Coronavirus Disease 2019 (COVID-19) has caused a global health crisis with a cumulative incidence of more than 622.5 million cases and 6.6 million deaths reported by October 2022. Latin American countries have been among the most affected, with Colombia ranking fourth in confirmed cases after Brazil, Argentina and Mexico, with 6.3 million confirmed cases and 141,000 deaths.1

Differences in clinical outcomes of patients between waves of the COVID-19 pandemic have been reported previously.2,3 The emerged omicron variant has been shown to have high transmissibility but not high virulence, resulting in lower hospitalization and mortality rates in population-based studies.4

To date, there is no widely accepted definition of a COVID-19 pandemic wave.5 Yet, a distinguishing feature that should be preserved throughout time is a rising or declining number of cases. Although the waves do not occur at the same time in all countries, the following waves are more severe and result in more deaths, especially in nations where immunization is not widely available.6

The COVID-19 immunization campaign in Colombia started in February 2021 and was carried out in stages, giving front-line healthcare providers and the elderly priority.7,8 All age groups began receiving vaccinations in July 2021, and by February 2022, more than 60 % of the Cali and the Colombian population had received the recommended vaccination scheme (two doses of any vaccine or one Ad26.COV2.S).7,8 Despite this, there is limited data on the clinical traits and prognoses of COVID-19 patients admitted to hospitals in Colombia during the several pandemic waves, both before and after immunization. Assessment of these traits is essential to determine if public health initiatives like immunization and modifications to clinical treatment practices were effective.9

This study aim is to compare the experiences and clinical outcomes of COVID-19 patients who were admitted in a Latin American university hospital in Cali, Colombia, during the first four waves of the pandemic. The results of this study will enable researchers, policy makers and health professionals to better understand the effects of the pandemic and to use this knowledge to inform future public health initiatives.

From June 2020 to February 2022, 6,039,022 COVID-19 cases were reported in Colombia of which 4275 patients were seen at FVL and 2733 were included (530 patients consulted in the first wave, 568 in the second wave, 1166 in the third wave and 469 in the fourth wave). The patterns of COVID-19 waves were very similar in our hospital compared to national patterns (Fig. 1, Panel A and B).

Fig. 1.

Distribution of COVID-19 cases and prevalence of circulating variants. Panel A: Incidence of COVID-19 in Colombia. Panel B: Incidence of COVID-19 in the study population. Panel C: Prevalence of variant/lineages circulating in Colombia.

(0.31MB).

Slightly over the half of the patients were male (55.6 %), with a mean age of 60 years, and 74.8 % had at least one comorbidity. Table 1 summarizes the characteristics of the hospitalized patients with SARS-CoV-2 infection stratified by wave. The second and third wave included a higher proportion of patients older than 50 years old. For the third wave, cases under 50-years-old were 38.4 % compared to the other waves which was less than 30 %. A decreasing proportion of males through waves were recorded. Hypertension (40.8 %), diabetes (20.6 %) and obesity (42.6 %) were common comorbidities during the four waves.

Patients admitted in the third wave had a lower prevalence of hypertension (34.8 %, compared to the other waves that were up to 40 %) and cerebrovascular diseases (2 %, in relationship with the trend higher than 3 %). In the fourth wave the history of cancer was significantly higher (22.2 %), while the trend in the other waves didn't overtake 10 %.

The vaccination was only available until the third wave. At the end of the study period the proportion of patients that received at least one dose of vaccination against COVID-19 was 56.7 %.

In addition, when contrasting the variants circulating in Colombia with respect to the waves of cases during the study period, it became evident that the end of the third wave coincided with a high circulation of B.1.617.2-Delta and B.1.621.1-Mu, while the fourth wave coincided with a higher circulation of BA.1.x-Omicron (Fig. 1, Panel C). Genomic surveillance in Colombia started later, so no public data were available regarding circulating variants during the first two waves.

Overall, ICU admissions were more frequent during the first wave (62.83 %), reflecting the initial challenges faced in managing severe cases of COVID-19. As the pandemic progressed, there was a significant reduction in the rate of ICU admissions across the successive waves: 51.23 % in the second wave, 58.23 % in the third wave, and finally down to 46.70 % in the fourth wave (p < 0.001). Additionally, the data revealed significant differences in ICU requirements among waves. Table 2 summarizes the outcomes of hospitalized patients stratified by age, highlighting these differences.

