Coronavirus Pandemic High mortality at the beginning of the COVID-19 pandemic in a referral center in the metropolitan area of Mexico City

Introduction: There is limited information about the coronavirus disease 2019 (COVID-19) disease in Latin-American countries. Our objective was to describe the clinical characteristics and outcomes of COVID-19 patients in Mexico. Methodology: We conducted a retrospective cohort study with 333 consecutive patients who were admitted to Hospital de Especialidades "Dr. Antonio Fraga Mouret" in Mexico City with COVID-19 between April 1, 2020, and June 30, 2020. Demographic, clinical, laboratory data, treatment details and 30-day outcomes were analyzed. Results: The patients studied included 52% men (172/233) and the median age was 45 years. Up to 75% (250/333) of patients were classified as overweight or obese. There were 185 (56%) inpatients; 85% (158/185) were hospitalized in the general ward, and 15% (27/185) in the Intensive Care Unit (ICU). Laboratory measurements showed significant differences between inpatients and outpatients such as lymphocyte-count (median 0.8 vs 1.2×10 9 /L, p < 0.001), LDH (median 650 vs 294 U/L, p < 0.001), CRP (median 147 vs 5 mg/L, p = 0.007), CK-MB (median, 15 vs 10 U/L, p = 0.008), ferritin (median, 860 vs 392 ng/mL, p = 0.02), and D-dimer (median, 780 vs 600 ng/mL, p = 0.15). These differences were seen between survivor and non-survivor patients as well. The rate of death in mechanically ventilated patients was 94% (67/71). Mortality at 30-day follow-up was 57% (105/185). Conclusions: We observed that majority of the non-survivors were obese and young. Complications leading to death was observed in majority of the cases.


Introduction
The twenty-first century witnessed two global health emergencies involving respiratory pathogens that changed the world. The coronavirus disease 2019 (COVID-19) spread through Asia in late 2019 and spread to Latin America in the begining of 2020. So far there are more than 219 million confirmed cases and over 4.55 million deaths worldwide. The first confirmed case in México was reported on February 28, 2020. Since then there have been more than 275,000 deaths [1]. Mexico currently faces an unprecedented healthcare crisis that has challenged public health policies. During the first quarter of the year, data from cohorts around the globe identified several epidemiological characteristics in patients diagnosed with COVID-19 who required inpatient treatment, especially those who became critically ill [2]. Multiple risk factors such as diabetes, hypertension, cardiovascular disease were identified [3]. In addition, being overweight and obese were associated with a more severe presentation of the disease in Mexico [4]. The prevalence of these chronic diseases may play a decisive role since they are medical issues associated with a more complex sociocultural phenomenon. These sociocultural factors, along with the lack of medical supplies and a collapsed healthcare system coexist in a pandemic and cannot be assessed separately [5]. The clinical presentation of COVID-19 is heterogeneous, and can range from the well-recognized devastating pulmonary damage and multi-organ failure to minor gastrointestinal, neurological, and even dermatological features. These symptoms are still poorly understood. Therefore, the analysis of local epidemiological and clinical data is essential for understanding COVID-19 in our population. We aimed to describe the epidemiological data, clinical characteristics and outcomes of COVID-19 patients treated in a tertiary referral center in the metropolitan area of Mexico City.

Methodology
This retrospective observational study was performed at the Hospital de Especialidades "Dr. Antonio Fraga Mouret", Instituto Mexicano del Seguro Social in México City. This is a Tertiary Referral Center and is a nine-floor hospital with 550 beds: 524 in the general ward and 26 Intensive care Unit (ICU) beds.
The hospital was converted into a hybrid hospital that received patients who required specialized attention, including COVID-19 suspected and confirmed cases. The beds in the wards were redistributed due to the pandemic to consist of two isolated floors of 70 beds each for suspected or confirmed patients and 12 ICU beds exclusively reserved for COVID-19 patients who required intensive care.
We followed the World-Health Organization (WHO) definitions for inclusion criteria [6]. Patients were divided into four groups: mild, moderate, severe, and critically ill. Mild patients were treated ambulatory, while moderate patients were treated depending on the medical criteria based on assessment.
Confirmation criteria were defined as a positive result of real time Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) assay of nasopharyngeal or lower respiratory tract swabs collected by a healthcare provider following the Center for Disease Control and Prevention (CDC) recommendations [7].

