Efficacy of nebulised vs intravenous tranexamic acid in management of non-massive haemoptysis: A randomised controlled trial

Dear Editor, We read with interest the randomised trial by Dinesh et al.[1] Comparing nebulised vs intravenous tranexamic acid (TXA) in non-massive haemoptysis. The study considers an important clinical question; however, several methodological issues warrant attention before extrapolating these outcomes to routine emergency practice. First, the open-label design with subjective outcome assessment is particularly problematic when comparing two active interventions in a small trial. Without blinding participants, clinicians, and outcome assessors, endpoints such as patient-reported time to cessation and patient-measured haemoptysis volume, measured using provided 100 mL transparent measuring cups with 10 mL markings, are susceptible to expectation and ascertainment bias, which may obscure true treatment differences.[2] Second, baseline imbalance in smoking status is clinically relevant. Current smokers comprised 34.8% in the nebulised group vs 8.7% in the intravenous group (absolute difference 26.1%; P = 0.071) [Table 1]. Smoking alters airway anatomy, mucociliary clearance, and bronchial vascularity, factors that may influence aerosol deposition and local drug effect. No multivariable adjustment was performed, leaving residual confounding (unadjusted in any multivariable model) that could bias comparative efficacy estimates. Third, although haemoglobin was monitored, trends were neither reported nor incorporated into blinded outcome adjudication.[3] Reliance on patient-measured expectorated volume prone to saliva admixture, technique variability, and recall bias without objective corroboration weakens confidence in the primary endpoint. The 72-hour telephone follow-up further limits the sensitivity for detecting delayed recurrence or complications. Fourth, the interpretation that nebulised TXA is “as effective as” intravenous TXA implicitly adopts an equivalence framework. Although the authors specified an equivalence margin in sample-size planning, outcomes were analysed using standard independent t-tests of mean difference rather than formal non-inferiority or equivalence methods. The reported confidence interval for the primary outcome (time to 8-hour cessation) spans −1.22 to +4.61 hours, encompassing clinically meaningful differences in emergency settings. In emergency care, even a delay of 2–3 hours in the cessation of haemoptysis may influence decisions regarding escalation to invasive airway or bronchoscopic interventions. Absence of statistical significance should not be equated with therapeutic equivalence.[4] Fifth, the 72-hour follow-up and the small sample size (n = 46) limit the ability to draw inferences about safety. Although the trial reported no adverse events in either group, this absence should be interpreted cautiously, as antifibrinolytic-associated thromboembolic complications are uncommon and may manifest beyond short observation windows. Additionally, the tuberculosis-predominant population (≈41%) constrains generalisability to settings where malignancy or bronchiectasis predominate, conditions with distinct airway pathology and aerosol deposition characteristics that may modify response to nebulised therapy.[3,5] Table 1: Baseline characteristics (reconciled with source)To summarise, while the study delivers preliminary comparative data on nebulised TXA, its open-label design, baseline imbalance, subjective outcome measurement, limited safety power, and absence of strict equivalence testing warrant prudent interpretation. The results merit tempered optimism; however, reliance on conventional frequentist P values to assert equivalence without strict evaluation against the prespecified 5% margin necessitates confirmation in adequately powered, blinded non-inferiority trials with longer follow-up and diverse haemoptysis aetiologies before routine clinical use. Author contributions Komatineni Himaja Ravikumar: Conceptualisation, drafting, correspondence. Khan Noman Zulfikar Ali: Literature review, verification. Sarbari Swaika: Supervision, critical review. Ethical statement This manuscript is a scholarly correspondence commenting on a previously published study; no new patient data were collected. Financial support and sponsorship Nil. Conflicts of interest There are no conflicts of interest.