Abstract
Background
Lymphopenia is a marker of poor prognosis in patients with community-acquired pneumonia (CAP), yet its impact on outcomes in patients with CAP and sepsis remains unknown. We aim to investigate the impact of lymphopenia on outcomes, risk of intensive care unit (ICU) admission, and mortality in CAP patients with sepsis.
Methods
This was a retrospective, observational study of prospectively collected data from an 800-bed tertiary teaching hospital (2005–2019).
Results
Of the 2203 patients with CAP and sepsis, 1347 (61%) did not have lymphopenia, while 856 (39%) did. When compared with the nonlymphopenic group, patients with sepsis and lymphopenia more frequently required ICU admission (P = .001), had a longer hospital length of stay (P ˂ .001), and presented with a higher rate of in-hospital (P ˂ .001) and 30-day mortality (P = .001). Multivariable analysis showed that C-reactive protein ≥15 mg/dL, lymphopenia, pleural effusion, and acute respiratory distress syndrome within 24 hours of admission were risk factors for ICU admission; age ≥80 years was independently associated with decreased ICU admission. In addition, age ≥80 years, chronic renal disease, chronic neurologic disease, being a nursing home resident, lymphopenia, and pleural effusion were independently associated with increased 30-day mortality, whereas pneumococcal vaccination, diabetes mellitus, and fever were independently associated with reduced 30-day mortality.
Conclusions
Lymphopenia was independently associated with risk of ICU admission and higher in-hospital and 30-day mortality in patients with CAP and sepsis. Early identification of lymphopenia could help identify septic patients with CAP who require or will shortly require critical care.
Keywords: infection, lymphopenia, outcomes, pneumonia, sepsis
Sepsis presents in approximately one-third of patients with severe community-acquired pneumonia (CAP), and ~74% of patients with sepsis present with lymphopenia [1–3]. Our group recently reported a particular immunophenotype of patients with CAP, which we named lymphopenic (<724 lymphocytes/mm3) CAP (L-CAP); this immunotype was found to be associated with increased severity and mortality [2]. We also observed that half of the patients with CAP showed lymphopenia upon hospital admission despite no history of immunosuppression [2], with similar findings previously reported [4, 5]. L-CAP is characterized by a depletion of CD4+ T lymphocytes, a greater inflammatory response, and low levels of IgG2, which were also correlated with greater severity in presentation and worse prognosis in patients with CAP [6]. Lymphopenia was also related to severity and poorer outcomes in patients with influenza virus–derived CAP and coronavirus disease 2019 (COVID-19) [7–9]. Lymphopenia was also reported to be an independent predictor of mortality in primary care pneumonia [10]. The same study furthermore showed that a low lymphocyte count (1–2 × 109 cells/L) was associated with an increase in short- and long-term mortality when compared with higher lymphocyte counts [10]. Finally, baseline lymphopenia was reported to be associated with an elevated risk of infections such as pneumonia in the general population [11]. Hence the importance of early identification of CAP patients with lymphopenia [12].
We hypothesize that lymphopenia in patients with CAP and sepsis is associated with higher severity and mortality. Therefore, the objectives of the present study were to investigate the impact of lymphopenia on risk factors for ICU admission and mortality in patients with CAP and sepsis, as well as patient outcomes.
METHODS
Study Design and Patients
This was a retrospective, observational study of data prospectively collected from the Hospital Clinic of Barcelona, Spain. We enrolled all consecutive, adult patients with a diagnosis of CAP admitted to the hospital via the emergency department between January 2005 and December 2019. We included patients from nursing homes, as we had demonstrated that microbial etiology in this population was similar to that of CAP in people residing in their own homes [13]. We excluded patients with severe immunosuppression due to, but not limited to, human immunodeficiency viral infection, active solid or hematologic malignancy who received chemotherapy, oral corticosteroid treatment with at least 20 mg of prednisone (or equivalent) per day for at least 2 weeks, or treatment with other immunosuppressive drugs. We also excluded individuals with active tuberculosis or a confirmed alternative diagnosis. Among all subjects with CAP, we selected patients with sepsis and performed a comparison between those with and without lymphopenia.
Patient Consent Statement
For publication purposes, the study was approved by the Ethics Committee of our institution (register: 2009/5451). The need for written informed consent was waived due to the noninterventional study design.
Definitions
Lymphopenic patients were defined as those with <724 lymphocytes/mm3 [2]. Pneumonia (CAP) was defined as the appearance of a new pulmonary infiltrate on chest x-ray during hospitalization, accompanied by symptoms and signs of a lower respiratory tract infection. Severe CAP was diagnosed by fulfilment of at least 1 major or 3 minor criteria per guidelines set by Infectious Disease Society of America/American Thoracic Society (IDSA/ATS) [14]. Polymicrobial pneumonia was defined as pneumonia due to >1 pathogen.
Prior antibiotic treatment was defined as the intake of antibiotics during the week before hospital admission. The appropriateness of empiric antibiotic treatment was determined according to multidisciplinary guidelines for the management of CAP [15].
Sepsis was defined as the presence of pneumonia and an increase of ≥2 points in the Sequential Organ Failure Assessment (SOFA) score per criteria of the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) [16]. The presence of sepsis was evaluated upon hospital admission during diagnosis of CAP. The presence of acute respiratory distress syndrome (ARDS) was evaluated within the first 24 hours of hospital admission based on the Berlin definition [17].
Data Collection, Evaluation, and Microbiologic Diagnosis
Demographic variables, comorbidities, and physiologic parameters were collected at the emergency department within 24 hours of admission. The Pneumonia Severity Index (PSI) and SOFA score were calculated at admission [18, 19]. We recorded whether patients had specific complications, including multilobar infiltration, pleural effusions, ARDS [17], septic shock [3], and acute renal failure [20] during hospitalization. Quantification of lymphocytes was performed on blood samples collected in ethylenediaminetetraacetic acid tubes. We used automatic analyzers available at the central laboratory following standard operating procedures approved for clinical use. All laboratory data were gathered at the time of patient admission. Further details are reported elsewhere [21]. All surviving patients were visited or contacted by telephone within 30 days of discharge; hospital records and the Catalunya Health Department database were reviewed at the 1-year mark.
