Risk factors associated with unintentional medication discrepancies at admission in an internal medicine department.

At admission, unintentional medication discrepancies (UMDs) can occur and may led to severe adverse events. Some of them are preventable through medication reconciliation (MR). As MR is a time-consuming activity, a better identification of high-risk patients of UMDs is mandatory. The objective was to identify risk factors associated with UMDs at admission in an internal medicine department. This prospective observational study was conducted from April 2017 to June 2019. At admission, inpatients had MR to obtain a complete list of home medications. This list was compared to prescriptions made at admission. All discrepancies were classified as intentional or UMDs. Univariate and multivariate analyses to identify the risk factors associated with UMDs were performed. MR was performed on 1157 patients (70.1 ± 16.8 years old); 550 MR (47.5%) contained at least one UMD. More than half of the UMDs (n = 892, 65.6%) corresponded to drug omission. The univariate analysis showed that age (> 60 years old), "living at home", medication preparation not performed by patient, medication-intake difficulties, number of sources consulted, MR duration, presence of a high-risk drug and the number of home medications were associated with UMDs. In the multivariate analysis, adjusted on the number of sources consulted, independent risk factors were "living at home" and the number of home medications. At admission to an internal medicine department, UMDs were frequent and associated with "living at home" and poly-medication. Our findings might help physicians to identify high-risk patients of UMDs since their admission.

CSF CA 15-3 in breast cancer-related leptomeningeal metastases.

UNLABELLED: The sensitivity of CSF cytology, the standard method for diagnosis of leptomeningeal metastases (LM), is low. Serum cancer antigen 15-3 (CA 15-3) is frequently used for the monitoring of patients with breast cancer (BC) and is a laboratory test available in most centers. The aim of the current study was to determine the feasibility of measuring CSF CA 15-3 and CA 15-3 CSF/serum ratio in patients with BC-related LM. Serum and CSF CA 15-3 values were evaluated in 20 BC patients with LM (Group 1), 20 patients with LM from other primary cancers (Group 2), 20 BC patients with parenchymal brain metastases only (Group 3) and 20 controls (Group 4). CSF and serum were collected on the same day. Serum and CSF CA 15-3 were assessed by an automatized immuno-enzymatic technology (TRACE(®) technology, KRYPTOR Automate, Brahms Society, France). In univariate analysis, BC patients with LM (Group 1) compared to other groups, a significantly elevated serum CA 15-3 (median 51 U/ml, range 12-2819) and CSF CA 15-3 (median 8.7 U/ml, range 0.1-251) was observed. Additionally, the CSF/serum ratio of CA 15-3 was significantly higher in this group of patients (median 0.18, range 0.002-4.40). Multivariate analysis identified a cut-off for CSF CA15-3 with 80 % sensitivity and 70 % specificity. CONCLUSIONS: The current study confirms the feasibility of determining CSF CA 15-3 using a widely available technology. Evaluation of the CSF CA 15-3 may be useful in the diagnosis and management of BC-related LM but further studies are needed.