Lymphopenia is Associated with Altered Pathogen Distribution and Increased Disease Severity in Viral Pneumonia: A Multicenter Retrospective Cohort Study Authors: Yanrui Wu, Xiaoqi Zhang, Lan Yang, Nan Su, Lijuan Li Published: 2026 ABSTRACT AbstractBackgroundLymphopenia is common in viral pneumonia and has been associated with adverse outcomes. However, its relationship with pathogen distribution, mixed infections, and opportunistic infections across different immune statuses remains incompletely characterized.MethodsWe conducted a multicenter retrospective cohort study including 2363 adult patients hospitalized with viral pneumonia, of whom 690 were immunocompromised. Patients were stratified into four groups according to immune status and lymphocyte count on admission. Clinical characteristics, laboratory parameters, pathogen profiles, and clinical outcomes were compared across groups. Multivariable logistic regression analyses were performed to identify factors independently associated with in-hospital mortality.ResultsPatients with lymphopenia had more severe clinical presentations, higher inflammatory and coagulation markers, and increased severity scores than those without lymphopenia. Across both immune-status strata, patients with lymphopenia had higher observed rates of respiratory failure, invasive mechanical ventilation, ECMO, and in-hospital mortality than those without lymphopenia. Pathogen analyses demonstrated marked differences in pathogen distribution across groups. Immunocompromised patients with lymphopenia showed significantly higher proportions of bacterial co-infections, fungal infections, and mixed infections, particularly involving Aspergillus flavus, Pneumocystis jirovecii, Acinetobacter baumannii, and Pseudomonas aeruginosa. Viral co-infections with human herpesvirus-1 and cytomegalovirus were also more frequent in this group. In multivariable analyses, lymphopenia, advanced age, elevated D-dimer levels, bacterial co-infection, and pre-admission corticosteroid use were independently associated with in-hospital mortality. Overall in-hospital mortality in the cohort was 16.7%, with the highest mortality observed in immunocompromised patients with lymphopenia (27.4%) and the lowest in immunocompetent patients without lymphopenia (7.3%).ConclusionLymphopenia is associated with altered pathogen distribution, increased mixed and opportunistic infections, and worse clinical outcomes in viral pneumonia, independent of underlying immune status. Lymphocyte count represents a readily available clinical marker reflecting immune dysfunction and infection complexity, highlighting the need for heightened vigilance and comprehensive pathogen evaluation in patients with viral pneumonia and lymphopenia.Keywords: bacterial co-infection, immune status, lymphopenia, respiratory failure, viral pneumonia Viral pneumonia remains a leading cause of morbidity and mortality worldwide, with older adults and immunocompromised populations being particularly vulnerable.1,2 Despite advances in diagnostic techniques and therapeutic strategies, clinical outcomes remain highly heterogeneous, reflecting variations in host immune competence and the complex contribution of co-infections. Lymphopenia, defined as a reduced peripheral blood lymphocyte count, is frequently observed in patients with severe viral infections and is widely regarded as a marker of immune dysregulation.3,4 Previous studies have primarily focused on lymphopenia as a prognostic indicator; however, its clinical implications across different immune backgrounds, particularly in relation to infection patterns and mixed infections, remain incompletely understood. Conventional severity scoring systems, such as the Pneumonia Severity Index (PSI) and the Confusion, Urea, Respiratory rate, blood pressure, and age ≥65 years (CURB-65) score, are commonly used for risk stratification in pneumonia. Nevertheless, these tools do not incorporate immunological parameters or explicitly account for co-infections, which may limit their applicability in patients with viral pneumonia.5,6 Furthermore, the presence of bacterial and opportunistic infections, along with immunomodulatory treatments such as corticosteroids, adds additional complexity to disease severity and outcomes.7,8 To address these gaps, we conducted a multicenter retrospective cohort study to systematically examine the interplay between immune status and lymphopenia in patients with viral pneumonia. We aimed to characterize differences in clinical features, pathogen distribution, treatment patterns, and disease severity across immune backgrounds, thereby providing a comprehensive clinical phenotype linking immune dysfunction to infection complexity in viral pneumonia. Mechanistically, virus-induced T-cell apoptosis, cytokine-driven lymphocyte exhaustion, and bone-marrow suppression have been proposed as key contributors to lymphocyte depletion during severe viral infections, providing a pathophysiological basis linking lymphopenia to impaired antiviral responses and adverse outcomes. In parallel, the increasing availability of unbiased molecular diagnostics such as targeted and metagenomic next-generation sequencing (tNGS/mNGS) now enables more comprehensive detection of bacterial, fungal, and viral co-pathogens in respiratory specimens, offering a unique opportunity