Endocarditis: Why fixed-dose ampicillin often misses the mark

Infective endocarditis, a severe infection of the heart's inner lining, carries substantial morbidity and mortality, particularly in vulnerable populations. Rheumatic heart disease, for instance, disproportionately affects Indigenous communities in Australia and New Zealand, leading to a cascade of cardiac complications including valve disease, heart failure, and endocarditis.¹ Similarly, pneumococcal endocarditis, while less common, presents a serious threat, especially in pediatric populations.² The challenge in treating these infections often lies not just in identifying the pathogen, but in ensuring adequate and sustained antibiotic exposure at the site of infection. Traditional fixed-dose regimens frequently fall short in achieving optimal drug concentrations due to individual patient variations in drug metabolism and clearance.

A recent proof-of-concept study explored the utility of a therapeutic drug monitoring (TDM)-guided strategy for optimizing continuous infusion ampicillin-based regimens in patients with enterococcal bloodstream infections and/or endocarditis.³ The study, published in Antibiotics (Basel), aimed to demonstrate the feasibility and effectiveness of this personalized approach in a real-world clinical setting. The investigators enrolled a series of patients, focusing on those with severe enterococcal infections where achieving specific pharmacokinetic/pharmacodynamic (PK/PD) targets was deemed essential for clinical success. This patient cohort often presents with complex comorbidities and altered physiology, making standard dosing less reliable.

The rationale for personalized dosing

Enterococci, particularly Enterococcus faecalis, are common culprits in infective endocarditis and bloodstream infections, often requiring prolonged courses of high-dose antibiotics. Ampicillin, a cornerstone of therapy for susceptible enterococcal infections, exhibits time-dependent killing, meaning its efficacy is best predicted by the duration for which its concentration remains above the minimum inhibitory concentration (MIC) of the pathogen. For beta-lactam antibiotics like ampicillin, the PK/PD target often sought is a free drug concentration above the MIC for 100% of the dosing interval (fT>MIC = 100%). Achieving this target consistently can be challenging with intermittent dosing, especially in critically ill patients with fluctuating renal function or altered volume of distribution.

Continuous infusion of ampicillin offers a theoretical advantage by maintaining steady drug concentrations, thereby maximizing the time above MIC. But even with continuous infusion, inter-patient variability in drug clearance can lead to significant deviations from the desired PK/PD target. This is where TDM becomes invaluable. By measuring actual drug concentrations in a patient's plasma, clinicians can adjust infusion rates to ensure the therapeutic target is met, or to avoid toxicity. The study's premise was that this individualized approach could improve clinical outcomes by optimizing antibiotic exposure, a concept that has gained traction in other difficult-to-treat infections.

What the study actually measured

The proof-of-concept study focused on a case series of patients diagnosed with enterococcal bloodstream infections and/or endocarditis.³ The investigators administered ampicillin via continuous infusion and used TDM to guide dose adjustments. The primary objective was to demonstrate the usefulness of this TDM-guided strategy in achieving the predefined PK/PD target of fT>MIC = 100%. They collected blood samples at specific time points after initiating the continuous infusion and after any dose adjustments, then measured ampicillin concentrations using validated analytical methods. These measured concentrations were then used to calculate the free drug concentration and compare it against the MIC of the isolated enterococcal strain.

The study meticulously documented the initial ampicillin dosing, subsequent TDM results, and any dose modifications made based on these results. They also recorded clinical outcomes, including resolution of infection and adverse events, though the small sample size of a case series meant these were secondary observations rather than statistically powered endpoints. The focus remained squarely on the pharmacokinetic success of achieving the fT>MIC target. This detailed approach allowed for a granular understanding of how individual patient characteristics influenced ampicillin pharmacokinetics and how TDM could effectively counteract this variability.

The numbers

The TDM-guided strategy successfully achieved the target fT>MIC = 100% in all patients included in the case series.³ This outcome demonstrates the feasibility of using TDM to personalize ampicillin dosing for enterococcal infections. The investigators observed that initial empiric dosing often required adjustment, highlighting the variability in ampicillin pharmacokinetics among patients. For instance, some patients required higher doses than initially prescribed to reach the target, while others achieved it with standard or even slightly lower doses. This individualized adjustment was critical, particularly in patients with endocarditis, where underdosing could lead to treatment failure and overdosing could increase the risk of adverse drug reactions.

