Piperacillin-tazobactam (Zosyn) extended-infusion dosing

Let's start with a patient case.  A 70 year old male patient is admitted to the hospital for a diabetic foot infection.  He has a past medical history of HTN, hyperlipidemia, CKD, gout, and obesity.  He weighs 100 kg and has a Scr of 2.8 mg/dL.  You recall that piperacillin-tazobactam is indicated for diabetic foot infections and are deciding what dosing regimen should be started.  Upon consulting Lexicomp or Micromedex, you will find that piperacillin-tazobactam has a wide dosing range depending on the type of infection being treated in addition to the extent of renal impairment.  Recommended doses range from 2.25 g to 4.5 g intravenously every 6 to 12 hours depending on these factors, leaving much room for uncertainty in many cases.

The following will briefly discuss piperacillin-tazobactam, its pharmacokinetics and pharmacodynamics, and how computer modeling has generated new dosing strategies.  If you just want a simplified way of dosing this medication, feel free to skip down to the 'Take home pearls' at the bottom of the article.

What is it?

First off, piperacillin-tazobactam is a product containing two medications that are always used in combination.  Piperacillin is the beta-lactam antibiotic and tazobactam is the beta-lactamase inhibitor, enabling the combination to be used for piperacillin-resistant beta-lactamase producing strains of various organisms.  This includes Gram-positives (not MRSA), Gram-negatives (including pseudomonas), and Gram-negative anaerobes.  They are prescribed in a fixed ratio in a way that you only have to order one dose.  A 4.5 g dose contains 4 g of piperacillin and 500 mg of tazobactam, a 3.375 g dose contains 3 g of piperacillin and 375 mg of tazobactam, and a 2.25 g dose contains 2 g of piperacillin and 250 mg of tazobactam.  There are 2.79 mEq (64 mg) of Na+ per gram which may become a factor in patients who require a salt restricted diet.

How does it work? (the pharmacodynamics)

Piperacillin-tazobactam, like all beta-lactam antibiotics, exhibits time-dependent killing of bacteria.  Studies have shown that increasing the concentration does not improve the rate of bacterial killing, as it does with fluoroquinolones and aminoglycosides.  The best predictor of bacterial killing in the case of beta-lactam antibiotics is the proportion of time of the dosing interval that the concentration of free drug exceeds the minimum inhibitory concentration (MIC), designated as %fT>MIC.  For piperacillin-tazobactam, this means that maximum bactericidal activity is achieved when the concentration exceeds the MIC for more than about 50% of the dosing interval.  (So theoretically, the best dosing regimen could be a continuous infusion (ie. 24 hours a day) - this however, is not a practical approach).

How does it work? (the pharmacokinetics)

Piperacillin-tazobactam is traditionally given over a half-hour intravenous infusion.  Both the piperacillin and tazobactam are eliminated by glomerular filtration, tubular secretion, and also into the bile with a half-life of 0.7 to 1.2 hours in healthy subjects.  Reflecting back to the example of a 6 hour dosing interval, this could mean that eight half-lives elapse (ie. 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2 = 0.39% of the dose is remaining).  This dosing strategy does not make optimal use of the pharmacokinetics and pharmacodynamics because piperacillin-tazobactam and most of our commonly use beta-lactam antibiotics were developed prior to our current understanding of these principles.  Therefore, many of the indicated dosing regimens have a low probability of attaining our goal (50% fT>MIC).

Why extended infusion?

Computer modeling has been used to develop the pharmacodynamic profile of a medication.  Taking into account population variance of pharmacokinetics, microbiologic surveillance, and pharmacodynamic targets, simulations can estimate the probability of attaining a goal (in the case of piperacillin-tazobactam, 50%fT>MIC) when administering a specific dosing regimen.  Extended infusion models (eg. infusion run over 4 hours) take advantage of sustaining a prolonged %fT>MIC compared to the usual half-hour infusion which has a higher peak but whose concentration drops precipitously after the infusion ends.  Also, less drug needs to be administered over the course of the day to achieve a similar %fT>MIC, meaning a cost savings from a purchasing standpoint, less sodium administered, and potentially fewer adverse effects (though not seen in studies).

