Comparison of Blood Test Data Between Patients With COVID-19 and Inuenza: a Retrospective Cross-sectional Survey in a Japanese University Hospital

Background: Characteristics features, including blood test data, of COVID-19 compared to those of inuenza have not been dened. Purpose: We aimed to compare clinical parameters, including blood test data, between COVID-19 and inuenza. Methods: This retrospective cross-sectional survey was conducted at Juntendo University Nerima Hospital, Tokyo, Japan. We recruited patients diagnosed with COVID-19 between January 1 and December 31, 2020, and all patients underwent blood tests. For comparison, we recruited patients diagnosed with inuenza. The number of patients with inuenza and those with COVID-19 was the same. All patients with inuenza underwent blood tests and the period was dated back from December 31, 2020 until the period that was the same number of COVID-19. Results: During the study period, 228 patients (male:female, 123 [54.0%]:105 [46.0%], age [mean±standard deviation], 54.68±18.98 years) were diagnosed with COVID-19. We recruited the same number of inuenza patients (228 patients; male:female, 129 [56.6%]:99 [43.4%], age [mean±standard deviation], 69.6±21.25). Age 15–70 years (vs. 71 years), diculty in breathing, malaise, and percentage of lymphocytes >20% were signicantly more frequent in COVID-19. However, nausea, body temperature >38.1°C, and white blood cell count >9000/μL were more frequent in inuenza. Conclusion: Our results are useful for differentiating COVID-19 from inuenza.


Introduction
In December 2019, a series of pneumonia cases of unknown cause were reported, involving clinical presentations that greatly resembled those of viral pneumonia [1][2][3]. The coronavirus disease (COVID- 19) pandemic has been a great threat to human life. Many case reports on the similarity between COVID-19 and in uenza have been published [4][5][6][7], but the disease presentation continues to vary among individuals. Symptoms that are different from those seen in past viral infections have also been reported.
For example, there are reports of patients presenting with major loss of smell and taste [8,9]. The wide range of features reported is thought to re ect effects on non-respiratory systems and indicate that signs may be observed in patients who are infected, but lack apparent respiratory symptoms [10].

Study design and study population
We recruited patients diagnosed with COVID-19 from January 1, 2020 to December 31, 2020, and all patients underwent blood tests. For comparison, we recruited patients diagnosed with in uenza; the number of patients with in uenza and those with COVID-19 was the same. All patients with in uenza also underwent blood tests and the period was dated back from December 31, 2020 until the period that was the same number of COVID-19. Furthermore, children (aged < 14 years) with COVID-19 and in uenza were excluded. COVID-19 was diagnosed using a polymerase chain reaction text for the detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) nucleic acid, and in uenza was diagnosed using a rapid in uenza diagnostic kit. Furthermore, more than two doctors were involved in each diagnosis. This retrospective, cross-sectional study was performed at Juntendo University Nerima Hospital, a 490-bed university-a liated hospital in Tokyo, Japan.
Clinical information was extracted by chart review. Data on the following variables were collected: age, sex, history of malignant diseases, asthma, heart disease including hypertension, diabetes mellitus, hemodyscrasia, human immunode ciency virus infection, use of immunosuppressive agents (including steroids), and general symptoms (chills, sore throat, cough, headache, diarrhea, nausea, joint pain, malaise, and di culty in breathing). We also extracted data on axillary body temperature, systolic and diastolic blood pressure, pulse rate, respiratory rate, oxygen saturation (room air), white blood cell (WBC) count along with the percentage of neutrophils and lymphocytes, hemoglobin level, platelet count, RDW, serum parameters (levels of total protein, albumin, lactate dehydrogenase, blood urea nitrogen, creatinine, sodium, potassium, chloride, glucose, aspartate aminotransferase, alanine aminotransferase, total bilirubin, glucose, hemoglobin A1c, and C-reactive protein [CRP]).

