key: cord-0915679-4xqfrsr8
authors: Matsunaga, Nobuaki; Hayakawa, Kayoko; Terada, Mari; Ohtsu, Hiroshi; Asai, Yusuke; Tsuzuki, Shinya; Suzuki, Setsuko; Toyoda, Ako; Suzuki, Kumiko; Endo, Mio; Fujii, Naoki; Suzuki, Michiyo; Saito, Sho; Uemura, Yukari; Shibata, Taro; Kondo, Masashi; Izumi, Kazuo; Terada-Hirashima, Junko; Mikami, Ayako; Sugiura, Wataru; Ohmagari, Norio
title: Clinical epidemiology of hospitalized patients with COVID-19 in Japan: Report of the COVID-19 REGISTRY JAPAN
date: 2020-09-28
journal: Clin Infect Dis
DOI: 10.1093/cid/ciaa1470
sha: ae17cf5dc2aba5026ae7d3414ae22167906f3b2a
doc_id: 915679
cord_uid: 4xqfrsr8

BACKGROUND: There is limited understanding of the characteristics of coronavirus disease 2019 (COVID-19) patients requiring hospitalization in Japan. METHODS: This study included 2638 cases enrolled from 227 health care facilities that participated in the COVID-19 Registry Japan (COVIREGI-JP). The inclusion criteria for enrollment of a case in COVIREGI-JP are both (1) a positive SARS-CoV-2 test and (2) inpatient treatment at a health care facility. RESULTS: The median age of hospitalized patients with COVID-19 was 56 years (interquartile range [IQR]: 40-71). More than half of the cases were male (58.9%, 1542/2619). Nearly 60% of the cases had close contact to confirmed or suspected cases of COVID-19. The median duration of symptoms before admission was 7 days (IQR: 4-10). The most common comorbidities were hypertension (15%, 396/2638) and diabetes without complications (14.2%, 374/2638). The number of non-severe cases (68.2%, n=1798) was twice the number of severe cases (31.8%, n=840) at admission. The respiratory support during hospitalization includes those who received no oxygen support (61.6%, 1623/2636), followed by those who received supplemental oxygen (29.9%, 788/2636), and IMV/ECMO (mechanical ventilation or extracorporeal membrane oxygenation) (8.5%, 225/2636). Overall, 66.9% (1762/2634) of patients were discharged home, while 7.5% (197/2634) died. CONCLUSIONS: We identified the clinical epidemiological features of COVID-19 in hospitalized patients in Japan. When compared with existing inpatient studies in other countries, these results demonstrated less comorbidities and a trend towards lower mortality.

Coronavirus disease 2019 (COVID-19) caused by the SARS-CoV-2 virus, has become a global public health crisis. [1, 2] In Japan, the first case of COVID-19 was reported on January 16, 2020 , and COVID-19 was classified as a designated infectious disease on February 1. [3] "Three pillars" of COVID-19 countermeasures including the early response to clusters have been formulated. [4, 5] The number of cases increased in late March, triggered by the influx of infected patients from overseas, and a state of emergency was declared on April 7. In May, the spread of the disease began to be temporarily controlled, and the declaration of a state of emergency was lifted on May 25. With the reactivation of social activities, the number of infected people has again rose since late June of 2020. [6] In Japan, as of July 7, 2020, there have been 19816 cases and 979 deaths, with fewer cases and deaths than in Western countries. [6, 7] The number of cases may be affected by the number of tests conducted; 6.0 tests performed per thousand were approximately 5-10% of those performed in the United States (US) or European countries. [6, 7] A total of 7.73 confirmed COVID-19 deaths per million in the population is about 2% the death rate of the US, less than 1.4% that of Italy and Spain, and less than 1.25% that of United Kingdom (UK). [6] [7] [8] The poor prognosis of COVID-19 may be related to a complex combination of factors, including number of patients, infrastructure of medical facilities, resources of medical personnel, and patient background. To determine the reasons for these differences in mortalities and to improve the management of COVID-19, a complete picture of the COVID-19-affected population in Japan is needed.

A nationwide COVID-19 inpatient registry, "COVID-19 REGISTRY JAPAN (COVIREGI-JP)", was started in March 2, 2020. [9] Using data from COVIREGI-JP, we conducted this study to identify the clinical epidemiological characteristics of COVID-19 inpatients in Japan.

A c c e p t e d M a n u s c r i p t

This is an observational study. Health care facilities which voluntarily participated in COVIREGI-JP enrolled the patients. Research collaborators in each facility manually input the data into the registry.

The research collaborators received funding from the research grant for each patient enrolled. The inclusion criteria for enrollment are (1) a positive SARS-CoV-2 test [10] and (2) inpatient treatment at a health care facility. Some eligible inpatients might not have been registered in COVIREGI-JP by the principal investigator"s decision (e.g., patients participating in other clinical studies whose data registration in COVIREGI-JP was deemed inappropriate, patients who refused to participate in the study by opting out, etc.). If one patient had a history of multiple COVID-19 hospitalizations and met the aforementioned inclusion criteria, each admission was included in the registry.

We modified the case report form (CRF) of the ISARIC (International Severe Acute Respiratory and Emerging Infection Consortium) to enable the collection of clinical epidemiological information and treatment data in Japan. [11] The study data were collected and managed using REDCap (Research Electronic Data Capture), a secure, web-based data capture application hosted at the JCRAC Data Center of the National Center for Global Health and Medicine.

[12]

We used data from cases that had entered all of the following major items as of July 7, 2020: demographics and epidemiological characteristics; comorbidities; signs and symptoms at admission; outcome at discharge; supportive care, history of drug administration, and complications during hospitalization. We did not treat parameters with the option "unknown" as missing values.

A c c e p t e d M a n u s c r i p t

"Severe disease at admission" was defined as participants meeting one or more of the following criteria: requiring invasive or non-invasive mechanical ventilation, requiring supplemental oxygen, SpO 2 ≤ 94% on room air, or tachypnea (respiratory rate ≥ 24 breaths per minute). [13] 2) Respiratory support during hospitalization "Respiratory support" during hospitalization was defined as follows:

No oxygen: patients who were never supported with supplemental oxygen during hospitalization, Oxygen: patients who were supported with non-invasive mechanical ventilation or supplemental oxygen (including high flow oxygen devices) during hospitalization, and IMV/ECMO: patients who were supported with IMV (invasive mechanical ventilation) or ECMO (extracorporeal membrane oxygenation).

Information on medications with antiviral effect against SARS-CoV-2, immunomodulatory and/or immunosuppressive effects against COVID-19, antibiotics, antifungal drugs, neuraminidase inhibitors, and anticoagulants were collected. These medications were included in the analysis if they were administered at least once during the hospitalization period. If more than one drug was used in one patient, each drug was counted separately regardless of whether they were administered concurrently. If one drug was administered during different periods of hospitalization, it was counted as one.

Continuous variables are described as the medians and interquartile range (IQR). As the number of missing values differed for each parameter, the number of cases in each parameter's severity/respiratory support category was used as the denominator for calculating the percentage. All statistical analyses are conducted using R, version 3.5.1 (R core Team).

A c c e p t e d M a n u s c r i p t

This study was approved by the NCGM ethics review (NCGM-G-003494-0). Informed consent was obtained in the form of opt-out on the registry website.

Data from 2638 cases from 227 facilities were included for the analysis. The available numbers of cases were different depending on each parameter due to the missing data. Non-severe cases at admission (68.2%, n=1798) were more than twice as prevalent as severe cases (31.8%, n=840). A c c e p t e d M a n u s c r i p t

Compared to non-severe cases, severe cases tended to have higher temperature, heart rate, and respiratory rate on admission, and were less alert. Pneumonia was much more prevalent by X-ray in severe cases than non-severe cases (86.9% (623/717) vs 52.1% (679/1303), respectively).

Overall, patients with severe disease on admission and those who received supplemental oxygen or IMV/ECMO during hospitalization tended to have more comorbid conditions than patients who were not severe /received no oxygen. Particularly, patients with underlying cardiovascular diseases, diabetes mellitus (DM), chronic obstructive pulmonary disease (COPD)/chronic lung diseases, and obesity were more frequently categorized as severe cases than those without these comorbidities. Additionally, 3.7% (95/2593) cases had immunosuppressive status.

Of the 2113 cases, 6.9% (145) reported no symptom, which accounted for 9.3% (132) of non-severe cases and 1.9% (13) of severe cases. About half of the patients had fever, cough, and fatigue.

Compared to non-severe cases, severe cases tended to present with fever, fatigue, cough, and shortness of breath. Conversely, dysgeusia and olfactory dysfunction were more common in nonsevere cases than severe cases. 

The most common medications used in this cohort were favipiravir, and ciclesonide (only available as an inhalant). Favipiravir, lopinavir/ritonavir, and hydroxychloroquine were more frequently used as the severity of the disease increased, while ciclesonide was used in half of the patients regardless of the severity of the disease. Steroids (excluding ciclesonid), nafamostat, and tocilizumab tended to be used more commonly in patients who received IMV/ECMO; this trend was particularly pronounced with steroid treatment (no oxygen: 1.7% [27/1605]; oxygen: 13.5% 

Acute respiratory distress syndrome (ARDS) was the most common complication in patients who received IMV/ECMO (70.2%, 158/225). Bacterial pneumonia was also prevalent (6.9%, 181/2633), Our results on demographics demonstrated that severe cases tended to be males, patients with past smoking history, and older patients, which are consistent with existing reports of risk for severe disease and death in COVID-19 patients.[14-16] Although the median age of our cases (56 [40-71]) was comparable to or slightly lower than existing inpatient data, [15, 17, 18] 12.9% of the cohort was in their 80s or older, reflecting the long life expectancy in Japan. [19] In patients in their 60s or older, the proportions of severely ill patients increased; however, the proportion of patients who received IMV/ECMO was lower among patients 80 or older, than patients in their 60s and 70s. This might partly be due to the less aggressive treatment approaches for this population.

As more than half of the inpatients had a history of close contact with confirmed or suspected COVID-19 cases, the strategy to target clusters of patients with history of close contacts was considered beneficial in terms of identifying the risk of infection. Although this does not necessarily imply healthcare-associated transmission, presence in a healthcare facility where COVID-19

infections have been managed was found in 11.5% of the cases; therefore, healthcare-associated facilities also seem to be an important target for controlling the spread of COVID-19. The proportion of healthcare workers in our cohort is higher than in other reports (3.8%-5.1%).

[20] This may be partly due to different patient populations (e.g. cases other than inpatients) included.

Comorbidities were also studied in this patient population. Cardiovascular disease, DM, chronic lung disease/COPD, and obesity were more prevalent in severe cases, consistent with previous reports. M a n u s c r i p t

In this study, fever, cough, dyspnea, and fatigue were more likely to be expressed in severely ill patients, whereas olfactory and dysgeusia were more likely to be expressed in those with nonsevere disease. The incidence of fever, cough, dyspnea, and fatigue was lower in our study than in the existing inpatient cohort, but this could be influenced by the phase of the epidemic and admission In this cohort, which included the early phase of cases of COVID-19 in Japan, favipiravir and ciclesonide were the predominant antiviral drugs administered. Patients who received IMV/ECMO tended to be treated more frequently with steroids and nafamostat. In Japan, as of July 29, 2020, some steroids (prednisolone and dexamethasone) were indicated for severe infections, but only remdesivir is approved for the treatment of COVID-19 (approved on 7 May 2020). All other drugs (listed in Figure 2 ) were used off-label. Regardless of whether or not the drug was used, COVID-19 inpatients could be enrolled voluntarily in this registry. However, favipiravir users were more likely to be included in the registry because healthcare providers are encouraged to participate when they contact the Ministry of Health, Labour, and Welfare about the provision of favipiravir. The preliminary report of the Favipiravir Observational Study in Japan included more elderly and more comorbid patients than that observed for this registry,[25] and reported higher in-hospital mortality (11.6%). As favipiravir users were more likely to be included in this registry, data of patients in this cohort may have been slightly more severely skewed than that of the general Japanese population hospitalized with COVID-19. As noted in the Methods, patients enrolled in other clinical studies may not have been included in this registry. It is also important to note that we did not count concomitant therapies in the current analysis. Evaluation of the efficacy of antiviral and immunomodulatory agents is outside the scope of this study and requires further studies. Anticoagulants were used in less than half of the patients who received IMV/ECMO; however, their use is expected to increase as the evidence for anticoagulation therapy is being recognized.

A c c e p t e d M a n u s c r i p t In our inpatient cohort, the mortality rate was 7.5% of cases, and 33.5% of the patients who received IMV/ECMO. The mortality in this study was lower than those of previously reported inpatient cohorts (approximately 15% to over 20%). [2, 14, 15, 17, 18 , 26] Additionally, mortality was lower even among intubated patients.[14, 15, 18] As some of these include patients with an undetermined outcome during hospitalization, the quantification of mortality may be even higher. The lower rate of mechanical ventilation than previous studies [14, 15, 17, 18] and lower number of severe cases at admission than a study using the same criteria for severity [13] suggest lower overall severity of hospitalized patients in Japan. Some of the studies with significantly higher mortality, especially among mechanically ventilated patients,[15, 18] may reflect the impact on health care infrastructure (e.g., unavailability to intubate timely) due to the rapid increase in patients in the region. The reasons for the relatively lower mortality in Japan might be due to factors such as lower number of comorbid patients, or lack of a drastic increase in patient numbers during the period of patient enrollment.

Approximately 70% of the patients in our cohort were non-severe on admission, and they would have been managed as outpatients in other countries. This may have also contributed to the low mortality observed in our cohort compared to that in other countries. Further investigation into the factors associated with severity of the disease would be necessary to expand on this.

Though this registry does not include the information on the exact cause of death, cases of patient death despite no oxygen use during hospitalization had underlying diseases; 5 cases had malignancy (3/5 were metastatic tumors), and two cases were in their 80s and 90s with dementia.

Some caveats regarding the interpretation of the results of this study are warranted. Although this study covered a large number of COVID-19 inpatients in Japan, there may have been some selection bias for the inclusion, as noted above, and also due to the manual input of the data. Thus, the study results may not completely reflect those of the general Japanese population hospitalized with COVID-19. DVT and PE were quite low compared to existing reports, and were likely to be underreported as they were based on diagnoses of each institution.

[27] Future validation using objective indicators such as d-dimer is therefore necessary. Although the data center provides a data input manual, an input check function, and inquiries to institutions, the accuracy of the data may be lower than that of clinical trials. In addition, due to the nature of the registry, data are updated daily, A c c e p t e d M a n u s c r i p t and it is possible that the findings using this registry may diverge from the presented results in the future. National policies for COVID-19 are in flux, and changes to regulations such as discharge criteria (i.e., from mandatory SARS-CoV-2 negativity to criteria based on the days from the disease onset) were made during the collection of this registry data, which may have affected the duration of hospitalization. [28, 29] We reveal the characteristics of diverse clinical epidemiological features of COVID-19 in hospitalized patients in Japan. Compared with existing inpatient studies in other countries, these data show that the patient population with COVID-19 in Japan had less comorbidities, and there was a trend toward lower mortality.

A c c e p t e d M a n u s c r i p t

We thank all the participating facilities for their care of COVID-19 patients, and cooperation on data entry to the registry. A c c e p t e d M a n u s c r i p t M a n u s c r i p t

International Severe Acute Respiratory and Emerging Infection Consortium (ISARIC). A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t 

Clinical progression of patients with COVID-19 in Shanghai

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Labour and Welfare. Current status of the novel coronavirus infection and the response of the MHLW

Expert Meeting on the Novel Coronavirus Disease Control. Views on the Novel Coronavirus Disease Control

Japanese strategy to COVID-19: How does it work?

Ministry of Health, Labour and Welfare. Current status of the novel coronavirus infection and the response of the MHLW

Ministry of Health, Labour and Welfare. Notification of physicians and veterinarians under the Infectious Diseases Act

A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t