(1)
Hôpital Ambroise Paré Service de Réanimation Médicale, Boulogne (Paris-West University), France
Severe dyspnea is one of the most distressing situations for a patient. Aiming at a therapy based on immediate diagnosis is a legitimate target.
The acute incapacity to breathe is one of the most distressing situations one can live [1]. The BLUE-protocol concentrates 18 years of efforts (mainly repeated submissions) aiming at promptly relieving these patients.
The idea of performing an ultrasound examination in time-dependent patients was not far from a blaspheme in 1985, definitely not envisageable according to the rules. Our approach possibly intrigued some doctors and nurses in the ERs of our institutions. During the management of these critical situations, time was not for quiet explanations. What the emergency doctors (who had to rush to the next patient in the overcrowded ER and eventually rushed after duty for a deserved nap, end of the story) did not fully see was that, after a few minutes, we were able to give to the nurse therapeutic options, while organizing the transfer to the ICU. And what they did not see at all (occupied by 1,000 other tasks, medical, administrative, familial, etc.: this was not time for international guidelines on lung ultrasound) was that these options were in accordance with the final diagnosis.
In the emergency setting, we use familiar tools since decades and centuries, mainly physical examination [2] and radiography [3], two basic tools, yet increasingly known for having limited precision. The crowded emergency room is not the ideal place for serene work, an acknowledged issue [4–8]. One-quarter of the patients of the BLUE-protocol in the first hour of management received erroneous or uncertain initial diagnoses, and many more received inappropriate therapy. The online document of Chest 134:117–125 details these 26 % of wrong diagnoses. CT seems a solution, but Chap. 29 will demonstrate its heavy drawbacks. One day, the community will maybe find this tool definitely too much irradiating [9].
We initiated this long work using an ADR-4000 (from 1982) then shifted for our Hitachi-405 (from 1992, last update 2008). Their 3 MHz sectorial probe and 5 MHz microconvex probe were perfectly suitable.
The Spirit of the BLUE-Protocol
Basically, the BLUE-protocol is a protocol. Yet it was designed for being a flexible one. Some protocols are possibly built for exempting doctors to think, but this one requires to keep on being a doctor. It is permanently “piloted.” In some situations, it will just confirm an obvious diagnosis. In others, it will confidently invalidate a diagnosis which looked the likely one.
For being perfectly understood (and anticipating remarks), the BLUE-protocol should be considered as an “intellectual exercise,” a tool just designed for using the minimal bunch of data for the maximal accuracy, when used alone. Countless articles are now using lung ultrasound, and many propose various algorithms including echocardiography and other items, advocating a “multiorgan approach.” This is not the spirit of our protocol: it associates these various items but does not include them (the difference is substantial). Comparing these studies with our approach would therefore make little sense. Regarding, for instance, the heart, see at the end of this chapter and at the end of Chap. 24 that we did not “forget” it (it is known that searchers may sometimes be absent minded, but up to forgetting the heart, there is a substantial step!); we just deleted it from our data. The accuracy of “BLUE-protocol plus echocardiography” is anyway featuring at end of this chapter, and we invite the readers to make an opinion.
Same remark for all clinical signs. Some academicians reproached to the BLUE-protocol to forget these precious signs [10] (don’t miss the discussion at the end of Ref. [10]). The clinical signs are, ironically, in the center of the Extended BLUE-protocol, for an improved accuracy (Chap. 35). The BLUE-profiles are here, available, up to this respected physician to integrate his favorite clinical data at will in his clinical approach.
The Design of the BLUE-Protocol
The BLUE-protocol was conceived in an observational study in a Parisian university-affiliated teaching hospital. We performed ultrasonography on admission, in the climax of dyspnea, on serial patients with acute respiratory failure. Acute respiratory failure was defined based on clinical criteria requiring admission to the ICU.
The gold standard was the final diagnosis considered in the hospitalization report, made by a medical ICU team (expert panel) who did not take into account the lung ultrasound data and used traditional approaches. Uncertain diagnoses, multiple diagnoses, and rare causes (frequency <3 %) were excluded (see Chap. 21).
After years necessary for the publication of the preliminary background (mostly lung terminology), we were able to propose the analysis of three items at the lung area – with dichotomous answer, collected at standardized points (upper and lower BLUE-points, PLAPS-point).
1.
Abolished anterior lung sliding (yes or no)
2.
Lung rockets at the anterior wall (present or absent)
3.
Alveolar and/or pleural syndrome (called PLAPS if posterior or/and lateral) (yes or no)
We added an adapted venous analysis (indicated in 54 % of cases). Note that the venous analysis takes the major time (2 min of a 3 min examination), which is nonetheless short, since we use a simple machine with fast start-up, the same microconvex probe, time-saving maneuvers, only one setting, and a contact product which allows major time savings. This will be repeated again, intentionally.
The BLUE-Profiles: How Many in the BLUE-Protocol?
A work of a “profiler,” based on analysis of hundreds of pre-data, was done during 7 years. From this observational work, the profiles of the BLUE-protocol were defined (Fig. 20.1).
Fig. 20.1
The decision tree of the BLUE-protocol. A decision tree using lung and venous ultrasound to guide the diagnosis of acute respiratory failure: the BLUE-protocol (Adapted from Lichtenstein and Mezière [11], with the authorization of Chest)
There are eight profiles. The anterior analysis, which initiates the BLUE-protocol, can describe six situations. Five of them conclude the protocol: the B-, B′-, A/B-, C-, and A′-profile. One of them is the A-profile (Video 10.1). The A-profile designates an anterior chest wall with predominant A-lines and lung sliding. The A-profile opens to three more profiles (A-DVT, A-no-V-PLAPS, and nude profile).
