A Comprehensive Analysis of COVID-19 Impact in Latin America

Background: Latin America has now become the epicenter of the global coronavirus disease 2019 (COVID-19) pandemic. In the ongoing COVID -19 pandemic, a profound burden of SARS-COV-2 infection has been reported in Latin America. In the present study, we aim to determine the profiles that are associated with this disease in Latin America. We analyzed symptoms, morbidities and gastrointestinal (GI) manifestations by country. Methods: We analyzed data from SARS-CoV-2 positive patients evaluated at healthcare centers and hospitals of 8 Latin American countries including Brazil, Peru, Mexico, Argentina, Colombia, Venezuela, Ecuador, and Bolivia between March 1 and July 30, 2020. These countries consist of a total population that exceeds 519 million. Demographics, comorbidities and clinical symptoms were collected. Statistical descriptive analysis and correlation analyses of symptoms, comorbidities and lethality were performed. Results: A total of 728,282 patients tested positive for COVID-19 across all the 8 Latin American countries. Of these, 52.6% were female. The average age was 48.4 years. Peru had the oldest cohort with 56.8 years old and highest rate of females (56.8%) while Chile had the youngest cohort (39 years old). Venezuela had the highest male prevalence (56.7%). Most common symptoms were cough with 60.1% (Bolivia had the highest rate 78%), fatigue/tiredness with 52.0%, sore throat with 50.3%, and fever with 44.2%. Bolivia had fever as the top symptom (83.3%). GI symptoms including diarrhea (highest in Mexico with 22.9%), nausea, vomiting, and abdominal pain were not associated with higher mortality. Hypertension was among the top (12.1%) comorbidities followed by diabetes with 8.3% and obesity 4.5%. In multivariable analyses, the leading and significant comorbidities were hypertension (r=0.83, p=0.02), diabetes (r=0.91, p=0.01), and obesity (r=0.86, p=0.03). Asthma (r=0.37, p=0.54) and increasing age (0.13 p=0.81) were not independently associated with higher mortality. Lethality was highest in Mexico (16.6%) and lowest in Venezuela (0.9%) among the analyzed cohorts. Conclusion: Nearly, 10.5%–53% of patients with COVID-19 have GI manifestations. Differential clinical symptoms were associated with COVID-19 in Latin America countries. Metabolic syndrome components were the main comorbidities associated with poor outcome. Country-specific management and prevention plans are needed. Country-specific management and prevention plans can be established from this meta-analysis.

In this study, we described the clinical characteristics, lethality, and symptoms of the con rmed COVID-19 cases o cially registered by countries that make pertinent information publicly available. Special emphasis was made on gastrointestinal symptoms and comorbidities as they relate to lethality of the disease in Latin America.

Search Strategy and Selection Criteria
We conducted a systematic literature search of published articles using electronic databases such as PubMed, OVID, Scopus, Google Scholar, LANCOVID (Latin America research network on COVID), and other resources from o cial health organizations of countries in Latin America such as Ministries of Health, National Institutes of Hygiene, or Hospitals from January 1 through July 30, 2020. We included the following terms in the search bar: COVID-19, Argentina, Brazil, Chile, Colombia, Ecuador, Mexico, Peru, Venezuela, Uruguay, Paraguay, Hispanic and Latin America or South America. Additionally, we searched for o cial and non-o cial press releases of patient and hospital data from government health institutions of the leading COVID-19 virus-infected Latin American countries. We searched leading reference databases, primary published sources and press releases from the various Ministries of Health of the respective Latin American countries. Case characteristics were described, including demographics, exposures, comorbidities and symptoms.
Selection and Identi cation of Relevant Literature: Using the listed inclusion and exclusion criteria, we rst sorted the Latin American COVID-19 studies by title and abstract; then we compiled the papers by relevance and conducted a new selection process by a thorough review of the data. Based on detailed insights derived from the most relevant papers, we revised our reference search criteria to obtain more re ned papers for our systematic review. We incorporated studies that reported patients' characteristics and symptoms of interest. From the selected papers, tables were generated for each dataset on Microsoft Excel. These tables included the following information for extracted data of each study: Study author (year, location, hospital or city, state and country), date of the report, location, con rmed cases, deaths, lethality rate, average age, sex, headache (cephalea), cough, myalgias, fever, odynophagia, anosmia, dyspnea, ageusia, diarrhea, thoracic pain, abdominal pain, tachypnea, cyanosis, sore throat, nasal congestion, nausea and vomiting, fatigue or tiredness, joint pain, hypertension, diabetes mellitus, asthma, obesity, cardiovascular disease, chronic cardiopathy, chronic obstructive pulmonary disease, chronic kidney disease, chronic neurological disease, cancer, in ammatory bowel disease, and tuberculosis ( Fig. 1).
Inclusion Criteria: The following inclusion criteria were adopted to validate article selection:

