This retrospective cohort study included patients admitted to the Centre Hospitalier Régional Lomé Commune (CHR-LC) in Togo from June 2021 to May 2022.

The CHR-LC is a public hospital that served as the national reference center for the exclusive care of patients with COVID-19 since the outbreak of the epidemic in Togo in March 2020 [13] . At the time of the study, this center had a capacity of 114 beds divided into 4 units: an intensive care and resuscitation unit (30 beds), a medical hospitalization unit (36 beds), a hemodialysis unit (2 beds), and an extension area for asymptomatic or convalescent cases (46 beds). Regarding the protocol of management, patients with critical and severe COVID-19 were admitted to the ICU. Moreover, in the absence of a high-dependency unit in our context, some patients with moderate conditions, but requiring continuous monitoring and care, due to a risk of rapid worsening, were also admitted to the ICU. The mild and moderate forms were admitted to general wards and managed with symptomatic treatment, such as paracetamol, and vitamin C, along with chloroquine and azithromycin. Patients with severe forms received standard oxygen, either by nasal cannula or by mask. Severe patients with suspected bacterial infection and critical patients received empirical antibiotics, as well as corticosteroids (dexamethasone), and antithrombotic prophylaxis (enoxaparin). Critical patients received specific management according to the complications, such as mechanical ventilation consisting of non-invasive ventilation or intubation for severe acute respiratory distress syndrome (ARDS), curative anticoagulation for pulmonary embolism, hemodialysis for acute renal impairment, fluids and vasopressors therapy for septic shock.

All patients aged 15 years and above, having a positive SARS-COV-2 rT-PCR test and who were hospitalized at the CHR-LC during the study period were eligible. Patients with incomplete medical records and those with unknown vaccination status were excluded.

We defined the severity of COVID-19 according to WHO clinical categorization [14] .

· Mild: patients with mild symptoms without signs of pneumonia or hypoxia.

· Moderate: patients with pneumonia and moderate hypoxia with SpO2 > 90% without signs of respiratory distress were tagged.

· Severe: patients with signs of severe pneumonia and hypoxia, with SpO2 < 90% requiring an oxygen flow greater than 10 liters per minute.

· Critical: patients with acute respiratory distress syndrome (ARDS), sepsis, shock, or other complications.

· Respiratory failure (distress) was defined as dyspnea with a respiratory rate above 22 cycles per minute, struggle signs, and hypoxia,

· Renal failure was defined as having a serum creatinine clearance of less than 30 ml/min/1.73 m² or other conditions requiring haemodialysis.

· Fully vaccinated: patient who received at least two doses of any type of COVID-19 vaccine, the last one being administrated at least 15 days before admission

· Unvaccinated: patient who did not receive any dose of COVID-19 vaccine, or who received only one dose or who received the second dose within 15 days before admission, irrespective of type of vaccine.

Data were collected using a data extraction form. Medical records and the electronic database of patients hospitalized at the CHR-LC were used as the basis for extraction. We collected, sociodemographic data (age, sex, profession, nationality), COVID-19 vaccination status based on patient’s declaration, number of doses, presence of co-morbidities (arterial hypertension, diabetes, HIV infection, asthma, cardiopathy, bronchopneumopathy, etc.), clinical characteristics of COVID-19 (severity, unit of hospitalization), length of hospitalization and outcome of hospitalization (death or discharge).

The primary outcomes were all-cause mortality and admission to ICU. The first day of hospitalization was considered as the start of follow-up and the date of discharge or death as the end of follow-up. Socio-demographic and clinical characteristics were used as explanatory variables to describe the factors associated with death during hospitalization.

A descriptive analysis was carried out. Categorical variables were presented as frequency. Quantitative variables were presented as medians with their interquartile range (IQR). Comparisons of medians were made by the Wilcoxon-Mann-Whitney test, and proportions were compared using the chi-square test or Fisher’s exact test. The 15-day and 30-day mortality rates were presented. To describe the factors associated with inpatient death, a Cox proportional hazard regression was performed. A multivariable model was constructed with the variables having a p-value < 0.20 in the univariable model. The results were presented as instantaneous hazard ratios (HR) with a 95% confidence interval (95%CI). All analyses were performed using R® software version 4.2.1. The significance level was set at 5%.

