Thoracentesis for pleural effusions from heart failure: does it help?

TAP-IT trial calls the practice into question

Thoracentesis is commonly performed to drain pleural effusions associated with heart failure.

A new randomized trial found no detectable benefits from this practice, although the sample size was small, patients were highly selected, and very large effusions were excluded.

The TAP-IT Trial

In three hospitals in Denmark, researchers enrolled 135 patients with chronic systolic heart failure (left ventricular ejection fractions ~25%) who were usually frail and elderly (~81 years old), who had “non-negligible” pleural effusions.

Those with effusions greater than two-thirds of the hemithorax were excluded, as were those with uncertain causes of their effusions (i.e., a need for diagnostic thoracentesis), contraindication to thoracentesis, severely impaired hemodynamics or respiratory failure, pulmonary infections, recent thoracentesis or other intrathoracic procedure, severe aortic stenosis, severe renal failure or dialysis.

They were randomized to receive either ultrasound-guided pleural pigtail catheter thoracentesis (with standard medical therapy) or medical therapy alone.

Over one liter was drained in more than half the patients in the intervention group (median 1,062 mL); two-thirds had at least 500 mL drained.

TAP-IT Trial: Outcomes

After 90 days of follow-up, there were no differences measured in:

• Days alive out of the hospital (84 with thoracentesis vs. 82, P=0.42), the primary outcome

• Mortality (13% in both groups)

• Hospital length of stay (5 days for both)

• Self-assessed health status, symptoms, functional limitations, and quality of life (KCCQ score), at 14 and 90 days.

• Diuretic requirements

• Weight loss

• Kidney function

• Status of medical therapy by hospital discharge.

The study was not powered to detect small differences in these outcomes.

There was significant crossover:

• 8% crossed over from control—>intervention (five control patients received thoracentesis).

• 10% of the participants in the thoracentesis group did not have a large enough effusion to tap. Two additional patients were formally crossed over to the medical therapy group for unclear reasons.

This resulted in an as-treated population with 64 (47%) undergoing thoracentesis and 71 (53%) treated with standard pharmacological therapy only.

Complications

Pneumothorax requiring intervention occurred in 1 out of 80 procedures (<1%) with three additional small pneumothoraces managed conservatively.

About one in four patients undergoing thoracentesis had “minor complications,” including accidental extubation in two, pain in nine (14%), six of whom received opioids, pneumothorax, repeated needlesticks, or minor bleeding at the site.

Caveats

This was a pragmatic and unblinded trial. Patients were offered enrollment by their attendings, creating the potential for selection bias if those offered enrollment differed systematically from those who were not.

For example, more-symptomatic patients might have been referred directly to thoracentesis (and experienced benefit), without being offered enrollment in TAP-IT. There’s no way to measure this factor.

Bias of exaggerated benefit is often a concern in trials of unblinded procedures, unless a sham procedure is included (it wasn’t here). Patients who get the procedure may be more likely to express subjective improvement, either to compensate their doctors or themselves for having had the procedure. Quality of life scores were lower with thoracentesis (non-significantly) here, but the small sample size limits any firm conclusions.

Thoracentesis was also performed pragmatically (“according to local practices”), usually with ultrasound-guided insertion of a 5-8 French catheter with passive drainage over a few hours. There was no standardized measurement of confirmation of consistency or completion of fluid drainage.

These were frail and elderly patients with advanced heart failure; almost one in seven had died within three months of trial enrollment.

Conclusions

Although thoracentesis is commonly performed for pleural effusions attributed to systolic heart failure, the TAP-IT trial calls that practice into question. Despite drainage of moderately large effusions, patients receiving thoracentesis had no fewer days in the hospital, improvement in symptoms or quality of life.

The small, pragmatic and unblinded trial was limited to mostly elderly, frail patients with severe heart failure, without severe renal failure, with moderate to moderately large effusions. Thus, wider extrapolation should be done with caution.

Many would say the primary outcome was misguided—thoracentesis wouldn’t be expected to improve survival, as symptom relief is the point of the procedure. There was no suggestion of benefit with thoracentesis on symptoms, but by enrolling only 136 patients, TAP-IT might have been inadequately powered to thoroughly test symptom relief and quality of life.

A larger trial in a broader cross-section of patients would be needed to conclude that thoracentesis brings no benefit whatsoever to patients with pleural effusion due to heart failure. Based on the results of TAP-IT, though, finding any such benefit appears to be unlikely.

Although serious complication rates were very low, discomfort and pain were common, and pneumothoraces did occur.

The safest procedure is always the one you don’t do, because you determine it’s not in the patient’s best interests. This trial will cause many physicians to think hard before ordering thoracentesis for pleural effusions due to heart failure.

Reference

Giargaard et al. A Randomized Controlled Trial of Thoracentesis in Acute Heart Failure. Circulation Volume 151, Number 16. https://doi.org/10.1161/CIRCULATIONAHA.124.073521