Dear Sir,
In this Journal, Gidari and colleagues showed that sera from BNT162b2 vaccinated humans had lower neutralization potency on the SARS-CoV-2 Gamma variant and that convalescent patients that had been infected by the Gamma variant were less protected from other SARS-CoV-2 strains.1
The Gamma variant has been considered the predominant SARS-CoV-2 lineage in Brazil during the first half of 2021.2 In our tertiary-care hospital in São Paulo, Brazil, all healthcare workers (HCW) with suspected COVID-19 are clinically evaluated by the HCW Health Services and tested using RT-PCR for SARS-CoV-2. We aimed to characterize the clinical features of COVID-19 caused by the Gamma variant in comparison with strains that were not variants of concern (non-VoC).
In this prospective cohort we included symptomatic COVID-19 cases among HCW from January 22 to May 15, 2021. All positive samples for SARS-CoV-2 RT-PCR underwent whole genome sequencing using the MinION platform (additional details on supplementary material).3 , 4 Clinical data were collected using a structured data formulary.
Proportions and medians were compared using the chi-square test and Mann-Whitney U test. Risk factors for Gamma variant infection were evaluated using a multivariate logistic regression with COVID-19 immunization, previous COVID-19, age, and month of diagnosis as independent variables. Risk factors for the presence of COVID-19 symptoms on diagnosis were evaluated using a multivariate logistic regression including Gamma variant infection, COVID-19 immunization, previous COVID-19, duration of symptoms on diagnosis, sex, and age as independent variables. Since we evaluated risk factors for 11 symptoms, these analyses were adjusted for multiple comparisons according to Bonferroni's correction. A HCW was considered to be fully immunized after at least 14 days after the second dose of any COVID-19 vaccine. Statistical analyses were two-tailed with an alpha error of 0.05. The software SPSS (version 17.0) and R (version 4.1.0) were used for the analyses. This study was approved by the Hospital's Ethics Committee (CAAE:42,708,721.0.0000.0068).
During the study period there were 523 symptomatic COVID-19 cases among HCW and 423 of them were included in the study, of which 415 (98%) had mild disease (Table 1 and Supplementary Fig. 1). The median (25th–75th percentile) age of our study group was 38 (29–48) years, 70% were female. Of the 175 (41%) fully immunized patients, 173 (99%) had received CoronaVac with a median (25th–75th percentile) interval between the second dose of the vaccine and onset of symptoms of 61 (37–75) days. Seventeen patients had previous COVID-19 with a median (25th–75th percentile) interval between the onset of symptoms of each episode of 245 (230–276) days. The most prevalent symptoms were coryza (73%) and headache (72%).
Table 1.
Clinical characteristics of 423 healthcare workers with COVID-19, and bivariate analysis comparing cases caused by the Gamma variant with cases caused by strains that are not variants of concern (non-VoC) (Hospital das Clínicas, São Paulo, Brazil. January 22 - May 15, 2021).
| Gamma variant (N = 313) | Non-VoC (N = 110) | pa | |
|---|---|---|---|
| N (%) or median (25th–75th percentile) | |||
| Age (years) | 37 (29–49) | 39 (31–46) | 0.452 |
| Age strata (years) | 0.890 | ||
| 16–40 | 189 (60%) | 67 (61%) | |
| 41–60 | 106 (34%) | 38 (35%) | |
| > 60 | 18 (6%) | 5 (5%) | |
| Female | 220 (70%) | 76 (69%) | 0.814 |
| Pre-existing conditions | |||
| Hypertension | 35 (11) | 12 (11) | 0.938 |
| Hyperlipidemia | 17 (5) | 7 (6) | 0.716 |
| Chronic pulmonary disease | 16 (5) | 2 (2) | 0.176 |
| Type 2 diabetes mellitus | 15 (5) | 4 (4) | 0.791 |
| Heart disease | 14 (5) | 6 (6) | 0.676 |
| Obesity | 7 (2) | 1 (1) | 0.686 |
