EFFECTIVENESS OF BASELINE AND POST-PROCESSED CHEST X-RAY IN NONEARLY COVID-19 PATIENTS

  1. Michele Gaeta
  2. Giuseppe Cicero
  3. Maria Adele Marino
  4. Tommaso D’Angelo
  5. Enricomaria Mormina
  6. Silvio Mazziotti
  7. Alfredo Blandino
  8. Giulio Siracusano
  9. Aurelio La Corte
  10. Massimo Chiappini
  11. Giovanni Finocchio

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Background

CT is a very sensitive technique to detect pneumonia in COVID-19 patients. However, it is impaired by high costs, logistic issues and high risk of exposure.

Chest x-ray (CXR) is a low-cost, low-risk, not time consuming technique and is emerging as the recommended imaging modality to use in COVID-19 pandemic.

This technique, although less sensitive than CT-scan, can provide useful information about pulmonary involvement.

Purpose

To describe chest x-ray features of COVID-19 pneumonia and to evaluate the sensitivity of this technique in detecting pneumonia. A further scope is to assess the effectiveness of a post-processing algorithm in improving lung lesions detectability.

Materials and Methods

72 patients with laboratory-confirmed COVID-19 underwent bedside chest X-ray.

Two radiologists were asked to express their opinion about: (i) presence of pneumonia (negative or positive); (ii) localization (unilateral or bilateral); (iii) topography (according to pulmonary fields); (iv) density (non consolidative ground-glass or inhomogeneous opacities; consolidative nodular-type or triangular; mixed consolidative e non-consolidative); and (v) presence of pleural effusion. The point (i) was evaluated separately, while the other points in consensus.

A quality assessment of post-processed x-ray images was performed by two different readers.

Results

The agreement about presence of pneumonia was almost perfect with K value of 0.933 and p < 0.001.

Sensitivity was 69%.

The following findings were seen: unilateral lung involvement in 50%; lower lung lesions in 54%; peripheral distribution in 48%; and non-consolidative pattern in 44%.

Post-processed images improved the detection of lesions in 7 out 72 patients (≅10%)

Conclusion

CXR owns a good sensitivity in detecting COVID-19 lung involvement. Use of post-processing algorithm can improve detection of lesions. Our data support recommendations of the Radiological Society of North America (RSNA) to consider chest x-ray as first step imaging examination in Covid-19 patients.

SUMMARY

Bedside CXR has a good sensitivity in evaluating COVID-19 lung involvement in hospitalized patients and should be considered as the first step imaging technique according to RSNA recommendations.

KEY RESULTS

  • Bedside CXR has a good sensitivity in evaluating COVID-19 lung involvement in non-early clinical cases.

  • The most common findings of lung involvement were slight different from the well-described CT-ones, with less common patterns of bilateral and peripheral distribution.

  • Post-processing algorithm enhances detection of pulmonary lesions.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2020.04.16.20061044: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    Institutional Review Board Statementnot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.
    Sex as a biological variableWe retrieved and analyzed CXRs obtained at the bed of 72 patients admitted in our COVID-center (25 females, 47 males, age range 43-100 years, mean age 68.8 years).

    Table 2: Resources

    No key resources detected.

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    This work has some limitations. First, our cohort, due to admission criteria, does not include asymptomatic or mild patients where CT scan can detect ultra-early and early findings of pneumonia [16], and then we cannot state which is the sensitivity of CXR in these patients. However, a recent research letter reported that CXR film can detect pneumonia also in some patients with mild COVID-19 disease [17]. Further studies are necessary on this direction. Second, we cannot report the outcome of patients because at the time of writing many of them have not concluded the disease cycle. Third, we have not obtained correlation with the known standard of reference for COVID-19 imaging which is chest CT since only a small sub-sets of our patients underwent CT. No attempt to correlate CXR and CT findings of this small sub-set of patients was made but some anecdotal cases have been used in our representative figures. However the almost perfect agreement between our two independent readers support the reliability of our results also considering similar results published by Wong et al. [11]. Quality of chest x-ray films obtained at bed in supine position cannot be perfect in a significant number of cases, particularly in old patients suffering of dyspnea. This limitation drove the development and clinical test of a post processing tool to improve the contrast resolution of CXR. Quality evaluation of post-processed CXRs was made either on the base of rigorous metrics (entropy) either on s...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2020.04.16.20061044v1 on medRxiv