Figure 1

Figure 1. Individual trajectories of 24 long-term care facility residents over 30-day study period in outbreak of SARS-CoV-2 B.1.1.7 lineage infections, Germany, February–March 2021. A) Four unvaccinated residents; B) 20 residents who...

On February 4, 2021, daily SARS-CoV-2 screening of employees yielded a positive antigen point-of-care test (AgPOCT) result in 1 caregiver in a LTCF in Berlin, Germany. Among 24 residents of the unit under their responsibility, 20 (83%) residents had received the second dose of BNT162b2 on January 29 or 30, 2021 (Figure 1). Four residents had not been vaccinated for nonmedical reasons (i.e., personal refusal or delayed provision of consent by legal guardian). AgPOCTs and reverse transcription PCR (RT-PCR) testing of all residents on February 4 detected SARS-CoV-2 infections in 3/4 unvaccinated and 10/20 vaccinated residents (Figure 1). At the time of testing, 2 vaccinated patients exhibited mild fatigue and one of those also had diarrhea; all other patients were asymptomatic.

The next week, testing detected 7 additional infections, resulting in 4/4 unvaccinated infected residents and 16/20 vaccinated infected residents. The remaining 4 vaccinated residents tested negative throughout the 30-day observation period (Figure 1).

In addition to residents, 11/33 (33%) staff members from the unit tested positive for SARS-Cov-2 by February 18; of those, none were twice-vaccinated staff members, 2/8 (25%) had received 1 dose of BNT162b, and 9/22 (40.9%) had not been vaccinated. No infected staff required hospital treatment.

Figure 2

Figure 2. Characteristics of outbreak of SARS-CoV-2 B.1.1.7 lineage infections after vaccination in long-term care facility, Germany, February–March 2021. A) After a positive test result in a healthcare worker, 16/20 (80.0%) vaccinated...

Respiratory symptoms, including cough and shortness of breath, occurred in 5/16 (31.3%) vaccinated patients and all 4 unvaccinated patients (Figure 2, panel A; Appendix Table). All 4 unvaccinated SARS-CoV-2–infected patients and 2/16 (12.5%) vaccinated patients required hospitalization (Figure 1; Figure 2, panel A). Supplemental oxygen therapy was required by 3/4 (75.0%) unvaccinated and 1/16 (6.3%) vaccinated patients (Figure 1; Figure 2, panel A). Two patients, 1/16 (6.3%) vaccinated persons and 1/4 (25.0%) unvaccinated persons, required intermittent oxygen therapy after discharge. One vaccinated patient with a history of hypertension and microvascular dementia died 6 days after testing positive by RT-PCR because of a hypertensive crisis with intracerebral hemorrhage. Another vaccinated patient died 16 days after testing positive by RT-PCR. Neither patient experienced respiratory symptoms during the infection (Figure 1).

Containment measures in place included mandatory use of FFP2 or N95 masks and daily AgPOCT screening for anyone entering the facility. Immediately after detection, the facility was closed to visitors and additional containment measures were put in place, including designated staff and separate entrance, elevator, and changing rooms. Staff were required to change personal protective equipment before entering each room. Residents of all 7 units of the LTCF underwent weekly AgPOCT for >3 weeks, and residents in the adjacent unit underwent AgPOCT every 2–3 days. The outbreak was contained within the unit; no further cases were detected.

All SARS-CoV-2 RNA-positive samples were tested for presence of SARS-CoV-2 VOCs by RT-PCR and complete genome sequencing (Appendix). RT-PCR suggested the presence of B.1.1.7, which was confirmed by sequencing in 11 patients for whom sufficient sequence information was available. In phylogenetic analysis, sequences form a monophyletic clade with additional sequences from Berlin interspersed (Appendix Figure 1), suggesting a common outbreak source, including infections outside the unit.

We performed serial RT-PCR testing of nasopharyngeal swab specimens from 22 patients. SARS-CoV-2 RNA concentrations peaked within 5 days (Appendix Figure 2). The median peak SARS-CoV-2 RNA concentration in vaccinated and unvaccinated patients overlapped concentrations detected at time of diagnosis in B.1.1.7 patients of similar ages (Figure 2, panel B). However, SARS-CoV-2 RNA concentration was lower among vaccinated residents than unvaccinated residents, although the difference was not statistically significant (6.45 vs. 8.15 log₁₀ copies/mL; p = 0.10). Furthermore, duration of SARS-CoV-2 RNA shedding was considerably shorter in vaccinated patients than in unvaccinated patients (7.5 [95% CI 7–17.3] days vs. 31 [95% CI 21.5–34.5] days; p = 0.003) (Figure 2, panel C). Peak SARS-CoV-2 RNA concentrations above 10⁶ copies per mL, below which virus isolation in cell culture is usually not successful, were detected in all 4 unvaccinated patients but only in 7/16 vaccinated patients (9).

We further assessed the level of infectiousness in 22 samples from 14 patients by virus cell culture (Appendix). One sample obtained from a vaccinated patient 7 days after the first positive RT-PCR test, which showed 9.32 log₁₀ SARS-CoV-2 RNA copies/mL, yielded a positive isolation outcome. Isolation attempts from samples of the same patient taken in the next 4 days and from 21 samples taken from 13 other patients were unsuccessful.

Five weeks after initial testing, 8/8 vaccinated and infected residents and 2/2 unvaccinated and infected residents showed robust antibody responses against SARS-CoV-2 spike antigens, virus neutralization capacity, and interferon-γ release of SARS-CoV-2–specific T cells (Figure 2, panels D, E; Appendix Figure 3). These results confirm the immune response capability in these patients.

We performed a longitudinal study of SARS-CoV-2 infections in a LTCF unit. Nearly all infected residents were symptomatic, including most residents that had received a second BNT162b2 dose the week before. The outbreak was caused by SARS-CoV-2 VOC lineage B.1.1.7, which might partly explain the high attack rate and lack of protection in vaccinated residents. Nevertheless, we reported a lower attack rate, a shorter duration of SARS-CoV-2 RNA shedding, and a lower proportion of symptomatic COVID-19 requiring hospitalization and oxygen support for vaccinated patients. However, despite the limited sample size and the short interval between second vaccination and infection, this outbreak raises questions about the effectiveness of the vaccination regimen in the elderly (3,8,10–12). A delayed and overall reduced immune response to BNT162b2 vaccination has been described in elderly persons (13,14), which might explain the reported outbreak and infections in LTCF described elsewhere (4,5).

This outbreak highlights that older adults have reduced protection <2 weeks after second BNT162b2 vaccination. Therefore, single-dose regimens and extended dosing intervals might be insufficient for fully protecting this population (15). Vaccination of LTCF residents and staff is likely effective in reducing the spread of SARS-CoV-2. However, regular SARS-CoV-2 screening, prompt outbreak containment, and nonpharmaceutical interventions (16) remain necessary for optimal protection in this setting.