Microbiologic diagnosis was performed in respiratory, urinary, and blood samples. Blood cultures, sputum cultures, and urine samples for Streptococcus pneumoniae and Legionella pneumophila antigen detection were obtained within 24 hours of hospital admission. When available, pleural fluid, tracheobronchial aspirates (TBAS), and bronchoalveolar lavage (BAL) samples were collected for gram and Ziehl-Nielsen staining and processed for detection of bacterial, fungal, and mycobacterial pathogens. Blood samples for serology of atypical pathogens and respiratory virus were collected at admission and thereafter between the third and sixth weeks.
Cultures were collected before starting empiric antibiotic therapy at the emergency department. Bacterial etiology was considered definite if 1 of the following criteria was met: (1) positive blood culture (in the absence of an apparent extrapulmonary focus); (2) positive bacterial culture of pleural fluid or transthoracic needle aspiration samples; (3) positive urinary antigen for Legionella pneumophila (Binax Now Legionella pneumophila Urinary Antigen Test; Trinity Biotech, Bray, Ireland); (4) positive urinary antigen for S. pneumoniae (Binax Now Streptococcus pneumoniae Urinary Antigen Test; Emergo Europe, the Hague, the Netherlands); (5) bacterial growth in cultures of TBAS (≥105 cfu/mL) in protected specimen brush (≥103 cfu/mL) or BAL (≥104 cfu/mL). More details about microbiologic diagnosis have been reported previously [21].
Respiratory viruses were diagnosed by serology, immunofluorescence assay (IFA), and isolation in cell cultures between 2005 and 2007. However, polymerase chain reaction (PCR) and/or cultures of nasopharyngeal swab samples were used in diagnosis between 2008 and 2019. Two independent, nested, multiplex real-time PCR tests were used to detect human influenza viruses (A, B, and C), respiratory syncytial virus, adenoviruses, parainfluenza viruses (1–4), coronaviruses (229E and OC43), enteroviruses and rhinoviruses (A, B, and C).
Outcomes
The primary outcome was 30-day mortality. Secondary outcomes were in-hospital mortality, ICU admission, ICU mortality, and the need for mechanical ventilation.
Statistical Analysis
We report the number and percentage of patients for categorical variables, the median (first quartile, third quartile) for continuous variables with non-normal distributions, and the mean (standard deviation) for continuous variables with normal distributions. Categorical variables were compared using the chi-square test or Fisher exact test, whereas continuous variables were compared using the t test or nonparametric Mann-Whitney U test.
Time to 30-day mortality was analyzed by Kaplan-Meier survival curves, which were then compared using the Gehan-Breslow-Wilcoxon test. Univariate and multivariable logistic regression analyses [22] were performed to identify variables associated with ICU admission. Factors showing an association in the univariate analyses (P < .20) were incorporated into the multivariable regression model. Final variable selection was performed using the backward stepwise selection method (likelihood ratio) (Pin < .05; Pout > .10). Odds ratios (ORs) and their 95% CIs were calculated. The Hosmer-Lemeshow goodness-of-fit test was performed to assess the overall fit of the model. Associations with 30-day mortality were also tested by univariate and multivariable analyses, with similar inclusion criteria applied for the Cox regression analysis (P < .20). Hazard ratios (HRs) and their 95% CIs were calculated. Proportional hazards assumptions were tested with log-minus-log plots. To investigate the fit of the final model, we evaluated deviance residuals. A subgroup analysis also examined 30-day mortality for patients admitted to the ICU. Areas under the receiver operating characteristic curve (AUC) of the multivariable models were then calculated to predict both ICU admission and 30-day mortality. The internal validity of the prediction models was assessed using ordinary nonparametric bootstrapping with 1000 bootstrap samples and bias-corrected, accelerated 95% CIs [23]. We used the multiple imputation method [24] for missing data in multivariable analyses. When analyzing factors associated with ICU admission and 30-day mortality, we excluded 236 patients with septic shock, as other studies had shown that septic shock was the main risk factor for mortality in patients with severe CAP [25, 26]. We also excluded an additional 211 patients who had do-not-resuscitate (DNR) orders.
Additional analyses on patient outcomes were performed considering lymphopenia as a total lymphocyte count ˂1000/mm3 and based on lymphocyte quartiles.
The level of significance was set at .05 (2-tailed), and all analyses were performed using IBM SPSS, version 25.0 (IBM Corp., Armonk, NY, USA).
RESULTS
We identified 4521 consecutive patients admitted to the emergency department for CAP during the study period, ultimately excluding 2318 (51%) from the analysis (Supplementary Figure 1). The final study population therefore comprised 2203 (49%) patients with CAP and sepsis.
Comparison of Patients With CAP and Sepsis, Presenting With and Without Lymphopenia
Of the 2203 patients with CAP and sepsis, 1347 (61%) did not have lymphopenia, while 856 (39%) did. Table 1 summarizes demographic and clinical characteristics of patients according to the presence or absence of lymphopenia. When compared with the nonlymphopenic group, patients in the lymphopenic group were more likely to be male, receive inhaled corticosteroids less often, had a lower rate of influenza vaccination, and presented with more chronic respiratory diseases and prior malignancies as comorbidities. Additionally, lymphopenic patients showed higher C-reactive protein and creatinine levels, as well as lower neutrophil and overall white blood cell counts within 24 hours of hospital admission. The lymphopenic group also presented with higher rates of severe CAP, bacteremia, multilobar infiltration, and septic shock.
Table 1.
