Pneumonia is an important cause of morbidity and mortality among adults and represents a major burden on healthcare systems worldwide. The clinical presentation of pulmonary infection commonly includes fever, cough, dyspnea, sputum production, chest discomfort, and radiological pulmonary infiltrates.1 However, these clinical and radiological manifestations are not sufficiently specific to reliably distinguish bacterial pneumonia from infections caused by viruses, fungi, or other respiratory pathogens. The distinction between bacterial and non-bacterial pulmonary infections is clinically important because treatment strategies differ substantially. Bacterial pneumonia is generally managed with antibacterial therapy, whereas viral and fungal infections may require antiviral or antifungal treatment, respectively. Inappropriate empirical antibiotic therapy may expose patients to unnecessary adverse effects, contribute to antimicrobial resistance, and increase healthcare costs.2–4 Respiratory microbiological investigations provide an opportunity to improve etiological diagnosis. Conventional investigations such as Gram staining and bacterial culture remain widely used because they are relatively accessible and provide information regarding bacterial identification and antimicrobial susceptibility. However, culture may have limitations related to prior antibiotic exposure, specimen quality, slow-growing organisms, and difficulties in distinguishing colonization from true infection.5,6 Molecular diagnostic techniques have expanded the diagnostic capabilities of respiratory microbiology. Multiplex respiratory PCR assays can detect multiple viral and bacterial targets within a relatively short period.7 These methods may be particularly useful when conventional cultures are negative or when viral infection is suspected. Similarly, fungal microscopy, culture, antigen detection, and molecular methods may contribute to diagnosis in selected patients, particularly those with underlying immunosuppression or radiological features suggestive of fungal disease.8 Despite the increasing availability of diagnostic technologies, microbiological investigations are often interpreted individually rather than as components of an integrated diagnostic pathway. Furthermore, a positive microbiological result does not always indicate causation, as respiratory specimens may contain colonizing organisms. The present study was therefore designed to evaluate the role of respiratory microbiological investigations in differentiating bacterial pneumonia from non-bacterial pulmonary infections and to assess whether combining conventional and molecular investigations improves diagnostic yield.

Study Design and Setting A prospective observational study was designed to evaluate respiratory microbiological investigations among adults presenting with suspected pulmonary infection. The study was conducted over a 12-month period in the departments of Pulmonary Medicine and Microbiology of a tertiary-care teaching hospital in India. The study protocol was approved by the institutional ethics committee before initiation. Written informed consent was obtained from all participants. Study Population Adults aged ≥18 years presenting with symptoms suggestive of lower respiratory tract infection and radiological evidence of a new pulmonary infiltrate were considered eligible. Inclusion Criteria Patients were included if they: • were aged ≥18 years; • had symptoms suggestive of acute pulmonary infection; • had a new infiltrate or consolidation on chest radiography or computed tomography; • had at least one systemic or respiratory feature suggestive of infection; and • were able to provide an appropriate respiratory specimen. Exclusion Criteria Patients were excluded if they: • had isolated non-infectious pulmonary disease; • had received prolonged antimicrobial therapy before respiratory sampling; • had inadequate respiratory specimens; • had pulmonary edema or isolated heart failure without evidence of infection; or • declined participation. Respiratory Specimen Collection Early-morning expectorated sputum was preferred for patients capable of producing adequate specimens. Samples were collected in sterile, leak-proof containers and transported promptly to the microbiology laboratory. For patients unable to produce satisfactory sputum, induced sputum or lower respiratory tract specimens were obtained according to clinical indication. Bronchoalveolar lavage was performed in selected patients when clinically warranted. Specimen quality was assessed using microscopic examination of epithelial cells and inflammatory cells. Samples considered grossly contaminated with oropharyngeal material were classified as inadequate and