Demographic and clinical characteristics
There were 728,282 con rmed cases in our study from 8 countries (  Cough and fatigue were the top symptoms in Latin American COVID-19 patients According to the combined overall weighted average, the most common symptoms reported in the 8 Latin American countries for positive cases of SARS-CoV-2 infection were cough with 60.1%, fatigue or tiredness with 52.0%, sore throat with 50.3%, and fever with 44.2% (Table 1). Bolivia was the country that reported the highest positivity for cough ( Fig. 2A) 78.8% and fever (Fig. 2B); at least 12% more than Mexico in second place. Ecuador reported the highest prevalence for fatigue followed by Peru and Bolivia with 52% and 51.5%, respectively (Fig. 2C).
Fever was not one of the main symptoms in Latin American COVID-19 patients While fever is probably one of the most common symptoms for COVID-19, that was not the case for the population we studied in Latin America. Fever was the 4th most common symptom with a mean of 44% and 3rd was Sore Throat (50.3%), dry cough and fatigue (Table 1). In our report, Bolivia was the region that reported 83.3% of fever, followed by Mexico with 76.6% and then at a much lower rate Peru with 49.4% (Fig. 2B).

Gastrointestinal manifestations were highest in Mexico among Latin America COVID-19 patients
Our cohort of patients reported a different array of gastrointestinal symptoms including diarrhea, abdominal pain, nausea, and vomiting. Diarrhea was the most reported symptom being present in 11.5% of patients (Table 3). There was no characterization of the type of diarrhea, whether or not it contained blood; mucus or pus. It is important to mention that only Argentina, Chile, Mexico and Peru (Fig. 3) reported diarrhea on their symptoms and Mexico was the country with the highest prevalence (22.9%), followed by Peru with 13.6%. Diarrhea was followed by nausea and vomiting in 7.2% of patients and lastly abdominal pain in 4.3% in the whole group analysis ( Table 2). As for gastrointestinal diseases, the prevalence of chronic hepatic disease was 0.31% in positive cases and as such the least common comorbidity ( Table 2). Hypertension was among the top comorbidities in Latin America COVID-19 patients Among comorbidities, the reports showed that hypertension was the most common with an average of 12.1% followed by diabetes with 8.3% and obesity with 4.5% (Table 2). Overall, 12.1% of the cohort had at least one comorbidity. It's also important to note that patients can have multiple symptoms and comorbidities. Differences for comorbidities were present among different countries. Mexico was the country with the highest prevalence of hypertension with almost 28.8% followed by Chile with 16% ( Fig. 4A) and for diabetes as well (Fig. 4C). Mexico reported the highest prevalence with 20.6% followed by Chile with 8%. As for obesity (Fig. 4B), Mexico had the highest prevalence with 25.6% followed by Brazil with a 23.7% of the total of their cohorts. As such, the cohort from Mexico was the most hypertensive, diabetic and obese of all 8 countries.

Lethality was highest in Mexico among Latin American countries
We analyzed the lethality rate for the 8 countries, and there are some important differences between them. Lethality is de ned as the number of deaths as end result of COVID-19 over number of positive COVID-19 (x100) (also known as Case Fatality Rate) in the population. The overall weighted average for the total population of our study (Table 1 Suppl.) showed that the lethality rate was 3.4%. In relation to the strati cation of the lethality rate per country (Fig. 8), there were mixed results, with Mexico reporting a lethality of 16.6%, followed by Brazil with 7.6% and then Bolivia with 5.6%. Venezuela reported the lowest lethality rate with 0.9%.
The variables result related to lethality rate (Fig. 5) showed that Mexico reported the highest rate with 16.7%, followed by Brazil with 7.6% and Bolivia with 5.6%. Venezuela reported the lowest lethality rate with just 0.95%. Overall, there were several differences in the reported lethality rates.