to characterize pathogen ecology across different immune phenotypes. Methods Study Design This multicenter study (Kunming Municipal Hospital of Traditional Chinese Medicine, Second People’s Hospital of Weifang, China-Japan Friendship Hospital) retrospectively enrolled 2363 patients diagnosed with viral pneumonia at the study institutions between August 2016 and July 2025. Inclusion criteria were: (1) adults aged ≥18 years; (2) hospitalized between August 2016 and July 2025; and (3) a confirmed diagnosis of viral pneumonia based on detection of viral nucleic acids by PCR, tNGS, or mNGS. Exclusion criteria were: (1) age <18 years; (2) hospital stay <24 hours; (3) incomplete clinical or microbiological data essential for the present analyses; (4) duplicate admissions during the study period (only the first admission was retained); and (5) pregnancy. Diagnostic Criteria The patients included in this study met the clinical and radiographic criteria for pneumonia, characterized by respiratory symptoms and new pulmonary infiltrates on imaging. Diagnosis of viral pneumonia includes the detection of viral nucleic acids by polymerase chain reaction (PCR), targeted next-generation sequencing (tNGS), or metagenomic next-generation sequencing (mNGS).9 Bacterial pneumonia was defined based on positive culture or nucleic acid evidence obtained through tNGS, or mNGS, along with compatible clinical and imaging features, consistent with established pneumonia diagnostic guidelines (Infectious Diseases Society of America/American Thoracic Society).10,11 Invasive fungal diseases were diagnosed according to the 2020 European Organisation For Research And Treatment Of Cancer/Mycoses Study Group Education and Research Consortium criteria using clinical imaging, serum and bronchoalveolar lavage fluid (BALF) galactomannan testing, and molecular diagnostics, including high-throughput sequencing, aimed at a more comprehensive assessment of invasive fungal infections.12 Immunocompromised patients were defined based on the following criteria: (1) solid-organ, stem-cell, or bone marrow transplant recipients; (2) patients undergoing chemotherapy for hematological malignancies (including acute lymphocytic leukemia, acute myeloid leukemia, chronic lymphocytic leukemia, myeloma, or lymphoma) or diagnosed with a solid tumor within 6 months before admission, or severe neutropenia (absolute neutrophil count <500 cells/mm3); (3) patients receiving chest radiation therapy within 3 months before admission; or (4) patients with autoimmune diseases (systemic lupus erythematosus, rheumatoid arthritis, polymyalgia rheumatica, interstitial lung disease, or others) treated with immunosuppressive therapies including chronic glucocorticoids (prednisone >10 mg/day for >3 weeks), methotrexate (>12.5 mg/week), cyclosporine, azathioprine, or biologics (etanercept, infliximab) within 3 months before admission.13,14 Patients with chronic conditions that impair immunity, such as liver cirrhosis, diabetes mellitus, or chronic renal failure, were included in the cohort but were not classified as immunocompromised solely on that basis. Data Collection Relevant clinical data were retrieved from the medical records, including: (1) demographics; (2) clinical presentation; (3) initial vital signs and lung examination; (4) disease severity (requirement for invasive or non-invasive mechanical ventilation, PSI score, and/or CURB-65 score5,6); (5) laboratory and microbiological findings (blood, sputum, and BALF analyses; bacterial and fungal cultures; viral nucleic acid detection; and antibiotic susceptibility profiles); (6) treatment details (administration of vasoactive agents, antimicrobials, glucocorticoids, or other immunosuppressive therapies); and (7) outcomes (survival status and in-hospital mortality rate). Microbiology Viral etiology was confirmed when respiratory viruses, including severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), cytomegalovirus (CMV), respiratory syncytial virus, influenza A and B, parainfluenza virus, rhinovirus, metapneumovirus, and adenovirus, were detected in sputum, endotracheal aspirate, BALF, or nasopharyngeal swabs using reverse transcription PCR. Serum and BALF galactomannan were measured by enzyme immunoassay. Bacterial and fungal pathogens in sputum and BALF were assessed by conventional culture. Respiratory specimens were inoculated and incubated according to standard laboratory procedures, and bacterial and fungal isolates were identified by the clinical microbiology laboratory using routine identification methods, including matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Antimicrobial susceptibility testing was performed and interpreted according to the current Clinical and Laboratory Standards Institute (CLSI) guidelines. Filamentous fungi were identified based on macroscopic and microscopic morphology, with internal transcribed spacer (ITS) region sequencing performed where required. Targeted next-generation sequencing (tNGS) and metagenomic next-generation sequencing (mNGS) were performed using validated commercial platforms to enable comprehensive pathogen detection. Pneumocystis jirovecii was detected by PCR and/or mNGS performed on BALF or induced sputum. Statistical Analysis Patient demographics, clinical characteristics, laboratory findings, and pathogen