Search

Background Tackling antimicrobial resistance by reducing unnecessary or inappropriate prescribing is a key priority. Objectives To review current clinical practice at Guy's and St Thomas’ Hospital around duration of IV antimicrobial therapy framed with Start Smart Then Focus national guidelines. Methods Newly started inpatient prescriptions for IV antimicrobials over three consecutive Mondays were reviewed. Prescriptions not for treatment of acute infections were excluded, as were patients admitted to critical care, obstetrics/gynaecology and paediatric wards. Each patient's record was reviewed at Day 0, 3 and 5. Data were analysed using IBM® SPSS Statistics version 27. Results In total, 80 patients with 89 antibiotic prescriptions (71 = β-lactam, 5 = aminoglycoside, 2 = glycopeptide, 1 = ciprofloxacin, 9 = metronidazole, 1 = linezolid) were included. Median age was 66 (IQR 53–75). Documented indications were for suspected or proven infection of urinary tract (15%), respiratory tract (25%), skin/soft tissue (14%), sepsis (1.25%), intra-abdominal (15%), bacteraemia (7.5%), bone and joint infections (2.5%) or other (20%). Median initial NEWS2 on Day 0 was 3 (IQR 1–4), WCC 11.45 (IQR 8.27–14.15) and CRP 121 (IQR 47–208). Median length of stay was 7.5 days. Median duration of IV antimicrobial was 4 (IQR 2–6) days. By D3, 3 patients had died and 16 had been discharged. Of 61 remaining inpatients on D3, 4 had their antimicrobials switched, 14 had been switched to PO antimicrobials, 43 were continued on IV antibiotics. By D5, 17 further patients had been discharged, 44 remained inpatients. Twenty-two had stopped antimicrobials with 22 continuing IV therapy. A number of patients could have been switched from IV to oral therapy by current guidance but were not (D3 x = 11, D5 n = 8). Higher median D0 CRP was seen in those who continued on IV therapy on D3 (139 versus 79 mg/dL) as was higher D3 WCC (13.9 versus 11.9 × 109/L). There was no significant relationship between a NEWS2 score ≥3 and IV to PO switch at D3 or D5. Thirty-two patients had input from the Infection team within 7 days of first prescription. Fifteen patients (21%) received IV antibiotics for more than 7 days, and 14/15 had Infection team input. Three patients were discharged with OPAT. Conclusions Median duration of IV therapy for suspected/proven infections in our centre is less than 5 days and only 28% of patients were still receiving IV antibiotics by D5. A number of missed opportunities for earlier switch to oral therapy were identified. The potential for earlier safe discontinuation of IV therapy using patient-specific measures should be explored. A minority of patients received IV therapy for >7 days, and this was almost universally associated with specialist Infection involvement in complex infections with a view to optimizing patient management and antimicrobial stewardship.

Objectives Despite the widespread use of vancomycin, optimization of dosing can be difficult. Increased vancomycin concentrations are associated with nephrotoxicity, while subtherapeutic levels may lead to inadequate antibacterial therapy and risk development of antimicrobial resistance. Recent guidelines have recommended the use of loading doses and switching to AUC24 TDM, targeting an AUC24:MIC ratio of 400–600 μg·h/mL to minimize toxicity while maintaining similar effectiveness for the treatment of serious infections. We aimed to review adherence to current local vancomycin dosing and conventional trough monitoring guidelines, as well as record time to, and factors related to, achieving target vancomycin troughs. Methods Fifty consecutive electronic prescriptions were screened for IV vancomycin use in treatment of active infections in a single acute Trust during April/May 2021. These were cross-referenced with microbiological, clinical and biochemical data from patient records. Variables were analysed by parametric or non-parametric methods on the basis of their normality distribution using GraphPad Prism. Results Forty-one patients were included of which 39 had a vancomycin level measured. Mean age was 57 years (range 19–88) and BMI 28 (range 17–44). Nineteen had unstable renal function, including one receiving intermittent haemodialysis. Median CLCR was 95 mL/min (IQR 66–129). Indications were included on the prescription for 26. Median duration of vancomycin was 4 days (range 2–12), and half of patients ultimately switched therapy to an alternate agent . Target troughs were documented in 29 (70%), 9 of which had a target trough ≥15 μg/mL. Twenty-four had positive microbiology samples, 19 of which were organisms against which vancomycin would have expected activity. Five (13%) failed to achieve trough levels >10, 6/39 (15%) achieved troughs of 10–15, 13/39 (33%) 15–20, and 15/39 (38%) had troughs of >20 μg/mL (Figure 1). Where it was achieved, the median time to level ≥10 μg/mL was 46 h (range 11–202) and time to level ≥15 μg/mL was 68 h (range 18–243). Twenty-five (60%) patients required titration above the currently recommended initial dose, and 22 required doses of 1500 mg twice daily or higher. BMI did not correlate with vancomycin level achieved. Conclusions A significant proportion of patients with serious infections fail to achieve or have a considerable delay in achieving therapeutic vancomycin trough. An even greater proportion have troughs above target and are at risk of nephrotoxicity. These findings support a potential switch to more personalized regimens using loading doses and AUC24 monitoring where vancomycin is the drug of choice. This would complement other stewardship measures to limit overall vancomycin use, achieve earlier and more complete source control, and optimize use of alternative antimicrobial therapies with more predictable PK/PD.