Saliva Could Hold Clues To How Sick You Will Get From COVID-19

Authors: Robert Service Science January 14, 2022

Science‘s COVID-19 reporting is supported by the Pulitzer Center and the Heising-Simons Foundation.

To the known risk factors for developing severe COVID-19—age, male sex, or any of a series of underlying conditions—a new study adds one more: high levels of the virus in your saliva. Standard COVID-19 tests sample the nasal passage. But several new tests look for SARS-CoV-2, the pandemic coronavirus, in saliva, and the new work finds a striking correlation between high virus levels there and later hospitalization or death. If the results are confirmed, saliva tests could help doctors prioritize which patients in the early stages of the disease should receive medicines that drive down levels of the virus.

“I thought it was pretty striking,” says Shane Crotty, a virologist at the La Jolla Institute for Immunology, who was not involved with the research. Crotty notes the results suggest virus levels in saliva reflect viral load deep in the lungs, where the disease does much of its damage in severe cases. “That is a fundamentally valuable insight,” Crotty says.

The new work isn’t the first to link the body’s coronavirus load and disease outcome. Several research groups have found a correlation between high viral levels in the nasal passages at the time of a patient’s hospital admission and ultimate disease severity. But other groups have failed to find that same link.

The standard test to detect SARS-CoV-2 samples nasal mucus using nasopharyngeal (NP) swabs. The procedure is unpleasant, but it is the customary way to sample respiratory pathogens. In recent months, however, several research groups have developed and received emergency use authorization from the U.S. Food and Drug Administration for tests detecting SARS-CoV-2 in saliva.

Yale University researchers were among the first, and the university’s hospitals have been using both saliva and NP swab tests. In both cases, labs analyze the samples using quantitative reverse transcription polymerase chain reaction tests, which can detect genetic material from SARS-CoV-2 and quantify the number of viral particles in each milliliter of sample.

Researchers led by Akiko Iwasaki, an immunologist at Yale, compared viral loads in saliva and NP swabs from 154 patients and 109 people without the virus. They divided the patients into groups that had low, medium, and high viral loads as determined by both types of test. Then they compared those results with the severity of symptoms the patients developed later.

They found that patients who developed severe disease, were hospitalized, or died were more likely to have had high virus loads in their saliva tests, but not in their NP swabs. Viral load in both saliva and nasal mucus declined over time in patients who recovered, but not in those who died.

When Iwasaki and her colleagues reviewed patients’ electronic medical records for markers of disease in the blood, they found that high saliva viral loads correlated with high levels of immune signals such as cytokines and chemokines, nonspecific molecules that ramp up in response to viral infections and have been linked to tissue damage. People with more virus in their saliva also gradually lost certain cells that mount an immune response against viral targets, had lower levels of antibodies targeting the spike protein that the virus uses to enter cells, and were slower to develop the strong immune response needed to knock down the virus in cases where they recovered. The team’s results appeared on 10 January in a preprint that has not been peer reviewed.

Iwasaki and her colleagues argue that saliva may be a better predictor of disease outcome than nasal mucus because the latter comes from the upper respiratory tract, whereas severe disease is associated with damage deep in the lungs. “Saliva may better represent what is going on in the lower respiratory tract,” Iwasaki says, because cilia lining the respiratory tract naturally move mucus up from the lungs into the throat, where it mixes with saliva; coughs have the same effect.

The results don’t have enough statistical power to reveal how much more likely a person with a high saliva viral load is to develop severe COVID-19, Iwasaki says. She is also eager for other groups to replicate the results, especially because efforts to link high NP swab viral loads with disease progression have had mixed results.

If other research confirms the finding, “it would clear away a lot of the fog” around this disease, Crotty says. Monica Gandhi, an infectious disease expert at the University of California, San Francisco, adds that if saliva tests are predictive, they could help doctors identify patients to treat early with either antibodies to reduce viral load or steroids to tamp down overactive nonspecific immune responses.

Saliva tests are cheaper and easier than NP tests, but much less widely available. So confirmation of the new results could bolster efforts to make saliva tests more readily available, says Sri Kosuri, CEO of Octant, Inc., a biotech company. “If this study happened in March, we’d be talking about whether we should be doing NP testing at all,” Kosuri says.

COVID-19 Update: The connection between local and global issues–the Pulitzer Center’s long standing mantra–has, sadly, never been more evident. We are uniquely positioned to serve the journalists, news media organizations, schools, and universities we partner with by continuing to advance our core mission: enabling great journalism and education about underreported and systemic issues that resonate now–and continue to have relevance in times ahead. We believe that this is a moment for decisive action. 

Digit ratios and their asymmetries as risk factors of developmental instability and hospitalization for COVID-19

Authors: A. Kasielska-TrojanJ. T. ManningM. JabłkowskiJ. Białkowska-WarzechaA. L. Hirschberg & B. Antoszewski  Scientific Reports volume 12, Article number: 4573 (2022) Cite this article Article Open Access Published: 


COVID-19 presents with mild symptoms in the majority of patients but in a minority it progresses to acute illness and hospitalization. Here we consider whether markers for prenatal sex hormones and postnatal stressors on developmental instability, i.e. digit ratios and their directional and unsigned asymmetries, are predictive of hospitalization. We focus on six ratios: 2D:3D; 2D:4D; 2D:5D; 3D:4D; 3D:5D; 4D:5D and compare hospitalized patient and control means for right, and left ratios, directional asymmetries (right–left) and unsigned asymmetries [|(right–left)|]. There were 54 patients and 100 controls. We found (i) patients differed in their digit ratios from controls (patients > controls) in all three ratios that included 5D (2D:5D, 3D:5D and 4D:5D) with small to medium effect sizes (d = 0.3 to 0.64), (ii) they did not differ in their directional asymmetries, and (iii) patients had greater |(right–left)| asymmetry than controls for 2D:4D (d = .74) , and all ratios that included 5D; 2D:5D (d = 0.66), 3D:5D (d = .79), 4D:5D (d = 0.47). The Composite Asymmetry of the two largest effects (2D:4D + 3D:5D) gave a patient and control difference with effect size d = 1.04. All patient versus control differences were independent of sex. We conclude that digit ratio patterns differ between patients and controls and this was most evident in ratios that included 5D. Large |(right–left)| asymmetries in the patients are likely to be a marker for postnatal stressors resulting in developmental perturbations and for potential severity of COVID-19.


Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) causes a respiratory and systemic illness (COVID-19) which may present as a severe pneumonia in 10–15% of patients. Severe disease can lead to acute respiratory distress and multi-organ failure often followed by intravascular coagulopathy1,2. Due to this variety and unknown severity and death risk factors, many studies and analyses have focused on identifying biomarkers of severe disease or poor outcomes in COVID-19 infections. Recent studies have shown that the clinical progress could be severe in cases of increased: neutrophil-lymphocytes ratio, C-reactive protein (CRP), troponin I, lactate dehydrogenase and that the troponin I, elder age and SO2 values are linked to in-hospital mortality. Across nations, there is variation in case fatality rates and in predictors of mortality3. For example, data from Belgium indicated severity was associated with older age, renal insufficiency, higher lactate dehydrogenase and thrombocytopenia and obesity4. Patterns of severity from Chinese studies included higher age, male sex, higher Body Mass Index, hypertension, lower T lymphocyte and B lymphocyte count, higher white blood cell count, higher D2 dimer, procalcitonin, CRP and aspartate aminotransferase. Among these variables age and weight appeared to be independent risk factors for disease severity5. Importantly, identifying these risk factors did not significantly change our understanding of the COVID-19 pandemic nor did it facilitate a reduction in mortality.

In many populations the severity of COVID-19 is sex dependent (males > females)6. The excess of male deaths has led to two opposing suggestions: (i) The androgen-driven COVID-19 pandemic theory7,8, and (ii) the male hypogonadism theory9. With regard to support for the former, viral entry to cells is androgen dependent, involving priming of the spike proteins and cleaving of angiotensin converting enzyme 2 (ACE2). Both processes are facilitated by trans-membrane protease, serine 2 (TMPRSS2)10. Androgen receptor activity is a requirement for the transcription of the TMPRSS2 gene, suggesting that testosterone facilitates SARS-CoV2 cell entry11. Thus the androgen-driven COVID-19 pandemic theory postulates that high mortality from SARS-CoV2 in men is related to hypergonadism. In contrast, proponents of the male hypogonadism theory point to theory-inconsistent relationships between testosterone and COVID-19 in males9. Thus, in men COVID-19 mortality rates increase with age but testosterone levels decrease12. The male hypogonadism theory gave a rationale for the analyses conducted by Manning and Fink9 who considered national values of digit ratios, in this case 2nd to 4th digit ratio, in relation to Covid-19 case fatality rates (CFR’s). The relative lengths of the second digit and fourth digit (digit ratio or 2D:4D) is sexually dimorphic (2D:4D males < 2D:4D females). It has been suggested that 2D:4D is a biomarker of prenatal sex steroids exposure, i.e. low 2D:4D correlates with high prenatal testosterone and low oestrogens, while high 2D:4D correlates with low foetal testosterone and high oestrogen. There is considerable support for the link between 2D:4D and prenatal sex steroids13 but for a contrasting view see McCormick & Carre, 202014. Manning and Fink found that nations with high CFR’s had high mean male 2D:4D9, thus supporting the hypogonadal theory (in support see Sahin, 2020 and for a critical view see Jones et al., 202015,16). With regard to right–left asymmetry of 2D:4D, i.e. directional asymmetry of 2D:4D, (Δ r–l 2D:4D) is also thought to be a negative correlate of high prenatal testosterone and low prenatal oestrogen9,17,18,19,20,21. In general unsigned asymmetries (such as that of single digit R–L asymmetries) may reflect developmental instability related to postnatal stressors including sex steroids and to correlates of low socio-economic status such as poor nutrition22,23.

Sex differences in digit ratios, with males < females, are present across a number of digits24,25,26,27. Here we focus on six ratios from digits 2 to 5, i.e. 2D:3D, 2D:4D, 2D:5D, 3D:4D, 3D:5D and 4D:5D (digit 1 is difficult to measure accurately). There is considerable evidence that prenatal sex steroids have an effect on 2D:4D. However, effect sizes for 2D:4D are likely to be linked to other ratios that share 2D or 4D. Right 2D:4D is stable during growth in children and adolescents supporting the contention that it retains information pertaining to prenatal sex steroids. However, other ratios, such as those that include 5D (and in particular 3D:5D), show sex differences (males < females) but are highly unstable during growth in children and adolescents. This instability is present in both hands but is expressed most intensely in the left hand24,27. The difference in stability across right and left hands suggests that R–L differences in ratios may contain important information which pertains to developmental instability rather than effects of prenatal sex steroids. Therefore, for each ratio we consider values from the right hand, left hand, Δ right–left (directional asymmetry) and |(right–left)|(FA). The purpose of this preliminary report was to focus on associations between digit ratios and severity of COVID-19, as evidenced by hospitalization of patients.

Following from the across-nation correlations between digit ratio and CFR’s we suggest that, in comparison to controls, patients hospitalized for COVID-19 will have: (i) high right and left hand digit ratios, and high Δ right–left directional asymmetry, indicating exposure to low prenatal testosterone and high prenatal oestrogen, and (ii) high |(right–left)| unsigned asymmetry (FA), indicating heightened levels of developmental instability arising from stressors such as pubertal sex steroids. With regard to these predictions we emphasize that there is potential for considerable inter-correlation between digit ratios. In this regard, 2D:4D has been shown to exhibit developmental stability while 3D:5D is particularly unstable during development24,27. Therefore, the patterns associated with 2D:4D and 3D:5D are least likely to be affected by inter-correlations between digit ratios. Thus, 2D:4D may contain information concerning prenatal influences and 3D:5D information concerning postnatal effects of developmental instability. Consequently, we suggest these two digit ratios should be the focus of greatest attention.


