Pleural Sliding During Valsalva and Muller Maneuvers

Background and aims of the study In a normally aerated lung, the only detectable structure is the pleura, visualized as a hyperechoic horizontal line. The pleural line moves synchronously with respiration: this dynamic horizontal movement is called lung sliding (1) representing movement of the two pleural layers as lungs expand during the respiratory cycle (2). Using B-mode, lung sliding appears as a pleural line oscillation from side to side which corresponds to the variation of the underlying artifacts. When M-mode is applied to the lung exam, the system displays a representation of tissue motion over time: in the case of normal lung sliding, the area on the M-mode tracing deep to the pleural line takes on a granular or "sandy" appearance while the relatively immobile superficial tissues appear smooth and linear. This has been termed "Seashore Sign," as the superficial tissues resemble waves and the deeper tissues the sandy beach (2, 3).

In point-of-care lung ultrasound, the sliding sign is of paramount importance in the diagnosis of pneumothorax (1-5), in which ultrasound has been found to be superior to chest radiography in the supine position (6-8). In fact, the presence of lung sliding in correspondence of the highest point of the hemithorax allows to rule out pneumothorax with a sensitivity close to 100% (7). However, absence of lung sliding does not necessarily imply to rule in pneumothorax (specificity 70%) (7). In fact, if lung sliding is abolished, there are several possibilities. (a)The patient is not ventilating as in cases of accidental esophageal intubation or intubation of a main-stem bronchus (9). Although sliding with respiration is not visualized, the pleural line can be seen to move slightly with each heartbeat as movement from the heart is transmitted to the pleural line. This has been termed the "lung pulse". (b)The patient is ventilating but the lung is not sliding due to a dense lobar consolidation. In these patients the underlying lung contains significant water, and as a result there will be a predominant B-line pattern. Since B-line patterns originate from the lung parenchyma, and the lung parenchyma cannot be visualized if there is air interposed between the pleura and the lung, the presence of B-lines also excludes the possibility of a pneumothorax at the interspace in question. (c)The patient is ventilating but the lung sliding is difficult to observe due to pleural adhesions, bullae, or pleurodesis. While these patients may have an associated B-line pattern, they may have an A-line pattern as well, which is why absence of lung sliding and B-lines is not sufficient to confirm the presence of a pneumothorax.

Recently, ultrasound artifacts mimicking lung sliding have been described in absence of the visceral pleura. This observation was carried out in a group of patients who underwent pneumonectomy (10). In these patients, lung sliding was clearly absent during mechanical ventilation in the operated hemithorax. During spontaneous breathing, although of lower intensity compared to the unoperated hemithorax, some artifacts mimicking lung sliding and seashore sign were found during each deep spontaneous breath (10). These findings have led to the hypothesis that these artifacts may be due to the contraction of the intercostal muscles.

The objective of the present study is to reproduce such artifacts during contraction of the inspiratory and expiratory muscles with closed glottis (Valsalva and Muller maneuvers): in this condition each subject is not ventilating and the two pleural layers are not sliding one over each other.

Materials and Methods Twenty healthy male volunteers aged between 18 and 65 years-old with no history of respiratory or cardiovascular pathologies will be enrolled. Females and all subjects with history of respiratory or cardiovascular diseases will be excluded. Volunteers will be enrolled among staff members of the Cardiac Surgical Intensive Care Unit of the "Agostino Gemelli" University Hospital in Rome. Each enrolled volunteer will be trained to perform Valsalva's and Muller's maneuvers.

Sonographic evaluation will be performed on both hemithorax in the sitting subject during the following phases lasting 10 seconds:

Apnea; Spontaneous regular breathing; Valsalva maneuver, during which each subject will be asked for performing three forced expirations with closed glottis after one forced inspiration; Muller maneuver, during which each subject will be asked for performing three forced inspirations after one forced expiration; Hyperventilation.

One certified anesthesiologist, proficient in lung ultrasound, will perform all sonographic evaluations using a MyLab Five® (Esaote, Italy) ultrasound device. A linear ultrasound transducer will be placed perpendicular to two consecutive ribs on the parasternal line of each hemithorax and the lung sliding will be evaluated using both B-mode and M-mode in each abovementioned phase lasting 10 seconds. B-mode evaluations will be registered as clips while M-mode ones will be registered as pictures.

If B-lines will be visible, their displacement with respect to the right lateral border of the sonographic image will be evaluated during Valsalva's and Muller's maneuvers and during spontaneous regular breathing.

Afterwards, images and clips registered and randomly named with a number known only to the investigator who performed sonographic exams, will be evaluated by two anesthesiologists and one radiologists all proficient in lung ultrasound. These three investigators will assign to each clip and image a score from 0 (no lung sliding) to 2 (if lung sliding is present). A score of 1 will be assigned in case of impossibility to determine presence or absence of lung sliding. In addition, each diagnostician will be asked for determining whether 'lung pulse' is easily distinguishable from lung sliding or not, assigning a score from 0 ('lung pulse' indistinguishable from lung sliding) to 1 ('lung pulse' clearly distinguishable from 'lung sliding' if present). All sonographic images and clips will be evaluated twice in order to assess reproducibility of the assessment by each investigator.

Data analysis Scores will be reported as means and intervals, or as percentages of the five possible values. Comparisons will be performed using Chi square test, Fisher test, and nonparametric tests such as Kolmogorov-Smirnov test, Kruskall-Wallis test and Friedman test.

The following hypotheses will be tested in both B-mode clips and M-mode images:

1. Apnea is associated with a score 0
2. Valsalva maneuver is associated with a score 0
3. Muller maneuver is associated with a score 0
4. Normal ventilation and hyperventilation are associated with scores > 0
5. Hyperventilation is associated with the highest score
6. Scores assigned to each condition are the same on both sides
7. Scores assigned by all investigators are similar to each other