Symptom Profile and Breakthrough Infections in Healthcare Workers Post Comirnaty Vaccine in a Tertiary General Hospital in Greece: A Narrative Review

Abstract

The COVID-19 pandemic has necessitated urgent measures to curb its spread, with vaccination emerging as a pivotal strategy. This prospective observational study evaluated breakthrough COVID-19 infections among healthcare workers (HCWs) vaccinated with Comirnaty (Pfizer-BioNTech) at a tertiary care hospital in Greece. Over an 8-month period, from February to September 2021, 1958 fully vaccinated HCWs were monitored. Rapid antigen tests and RT-PCR tests were conducted weekly for asymptomatic HCWs. Contact tracing and whole-genome sequencing were performed. Results showed that 2.75% (54 cases) of HCWs experienced breakthrough infections. Among these, 25 (45%) were asymptomatic, mild symptoms occurred in 21 (37%), while 7 (13%) had a fever (≥38 °C) alone and 3 (5%) developed high fever (≥39 °C) with respiratory symptoms. Physicians and nursing staff were the most affected groups. Dominant SARS-CoV-2 variants detected included Delta, British, and Wild type variants. Comparison with existing literature underscored the effectiveness of Comirnaty in reducing breakthrough infections. The findings emphasize the importance of continued booster vaccinations and ongoing surveillance to mitigate breakthrough infections among HCWs.

1. Introduction

The COVID-19 pandemic has posed significant global challenges, necessitating effective measures to limit its spread and mitigate its impact. Among the strategies implemented, large-scale vaccination of high-risk groups, followed by broader population coverage, emerged as a critical approach [1,2]. In Greece, vaccination against COVID-19 with Comirnaty (Pfizer-BioNTech, New York, NY, USA) commenced in January 2021 amid a renewed lockdown, prioritizing high-risk individuals and healthcare workers (HCWs). HCW vaccination began in the first week of February 2021 with a two-dose regimen administered three weeks apart.

While vaccination has significantly reduced the risk of severe disease, breakthrough infections (VBT infections) have increasingly been reported, particularly with the emergence of SARS-CoV-2 variants such as Delta and, more recently, Omicron, which exhibit immune evasion potential [3]. However, there remains limited literature specifically assessing the long-term incidence and symptomatology of VBT infections in HCWs, particularly in relation to evolving viral variants.

Therefore, this study aimed to evaluate the incidence and symptom profile of breakthrough COVID-19 infections among fully vaccinated HCWs over an 8-month monitoring period in a tertiary care hospital in Greece. Additionally, we sought to examine potential correlations between the time since vaccination and antibody titers, providing insights into the durability of vaccine-induced immunity.

2. Materials and Methods

The study was conducted at “Hygeia”, a Private Tertiary General Hospital in Athens, Greece, in accordance with the 1964 Declaration of Helsinki standards and with approval from the Hospital Scientific Committee. This was a prospective observational study. “Hygeia’’ has a capacity of 300 beds, including 2 general ICUs with 12 and 14 beds, respectively. However, since January 2020 according to the Greek Ministry of Health instructions, patients suffering from COVID-19 were not hospitalized in private hospitals, but only in Public General Health Care Facilities. The latter were remodeled to organize COVID-19 Clinics under the auspices of Internal Medicine Depts or/and COVID-19 ICUs with specifically trained personnel regarding the needs of the new pandemic as well as for the strict application of protection.

A VBT infection was defined according to the CDC as the detection of the SARS-CoV-2 RNA or antigen in a respiratory specimen collected from a person ≥ 14 days after completion of the two recommended doses of the Comirnaty vaccine [4]. HCWs were considered fully vaccinated against COVID-19 if ≥14 days had lapsed after receiving the 2nd Comirnaty dose.

During the study period, two diagnostic tests were applied by the Central Labs of Hygeia Hospital:

• A Rapid Test for the qualitative detection of SARS-CoV-2 antigen on nasopharyngeal swabs (by Biosynex Swiss SA Illkirch-Graffenstaden France). The Biosynex COVID-19 Antigen BSS (as of February 2021) is a qualitative membrane-based immunoassay that uses colloidal—gold conjugated particles with monoclonal antibodies to detect nucleocapsid protein of SARS–CoV-2. An internal control is included in the assay to indicate the proper volume of sample added to the test. The limit of detection of the assay is 1.15 × 10² TCID50/ML. External validation data reported a sensitivity of 96% and specificity of 98% in clinical settings [5]. It is important to note that the same test kits were consistently used throughout the 8-month study to ensure uniformity in results.
• A real-time PCR (RT-PCR) based on the qualitative determination of SARS-CoV-2 on an RT-PCR method using specific primers for N, E, or RdRp genes of SARS-CoV-2 (Youseq Winchester, UK). The detection limit of this method was 0.58 viral copies /mL of the sampling swab after saline buffer washing. However, the negative result cannot exclude the presence of inhibitors or the existence of a very low percentage of viral RNA below the detection limit of the method. External validation data indicated a sensitivity of 99.5% and a specificity of 99.9% in clinical evaluations [6].

