March 25, 2020

EJRC - RCT of antiviral treatment for hospitalised hypoxic COVID-19 patients

A Trial of Lopinavir–Ritonavir in Adults Hospitalized with Severe Covid-19

COVID-19 was declared a pandemic by the WHO in March 2020.  Despite supportive care, the high mortality means there is a genuine need for a specific anti-viral treatment. Many ongoing studies are based on the experience gained from two previous Coronaviral infections: SARS in 2003 and MERS-CoV in 2012.

Cao et al conducted a randomized, controlled, open-label trial on antiviral treatment (lopinavir–ritonavir) for hospitalised hypoxic COVID-19 patients. (1) Lopinavir is HIV aspartate protease inhibitor which had been combined with Ritonavir to increase efficiency through the inhibition of cytochrome P450.

Eligible patients were adults (>18 years), non-pregnant, hypoxic patients (SaO2≤94% or PaO2/FiO2 <300 mmHg) manifesting pneumonia on the CXR. They excluded patients with HIV, allergy/hypersensitivity, severe liver disease, on medications contraindicated with lopinavir–ritonavir or breast-feeding. Treating physicians also had the right for exclusion based on ‘best interest’.

The authors reported no significant baseline difference between both arms – However, more patients in the lopinavir–ritonavir group had tachypnea, low systolic blood pressure, and higher viral load at day 1.

The primary outcome was time to clinical improvement based on either an ordinal scale or discharge from hospital.  The authors concluded no difference between both arms (median: 16 days in both groups; 95% confidence interval [CI], 0.95 to 1.85; p = 0.09). However, lopinavir–ritonavir group showed a better trend in many of the secondary outcome parameters:

  • The intention and modified intention to treat analysis of the 28-days mortality: (19.2% vs. 25.0%; 95% CI, −17.3 to 5.7; and 16.7% vs. 25.0%; 95% CI, −19.6 to 3.0 respectively).
  • Shorter stay in ICU (median, 6 vs. 11 days; 95% CI, −9 to 0).
  • Shorter time from randomisation to hospital discharge (median, 12 vs. 14 days; 95% CI, 0 to 3).
  • Better Clinical improvement at day 14 according to the 7-category ordinal scale (45.5% vs. 30.0%; 95% CI, 2.2 to 8).

Also, the lopinavir–ritonavir group received less vasopressors, RRT, NIV and invasive ventilation during the trial (Statistical significance not reported). However, most of those potential superiorities of lopinavir–ritonavir were not statistically significant.

Concerning the adverse effects, 14% of lopinavir–ritonavir recipients were not able to continue the treatment mainly due to GI side effects.



  1. RCT
  2. Addressing a very important research question in a time of a fatal pandemic (High impact article).
  3. The sample size (n=199) is very convenient for the short period and pressure to publish results.
  4. Theory generating for future research. The authors did address many limits in order to guide such studies.


1. Treatment was largely empirical, as there were no pre-clinical or early phase trials (in a context of time pressure and emergency fatal pandemic):

a) Late start of treatment (13 days median time between symptoms and randomization: Argument of benefit if early treatment before lung injury. In a post-hoc subgroup analysis, mortality was higher when treatment started 12 days after the onset of symptoms.

2. Trial methodology:

a) Open label (not blind), no placebo, single centre raising bias concerns.

– However, randomization was done by a statistician not involved in the trial, and allocation was concealed (advantage).

b) The treatment dose and duration were based on previous experience for other infection.

c) Subjective exclusion criteria: treating physician was able to exclude patients based on ‘best interest’ can cause selection bias.

d) End point at 28-days can miss mortality and side effects of treatment.

– However, it is acceptable due to the importance of the results and the authors aim to do further follow-up.

3. Measurement (Viral Oropharyngeal swabs) – 2 points need to be addressed:

a) Sensitivity in cases of pneumonia (versus lower respiratory samples).

b) Correlation with clinical cure/carrier state.

4. External validity and generalisability can be affected by:

a) Single centre trial

b) Younger patients (Median 58 years in comparison to more aged population in more developed countries).

c) Higher overall mortality in this trial (22.1%) compared to other descriptive studies of Covid-19. (2)

d) Steroids were administered in about third of the patients in both groups – a practice which can deviate from country to country.


The treatment by lopinavir–ritonavir failed to show earlier clinical improvement in this open label randomised single-centre controlled trial. However, it showed a potential advantage in many secondary outcome parameters. The conclusion and limits of the study can benefit and guide further research.  Such research should consider the earlier use, different dose and duration, clinical end point, blind RCT, combination therapy and larger sample size.

This article review was prepared and submitted by Dr Ashraf Roshdy, Queen Elizabeth Hospital (Lewisham and Greenwich NHS trust), London, United Kingdom, on behalf of the ESICM Journal Review Club.

Comment on this news