The Effects of SARS-CoV-2 Virus Type and of Vaccination Status on Causes of Death Over the Course of the COVID-19 Pandemic

Evolutionary processes have meant that the original virus variant of SARS-CoV-2 was largely replaced during the course of the pandemic by several variants of concern (VOC). In Germany, the alpha, delta, and omicron subtypes were consecutively epidemiologically dominant or indeed they still are. VOCs in their pathogenic characteristics—such as transmissibility, virulence, or susceptibility—show differences vis-a-vis the immune response of recovered or vaccinated persons.

Coinciding with the change from the alpha subtype to the delta subtype, large parts of the population without risk factors started to be vaccinated against SARS-CoV-2. The present study aimed to investigate and compare the causes of death in SARS-CoV-2 positive patients who died in a university hospital, especially in relation to the different viral variants.

Methods

We included all patients who died from March 2020 through April 2021 at the University Medical Center Hamburg-Eppendorf (UKE) with a SARS-CoV-2 infection (positive qPCR test) and in whom the virus had been typed using qPCR. We analyzed among others demographic baseline characteristics, vaccination and booster status, risk factors for a severe course of COVIC-19 disease (in analogy to the classification of the US Centers for Disease Control and Prevention [1]), the clinical course, microbiological findings, and results of the postmortem exam. All cases were classified by using the structured DELPHI procedure regarding the actual cause of death according to Fitzek et al. (2). If the COVID-19 infection was a part of the causal chain that directly led to the death, the death was classified as „FROM SARS-CoV-2 infection.“ If the infection did was not relevant to the cause of death, the death was classified according to the WHO classification (3) as „WITH SARS-CoV-2 infection.“ In the latter scenario, the cause of death was categorized into cardiovascular, infectious, hemato-oncological, trauma-related, or other causes.

Results

During the observation period, 234 patients in the UKE died in the context of their inpatient stay with a positive qPCR test for SARS-CoV-2-RNA. The autopsy rate in this population was 26.4%. In seven cases, no technical typing was possible, with the result that 227 patients were included in the analysis. 117 deceased patients were infected with wild-type SARS-CoV-2, 33 with the alpha variant, 38 with the delta variant, and 19 with the omicron subtype. The rate of patients who died FROM SARS-CoV-2 infection was as follows for the individual virus subtypes: 85% (wild type), 94% (alpha), 82% delta, and 46% (omicron). Of these, 24% of patients infected with the delta subtype had been vaccinated/received a booster vaccine, of which 16% experienced a severe course of COVID-19 disease. Of those patients infected with the omicron strain, the proportion of deaths in patients who had been vaccinated/received a booster was 41%, in all of whom COVID-19 took a severe course. The median age of the alpha group was younger to a statistically significant degree, the remaining groups did not differ significantly in terms of age. The Figure and Table summarize the results; the latter also shows the causes of death of the patients who died WITH confirmed SARS-CoV-2.

Figure.

Ratio of patients who died FROM versus WITH SARS-CoV-2 infection as a function of virus subtype. non-VOC: non-variant of concern

Table. Demographic characteristics of patients who died after testing positive for SARS-CoV-2 on polymerase chain reaction, by virus subtype, vaccination status, and presence of risk factors for t severe course of their COVID-19 infection.

	Viral variant
	non-VOC n = 117	Alpha n = 33	Delta n = 38	Omikron n = 39
Female/male	40/77	10/23	12/26	15/24
Age (median/interquartile range)	70 (59–82)	60 (50–68)	63.5 (55–71)	71 (57.5–81)
Died FROM SARS-CoV-2 infection – Not vaccinated – Vaccinated/received booster, with risk factor – Vaccinated/received booster, without risk factor	99 (85 %) 0 0	31 (94 %) 0 0	22 (58 %) 6 (16 %) 3 (8 %)	2 (5 %) 16 (41 %) 0
Died WITH SARS-CoV-2 infection – Not vaccinated – Vaccinated/received booster, with risk factor – Vaccinated/received booster, without risk factor	18 (15 %) 0 0	2 (6 %) 0 0	3 (8 %) 4 (10 %) 0	4 (10 %) 17 (44 %) 0
Causes of death – COVID – Cardiovascular –Infectious (non-COVID) – Hemato-oncological – Trauma-related – Other	99 3 9 3 1 2	31 2 0 0 0 0	31 2 1 0 0 4	18 5 6 3 2 5

non-VOC, non-variant of concern

Discussion

The analysis of the data shows two results:

• The proportion of patients who died with a positive SARS-CoV-2 PCR test who died FROM SARS-CoV-2 infection fell over the course of the pandemic in our population from 85% to 46%.

• Furthermore, the rate of vaccinated patients without risk factors who died in the phase of the delta subtype fell to 8% and since the transition to the omicron subtype, to 0% of deaths.

These results underline the good effectiveness of the licensed COVID-19 vaccines as regards their potential for preventing fatal outcomes (4). The extent to which reduced pathogenicity of the omicron subtype may also be a reason for the reduced rate of deaths in patients who died FROM SARS-CoV-19 infection cannot be assessed on the basis of our data. Some studies (5) that compared the disease course of patients with the omicron subtype with the delta subtype support this interpretation, but there are also preliminary reports (non-peer-reviewed preprints) that put this into question. Vaccinated persons without risk factors have an extremely low risk of a fatal course if infected. The majority of deaths occurred in the group of unvaccinated persons or the group of vaccinated persons with risk factors. In our population, this concerned for the omicron subtype primarily patients with severe immunosuppression in hemato-oncological or autoimmune disorders. Age distribution is not the determinant.

These insights are positive for vaccinated patients without risk factors for COVID-19 infection with a fatal outcome, but they also show that patients with risk factors, preventive non-medication related measures (for example, wearing facial coverings over mouth and nose), passive immunization (administration of monoclonal antibodies), and early use of antiviral drugs) are still important components of reducing COVID-19 associated mortality. It remains to be seen to what extent the preparations that are currently available for passive immunization will stay effective against future viral variants.

The limitation of our study is the single-center data collection at one university hospital. To our knowledge, this is the only data collection in Germany regarding the question of whether patients died FROM or WITH SARS-CoV-2 in a setting of varyingly dominant variants of concern during the course of the pandemic. The high autopsy rate in our population supports the results of our analysis of deaths.