The study did not provide specific hazard ratios or p-values for clinical outcomes, as it was designed as a proof-of-concept for pharmacokinetic optimization rather than a randomized controlled trial of clinical efficacy. But the consistent achievement of the PK/PD target across all cases suggests a strong foundation for future larger trials. The ability to maintain optimal drug exposure is a prerequisite for antibiotic efficacy, especially against challenging pathogens like enterococci in biofilm-rich environments such as heart valves. This approach aligns with broader efforts to combat antibiotic resistance by ensuring every dose counts.

Where it falls short

The primary limitation of this work is its design as a case series. While it effectively demonstrated the proof of concept for TDM-guided ampicillin dosing, it did not provide comparative data on clinical outcomes against standard dosing regimens. The small number of patients (N=?) also restricts the generalizability of the findings and prevents any definitive conclusions regarding improved patient survival or reduced treatment failure rates. Larger, randomized controlled trials are necessary to establish the clinical superiority of a TDM-guided approach over empirical dosing in terms of hard clinical endpoints. The study also did not examine specific patient subgroups, such as those with renal impairment or obesity, where pharmacokinetic variability might be even more pronounced and TDM potentially more impactful. The absence of detailed demographic and clinical characteristics for each case also limits the ability to draw correlations between patient factors and dosing requirements. Still, the consistent achievement of the PK/PD target in all cases provides a strong signal for further investigation.

The logistical challenges of implementing TDM in routine clinical practice are also worth considering. TDM requires specialized laboratory infrastructure for drug concentration measurements and expertise in pharmacokinetic interpretation to guide dose adjustments. Not all institutions have these resources readily available, which could limit the widespread adoption of such a personalized strategy. The cost-effectiveness of TDM-guided therapy, particularly in the context of prolonged treatment for endocarditis, also warrants further economic analysis. Despite these practical considerations, the principle of optimizing drug exposure for severe infections remains compelling. For clinicians managing complex cases, a comprehensive reference like the Oxford Handbook of Infectious Diseases and Microbiology can provide valuable insights into antimicrobial therapy and resistance patterns.

The study also did not explicitly address potential drug-drug interactions that might influence ampicillin pharmacokinetics, a common concern in polymedicated patients with endocarditis. Many patients with IE have underlying cardiac conditions and may be on multiple medications, including anticoagulants, antiarrhythmics, and other cardiovascular drugs. The impact of these concomitant medications on ampicillin clearance and distribution could be significant, and a TDM-guided approach would inherently account for these interactions by measuring the net effect on drug concentrations. But the study did not detail the co-medications of the included patients, which is a missed opportunity to highlight the robustness of TDM in complex polypharmacy scenarios. Future research should explicitly explore these interactions and how TDM can mitigate their impact on antibiotic efficacy.

Another area not fully explored was the potential for resistance development under suboptimal dosing. While the study focused on achieving the fT>MIC target, the long-term implications of maintaining these concentrations on preventing the emergence of resistance were not assessed. In enterococcal infections, particularly those involving prosthetic valves or devices, the risk of resistance is a constant concern. Ensuring optimal drug exposure from the outset is a key strategy in preventing resistance, but this study did not follow patients long enough to observe such outcomes. The broader context of antibiotic resistance demands that every therapeutic decision be made with an eye toward preserving the efficacy of existing agents.

The study also did not specify the MIC values for the enterococcal strains isolated from each patient. While the fT>MIC target was stated, the actual MICs are important for understanding the magnitude of the therapeutic challenge. Higher MICs would necessitate higher drug concentrations to achieve the same fT>MIC target, potentially pushing doses into ranges where toxicity becomes a greater concern. Providing the range or median MICs would have added valuable context to the dosing adjustments made. This information is particularly relevant for clinicians who might encounter strains with varying susceptibility profiles in their practice. The absence of this detail means the reader must assume that the MICs were within a treatable range, which is not always the case in real-world scenarios.

Finally, the study did not discuss the specific methods used for determining the free drug concentration of ampicillin. While it stated that free drug concentration was considered, the practical aspects of how this was achieved (e.g., ultrafiltration) were not detailed. The free drug concentration is the pharmacologically active component, and accurately measuring it is critical for precise TDM. Total drug concentration measurements can be misleading, especially in patients with altered protein binding, which is common in critical illness. Clarifying this methodological detail would have strengthened the scientific rigor of the proof-of-concept. The next step for this research would be a randomized trial comparing TDM-guided continuous infusion to standard intermittent dosing, powered for clinical endpoints like mortality and treatment success.