Characteristics of Patients With CAP and Sepsis According to the Presence or Absence of Lymphopenia
| Lymphopenia (<724 Lymphocytes/mm3) | |||
|---|---|---|---|
| Variable | No (n = 1347) | Yes (n = 856) | P Value |
| Age, mean (SD), y | 72.1 (15.8) | 72.4 (16.3) | .303 |
| Male sex, No. (%) | 830 (62) | 583 (68) | .002 |
| Current smoker, No. (%) | 262 (20) | 157 (19) | .555 |
| Current alcohol user, No. (%) | 188 (14) | 107 (13) | .353 |
| Previous antibiotic in last week, No. (%) | 297 (23) | 165 (20) | .117 |
| Influenza vaccine, No. (%) | 635 (53) | 329 (45) | ˂.001 |
| Pneumococcal vaccine, No. (%) | 236 (20) | 151 (21) | .758 |
| Previous inhaled corticosteroids, No. (%) | 308 (23) | 155 (18) | .007 |
| Previous systemic corticosteroids, No. (%) | 72 (6) | 41 (5) | .569 |
| Prior pneumonia (last year), No. (%) | 192 (15) | 123 (15) | .843 |
| Comorbidities, No. (%)a | 1020 (76) | 630 (74) | .203 |
| Chronic respiratory disease | 570 (43) | 324 (39) | .048 |
| Chronic cardiovascular disease | 232 (17) | 123 (14) | .079 |
| Diabetes mellitus | 347 (26) | 212 (25) | .685 |
| Chronic neurologic disease | 295 (23) | 174 (21) | .284 |
| Chronic renal disease | 129 (10) | 84 (10) | .859 |
| Chronic liver disease | 65 (5) | 44 (5) | .728 |
| Previous malignancy | 93 (7) | 107 (13) | <.001 |
| Nursing home resident, No. (%) | 139 (10) | 85 (10) | .799 |
| Dyspnea, No. (%) | 1010 (77) | 611 (73) | .066 |
| Pleuritic pain, No. (%) | 420 (32) | 238 (29) | .119 |
| Fever, No. (%) | 953 (72) | 616 (73) | .468 |
| Deterioration in sensorium, No. (%) | 374 (28) | 221 (26) | .328 |
| Respiratory rate, median (IQR), breaths/min | 24 (21–32) | 24 (21–32) | .632 |
| C-reactive protein, median (IQR), mg/dL | 17.7 (8.9–26.7) | 19.5 (8.8–28.8) | .012 |
| Neutrophils, median (IQR), cells/mm3 | 11.4 (8.1–16.3) | 7.9 (3.8–12.3) | <.001 |
| Creatinine, median (IQR), mg/dL | 1.2 (0.9–1.6) | 1.3 (0.9–1.7) | .024 |
| White blood cell count, median (IQR), cells/mm3 | 14.2 (10.3–19.6) | 9.3 (4.9–13.7) | <.001 |
| PSI risk class IV–V, No. (%)b | 634 (67) | 401 (72) | .053 |
| Severe CAP, No. (%) | 407 (38) | 330 (49) | <.001 |
| Bacteremia, No. (%)c | 97 (10) | 138 (21) | <.001 |
| Pleural effusion, No. (%) | 182 (14) | 118 (14) | .789 |
| Multilobar infiltration, No. (%) | 363 (27) | 273 (32) | .013 |
| Acute respiratory distress syndrome, No. (%) | 83 (6) | 68 (8) | .106 |
| Acute renal failure, No. (%) | 471 (36) | 322 (38) | .183 |
| Septic shock, No. (%) | 129 (10) | 107 (13) | .028 |
| Do-not-resuscitate order, No. (%) | 121 (9) | 90 (11) | .162 |
| Empiric antibiotic therapy, No. (%) | |||
| Monotherapy | 432 (33) | 205 (25) | <.001 |
| Fluoroquinolones | 305 (23) | 126 (15) | <.001 |
| β-lactams | 121 (9) | 76 (9) | .933 |
| Other therapy | 6 (0.4) | 3 (0.3) | >.999 |
| Combination therapies | 882 (67) | 630 (75) | <.001 |
| β-lactams plus fluoroquinolones | 402 (31) | 284 (34) | .098 |
| β-lactams plus macrolides | 358 (27) | 254 (30) | .112 |
| Other combination therapies | 122 (9) | 92 (11) | .191 |
| Appropriate empiric treatment, No. (%) | 1123 (95) | 699 (95) | .756 |
Percentages were calculated with nonmissing data only.
Abbreviations: CAP, community-acquired pneumonia; IQR, interquartile range; PSI, pneumonia severity index.
aMay have >1 comorbid condition.
bStratified according to 30-day mortality risk for CAP: classes I–III (≤90 points) had low mortality risk while classes IV–V (>90 points) had the highest mortality risk.
cCalculated only for patients with blood samples (983 patients in the nonlymphopenic group and 673 patients in the lymphopenic group were used to calculate percentages).
An etiologic diagnosis was achieved more often in the lymphopenic group (47% vs 39%; P < .001). Atypical pathogens were more frequently detected in the nonlymphopenic group (2% vs 6%; P = .001); Legionella pneumophila was more common in the lymphopenic group (7% vs 3%; P = .022) than in the nonlymphopenic group.
Data were available for empiric antibiotic treatment in 2149 (98%) patients. The most frequent regimens were ß-lactam plus fluoroquinolone (32%) and ß-lactam plus macrolide (28%). No differences in empiric antibiotic therapy were present between groups, except for a more frequent administration of fluoroquinolones in monotherapy in the nonlymphopenic group (23% vs 15%; P < .001). Inappropriate empiric antibiotic therapy rates were comparable (5% and 5%; P = .756).
Outcomes of Patients With CAP and Sepsis, Presenting With and Without Lymphopenia
Patients with lymphopenia required ICU admission more frequently (25% vs 32%; P = .001), had a longer hospital length of stay (8 days vs 9 days; P < .001), and presented with higher in-hospital (8% vs 12%; P < .001) and 30-day mortality (8% vs 12%; P = .001) (Table 2). Per the presence or absence of lymphocytopenia, Kaplan-Meier curves for 30-day survival in the entire population and in the subgroup of ICU patients are shown in Figures 1 and 2, respectively.
Table 2.
Clinical Outcomes According to Lymphopenia in Patients With CAP and Sepsis
| Lymphopenia (<724 Lymphocytes/mm3) | |||
|---|---|---|---|
| Variable | No (n = 1347) | Yes (n = 856) | P Value |
| Hospital length of stay, median (IQR), d | 8 (6; 13) | 9 (6; 15) | <.001 |
| In-hospital mortality, No. (%) | 104 (8) | 105 (12) | <.001 |
| ICU admission, No. (%) | 340 (25) | 270 (32) | .001 |
| ICU mortality, No. (%)a | 26 (8) | 24 (9) | .579 |
| Mechanical ventilation, No. (%)b | .122 | ||
| Noninvasive | 84 (7) | 67 (10) | .118 |
| Invasive | 103 (9) | 76 (11) | .245 |
| 30-d mortality, No. (%) | 103 (8) | 101 (12) | .001 |
Abbreviations: CAP, community-acquired pneumonia; ICU, intensive care unit; IQR, interquartile range.
aCalculated only for patients admitted to the ICU (340 patients in the nonlymphopenic group and 270 patients in the lymphopenic group were used to calculate the percentages).
bPatients who received noninvasive ventilation initially but later needed subsequent intubation were included in the invasive mechanical ventilation group.
Figure 1.
Kaplan-Meier survival curves for 30-day mortality in patients with community-acquired pneumonia and sepsis in relation to their lymphocyte counts (<724 vs ≥724 lymphocytes/mm3).
Figure 2.