The Rapid Test was conducted weekly for all asymptomatic HCWs, including those with a prior history of COVID-19 infection, as well as for all hospitalized patients. This testing frequency was implemented to promptly identify and isolate asymptomatic carriers, thereby minimizing the risk of nosocomial transmission. Surgeons, anesthesiologists, gastroenterologists (for procedures like colonoscopy and gastroscopy), and pulmonologists (for bronchoscopy and spirometry) underwent additional RT-PCR testing within 24 h before performing these procedures. Hospitalized patients displaying suspicious symptoms but testing negative on the Rapid Test were also subjected to RT-PCR testing. In the Emergency Department, RT-PCR tests were performed on patients requiring hospitalization for life-threatening conditions or presenting symptoms consistent with COVID-19. Individuals with positive test results were either transferred to a public hospital after receiving initial care or isolated at home until a subsequent RT-PCR test confirmed a negative result, based on their clinical condition. In cases of exposure to a confirmed COVID-19 case among HCWs, the Infection Control Committee Head Nurse at ’Hygeia’ hospital conducted meticulous contact tracing to identify and manage potential risks.

SARS-CoV-2 variants derived from HCWs who experienced VBT infections were subjected to whole genome sequencing by applying the CoviDetectTM Assay (PentaBase Aps, DK-5000 Odense, Denmark) for CoV-2 B.1.1.7, B.1.351, P.1 and P.2. When new variants were detected the Allplex-SARS-CoV-2 Master Assay, as well as the Allplex-SARS-CoV-2 Variants I and II Assays (by Seegene Inc, Seoul, Republic of Korea) were applied. The CoviDetectTM Assays are PCR assays intended for the detection of genetic variations from positive SARS-CoV-2 samples followed by a DNA melt analysis. The Allplex-SARS-CoV Assays are multiplex RT-PCR that provide Ct values of multiple targets.

Additionally, antibody testing was performed at the time of the positive COVID-19 test for some HCWs with VBT infections, based on sample availability, but without a predefined selection pattern. Two different types of serum antibodies were detected by the following methodology:

• The RBD-specific antibody detection.

Serum samples were tested using the Elecsys^® Anti-SARS-CoV-2 S (Roche Diagnostics, Basel, Switzerland) reagent on a Cobas e 411 immunoassay analyzer for the semiquantitative detection of TAbs-RBD (IgA, IgM, and IgG) of S1 subunit of SARS-CoV-2 spike protein according to the manufacturer’s instructions. The Elecsys^® Anti-SARS-CoV-2 S is an Electrochemiluminescence Immunoassay (ECLIA), which is based on a double-antigen sandwich Enzyme-linked Immunosorbent Assay (ELISA) methodology. Calibration standards and inter-assay controls were routinely employed to ensure reliability, and the laboratory performing these assays was accredited under ISO 15189 [ISO 15189:2022 Medical laboratories—Requirements for quality and competence, International Organization for Standardization, Geneva, Switzerland]. Values of ≥0.8 U/mL are considered as positive.

• The Anti-RBD-specific neutralization assay

Determination of anti-RBD neutralization titers was conducted using the commercially available and Food and Drug Administration-approved cPassTM SARS-CoV-2 Neutralization Antibody Detection kit (GenScript, Biotech Corporation, Piscataway, NJ, USA) according to the manufacturer’s instructions (as of 20/07/2020). Potential cross-reactivity or false positives were minimal but possible, as indicated by the manufacturer. Calibration standards were used, and the laboratory was accredited under ISO 15189 for these assays. This kit is based on a blocking ELISA using Horseradish Peroxidase conjugated recombinant SARS-CoV-2 RBD fragment and the human Angiotensin-Converting Enzyme 2 (ACE2) receptor protein. The optical density (OD) was measured at 450 nm in the Labtech LT-4500 microtiter plate reader and the percentage of RBD-specific neutralization antibodies was calculated by the following type: Percentage signal inhibition (%) = (1 − OD value of sample/OD value of negative control) x 100. Percentages of ≥30% were considered positive.

