Should manometry be routinely used during thoracentesis? Yes, but not without some basic physiologic understanding!

KEYWORDS:

• Pleural effusion
• manometry
• reexpansion pulmonary edema
• pneumothorax
• thoracentesis

Pleural effusions are one of the most common entities seen by pulmonologists. An estimated 1.5 million new pleural effusions occur annually, 200,000 of which will ultimately prove malignant, leading to approximately 175,000 diagnostic and therapeutic thoracenteses in the US alone [1]. This past decade has witnessed unprecedented efforts in clinical pleural research, which have led to significant changes in the way we approach, diagnose and treat pleural effusions. The exponential use of ultrasound, for example, has unequivocally been demonstrated to improve the yield and safety of pleural procedures in general, and thoracentesis in particular. Similarly, pleural manometry has enjoyed in the past several years a regain of interest from the pulmonary community. Yet, unlike ultrasound, manometry remains to be widely accepted as a valuable diagnostic tool during therapeutic thoracentesis, and continues to generate interesting debates among pleural experts [2,3].

Attention to pleural pressures is not a novel concept. It was described more than 100 years ago and was commonly used in the era of therapeutic lung collapse for tuberculosis, to alert clinicians of adequate needle entry into the pleural space when physiologic negative pleural pressures were identified [4]. Similar to the use of thoracoscopy for the evaluation of pleural disease, the use of pleural manometry fell out of favor with the discovery of effective anti-tuberculous agents, only to be ‘re-discovered’ in the 1970s as a window into the physiology of pleural effusions [5]. Multiple techniques have been described since the days of undamped U-shaped manometers, the ‘poor man’s manometry’, such as the interposition of a resistor to damp the signal (limiting respiratory swings for more accurate measurements), and the use of electronic, digital hand-held and high-temporal resolution manometers [4,6,7]. In spite of a prolific body of literature, manometry remains under-utilized. One proposed explanation is that patient outcomes have not definitively been shown to improve with manometry. More likely, however, we believe that manometry and the physiological derangements it helps uncover have perhaps simply been generally poorly understood. A few basic physiologic clarifications are therefore needed.

In normal human physiologic conditions, the pleural space is approximately 10–20 μm wide and contains 10–20 mL of pleural fluid [1]. The laws of fluid dynamics in open channels (such as the channel bordered by visceral and parietal pleura) dictate that the narrower the channel, the less the contribution of hydrostatic pressure gradient due to the height of the fluid column, balanced as it is by pressure losses due to viscous flow [8]. Accordingly, in normal resting physiologic conditions, pleural pressure should be fairly similar anywhere in the pleural space, approaching −3 to −5 cmH₂O, a combination of the elastic recoil of the lung and the tendency of the chest wall to expand outward. The accumulation of pleural fluid results in a greater separation of the two pleural surfaces and allows the development of a hydrostatic pressure gradient which, at resting conditions, should result in a pressure liquid gradient of 1 cmH₂O/cm of height. Given their respective pressure volume curves, the elastic forces of the lungs and chest wall are unlikely to significantly contribute to the liquid pleural pressure (except in very large effusions with tension mechanism), which for all intents and purposes should approximate the pressure generated by the column of fluid present.

The clinical significance of these physiological principles is that as fluid accumulates in the pleural space (typically from a combination of increased production and decreased reabsorption), pleural pressure should increase from a slightly subatmospheric level to a positive value that correlates with the volume of the effusion (precisely the height of the column of fluid). Drainage of the effusion should, conversely, result in a gradual decline back to the initial level. If pressure measurements are obtained throughout the drainage procedure and plotted on a pressure–volume curve, a linear decline in pleural pressure can be observed, given a normal compliance of the respiratory system (lung and chest wall). This situation is traditionally described as a normal pleural elastance curve, and suggests that both lung and chest wall gradually and freely return to their original position during pleural drainage. It is therefore intuitively obvious that obtaining pleural pressure measurements during thoracentesis can provide invaluable data on the capacity of the lung to re-expand, and of the chest wall to retract to its original position [5]. Practically, this retraction of the chest wall is primarily driven by the diaphragm re-assuming its normal dome-shaped configuration. It is, interestingly, this change in diaphragm conformation that may be most responsible for the relief of breathlessness associated with drainage of large pleural effusions, a finding easily visualized by ultrasound, according to the concept of ‘length-tension inappropriateness’ [9–11].

