Pneumonia in medical professionals during COVID-19 outbreak in cardiovascular hospital.

OBJECTIVES: The pandemic of coronavirus associated disease (COVID-19) placed the health care workers at high risk. We investigated clinical and treatment characteristics of infected medical professionals in a cardiovascular hospital.
METHODS: The study was retrospective, conducted in tertiary cardiovascular hospital and included employees with confirmed coronavirus infection. They filled out a questionnaire about health status, symptoms, admission to hospital and treatment. The vaccination status against tuberculosis, hepatitis B and seasonal influenza was assessed. Pneumonia was defined as CT finding of ground glass opacifications (GGO) with consolidations typical for COVID-19.
RESULTS: The study included 107 confirmed cases of COVID - 19 out of 726 employees (15%). Most of the infected were from cardiac surgery department (74/107, 69%). Substantial number of employees did not have any symptoms [31 (28.9%)] and 38 patients (35.5%) were admitted to hospital. The average hospital length of stay was 8.1 ± 5.6 days. Seventy-five of 107 (70.1%) received seasonal influenza vaccine. Pneumonia with CT features of GGO and consolidation occurred in 25/107 (23.4%) patients of which 14/107 (13.1%) had bilateral involvement. In multivariate logistic regression analysis including recognized characteristics associated with worse outcomes in COVID-19 (obesity, diabetes mellitus, coronary artery disease, cerebrovascular disease, current smoking, heart failure, influenza immunization), only influenza immunization remained an independent predictor of occurrence of bilateral pneumonia (OR 0.207; 95%CI[0.050 - 0.847]; p = 0.029).
CONCLUSIONS: The association of influenza immunization and less aggressive form of pneumonia might provide a finding that supports the institution of preventive measures that can be beneficial in reduction of global coronavirus burden.

Introduction

Coronavirus disease (COVID-19) caused by RNA virus has caused a worldwide pandemic. The infection can present with various clinical scenarios, ranging from completely asymptomatic course to life-threatening forms of acute respiratory distress syndrome and multi-organ failure (Guan et al., 2020). The disease was first reported as an unusual form of pneumonia in Wuhan, Hubei province, China in December 2019 (Huang et al., 2020). Multi-slice CT has been extensively used for early identification of COVID-19 pneumonia. Typical CT features include peripheral, frequently bilateral, ground glass opacifications (GGO) with multifocal distribution, followed by linear consolidations. These consolidations tend to rapidly progress towards organized pneumonia patterns on follow up examinations (Huang et al., 2020, Lomoro et al., 2020, Hani et al., 2020). These distinctive CT features of COVID-19 can be useful to differentiate it from other common causes of pneumonia, like bacterial infection.

The pandemic has placed the health care workers at high risk of being infected with a new variety of coronavirus. There were reports of more than 3300 healthcare workers being infected in China, with variable rates between 2 and 20% of all individuals with COVID-19, many of which have died. In Italy, a country also heavily struck by COVID-19, around 20% of all medical professionals have been infected by the virus (The Lancet 2020; Heymann and Shindo (2020)).

In Serbia, the rates of COVID-19 infections were relatively low, with total number of cases of 12310, out of whom 254 persons died due to the infection (1409 infected per 1 million inhabitants) (COVID-19 Pandemic 2020). Medical workers in Serbia were also affected, but the total number of them cannot be estimated. So far, the reports of medical institutions in which there was an outbreak of COVID-19 were scarce. We, therefore, investigated the demographic, clinical, diagnostic and treatment characteristics of infected health care workers in a large tertiary cardiovascular hospital struck by an epidemic.

Methods

The study was retrospective and conducted in a single, high-volume university cardiovascular center with 24 hours cardiac surgery and interventional cardiology service. Study included healthcare workers currently at service in this institution that were infected with coronavirus and had confirmed diagnosis of COVID-19 using polymerase chain reaction (PCR). The participants gave consent to take part in the study and were asked to fill out the predefined questionnaire. The study was conducted in accordance with Helsinki declaration.

