Variation of Diagnostic Ultrasound Induced Pulmonary Capillary Hemorrhage with Fraction of Inspired Oxygen

Abstract

Pulmonary capillary hemorrhage (PCH) induction by diagnostic ultrasound (DUS) was investigated for the influence of the fraction of inspired oxygen (FiO₂). Sprague Dawley rats were anesthetized with Telazol Only (TO) or of Telazol plus Xylazine (TX), which can enhance the DUS-PCH. A linear array probe (10L, GE Vivid 7 Dimension) was used in B mode at 7.5 MHz to expose the right lung. FiO₂ of 10%, 20%, 60% and 100% were delivered by a nose cone. The ultrasound images showed the PCH effect as growing comet tail (B-line) artifacts and also revealed subpleural consolidated segments (SCS) for higher FiO₂. PCH for TO with 20% and 60% FiO₂ were significantly greater (P<0.05) than for the 10% FiO₂. PCH for TX with 10% and 20% FiO₂ were significantly greater (P<0.02) than for TO anesthesia. Added xylazine and high % FiO₂ reduced PCH thresholds, but xylazine plus high % FiO₂ together did not lower the PCH threshold further. The lowest threshold observed of 1.4 MPa, corresponding to an in situ Mechanical Index = 0.5.

Introduction

Diagnostic ultrasound is commonly used for clinical pulmonary examinations (Ahmad and Eisen, 2015; Dietrich et al. 2017). Pulmonary ultrasound is valuable in the diagnosis of pneumonia, pulmonary edema, pulmonary embolism, atelectasis, diffuse parenchymal disease, respiratory distress syndrome, and lung cancer (Sartori and Tombesi (2010). The ultrasound images of the lung surface often are examined for the presence and configuration of comet tail (B line) artifacts (CTA), which extend inward from the bright reflection of the pleura. The B lines are indicative of edema or interstitial lung disease (Copetti et al. 2008; Ahmad and Eisen, 2015), and the analysis of the CTA is comparable in diagnostic efficacy with CT findings (Martelius et al. 2016). The assessment of CTA and other image features are valuable in neonatal examinations for diagnosis of respiratory distress syndrome (Chen et al 2017), assessing surfactant treatment (Oktem et al 2019), and pneumothorax (Liu et al 2017). Another sign known as the “shred sign”, is indicative of pulmonary consolidation in localized regions (Lichtenstein, 2017). This may result from fluid in a group of alveoli and can be seen, for example, in neonatal pulmonary exams with lung fluid (Chen et al. 2017; Liu et al 2019), including pulmonary hemorrhage in the newborn (Ren et al 2017).

The ultrasound exposure of lung also has a safety issue: induction of pulmonary capillary hemorrhage (PCH). The induction of PCH by pulsed ultrasound in animal models was first reported in 1990 (Child et al 1990), and this biological effect has received significant research attention (Church et al 2008). We have recently studied PCH induction by diagnostic B mode ultrasound imaging in rats (Miller, 2012). The ongoing induction of the capillary hemorrhage bioeffect was evident in the diagnostic ultrasound image as growing CTA at the pleural surface. These CTA project inward perpendicular to the lung surface image, giving essentially the same qualitative indication used for diagnosis of pulmonary edema and interstitial edema. This coincidence between a clinical image diagnostic sign, and a biological effect indication, presents an important uncertainty in diagnostic ultrasound safety. Sonographers conceivably could inadvertently induce PCH while examining some patients at a diagnostic ultrasound Mechanical Index (MI). Clinically, the image signs of PCH potentially could be misinterpreted as a diagnostic indication of a condition investigated by the exam and not recognized as a pulmonary ultrasound safety issue.

The PCH effect of diagnostic ultrasound follows a threshold exposure-response with dosimetric parameters including peak rarefaction pressure amplitude (PRPA), MI or pulse average intensity. The effect also depends on physical parameters such as exposure duration (Miller et al 2016a), ultrasound mode (e. g. scanned B mode versus fixed M mode) (Miller et al. 2018a) but has little dependence on ultrasound frequency (Miller et al. 2015a). The thresholds and magnitudes also depend on animal physiology, such as age (Dalecki et al. 1997; O’Brien et al. 2003). Specific anesthetics (Miller et al. 2015b) and even common patient sedatives (Miller et al. 2016b) can enhance the susceptibility to PCH, while infusion of intravenous (IV) fluids, such as normal saline can reduce susceptibility (Miller et al 2018b).

