Yasunaga et al. [5] reported the side effect of cefminox sodium on coagulation function for the first time in 1990. PT and APTT were prolonged in 0.29% of patients, fibrinogen (FIB) was decreased in 0.15% of patients, and no thrombocytopenia was found. Patients with underlying chronic diseases, especially those with malignant tumors, were more likely to have coagulation dysfunction after using the drug. In general, coagulopathy caused by cefminox sodium is rare, and few clinical cases have been reported.

As a kind of β-lactam antibiotics, the main cause of coagulation dysfunction caused by cefminox sodium is vitamin K (VitK)-dependent hypoprothrombinemia [6, 7]. Human VitK comes from food intake on the one hand, and from intestinal bacteria synthesis on the other hand. β-lactam antibiotics can inhibit and kill normal intestinal flora, resulting in the reduction of VitK synthesis and the disorder of VitK-dependent coagulation factors (II, VII, IX and X), which results in coagulopathy. Coagulation dysfunction caused by cefoperazone/sulbactam sodium is the most common [8, 9]. In a large retrospective study of 23,242 patients involved, prolonged PT and coagulopathy with cefoperazone/sulbactam were found in 5.3% and 9.2%, respectively, with thrombocytopenia in 15.7% and bleeding in 4.2% of patients [10]. Severe gastrointestinal bleeding can occur in a very small number of patients after the use of cefoperazone [11]. The second reason of coagulation dysfunction caused by cefminox sodium is the N-methylthiotetrazole (NMTT) group on the 3rd side chain of cefminox sodium, which directly inhibits mitochondrial shuttling activity or VitK oxidoreductase and affects the VitK-dependent coagulation factor carboxylation, causing VitK-dependent coagulation factor deficiency [12].

No bacterial infection occurred in our reported patient during the previous multiple hospitalizations, and no cefminox sodium was used before. In June 2018, the first time bacterial infection occurred during her hospitalization. According to the patient's abdominal discomfort symptoms, abdominal suspicious tenderness, elevated blood bacterial infection index (total white blood cells, neutrophil ratio, C-reactive protein, procalcitonin), ascites showed by ultrasound, peritonitis was considered. Because the patient refused to perform abdominal paracentesis to extract ascites for laboratory test, she was empirically treated with cefminox sodium. After treatment, the patient's fatigue increased progressively, spirit and appetite were very poor, and the deterioration of coagulation function was mistaken for the deterioration of liver disease and liver failure. After stopping cefminox sodium, the patient's fatigue gradually alleviated, the spirit and appetite gradually improved, and the coagulation function improved, which was mistakenly thought the improvement was caused by plasma transfusion, and then the general condition improved. Therefore, on August 4, when the patient suffered from peritonitis again, cefminox sodium was applied again for anti-infective treatment according to experience, and the same performance appeared as the first application of cefminox sodium. At the same time, it was found that the coagulation function was still deteriorating sharply during the second application of cefminox sodium with supportive treatment. However, after cefminox sodium was discontinued, the coagulation function was improved very quickly even without plasma transfusion. During the first and second application of cefminox sodium, the serum bilirubin levels didn’t change significantly when the coagulation function deteriorated sharply (Fig. 2), which suggested that the deterioration of coagulation function wasn’t caused by the deterioration of liver disease, but by cefminox sodium. Unfortunately, we mistakenly diagnosed that the deterioration of coagulation function was related to liver failure, so we only treated the patient with intermittent plasma transfusion. We didn't supplement vitamin K because we didn't realize the deterioration of coagulation function caused by vitamin K deficiency. In addition, our patient's liver function was classified as Child Pugh C, and vitamin K usage was forbidden for those with severe liver diseases or poor liver function. We didn’t change antimicrobials because we didn’t realize the coagulopathy side effect of cefminox sodium. Fortunately, we didn’t use cefminox sodium for a long time due to its good anti-infective effect. After stopping the drug, the patient's coagulation function quickly recovered and there were no fatal consequences happened.

Most of the coagulation factors or their intermediates are synthesized by the liver. FIB is synthesized by hepatocytes, prothrombin is synthesized by hepatocyte microsomes, and coagulation factors VII, X, VIII, IX and V may be synthesized in the liver. Patients with liver diseases, especially those with cirrhosis and liver failure, are often accompanied with coagulation dysfunction [13], and such patients are often complicated with peritonitis. In cases of further deterioration of coagulation function caused by β-lactam antibiotics represented by cefminox sodium, the Naranjo score [14] and other liver function indexes (especially serum bilirubin levels) of patients can be combined to determine whether the deterioration of coagulation function caused by liver disease deterioration or by antibiotics. If it is caused by drugs, the drug should be stopped or other kinds of antibiotics should be replaced in time.

We should not regard all cases of the coagulation function deterioration (such as PT prolonging, PTA declining, INR decreasing, etc.) as the aggravation of liver disease or liver failure in medical practice. If the patient is in the course of β-lactam antibiotics administration, his/her coagulation function deteriorates rapidly but there are no matching performances such as rapid increase of serum bilirubin levels and aggravation of general conditions, the coagulation dysfunction caused by antibiotics should be considered. Protein induced by vitamin K absence or antagonist-II (PIVKA-II) can be detected to judge whether it is VitK-dependent hypoprothrombinemia caused by antibiotics [15, 16]. If the side effect of antibiotics is confirmed or highly suspected, stopping or changing antibiotics in time can effectively avoid the occurrence of malignant events. When liver function of the patient permits, VitK injection can promote the rapid recovery of coagulation function [17]. This case provides a warning role to our doctors, especially hepatologists. We should break the mindset and broaden the thinking model in the clinical diagnosis and treatment process.