Data collection
Patients with confirmed COVID-19 admitted to the hospital between April 1, 2020, and May 30, 2020 were included in our study; we included subjects who had available outcomes and their 30-day follow-up assessment. The date of the final follow-up was June 30, 2020. Data was collected from the electronic health records from all patients. The study population was dichotomized into ambulatory and hospitalized groups. The latter group of patients was divided according to hospitalization in the general ward or ICU.
We assessed demographic and epidemiological data, medical history, radiological characteristics, and laboratory values (at admission and 72 hours after) for all patients. We calculated severity scores and scales commonly used in the COVID-19 pandemic context such as qSOFA, NEWS2, CALL Score, CURB-65 [8][9][10][11] for all enrolled subjects. In-hospital evolution, management, complications, and outcomes were captured. Both outpatients and inpatients had a 30-day follow-up. This study was reviewed and approved by the hospital's ethics committee (Registry number R-2020-3501-104).

Statistical Analysis
The demographic data, clinical characteristics, and laboratory measurements were analyzed with descriptive statistics. Comparisons among survivors and non-survivor patients were performed with the Fisher's exact test or Chi square test for categorical variables when appropriate. The continuous variables were analyzed with Student t test or the Wilcoxon rank sum tests according to the data distribution. A p value < 0.05 was considered statistically significant. Analyses were performed using R Project for Statistical Computing; R Foundation, version 3.6.1.

Results
A total of 955 patients were tested for COVID-19, 415 of them were positive and 540 were negative. We also included 333 patients who tested positive in the final analysis and whose 30-day outcomes were available.

Ambulatory COVID-19 patients
Among the patients included in the study, 148 (45%) were ambulatory patients and were younger than the inpatient group. Median age of the ambulatory patients was 39 yrs whereas median age of the inpatients was 54 yrs (p < 0.001). The ambulatory patients included 57% (85/148) health care workers and their main comorbidities were type 2 diabetes (11%, 16/148) and hypertension (11%, 16/148); 7% (11/148) of them had both type 2 diabetes and hypertension. The CALL Score indicated high risk in only two patients (10 and 13 points, respectively), and the median CALL score of all ambulatory patients indicated low risk (5 points). Almost 99% of the patients (146/148) had a NEWS2 score of low risk with a median score of 2. The most common symptoms were fatigue (92%), myalgias/arthralgias (89%), cough (82%), and headache (79%). Baseline characteristics of the patients and outcomes are shown in Table 1.