Participants were recruited from a Department of Infectious Diseases and Liver Diseases of a Medical University. All consecutive patients with diagnosed COVID-19 who were hospitalized in the Department due to the severe or high risk of severe COVID-19 were included. During a first wave of the Covid-19 pandemic (March–August 2020) there were 54 (28 men and 26 women) patients who met the study criteria (Inclusion Criteria: 1. admitted to hospital because of Covid-19, 2. positive PCR test, 3. conscious and able to give informed written consent for participation; Exclusion Criteria: 1. unconscious, unable to give written consent for participation, 2. Covid-19 positive patients hospitalized because of other than Covid health issues, pregnant women, children (< 18 years), patients after transplantations, during immunotherapy and with renal failure requiring dialyses). Of these, there were 51 for whom the right hand ratios could be measured, 52 for the left hand and 49 for whom R–L measurements were possible (one hand only was available for measurement for 5 patients due to a hand injury and/or finger contractures). The protocol of the study included a clinical questionnaire based on medical records (age, symptoms) severity of the disease (scale 0–4; 0 -no symptoms, 1—mild, 2—medium, 3—severe, 4—critical), length of hospitalization and oxygen therapy, days in intensive care unit, concomitant diseases, history of smoking and occupational exposure, and laboratory test results (white blood count, fibrinogen, d-dimers, platelets count, oxygen saturation, procalcitonin) and anthropometric measurements.

Controls, 47 women (mean age 51.3 ± 16.1 years) and 53 men (mean age 52.2 ± 14.4 years) were recruited from a Plastic Surgery Out-patient Clinic (approximately 80% of the sample) and among other volunteers after the first wave of COVID-19. We consider our sample to be representative of the general population. Thus, the Out-patient Clinic is state-funded, attendees are from a variety of backgrounds and ages, and they present with a variety of needs such as removal of scars, moles and eyelid disorders (ptosis, ectropion, entropion). We did not include women after breast cancer who come for breast reconstruction, patients with skin cancer, post-bariatric patients, any patient who has immunosuppression. Controls were included based on a negative history of COVID-19 (non-infected or non-symptomatic subjects). One woman reported injury of the 3rd finger of the left hand and was included in the study after exclusion of this finger measurement. All the participants were White (based on patients’ medical data and controls recruitment).

Ethical statement

The protocol was agreed by the Bioethical Committee of the Medical University of Lodz (RNN/152/20/KE). All methods were performed in accordance with the relevant guidelines and regulations. Written informed consent was obtained from all participants.

Hand images

With regard to the measurement of digit ratio, our preference would have been for direct measurement of fingers. However, it was difficult to measure digit length directly from the hands of the patients because many of them were very ill and measurers were hampered by personal protective equipment. Moreover, direct digit measurement requires a period of time during which the patient and measurer are in close proximity. This is to be avoided with an infectious viral agent. Indirect methods such as the use of photocopies or scanners, give a permanent record of digit lengths. Against this, it was felt that the use of photocopiers or scanners was not appropriate as repeated use of such machines may result in cross infection resulting from virus particles being left on surfaces. Moreover, in comparison to directly measured digits, indirect images yield lower 2D:4D ratios24,28,29 with magnitudes that may vary by sex and hand30. These effects may extend to asymmetries also, and the accuracy of asymmetries measured from photocopies has been questioned31. Therefore, it was decided to photograph the hands of the patients. Typically, the patient was sitting up in bed and he/she was instructed to place their hands horizontally with the palms uppermost, the digits straight and together. In order to minimize inconvenience to the patient it was decided not to use a tripod with the camera. Rather, the experimenter held the camera approximately 30 cm above the patient’s hand. This protocol was felt to be appropriate as it would minimise the amount of proximity necessary between experimenter and patient. Moreover, it gives a permanent image of the supine hand which did not involve potential distortions resulting from digit contact with glass surfaces. It is to be noted that the relative lengths of digits within a hand can be obtained in this way but R–L contrasts of absolute measures of digit length are likely to be unreliable as they will be influenced by small vertical differences in distance between hand and camera. Photographs were checked for definition at the tips of the digits and at the metacarpophalangeal crease at the base of the digits. A second photograph was taken if the first was not deemed to be of sufficient quality.


Eight measurements were taken from patients’ and controls’ hand photographs: second, third, fourth and fifth digits’ lengths (2D, 3D, 4D and 5D) (right (R) and left hand (L)). On the basis of the these parameters the following ratios were calculated: 2D:3D, 2D:4D, 2D:5D, 3D:4D, 3D:5D, 4D:5D for the right (R) and left (L) hand (D length [mm]/D length [mm]) in addition to the ratios’ directional asymmetries (right ratio–left ratio: Δ2D:3D, Δ2D:4D, Δ2D:5D, Δ3D:4D, Δ3D:5D, Δ4D:5D) and unsigned asymmetries (FAs) (|(right–left)|). We also calculated two composite asymmetries by summing (i) all six (|(right–left)|) asymmetries and (ii) and asymmetries for the “independent” ratios of 2D:4D and 3D:5D. We refer to the latter as the “Clinical Composite Asymmetry” in the Results section. All measurements (in patients and controls) were made twice by AKT using the GNU Image Manipulation Program (GIMP) version 2.10.20. For a subset of measurements, a sliding calliper was used directly on the image of the fingers on the computer screen (by JTM). Measurements were performed on the palmar side of the hand using anthropometric points lying on the digit axis: pseudophalangion—the most proximal point in the finger metacarpophalangeal crease, dactylion—the most distal point on the fingertip32. There was high repeatability of digit ratios within and between observers. The final ratios were calculated as a mean of two ratios obtained from the GIMP program. These ratios were used in the further analysis of the data.

Statistical analysis

Analysis was conducted on the differences in the digit ratios and their directional (right–left) and unsigned [|(right–left)|] asymmetries between patients hospitalized due to Covid-19 and controls. The normality of distribution of the tested variables was examined (using Shapiro–Wilk test) and the homogeneity of variances was checked (using the Bartlett test). With both assumptions met we applied univariate t-tests for differences between means in addition to two-way analysis of variance (ANOVA). If any of these assumptions were not met then non-parametric tests were used. Logistic regression was used to evaluate the relationship between the asymmetry index being the sum of the unsigned asymmetries of the ratios of the largest effect sizes estimated with omega-squared for ANOVA (“Clinical Composite Asymmetry”) and the risk of hospitalization due to Covid-19. Finally, logistic regression model included the following variables: the sum of asymmetries of 2D:4D and 3D:5D (dependent variable) and the group (patient vs. control) and sex (independent variables). Effect size for inter-group differences was evaluated with Cohen’s d for t-tests and omega-squared (ω2) for ANOVA. The interpretation of descriptors of magnitude for the former were small 0.20, medium 0.50 and large 0.80 and for ω2 > 0.01 —weak, > 0.06—medium, > 0.14—strong effect. The probability of p < 0.05 was accepted as a level of significance.


Characteristics of Covid-19 patients

Among 54 patients there were 28 men (mean age 54.7 ± 14.7 years) and 26 women (mean age 59.3 ± 18.2 years). The group of patients did not differ in age and frequency of males and females from the controls (F = 1.085; p = 0.299). Specific characteristics (i.e. BMI, comorbidity, smoking status) and Covid-19 symptoms and severity are shown in Table 1.Table 1 Characteristics of patients hospitalized because of Covid-19.Full size table

Reliability of measurements

First we checked intra-observer reliability for all twelve ratios (ratio 1 versus ratio 2) for observer AKT. The coefficient of reliability for raw measurements (R) ranged from 96.07% (for 3D:4D L) to 99.66% (for 2D:5D R). Intra-class correlation coefficients were also very high Table 2). Repeatability of signed asymmetries can be low because they contain the measurement error of four digits. However, for the signed asymmetries (R–L) and the unsigned asymmetries (|R–L|) the R ranged from 99.86% (for 2D:3D |R–L| to 99.97% (for 2D:4D R–L) also with high ICC’s (Table 2). Further analysis included mean values of ratio 1 and 2. Then, inter-observer reliability was checked (observer AKT versus observer JTM), for two ratios: 2D:4D R and 2D:4D L and their signed and unsigned asymmetries. Due to the high reliability between observers (2D:4D R: TEM = 0.0089, R = 99.66%, ICC = 98.09%; 2D:4D L: TEM = 0.0118, R = 98.91%, ICC = 97.93%; R–L: TEM = 0.0076, R = 98.66%, |R–L|: TEM = 0.0076, R = 96.13%, ICC = 96.43%) final analysis included data from AKT.Table 2 Technical error measurement (TEM) and the coefficient of reliability for raw measurements (R) for ratios and for R–L and |R–L| of six ratios for observer 1.Full size table

Digit ratios: patients vs. controls

There were no relationships between age and digit ratios in any of the twelve tests (values of r varied from − 0.14 for right 2D:3D to 0.1 for right 4D:5D, all p > 0.05).

Patient and control means and SD’s for six ratios and 12 effects (right and left ratios) are given in Table 3. Values of p and Cohen’s d are included from t-tests. There were five significant effects ranging from small to medium in magnitude. Four of these showed higher values in the patients compared to the controls, i.e. 3D:5D right d = 0.55, left d = 0.37; 4D:5D right d = 0.64, left d = 0.58. One effect showed mean patient < control (right 2D:5D d = 0.38). We note that all five significant effects were present in ratios that included 5D. Correction for multiple tests is inappropriate across Table 3 as the variables are not independent, i.e. the length of each digit is present in three ratios. We considered the effect of sex on these patient/control differences by performing two-factor ANOVA’s (independent variables: group [patients, controls], sex [males, females] with dependent variable digit ratio). All five remained significant (see effect sizes [ω2]), There were no effects of sex and no significant interactions (Table 4).Table 3 Patient and control means and SD’s for six digit ratios (2D:3D; 2D:4D; 2D:5D;3D:4D; 3D:5D; 4D:5D) and their signed and unsigned asymmetries.Full size tableTable 4 Differences in digit ratios and their asymmetries between patients and controls—(ANOVA) controlled for sex.Full size table

Digit ratio asymmetries: patients vs. controls

Two associations between age and asymmetry were significant (|R–L| 2D:4D, R = 0.17, p = 0.03 and |R–L| 4D:5D, R = 0.24, p = 0.03). However, there were no relationships between age and asymmetries in ten of the twelve tests (values of R varied from − 0.13 for R–L 2D:3D to 0.16 for |R–L| 2D:3D, all p > 0.05).

There were no significant differences in directional asymmetries (R–L) between patients and controls (Tables 3 and 4). There is some evidence in the literature that directional asymmetry of 2D:4D shows sex differences (males < females). Therefore we checked for directional asymmetry (deviations from a mean of zero) in (R–L) in patients and controls for all six ratios split by sex. For male patients (n = 23) one-sample t-tests with mean set at zero showed there were no significant deviations from zero in any ratio (means varied from 0.002 for 3D:4D to 0.049 for 2D:5D, all p > 0.05). For female patients (n = 26), for five ratios means varied from − 0.014 for 2D:4D to 0.031 for 3D:5D, all p > 0.05. For female 4D:5D there was directional asymmetry with mean of 0.034, t = 2.20, p = 0.04. With regard to controls (males n = 53, females n = 47) there was a similar pattern with evidence of directional asymmetry in 4D:5D (males: mean = 0.018, t = 2.44, p = 0.02 and females: mean = 0.016, t = 2.31, p = 0.03). For the remaining ratios means varied from − 0.006 to 0.010, all p > 0.05. Therefore, there was no evidence of significant directional asymmetry in male and female mean (R–L) ratios with the exception of 4D:5D which showed some evidence of higher ratios in the right hand compared to left hand. This suggests that the ratios we consider here (with the exception of 4D:5D) have a mean that does not significantly deviate from zero, i.e. they have the properties of ideal fluctuating asymmetry.