testing results were summarized as means with standard deviations, medians with interquartile ranges, or frequencies with percentages, as appropriate. Comparisons between groups were performed using the Student’s t test or Mann–Whitney U-test for continuous variables, depending on data distribution, and the χ2-test or Fisher’s exact test for categorical variables. Missing data were handled using complete-case analysis. Variables with substantial missingness were excluded from multivariable analyses to reduce potential bias. The association between peripheral blood lymphocyte count and in-hospital mortality was explored using receiver operating characteristic (ROC) curve analysis, and the optimal cut-off value was determined by maximizing the Youden index. Patients were subsequently classified into lymphopenia and non-lymphopenia groups based on this threshold. Univariate logistic regression analyses were first conducted to identify factors associated with in-hospital mortality. Variables with clinical relevance or a P value <0.10 in univariate analyses were entered simultaneously into multivariable logistic regression models (enter method) to identify independent associations; established comorbidities were additionally included as adjustment covariates. Of the 2363 patients, 2262 had complete data on all model covariates and were included in the multivariable models; the remaining patients were excluded owing to missing covariate data under complete-case analysis. Subgroup analyses were performed according to immune status to assess the consistency of associations across different immune backgrounds. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). All statistical analyses were performed using IBM SPSS Statistics for Windows, version 26.0 (IBM Corp., Armonk, NY, USA). All tests were two-sided, and a P value <0.05 was considered statistically significant. Results Patient Grouping and Baseline Characteristics This study included 2363 patients with viral pneumonia, of whom 690 were immunocompromised. Based on immune status and lymphocyte count, we classified the patients into four groups: immunocompromised with lymphopenia (n=318), immunocompromised without lymphopenia (n=372), immunocompetent with lymphopenia (n=730), and immunocompetent without lymphopenia (n=943). Based on ROC curve analysis, the optimal lymphopenia cut-off values for the immunocompromised and immunocompetent groups were 0.7×109/L and 0.865×109/L, respectively. We observed significant differences in sex and age distributions among the four groups (all P<0.001). Patients with lymphopenia exhibited more severe clinical presentations, including a higher prevalence of dyspnea and disturbance of consciousness (P<0.001 and P=0.010, respectively). Laboratory parameters showed lower lymphocyte and monocyte counts, with elevated neutrophil, lactate dehydrogenase, troponin T, C-reactive protein, and D-dimer levels (all P<0.05) in the lymphopenia groups. Moreover, severity scores, including PSI and CURB-65 >1, were significantly higher in patients with lymphopenia (P<0.001). Medical History and Comorbidities The immunocompromised group showed a higher burden of underlying diseases, including connective tissue diseases, malignancies, hematologic disorders, chronic renal failure, and solid organ transplantation compared with the immunocompetent group (all P<0.001). Severe pneumonia occurred more frequently in patients with lymphopenia (P<0.001). Treatments and Outcomes The use of corticosteroids and immunosuppressants was significantly higher in immunocompromised patients (both P<0.001). Lymphopenia was associated with an increased incidence of respiratory failure, invasive mechanical ventilation, and extracorporeal membrane oxygenation (ECMO) (all P<0.01). In-hospital mortality differed significantly among the groups, with the highest mortality observed in immunocompromised patients with lymphopenia (P<0.001) (Table 1). Table 1Clinical Characteristics of Viral Pneumonia Stratified by Immune Status and Lymphocyte CountVariablesTotal, N=2363Immunocompromised, Lymphopenia# Group, n=318Immunocompromised, Non-Lymphopenia Group, n=372Immunocompetent, Lymphopenia# Group, n=730Immunocompetent, Non-Lymphopenia Group#, n=943P-valueSex, female, n (%)848 (35.9)123 (38.7)126 (33.9)216 (29.6)383 (40.6)<0.001Age, median (IQR)68.0 (58.0, 78.0)64.0 (56.0, 70.0)63.0 (53.0, 70.0)73.0 (63.0, 82.0)68.0 (58.0, 78.0)<0.001Symptoms and signs, n (%) Fever1904 (80.6)258 (81.1)277 (74.5)625 (85.6)744 (78.9)<0.001 Cough2077 (87.9)277 (87.1)331 (89.0)638 (87.4)831 (88.1)0.847 Expectoration2027 (85.8)276 (86.8)329 (88.4)613 (84.0)809 (85.8)0.223 Dyspnea1450 (61.4)212 (66.7)219 (58.9)518 (71.0)501 (53.1)<0.001 Consciousness disturbance177 (7.5)20 (6.3)18 (4.8)73 (10.0)66 (7.0)0.010Laboratory examination, median (IQR) White blood cell, ×109/L7.22 (5.16, 10.25)6.36 (4.20, 9.59)7.87 (5.56, 11.43)7.28 (4.55, 10.48)7.20 (5.63, 9.70)0.001 Neutrophils, ×109/L5.52 (3.55, 8.57)5.59 (3.48, 8.86)5.75 (3.60, 8.98)6.29 (3.71, 9.58)5.01 (3.52, 7.60)<0.001 Lymphocyte, ×109/L0.91 (0.55, 1.40)0.43 (0.29, 0.58)1.21 (0.93, 1.71)0.55 (0.38, 0.69)1.36 (1.08, 1.75)<0.001 Monocyte, ×109/L0.40 (0.26, 0.58)0.28 (0.17, 0.41)0.51 (0.37, 0.71)0.33 (0.18, 0.49)0.48 (0.34, 0.64)<0.001 Platelet, ×109/L200.0 (143.0, 268.0)162.0 (118.0, 217.0)192.0 (139.0, 253.5)177.0 (127.0, 235.5)233.5 (177.0, 308.0)<0.001 Hemoglobin, g/L122.0 (106.0, 136.0)109.0 (88.0, 127.0)115.0 (100.0, 132.0)123.0 (108.0, 137.0)126.0 (113.0, 138.0)<0.001 Lactate dehydrogenase (LDH), U/L264.0 (200.0, 384.0)340.0 (254.0, 479.0)271.0 (206.0, 380.0)301.0 (228.0, 441.0)222.0 (181.3, 306.8)<0.001 Troponin T, ng/mL0.015 (0.009, 0.031)0.016 (0.010, 0.040)0.012 (0.008, 0.026)0.019 (0.010, 0.043)0.012 (0.007, 0.023)<0.001 C-reactive protein (CRP), mg/L40.29 (12.00, 95.17)58.4 (26.88, 109.07)30.25 (7.18, 68.40)71.63 (28.80, 131.1)20.64 (4.25, 64.93)<0.001 D-Dimer, mg/L1.20 (0.57, 2.64)1.44 (0.77, 2.92)1.07 (0.52, 2.32)1.65 (0.82, 4.12)0.90 (0.44, 1.94)<0.001 Procalcitonin, ng/mL0.20 (0.10, 0.35)0.20 (0.10, 0.55)0.20 (0.10, 0.31)0.20 (0.10, 0.66)0.13 (0.10, 0.25)0.013Pneumonia Severity Index score82.0 (64.0, 105.0)95.0 (70.0, 116.3)76.5 (60.0, 99.0)96.0 (75.0, 116.0)74.0 (59.0, 92.0)<0.001CURB65 score>1894 (37.8)126 (39.6)108 (29.0)394 (54.0)266 (28.2)<0.001Underlying Diseases, n (%) Hypertension1075 (45.5)140 (44.0)141 (37.9)387 (53.0)407 (43.2)<0.001 Coronary Heart Disease473 (20.0)44 (13.8)49 (13.2)178 (24.4)202 (21.4)<0.001 Chronic heart failure176 (7.5)20 (6.3)17 (4.6)79 (10.8)60 (6.4)<0.001 Cerebrovascular disease315 (13.3)27 (8.5)35 (9.4)122 (16.7)131 (13.9)<0.001 Diabetes mellitus778 (32.9)120 (37.7)106 (28.5)268 (36.7)284 (30.1)0.002 Tumor201 (8.5)60 (18.9)52 (14.0)37 (5.1)52 (5.5)<0.001 Anemia573 (24.2)117 (36.8)88 (23.7)201 (27.5)167 (17.7)<0.001 Connective tissue disease*240 (10.2)99 (31.1)112 (30.1)9 (1.2)20 (2.1)<0.001 Interstitial lung disease368 (15.6)78 (24.5)146 (39.2)53 (7.3)91 (9.7)<0.001 Bronchiectasis106 (4.5)6 (1.9)21 (5.6)18 (2.5)61 (6.5)<0.001 Bronchial asthma77 (3.3)7 (2.2)11 (3.0)20 (2.7)39 (4.1)0.245 Chronic obstructive pulmonary disease248 (10.5)20 (6.3)28 (7.5)87 (11.9)113 (12.0)0.004 Leukemia22 (0.9)7 (2.2)15 (4.0)0 (0)0 (0)- Lymphoma69 (2.9)34 (10.7)35 (9.4)0 (0)0 (0)- Nephritic syndrome28 (1.2)9 (2.8)11 (3.0)5 (0.7)3 (0.3)0.001 Chronic renal failure223 (9.4)64 (20.1)41 (11.0)79 (10.8)39 (4.1)<0.001 Solid organ transplant176 (7.5)89 (28.0)87 (23.4)0 (0)0 (0)- Liver cirrhosis19 (0.8)8 (2.5)11 (3.0)0 (0)0 (0)- Ex-smoker or Current smoker773 (32.7)107 (33.6)139 (37.4)250 (34.2)277 (29.4)0.025Disease severity Severe pneumonia1256 (53.2)202 (63.5)161 (43.3)525 (71.9)368 (39.0)<0.001Treatment, n (%) Glucocorticoids645 (27.3)244 (76.7)287 (77.2)59 (8.1)55 (5.8)<0.001 Immunosuppressants442 (18.7)202 (63.5)240 (64.5)0 (0)0 (0)<0.001Complications, n (%) Noninvasive ventilation494 (20.9)97 (30.5)60 (16.1)207 (28.4)130 (13.8)<0.001 Invasive mechanical ventilation423 (17.9)79 (24.8)64 (17.2)193 (26.4)87 (9.2)<0.001 Respiratory failure during admission1194 (50.5)202 (63.5)161 (43.3)492 (67.4)339 (35.9)<0.001 Extracorporeal membrane oxygenation87 (3.7)14 (4.4)14 (3.8)40 (5.5)19 (2.0)0.002 Hospital mortality395 (16.7)87 (27.4)57 (15.3)182 (24.9)69 (7.3)<0.001Notes: *Connective tissue disorders: rheumatoid arthritis, systemic lupus erythematosus, dermatomyositis, polymyositis, systemic sclerosis, Sjogren’s syndrome, etc. #Lymphopenia was defined using immune status–specific cut-off values determined by ROC analysis: 0.7×109/L for immunocompromised patients and 0.865 ×109/L for immunocompetent patients. Clinical Characteristics of Viral Pneumonia Stratified by Immune Status and Lymphocyte Count Notes: *Connective tissue disorders: rheumatoid arthritis, systemic lupus erythematosus, dermatomyositis, polymyositis, systemic sclerosis, Sjogren’s syndrome, etc. #Lymphopenia was defined using immune status–specific cut-off values determined by ROC analysis: 0.7×109/L for immunocompromised patients and 0.865 ×109/L for immunocompetent patients. Pathogen Distribution Among the 2363 patients, those who were immunocompromised with lymphopenia exhibited significantly higher proportions of bacterial co-infections (39.9%) and Aspergillus co-infections (33.6%) (all P<0.001). The rates of mixed bacterial, fungal, and viral infections were also highest in the immunocompromised lymphopenia group (21.4%, P<0.001). Additionally, the proportion of patients co-infected with human herpesvirus-1 or CMV was significantly higher among immunocompromised patients (23.0%, P<0.001). SARS-CoV-2 infection was most prevalent in the immunocompetent lymphopenia group (64.1%, P<0.001), whereas respiratory syncytial virus (RSV) was more prevalent in the immunocompromised non-lymphopenia group (14.0%, P<0.001). Among bacterial species, Pseudomonas spp., Acinetobacter baumannii, and Enterobacter cloacae were observed significantly more frequently in immunocompromised patients with lymphopenia (all P<0.01). Pneumocystis was most frequent in the immunocompromised lymphopenia group (7.2%; overall P<0.001), whereas Aspergillus fumigatus was most frequent in the immunocompromised non-lymphopenia group (15.6%; overall P=0.003). (Table 2 and Figure 1). Table 2The Co-Pathogens of Viral Pneumonias Between Several GroupsVariables, n (%)Total Patients, n=2363Immunocompromised, Lymphopenia Group, n=318Immunocompromised, Non-Lymphopenia Group, n=372Immunocompetent, Lymphopenia Group, n=730Immunocompetent, Non-Lymphopenia Group, n=943P-valueNon-Survivor Group, n=395Survivor Group, n=1968P-valueCo-infection with bacterial672 (28.4)127 (39.9)128 (34.4)218 (29.9)199 (21.1)<0.001197 (49.9)475 (24.1)<0.001Co-infection