excluded from culture-based analysis. Microbiological Investigations Gram Staining Direct Gram staining was performed on acceptable respiratory specimens. The presence, morphology, and quantity of bacteria, together with the inflammatory cellular response, were recorded. A result was considered suggestive of bacterial infection when abundant morphologically compatible bacteria were observed in association with significant neutrophilic inflammation and an acceptable specimen. Bacterial Culture Respiratory specimens were cultured on appropriate routine bacterial media under standard laboratory conditions. Organisms were identified using conventional biochemical methods and/or automated identification systems. Potential respiratory pathogens included: • Streptococcus pneumoniae; • Haemophilus influenzae; • Staphylococcus aureus; • Enterobacterales; • Pseudomonas aeruginosa; • other clinically relevant Gram-negative bacilli; and • selected atypical bacterial pathogens when specifically investigated. Antimicrobial susceptibility testing was performed according to contemporary laboratory standards at the study place. Respiratory Viral PCR Respiratory specimens were tested using a multiplex respiratory viral PCR panel. The panel included commonly encountered respiratory viruses such as: • influenza A and B; • respiratory syncytial virus; • adenovirus; • parainfluenza viruses; • human metapneumovirus; and • rhinovirus/enterovirus. SARS-CoV-2 PCR was also available according to the prevailing clinical testing algorithm. Fungal Investigations Fungal investigations were performed in patients with clinical or radiological features suggesting fungal pulmonary disease or in those with relevant immunocompromising conditions. Testing included direct microscopy, fungal culture, and selected antigen/molecular investigations according to clinical indication. Additional Molecular Investigations Targeted molecular testing for selected bacterial pathogens was performed when routine culture was negative despite a high clinical suspicion or when an atypical pathogen was suspected. Diagnostic Classification Patients were classified into diagnostic categories based on an integrated assessment of clinical, radiological, and microbiological findings, with microbiological results interpreted in the overall clinical context. Bacterial Pneumonia Bacterial pneumonia was defined by compatible clinical and radiological findings together with isolation or molecular detection of a clinically relevant bacterial pathogen from an appropriate respiratory specimen, supported by microscopy and/or other clinical evidence when appropriate. Viral Pneumonia Viral pneumonia was defined as compatible clinical and radiological disease with detection of a respiratory viral pathogen and no convincing evidence of bacterial infection. Fungal Pulmonary Infection Fungal infection was classified as probable or proven according to the combination of host factors, radiological findings, microbiological evidence, and clinical response. Mixed Infection Patients demonstrating convincing evidence of more than one clinically relevant pathogen were categorized as having mixed infection. Undetermined etiology Patients with clinical and radiological evidence of pulmonary infection but without an adequately supported microbiological diagnosis were categorized as having undetermined etiology. Outcome Measures The primary outcome was the diagnostic yield of respiratory microbiological investigations in differentiating bacterial pneumonia from non-bacterial pulmonary infections. Secondary outcomes included: 1. diagnostic yield of individual microbiological investigations; 2. distribution of respiratory pathogens; 3. overall etiological diagnostic yield of combined microbiological testing; 4. association between specimen quality and microbiological yield; 5. effect of prior antibiotic exposure on bacterial culture positivity; and 6. changes in antimicrobial therapy following microbiological results. Statistical Analysis Data were analyzed using Statistical Package for the Social Sciences (SPSS) version 25.0 (IBM, USA). Continuous variables were summarized using mean ± standard deviation or median with interquartile range, as appropriate. Categorical variables were presented as frequencies and percentages. Diagnostic yield was calculated as the proportion of patients with a positive or diagnostic microbiological result among those evaluated for the respective investigation. The chi-square test or Fisher's exact test was used for categorical comparisons. Continuous variables were compared using the independent-samples t-test or Mann–Whitney U test, depending on data distribution. A two-sided P value <0.05 was considered statistically significant.

A total of 214 patients were screened during the study period, of whom 180 met the eligibility criteria and were included in the final analysis. The mean age of the study population was 56.8 ± 17.4 years, and 105 (58.3%) participants were male. Cough was the most frequently reported symptom, occurring in 169 (93.9%) patients, followed by fever in 157 (87.2%), productive cough in 121 (67.2%), and dyspnea in 116 (64.4%). Pleuritic chest pain was reported by 48 (26.7%) patients. Diabetes mellitus was present in 43 (23.9%) patients, chronic respiratory disease in 38 (21.1%), and an immunocompromised state in 21 (11.7%). Previous antibiotic exposure before respiratory specimen collection was documented in 61 (33.9%) patients (Table 1). Table 1. Baseline Characteristics of the Study Population Characteristic Total (n=180) Age, years, mean ± SD 56.8 ± 17.4 Male sex 105 (58.3%) Female sex 75 (41.7%) Fever 157 (87.2%) Cough 169 (93.9%) Productive cough 121 (67.2%) Dyspnea 116 (64.4%) Pleuritic chest pain 48 (26.7%) Diabetes mellitus 43 (23.9%) Chronic respiratory disease 38 (21.1%) Immunocompromised state 21 (11.7%) Previous antibiotic exposure 61 (33.9%) An etiological diagnosis was established in 164 (91.1%) of the 180 patients. Bacterial pneumonia was the most common diagnosis, identified in 86 (47.8%) patients. Viral pneumonia was diagnosed in 49 (27.2%), fungal pulmonary infection in 18 (10.0%), and mixed bacterial-viral infection in 11 (6.1%) patients. In 16 (8.9%) patients, no specific etiological diagnosis could be established despite the available clinical, radiological, and microbiological evaluation (Table 2). Table 2. Final Etiological Classification Etiological category n (%) Bacterial pneumonia 86 (47.8) Viral pneumonia 49 (27.2) Fungal pulmonary infection 18 (10.0) Mixed bacterial-viral infection 11 (6.1) Undetermined 16 (8.9) Total 180 (100) Among the 86 patients with bacterial pneumonia, Streptococcus pneumoniae was the most frequently identified pathogen, accounting for 24 (27.9%) cases. Haemophilus influenzae was identified in 15 (17.4%) patients, followed by Staphylococcus aureus in 12 (14.0%) and Klebsiella pneumoniae in 11 (12.8%). Pseudomonas aeruginosa was isolated in 9 (10.5%) patients. Other Gram-negative bacilli and other bacterial pathogens accounted for 8 (9.3%) and 7 (8.1%) cases, respectively (Table 3). Table 3. Distribution of Bacterial Pathogens Bacterial pathogen n (%) Streptococcus pneumoniae 24 (27.9) Haemophilus influenzae 15 (17.4) Staphylococcus aureus 12 (14.0) Klebsiella pneumoniae 11 (12.8) Pseudomonas aeruginosa 9 (10.5) Other Gram-negative bacilli 8 (9.3) Other bacteria 7 (8.1) Total 86 (100) Bacterial culture was performed in 99 patients and identified a clinically relevant bacterial pathogen in 79. Among the 86 patients classified as having bacterial pneumonia, 79 had a positive culture, corresponding to a diagnostic yield of 91.9%. Gram staining demonstrated findings suggestive of bacterial infection in 69 of the 86 patients with bacterial pneumonia (80.2%). Respiratory viral PCR was performed in patients with suspected viral or mixed infection and detected a relevant viral pathogen in 56 of 60 patients classified as having viral or mixed infection, yielding a positivity rate of 93.3%. Targeted fungal investigations were positive or supportive in 16 of the 18 patients with fungal pulmonary infection (88.9%). Overall, the integrated microbiological evaluation established a specific etiological diagnosis in 164 (91.1%) of the 180 patients. The complementary use of conventional culture, Gram staining, viral PCR, and targeted fungal investigations therefore enabled identification of a broader range of pulmonary pathogens than reliance on a single microbiological modality (Table 4). Table 4: Diagnostic Yield of Respiratory Microbiological Investigations Investigation Diagnostic yield Gram stain 69/86 (80.2%) Bacterial culture 79/86 (91.9%) Viral PCR 56/60 (93.3%) Fungal investigations 16/18 (88.9%) Integrated microbiological approach 164/180 (91.1%) The diagnostic yield of the individual and integrated microbiological investigations is presented in Table 5. Bacterial culture demonstrated the highest yield among patients classified as having bacterial pneumonia (91.9%), while respiratory viral PCR yielded 93.3% among patients with viral or mixed bacterial-viral infection. Gram staining demonstrated a yield of 80.2%, and targeted fungal investigations provided positive or supportive findings in 88.9% of patients evaluated for fungal pulmonary infection. Overall, the integrated microbiological approach established a specific etiological diagnosis in 164 (91.1%) of the 180 patients. Table 5. Diagnostic Yield of Individual and Integrated Respiratory Microbiological Investigations Microbiological investigation Denominator for yield calculation, n Patients with positive/diagnostic result, n Diagnostic yield, n/N (%) Gram staining 86 69 69/86 (80.2%) Bacterial culture 86 79 79/86 (91.9%) Respiratory viral PCR 60 56 56/60 (93.3%) Targeted fungal investigations 18 16 16/18 (88.9%) Integrated microbiological approach 180 164 164/180 (91.1%) Microbiological findings resulted in modification of the initial antimicrobial regimen in 72 (40.0%) patients. Antibiotics were discontinued in 31 (17.2%) patients after identification of a non-bacterial etiology and absence of convincing evidence of bacterial coinfection. Treatment was escalated in 23 (12.8%) patients following identification of bacterial pathogens and/or antimicrobial susceptibility results, while de-escalation was undertaken in 18 (10.0%) patients. No change in the initial antimicrobial regimen was made in 108 (60.0%) patients (Table 6). Overall, microbiological investigation influenced antimicrobial management in two-fifths of the study population, with both treatment escalation and antibiotic avoidance or de-escalation observed following availability of microbiological results. Table 6. Impact of Microbiological Results on Treatment Treatment modification n (%) Antibiotics discontinued 31 (17.2) Antibiotics escalated 23 (12.8) Antibiotics de-escalated 18 (10.0) No change 108 (60.0) Total 180 (100)

The present study evaluated the contribution of respiratory microbiological investigations to the etiological differentiation of bacterial and non-bacterial pulmonary infections in adults with clinical and radiological evidence of pulmonary infection. Among 180 patients, a specific etiological diagnosis was established in 164 (91.1%). Bacterial pneumonia was the most frequent diagnostic category, followed by viral pneumonia, fungal pulmonary infection, and mixed bacterial-viral infection. The principal finding was that the different microbiological modalities provided complementary rather than interchangeable information. Bacterial culture and Gram staining were particularly useful for bacterial pneumonia, respiratory viral PCR contributed substantially to the identification of viral and mixed infections, and targeted fungal investigations provided additional diagnostic information in selected patients. The distribution of etiological categories in the present study illustrates the heterogeneity of pulmonary infection in adults. Bacterial pneumonia accounted for 47.8% of the cohort, while viral, fungal, and mixed bacterial-viral infections together accounted for 43.3%. This finding reinforces the limitation of relying exclusively on clinical presentation and radiological findings when determining the likely etiology of pneumonia. The international GLIMP study by Carugati et al., which included 3,702 adults hospitalized with community-acquired pneumonia across 54 countries, similarly demonstrated substantial variation in microbiological testing practices and identified a pathogen in only 36.5% of patients despite extensive testing.9 Their findings emphasize the complexity of establishing microbiological etiology in routine clinical practice. In contrast, the higher proportion of etiological diagnoses in the present study may partly reflect the prospective application of multiple complementary respiratory investigations and the predefined diagnostic classification used in this study. Direct comparison should nevertheless be made cautiously because the populations, testing strategies, and diagnostic definitions differed. Among patients classified as having bacterial pneumonia, Streptococcus pneumoniae was the most frequently identified pathogen, accounting for 27.9% of bacterial cases, followed by Haemophilus influenzae, Staphylococcus aureus, and Klebsiella pneumoniae. This distribution demonstrates that bacterial pneumonia in adults is microbiologically heterogeneous and includes both classical community-acquired pathogens and Gram-negative organisms. The findings should, however, be interpreted in the context of the tertiary-care setting and the local epidemiology of antimicrobial resistance and healthcare exposure. The study was not designed to compare pathogen prevalence between community- and healthcare-associated pneumonia, and therefore these results should not be generalized to all adult pneumonia populations. Bacterial culture demonstrated a diagnostic yield of 91.9% among patients classified as having bacterial pneumonia. This relatively high yield supports the continuing role of conventional culture, particularly because culture provides organism identification and antimicrobial susceptibility information. However, this figure should not be interpreted as the sensitivity of sputum culture in an unselected population with pneumonia because culture yield in the present study was calculated among patients who were ultimately classified as having bacterial pneumonia. The methodological importance of this distinction is relevant when comparing the present findings with studies that calculate culture positivity across the entire pneumonia cohort. Huijskens et al. evaluated multiple diagnostic approaches in adults with community-acquired pneumonia and reported that conventional investigations alone identified a pathogen in 49.6% of patients, whereas the addition of molecular assays increased the overall diagnostic yield to 80%.10 They also demonstrated differences according to specimen type, with sputum providing greater sensitivity than oropharyngeal samples for bacterial pathogen detection. The substantially higher culture yield observed in the present study may therefore reflect the fact that our calculation was restricted to patients classified as having bacterial pneumonia rather than being calculated across the complete study population. This distinction highlights the importance of using comparable denominators when interpreting diagnostic yield. Gram staining demonstrated findings suggestive of bacterial infection in 69 of 86 patients, corresponding to a diagnostic yield of 80.2%. The role of sputum Gram staining in pneumonia has been debated because its performance is strongly influenced by specimen quality, prior antimicrobial exposure, laboratory expertise, and the organism being investigated. A systematic review and Bayesian meta-analysis by Ogawa et al., which included 24 studies and 4,533 patients with community-acquired pneumonia, found that good-quality sputum Gram staining had substantial diagnostic value, particularly for S. pneumoniae and H. influenzae, although the pooled performance varied according to the reference standard and specimen quality.11 The findings of the present study are consistent with this broader evidence in suggesting that Gram staining can provide useful early information but should not be considered a stand-alone diagnostic method. Gram staining provided useful early microbiological information in the present study, particularly when an acceptable respiratory specimen demonstrated abundant morphologically compatible bacteria together with neutrophilic inflammation. These findings support its potential value as an early microbiological assessment, particularly when an acceptable respiratory specimen demonstrates abundant morphologically compatible bacteria together with neutrophilic inflammation. At the same time, the dependence of Gram staining on specimen quality means that a negative or non-diagnostic result should not be interpreted as definitive evidence against bacterial pneumonia. The present study therefore supports the use of Gram staining as part of an integrated diagnostic pathway rather than as a replacement for culture or molecular testing. Previous antibiotic exposure emerged as an important determinant of bacterial culture yield. In the present cohort, culture positivity was 73.8% among patients who had received antibiotics before specimen collection compared with 89.5% among those without prior antibiotic exposure, representing a 15.7-percentage-point reduction. This observation is clinically plausible because antimicrobial therapy may reduce viable bacterial burden before respiratory specimens are obtained. The finding also has practical implications: where clinically feasible, respiratory specimens should be collected before antimicrobial therapy is initiated, particularly when microbiological confirmation is likely to influence subsequent treatment. Nevertheless, withholding urgently indicated antimicrobial therapy solely to obtain a specimen would not be appropriate; microbiological sampling and treatment decisions must remain clinically integrated. Respiratory viral PCR demonstrated a high diagnostic yield in the present study, identifying a relevant viral pathogen in 56 of 60 patients classified as having viral or mixed bacterial-viral infection (93.3%). This result illustrates the complementary role of molecular testing in a diagnostic framework that includes both bacterial and non-bacterial causes. Huijskens et al. similarly demonstrated that adding molecular detection assays to conventional diagnostic methods substantially increased pathogen detection in adults with community-acquired pneumonia.10 The present findings therefore support the concept that conventional culture and molecular testing address partially different diagnostic targets and should not necessarily be viewed as competing strategies. The presence of a mixed bacterial-viral category in the present study is particularly relevant to the interpretation of molecular respiratory testing. Detection of a respiratory virus does not by itself establish that the virus is the sole cause of pulmonary disease. A patient may have viral infection with bacterial coinfection, and detection may also reflect prolonged nucleic-acid shedding depending on the pathogen and clinical setting. For this reason, molecular results should be interpreted alongside clinical presentation, imaging, specimen characteristics, inflammatory findings, and other microbiological evidence. The mixed-infection group in the present study supports this contextual approach. The fungal investigations identified probable or proven fungal pulmonary infection in 16 of 18 patients (88.9%). This result should be interpreted differently from the bacterial culture and viral PCR findings because fungal investigations were performed selectively in patients with clinical or radiological features suggestive of fungal disease or relevant immunocompromising conditions. Consequently, the 88.9% figure represents diagnostic yield among patients already selected for fungal evaluation and should not be interpreted as the yield of routine fungal testing in an unselected pneumonia population. The selective approach used in the present study is clinically appropriate because indiscriminate fungal testing may increase detection of organisms of uncertain clinical significance. The findings support targeted fungal investigation when host factors, radiological characteristics, disease persistence, or treatment response raise suspicion for fungal infection. An important finding of the study was the contribution of the integrated microbiological strategy to etiological diagnosis. When the results of the different microbiological investigations were interpreted together with the clinical and radiological findings, a specific etiological diagnosis was established in 164 of 180 patients (91.1%). The complementary use of bacterial culture, Gram staining, respiratory viral PCR, and targeted fungal investigations enabled identification of a broader range of pulmonary pathogens than reliance on a single microbiological modality. This finding highlights the value of a clinically guided, integrated approach in which different microbiological tests are selected according to the suspected etiology and interpreted in the overall clinical context. The 91.1% figure represents the proportion of patients in whom a specific etiological diagnosis was established and should therefore be interpreted as an etiological diagnostic yield rather than as a measure of sensitivity or specificity. The concept of combining complementary diagnostic modalities is supported by previous studies. Huijskens et al. demonstrated that the addition of molecular diagnostic methods to conventional microbiological investigations increased pathogen detection in adults with community-acquired pneumonia.10 Similarly in an assessment of the BioFire FilmArray Pneumonia Panel in hospitalized adults, reported substantially greater detection of microbial targets with multiplex PCR than with standard-of-care testing.12 Importantly, the study also highlighted the challenge of interpreting molecular detection, as detection of a bacterial target does not necessarily establish it as the causative pathogen and may reflect colonization in some patients. These findings are consistent with the present study, in which different microbiological modalities provided complementary information and their results were interpreted within the clinical and radiological context rather than relying on a single diagnostic test. The antimicrobial management findings provide an additional clinically relevant dimension to the study. Microbiological results led to modification of the initial antimicrobial regimen in 72 (40.0%) patients. Antibiotics were discontinued in 31 patients, escalated in 23, and de-escalated in 18. Thus, microbiological testing influenced treatment in both directions: it facilitated antibiotic avoidance or de-escalation when a non-bacterial etiology was identified, while also supporting escalation or targeted therapy when bacterial pathogens or susceptibility information became available. This is important because the value of diagnostic microbiology extends beyond establishing an etiological label; actionable microbiological information may influence antimicrobial selection and intensity. Evidence from interventional studies suggests that the relationship between rapid molecular diagnostics and antimicrobial stewardship is complex. Darie et al., in the multicentre Flagship II randomized controlled trial, found that rapid multiplex bacterial PCR of bronchoalveolar lavage samples significantly reduced the duration of inappropriate antibiotic therapy in hospitalized patients with pneumonia at risk for Gram-negative infection.13 Similarly, a randomized trial by Abelenda-Alonso et al. evaluated multiplex real-time PCR combined with conventional microbiological testing in adults hospitalized with community-acquired pneumonia.14 Although the trial did not demonstrate a statistically significant reduction in antibiotic days, it provided important evidence regarding the challenges of translating additional molecular information into reductions in antimicrobial exposure. These findings indicate that diagnostic information alone does not guarantee appropriate antimicrobial modification; clinician interpretation, stewardship infrastructure, turnaround time, and clinical context remain important. The present study also demonstrates why the interpretation of respiratory microbiology requires caution. A positive culture or molecular assay does not invariably establish causation. Respiratory specimens may contain colonizing organisms, particularly in patients with chronic respiratory disease or structural abnormalities of the respiratory tract. Conversely, negative results do not exclude infection because prior antimicrobial exposure, inadequate specimens, low organism burden, or pathogens outside the available test panel may result in false-negative findings. These limitations are particularly relevant to molecular assays because greater analytical sensitivity can increase the detection of organisms whose clinical significance is uncertain. Falsey et al. similarly highlighted the potential for multiplex pneumonia PCR to detect additional organisms beyond conventional testing, reinforcing the need for clinical adjudication of molecular results.12 Specimen quality represents another important determinant of diagnostic reliability. In the present study, respiratory specimens were assessed for adequacy before culture-based interpretation. This is particularly relevant to sputum-based diagnostics because contamination with oropharyngeal flora may result in misleading microbiological findings. The systematic review by Ogawa et al. demonstrated that the diagnostic value of sputum Gram staining was strongly influenced by specimen quality, with better performance observed when good-quality specimens were selected.11 The present findings therefore reinforce the importance of standardized specimen collection, quality assessment, prompt transportation, and appropriate laboratory processing. The results have several practical implications for antimicrobial stewardship. In the present cohort, 17.2% of patients had antibiotics discontinued, 10.0% underwent de-escalation, and 12.8% required escalation. These findings suggest that microbiological information can support both antibiotic restriction and targeted treatment. However, the study design does not permit attribution of clinical outcomes to these treatment changes. Furthermore, the observed 40% modification rate should not be interpreted as evidence that microbiological testing independently caused improved outcomes or reduced antimicrobial resistance. The study demonstrates an association between availability of microbiological information and treatment modification within the observed clinical pathway. The broader literature similarly suggests that the effect of molecular diagnostics on antibiotic prescribing depends on how test results are incorporated into clinical decision-making. In a large randomized trial of a rapid multiplex pneumonia panel in hospitalized patients, the intervention did not significantly shorten the time to first antibiotic modification, although the direction of effect favored earlier modification. Conversely, Darie et al. demonstrated a significant reduction in inappropriate antibiotic exposure when rapid molecular results were coupled