Kaplan-Meier survival curves for 30-day mortality in intensive care unit patients with community-acquired pneumonia and sepsis in relation to their lymphocyte counts (<724 vs ≥724 lymphocytes/mm3).
When lymphopenia was defined as ˂1000/mm3, it was associated with a longer length of hospital stay and higher in-hospital and 30-day mortality rates (Supplementary Table 1). We also analyzed patient outcomes according to lymphocyte quartiles (<531 vs 531–874.4 vs 874.4–1350 vs ≥1350 lymphocytes/mm3) (Supplementary Table 2). In relation to lymphocyte quartiles, Kaplan-Meier survival curves for 30-day mortality in the entire population and in the subgroup of ICU patients are shown in Supplementary Figures 2 and 3, respectively. Similar outcomes according to lymphocyte quartiles were observed among patients.
Risk Factors for ICU Admission in Patients With CAP and Sepsis
The univariate logistic regression analysis identified several variables significantly associated with ICU admission in patients with CAP and sepsis (Table 3). In the multivariable analysis, C-reactive protein ≥15 mg/dL, lymphopenia, pleural effusion, and ARDS were risk factors for ICU admission in patients with sepsis, while age ≥80 years was independently associated with decreased ICU admission. The AUC was 0.70 (95% CI, 0.67–0.73) for the predictive model of ICU admission. Internal validation of the final model was conducted with a bootstrapping procedure with 1000 samples, demonstrating robust results inasmuch as all variables remained significant with small 95% CIs around the original coefficients.
Table 3.
Univariate Logistic Regression Analysis for Variables Associated With ICU Admission and Independent Predictors of ICU Admission Determined by Multivariable Logistic Regression Analysis (n = 1664)a
| Univariate | Multivariableb | |||||
|---|---|---|---|---|---|---|
| Variable | OR | 95% CI | P Value | OR | 95% CI | P Value |
| Age ≥80 y | 0.32 | 0.24–0.42 | <.001 | 0.31 | 0.23–0.41 | <.001 |
| Male sex | 1.33 | 1.04–1.71 | .024 | - | - | - |
| Previous inhaled corticosteroids | 0.98 | 0.74–1.29 | .884 | - | - | - |
| Previous systemic corticosteroids | 1.16 | 0.66–2.03 | .606 | - | - | - |
| Antibiotic use in the last week | 0.90 | 0.68–1.19 | .442 | - | - | - |
| Chronic respiratory disease | 0.96 | 0.76–1.21 | .708 | - | - | - |
| Chronic cardiovascular disease | 0.76 | 0.54–1.05 | .099 | - | - | - |
| Chronic renal disease | 0.63 | 0.40–0.99 | .046 | - | - | - |
| Chronic liver disease | 1.16 | 0.67–2.00 | .593 | - | - | - |
| Diabetes mellitus | 0.85 | 0.65–1.11 | .224 | - | - | - |
| Chronic neurologic disease | 0.63 | 0.46–0.87 | .005 | - | - | - |
| Previous pneumonia | 0.70 | 0.49–0.99 | .043 | - | - | - |
| Nursing home resident | 0.62 | 0.39–0.99 | .044 | - | - | - |
| Fever | 0.91 | 0.70–1.18 | .474 | - | - | - |
| Deterioration in sensorium | 0.90 | 0.68–1.19 | .464 | - | - | - |
| C-reactive protein ≥15 mg/dL | 1.55 | 1.21–1.98 | <.001 | 1.39 | 1.07–1.79 | .012 |
| Lymphopenia (<724 lymphocytes/mm3) | 1.32 | 1.04–1.67 | .022 | 1.37 | 1.07–1.76 | .013 |
| Pleural effusion | 1.85 | 1.35–2.53 | <.001 | 1.81 | 1.30–2.52 | <.001 |
| Acute respiratory distress syndrome | 5.74 | 3.63–9.09 | <.001 | 6.14 | 3.78–9.96 | <.001 |
| Acute renal failurec | 1.22 | 0.96–1.55 | .108 | - | - | - |
| Streptococcus pneumoniae | 1.36 | 1.03–1.79 | .033 | - | - | - |
| Respiratory virus | 1.23 | 0.85–1.77 | .267 | - | - | - |
Data are shown as estimated ORs (95% CIs) of the explanatory variables in the sepsis group. The OR represents the odds that the presence of ICU admission will occur given exposure of the explanatory variable, compared with the odds of the outcome occurring in the absence of that exposure. The P values are based on the null hypothesis that all ORs relating to an explanatory variable equal unity (no effect).
aExcluded 236 patients with septic shock, 34 patients with missing data regarding septic shock, 211 patients who had do-not-resuscitate orders, and 58 with missing data regarding a do-not-resuscitate order.
Abbreviations: ICU, intensive care unit; OR, odds ratio.
bHosmer-Lemeshow goodness-of-fit test, P = .597.
cVariable highly correlated with another independent variable and therefore not included in the multivariable model.
When using ˂1000 lymphocytes/mm3 as a cutoff for lymphopenia, the latter was not found to be a risk factor for ICU admission (Supplementary Table 3). Conversely, the analysis according to lymphocyte quartile showed that presenting with lymphocyte counts <531 lymphocytes/mm3 constituted a risk factor for ICU admission (Supplementary Table 4).
Factors Associated With 30-Day Mortality in Patients With CAP and Sepsis
In the multivariable analysis (Table 4), age ≥80 years, chronic renal disease, chronic neurologic disease, nursing home resident, lymphopenia, and pleural effusion were independently associated with increased 30-day mortality, while pneumococcal vaccine, diabetes mellitus, and fever were independently associated with decreased 30-day mortality. The AUC was 0.59 (95% CI, 0.53–0.66) for the multivariable model of 30-day mortality. In the subgroup of patients requiring ICU admission, age ≥80 years, chronic respiratory disease, chronic liver disease, lymphopenia, and acute renal failure were factors associated with 30-day mortality (AUC, 0.68; 95% CI, 0.56–0.81) (Table 5). Internal validation demonstrated robust results for all included variables in both multivariable models, with small 95% CIs around the original coefficients. In an additional analysis of factors associated with the risk of 30-day mortality, lymphopenia with a cutoff of <1000 lymphocytes/mm3 was not a risk factor for such mortality in the entire population or in the subgroup of ICU patients (Supplementary Tables 5 and 6). However, when we repeated the analysis including the variable according to lymphocyte quartiles, lymphopenia (<531 lymphocytes/mm3) was a risk factor for 30-day mortality in the entire population, yet not in the subgroup of ICU patients (Supplementary Tables 7 and 8).
