FormalPara Key Summary points

Why carry out this study?

This is the first study to evaluate the effect of pharmacist-driven dosing and monitoring services on the outcomes of patients under teicoplanin treatment.

It has been well established that pharmacist dosing and monitoring services can enhance vancomycin and aminoglycoside treatments. Nevertheless, in contrast to the aforementioned two drugs, an extremely high variability in serum concentration was found in teicoplanin. Only very few studies with small sample sizes to date have evaluated the efficacy of pharmacist-driven dosing and monitoring service on the outcomes of patients treated with teicoplanin.

Therefore, our team hypothesized that pharmacist-driven dosing and monitoring services may improve the clinical and economic outcomes of patients treated with teicoplanin.

What was learned from the study?

Our study found that pharmacist-driven dosing and monitoring services resulted in a higher likelihood of attaining target concentration (54% versus 16%) and significantly improved clinical and economic outcomes. A significantly lower rate of sepsis, septic shock, and shorter length of hospital stay were also observed. Economic benefits discovered during the study included lower drug costs and lower total hospitalization costs. No difference was found in the incidence of nephrotoxicity.

Introduction

Pharmacist-driven (PD) dosing and monitoring services in antibiotics based on pharmacokinetics have been shown to improve clinical outcomes [1,2,3,4,5]. The Infectious Diseases Society of America recommends implementing pharmacokinetic dosing services for patients as part of an antimicrobial stewardship program, and pharmacists now play a key role in antimicrobial stewardship to optimize antimicrobial selection, dose, and duration [6, 7]. Since the 1980s, the pharmacist-managed vancomycin and aminoglycoside strategies have been demonstrated to provide benefits in both clinical and economic outcomes [8, 9]. Pharmacist-managed pharmacokinetic dosing for vancomycin and aminoglycosides can reduce nephrotoxicity and increase target concentration attainment [1,2,3,4,5]. Further studies are needed for PD antibiotic therapies, other than vancomycin and aminoglycosides.

Following decades of clinical application, teicoplanin is now regarded as an alternative to vancomycin for treating infections caused by Gram-positive bacteria, particularly those caused by methicillin‐resistant Staphylococcus aureus (MRSA) [10]. Despite that the elimination of teicoplanin occurs primarily through the kidneys, teicoplanin has been reported to be less nephrotoxic than vancomycin [11, 12]. The Japanese Society of Chemotherapy and the Japanese Society of Therapeutic Drug Monitoring recently published clinical practice guidelines and recommended a target Cmin value of 15–30 mg/L in patients with non-complicated MRSA infections [13]. However, teicoplanin’s Cmin in actual clinical practice has been reported to range from 3.6 to 24.2 mg/L [14], indicating that unstable drug exposure is frequently observed during teicoplanin therapy. This high variability may lead to insufficient exposure to teicoplanin in patients with varying pathophysiological statuses [15,16,17]. Therefore, the dosing strategy should be calculated and personalized according to individual patients’ pathophysiological statuses and teicoplanin’s pharmacokinetics. Additionally, teicoplanin dosage regimens should be adjusted in a timely manner by monitoring the serum concentrations and clinical response [18]. Pharmaceutical care and therapeutic drug monitoring (TDM) provided by clinical pharmacists during teicoplanin therapy may help to improve the clinical outcomes in patients with MRSA infections. Besides the clinical benefits, in 2005, a nationwide analysis carried out in the USA showed that the presence of pharmacist-driven vancomycin or aminoglycoside dosing and monitoring service was associated with improvement of clinical and economic outcomes for patients who received these drugs [9]. An analysis that included 303 patients under vancomycin treatment demonstrated an overall cost saving of 5.1% under the care of the pharmacist [19]. This means that while improving the clinical outcomes of patients, the burden of medication cost may also be relieved. Our study focused on non-critically ill patients, since vancomycin is still considered to be the first-line choice of treating infections caused by Gram-positive bacteria in critically ill patients.

Referring to the above, this study aims to investigate the effect of pharmacist-driven teicoplanin therapy and pharmaceutical care on the clinical and economic outcomes of non-critically ill patients.

