Association of vaccination status and severity of SARS-CoV-2 infections in hospitalized patients

Abstract

Vaccination is considered the most promising approach for addressing the COVID-19 pandemic. However, even vaccinated people remain at risk. In this study, we examined the association between levels of vaccination and clinical outcomes in hospitalized patients. We conducted a retrospective review of adults hospitalized with COVID-19 infection. Of 484 patients, fully vaccinated (OR = 0.49, p = 0.001) and updated patients (OR = 0.46, p = 0.004) had significantly lower probability of critical severity compared to unvaccinated. Vaccination status is significantly related with 30-day mortality (p = 0.005) but not significantly associated with need for respiratory support or ICU stay. Mean length of stay (LOS) of 6.6 days among boosted patients is significantly lower than patients with no vaccination status (10.7 d, p < 0.001). Our study findings provide real-world evidence of the benefit of booster vaccinations against critical infection and death as well as shortcomings in ICU stay, length of stay or need for ventilatory support.

Introduction

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was first identified in Wuhan, Hubei Province in December of 2019 [1] Coronavirus disease 2019 (COVID-19) infection then rapidly spread across countries, resulting in countless infections and deaths [2]. In response to the disease's rapid spread and ease of transmission, vaccination research and production was prioritized allowing for the rapid distribution of mRNA (Pfizer and Moderna) and adenovirus-based (Janssen) vaccines [3]. Clinical trials supported the emergency use of these mRNA vaccines following the estimated efficacy of the two-dose vaccination series to be >90 % effective in preventing transmission [4], [5], [6].

SARS-CoV-2 is a single stranded, positive sense RNA that has a genome composed of 29,903 nucleotides [7]. The virus has four known structural proteins including spike (S), envelope (E), membrane (M), and nucleocapsid (N) proteins. Most designed vaccines inhibit viral pathogenesis by way of targeting the S protein, which utilizes the human host cell surface receptor angiotensin-converting enzyme 2 (ACE-2) to gain entry into the host. Despite our best vaccination efforts, SARS-CoV-2 has undergone multiple mutations resulting in higher transmissibility and resistance to neutralization by host defenses [8], [9], [10], [11].

The Delta variant of COVID-19 emerged in India in December of 2020. Other variants including D614G and Alpha were also noted around this time [8]. Despite these mutations, the monovalent vaccines continued to confer immunity, as evident by the case-control analysis completed by Tenforde et al. including 10,078 patients, which showed mRNA vaccination efficacy against the Delta, D614G, and Alpha variants of SARS-CoV-2 in hospitalized patients as far as 9-months after receiving the vaccination [12], [13]. In contrast, numerous studies have found waning immunity to SARS-CoV-2 beginning as early as 3–6 months post vaccination with the two monovalent vaccines [14], [15]. For this reason and the continued rise in cases around winter months, the Food and Drug Administration (FDA) and Centers for Disease Control and Prevention (CDC) began recommending monovalent boosters in fully vaccinated patients. This monovalent booster contained the original mRNA strain against SARS-CoV-2.

At the time of this study, the dominant strain in the community is the Omicron variant, or B.1.1.529, of which three lineages exist: BA.1, BA.2, and BA.3 [9]. It was first detected in Southern Africa in November 2021 [8]. While it is still heavily reliant on ACE-2 receptors to enter host cells, it tends to replicate at a higher rate in epithelial cells of the upper respiratory tract. This variant is more likely to cause milder disease, yet potentially more easily transmissible as well as more resistant to treatment and vaccinations [9]. This variant contained more than 30 mutations, making it the most genetically different from the wild type virus. Emergence of these variants place even fully vaccinated people at risk of COVID-19. This prompted a second monovalent booster dose recommendation targeting the original SARS_CoV-2 strain for US adults aged ≥50 years or for individuals who are immunocompromised [4].

In this study, we examined association between different levels of vaccination specifically unvaccinated, partial vaccination and updated vaccination and clinical outcomes including 30-day mortality, severity of illness and need for intensive care unit (ICU) stay or respiratory support in hospitalized patients with COVID-19.

Methods

Patient and vaccination status

Data for this retrospective chart review was obtained from the electronic medical records of 484 adult patients who were admitted to Sanford Health and tested positive for COVID-19 in the month of January 2022. The facility is a 284-bed hospital serving patients in rural North Dakota and western Minnesota. In the fall of 2020, the state of North Dakota had the highest COVID-19 related death rate and cases in the United States. The state has over 780,000 population with over 86 % Caucasian, 3.6 % Black or African American while 5.3 % identify as American Indian and Alaskan Native. It is a rural state located in the geographic center of the US. The study was conducted in this setting when the world saw the Omicron effect that peaked at the end of January 2022. Omicron was regarded as being significantly more transmissible than Delta variant such that we saw another big surge of cases, hospitalizations and deaths.