Table 4.
Univariate Cox Regression Analysis for Variables Associated With 30-Day Mortality and Independent Predictors of 30-Day Mortality Determined by Multivariable Cox Regression Analysis (n = 1664)a
| Univariate | Multivariable | |||||
|---|---|---|---|---|---|---|
| Variable | HR | 95% CI | P Value | HR | 95% CI | P Value |
| Age ≥80 y | 3.99 | 2.55–6.24 | <.001 | 2.87 | 1.80–4.57 | <.001 |
| Male sex | 1.12 | 0.73–1.73 | .596 | - | - | - |
| Influenza vaccine | 0.89 | 0.59–1.33 | .565 | - | - | - |
| Pneumococcal vaccine | 0.57 | 0.31–1.05 | .071 | 0.49 | 0.27–0.92 | .025 |
| Previous inhaled corticosteroids | 1.31 | 0.83–2.05 | .247 | - | - | - |
| Previous systemic corticosteroids | 0.75 | 0.24–2.36 | .620 | - | - | - |
| Antibiotic use in the last week | 1.21 | 0.77–1.92 | .408 | - | - | - |
| Chronic respiratory disease | 1.00 | 0.67–1.51 | .993 | - | - | - |
| Chronic cardiac disease | 1.57 | 0.97–2.53 | .064 | - | - | - |
| Chronic renal disease | 2.78 | 1.69–4.55 | <.001 | 2.64 | 1.59–4.38 | <.001 |
| Chronic liver disease | 0.97 | 0.36–2.65 | .956 | - | - | - |
| Diabetes mellitus | 0.48 | 0.27–0.85 | .011 | 0.42 | 0.24–0.75 | .003 |
| Chronic neurologic disease | 3.33 | 2.12–5.00 | <.001 | 2.48 | 1.61–3.81 | <.001 |
| Previous pneumonia | 0.88 | 0.49–1.58 | .663 | - | - | - |
| Nursing home resident | 3.56 | 2.21–5.74 | <.001 | 2.17 | 1.31–3.61 | .003 |
| Fever | 0.44 | 0.30–0.67 | <.001 | 0.49 | 0.32–0.73 | .001 |
| Deterioration in sensorium | 1.99 | 1.31–3.02 | .001 | - | - | - |
| C-reactive protein ≥15 mg/dL | 0.72 | 0.48–1.08 | .111 | - | - | - |
| Lymphopenia (<724 lymphocytes/mm3) | 1.73 | 1.15–2.59 | .008 | 1.94 | 1.29–2.93 | .001 |
| Pleural effusion | 1.81 | 1.10–2.97 | .018 | 1.90 | 1.15–3.14 | .013 |
| Acute respiratory distress syndrome | 1.63 | 0.76–3.53 | .211 | - | - | - |
| Acute renal failureb | 2.72 | 1.81–4.09 | <.001 | - | - | - |
| Appropriate empiric antibiotic treatment | 0.99 | 0.36–2.68 | .977 | - | - | - |
| Streptococcus pneumoniae | 1.05 | 0.64–1.74 | .846 | - | - | - |
| Respiratory virus | 0.68 | 0.32–1.48 | .334 | - | - | - |
Data are shown as estimated HRs (95% CIs) of the explanatory variables in the 30-day mortality group. The HR is defined as the ratio of the hazard rates corresponding to the conditions described by 2 levels of an explanatory variable (the hazard rate is the risk of death [ie, the probability of death], given that the patient has survived up to a specific time). The P value is based on the null hypothesis that all HRs relating to an explanatory variable equal unity (no effect).
Abbreviation: HR, hazard ratio.
aExcluded 236 patients with septic shock, 34 patients with missing data regarding septic shock, 211 patients who had do-not-resuscitate orders, and 58 with missing data regarding a do-not-resuscitate order.
bVariable highly correlated with another independent variable and therefore not included in the multivariable model.
Table 5.
Univariate Cox Regression Analysis for Variables Associated With 30-Day Mortality and Independent Predictors of 30-Day Mortality Determined by Multivariable Cox Regression Analysis in ICU Patients (n = 368)a
| Univariate | Multivariable | |||||
|---|---|---|---|---|---|---|
| Variable | HR | 95% CI | P Value | HR | 95% CI | P Value |
| Age ≥80 y | 5.49 | 2.31–13.03 | <.001 | 6.75 | 2.60–17.50 | <.001 |
| Male sex | 0.71 | 0.30–1.72 | .452 | - | - | - |
| Influenza vaccine | 1.00 | 0.41–2.40 | .992 | - | - | - |
| Pneumococcal vaccine | 0.58 | 0.13–2.47 | .457 | - | - | - |
| Previous inhaled corticosteroids | 2.17 | 0.90–5.24 | .084 | - | - | - |
| Previous systemic corticosteroids | 2.20 | 0.51–9.43 | .290 | - | - | - |
| Antibiotic use in the last week | 1.46 | 0.57–3.76 | .435 | - | - | - |
| Chronic respiratory disease | 2.32 | 0.96–5.60 | .061 | 2.61 | 1.08–6.33 | .033 |
| Chronic cardiac disease | 2.58 | 1.00–6.65 | .050 | - | - | - |
| Chronic renal disease | 3.66 | 1.23–10.88 | .020 | - | - | - |
| Chronic liver disease | 3.46 | 1.02–11.75 | .047 | 8.53 | 2.18–33.48 | .002 |
| Diabetes mellitus | 1.00 | 0.36–2.72 | .993 | - | - | - |
| Chronic neurologic disease | 1.89 | 0.69–5.15 | .216 | - | - | - |
| Previous pneumonia | 0.77 | 0.18–3.32 | .729 | - | - | - |
| Nursing home resident | 1.75 | 0.41–7.53 | .450 | - | - | - |
| Fever | 1.27 | 0.47–3.48 | .637 | - | - | - |
| Deterioration in sensorium | 1.43 | 0.56–3.70 | .456 | - | - | - |
| C-reactive protein ≥15 mg/dL | 0.54 | 0.23–1.27 | .157 | - | - | - |
| Lymphopenia (<724 lymphocytes/mm3) | 2.30 | 0.95–5.55 | .064 | 2.57 | 1.06–6.24 | .037 |
| Pleural effusion | 1.32 | 0.48–3.61 | .586 | - | - | - |
| Acute respiratory distress syndrome | 1.15 | 0.34–3.90 | .825 | - | - | - |
| Acute renal failure | 3.66 | 1.48–9.06 | .005 | 3.52 | 1.38–8.95 | .008 |
| Appropriate empiric antibiotic treatment | 1.62 | 0.22–12.08 | .637 | - | - | - |
| Streptococcus pneumoniae | 1.32 | 0.51–3.41 | .562 | - | - | - |
| Respiratory virus | 0.36 | 0.05–2.70 | .322 | - | - | - |
Data are shown as estimated HRs (95% CIs) of the explanatory variables in the 30-day mortality group. The HR is defined as the ratio of the hazard rates corresponding to the conditions described by 2 levels of an explanatory variable (the hazard rate is the risk of death [ie, the probability of death], given that the patient has survived up to a specific time). The P value is based on the null hypothesis that all HRs relating to an explanatory variable equal unity (no effect).