Methods

Study Design and Patient Enrollment

This was a retrospective study of non-critically ill patients treated with intravenous teicoplanin who were admitted to the Nanjing Drum Tower Hospital (China), a university-affiliated tertiary care medical center, from January to December 2019. A preauthorization strategy of antimicrobial stewardship led by pharmacists and infectious disease physicians was established at Nanjing Drum Tower Hospital since 2012. All doctors were routinely recommended to perform teicoplanin TDM on the fourth day of the therapy after teicoplanin was prescribed, but the adherence to the recommendation differs greatly between different wards. Clinical pharmacists prompted TDM and rational use of teicoplanin in the wards they were working in. The clinical pharmacist used a personalized dosing strategy and adjustments during pharmaceutical care. In 2019, a total of 62 wards were registered at Nanjing Drum Tower Hospital. Among them, 40 wards were staffed by a total of only 33 clinical pharmacists, while the other 22 wards had no clinical pharmacists on staff. Non-critically ill adult patients (over 18 years of age) admitted to the hospital with infections treated with intravenous teicoplanin were included. Patients with isolation of MRSA from specimens or the detection of Gram-positive cocci and a suspicion of MRSA infection were included. The exclusion criteria were as follows: patients who received teicoplanin for less than 72 h, cases of systemic antimicrobial therapy with the activity for MRSA before teicoplanin therapy, patients admitted to the intensive care unit (ICU) before teicoplanin treatment, transfer to ICU within 72 h after teicoplanin therapy started, the presence of sepsis or septic shock before teicoplanin therapy, patients recruited by another study, patients that did not receive TDM, and patients with incomplete medical records. Given the nature of this retrospective observational study, no intervention was made to standardize care. The clinical protocols and the treatment of the patients were determined by the clinical team that cared for the patient. Teicoplanin regimens used were recorded. The loading dose was administered on day 1, twice daily according to the local consensus and the drug package inserts. If a TDM was performed, the trough concentration was routinely determined before administration on the fourth day according to clinical practice. Clinical outcomes were compared between the pharmacist-driven teicoplanin dosing and monitoring group (PD group) and the non-pharmacist-driven teicoplanin dosing and monitoring group (NPD group). The cost of teicoplanin, cost of medications and the total cost were recorded. The institutional ethics committee review board approved this study (Nanjing Drum Tower Hospital Project number: 2020-085-01). The study was performed in accordance with the ethical standards of the Declaration of Helsinki. Informed consent was not feasible due to the retrospective nature of the study. This study was registered in the Chinese Clinical Trial Registry (ChiCTR2000033521).

Outcome Measures

The primary outcomes were defined as the attainment of the target trough concentrations (Cmin 15–30 mg/L) at the initial TDM, a composite endpoint of all-cause mortality, and admission to the ICU or development of sepsis or septic shock during the hospital stay or within 30 days after hospital admission. Readmission within 10 days after initial hospital discharge was considered part of the index admission.

Secondary outcomes included the proportions of admissions to the ICU, sepsis, septic shock, and mortality. The clinical response at the end of treatment was considered a secondary outcome, for attenuation of the signs and symptoms was hypothesized to be part of the beneficial effect. By definition, a successful clinical response had to meet all of the following criteria: clinical response of “cured” [resolution of clinically significant signs and symptoms, defined as any criteria of systemic inflammatory response syndrome (SIRS)] or “improved” (partial resolution of clinically significant signs and symptoms); the patient did not receive a concomitant antibacterial agent that could potentially have been effective against MRSA during treatment. Also, the length of hospital stay was considered a secondary outcome.

