COVID-19 drug development in the critically ill – a “special population”

Commentary by Carl Kirkpatrick, PhD. Professor at Monash University

As the number of patients diagnosed with COVID-19 exceeded the 2.5 million worldwide, the impact on the hospital systems in many countries has been very apparent.   Given the limited number of Intensive Care Unit (ICU) beds per hospital or region, one of the key modelling factors for health systems has been to understand the likely number of infected patients that will ultimately require critical care support and the resource allocation required to support those predicted patient numbers.

Following the well described initial COVID symptoms, approximately 15% of patients develop pneumonia (~ 5 days later) and a smaller percentage (~5-10%) need critical care support in an ICU. With ICU admissions often occurring 7-12 days after initial symptom onset.  It is suggested that the observed variance in ICU admission rate from country to country or state to state has been determined by bed availability and local infection rates.  This aside, there is reasonable evidence that patients requiring ICU care are generally older with multiple comorbidities, including hypertension, diabetes, obesity, and respiratory disorders e.g. asthma or COPD (1) .

There have been two recent publications describing the global COVID-19 ICU learnings for patient triage into the ICU, clinical management and support, best practice for patient isolation, and staff protection from COVID transfer and infection. (1,2).  They have also raised the key unanswered clinical and therapeutic management questions that will require robust research, patient registries or well-designed RCTs.

The aim of this short article is to initiate discussion about the clinical pharmacology and dose considerations of current COVID-19 treatments in ICU rather than usual ICU supportive care. Many of the proposed antiviral compounds and immune-supportive therapies currently being used or proposed for COVID-19 are not typically used in the ICU setting. Furthermore, new agents will be proposed and be introduced into clinical practice in the coming months.  This forum is an ideal place to ask questions and seek guidance on dosing, drug interactions and dose optimization in ICU for these newer agents.

Pharmacotherapy in ICU is rarely easy; it is complicated by changes in physiology as a result of shock, myocardial dysfunction, and acute kidney injury associated with COVID-19 and secondary bacterial infections (1,3)).  As a result of these physiological changes, the following pharmacokinetic changes can occur: larger volumes of distribution, reduced hepatic blood flow altering the clearance of metabolized drugs, and lower clearances of renally eliminated medications due to acute kidney injury.  For example, Vancomycin has a larger volume of distribution in the critically ill resulting in lower concentrations for a given dose, which combined with reduced renal elimination i.e. lower clearance, means that time to reach steady state will be longer i.e. longer half-life.   Dosing strategies to overcome these issues are well known for standard critically ill therapies and anti-infectives.

The key agents currently under investigation (and with the most evidence for achieving effective concentrations in plasma or lungs) are the combination HIV treatment containing Lopinavir plus Ritonavir (800mg/200 mg) twice daily (, Hydroxychloroquine (HCQ) utilized in a variety of different doses and dosing regimens, and Azithromycin (500mg for up to 5-9 days).  With the exception of azithromycin, none of these agents are typically used in the critical care setting and therefore no advice is available on necessary dose adjustments for their use in these patients.

An understanding of the pharmacokinetics of these agents relative to altered physiology in the critically ill and possible pharmacokinetic and pharmacodynamics drug-drug interactions with standard concomitant ICU supportive therapy is essential. The mantra of needing the right drug and the right dose for the patient, means that individualized dosing decisions using first principles given the patients current clinical situation will be required.

Some points for consideration when making drug and dosing decisions:

  1. Sponsor generated product information (PI) or independent drug monographs that provide guidance to prescribers on dosing in critical care settings may not be available;
  2. Lopinavir is a significant “perpetrator” of pharmacokinetic drug-drug interactions, but its effects are complex, being concentration and time-dependent. Indeed, it acts as a strong inducer of CYP2C19 but also as a strong mechanism-based inhibitor of CYP3A enzymes. Ritonavir is also a strong mechanism based inhibitor of CYP3A enzymes used to boost the bioavailability of Lopinavir. Therefore many of the concomitant drugs used to treat the critically ill may are potential “victims” for lopinavir/ritonavir-mediated drug-drug interactions, particularly those with narrow therapeutic indices such as anticoagulants, short acting benzodiazepines eg midazolam, immunosuppressants and anti-convulsants; (4)
  3. Despite being registered since 1955, the molecular determinants of HCQ pharmacokinetics are incompletely understood e.g., the contributions of CYP enzymes involved in the generation of active metabolites. Likewise, there is large between patient variability in the steady-state concentrations of hydroxychloroquine. It would seem likely that combinations with pan CYP inhibitors are likely to increase the concentrations of HCQ through drug-drug interactions.  It is associated with cardiac arrhythmias and ECG monitoring will be required.
  4. Many of the antiviral agents recommended for treating COVID-19, have significant cardiac side effects or toxicities. These will become more apparent as doses are increased to the upper range or drug clearance decrease due to altered physiology caused by infection. Thus, there is significant potential for pharmacodynamics drug interactions particularly cardiac arrhythmias with combination anti-viral therapies, such as QT prolongation. Concomitant therapies used to treat myocardial dysfunction/arrhythmias e.g. amiodarone and flecainide are not recommended even with ECG monitoring (4).
  5. Azithromycin being part of the macrolide family could be expected to have CYP drug-drug interaction potential. However, Azithromycin is thought to only a mild inhibitor and not lead to significant “perpetrator” drug-drug interactions.  It may also cause some QT prolongation, but not lead to torsades de pointes, ECG monitoring would seem logical.
  6. Renal Replacement Therapy (RRT) may be required due to acute kidney failure. Indeed, principles exist around hydrophilicity, molecular weight and protein binding as to the extent a drug will be removed by various renal replacement modalities.  It would seem unlikely that HCQ or Lopinavir will require dose adjustment due to very large volumes of distribution and high protein binding respectively.  Further data on the effect of RRT on other anti-viral and immune-supportive therapies that may be used to treat COVID-19 can be found here (4).

Pharmacotherapy in the ICU poses many challenges for clinicians and drug developers at the best of times. Choosing the right drug and dose while the patient has altering physiology and therefore pharmacokinetics, means that integrated knowledge of drug pharmacology, pharmacokinetics and making decisions based on first principles is key.

Please continue this discussion in the “Dosing for Special Patient Populations” section of the forum.

  1. Phua J, Weng L, Ling L et al. Intensive care management of coronavirus disease 2019 (COVID-19): challenges and recommendations.  The Lancet Respiratory Medicine 2020
  2. Alhazzani, W., Møller, M.H., Arabi, Y.M. et al. Surviving Sepsis Campaign: guidelines on the management of critically ill adults with Coronavirus Disease 2019 (COVID-19). Intensive Care Med (2020).
  3. Cotta M, Roberts JA, Lipman J. Antibiotic dose optimization in critically ill patients.  Medicina Intensiva 2015;39(9):563-72.
  4. Zeitlinger, M, Koch B, Bruggemann R, et al., Pharmacokinetics-pharmacodynamics of antiviral agents to treat SARS-COV-2 and their potential interactins with drugs and other supportive measures: A comprehensive review by the PK/PD anti-infectives study group of the European Society of Antimicrobial Agents.


Leave a Reply

Your email address will not be published. Required fields are marked *