Study population

The healthcare workers, after consenting to participate in the study, were asked questions about their previous medical conditions, ongoing diseases and prescribed medications. The questionnaire also included segments about symptoms related to COVID-19, dates of hospital admissions, diagnostic procedures and therapies they were subjected to. Also, the study explored the use of supplemental oxygen, mechanical ventilation and antibiotics in hospitalized patients. The medical records have been collected from the institutions that treated them. After completing the study, vital status of the participants was assessed via the telephone interview conducted by one of the investigators.

The diagnosis was made after collecting throat and nose swabs that have been tested for presence of coronavirus RNA using reverse transcriptase polymerase chain reaction (RT PCR). The tests were done using real time fluorescent RT-PCR kit for detecting 2019-nCoV (BGI Genomics, Wuhan, China). The tests were repeated after 7 and 14 days.

Data collection

The data collected from the patients included age, gender, work position, health status prior to being infected with coronavirus, presence of chronic disease and prescribed medications. The diseases that could influence the clinical course severity of COVID-19 were active or previous cancer, hypertension, diabetes mellitus, coronary artery disease (defined as previous myocardial infarction, percutaneous or surgical revascularization or known severe coronary artery disease not amenable to treatment), cerebrovascular disease (defined as previous cerebrovascular accident or known severe carotid artery disease and/or treated with surgery or stent implantation), heart failure (defined as left ventricular ejection fraction of less than 35%), chronic renal failure (defined as glomerular filtration rate of less than 60 ml/min/m2) chronic obstructive pulmonary disease (COPD), bronchial asthma, obesity (defined as body mass index greater than 30 kg/m2), smoking status (current or former). COVID-19 symptoms were also assessed: sore throat, cough, rhinitis, fever, myalgia, loss of sense of smell or taste, diarrhoea, headache, shortness of breath and tiredness. The vaccination status was also assessed focusing on immunization against tuberculosis, hepatitis B and seasonal influenza.

Pneumonia was defined as CT finding of GGO with consolidations typical for COVID-19 pneumonia together with presence of symptoms and was further classified as unilateral or bilateral based on the distribution of CT findings (Huang et al., 2020, Lomoro et al., 2020, Hani et al., 2020).

The medical records of hospitalized patients were used to collect the data regarding duration of hospitalization, use of supplemental oxygen, need for non-invasive or invasive mechanical ventilation, intensive care unit treatment, use of antibiotics, antimalarial drug chloroquine or immunomodulatory medications.

Statistical analysis

Continuous data were summarized as the means ± standard deviation (SD). Categorical data were summarized as counts and percentages. T-test was used for comparing continuous variables if their distribution was normal and Mann-Whitney U test was used for continuous variables that did not have normal distribution. The chi-squared test and Fisher’s exact test were used for testing categorical variables. The effects of patient’s characteristics on endpoint of confirmed bilateral pneumonia were explored using multivariate logistic regression, and their impact was expressed as odds ratio with 95% confidence interval. Forward stepwise regression was used to determine independent predictors of occurrence of bilateral pneumonia in a multivariate analysis model. The stepwise selection iteratively selected the most significant variable with multivariate p-value <0.25, to start the model. At each step, another significant variable is added and after running the model, a check was performed to remove the variable with a multivariate p-value >0.10. This was repeated with the complete set of variables until no more variables could be entered and no variables could be dropped. The p value of <0.05 was considered statistically significant. All statistical analyses were performed using PASWStatistics 18.0 statistical software (SPSS Inc, Chicago, Illinois, USA).