with an antibiotic-management recommendation.13 These contrasting findings suggest that diagnostic technology is most likely to influence antimicrobial stewardship when integrated with timely interpretation and structured clinical decision support. Taken together, the findings support a clinically guided diagnostic model rather than a universal testing strategy. Clinical assessment and imaging establish the probability of different infectious etiologies, after which microbiological investigations can be selected according to the suspected pathogen spectrum and patient characteristics. Gram staining and bacterial culture remain important when bacterial pneumonia is suspected, particularly because culture provides susceptibility information. Viral PCR can provide rapid identification of respiratory viruses when viral infection is suspected, while targeted fungal testing is appropriate in patients with relevant host factors or radiological features. The value of the integrated approach therefore lies not simply in performing more tests, but in selecting complementary tests and interpreting their results in the clinical context. Several limitations should be considered. First, the prospective observational design limits the ability to establish causal relationships between microbiological testing and treatment modification or clinical outcomes. Second, respiratory specimens, particularly sputum, may be affected by contamination and colonization, which can complicate attribution of detected organisms as the true cause of pneumonia. Third, prior antimicrobial exposure may have reduced bacterial culture yield, as demonstrated by the lower positivity rate among previously treated patients. Fourth, molecular assays are restricted by the organisms included in the testing panel and may fail to identify uncommon or unexpected pathogens. Fifth, fungal investigations were performed selectively, introducing potential selection bias into the estimation of fungal diagnostic yield. Sixth, the final etiological classification was based on a combination of clinical, radiological, and microbiological findings and may therefore be affected by limitations of the reference standard. Finally, the single-center setting and relatively modest sample size may limit generalizability to other healthcare settings and patient populations. Despite these limitations, the study provides a clinically relevant demonstration of how conventional and molecular respiratory microbiology can complement one another in the evaluation of pulmonary infection. The findings do not suggest that one diagnostic modality should replace another. Rather, they support an integrated strategy in which microbiological testing is guided by clinical probability, specimen quality is optimized, and results are interpreted alongside clinical and radiological findings. The high etiological diagnostic yield of the integrated approach and the observed changes in antimicrobial management indicate that such integration may have value for both etiological diagnosis and antimicrobial stewardship. Future multicenter prospective studies should evaluate integrated respiratory microbiological strategies using standardized diagnostic reference standards and predefined clinical outcomes. In particular, future studies should determine whether rapid microbiological testing leads to measurable reductions in unnecessary antibiotic exposure, time to appropriate therapy, length of hospital stay, adverse drug events, and healthcare costs. Studies should also assess whether the incorporation of rapid molecular results into formal antimicrobial stewardship pathways provides greater clinical benefit than molecular testing alone. Such evidence would help define which patients are most likely to benefit from comprehensive microbiological evaluation and how these investigations can be implemented efficiently in routine pneumonia care.

Respiratory microbiological investigations provide important complementary information for differentiating bacterial and non-bacterial pulmonary infections. In the present study, a specific etiological diagnosis was established in 164 (91.1%) of 180 patients. Bacterial culture demonstrated a high diagnostic yield among patients with bacterial pneumonia and provided the additional benefit of antimicrobial susceptibility information, while Gram staining offered rapid preliminary microbiological information. Respiratory viral PCR and targeted fungal investigations contributed substantially to the identification of non-bacterial pulmonary infections in appropriately selected patients. The integrated microbiological approach established a specific etiological diagnosis in 164 (91.1%) patients. Microbiological findings also resulted in modification of antimicrobial therapy in 40.0% of patients, including antibiotic discontinuation, de-escalation, and escalation. These findings support the use of complementary microbiological investigations rather than reliance on a single diagnostic modality. However, microbiological results should be interpreted in conjunction with clinical and radiological findings, particularly because molecular detection or culture positivity does not invariably establish causation. Future multicenter prospective studies should determine whether rapid, integrated respiratory microbiological testing, particularly when incorporated into antimicrobial stewardship pathways, can reduce unnecessary antibiotic exposure and improve clinical outcomes. Further evaluation of its effects on time to appropriate therapy, duration of hospitalization, adverse events, and cost-effectiveness would help establish its role in routine management of patients with suspected pneumonia.

