Ultrasound-induced lung hemorrhage in the monkey

Ultrasound in Med.& Biol., Wol. 20, No. 1, pp. 65-72, 1994 Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0301-5629/94 $6.00 + .00

Pergamon

OOriginal Contribution ULTRASOUND-INDUCED LUNG HEMORRHAGE IN THE MONKEY ALICE F. TARANTAL a n d DON R. CANFIELD California Regional Primate Research Center, University of California, Davis, CA, USA (Received 5 February 1993; in finalform 4 August 1993)

AbstractmStudies with the mouse have shown that lung hemorrhage can result from exposure to ultrasound at a peak pressure of ~1 MPa at 4 MHz (Mechanical Index [MI] ~0.5). In order to determine whether a comparable outcome could occur in a larger animal with characteristics similar to humans, studies were performed with monkeys using a clinical scanner under maximum output conditions (imaging + pulsed and color Doppler; derated Pr of 3.7 MPa [4.5 MPa, measured in water], 4 MHz; MI ~ 1.8) (N = 57). Monkeys ranged in age from 1 day of life to 16 years with exposures limited to the right lung lobes (5 min cranial, 5 rain caudal; N = 41 exposed, N = 12 sham-exposed controls, N = 4 colony controls). Results showed that animals ranging in age from 3 months to 5 years (mean age of 2.5 years) had a greater propensity for the occurrence of multiple well-demarcated circular hemorrhagic foci (0.1-1.0 cm), which were not observed in either control group. These lesions were characterized by marked congestion of alveolar capillaries with accumulation of red blood cells within the alveolar spaces, and were unassociated with major vessels or respiratory bronchioles. Further studies will be required in order to determine the relevance of these findings to the human, although it was concluded that ultrasound-induced lung hemorrhage in the monkey is of a significantly lesser degree when compared to the mouse. Key Words: Ultrasound, Cavitation, Bubble activity, Lung, Hemorrhage, Macaques, M. rad/ata, M. fascicularis.

(Rooney 1970) appears to be greatest at tissue interfaces that involve muscle, connective tissue and related vasculature (Hynynen 1991). Although a proven phenomenon in vitro (Holland et al. 1992) the occurrence of cavitation in vivo has been difficult to document in mammalian systems primarily due to the transient nature of its occurrence (/.ts) and the focal nature of the resultant effects (#m) (AIUM 1993). A simple theoretical relationship has been described between acoustic pressure and the onset of bubble activity, which has provided the basis for the adoption of an Output Display Standard (ODS) (AIUM/NEMA 1992) for clinical imaging systems, which predicts the likelihood of the occurrence of cavitation in vivo, specifically for clinical purposes. In the proposed ODS, the Mechanical Index (MI) is defined as the derated peak rarefactional pressure (Pr) in MPa divided by the square root of the ultrasonic center frequency in MHz. The MI assumes a "worst case" scenario under clinical scanning conditions. The intended use of the MI is to provide a real-time display that will allow the operator to adjust acoustic output, thereby minimizing patient exposure and employing the ALARA (as low as reasonably achievable) principle.

INTRODUCTION Many studies have been conducted both in vivo and in vitro in an effort to identify the biological interactions ultrasound may have with soft tissues, both from a thermal (NCRP 1992; Tarantal et al. 1993 a and b) and acoustic cavitation perspective (AIUM 1993). Although a significant amount of data has been generated on heating, there is limited information available on the occurrence of cavitation in vivo. The conditions under which cavitation nuclei (bubbles) may develop in vivo are not well understood, although it has been established in vitro that when they do occur they can be driven to oscillate by the presence of an acoustic field. Under some conditions a sudden change in pressure amplitude can result in expansion and violent collapse of these bubbles, with subsequent damage due to the generation of high temperatures (Flint and Suslick 1991) and potentially toxic free radical formation (Kehrer et al. 1988; Kondo and Kano 1988; Suslick 1989). Localized impact of these events on cell structure Address correspondence to: Alice F. Tarantal, Ph.D., California Regional Primate Research Center, University of California at Davis, Davis, CA 95616-8542, USA. 65