With regard to unsigned asymmetries (|R–L|), the distributions are “half-normal”. It may be that t-tests of means for patients versus controls are robust enough to give meaningful p values. However, in order to consider such differences in a conservative manner we applied Mann–Whitney U tests. There were four significant effects (2D:4D, d = 0.74; 2D:5D, d = 0.66; 3D5D, d = 0.79; 4D:5D, d = 0.47) and all showed patients > controls. We note that three effects are for variables that include digit 5D. Summing the unsigned asymmetries across all six ratios we found this composite measure of asymmetry was higher in the patients compared to controls (d = 0.8). We then focused on |R–L| in the two “independent” ratios with the highest effect size (i.e. 2D:4D and 3D:5D) and found they were not correlated (r = − 0.047). A composite of these two variables, a “Clinical Composite Asymmetry showed the highest effect size of all with patients > controls, d = 1.04 (Fig. 1, Table 3).

figure 1
Figure 1

We further considered the effect of sex on these patient/control differences by performing two-factor ANOVA’s (independent variables: group [patients, controls], sex [males, females] with dependent variable digit ratio). There were high effect sizes (ω2) for |Δ2D:4D| = 0.115; |Δ2D:5D| = 0.105; |Δ3D:5D| = 0.155; |Δ4D:5D| = 0.055. The effect size for the “Clinical Composite Asymmetry” of 2D:4D and 3D:5D was 0.231. Logistic regression indicated that the “Clinical Composite Asymmetry”, regardless of sex, correlates with the risk of hospitalization due to Covid-19. The area under an ROC curve (AUC) is 0.787, which shows that this classifier is better than a random classifier (AUC = 0.5) with the cut-off point of 0.087. A “Clinical Composite Asymmetry” that is higher than 0.087 discriminates hospitalized patients (sensitivity—71% and specificity 75%) (Fig. 2). The risk of hospitalization in case of the index > 0.087 is 3.5 times higher than in those with lower “Clinical Composite Asymmetry” (OR 3.667).

figure 2
Figure 2

“Clinical Composite Asymmetry” did not correlate with Covid-19 severity (R = − 0.075, p = 0.61) or with length of hospitalization (R = 0.137, p = 0.35).


This study focused on associations between digit ratios and severity of COVID-19, as evidenced by hospitalization of patients. Our results indicate that digit ratios, and their asymmetries may be regarded as simple clinical markers of the possible risk of hospitalization due to Covid-19. Additionally, the study aimed to examine the role of prenatal sex steroids and that of postnatal developmental instability on the course of Covid-19. We have found evidence for digit ratio and digit ratio asymmetry differences between hospitalized patients with COVID-19 and controls. For digit ratios the magnitude of the effect sizes was small to medium (d = 0.3–0.64) and involved all ratios that included 5D, i.e. 2D:5D, 3D:5D and 4D:5D (patients > controls). There were no significant differences between patients and controls for directional (right-left) asymmetry. The largest effect sizes (medium to large) were found for measures of developmental instability, i.e. differences in unsigned asymmetries between patients and controls (patients > controls). These included 2D:4D (d = 0.74), and all ratios that involved 5D, i.e. 2D:5D (d = 0.66), 3D:5D (d = 0.79) and 4D:5D (d = 0.47). There are likely to be inter-correlations between these asymmetry effect sizes, for example 2D is present in two of them, as is 4D and 5D is present in three. The two largest effect sizes were found in ratios that may be independent of each other in the sense that they do not share digits (i.e. 2D:4D and 3D:5D). Summing the unsigned asymmetries of 2D:4D and 3D:5D gave a composite asymmetry with a large effect size (patient > control) of d = 1.04. Removing the effect of sex in a two-factor ANOVA had little effect on the magnitude of the effect size which remained large (ω2 = 0.231). We suggest that the unsigned composite asymmetry of 2D:4D and 3D:5D may have utility in identifying individuals that are of high risk for hospitalization resulting from COVID-19. Therefore, we have referred to it as a „Clinical Composite Asymmetry”. The utility of Clinical Composite Asymmetry as a classifier was characterized by AUC = 0.787 (good classifier). In addition, the optimum cut off point ≤ 0.087 was determined, for which sensitivity and specificity were 71% and 75% respectively with OR over 3.5. Regression analysis showed that the index > 0.087 may be a prognostic factor for hospital care for patients with Covid-19. However, to verify the prognostic value of the suggested index further studies based on larger populations in different ethnic groups are needed.

Much of the work concerning effects of prenatal sex steroids on digit ratio has concentrated on 2D:4D. However, effect sizes for 2D:4D are likely to be linked to other ratios that share 2D or 4D (i.e. 2D:3D; 2D:5D; 3D:4D; 4D:5D). The 3D:5D ratio has also been described as sexually dimorphic (males < females) and may show effects that are independent of 2D:4D24,25,26. Importantly, 3D:5D is not stable during development across age ranges from 2 to 18 years. Rather it shows a reduction with age which suggests that it may be influenced by postnatal production of androgens24. Comparisons between digit ratios of hospitalized patients versus controls gave small to medium effect sizes for 2D:5D, 3D:5D and 4D:5D. In so far as these digit ratios are influenced by sex steroids, this may be evidence for a link between severity of COVID-19 and prenatal (2D:4D) and postnatal (3D:5D) testosterone and oestrogen. Studies in humans and with an animal model (Golden Hamsters) have reported that SARS-CoV2 upregulates the enzyme CYP19A1 (oestrogen synthetase) leading to a profound reduction in testosterone and an increase in oestrogen in the lungs and other organs. Dysregulated sex hormones and interferon gamma (IFN-γ) levels are associated with critical illness in Covid-19 patients. In this regard, both male and female Covid-19 patients, present elevated oestradiol levels which positively correlates with IFN-γ levels (for humans33, for an animal model34). Manning and Fink9 reported that national values of male 2D:4D are positively related to national COVID-19 CFR’s. This led them to suggest that nations with high COVID-19 mortality have male populations that have experienced low prenatal testosterone relative to oestrogen.

However, it is more likely that the differences between patients and controls have arisen as the result of elevated levels of developmental instability in the former compared to the latter. Manning24 has considered the stability of all six digit ratios during growth between the ages of 2 years and 18 years. Right 2D:4D was stable but left 2D:4D was not. All ratios that included 5D showed growth-linked instability for both the right and left hands. We suggest that ratios that include 5D are „hotspots” for developmental instability that may be triggered by stressors that include rapid growth22,24. Recently a syndemic approach, which includes biological and social interactions for prognosis, treatment, and health policy, has been proposed. Interaction between infection with SARS-CoV-2 and an array of non-communicable diseases strongly associated with poverty, including obesity, hypertension, diabetes, cardiovascular and chronic respiratory diseases, and cancer is now considered. Moreover, syndemics are characterised by biological and social interactions between conditions and states, which increase one’s susceptibility to poor health outcomes35. In this respect, considering morphological signs of exposure to prenatal sex steroids and developmental instabilities (interaction between rapid early growth and stressors such as poor maternal and childhood nutrition) in patients with severe or fatal course of COVID-19 may give insight into the syndemic nature of Covid-19.

In contrast to right and left digit ratios, differences in the magnitude of digit ratio asymmetries between the right and left hand gave medium to large effect sizes. This was not apparent in directional asymmetries (R–L), perhaps because they comprise subtle deviations from perfect symmetry. Such asymmetries have been described as weakly sexually dimorphic for right-left 2D:4D (or Dr-l: with male Dr-l < female Dr-l36). Removing the signs from directional asymmetry (|R–L|) gave us variables that showed medium to high effect sizes in comparisons between patients and controls. Digit ratio (|R–L|) is a measure which is equivalent to asymmetry differences in digit length22. However, in this case we are dealing with R–L differences in morphological patterns involving two digits rather than differences between single digits of the right and left hands. It is not known whether |R–L| is sexually dimorphic across the six digit ratios. There is evidence that the phenotype of Dr-l is influenced by variation in the gene for the enzyme CYP19A1. Thus, Zhanbing et al. (2019) have reported a CYP19A1 single-nucleotide polymorphism (rs4775936) is related to variation in Dr-l in a Chinese sample36. CYP19A1 is important in the conversion of testosterone to oestrogen and SNP rs4775936 has been linked to the incidence of breast cancer. It may not be coincidental that up-regulation of CYP19A1 occurs in the lungs and other organs of COVID-19 patients leading to dysregulation of sex hormones (acute reduction in testosterone and an increase in oestrogen) and a marked increase in interferon gamma (IFN-γ) levels. Both are associated with critical illness in Covid-19 patients33,34. Further work is indicated to investigate patterns of age in digit ratio asymmetry (|R–L|). However, if it takes the form of single trait asymmetry, such as that of digit length asymmetry, it may show high levels in children which reduce with increasing age22. Associated with these age changes in digit asymmetry we find age dependent instability of all digit ratios that include 5D24. Such a pattern would suggest unsigned asymmetries of digit ratios involving 5D are sensitive correlates of developmental instability which may be negatively associated with immune system function. Such an interpretation is consistent with the view that increased asymmetry is the result of a combination of deleterious genetic and environmental factors and is defined as small, random deviations from perfect bilateral symmetry regarded as a measure of the developmental stability of the individual and phenotypic and genetic quality23.

We acknowledge our study has limitations: (i) Our sample size of 54 hospitalized patients is small. Obtaining good quality photographs of patients’ hands during hospitalization was sometimes challenging. A larger number would have been possible if the patient’s hands had been photocopied or scanned at discharge. However, with this latter methodology one risks missing severe cases of COVID-19 that are never discharged. We plan to extend our study, adding numbers of hospitalized patients and remeasuring digit lengths at discharge. With an increase in sample size we will consider relationships between clinical variables and digit ratios and their asymmetries. It is to be noted that within Table 4 we control for sex and report effect sizes for patients versus controls. With regard to the latter differences the effect sizes are medium to strong but none of the former are significant. This may be because sex differences in digit ratios have small to medium effect sizes. If we are correct in this we would expect that larger samples to show significant effects for sex in addition to differences between patients and controls. (ii) Additionally a further confounder may be the unknown infection status of the controls. They were recruited during the same time frame as patients hospitalized due to Covid-19 among other out-patient patients age-matched and based on their negative history of any symptoms of Covid-19 infection. Such appointment of controls may have resulted in a heterogeneity of this group such that they may have included non-infected individuals as well as non-symptomatic but infected or past-infected individuals. However, this does not invalidate the idea of this study which was focused on “markers” of symptomatology related to hospitalization. It would be beneficial to perform such analysis comparing symptomatic versus asymptomatic (or mildly symptomatic treated out-patients) but infected patients. However, this was not possible during the first waves of Covid-19 as there was no nation-wide testing in Poland (in general only symptomatic individuals were tested). In this regard our results may be biased by behavioral factors—i.e. the way participants prevented infection. (iii). We were not able to make comparisons between individuals who had been vaccinated and those who have not. This was because our data were obtained during the first wave of Covid-19, so none of the participants (patients and controls) had been vaccinated. In this regard, it would be of great interest to compare the efficacy of the vaccine in individuals with high and low values of the „Clinical Composite Asymmetry”. The prediction would be that vaccine efficacy would be low when „Clinical Composite Asymmetry” is high and high when „Clinical Composite Asymmetry” is low.

