How non-adherence encourages antibiotic tolerance

New research on antibiotic-tolerant bacteria has underlined the importance of medicines compliance

Researchers have developed a simple new method for measuring the time it takes to kill bacteria that have become tolerant to antibiotic drugs.

Tolerance is the first step towards diseases developing full resistance to antibiotics.

The new findings should allow doctors to ensure they are prescribing antibiotics for the correct length of time to prevent antibiotic resistance developing, say the Israeli and American researchers.

“These findings allow measurement of tolerance, which has previously been largely overlooked in the clinical setting,” says senior study author Nathalie Balaban of The Hebrew University of Jerusalem.

“Routinely measuring tolerance could supply valuable information about the duration of antibiotic treatments, reducing the chance of both under- and over-treatment.

“Furthermore, data compiled from such measurements could give an estimate of how widespread the phenomenon of tolerance really is, which is currently a complete unknown.”

Due to selective pressure, pathogens acquire resistance through mutations that make the antibiotic less effective, for example, by interfering with the ability of a drug to bind to its target. Currently, clinicians determine which antibiotic and dose to prescribe by assessing resistance levels using the routine metric minimum inhibitory concentration (MIC)—the minimal drug concentration required to prevent bacterial growth.

Although resistant strains continue to grow despite exposure to high drug concentrations, tolerant strains can survive lethal concentrations of an antibiotic for a long period of time before succumbing to its effects. Tolerance is often associated with treatment failure and relapse, and it is considered a stepping stone toward the evolution of antibiotic resistance.

But unlike resistance, tolerance is poorly understood and is currently not evaluated in healthcare settings.

“The lack of a quantitative measure means that this aspect of the treatment relies largely on the experience of the individual physician or the community,” says first author Asher Brauner, a PhD student in Prof Balaban’s laboratory.

“This can lead to treatment being either too short, increasing the risk of relapse and evolution of resistance, or much too long, unnecessarily causing side effects, release of antibiotic waste into the environment, and additional costs.”

The team developed a tolerance metric called the minimum duration for killing 99% of the population (MDK99).

The  protocol, which can be performed manually or using an automated robotic system, involves exposing populations of approximately 100 bacteria in separate microwell plates to different concentrations of antibiotics for varied time periods, while determining the presence or lack of survivors.

The researchers applied MDK99 to six Escherichia coli strains, which showed tolerance levels ranging from two to 23 hours under ampicillin treatment.

MDK99 also facilitates measurements of a special case of tolerance known as time-dependent persistence—the presence of transiently dormant subpopulations of bacteria that are killed more slowly than the majority of the fast-growing population. Like other forms of tolerance, time-dependent persistence can lead to recurrent infections because the few surviving bacteria can quickly grow to replenish the entire population once antibiotic treatment stops.

“A take-home message from this is that it is important to complete a course of antibiotic treatment as prescribed, even after the disappearance of the symptoms,” Prof Balaban says.

“Partial treatment gives tolerance and persistence mutations a selective advantage, and these, in turn, hasten the development of resistance.”

In future studies, Prof Balaban and her team plan to use MDK99 to study the evolution of tolerance in patients. Moreover, the ability to systematically determine the tolerance level of strains in the lab could facilitate research in the field, they say.

“If implemented in hospital clinical microbiology labs, MDK99 could enable the efficient classification of bacterial strains as tolerant, resistant, or persistent, helping to guide treatment decisions,” Prof Balaban says.

“In the end, understanding tolerance and finding a way to combat it could significantly reduce the ever-growing risk of resistance.”

Previous Pharmacy under more pressure from Amazon
Next Should the US emulate our S3 schedule?

NOTICE: It can sometimes take awhile for comment submissions to go through, please be patient.


  1. Robyn Ingram

    Pharmacists need to be aware that the methods by which microbes develop antimicrobial resistance are complex, specific for each drug-bug pair, and depend on the interplay of a wide variety of factors. This article has presented the information in a highly simplified manner and the recommendations should be interpreted with caution.