In general, younger patients (aged 49 years or younger) had a higher ICU and IMV requirement in the third wave, in contrast to the older groups (aged 60-years or older) who mainly were admitted to ICU during the second wave. The trend for IMV requirement mirrored this pattern, with the younger cohort showing a higher proportion of IMV use during the third wave. Notably, the overall use of IMV was more frequent in the first wave at 39.3 %, which then decreased to 32.2 % in the second wave, 31.2 % in the third wave, and significantly dropped to 21.6 % in the fourth wave (p < 0.001). This observed decrease in IMV usage across the waves aligns with the overall trend of reduced ICU admissions.

The hospital LOS was longest during the first wave, with a median of 9 days, and remained consistent at 8 days for the second and third waves. A minor uptick to a median of 8.5 days occurred in the fourth wave. Among different age groups, patients between 60 and 79 years experienced the most extended LOS. However, during the third wave, the 50‒59 age range saw the lengthiest hospital stays, whereas there was a gradual decrease in LOS for the eldest group (80 years or older) throughout the waves. The in-hospital mortality remained stable through this wave analysis, with rates oscillating between approximately 18‒20 %. However, when stratified by age, we found that patients aged 50–59 years-old had lower mortality during the first two waves, followed by a significant increase in the third and fourth wave. The distribution of mortality according to age and waves was illustrated in Table A1.

ICU mortality was 82.3 %, with a significant difference comparing non-ICU patients (17.7 %), see Table 2.

A higher utilization of corticosteroids was noted during the first wave, with a subsequent pronounced decrease across the later waves: 86.3 % in the first wave, 79.2 % in the second, 74.9 % in the third, and 57.8 % in the fourth (p < 0.001). Similarly, the incidence of secondary infections was highest initially at 23.8 %, dropping to 16 % in the second wave, further declining to 13.2 % in the third, with a slight increase to 18.4 % in the fourth wave (p < 0.001).

Among ICU patients (Table 3), a higher proportion of patients had a history of diabetes and obesity, and in-hospital mortality; however, chronic kidney disease, cancer, and solid organ transplantation were mainly managed at the general ward. Furthermore, there were no significant differences among non-ICU and ICU patients regarding the vaccination status; 16.7 % of ICU patients had at least one dose of vaccine against SARS-CoV-2 vs. 15.9 % admitted to the general ward.

From all admitted patients in third wave with at least one COVID-19 vaccine, 60 % required ICU admission, 36 % IMV and 21 % had a fatal outcome.

This study described the COVID-19 dynamics among hospitalized patients across four consecutive waves from 2020 to 2022, observing the most significant peaks in July 2020, December 2020, June 2021, and January 2022. There were differences in age, chronic diseases frequency, ICU admission, IMV requirement, and LOS among waves. In contrast, there were no differences in overall mortality, which may be explained by the unavailability of new specific drugs against SARS-CoV2 in Colombia during the study period and an incomplete coverage of vaccination in Colombia at the end of the fourth wave. When sub analysis were performed, some differences were observed in the proportion of mortality according to age groups and between ICU and non-ICU patients.

In this study, we observed a marked increase of the cases during the third wave, which can be explained by the emergence of SARS-CoV-2 variants such as gamma and alpha variants circulating by June 2021 concurrently with the third wave of the pandemic in Colombia.11 In the fourth wave, the BA.1.x-Omicron variant predominated; however, there was no significant increase in cases despite the most virulent variant circulating. The higher proportion of vaccinated individuals and the lower association of severe cases with this lineage could explain this observation. These variants have increased transmissibility, virulence, and reduced vaccine efficacy as demonstrated in previous studies.12

Despite the increase in cases in the third wave, the length hospital stays, admission to ICU, and the requirement for IMV were lower than in the first wave. Likewise, mortality was maintained concerning previous waves, and the younger population predominated. Possible underlying causes of these changing patterns are the lower exposure to infection among older individuals, since the youngest was the first to rejoin work and social activities and has been associated with better clinical outcomes respect to the other age groups.13,14

In the fourth wave, there was a substantial decrease in COVID-19 cases. However, it is noteworthy that the most affected population were older than 70 years old since it was expected that the cases in this population would be lower because of being the first to receive the vaccination and boosters. Arregocés-Castillo et al. observed that the effectiveness of vaccines declined with increasing age, probably related to the greater probability of pre-existing conditions in older people, in conjunction with age-related frailty and immunosenescence.15

One of the findings from our study was that, by the end of the fourth wave, vaccination coverage was below 60 %, and among patients admitted to the ICU in the third wave, more than 80 % were not immunized against SARS-CoV-2. Upon comparison, there were no significant differences in vaccination status between ICU and non-ICU patients. Similarly, other studies conducted around the same time reported that approximately 88 % of ICU-admitted patients were unvaccinated.16 Despite the suboptimal vaccination coverage, there was an increase in vaccination rates in comparison to previous waves, which appeared to correlate with a decreased demand for ICU and mechanical ventilation during this period. The low vaccination rates at this point can be explained by various sociocultural and geopolitical factors unique to Colombia at that stage of the pandemic. These include significant delays in vaccine delivery at the pandemic's onset, logistical and socio-political challenges, including protests that disrupted vaccination efforts, and an inclusive yet ambitious strategy to vaccinate the extensive Venezuelan migrant population, which was ineffective in achieving the intended coverage by this point.17

Overall, we observed an improvement in outcomes for patients admitted during the second, third and fourth wave than in the first wave, with a reduction in ICU admission, IMV requirements, and shorter Length of Stay (LOS), similarly to the reported by Fluck et al.18 and Zuil et al.2 but contrary to finding in another study which showed a more severe second wave than the first one in Brazil regarding ICU admission and IMV requirement.3 These differences in severity could be secondary to the new clinical protocols and the experience gained from the first two waves, and the positive effects of immunization. Although we found a higher proportion of patients admitted to the ICU than in other studies,2,19 mortality among hospitalized patients was similar, which may be an indicator of appropriate interventions in patients who required ICU in our institution.

In our study population no patients receive specific therapies approved for COVID-19. Despite several clinical trials and studies have suggested using antivirals such as remdesivir, molnupiravir, and Nirmatrelvir-Ritonavirm, as they have had an impact on mortality in patients with COVID-19.20–22 It was not until March 2022 that molnupilavir was approved for use in Colombia for unvaccinated individuals with no history of COVID-19 infection and risk factors for progression to severe disease.23 This could be one of the factors that explain the lack of variation along the waves in mortality in our population since.

Corticosteroids started to be used in severe cases of COVID-19 for ARDS, given the findings reported by the RECOVERY study.24 This type of medication has been implied in important pathological processes as viral clearance and the modulation of the immune system functions, which are critical for the worsening or recovery of the critically ill COVID-19 patients depending on the therapy timely and dosage.25 De Bruyn et al.26 also found a cumulative dose of corticosteroids administered in the ICU and male sex as risk factors associated with the acquisition of secondary infection. In our study, higher use of corticosteroids was observed among patients who developed secondary infections, mainly in the first wave. However, there was a lack of evidence regarding corticosteroids adequate dosages at the beginning of the pandemic.4,27,28

A major finding in this cohort was that IMV requirement was lower than ICU admission. IMV requirements in our population may reflect the implementation of alternative oxygen therapy strategies such as High-flow nasal cannula, where high-flow oxygen through a nasal cannula decreased the need for mechanical ventilation support and clinical recovery time.29

This study has some limitations. Data were collected from clinical record, thus there could be missing data due to the accuracy and quality of data collection. This study is a single-center study, which may limit external validity and have a potential selection bias. Although we attempted to use national public data to understand infection and reinfection dynamics in our study as an approximation of an ecological study, we do not know exactly the impact of new variants of SARS-CoV-2 and vaccination in our population. In this study data regarding the onset, duration and dosages of steroids were not evaluated. Nonetheless, we believe that this study may represent the region's situation because our hospital receives patients from broad coverage in the country; no similar studies were found locally. The information on the fifth wave is not available, since it was not collected in the registry.

During the first four waves of the pandemic, there were differences in the characteristics of patients hospitalized with COVID-19. The potential appearance of SARS-CoV-2 serotypes showed a variation in the number of cases, but not with an increase in mortality, suggesting greater transmissibility but not greater virulence. Vaccination and the experience gained by health professionals in managing patients with SARS-CoV-2 have led to improved clinical outcomes. Efforts are needed to maintain the prevention of the spread of SARS-CoV-2, understand the risk of the new variants, and the effects of available vaccines.

This study was approved by the Comité de Ética en Investigación Biomédica at Fundación Valle de Lili (Approval number: IRB/EC nº 1566). It followed the ethical principles for clinical research by the Declaration of Helsinki. As research without risk according to Colombian law, it was considered exempt from applying informed consent (Resolution 8430 of 1993).

During the preparation of this work the authors used Chat-GPT in order to improve English writing. After using this tool/service, the authors reviewed and edited the content as needed and takes full responsibility for the content of the publication.