In conclusion, we have found differences in digit variables between patients hospitalized for COVID-19 and controls. Overall, our findings point to high levels of developmental instability in the former compared to the latter. Our focus was on six digit ratios and for each we considered right and left ratios and their asymmetries (signed and unsigned). We found differences in digit ratios between patients and controls that were focused on ratios that included 5D. The effect sizes were small to medium. Unsigned asymmetries of four digit ratios, including three that involved 5D, yielded medium to large effect sizes with patients > controls. The largest of these asymmetries were for 2D:4D and 3D:5D. A „Clinical Composite Asymmetry” for these two variables gave an effect size which may have some utility in identifying individuals who have experienced high developmental instability. Thus, this „Clinical Composite Asymmetry” may enable us to identify individuals who are likely to experience severe COVID-19 and those who may not.


  1. WHO-China Joint Mission, Report of the WHO-China Joint Mission on Coronavirus Disease 2019 (COVID-19) (2020).
  2. Mattiuzzi, C. & Lippi, G. Which lessons shall we learn from the 2019 novel coronavirus outbreak?. Ann. Transl. Med. 8, 48 (2020).CAS Article Google Scholar 
  3. Tahtasakal, C. A. et al. Could we predict the prognosis of the COVID-19 disease?. J. Med. Virol. 93, 2420–2430 (2020).Article Google Scholar 
  4. van Halem, K. et al. Risk factors for mortality in hospitalized patients with COVID-19 at the start of the pandemic in Belgium: A retrospective cohort study. BMC Infect. Dis. 20, 897 (2020) (Erratum in: BMC Infect. Dis. 20, 956 (2020)).Article Google Scholar 
  5. Yi, P. et al. Risk factors and clinical features of deterioration in COVID-19 patients in Zhejiang, China: A single-centre, retrospective study. BMC Infect. Dis. 20, 943 (2020).CAS Article Google Scholar 
  6. Guan, W. J. et al. Clinical characteristics of coronavirus disease 2019 in China. N. Engl. J. Med. 382, 1708–1720 (2020).CAS Article Google Scholar 
  7. Wambier, C.G., A. Goren, A. Ossimetha, G., & Nau, A.A. Qureshi, Androgen-driven COVID-19 pandemic theory. Preprint at (2020).
  8. Wambier, C. G. & Goren, A. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection is likely to be androgen mediated. J. Am. Acad. Dermatol. 83, 308–309 (2020).CAS Article Google Scholar 
  9. Manning, J. T. & Fink, B. Understanding COVID-19: Digit ratio (2D:4D) and sex differences in national case fatality rates. Early Hum. Dev. 146, 105074 (2020).CAS Article Google Scholar 
  10. Heurich, A. et al. TMPRSS2 and ADAM17 cleave ACE2 differentially and only proteolysis by TMPRSS2 augments entry driven by the severe acute respiratory syndrome coronavirus spike protein. J. Virol. 88, 1293–1307 (2014).Article Google Scholar 
  11. Hoffmann, M. et al. SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor. Cell 181, 271–280 (2020).CAS Article Google Scholar 
  12. Ferrini, R. L. & Barrett-Connor, E. Sex hormones and age: A cross-sectional study of testosterone and estradiol and their bioavailable fractions in community-dwelling men. Am. J. Epidemiol. 147, 750–754 (1998).CAS Article Google Scholar 
  13. Swift-Gallant, A., Johnson, B. A., Di Rita, V. & Breedlove, S. M. Through a glass, darkly: Human digit ratios reflect prenatal androgens, imperfectly. Horm. Behav. 120, 104686 (2020).Article Google Scholar 
  14. McCormick, C. M. & Carre, J. M. Facing off with the phalangeal phenomenon and editorial policies: A commentary on Swift-Gallant, Johnson, Di Rita and Breedlove 2020. Horm. Behav. 120, 104710 (2020).Article Google Scholar 
  15. Sahin, H. A further analysis of Manning and Fink. Early Hum. Dev. 105121 (2020) (Online ahead of print).
  16. Jones, A. L., Jaeger, B. & Schild, C. No credible evidence for 2D:4D and COVID-19 outcomes: A probabilistic perspective on digit ratio, ACE variants, and national case fatalities. Early Hum Dev. 152, 105272 (2020).Article Google Scholar 
  17. Manning, J. T., Scutt, D., Wilson, J. & Lewis-Jones, D. I. The ratio of 2nd to 4th digit length: A predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen. Hum. Reprod. 13, 3000–3004 (1998).CAS Article Google Scholar 
  18. Manning, J. T. Resolving the role of prenatal sex steroids in the development of digit Ratio. Proc. Natl. Acad. Sci. U. S. A. 108, 16143–16144 (2011).ADS CAS Article Google Scholar 
  19. Manning, J. T. & Fink, B. Digit ratio and personality and individual differences. In The SAGE Handbook of Personality and Individual Differences (eds Zeigler-Hill, V. & Shackelford, T. K.) (SAGE Publications Ltd, 2018).Google Scholar 
  20. Manning, J. T. Digit Ratio: A Pointer to Fertility, Behavior, and Health (ed. Manning, J.T) 24–41 (Rutgers University Press, 2002).
  21. Manning, J. T. The Finger Ratio (ed. Manning, J.T) 5–10 (Faber and Faber, 2008).
  22. Wilson, J. M. & Manning, J. T. Fluctuating asymmetry and age in children: Evolutionary implications for the control of developmental stability in children. J. Hum. Evol. 30, 529–537 (1996).Article Google Scholar 
  23. Thornhill, R. & Møller, A. P. Developmental stability, disease and medicine. Biol. Rev. Camb. Philos. Soc. 72, 497–548 (1997).CAS Article Google Scholar 
  24. Manning, J.T. Sex differences and age changes in digit ratios: Implications for the use of digit ratios in medicine and biology. In Handbook of Anthropometry, 841–851 (Springer, 2012).
  25. Manning, J. T., Callow, M. & Bundred, P. E. Finger and toe ratios in humans and mice: Implications for the aetiology of diseases influenced by HOX genes. Med. Hypotheses. 60, 340–343 (2003).CAS Article Google Scholar 
  26. McFadden, D. & Shubel, E. Relative lengths of fingers and toes in human males and females. Horm. Behav. 42, 492–500 (2002).Article Google Scholar 
  27. Trivers, R., Manning, J. & Jacobson, A. A longitudinal study of digit ratio (2D:4D) and other finger ratios in Jamaican children. Horm. Behav. 49, 150–156 (2006).Article Google Scholar 
  28. Ribeiro, E., Neave, N., Morais, R. N. & Manning, J. T. Direct versus indirect measurement of digit ratio (2D: 4D) a critical review of the literature and new data. Evol. Psychol. 14, 1474704916632536 (2016).Article Google Scholar 
  29. Fink, B. & Manning, J. T. Direct versus indirect measurement of digit ratio: New data from Austria and a critical consideration of clarity of report in 2D:4D studies. Early Hum. Dev. 127, 28–32 (2018).Article Google Scholar 
  30. Manning, J. T. et al. Photocopies yield lower digit ratios (2D:4D) than direct finger measurements. Arch. Sex. Behav. 34(3), 329–333 (2005).Article Google Scholar 
  31. Manning, J. T., Fink, B., Neave, N. & Caswell, N. The second to fourth digit ratio and asymmetry. Ann. Hum. Biol. 33, 480–492 (2006).Article Google Scholar 
  32. Kasielska-Trojan, A. & Antoszewski, B. Can digit ratio (2D:4D) studies be helpful in explaining the aetiology of idiopathic gynecomastia?. Early Hum. Dev. 91, 57–61 (2015).Article Google Scholar 
  33. Schroeder, M. et al. Sex hormone and metabolic dysregulations are associated with critical illness in male Covid-19 patients. Preprint at (2020).Article Google Scholar 
  34. Stanelle-Bertram, S., et al. SARS-CoV-2 induced CYP19A1 expression in the lung correlates with increased aromatization of testosterone-to estradiol in male golden hamsters. Preprint at (2020).
  35. Horton, R. Offline: COVID-19 is not a pandemic. Lancet 396, 874 (2020).CAS Article Google Scholar 
  36. Zhanbing, M. et al. Association of CYP19A1 single-nucleotide polymorphism with digit ratio (2D:4D) in a sample of men and women from Ningxia (China). Early Hum. Dev. 132, 58–65 (2019).Article Google Scholar 

Download references

Author information


  1. Plastic, Reconstructive and Aesthetic Surgery Clinic, Institute of Surgery, Medical University of Lodz, Kopcinskiego 22, 90-153, Lodz, PolandA. Kasielska-Trojan & B. Antoszewski
  2. Applied Sports, Technology, Exercise, and Medicine (A-STEM), Swansea University, Swansea, UKJ. T. Manning
  3. Department of Infectious and Liver Diseases, Medical University of Lodz, Lodz, PolandM. Jabłkowski & J. Białkowska-Warzecha
  4. Department of Gynecology and Reproductive Medicine, Karolinska University Hospital, Stockholm, SwedenA. L. Hirschberg


Conception or design of the work: J.T.M., A.K.T., B.A., A.H. Data acquisition: J.B.W., M.J. Data analysis: J.T.M., A.K.T. Interpretation of data: J.T.M., A.K.T., B.A. Drafting and revising the ms: J.T.M., A.K.T., B.A., A.H., J.B.W., M.J. All authors have approved the submitted version.

The length of your fingers may determine how sick you get from COVID-19

Authors: Chris Melore March 28, 2022

Your risk of ending up in the hospital with COVID-19 may literally be in your own hands. A new study finds finger length displays a link to a person’s sex hormone levels. What does this have to do with COVID-19? Researchers at Swansea University say a patient’s testosterone levels play a key role in how sick they get after infection.

Previous studies show that having a longer ring finger is a sign of higher testosterone levels in the womb. On the other hand, a longer index finger signals higher levels of estrogen. Typically, men have longer ring fingers and women have longer index fingers.

The new study examined this link between the sex hormones before birth and during puberty and the rate of COVID hospitalizations. Their findings reveal that people with “feminized” short little fingers in comparison to their other digits end up suffering more severe cases of COVID-19. Moreover, people who have larger size differences between the fingers on their left and right hands are at even greater risk.

The link between testosterone and coronavirus

Although most people only experience mild COVID symptoms, the elderly and men are more likely to have a severe case that requires urgent care. This has led scientists to wonder if a man’s testosterone levels play a role in disease severity.

One theory is that high testosterone levels cause COVID to worsen. However, another study links low levels in elderly men to a severe case of the virus.

To figure out which is right, the team examined the size ratios of the 2nd, 3rd, 4th, and 5th digits on the hands of over 150 people. Fifty-four of these individuals were COVID-19 patients, while the others served as a healthy control group.

Specifically, the results show bigger differences between the 2D:4D and 3D:5D ratios on each person’s hands had a connection to a more severe case of COVID-19.

“Our findings suggest that COVID-19 severity is related to low testosterone and possibly high estrogen in both men and women,” says Professor John Manning in a university release.

“’Feminized’ differences in digit ratios in hospitalized patients supports the view that individuals who have experienced low testosterone and/or high estrogen are prone to severe expression of COVID-19. This may explain why the most at-risk group is elderly males,” the researcher continues. “This is significant because if it is possible to identify more precisely who is likely to be prone severe COVID-19, this would help in targeting vaccination. Right-Left differences in digit ratios (particularly 2D:4D and 3D:5D) may help in this regard.”

