Combined NGS and lung biopsy for refractory respiratory failure: a case of HRV and secondary bacterial pneumonia mimicking organizing pneumonia

Abstract

Background

Some studies of community-acquired pneumonia (CAP) have reported that human rhinovirus (HRV) is the most common virus in viral pneumonia in immunocompetent adults. Secondary bacterial and fungal infections are increasingly recognized complications of HRV infection that have substantial morbidity and mortality. We report a novel case of a co-infection of Streptococcus pneumoniae (S. pneumoniae) associated with HRV pneumonia that had successful diagnosis with combined target next generation sequencing (NGS) and percutaneous lung puncture biopsy (PLPB).

Case presentation

A 62-year-old female was admitted with productive cough, dyspnea and respiratory failure. She was initially diagnosed with severe pneumonia caused by HRV infection by targeted NGS from bronchial-alveolar lavage fluid. After initial clinical improvement treated by high flow nasal cannula (HFNC) and antibiotics, the patient’s condition worsened again after her discharge, with persistent dyspnea and refractory hypoxemia. Chest computed tomography showed areas of consolidation and ground glass opacification. Despite empirical antibiotics for a suspected secondary co-infection, her condition showed no significant improvement. A PLPB was performed, and targeted NGS for the lung tissue was positive only for S. pneumoniae. Targeted NGS of her sputum was positive for S. pneumoniae, Aspergillus fumigatus and type A HRV. The patient was treated with linezolid, voriconazole and methylprednisolone. HFNC was weaned on day 57, and she was discharged with good lung recovery.

Conclusions

Our case demonstrates the diagnostic utility of combined targeted NGS and CT-guided PLPB in resolving refractory pneumonia with overlapping viral and bacterial etiologies. Co-infection with these two pathogens should be considered in the differential diagnosis of patients with consolidation, wheezing and respiratory failure following severe HRV infection. The combination of targeted NGS and CT-guided PLPB should be reserved for diagnostically challenging cases refractory to conventional methods.

Background

Recent studies have indicated that viruses account for 29–55% of community-acquired pneumonia (CAP) cases among adults, and the most common virus detected is human rhinovirus (HRV) (4.9 ∼ 30.6%) [1,2,3]. HRV is also reported as the most frequently detected respiratory virus in chronic obstructive pulmonary diseases (COPD) patients [4]. The development of secondary bacterial infection has been reported in patients with HRV pneumonia, and a high mortality of 8.2% of secondary Aspergillus infection in HRV pneumonia was reported in one case series [5].

In this report, we describe a diagnostically challenging case of severe pneumonia and acute respiratory failure caused by co-infection of Streptococcus pneumoniae (S. pneumoniae) secondary to HRV pneumonia diagnosed by combined target next generation sequencing (NGS) and percutaneous lung puncture biopsy (PLPB) in an immunocompetent adult COPD patient.

