Let's Talk Vancomycin: Part 2

Welcome back everyone! I’m so glad you could join us for Part 2 of Let’s Talk Vancomycin (Vanc for short). Last time in Part 1, we reviewed a lot of info including the basic structure, mechanism, indications, administration, pharmacokinetics, safety, and resistance of vanc. Today, we’ll continue our deep dive with vanc and learn more about how we might approach dosing and calculations!

What this post IS:

• A simple review on how to approach calculations and dosing for vanc therapeutic regimens.
  • Manual calculations (Trapezoidal Method)
  • Bayesian dosing software and online calculators
  • Practical dosing (in an inpatient setting)
  • Loading doses (To load or not to load, that is the question.)

What this post IS NOT:

• An official treatment guideline for vanc dosing.

Always refer to and review your specific institution’s guidance on vanc dosing (almost all institutions have their own protocols). Now without further ado, let’s buckle up and get ready for a wild ride!

In 2009, the Infectious Diseases Society of America (IDSA), American Society of Health-System Pharmacists (ASHP), and Society of Infectious Diseases Pharmacists (SIDP) came together to create therapeutic guidelines on vanc¹. The primary recommendation was to dose vanc to a goal trough serum level of 15-20 mg/L for Staphylococcus aureus infections. This was thought to improve the chances of reaching a goal area under the curve over 24 hours to minimum inhibitory concentration ratio (AUC/MIC) of 400 mg*hour/L, thereby increasing efficacy. Trough was essentially used as a surrogate for the AUC/MIC (our true goal) since it was way easier to manage, dose adjust, and monitor.

However, since the 2009 recommendations came out numerous studies and reports have concluded that trough level monitoring actually results in increased nephrotoxicity (acute kidney injury; AKI) in both adults and pediatrics².  With damaged kidneys here, there, and everywhere what were we going to do for patients who needed vanc so desperately?!

Fast forward to 2020, the IDSA, ASHP, SIDP, and the Pediatric Infectious Diseases Society (PIDS) realized the shortcomings of trough monitoring and came back together to revise our vanc guidance (we’re saved!)³. This newly updated guideline changed the recommendations from a trough goal, to dosing vanc using a goal AUC of 400-600 mg*hr/L for serious methicillin-resistant Staphylococcus aureus (MRSA) infections. Note that this is assuming an MIC of 1 mg/L. Ultimately, this new AUC recommendation helped to improve and reduce rates of AKI and nephrotoxicity^4,5.

Generally speaking, 400-600 mg*hr/L is what you want to aim for with any vanc regimen especially those with a confirmed MRSA infection. Different institutions, however, may have different ranges for various types of infection or organisms (e.g., a gram-positive bacteremia caused by Streptococcus spp. and when beta-lactams are contraindicated).

I’ve thrown around this “area under the curve” (AUC) term and many of you have likely heard of it already. It’s exactly as the name suggests… it is the area underneath the pharmacokinetic concentration vs. time curve of a drug. Every drug has a concentration/time curve which outlines the total drug exposure in a person’s body within a given period of time after a dose. It tells the story of how a drug is absorbed into the body, and how it is eliminated across time.

The AUC is the entire grey shaded area. When we talk about efficacy though, we’re only referring to the AUC/MIC (specifically the blue-line shaded area). Anything below the MIC will not be effective for the organism we’re treating (it is below the minimum inhibitory concentration).

Timing is also important here. Per guidelines, we want to achieve time above MIC daily for effective treatment. Sometimes people may tack on an extra “24” to the AUC to get “AUC₂₄”. This isn’t always the case so with vanc AUC, generally note that when we talk about AUC it’s the AUC across 24 hours (1 day).

Now that we know what AUC is and what our goal range for dosing is, how do we calculate it?

There’s two ways to calculate AUC:

1. Manual calculations (Trapezoidal method)
2. Calculators (Bayesian calculations) either online or with special software

Let’s get down and dirty and start with the manual way.

Here’s what you’ll need to hand-calculate the AUC via the trapezoidal method:

• 2 (two) steady state vanc concentrations within one dosing interval
  • One concentration ~1-2 hours after the infusion has ended (Peak)
  • One concentration ~30 minutes before the next dose (Trough)

Notice here we need steady state vanc concentrations to calculate the AUC.

“Wait! If steady state doesn’t occur until after the 3^rd or 4^th dose how are we supposed to calculate the initial vanc dose?”

Great question! We’ll discuss this a bit later in the practical dosing section so hold on to that. For now, you may have correctly deduced that these AUC calculations are primarily performed for dosing adjustments AFTER the initial dose.