| Chronic kidney disease | 3 (1) | 0 | 0.571 |
| Immunosuppressive drug use | 2 (1) | 2 (2) | 0.278 |
| Cancer | 2 (1) | 1 (1) | 1.000 |
| Solid-organ transplantation | 1 (0) | 0 | 1.000 |
| Pregnancy | 4/220 (2) | 1/76 (1) | 0.932 |
| Work category | 0.628 | ||
| Nursing technician | 79 (25) | 25 (23) | |
| Administrative staff | 66 (21) | 23 (21) | |
| Physician | 58 (18) | 16 (14) | |
| Auxiliary health teamb | 43 (14) | 20 (18) | |
| Nurse | 27 (9) | 11 (10) | |
| Multidisciplinary health teamc | 18 (6) | 10 (9) | |
| Others | 22 (7) | 5 (5) | |
| Vaccination status and type of vaccine | < 0.001 | ||
| Unvaccinated | 44 (14) | 22 (20) | |
| Partially immunized d | 99 (32) | 83 (75) | |
| CoronaVac | 71 (72) | 81 (98) | |
| ChAdOx1 | 19 (19) | 1 (1) | |
| Unknown | 9 (9) | 1 (1) | |
| Fully immunized e | 170 (54) | 5 (5) | |
| CoronaVac | 168 (98) | 5 (100) | |
| ChAdOx1 | 1 (1) | 0 | |
| BNT162b2 | 1 (1) | 0 | |
| Symptoms on diagnosis | |||
| Coryza | 236 (75) | 71 (65) | 0.028 |
| Headache | 230 (74) | 76 (69) | 0.376 |
| Cough | 227 (73) | 74 (67) | 0.296 |
| Sore throat | 170 (54) | 52 (47) | 0.203 |
| Myalgia | 159 (51) | 43 (39) | 0.034 |
| Asthenia | 137 (44) | 43 (39) | 0.393 |
| Fever | 94 (30) | 43 (39) | 0.81 |
| Hyposmia/anosmia f | 78 (25) | 51 (46) | < 0.001 |
| Dysgeusia f | 66 (21) | 42 (38) | < 0.001 |
| Gastrointestinal tract symptomsg | 63 (20) | 33 (30) | 0.033 |
| Dyspnea | 31 (10) | 8 (7) | 0.412 |
| Duration of symptoms (days) | 3 (2 - 4) | 3 (2–4) | 0.823 |
| COVID-19 diagnosed previously | 7 (2) | 10 (9) | 0.002 |
| Previous positive RT- PCR | 3 (43) | 0 | |
| Previous positive serology | 4 (57) | 10 (100) | |
| Hospitalization | 5 (2) | 3 (2) | 0.434 |
| Oxygen supplementation | 4 (1) | 3 (3) | 0.383 |
| Admission to intensive care unit | 3 (1) | 2 (2) | 0.608 |
| Invasive mechanical ventilation | 3 (1) | 0 | 0.571 |
| Vasoactive drug support | 2 (1) | 0 | 1.000 |
| Thromboembolic event | 2 (1) | 1 (1) | 0.772 |
| Death | 3 (1) | 0 | 0.571 |
Calculated by Chi-square test or Mann Whitney U test.
Included workers of pharmacy, laboratory, radiology, security, and medical students.
Included physiotherapist, dentist, psychologist, nutritionist, and social services.
Recipients of at least one dose of a COVID-19 vaccine and recipients of 2nd dose in the last 13 days.
Recipients of two doses of a COVID-19 vaccine at least 14 days after the second dose.
The simultaneous dysfunction of taste and smell was present in 93 (30%) Gamma variant cases and 58 (53%) non-VoC cases.
Abdominal pain, diarrhea, nausea, or vomiting.
Among the 423 SARS-CoV-2 isolates, 313 (74%) were of the Gamma variant and 110 (26%) were non-VoC. The distribution of SARS-CoV-2 lineages varied over the study period with predominance of non-VoC cases in the first months and a posterior increase in cases caused by the Gamma variant (Supplementary Fig. 2).
Although in the bivariate analysis COVID-19 immunization was associated with infection by the Gamma variant, in the multivariate analysis immunization status was not associated with Gamma variant infection after adjustment for the month of occurrence of COVID-19, age, and history of previous COVID-19 (Supplementary Table 1 and Fig. 3). The month in which COVID-19 occurred was the only factor associated with having an infection by the Gamma variant.
In the unadjusted bivariate analysis, coryza and myalgia were more frequent in Gamma variant cases; and hyposmia/anosmia, dysgeusia, and gastrointestinal tract symptoms in non-VoC cases. However, in the multivariate analysis after adjustment for multiple comparisons, hyposmia/anosmia (OR = 0.304, adj p < 0.001) and dysgeusia (OR = 0.385, adj p = 0.011) were the only symptoms significantly associated with the Gamma variant (Table 2 and Supplementary Table 2).
