriaxone antibodies.16 We therefore suggest that pre-existing cross-reactive antibodies to sulperazon may be responsible for this rapid onset of thrombocytopenia. The exact pathogenesis of immune thrombocytopenia induced by beta-lactam drugs is not fully understood. Studies have suggested that it is mediated by IgG and IgM immunoglobulins that recognise membrane glycoproteins expressed on platelets, ultimately activating the complement system and promoting platelet destruction.17 Although DITP has a mild clinical course in most cases, the fact that in this case it was severe may be due to cross-reaction between cefoperazone and ceftriaxone. This cross-reaction can be explained by the reaction of side chains shared by several different beta-lactam drugs with the patient’s antibody, resulting in the specificity of the platelet glycoprotein target.18

Based on these facts, we conclude that the self-limited acute reactive thrombocytopenia was most likely induced by sulperazon. Because sulperazon is well tolerated by most patients, a rapid decrease in the platelet count in a short time may be related to individual factors. One limitation of this report is the absence of any further experimental methods for detecting drug-induced antibodies. Regardless of this, the present case highlights that DITP must be considered in the setting of unexplained thrombocytopenia.