with Aspergillus548 (23.2)107 (33.6)113 (30.4)148 (20.3)180 (19.1)<0.001125 (31.6)423 (21.5)<0.001Co-infection with Mucorales33 (1.4)5 (1.6)4 (1.1)10 (1.4)14 (1.5)0.9398 (2.0)25 (1.3)0.243Mixed bacterial, fungal, and viral infection308 (13.0)68 (21.4)73 (19.6)82 (11.2)85 (9.0)<0.00189 (22.5)219 (11.1)<0.001Co-infection with Human herpesvirus-1 or Cytomegalovirus233 (9.9)73 (23.0)74 (19.9)31 (4.2)55 (5.8)<0.00166 (16.7)167 (8.5)<0.001Two or more respiratory viruses130 (5.5)26 (8.2)31 (8.3)30 (4.1)43 (4.6)0.00226 (6.6)104 (5.3)0.302Virus SARS-CoV-21319 (55.8)185 (58.2)164 (44.1)468 (64.1)502 (53.2)<0.001221 (55.9)1098 (55.8)0.954 Human Rhinovirus98 (4.1)15 (4.7)14 (3.8)20 (2.7)49 (5.2)0.08412 (3.0)86 (4.4)0.226 Influenza A virus561 (23.7)73 (23.0)93 (25.0)172 (23.6)223 (23.6)0.929107 (27.1)454 (23.1)0.087 Influenza B virus98 (4.1)14 (4.4)20 (5.4)24 (3.3)40 (4.2)0.41611 (2.8)87 (4.4)0.137 Parainfluenza virus108 (4.6)22 (6.9)27 (7.3)17 (2.3)42 (4.5)<0.00119 (4.8)89 (4.5)0.803 Adenovirus70 (3.0)4 (1.3)15 (4.0)17 (2.3)34 (3.6)0.07010 (2.5)60 (3.0)0.580 Coronavirus61 (2.6)13 (4.1)11 (3.0)11 (1.5)26 (2.8)0.0889 (2.3)52 (2.6)0.677 Respiratory syncytial virus 157 (6.6)15 (4.7)52 (14.0)28 (3.8)62 (6.6)<0.00128 (7.1)129 (6.6)0.698 Human herpesvirus-154 (2.3)11 (3.5)11 (3.0)11 (1.5)21 (2.2)0.19414 (3.5)40 (2.0)0.067 Cytomegalovirus208 (8.8)66 (20.8)68 (18.3)28 (3.8)46 (4.9)<0.00160 (15.2)148 (7.5)<0.001 Epstein-Barr virus106 (4.5)28 (8.8)18 (4.8)23 (3.2)37 (3.9)0.00123 (5.8)83 (4.2)0.160 Human herpesvirus 6A or 714 (0.6)2 (0.6)2 (0.5)3 (0.4)7 (0.7)0.8513 (0.8)11 (0.6)0.636 Human metapneumovirus26 (1.1)5 (1.6)5 (1.3)0 (0)16 (1.7)0.0073 (0.8)23 (1.2)0.477Bacteria Streptococcus pneumoniae48 (2.0)10 (3.1)9 (2.4)9 (1.2)20 (2.1)0.2006 (1.5)42 (2.1)0.429 Haemophilus influenzae38 (1.6)3 (0.9)5 (1.3)8 (1.1)22 (2.3)0.1455 (1.3)33 (1.7)0.553 Staphylococcus aureus78 (3.3)12 (3.8)18 (4.8)24 (3.3)24 (2.5)0.19812 (3.0)66 (3.4)0.749 Escherichia coli24 (1.0)9 (2.8)7 (1.9)2 (0.3)6 (0.6)<0.0014 (1.0)20 (1.0)0.995 Enterobacter cloacae27 (1.1)10 (3.1)5 (1.3)6 (0.8)6 (0.6)0.0039 (2.3)18 (0.9)0.021 Klebsiella pneumoniae138 (5.8)23 (7.2)32 (8.6)41 (5.6)42 (4.5)0.02245 (11.4)93 (4.7)<0.001 Pseudomonas aeruginosa191 (8.1)43 (13.5)40 (10.8)50 (6.8)58 (6.2)<0.00159 (14.9)132 (6.7)<0.001 Acinetobacterbaumannii217 (9.2)44 (13.8)26 (7.0)97 (13.3)50 (5.3)<0.00193 (23.5)124 (6.3)<0.001 Burkholderia41 (1.7)4 (1.3)7 (1.9)20 (2.7)10 (1.1)0.06323 (5.8)18 (0.9)<0.001 Stenotrophomonas maltophilia64 (2.7)10 (3.1)10 (2.7)25 (3.4)19 (2.0)0.33821 (5.3)43 (2.2)<0.001 Enterococcus spp.84 (3.6)23 (7.2)16 (4.3)23 (3.2)22 (2.3)0.00125 (6.3)59 (3.0)0.001 Nocardia spp.11 (0.5)3 (0.9)1 (0.3)3 (0.4)4 (0.4)0.5811 (0.3)10 (0.5)0.497 Serratia marcescens11 (0.5)2 (0.6)1 (0.3)6 (0.8)2 (0.2)0.2845 (1.3)6 (0.3)0.010 Corynebacterium striatum91 (3.9)19 (6.0)12 (3.2)35 (4.8)25 (2.7)0.02132 (8.1)59 (3.0)<0.001 Moraxella catarrhalis4 (0.2)0 (0)1 (0.3)2 (0.3)1 (0.1)0.6931 (0.3)3 (0.2)0.657 Mycobacterium tuberculosis23 (1.0)6 (1.9)3 (0.8)4 (0.5)10 (1.1)0.2302 (0.5)21 (1.1)0.300 Nontuberculosis mycobacteria11 (0.5)1 (0.3)4 (1.1)1 (0.1)5 (0.5)0.1772 (0.5)9 (0.5)0.896Fungus Pneumocystis jirovecii53 (2.2)23 (7.2)16 (4.3)6 (0.8)8 (0.8)<0.00118 (4.6)35 (1.8)0.001 Aspergillus fumigatus271 (11.5)47 (14.8)58 (15.6)74 (10.1)92 (9.8)0.00353 (13.4)218 (11.1)0.183 Aspergillus niger46 (1.9)9 (2.8)10 (2.7)11 (1.5)16 (1.7)0.3319 (2.3)37 (1.9)0.601 Aspergillus flavus113 (4.8)26 (8.2)21 (5.6)33 (4.5)33 (3.5)0.00718 (4.6)95 (4.8)0.818 Aspergillus terreus15 (0.6)2 (0.6)3 (0.8)5 (0.7)5 (0.5)0.9476 (1.5)9 (0.5)0.015 Aspergillus nidulans14 (0.6)3 (0.9)2 (0.5)4 (0.5)5 (0.5)0.8572 (0.5)12 (0.6)0.807 Aspergillus oryzae14 (0.6)4 (1.3)4 (1.1)3 (0.4)3 (0.3)0.1401 (0.3)13 (0.7)0.336 Rhizopus oryzae11 (0.5)0 (0)2 (0.5)5 (0.7)4 (0.4)0.5082 (0.5)9 (0.5)0.896 Rhizopus microsporus7 (0.3)1 (0.3)2 (0.5)1 (0.1)3 (0.3)0.7102 (0.5)5 (0.3)0.400 Rhizomucor pusillus13 (0.6)2 (0.6)1 (0.3)5 (0.7)5 (0.5)0.8445 (1.3)8 (0.4)0.035 Cryptococcus5 (0.2)0 (0)4 (1.1)1 (0.1)0 (0)-2 (0.5)3 (0.2)0.162Chlamydia2 (0.1)0 (0)0 (0)1 (0.1)1 (0.1)-1 (0.3)1 (0.1)0.207Mycoplasma44 (1.9)2 (0.6)6 (1.6)13 (1.8)23 (2.4)0.2119 (2.3)35 (1.8)0.502Legionella9 (0.4)3 (0.9)3 (0.8)2 (0.3)1 (0.1)0.0891 (0.3)8 (0.4)0.652 The Co-Pathogens of Viral Pneumonias Between Several Groups Key co-infections and mixed infections according to immune status, lymphopenia, and clinical outcome. (A) shows the distribution of selected co-infections and mixed infections across patient groups stratified by immune status and lymphocyte count. (B) shows the distribution of the same infections in non-survivors and survivors. Data are presented as percentages. The predominant bacteria included Acinetobacter baumannii (23.5% vs 6.3%, P<0.001), Pseudomonas aeruginosa (14.9% vs 6.7%, P<0.001), Klebsiella pneumoniae (11.4% vs 4.7%, P<0.001), and Corynebacterium striatum (8.1% vs 3.0%, P<0.001), these were more prevalent in non-survivors (P<0.05). Enterobacter cloacae and Enterococcus spp. were also significantly elevated in non-survivors (all P<0.05). Pneumocystis jirovecii infection was concentrated in the immunocompromised lymphopenia group (7.2%, P<0.001). This group also showed significantly increased proportions of invasive Aspergillus infections (33.6% vs 19.1%, P<0.001). Rhizomucor pusillus infection occurred slightly more frequently