Content uploaded by Alain Boussuges
Author content
All content in this area was uploaded by Alain Boussuges on Mar 24, 2016
Content may be subject to copyright.
1
CHAPTER 6: BUBBLE MEASUREMENT TECHNIQUES
Authors: Alain Boussuges, Frauke Tillmans
Take home message
· Ultrasonic techniques allow the visualisation of microbubbles in the heart and blood
vessels after diving
· Techniques to detect bubbles include Continuous Wave Doppler, Transthoracic
Echocardiography, Pulsed Doppler Detection, and Transesophageal Echocardiography
· Counting and grading the amount of bubbles following diving is a powerful tool to
validate current dive tables and algorithms, assess the effect of different preconditioning
techniques, and therefore increase diving safety
· There are different counting and scoring scales to categorize bubble quantities
· Although useful for research applications, counting bubbles has its limitations in clinical
application and therefore doubtable diagnostic value for the individual DCS risk
assessment.
Abstract
Nitrogen desaturation in SCUBA divers leads to the production of circulating gas bubbles
during the decompression period and over the ensuing minutes. Since the 1970s and the
studies of Guillerm and Masurel in France, and Smith and Spencer in USA, these bubbles
have been detected using ultrasonographic techniques. A system of classification to assess the
quantity of circulating bubbles has been developed by several research groups. The bubble
scores or scales are based on the number of bubbles with respect to the cardiac cycle,
commonly referred to as heartbeat. Several studies have investigated the correlation between
the bubble rating and the probability of a decompression accident. Although the significance
of this correlation is debated, all of these studies agree that the risk of DCI is low in the
absence of circulating bubbles. Bubble screening can therefore mainly be used as a safety
indicator for validating diving profiles. A decompression profile that does not induce the
production of bubbles in a large population of divers may be considered a safe profile. The
main quality of a bubble screening procedure is therefore its sensitivity. On the other hand,
2
echo techniques have a low specificity to predict DCS. Indeed circulating bubbles are
frequently observed after diving in subjects without any resulting clinical disorders. In this
respect, in addition to avoiding DCS, the aim here is to limiting decompression stress
evaluated against circulating bubbles post dive imaged using ultrasound. The aim of this
chapter is to discuss various techniques used to screen circulating bubbles, their applications,
and their limitations.
1. Ultrasonographic techniques used to screen for circulatory bubbles
Ultrasonographic techniques are also widely referred to as ultrasound, sono, echography or
simply echo. The idea behind this method is to use high frequency sound waves and their
echoes to locate and display organs or tissues in the human body. This technique is
comparable to SONAR used in submarines or the echolocation of whales. The frequencies
used in medical diagnostics, treatment and follow-up of patients are ultrasounds, ranging
between 2 and 18MHz (sounds above the threshold of human hearing) depending on the
target organ or tissue. The following techniques have been proved suitable to screen
circulating bubbles following a dive.
1.1. Continuous Wave Doppler (CW Doppler)
Continuous Wave Doppler with “blind” positioning of the transducer (i.e. without “seeing”
the blood vessel over which the probe is placed) was the first method used to screen
circulating bubbles. This procedure has been extensively studied in the 70's by Spencer. The
signal is recorded as an audio-file, which can then be evaluated by different observers.
Several anatomical sites can be explored including the pulmonary artery, subclavian vein, and
the lower vena cava. The pulmonary artery outflow tract is the most appropriate site for
studying bubble quantities induced by a dive, as all of the venous blood passes here. The
studies of Spencer, Kisman and Masurel using the Doppler effect resulted in classification
systems for quantifying the circulating bubbles (see Table 1 and 2). The bubble grades rate
the number and frequency of bubbles compared with the heartbeat. The Kisman-Masurel code
is somewhat more complicated than the Spencer scale, as it is composed of three parameters
(frequency, loudness, duration) in various combinations (see Table 2). However, a
“conversion” scale from KM code to Spencer scale has been developed. The limitation of CW
Doppler is the bubble detection sensitivity (diameter of detectable bubbles) and the fact that it
3
takes a lot of training to interpret the bubble signals correctly. Therefore, there may be a
variation in results according to the experience of the observer. Consequently, interpretation
of statistical results, especially when comparing several series, must be taken with caution.
GRADE
Auditory Assessment of Continous Wave Doppler Recordings
Grade 0
Complete lack of bubble signals
Grade 1
Occasional bubble signals, the great majority of cardiac cycles are free of
bubble signals
Grade 2
Many, but less than half of the cardiac cycles contain bubble signals, singly
or in groups.
Grade 3
All of the cardiac cycles contain bubble signals, but not overriding the
cardiac signals.
