Archivos de Bronconeumología 58 (2022) 520–522

www.archbronconeumol.org

Scientific Letter

Outcomes of Critically Ill Very Old Patients With admitted to the ICU; 448 (52%) were not ventilated, 123 (14%)
Community-Acquired Pneumonia and Acute received NIMV, and 297 (34%) received IMV; and 1976 (30%) were
Respiratory Distress Syndrome VOP, 197 (10%) of them were admitted to the ICU; 95 (48%) were
not ventilated, 38 (19%) received NIMV, and 64 (32%) received IMV
To the Director, (Fig. 1, Panel A). Overtime, the proportion of VOP admitted to the
ICU changed significantly between 14% and 45% (p < 0.001) (Fig. 1,
During the last decades the overall and relative number of Panel B).
patients with community-acquired pneumonia (CAP) requiring One hundred and fifteen (27%) patients < 80 years old met the
intensive care management has increased globally, especially Berlin ARDS criteria, and 305 (73%) did not. Twenty-seven VOP
among elderly people.1 The percentage of intensive care unit (ICU) (26%) met the Berlin ARDS criteria, and 75 cases (74%) did not.
admissions attributable to elderly patients ranges between 9 and In patients younger than 80 years, those with ARDS presented
19% in European studies2,3 and 20–30% in American studies.4 higher rates of comorbidities (79% vs. 65%, p = 0.004) and higher
Severe CAP is the most common cause of acute respiratory dis- median PSI score (120 vs. 108, p = 0.012), than patients without
tress syndrome (ARDS), which occurs in approximately 7–10% ICU ARDS, while there were not significant differences regarding in-
patients with CAP,5 although a recent study found 13% of ARDS hospital (23% vs.27%, p = 0.443), 30-day (21% vs. 21%, p = 0.974) and
among ICU patients with CAP, reaching 29% in those requiring 1-year mortality (29% vs. 30%, p = 0.809) between groups. Interest-
mechanical ventilation.6 Interestingly, the number of quadrants ingly, when we compared ARDS patients <80 years old with ARDS
on chest imaging seems to be associated with an increased risk of VOP, we found that VOP with ARDS had a higher PSI score (108 vs.
death in patients with acute hypoxaemic respiratory failure requir- 140, p < 0.001), higher in-hospital (27% vs. 52%, p = 0.012), 30 days
ing mechanical ventilation.7 However, there is limited information (21% vs. 56%, p < 0.001) and 1-year mortality (30% vs. 68%, p < 0.001)
on ARDS in very old patients (VOP ≥80 years old) with CAP. We than patients < 80 years old with ARDS.
aimed to assess the prevalence, clinical characteristics, outcomes Our study population therefore comprised 102 VOP patients
and risk factors of ARDS in VOP with CAP. (Fig. 1, Panel A). Amongst the 27 VOP with ARDS, 9 (35%), 13
Prospective observational cohort study of consecutive adult (50%), and 5 (15%) patients had mild, moderate, and severe ARDS,
patients with CAP admitted to the ICU within 24 h from hospital respectively. Seventy-two (84%) non-ARDS VOP presented unilat-
admission, between November 1996 and December 2019. Inclusion eral infiltrates (1 quadrant 84%; 2 quadrants 12%).
criteria were: (1) hospitalized patients ≥80 years old with diagno- Compared to VOP without ARDS, VOP with ARDS had more
sis of CAP; (2) severe CAP (according to ATS/IDSA criteria)8 ; (3) ICU frequently received previous antibiotic therapy, and had higher
admission; and (4) either invasive (IMV) or non-invasive mechan- median C-reactive protein values (Table 1). An etiologic diagno-
ical ventilation (NIMV) within 24 h from admission. Patients were sis was obtained in 52 VOP (51%). The most frequent pathogen in
excluded if they had severe immunosuppression or active tubercu- both groups was S. pneumoniae (22 out of 40 patients in the non-
losis. ARDS diagnosis was based on Berlin definition.9 Chest-X-ray ARDS VOP group [55%] vs. 8 out 12 VOP in the ARDS group [67%],
involvement was analyzed as follows: involvement of one quadrant p = 0.473). The most frequent antibiotic regimens were ␤-lactam
was considered unilateral, while two quadrants could be unilateral plus either a respiratory fluoroquinolone (46%) or a macrolide
or bilateral; involvement of three or four quadrants was considered (25%). There were not differences in the empiric treatment between
bilateral.7 ARDS and non-ARDS VOP.
Descriptive statistics were used for basic features of study ICU, 30-day and 1-year mortality were significantly higher in
data and appropriate statistical tests were performed to compare the ARDS VOP group (p = 0.008, 0.031 and p = 0.024, respectively).
groups. A propensity score for VOP with ARDS was developed by Main causes of death were respiratory failure (non-ARDS VOP: 56%
means of a logistic regression model. The score was entered as a vs. ARDS VOP: 55%) and refractory shock with multi-organ failure
continuous variable into four logistic regression analyses to assess (non-ARDS VOP: 39% vs. ARDS VOP 36%). The propensity-adjusted
association between ARDS and outcomes (i.e., in-hospital, ICU, 30- analyses showed that ARDS patients had higher risk of in-hospital,
day, and 1-year mortality). A similar analysis was performed to ICU, 30-day and 1-year mortality compared to non-ARDS VOP
assess association between lung imaging quadrants and outcomes. (odds ratio [OR] 3.13 [95% confidence interval (CI) 1.02–9.61],
For publication purposes, the study was approved by the Comité OR 4.12 [95%CI 1.22–13.91], OR 3.32 [95%CI 1.09–10.13], and OR
Ètic d’Investigació Clínica, register: 2009/5451. The need for written 4.80 [95%CI 1.40–16.43], respectively), and confirmed by internal
informed consent was waived due to the non-interventional study validation using bootstrapping with 1000 bootstrap samples and
design. bias-corrected.
Among 6547 CAP patients admitted during the study period, VOP with bilateral infiltrates presented significantly higher ICU
4571 patients (69.8%) were < 80 years old, 904 (24%) of them were mortality rates than patients with unilateral/1-quadrant infiltrates