Could testosterone drugs defeat the pandemic?

Currently, study authors say there are several trials examining anti-androgen (testosterone) drugs which may help treat COVID-19. At the same time, scientists are also looking at testosterone as a possible anti-viral medication against COVID.

“Our research is helping to add to understanding of Covid-19 and may bring us closer to improving the repertoire of anti-viral drugs, helping to shorten hospital stays and reduce mortality rates,” Prof. Manning adds. “The sample is small but ongoing work has increased the sample. We hope to report further results shortly.”

This isn’t the first study to link finger length to seemingly unrelated topics. A previous study connected children’s finger length to their mother’s income as well as vulnerability to childhood illnesses.

The study is published in the journal Scientific Reports.

Baricitinib for patients with severe COVID-19—time to change the standard of care?

AuthorsAlexander Supadya,c,d and Robert Zeiserb Lancet Respir Med. 2022 Feb 3doi: 10.1016/S2213-2600(22)00021-2 [Epub ahead of print]PMCID: PMC8813061PMID: 35123659

Mortality is high among patients with severe COVID-19 who require invasive mechanical ventilation (IMV) or extracorporeal membrane oxygenation (ECMO).12 Therefore, further specific treatment options for these patients are urgently needed.

Early in the COVID-19 pandemic, Janus kinase (JAK) inhibitors were identified as potential therapeutic agents for the treatment of SARS-CoV-2 infections.34 JAK inhibitors, such as ruxolitinib, fedratinib, and baricitinib, have strong anti-inflammatory properties and are already in clinical use for the treatment of graft-versus-host disease, myelofibrosis, and rheumatoid arthritis.567

Two large, randomised, controlled trials have assessed the use of baricitinib in hospitalised patients with COVID-19.89 The Second Adaptive COVID-19 Treatment Trial (ACTT-2)8 showed that administration of baricitinib in combination with remdesivir shortened the time to recovery in hospitalised patients with COVID-19 compared with remdesivir alone. The COV-BARRIER trial9 did not show a significant benefit of baricitinib plus standard of care compared with placebo plus standard of care in the primary endpoint of disease progression by day 28. Nevertheless, although these were only secondary endpoints, 28-day mortality and 60-day mortality were significantly lower in patients who received baricitinib than in those who received placebo.

However, important questions remain, especially regarding the effect of baricitinib in patients with severe COVID-19 who require IMV or even ECMO. In the ACTT-2 trial, only 111 (11%) of 1033 patients were receiving IMV or ECMO at study inclusion; whereas in the primary COV-BARRIER trial, these severely affected patients were excluded.

In The Lancet Respiratory Medicine, E Wesley Ely and colleagues10 report the findings of an exploratory trial that followed the design of the COV-BARRIER trial and aimed to evaluate the efficacy and safety of baricitinib in addition to standard of care in critically ill hospitalised adults with COVID-19 who were receiving IMV or ECMO. Patients were randomly assigned to receive baricitinib 4 mg (n=51) or matched placebo (n=50) once daily for up to 14 days in addition to standard of care. All-cause mortality by day 28 was significantly lower in patients who received baricitinib than in those who received placebo (20 [39%] of 51 participants died in the baricitinib group vs 29 [58%] of 50 in the placebo group; hazard ratio [HR] 0·54 [95% CI 0·31–0·96]; p=0·030); this finding persisted through day 60 (23 events [45%] vs 31 [62%]; HR 0·56 [95% CI 0·33–0·97]; p=0·027). The authors concluded that baricitinib might represent a novel option for the reduction of mortality in patients with COVID-19, even in progressed disease stages (ie, when already receiving IMV or ECMO).

Although the assignment of participants to the treatment groups was randomised and masked, the mortality benefit determined in this exploratory trial should be interpreted with caution. Important baseline data for the assessment of disease severity and comparison of the study cohort with other cohorts were not provided, including duration of IMV and ECMO before study inclusion, ventilator settings, ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen, and prognostic scores, such as Sequential Organ Failure Assessment (SOFA), Simplified Acute Physiology Score (SAPS) II, or Respiratory Extracorporeal Membrane Oxygenation Survival Prediction (RESP). Without these supporting data, it cannot be assumed that the study groups were balanced in terms of baseline disease severity and that quality of care was consistent across study locations. Therefore, the results from this study cannot match the level of evidence from a phase 3 randomised, controlled trial with well-defined primary and secondary endpoints, accompanying sample size estimations, and a prespecified statistical analysis plan.

Interestingly, the all-cause mortality by day 60 in the placebo group in this study was high (62%), despite the baseline use of corticosteroids in 44 (88%) of 50 patients in this group.10 A meta-analysis summarising data from more than 50 000 patients with COVID-19 receiving IMV showed a mortality of only 45%, although a large number of these patients were treated without corticosteroids at the beginning of the pandemic.1 Furthermore, by contrast with the findings from this trial, in ACTT-2, in the subgroup of patients receiving IMV or ECMO at baseline, who were randomly assigned to receive baricitinib (n=54) or placebo (n=57) in addition to remdesivir but without corticosteroids, there was no significant difference between the treatment groups with respect to the outcomes of recovery time or survival by day 28.8 These observations contradict the results of the study by Ely and colleagues. At this time, we can only speculate about the reasons for these conflicting results; possible explanations are a potential influence of the concomitant treatment with remdesivir, the effect of corticosteroids, or any other differences between the groups or the treatments they received.

Taken together, we believe that the level of evidence provided by the results from this study is not sufficient to change standard of care and introduce baricitinib into clinical routine for COVID-19 patients with severe respiratory failure. However, the results of this exploratory trial and the data from COV-BARRIER and ACTT-2 reflect clinical equipoise for the addition of baricitinib to standard of care for patients with severe COVID-19 requiring IMV or ECMO, and provide a sound basis for a well-designed phase 3 trial to confirm these findings.

Open in a separate windowCopyright © 2022 Dr Barry Slaven/Science Photo Library

AS reports research grants and lecture fees from CytoSorbents and lecture fees from Abiomed, outside of the submitted work. RZ reports lecture fees from Novartis, Incyte, and Mallinckrodt, outside of the submitted work.


1. Lim ZJ, Subramaniam A, Ponnapa Reddy M, et al. Case fatality rates for patients with COVID-19 requiring invasive mechanical ventilation. A meta-analysis. Am J Respir Crit Care Med. 2021;203:54–66. [PMC free article] [PubMed] [Google Scholar]

2. Barbaro RP, MacLaren G, Boonstra PS, et al. Extracorporeal membrane oxygenation for COVID-19: evolving outcomes from the international Extracorporeal Life Support Organization Registry. Lancet. 2021;398:1230–1238. [PMC free article] [PubMed] [Google Scholar]

3. Richardson P, Griffin I, Tucker C, et al. Baricitinib as potential treatment for 2019-nCoV acute respiratory disease. Lancet. 2020;395:e30–e31. [PMC free article] [PubMed] [Google Scholar]

4. Stebbing J, Phelan A, Griffin I, et al. COVID-19: combining antiviral and anti-inflammatory treatments. Lancet Infect Dis. 2020;20:400–402. [PMC free article] [PubMed] [Google Scholar]

5. Zeiser R, von Bubnoff N, Butler J, et al. Ruxolitinib for glucocorticoid-refractory acute graft-versus-host disease. N Engl J Med. 2020;382:1800–1810. [PubMed] [Google Scholar]

6. Genovese MC, Kremer J, Zamani O, et al. Baricitinib in patients with refractory rheumatoid arthritis. N Engl J Med. 2016;374:1243–1252. [PubMed] [Google Scholar]

7. Harrison CN, Schaap N, Vannucchi AM, et al. Janus kinase-2 inhibitor fedratinib in patients with myelofibrosis previously treated with ruxolitinib (JAKARTA-2): a single-arm, open-label, non-randomised, phase 2, multicentre study. Lancet Haematol. 2017;4:e317–e324. [PMC free article] [PubMed] [Google Scholar]

8. Kalil AC, Patterson TF, Mehta AK, et al. Baricitinib plus remdesivir for hospitalized adults with COVID-19. N Engl J Med. 2021;384:795–807. [PMC free article] [PubMed] [Google Scholar]

9. Marconi VC, Ramanan AV, de Bono S, et al. Efficacy and safety of baricitinib for the treatment of hospitalised adults with COVID-19 (COV-BARRIER): a randomised, double-blind, parallel-group, placebo-controlled phase 3 trial. Lancet Respir Med. 2021;9:1407–1418. [PMC free article] [PubMed] [Google Scholar]

10. Ely EW, Ramanan AV, Kartman CE, et al. Efficacy and safety of baricitinib plus standard of care for the treatment of critically ill hospitalised adults with COVID-19 on invasive mechanical ventilation or extracorporeal membrane oxygenation: an exploratory, phase 3, randomised, double-blind, placebo-controlled trial. Lancet Respir Med. 2022 doi: 10.1016/S2213-2600(22)00006-6. published online Feb 3. [CrossRef] [Google Scholar]

Risk of severe COVID-19 disease with ACE inhibitors and angiotensin receptor blockers: cohort study including 8.3 million people

  1. Julia Hippisley-Cox1, Duncan Young2,3, Carol Coupland4, Keith M Channon5, Pui San Tan6, David A Harrison7, Kathryn Rowan8,  Paul Aveyard6, Ian D Pavord9, Peter J Watkinson5,10
  2. Correspondence to Prof Julia Hippisley-Cox, Primary Care Health Sciences, University of Oxford, Oxford OX1 



There is uncertainty about the associations of angiotensive enzyme (ACE) inhibitor and angiotensin receptor blocker (ARB) drugs with COVID-19 disease. We studied whether patients prescribed these drugs had altered risks of contracting severe COVID-19 disease and receiving associated intensive care unit (ICU) admission.


This was a prospective cohort study using routinely collected data from 1205 general practices in England with 8.28 million participants aged 20–99 years. We used Cox proportional hazards models to derive adjusted HRs for exposure to ACE inhibitor and ARB drugs adjusted for sociodemographic factors, concurrent medications and geographical region. The primary outcomes were: (a) COVID-19 RT-PCR diagnosed disease and (b) COVID-19 disease resulting in ICU care.


Of 19 486 patients who had COVID-19 disease, 1286 received ICU care. ACE inhibitors were associated with a significantly reduced risk of COVID-19 disease (adjusted HR 0.71, 95% CI 0.67 to 0.74) but no increased risk of ICU care (adjusted HR 0.89, 95% CI 0.75 to 1.06) after adjusting for a wide range of confounders. Adjusted HRs for ARBs were 0.63 (95% CI 0.59 to 0.67) for COVID-19 disease and 1.02 (95% CI 0.83 to 1.25) for ICU care.

There were significant interactions between ethnicity and ACE inhibitors and ARBs for COVID-19 disease. The risk of COVID-19 disease associated with ACE inhibitors was higher in Caribbean (adjusted HR 1.05, 95% CI 0.87 to 1.28) and Black African (adjusted HR 1.31, 95% CI 1.08 to 1.59) groups than the white group (adjusted HR 0.66, 95% CI 0.63 to 0.70). A higher risk of COVID-19 with ARBs was seen for Black African (adjusted HR 1.24, 95% CI 0.99 to 1.58) than the white (adjusted HR 0.56, 95% CI 0.52 to 0.62) group.


