The Need for Early Management in Patients With COVID-19

In March 2020, the IHU Méditerranée Infection set up a screening and treatment center for patients with COVID-19, a system that has been ultimately recommended by French public health authorities. The recent publication of the pro�les of patients hospitalized in France published by the Directorate for Research, Studies, Evaluation and Statistics gives us the opportunity to measure the impact of this multidisciplinary early management system coupled with screening on mortality at 90 days. Analysis of the data shows that the system established at IHU-MI was associated with lower mortality, taking age and sex into account. Regarding the age-standardized mortality rate, mortality rates were lower than national data regardless of the period of the epidemic. Early management seems to have signi�cantly decreased the mortality rate in the under-60 age group, suggesting the importance of early management, regardless of age. In addition, these patients had pejorative clinical criteria (high NEWS-2 score, ICU visits, oxygen saturation below 95%) requiring hospitalization, and co-morbidities that are now known to be aggravating factors [7]. This reinforces the need to care for all individuals, regardless of age. Early medical care, as part of a system integrating a screening center and a day hospital, may explain the lower mortality rates.


Introduction
In December 2019, a new virus of the coronaviridae family called SARS-CoV-2 emerged in Wuhan, Hubei region, China.It spread rapidly to the rest of the world and was declared a pandemic in March 2020.As of November 16, 2020, there were 1,319,267 patient deaths from COVID-19 [1].
The management of patients with COVID-19 has evolved over time, particularly in France.Indeed, when the rst cases appeared in February/March 2020, the only individuals screened were "individuals presenting clinical signs of acute respiratory infection with documented or subjective fever and who had traveled or stayed in a high-risk exposure zone within 14 days prior to the date of clinical signs, or individuals who have had close contact with a con rmed case of COVID-19 or any person with signs of pneumonia or acute respiratory distress" [2].Such management did not include any recommendation for mass screening that were already in plance in countries like Iceland or South Korea [3,4].Indeed, the French government explicitly indicated that screening during the epidemic phase was not necessary [5].
By March 17, the French authorities had implemented a 55-day population lockdown as a health measure.Patients with COVID were instructed to consult emergency services only in case of respiratory di culties [6].During the rst wave of the new virus, only one treatment was o cially recommended to reduce fever in COVID-19 cases: paracetamol [6].
At the same time, the Institut Hospitalo-Universitaire Méditerranée Infection (IHU-MI) based in Marseilles, South-Eastern France offered an alternative management system.The IHU, created in 2011 and funded by the Ministry of Research, is the only research and care facility of this kind dedicated to the ght against infectious diseases in France (https://www.mediterranee-infection.com/).It includes a biology laboratory, 75 hospital beds and research and development teams.In March 2020, the IHU-MI set up a screening and treatment center for patients with COVID-19, a system that has been ultimately recommended by French public health authorities [7].The IHU offered rapid screening, with results in less than 24 hours, to any individual presenting at the center, as well as outpatient treatment for patients who were positive for SARS-CoV-2.The IHU standardized clinical protocol [8] included: a medical examination with measurement of pulse, blood pressure, respiratory rate and ambient air saturation to evaluate the NEWS-2 score [9], a biological assessment, a low-dose chest CT scan according to age and/or desaturation criteria [10,11].As regards drug treatment, treatment with hydroxychloroquine-azithromycin in the absence of contraindications with the addition of broad-spectrum antibiotics (ceftriaxone or ertapenem) in patients with a NEWS-2 score greater than 5 was proposed [8].When patients had an oxygen saturation below 95% or other clinical signs demonstrating deterioration of the individual's health status, they were then hospitalized at IHU-MI, mostly when they were contagious, to avoid the spread to non-COVID patients and staff.At the peak of the epidemic in April and bed saturation, once they were RT-PCR-negative, patients were transferred to a conventional COVID unit for their remaining care.For outpatients, follow-up was performed at the beginning of the epidemic at D2, D6 and D10 and from 03/2020 onwards only at D10, due to the large number of patients [8].
Patient observation and massive early diagnosis (4,021) made it possible to adapt patient management, which has evolved in line with the knowledge acquired through multidisciplinary collaboration involving cardiologists, radiologists, infectious disease specialists, intensivists and ENT specialists [11][12][13][14][15][16].For example, the observation of 'happy hypoxia' has led to the recommendation of ambulatory use of pulse oximeters [13] and the search for high D-dimer anticoagulation levels in patients at risk [14].This management has been the subject of several publications on clinical and therapeutic results [11,16].
The recent publication of the pro les of patients hospitalized in France published by the DRESS [17] gives us the opportunity to measure the impact of this multidisciplinary early management system coupled with screening on mortality at 90 days.

