Intraperitoneally Administered Vancomycin Results in Suboptimal Serum and Peritoneal Effluent Drug Levels in Patients with PD-related Peritonitis

Introduction

Peritonitis is a critical complication of peritoneal dialysis (PD) and is associated with significant morbidity and mortality.^1,2 Vancomycin is commonly employed in the treatment of PD-related peritonitis. The International Society of Peritoneal Dialysis (ISPD) recommends a dose of 15–30 mg/kg every 5–7 days administered intraperitoneally (IP) Peritoneal membrane characteristics, dwell time, and peritoneal inflammation due to peritonitis affect the bioavailability and equilibration half-life of IP vancomycin.³ Serum vancomycin levels of 15 mg/L or AUC/MIC ratio of 400 is recommended for efficacy and to prevent the development of resistance.⁴ Therapeutic drug level monitoring (TDM) is not routinely available, and the turnaround time does not permit real-time decision-making. Recommended IP doses of most antibiotics are based on published clinical experience and not pharmacokinetic studies. These recommendations do not consider residual renal function. The importance of residual renal function in antibiotic dosing was highlighted in a prospective study of 339 episodes of peritonitis. Patients with a urinary creatinine clearance of more than 5 mL/min had higher rates of treatment failure.⁵ We studied whether sufficient serum levels were achieved in patients with PD-related peritonitis treated with the recommended dose of vancomycin.

Materials and Methods

Consecutive patients presenting to the Department of Nephrology during 19 months with PD-related peritonitis were included. Those with malignancy or those who had received antibiotics within 48 h before presentation and aged less than 18 years were excluded. Ethical approval and patient consent was obtained.

Once a diagnosis of peritonitis was made, patients were asked to perform PD exchange and the drain bag was disconnected. Under sterile precautions, three PD effluent samples of 50 mL each were collected. Two of these samples were centrifuged at 3000 rpm for 15 min, one was sent for Gram stain and inoculated in conventional aerobic solid culture media. Sensitivity was analyzed using the Kirby-Bauer disc diffusion method. The second sample was reconstituted, suspended in 10 mL sterile saline, and inoculated in BACTEC BD aerobic blood culture media. The third sample was stored for further microbiologic studies.

Patients were started on an intermittent regimen of empirical antibiotics administered IP. Vancomycin 15 mg/kg Q96H (Vancocin Chandra Bhagat Pharma Pvt. Ltd.) and amikacin 2 mg/kg q24h with a dwell time of 8 h were used. Patients with urine volume >100 mL/day received a 25% higher initial dose of antibiotics.⁶ Five milliliters of blood samples were collected at 1, 12, 24, and 96 h after the initial administration of antibiotics. Five milliliters of PD effluent fluid was collected 24 and 96 h after administration of vancomycin. Blood samples were settled to clot, centrifuged at 3000 rpm for 10 min, and the serum was separated and stored at −80°C without delay.

Results

During the study period, there were 45 episodes of PD-related peritonitis, and 41 episodes in 27 patients treated with IP antibiotics were included in the analysis. Figure 1 depicts patient screening, enrollment, and outcomes. The baseline demographic data, etiology of kidney disease, comorbidities, and CAPD treatment details are summarized in Table 1.

Close

Figure 1:

Patient screening, enrolment and outcomes. PD: peritoneal dialysis, HD: Hemodialysis, IP: Intraperitoneal

Export to PPT

Discussion

In patients with PD-related peritonitis treated with IP vancomycin dose of 15 mg/kg Q96H, we found that the serum and PD fluid levels at 1, 12, 24, and 96 h were deficient compared to the recommended trough levels.

Vancomycin undergoes negligible metabolism, and its elimination is mainly through glomerular filtration. This leads to a significant reduction in vancomycin clearance in patients with CKD, resulting in an elimination half-life of 7.5 days in contrast to 4–6 h in patients with normal GFR. The amount of vancomycin reaching systemic circulation from the peritoneal cavity relative to intravenous dosing is 50% in normal subjects and 70–90% in patients with peritonitis.⁷ The serum level of vancomycin during treating an episode of PD peritonitis depends on peritoneal solute transfer rate, PD prescription, RRF, and other patient characteristics such as weight.³

Our patients generally have a low lean body mass (mean body weight was 55 kg), and the standard prescription was three PD exchanges per day. This is different from the practice in high-income countries where most patients perform four PD exchanges daily and have relatively higher body weight compared to our patient population. The dose of vancomycin used in patients with End Stage Renal Disease (ESRD) and either on HD or PD at our center was 15mg/kg Q96H in the absence of access to TDM. We wanted to test whether this dose achieved therapeutic MIC in our patients with PD-related peritonitis.

