The feasibility of nurse practitioner-performed, telementored lung telesonography with remote physician guidance - ‘a remote virtual mentor’

A RTMUS system was composed of easily accessible off-the-shelf hardware:

Using this RTMUS system, the remote mentor was able to view the patient and examiner's hands holding the ultrasound probe while simultaneously viewing the resultant ultrasound image in the same field of view displayed over Skype. Both the examiner and the mentor could verbally communicate over Skype. All sessions were conducted with the mentor using a laptop computer (Hewlett Packard ProBook 4520 s, Hewlett Packard, Mississauga, Ontario, Canada), through which the mentor could be viewed through the mentor's embedded web camera. After the initial introductions, however, the mentor's video display was deactivated to increase connectivity speed.

All examinations were conducted after TT removal and prior (and thus blind) to CXR. All examinations were conducted over Skype, with the involved nurse initiating a video call once connectivity had been established following secure log-in. All actual examinations were conducted by a credentialed NP except for the final examination, which was conducted by the patient's assigned registered nurse. Neither nurse had any formal ultrasound training or prior ultrasound experience, while the remote mentor had extensive experience as a clinical ultrasonographer and teacher. The remote mentor was neither informed of the clinical history of the patient being examined nor of the side from which the TT had been removed.

Using lung ultrasound to exclude PTX relies on the concept that if movement of the visceral upon the parietal pleura is seen, then the two pleural surfaces are contiguous and thus cannot be separated by air. Therefore, by definition, PTX is excluded. The mentor therefore guided the nurse to manipulate the probe to demonstrate the ‘bat sign of Lichtenstein’ (Figure 1) representing the pleural interface of the visceral and the parietal pleura beneath two contiguous rib shadows [8, 9]. Thereafter, the mentor and nurse collaboratively aimed to discern whether lung sliding was present, utilizing the diagnostic algorithms for pneumothorax detection promulgated by the World International Network Focused on Critical Ultrasound [10]. If present, this was documented through capturing still images demonstrating a color power Doppler signal at the pleural interface (power-slide sign) [11]. After the NanoMaxx ultrasound machine underwent a software upgrade in January 2012, allowing the additional modality of time-motion mode, the examination included documenting visible sliding through capture of a typical ‘seashore’ image using M-mode [9, 12]. If typical sliding was not seen during the examination, its absence was attempted to be documented through the archiving of an absent power-slide [11, 12] at the pleural interface and through capture of the ‘stratosphere sign’ [9] after the software upgrade. With absent lung sliding, the lung-point sign was also sought as a definitive sign of PTX [9, 13]. Subcutaneous emphysema was also sought and defined as the presence of typical ultrasonographic air detected at an anatomic location superior to the pleural interface. Further signs sought to enable detection or exclusion of PTXs were B-lines and the lung pulse [10, 14]. At the completion of each examination, the mentor and nurse discussed the totality of the findings and made and documented a clinical ‘call’ concerning the presence or absence of a post-TT removal PTX (Table 1). Formal examination times were not recorded. The study was conducted under the auspices of the Conjoint Ethics Board (Ethics IDs 17239 and 20949).

Ultrasound findings on all lung fields were tabulated as positive or negative for PTXs and compared to the subsequent upright CXR report as the reference standard. Dichotomous data were analyzed using Stata version 12.0 (Stata Corp., College Station, TX, USA) by calculating rates of true positives, true negatives, sensitivity, and specificity.

Thirteen patients (26 lung fields) were examined. The remote exam was possible in all cases with ultimately good connectivity including one trans-Atlantic interpretation (Table 1). Compared to the subsequent upright CXR report as the perceived reference standard, there were 4 true-positive remotely diagnosed PTXs, 2 false-negative diagnoses, and 20 true-negative diagnoses for 66% sensitivity (95% confidence interval (CI) 0.22 to 0.96), 100% specificity (95% CI 0.59 to 1.00), 100% positive predictive value (95% CI 0.40 to 0.97), and 92% accuracy for remotely guided chest examination. The likelihood ratio of a positive test was infinite, and the likelihood ratio of a negative test was 0.33 (95% CI 0.11 to 1.03). No patient, including those diagnosed with a post-TT removal PTX, required re-insertion of a pleural drain.

Specific cases are illustrative and novel, constituting the first reported examples of these techniques to our knowledge. There were two false-negative examinations. In case 1, the final radiology report noted ‘a tiny residual left apical pneumothorax’ on CXR (Figure 2). In case 7 which was conducted across the Atlantic Ocean, the final CXR report was of ‘a small focal lucency noted at the right apex that could represent a small loculated pneumothorax,’ noting this was opposite from the side of TT removal (Figure 3). In both these cases, the remote expert did not appreciate a PTX as there appeared to be pleural apposition and lung sliding (Figures 4 and 5). Case 7 involved the greatest geographic separation with the examiner and patient being in Calgary, while the mentor was online in Heathrow, England (7,058 km distant). In this case, obvious lung sliding was not apparent, but the final clinical call was ‘no true sliding seen but comet-tails and lung pulse - designated No PTX,’ suggesting a possible error in interpretation rather than remote imaging accuracy.

There were also four true-positive examinations, in which abnormalities of pleural sliding were correctly diagnosed, corresponding to four reported post-TT removal CXRs with a 15-mm pleural separation (Figure 6; case 2), a trace PTX with 5 to 6 mm of separation (case 8), an 8-mm pleural separation (case 10), and a ‘tiny’ PTX (case 13). The last diagnosis constituted a remarkable case in which a registered nurse who had no prior ultrasound experience and had never picked up an ultrasound probe previously was mentored to detect the subtle and tiny PTX which was denoted by clear delineation of a lung point at the lung apex. This single case has thus been reported separately [15]. The lack of sliding was confirmed both visually and through M-mode interrogation. There appeared to be a persistent junction between areas where the lung was clearly sliding with respiration and not on either side of the hyperlucent pleural line interpreted to represent a small focus of intrapleural air, which is by definition a loculated PTX. This description of a localized small PTX has recently been corroborated and termed the ‘double lung point.’