Take a look at the probability of target attainment curve with different regimens given different MICs here:

We can see that the 4 hour infusion given every 8 hours has a higher probability of achieving the target 50%fT>MIC when compared to the traditional dosing, even when it is given every 4 hours (that's twice as much drug being administered).

Some clinical evidence

Several studies have been performed evaluating extended infusion versus traditional dosing on clinical endpoints.  All studies have consistently demonstrated higher pharmacodynamic target achievement compared to traditional dosing.  As far as clinical endpoint improvement, one study of patients with documented pseudomonal infections demonstrated reduced mortality (12.2 vs. 31.6%, p=0.04) and shorter length of stay (21 vs. 38 days, p=0.02) in those patients who were critically ill (APACHE II score > or = 17) with an extended infusion regimen of 3.375 g every 8 hours over 4 hours compared to patients who previously received a traditional half-hour infusion of 3.375 g every 4 or 6 hours.

What about renal impairment?

A subsequent simulation study was performed to assess the likelihood of pharmacodynamic target attainment (50%fT>MIC) using different regimens with varying degrees of renal impairment.  Several doses, frequencies, and infusions were studied with different creatinine clearances.  At a creatinine clearance of 100 mL/min, the 3.375 g every 8 hours over a 4 hour infusion was found to have a comparable attainment of target as 3.375 g every 12 hours at 20 mL/min.  Presumably, even lower clearances could only have improved target attainment as would increasing the infusion time.

Back to the patient case

After calculating the creatinine clearance to be ~35 mL/min, you see that based on the prescribing information, the original dose of 3.375 g every 6 hours should be decreased to 2.25 g every 6 hours for a creatinine clearnace between 20-40 mL/min.  Other questions come into play when determining the creatinine clearance though, such as, "Should I use the actual, ideal, or adjusted body weight?" and "Is this Scr stable, acutely elevated, or possibly downtrending?".  Depending on the answer to each of these questions, the calculated creatinine clearance could easily move outside of the 20-40 mL/min range that we've just adjusted for.  Altering the regimen to take advantage of the pharmacodynamics and pharmacokinetics of piperacillin-tazobactam can simplify our dosing and eliminate much of the dosing uncertainty and the need to change the dose on a daily basis as the Scr changes.  At 35 mL/min (or any creatinine clearance >20 mL/min) the 3.375 g every 8 hours over a 4 hour infusion can be used.

Take home pearls:

• Extended infusion regimen takes advantage of what we now know about pharmacokinetics and pharmacodynamics.
• The half-life of piperacillin-tazobactam is about 1 hour in healthy subjects.
• 3.375 g given over 4 hours has a high probability to achieve the desired time over the MIC when:
• Dosed every 8 hours for creatinine clearance 20 mL/min or greater 
• Dosed every 12 hours for lower clearances (including hemodialysis)
• At North Shore University Hospital, these two extended infusion regimens are the only two regimens that can be ordered.
• The first dose is an exception as it can be given over a half-hour.

References

1.  Zosyn (R).  [package insert].  Philadelphia, PA: Wyeth Pharmaceuticals Inc; 2012.

2.  Lodise TP, Lomaestro BM, Drusano GL, et al.  Application of antimicrobial pharmacodynamic concepts into clinical practice: focus on beta-lactam antibiotics: insights from the Society of Infectious Disease Pharmacists.  Pharmacotherapy 2006;26(9):1320-32.

3.  Lodise TP, Lomaestro BM, Drusano GL.  Piperacillin-tazobactam for pseudomonas aeruginosa infection: clinical implication of an extended-infusion dosing strategy.  Clin Infect Dis 2007;44(3):357-63.

4.  Patel N, Scheetz MH, Drusano GL, et al.  Identification of optimal renal dosage adjustments for traditional and extended-infusion piperazillin-tazobactam dosing regimens in hospitalized patients.  Antimicrob Agents Chemother  2010;54(1):460-5.

photo by Unhindered by Talent