Impact of therapeutic drug monitoring of antibiotics in the management of infective endocarditis.

Treatment of infective endocarditis (IE) is based on high doses of antibiotics with a prolonged duration. Therapeutic drug monitoring (TDM) allows antibiotic prescription optimization and leads to a personalized medicine, but no study evaluates its interest in the management of IE. We conducted a retrospective, bicentric, descriptive study, from January 2007 to December 2019. We included patients cared for IE, defined according to Duke's criteria, for whom a TDM was requested. Clinical and microbiological data were collected after patients' charts review. We considered a trough or steady-state concentration target of 20 to 50 mg/L. We included 322 IE episodes, corresponding to 306 patients, with 78.6% (253/326) were considered definite according to Duke's criteria. Native valves were involved in 60.5% (185/306) with aortic valve in 46.6% (150/322) and mitral in 36.3% (117/322). Echocardiography was positive in 76.7% (247/322) of cases. After TDM, a dosage modification was performed in 51.5% (166/322) (decrease in 84.3% (140/166)). After initial dosage, 46.3% (82/177) and 92.8% (52/56) were considered overdosed, when amoxicillin and cloxacillin were used, respectively. The length of hospital stay was higher for patient overdosed (25 days versus 20 days (p = 0.04)), and altered creatinine clearance was associated with overdosage (p = 0.01). Our study suggests that the use of current guidelines probably leads to unnecessarily high concentrations in most patients. TDM benefits predominate in patients with altered renal function, but probably limit adverse effects related to overdosing in most patients.

[Para valvular leak closure in TAVI]. / Fermeture des fuites para-prothétiques des TAVI.

Literature concerning transcutaneous symptomatic para valvular cardiac leaks closure (PVLC) after trans aortic valve implantation (TAVI) is relatively scarce. Hereby we present 2 clinical cases, one on an Edwards® Sapien 3 valve and the other one on a Medtronic® Evolut R valve. We present also the preliminary results of the 7 PVLC on TAVI included in our prospective FFPP registry during the 2 first years of enrolment (2017-2018), for a total of 158 inclusions for all valves. Seven procedures were performed on 8 leaks, using a majority of vascular plugs (3 Abbott® Amplatzer Vascular Plugs 2 (AVP2), 3 AVP3, 1 AVP4, and 1 muscular Ventricular Septal Defect (VSD) occluder). All procedures were successful without complication. At 1-month follow-up, all patients became asymptomatic. One-year follow-up was already available for 4 patients: 3 of them were symptoms free, and one-who had a second leak not suitable for PVLC-, underwent a « TAVI in TAVI ¼ procedure 2 months after PVLC. This short experience demonstrates the feasibility, the efficacy and the safety of PVLC on TAVI. We expect to be able to offer more in depth information at the end of our prospective ongoing study.

Crystalloids versus colloids for fluid resuscitation in critically-ill patients.

The choice of crystalloid or colloid for fluid resuscitation has been debated for the last few years. Although colloids seems to be more interesting when taking into account their physiological properties, their effect on mortality is not better than crystalloids if they are used in an adequate amount. Moreover, colloids' side effects are far more important than those of crystalloids. Several randomised studies pointed out the renal effects of colloids including acute renal injury with an increased need of renal replacement therapy. An unacceptably high rate of renal side effects has resulted in premature termination of some clinical trials. In addition, homeostatic and anaphylactoid effects of colloids on coagulation and on anaphylaxis may increase the risk of death associated with their use. Finally, colloids are much more expensive than crystalloids. For all these reasons, we conclude that crystalloids should be preferred to colloids for fluid resuscitation.