Small perturbations on the air-side of the blood-air barrier appear to have a strong influence on ultrasound induced PCH, comparable to those of perturbations on the blood-side. On the air side, scanned anesthetized rats which had intubation for intermittent positive pressure ventilation, showed varied susceptibility for varied pressures (Miller et al. (2018c). When rather small end expiratory pressures of ±4 cm H₂O were applied, the PCH was virtually eliminated by positive pressure, but strongly enhanced by negative pressure. Similar important variations in diagnostic ultrasound induced PCH may also be caused by other medical interventions or treatments, which have yet to be studied.

A common clinical intervention is the enhancement of inhaled oxygen for patient assistance. A reduced oxygen fraction can also occur in some patients leading to hypoxia. The reduced oxygen and air pressure at high altitudes can injure lung, and analysis of B lines aids in characterization of high altitude pulmonary edema (Fagenholz et al. 2007). Conversely, high fractional inspired oxygen (FiO₂) can be used for therapeutic intervention in many patients especially during surgery anesthesia (Edmark et al 2003; Baum, 2004). High FiO₂ can produce atelectasis (Magnusson and Spahn, 2003) and a moderate FiO₂ may be optimal (Agarwal et al. 2002; Kabon and Kurz, 2006). Atelectasis can typically be reversed by positive pressure ventilation and positive peek end expiratory pressure (Kabon and Kurz, 2006). These considerations suggest that varied FiO₂ similar to the clinical situations might be an important factor for perturbing the susceptibility of lung to ultrasonic PCH. This study was designed to investigate the variation of the diagnostic ultrasound induction of PCH and exposure-response thresholds with variation in FiO₂.

Methods

Animal preparation

All in vivo animal procedures were conducted with the approval and guidance of the Institutional Animal Care and Use Committee, University of Michigan, Ann Arbor, MI. Female Sprague Dawley rats (Charles River, Wilmington, MA, USA) were used for this research and maintained in the animal housing rooms of the Unit of Laboratory Animal Medicine. The rats weighed an average of 242 ± 13 g at the time of testing. Each rat was anesthetized with an intraperitoneal (IP) injection either of Telazol (Zoetis Inc., Kalamazoo, MI USA) 90 mg/kg only (TO) or of Telazol plus xylazine (XylaMed™ xylazine injection, MWI, Boise, ID, USA) 9 mg/kg IP (TX). The two different anesthetic techniques were used to explore influence of FiO₂ for both the relatively low and high PCH bioeffects obtained without and with added xylazine, as described previously (Miller et al. 2018b). The right thorax of each rat was shaved and depilated for ultrasound transmission. The rats were mounted on a plastic board fitted with a nose cone connected to an anesthesia machine (Surgivet Isotec 4, Smiths Medical ASD Inc., St. Paul, MN, USA). The inhalation anesthetic was turned off, so that the anesthesia machine only served as a calibrated means to deliver the desired inspiration gases without mechanical ventilation. For exposure, the rat and mounting board was aligned vertically in a 38 °C water bath. The water bath setup allowed precise aiming of the diagnostic ultrasound exposure probe, and maintained the body temperature of the anesthetized rats.

Ultrasound

The diagnostic ultrasound machine for this study was a clinical GE Vivid 7 Dimension (GE Vingmed Ultrasound AS, Horten, Norway) with the 10L linear array probe. The machine was used in B mode at 7.5 MHz frequency with 3 cm image depth and single focus at 1.3 cm depth. The probe was held with a gimbal mount in the heated water bath filled with vacuum degassed water. The ultrasound pulse parameters were measured at the positon of the maximum PRPA (1.3 cm depth on the probe axis) using a calibrated hydrophone with a 0.2 mm sensitive spot (model HGL-0200, Onda Corp., Sunnyvale, CA) and are listed in Table 1. The measured frequency was 7.3 MHz, with 60.2 frames per second and 96 μs pulse repetition period with 270 ns pulse duration. The −6 dB width of the beam was 1.2 mm. The digitized pulse pressure waveform, measured in water, was derated by an attenuation factor of 1.2 dB/cm-MHz (Miller et al. 2015b), which was – 4.4 dB for the approximate chest wall thickness of 0.5 cm, and ultrasound frequency of 7.3 MHz. This attenuation derating approximated in situ values at the lung surface, listed in Table 1. The output setting of the machine was used to adjust exposure settings −2 dB apart for assessment of the exposure–response trends. The probe was aimed between the ribs on the right side to show the lung surface image at approximately the focal depth. A −20 dB power setting (pressure amplitudes reduced by a factor of ~10) was used for aiming and recording before and after ultrasound images. For exposure to the desired setting, the power was quickly raised and timed for 5 min, after which the power was quickly lowered again to −20 dB.

