How has the Hospital Clínic managed to reduce covid-19 mortality from 11.6% to 1.4%?

It comes as no surprise that as time has gone by and more has been learned about covid-19 and its clinical impact, more patients are able to leave hospital ICUs and overcome the acute phase of the disease. Beyond learning, however, there are other factors that, combined, can achieve dramatic reductions in hospital mortality. This is the case of Hospital Clínic in Barcelona, whose mortality rates have decreased from 11.6% to 1.4% in the period from March to September. In other words, if at the end of March one in ten people died in the health centre, in September it was one in a hundred, one of the lowest hospital mortality rates in Europe. The results of the study, led by Carolina García-Vidal, a member of the Infectious Diseases Service and researcher at IDIBAPS, have been published in the journal The Lancet.

How can mortality be reduced so drastically in such a short time with a disease surrounded by more questions than certainties? "There is not just one factor, but it is the combination of many elements that makes it possible," responds García-Vidal. "Learning is one, but it is by no means the only one," he adds. The researcher attributes much of the success achieved to the teamwork of multiple specialists.

"The first step is to have enough ICU beds, ventilators and all the technological equipment needed for each patient," she says. In addition, it is necessary to have the pharmacological treatments that are working at this point and to administer them "at the right time". In essence, antivirals that help contain the spread of the coronavirus such as remdesivir, which gives good results in specific phases of the disease, and anti-inflammatory drugs with sufficient power to limit respiratory involvement.

In addition to all these elements, the Hospital Clínic has developed a computer model in conjunction with the Barcelona SuperComputing Center (BSC) that predicts with high levels of precision the evolution of a specific patient and thus can anticipate specific episodes of the disease in which a specific intervention would be necessary. The model, based on artificial intelligence, takes as its starting point the characteristics of each patient, including the existence of any previous pathology. "It is a tool to help the personalised treatment of each patient," says Garcia-Vidal. The personalised treatment helps, on the other hand, to "act early".

However, for the specialist it is essential to "avoid the collapse" of the system, since if any of these elements were lost, it would be difficult to maintain a low mortality rate. "Without ICU beds, ventilators, drugs and nursing, medical and cleaning equipment, none of this would be possible," she says.

Unprecedented work

The work published in The Lancet includes 1,645 patients of both sexes with a seven-month follow-up from March 1 to September 30, 2020, admitted to Hospital Clínic. Overall mortality, both on the hospital ward and in the ICU, reached 11.6% in March. In September the rate dropped to 1.4%. The study is one of the first in the world to collect this set of data.

Throughout this period both the age of patients and the diversity of previous pathology "have been changing", says the researcher, while the mortality rate has been reduced, which is why she considers it "reasonable" to attribute the decline to improvements in treatment, the availability of ICUs and also professionals from different fields. Similarly, faster hospitalisation from the appearance of first symptoms is also associated with lower mortality.

Garcia-Vidal considers that the combination of all these elements is key to successfully facing this third wave of covid-19. "Applying more aggressive treatments gives good results," he says. "When in other centers they were moving between 30% and 40% mortality, we were already in the vicinity of 10%. Applying the guidelines implemented at the Clinic now "we can say that we have a mortality of 1.4%". The guidelines, says the researcher, can serve as a guide or a "quality standard" in treatment. If so, it would be one of the first in the world.