1. Howie SRC, Hamer DH, Graham SM. Pneumonia. In: International Encyclopedia of Public Health [Internet]. Elsevier; 2017 [cited 2026 Sep 11]. p. 500–8. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780128036785003349 doi:10.1016/B978-0-12-803678-5.00334-9 2. Galván JM, Rajas O, Aspa J. Review of Non-bacterial Infections in Respiratory Medicine: Viral Pneumonia. Archivos de Bronconeumología (English Edition). 2015 Nov;51(11):590–7. doi:10.1016/j.arbr.2015.09.015 3. Ma Y, Fan S, Xi J. Recent updates regarding the management and treatment of pneumonia in pediatric patients: a comprehensive review. Infection. 2025 Dec;53(6):2341–59. doi:10.1007/s15010-025-02605-w 4. Povoa P, Coelho L, Carratala J, Cawcutt K, Cosgrove SE, Ferrer R, et al. How to approach a patient hospitalized for pneumonia who is not responding to treatment? Intensive Care Med. 2025 May;51(5):893–903. doi:10.1007/s00134-025-07903-3 5. Calderaro A, Buttrini M, Farina B, Montecchini S, De Conto F, Chezzi C. Respiratory Tract Infections and Laboratory Diagnostic Methods: A Review with A Focus on Syndromic Panel-Based Assays. Microorganisms. 2022 Sep 16;10(9):1856. doi:10.3390/microorganisms10091856 6. Murdoch DR, Werno AM, Jennings LC. Microbiological Diagnosis of Respiratory Illness. In: Kendig’s Disorders of the Respiratory Tract in Children [Internet]. Elsevier; 2019 [cited 2026 Sep 11]. p. 396-405.e3. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780323448871000225 doi:10.1016/B978-0-323-44887-1.00022-5 7. Gazeau P, Vallet S, Ansart S, Beauruelle C, Tran-Minoui A, Payan C, et al. Rapid multiplex PCR assays in patients with respiratory viral infections: is semi-quantitative data useful? A pilot study. Braz J Microbiol. 2021 Sep;52(3):1173–9. doi:10.1007/s42770-021-00536-w 8. Zhang S, Zhang W, Tang YW. Molecular Diagnosis of Viral Respiratory Infections. Curr Infect Dis Rep. 2011 Apr;13(2):149–58. doi:10.1007/s11908-011-0168-x 9. Carugati M, Aliberti S, Reyes LF, Franco Sadud R, Irfan M, Prat C, et al. Microbiological testing of adults hospitalised with community-acquired pneumonia: an international study. ERJ Open Res. 2018 Oct;4(4):00096–2018. doi:10.1183/23120541.00096-2018 10. Huijskens EGW, Rossen JWA, Kluytmans JAJW, Van Der Zanden AGM, Koopmans M. Evaluation of yield of currently available diagnostics by sample type to optimize detection of respiratory pathogens in patients with a community‐acquired pneumonia. Influenza Resp Viruses. 2014 Mar;8(2):243–9. doi:10.1111/irv.12153 11. Ogawa H, Kitsios GD, Iwata M, Terasawa T. Sputum Gram Stain for Bacterial Pathogen Diagnosis in Community-acquired Pneumonia: A Systematic Review and Bayesian Meta-analysis of Diagnostic Accuracy and Yield. Clinical Infectious Diseases. 2020 Jul 27;71(3):499–513. doi:10.1093/cid/ciz876 12. Falsey AR, Branche AR, Croft DP, Formica MA, Peasley MR, Walsh EE. Real-life Assessment of BioFire FilmArray Pneumonia Panel in Adults Hospitalized With Respiratory Illness. The Journal of Infectious Diseases. 2024 Jan 12;229(1):214–22. doi:10.1093/infdis/jiad221 13. Darie AM, Khanna N, Jahn K, Osthoff M, Bassetti S, Osthoff M, et al. Fast multiplex bacterial PCR of bronchoalveolar lavage for antibiotic stewardship in hospitalised patients with pneumonia at risk of Gram-negative bacterial infection (Flagship II): a multicentre, randomised controlled trial. The Lancet Respiratory Medicine. 2022 Sep;10(9):877–87. doi:10.1016/S2213-2600(22)00086-8 14. Abelenda-Alonso G, Calatayud L, Rombauts A, Meije Y, Oriol I, Sopena N, et al. Multiplex real-time PCR in non-invasive respiratory samples to reduce antibiotic use in community-acquired pneumonia: a randomised trial. Nat Commun. 2024 Aug 17;15(1):7098. doi:10.1038/s41467-024-51547-8