ACE inhibitors and ARBs are associated with reduced risks of COVID-19 disease after adjusting for a wide range of variables. Neither ACE inhibitors nor ARBs are associated with significantly increased risks of receiving ICU care. Variations between different ethnic groups raise the possibility of ethnic-specific effects of ACE inhibitors/ARBs on COVID-19 disease susceptibility and severity which deserves further study.

Link between fever, diarrhea, severe COVID-19, and persistent anti-SARS-CoV-2 antibodies

Authors: By Dr. Liji Thomas, MD Jan 7 2021

Ever since the coronavirus disease 2019 (COVID-19) pandemic began, there have been many attempts to understand the nature and duration of immunity against the causative agent, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2).

A new preprint research paper appearing on the medRxiv* server describes a link between the persistence of neutralizing antibodies against the virus, disease severity, and specific COVID-19 symptoms.

Permanent immunity is essential if the pandemic is to end. In the earlier SARS epidemic, antibodies were found to last for three or more years after infection in most patients. With the current virus, it may last for six or more months at least, as appears from some reports. Other researchers have concluded that immunity wanes rapidly over the same period, with some patients who were tested positive for antibodies becoming seronegative later on. This discrepancy may be traceable to variation in testing methods, sample sizes and testing time points, as well as disease severity.

Study details

The current study looked at a population of over a hundred convalescent COVID-19 patients, testing most of them for antibodies at five weeks and three months from symptom resolution.

The researchers used a multiplex assay that measured the Immunoglobulin G (IgG) levels against four SARS-CoV-2 antigens, one from SARS-CoV, and four from circulating seasonal coronaviruses. In addition, they carried out an inhibition assay against SARS-CoV-2 spike receptor-binding domain (RBD)-angiotensin-converting enzyme 2 (ACE2) binding and a neutralization assay against the virus. The antibody titers were then plotted against various clinical features and demographic factors.

Antibody titers higher in COVID-19 convalescents

The researchers found that severe disease is correlated with advanced age and the male sex. Patients with underlying vascular disease were more likely to be hospitalized with COVID-19, but those with asthma were relatively spared.

Convalescent COVID-19 patients had higher IgG levels against all four SARS-CoV-2 antigens, relative to controls, and in 98% of cases, at least one of the tests was likely to show higher binding compared to controls. IgGs targeting the viral spike and RBD were likely to be much more discriminatory between SARS-CoV-2 patients and controls. Interestingly, anti-SARS-CoV IgG, as well as anti-seasonal betacoronavirus antibodies, were likely to be higher in these patients.

Anti-spike and anti-nucleocapsid IgG levels, as well as neutralizing antibody titers, were higher in convalescent hospitalized COVID-19 patients than in convalescent non-hospitalized patients, and the titers were positively associated with disease severity.Antibodies against SARS-CoV-2 persist three months after COVID-19 symptom resolution. Sera from COVID-19 convalescent subjects (n=79) collected 5 weeks (w) and 3 months (m) after symptom resolution were subjected to multiplex assay to detect IgG that binds to SARS-CoV-2 S, NTD, RBD and N antigens (A), to RBD-ACE2 binding inhibition assay (B), and to SARS-CoV-2 neutralization assay (C). Dots, lines, and asterisks in red represent non-hospitalized (n=67) and in blue represent hospitalized (n=12) subjects with lines connecting the two time points for individual subjects (*p<0.05 and **p<0.01 by paired t test).Antibodies against SARS-CoV-2 persist three months after COVID-19 symptom resolution. Sera from COVID-19 convalescent subjects (n=79) collected 5 weeks (w) and 3 months (m) after symptom resolution were subjected to multiplex assay to detect IgG that binds to SARS-CoV-2 S, NTD, RBD and N antigens (A), to RBD-ACE2 binding inhibition assay (B), and to SARS-CoV-2 neutralization assay (C). Dots, lines, and asterisks in red represent non-hospitalized (n=67) and in blue represent hospitalized (n=12) subjects with lines connecting the two time points for individual subjects (*p<0.05 and **p<0.01 by paired t test).

Clinical correlates of higher antibody titer

Related Stories

When antibody titers in non-hospitalized subjects were compared with clinical and demographic variables, they found that older males with a higher body mass index (BMI) and a Charlson Comorbidity Index score >2 were likely to have higher antibody titers. COVID-19 symptoms that correlated with higher antibody levels in these patients comprise fever, diarrhea, abdominal pain and loss of appetite. Chest tightening, headache and sore throat were associated with less severe symptoms.

The link between the specific symptoms listed above with higher antibody titers could indicate that they mark a robust systemic inflammatory response, which in turn is necessary for a strong antibody response. Diarrhea may mark severe disease, but it is strange that in this case, it was not more frequent in the hospitalized cohort. Alternatively, diarrhea may have strengthened the immune antibody response via the exposure of the virus to more immune cells via the damaged enteric mucosa. More study is required to clarify this finding.

Potential substitute for neutralizing assay

The binding assay showed that the convalescent serum at five weeks inhibited RBD-ACE2 binding much more powerfully than control serum. Neutralizing activity was also higher in these sera, but in 15% of cases, convalescent patients showed comparable neutralizing antibody titers to those in control sera. On the whole, however, there was a positive association between neutralizing antibody titer, anti-SARS-CoV-2 IgG titers, and inhibition of ACE2 binding.

Persistent immunity at three months

This study also shows that SARS-CoV-2 antibodies persist in these patients at even three months after symptoms subside, with persistent IgG titers against the SARS-CoV-2 spike, RBD, nucleocapsid and N-terminal domain antigens. Binding and neutralization assays remained highly inhibitory throughout this period. The same was true of antibodies against the other coronaviruses tested as well, an effect that has been seen with other viruses and could be the result of cross-reactive anti-SARS-CoV-2 antibodies. Alternatively, it could be due to the activation of memory B cells formed in response to infection by the seasonal beta-coronaviruses.


IgG titers, particularly against S and RBD, and RBD-ACE2 binding inhibition better differentiate between COVID-19 convalescent and naive individuals than the neutralizing assay,” the researchers concluded.

These could be combined into a single diagnostic test, they suggest, with extreme sensitivity and specificity. The correlation with neutralizing antibody titers could indicate that the neutralizing assay, which is more expensive, sophisticated and expensive, as well as more dangerous for the investigators, could be replaced by the other antibody tests without loss of value.

In short, the study shows that specific antibodies persist for three months at least following recovery; antibody titers correlate with COVID-19-related fever, loss of appetite, abdominal pain and diarrhea; and are also higher in older males with more severe disease, a higher BMI and CCI above 2. Further research would help understand the lowest protective titer that prevents reinfection, and the duration of immunity.

*Important Notice

medRxiv publishes preliminary scientific reports that are not peer-reviewed and, therefore, should not be regarded as conclusive, guide clinical practice/health-related behavior, or treated as established information.Journal reference:

The Epidemiology, Transmission, and Diagnosis of COVID-19

Authors: By: Neesha C. Siriwardane & Rodney Shackelford, DO, Ph.D. April 15, 2020

Introduction to COVID-19

Coronaviruses are enveloped single-stranded RNA viruses of the Coronaviridae family and order Nidovirales (1). The viruses are named for their “crown” of club-shaped S glycoprotein spikes, which surround the viruses and mediate viral attachment to host cell membranes (1-3). Coronaviruses are found in domestic and wild animals, and four coronaviruses commonly infect the human population, causing upper respiratory tract infections with mild common cold symptoms (1,4). Generally, animal coronaviruses do not spread within human populations, however rarely zoonotic coronaviruses evolve into strains that infect humans, often causing severe or fatal illnesses (4). Recently, three coronaviruses with zoonotic origins have entered the human population; severe acute respiratory syndrome coronavirus-2 (SARS) in 2003, Middle Eastern respiratory syndrome (MERS) in 2012, and most recently, coronavirus disease 2019 (COVID-19), also termed SARS-CoV-2, which the World Health Organization declared a Public Health Emergency of International Concern on January 31st, 2020 (4,5). 

COVID19 Biology, Spread, and Origin

COVID-19 replicates within epithelial cells, where the COVID-19 S glycoprotein attaches to the ACE2 receptor on type 2 pneumocytes and ciliated bronchial epithelial cells of the lungs. Following this, the virus enters the cells and rapidly uses host cell biochemical pathways to replicate viral proteins and RNA, which assemble into viruses that in turn infect other cells (3,5,6). Following these cycles of replication and re-infection, the infected cells show cytopathic changes, followed by various degrees of pulmonary inflammation, changes in cytokine expression, and disease symptoms (5-7). The ACE2 receptor also occurs throughout most of the gastrointestinal tract and a recent analysis of stool samples from COVID-19 patients revealed that up to 50% of those infected with the virus have a COVID-19 enteric infection (8).

COVID-19 was first identified on December 31st, 2020 in Wuhan China, when twenty-seven patients presented with pneumonia of unknown cause. Some of the patients worked in the Hunan seafood market, which sold both live and recently slaughtered wild animals (4,9).  Clusters of cases found in individuals in contact with the patients (family members and healthcare workers) indicated a human-to-human transmission pattern (9,10). Initial efforts to limit the spread of the virus were insufficient and the virus soon spread throughout China. Presently COVID-19 occurs in 175 countries, with 1,309,439 cases worldwide, with 72,638 deaths as of April 6th, 2020 (4). Presently, the most affected countries are the United States, Italy, Spain, and China, with the United States showing a rapid increase in cases, and as of April 6th, 2020 there are 351,890 COVID-19 infected, 10,377 dead, and 18,940 recovered (4).  In the US the first case presented on January 19th, 2020, when an otherwise healthy 35-year-old man presented to an urgent care clinic in Washington State with a four-day history of a persistent dry cough and a two-day history of nausea and vomiting.  The patient had a recent travel history to Wuhan, China. On January 20th, 2020 the patient tested positive for COVID-19.  The patient developed pneumonia and pulmonary infiltrates, and was treated with supplemental oxygen, vancomycin, and remdesivir. By day eight of hospitalization, the patient showed significant improvement (11). 

Sequence analyses of the COVID-19 genome revealed that it has a 96.2% similarity to a bat coronavirus collected in Yunnan province, China. These analyses furthermore showed no evidence that the virus is a laboratory construct (12-14). A recent sequence analysis also found that COVID-19 shows significant variations in its functional sites, and has evolved into two major types (termed L and S). The L type is more prevalent, is likely derived from the S type, and may be more aggressive and spread more easily (14,15). 


While sequence analyses strongly suggest an initial animal-to-human transmission, COVID-19 is now a human-to-human contact spread worldwide pandemic (4,9-11). Three main transmission routes are identified; 1) transmission by respiratory droplets, 2) contract transmission, and 3) aerosol transmission (16). Transmission by droplets occurs when respiratory droplets are expelled by an infected individual by coughing and are inhaled or ingested by individuals in relatively close proximity.  Contact transmission occurs when respiratory droplets or secretions are deposited on a surface and another individual picks up the virus by touching the surface and transfers it to their face (nose, mouth, or eyes), propagating the infection. The exact time that COVID-19 remains infective on contaminated surfaces is unknown, although it may be up to several days (4,16). Aerosol transmission occurs when respiratory droplets from an infected individual mix with air and initiate an infection when inhaled (16). Transmission by respiratory droplets appears to be the most common mechanism for new infections and even normal breathing and speech can transmit the virus (4,16,17). The observation that COVID-19 can cause enteric infections also suggests that it may be spread by oral-fecal transmission; however, this has not been verified (8). A recent study has also demonstrated that about 30% of COIVID-19 patients present with diarrhea, with 20% having diarrhea as their first symptom. These patients are more likely to have COVID-19 positive stool upon testing and a longer, but less severe disease course (18).  Recently possible COVID-19 transmission from mother to newborns (vertical transmission) has been documented. The significance of this in terms of newborn health and possible birth defects is currently unknown (19). 