Grade 4
Bubble signals sounding continuously throughout systole and diastole of
every cardiac cycle, and overriding the normal cardiac signals.
Table 1. Grading with the Spencer scale (first established in 1974)
CODE
FREQUENCY
Bubbles per
cardiac cycle
PERCENTAGE
Cardiac cycles at
rest
DURATION
Movement, duration
cardiac cycles
AMPLITUDE
0 0 0 0 No bubbles
discernible
1
1
-
2
1
-
10
1
-
2
Barely perceptible
2 Several, 3-8 10-50 3-5 Moderate
amplitude
4
3
Rolling, 9+
50
-
99
6
-
10
Loud
4 Continuous
sound 100 10+ Maximal
Table 2. Kisman-Masurel code (originally published in 1983)
1.2. Transthoracic echocardiography (TTE)
Transthoracic echocardiography is a good tool for screening and viewing circulating bubbles
in the heart chambers. Gas bubbles appear as high intensity blobs on the images. 2D
echocardiography is currently the most frequently used modality for screening circulating
bubbles. Images can be obtained from the parasternal view (long axis and short axis of the
heart) while the diver is lying on his left side (see below), and from an apical four chamber
view (Figure 1 and 2), where the diver is lying on his back. These recordings can be
performed at rest, during movement, or during a contraction of the quadriceps or other big
muscles to dislodge bubbles that did remain stuck at blood vessel walls and then start
circulating.
Bubbles
(right cavities)
RV
LV
RA LA
5
Figure 1. Apical four chamber view: Circulating bubbles in the right cavities of the heart
(note: ultrasonic recordings result in a mirror inversion of right and left side of the heart)
Figure 2. Apical four chamber view: Circulating bubbles in the right cardiac cavities with
passage in the left cavities
In 2007, Brubakk & Eftedal have proposed and validated a scoring scale which is easily
learned and executed (Table 3).
GRADE
Visual Observation in Echocardiograph Recordings
(Images)
Grade 0
No
observable bubbles
Grade 1
Occasional bubbles
Grade 2
At least 1 bubble every 4
cardiac
cycles
Grade 3
At least 1 bubble every
cardiac
cycle
Bubbles
(right cavities)
Bubbles
(left cavities)
RV
LV
RA
LA
6
Grade 4
At least 1 bubble per cm
2
in every image
Grade 5
“White
-
out”, single bubbles cannot be discriminated
Table 3. Brubakk & Eftedal scale (2007)
7
Very recently, a new counting method was developed and validated by Germonpre et al.
(2014) whereby actual bubble signals are counted in standardised TTE image sequences. This
method permits to average bubble signals in the right heart cavities over 10-15 heartbeats,
analysing only a few frames per heartbeat (just before the contraction of the right ventricle,
see Figure 3). The result is a quantitative, not a categorized figure which may be used to
evaluate decompression stress in medium-to-low bubble dives, with very good agreement
between raters even for non-medical personnel. Although this technique requires a rigorous
fixed echocardiographic visualisation and is thus more difficult to learn than a “standard” 2D
evaluation, it opens the possibility for automatic “bubble counting” by intelligent computer
learning algorithms. These are actually in development showing as good an agreement with
humans as human to human raters on the same videos as the Germonpre et al. 2014 and will
permit the continuous counting of bubbles in a one to 2 hour period after a dive. Integration of
the counted bubbles would then become a useful tool for evaluation of the “decompression
stress” and the influence of a modified dive profile or of a pre-dive intervention (pre-
conditioning).
Figure 3. Frame-based bubble counting (artificial gold colouring used to improved contrast).
Manual counting, bubbles outlined in green.
1.3. Pulsed Wave Doppler detection
8
Circulating bubbles can also be detected using the so-called pulsed-wave (PW) Doppler
effect. It is based on the same technique as the CW Doppler, but the probe will not send a
constant signal (continuous wave) but intermitted signals (pulsed) whose echoes are then
recorded back. Both techniques have different applications in diagnostics. The transducer can
be positioned in a blind manner and the investigation methods are similar to those described
for continuous wave Doppler. Furthermore, two-dimensional echocardiography can be used to
guide the Pulsed Doppler study of the pulmonary artery blood flow (Figure 4).
Figure 4. Parasternal short axis view: study of pulmonary blood flow by pulsed Doppler. The
sample volume (between the two markers on the dotted line) is placed in the outflow area of
the right ventricle just below the pulmonary valve. Ao: Aorta, RA: Right atrium, RV: Right
ventricle, PA: Pulmonary artery
The sample volume is placed in the outflow area of the right ventricle. Circulating bubbles are
visualized in the flow spectrum as bright spots (Figure 5 and 6). A specific
echocardiographic and pulsed Doppler bubble grade scale has been derived from the scales of
Spencer (Table 4).