in non-survivors than in survivors (1.3% vs 0.4%, P=0.035). Co-infection with human herpesvirus-1 and CMV was also more common in this group (23.0%, P<0.001) (Table 2 and Figure 1). Prognostic Factors and Multivariable Analysis Multivariable logistic regression analysis identified age, lymphopenia, elevated D-dimer level (>1.765 mg/L), pre-admission corticosteroid use, and bacterial co-infection as independent predictors of in-hospital mortality in the overall cohort. In the immunocompromised subgroup, the significant predictors were age, lymphopenia, elevated D-dimer levels, Acinetobacter baumannii infection, Pneumocystis pneumonia, and chronic heart failure. Among immunocompetent patients, age, lymphopenia, elevated D-dimer levels, bacterial co-infection, and corticosteroid use before admission were independently associated with increased risk. Among immunocompetent patients without lymphopenia, elevated D-dimer levels showed the strongest association with mortality, along with Acinetobacter baumannii and Pseudomonas aeruginosa infections, chronic heart failure, and interstitial lung disease (Table 3, Figures 2 and 3). Table 3Logistic Regression Analysis of Prognostic Factors in Viral Pneumonia PatientsVariablesTotal Cohort (N=2262, Deaths=379)Immunocompromised (N=670, Deaths=138)Immunocompetent (N=1592, Deaths=241)Non-Lymphopenic Immunocompetent Group (N=891, Deaths=66)aOR95% CIP valueaOR95% CIP valueaOR95% CIP valueaOR95% CIP valueAge≥65 years old1.6391.238–2.169<0.0011.9201.231–2.9950.0041.4971.040–2.1560.0301.0700.570–2.0080.833Lymphopenia2.5071.935–3.250<0.0011.8051.165–2.7950.0183.0582.190–4.270<0.001---D-Dimer>1.765 mg/L3.4702.692–4.474<0.0013.0691.984–4.747<0.0013.7482.716–5.172<0.0016.3503.486–11.564<0.001Systemic corticosteroid use before admission1.8941.336–2.685<0.0011.2660.712–2.2500.4221.8981.224–3.1910.0161.2170.449–3.3000.699Co-infection with bacterial1.8181.300–2.541<0.0011.5250.876–2.6540.1361.8791.224–2.8850.0041.6220.744–3.5350.224Acinetobacter baumannii infection1.8201.240–2.6720.0022.1701.145–4.1120.0171.6951.031–2.7870.0372.7651.138–6.7210.025Pseudomonas aeruginosa infection1.2680.838–1.9190.2621.4890.777–2.8540.2311.1860.683–2.0600.5452.4641.010–6.0080.047Pneumocystis pneumonia infection2.1331.076–4.2300.0302.4311.094–5.4030.0181.8950.446–8.0540.3871.3180.114–15.2720.825Chronic heart failure1.8091.212–2.7000.0042.6581.218–5.8000.0141.5840.976–2.5700.0634.5821.866–11.249<0.001Female0.6210.471–0.818<0.0010.6770.427–1.0740.0980.6110.429–0.8700.0060.8440.460–1.5470.583Cerebrovascular disease1.4171.008–1.9910.0450.7140.330–1.5440.3921.7761.202–2.6230.0021.9850.928–4.2440.077Interstitial lung disease1.5261.079–2.1580.0171.4360.881–2.3390.1461.6540.999–2.7370.0504.0651.895–8.716<0.001Notes: Of the 2363 patients in the overall cohort, 2262 had complete data on all covariates and were included in the multivariable logistic regression models; the remaining 101 patients were excluded owing to missing covariate data under complete-case analysis.Abbreviations: aOR, adjusted odds ratio; CI, confidence interval. Logistic Regression Analysis of Prognostic Factors in Viral Pneumonia Patients Notes: Of the 2363 patients in the overall cohort, 2262 had complete data on all covariates and were included in the multivariable logistic regression models; the remaining 101 patients were excluded owing to missing covariate data under complete-case analysis. Abbreviations: aOR, adjusted odds ratio; CI, confidence interval. Factors associated with in-hospital mortality in the overall cohort. Forest plot showing adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for factors associated with in-hospital mortality in the multivariable logistic regression model. The vertical dashed line indicates an odds ratio of 1. Bold P values denote statistically significant associations (P < 0.05). Variables associated with lower and higher mortality are shown to the left and right of the reference line, respectively. Subgroup analyses of factors associated with in-hospital mortality. Forest plots showing adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for factors associated with in-hospital mortality in immunocompromised patients, immunocompetent patients, and non-lymphopenic immunocompetent patients. Discussion In this large multicenter cohort of 2363 patients with viral pneumonia, we found that lymphopenia was consistently associated with greater disease severity and higher in-hospital mortality across different immune backgrounds. By stratifying patients according to both immune status and lymphocyte count, our analysis demonstrated that the association between lymphopenia and adverse clinical outcomes was evident in both immunocompromised and immunocompetent individuals. These findings suggest that lymphocyte depletion reflects a clinically relevant state of immune dysregulation during viral pneumonia, rather than serving merely as a surrogate of underlying immunosuppression. Notably, the association between lymphopenia and poor outcomes was evident across both immune-status strata, patients with lymphopenia exhibited a substantially higher burden of bacterial, opportunistic, and mixed infections, including multidrug-resistant organisms and invasive fungal pathogens. Together, these observations highlight peripheral lymphocyte count as a readily available clinical