Population study
Our study is based on a comparison between patients hospitalized at IHU Méditerranée Infection (IHU-MI) in Marseille, (France) and the inpatient population in France analyzed in the report of the Directorate for Research, Studies, Evaluation and Statistics (Direction de la Recherche, des Études, de l'Evaluation et des Statistiques, DRESS) [17].
The period covers patients hospitalized between March 1 and June 15, 2020.Diagnosis of COVID-19 disease is based on the same criteria for both populations: RT-PCR testing and/or COVID-speci c images of COVID disease on chest CT.However, RT-PCR was the essential criterion for the diagnosis of COVID-19 in our Institute.

Criteria for comparison
The comparison between the two hospitalized populations with COVID-19 focuses on age, gender and mortality at 90 days.Hospital mortality was sought for all patients hospitalized at IHU-MI 90 days after admission using the Medical Information Department (DIM) of the Assistance Publique-Hôpitaux de Marseille (AP-HM).This updated death census was carried out on October 21, 2020.However, only deaths that took place in the hospital could be documented.
The clinical characteristics (NEW-2 severity score, oxygen saturation, clinical symptoms on admission and associated comorbidities) of patients hospitalized at IHU-MI were collected but were not compared with the DRESS population, as these data were not available in this dataset.
The percentage of Intensive Care Unit (ICU) visits was observed and compared between the two populations.However, the status of ICU visits was not further analyzed due to the inability to identify ICU and critical care patients in the DRESS study.

Statistical analysis
Categorical variables were presented as n (%) and continuous variables as mean(std) q1-median-q3.We used Fisher's exact test and the Wilcoxon-Mann-Whitney test to compare distributions of categorical and continuous attributes between different categories of patients.One-sided exact binomial tests were performed (when appropriate) to determine if the proportions observed in our cohort were signi cantly lower than national estimates.Two sided 95% con dence intervals were also calculated.To compare death rates at 90 days in our institute with national estimates, we also used direct age standardization.
The reference population was all patients hospitalized for COVID-19 between March 1 and June 15 in France (n = 91,061).A two-sided p-value of less than 0.05 was considered statistically signi cant.
Analyses were carried out using SAS 9.4 statistical software (SAS Institute, Cary, NC).

Mortality rate 90 days after admission
In France, 17,367 (19%) inpatients died within 90 days of admission compared to 6.6% of patients hospitalized at IHU-MI (p < 0.0001) (Fig. 1).The majority of deaths occurred in the over-80 age group (Fig. 1).No deaths occurred in the under-50 age group at MI HUI, while deaths ranged from 1.3-2.1% in the 0-40 age group and 3.6 to 4.6% in the 41-50 age group nationally.In older age groups, one sided exact binomial tests indicated that the mortality rates were signi cantly lower among women aged 71-80 years at IHU-MI (7.7% vs 18.6% -p = 0.0400) and among patients aged > 80 years (18.9% vs 39.4% -p = 0.0011 and 17.2% vs 27.9% -p = 0.0133 for men and women, respectively).
Age-standardized mortality rates at HI-MI for the months of March, April and May-June are still lower than those observed in the French national data (Fig. 2).The gap between the mortality rates for these two populations narrowed in May-June (7.8% at IHU-MI vs. 12.2% in France).Hospital mortality rates in France decreased over time, from 24.6% in March to 12.2% in May-June.This trend is less marked at the level of our Institute.

Clinical characteristics of patients under 60 years of age hospitalized at IHU-MI
Patients hospitalized at IH-MI were younger.Indeed, the most represented age group was 51-60 years old (23.1% versus 14% in France) and only 19.9% were over 80 years old, versus 31% for France (Table 1).
Pro le analysis of patients aged 60 (47%) and under showed a pejorative clinical pro le, with 13.9% having a NEWS-2 score greater than or equal to 7; 20.0% of subjects aged 60 and under had an oxygen saturation below 95%, and 9.7% had a stay in the ICU (Table 2).Nearly half of the 47% had a comorbidity.The same proportion of individuals with dyspnea is found in those under 60 years and those 60 years and older (36%).

