Penicillins are one type of the most important antibiotics used in the clinic.Control of drug impurity profiles is an important part of ensuring drug safety.This is particularly important in penicillins where polymerization can lead to polymers as elicitors of passive cutaneous anaphylaxis.The current under-standing of penicillin polymerization is based on reactions with amino groups,but no comprehensive mechanistic understanding has been reported.Here,we used theoretical calculations and column switching-LC/MS techniques to study penicillin dimerization.Ampicillin and benzylpenicillin were selected as representative penicillins with or without amino groups in the side chain,respectively.We identified four pathways by which this may occur and the energy barrier graphs of each reaction process were given.For benzylpenicillin without an amino group in the 6-side chain,dimerization mode A is the dominant mode,where the 2-carboxyl group of one molecule reacts with the β-lactam of another molecule.However,ampicillin with an amino group in the 6-side chain favors dimerization mode C,where the amino group of one molecule attacks the β-lactam of another molecule.These findings can lead to a polymer control approach to maintaining penicillin antibiotics in an active formulation.

Objective:To investigate the characteristics and relationship between the location of lower extremity deep vein thrombosis (DVT) and the site of pulmonary embolism in hospitalized patients.Methods:The data of patients with lower extremity DVT diagnosed by ultrasound examination and pulmonary embolism diagnosed by CT pulmonary angiography from December 2017 to December 2021 were analyzed retrospectively. According to the location of lower extremity DVT, the patients were divided into mixed DVT, proximal DVT, and distal DVT which was further divided into anterior/posterior tibial vein or peroneal vein thrombosis and calf muscular venous thrombosis. Mixed DVT was referred to the presence of both proximal and distal DVT. According to the involved site of pulmonary artery, pulmonary embolism was divided into three types: main pulmonary artery, left or right pulmonary artery trunk embolism, lobar pulmonary artery embolism and segmental pulmonary artery embolism. The location of lower extremity DVT, the site of pulmonary embolism, the clinical manifestation (shortness of breath, chest tightness, chest pain, hemoptysis, cough, lower limb swelling, lower limb pain, syncope, fever) and risk factors (fracture/trauma, tumor, diabetes, hypertension, atrial fibrillation, infection, surgery, autoimmune diseases, paralysis, pregnancy) of venous thromboembolism (VTE), and the level of D-dimer were analyzed.Results:A total of 209 patients were enrolled finally, including 127 patients with left lower extremity DVT (60.8%) and 82 with right lower extremity DVT (39.2%). Mixed DVT accounted for 39.2%, proximal DVT accounted for 17.3%, and distal DVT accounted for 43.5% (anterior/posterior tibial vein and peroneal vein thrombosis accounted for 14.8%, calf muscular venous thrombosis accounted for 28.7%). The incidences of main pulmonary artery embolism, left or right pulmonary artery trunk embolism in the mixed DVT and proximal DVT were significantly higher than those in the anterior/posterior tibial vein or peroneal vein thrombosis and calf muscular venous thrombosis [41.5% (34/82), 38.8% (14/36) vs. 16.2% (5/31), 10.0% (6/60)], with statistically significant differences (all P < 0.05). The incidences of pulmonary segmental artery embolism in the anterior/posterior tibial vein or peroneal vein thrombosis were higher than those in the mixed DVT and proximal DVT [41.9% (13/31) vs. 26.8% (22/82), 30.6% (11/36)], but the difference was not statistically significant (both P > 0.05). The incidences of pulmonary segmental artery embolism in the calf muscular venous thrombosis were significantly higher than those in the mixed DVT and the proximal DVT [66.7% (40/60) vs. 26.8% (22/82), 30.6% (11/36)], and the difference was statistically significant (both P < 0.05). The levels of D-dimer in patients with calf muscular venous thrombosis combined with main pulmonary artery embolism, left or right pulmonary artery trunk embolism were significantly higher than those in patients with calf muscular venous thrombosis combined pulmonary segmental artery embolism (mg/L: 6.08±3.12 vs. 3.66±2.66, P < 0.05). There were no significant differences in D-dimer levels in other patients with DVT combined with pulmonary embolism in different sites. In terms of the clinical manifestations of VTE, the incidences of lower limb swelling in the mixed DVT and proximal DVT were significantly higher than those in the anterior/posterior tibial vein or peroneal vein thrombosis and calf muscular venous thrombosis [54.9% (45/82), vs. 29.0% (9/31), 15.0% (9/60), both P < 0.05], the incidences of lower limb swelling in the proximal DVT were significantly higher than those in the calf muscular venous thrombosis [41.7% (15/63) vs. 15.0% (9/60), P < 0.05], there were no significant difference in the other clinical manifestations among the DVT groups. There was no significant difference in the incidence of VTE risk factors among the groups. Conclusions:The DVT of inpatients mostly occurred in the left lower limb, and the incidence of distal DVT was higher than that of proximal DVT. Mixed DVT and proximal DVT combined with pulmonary embolism mostly occurred in the main pulmonary artery, left or right pulmonary artery trunk, while distal DVT combined with pulmonary embolism mostly occurred in the pulmonary segmental artery. The levels of D-dimer in patients with lower extremity DVT combined with main pulmonary artery or left and right pulmonary artery trunk embolism were higher than those in patients with pulmonary lobe and segmental artery embolism. The incidence of lower extremity swelling in patients with mixed DVT and proximal DVT was higher than that in patients with distal DVT.