Results

The study included 107 COVID-19 confirmed cases out of 726 employees (15%) in active service when the diagnosis was established. The cases were recorded from March 20th till April 22nd 2020 at the Institute for cardiovascular diseases Dedinje, Belgrade, Serbia.. The professional profile of patients was 28 physicians (26.1%), 64 nurses (59.8%), 5 radiology technicians (4.6%), 4 physical therapists (3.8%) 2 cardiovascular perfusionists (1.9%) and 4 persons from hospital’s non-medical services (3.8%). Most of the patients came from department of cardiac surgery with complementary services of anaesthesiology, intensive care unit and postoperative ward (74/107, 69.1%). while cardiology department and services that spanned across entire hospital, like radiology and physical rehabilitation had fewer cases of infection (Table 1 ).The average age of infected personnel was below 40 years and almost 30% of them were active smokers. The clinical data regarding the patients are displayed in Table 2 .

Table 1

Distribution of infected medical professionals by department

Department	Infected	At service	%
Cardiac surgery	13	109	11.9
Anaesthesiology & ICU	36	151	23.8
Postoperative department	25	196	12.7
Cardiology n (%)	15	122	12.3
Radiology n (%)	8	56	14.3
Physical therapy	6	20	30.0
Non-medical service	4	72	5.6
Total	107	726	15.0

ICU-intensive care unit

Table 2

Clinical characteristics of all patients, hospitalized vs non-hospitalized

Variable	Total (n = 107)	H (n = 38)	NH (n = 69)		P value
Age (years)	39.1 ± 11.4	38.1 ± 9.8		39.9 ± 12.4	0.471
No previous disease n (%)	71 (66.3)	21 (55.3)		50 (72.2)	0.043
Malignancy n (%)	2 (1.9)	1 (2.6)		1 (1.5)	0.587
Hypertension n (%)	10 (9.3)	5 (13.1)		5 (7.2)	0.736
Coronary artery disease n (%)	4 (3.8)	1 (2.8)		3 (4.3)	0.640
Heart failure	1 (0.9)	0 (0.0)		1 (1.5)	0.587
Diabetes mellitus n (%)	3 (2.8)	1 (2.8)		2 (2.9)	1.000
Bronchial asthma n (%)	1 (0.9)	0 (0.0)		1 (1.5)	0.587
COPD n (%)	0 (0.0)	0 (0.0)		0 (0.0)	-------
Chronic renal failure n (%)	0 (0.0)	0 (0.0)		0 (0.0)	-------
Previous CVD n (%)	2 (1.9)	0 (0.0)		2 (2.9)	0.510
Thyroid disease n (%)	8 (7.5)	7 (18.4)		1 (1.5)	0.007
Arrhythmia n (%)	4 (3.8)	2 (5.2)		2 (2.9)	0.567
BMI > 30 kg/m2 n (%)	5 (4.7)	3 (7.9)		2 (2.9)	0.337
Current smoking n (%)	32 (29.9)	11 (28.9)		21 (30.4)	0.378

BMI – body mass index; COPD – chronic obstructive pulmonary disease; CVD – cerebrovascular disease; H – hospitalized, NH – non-hospitalized

The epidemic started on March 20th when first two cases among medical staff were recorded and on April 04th there was a sharp rise in the number of infected personnel reaching its peak by April nd.. In the same time frame there was an increase in number of infected patients that were hospitalized at cardiac surgery department (Fig. 1 ).

Fig. 1

COVID-19 cases among patients and staff. The diagram depicts temporal trends in cumulative infection of medical professionals and cardiac surgery patients.

Regarding manifestations of COVID-19 infection, substantial number of employees did not have any complaints [31 (28.9%)]. Symptomatic individuals predominantly complained of sore throat, general weakness and fever (Table 3 ). Total number of 38 patients (35.5%) were admitted to hospital for treatment. The average hospital length of stay was 8.1 ± 5.6 days. Most of others, that did not require hospital treatment, were quarantined at home (41/107, 38.3%) or at one of the facilities for collective quarantine (28/107, 26.2%). The hospitalized medical professionals were less likely to be free of disease prior to coronavirus infection and in this group there were more persons having thyroid disorders (Table 2). In hospital t most of them received chloroquine 21/38 (55.2%) and underwent treatment with various intravenous antibiotics (Table 4 ). Three patients [3/38 (7.9%)] were admitted to intensive care unit because of prolonged hypoxia and one of them was intubated and treated with positive pressure ventilation for 6 days. This group received Interleukin 6 (IL-6) inhibitor tocilizumab.There were no deaths in the whole cohort of patients (Fig. 2 ).