https://doi.org/10.1016/j.arbres.2021.05.027
0300-2896/© 2021 SEPAR. Published by Elsevier España, S.L.U. All rights reserved.
C. Cillóniz, J.M. Pericàs, H. Peroni et al. Archivos de Bronconeumología 58 (2022) 520–522

Fig. 1. Flow chart of study population (Panel A) and rate of VOP admitted to the ICU per year (Panel B).

Table 1
Patients characteristic and outcomes according to the presence of ARDS.

Variable ARDS

No Yes p-Valuea
(n = 75) (n = 27)

Age, median (Q1; Q3), years 83 (81; 85) 83 (81; 86) 0.361
Male sex, n (%) 52 (69) 15 (56) 0.196
Previous antibiotic, n (%) 10 (15) 9 (38) 0.022
Previous episode of pneumonia, n (%) 11 (16) 1 (4) 0.130

Comorbidities, n (%)b 61 (81) 20 (74) 0.424

Chronic respiratory disease 33 (45) 12 (48) 0.317
Chronic cardiovascular disease 14 (19) 5 (19) 0.995
Diabetes mellitus 29 (39) 8 (30) 0.402
Neurological disease 14 (20) 4 (16) 0.661
Chronic renal disease 12 (16) 4 (15) 0.864
Chronic liver disease 2 (3) 2 (7) 0.276

Nursing-home, n (%) 7 (10) 1 (4) 0.356

C-reactive protein, median (Q1; Q3), mg/dL 15.0 (6.3; 22.4) 25.1 (13.2; 28.9) 0.015
PSI score, median (Q1; Q3) 140 (121; 160) 141 (121; 179) 0.620
Severe CAP, n (%) 61 (82) 23 (88) 0.552
SOFA score, median (Q1; Q3) 5.5 (3; 7) 4.5 (3; 6) 0.525

CXR quadrants involved, n (%) <0.001

1 quadrant 63 (84) 0 (0) <0.001
2 quadrants 11 (15) 16 (59) <0.001
>2 quadrants 1 (1) 11 (41) <0.001