Here are the profiles, which were assimilated to specific diseases. The term “profile” assumes an association of signs (two), plus a location (Fig. 20.2).
Fig. 20.2
Regular distributions. The main regular profiles of the BLUE-protocol. Note that this figure uses a particular representation of the B-profile and the B′-profile in static images: on M-mode, lung sliding is materialized by this succession of vertical white and black stripes (it reminds real-time images done with vascular probes), since the B-lines come and go through the shooting line of the M-mode. Abolished lung sliding generates a homogeneous MM-space in M-mode: hyperechoic if the shooting line strings a B-line (like here) and hypoechoic if done between two B-lines. Note for the C-profile than only one point is required. The A′-profile does not feature for space management (already dealt with)
1.
The A-profile plus DVT was assimilated to pulmonary embolism. The term “DVT” or “no DVT” was fully detailed in Chap. 18, because it is integrated in a specific, adapted protocol.
2.
The A-no-V-PLAPS profile is a temporary label which designates an A-profile with no DVT and with a PLAPS (uni- or bilateral). Called in some of our articles the A-V-PLAPS-profile, it is now slightly longer but more logical, thus hopingly easier to remember. When “A-no-V-PLAPS” is spelled slowly, we can understand “A,” i.e., no pneumothorax and no pulmonary edema; then “no V,” i.e., schematically, pulmonary embolism unlikely; and then “PLAPS,” making at this step COPD/asthma unlikely. The A-no-V-PLAPS profile was assimilated to pneumonia.
3.
The nude profile is a normal profile, i.e., A-profile with no DVT and no PLAPS. It was assimilated to asthma or COPD, two bronchial diseases put together because of a same origin (bronchial obstruction), a roughly same therapy, and a same pathophysiological absence of interstitial, alveolar, pleural, or venous signs.
4.
The B-profile designates anterior predominant bilateral lung rockets associated with lung sliding (Video 13.1). It was assimilated to hemodynamic pulmonary edema.
5.
The B′-profile is a B-profile with abolished lung sliding (Video 13.2). It was assimilated to pneumonia.
6.
The A-/B-profile designates anterior predominant lung rockets at one side and predominant A-lines at the other. It was assimilated to pneumonia.
7.
The C-profile designates anterior lung consolidation, regardless of size and number. The C-profile was assimilated to pneumonia.
8.
The A′-profile is an A-profile with abolished lung sliding (Video 14.2). When a lung point was associated, it was assimilated to pneumothorax.
Once these profiles were predefined as written, the study could begin. We then assessed the concordance between profiles and diseases.
Some Terminology Rules
We specify the precise language used in the BLUE-protocol for enabling other teams to reproduce our results.
When the first of the four anterior BLUE-points shows lung sliding with A-lines, labeling it a “quarter of A-profile” indicates that the user has understood that the “A-profile” is defined on the four anterior points. We prefer to read that a given patient had “four quarters of B-profile” (i.e., a B-profile, clearly expressed).
One of the four anterior points with a lung consolidation, even minute (C-line), makes a C-profile.
One isolated B-line visible at all four anterior BLUE-points: this is such a rare pattern that we do not know its clinical relevance. We should temporarily consider this profile as an A-profile.
Some profiles should not generate too much troubles (Fig. 20.3).
Fig. 20.3
Atypical distributions. Some atypical distributions of anterior lung rockets
A quarter of B-profile visible on three of the four anterior BLUE-points makes sensu stricto “three-quarters of a B-profile.” It should probably be linked to a B-profile.
A quarter of B-profile at the right upper BLUE-point with a quarter of B-profile at the left lower BLUE-point and a quarter of B-profile at the right upper BLUE-point with a quarter of B-profile at the left upper BLUE-point are rare profiles, rare enough for not having been seen in the BLUE-protocol. We think wise and logical to link such profiles to an irregular A/B-profile (much more than a B-profile) – suggesting pneumonia/ARDS.
Two-quarters of B-profile at the two lower BLUE-points: this profile, seen in 5 % of cases of hemodynamic pulmonary edema, should probably be considered as an irregular B-profile (patient under beginning of depletive therapy?).
Three-quarters of A-profile with one-quarter of B-profile must be assimilated to an A-profile. This is usually a pneumonia, which will be recognized using the long sequence of the BLUE-protocol: no venous thrombosis and a PLAPS usually present: “A”-no-V-PLAPS-profile.
The Results
At the submission of the manuscript, 302 patients were analyzed. After exclusion of 16 patients for unknown diagnosis, 16 for double diagnosis, and 9 for rare diagnosis, 260 dyspneic patients with one definite diagnosis were considered. The main causes of acute respiratory failure seen in our walls were pneumonia (31 %), pulmonary edema (24 %), decompensated COPD without cause (18 %), severe asthma (12 %), pulmonary embolism (8 %), and pneumothorax (3 %). Table 20.1 details our results.
Table 20.1
Accuracy of the BLUE-protocol
Mechanism of dyspnea | Profiles of BLUE-protocol | Sensitivity | Specificity | Positive predictive value | Negative predictive value |
|---|---|---|---|---|---|
Acute hemodynamic pulmonary edema | B–profile
Related posts:
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|