Table 1.

The pulse parameters used for the 6 exposure power levels (dB), all with an observed frequency of 7.3 MHz and derated for attenuation in situ.

Setting	MIOS	PRPA	PMPA	MIIS	ISPPA
dB	MPa MHz−½	MPa	MPa	MPa MHz−½	W cm−2
0	1.2	3.1	5.0	1.14	446
−2	1.0	2.7	4.4	0.96	335
−4	0.8	2.3	3.6	0.82	222
−6	0.6	1.8	2.8	0.66	141
−8	0.5	1.5	2.1	0.56	86
−20	0.12	0.37	0.40	0.13	4.1

PRPA, peak rarefactional pressure amplitude; PMPA, pulse mean pressure amplitude; I_SPPA, spatial peak pulse average intensity; MI_OS; MI_IS, Mechanical Index calculated in situ value.

This study was planned to assess the influence of FiO₂ on the PCH bioeffect of diagnostic ultrasound. Four fractions of inspired oxygen were delivered from a medical gas vendor (Cryogenic Gases Inc., Ann Arbor MI). These were 10%, 20% (medical air), 60% and 100% FiO₂. The remaining gas percentage was nitrogen for all FiO₂ values. A pulse oximeter probe (Kent Scientific Inc., Torrington, CT,USA) was placed on a paw and used to measure the heart rate (HR) and percent oxygen saturation (%SpO₂) after initial anesthesia, but before starting the inspired oxygen delivery, and after approximately 15 min of inspired oxygen treatment including the 5 min of ultrasound exposure. The 10% FiO₂ was similar to the effective oxygenation (partial pressure of oxygen) at an altitude of 5,800 m (Peacock, 1998), although the total gas pressure with nitrogen was approximately sea level. The 60% FiO₂ was used as a representative value for elevated oxygen delivery that typically ranges from 30–80% FiO₂ (Smith et al. 2020). In a previous study, 100% FiO₂ seemed to produce spontaneous comet tail artifacts and areas of atelectasis on some rat’s lungs (Miller et al, 2018c).

During and after ultrasound exposure, the images provide information about the PCH effect by displaying CTA. In this research the CTA gave a qualitative indication of PCH induced by the ultrasound exposure. For low exposure PRPAs, none or a few small image perturbations might be identified, while for the highest exposure in susceptible physiological conditions, CTA fill the bright line lung surface image in a water-fall like effect. The width of the bright-line lung surface image was measured, together with the width which was involved with the CTAs. The percentage width of CTAs was calculated as one measure of the PCH effect. In addition, subpleural consolidated segments (SCS) were noted with use of 60% or 100% FiO₂ in the post-exposure ultrasound images as the “shred sign” (Liu et al 2015; Song et al 2017). These signs, which actually imaged the consolidated regions (not artifacts due to multiple reflections as were the B lines) also were assessed as a width relative to the bright surface image.

Five min after exposure, each rat was removed from the water bath and the gas delivery nose cone, and sacrificed under anesthesia by exsanguination of the inferior vena cava. The trachea was occluded, the thorax was opened and the heart and lungs were removed together. The right cranial and medial lobes were examined and photographed using a stereomicroscope with digital camera (Spot Flex, Diagnostic Instruments Inc., Sterling Heights, MI USA). The lung photomicrographs were used to measure the approximate diameter and area of the regions of PCH on the lung surface using image analysis software (Spot v. 5.1, Diagnostic Instruments, Inc., Sterling Heights, MI USA). In addition, lungs were fixed by immersion in 4% buffered formaldehyde and suspended from a tether in an upside down 50 ml tube for a minimum of two weeks for later histological examination. Histological slides were made as 5 micron paraffin sections with hematoxylin and eosin staining by the Research Histology and Immunoperoxidase Laboratory of the University of Michigan Comprehensive Cancer Center.