Association of lethality and comorbidities in COVID-19 patients
We explored the association of lethality rate as a dependent variable in univariate and multivariate analyses with several independent variables such as obesity ( Table 2 Suppl), hypertension, diabetes, and asthma ( Table 2 Suppl.). Each of these analyses was statistically signi cant except for asthma (Table 2 Suppl.). This data shows that hypertension, diabetes and obesity are associated with lethality in COVID-19 patients in Latin America. A multiple regression analysis shows that the most in uential factor is obesity followed by hypertension and diabetes.

Lethality and GI manifestations in COVID-19 patients in Latin America
We explored the association of lethality rate as a dependent variable in univariate and multivariate analyses with several independent variables such as GI symptoms. None of the symptoms including diarrhea, abdominal pain, nausea/vomiting individually or collectively was associated with lethality in Latin America.

COVID-19 is a consequence of infection with the SARS-CoV-2 virus. It is well known that previous
Coronavirus infections attack mainly the respiratory system [15]. Probably, because of the con ned geographical nature and short-lived SARS-CoV-1 and MERS epidemics, the medical and scienti c community did not explore other potential and secondary targets of coronaviruses. In this regard, the expansion of the current pandemic clearly showed that SARS-CoV-2 can target other systems that display ACE2 receptor. This receptor is expressed in the gastrointestinal (GI) system which makes it a potential target for the virus and can lead to GI symptoms [16]. Indeed, viral RNA particles have been detected in stool samples which raised another issue that pertains to fecal-oral transmission, be it in con ned clinical and domestic settings or through sewage systems [16]. However, we do not know the prevalence and frequency of SARS-CoV-2-associated gastrointestinal manifestations in Latin America and the virusassociated symptoms and outcomes, especially in the presence of comorbidities.
There are few reports or publications made by health institutions or investigators from Latin America. Nevertheless, at this time that we are writing this paper, one of the epicenter of the disease is located in this region of the world. Taking into consideration some factors such as di cult access to healthcare in several regions of Latin America, the high index of poverty, the low availability of clean water, and waste management that further exacerbate the pandemic's outcome [17], it would be prudent to study the most common symptoms and comorbidities to distinguish the high risk categories in order to utilize the scarce resources intelligently and wisely in every day practice in a country/population-speci c manner. Here we report the results of our systematic analysis using the published and available data to date, and demonstrate that gender distribution, age range, symptoms and comorbidities are different among the 8 Latin American cohorts. Hypertension and obesity were the predominant comorbidities and diarrhea the top GI symptom.
Our study revealed that COVID-19 manifests differently in Latin America among the various regions.
These differences are important to consider since their impact can in uence management, treatment and prevention approaches for better outcomes when addressing the infection. In our study, the most common symptoms in Latin America are dry cough, fatigue, sore throat and fever. Also, the most common comorbidities are obesity, hypertension and diabetes [18,19]. This contrasts with what was reported by Maechler et al, from a Berlin testing center that indicates that the most common symptoms were fatigue, myalgia and cough while top comorbidities were chronic lung disease and chronic heart disease [14]. These discrepancies could emanate from the methodologies or sources of data collection in the two studies. Furthermore, our data are more recent and include larger cohorts. The conclusive evidence gathered in our systematic review can make a signi cant impact on screening, tagging, and separating patients into suspected and hospitalization candidate cases that are amenable for follow-ups such as contact-tracing. It is worth noting that these symptoms and comorbidities did not distribute similarly among the eight studied nations. Mexico and Chile followed a similar pro le as far as comorbidities are concerned.
There is a special mention about GI symptoms in our study. The prevalence of diarrhea was much higher than reported in other investigations [20][21][22]. This is probably because a higher prevalence of diarrhea in Latin American populations exists prior to SARS-CoV-2 outbreak, due to the consumption of contaminated foods [23]. In our study, we examined the clinical aspects of the disease and also compared the different demographics from every population and analyzed their association with lethality, incidence and mortality. According to our research, the cumulative lethality in Latin America is 3.4%, very close to the one reported worldwide based on the current data published by John Hopkins University [7]. We also understand that the current lethality rate (a.k.a. case fatality rate) may be in ated compared to the actual risk of death because the people most likely to get tested for COVID-19 are those with severe symptoms and not those with mild symptoms or asymptomatic patients. This rate might also be de ated as many COVID-19 related deaths in Latin American countries are very likely under-reported.
