Continuous infusion (CI) of beta-lactams could optimize their pharmacokinetic/pharmacodynamic indices, especially in difficult-to-treat infections. Purpose To validate an easy-to-use method to guide beta-lactams dosage in CI (formula). Methods A retrospective analysis was conducted of a prospectively collected cohort ( n  = 24 patients) with osteoarticular infections caused by Gram-negative bacilli (GNB) managed with beta-lactams in CI. Beta-lactams dose was calculated using a described formula (daily dose = 24 h × beta-lactam clearance × target “steady-state” concentration) to achieve concentrations above the MIC. We correlated the predicted concentration ( C pred  = daily dose/24 h × beta-lactam clearance) with the patient’s observed concentration ( C obs ) measured by UPLC–MS/MS ( Spearman’s coefficient). Results The most frequent microorganism treated was P. aeruginosa (21 cases; 9 MDR). Beta-lactams in CI were ceftazidime ( n  = 14), aztreonam (7), and piperacillin/tazobactam (3), mainly used in combination (12 with colistin, 5 with ciprofloxacin) and administered without notable side effects. The plasma C obs was higher overall than C pred ; the Spearman correlation between both concentrations was rho = 0.6 (IC 95%: 0.2–0.8) for all beta-lactams, and rho = 0.8 (IC 95%: 0.4–1) for those treated with ceftazidime. Conclusions The formula may be useful in clinical practice for planning the initial dosage of beta-lactams in CI, while we await a systematic therapeutic drug monitoring. The use of beta-lactams in CI was safe.