key: cord-0962140-5b336kv4
authors: Covino, Marcello; De Matteis, Giuseppe; Burzo, Maria Livia; Franceschi, Francesco; Sandroni, Claudio
title: Reply to: Comment on Predicting In‐Hospital Mortality in COVID‐19 Older Patients with Specifically Developed Scores
date: 2021-01-22
journal: J Am Geriatr Soc
DOI: 10.1111/jgs.17031
sha: 499a0712904af3cb95a1603efc8549104daa47f4
doc_id: 962140
cord_uid: 5b336kv4

This letter comments on the letter by Alain Putot

To the Editor: We thank Odille et al 1 for the interest in our research and for bringing additional relevant updates to improve our work. In our manuscript, as the researchers outlined, the main finding is that the prognosis of COVID-19 older patients is mainly driven by respiratory status at admission. 2 Consequently, the severity scores including a respiratory function evaluation, such as those adopted for sepsis and pneumonia, 3-5 the general early warning scores (EWS), 6 and the specifically developed EWS, are able to stratify prognostic risk of COVID-19 patients. Likewise, as suggested by Odille et al, 1 the World Health Organization (WHO) pragmatic severity scale, which is based solely on respiratory evaluation (S1: no pneumonia; S2: pneumonia, with SpO2 ≥ 90% on room air; S3: severe pneumonia, with respiratory rate > 30 breaths/min or SpO2 < 90% on room air; S4: critical disease with acute respiratory distress syndrome), 7 could be useful for a quick patient stratification, also in emergency department setting. Nevertheless, the authors underlined that a WHO pragmatic severity scale validation in COVID-19 older patients is lacking.

Therefore, we expanded our work analyzing the performance of this index in our cohort of 210 patients aged 65 years or more. Globally, among deceased patientsaccording to WHO severity scale-1/23 (4.3%) was in S1 group, 18/142 (12.7%) were in S2 group, 13/27 (48.1%) were in S3 group, and 10/18 (55.6%) were in S4 group. Overall, the scale had an area under receiver operating characteristic curve (AUROC) of 0.705 (0.636-0.767), with WHO score greater or equal to 2 having 47.2% (30.4-64.5) sensitivity and 89.0% (83. 2-93.4 ) specificity for prediction of in-hospital death (Figure 1 ). Consequently, the WHO index showed a fairly good predictive value with respect to in-hospital death in our population, similarly to NEWS (difference between ROC areas 0.025 SE 0.038, P = .508), qCSI score (difference between ROC areas 0.016 SE 0.037, P = .659), ISARIC-4c (difference between ROC areas 0.086 SE 0.050, P = .088), and COVID-GRAM (difference between ROC areas 0.081 SE 0.050, P = .106).

To date, mainly due to age-related vulnerability and to the several complications, 8 the older patients are facing a more severe COVID-19. Therefore, the choice of an effective and quick-to-use tool could be crucial in identifying patients who could experience worse outcome. Accordingly, since all the evaluated scores had a similar performance for death risk stratification in older COVID-19 patients, we agree with Odille et al 1 that adopting the simplest index would be the best option in this population.

Comment on: predicting in-hospital mortality in COVID-19 older patients with specifically developed scores

Reply to: Comment on: predicting in-hospital mortality in COVID-19 older patients with specifically developed scores

Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study

The third international consensus definitions for sepsis and septic shock (Sepsis-3)

A prediction rule to identify low-risk patients with community-acquired pneumonia

Predicting intensive care unit admission and death for COVID-19 patients in the emergency department using early warning scores

Clinical management of COVID-19

Clinical characteristics and prognostic factors in COVID-19 patients aged ≥80 years