Experimental Plan and Statistics

Eight exposure conditions included the four FiO₂ values for the two anesthesia techniques. For each exposure condition, groups six rats, except 8 rats for shams, were exposed at different power settings. 9 groups were completed using TO anesthesia: 10% FiO₂ for 0 and −2 dB, 20% at 0 dB, 60% at 0, −2, −4, −6 dB and sham, and 0 dB at 100% FiO₂. 12 groups were completed using TX anesthesia: 10% FiO₂ for 0, −2, −4, and −6 dB, dB, 20% at 0 dB, 60% at 0, −2, −4, −6, −8 dB and sham, and 0 dB at 100% FiO₂. A general trend for this range (10% to 100%) of FiO₂ was assessed by testing each exposure condition at 0 dB. For threshold determination, exposures were performed at successively lower power settings (dB) until no significant PCH was seen, indicating the threshold, and this lowest power group was considered to be in lieu of an additional sham group (to minimize unnecessary animal use). The PCH effect is normally limited to the scan plane, with other (essentially sham exposed) lung outside the ultrasound exposed area unaffected. However, in this study some consolidation was seen outside the ultrasound-scanned area (see Results), and sham groups were performed for the TO and TX conditions with 60% FiO₂. Statistical analysis was performed to compare groups using SigmaPlot for Windows V. 14.0 (Systat Software Inc., San Jose CA, USA). The Mann-Whitney Rank Sum test was used to compare means of measured PCH area between exposure groups and shams, and the t-test was used for other comparisons with statistical significance assumed at P<0.05. Thresholds for the 10% and 60% FiO₂ conditions were determined by linear regression of data points selected to estimate the zero PCH intercept PRPA.

Results

The results for the HR and %SpO₂ are listed in Table 2. The TO group with 10% FiO₂ showed reduced HR and %SpO₂ at the post-exposure time point, relative to the initial readings for normal air breathing at anesthesia, as expected from the low FiO₂. The HR rate was normal and the SpO₂ increased to normal values for the higher FiO₂ values. The addition of Xylazine for the anesthesia reduced the HR (mean 311 ± standard deviation 32) and %SpO₂ (78.5 ± 7.2) for the initial measurements at anesthesia compared to the TO groups with HR 442 ± 40 and SpO₂ 83.2 ± 5.9: HR difference was P<0.001 (n= 73 and 93), and SpO₂ difference was P<0.001 (n=73 and 93). The post exposure HR measurements for TX were further reduced (P<0.001) compared to the TO groups. The TX anesthesia reduced the initial SpO₂ below the normal value of about 95 % (Table 2). The 10% FiO₂ further reduced the %SpO₂ (post exposure), but the higher % FiO₂ values succeeded in elevating the %SpO₂ to the normal range. These results showed that the different anesthesia and % FiO₂ values gave the desired variation in physiological conditions.

Table 2.

Physiological parameters, mean (standard deviation), of heart rate (HR) and percent oxygen saturation (%SpO₂) at initial anesthesia and post-exposure for the different anesthetic Telazol Only (TO) or Telazol plus Xylazine (TX) and fraction of inspired oxygen as 10%, 20% or 60%. The initial and post results are compared by the p statistic from paired t-tests (no significant difference, nsd).

Condition	n	Initial HR (bpm)	Post HR (bpm)	P	%SpO2	%SpO2	P
10% TO	13	454 (42)	347 (43)	<0.001	82 (6)	62 (8)	<0.001
20% TO	6	434 (32)	439 (74)	nsd	86 (4)	91 (2)	0.03
60% TO	33	438 (42)	412 (95)	nsd	83 (7)	96 (3)	<0.001
100% TO	7	444(34)	409 (81)	nsd	83 (3)	97 (2)	<0.001
10% TX	24	313 (32)	248 (30)	<0.001	77 (8)	62 (6)	<0.001
20% TX	6	323 (25)	266 (39)	<0.001	75 (3)	83(6)	0.02
60% TX	38	310 (33)	234 (22)	<0.001	80 (7)	95 (6)	<0.001
100% TX	7	309 (34)	240 (27)	0.02	79 (6)	96 (2)	<0.001