Alright, let’s say we’ve achieved steady state. Here’s what our expected concentration vs. time curve may look like (please excuse my crude drawings) for a complete therapeutic regimen over multiple doses:

Every point where the curve moves upward represents a dose and the absorption of vanc. Every point where the curve moves downward represents the elimination phase of vanc. The more doses we give, the higher the concentration is overall. That is, until it reaches steady state!

Let’s focus on just one of those dosing intervals at steady state:

Ok, you’re telling me we need calculate everything under that red curve?? Yes… and no. You don’t need to be a math wizard to know that’s probably going to be quite difficult given the curvature of the lines. Us pharmacists love our simplicity and the black and white (even though we may not get it all the time)!

So, here’s what we’re doing to do instead:

Much easier! Yes I know, we’re clearly missing a chunk of the AUC when we place our straightened “curve”. When doing manual calculations, you’re really just estimating what the AUC is, so you’re going to have a natural underestimation of the true AUC value using the trapezoidal method.

Next, here's where our "trapezoids" come into play:

Essentially what we’re trying to do here is calculate the area of the two trapezoids (green and pink) and add them together to get our estimated AUC. Simple, right?

Ok, stay with me here as we travel back to the olden days of geometry class… How do we calculate the area of a trapezoid?

Let’s incorporate this formula into our figure:

Notice our Cmin (obtained trough level) is the same at both the start of the infusion and the at the end right before the next infusion. Why is that? Because we’re at steady state! Remember, the rate of drug in equals the rate of drug out.

Let’s put everything together and create our adapted trapezoidal area equations:

The green trapezoid follows all the rules of your typical trapezoid equation. Cmin (trough level) is our base 1. Cmax (peak level) is our base 2. The time of the infusion (Tinf) is our height.

The pink trapezoid is where it gets a little more challenging. While I did draw a straight line to simplify things, note that during the elimination phase of vanc we must consider its logarithmic nature. This does mean that the area of the pink trapezoid does NOT follow your typical equation. I won’t get into the nitty gritty details here but know that we use the natural log for calculating the area of the pink trapezoid.

Please note that the above equations can only be used in a perfect world, where the peak level we obtain is the TRUE peak and the trough level we obtain is the TRUE trough. In the real world, the timing of the peaks and troughs that we obtain may look a bit more like this:

Recall what components are needed for the trapezoidal method. Peaks are usually taken ~1-2 hours POST-infusion, and troughs are usually taken ~30 minutes before the next infusion. What you absolutely DON'T want are peak or trough levels obtained during the infusion (BIG no no). If that happens, you'll need to start over with obtaining new levels (and no one wants that!).

Since our peak and trough levels are never going to be perfect, we need a few more equations to back-calculate the “true” peaks and troughs. You may have seen or learned about them before:

Now that we can calculate the Cmax (true peak) and Cmin (true trough) from our measured concentrations, all we need to do is plug and chug into the original equations, add them together and get the AUC!

We’ve got our AUC, now what? Practically speaking (which we'll go through later), you’d want to compare your manually-calculated AUC with that of an online software or calculator. Otherwise, you take your calculated AUC and see whether it falls within the desired goal range (400-600 mg*hr/L). If it falls below, you'd want to increase the dose or frequency (more frequent dosing). Above and you'd want to decrease the dose or frequency.

That marks the end for the manual calculations section. Keep reading to learn more about online calculators and Bayesian dosing.

Calculators have seriously made our lives way easier so let’s give thanks to these great inventions. Compared to manual calculations, all you need (at minimum) is one vanc level and you’re good to go! So, how’s this all possible? You may or may not have heard of the term “Bayesian” throughout your learning. Put simply, Bayesian dosing is a method that uses patient data and laboratory results to estimate a patient’s ability to absorb, process, and clear a drug from their system^{9, 10}.

Here’s how it works: Bayesian software first gathers individualized patient data and pharmacokinetic parameters (e.g., volumes of distribution, clearance) from a general population. It then uses this data to create a “population PK model”. Think of this as a generalized tool to predict how vanc will move through someone’s body (the “PK”). Whenever we input new patient data into the system, it takes this new data (including observed vanc concentrations) and combines it with the population model to predict and provide an optimized patient-specific dosing regimen. Essentially, we’re using prior knowledge (population model) to predict outcomes for new events (new patient).

The 2020 vanc therapeutic guidelines encourage the importance of using Bayesian software programs and recommend these programs over traditional calculations³.

What are these Bayesian software programs you may ask? Here’s a list of few you might’ve seen or heard of (or even used):

• DoseMe Rx
• Insight Rx
• Precise PK

Though the foundational concepts being used in each software for Bayesian dosing are the same, each employs their own unique way of optimizing individualized patient regimens. Some provide therapeutic monitoring and dosing for drugs other than vanc, including our beta-lactam antibiotics.