Table 2.
Evaluation of factors associated with hyposmia/anosmia and dysgeusia in 423 healthcare workers with COVID-19 (Hospital das Clínicas, São Paulo, Brazil. January 22 - May 15, 2021).
| Hyposmia/anosmia |
Dysgeusia |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Bivariate analysis |
Multivariate analysis |
Bivariate analysis |
Multivariate analysis |
|||||||||
| OR(95%CI)a | p | Adj pb | OR(95%CI)a | p | Adj pb | OR(95%CI)a | p | Adj pb | OR(95%CIa | p | Adj pb | |
| Gamma variant | 0.433(0.270–0.693) | < 0.001 | < 0.001 | 0.303(0.176–0.522) | < 0.001 | < 0.001 | 0.384(0.244–0.605) | < 0.001 | < 0.001 | 0.385(0.221–0.672) | 0.001 | 0.011 |
| Previous COVID-19 | 1.630(0.606–4.381) | 0.333 | 1.000 | 1.317(0.465–3.734) | 0.604 | 1.000 | 1.626(0.586–4.507) | 0.350 | 1.000 | 1.321(0.458–3.806) | 0.607 | 1.000 |
| Fully immunizedc | 0.856(0.561–1.306) | 0.470 | 1.000 | 1.657(0.989–2.779) | 0.055 | 0.605 | 0.784(0.500–1.230) | 0.290 | 1.000 | 1.334(0.779–2.286) | 0.294 | 1.000 |
| Duration of symptoms | 1.084(0.990–1.186) | 0.081 | 0.891 | 1.119(1.020–1.227) | 0.017 | 0.187 | 1.097(1.000–1.204) | 0.50 | 1.000 | 1.125(1.023–1.236) | 0.015 | 0.165 |
| Female | 1.250(0.801–1.951) | 0.326 | 1.000 | 0.802(0.503–1.279) | 0.354 | 1.000 | 0.975(0.605–1.573) | 0.918 | 1.000 | 1.071(0.652–1.759) | 0.788 | 1.000 |
| Age strata (years) | ||||||||||||
| 16–40 | 1 (ref) | 1 (ref) | 1 (ref) | 1 (ref) | ||||||||
| 41–60 | 0.734(0.468–1.151) | 0.178 | 1.000 | 0.639(0.397–1.026) | 0.064 | 0.704 | 0.821(0.512–1.318) | 0.414 | 1.000 | 0.730(0.446–1.196) | 0.212 | 1.000 |
| >60 | 0.416(0.137–1.261) | 0.121 | 1.000 | 0.281(0.083–0.947) | 0.041 | 0.451 | 0.559(0.184–1.702) | 0.306 | 1.000 | 0.408(0.121–1.377) | 0.149 | 1.000 |
Odds ratio (95% confidence interval).
Adjusted p-value for multiple comparisons (11 symptoms) according to Bonferroni's correction.
Recipients of two doses of a COVID-19 vaccine as of 14 days after the second dose.
We showed important differences in the clinical presentation between Gamma variant and non-VoC infection with a decreased frequency of hyposmia/anosmia and dysgeusia in Gamma variant cases. The increased occurrence of taste and smell disorders in COVID-19 has been considered a useful tool for the clinical triage of respiratory infections during the pandemic, with a frequency varying from 5 to 87% in COVID-19.5, 6, 7 Thus, COVID-19 caused by the Gamma variant may present more often with cold-like symptoms. It has also been described that hyposmia/anosmia and dysgeusia are more common among younger and female patients.6 , 8 However, there was no association between these symptoms and age or sex in our study.
The increase in the proportion of COVID-19 cases caused by the Gamma variant in early 2021 was temporally associated with the beginning of the vaccination campaign in Brazil.2 This context raised the concern that Gamma variant could evade previous SARS-CoV-2 immune response, as suggested in a previous in vitro study.9 Our results do not corroborate this hypothesis since we found no association between immunization status or previous COVID-19 diagnosis, and Gamma variant infection. However, our study does not have enough power to analyze this association due to restricted circulation of Gamma variant soon after the vaccination campaign.
The rise in Gamma variant cases in Brazil was also coincident with an increased COVID-19 incidence in younger patients.10 Nonetheless, there was no association between age and Gamma variant infection in our study. This difference is probably explained by the COVID-19 vaccination rollout in Brazil, which prioritized elderly and HCW regardless of age.