The percentage of the width of the bright line image of the lung surface which is involved in CTA (B lines) provides an indicator of PCH effect. The percentage of CTA involvement is plotted in Figure 1 for the 60% FiO₂ with TO and TX, and 10% FiO₂ with TX. This indicator increased from zero above a threshold PRPA, see Table 3. The mean percentages of CTA involvement were less for TO than for TX anesthesia for all exposure settings, but the difference was statistically significant only for the −2 and −4 dB settings (P<0.05). For comparison to the other % FiO₂ values for TO at 0 dB, % CTA was 26.4 % (mean) (17.5%, standard deviation), 59.4 % (11.8 %) and 50.1% (12.3 %) for 10, 20 and 100% FiO₂, respectively. For TX at 0 dB, % CTA was 70.4 % (12.0%), 87.5 % (6.7 %) and 92.9% (4.5 %) for 10, 20 and 100% FiO₂, respectively. The general trend was for increasing % FiO₂ to yield greater extent of CTA indication of PHC occurrence.

Figure 1.

A plot of the measurement (mean and standard error bars) of ultrasonic image signs for the dose response experiments at 10% (blue) and 60% (orange) FiO₂ as the percentage of the bright pleural image with the artifact: SCS subpleural consolidated segments and CTA comet tail artifacts. Dashed lines represent linear regression data from the Telazol only anesthesia, while solid lines represent linear regression data from the Telazol plus xylazine anesthesia.

Table 3.

Threshold determinations for results using 10% FiO₂ and 60% FiO₂ as the peak rarefactional pressure amplitude (PRPA, MPa) zero cross-over of linear regressions (with coefficient of determination r²) for comet tail artifacts (CTA) and subpleural consolidated segments (SCS) in the ultrasound images, and for pulmonary capillary hemorrhage (PCH) areas measured on fresh lung samples.

	CTA		SCS		PCH
Condition	MPa	r2	MPa	r2	MPa	r2
10% TO	ND		none		2.8	0.11
60% TO	1.7	0.36	1.5	0.12	1.8	0.21
10% TX	1.7	0.52	none		1.5	0.21
60% TX	1.4	0.79	1.4	0.48	1.4	0.63

The induction of changes giving a percentage of the bright lung surface image which showed SCS was a new finding associated only with the increased % FiO₂ values of 60% and 100%. Examples of the appearance in the ultrasound images with CTA only and CTA with SCS are presented in Fig. 2. The SCS indicator also increased above a threshold, see Fig. 1 and Table 3. SCS was also less for TO than TX anesthesia but with statistical significance only for −2 dB exposure (P<0.05). For comparison, the 100% FiO₂ values of percentage SCS were 19.4 % (11.4 %) for TO and 84.0 % (5.6%) for TX anesthesia.

Figure 2.

Examples of the ultrasound images and the artifact signs with Telazol plus Xylazine anesthesia for 20% and 60% FiO₂ and 0 dB before (top) and after (bottom) exposure.

The results for PCH area at 0 dB are presented in Figure 3 as means with standard error bars. The sham groups of 8 rats each for the TO and TX conditions at 60% FiO₂ had zero PCH. The 0 dB groups for TO and TX conditions at 60% FiO₂ had PCH area highly significantly (P<0.01) different from the zero sham results. The 20% FiO₂ conditions at 0 dB would also be highly statistically significant when compared to a zero result, as has been observed in previous studies (Miller et al. 2018a, 2018b; 2018c). A zero assumption was supported by the absence of PCH effects outside the exposure scan plane. There were few outstanding differences in the PCH for the different conditions. The PCH for TO 20% and 60% FiO₂ were significantly greater than for the 10% FiO₂ (P<0.05). The 10% and 20% FiO₂ for TX were significantly greater than the corresponding values for TO anesthesia (P<0.02). The previously observed greater PCH effect for added xylazine (Miller et al. 2015b) was confirmed for the low 10% and 20% FiO₂ values (P<0.02), but not for the 60% and 100% FiO₂.(P>0.1). Therefore, the increasing oxygen enhanced the susceptibility to ultrasound induced PCH for the TO, but not for the TX anesthesia.

Figure 3.