66

Ultrasound in Medicine and Biology

The importance of the use of the MI relates specifically to the fact that current diagnostic ultrasound equipment has peak pressures for imaging (measured in water) to range from 0.8 to 8.8 MPa (mean = 4.1), whereas for color Doppler units, pressures of 2.4 to 6.4 MPa (mean = 4.7) have been calculated (Duck and Martin 1991). Although several scanners marketed in other countries were included in this study (some with higher pressures than commercial systems manufactured in the United States), this implies that current commercial ultrasound systems can operate at MI levels of 0.5 to 1.8. Studies with small nonmammalian species such as Drosophila have shown that cavitation is directly associated with death of larvae (Carstensen et al. 1990), with roughly half of the larvae killed during exposure to positive pressures of 2 - 3 MPa; Drosophila larvae are known to contain small stabilized gas bodies within their developing respiratory systems that make them susceptible to the effects of cavitation. Transient cavitation has been proposed as the mechanism responsible for lung hemorrhage and destruction of the blood-air barrier (producing direct continuities between capillary lumina and alveolar spaces) in young mice exposed to ultrasound at peak pressures of 0 . 8 4.0 MPa ( 1 - 4 MHz, 1 and 10 #s pulse durations, pulse repetition frequency [PRF] of 10-100 Hz) (Child et al. 1990; Penney et al. 1993). Further attempts to confirm these findings have shown that gravid mice develop significant lung hemorrhage after exposure to pressures of up to 20 MPa, although their 18 day gestation fetuses did not appear to have a similar response when exposed to comparable levels in utero (Hartman et al. 1990). However, potential effects on the fetal lung due to cavitation nuclei have not been fully investigated. Although overall organization and cellular composition of the lung are conserved among mammals, there are significant physiologic and anatomic differences related to organization of the distal conducting airways, acinar composition, and blood supply, to name a few (Tyler and Julian 1992). Variations related to the structure of the gas-exchange area are also known, and may be of importance when considering the potential for cavitation and the resultant damage that it may cause. Based on these differences, studies were proposed with the goal of identifying whether ultrasound-induced lung hemorrhage reported in the mouse could occur in the monkey. The macaque has served as an excellent model for the human under a variety of conditions, and has been incorporated in bioeffects studies where thermal mechanisms of ultrasound have been investigated (Tarantal and Hendrickx 1989a and b; Tarantal et al. 1993a and b). The purpose of these studies was, therefore, to determine whether

Volume20, Number 1, 1994 lung hemorrhage could result in an animal model with similar physiologic and anatomic features when compared to the human, while using a commercial imaging system set at maximum output settings. This included setting the system at the maximum MI attainable (ODS 1992) in order to correlate observed findings with the predictive value of the MI.

MATERIALS AND METHODS

Ultrasound system A state-of-the-art commercial ultrasound system was used for these studies (Advanced Technology Laboratories, Inc. [ATL] Ultramark 9 [UM9] with HDI ®) with output characteristics similar to many scanners currently in clinical use. All exposures were performed with a linear array transducer (ATL, L5; 38 mm footprint) which operates in scan mode at 5 MHz and in pulsed and color Doppler modes at 4 MHz. The UM 9 was placed in the vascular imaging setups mode in "triple m o d e " (two-dimensional imaging + color and pulsed Doppler) with the smallest color box achievable, a 1 mm Doppler sample volume, and a scanning depth of 2.8 cm with a sample volume depth of 1.2 cm, or at a 3.5 cm scanning depth with a 2.0 cm sample volume depth. The PRF was set at 1515 kHz, and the maximum output setting was adjusted to achieve the maximum MI obtainable for the system (1.8); derated Pr at this setting was 3.7 MPa (4.5 MPa, measured in water). Output characteristics at these settings were measured (derated and water maximum intensity depth of 1.43 cm; measurements performed by ATL, Inc.) and are presented in Table 1. Animals and exposure conditions All animal procedures employed within the study conformed to the requirements of the Animal Welfare Act. The California Regional Primate Research Center (CRPRC) is fully accredited by the Association for Accreditation of Laboratory Animal Care; all study protocols are approved prior to implementation by the Institutional Animal Use and Care Committee at the University of Califomia at Davis under the auspices of the Campus Veterinarian. Activities related to animal care (diet, housing), immobilization procedures, and necropsy were performed as per CRPRC standard operating procedures. Fifty-three bonnet macaques (Macaca radiata) and four long-tailed macaques (Macaca fascicularis) were included in these studies. Animals ranged in age from 1 day to 16 years with a roughly equivalent number of males (N = 29) and females (N = 28) evaluated (Table 2). All bonnet macaques were group-housed in an outdoor cage, whereas the long-tailed macaques