Table 3

Complaints of symptomatic patients with COVID-19.

Symptom	n = 76
Sore throat n (%)	38 (50.0)
Rhinorrhea n (%)	23 (30.2)
Fever n (%)	43 (56.6)
Fever < 37.3⁰ C n (%)	10 (13.2)
Fever 37.3 – 38.0⁰ C n (%)	23 (30.2)
Fever 38.1 – 39⁰ C n (%)	10 (13.2)
Fever > 39.0⁰ C n (%)	0 (0.0)
Cough n (%)	31 (40.8)
General weakness n (%)	37 (48.7)
Headache n (%)	20 (26.3)
Diarrhea n (%)	21 (27.6)
Myalgia n (%)	22 (28.9)
Breathlessness n (%)	13 (17.1)
Loss of smell or taste n (%)	20 (26.3)

Table 4

Choice of antibiotics in hospitalized COVID-19 patients.

Antibiotic	n = 38
Azithromycin n (%)	19 (50.0)
Ceftriaxone n (%)	16 (42.1)
Levofloxacin n (%)	3 (7.9)
Ciprofloxacin n (%)	1 (2.6)
Cefixime n (%)	1 (2.6)
Vancomycin n (%)	2 (5.2)
Amoxycillin n (%)	1 (2.6)
Metronidazole n (%)	1 (2.6)

Fig. 2

Forest plot of risk factors for bilateral pneumonia. The diagram displays impact of patients’ characteristics in multivariable regression on occurrence of bilateral pneumonia. Odds ratio (OR) and 95% confidence interval (CI) are given for each patients’ characteristics.

Vaccinationl status was assessed in all patients. Each of them had been immunized against tuberculosis Bacillus Calmette Guerin (BCG), since it is still mandatory for every new-born in Serbia. In this group 73/107 (68.2%) have been immunized against Hepatitis B. Seasonal influenza vaccine received 75/107 (70.1%) of infected medical professionals. There were no differences in immunization status between the hospitalized (H) and non-hospitalized (NH) patients [influenza H 28/38 (71.1%) vs NH 47/69 (68.1%), p = 0.414; hepatitis B H 28/38 (73.7%) vs NH 45/69 (65.2%), p = 0.457].

Pneumonia with CT features of GGO and consolidation occurred in 25/107 (23.4%) patients of which 14/107 (13.1%) had bilateral involvement. Bilateral pneumonia was less frequent in patients that had received seasonal influenza vaccine [influenza vaccine 6/75 (8.0%) vs no influenza vaccine 8/32 (25%), p = 0.07] In a multivariate logistic regression analysis that included previously recognized patients’ characteristics that were associated with worse outcomes in COVID-19 ( obesity, diabetes mellitus, coronary artery disease, cerebrovascular disease, current smoking, heart failure, seasonal influenza immunization), only seasonal influenza immunization remained an independent predictor of occurrence of bilateral pneumonia verified on CT (OR = 0.207; 95%CI [0.050, 0.847]; p = 0.029) (Table 5 ) (Shi et al., 2020, Hajifathalian et al., 2020, Wang et al., 2020).

Table 5

Univariate and multivariate predictors of CT confirmed bilateral pneumonia.

	Univariate			Multivariate
	OR [95% CI]	P-value		OR [95% CI]	P-value
Diabetes mellitus	7.182 [0.418 – 123.251]	0.174		2.9E [0.000 – ]	1.000
Hypertension	1.800 [0.334 – 9.704]	0.494		- - -	- - -
Coronary artery disease	2.333 [0.223 – 24.459]	0.480		- - -	- - -
Seasonal influenza shot	0.297 [0.082 – 1.074]	0.064		0.207 [0.050 – 0.847]	0.029
Obesity	5.133 [0.763 – 34.549]	0.093		15.644 [0.732 – 334.336]	0.078
Current smoking	3.000 [0.615 – 14.626]	0.174		10.532 [0.976 – 113.628]	0.052
Heart failure	0.000 [0.000 – ]	1.000		- - -	- - -

CT – computerized tomography; OR – odds ratio.