Results for PCH areas measured on the fresh lung samples for the maximal 0 dB exposure given as the mean and standard error of groups of 6 rats with the indicated FiO₂ values. Open bars represent data from the Telazol only anesthesia, while filled bars represent linear regression data from the Telazol plus xylazine anesthesia.

The PCH area exposure-response results for the 10 and 60% FiO₂ groups is shown in Figure 4. For TO, results were relatively low, particularly for the 10% FiO₂. Although PCH was observed in 3 of 6 rats for the 10% FiO₂, this was not a statistically significant occurrence (z test P=0.18). The PCH was much larger for TX with 60% FiO₂ than for 10% FiO₂, for the moderate exposures of −2 (P< 0.002) and −4 dB (P< 0.003). The data was fitted by linear regression to estimate the PCH threshold by the zero crossing, see Table 3.

Figure 4.

A plot of the measurements (mean and standard error bars) of diagnostic ultrasound induced PCH for the dose response experiments at 10% (blue) and 60% (orange) FiO₂ for Telazol only (TO) or Telazol plus Xylazine (TX) anesthesia with linear regression of the means to estimate the thresholds for each condition (see text). Dashed lines represent linear regression data from the Telazol only anesthesia, while solid lines represent linear regression data from the Telazol plus xylazine anesthesia.

The fresh lung samples showed the characteristic PCH across the scan plane shown previously in fresh tissue samples and histology (Miller et al. 2018). The SCS indication on the ultrasound images was associated with a somewhat different appearance of the fresh lung samples and histology. Examples are shown in Fig. 5 for fresh lung samples with 20% or 60% FiO₂, which were the lungs scanned for the ultrasound images shown in Fig. 2 (20TX0 and 60TX0). For 20% FiO₂, the PCH along the scan line has an appearance of containing numerous gas filled alveoli and limited approximately to the beam dimensions. For 60% FiO₂, the apparent PCH region is rather wide, exceeding the scan beam width in one area which extended to the rim of the medial lobe. In addition, the PCH region has relatively few remaining alveolar gas regions and seems relatively transparent (compared to whole blood hemorrhage). This phenomenon seems to be flooding of the lung area with PCH blood and clear liquid, which appears to be congestion or consolidation from atelectasis and edema. The histology in Fig. 5 reveals the cross section of the PCH scan lines described above with numerous remaining clear (gaseous) regions (probably alveoli) for the 20% FiO₂ case, and virtually no retained gas for the 20% FiO₂ case. This flooding effect was observed only for 60% FiO₂ and 100% FiO₂, for which an entire lobe was involved for some samples. The occurrence of SCS and flooding of adjacent regions is listed in Table 4. These findings represent a different type of impact, for which some hemorrhage and fluid extends beyond the immediate scan plane area and reduces the oxygenating function of the large volumes of lobe.

Figure 5.

Fresh lung samples with Telazol plus Xylazine anesthesia and 0 dB scanning (see Fig. 2) for 20% FiO₂ (top left, scale bar 5 mm) and 60% FiO₂ (top right, scale bar 5 mm). Corresponding histological sections with H&E stain are presented for sections cut across the scan planes for 20% FiO₂ (bottom left, scale bar 0.5 mm) and 60% FiO₂ (bottom right, scale bar 0.5 mm).

Table 4.

The proportions of occurrence for CTA/PCH, SCS and alveolar flooding for the 60% and 100% FiO₂ groups, which had subpleural consolidation and alveolar flooding. Relative to 0/8 shams, 4/6 or higher were statistically significant proportions.

Condition	dB	CTA/PCH	SCS	Flooding
60% TO	0	5/6	4/6	0/6
60% TO	−2	6/6	1/6	0/6
60% TO	−4	2/6	2/6	0/6
60% TO	−6	1/6	1/6	0/6
100% TO	0	6/6	3/6	2/6
60% TX	0	6/6	6/6	3/6
60% TX	−2	6/6	6/6	4/6
60% TX	−4	6/6	6/6	2/6
60% TX	−6	3/6	3/6	0/6
60% TX	−8	3/6	0/6	0/6
100% TX	0	6/6	6/6	6/6