Ultrasound-induced lung hemorrhage• A. F. TARANTALand D. R. CANFIELD Table 1. Measured output characteristics for the ATL Ultramark 9 with HDI® Ultrasound System under "triple mode" conditions (two-dimensional imaging, color and pulsed Doppler) at 1.43 cm. Linear array scanhead (L5; 38 mm) at a center frequency of 4 MHz (0.65/zs pulse duration; duty cycle of 0.10%, pulse repetition frequency of 1515 kHz). Condition 1 (Focal point 1.2 cm) Parameter ISPPA(W/cm2) Measured Extrapolated IsPrA(mW/cm2) Measured Extrapolated /max(W/cm2) Measured Extrapolated Pr (MPa) Measured Extrapolated Pc (MPa) Measured Extrapolated MI Measured Extrapolated

Water

Condition 2 (Focal point 2.0 cm)

Derated

Water

Derated

933 1144

628 742

866 1084

511 588

958 1143

645 745

891 1030

527 561

1326 1760

890 1140

1188 1812

707 992

4.5 4.9

3.7 4.0

4.2 4.1

3.2 3.0

6.1 5.0

5.0 6.7

6.1 9.1

4.7 6.7

2.3 2.4

1.9 2.0

2.1 2.0

1.6 1.5

I = intensity; SPPA = spatial peak pulse average; SPTA = spatial peak temporal average; Pr = peak rarefactional pressure; Pc = peak compressional pressure; MI = Mechanical Index.

were housed in pairs in indoor cages. A total of 41 animals were exposed to ultrasound; 12 served as sham-exposed controls, whereas 4 were colony controis with no sham procedures performed. Bonnet macaques housed outdoors were brought to indoor quarters 24 h before study initiation. On the day of the procedure, animals were sedated with ketamine hydrochloride (10 mg/kg; intramuscular), and brought individually or in groups of two to the ultrasound suite. A 1 mL blood sample was collected from a peripheral vessel for a complete blood count, and all animals were auscultated in order to document normal chest sounds. Animals were placed in a supine position on the scanning table in modified dorsal/left lateral recumbency, with the fight arm elevated to allow access to the fight thorax. Hair was shaved from the thoracic and axillary regions in preparation for the exposures. Acoustic gel was applied (Aquasonic ®, Parker Laboratories, Inc., Orange, NJ), and the transducer was placed intercostally and parallel to the fibs (roughly T 3 - 5 for cranial lobe exposure and approximately T 6 - 7 for caudal lobe exposure) with the beam positioned obliquely to the lung lobes such that cranial exposures were performed with the beam pointing slightly caudad, and caudal

