key: cord-0947211-3yr9iuya
authors: Niv, Yaron; Eliakim-Raz, Noa; Bar-Lavi, Yaron; Green, Manfred; Dreiher, Jacob; Hupert, Amit; Freedman, Laurence; Weiss, Yoram; Zetland, Riki; Luz, Shirli; Menachemi, Doron; Kuniavski, Michael; Rahav, Gaila; Sagi, Ram; Goldschmidt, Nethanel; Mahalla, Hanna
title: Comparing Covid-19 pandemic waves in hospitalized patients – a retrospective, multicenter, cohort study
date: 2022-02-10
journal: Clin Infect Dis
DOI: 10.1093/cid/ciac119
sha: e788ea8c577a907f3b375a6cc4e6e20cc2941333
doc_id: 947211
cord_uid: 3yr9iuya

BACKGROUND: Covid-19 disease was first diagnosed in Israel at the end of February 2020. Until the end of June 2021 842,536 confirmed cases and 6428 deaths were accumulated. The aim of our multicenter retrospective cohort study is to describe the demographic and clinical characteristics of hospitalized patients and to compare the pandemic waves before immunization. METHODS: Out of 22302 patients hospitalized in general medical centers we randomly selected 6329 admissions for the study. Of these, 3582 and 1106 were eligible for the study in the first period (1 (st) & 2 (nd) waves), and in the second period (3 (rd) wave), respectively. RESULTS: Thirty-day mortality was higher in the 2nd period than in the 1st period, 25.20% versus 13.68% (P<0.001). Invasive mechanical ventilation supported 9.19% and 14.21% of the patients in the 1st period and 2nd period, respectively. Extracorporeal Membrane Oxygenation (ECMO) was used more than twice as often on the 2nd period . CONCLUSIONS: Invasive ventilation, use of ECMO and mortality rate were 1.5 to 2 times higher on the 2 (nd) period than in the 1 (st) period. Patients of the 2 (nd) period had a more severe presentation and higher mortality than those of the 1 (st) period.

Severe acute respiratory syndrome coronavirus-2 (SARS-SoV-2) that causes coronavirus disease 2019 (Covid- 19) was first diagnosed in Israel at the end of February 2020. Since then and up to June 2021 there were 842536 confirmed Covid-19 cases and 6428 deaths 1 . Three waves of the pandemic were clearly defined in Israel; each subsided following social distancing, use of masks, isolation of corona cases, and finally strict lock down 2 .

The Israel Ministry of Health has approved both mRNA-based vaccines (Moderna/NIH and Pfizer/BioNTech), and national immunization program has started vigorously on December 19 , 2020 (with the Pfizer/BioNTech vaccines, mRNA-BNT162b2, which require 2 doses). The national immunization program prioritized elderly adults and other populations with higher risk for severe COVID-19 followed by the general population. On July 3, 2021 5174406 Israelis, (55.64% of the total Israeli population) were immunized with 2 injections of BNT162b2 RNA vaccine of Pfizer. High efficiency of prevention of disease, symptoms, hospitalization and severe illness was reported in Israel, with 92%, 94%, 87% and 92% rates, respectively 3 .

Reports from China, Italy, Great Britain and the USA at the beginning of the first wave suggested high morbidity, mortality, stressed hospitals, and intensive care units utilization 4 .

When 500 new cases were diagnosed in Israel, a tight lock down of 32 days was issued and the 1 st wave resulted in 4000 hospitalizations and 329 deaths 5 . The second and third waves were severe and ended with 40000 hospitalizations and 6099 more deaths 6 .

The epidemiological and clinical presentations were definitely changed with the successful immunization project. The aim of our multicenter report is to describe the demographic and clinical characteristics, underlying diseases, laboratory findings and outcomes among hospitalized non-immunized patients with Covid-19 in Israel, using in-A c c e p t e d M a n u s c r i p t 4 depth record of about 25% of hospitalized patients, creating a valid and representative retrospective nation-wide cohort, and to compare the 2 periods of the pandemic. 

Patients' data were censored and anonymized at the time of collection. Acute respiratory distress syndrome (ARDS) was defined when oxygen saturation was less than 93%, more than 30 breaths per minute and bilateral pulmonary infiltrates on chest imaging. Age distribution was plotted against duration of hospitalization, frequency of invasive ventilation and mortality. The duration of time between RT-PCR test for SARS-CoV-2 virus and the day of hospitalization was recorded for every patient in both periods. The case fatality rate per hospital per period was defined as the percentage of patients admitted for Covid-19 who died in that hospital during that period. The range of case-fatality rates (CFRs) was 6.7% -19.9% with a median of 13.0%. A low level of CFRs was considered to be 6.7% -13.0%, and a high level 13.4% -19.9%. Data of the 1 st period (1 st and 2 nd waves combined) was compared with data of the 2 nd period (3 rd wave). We performed multivariate analysis to correlate background diseases with invasive ventilation and mortality. Logistic regression models were used for invasive ventilation and mortality, controlling for underlying diagnoses and demographics, with comparison between the two periods. We also compared hospitals, which had low mortality rate, with those with high mortality rate in both periods, looking for possible explanation for the differences in background diseases and their number per patient, symptoms at admission, pathological results of laboratory tests and treatments.

The clinical status was severe in 19.35% of the patients in the 1 st period and 35.91% in the 2 nd period. Patients were older on the 2 nd period than in the 1 st period, 71.01% and 58.23% were older than 60 years, respectively (Table 1) . Patients admitted during the 2 nd period were more likely to be male and had a higher prevalence of all underlying diseases except for active smoking ( Table 1 ). Fever of 38 0 C or higher and dyspnea on admission were found in 28.84% and 46.01%, 32.47% and 54.69% in the 1 st period and the 2 nd period, respectively.

In the 1 st period 53% underwent RT-PCR testing for SARS-CoV-2 virus within 24h of admission, and 84% within 7 days before admission. The respective data for the 2 nd period A c c e p t e d M a n u s c r i p t 7 were 52% and 83%. Thus, duration of disease before admission could not explain the difference.

Hospital length of stay directly related with severity of illness and need for mechanical ventilation, and was similar in all waves (Table 1) 

Laboratory tests results on patients' admission were more likely to be pathological during the 2 nd period than during the 1 st period, except for the rate of hypokalemia (Table 1) . Hypoxia, anemia, hypoalbuminemia and elevated D-dimer levels were presented in 10% more patients in the 2 nd period than in the 1 st period.

RT-PCR testing was performed on every admitted patient, and gene sequencing was done whenever needed according to epidemiological information. On the 1 st period all the patients had the original virus imported from China and Europe. This variant changed to the British variant that expanded to 80% of patients on the 2 nd period.

Patients were treated with steroids, remdesivir, convalescent plasma, vitamin D and anticoagulants in both periods (Table 1) . Therapeutic use of anticoagulants, steroids and vitamin D was higher in the 2 nd period than the 1 st period, 79.79%, 75.69%, 20.96%, versus 66.19%, 48.99%, 13.87% of the cases, respectively.

Half of the patients on high flow oxygen or invasive ventilation received remdesivir.

Remdesivir was associated with reduced mortality in these patients with an estimated OR of 0.53 for mortality (P=0.0013). The inverse association between remdesivir treatment and A c c e p t e d M a n u s c r i p t 8 mortality was seen in both the 1 st and 2 nd periods, which comprise patient populations that are quite different in several aspects, including the latter period having more elderly and more severely ill patients. However, within the group of those receiving high flow oxygen and invasive ventilation, the pattern of remdesivir prescription was opposite to that expected.

Those with better prognosis were more likely to receive remdesivir: in the 1 st period the percentage of patients receiving remdesivir among those older than 70 years and those with heart disease, lung disease, kidney disease and active malignancy versus patients younger than 70 years and those without these diseases, were 39%, 38%, 41%, 25%, and 39%, versus 61%, 57%, 51%, 55%, and 50%, respectively. Similar results were found in the 2 nd period -16%, 13%, 11%, 8%, and 11% versus 30%, 28%, 26%, 27% and 24%, respectively.

Similarly, in the 1 st period, the percentage of patients on high flow oxygen or invasive ventilation receiving remdesivir among those with low blood pressure, anemia, leukocytosis and high creatinine level was 26%, 41%, 39% and 32% versus 50%, 56%, 53% and 61% among those without these conditions, respectively; and in the 2 nd period, these percentages were 12%, 11%, 18%, and 10% versus 23%, 34%, 24%, and 29%, respectively.

Half of the patients in the 1 st period needed oxygen therapy in comparison with 2/3 of the patients in the 2 nd period . High flow oxygen by nasal cannula was given to 19.16% and 31.90% of patients in the 1 st period and 2 nd period, respectively. Invasive mechanical ventilation supported 9.19% and 14.21% of the patients in the 1 st period and 2 nd period, respectively. Nitric oxide inhalation and prone positioning were similarly practiced in both periods. ECMO was used more than twice as often on the 2 nd period ( (Table 2) . A positive correlation was found between the need for invasive ventilation and the number of background diseases per patients in both periods (Figure 1, Figure 2 ).

In hospital mortality and 30-day mortality were significantly higher on the 2 nd period than on the 1 st period ( Table 2) . A positive correlation was found between 30-day mortality and the number of background diseases per patient in both periods ( Figure 1, Figure 2 , Table   2, and Table 3) .

Case-fatality rates were higher in some hospitals more than in others, where patients were older, had worse laboratory tests results, and with more background diseases (Table 3 ). This was found in both periods.

We describe a cohort of 4688 Covid-19 Israeli patients, hospitalized between February 28 th 2020 and January 15 th 2021, comparing 2 periods of the Covid-19 pandemic. Invasive ventilation, use of ECMO and mortality rate were 1.5 to 2 times higher on the 2 nd period than A c c e p t e d M a n u s c r i p t 01 in the 1 st period. Since no breakthrough in patients' management happened, severe outcome was probably correlated with male gender, a worse clinical presentation (fever and dyspnea), more background diseases, poorer results of laboratory tests on admission, and higher prevalence of the British variant, which was present in 90% of the patients on the 2 nd period.

This observation may also explain the higher use of certain drugs on the 2 nd period, such as anticoagulants, steroids and vitamin D. The duration of disease before admission was similar in both periods and could not explain the difference in patient clinical status and outcome.

The range of mortality rates in different countries changed significantly between cohorts and correlated with the patients' age, gender, clinical status on admission and with the capabilities of the local health system . In our cohort, we had more men than women, and 38.45% - Our study has limitations, being retrospective and descriptive, and not including all the admitted patients. However, our cohort was composed of hospitalized patients in referral, tertiary, and academic centers as well as smaller rural, district hospitals, and we expect it was representative of the national experience. It seems that, within the group of more seriously ill patients there was a tendency to prescribe remdesivir to those patients with a better prognosis.

Thus, if we did a straight analysis of remdesivir versus no remdesivir with no statistical adjustment, we would not be surprised to see that patients receiving remdesivir have lower mortality. However, even after statistical adjustment for prognostic factors, we found a survival advantage to those receiving remdesivir. It would therefore be tempting to conclude that remdesivir does reduced the risk of death in patients receiving high flow oxygen or invasive ventilation. However, it is well known that statistical adjustment for prognostic factors usually does not completely remove the biases caused by selection, and the observed benefit to the remdesivir patients could easily be due to residual confounding, i.e. incomplete adjustment for the prescribing pattern.

In conclusion, we describe the three waves of Covid-19 pandemic in Israel. Patients' clinical status on admission, fever, dyspnea, background diseases and pathological laboratory results may be predictors for invasive mechanical ventilation and for mortality. The patients of the 3 rd wave (second period) had a more severe presentation and higher mortality than the 1 st & 2 nd waves (first period). M a n u s c r i p t 07 A c c e p t e d M a n u s c r i p t 

Guidelines, policies and information for staff

BNT162b2 mRNA Covid-19 vaccine in a nationwide mass vaccination setting

Covid-19 coronavirus pandemic

Ministry of Health, First wave information

Second and third wave information

Covid-19 in critically ill patients in the Seattle region -case series

Deep phenotyping of 34128 adult patients hospitalized with Covid-19 in an international network study

Medical treatment expert group for COVID-19

Development and validation of a clinical risk score to predict the occurrence of critical illness in hospitalized patients with COVID-19

Risk factors associated with mortality among patients with COVID-19 in intensive care units in Lombardy

Risk factors associated with in-hospital mortality in a US national sample of patients with COVID-19

factors associated with mortality among hospitalized patients with COVID-19: A retrospective cohort study

Distinct clinical characteristics and risk factors for mortality in female in patients with Coronavirus disease 2019 (COVID-19): A sexstratified, large-scale cohort study in Wuhan, China

Survival in adult inpatients with COVID-19

Clinical mortality in a large COVID-19 cohort: observational study

Cohort profile: SARS-CoV-2/COVID-19 hospitalized patients in Switzerland

3219 hospitalized patients with COVID-19 in Southeast Michigan: a retrospective case cohort study

Hospitalised COVID-19 patients of the Mount Sinai Health System: a retrospective observational study using the electronic medical records

Baseline demographics and clinical characteristics among 3471 US patients hospitalized with COVID-19 and Pulmonary involvement: A retrospective study

Presenting characteristics, comorbidities, and otcomes among 5700 patients hospitalized with COVID-19 in the New York City area

et al Patient characteristics and outcomes of 11,721 patients with COVID19 hospitalized across the United States

Survival of hospitalized COVID-19 patients in Northern Italy: A population-based cohort study by the ITA

Clinical characteristics of patients hospitalized with COVID-19 in Spain: Results from the SEMI-COVID-19 Registry

Outcomes and Mortality Among Adults Hospitalized With COVID-19 at US Medical Centers

Outcomes among 6721 Hospitalized COVID-19 Patients across the New York City Public Hospital System: a Retrospective Cohort Study

Characteristics and outcomes of COVID-19 patients in New York City's public hospital system

A c c e p t e d M a n u s c r i p t 09 Table 3 . Comparison between medical centers with low fatality rate of 6.7% -13.0%, and high fatality rate of 13.4% -19.9% (%).