U-shaped-aggressiveness of SARS-CoV-2: Period between onset of nonspecific-specific symptoms for COVID-19. A population-based cohort study

Background: Early identification of different COVID-19 clinical presentations may depict distinct pathophysiological mechanisms and guide management strategies. Objective: To determine the aggressiveness of SARS-CoV-2 using symptom progression in COVID-19 patients. Design: Historic cohort study of Mexican patients. Data from January-April 2020 were provided by the Health Ministry. Setting: Population-based. Patients registered in the Epidemiologic Surveillance System in Mexico. Participants: Subjects who sought medical attention for suspicion of COVID-19. All patients were subjected to RT-PCR testing for SARS-CoV-2. Measurements: We measured the period between onset of nonspecific to specific symptoms for COVID-19 (PONSS) and compared it to the primary outcomes (mortality and pneumonia). Results: 65,500 patients were included. Reported fatalities and pneumonia were 2176 (3.32%), and 11568 (17.66%), respectively. According to the PONSS, patients were distributed as follows: 14.89% in <24 hours, 43.25% between 1-3 days, 31.87% between 4-7 days and 9.97% >7 days. The distribution for mortality and pneumonia was 5.2% and 22.5% in <24 hours, 2.5% and 14% between 1-3 days, 3.6% and 19.5% between 4-7 days, 4.1% and 20.6% >7 days, respectively (p<0.001). Adjusted-risk of mortality was (OR [95% CI], p-value): <24 hours= 1.75 [1.55-1.98], p<0.001; 1-3 days= 1 (reference value); 4-7 days= 1.53 [1.37-1.70], p<0.001; >7 days= 1.67 [1.44-1.94], p<0.001. For pneumonia: <24 hours= 1.49 [1.39-1.58], p<0.001; 1-3 days= 1; 4-7 days= 1.48 [1.41-1.56], p<0.001; >7 days= 1.57 [1.46-1.69], p<0.001. Limitations: Using a database fed by large numbers of people carries the risk of data inaccuracy. However, this imprecision is expected to be random and data are consistent with previous studies. Conclusion: The PONSS shows a U-shaped SARS-CoV-2 aggressiveness pattern. Further studies are needed to corroborate the time-related pathophysiology behind these findings.


49
Coronaviruses are single-stranded RNA organisms capable of infecting humans and other animal species (1,2). The 50 most recently discovered coronavirus, Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), is cause 51 of the clinical entity denominated Coronavirus Disease 2019 . This virus initially spread in the Wuhan 52 province in China and later to the rest of the world, causing a pandemic (3). Reported worldwide cases are 53 continuously growing and currently (as of July 3 rd , 2020) there are over 10 million infected people confirmed and 54 over 500,000 fatalities. Global reports reveal case-fatality rate of 4.8% and more than half of the cases are in the 55 Americas region. In Mexico, over 230,000 cases have been reported, with over 28,000 fatalities and a case-fatality 56 rate of 12.3%, which surpasses by far the global estimate (4).

57
Every human in the world is susceptible to infection, for as the mean age of infected patients is 47 years, 87% of 58 patients lie between 30 and 79 years old. COVID-19 behaves more aggressively in older patients and in patients 59 undergoing chronic medical conditions such as obesity, diabetes, hypertension and other cardiovascular diseases, 60 increasing the risk of mortality in these populations (5,6). Approximately 80% of cases are asymptomatic with a 61 mild disease course, while the other 20% can be accompanied of severe complications such as pneumonia, Acute 62 Respiratory Distress Syndrome (ARDS) and other secondary infections. Among these severe cases, 80% correspond 63 to people over 60 years. Many of these cases can be attributed to a severe clinical entity known as "cytokine 64 storm", which causes a rise in serum levels of many pro-inflammatory mediators and provokes massive tissue 65 damage in several vital organs (5,7).

66
In patients who developed severe symptoms, dyspnea was reported between 8-12 days after onset of symptoms, and 67 some patients deteriorate into severe disease during the first week after onset of symptoms. This accelerated 68 worsening has been hypothesized to be caused by the cytokine storm and to thrombotic events that may be caused 69 by infection with SARS-CoV-2 (8).