Our study has limitations. Our results might not be generalizable to severe cases of COVID-19 since our cohort comprised mainly mild cases. The temporal association between the vaccination campaign and the rise in Gamma variant cases might have impaired the evaluation of the association between vaccination status and the occurrence of Gamma variant infection by unadjusted or residual confounders.
In conclusion, COVID-19 caused by the Gamma variant presents with different symptoms compared to non-VoC infection. The increase in Gamma variant cases should raise the awareness that COVID-19 may present more often with cold-like symptoms because of a decrease in the frequency of hyposmia/anosmia and dysgeusia.
Declaration of Competing Interest
None.
Acknowledgments
We thank all the physicians of the HCW Health Services for helping with the collection of the clinical data and the LIM-46 team of the Instituto de Medicina Tropical da Faculdade de Medicina da Universidade de São Paulo, Brazil, for performing the SARS-CoV-2 whole genome sequencing. This study was supported by the Itau Unibanco “Todos pela saúde” program.
Footnotes
Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jinf.2021.09.005.
Appendix. Supplementary materials
References
- 1.Gidari A., Sabbatini S., Bastianelli S., Pierucci S., Busti C., Monari C., et al. Cross-neutralization of SARS-CoV-2 B1.1.7 and P.1 variants in vaccinated, convalescent and P.1 infected. J Infect. 2021 doi: 10.1016/J.JINF.2021.07.019. [Preprint] In press. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Silva J.R.D, Pena L. Collapse of the public health system and the emergence of new variants during the second wave of the COVID-19 pandemic in Brazil. One Health. 2021;13 doi: 10.1016/j.onehlt.2021.100287. (Amsterdam, Netherlands) [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tyson J.R., James P., Stoddart D., Sparks N., Wickenhagen A., Hall G., et al. Improvements to the ARTIC multiplex PCR method for SARS-CoV-2 genome sequencing using nanopore. [Preprint] bioRxiv. 2020 doi: 10.1101/2020.09.04.283077. [DOI] [Google Scholar]
- 4.Pater A.A., Bosmeny M.S., Parasrampuria M., Eddington S.B., Ovington K.N., White A.A., et al. High throughput nanopore sequencing of SARS-CoV-2 viral genomes from patient samples. [Preprint] Available at: bioRxiv 2021.02.09.430478; doi: 10.1101/2021.02.09.430478 [DOI] [PMC free article] [PubMed]
- 5.Bénézit F., Le Turnier P., Declerck C., Paillé C., Revest M., Dubée V., et al. RAN COVID Study Group Utility of hyposmia and hypogeusia for the diagnosis of COVID-19. Lancet Infect Dis. 2020;20:1014–1015. doi: 10.1016/S1473-3099(20)30297-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zahra S.A., Iddawela S., Pillai K., Choudhury R.Y., Harky A. Can symptoms of anosmia and dysgeusia be diagnostic for COVID-19? Brain Behav. 2020;10(11):e01839. doi: 10.1002/brb3.1839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lechien J.R., Chiesa-Estomba C.M., Hans S., Barillari M.R., Jouffe L., Saussez S. Loss of smell and taste in 2013 European patients with mild to moderate COVID-19. Ann. Intern. Med. 2020;173:672–675. doi: 10.7326/M20-2428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Lee Y., Min P., Lee S., Kim S.W. Prevalence and duration of acute loss of smell or taste in COVID-19 patients. J Korean Med Sci. 2020;35(18):e174. doi: 10.3346/jkms.2020.35.e174. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Souza W.M., Amorim M.R., Sesti-Costa R., Coimbra L.D., Brunetti N.S., Toledo-Teixeira D.A., et al. Neutralisation of SARS-CoV-2 lineage P.1 by antibodies elicited through natural SARS-CoV-2 infection or vaccination with an inactivated SARS-CoV-2 vaccine: an immunological study. Lancet Microbe. 2021 doi: 10.1016/S2666-5247(21)00129-4. [published online ahead of print, 2021 Jul 8] [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Boletim Observatório COVID-19 – Fiocruz, Brasil. Pages 7–9. Available at: https://portal.fiocruz.br/documento/boletim-do-observatorio-covid-19-semanas-epidemiologicas-16-e-17-de-2021 (Accessed July 20, 2021).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.