Case presentation

First admission

A 62 year-old female, non-smoker, with a history of long-term biomass fuel exposure and a history of old pulmonary tuberculosis and suspected COPD ten years before (the initial documented lung function tests was not provided at this admission) and no contact history with other sick individuals prior to admission, was admitted to our ward on November 1, 2021 (considered as day 1 for the case timeline) with repeated productive cough and short of breath for 7 years (with no acute exacerbations or hospitalizations in the preceding 3 months), deteriorated for fourteen days. She had no prior history of chronic respiratory failure or home oxygen use. Physical examination showed T 36.2℃, P 98 beats/min, RR 24 breaths/min, BP 145/85mmHg, SPO₂ 89% (on room air), BMI:18.07 kg/m², and bilateral moist crackles and wheezing in the right lung. Laboratory tests showed white blood cell count (WBC) was 6.94 × 10⁹/L, C reactive protein (CRP) was 13.7 mg/L, and procalcitonin (PCT) was<0.05 ng/ml. Arterial blood gas analysis (ABG) (on high-flow nasal cannula (HFNC), FiO₂ 0.3) showed pH 7.44, PCO₂37mmHg, PO₂68mmHg (PaO₂/FiO₂ ratio 227mmHg), HCO₃⁻25.2mmol/L. Chest high resolution computed tomography (HRCT) showed bilateral ground-glass opacities (GGO) and consolidation (Fig. 1). Amoxicillin-clavulanate (1.2 g IV infusion every 8 h) and moxifloxacin 400 mg daily were administered for 9 days, methylprednisolone 40 mg qd was administered for 4 days, budesonide 1 mg via nebulization q8h was given for 14 days, and high-flow nasal cannula (HFNC, T 34℃, flow 25 L/min, FiO₂ 0.3) was used for her respiratory support. And her condition deteriorated with dyspnea and severe respiratory failure. Repeat of HRCT on day 8 showed bilateral GGO with aggravating consolidation on new areas (Fig. 1). Due to clinical deterioration and radiological progression despite initial therapy, cefoperazone-sulbactam was initiated to broaden coverage against suspected Pseudomonas aeruginosa infection. Bronchoscopy was done on day 10 due to respiratory instability on day 8, and lots of thin secretion in bilateral bronchi were observed (Fig. 2). Bronchial-alveolar lavage fluid (BALF) analysis via Gram stain, fungal stains, and cultures showed no bacterial or fungal growth, while targeted NGS reveled high reads for type A HRV. The patient was diagnosed with acute exacerbation of COPD (AECOPD) triggered by HRV pneumonia. As she showed severe short of breath and wheezing, methylprednisolone was administered on day15, and her symptoms were significantly relieved. Repeated HRCT on day 18 showed that the lesions in the right middle and lower lobes were absorbed, but the GGO in the left lower lobe progressed (Fig. 1). She was discharged on day 19.

Fig. 1

Chest computed tomography (CT) on Day1, Day 8, Day 18, Day 41, Day 45 (including the slice of percutaneous puncture lung biopsy) Day 54 and Day63. Three representative slices of the upper, middle and lower lobe were chosen

Fig. 2

Left and right upper bronchi under bronchoscopy examination on Day 10. Lots of thin secretion in bilateral bronchi and mucous membranes hyperemia and edema were observed

Second admission

Her symptoms recurred on day 36, showed as cough with blood sputum and short of breath. Blood count tested in the outpatient showed a high level WBC of 14.74 × 10⁹/L, with a neutrophil of 86.7%. Levofloxacin was given but her condition was deteriorated and was readmitted to our ward on day 40. Physical examination showed T 36.2℃, P 92 beats/min, RR 28 breaths/min, BP 109/72mmHg, SPO₂ 85% (on room air), wheezing and scattered moist crackles could be heard in both lungs. Laboratory tests showed WBC of 6.94 × 10⁹/L with a neutrophil of 74.5%, and her CRP was 127 mg/L. Her PCT rose to 0.35ng/ml, and her serum galactomannan (GM) was 0.12. Blood gas (on room air) showed pH 7.50, PCO₂28mmHg, PO₂45mmHg (PaO₂/FiO₂ ratio 214mmHg), suggested acute hypoxemic respiratory failure. Repeated chest HRCT showed right middle lobe consolidation and multiple GGO with local consolidation in the right middle lobe and both lower lobes of the lungs (Fig. 1). She was put on HFNC (T 34℃, flow 25 L/min, FiO₂ 0.35) and a combined antibiotic therapy of moxifloxacin (400 mg IV infusion once daily), cefoperazone-sulbactam (3 g IV infusion every 12 h) and ribavirin. Targeted NGS of her sputum on day 44 was positive for S. pneumoniae, Aspergillus fumigatus, and type A HRV. However, standard microbiological tests of the same sample–including Gram stain (revealing moderate Gram-positive cocci) and sputum culture (showing normal respiratory flora without pathogenic isolates after 48 h)–failed to identify these pathogens. The differential diagnoses of viral pneumonia, bacterial pneumonia or organizing pneumonia were considered. Her deteriorated respiratory failure made another bronchoscopy hard to perform. Therefore, PLPB guided by CT was done on day 45. The histological examination suggested a necrosis, and the targeted NGS from the lung tissue was only positive for S. pneumoniae. Her serum total IgE concentrations were 140.7IU/mL (normal range, 0 ∼ 100IU/mL). But the serum concentrations of IgG and IgM of Aspergillus were normal. Linezolid 600 mg twice daily, voriconazole 200 mg twice daily and methylprednisolone 40 mg twice daily were started on day 46. Her oxygenation improved, and her wheezing was relieved on day 57. A repeated chest CT on day 63 showed absorption of the consolidated lung in the right middle lobe and obvious alleviation of the focal infections in both lower lobes (Fig. 1). She was discharged on day 64. A time line of the examinations, antibiotics and steroids use is illustrated in Fig. 3.

Fig. 3

Time line of the examinations, antibiotics and steroids use

Literature review

We conducted a search in PubMed using the terms “Rhinovirus pneumonia” and “respiratory failure”, excluding literature related to “pediatric” and “immunosuppressed” populations. The initial search yielded 31 relevant articles, which, after rigorous evaluation, were refined to a final selection of 10 articles, consisting of 1 cohort study and 9 individual case reports (Table 1).

Table 1 Reported case series of rhinovirus-associated severe pneumonia in immunocompetent adults

The 10 articles were published between 2014 and 2024, collectively encompassing a collective total of 42 cases. The patients’ ages ranged from 20 to 95 years, with 4 cases (10%) under 30 years. Among these, two (4.8%) reported no comorbidities. Chronic lung diseases (n = 21) were the most frequent pre-existing condition, followed by nutritional and metabolic disorders (n = 16), such as obesity and diabetes. The clinical symptoms of HRV pneumonia with respiratory failure lacked specificity, with cough (n = 37), sputum (n = 29), and fever (n = 17) being the most common symptoms. The most frequent imaging feature was consolidation (n = 9), followed by interstitial lung changes (n = 8) and ground glass opacity (GGO)(n = 5). In terms of treatment, there is currently no specific antiviral therapy for HRV, necessitating a reliance on comprehensive supportive care in clinical practice. Among the cases reviewed, 24 cases received systemic corticoids, 22 cases were treated with Oseltamivir, and 1 case was treated with Ribavirin. The treatment for respiratory failure primarily involved non-invasive mechanical ventilation (NIV) (n = 7), supplemented by invasive mechanical ventilation (IMV) (n = 5) and high-flow nasal cannula (HFNC) (n = 4). Unlike other instances, viral diagnosis in our case was conducted using NGS, specifically targeted NGS for sputum, alveolar lavage fluid, and in vivo lung tissue samples collected via PLPB. The studied cases had a mortality rate of 12% (5 out of 42 cases), comparable to the hospital mortality rate of 12.5% reported in another study on HRV -associated CAP [5].

Discussion and conclusions

A prolonged refractory respiratory failure and consolidations caused by S. pneumoniae pneumonia secondary to HRV pneumonia in an immunocompetent COPD patient confirmed by the combined targeted NGS and PLPB is first reported in this case.

The debate about the role of HRV in CAP is still ongoing, especially in immunocompetent adult individuals. However, HRV is increasingly detected in immunocompetent adult patients with CAP [3, 6], and COPD exacerbations [4]. Recent studies suggested that in immunocompetent hosts, HRV can cause clinically significant respiratory disease [2, 3, 7, 8].

Clinical data using BALF specimens from immunocompetent adult patients with HRV-associated CAP arescarce [2, 5, 9, 10]. The use of NGS has enhanced the detection of respiratory virus infection. However, the identification of virus does not necessarily mean that it is a CAP infection pathogen. Moreover, when immunocompetent adult patients with underlying diseases suffer from severe pneumonia, and when HRV was the only pathogen detected in BALF, primary HRV pneumonia should be considered [7, 11]. A major strength of our case was that HRV-infection was confirmed the only pathogen via BALF samples obtained from this patient of her first admission, which indicated that HRV may have an essential role in the pathogenesis of the pneumonia and respiratory failure, even in mono-infection. And her underlying disease of COPD maybe a risk factor for her susceptability to HRV pneumonia.

In our case, a refractory respiratory failure, severe wheezing and a consolidation on CT were shown during her second admission, and a repeated targeted NGS from her sputum showed positive for HRV type A, S. pneumoniae and A. fumigatus. Previous report suggested that the most common radiological finding was consolidation, and S. pneumoniae was the most common co-pathogen among patients with severe CAP caused by HRV [5, 12, 13]. In our case, the secondary S. pneumonia pneumonia was further confirmed by lung biopsy and positive NGS test from her lung tissue. However, HRV was only found in her sputum but not in her lung tissue. It remains unclear whether the S. pneumoniae pneumonia was directly caused by viral pathogenicity or if the viral infection increased the patient’s susceptibility to bacterial pneumonia. On one hand, HRV infection induces epithelial cell damage, inflammatory responses and immunosuppression, and significantly enhances the adherence of S. pneumoniae toepithelial cells. Consequently, this increases the risk of secondary bacterial infections, particularly those caused by S. pneumoniae [11,12,13,14,15]. On the other hand, secondary S. pneumoniae pneumonia was an independent risk factor associated with worse outcome in HRV infected patients [16]. Moreover, clinical deterioration was more common in viral-bacterial co-infection patients with higher incidence of septic shock, longer ICU length of stay, longer hospital length of stay and higher 28-day mortality [5]. Our patient had a secondary S. pneumoniae pneumonia after 2 weeks of her HRV pneumonia, and the HRV was still positive in her sputum at that time. A recent study suggested that airway epithelial cell immunity is delayed during HRV infection in asthma and COPD [15]. Therefore, the viral-bacterial co-infection could be one connected process. However, it is unclear how long HRV persists in the respiratory system and how it should be considered as a risk factor for secondary bacterial infections.

For our patient, the consolidation was also suspected for an organizing pneumonia (OP) after HRV pneumonia, as OP was reported in several viral pneumonia, such as influenza viruses [16] and coronaviruses [17], which could lead to persistent respiratory failure. A prior HRV pneumonia case with transbronchial lung biopsy (TBLB) findings of fibrin exudates and mild alveolitis further suggested OP potential [18]. Given the patient’s critical respiratory status (PaO2/FiO2 214 mmHg) precluding safe bronchoscopy, CT-guided PLPB was prioritized for definitive diagnosis. The primary purpose of CT-guided PLPB, however, was to exclude OP given the patient’s steroid-responsive radiographic progression, rather than to diagnose infection. Histopathological examination of lung tissue demonstrated necrotizing inflammation without organizing pneumonia features, while targeted NGS confirmed S. pneumoniae co-infection. This dual-path approach (histological exclusion of OP + molecular pathogen detection) highlights the value of integrating PLPB and targeted NGS in critically ill patients with peripheral consolidations and contraindications to bronchoscopy, where conventional diagnostics fail to resolve ambiguity.

Our patient present with persistent wheezing and dyspnea during hospitalization. While HRV induced severe COPD exacerbation was the primary consideration, Aspergillus tracheobronchitis (ATB) was also suspected. HRV is a well-established trigger for exacerbations in asthma and COPD [4, 19, 20]. However, fungal co-infection was only reported in two patients with HRV pneumonia in one study [11]. In our case, targeted NGS of sputum detected Aspergillus fumigatus, and the elevated serum total IgE (140.7 IU/mL) alongside clinical improvement with voriconazole and corticosteroids supported possible diagnosis of allergic bronchopulmonary aspergillosis (ABPA). Regarding ABPA, the CT findings on Day 1 showed bronchiectatic dilatation with a mucus plug in the left lobe, transient IgE elevation and steroid responsiveness which suggest a possible ABPA. However, the patient did not have a history of asthma, her blood eosinophilia was normal, and Aspergillus-specific IgE was not detected [21]. Moreover, invasive pulmonary aspergillosis (IPA) was excluded by negative serum galactomannan (GM) and negative fungal cultures, ATB in critically ill COPD patients often lacks classic diagnostic markers (e.g., GM positivity, radiological cavitation) and may present solely with airway-centric symptoms [22, 23]. As emphasized in consensus guidelines, definitive diagnosis of ATB requires bronchoscopic visualization of airway invasion (e.g., pseudomembranes, ulcers) and histopathological confirmation of Aspergillus hyphae in tracheobronchial tissue [24]. The rapid resolution of dyspnea likely reflects synergistic anti-inflammatory (methylprednisolone) and antifungal (voriconazole) effects on HRV-induced bronchial hyperreactivity and localized fungal airway involvement. This case underscores the diagnostic challenges of ATB in immunocompetent hosts, where conventional methods (e.g., fungal culture, GM assay) may lack sensitivity; while NGS provides enhanced pathogen detection capability, its limitations in distinguishing colonization from active infection necessitate integration with clinical and histopathological findings. Whether Aspergillus infection is a common co-infection pathogen in HRV as frequent as influenza [11] warrant further investigation. Meanwhile, the incidence and recognition of severe viral-associated pulmonary aspergillosis (VAPA) are rising. The 2024 consensus definitions from ESGCIP, EFISG, ESICM, ECMM, MSGERC, ISAC, and ISHAM provide valuable guidance on the diagnosis and management of invasive fungal diseases in critically ill adult ICU patients. However, the underlying mechanisms contributing to this phenomenon require further exploration [25].

Cao and colleagues also conducted a study on rhinovirus pneumonia, involving both immunocompetent and immunocompromised patients [16]. A total of 694 pneumonia cases were analyzed in this study, of which 49 (7% of the total) tested positive for HRV. Respiratory failure occurred in 16 (32.7%) of the HRV-positive cases, and septic shock in 2 (4.1%), resulting in an 8.2% (4/49) hospital mortality rate. These findings highlight that HRV can cause pneumonia in immunocompetent adults and may lead to critical illness. The frequent presence of chronic lung disease and diabetes as underlying conditions in this cohort aligns with our literature review, suggesting that patients with these conditions, including those who are immunocompromised, should be vigilant about the possibility of HRV infection and consider it in their differential diagnosis.

In instances where microbiologic results were obtainable, HRV was predominantly identified as the sole causative agent, with only one instance of a mixed infection profile, highlighting a remarkably low incidence of co-infection. Contrastingly, these findings diverge from data from two other cohort studies on HRV pneumonia, which reported co-infection rates of 40% [5] and 69% [13], respectively. This discrepancy may be attributed to the focus of the aforementioned studies on HRV infections, while our review narrowly focused on cases with a confirmed clinical diagnosis of HRV pneumonia. In the clinical context, diagnosing HRV pneumonia requires careful consideration; notably, a positive test for HRV in respiratory specimens by itself does not suffice for a definitive diagnosis. It is essential to also exclude other potential pathogenic microorganisms that could contribute to the pneumonia. For HRV pneumonia, while typical shedding from the upper respiratory tract lasts 7–10 days in healthy individuals [26], in immunocompromised or those with chronic respiratory disease (such as COPD or asthma), viral shedding could last up to 2–3 weeks or longer [4, 14, 26].

Compared to traditional comprehensive microbiological tests, metagenomic NGS significantly enhances the diagnostic efficiency for pneumonia pathogenesis, proving invaluable in identifying patients infected with novel, rare, and atypical pathogens [27]. In our case, sputum cultures repeatedly showed normal flora, while targeted NGS detected S. pneumoniae and Aspergillus fumigatus, pathogens later confirmed by lung biopsy, highlighting the limitations of conventional methods in antibiotic-pretreated patients. However, its adoption in clinical settings is hindered by its high cost. Targeted NGS, unlike metagenomic NGS, which sequences all genetic material in a sample without prior selection, focuses on targeted regions, providing higher sensitivity and specificity for specific pathogens. Targeted NGS uses specific primers or probes to enrich and sequence target pathogens, offering high sensitivity and specificity for detecting multiple pathogens, including bacteria, viruses, and fungi. In terms of result interpretation, we follow expert consensus guidelines [16]. The final results should be interpreted in conjunction with clinical symptoms, laboratory tests, and other diagnostic methods to ensure accuracy [28, 29]. Li et al. [30] conducted a comparative study on the detection efficacy of targeted NGS and metagenomic NGS on respiratory samples, finding that targeted NGS was effective. The study by Zhang et al. [31] evaluated the clinical application of targeted NGS in severe pneumonia, demonstrating its high concordance with clinical diagnoses and its efficacy in detecting pathogens, even those missed by traditional culture methods.

While NGS identifies microbial nucleic acids, it cannot distinguish between viable pathogens, non-viable debris, or transient colonization. In our case, the detection of S. pneumoniae DNA in lung tissue (despite negative cultures) likely represents residual bacterial DNA post-antibiotic lysis rather than active replication. Similarly, A. fumigatus DNA in sputum may reflect airway colonization, as invasive aspergillosis requires histopathological evidence absent here. This underscores a fundamental principle: NGS findings must be contextualized within clinical, radiological, and histopathological data to avoid overdiagnosis.

Our case, which initially improved but later deteriorated, underscores the challenges faced by patients with viral infections. Such patients are at higher risk for developing severe and persistent pneumonia, requiring not only robust nutritional and immunological support but also adequate and prolonged antimicrobial therapy. The complexity of the illness, including overlapping viral and bacterial infections and potential bacterial resistance, not only delays microbiological diagnosis but also complicates therapeutic decision-making, ultimately prolonging treatment duration and reducing clinical efficacy. This case emphasizes the importance of timely and comprehensive treatment strategies.

While NGS and PLPB offer significant advantages in diagnosing complex respiratory infections, it is important to acknowledge their limitations. Although targeted NGS has notably reduced detection costs, metagenomic NGS remains costly, particularly when both DNA and RNA analyses are required. Despite its high sensitivity, NGS can sometimes detect low-abundance pathogens that may not be clinically significant, potentially leading to overdiagnosis or misinterpretation of results. Given the potential for bronchoscopy to cause complications such as nosocomial infections, it is crucial to carefully select patients, particularly those who are immunocompromised. PLPB requires specialized imaging, advanced equipment, and skilled operators, which limits its accessibility in some healthcare settings. While CT-guided PLPB provided critical diagnostic insights, the number of specimens collected was limited, thereby may constraining the scope of the diagnosis. Additionally, the risk of complications such as pneumothorax or bleeding must be carefully considered. These limitations highlight the need for a balanced approach when selecting diagnostic tools, especially in resource-limited settings.

Our case demonstrates the diagnostic utility of combined targeted NGS and CT-guided PLPB in resolving refractory pneumonia with overlapping viral and bacterial etiologies. Co-infection with these two pathogens should be considered in the differential diagnosis of patients with consolidation, wheezing and respiratory failure following severe HRV infection. The combination of targeted NGS and CT-guided PLPB should be reserved for diagnostically challenging cases refractory to conventional methods (e.g., culture-negative, radiologically ambiguous, or immunocompromised hosts), as recommended by IDSA/ATS guidelines for community-acquired pneumonia [32, 33]. Lung infections with primary viral pneumonia often lead to secondary bacterial pneumonia if they remain inadequately evaluated. Secondary bacterial pneumonia is more frequently reported in cases of pneumonia caused by viruses such as influenza, parainfluenza, RSV, rhinovirus, coronavirus, and measles.