Abbreviations: HR, hazard ratio; ICU, intensive care unit.
aExcluded 187 patients with septic shock, 11 patients with missing data regarding septic shock, 30 patients who had do-not-resuscitate orders, and 27 with missing data regarding a do-not-resuscitate order.
DISCUSSION
In this large cohort of hospitalized patients with CAP and sepsis, when compared with nonlymphopenia patients, we found that lymphopenia was associated with an increased risk of ICU admission, mechanical ventilation, longer length of stay, and in-hospital and 30-day mortality. We further found that 39% of patients with CAP and sepsis had lymphopenia, and such a high percentage is suggestive of lymphopenia being associated with a more severe clinical presentation of CAP.
The observation that lymphopenia was associated with higher 30-day mortality in our study population is in accordance with a prior study that evaluated the usefulness of lymphopenia in predicting the short-term outcome of patients with sepsis [27]. The authors found that lymphopenia was independently associated with increased 28-day mortality and a significantly higher requirement for ICU admission [27]. Prior, Drewry et al. [28] had studied 335 adult patients with bacteremia and sepsis and reported that the median lymphocyte count on day 4 after admission was an independent variable predictor for 28-day mortality. More recently, Wagner et al. [29] found a significant association between lymphocytopenia and disease severity in patients with COVID-19, which was later validated in further studies and meta-analyses [30]. When we used lymphopenia at a cutoff of <1000 lymphocytes/mm3, it was associated with higher in-hospital and 30-day mortality; however, it was not a risk factor for ICU admission [2]. Lymphopenia values <1000 /mm3 are perhaps more accurate when considering the number of lymphocytes for prediction of different outcomes in CAP.
The main hypothesis presented by most investigators to explain the association between lymphopenia and disease severity in CAP is a deregulated immune response to infection, including the activation of different immune cells, secretion of various cytokines, and subsequent activation of cellular apoptosis mechanisms (both intrinsic and extrinsic pathways) so as to cause impaired inflammatory responses [31, 32]. Such a deregulated immune response would favor uncontrolled lymphocyte migration to the lungs and extrapulmonary tissues alongside apoptosis and lead to secondary lymphopenia and its subsequent persistence [33, 34]. These mechanisms might therefore place patients in a state of immunosuppression, leading to an increased risk of further severity and higher mortality. However, an accurate understanding of the mechanisms underlying lymphopenia is still lacking. The massive migration of lymphocytes to the lung, the adhesion to the vascular endothelium, the impaired production in the bone marrow, and an increase in apoptosis pathways during the acute phase of pneumonia may contribute to lymphopenia [8]. Immunosenescence—comprising a set of changes occurring in peripheral T lymphocytes—and the presence of chronic comorbidities may induce chronic endothelial dysfunction that could help explain lymphopenia in an elderly population with severe COVID-19 [35–37].
In addition, elevated levels of C-reactive protein, pleural effusion, and ARDS were associated with risk of ICU admission. This is in accordance with previous studies that reported an association between these variables, treatment failure, and increased need for critical care in patients with CAP [38–40]. In our study, age ≥80 years was associated with a lower risk of ICU admission, thus reflecting the controversial and clinically difficult decision-making process held when considering ICU admission in elderly patients. Frailty, larger burdens of comorbidities, and relatively scant studies evaluating the prognosis of very old patients with CAP admitted to the ICU are some of the major underlying causes of such complex clinical scenarios [41–43].
On the other hand, pneumonia due to atypical pathogens mostly occurs in young patients with few comorbidities and has milder clinical presentation. The fact that in our study atypical pathogens were more frequently detected in nonlymphopenic patients, who indeed were younger and had fewer comorbidities, confirms prior observations [44, 45]. Inversely, we observed that L. pneumophila was most commonly detected in lymphopenic patients, which is consistent with Legionella usually causing rapidly progressive and severe forms of pneumonia [46, 47].
Age ≥80 years, chronic renal disease, chronic neurologic disease, being a nursing home resident, and pleural effusion were independently associated with increased 30-day mortality in the overall cohort. Meanwhile, in the subgroup of patients requiring ICU admission, age ≥80 years, chronic respiratory disease, chronic liver disease, lymphopenia, and acute renal failure were associated with 30-day mortality. Most of these factors except lymphopenia are already well-known risk factors of mortality and severity in patients with CAP and sepsis; for example, the PSI score includes many of them. Furthermore, advanced age, residing in a nursing home, and chronic renal and neurologic diseases favor the development of delirium in sepsis, which is associated with a prolonged hospital length of stay and increased mortality [48, 49]. However, the role of lymphopenia as a risk factor for mortality in CAP with sepsis is a novel finding. Additionally, we found that previous pneumococcal vaccination, diabetes mellitus, and fever were associated with a lower risk of 30-day mortality. The impact of diabetes mellitus on the outcomes of patients with sepsis remains controversial. In a study carried out in the Netherlands, no significant differences were found in short- or long-term outcomes, inflammatory biomarker levels, coagulation factor, or endothelial activation when comparing mortality in 241 diabetic and 863 nondiabetic patients with sepsis [50]. Another large study including 1.5 million critically ill patients suggested that diabetes may have a protective effect [51]. The reason could be related to a greater tolerance of sustained levels of moderate hyperglycemia, as well as better adaptability to marked fluctuations in blood glucose levels in patients with diabetes. On the other hand, with an increased likelihood of multiorgan dysfunction, patients without diabetes may be at a disadvantage due to a compromised immune response and altered microvasculature [52]. With respect to vaccinations, S. pneumoniae is well known to be the main bacterial pathogen in patients with CAP and sepsis. This, therefore, highlights the importance of vaccinations in pneumonia and sepsis prevention.
Our study has some limitations, though, beginning with its retrospective nature. However, the data of all patients included were collected consecutively and prospectively per our study protocol in CAP. Second, the study was carried out in a single-center teaching hospital in Spain. The identified cutoff for lymphopenia validated in our center should be confirmed and validated in future studies to increase external validity.
In conclusion, patients with CAP and sepsis presenting with lymphopenia have higher rates of ICU admission and mortality. It is, therefore, critical to identify lymphopenia in hospitalized patients with CAP and sepsis, as lymphopenia might serve to prioritize patients with CAP and sepsis who require or will shortly require critical care. Moreover, early identification of lymphopenia could impact treatment optimization and the need for complementary treatments with immune modulators and drug-inducing expansion of lymphocyte counts [53, 54]. Lastly, including lymphocyte count in bundles of care for patients with CAP might be an appropriate way of improving management of such patients.
Supplementary Data
Supplementary materials are available at Open Forum Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Acknowledgments
We are indebted to all participating medical and nursing colleagues for their assistance and cooperation in this study. Thank you to Anthony Armenta for providing medical editing assistance for the manuscript at hand.
Financial support. This study was supported by CIBER de Enfermedades Respiratorias (CIBERES CB06/06/0028) and by 2009 Support to Research Groups of Catalonia 911, IDIBAPS. Dr Cillóniz is the recipient of the SEPAR fellowship 2018 and a grant from the Fondo de Investigación Sanitaria (PI19/00207). The sponsor had no role in the design of the study, collection and analysis of the data, or preparation of the manuscript.
Potential conflicts of interest. The authors declare that they have no conflicts of interest. All authors: no reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References
- 1. Shankar-Hari M, Fear D, Lavender P, et al. Activation-associated accelerated apoptosis of memory B cells in critically Ill patients with sepsis. Crit Care Med 2017; 45:875–82. [DOI] [PubMed] [Google Scholar]
- 2. Bermejo-Martin JF, Cilloniz C, Mendez R, et al. ; NEUMONAC group. Lymphopenic community acquired pneumonia (L-CAP), an immunological phenotype associated with higher risk of mortality. EBioMedicine 2017; 24:231–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Rhodes A, Evans LE, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016. Intensive Care Med 2017; 43:304–77. [DOI] [PubMed] [Google Scholar]
- 4. Fantin B, Joly V, Elbim C, et al. Lymphocyte subset counts during the course of community-acquired pneumonia: evolution according to age, human immunodeficiency virus status, and etiologic microorganisms. Clin Infect Dis 1996; 22:1096–8. [DOI] [PubMed] [Google Scholar]
- 5. Schleicher GK, Hopley MJ, Feldman C. CD4 T-lymphocyte subset counts in HIV-seropositive patients during the course of community-acquired pneumonia caused by Streptococcus pneumoniae. Clin Microbiol Infect 2004; 10:587–9. [DOI] [PubMed] [Google Scholar]
- 6. Méndez R, Menéndez R, Amara-Elori I, et al. Lymphopenic community-acquired pneumonia is associated with a dysregulated immune response and increased severity and mortality. J Infect 2019; 78:423–31. [DOI] [PubMed] [Google Scholar]
- 7. Bellelli V, d’Ettorre G, Celani L, et al. Clinical significance of lymphocytopenia in patients hospitalized with pneumonia caused by influenza virus. Crit Care 2019; 23:330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jafarzadeh A, Jafarzadeh S, Nozari P, et al. Lymphopenia an important immunological abnormality in patients with COVID-19: possible mechanisms. Scand J Immunol 2021; 93:e12967. [DOI] [PubMed] [Google Scholar]
- 9. Tan L, Wang Q, Zhang D, et al. Lymphopenia predicts disease severity of COVID-19: a descriptive and predictive study. Signal Transduct Target Ther 2020; 5:33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hamilton F, Arnold D, Payne R. Association of prior lymphopenia with mortality in pneumonia: a cohort study in UK primary care. Br J Gen Pract 2021; 71:e148–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Warny M, Helby J, Nordestgaard BG, et al. Lymphopenia and risk of infection and infection-related death in 98 344 individuals from a prospective Danish population-based study. PLoS Med 2018; 15:e1002685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Marrie TJ, Wu L. Factors influencing in-hospital mortality in community-acquired pneumonia: a prospective study of patients not initially admitted to the ICU. Chest 2005; 127:1260–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Polverino E, Torres A, Menendez R, et al. ; HCAP Study investigators. Microbial aetiology of healthcare associated pneumonia in Spain: a prospective, multicentre, case-control study. Thorax 2013; 68:1007–14. [DOI] [PubMed] [Google Scholar]
- 14. Mandell LA, Wunderink RG, Anzueto A, et al. ; Infectious Diseases Society of America; American Thoracic Society. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis 2007; 44(Suppl 2):S27–72. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Torres A, Barberán J, Falguera M, et al. Multidisciplinary guidelines for the management of community-acquired pneumonia [in Spanish]. Med Clin (Barc) 2013; 140:223.e1–19. [DOI] [PubMed] [Google Scholar]
- 16. Singer M, Deutschman CS, Seymour CW, et al. The Third International consensus definitions for sepsis and septic shock (Sepsis-3). JAMA 2016; 315:801–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Ranieri VM, Rubenfeld GD, Thompson BT, et al. ; ARDS Definition Task Force. Acute respiratory distress syndrome: the Berlin definition. JAMA 2012; 307:2526–33. [DOI] [PubMed] [Google Scholar]
- 18. Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med 1996; 22:707–10. [DOI] [PubMed] [Google Scholar]
- 19. Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997; 336:243–50. [DOI] [PubMed] [Google Scholar]
- 20. Bellomo R, Ronco C, Kellum JA, et al. ; Acute Dialysis Quality Initiative workgroup. Acute renal failure—definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004; 8:R204–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Cillóniz C, Ewig S, Polverino E, et al. Microbial aetiology of community-acquired pneumonia and its relation to severity. Thorax 2011; 66:340–6. [DOI] [PubMed] [Google Scholar]