67

exposures performed with the beam pointing slightly craniad. In all cases, the beam was focused on the fight lung lobes. An image of lung tissue was obtained immediately prior to initiating the triple mode exposures, which lasted for 5 min duration for each location (cranial and caudal; total exposure of 10 min). There was an approximate 2 - 3 min interval between exposures to allow for repositioning. The transducer was manually held in place without significant pressure on the chest wall, thereby permitting normal respiratory chest excursions. All exposures were videotaped throughout the length of the exposures, and photographic documentation of the beam location and output settings was obtained after each experiment was completed. At the termination of the study period, each animal was transported to the necropsy room and euthanized with an intravenous overdose of pentobarbital. An initial evaluation of gross structures related to the chest wall, lung, heart, and pleural cavity was performed in situ, followed by measurement of chest wall thickness (skin, fat, muscle) and removal of intact lung tissues (within 10 min postmortem). The lungs were rinsed in 0.9% saline, any apparent gross lesions were described (including size, position, and lung lobe location), and photographs were taken from dorsal and ventral aspects of the intact lungs. All lungs were then immersed in 10% neutral buffered formalin prior to sectioning. Portions of cranial and caudal, fight and left lung lobes were obtained for histopathologic evaluation (sectioned at 6 #m), with comparable samples collected for all animals studied. Representative sections were stained with hematoxylin and eosin in preparation for evaluation. All gross and histopathologic descriptive evaluations were performed in a blinded manner. A system for scoring all lung tissue pathology was devised based upon the type and volume of hemorrhage noted (see Table 3). Scores ranged from zero (no hemorrhage either grossly or histologically) to six (severe hemorrhage encompassing -> 1/2 of both the fight and left lung lobes). RESULTS The outcome for these studies is presented in Tables 4 and 5. For the 41 animals exposed to ultrasound, Table 2. Animal age and body weight by group mean ___standard deviation (Range). Group (N) Control (4) Sham-exposed (12) Exposed (41)

Age

Body weight (kg)

9.9 y _ 6.2 (4 y-15 y 8 mo) 5.5 y 4- 5.4 (3 wk-16 y) 4.8 y -4- 4.6 (1 d-15 y)

9.9 4- 5.9 (3.6-15.9) 4.6 4- 3.7 (0.5-10.2) 4.7 4- 3.8 (0.3-14.7)

68

Ultrasound in Medicine and Biology

Volume 20, Number 1, 1994

Table 3. S c o r i n g s y s t e m for evaluating lung tissues in m a c a q u e s p o s t m o r t e m . Score

Table 5a. Results o f e x p o s u r e o f m a c a q u e s at a 1.2 c m focal z o n e ( N = 21).

Description

0

No discoloration noted grossly; no hemorrhage seen histologically. No discoloration noted grossly; hemorrhage identified histologically. Generalized tissue mottling grossly; mild hemorrhage histologically. O n e - t w o foci of discoloration grossly; mild hemorrhage histologically. Multifocal to coalescing discoloration grossly; mild to moderate hemorrhage histologically. Large, irregular patches of discoloration grossly; moderate to severe hemorrhage histologically. Marked discoloration grossly; severe hemorrhage histologically ~ ~ all lung lobes.

1 2 3 4 5 6

33 (80.5%) showed some evidence of acute pulmonary hemorrhage. The greatest percentage of animals scored a four (15/41; 36.6%); characteristic lesions were composed of multiple discrete to coalescing, well-delineated round to oval foci of erythema with smooth borders identified from the pleural surface (Fig. 1). Foci measured approximately 0.1-1.0 cm in greatest dimension. Five of 41 (12.2%) scored either a three ( 1 2 circular hemorrhagic foci), a two (generalized mottling), or a one (hemorrhage noted histologically only; total of 15 animals). An additional three exposed animals showed larger, less well-delineated areas of dark red discoloration in dependent regions of lung (scored as a five or six). Generally, these animals were older (10-13 years), obese, and had pleural adhesions; based on gross and histologic findings (see below), these features were considered to be the result of congestion rather than hemorrhage. The remaining 8 of 41 in this group (19.5%) did not exhibit signs of hemorrhage either grossly or histologically and, therefore, were scored with a zero. For sham-exposed controls, 10 of 12 (83.3%) showed some evidence of lung hemorrhage, with the

Score

N (%)

HemR (N)

HemL (N)

R +L (N)

0 1 2 3 4 5 6

2 (9.5) 2 (9.5) 2 (9.5) 2 (9.5) 13 (62.0) 0 0

-0 0 2 6 . .