Statistical Analysis

Baseline characteristics were summarized using descriptive statistics including mean along with standard deviation for continuous variables and absolute frequency along with percentage for categorical ones. Time gaps are presented as medians and interquartile ranges (IQRs).

Multicollinearity was assessed using the Variance Inflation Factor (VIF), and no variables exceeded the recommended threshold of 5. Linearity in the logit for continuous variables was evaluated using the Box-Tidwell test, confirming that antibody levels and age met this assumption. Goodness-of-fit was checked through the Hosmer–Lemeshow test and ROC curve analysis, with the model demonstrating good discrimination.

Missing values in antibody levels and demographic data were systematically evaluated, and a complete-case analysis approach was used, ensuring that only individuals with complete data were included in the final models. Sensitivity analyses confirmed that this approach did not significantly alter the results.

The statistical significance threshold was predefined at p < 0.05, with results between p = 0.05 and p = 0.1 considered suggestive but not conclusive. Given the multiple subgroup analyses performed, including gender, antibody type, and variant detection, adjustments for multiple comparisons were considered, and Bonferroni correction was applied where necessary. Adjusted significance thresholds for multiple testing were explicitly reported to ensure robust interpretation.

All analyses have been performed using STATA MP13.

3. Results

The distribution of vaccinated personnel across various HCW specialties is presented in

Table 1. Vaccination rates with Comirnaty among HCWs according to their specialty.

Personnel Category	Total No of Personnel	Vaccinated	Non Vaccinated	Vaccination Rate (%, 95% CI)
Physicians	1057	1048	9	99.15% (98.15–99.6%)
Non medical staff *	313	298	15	95.21% (95.21–97.4%)
Nurses	523	468	55	89.48% (87.00–91.8%)
GDA **	157	145	12	92.36% (89.5–94.9%)
Total	2049	1958	91	95.56% (94.9–96.2%)

* administration personnel; ** technicians, sanitation workers, kitchen staff, and housekeeping staff.

. It should be noted that 91 HCWs (4.44%) were unvaccinated, either due to a prior COVID-19 infection, in accordance with the official 2021 guidelines of the Hellenic National Vaccination Committee, or because they were off duty due to underlying health conditions or pregnancy. Furthermore, according to Greek Ministry of Health policies, unvaccinated HCWs were not permitted to work during the study period and were therefore not included in the statistical evaluation. As shown in Table 1, 95.56% of HCWs were fully vaccinated, including 99.15% of medical personnel and 92.36% of general duty assistants (GDAs) working in various hospital departments, such as technical services, sanitation, kitchen, and housekeeping staff.

Regarding demographics, gender distribution varied significantly among healthcare worker categories (p < 0.001). Specifically, the majority of male HCWs were physicians (64%), whereas the majority of female HCWs were nurses (51.7%). Among vaccinated personnel, 54 (2.75%) VBT infections were confirmed through both rapid tests and RT-PCR. The median time between vaccination and infection was 4.4 months (range 2.3 to 6.3 months). Males experienced infection after a shorter time interval compared to females (2.8 months vs. 5.3 months, p = 0.1). In 25 cases (45%), patients were mostly asymptomatic, experiencing only nasal congestion, which was mistakenly attributed to allergic rhinitis. COVID-19 infection in these cases was incidentally diagnosed through weekly Rapid tests conducted for all Hygeia personnel. Mild symptoms were reported by 21 patients (37%), who complained of headache, low-grade fever (37.1–37.4 °C), and fatigue, while 7 patients (13%) had fever (≥38 °C) without other symptoms. Only 3 patients (5%) experienced high fever (≥39 °C), along with cough, dyspnea, and hypoxemia, with abnormal x-ray findings showing scattered opacities. Two of these patients did not require hospitalization and managed their symptoms at home with nasal oxygen and their clinical condition improved quickly. Only one patient required hospitalization receiving high-flow oxygen supplement without the need for mechanical ventilation and being discharged after 3 days.

Table 2. Distribution of 54 confirmed COVID-19 patients by gender, age, and personnel category.