Basically, the whole idea and purpose for Bayesian dosing is to provide quicker, more appropriate, individualized therapy for our patients. See a great summary table below outlining the main differences between traditional vs. Bayesian dosing:

While all of this sounds like the perfect method for calculating a perfect dose for our patients, there are some considerations to keep in mind if you or your institution decides to use a Bayesian dosing software:

• First, since Bayesian dosing relies heavily on a general population, not every patient you’re treating will fall within the specific pop model the calculations are based on. For example, say you have an obese pediatric patient requiring vanc. You want to use Bayesian dosing, but many of the pop models only integrate normal, healthy adults into their calculations. An obese pediatric patient is going to have very specific metrics (PK-wise), so can we rely on the Bayesian software that uses adult data to predict a dose?

• Second, many institutions have expressed cost as a limiting factor on why they can't integrate these programs into their practice. According to a 2021 cost-benefit analysis by Lee and colleagues, Bayesian software programs vary widely in price from $10,000 per year to up to $50,000 per year¹¹!
• Third, even though these software programs make it incredibly easy to calculate specific therapeutic regimens for patients, there is still a need to educate people on how to use them. For 11 years (and counting), many have been utilizing trough monitoring to dose vanc, so there may be some pushback when it comes to education on why Bayesian dosing can be beneficial and how to use it effectively.

Bayesian dosing is a strong tool that provides a very promising method to help clinicians design an optimal dosing regimen for individual patients. Whether you’re a student, a doctor, a nurse, or a pharmacist, I encourage you to see (if you haven’t already) if your institution uses Bayesian dosing!

Congrats on making it this far! I promised I’d go over practical dosing and how we may perform initial doses of vanc, so let’s dive right in. I’ll be using an example institutional guideline from Northwestern Memorial Hospital located in Chicago, Illinois¹².

If it’s decided that empiric vanc (an initial dose) is needed, here is a nice table that provides suggested guidance based on creatinine clearance (CrCL):

At Northwestern, empiric dosing is based on the patient’s actual body weight (ABW) regardless of if they’re obese (though this is an unpredictable area based on differing pharmacokinetic studies of vanc). Notice an initial dosing of 15 mg/kg is the same across all CrCL groups. This also follows the 2020 vanc therapeutic guidelines which recommend a dosing regimen of 15-20 mg/kg (in patients with suspected or definitive MRSA infection). For obese patients, the initial dose is capped at 2,000 mg, but this cap may be different at other institutions. We round doses to the nearest 250 mg (e.g., 1000, 1250, 1500, etc.) to make it easier to compound (and our technicians happy).

Once the empiric dose and frequency are decided (based on CrCL), this is when you’d employ the AUC monitoring that we learned all about above. If we anticipate the patient will be on vanc for > 72 hours, we’ll need to obtain a steady-state trough level before the 3^rd or 4^th dose. Then, we’d go ahead and calculate the AUC to determine whether the current dose should be adjusted.

Determination/calculation of the AUC is a little different at Northwestern. Hand-calculations are not used (unless necessary). Instead, the “kinetics navigator” function embedded into Epic is used to calculate an AUC based on a single steady-state trough concentration. For those who’ve worked in Epic before, you may be using this function yourself! If not, here is a snapshot of what it looks like:

As stated by step H in the figure above, you can adjust the dose if needed after calculating the AUC through the kinetics navigator. Continue the same dose and frequency if the AUC falls within the desired range. Otherwise, you may use the navigator to calculate "test doses" and "test frequencies" to see what the predicted AUC may be if you adjust the dose accordingly to fit the patient's needs.

Sometimes, patients will require a “loading dose” of vanc. Put simply, it’s a large dose typically administered as a one-time bolus at the onset of therapy. This is different than the initial empiric dosing discussed above, often in addition to the initial empiric therapy. The recommendations for a load can vary. Per the 2020 guidelines, it’s recommended to give a 20-35 mg/kg dose with a max of 3,000 mg. Northwestern adopts a more stringent range, recommending 20-25 mg/kg with a max of 2,000 mg.

The purpose of a loading dose is to allow the patient to rapidly achieve therapeutic vanc blood levels in a timely manner. This can be helpful emergency situations and in those who may be critically ill. There’s also been multiple previous studies and systematic reviews that outline the benefit of providing loading doses to patients^13-15. See this diagram below for a nice visual representation:

We want to ensure that drugs are achieving therapeutic concentrations inside the body quickly, especially in critically ill patients. If we DON’T load, you can clearly see it takes a lot more doses to achieve the effective concentration (green line). If we DO load, you can see that the curve shoots straight up and above that effective concentration line after just one dose (red line).