The therapeutic peak level so far from published literature varies from 15 to 30 µg/mL against a dose of 15–30 mg/kg measured at 4 h, 6 h, 48 h.^8,9 In a recent retrospective cohort study from a single center, a loading dose of 2 g of IP vancomycin led to sub-therapeutic drug levels in one-third of the patients. GFR and weight were independently associated with a likelihood of lower therapeutic levels, and the possibility of achieving a therapeutic level was approximately four times higher with an initial dose of 25 mg/kg compared to <25 mg/kg of dosing. The trough levels in PD fluid range from >2 µg/mL after the first exchanges to 7.7–8.7 µg/mL on the fifth day or seventh day with similar dosing regimens.^9,10

In a study by Mulhern et al., a trough level of >12 µg/mL at the end of the first week predicted less number of relapses after PD-related peritonitis (0/13 episodes) compared to levels <12 µg/mL (9/13 episodes) among 31 peritonitis patients on IV vancomycin 15 mg/kg every seventh day.¹¹

It is a matter of concern that most patients did not attain the MIC of >15 mcg/mL in our study. This was very unlikely due to the poor quality of drugs as we tested the study drug from the hospital pharmacy against five other popular brands procured from the open market. We did not analyze transport status in our patients because the transport parameters during an episode of peritonitis are expected to be different from when there was no active peritoneal infection. Our patients had a mean body weight of 55 kg, we accounted for RRF in determining the drug dose, and our patients were performing only three PD exchanges per day, potentially reducing peritoneal drug clearance. The blood and peritoneal fluid samples collected were stored at −80°C, and analysis was performed every month. Vancomycin is known to remain stable without degradation for up to 85 days when stored at minus 85°C.¹² Despite all these, the observed serum and peritoneal drug levels were deficient. All these suggest that the dose of vancomycin used was insufficient, and a higher dose or more frequent dosing interval is likely required to achieve serum levels of >15 mcg/mL in patients with PD-related peritonitis.

Despite the deficient serum and peritoneal effluent levels of vancomycin, we did not find any correlation between the drug levels and clinical outcomes and the incidence of complications. This is likely because gram-positive organisms accounted for only 11 episodes (24.4%) of peritonitis, all of which, except for one, were from MSSA. The only episode of MRSA peritonitis was refractory and resulted in catheter removal. The MIC for MSSA is less than 15 mcg/mL. The rest of the infections were from gram-negative organisms, fungi, or from mycobacteria. In this regard, it is emphasized that the study was designed primarily to assess the drug levels from the commonly used dose of vancomycin and not to analyze the association of drug levels with clinical outcomes.

Our study had some limitations. Serum samples were collected only at 1, 12, 24, and 96 h, precluding the calculation of AUC for vancomycin. The reason why vancomycin peak level monitoring was set at 1 h and 12 h, respectively, after IP administration, was based on earlier pharmacokinetics studies to document the peak levels. Samples were collected at 24 and 96 h to assess the trough levels. More frequent samples would have required hospitalization of the patients which was logistically challenging. A limited sampling strategy was designed to allow studying the peak and trough levels and also to avoid multiple venipunctures and inconvenience to the patient. The first sample was collected at 1 h after IP administration of vancomycin in our study. This might have been too early to study the peak level. When administered IP, the equilibration half-life of vancomycin in PD patients with peritonitis was 1.6 to 2.9 h, and it was 2.9 h in those without peritonitis.³ We used a vancomycin dose of 15 mg/kg (instead of a higher dose); however, this was in line with the standard practice at our center and most other centers in our region where TDM is not freely available or accessible. Our primary intention was to check whether the commonly employed dose at our center was appropriate. Characteristics and PD prescriptions in our patients were different from those in most high-income countries, and findings may not be generalizable to all PD patients.

Based on our observations, it may be safe to conclude that in the absence of TDM, using a fixed dose of IP vancomycin 15 mg/kg Q96H is likely to be ineffective in treating MRSA. More studies on AUC of vancomcyin with various doses need to be undertaken to improve the vancomycin dose recommendation in treating PD-related peritonitis. TDM should be more widely used in monitoring vancomycin treatment in this situation until more data are available on the dosage of vancomycin required to attain an AUC of 400–600 mg·h/L without TDM.

Conclusion

In patients with PD-related peritonitis, intermittent intraperitoneal administration of 15 mg/kg of vancomycin every 96 h does not attain the recommended serum and peritoneal effluent levels of vancomycin in the treatment of MRSA. Therapeutic drug monitoring may be necessary to ensure sufficient serum levels of the drug. There is an urgent need for pharmacokinetic studies on commonly used IP antibiotics in the PD population to facilitate correct dose recommendations based on scientific information.