Treatment and outcomes
Macrolides as COVID-19 therapy were administered to 23 (16%) of the ambulatory patients and 57% (106/185) inpatients. In addition, they received a high percentage of hydroxychloroquine 38% (71/185) and corticosteroids 16% (30/185). There were no significant differences between the survivors and non-survivors with regards to the treatments given with azithromycin, hydroxychloroquine, and corticosteroids (Table 3). Empiric antibiotic therapy was administered to 71% (132/185) of the inpatients, of whom 77% (102/132) received a third-generation cephalosporin. Prophylactic antithrombotic treatment (enoxaparin) was given to 79% (146/185) of the inpatients. Thrombosis as a complication occurred in 3% (11/185) of inpatients. There was no significant difference Table 2. Laboratory measurements at the admission of all the patients with COVID-19 classified by place of hospitalization.
The procalcitonin levels were significantly higher in nonsurvivor patients during admission; however, we did not document co-infections at the presentation.
Among the hospitalized patients, 135 patients (73%) presented with overweight or obesity and these were the most prevalent comorbidities in our population. Obesity has been described as the main comorbidity associated with COVID-19 in the Mexican population [17]. This is not surprising since in the last evaluation, ENSANUT 2018 [12], 75.2% of Mexican adults aged ≥ 20 yrs were overweight or obese. Overweight and obesity are related to a multifactorial chronic inflammatory state and immune dysregulation, a risk factor that would explain the high mortality in areas with a high prevalence of overweight and obesity [18].
Other significant comorbidities found were type 2, diabetes and hypertension, present in 28.5% and 29% of the hospitalized patients, which was similar to observations in other parts of the world [13][14]. Despite these being described as risk factors for mortality in other studies [19], we did not find a high proportion among the nonsurvivor patients. However, its prevalence was higher in patients who required hospitalization. Interestingly, Palaiodimos et al. [20] reported a multivariate analysis, including these factors, and the results indicated that obesity may be the underlying link between them.
Prognostic admission scores were significantly higher in nonsurvivors compared to survivors. This data suggests that subjects with higher scores could benefit from ICU management; however in Mexico and other Latin American countries this was not possible because of lack of ICU beds (especially at the beginning of the pandemic) [5].
The nonsurvivor patients presented significant laboratory findings such as marked lymphopenia, high LDH, D-dimer, creatine kinase MB, C-reactive protein, ferritin, procalcitonin, and neutrophil/lymphocyte ratio. These findings were similar to previous studies in which these markers have helped determine disease severity and mortality. Some markers such as D-Dimer, ferritin, and procalcitonin were not available in all parts of our country. However, markers such as lymphocytes, DHL or neutrophil/lymphocyte ratio are available in most hospitals and can be used as prognostic markers of the disease [19]. The main complications included ARDS (45%), AKI (30%), and all-cause shock (29%) and were significantly higher in our studies than in the reports from NY [13] and China [14].
Following internal emergency protocols, a high percentage (71%) of patients received empirical antimicrobial therapy, mainly including beta-lactam derivatives, even though we did not find co-infections at admission. Our secondary infection rate was lower (8%) than that reported in a systematic review (15%) [21]. Broad-spectrum antibiotics are commonly administered to COVID-19 patients; it has been reported that up to 75-100% of critically ill patients are administered antibiotics, despite low rate of secondary infections [22]. The disproportionate use of these will result in increase in antimicrobial resistance, mainly in the COVID-19 units, and the risk of novel resistance will play an important role that could be evaluated in the future.
In this study, we did not find significant differences in the treatments between survivor and nonsurvivor patients. Despite the efforts of global researchers, there is no specific treatment to date, although several clinical trials are in progress and the results are promising, and many vaccines are being developed that may help control the global pandemic [3].
The mortality at 30-day follow-up of all patients was 34%, higher than that reported from metropolitan Detroit, USA, that included ambulatory and inpatients and reported a mortality of 16% [23]. The rate of death among hospitalized patients was 54%; this is significantly higher than the cohorts that included hospitalized patients in NY (21%) [13], China (21.9%) [24], and Italy (20.6%) [25].
Patients with mechanical ventilation are challenging for non-intensive care physicians to care for these patients in the general ward. We reported a 94% death rate among patients with mechanical ventilation. This rate is similar to studies in other parts of the world such as New York (88%) [13], China (92%) [26], and US (93.5%) [27]. Studies have reported varied ranges of mortality that can be misinterpreted. Some studies focused on hospitalized patients, while others included all types of patients. Due to the rapid need for information, most cohorts included patients who had not completed outcomes. More specific information can only be objectively assessed over time.

Limitations
Our study has several limitations. First, this is a retrospective study; most data were acquired via electronic files or phone calls. Furthermore, we did not have a close follow-up in the case of ambulatory patients; not every subject had all laboratory or image studies; some patients were assessed with routine laboratories and a positive SARS-CoV-2 RT-PCR test and were isolated at home and followed remotely during the 30-day period. We did not make adjustments for multiple confounders, so it is impossible to obtain conclusions about factors associated with mortality.
In conclusión, this study shows the characteristics and 30-day outcomes of the first consecutive COVID-19 patients who visited a referral center in Mexico City, the epicenter of the pandemic in Mexico. A high proportion of patients had comorbidities such as obesity, diabetes and hypertension. Critically ill patients requiring mechanical ventilation in the general ward had a high death rate.