Regarding the demographics of our population, there is a notable difference between sexes, females being the most affected. This is different than what Conti et al. reported on their study where males were the primary target; which they explained as an indication that women were less susceptible to viral infections due to a difference in innate immunity, steroid hormones and the presence of two X chromosomes and the immune regulatory genes associated with chromosome X [24]. Here again, we must note that underlying factors such as age, sex and pre-existing health conditions make the individual risk different from the overall risk. This is important to highlight because while certain reports of case fatality rates include some of these demographics, these numbers also have the same context-based biases as the overall lethality or case fatality rates discussed above.
Since lethality is the main outcome to avoid in this stage of the pandemic, we performed different analyses in order to compare the impact of the top comorbidities on lethality. We found that obesity, hypertension and diabetes were highly correlated with lethality, and that the presence of these conditions exacerbates the severity and outcome of COVID-19 infection. This is similar to what was reported by other studies [18,19].
Another pertinent element to highlight, is that the prevalence of tuberculosis in our study was above 1.2%.
There is a well-known high prevalence of tuberculosis in the region and the usage of BCG vaccination in Latin America. Many studies reported a potential effect of the BCG vaccine in providing heterologous protection [25] against other infections like the one from SARS-CoV-2. While the BCG vaccine role is still debated, future large and well-designed comparative studies might reveal whether it played any role in the pandemic's outcome in Latin America.
Hospital/healthcare structures and some of the observed differences between countries have to do with pandemic management and data availability. Indeed, Venezuela for example showed the least lethality, which might be explained by free access to healthcare system and the fact that the treatment protocol includes hospitalizing every positive case reported [26]. This is in contrast to other countries such as Mexico where only severe disease patients are hospitalized and less access to advanced care is available for the general population [27]. In addition, the lockdown strategies differed between countries which affected the dynamics of transmission and infection rates. The availability of medical records of the patients is very critical for the precise interpretation of the data. Some publications like the one from Chile presented complete clinical data, while the same detailed data was not available for the other regions since most of them use manual medical records and managing such data is di cult.
A limitation of our study is its retrospective nature. Some of our reports are from several months ago when the pandemic started, also we only listed studies that were available as published reports (Table 1) or on the web. Many of the health authorities in these countries have limited data sharing capability to provide solid recommendations to the healthcare sector on how to handle the pandemic. Furthermore, some of them reported merely symptoms, others only comorbidities. Therefore, there were disparities in the way they reported patients' information. There are many variables such as immune reaction and blood test, imaging, treatment, liver enzymes and other underlying diseases that are very important for the analysis but are missing in the analyzed data. There is a need for standard collection of data for such a massive and lethal pandemic. For global standardizations, organizations such as WHO or PAHO, can recommend/provide useful tools for both electronic and manual medical records in order to collect precise data for dealing with either endemic or pandemic infections. It is relevant to comprehensively study the clinical conditions of the pandemic so each country can evaluate the most common trends in their population and get better management and treatment protocols.
GI and liver manifestations such as anorexia, dysgeusia, nausea, vomiting, diarrhea, abdominal pain, and discomfort are common among patients with COVID-19. Hepatic involvement in this condition is re ected by abnormal liver enzymes and much less frequently elevated bilirubin values. Few studies [28,29] showed that patients with the severe disease more often have GI symptoms and hepatic involvement.
While the mechanisms of GI manifestations elicited by SARS-CoV-2 are still unknown, it is clear that ACE2 expression in the esophagus, small and large intestines, plays a role. Indeed, ACE2 was reported to be coexpressed with TMPRSS2 that is needed for the cleavage of the spike protein and the facilitation of the virus entry into the host GI system aggravating systemic effects of the virus and elevating the rates of transmission. While millions of patients have been discharged and tested negative in their nasopharyngeal specimens, a sizable portion of them still tested positive in their stools [30]. This group of patients should be the focus of next stringent monitoring methods as they are likely to play a role in the dissemination and recurrence of the disease. It will also be worth monitoring whether SARS CoV-2 persistence in the GI system will impact overall GI system functions.
In conclusion, while the pandemic is still raging and vaccines and therapeutics are still in development, the only direct interventions readily available now are modi cations of human behavior measures such as hygiene, social distancing and mask-wearing to limit exposure and future infections. Since this disease is here to stay in the foreseeable future, strict measures need to be taken in Latin American countries to act on comorbidities such as obesity, hypertension and diabetes, all part of the metabolic syndrome. Furthermore, public health interventions to modify these comorbidities will have a positive impact on COVID-19 overall outcome in Latin America. Comparison of (%) mean A) Cough, B) Fever symptom, and C) Fatigue and tiredness, in selected studies from different countries.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.