70
Hospitalized patients have been thoroughly described and analyzed, with an average time between onset of 71 symptoms to intubation of 14.5 days, and a time from intubation to death ranging from 4-5 days (5,7,9). A longer 72 period between onset of symptoms and first contact seeking medical attention has been associated with a poorer 73 outcome in these patients. However no in-depth studies have been conducted (10).

74
Until now, studies have been focused on patient-centered risk factors, while SARS-CoV-2 aggressiveness has been 75 stablished as provoking 20% of severe and critic patients, however, there are still many unanswered questions 76 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted November 3, 2020.

104
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted November 3, 2020. ; https://doi.org/10.1101/2020.10.28.20221697 doi: medRxiv preprint To establish the date of symptoms specific to COVID-19, we assumed that the population followed the 105 recommendations of the government to seek medical attention only when presenting COVID-19 specific-symptoms.

106
These recommendations were: Stay-at-home orders unless two of these three symptoms were present: 1) cough, 2) 107 fever or 3) headache, plus one or more of the following: 1) breathing difficulty, 2) sore or burning throat, 3) runny 108 nose, 4) red eyes, 5) pain in muscles or joints, or 6) being part of these high-risk groups: pregnancy, <5 or ≥60 years 109 old, or having a chronic disease such as hypertension, diabetes mellitus, cancer or HIV. These indications were 110 broadcast by television, radio, newspapers and internet since the beginning of the pandemic until July 1, 2020; up to 111 this date the government issued nationwide stay-at-home orders and only essential activities were permitted. 112 Therefore, the period between onset of nonspecific to specific symptoms for COVID-19 (PONSS) was calculated 113 between the date of onset of symptoms and the date the patients sought medical attention. Initially, PONSS was 114 categorized in days (<1, 1, 2, 3, etc.), but to improve comprehension and data management, adjacent days whose 115 frequency of death remained in similar proportions, were grouped into 4 categories (<24 hours, 1-3 days, 4-7 days 116 and >7 days). The primary outcomes were mortality and pneumonia. The presence of pneumonia was used as an 117 indicator of severe disease as reported in previous studies (5,7).

118
Missing data were substituted using the mode for the following categorical variables: Health Sector (338 patients,  Finally, the four categories of PONSS against primary outcomes -mortality and pneumonia-, were compared using a 131 multivariable logistic regression model. The model was adjusted in five steps for the following variables: age, 132 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted November 3, 2020.

146
The study population included 65,500 patients. Among them, the average age was 41±17 years, 50.2%, were 147 women, 55.8% belonged to a high socioeconomic level, 27.7% to a medium and 16.5% to a low one, 4.6% of 148 patients were treated on a private health institution, 37.7% in a facility for patients with social security and 57.7% 149 attended to a public hospital for patients without social security. Of all the patients, 41% had at least one 150 comorbidity, hypertension being the most frequent in 17%, followed by obesity in 15.6% and diabetes 12. 8%

160
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted November 3, 2020. ; https://doi.org/10.1101/2020.10.28.20221697 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted November 3, 2020.  Table 3 and Figure 1 show the risks for mortality and pneumonia related to PONSS. A "U-shaped distribution" was 178 observed according to PONSS (<24 hrs., followed by 1-3 days, 4-7, and >7 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted November 3, 2020.

194
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review)
The copyright holder for this preprint this version posted November 3, 2020.

200
In this study, we found an association concerning the period between onset of nonspecific to specific symptoms for  CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted November 3, 2020. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted November 3, 2020. ; https://doi.org/10.1101/2020.10.28.20221697 doi: medRxiv preprint Supplementary Material Section Captions 302 S1 Table. Stepwise analyses for binary logistic regression with Mortality as outcome for all patients in the study. 303 S2 Table. Stepwise analyses for binary logistic regression with Pneumonia as outcome for all patients in the study.

304
. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted November 3, 2020. ; https://doi.org/10.1101/2020.10.28.20221697 doi: medRxiv preprint