Recent guidelines [4, 5] suggest a loading dose of 25–30 mg/kg to achieve the targeted range for treatment with vancomycin. In our study, however, the proportions of patients achieving 10–20 µg/mL and 15–20 µg/mL at 48 h after the first dose were 56.9% and 5.6%, respectively, in patients with an eGFR ≥ 90 mL/min/1.73 m2 who received a regimen with a loading dose of 25 mg/kg. There was no significant difference in the Cmin between the regimen with and without a loading dose. Rosini et al.  reported that the 12-h Cmin allowed a significantly greater proportion of patients to achieve 15 μg/mL among patients with a loading dose of 30 mg/kg followed by 15 mg/kg twice daily compared with those without a loading dose. However, the effect of the loading dose was attenuated at the steady state, and only 20% of patients with a loading dose attained the target Cmin at steady state. Most of their study patients also had a high CLcr (mean 93 mL/min). However, in a study including patients with deteriorated renal function, a loading dose resulted in a higher concentration at steady state [16, 17]. In patients with deteriorated renal function, not only a loading dose but also a maintenance dose had a significant impact on the Cmin.
Why the Cmin was low in our study might be explained by only three obese patients being included. Pharmacokinetic (PK) data indicated a larger volume of distribution and accelerated renal clearance of vancomycin in obese patients , which might justify a weight-based dosing regimen. However, a study using weight-based regimens by Reynolds et al.  reported a Cmin > 20 μg/mL in approximately 50% of obese patients. Richardson et al.  reported that increasing BMI categories were associated with a Cmin >20 μg/mL. Leong et al.  demonstrated that an adjusted body weight was more accurate at predicting vancomycin clearance in obese individuals.
A loading dose alone may not be sufficient to increase the Cmin measured 48 h after the initial dose in patients without decreased vancomycin clearance. However, the main purpose of a loading dose is not to obtain the target Cmin at steady state, but the rapid achievement of a therapeutic concentration within 12–24 h . To maintain a high concentration thereafter, an increase in the maintenance dose from 15 mg/kg to 20 mg/kg might be required . Many studies reported a higher achievement of a Cmin >15 μg/mL after the first dose in patients with a loading dose compared with patients without a loading dose [10, 15, 23, 24]. Casapao et al.  evaluated the association between the day 1 vancomycin exposure profile and outcomes among patients with MRSA infective endocarditis, and an AUC/MIC of < 600 was independently associated with failure.
Few studies have focused on the use of a loading dose to increase clinical efficacy for the treatment of MRSA infections. Jeon et al.  reported an initial daily dose per weight < 40 mg/kg was a strong risk factor for vancomycin non-responsiveness in patients with MRSA pneumonia. Wesolek et al.  reported that an initial dose > 20 mg/kg led to a faster resolution of systematic inflammatory response syndrome (> 20 mg/kg: 67 h vs ≤ 20 mg/kg: 109 h). In our study, a loading dose increased the rate of early clinical response at 48–96 h after the start of therapy. However, there was no significant difference in clinical success at EOT between the regimen with and without a loading dose. Clinical response beyond 72 h of treatment can be affected by changes in the vancomycin dose later in therapy, based on the initial TDM conducted. Indeed, the achievement rate of target Cmin of 10–20 μg/mL was increased from 7.4% to 74.1% of patients in whom the dosage was increased after initial TDM in our study.
PK analysis indicated that a loading dose of 25–30 mg/kg could be applied irrespective of renal function . Clinicians, however, tend to be reluctant to use a loading dose, especially in patients with decreased renal function, for fear of causing renal injury. Alvarez et al.  demonstrated that all patients with a loading dose who presented with decreased vancomycin clearance before the administration of vancomycin had an elevated serum concentration (>28 μg/mL) in the first 24 h of treatment. These patients, however, did not have nephrotoxicity. Although prolonged exposure to elevated Cmin causes nephrotoxicity, a temporary increase in the Cmin with an initial loading dose might not cause nephrotoxicity. Marvin et al.  reported that a loading dose > 20 mg/kg was not associated with increased nephrotoxicity in patients with a CLcr < 30 mL/min compared with a standard dosing regimen.
This study had some limitations. First, we did not measure the Cmin 24 h after the first dose, and the early attainment of the target Cmin with a loading dose was not confirmed. Second, early clinical responses were evaluated at 48–72 h after the first dose. Although an early clinical response was defined as improvement at 48–72 h after the start of therapy in patients with acute bacterial skin and skin-structure infections , a significantly lower response rate in respiratory tract infection in our study might be caused by this early clinical response evaluation. Third, APACHE II scores were approximately 8.0, indicating mild-to-moderate severity. Although the Infectious Diseases Society of America guidelines3 recommend a loading dose for seriously ill patients, the efficacy of a loading dose was not confirmed in severely ill patients in our study. Fourth, patients with an eGFR ≥90 mL/min/1.73m2 were included in our study, which might explain the low Cmin, value observed, and thus exaggerate the weakness of the regimen without a loading dose.