Vancomycin TDM: Therapeutic Drug Monitoring or Just Fooling Ourselves?

Abstract & Commentary

Synopsis: Despite many years of practice, a recent questionnaire about therapeutic drug monitoring of vancomycin revealed marked variability and a lack of consensus regarding postdose assay sampling times, target ranges, and what constituted a toxic level.

Source: Tobin CM, et al. Vancomycin therapeutic drug monitoring: Is there a consensus view? The results of a UK National External Quality Assessment Scheme (UK NEQAS) for Antibiotic Assays questionnaire. J Antimicrob Chemother. 2002;50:713-718.

A questionnaire about vancomycin therapeutic drug monitoring (TDM) was sent to 310 subscribers of the United Kingdom National External Quality Assessment Scheme (UK NEQAS) for antibiotic assays inquiring about the methodology, dosage regimen, and toxicity. Just more than half (178) responded, and, of these, 49% provided a 24-hour, 7-days-a-week service. A quarter of the laboratories received fewer than 100 requests annually, 38% received 100-500 requests, 13% received 500-1000, and 24% received more than 1000. Four in 5 had guidelines for vancomycin use based mostly on the manufacturer’s labeling or the British National Formulary. Only 11% deviated from a twice-daily dosage regimen. Assays after 2-3 days of treatment were recommended by 71% of respondents. Trough samples were taken within 10 minutes of the dose by 92% of respondents. By contrast, postdose samples (ie, "peak samples") were taken 1 hour after the dose by 44% and 2 hours after the dose by 28%, with the remainder not mentioning any specific time at all. Only 11% of respondents stated that the sample time for a peak concentration referred to that observed after the end of the infusion. Anticipated trough levels ranged from 3 mg/L to 15 mg/L while corresponding peak levels ranged from 15 mg/L to 50 mg/L. Three out of 4 regarded trough levels ³ 10 mg/L to be toxic, whereas 20%, 40%, and 13% considered postdose levels of 30 mg/L, ³ 40 mg/L, and ³ 50 mg/L, respectively, to constitute a toxic concentration (see Table 2, below). A dose reduction would be advised by 35%, omitting the next dose by 30% and extending the dosing interval by 25%. Microbiologists reported 47% of results, biomedical scientists 46%, and biochemists or pharmacists the remainder. All but 5% found the assays appropriate, with a quarter of respondents believing them to be useful and a further 25% regarding them as essential for patients given dialysis.


Reported Peak and Trough Vancomycin Targets

peak N = 111 trough N = 115
conc [mg/L]  conc [mg/L]  %
50 1 1 15 1 1
40 1 1 12 1 1
35 to 45 3 3 10 54 47
35 1 1 8 2 2
30 to 50 1 1 5 to 15 1 1
30 to 40 2 2 5 to 12 1 1
30 8 7 5 to 10 52 45
26 1 1 5 1 1
25 to 40 12 11 3 to 10 2 2
25 to 35 3 3      
25 to 30 4 4
25 1 1      
20 to 50 1 1
20 to 45 1 1      
20 to 40 31 28
20 to 25 1 1      
20 to 30 17 15
20 1 1      
18 to 30 3 3
18 to 26 13 12      
18 to 25 2 2
18 to 24 1 1      
15 to 25 2 2

Comment by J. Peter Donnelly, PhD

Any newcomer to this field might well be quite taken aback by the fact that after more than 20 years of widespread use there seems to be no standard of TDM for vancomycin or even a consensus. Aside from the shortcomings of such surveys, which never elicit responses from all concerned, these results demonstrate a clear need for developing practice guidelines based on evidence. A cursory glance at the Table is enough to convince. While most (92%) used trough levels of < 10 mg/L, the best that can be said of anticipated peak levels is "somewhere between 15 mg/L and 50 mg/L." Hardly a good result. We do not know if the larger providers were more in agreement than the intermittent ones nor if the microbiologists had less-divergent practices than the biomedical scientists, but if TDM was a product there would be very few customers indeed. Imagine taking your car to several garages for its mandatory annual check-up with all offering the same service but with markedly different ways of interpreting this. Some are open all hours, while others are open only during office hours. Some tell you to run the engine for at least 2 hours before coming in, while others tell you to park it outside overnight so that it can be tested cold. All use the same diagnostic equipment, but some feel the idle speed should be 750 rpm, others 1000 rpm, and some even 1500 rpm. Different ones feel that the rpm should be 2500 rpm at a steady 50 mph, others 3000 rpm, and again a few 3500 rpm. Then each one interprets the same report differently. At the very least, you would be confused. You might even be sufficiently motivated to submit a complaint along with thousands of others to your local consumer watchdog, which would be up in arms and demand immediate legislation.

It would be unfair and naïve to assume that only microbiology suffers from a lack of standards. But the example is revealing especially at a time when standards are being demanded and good laboratory practice is accepted as a goal worth pursuing. Clearly a standard for vancomycin TDM is needed but that is only the beginning. Once drawn up, it has to be implemented and tested on a regular basis to ensure compliance. Achieving this within the United Kingdom is a tall order even for the most dedicated but still feasible with commensurate funding and support. Adopting it throughout the European Union may be asking for nothing short of a miracle.

Dr. Donnelly is Clinical Microbiologist University Hospital Nijmegen, The Netherlands Section Editor, Microbiology.