The basic reproductive number or R0, measures the expected number of cases generated by one infection case within a population where all the individuals can become infected. Any number over 1.0 means that the infection can propagate throughout a susceptible population (4). For COVID-19, this value appears to be between 2.2 and 4.6 (4,20,21). Unpublished studies have stated that the COVID10 R0 value may be as high as 6.6, however, these studies are still in peer review. 

COVID-19 Prevention

There is no vaccine available to prevent COVID-19 infection, and thus prevention presently centers on limiting COVID-19 exposures as much as possible within the general population (22). Recommendations to reduce transmission within community include; 1) hand hygiene with simultaneous avoidance of touching the face, 2) respiratory hygiene, 3) utilizing personal protective equipment (PPE) such as facemasks, 4) disinfecting surfaces and objects that are frequently touched, and 5) limiting social contacts, especially with infected individuals  (4,9,17,22). Hand hygiene includes frequent hand-washing with soap and water for twenty seconds, especially after contact with respiratory secretions produced by activities such as coughing or sneezing. When soap and water are unavailable, hand sanitizer that contains at least 60% alcohol is recommended (4,17,22). PPE such as N95 respirators are routinely used by healthcare workers during droplet precaution protocols when caring for patients with respiratory illnesses. One retrospective study done in Hunan, China demonstrated N95 masks were extremely efficient at preventing COVID-19 transfer from infected patients to healthcare workers (4,22-24). It is also likely that wearing some form of mask protection is useful to prevent COVID19 spread and is now recommended by the CDC (25). 

Although transmission of COVID-19 is primarily through respiratory droplets, well-studied human coronaviruses such as HCoV, SARS, and MERS coronaviruses have been determined to remain infectious on inanimate surfaces at room temperature for up to nine days. They are less likely to persist for this amount of time at a temperature of 30°C or more (26). Therefore, contaminated surfaces can remain a potential source of transmission. The Environmental Protection Agency has produced a database of appropriate agents for COVID-19 disinfection (27). Limiting social contact usually has three levels; 1) isolating infected individuals from the non-infected, 2) isolating individuals who are likely to have been exposed to the disease from those not exposed, and 3) social distancing. The later includes community containment, were all individuals limit their social interactions by avoiding group gatherings, school closures, social distancing, workplace distancing, and staying at home (28,29). In an adapted influenza epidemic simulation model, comparing scenarios with no intervention to social distancing and estimated a reduction of the number of infections by 99.3% (28). In a similar study, social distancing was estimated to be able to reduce COVID-19 infections by 92% (29). Presently, these measured are being applied in many countries throughout the world and have been shown to be at least partially effective if given sufficient time (4,17,30). Such measures proved effective during the 2003 SARS outbreak in Singapore (30). 

Symptoms, Clinical Findings, and Mortality 

On average COVID-19 symptoms appear 5.2 days following exposure and death fourteen days later, with these time periods being shorter in individuals 70-years-old or older (31,32). People of any age can be infected with COVID-19, although infections are uncommon in children and most common between the ages of 30-65 years, with men more affected than women (32,33). The symptoms vary from asymptomatic/paucisymptomatic to respiratory failure requiring mechanical ventilation, septic shock, multiple organ dysfunction, and death (4,9,32,33). The most common symptoms include a dry cough which can become productive as the illness progresses (76%), fever (98%), myalgia/fatigue (44%), dyspnea (55%), and pneumoniae (81%), with less common symptoms being headache, diarrhea (26%), and lymphopenia (44%) (4,32,33). Rare events such as COVID-19 acute hemorrhagic necrotizing encephalopathy have been documented and one paper describes conjunctivitis, including conjunctival hyperemia, chemosis, epiphora, or increased secretions in 30% of COVID-19 patients (34,35). Interestingly, about 30-60% of those infected with COVID-19 also experience a loss of their ability to taste and smell (36). 

The clinical features of COVID-19 include bilateral lung involvement showing patchy shadows or ground-glass opacities identified by chest X-ray or CT scanning (34). Patients can develop atypical pneumoniae with acute lung injury and acute respiratory distress syndrome (33). Additionally, elevations of aspartate aminotransferase and/or alanine aminotransferase (41%), C-reactive protein (86%), serum ferritin (63%), and increased pro-inflammatory cytokines, whose levels correlate positively with the severity of the symptoms (4,31-33,37-39).

About 81% of COVID-19 infections are mild and the patients make complete recoveries (38). Older patients and those with comorbidities such as diabetes, cardiovascular disease, hypertension, and chronic obstructive pulmonary disease have a more difficult clinical course (31-33,37-39). In one study, 72% of patients requiring ICU treatment had some of these concurrent comorbidities (40). According to the WHO 14% of COVID-19 cases are severe and require hospitalization, 5% are very severe and will require ICU care and likely ventilation, and 4% will die (41). Severity will be increased by older age and comorbidities (4,40,41). If effective treatments and vaccines are not found, the pandemic may cause slightly less than one-half billion deaths, or 6% of the world’s population (41). Since many individuals infected with COVID-19 appear to show no symptoms, the actual mortality rate of COIVD-19 is likely much less than 4% (42). An accurate understanding of the typical clinical course and mortality rate of COVID-19 will require time and large scale testing.         

COVID-19 Diagnosis

COVID-19 symptoms are nonspecific and a definitive diagnosis requires laboratory testing, combined with a thorough patient history.  Two common molecular diagnostic methods for COVID-19 are real-time reverse polymerase chain reaction (RT-PCR) and high-throughput whole-genome sequencing.  RT-PCR is used more often as it is cost more effective, less complex, and has a short turnaround time. Blood and respiratory secretions are analyzed, with bronchoalveolar lavage fluid giving the best test results (43). Although the technique has worked on stool samples, as yet stool is less often tested (8,43). RT-PCR involves the isolation and purification of the COVID-19 RNA, followed by using an enzyme called “reverse transcriptase” to copy the viral RNA into DNA. The DNA is amplified through multiple rounds of PCR using viral nucleic acid-specific DNA primer sequences. Allowing in a short time the COVID-19 genome ti be amplified millions of times and then easily analyzed (43). RT-PCR COVID-19 testing is FDA approved and the testing volume in the US is rapidly increasing (44,45). The FDA has also recently approved a COVID-19 diagnostic test that detects anti-COVID-19 IgM and IgG antibodies in patient serum, plasma, or venipuncture whole blood (43). As anti-COVID-19 antibody formation takes time, so a negative result does not completely preclude a COVID-19 infection, especially early infections. Last, as COVID-19 often causes bilateral pulmonary infiltrates, correlating diagnostic testing results with lung chest CT or X-ray results can be helpful (4,31-33,37-39).  

Testing for COVID-19 is based on a high clinical suspicion and current recommendations suggest testing patients with a fever and/or acute respiratory illness. These recommendations are categorized into priority levels, with high priority individuals being hospitalized patients and symptomatic healthcare facility workers. Low priority individuals include those with mild disease, asymptomatic healthcare workers, and symptomatic essential infrastructure workers. The latter group will receive testing as resources become available (41,46,47). 

COVID-19 Possible Treatments

Presently research on possible COVIS-19 infection treatments and vaccines are underway (48). At the writing of this article many different drugs are being examined, however any data supporting the use of any specific drug treating COVID-19 is thin as best. A few drugs that might have promise are:  


Hydroxychloroquine has been used to treat malarial infections for seventy years and in cell cultures it has anti-viral effects against COVID-19 (49). In one small non-randomized clinical trial in France, twenty individuals infected with COVID-19 who received hydroxychloroquine showed a reduced COVID-19 viral load, as measured on nasopharyngeal viral carriage, compared to untreated controls (50). Six individuals who also received azithromycin with hydroxychloroquine had their viral load lessened further (50). In one small study in China, a similar drug (chloroquine) was superior in reducing COVID-19 viral levels in treated individuals compared to untreated control individuals (51).  These results are preliminary, but promising. 


Remdesivir is a drug that showed value in treating patients infected with SARS (52). COVID-19 and SARS show about 80% sequence similarity and since Remdesivir has been used to treat SARS, it might have value in treating COVID-19 (52). These trials are underway (48). Remdesivir was also used to treat the first case of COIVD-19 identified within the US (11). There are many other drugs being examined to treat COVID-19 infections, however, the data on all of them is presently slight to none, and research has only begun. There is an enormous research effort underway, and progress should be rapid (48). 


Our understanding of COVID-19 is changing extremely rapidly and new findings come out daily. Combating COVID-19 effectively will require multiple steps; including slowing the spread of the virus through socially isolating and measures such as hand washing. The development of effective drug treatments and vaccines is already a priority and rapid progress is being made (48). Additionally, many areas of the world, such as South American and sub-Saharan Africa, will be affected by the COVID-19 pandemic and are likely to have their economies and healthcare systems put under extreme stress. Dealing with the healthcare crisis in these countries will be very difficult. Lastly, several recent viral pandemics (SARS, MERS, and COVID-19) have come from areas where wildlife is regularly traded, butchered, and eaten in conditions that favor the spread of dangerous viruses between species, and eventually into human populations. The prevention of new viral pandemics will require improved handling of wild species, better separation of wild animals from domestic animals, and better regulated and lowered trade in wild animals, such as bats, which are known to be a risk for carrying potentially deadly viruses to human populations (53). 