Case No	Gender	Personnel Category	Age	Case No	Gender	Personnel Category	Age	Case No	Gender	Personnel Category	Age
1	M	nurses	51	19	M	doctors	42	37	F	nurses	29
2	F	nurses	42	20	F	nurses	52	38	M	doctors	29
3	F	doctors	65	21	F	GDAs	56	39	F	non medical staff	53
4	M	non medical staff	52	22	F	non medical staff	51	40	F	nurses	35
5	M	doctors	49	23	M	non medical staff	65	41	F	doctors	34
6	M	doctors	49	24	F	nurses	43	42	F	nurses	32
7	M	doctors	61	25	F	nurses	39	43	F	non medical staff	35
8	F	nurses	52	26	M	nurses	26	44	F	nurses	48
9	M	doctors	61	27	M	non medical staff	46	45	F	nurses	27
10	F	doctors	60	28	F	non medical staff	41	46	F	non medical staff	51
11	F	nurses	39	29	M	nurses	38	47	M	doctors	56
12	F	GDAs	53	30	M	doctors	46	48	F	doctors	44
13	M	doctors	48	31	F	nurses	36	49	M	doctors	41
14	F	nurses	62	32	M	GDAs	29	50	F	nurses	49
15	M	doctors	52	33	M	doctors	40	51	M	doctors	53
16	M	doctors	65	34	F	non medical staff	45	52	F	non medical staff	48
17	F	nurses	63	35	M	doctors	65	53	M	GDAs	50
18	M	doctors	48	36	M	nurses	26	54	F	non medical staff	56

GDAs: general duty assistants, non-medical staff: administration personnel; The percentage distribution of vaccine-breakthrough (VBT) cases among healthcare workers (HCWs) shows that physicians accounted for 37.04% of cases, nurses for 35.19%, non-medical staff for 20.37%, and general duty assistants (GDA) for 7.4%, with a total of 54 confirmed VBT cases.

displays the distribution of VBT infections among all HCWs, categorized by their respective specialties. Physicians and nursing staff represented the largest groups affected, accounting for 37.3% and 35.18% of VBT cases, respectively, while administrative staff and GDAs accounted for 20.37% and 7.40% of the affected individuals, respectively. Notably, all three seriously ill HCWs belonged to the nursing staff.

Table 3. Monthly spread of VBT infections among 54 HCWs in relation to the number of COVID-19-positive individuals in the Greek community.

Month (2021)	No of Cases
	HCWs	Community
February	2	34,143
March	5	72,589
April	14	81,344
May	3	57,273
June	1	20,150
July	11	70,850
August	14	94,650
September	6	67,803

highlights the correlation between VBT infections, and the number of COVID-19 cases reported each month in the Greek community. As the number of COVID-19 cases increased, so did the number of VBT infections, and vice versa [data obtained from National Public Health Organization (NPHO)] [7]. None of the VBT infection cases belonged to a high-risk group; however, 21 (38.8%) were over 50 years old, and 9 (16.6%) were between 61 and 65 years old.

Tracking of the victims revealed that there was no high-risk contact inside the Hygeia Hospital, whereas exposure to SARS-CoV-2 occurred: (a) at home from already infected members in their families; (b) through exposure to other colleagues at work who were at the incubation period; (c) after unknown contact during their summer vacation; (d) during social events.

Table 4. Distribution of 24 HCWs by SARS-CoV2 variant, antibody titers, and interval of VBT infection post Comirnaty vaccination.

				Antibodies
Gender	Personnel Category	Age (Years)	Onset of Illness After Vaccination (Days)	AntiSpike (U/mL)	Neutralizing (%)	COVID Variant	Reason for na
M	non medical staff *	52	32	52	na	BRITISH (B.1.1.7)	assay failure
F	physician	60	54	4150	95	BRITISH (B.1.1.7)
M	physician	42	61	1248	93.81	UNKOWN
M	non medical staff *	65	67	973	92.14	WILD TYPE (WIV04)
F	non medical staff *	51	94	320.3	79.05	BRITISH (B.1.1.7)
F	nurse	39	96	3201	92.62	BRASIL (P.1)
M	nurse	26	108	na	na	DELTA (B.1.617.2)	insufficient serum volume
F	nurse	43	116	836.5	87.38	WILD TYPE (WIV04)
M	non medical staff *	46	120	136.4	77.17	DELTA (B.1.617.2)
M	nurse	38	134	1862	90	WILD TYPE (WIV04)
M	physician	46	136	643.5	81.91	DELTA (B.1.617.2)
F	non medical staff *	41	140	665.7	82.14	DELTA (B.1.617.2)
M	physician	40	149	1615	79.05	WILD TYPE (WIV04)
F	non medical staff *	45	151	418.6	59.76	DELTA (B.1.617.2)
M	physician	65	154	913.2	91.91	WILD TYPE (WIV04)
M	physician	29	164	588.8	82.14	DELTA (B.1.617.2)
F	nurse	35	171	2784	97.86	DELTA (B.1.617.2)
M	nurse	32	176	641.1	76.91	WILD TYPE (WIV04)
M	nurse	48	180	91	61.19	DELTA (B.1.617.2)
M	non medical staff *	51	187	25,000	92.5	DELTA (B.1.617.2)
M	physician	41	193	na	na	DELTA (B.1.617.2)	assay failure
F	nurse	49	196	na	na	DELTA (B.1.617.2)	insufficient serum volume
M	GDA **	50	215	772.4	72.72	DELTA (B.1.617.2)
F	non medical staff *	56	218	14,875	90	DELTA (B.1.617.2)