The following comments can therefore be logically inferred:

1. In the presence of pleural effusion, opening pleural pressure in resting conditions (functional residual capacity, or end-expiration) should be positive, as it reflects the height of the column of fluid. A negative opening pressure, in the absence of inspiratory effort, should suggest an ex vacuo mechanism for the pleural effusion. Therapeutic drainage would not be expected to improve symptoms.

2. The slope of the pleural elastance curve (pleural pressure/volume removed) should be relatively constant throughout the procedure, reflecting a gradual and unhindered return of lung and chest wall to their original positions. An abrupt pressure decline should alert the clinicians of the impossibility of the lung to expand and/or the chest wall to retract any further. Continued drainage beyond this inflection point will not benefit the patient.

3. A sharp decline in pleural pressures may occur at any time, but unless it occurs at the very end of the procedure at which point pleural deformation forces may affect pressure measurements, such a decline generally signifies that the lung can no longer expand and the procedure should be stopped. We therefore prefer the term ‘unexpandable lung’ to the often confusing terms of trapped (absolutely unexpandable) and entrapped lung (partially expandable). There are a variety of causes of unexpandable lung which have been reviewed elsewhere [12].

The importance of early identification of unexpandable lung needs to be emphasized. Continued drainage beyond maximal expansion of the lung has clearly been associated with chest discomfort, which is often severe [13]. While re-introduction of atmospheric air into the pleural space can result in prompt symptom relief (therapeutic pneumothorax), this intervention is in practice rarely performed [12]. An informed use of manometry may therefore help mitigate this problem. Perhaps more importantly, there is indirect evidence that complications of therapeutic thoracentesis, such as re-expansion pulmonary edema and pneumothorax ex vacuo, likely result from excessively negative pleural pressures, driving an increase in transpulmonary pressure leading to stress injury [14–16]. Finally, as pleurodesis requires extensive and maintained apposition of visceral and parietal pleura, excessively negative pressures (as seen with unexpandable lung) would seem a poor predictor of successful pleurodesis, a hypothesis that finds some support in published data [17].

As mentioned earlier, however, the use of pleural manometry has not, in fact, been associated with improved clinical outcomes. While no prospective controlled studies have been performed, limited retrospective data suggest that re-expansion pulmonary edema does occur in spite of the use of manometry at a rate similar to that of studies not using manometry, independently of measured pleural pressures and elastance [18]; pneumothorax ex vacuo occurs in the absence of detectable excessively negative pleural pressures [14]; and change in chest discomfort before and after drainage is not different in manometry- versus symptom-guided thoracentesis, regardless of the volume drained [19].

These disappointing observations may simply reflect the lack of quality randomized controlled studies, and the fact that physicians who perform significant numbers of thoracenteses cannot blind themselves to pleural pressure changes during the procedure. It is certainly possible that outcomes could have been worse in the absence of manometry. More likely, it should be evident from the physiologic points made above that the slope of the pleural elastance curve should be a more important guide than the arbitrarily chosen absolute pressure value of –20 cmH2O used in these studies, a proposed conservative cutoff loosely based on limited animal data [15,16]. Other potential explanations for this apparent lack of clinical benefit include the necessarily intermittent pressure measurements (typically done every 200 mL of fluid drained), which leave substantial periods of ‘blind time’ during which abrupt changes in pleural compliance may go unnoticed, and the fact that lungs may be heterogeneous, leading to different levels of stress for a given transpulmonary pressure, exposing less compliant areas to unpredictable regional complications.

A more likely explanation, however, is that, except for chest discomfort, complications related to excessively negative pressures are rare, and adequately powered studies addressing these complications as end points would not be practically feasible [20]. A randomized controlled study, powered to identify an improvement in chest discomfort with the use of manometry during large volume thoracentesis is ongoing, and may help clarify its diagnostic utility (https://clinicaltrials.gov/show/NCT02677883).

Regardless of these results, the case for pleural manometry as a useful diagnostic tool to identify unexpandable lung is strong. Manometry provides useful information on the pathophysiology of pleural effusions that is not otherwise available. Correct interpretation of the data is contingent upon a clear understanding of basic physiologic principles without which the information provided would merely be reduced to another set of meaningless numbers. Numbers do not improve patient outcomes, informed physicians do.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

Fabien Maldonado has received an unrestricted research grant from Centurion Medical Medical Products.