  1. Origin and evolution of pathogenic coronaviruses. Nat Rev Microbiol 2019;17:181-92. 
  2. Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation. Science 2020;367:1260-3.
  3. Structural basis for the recognition of SARS-CoV-2 by full-length human ACE2. Science 2020;367:1444-8. 
  4. CDC. 2019 Novel coronavirus, Wuhan, China. 2020. Accessed 6 April 2020. 
  5. SARS and MERS: recent insights into emerging coronaviruses.Nat Rev Microbiol 2016;14:523-34. 
  6. Coronaviruses post-SARS: update on replication and pathogenesis. Nat Rev Microbiol. 2009;7:439-50.
  7. The Novel Coronavirus: A Bird’s Eye View. Int J Occup Environ Med. 2020;11:65-71. 
  8. Evidence for Gastrointestinal Infection of SARS-CoV-2. Gastroenterology 2020;S0016-5085:30282-1. 
  9. Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia. N Engl J Med. 2020;382:1199-1207. 
  10. A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating a person-to-person transmission: a study of a family cluster. Lancet 2020;395:514-23. 
  11. First Case of 2019 Novel Coronavirus in the United States.N Engl J Med. 2020;382:929-36. 
  12. A pneumonia outbreak associated with a new coronavirus of probable bat origin. Nature 2020;579:270-3. 
  13. Full-genome evolutionary analysis of the novel coronavirus (2019-nCoV) rejects the hypothesis of emergence as a result of a recent recombination event. Infect Genet Evol. 2020;79:104212. 
  14. The proximal origin of SARS-CoV-2. Nat Med. 2020.
  15. On the origin and continuing evolution of SARS-CoV-2 Natl Sci Review 2020. 
  16. National Health Commission of People’s Republic of China. Prevent guidelines of 2019-nCoV. 2020. Accessed 6 April 2020.
  17. Transmission Potential of SARS-CoV-2 in Viral Shedding Observed at the University of Nebraska Medical Center
  18. Digestive symptoms in CVOID-19 patients with mild disease severity: Clinical presentation, stool viral RNA testing, and outcomes. 
  19. Neonatal Early-Onset Infection With SARS-CoV-2 in 33 Neonates Born to Mothers With COVID-19 in Wuhan, China. JAMA Pediatr.  doi:10.1001/jamapediatrics.2020.0878. 
  20. Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia. N Engl J Med. 2020;382:1199-1207. 
  21. Data-based analysis, modelling and forecasting of the COVID-19 outbreak. PLoS One 2020;15:e0230405. 
  22. Covid-19 – Navigating the Uncharted. N Engl J Med. 2020;382:1268-9. 
  23. Rational use of face masks in the COVID-19 pandemic. Lancet Respir Med. 2020;S2213-2600(20)30134-X. 
  24. Association between 2019-nCoV transmission and N95 respirator use. J Hosp Infect. 2020i:S0195-6701(20)30097-9. 
  25. Recommendation Regarding the Use of Cloth Face Coverings, Especially in Areas of Significant Community-Based Transmission.
  26. COVID-19 outbreak on the Diamond Princess cruise ship: estimating the epidemic potential and effectiveness of public health countermeasures. J Travel Med. 2020 Feb 28. 
  28. Interventions to mitigate early spread of SARS-CoV-2 in Singapore: a modelling study. Lancet Infect Dis. 2020;S1473-3099(20)30162-6.
  29. The effect of control strategies to reduce social mixing on outcomes of the COVID-19 epidemic in Wuhan, China: a modelling study. Lancet, Public Health
  30. SARS in Singapore–key lessons from an epidemic. Ann Acad Med Singapore 2006;35:301-6. 
  31. Early Transmission Dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia. N Engl J Med. 2020;382:1199-207. 
  32. Updated understanding of the outbreak of 2019 novel coronavirus (2019-nCoV) in Wuhan, China. J Med Virol. 2020;92:441-7. 
  33. Clinical features of patents infected with novel 2019 coronavirus in Wuhan, China. Lancet 2020;395:497-506.
  34. COVID-19-associated Acute Hemorrhagic Necrotizing Encephalopathy: CT and MRI Features. Radiology 2020:201187. 
  35. Characteristics of Ocular Findings of Patients With Coronavirus Disease 2019 (COVID-19) in Hubei Province, ChinaJAMA Ophthalmol. 2020. doi: 10.1001/jamaophthalmol.2020.1291. 
  36. A New Symptom of COVID-19: Loss of Taste and Smell. 38. Obesity. 2020. doi: 10.1002/oby.22809.
  37. Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl J Med. 2020 DOI: 10.1056/NEJMoa2002032.
  38. Clinical course and mortality risk of severe COVID-19. Lancet 2020;395:507-13. 
  39. The cytokine release syndrome (CRS) of severe COVID-19 and Interleukin-6 receptor (IL-6R) antagonist Tocilizumab may be the key to reduce the mortality. Int J Antimicrob Agents 2020;28:105954. 
  40. Clinical Characteristics of 138 Hospitalized Patients with 2019 Novel Coronavirus–Infected Pneumonia in Wuhan, China.  JAMA 2020; 323:1061-9. 
  41. Unknown unknowns – COVID-19 and potential global mortality. Early Hum Dev. 2020;144:105026. 
  42. Substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (SARS-CoV2). Science 2020 Mar 16. pii: eabb3221.
  43. Detection of 2019 novel coronavirus (2019-nCoV) by real-time RT-PCREuro Surveill. 2020;25. 44.
  45. Centers for Disease Control and Prevention. Evaluating and Testing Persons for Coronavirus Disease 2019 (COVID-19)
  46. Infectious Diseases Society of America. COVID-19 Prioritization of Diagnostic Testing.
  47. Race to find COVID-19 treatments accelerates. Science  2020:367;1412-3.
  48. Hydroxychloroquine, a less toxic derivative of chloroquine, is effective in inhibiting SARS-CoV-2 infection in vitro. Cell Discov. 202018;6:16. 
  49. Hydroxychloroquine and azithromycin as a treatment of COVID-19: results of an open-label non-randomized clinical trial. Int J Antimicrob Agents 2020:105949. 
  50. Breakthrough: Chloroquine phosphate has shown apparent efficacy in treatment of COVID-19 associated pneumonia in clinical studies. Biosci Trends 2020;14(1):72-73.
  51. Remdesivir for severe acute respiratory syndrome coronavirus 2 causing COVID-19: An evaluation of the evidence. Travel Med Infect Dis. 2020 2:101647.
  52. Permanently ban wildlife consumption.

Clinical determinants of the severity of COVID-19: A systematic review and meta-analysis




We aimed to systematically identify the possible risk factors responsible for severe cases.


We searched PubMed, Embase, Web of science and Cochrane Library for epidemiological studies of confirmed COVID-19, which include information about clinical characteristics and severity of patients’ disease. We analyzed the potential associations between clinical characteristics and severe cases.


We identified a total of 41 eligible studies including 21060 patients with COVID-19. Severe cases were potentially associated with advanced age (Standard Mean Difference (SMD) = 1.73, 95% CI: 1.34–2.12), male gender (Odds Ratio (OR) = 1.51, 95% CI:1.33–1.71), obesity (OR = 1.89, 95% CI: 1.44–2.46), history of smoking (OR = 1.40, 95% CI:1.06–1.85), hypertension (OR = 2.42, 95% CI: 2.03–2.88), diabetes (OR = 2.40, 95% CI: 1.98–2.91), coronary heart disease (OR: 2.87, 95% CI: 2.22–3.71), chronic kidney disease (CKD) (OR = 2.97, 95% CI: 1.63–5.41), cerebrovascular disease (OR = 2.47, 95% CI: 1.54–3.97), chronic obstructive pulmonary disease (COPD) (OR = 2.88, 95% CI: 1.89–4.38), malignancy (OR = 2.60, 95% CI: 2.00–3.40), and chronic liver disease (OR = 1.51, 95% CI: 1.06–2.17). Acute respiratory distress syndrome (ARDS) (OR = 39.59, 95% CI: 19.99–78.41), shock (OR = 21.50, 95% CI: 10.49–44.06) and acute kidney injury (AKI) (OR = 8.84, 95% CI: 4.34–18.00) were most likely to prevent recovery. In summary, patients with severe conditions had a higher rate of comorbidities and complications than patients with non-severe conditions.


Patients who were male, with advanced age, obesity, a history of smoking, hypertension, diabetes, malignancy, coronary heart disease, hypertension, chronic liver disease, COPD, or CKD are more likely to develop severe COVID-19 symptoms. ARDS, shock and AKI were thought to be the main hinderances to recovery.

For More Information:

Anosmia and dysgeusia in SARS-CoV-2 infection: incidence and effects on COVID-19 severity and mortality, and the possible pathobiology mechanisms – a systematic review and meta-analysis

Authors: Endang Mutiawati, Conceptualization, Data Curation, Resources, Validation, Writing – Original Draft Preparation, Writing – Review & Editing,a,1,2Marhami Fahriani, Conceptualization, Data Curation, Investigation, Methodology, Validation, Writing – Original Draft Preparation, Writing – Review & Editing,3Sukamto S. Mamada, Data Curation, Investigation, Validation, Writing – Review & Editing,4Jonny Karunia Fajar, Conceptualization, Formal Analysis, Investigation, Methodology, Writing – Review & Editing,3,5Andri Frediansyah, Data Curation, Investigation, Writing – Original Draft Preparation, Writing – Review & Editing,6Helnida Anggun Maliga, Data Curation, Investigation, Validation, Writing – Review & Editing,7Muhammad Ilmawan, Data Curation, Investigation, Validation, Writing – Review & Editing,7Talha Bin Emran, Validation, Writing – Review & Editing,8Youdiil Ophinni, Investigation, Validation, Writing – Review & Editing,9Ichsan Ichsan, Validation, Writing – Review & Editing,3,10Nasrul Musadir, Validation, Writing – Review & Editing,1,2Ali A. Rabaan, Validation, Writing – Review & Editing,11Kuldeep Dhama, Supervision, Validation, Writing – Review & Editing,12Syahrul Syahrul, Supervision, Validation, Writing – Review & Editing,1,2Firzan Nainu, Data Curation, Investigation, Supervision, Validation, Writing – Review & Editing,4 and Harapan aPreparation, Writing – Review & Editing3,10,13


Background: The present study aimed to determine the global prevalence of anosmia and dysgeusia in coronavirus disease 2019 (COVID-19) patients and to assess their association with severity and mortality of COVID-19. Moreover, this study aimed to discuss the possible pathobiological mechanisms of anosmia and dysgeusia in COVID-19.

Methods: Available articles from PubMed, Scopus, Web of Science, and preprint databases (MedRxiv, BioRxiv, and Researchsquare) were searched on November 10th, 2020. Data on the characteristics of the study (anosmia, dysgeusia, and COVID-19) were extracted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline. Newcastle–Ottawa scale was used to assess research quality. Moreover, the pooled prevalence of anosmia and dysgeusia were calculated, and the association between anosmia and dysgeusia in presence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was assessed using the Z test.

Results: Out of 32,142 COVID-19 patients from 107 studies, anosmia was reported in 12,038 patients with a prevalence of 38.2% (95% CI: 36.5%, 47.2%); whereas, dysgeusia was reported in 11,337 patients out of 30,901 COVID-19 patients from 101 studies, with prevalence of 36.6% (95% CI: 35.2%, 45.2%), worldwide. Furthermore, the prevalence of anosmia was 10.2-fold higher (OR: 10.21; 95% CI: 6.53, 15.96, p < 0.001) and that of dysgeusia was 8.6-fold higher (OR: 8.61; 95% CI: 5.26, 14.11, p < 0.001) in COVID-19 patients compared to those with other respiratory infections or COVID-19 like illness. To date, no study has assessed the association of anosmia and dysgeusia with severity and mortality of COVID-19.

Conclusion: Anosmia and dysgeusia are prevalent in COVID-19 patients compared to those with the other non-COVID-19 respiratory infections. Several possible mechanisms have been hypothesized; however, future studies are warranted to elucidate the definitive mechanisms of anosmia and dysgeusia in COVID-19.

For More Information:

Overweight/obesity as the potentially most important lifestyle factor associated with signs of pneumonia in COVID-19

  • Published: November 18, 2020



The occurrence of pneumonia separates severe cases of COVID-19 from the majority of cases with mild disease. However, the factors determining whether or not pneumonia develops remain to be fully uncovered. We therefore explored the associations of several lifestyle factors with signs of pneumonia in COVID-19.


Between May and July 2020, we conducted an online survey of 201 adults in Germany who had recently gone through COVID-19, predominantly as outpatients. Of these, 165 had a PCR-based diagnosis and 36 had a retrospective diagnosis by antibody testing. The survey covered demographic information, eight lifestyle factors, comorbidities and medication use. We defined the main outcome as the presence vs. the absence of signs of pneumonia, represented by dyspnea, the requirement for oxygen therapy or intubation.


Signs of pneumonia occurred in 39 of the 165 individuals with a PCR-based diagnosis of COVID-19 (23.6%). Among the lifestyle factors examined, only overweight/obesity was associated with signs of pneumonia (odds ratio 2.68 (1.29–5.59) p = 0.008). The observed association remained significant after multivariate adjustment, with BMI as a metric variable, and also after including the antibody-positive individuals into the analysis.


This exploratory study finds an association of overweight/obesity with signs of pneumonia in COVID-19. This finding suggests that a signal proportional to body fat mass, such as the hormone leptin, impairs the body’s ability to clear SARS-CoV-2 before pneumonia develops. This hypothesis concurs with previous work and should be investigated further to possibly reduce the proportion of severe cases of COVID-19.

For More Information: