Determinants of COVID-19 Pneumonia (MC-19)

Overview

Molecular testing (e.g PCR) of respiratory tract samples is the recommended method for the identification and laboratory confirmation of COVID-19 cases. Recent evidence reported that the diagnostic accuracy of many of the available RT-PCR tests for detecting SARS-CoV2 may be lower than optimal. Of course, the economical and clinical implications of diagnostic errors are of foremost significance and in case of infectious outbreaks, namely pandemics, the repercussions are amplified. False positives and false-negative results may jeopardize the health of a single patient and may affect the efficacy of containment of the outbreak and of public health policies. In particular, false-negative results contribute to the ongoing of the infection causing further spread of the virus within the community, masking also other potentially infected people.

As demonstrated by a study conducted by Ai et al., including 1014 suspect COVID-19 cases who underwent multiple RT-PCR testing and chest-CT, overall 88% of patients had positive CT scans while RT-PCR positivity was found only in 59% of all cases. Also, as reported by Yang et al, the total positive rate of RT-PCR for throat swab samples was reported to be about 30% to 60% at initial presentation. Thus, a negative result does not exclude the possibility of infection and should not be used as the only criterion for treatment of patient and management decisions. Reasons for false negative RT-PCR may include the lack of identification or inadequate procedures for specimen collection, handling and storage, as well as active viral recombination or testing carried out of the diagnostic window. From preliminary studies has emerged that patients may show very early but significant CT changes even before RT-PCR studies. Hence, the necessity for developing a combined approach for the diagnosis of these particular patients who present with negative RT-PCR test results. The investigators hypotheses is that several patients who presented with pneumonia confirmed at CT scan during the Coronavirus outbreak, and who tested negative for SARS-CoV2 at RT-PCR could probably be affected by the disease and need to be carefully observed. Primary end-point The primary end-point of our prospective, observational study is to assess if inpatients who presented with pneumonia but had a negative test for Covid-19 are positive at the serology for SARS-CoV-2. Secondary end-points Among the other secondary end-points, the investigators aim is to find if the combination of CT scan and serology could help in the identification of those patients who were initially negative at laboratory testing alone. Other secondary end-points are the efficacy of different pharmaceutical treatments against Covid-19 that were empirically started in those highly suspicious cases and the development of an approach useful for those patients who initially tested negative for Covid-19 infection. Methods Before starting the study, the protocol will be submitted to and approved by the local Ethical Committees at the Fondazione Policlinico Universitario A. Gemelli IRCCS, Catholic University, Rome, Italy. Before enrollment each subject will sign the informed consent. Inclusion criteria: hospitalized subjects of both sexes aged 18 years or older with diagnosis of pneumonia, confirmed by chest imaging and oxygen saturation (SaO2) ≤ 94% in ambient air, Covid-19 test negative, given informed consent to data collection from the patient or from the patient's legal representative if the patient is too unwell to provide consent. Exclusion criteria: age lower than 18 years, pregnancy or breast-feeding. Nasopharyngeal swab samples will be taken for quantitative real-time polymerase chain reaction to make diagnosis of Covid19 (2 repeated tests). Data collected include time of symptoms (cough, fever, dyspnea, conjunctivitis, diarrhea, asthenia, arthralgia) age, sex, height, weight, education, alcohol and smoking habits, morbidities, plasma glucose, creatinine, transaminases, γ-GT, total cholesterol, HDL-cholesterol, triglycerides, complete blood count, D-dimer, lactic acid dehydrogenase (LDH), high-sensitivity C-reactive protein (hs-CRP), creatinkinase (CK), ferritin, HbA1c, chest X rays, chest CT scan, cultures, therapy for pneumonia, other treatments including anti-hypertensive and anti-hyperglycemic agents, body temperature, blood pressure, and oxygen flow rate or other types of oxygen treatment. Five ml of plasma divided in aliquots of 1 ml each will be also obtained and stored at -80°C in anonymized way for future analysis, including third parties. Sample size If there is truly no difference between the standard and experimental treatment (16% in both groups), then 260 patients are required to be 90% sure that the limits of a two-sided 90% confidence interval will exclude a difference between the standard and experimental group of more than 15% Significance (α) = 0.05 Power (1-β)= 90% Percentage deaths in both control and experimental group = 16% Equivalence limit = 15% Calculation based on the formula: n = 2 × f(α, β/2) × π × (100 - π) / d2 where π is the true percent 'success' in both the control and experimental treatment groups, and f(α, β) = \[Φ-1(α) + Φ-1(β)\]2 Φ-1 is the cumulative distribution function of a standardised normal deviate. Statistics The association between recovery and patient groups will be tested by means of a Fisher exact test. A Cox Proportional-Hazard regression will be used to compare survival curves (times to improvement) among the studied groups by correcting for the administered therapy and for all the quantitative collected variables. Quantitative variables, measured at hospital admission, will be compared among groups using ANOVA. In univariable analyses, categorical variables, as gender, education, alcohol consumption and smoke habits will be analysed by means of a Chi-Squared test to study their association with the recovery, while a logistic regression model will be used to test possible quantitative predictors of recovery. A multivariable logistic model, with a stepwise selection procedure, will be then used to test all the variables that are significant in a univariable analysis.