Discussion

The coronavirus pandemic has heavily harmed the healthcare system in every country. Its impact was partly caused by a rapid spread of the disease in the population and partly by the fact that substantial number of infected required medical attention and hospitalization. Due to its effect of causing profound hypoxia in great number of infected individuals, leading to the need for supplemental oxygen and mechanical ventilation, it has placed a heavy burden on hospitals, especially intensive care units and their staff (Guan et al., 2020, Covid-19 Coronavirus Pandemic, 2020). Highly pathogenic and infective nature of coronavirus, along with increased number of diseased patients admitted to hospitals, has generated an outbreak of infection among health care workers, further hampering the function of already stressed healthcare systems (Wang et al., 2020, Otter et al., 2016).

In the study we have explored the outbreak of COVID-19 in tertiary cardiovascular institution. The first infected persons among healthcare professionals were detected on March 20th when two tested positive. This led to a rapid spread of infection and sharp increase in the number of infected individuals at the beginning of April, corresponding with the incubation period of coronavirus (Guan et al., 2020, Huang et al., 2020). The possible source of infection could have been a female patient that was admitted to the hospital on March 8th presenting as ST elevation myocardial infarction with normal coronary angiogram and history of fever. The patient was resuscitated, but unfortunately succumbed on the day of admission. At the same time, first of the infected from cardiac surgery department reported travelling abroad to Austria for winter vacation where he could have contracted the coronavirus. After this, PCR test was done in every hospitalized cardiac surgery patient including ones admitted and waiting for surgery and in medical staff that were in contact with the infected persons. On April 5th 9 of cardiac surgery patients tested positive, including 5 in the preoperative ward who were completely asymptomatic (Fig. 1). All this led to a complete shutdown of cardiac surgery department together with intensive care unit and postoperative care ward.

The highest percentage of infected individuals has been recorded at the anaesthesia with intensive care unit (ICU) and physical therapy departments. This can be explained by the close contacts with the patients and co-workers that these persons have in their daily work. Airway management, central venous line placement and thoracic drainage are procedures associated with increased risk of viral dissemination (Cook et al., 2020). On the other hand, physical therapy of cardiovascular patients carries an increased risk of infection for a person performing the treatment. Additionally, their work consists of visits to different departments which increases the number of contacts with patients and other employees further aggravating the possibility of infestation (Felten-Barentsz et al., 2020).

Most of the medical professionals – patients were young with more than half of them without any previous medical condition. Interestingly, almost 30% were completely asymptomatic, which was not uncommon especially in the population of young, previously healthy persons (Bai et al., 2020, Hu et al., 2020). The reported rates of asymptomatic individuals varied substantially between the studies from as low as 1% in large cohort of Chinese patients to more than 50% as reported in the study from the cruiser ship Diamond Princess. The need to treat an asymptomatic carrier of coronavirus has been a matter of debate (Gao et al., 2020, Mizumoto et al., 2020). In our group, asymptomatic individuals were not treated neither with antiviral agents, nor with antibiotics or chloroquine and their disease course was uneventful. The significance of asymptomatic patients as a source of infection depends on their distribution in the population and the amount and duration of virus dispensation from the upper respiratory tract (Gao et al., 2020). In our group viral dissemination among close contacts, as estimated by the patients, was very low in the whole group (0.18 ± 0.61 individuals), but it has to be noted that this was self-estimate that hasn’t been further explored with systematic testing of the families and close contacts. This could be also attributed to the fact that the patients were medically educated and followed the measures of self-isolation and hygiene from the moment of COVID-19 diagnosis..