* administration workers; ** sanitation workers, kitchen staff.

details a subgroup of 24 HCWs (among the 54 presented in Table 2), showing the time in days between full vaccination and VBT infection, along with exact variant detection and anti-spike and neutralizing antibody titers. The selection was based on sample availability to ensure representative variant detection. The majority, 13/24 (54.2%), were infected by the Delta variant, with a median time from 14 days post-vaccination to infection of 171 days (range 140–193). Six HCWs (25%) were infected with the Wild type, with a median interval of 133 days (range 67–154), while 3 (12.5%) were infected with the British variant, with a median time of 54 days (range 32–94). One individual (4.1%) was infected with the Brazilian variant, 96 days after full vaccination. For one individual (4.1%), the specific SARS-CoV-2 variant could not be identified. An analysis of the correlation between SARS-CoV-2 variants and both total anti-spike and neutralizing antibody titers in the same group of 24 vaccinated HCWs with VBT infections revealed high antibody levels in all cases, including the three individuals with severe infections. Additionally, as shown in Table 4, the number of VBT infections in this specific subgroup of HCWs, showed a significant increase with the extension of the post-vaccination interval. Doctors and nurses were the most affected groups, each comprising 8 cases, while non-medical staff accounted for 7 cases.

It is noteworthy that, during the spread of the Delta variant, the highest levels of neutralizing antibodies were observed, indicating a strong immune response. Furthermore, throughout the eight-month study period, all measured antibody titters remained significantly protective.

Among the 54 infected HCWs, antibodies against either the spike protein or neutralizing antibodies were evaluated in 26 individuals (48.2%). Both types of antibodies were measured in 21 (38.9%) of the infected HCWs. The median spike antibody level was 654.6 (IQR: 236.5–1615), while the median neutralizing antibody level was 82.1 (IQR: 77.2–92.1). The distribution of neutralizing antibodies appeared to be normal, whereas the distribution of spike antibodies was not normal (Figure 1 and Figure 2).

Notably, logistic regression analysis indicated an association between the detection of neutralizing antibodies and the Delta variant (OR: 0.9, p = 0.1). When adjusted for the co-detection of spike protein antibodies, the odds ratio remained relatively stable (OR: 0.8, p < 0.05). Finally, a multivariable model (

Table 5. Multivariable model assessing different parameters associated with delta variant.

Delta Variant Present	OR	Standard Error	p-Value	[95% CI]
Female	0.6	0.7	0.63	0.1–5.6
Age	0.9	0.1	0.39	0.8–1.1
Neutralizing antibodies	0.9	0.1	0.04	0.7–1.0
Spike protein antibodies	1.0	0.0	0.23	1.0–1.0

p < 0.01, p < 0.05, p < 0.1; This model is interpreted as patients for whom delta was detected vs those who didn’t; p-value = 0.06.

, Figure 3) identified neutralizing antibodies as a key prognostic marker for non-Delta variants, even after adjusting for age, gender, and the detection of spike antibodies.