-2 0 0 0

-0 2 0 7

Score

N

Exposed

Sham

CC

0

13

1

7

2 3 4 5 6

8 9 15 2 3

8/41 5/41 5/41 5/41 15141 1/41 2/41

2/12 2/12 2/12 4/12 0/12 1/12 1/12

3/4 0/4 1/4 0/4 0/4 0/4 0/4

5/16 2/16 3/16 4/16 0/16 1/16 1/16

CC = colony controls

Age range 4 y-16 y 2 mo-15 y

1 d-14 y 1 mo-3 y 3 rot-5 y 10 y-13 y 5 y-10 y

Mean age (y) 8 4.3 7 1 2.5 11.5 8

. .

BW (kg)

1.5-14.5 y 1.5-15 y 14.5 y 3.5-10.5 y 3 mo-4.5 y . . . .

1.4-10.6 2.1-7.9 10.2-14.7 3.2-6.5 0.9-5.4

Hem R = hemorrhage noted right lung lobes; Hem L = hemorrhage noted left lung lobes; R = right, L = left.

largest percentage of animals receiving a score of three (4/12; 3 3 . 3 % - - 1 - 2 small foci of erythema) and 2 of 12 (16.7%) receiving either a zero, one, or two (total of six animals). The remaining two anirr~als scored either a five or six. For the four untreated colony controis, 3 of 4 (75%) scored a zero, whereas the remaining one animal (25%) scored a one. All lungs, except those of very young animals, showed a variable tendency toward failure to collapse upon opening the chest cavity as well as some degree of black stippling suggestive of pneumoconiosis, which is a common finding for animals housed in outdoor cages. Overall, although all groups of animals showed some degree of lung hemorrhage, only animals exposed to ultrasound were noted with the multiple, welldelineated hemorrhagic foci associated with a score of four. The age range for animals within this group was 3 months to 5 years, with a mean age of 2.5 years (Table 4). These animals ranged in weight from 0 . 9 6.3 kg, with a chest wall thickness of 0.3-1.2 cm. Histopathologically, these lesions were defined as discrete areas of hemorrhage, usually near the pleural surface (Fig. 2), and were characterized by marked congestion of associated alveolar capillaries with an accumulation of red blood cells within the alveolar

Table 5b. Results of exposure of macaques at a 2.0 cm focal zone (N = 20).

Table 4. Overall outcome for exposed, sham-exposed, and colony controls. Total controls

. .

Age (range)

Score 0 1 2 3 4 5 6

N (%) 5 3 4 3 2 1 2

(25) (15) (20) (15) (10) (5) (10)

HemR (N)

HemL (N)

R+ L (N)

Age (range)

BW (kg)

-0 0 1 1 0 0

-1 1 2 0 0 0

-2 2 0 1 1 2

4-9 y 2 mo-7 y 1 d-ll y 1-2 mo 3-5 y 13 y 5-8 y

4.7-10.2 0.3-9.6 0.3-10.8 0.6-0.7 3.5-5.6 13.6 4.9-5.6

Hem R = hemorrhage noted right lung lobes; Hem L = hemorrhage noted left lung lobes; R = right, L = left.

Ultrasound-induced lung hemorrhage• A. F. TARANTALand D. R. CANFIELD

69

spaces. Intra-alveolar hemorrhage varied from scattered individual red cells to total obliteration of the alveolar space by blood. Involvement of subpleural alveoli was rare and occurred in only a small percentage of cases evaluated. The hemorrhage did not correlate with major arterioles, venules, or distal airways and therefore appeared to originate from alveolar capillaries. Although there were a greater number of animals noted with hemorrhage within the right lung lobes, a significant number also contained hemorrhagic foci in the left lobes with comparable histologic findings (Table 5a and b). All lungs except those from the younger animals showed variable amounts of chronic inflammation comprising aggregates of lymphocytes and more diffuse accumulation of lymphocytes and histiocytes in the pulmonary interstitium. Some of this inflammation appeared to be associated with the accumulation of the black granular deposits characteristic of inhaled pollutants (pneumoconiosis), although there was no association of hemorrhage with background inflammation. DISCUSSION

Fig. 1. Photographic documentation of lung tissue from representative bonnet macaque exposed to ultrasound (5 min cranial, 5 min caudal lung lobes). Note multiple hemorrhagic foci (score of four).

The results of these studies have shown that focal, intra-alveolar hemorrhage can occur in lung tissue of macaques after exposure to ultrasound from a diagnostic system at maximum (worst case) output settings (derated Pr 3.7 MPa, MI ~ 1.8). Although the mechanism for these effects has not been established, current knowledge suggests that it may be due to bubble activity, although a thermal effect cannot be ruled out at this time (ALUM 1993). Some of the controversy asso-

Fig. 2. Histologic appearance of lung tissue from representative bonnet macaque with a score of four, exposed to ultrasound. Note intra-alveolar hemorrhage (arrowheads). Bar = 0.2 mm.

70

Ultrasound in Medicine and Biology

ciated with the cavitation phenomenon in vivo includes identification of the method whereby microbubbles are generated, the conditions under which they may occur, and their exact location (Carstensen 1987). Contrary to reports in the mouse (Child et al. 1990), lung lesions were noted in monkey sham-exposed controls. Although this suggests some effect induced by the scanning procedure, it is important to note that none of the animals in either the sham or untreated control groups displayed the characteristic lesions attributed to ultrasound exposure (score of four) (Fig. 3). Although the data suggest that ultrasound-induced lung hemorrhage can occur in the macaque, it is important to note that it was of a significantly lesser degree than the hemorrhage reported in the mouse. There are several possible explanations for this variation in outcome, including anatomic differences between the rodent and the primate. These include an overall size differential in relation to the sound field as well as differences in lung compliance, the structure of terminal bronchioles, alveolar size, and blood supply. Although the numbers of alveoli within the pulmonary acinus are relatively constant between the species, there is a distinct difference in the structure and arrangement of the bronchioles supplying the gas-exchange area, including the type of terminal bronchioles and the extent of smooth muscle incorporated along the length of the alveolar ducts (Mercer and Crapo 1992). For human and nonhuman primates, smooth muscle fibers extend along the terminal generations of

80

60

[] Exposed [] []

o~

Sham Control

4o

20

0 0

1

2

3

4

5

6

Score Fig. 3. Percent (%) of animals in each study group (exposed, sham-exposed, colony controls) with scores ranging from zero to six. Only animals exposed to ultrasound achieved a score of four.

Volume20, Number 1, 1994 the alveolar ducts, whereas in the mouse these fibers do not extend past the bronchiole-alveolar duct junction, which has been shown to end abruptly. This may suggest that there is a greater potential for hemorrhage and tissue destruction in the mouse lung, particularly where the terminal airways are shorter, thinner, and perhaps less distensible. Morphologic studies have shown that capillaries within the alveolar septa of most mammals are arranged as a single layer separated from the air spaces by a very thin cellular barrier (alveolar epithelium + capillary endothelium ~ 0.1-0.2 #m). Due to this anatomic configuration, it is possible that these regions are more susceptible to conditions where bubble oscillation and rupture may occur, with a greater degree of damage possible when pulmonary arterial and venous pressures are greater and the lung tissues are less able to expand. Another important factor specific to the lung may be the layer of surfactant within the alveoli. As part of the alveolar lining fluid, surfactant is responsible for modifying surface tension in order to promote lung expansion and prevent lung collapse (Hawgood 1991). It is possible that during exposure to ultrasound, small microbubbles are created within comers of the surfactant-rich alveolus. These microbubbles may oscillate, with localized disruption of the epithelial/endothelial barrier and subsequent leakage of red blood cells. In species where the alveoli are small (i.e., adult mouse ~ 40 #m in diameter) and less distensible, damage may be greater, which could explain why lesions in the mouse are more severe than in the monkey (alveolus - 120 #m in diameter). Of interest is the fact that a direct correlation has been shown between reduced surface tension and a reduction in the threshold for cavitation (Holland and Apfel 1989). Another factor worthy of consideration is the age of the experimental animals. It has been shown that the vascular and hemodynamic characteristics of the lung are age-dependent, and that postnatal lung development occurs over a long period of time, which encompasses puberty (Ten Have-Opbroek and Plopper 1992). The mice incorporated in the Child et al. study (1990) were 7 weeks of age (pubertal); the greatest number of animals affected by ultrasound exposure in the studies described herein were from 3 months to 5 years of age (infants to young adults; puberty ~ 3.54 years). This may imply that characteristics specific to early life render the lung more susceptible to these effects; such characteristics may include the distensibility of the chest wall, the degree of attenuation of the sound beam that occurs in the lungs of these young individuals, and the compliance of the lung tissues. In the macaque study it was noted that hemorrhage occurred in both the right and left lung lobes of many of

