Early Immunosuppressive Therapy for Idiopathic Pneumonia Syndrome Prior to Bronchoscopy Is Associated with Favourable Outcomes

Introduction: Idiopathic pneumonia syndrome (IPS) is a dangerous complication following haematopoietic stem cell transplantation (HSCT). Immunosuppressive therapy (IST) using high-dose methylprednisone and etanercept produces complete response (CR) in the majority of IPS cases. However, exclusion of respiratory pathogens by bronchoscopy prior to IST commencement carries procedural risks that may worsen respiratory failure and IPS severity, with potential negative impact upon IST response and survival. Given this, we perform bronchoscopy post-IST commencement only in those patients with only mild (nasal oxygen) or severe (already intubated) respiratory failure. We commence IST within 24 hours of clinical and radiological suspicion of IPS, and cease IST if alternative aetiologies are subsequently proven.

Aim: We aimed to evaluate the safety and efficacy of our local early-intervention IST strategy in patients with new-onset suspected IPS following HSCT.

Methods: We performed a retrospective, single-centre descriptive cohort study of patients who received IST for IPS between 2014-2019. Details collected included: patient demographics, IPS diagnostic details/severity, IST response, and survival. The primary study objectives were the incidence of complete response (CR), defined by cessation of supplemental oxygen for at least 48 hours, and overall survival (OS) following IST. IST comprised high-dose corticosteroids (methylprednisolone 2mg/kg/day or equivalent) and etanercept (25mg twice weekly for 4 weeks, then weekly to complete 12 total doses).

Results: Of the 480 HSCT performed during the study period, 17 patients developed suspected IPS and received IST. At IPS diagnosis, 10 (59%) required intensive care admission. Non-invasive (NIV) and invasive (IV) ventilation was required in 4 (24%) and 5 (29%) patients, respectively. All patients met clinical and radiological American Thoracic Society (ATS) criteria for IPS. Bronchoscopy was only performed in 9 (53%), comprising all 5 receiving IV and 4 who were only requiring supplemental nasal oxygen. All 4 patients receiving NIV were considered too high-risk for post-bronchoscopy intubation, so did not proceed. Of those who underwent bronchoscopy, none experienced worsening respiratory failure, however 1 received a revised diagnosis of bacterial pneumonia and ceased IST within 3 days of commencement, and survived. In the remaining 16 patients, CR to IST was achieved in 14 (88%) patients, with 1-year OS of 44%. However, at a mean 17 months follow-up, OS for the whole cohort was 31% (Figure 1), reflecting a high incidence of non-relapse mortality (NRM) at 63% (Table 1). There was no significant difference in response or survival between patients who did or did not undergo bronchoscopy.

Conclusion: Our early-intervention IST strategy for IPS appears to produce similar response rates and survival as those reported in other prospective and retrospective series, without any obvious safety concerns. Although adherence to ATS diagnostic criteria and exclusion of occult infectious aetiologies is preferred, bronchoscopy may be safely delayed or omitted in a subset of patients. Early intervention may arrest the respiratory failure and ultimately avoid mechanical ventilation.