Nebulized TXA for Hemoptysis (Wand 2018)

Hemoptysis can be scary. I covered the emergency medicine approach to massive hemoptysis in one of the early posts on First10EM. However, even when the hemoptysis is small volume, it can be quite distressing to patients and clinicians alike. This paper asks whether using nebulized tranexamic acid (or TXA) for hemoptysis results in less bleeding.

The paper

Wand O, Guber E, Guber A, Epstein Shochet G, Israeli-Shani L, Shitrit D. Inhaled Tranexamic Acid TXA for Hemoptysis Treatment: A Randomized Controlled Trial. Chest. 2018; 154(6):1379-1384. PMID: 30321510

The Methods

This is a single centre, randomized, placebo controlled trial. ClinicalTrials.gov NCT01496196.

Patients

Adult patients admitted to a pulmonary ward for hemoptysis that started within 24 hours of the admission.

• Exclusions: massive hemoptysis; respiratory or hemodynamic instability; pregnancy; renal or hepatic failure; coagulopathy; or previous use of, or sensitivity to, TXA.

Intervention

Nebulized TXA (500mg/5mL) 3 times daily.

Comparison

Nebulized normal saline.

Outcome

Primary outcomes were resolution of hemoptysis within 5 days and volume of daily expectorated blood.

The Results

They included 47 patients (out of 55 screened) with a mean age of 66 years. The group was 74% male, 64% smokers, and 66% had known lung disease.  

The trial was stopped early for efficacy.

Resolution of hemoptysis occured in 96% of the TXA group and 50% of the placebo group (p<0.005).

Mortality was not statistically different, but occured in 9% of the placebo group as compared to 0% of the TXA group (p=0.21). As you might expect, the mortality numbers evened out by 1 year (16% vs 18%).

30 day recurrence was not statistically significant, but occurred in 27% of the placebo group and 8% of the TXA group (p=0.12). Recurrence was statistically significant by 1 year (50% vs 12%; p <0.01).

My thoughts

There are a large number of problems with this study that leave me uncertain about the results, despite the rather drastic difference between the two groups. It is a single center study with small numbers. These are patients admitted to a pulmonary ward, excluding patients with massive hemoptysis, and so will only represent a small subset of those we see in the emergency department. Although the trial was blinded, they don’t mention allocation concealment anywhere in the manuscript, which, if done improperly, can result in significant selection bias. Furthermore, they have two primary outcomes when there should only be one. (Even worse, the clinicaltrials.gov listing only mentions one primary outcome, so total amount of blood was presumably a secondary outcome elevated to primary status only in the manuscript.)

The trial was stopped early, but they don’t say whether this was a pre-planned analysis or stopping point. Without clear criteria for interim data checks and trial stoppage, p hacking (or performing multiple analyses and only stopping when a statistically significant p value is seen) is possible, limiting the value of the results. Even if the trial was stopped according to a specified protocol, stopping trials early for benefit will frequently overestimate the magnitude of the benefit.

Although there were no statistical differences between the two groups at baseline, there were a number of differences that could have influenced the results. The TXA group had fewer smokers (56% vs 73%), less cancer (36% vs 41%), and less antiplatelet use (32% vs 50%). On the other hand, the placebo group had less anticoagulant use (9% vs 24%) and a smaller volume of bleeding at admission (35 mL vs 51 mL). These discrepancies in randomization are one of the big problems with small trials. It isn’t clear what effects these differences have on the results. We will need to see the study replicated to be sure there wasn’t a confounder influencing the positive results with TXA.

A major red flag for me is that the math in this manuscript is atrocious. For example, there are multiple simple arithmetic errors in table 3, including the total number of procedures being less at 1 year than they were at 30 days, which is clearly impossible. These types of errors, especially when they are so obvious, make me very concerned about possible bigger mistakes at the research level, and also about the editorial process. It also makes it hard to trust the reported p values when simple division was done incorrectly in multiple places.

Despite these many problems, there is reason to believe that TXA could be helpful in hemoptysis. A Cochrane review on the subject found 2 RCTs and concluded that TXA may reduce the duration of hemoptysis, although the quality of evidence was considered low. (Prutsky 2016) Subsequently, another small pilot RCT showed a slightly improvement in perceived hemoptysis when comparing IV TXA to placebo. (Bellam 2016)

It isn’t clear that nebulized TXA would be any better than IV or oral administration. This study specifically excluded patients the massive hemoptysis. I think nebulized TXA is much less likely to work in this group (because of the blood blocking the airway) and is likely to interfere with other important procedures, like adequate preoxygenation. If you are going to use TXA in a patient with massive hemoptysis, I think it makes a lot more sense to give it intravenously.

Bottom line

This is a small trial with lots of issues, but it hints that in stable inpatients with small volume hemoptysis, the use of nebulized TXA might reduce bleeding. The trial is too small to comment on harms. We need to see this trial replicated, preferable in a larger multicentre RCT, before this becomes standard care, but I think it would be reasonable to use this therapy while waiting for more trials to be completed.

Other FOAMed

REBEL EM: TXA for Everyone: Inhaled TXA for Hemoptysis

EM Cases Quick Hits Episode 2 – Swami discusses this paper and TXA for everything

REBEL EM: Tranexamic Acid (TXA) for Everything that Bleeds?

EM PharmD: A Breath of Fresh Air: Nebulized TXA for Hemoptysis

References

Bellam BL, Dhibar DP, Suri V, et al. Efficacy of tranexamic acid in haemoptysis: A randomized, controlled pilot study. Pulmonary pharmacology & therapeutics. 2016; 40:80-3. [pubmed]

Prutsky G, Domecq JP, Salazar CA, Accinelli R. Antifibrinolytic therapy to reduce haemoptysis from any cause. The Cochrane database of systematic reviews. 2016; 11:CD008711. [pubmed]

Wand O, Guber E, Guber A, Epstein Shochet G, Israeli-Shani L, Shitrit D. Inhaled Tranexamic Acid TXA for Hemoptysis Treatment: A Randomized Controlled Trial. Chest. 2018; 154(6):1379-1384. PMID: 30321510