So, who are the patients that we DO load vanc for?

According to the 2020 guidelines, loading doses are suggested for the following patients:

• Critically ill or in the ICU
• Require dialysis or renal replacement therapy
• Receiving vanc continuous infusion therapy

At Northwestern, there are similar suggestions with a few differences:

• Hemodynamic instability or critically ill, and ECMO
• Documented severe MRSA infections
• Suspected meningitis, endocarditis, or pneumonia
• A loading dose is NOT recommended for hemodynamically stable patients or non-ICU patients

The concepts are the same. Generally, we want to load patients who are critically ill or on some sort of renal replacement therapy. Institutional guidance on loads may vary.

I know there was a TON of information in today’s article, so let’s summarize everything we went through:

Bear in mind that these concepts are for more generalized methods of dosing vanc. There’s still so much more that we could go through including dosing in specific populations, details on the correlation of trough values to AUC, specific AUC goals for different sites of infections and causes, etc.

Let us know your thoughts on how to approach vanc dosing, and if you want more topics like this or practice examples comment down below!

1. Rybak MJ, Lomaestro BM, Rotschafer JC, et al. Vancomycin therapeutic guidelines: a summary of consensus recommendations from the infectious diseases Society of America, the American Society of Health-System Pharmacists, and the Society of Infectious Diseases Pharmacists. Clin Infect Dis 2009; 49(3): 325-7.
2. van Hal SJ, Paterson DL, Lodise TP. Systematic review and meta-analysis of vancomycin-induced nephrotoxicity associated with dosing schedules that maintain troughs between 15 and 20 milligrams per liter. Antimicrob Agents Chemother 2013; 57(2): 734-44.
3. Rybak MJ, Le J, Lodise TP, et al. Therapeutic monitoring of vancomycin for serious methicillin-resistant Staphylococcus aureus infections: A revised consensus guideline and review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists. Am J Health Syst Pharm 2020; 77(11): 835-64.
4. Finch NA, Zasowski EJ, Murray KP, et al. A Quasi-Experiment To Study the Impact of Vancomycin Area under the Concentration-Time Curve-Guided Dosing on Vancomycin-Associated Nephrotoxicity. Antimicrob Agents Chemother 2017; 61(12).
5. Abdelmessih E, Patel N, Vekaria J, et al. Vancomycin area under the curve versus trough only guided dosing and the risk of acute kidney injury: Systematic review and meta-analysis. Pharmacotherapy 2022; 42(9): 741-53.
6. https://www.tldrpharmacy.com/content/the-complete-but-practical-guide-to-dosing-vancomycin-based-on-aucmic-targets
7. Pai MP, Neely M, Rodvold KA, Lodise TP. Innovative approaches to optimizing the delivery of vancomycin in individual patients. Adv Drug Deliv Rev 2014; 77: 50-7.
8. https://www.certara.com/knowledge-base/calculating-auc-linear-and-log-linear/
9. https://www.wolterskluwer.com/en/solutions/sentri7-clinical-surveillance/medication-management/bayesian-dosing#:~:text=Bayesian%20dosing%20uses%20patient%20data,a%20drug%20from%20their%20system.
10. https://doseme-rx.com/bayesian-dosing/science-behind-doseme
11. Lee BV, Fong G, Bolaris M, et al. Cost-benefit analysis comparing trough, two-level AUC and Bayesian AUC dosing for vancomycin. Clin Microbiol Infect 2021; 27(9): 1346.e1-.e7.
12. https://adsp.nm.org/uploads/1/4/3/0/143064172/nm_vancomycin_protocol.pdf
13. Rosini JM, Laughner J, Levine BJ, Papas MA, Reinhardt JF, Jasani NB. A randomized trial of loading vancomycin in the emergency department. Ann Pharmacother 2015; 49(1): 6-13.
14. Mei H, Wang J, Che H, Wang R, Cai Y. The clinical efficacy and safety of vancomycin loading dose: A systematic review and meta-analysis. Medicine (Baltimore) 2019; 98(43): e17639.
15. Hodiamont CJ, Juffermans NP, Berends SE, et al. Impact of a vancomycin loading dose on the achievement of target vancomycin exposure in the first 24 h and on the accompanying risk of nephrotoxicity in critically ill patients. J Antimicrob Chemother 2021; 76(11): 2941-9.

*Information presented on RxTeach does not represent the opinion of any specific company, organization, or team other than the authors themselves. No patient-provider relationship is created.