9
Grade 0
No detectable bubble signal (2D echocardiography and pulsed Doppler
ultrasonography)
Grade 1
Occasional bubbles; the great majority of cardiac cycles are free of bubbles (2D
echocardiography and pulsed Doppler ultrasonography)
Grade 2
Flow of bubbles (2D echocardiography); many but less than half of the cardiac
periods contain bubble signals singularly or in group (pulsed Doppler
ultrasonography)
Grade 3
Flow of bubbles (2D echocardiography); the majority of the cardiac periods
contain bubble signals singularly or in group (pulsed Doppler ultrasonography)
Grade 4
Bubbles fill cardiac chambers (2D echocardiography); all the cardiac periods
contain bubble signals in group (pulsed Doppler ultrasonography)
Table 4. 2D Echocardiographic and Pulsed Doppler Grade (Boussuges 1998)
10
Figure 5 Normal pulmonary blood flow as shown by PW Doppler
Figure 6 Circulating bubbles detected in the pulmonary blood flow: the arrows indicate
bubble signals in each cycle overriding the pulmonary blood flow
11
1.4. Transoesophageal echocardiography (TOE, am. engl. TEE)
Transoesophageal 2D mode ultrasonography is a so-called “invasive” procedure in which the
transducer is inserted into the mouth and oesophagus of the subject (Figure 7). The image is
generally more accurate than transthoracic echography because of very limited tissue
interference between the transducer and the heart; in the latter, ribs and chest muscles produce
artefacts and the cardiac structures are much more distant from the probe. Some major
limitations of the transoesophagal imaging are that it is uncomfortable and more time
consuming than non-invasive techniques, it needs medical personnel on site and requires the
subject not to eat or drink several hours before the screening. Although it has been shown to
be useful in screening bubbles, its actual use is thus limited to experimental protocols in
anaesthetised animals.
12
Figure 7. Possible Positioning of the transducer for TTE: apical view (A) and
parasternal view (B) and Transesophagal echocardiography (C)
A
B
C
13
2. Comparative studies
Some studies have compared the various techniques established to screen circulating bubbles.
Norwegian scientists for instance found that researchers would come to the same conclusion
on the quantity of circulating bubbles in the image grading system for 2D echocardiography
and the Spencer code for Doppler signals in resting subjects. However the results between the
two techniques derived strongly from one another if the screening was done on moving
subjects. In divers with poor quality imaging (due to movement or body composition), 2D
ultrasonography and Doppler methods can be considered as complementary. It has been
reported that pulsed Doppler guided by 2D imaging is the superior method for the detection of
venous gas emboli compared to 2D echocardiography alone. In a comparison of CW Doppler
with blind positioning of the transducer and pulsed Doppler guided by 2D-echocardiography,
both methods were found to be equivalent in terms of sensitivity when the ultrasound
exploration is easy to perform and yields a good Doppler signal (i.e. in young athletic divers).
In cases where the examination is more difficult to perform, 2D images guided by pulsed
Doppler are more precise. Using imaging to guide the pulsed wave Doppler facilitates good
quality recordings of the blood flow in the pulmonary artery and optimizes the screening
procedure for circulating bubbles.
3. Application of circulating bubble screening-recommendations, interest, limitations
After a dive, the desaturation process may last several hours with an estimated maximum rate
of circulating bubble detection occurring around 30-40 minutes after surfacing. Continuous or
repeated detections performed at intervals every few minutes are needed to estimate the whole
bubble quantity and maximal bubble grade. Furthermore, circulating bubble grade may vary
over a very short time frame. An automated assessment method based on detecting and
counting bubbles in real time would assist analysis and advances in this field are on-going.
Recordings must not exceed one minute in a resting diver to limit bias caused by the subject
or the examiner. It is known that circulating bubbles can sometimes pass into the arterial
circulation via a right-to-left shunt, such as the inter-atrial PFO or pulmonary shunts.
Consequently, an apical four-chamber view (see Figure 1 and 2) and/or aortic blood flow
14
recording should be systematically performed to detect circulating bubbles in the left (arterial)
cavities. To improve the significance of the findings, the standard protocol is to record the
explorations as a video file for later analysis by two independent investigators. Bubble grades
provide only a semi-quantitative evaluation of the number of bubbles due to the categorical
nature of the data. This means for example grade 2 does not necessarily show twice as many
bubbles as grade 1 and bubbles do not appear regularly in respect to time interval and
quantity. Because of these limitations, a sort of linearization such as the Kisman Integrated
Severity Score (KISS) has been proposed, where the results of three or four measurements at
various time intervals are integrated into a single number. These mathematical methods
facilitate the statistical treatment of these data but it should be kept in mind that they only
provide an estimate of the circulating bubble production.