Discussion

The fraction of inspired oxygen (FiO₂) was investigated for its influence on DUS-PCH induction. In order to investigate both possible reduction and enhancement of PCH susceptibility, anesthesia was either IP injection of Telazol only (TO) or of Telazol plus xylazine (TX). Xylazine appears to enhance PCH induction relative to anesthesia with only ketamine (Miller et al 2015b) or Telazol (Miller et al 2018b). Fractions of inspired oxygen of 10%, 20% (medical air), 60% and 100% FiO₂ were delivered for 15 min before and during exposures. The ultrasound B mode images were examined for the extent of CTA (B lines), indicative of edema (Dietrich et al 2016) and also the extent of the shred sign for subpleural consolidated segments (SCS). Use of 60% or 100% FiO₂ was associated with SCS, but the 10% or 20% FiO₂ were not (Figure 1). Low 10% FiO₂ oxygenation can simulate some effects of high altitude (similar to 5,800 m (Peacock, 1998)) or a poor breathing environment while high oxygenation for patient support has been associated with atelectasis for long duration treatment (Magnusson and Spahn, 2003). Atelectasis can be diagnosed with diagnostic ultrasound and has an appearance like the shred sign (Liu et al 2015; Song et al 2017), which can be similar to the appearance of the SCS in our rat images (Fig. 2). The lung samples with SCS also often had areas of flooding edema, or possibly atelectasis, extending beyond the region within the scan plane (Fig. 5, Table 4), suggesting that the diagnostic ultrasound exposure could trigger the extensive edema. This consequence of pulmonary diagnostic ultrasound exposure might be important for some patients with oxygenation treatment and possibly trigger more extensive lung injury than might be expected from the size of the scan plane.

PCH for the maximal exposure of 3.1 MPa (Fig. 3) for TO 20% and 60% FiO₂ were significantly greater than for the 10% FiO₂. PCH for TX with 10% and 20% FiO₂ were significantly greater than for TO anesthesia. The increased oxygen appeared to enhance the susceptibility to ultrasound induced PCH for the TO, with effects increasing for 60% FiO₂ relative to 10% FiO₂, but not for the TX anesthesia. The presence of xylazine may cause enhanced susceptibility to ultrasound, which is not exceeded by the higher FiO₂. The PCH exposure-response results (Fig. 4, Table 3) for TO were relatively low, particularly for the 10% FiO₂, but the 60% FiO₂ increased the magnitude and lowered the threshold for PCH. For TX the PCH magnitudes were greater with 60% FiO₂ than with 10% FiO₂, but the thresholds were not significantly different. PCH thresholds at 60% FiO₂ for TO and TX were not significantly different. Thresholds for TO were 2.8 MPa and 1.8 MPa for 10% and 60% FiO₂ and for TX thresholds were 1.5 MPa and1.4 MPa for 10% and 60% FiO₂. The higher FiO₂ tended to enhance the susceptibility of the lungs to PCH and to SCS. The presence of Xylazine also enhanced susceptibility to PCH. However, enhancement engendered by high oxygenation and by xylazine, while roughly additive in producing greater PCH magnitude, was not synergistic in regard to the minimum thresholds, which were about the same and similar to previous results (Miller et al 2018a). The lowest threshold observed was 1.4 MPa, which corresponded to an in situ MI_IS = 0.5.

Limitations

This study explored the influence of varied FiO₂ on the PCH bioeffect of diagnostic ultrasound. The use of a low 10% FiO₂ does not fully simulate high altitude hypoxia because it was delivered at atmospheric pressure. The use of our rat model of pulmonary diagnostic ultrasound limits the direct applicability of the results to clinical ultrasound. However, this can be partly translated to clinical conditions by considering the in situ MI, which relates to the clinical on-screen MI value.

Conclusions

The variation of % FiO₂ varied the susceptibility of the lung to diagnostic ultrasound induced PCH. In addition, the higher % FiO₂ values of 60% and 100% led to increased edema and imaging of SCS, which extended beyond the scan plan exposure area. While both the increased oxygenation and xylazine separately enhanced the PCH, they did not combine to produce a very low threshold for PCH. Clinically, the exposure duration and the acoustical output should be kept as low as is reasonably achievable (ALARA), consistent with collection of diagnostically acceptable images by sonographers. This bioeffect can readily be addressed by the practice of ALARA for pulmonary ultrasound with MI<0.4 (AIUM, 2015).