Unilateral/bilateral, n (%) <0.001

Unilateral – 1 quadrant 63 (84) 0 (0) <0.001
Unilateral – 2 quadrants 9 (12) 0 (0) <0.001
Bilateral 3 (4) 27 (100) <0.001

Appropriate empiric treatment, n (%) 49 (82) 23 (96) 0.165

Mechanical ventilation, n (%)c 0.368

Non-invasive 26 (35) 12 (44)
Invasive 49 (65) 15 (56)

Length of hospital stay, median (Q1; Q3), days 16 (11; 26) 17 (9; 23) 0.663
In-hospital mortality, n (%) 24 (32) 14 (52) 0.067
ICU mortality, n (%) 12 (16) 11 (41) 0.008
30-day mortality, n (%) 24 (32) 15 (56) 0.031
1-year mortality, n (%) 31 (42) 17 (68) 0.024

Abbreviations: ARDS indicates acute respiratory distress syndrome; CAP, community acquired pneumonia; CXR, chest X-ray; ICU, intensive care unit; PSI, pneumonia severity
index; Q1, first quartile; Q3, third quartile; SOFA, sequential organ failure assessment. We excluded patients with severe immunosuppression, active tuberculosis, CAP with
sepsis that development septic shock, and unavailable data. Percentages calculated on non-missing data.
a
Categorical variables were compared with the Chi-square or the Fisher exact test when necessary; continuous variables were compared with the non-parametric
Mann–Whitney test.
b
May have > 1 comorbid condition.
c
Patients who received initially non-invasive ventilation but needed subsequently intubation were included in the invasive mechanical ventilation group.