    The purpose of antibiotic treatment is not necessarily to reduce the bacterial count to zero. By impeding microbe growth the antibiotic assists the patient’s own immune system to overcome the infection. It is sufficient to slow the rate at which the bacterial colony is growing for antibiotic treatment to be successful. It’s also worth remembering that antibiotics don’t just have an impact on the targeted bacteria but affect the patient microbiome, the ramifications of which we are only just beginning to understand.

    Recommendations to “complete the course” of antimicrobials are only appropriate if the prescribed treatment duration is appropriate for the condition – however we often see excessively lengthy antibiotic courses, involving unnecessary repeat prescriptions, prescribed in the community setting. Prolonged antimicrobial courses greatly increase the risk of antibiotic resistance developing and are rarely necessary in the treatment of acute infections in community practice. Advising a patient to continue taking antibiotics once their symptoms have improved, and beyond the course lengths recommended in the Therapeutic Guidelines: Antibiotic, cannot be recommended. Exceptions include if the patient truly warrants antimicrobial prophylaxis or is under the care of an infectious diseases specialist.

    Patient non-adherence, as it relates to skipping medication doses or erratic treatment, should be discouraged and pharmacists are well placed to advise patients about methods to assist with treatment compliance.

    In general, the prescribing and administration of antimicrobials should follow the recommendations in the Therapeutic Guidelines: Antibiotic. Pharmacists encountering prescribing that deviates from this best practice guideline should contact the prescriber to discuss the issue.

    Robyn Ingram
    Program Manager – National Antimicrobial Prescribing Survey

    • Jarrod McMaugh

      G’Day Robyn

      I’m going to start by saying that I don’t disagree with the message you are delivering here, and that controlling the prescribing of antibiotics is extremely important for future health of all.

      That being said, there are a few points I would make.

      First, the articles on AJP aren’t intended to be technical in nature – they are news-type articles and as such in-depth discussions are not always warranted… it’s about raising awareness of the issue, and expectation that the pharmacist will do further reading on the matter. Therefore your comment that this article is presenting the matter too simplistically may be a bit unfair to the author.

      Secondly, while the nature of the guidance from TG: Antibiotic is rigorous, the nature of communication between prescribers and pharmacists in Australia is not such that we would know the causative organism, or even whether sensitivity to an antibiotic has been determined. Much of the prescribing is empirical in nature. In addition, we don’t know the intent of the prescriber in 90% of the antibiotic scripts that we do….. so our main drive is to reinforce what the prescriber has advised the patient. We can’t confirm the correct treatment in TG if we don’t have access to all of the clinical information. While we have the clinical capacity to do this, we don’t have the infrastructure in place to allow it to happen (perhaps MyHealthRecord will assist with this)

      Thirdly, while it is our role to ensure that prescribers are not making errors, we also have to operate in a way that gives the benefit of the doubt to the presriber’s knowledge for those antibiotic prescriptions that we receive that are seemingly the accepted treatment protocols – this means we aren’t ringing presribers for every antibiotic prescription to clarify the correct dose, active ingredient, form, and course of treatment….. perhaps this is our future though given the way antibiotic resistance is heading.

      • Ronky

        Jarrod, have you ever in your life known of a doctor in private practice to decide the quantity of antibiotic he prescribes based on how much he thought was necessary to treat the infection rather than “What’s the maximum quantity (and maximum number of repeats) that I can get the Government to pay for?”
        The days of “You MUST finish the entire prescribed course of antibiotics even if you’re completely well!” are over. There, is that simple and newsy enough for you?

        • Jarrod McMaugh

          yeah heaps Ronky. I know plenty of GPs who do both.

          Completely agree that there is no place for prescribing a quantity or course length just because that’s the pack size, but similarly I would not judge every antibiotic prescription I receive to be based on the pack size….

Leave a reply