Statistical analysis
Bivariate comparisons of each variable between COVID-19 and in uenza were performed using independent samples t-tests for continuous data or chi-square tests for categorical data. Differences with p values of < 0.05 were de ned as statistically signi cant. Variables with a p value of < 0.05 in bivariate analysis were entered into multivariable logistic regression models. The accuracy of the logistic regression models was assessed using the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. All statistical analyses were performed using EZR (Saitama Medical Center; Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). EZR is a modi ed version of R Commander designed to add statistical functions frequently used in biostatistics [19].
This study was conducted in accordance with relevant guidelines and regulations. This retrospective study was approved by the ethics committee of Juntendo University Nerima Hospital, Tokyo, Japan (approval number: 2020066).  Table 1 shows the characteristics of patients with COVID-19 and in uenza and the results of the bivariable analysis. Age, cancer, asthma, heart diseases, di culty in breathing, chills, nausea, malaise, body temperature, systolic blood pressure, heart rate, respiratory rate, saturation, WBC count, percentage of neutrophils, percentage of lymphocytes, hemoglobin level, RDW, platelet count, aspartate aminotransferase level, sodium level, chloride level, CRP level, creatine phosphokinase level, and creatine phosphokinase level were signi cantly different between the groups (Table 1). We conducted multivariable logistic regression analysis based on bivariable analysis ( Table 2). The following variables were considered for multivariable regression analysis: age, chills, di culty in breathing, nausea, malaise, body temperature, WBC count, percentage of lymphocytes, and creatine phosphokinase level. We incorporated these factors as they are thought to be clinically useful. For example, chloride, we excluded the factor which a signi cant difference was over in normal range even if signi cantly different. Based on the cutoff value, body temperature was categorized as hypothermia (< 35.9°C) and hyperthermia (> 38.1°C), and a body temperature of 36°C-38°C was used as the reference range. Furthermore, the cutoff WBC count was > 9000/µL, which was higher than the normal upper limit. For a similar reason, the cutoff value for the percentage of lymphocytes was > 20%, and that for the creatine phosphokinase level were > 300 U/L. Because the percentage of neutrophils and lymphocytes was highly correlated with each other, the percentage of lymphocytes was used in multivariable analysis as a representative of these two variables. The signi cant factors in multivariable analysis were age 15-70 years (vs. 71 years) (odds ratio [OR] = 5.53, 95% con dence interval [CI]: 2.86-10.70, p < 0.001), di culty in breathing (OR = 3.20, 95% CI: 1.37-7.47, p < 0.01), nausea (OR = 0.10, 95% CI: 0.01-0.97, p = 0.04), malaise (OR = 2.43, 95% CI:

Discussion
To the best of our knowledge, this is the rst study to directly compare patients with COVID-19 and in uenza using laboratory test data in a university hospital. Age 15-70 years (vs. 71 years), di culty in breathing, malaise, and percentage of lymphocyte > 20% were signi cantly more frequent in patients with COVID-19 than in those with in uenza. However, nausea, body temperature > 38.1°C, and WBC count > 9000/µL were more frequent in patients with in uenza than in those with COVID-19.
Other studies have reported that age at the onset of COVID-19 is lower than that at the onset of in uenza [14,20]. The results of these studies support the results of the present study.
SARS-CoV-2 mainly invades respiratory epithelial cells by adhesion to angiotensin-converting enzyme 2; thus, infected patients may develop mild-to-severe in ammatory responses and acute lung injury [21]. Another study has reported that COVID-19 causes acute respiratory failure in the predominance of in uenza [14]. These studies support the results of our study, revealing a higher number of cases of dyspnea cases than those of in uenza.
Similarly, malaise is a symptom that has been reported in many patients with COVID-19 [14]. According to a previous study, 63% patients with COVID-19 have malaise [22]. These studies support the results of our study, revealing a higher number of cases of malaise than those of in uenza cases. Malaise may be related to an increase in viral load and immune response to the infection [23]. In addition, insu cient energy production to meet the required metabolic demands is related to malaise [24].
Many studies have reported low lymphocyte counts in patients with COVID-19 [25,26]. The host immune responses strongly try to involve all potential cells and cytokines. In chronic COVID-19 cases, natural killer cells and T cells become exhausted, and a decrease in their count leads to lymphopenia. The inability to eradicate the infection in the affected organ causes hyper-initiation of the immune system, which releases excessive amounts of in ammatory cytokines to compensate for the exhausted ones and the low lymphocyte count [25]. Similarly, a low lymphocyte count has been reported in in uenza [27,28]. However, our study shows that the lymphocyte counts are lower in in patients with COVID-19 than in patients with in uenza. Furthermore, regarding WBC count, a study has shown that low leukocyte counts are markers of COVID-19 [29]. In our study, WBC counts > 9000/µL were more frequently associated with in uenza than with COVID-19. This nding suggests the likelihood of it being useful for the differentiation of COVID-19 from in uenza. In addition, our study showed that CRP levels are not useful for the differentiation of COVID-19 from in uenza. Similarly, a previous study showed that CRP levels were not an effective discriminator of COVID-19 and non-COVID-19 cases [29]. However, a signi cantly higher level of CRP was observed in the severe COVID-19 group than in the non-COVID-19 group, which con rms the ndings of previous studies regarding the clinical utility of CRP levels as an indicator of severe disease and progressive in ammation [30,31]. In our study, a signi cant difference in CRP levels was possibly not observed because none of the patients with COVID-19 had severe disease. In contrast to a previous meta-analysis [32], a recent meta-analysis revealed that procalcitonin levels were not signi cantly different between the severe and non-severe groups. The procalcitonin level was only assessed in a few cases in this study; therefore, we did not add this in the extraction item; however, we think that it is the contents that procalcitonin level may be argued comparison between COVID-19 and in uenza in the future.
Body temperature > 38.1°C was more frequent in patients with in uenza than in those with COVID-19. Many cases not involving fever have been reported in COVID-19 [33,34]. We believe that these ndings support our results. However, the result may be effects that patients receiving antipyretics could not be excluded. Furthermore, many patients with COVID-19 had to be hospitalized for isolation owing to the high infectious control, and they underwent laboratory tests even if they were asymptomatic.
Nausea is more frequently associated with in uenza than with COVID-19. In a study, nausea occurred in approximately 3.9% patients with COVID-19 [33]. Similarly, the ratio of nausea is unknown, but it can cause digestive symptoms including nausea in in uenza. The mechanism to cause nausea is unclear and requires further investigation. In ammatory reactions may occur when a virus infects gastrointestinal mucosal bleeding cells, and digestive symptoms such as nausea are thought to occur during these reactions. In this study, signi cantly fewer symptoms of nausea in patients with COVID-19 may re ect the greater strength of the respiratory tract infection than the gastrointestinal cells of COVID-19. Di culty in breathing was signi cantly more frequent in patients with COVID-19 than in those with in uenza, which may support our hypothesis.
There are some limitations to this study. First, there is a greater possibility that patients with COVID-19 are hospitalized as per law in Japan. Many patients with COVID-19 are hospitalized for isolation, which can help in controlling the spread of the highly infectious disease, even if present in young and asymptomatic patients. Furthermore, an in uenza extraction period of several years passed (from December 31, 2020 to January 11, 2015). Annually, the seasonal u may vary in symptoms, and this effect may have in uenced the results of our study. However, the type of in uenza did not in uence the results, as there were no patients with in uenza type B who underwent blood tests. In addition, we did not extract the data of three symptoms (taste disturbance, dysosmia, and conjunctival hyperemia) as they were inaccurate; therefore, these symptoms were not included in the chart review of patients with in uenza. Finally, the patient population enrolled in this study was limited to that from a single hospital. Additionally, this was a retrospective study. Hence, a multicenter prospective study should be conducted with a larger number of patients to verify our results.
In conclusion, our results are useful for differentiating patients with COVID-19 from those with in uenza. Age 15-70 years (vs. 71 years) years, di culty in breathing, malaise, and percentage of lymphocytes > 20% were signi cantly more frequent in COVID-19 than in in uenza. However, nausea, body temperature > 38.1°C, and WBC counts > 9000/µL were more frequent in in uenza than in COVID-19.

Declarations
Ethics approval and consent to participate The Ethics Committee of Juntendo University Nerima Hospital approved this study (Approval Number 2020066). This study was an observational study and written informed consent was waived due to a public health outbreak investigation by the Ethics Commission (Juntendo University Nerima Hospital).

Consent for publication
All authors agreed to publish this study.

Availability of data and materials
The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Competing interests
Nothing to disclose for all authors Funding This research received no speci c grant from any funding agency in the public, commercial or not-forpro t sectors.
Authors' contributions SF, AI, MS and TN contributed to study concept and design. SF and AI were involved in acquisition of participants and data. SF and DK were involved in analysis and interpretation of data. SF, AI, MS, DK and TN were involved in preparation of the manuscript.