marker that captures the complexity of immune dysfunction and infection patterns in viral pneumonia, especially when interpreted in conjunction with baseline immune status. Accumulating evidence suggests that virus-induced lymphocyte apoptosis and functional exhaustion contribute to lymphocyte depletion during severe viral infections, which may impair antiviral immune responses and is associated with disease progression.15,16 In this context, the association between lymphopenia and adverse clinical outcomes appears to be more pronounced among immunocompromised patients, who also exhibit a higher burden of bacterial and opportunistic infections, including multidrug-resistant organisms and invasive fungal pathogens. These observations are consistent with the concept that impaired immune surveillance may predispose vulnerable hosts to secondary infections, thereby contributing to worse clinical trajectories.17,18 Lymphocytes, particularly CD4⁺ and CD8⁺ T cells, play central roles in antiviral defense, including viral clearance, coordination of adaptive immune responses, and immune regulation. Consequently, lymphopenia may reflect a broader state of immune dysregulation that extends beyond numerical depletion, encompassing functional exhaustion and impaired cellular responses, as demonstrated in prior experimental and clinical studies. Previous investigations, most notably in coronavirus disease 2019 (COVID-19), have linked lymphopenia and T-cell dysfunction with increased disease severity and poor outcomes.19–22 Our findings extend these observations by demonstrating that lymphopenia is associated with adverse outcomes even among immunocompetent patients, suggesting that acute virus-induced immune exhaustion may transiently mirror features of immunosuppression.23 This vulnerability may be further exacerbated in immunocompromised individuals, which is consistent with the higher rates of bacterial and opportunistic fungal infections observed in this group.24,25 Following viral infection, an imbalance between innate and adaptive immune responses may further contribute to immune dysfunction, potentially increasing susceptibility to secondary infections.26 Together, these findings support a conceptual framework in which lymphopenia serves as a clinical marker of immune dysregulation in viral pneumonia, while underscoring the need for future studies to elucidate underlying mechanisms and therapeutic implications. In the present study, pre-admission use of systemic corticosteroids was independently associated with worse clinical outcomes in patients with viral pneumonia. Although corticosteroids may attenuate excessive inflammatory responses in selected cases of severe viral pneumonia or acute respiratory distress syndrome, their immunosuppressive effects may also impair lymphocyte function, delay viral clearance, and increase susceptibility to secondary bacterial and fungal infections, as suggested by prior studies.27 These observations highlight the complex and context-dependent effects of corticosteroids in viral pneumonia, particularly among patients with pre-existing or virus-induced lymphopenia. Emerging evidence indicates that the timing, dose, and duration of corticosteroid exposure may influence clinical outcomes; however, the present study was not designed to evaluate corticosteroid treatment strategies. Further prospective studies are therefore warranted to clarify the optimal use of corticosteroids across different clinical settings and immune backgrounds.28,29 Our findings are consistent with a recent real-world Asian cohort by Rosdiana et al, which independently validated the CURB-65 score, vasopressor use, mechanical ventilation, and corticosteroid exposure as significant predictors of in-hospital mortality among hospitalized community-acquired pneumonia patients,30 thereby supporting the generalizability of the prognostic factors identified in our multivariable analyses beyond the Chinese population. Despite several strengths, including a large sample size and a multicenter design, this study has important limitations. First, the retrospective nature of the study may introduce selection and information biases, particularly due to incomplete data on variables such as corticosteroid dosage, duration, and indications. Second, heterogeneity in diagnostic approaches across participating centers, including variability in next-generation sequencing (NGS) platforms and pathogen detection protocols, may have influenced the sensitivity of microbiological identification. Although NGS was not performed for all patients, routine diagnostic testing for common respiratory viruses, bacterial pathogens, and fungal cultures was conducted according to local clinical practice. In addition, the absence of detailed lymphocyte subset analyses and functional immune assessments limited our ability to explore underlying immunological mechanisms. Consequently, the findings should be interpreted primarily as clinical associations rather than mechanistic evidence. Future prospective studies incorporating standardized pathogen detection strategies and comprehensive immune phenotyping are warranted to further elucidate the role of lymphopenia in viral pneumonia across different immune backgrounds. In addition, although microbiological work-up was harmonized across centers, the specific next-generation sequencing platforms (tNGS/mNGS) and detection protocols varied among the participating hospitals, which may introduce some inter-center heterogeneity in pathogen identification; however, conventional culture-based and serological methods were performed using uniform CLSI-based standards, mitigating this potential bias. Finally, missing data were handled by complete-case analysis; although the proportion of missingness for key clinical and microbiological variables was low in our cohort, we acknowledge that this approach may introduce minor bias for variables with higher missingness (eg, corticosteroid dosing details), and future prospective studies incorporating sensitivity analyses or multiple imputation are warranted. Interpretation This multicenter study demonstrates that lymphopenia is consistently associated with greater disease severity and altered pathogen patterns in patients with viral pneumonia, particularly among those with underlying immune compromise. Assessment of lymphocyte counts in conjunction with immune status may help identify immune dysfunction and vulnerability to mixed bacterial, fungal, and viral infections. These findings highlight the importance of comprehensive pathogen evaluation and cautious use of immunomodulatory therapies in patients with viral pneumonia and lymphopenia, while underscoring the need for prospective studies across different immune backgrounds. Notably, patients with lymphopenia, particularly those who were immunocompromised, more frequently received corticosteroids and immunosuppressants and presented with more severe clinical manifestations, further underscoring the clinical relevance of integrating lymphocyte count with immune status in risk assessment. Future prospective studies incorporating lymphocyte subset analyses (eg, CD4⁺ and CD8⁺ T-cell counts and functional markers) are warranted to further elucidate the immunological mechanisms underlying these associations and to refine personalized risk stratification in viral pneumonia. Acknowledgments We sincerely thank the clinical teams, research nurses, and laboratory personnel from all participating centers for their invaluable support in patient enrollment, data collection, sample processing, and patient management. We also appreciate the assistance provided by the institutional biostatistics and data analysis units. Funding Statement This work was supported by the China–Japan Friendship Hospital Cross-Sectional Study (grant numbers: 2024-ZF-37, 2025-ZF-23, 2025-ZF-38) and the Yunnan Provincial Clinical Talent Training Program of Traditional Chinese Medicine (No. Yun Wei Zhong Yi Zong He Fa [2024] 1). Patient and Public Involvement Patients and the public were not involved in the design, conduct, reporting, or dissemination of this retrospective study. Abbreviations ARDS, Acute respiratory distress syndrome; AUC, Area under the curve; BALF, Bronchoalveolar lavage fluid; CMV, Cytomegalovirus; COVID-19, Coronavirus disease 2019; CURB-65, Confusion, Urea, Respiratory rate, Blood pressure, and age ≥65 years; ECMO, Extracorporeal membrane oxygenation; IPA, Invasive pulmonary aspergillosis; mNGS, Metagenomic next-generation sequencing; MuLBSTA, Multilobular infiltration, hypo-lymphocytosis, bacterial co-infection, smoking history, hypertension and age; NGS, Next-generation sequencing; PCR, Polymerase chain reaction; PSI, Pneumonia Severity Index; ROC, Receiver operating characteristic; RSV, Respiratory syncytial virus; SARS-CoV-2, Severe acute respiratory syndrome coronavirus 2; tNGS, Targeted next-generation sequencing. Data Sharing Statement The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request. Ethics Approval and Consent to Participate This study was conducted in accordance with the Declaration of Helsinki. This study was approved by the Ethics Committees of the study institution (Kunming Municipal Hospital of Traditional Chinese Medicine, Ethics approval number: 2026-010-01; Second People’s Hospital of Weifang, Ethics approval number: KY2025-003-01, China-Japan Friendship Hospital, Ethics approval number: 2025-KY-487). Given the retrospective design of the study, the ethics committees of the participating hospitals waived the requirement for informed consent. Author Contributions Study design: YW and LL. Data collection: YW, XZ and LY. Statistical analyses: NS and LL. Writing: NS and LL. All the authors take full responsibility for the study design, data analysis and interpretation, and manuscript preparation. All authors have read and approved the manuscript, meet the ICMJE requirements for authorship, and believe that the manuscript represents an honest account of this study. All authors contributed to data analysis, drafting or revising the article, have agreed on the journal to which the article will be submitted, gave final approval of the version to be published, and agree to be accountable for all aspects of the work. Disclosure Dr. LJ Li serves as the guarantor of this work and takes full responsibility for the integrity of the study, the accuracy of the data analysis, and the completeness of the reporting. The authors report no conflicts of interest in this work. References Associated Data Data Availability Statement The datasets used and/or analyzed in the current study are available from the corresponding author upon reasonable request.