Discussion
Analysis of the data shows that the system established at IHU-MI (systematic screening with rapid reporting of results and comprehensive management of positive results) was associated with lower mortality, taking age and sex into account.Regarding the age-standardized mortality rate, mortality rates were lower than national data regardless of the period of the epidemic.There was also a decrease in the age-standardized mortality rate between March and April, as observed at the national level.
The elderly (over 80 years of age) are the individuals who most frequently die in the context of COVID-19.This was observed in our cohort as well as in France and worldwide.However, the difference in mortality between the IHU population and the DRESS population is greatest in the under-60 age group.Only two deaths (0.6%) in the under-60 group (one 59-year-old and one 60-year-old patient) were noted in our population, compared to 26.6% nationally.Overall, early management seems to have signi cantly decreased the mortality rate in the under-60 age group, suggesting the importance of early management, regardless of age.In addition, these patients had pejorative clinical criteria (high NEWS-2 score, ICU visits, oxygen saturation below 95%) requiring hospitalization, and co-morbidities that are now known to be aggravating factors [7].This reinforces the need to care for all individuals, regardless of age.In France, the management of so-called "young" patients has probably been underestimated, given the rst available severity criteria.They were not considered to be at risk at the time.Today, recommendations include the existence of co-morbidities as a factor of severity, regardless of age [7].
One of the explanations for these positive results in terms of mortality is undoubtedly access to the exceedingly early care system facilitated by the IHU-MI, which screens and manages COVID-19 patients within the same structure.The implementation of generalized screening open to all; i.e., both symptomatic and asymptomatic individuals, has made it possible to quickly and easily enter a marked care pathway.Indeed, the time required for treatment of COVID-19 is extremely long and has a strong impact on mortality, similar to oncology.Management is modi ed according to the disease stage of the patient [8].The rst stage is the appearance of lung lesions and the rst clinical signs associated with a high viral load after the incubation period.During this stage, an antiviral was given to the patient if there were no contraindications and was usually combined with the use of a broad-spectrum antibiotic.The second phase corresponds to the persistence of the virus and an immune reaction during which patients, particularly those with risk factors, were particularly monitored.Lymphocytopenia, eosinopenia, elevated troponin or D-dimers greater than 0.5 µg/L were observed during this second phase.Thrombotic complications were monitored.The third stage corresponds to the in ammatory phase, which occurs between day 7 and 10 and is linked to the release of pro-in ammatory cytokines associated with a high risk of transfer to ICU.Severe acute respiratory syndrome (SARS) is the last phase and requires ICU management.It is preferable to manage these patients before this in ammatory phase, when patients who have decompensated are found.In France, many patients who did not have access to a center such as the IHU-MI in the rst phase of the epidemic undoubtedly complied with the recommendations of the General Health Directorate and went to the emergency department.Unfortunately, the lockdown and 'happy hypoxia' certainly delayed the management of the patient presenting in a signi cant in ammatory phase, leading to frequent recourse to the ICU.At IHU-MI, a complete medical examination, including oxygen saturation on ambient air, low-dose thoracic CT scan, and a biological control made it easier to identify patients with no clinical signs of severity but whose lungs were badly damaged.Early medical care, as part of a system integrating a screening center and a day hospital, may explain the lower mortality rates.
The complexity of the health situation; i.e., faced with an unknown disease in the context of a hospital crisis [18] and a non-operational crisis mechanism [19] has disrupted the health management of this crisis.The IHU-MI model was able to set up an e cient organization; the massive reception (33,503) of patients made it possible to build up a database of observations and research which allowed better understanding of the pathophysiological mechanisms of this disease.The signi cant difference in mortality rates shows the effectiveness of the IHU-MI model and the need for more in-depth feedback on the different methods of management of SARS-CoV-2 positive patients in order to identify areas for improvement, particularly in the treatment pathway.

Declarations Ethic declaration
Data from our cohort were collected retrospectively from the routine care setting using the electronic health recording system of the hospital.Our institutional review board committee (Mediterranée Infection