Safety outcome measures were defined as the proportion of nephrotoxicity, thrombocytopenia, hepatotoxicity, and any other adverse events recorded during teicoplanin therapy or within 72 h after the teicoplanin therapy ended. Nephrotoxicity was defined as acute kidney injury determined by serum creatinine (SCr) increase of more than 0.3 mg/L within 48 h, or 50% increase from the baseline within 7 days, or a urine volume less than 0.5 mL/kg/h for 6 h [20]. Thrombocytopenia was defined as a platelet count of < 100 × 103/µL combined with a reduction of more than 25% from the baseline count [21]. Hepatotoxicity was defined as any one of the following: serum alanine aminotransferase (ALT) or aspartate aminotransferase (AST) greater than five times the upper limit of normal (UPN), or alkaline phosphatase (ALP) greater than two times of the UPN on two separate occasions at least 24 h apart, as well as total serum bilirubin > 2.5 mg/dL, along with elevated serum AST, ALT, or ALP level, or international normalized ratio > 1.5 with elevated serum AST, ALT or ALP [22]. Other adverse events were identified and recorded by the clinical team looking after the patient.

All patients’ costs during hospital stays were calculated. The total cost of teicoplanin therapy, cost of TDM, cost of medications, and aggregated charges were recorded. The cost of TDM of teicoplanin was $14 at our center. All costs were obtained in the Chinese official currency (Chinese Yuan) and then converted to US dollars at an exchange rate of 0.141 (1.00 US dollar equivalent to 7.11 Chinese Yuan; October 31, 2022) when the data were analyzed.

Statistical Analysis

Continuous variables were tested for normal distribution using the Kolmogorov–Smirnov test and expressed as the mean ± standard deviation (SD) for the normally distributed data and as the median and quartiles (25–75) for skewed data distributions. The two-tailed Student’s t-test and the Mann–Whitney test were used to analyze continuous data when appropriate. Categorical variables were presented as the number of cases. The Pearson chi-squared (χ2) test and Fisher’s exact test were used to analyze the categorical variables. The effects of PD dosing service and pharmaceutical care on the primary outcomes were evaluated by logistic regression. The results were reported as the odds ratio (OR) with a 95% confidence interval (CI). Subgroup analyses of the primary outcomes were performed to assess the interactions between different characteristics. The following predefined baseline characteristics were analyzed: sex (male versus female), age-adjusted Charlson Comorbidity Index (≥ 3 versus < 3), chronic kidney disease (CKD) (with CKD versus without CKD), hypoalbuminemia (serum album < 25 g/L versus  ≥ 25 g/L), and primary infection (pneumonia versus others). p < 0.05 was required for statistical significance. The IBM SPSS Statistics software (Statistics for Windows, version 25, IBM Corporation, Armonk, NY, USA) and GraphPad Prism 9.2 (GraphPad Software, La Jolla, CA, USA) were used for statistical analysis and plotting graphs.

Results

Patient Characteristics

A total of 1095 patients received intravenous teicoplanin therapy. Of these, 473 were critically-ill patients admitted to the ICU, and the other 622 were not critically ill, while 451 of these cases had either MRSA cultures isolated from patient specimens, or gram-positive cocci were identified in the patients’ specimen and the patients were clinically suspected to have MRSA. Seventy cases received systemic antimicrobial therapy with MRSA activity prior to teicoplanin therapy. Twenty-seven patients were transferred to the ICU within 72 h after starting teicoplanin therapy. Twenty-five patients were excluded because they received less than 72 h of intravenous teicoplanin therapy and 49 were excluded for other reasons. Among them, 163 patients received at least one TDM and were finally included in the final analysis. Clinical pharmacists handled seventy cases and provided pharmaceutical care, while the remaining 93 cases did not receive such intervention. No differences were found in loading dose regimens or maintenance dose regimens. The patient characteristics are presented in Table 1.

Table 1 Patient characteristics
Full size table

Clinical Outcomes

As shown in Table 2, the primary composite endpoint of all-cause mortality, admission to the ICU, or development of sepsis or septic shock occurred in 18 (26%) patients in the PD group, as compared with 46 (50%) patients in the NPD group (p = 0.002). The results of logistic regression multivariate analysis showed that PD dosing and pharmaceutical care (OR 0.12, 95% CI 0.09–0.32, p < 0.001) was associated with reduced risk of reaching primary composite endpoint, while baseline procalcitonin (PCT) > 0.5 ng/mL (OR 17.00, 95% CI 6.30–45.83, p < 0.001) and pneumonia as primary infection site (OR 4.13, 95% CI 1.82–9.41, p < 0.001) were associated with increased risk of reaching a primary composite endpoint, detailed in Table S1. The results of the primary composite endpoint subgroup analyses are reported in Fig. S1. The effects of the PD teicoplanin therapy provided a clinical benefit for the primary composite endpoint in most subgroups except patients with primary infection site outside pneumonia, but the interaction effect did not reach statistical significance (p for interaction = 0.299). The prevalence of attaining target Cmin differed significantly between groups (p < 0.001). Thirty-eight patients (54%) in the PD group and 15 patients (16%) in the NPD group attained the target Cmin. Pharmacist-driven dosing services and pharmaceutical care were associated with an increased probability of target concentration attainment (OR 5.20, 95% CI 2.44–11.05, p < 0.001); detailed results of the logistic analysis are shown in Table S2. The effect was consistent across subgroups (Fig. S2). Subgroup analyses showed that pharmacist-driven dosing and pharmaceutical care significantly increases teicoplanin target concentration attainment across subgroups.

Table 2 Comparison of clinical outcomes of non-critically ill patients treated by teicoplanin who received PD dosing and pharmaceutical care and who did not
Full size table

Concerning secondary outcomes, the PD group experienced a 6% hospital mortality rate, in contrast to the NPD group’s 14% (4 versus 13, p = 0.087), with the majority of fatalities taking place in the ICU. No statistical differences were found in the proportion of admission to the ICU between groups (14% versus 26%, p = 0.073). Sepsis occurred in 17 (24%) of the patients in the PD group and 37 (40%) of those in the NPD group (p = 0.037). Four (6%) patients developed septic shock in the PD group and 15 (16%) in the NPD group (p = 0.040). A clinical response was observed at the end of treatment in 60 patients (86%) of the PD group and 75 patients (81%) of the NPD group (p = 0.396). The median hospital length of stay among the patients was 22 days in the PD group and 35 days in the NPD group (p < 0.001). The incidence rate of nephrotoxicity was 14% in the PD group and 16% in the NPD group; no statistical differences were found between the two groups (p = 0.746). Two patients (3%) in the PD group and seven patients (8%) in the NPD group experienced thrombocytopenia (p = 0.196). Four events (6%) of hepatotoxicity were observed in the PD group, while 11 (12%) were observed in the NPD group (p = 0.181). Other adverse events also showed no differences between groups (16% versus 24%, p = 0.212). Detailed events are shown in Table S3.

Economic Outcomes

As shown in Table 3, the total cost during hospital stay was lower in the PD group (p < 0.001). The median total cost of patients in the PD group was $17,763 with an interquartile range from $10,925 to $20,805, while the median aggregated cost of patients in the NPD group was $20,129 with an interquartile range from $15,241 to $25,890. A cost increase in median total charge was observed at a level of $2366 (13%). Cost of medications also differed greatly between groups, significantly lower drug cost was observed in the PD group ($5456 [3664–7279] vs $6702 [5441–7814], p < 0.001). When comparing the medians of drug cost between groups, an increase of $1246 (22%) was found. No differences were found when comparing teicoplanin cost [$605 (534–679) versus 644 (557–773)], p = 0.078).

Table 3 Cost analysis of non-critically ill patients treated by teicoplanin who received PD dosing services and pharmaceutical care and who did not
Full size table

Target Trough Concentration Attainment

TDM was performed 244 times for 163 patients. Unfortunately, only 33% of patients attained the target Cmin according to the initial TDM during teicoplanin therapy. The prevalence attaining target concentration in follow-up TDM was higher (42%) after adjustment of teicoplanin dosage regimens.

As demonstrated in Fig. 1, the PD group showed a significantly higher trough concentration (median initial Cmin: 16.3 mg/L versus 11.1 mg/L, p < 0.001) and a higher proportion of attaining the target Cmin than the NPD group. In the PD group, 9 (13%) adjustments of dosing strategies were made during teicoplanin therapy, while only 4 (4%) adjustments were made in the NPD group (p = 0.046), respectively. Finally, after dosing adjustments during teicoplanin therapy, 40 patients (59%) from the PD group finally reached the target Cmin, while only 28 patients (30%) of the NPD group reached the target Cmin (p < 0.001).

Fig. 1
figure 1

Initial trough concentrations between the PD group and NPD group. Each black dot on the left represents one case in the PD group. Each gray square on the right represents one case in the NPD group. The median initial concentration was 16.3 mg/L with an interquartile range from 10.3 mg/L to 21.5 mg/L in the PD group (black solid line with error bars), and 11.1 mg/L with an interquartile range from 7.2 mg/L to 15.6 mg/L in the NPD group (gray solid line with error bars), p < 0.001

Full size image

Figure 2 shows the distribution and proportion of different initial Cmin levels in the PD group and the NPD group. The incidence of subtherapeutic initial trough concentrations was significantly lower in the PD group than in the NPD group [27 (39%) vs 70 (75%), respectively, p < 0.001]. The proportion of patients with high initial Cmin (> 30 mg/L) was 7% (5 patients) in the PD group and 9% (8 patients) in the NPD group, respectively; no statistical differences were found (p = 0.734). Unfortunately, subtherapeutic trough concentrations still occurred in 26 patients (37%) in the PD group and in 59 patients (63%) in the NPD group after dosage adjustment during teicoplanin therapy (p < 0.001). The incidence of unexpectedly high trough concentration did not show any statistical differences between groups after adjustment of the dosage regimen [4 (6%) versus 6 (7%), respectively, p = 0.846].

Fig. 2
figure 2

The initial trough concentrations distribution of patients from PD group (blue) and NPD group (red) are shown above. Most patient from NPD group failed to attain the target Cmin. Significant differences (p < 0.001) were observed between groups in patients with subtherapeutic concentration (< 15 mg/L) and patients attaining the target Cmin (15–30 mg/L). No significant difference was found in patients with unexpectedly high trough concentration (> 30 mg/L)

Full size image

Discussion

Despite the increased use of teicoplanin, achieving initial therapeutic serum concentrations still poses a challenge to clinicians due to the significant variability in pharmacokinetic parameters and clinical heterogeneity observed across different patient populations [15,16,17]. In recent decades, pharmacists have played an increasingly important role in the procurement, distribution, review and safety handling of drugs [23]. Pharmacist-driven dosing and monitoring services have been found to be beneficial in vancomycin and aminoglycoside therapy, especially when applied to attain an effective target concentration or area under the curve [24]. According to our study, a pharmacist-driven teicoplanin therapy and pharmaceutical care can increase the likelihood of patients achieving the target initial therapeutic serum concentration, which improved the clinical outcomes of non-critically ill patients.

Teicoplanin has a high protein binding rate, a low volume of distribution, and a very long elimination half-life of 83–163 h [25, 26]. As a result, variations in serum albumin levels and renal function can cause significant changes in serum concentrations of teicoplanin. These characteristics highlight the importance of TDM in teicoplanin treatment. The pharmacist-implemented teicoplanin dosing strategy and monitoring can take advantage of pharmacokinetic data and tailor the dosage to each individual patient, which enhanced the percentage of patients achieving the target Cmin, particularly among those with low serum albumin and compromised renal function. According to the subgroup analyses of our study, patients with low serum albumin levels significantly benefit from pharmacist-driven teicoplanin dosing services and pharmaceutical care. Additionally, the PD group encompassed a greater number of patients with compromised renal function, and their clinical outcomes also saw improvement as a result of the pharmacist-guided dosing and monitoring services. However, we did not witness a decrease in the incidence of nephrotoxicity, hepatotoxicity, or other adverse events. One potential explanation is that teicoplanin’s recommended target trough concentration for clinical efficacy is 15–30 mg/L, which is significantly lower than the concentrations causing adverse effects (thrombocytopenia  ≥ 40 mg/L, nephrotoxicity  ≥ 60 mg/L) [21, 27,28,29]. Owing to the large therapeutic window, the incidence of unexpectedly high trough concentrations reaching the toxicity threshold was relatively low in both groups. Therefore, teicoplanin’s TDM was primarily administered to ensure the attainment of the trough concentration for clinical effectiveness, rather than to prevent adverse effects.