Fewer non-hospitalized medical professionals had previous diseases and less frequently had thyroid disorders (Table 2). The hospitalized ones had higher incidence of hypertension and obesity that, albeit statistically non-significant, could be associated with higher incidence of hospital admission for COVID-19 (Zhou et al., 2020, Driggin et al., 2020). The higher incidence of thyroid disease among hospitalized individuals could be caused by susceptibility of immune system to coronavirus and excessive and disproportionate response to viral infection. The autoimmune response to viral particles has been previously associated with thyroid disease (Hashimoto thyroiditis). Increased immune response to coronavirus in a person with autoimmune thyroid disorder could lead to more severe clinical manifestations and increased risk of hospital admission (Desailloud and Hober, 2009, Shukla et al., 2018, Gasparyan et al., 2020).

The low rates of death and respiratory distress syndrome could be attributed to overall profile of the infected population consisting of young, relatively healthy individuals. The only patient requiring mechanical ventilation was a 63-year-old radiology technician who was obese, a smoker with multiple comorbidities (hypertension, diabetes mellitus andcoronary artery disease). This patient’s characteristics complied with the risk profile for intensive care admission and poor outcome in COVID-19 (Driggin et al., 2020).

The principal finding of this report is a lower incidence of bilateral pneumonia, confirmed by CT, in the group of patients previously immunized with seasonal influenza vaccine. So far, there have been no studies suggesting the cross-reactivity to SARS-CoV-2 in persons immunized against influenza. The idea is based on the premise that individuals vaccinated against a viral disease produce antibodies that can cross-react against the proteins of various viral species and thus prevent other viral infections or cause individuals to have mild form of the illness. This hypothesis was used to explain the lower rates of COVID-19 in children in the countries where vaccination is mandatory against measles and rubella (Gasparyan et al., 2020, Salman and Salem, 2020). The basis for this could be the similar ontogenesis of various viral species. Coronaviridea, to which SARS-CoV-2 belongs, and Orthomyxoviridae, a group to which influenza virus belongs, both RNA viral species, express similarity in the structure of hemagglutinin esterases that seems to be a consequence of a process named lateral gene transfer that occurred 8000 years ago. This process supposedly originated from influenza C virus, after separating from influenza A and B, to one of the viruses from Coronaviridae genera (Zeng et al., 2008, Wang et al., 2008). The seasonal influenza vaccine for 2019/2020 contained three or four strains of influenza A and B which can potentially exhibitcross-reactivity with some of the proteins of SARS-COV-2 virus.

Another possible mechanism is the so called “trained” immunity phenomenon, meaning that Influenza vaccine could modulate and improve immune response to another virus. This mechanism has already been attributed to mandatory vaccination against tuberculosis. It has been hypothesized that in countries where persons have been immunized with multiple shots of BCG they have acquired improved immune response to other pathogens, including SARS-CoV-2, leading to lower rates of infection and severe complications like acute respiratory distress syndrome and death (Gasparyan et al., 2020).

The presented study has several limitations. The study group is relatively small and the incidence of bilateral pneumonia is low. This can limit the validity of conclusions that can be drawn regarding the positive effects of seasonal influenza immunization as an adjunctive measure in fighting global burden of COVID-19. This association between previous immunization status and susceptibility to coronavirus infection is hypothesis generating and should be further explored.

Conclusions

Coronavirus pandemic posed an immense risk for medical professionals in hospitals where cardiovascular surgery and interventions are performed. The possible association of seasonal influenza immunization and less aggressive form of COVID-19 pneumonia in infected medical staff might provide a finding that supports the institution of preventive measures that can be beneficial in treatment and reduction of global SARS-CoV-2 burden.

Disclosure statement: The authors report no financial relationships, source of funding or conflicts of interest regarding the contents of the manuscript.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding source

This research project has not been sponsored or partially funded by any agency or company.

Ethical approval

The manuscript has been approved by the Institutional Ethics committee.