Each patient was assigned a vaccination status, with the options being fully vaccinated, partially vaccinated, unvaccinated, or boosted. For patients who got vaccinated, the type of vaccine was also recorded. All vaccinated patients received one of 3 possible vaccines used in the United States: Pfizer, Moderna, or Johnson & Johnson. Every patient who was boosted at the time of testing positive had received only one additional monovalent vaccination in addition to the primary series. Demographic information that was recorded included age, sex, race, and smoking history. Each patient’s medical history was examined to see if they had a history of conditions including chronic lung disease, diabetes, cardiovascular disease, HIV, immunocompromised state other than HIV, and hypertension. Outcomes of interest included severity of infection, 30-day mortality as well as hospital interventions including respiratory support with bilevel positive airway pressure (BiPAP) and mechanical ventilation, and need for intensive care unit (ICU) stay were recorded.

Severity of illness

Patients were assigned into one of three categories for illness severity during their admission: mild-to-moderate, severe, and critical. The COVID-19 clinical spectrum, included below, from the National Institute of Health was utilized to place patients into a category, along with clinical judgement [16].

Moderate illness: Individuals who show evidence of lower respiratory disease during clinical assessment or imaging and who have an oxygen saturation measured by pulse oximetry (SpO2) ≥94 % on room air at sea level.

Severe illness: Individuals who have SpO2 <94 % on room air at sea level, a ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PaO2/FiO2) <300 mm Hg, a respiratory rate >30 breaths/min, or lung infiltrates >50 %.

Critical illness: Individuals who have respiratory failure, septic shock, and/or multiple organ dysfunction.

Statistical analysis

Outcomes including 30-day-mortality, respiratory support with BIPAP or mechanical ventilation, ICU stay are all dichotomous distributed variables. Logistic regression models were applied to all these three variables. Logistic ordinary model was used for severity of illness which has three categories. Patient’s age, gender, race and number of comorbidities were included in all the models as covariates. All analyses were done by SPSS V 25, and p < 0.05 was considered as statistical significance.

Results

Description of the patient population

Of 484 patients included, 256 (53 %) were unvaccinated against COVID-19, 131 (27 %) were fully vaccinated, and 82 (17 %) has received a booster. Only 15 (3 %) were partially vaccinated with one dose of either the Pfizer or Moderna vaccine. Among all the patients, gender was evenly distributed (51 % male), and the majority were Caucasian (84 %). Age ranged from 19 to 97 with an average age of 64 years. Additional demographics of the study are shown in Table 1 below. A total of 80 (16.5 %) patients died within 30 days of their initial admission to the hospital while a majority of patients (81 %) presented with severe or critical severity of illness. There were 74 (15.3 %) patients admitted to the ICU. The average number of comorbidities of the patients is 7, ranging from 0 to 18. The mean inpatient length of stay is 9 days ranging from 1 to 114 days (Table 1).

Table 1.

Characteristic of patients hospitalized with COVID-19.

Patient characteristics (n = 484)		N(%)
Sex	Male	247 (51)
	Female	237 (49)
Age, years (mean ± SD)		64 ± 18
Vaccination Status	Unvaccinated	256 (53)
	Fully vaccinated	131 (27)
	Boosted	82 (17)
	Partial vaccination	15 (3)
Vaccine type	Pfizer	131 (27)
	Moderna	78 (16)
	J&J	19 (4)
Race	Caucasian/White	406 (84)
	American Indian or Alaskan Native	66 (14)
	African American/Black	6 (1)
	Other	6 (1)
Number of comorbidities* (mean ± SD)		7 ± 4
30-day mortality		80(17)
Severity of illness	Mild-moderate	92 (19)
	Severe	117 (24)
	Critical	277 (57)
ICU stay		74 (15)
Length of stay in days (mean ± SD)		9 ± 12

Upon evaluation of severity of infection, fully vaccinated (OR = 0.49, p = 0.001) and boosted patients (OR = 0.46, p = 0.004) had significantly lower probability of critical severity compared to unvaccinated patients. Older patients (beta = 0.014, p = 0.015) and patients with more comorbidities (beta = 0.114, p < 0.001) had higher probability of severe or critical condition. Males also had a higher chance to be in critical condition than females (OR = 1.58, and p = 0.014).

Assessment of main outcomes specifically 30-day mortality, need for respiratory support and ICU stay is summarized in Table 2. Vaccination status was significantly related with 30-day mortality (p = 0.005). Fully vaccinated patients had an estimated probability of death of 0.06 and the chance of death was 70 % lower for the fully vaccinated patients than patients without vaccination (OR = 0.17). Older patients and patients with more comorbidities had higher probability of death within 30 days (p < 0.001 and p = 0.002, respectively).

Table 2.

Relationship of Vaccination Status to Mortality, Need for Respiratory Support and Intensive Care Unit Stay.