- 22. Hosmer DW, Lemeshow S.. Applied Logistic Regression. New York: John Wiley & Sons, Inc; 1989. [Google Scholar]
- 23. Efron B, Tibshirani RJ.. An Introduction to the Bootstrap. Chapman and Hall/CRC, UK; 1994. [Google Scholar]
- 24. Sterne JA, White IR, Carlin JB, et al. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ 2009; 338:b2393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Valencia M, Badia JR, Cavalcanti M, et al. Pneumonia Severity Index Class V patients with community-acquired pneumonia: characteristics, outcomes, and value of severity scores. Chest 2007; 132:515–22. [DOI] [PubMed] [Google Scholar]
- 26. Ferrer M, Travierso C, Cilloniz C, et al. Severe community-acquired pneumonia: characteristics and prognostic factors in ventilated and non-ventilated patients. PLoS One 2018; 13:e0191721. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Sheikh Motahar Vahedi H, Bagheri A, Jahanshir A, et al. Association of lymphopenia with short term outcomes of sepsis patients; a brief report. Arch Acad Emerg Med 2019; 7:e14. [PMC free article] [PubMed] [Google Scholar]
- 28. Drewry AM, Samra N, Skrupky LP, et al. Persistent lymphopenia after diagnosis of sepsis predicts mortality. Shock 2014; 42:383–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Wagner J, DuPont A, Larson S, et al. Absolute lymphocyte count is a prognostic marker in Covid-19: a retrospective cohort review. Int J Lab Hematol 2020; 42:761–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Huang I, Pranata R. Lymphopenia in severe coronavirus disease-2019 (COVID-19): systematic review and meta-analysis. J Intensive Care 2020; 8:36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Hotchkiss RS, Coopersmith CM, Karl IE. Prevention of lymphocyte apoptosis—a potential treatment of sepsis? Clin Infect Dis 2005; 41(Suppl 7):S465–9. [DOI] [PubMed] [Google Scholar]
- 32. Girardot T, Rimmelé T, Venet F, Monneret G. Apoptosis-induced lymphopenia in sepsis and other severe injuries. Apoptosis 2017; 22:295–305. [DOI] [PubMed] [Google Scholar]
- 33. Marshall JC, Charbonney E, Gonzalez PD. The immune system in critical illness. Clin Chest Med 2008; 29:605–16, vii. [DOI] [PubMed] [Google Scholar]
- 34. Rubio I, Osuchowski MF, Shankar-Hari M, et al. Current gaps in sepsis immunology: new opportunities for translational research. Lancet Infect Dis 2019; 19:e422–36. [DOI] [PubMed] [Google Scholar]
- 35. Linton PJ, Dorshkind K. Age-related changes in lymphocyte development and function. Nat Immunol 2004; 5:133–9. [DOI] [PubMed] [Google Scholar]
- 36. Qin L, Jing X, Qiu Z, et al. Aging of immune system: immune signature from peripheral blood lymphocyte subsets in 1068 healthy adults. Aging (Albany NY) 2016; 8:848–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Bermejo-Martin JF, Martín-Fernandez M, López-Mestanza C, Duque P, Almansa R. Shared features of endothelial dysfunction between sepsis and its preceding risk factors (aging and chronic disease). J Clin Med 2018; 7: 400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Menéndez R, Cavalcanti M, Reyes S, et al. Markers of treatment failure in hospitalised community acquired pneumonia. Thorax 2008; 63:447–52. [DOI] [PubMed] [Google Scholar]
- 39. Thompson BT, Chambers RC, Liu KD. Acute respiratory distress syndrome. N Engl J Med 2017; 377:562–72. [DOI] [PubMed] [Google Scholar]
- 40. Cilloniz C, Ferrer M, Liapikou A, et al. Acute respiratory distress syndrome in mechanically ventilated patients with community-acquired pneumonia. Eur Respir J 2018; 51: 1702215. [DOI] [PubMed] [Google Scholar]
- 41. Flaatten H, de Lange DW, Artigas A, et al. The status of intensive care medicine research and a future agenda for very old patients in the ICU. Intensive Care Med 2017; 43:1319–28. [DOI] [PubMed] [Google Scholar]
- 42. Andersen FH, Flaatten H, Klepstad P, et al. Long-term survival and quality of life after intensive care for patients 80 years of age or older. Ann Intensive Care 2015; 5:53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Cillóniz C, Dominedò C, Pericàs JM, et al. Community-acquired pneumonia in critically ill very old patients: a growing problem. Eur Respir Rev 2020; 29: 190126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Cillóniz C, Polverino E, Ewig S, et al. Impact of age and comorbidity on cause and outcome in community-acquired pneumonia. Chest 2013; 144:999–1007. [DOI] [PubMed] [Google Scholar]
- 45. Jain S, Self WH, Wunderink RG, et al. ; CDC EPIC Study Team. Community-acquired pneumonia requiring hospitalization among U.S. adults. N Engl J Med 2015; 373:415–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Viasus D, Di Yacovo S, Garcia-Vidal C, et al. Community-acquired Legionella pneumophila pneumonia: a single-center experience with 214 hospitalized sporadic cases over 15 years. Medicine (Baltimore) 2013; 92:51–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Cillóniz C, Torres A, Niederman M, et al. Community-acquired pneumonia related to intracellular pathogens. Intensive Care Med 2016; 42:1374–86. [DOI] [PubMed] [Google Scholar]
- 48. Cilloniz C, Dominedò C, Ielpo A, et al. Risk and prognostic factors in very old patients with sepsis secondary to community-acquired pneumonia. J Clin Med 2019; 8:961. [DOI] [PMC free article] [PubMed]
- 49. Brummel NE, Boehm LM, Girard TD, et al. Subsyndromal delirium and institutionalization among patients with critical illness. Am J Crit Care 2017; 26:447–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. van Vught LA, Scicluna BP, Hoogendijk AJ, et al. Association of diabetes and diabetes treatment with the host response in critically ill sepsis patients. Crit Care 2016; 20:252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Graham BB, Keniston A, Gajic O, et al. Diabetes mellitus does not adversely affect outcomes from a critical illness. Crit Care Med 2010; 38:16–24. [DOI] [PubMed] [Google Scholar]
- 52. Krinsley JS. Moving closer to untangling a sweet web: hyperglycemia, diabetic status, and mortality in the critically ill. Crit Care Med 2010; 38:295–6. [DOI] [PubMed] [Google Scholar]
- 53. Shindo Y, Unsinger J, Burnham CA, et al. Interleukin-7 and anti-programmed cell death 1 antibody have differing effects to reverse sepsis-induced immunosuppression. Shock 2015; 43:334–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 54. Shindo Y, McDonough JS, Chang KC, et al. Anti-PD-L1 peptide improves survival in sepsis. J Surg Res 2017; 208:33–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.