4. Discussion

Since the World Health Organization declared COVID-19 a pandemic on 11 March 2020, the infection has impacted millions of people worldwide. By July 2023, SARS-CoV-2 had been linked to over 690 million cases and approximately 7 million deaths [8]. However, these numbers would likely have been much higher if not for the rapid development of vaccines, particularly the mRNA vaccines [1]. The first study assessing the safety and efficacy of the Pfizer-BioNTech Comirnaty mRNA COVID-19 vaccine was published on 31 December 2020. This study involved 43,548 participants, with 21,720 receiving the Comirnaty vaccine and 21,728 receiving a placebo. The two-dose regimen demonstrated 95% effectiveness in individuals aged 16 and older (8 cases among vaccinated individuals compared to 162 cases in the placebo group) at least seven days after the second dose and 52% effectiveness starting 12 days after the first dose [2]. However, as the pandemic progressed, it became clear that the primary benefit of the Pfizer and Moderna mRNA vaccines lay in their ability to protect against severe manifestations of the disease. These severe cases often required hospitalization, intubation, and mechanical ventilation, with a mortality rate exceeding 70% during the dominance of the Wuhan, Alpha (English), and Delta variants. Among individuals who had completed the two-dose vaccination regimen, the vaccines offered significantly reduced protection against infection. However, in most cases, these infections manifested as mild symptoms—such as low-grade fever, sore throat, and headache—or were entirely asymptomatic [1].

Considering that individuals with underlying risk factors are at the highest risk of severe or fatal disease, it was deemed essential to prioritize HCWs in high-risk groups for vaccination against SARS-CoV-2. While limited studies have explored vaccine VBT infections among hospital staff, more data is needed to address several key aspects. These include identifying the specific hospital staff groups most vulnerable to VBT infections and determining the critical period of susceptibility by analyzing the time interval between vaccination and VBT occurrence. Further investigation should also examine the relationship between humoral immune responses and VBT infections and the impact of the isolated viral variant on the immune response and disease severity.

The current study was designed to address the previously mentioned issues, albeit with the limitation that patients with COVID-19 were not routinely hospitalized in any private hospital in Greece, including “Hygeia” hospital. Despite this, HCWs in such facilities face numerous opportunities for potential exposure to the virus. High-risk areas include the Emergency Room, Outpatient Clinics, and Private Consulting Rooms where numerous undiagnosed but suspicious cases of COVID-19 are often encountered. Additional areas of potential exposure include Outpatient Radiology Departments, various laboratories, and waiting halls. On the other hand, the duration of efficient immunity in relation to observed VBT infections, particularly among physicians and nursing personnel, is of major importance for assessing the need for a third booster dose of either mRNA vaccine [9].

The current study lasted eight months (1 February until 30 September 2021) and, to our knowledge, represents the longest post-vaccination follow-up study in the literature. During this period, the Delta variant—recognized for its higher transmissibility and virulence compared to earlier variants [10,11,12]—emerged in Greece in March and became dominant from July to December 2021. In this study, among 2049 HCWs, 95.56% were fully vaccinated. Vaccination rates were highest among medical and nursing staff, as well as non-medical personnel (e.g., hospital administrators) and GDAs at 99.15%, 89.48%, 95.21%, and 92.36%, respectively (Table 1). The participants were almost evenly distributed by gender, with 45% men and 55% women. The Hospital Infection Control Committee implemented a comprehensive schedule to minimize the spread of SARS-CoV-2, yielding notable results: only 54 (2.75%) VBT infections were observed among the 1958 fully vaccinated HCWs (Table 2). However, a comparison with non-vaccinated staff that should have been of interest, could not be conducted, as the majority of them were off duty during the study period, following directives from the Greek Ministry of Health.

The first study in HCWs aiming to evaluate the incidence of VBT infections following full vaccination against SARS-CoV-2 was conducted by two Universities of California, San Diego (UCSD), and Los Angeles (UCLA) University of California, mandating all asymptomatic HCWs to undergo weekly nasal PCR testing [2]. Between 1 December 2020, and 9 February 2021, a total of 28,184 HCWs were fully vaccinated with the Comirnaty vaccine. Among these, 8 HCWs tested positive for COVID-19 8–14 days after the second dose, and 7 tested positive ≥15 days post-second dose, resulting in an overall positivity rate of 0.05%. The absolute risk of testing positive was 1.19% at UCSD and 0.97% at UCLA. By the end of the study, 5455 HCWs at UCSD and 9535 at UCLA had received their second dose at least two weeks prior, maintaining the same positivity rate of 0.05% [13]. However, the study’s findings were limited by the short observation period of seven weeks, offering no insights into the long-term efficacy of the vaccines. Notably, these observed positivity rates were higher than those reported in the official trials of the Moderna mRNA-1273 and Pfizer BNT162b2 vaccines [14,15]. This discrepancy could be attributed to factors such as regular testing of both symptomatic and asymptomatic staff, a regional surge in COVID-19 cases, as well as the different demographic characteristics, i.e., HCWs are generally younger and face a higher overall risk of exposure to SARS-CoV-2 compared to clinical trial participants. Importantly, the study highlighted that 57 cases occurred between days 15–21 after the first dose, compared to 30 VBT infections observed during days 1–7 and 8–14 after the second dose. These findings underscore the critical need to maintain health mitigation measures until herd immunity is achieved [13].

Interestingly, in an early short-term post-vaccination study in Israel (20 December 2020–2 January 2021), involving 4081 HCWs prioritized for one dose of Comirnaty, 0.54% (22 individuals) developed COVID-19 within 1–10 days post-vaccination (median of 3.5 days). This underscores the importance of not dismissing early symptoms as vaccine-related without testing, to prevent secondary exposures in hospital settings [16]. Studies conducted during the predominance of the Delta variant revealed a higher likelihood of VBT infections [11]. Although hospitalization due to VBT infections is now infrequent, recent research associates hospitalization risk with factors such as Hispanic ethnicity, obesity, employment type, and comorbidities including asthma, hypertension, and immunosuppression [10].

In our study, the detection of high titers of neutralizing antibodies was associated with a higher likelihood of infection by non-Delta variants, underscoring the critical role of vaccination in combating the Delta variant, which predominated in Greece during the post-vaccination period. A review examining the relationship between neutralizing antibodies produced after vaccination and infection found that the Comirnaty vaccine was highly effective in eliciting these antibodies, and that high neutralizing antibody levels generally correlate with protection against Delta and Omicron subvariants. However, our study was not designed to determine variant-specific vaccine efficacy rates, and we acknowledge that vaccine effectiveness against different variants may vary due to immune evasion mechanisms. Recent evidence suggests that this paradoxical association may be explained by the immune evasion properties of emerging variants. Several studies have documented that variants such as Delta—and more prominently, Omicron—possess multiple mutations in the receptor-binding domain, which can diminish the binding affinity of vaccine-induced antibodies. In some cases, high titers may reflect a robust anamnestic response post-exposure rather than pre-existing protection, indicating that despite a quantitatively strong antibody response, the functional quality of these antibodies in neutralizing variant strains may be compromised. Moreover, robust T-cell-mediated immunity likely plays a critical role in controlling disease severity, which could explain why breakthrough infections remained mostly mild among vaccinated healthcare workers during the Delta predominance [10,17].

The SIREN study [18], a prospective multicenter cohort study conducted from 7 December 2020 to 5 February 2021, in UK hospitals during the dominance of the British variant, evaluated the effectiveness of complete Comirnaty vaccination in HCWs. The study included 23,324 participants from 104 sites across England. Among the unvaccinated cohort, 14 infections per 10,000 person days were reported, compared to 8 VBT infections in the vaccinated cohort. Vaccine effectiveness was estimated at 85% (74–96%) seven days after two doses, with follow-up extending up to two months post-vaccination.

Given the limited information on vaccine effectiveness in HCWs, who represent a high-risk group for SARS-CoV-2 exposure, it is pertinent to highlight the findings of an observational retrospective study conducted at a tertiary medical center in Tel Aviv. This study aimed to assess the association between Comirnaty vaccination and VBT infections in HCWs, with a follow-up period of 63 days after complete vaccination [14]. The study evaluated the rates of symptomatic and asymptomatic infections, recognizing that asymptomatic infections play a significant role in the transmission of SARS-CoV-2, providing a probable explanation for the rapid spread of the pandemic [2,14]. Among 5372 fully vaccinated and 696 unvaccinated HCWs, the study found that Comirnaty vaccination was associated with an adjusted incidence rate ratio of 0.03 for symptomatic infections and 0.14 for asymptomatic infections more than seven days after the second dose. Both incidence rate ratios were statistically significant.

After reviewing the current literature on the incidence of VBT infections in HCWs and the protective efficacy of the Comirnaty mRNA vaccine, the Greek incidence of VBT infections—2.75% (54 HCWs)—after statistical evaluation, appears to align well with, or even be lower than, the rates reported in other studies. The relatively lower rate of VBT infections observed in the Greek study can likely be attributed to several factors:

• COVID-19 patients were not hospitalized in private hospitals in Greece during the eight-month study period (1 February–30 September 2021), which may have reduced HCWs’ exposure to the virus. Prolonged intrafamilial exposure, particularly during the lockdown, and participation in crowded social events after lockdown restrictions were ended were identified as contributing factors to VBT infections.
• A comprehensive screening strategy was implemented to detect infections early. This included weekly rapid antigen testing for all asymptomatic personnel and hospitalized patients, weekly RT-PCR testing for physicians working in Emergency Rooms, surgeons, and anesthesiologists, and additional RT-PCR testing for any HCW exhibiting COVID-19 symptoms. This approach minimized exposure to asymptomatic HCWs in the incubation period and asymptomatic COVID-19-positive hospitalized patients admitted for non-COVID-related conditions. This is particularly important given that asymptomatic individuals can carry viral loads comparable to symptomatic cases, making them a significant source of virus transmission. However, studies on viral dynamics suggest that viral loads decline more rapidly in vaccinated individuals, and the virus they shed is less likely to be culture-positive. As a result, fully vaccinated individuals are not only at a lower risk of infection but, if infected, have a shorter period of contagiousness [19,20].
• The absence of significant underlying risk factors among the study group, except for five HCWs aged 61–65 with potentially higher risk, likely contributed to the findings. Notably, aside from three HCWs who experienced high fever and showed positive chest X-rays—one of whom required brief hospitalization—all others recovered at home and tested negative. Age distribution at least in those HCWs who became positive in the first three months post-vaccination seems to correlate with a threshold of ≥50 years of age.
• Since July 2021, the Delta variant became the predominant strain, yet no significant differences were observed in the clinical presentation of infected HCWs, who generally experienced mild symptoms. This supports that the Comirnaty vaccine was equally effective against the Wild, British, and Delta variants [21].
• As observed in this study and corroborated by other researchers, the incidence of VBT infections increased as the time since the administration of the second vaccine dose lengthened, with a notable threshold emerging around seven months (Table 4), despite that, the levels of both IgG anti-spike and neutralizing antibodies remained consistently high, showing no significant variation beyond the seven-month mark and being evenly elevated among physicians, nurses, and other medical personnel. Notably, previous studies have indicated that anti-spike IgG levels exceeding 1000 AU/mL are predictive of infection within the preceding three months, while levels below this threshold demonstrated a 99.5% negative predictive value for COVID-19 infection during the same period [22]. Thus, anti-IgG serology can be a valuable tool for confirming or excluding prior infections within the previous three months when assessing vaccination timing. It is important to note that VBT infections, as observed in this study and reported elsewhere, have also been documented within the initial weeks after vaccination. Furthermore, with the emergence of the BA2 and BA5 Omicron variants—against which the effectiveness of both mRNA vaccines is reduced—it remains imperative to maintain preventive measures. Social distancing, mask-wearing, and regular handwashing are essential to mitigate the risk of COVID-19 infections caused by these variants among healthcare workers.

It is crucial to further examine VBT infections to better understand the interplay between neutralizing antibodies, immune evasion by variants, and cellular immunity. Over time, the emergence of new variants, the development of updated vaccines, and longer follow-up data will enhance our understanding of the mechanisms driving VBT infections and the factors influencing symptom severity.

5. Limitations

This study has several limitations. As COVID-19 patients were not hospitalized in private hospitals in Greece, including “Hygeia” hospital, HCWs may have had lower exposure risk than those in public hospitals, potentially influencing breakthrough infection rates. Despite weekly testing, asymptomatic or transient infections may have been underestimated due to short-lived viral shedding. The eight-month follow-up, though one of the longest available, may not fully capture long-term vaccine efficacy, especially against newer variants like Omicron. Potential biases include recall bias in self-reported symptoms, selection bias in the subgroup analysis (Table 4), and underreporting of mild or asymptomatic cases. Additionally, the analysis of antibody levels and their correlation with variants was performed on a small subgroup (24 HCWs), as sample availability limited testing for all cases. This may impact the generalizability of our findings, and future studies with larger datasets are needed to confirm these observations. Longer follow-up studies are needed to assess vaccine effectiveness and breakthrough infections in HCWs.

6. Conclusions

As the novel introduced monovalent anti-XBB.1.5 COVID-19 mRNA vaccine is already on the market, specifically engineered to provide protection against the Omicron BA2 variants, it is clear—based on the literature reviewed and the findings of this study—that vaccination conferring significant benefits to HCWs should be continued. These benefits include a reduction in the incidence and milder symptomatology of VBT infection. This highlights the ongoing necessity of vaccination, even in the current context where the WHO has declared the pandemic over. For high-risk groups, including HCWs, vaccination remains an essential measure to safeguard their health and maintain the resilience of healthcare systems. Additionally, booster doses are crucial to sustaining immunity, and continued surveillance of breakthrough infections is necessary to adapt vaccination strategies and ensure long-term protection for HCWs