Since the 80’s, there have been quite a few studies that reported a correlation between
detected bubble quantity and DCS. Maximal observed bubble grade has also been considered
as an estimate for decompression stress. The interpretations of these studies are debated
because correlations between high bubble grade and DCS risk are more frequently based on
dry hyperbaric chamber dives and less frequently on real open sea diving. Furthermore, the
clinical disorders reported in the chamber are articular pain or cutaneous manifestations and
the incidence of these disorders seems to be particularly elevated in comparison with real
SCUBA diving. Finally, high bubble grades (bubbles throughout the cardiac cycle) are
frequently observed in SCUBA divers showing no clinical symptoms of DCS; and perhaps
even more astonishing are recordings from asymptomatic divers which clearly show bubbles
in the left cavities. On the other hand, when few bubbles are detected during the whole
desaturation period, the risk of DCS seems to be low but not nil. Limits of the interpretation
of circulating bubble detection are also supported by the studies in altitude chambers. Indeed,
women formed fewer nitrogen bubbles than men for a given exposure, but still seemed to
have a higher incidence of altitude decompression sickness.
These observations indicate that although bubbles are commonly considered to be the trigger
of DCS, some clinical manifestations of DCS are secondary to non-circulating tissue bubbles,
are often related to inflammation and it is yet not completely understood which observations
linked to decompression stress are caused by, and which are independent of (detectable)
bubbles. Therefore, in an individual case of a diver with symptoms, detection or not of
circulating bubbles has no real diagnostic value.
15
Conclusions
Circulating bubble detection provides a valuable tool for evaluation of the safety of a dive
profile in a large population. Calculating the time integral of bubbles during the whole post-
dive observation period would give an estimate of the total volume of intravascular gas. For a
large population, when few bubbles are detected the dive profile can be considered safe. On
the other hand, the use of circulating bubble detection in individual decompression stress or
DCS risk in small populations is debatable. As detection techniques for circulating bubbles
are getting more and more technologically advanced, these conclusions may (or may not)
change in the future.
16
Related Sources
· M.F. Gillis, P. Petersen, M.T. Karangianes (1968) In vivo detection of circulating
gas emboli with decompression sickness using the Doppler flowmeter. Nature, 217,
965-967.
· M.P. Spencer (1976) Decompression limits for compressed air determined by
ultrasonically detected blood bubbles. J Appl Physiol, 40(2), 229-235.
· T.W. Beck, S. Daniel, W.D.M. Paton, E.B. Smith (1978) Detection of bubbles in
decompression sickness. Nature, 276, 173-174.
· A. Boussuges, D. Carturan, P. Ambrosi, G. Habib, J.M. Sainty, R. Luccioni
(1998) Decompression induced venous gas emboli in sport diving : detection with 2D
echocardiography and pulsed Doppler. Int J Sports Med, 19(1), 7-11.
· J.C. Buckey, D.A.Knaus, D.L. Alvarenga, M.A. Kenton, P.J. Magari (2005) Dual-
frequency ultrasound for detecting and sizing bubbles. Acta Astronautica, 56, 1041-
47.
· G. Swan, B. D. Bollinger, T. G. Donoghue, J.C. Wilbur, S. D. Phillips, D.L.,
Alvarenga, D. A. Knaus, P.J. Magari, and J. C. Buckey (2011) Microbubble
detection following hyperbaric chamber dives using Dual–Frequency ultrasound. J
Appl Physiol 111(5): 1323-1328.
· L. Blogg, M. Gennser (2011) The need for optimisation of post-dive ultrasound
monitoring to properly evaluate the evolution of venous gas emboli. Diving Hyperb
Med, 41(3), 139-146
· A. Brubakk, T. Neumann (2003) Bennett and Elliotts’ physiology and medicine of
diving, 5th ed., Saunders Ltd.
· Parlak IB, Egi SM, Ademoglu A, Germonpre P, Esen OB, Marroni A, Balestra C.
(2014) Bubble stream reveals functionality of the right-to-left shunt: Detection of a
potential source for air embolism. Ultrasound in Medicine & Biology 40, 330-340.
· Eftedal OS (2007) Ultrasonic detection of decompression induced vascular
microbubbles. thesis for the degree of doctor philosophiae, NTNU, Norway
· Germonpre P, Papadopoulou V, Hemelryck W, Obeid G, Eckersley RJ, Tang
MX, Balestra C. (2014) The use of portable 2D echocardiography and "frame-based"
bubble counting as a tool to evaluate diving decompression stress. Diving Hyperb Med
44, 5-13.