30 day mortality
Vaccination status	Probability of death	Odds ratio (95 % CI)
Fully vaccinated	0.06	0.31 (0.15–0.64)
Boosted	0.10	0.54 (0.27–1.09)
Partial vaccinated	0.27	0.53 (0.56–6.35)
Not vaccinated	0.17	reference level
Respiratory support (BIPAP or mechanical ventilation)
Vaccination status	Probability of respiratory support	Odds ratio (95 % CI)
Fully vaccinated	0.08	0.54 (0.27–1.06)
Boosted	0.08	0.54 (0.25–1.12)
Partial vaccinated	0.23	1.71 (0.50–5.87)
Not vaccinated	0.15	Reference level
ICU stay
Vaccination status	Probability of ICU stay	Odds ratio (95 % CI)
Fully vaccinated	0.15	0.84 (0.45–1.55)
Boosted	0.14	0.77 (0.36–1.63)
Partial vaccinated	0.06	0.29 (0.04–2.29)
Not vaccinated	0.17	Reference level

Fully vaccinated and boosted patients had lower probability of respiratory support with either BIPAP or mechanical ventilation, both at 0.8 but did not reach limit of statistical significance.. The patients with higher number of comorbidities had increased probability of requiring support (p = 0.008, OR = 1.118). Similarly, vaccination status and all the covariates were not statistically related with ICU admission (p > 0.05).

Discussion

The SARS-CoV-2 virus responsible for the most recent pandemic is believed to be on its way to endemicity over time. Vaccines have been a valuable tool in addressing the pandemic. However, given the dynamic nature of COVID-19 infection as new variants emerge, research and development of the vaccines will need to be an ongoing effort as the situation surrounding the pandemic changes. With the rapid mutation rate of the virus, vaccination will be necessary to maintain population safety [17], [18], [19], [20]. This study demonstrated some benefits of COVID-19 vaccination in terms of severity of illness and 30 day mortality. Fully vaccinated and updated patients had significantly lower probability of critical severity compared to unvaccinated patients. The odds of a critically severe infection is reduced more than half upon receipt of full vaccination with or without a booster. We have found that increased age and presence of more concomitant conditions predispose to increased severity of illness. Immunogenicity and subsequent vaccine effectiveness appear lower on patients with immunocompromising conditions. Increased age and concomitant comorbidities as described in this study likely reflect the predisposition to more severe infections. Identification of risk factors and conditions that may affect vaccine responsiveness and therefore possibility of more severe infections prompted special considerations and schedules on vaccination for special populations including the immunosuppressed. Booster vaccinations were recommended to restore protection against severe infections when the effectiveness of the vaccine wanes over time.

In this study, there were no benefits seen in ICU stay nor need for respiratory support. This study emphasizes the benefit and shortcomings of vaccine in the real world setting and justifies the recommendations for constantly updating vaccines since the virus that causes COVID-19 is always changing and protection from COVID-19 vaccines decline over time [14], [21].

The bivalent booster was developed to improve protection against viral mutations including the Omicron variant. The bivalent booster has been shown to be effective in providing a modest amount of protection in preventing symptomatic COVID-19 requiring medical attention when compared to patients who have no vaccination or 2–4 doses of monovalent vaccine doses [13]. The bivalent vaccine provides more diverse antibody production and better protection against the omicron variant (FDA). Most recently, the US Food and Drug Administration and Centers for Disease Control and Prevention have updated the COVID-19 vaccine authorizations and recommendations that the 2023–2024 formula have been updated to target the Omicron variant XBB.1.5.

The limitations of this study include a small population size. Due to the duration of this study, our sample size was 484 patients. Additionally, confounding factors were not taken into account including comorbidities and severity of illness.

There were only 15 patients were in partial vaccination category, the sample size is quite limited. In addition, among these 15 patients, 5 patients died within 30 days, and 9 of them were in critical condition. Results in this category may not able generalizable to the population due to sample size. In spite of these, this study reflects a specific perspective in the timeline of COVID-19 history. It was conducted at the peak of Omicron variant in a rural facility at a state that used to have the highest infection rate per capita in the world.

Conclusions

At the peak of COVID-19 infection with Omicron variant (B.1.1.529), this study findings provided real-world evidence in an upper midwest US state that full and updated booster vaccinations substantially increase protection against critical infection and death in hospitalized patients with COVID-19 infection.

However, there were no perceivable benefit in need for respiratory support or ICU stay justifying the need for updating components of the vaccine as circulating strains are also dynamic.

Funding Information

None.

CRediT authorship contribution statement

Dubert Guerrero: Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Supervision, Writing – original draft, Writing – review & editing. Thomas Baker: Conceptualization, Investigation, Methodology, Project administration, Resources, Validation, Writing – original draft, Writing – review & editing. Megan Corn: Investigation, Methodology, Writing – original draft, Writing – review & editing. Sean Keup: Conceptualization, Funding acquisition, Investigation, Project administration, Writing – original draft, Writing – review & editing. Austin Nickell: Conceptualization, Data curation, Resources, Writing – original draft, Writing – review & editing. Li Cao: Data curation, Formal analysis, Investigation, Methodology, Writing – original draft.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Dubert Guerrero reports statistical analysis was provided by Sanford Health.

Data availability

The data that has been used is confidential.