521
C. Cillóniz, J.M. Pericàs, H. Peroni et al. Archivos de Bronconeumología 58 (2022) 520–522

(p < 0.05). Propensity-adjusted analyses showed that VOP with 2. Haas LEM, Karakus A, Holman R, Cihangir S, Reidinga AC, de Keizer NF. Trends
bilateral infiltrates showed an increased risk of 1-year mortality in hospital and intensive care admissions in the Netherlands attributable to the
very elderly in an ageing population. Crit Care. 2015;19:353.
compared to VOP with unilateral/1-quadrant infiltrates (OR 4.01 3. Cillóniz C, Ewig S, Polverino E, Marcos MA, Esquinas C, Gabarrús A, et al. Micro-
[95%CI 1.27–12.62]), confirmed by internal validation using boot- bial aetiology of community-acquired pneumonia and its relation to severity.
strapping with 1000 bootstrap samples and bias-corrected. Thorax. 2011;66:340–6.
4. Wunsch H, Guerra C, Barnato AE, Angus DC, Li G, Linde-Zwirble WT. Three-
The main findings of our study are as follows. First, 10% of year outcomes for Medicare beneficiaries who survive intensive care. JAMA.
patients with CAP admitted to the ICU were VOP and there was 2010;303:849–56.
an increase in the proportion of VOP admitted to the ICU overtime, 5. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, et al. Epidemiology, pat-
terns of care, and mortality for patients with acute respiratory distress syndrome
which is in accordance with reports showing a progressive increase
in intensive care units in 50 countries. JAMA. 2016;315:788–800.
of critically ill VOP worldwide.1,2,10,11 Second, 52% of VOP admitted 6. Cilloniz C, Ferrer M, Liapikou A, Garcia-Vidal C, Gabarrus A, Ceccato A, et al.
to ICU received MV, which is also consistent with prior studies. For Acute respiratory distress syndrome in mechanically ventilated patients with
community-acquired pneumonia. Eur Respir J. 2018:51.
instance, Storms et al.12 found that 42% of patients admitted to the
7. Pham T, Pesenti A, Bellani G, Rubenfeld G, Fan E, Bugedo G, et al. Outcome of acute
ICU due to CAP received MV, whereas in a study including 930 with hypoxaemic respiratory failure. Insights from the lung safe study. Eur Respir J.
CAP admitted to the ICU, we observed that 46.5% received MV.6 2020.
Third, ARDS developed in 26% of very old CAP patients treated in 8. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, et al.
Infectious Diseases Society of America/American Thoracic Society consensus
ICU with either IMV or NIMV, which is similar to the overall per- guidelines on the management of community-acquired pneumonia in adults.
centage of ARDS found in studies also including non-VOP adults.6,13 Clin Infect Dis. 2007;44 Suppl. 2:S27–72.
Yet, some of these studies showed an age-dependent gradient in the 9. Ferguson ND, Fan E, Camporota L, Antonelli M, Anzueto A, Beale R, et al. The
Berlin definition of ARDS: an expanded rationale, justification, and supplemen-
incidence of ARDS in the general population, with a trend towards tary material. Intensive Care Med. 2012;38:1573–82.
higher percentages of ARDS amongst older patients.13,14 Fourth, 10. Muessig JM, Masyuk M, Nia AM, Franz M, Kabisch B, Kelm M, et al. Are we ever
ARDS was associated with significantly higher risk of both short- too old? Characteristics and outcome of octogenarians admitted to a medical
intensive care unit. Medicine (Baltimore). 2017;96:e7776.
term and long-term mortality, which provides insight for clinical 11. Roch A, Wiramus S, Pauly V, Forel J-M, Guervilly C, Gainnier M, et al. Long-term
decision-making, i.e., early implementation of measures to prevent outcome in medical patients aged 80 or over following admission to an intensive
ARDS development. Interestingly, in patients under 80 years old we care unit. Crit Care. 2011;15:R36.
12. Storms AD, Chen J, Jackson LA, Nordin JD, Naleway AL, Glanz JM, et al. Rates and
did not observe differences in outcomes between ARDS and non- risk factors associated with hospitalization for pneumonia with ICU admission
ARDS, which is consistent with prior findings of our group.6 Finally, among adults. BMC Pulm Med. 2017;17:208.
patients with unilateral infiltrates had lower severity than patients 13. Manzano F, Yuste E, Colmenero M, Aranda A, García-Horcajadas A, Rivera R, et al.
Incidence of acute respiratory distress syndrome and its relation to age. J Crit
with bilateral infiltrates while patients with bilateral infiltrates had
Care. 2005;20:274–80.
higher ICU mortality than patients with unilateral infiltrates. Our 14. Rubenfeld GD, Caldwell E, Peabody E, Weaver J, Martin DP, Neff M, et al. Incidence
results contrast with those of Pham et al.,7 who found that patients and outcomes of acute lung injury. N Engl J Med. 2005;353:1685–93.
with unilateral-infiltrate had mortality rates comparable to that of
patients with ARDS of similar severity. Catia Cillóniz a,∗ , Juan M. Pericàs b,c , Héctor Peroni d ,
Our study is limited by its small sample size, which precluded Enric Barbeta a , Albert Gabarrús a , Antoni Torres a,∗
further relevant sub-analysis such as the impact of ARDS severity
a Department of Pneumology, Institut Clinic del Tòrax, Hospital Clinic
on mortality. However, to our knowledge ours is the first study
providing information on ARDS due to CAP in VOP admitted to ICU. of Barcelona, Institut d’Investigacions Biomèdiques August Pi i Sunyer
In conclusion, ARDS in VOP with CAP admitted to ICU and ven- (IDIBAPS), University of Barcelona (UB) – SGR 911 – Ciber de
tilated is associated with higher risk of morbidity and mortality. Enfermedades Respiratorias (Ciberes), Barcelona
b Department of Infectious Diseases, Hospital Clinic of Barcelona,
Further research is required in order to enhance clinical decision-
making in VOP with severe CAP. Barcelona, Spain
c Vall d’Hebron Institute for Research (VHIR), Barcelona, Spain
d Internal Medicine Department, Respiratory Medicine Unit and
Conflicts of interest
Emergency Department, Hospital Italiano de Buenos Aires, Buenos
The authors declare that they have no conflicts of interest. Aires, Argentina

∗ Corresponding author.
References
E-mail addresses: [email protected] (C. Cillóniz),
1. Laporte L, Hermetet C, Jouan Y, Gaborit C, Rouve E, Shea KM, et al. Ten-year [email protected] (A. Torres).
trends in intensive care admissions for respiratory infections in the elderly. Ann
Intensive Care. 2018;8:84.

522