In our study, we found that more than half of the patients underwent teicoplanin treatment without the assistance of clinical pharmacists, owing to a shortage of qualified clinical pharmacists. More clinical pharmacists are needed to promote the rational use of medications. In addition to the dosing service of teicoplanin, the pharmaceutical care provided by clinical pharmacists can bring other benefits that were not directly measured in our research. One such example is comorbidity. Among the patients in our study, we found a median age-adjusted Charlson Comorbidity Index of 3, which indicates a considerable level of comorbidities. Comorbidities are common among hospitalized patients, which means that polypharmacy and potential drug interactions frequently occur among these patients. Pharmacist-led interventions have been documented to decrease the likelihood of drug–drug interactions. For interactions with high severity, the reduction rate reached up to 81% [30]. Pharmacists can also improve the clinical outcomes by improving the adherence to medications. An analysis of 54,322 hospitalizations for heart failure revealed that nonadherence to medications contributed to hospital admission in 7.9% subjects [31]. Pharmacists can help the medical team and the patients identify barriers to medication adherence. To achieve this goal, pharmacists may use strategies such as patient education, drug regimen simplification, and identifying lower-cost alternatives where appropriate [32]. Additionally, the primary aspects of pharmaceutical care included clarifying drug orders and identifying and addressing drug-related problems, resulting in a reduction of medication errors and adverse events. These services have been reported to decrease annual institutional costs by $25,140 to $270,000 [33]. Our study revealed that pharmacist-driven teicoplanin dosing and pharmaceutical care also improved patients’ economic outcomes. We observed cost savings of 13% in total expenses and 22% in overall drug costs, with no difference in teicoplanin costs between groups. This suggests that promoting rational teicoplanin use and implementing appropriate pharmaceutical care by clinical pharmacists were crucial factors in avoiding unnecessary costs to the patients.

In addition to infectious diseases, clinical pharmacists who specialize in other specialties have also demonstrated the benefits of their care in patients’ clinical outcomes in different populations other than patients with infectious disease. Koshman reviewed the published randomized trials and found that the involvement of pharmaceutical care in the treatment of patients with heart failure greatly reduces the risk of all-cause hospitalizations [34, 35]. Another study of 335 patients in 32 medical offices demonstrated that a pharmacist–physician collaborative intervention improved blood pressure control in patients with diabetes and/or chronic kidney disease [36]. In a postoperative cohort of patients with colorectal cancer, a pharmacist-led parenteral nutrition standardization program reduced the incidence of postoperative infection and many other complications [37]. In the future, we hope that advancing pharmaceutical care will enable clinical pharmacists to overcome existing barriers and contribute to clinical decision-making. This involvement would enhance the effectiveness of drug therapy, decrease drug-related expenses, and ultimately improve the quality of patient care [38].

Despite the fact that this study encompassed a considerable number of non-critically ill patients and yielded statistically significant, conclusive results, it does have multiple limitations. First, this was a single-center retrospective study, which means that the effects of pharmaceutical care might differ in other healthcare facilities. Second, clinical pharmacists working at our center gave priority to patients with specific pathophysiology (i.e., chronic kidney disease), which may have resulted in distinct baseline characteristics between the two groups. However, to minimize bias as much as possible, we have taken some measures such as performing subgroup analyses. Finally, no interventions occurred during the study. Future prospective randomized controlled studies are needed to further investigate these findings.

Conclusions

Pharmacist-driven teicoplanin dosing services and pharmaceutical care increased the likelihood of reaching the target concentration and for providing follow-up dose adjustment during therapy to improve the clinical and economic outcomes for non-critically ill patients. In summary, our study demonstrated a reduced percentage of patients with sepsis or septic shock, a higher likelihood of achieving to reach target concentration, shorter length of hospital stay, and lower medication costs as a result of pharmacist-driven teicoplanin dosing services and pharmaceutical care.