The study was authorized by the Direction of the CHR-LC and was approved by the “Comité de Bioéthique pour la Recherche en Santé (CBRS)” (Bioethics Committee for Health Research) from the Togo Ministry of Health (014/2022/CBRS of May 24, 2022). Confidentiality was ensured through the anonymity of patients’ data. Compliance with the rules of good research practice was also assured. Patient consent was not required as it was a retrospective study.

During the study period, 610 patients were hospitalized. Of these, 6 were excluded from the analysis for being under 15 years of age (n = 2) and for unknown vaccination status (n = 4). Finally, 604 patients were included (Figure 1).

The median age of patients was 54 years (IQR: 41 - 68) with a sex ratio of 1:1. Hypertension and diabetes were the two main comorbidities observed in hospitalized patients (40.8% and 28.2%, respectively) (Table 1).

Overall COVID-19 vaccination coverage was 9.1% (55/604). The proportion of vaccinated patients varied according to socio-demographic profiles and comorbidities. Vaccination coverage was significantly lower among women than men (6.0% vs. 12.3%, p = 0.007) but was not different across age groups and comorbidities (Table 1).

¹Median (25%-75%); n (%). ²Wilcoxon rank sum test; Fisher’s exact test; Pearson’s Chi- squared test; ³IQR: interquartile range. *At least two doses of vaccination against COVID-19.

Overall, 41.1% (248/604) and 12.1% (73/604) of patients were admitted at the severe or critical stage (Table 2, Table S1). The proportion of severe forms was significantly lower in fully vaccinated patients than in unvaccinated ones (30.9% vs. 42.1%, p < 0.001). Similarly, none of fully vaccinated patient was admitted with a critical stage of COVID-19 compared to unvaccinated ones (0% vs. 13.2%, p < 0.001).

During the hospitalization, 10.9% (66/604) of patients experienced one or more complications. Respiratory and renal failure were the most common complications, observed in a proportion of 3.6% (22/604) for both. None of these complications was significantly associated with the vaccination status as shown in Table 2.

Around 3 out of 5 patients were admitted to the intensive care unit. ICU admission was significantly more frequent in unvaccinated patients than in fully vaccinated ones (63.0% vs. 38.2%; p < 0.001). Table 2 shows the clinical characteristics of hospitalized patients according to vaccination status.

When considering the whole sample, median hospital stay was 10 days (IQR: 6 -

¹n (%). ²Pearson’s Chi-squared test; Fisher’s exact test.

15), and did not vary according to patients’ vaccination status (p = 0.741).

Among those patients who were discharged, hospitalization was significantly shorter in fully vaccinated patients (median 10 vs. 13 days, p = 0.012). The duration of hospital stay was not different according to vaccination status among deceased patients as death occurred within the first week of hospitalization in both groups (Figure 2, Table S2).

Overall, 172 patients died, representing an in-hospital mortality rate of 28.5% (Table 2). The 15 and 30-day mortality rates were 26.1% (158/604) and 28.3% (171/604) respectively. Table 3 presents the fifteen and thirty-day mortality rates in patients hospitalized.

Mortality was significantly lower in fully vaccinated patients than in unvaccinated

¹Median (25%-75%); n (%); ²Wilcoxon rank sum test; Fisher’s exact test; Pearson’s Chi-squared test.

ones (9.1% vs. 30.4%; p < 0.001) (Table 2). The same trend was observed for 15 days’ mortality rate (9.1% vs. 27.9%; p = 0.003) and 30 days’ mortality rate (9.1% vs. 30.2%; p < 0.001).

The instantaneous risk of death at any time after hospital admission was 73.0% lower (adjusted HR (aHR): 0.27 (0.11 - 0.67)) in fully vaccinated patients, independently of other risk factors. In contrast, age equal to or greater than 55 years (aHR: 1.82; 95% CI: 1.32 - 2.52), diabetes (aHR: 1.60; 95% CI: 1.17 - 2.19), and renal failure (aHR: 3.94; 95% CI: 1.73 - 8.98) were significantly associated with an increased risk of mortality (Figure 3).