Ultrasound-inducedlung hemorrhage • A. F. TARANTALand D. R. CANFIELD the y o u n g e r monkeys. Since these y o u n g e r animals have a smaller, thinner-walled thoracic cavity (chest wall thicknesses of 0 . 3 - 1 . 2 cm), this m a y imply that a larger volume of lung tissue may have been exposed to a greater percentage of the beam. In direct contrast were the large, obese animals, where thicker chest walls ( 0 . 8 - 1 . 8 cm) most likely resulted in a greater degree of attenuation. In summary, it is evident that further studies will be required in order to map the precise distribution of hemorrhage in relation to the scan field, and to investigate the range of anatomic and physiologic effects that correlate with these findings. With the results obtained to date, however, several areas worthy of consideration are apparent. First, although it is clear that lung hemorrhage can result in the macaque after exposure to a clinical scanner, the degree of hemorrhage detected is considered anatomically mild, particularly when compared to the reported outcome in the mouse. Although the threshold for these effects is not known, it may be possible that different settings (with lower MIs), which may be more clinically relevant (i.e., more appropriate for optimal two-dimensional imaging and Doppler performance) would not result in equivalent levels of damage. However, although perceived as mild, the physiologic ramifications of this hemorrhage are not known, nor are the conditions under which it may be induced to occur. It is also not k n o w n if repetitive exposures would result in a more severe outcome with the potential for sequelae of greater physiologic importance. Second, from a scanning perspective, and based on species differences, it is anticipated that if lung hemorrhage could occur in the human, it would be more similar in degree to the foci noted in the monkey as compared to the mouse. Based on study outcome, it is also concluded that infants may be most susceptible to these changes. It is currently not k n o w n if respiratory conditions/compromise c o m m o n to premature infants would enhance the potential for this effect, particularly if receiving exogenous surfactants. The factors considered relevant for the occurrence of this phen o m e n o n in vivo are overall size, presence of surfactant, and lung distensibility or compliance. It is evident that continued studies with relevant animal models such as the macaque will help to answer some of these questions, while attempting to identify safe scanning levels. Correlation of any observed effects with the calculated M I will also aid in providing margins of safety, which are directly applicable for the clinical user.

SUMMARY Studies with the macaque indicate that lung hemorrhage can result after exposure to a commercial im-

71

aging system, with a derated Pr of 3.7 M P a at an MI of 1.8. The hemorrhage was multifocal, punctate, and involved alveolar capillaries. It is hypothesized that bubble activity may be responsible for the observed findings, and that young infants may be most susceptible to its occurrence. Acknowledgements--This research was supported by NIH Grant #RR00169. The authors would like to thank Ms. Kimberli A. Schmidt for her expert technical assistance, and Drs. Christy K. Holland and William D. O'Brien for review of the manuscript.

REFERENCES American Institute of Ultrasound in Medicine/National Electrical Manufacturers Association. Standard for real-time display of thermal and mechanical acoustic output indices on diagnostic ultrasound equipment. AIUM/NEMA; 1992. American Institute of Ultrasoundin Medicine. Bioeffects and safety of diagnostic ultrasound. Bethesda, MD: AIUM; 1993. Carstensen, E. L. Acoustic cavitation and the safety of diagnostic ultrasound. Ultrasound Med. Biol. 13:597-606; 1987. Carstensen, E. L.; Campbell,D. S.; Hoffman,D.; Child, S. Z.; AymeBellegarda, E. J. Killing of Drosophila larvae by the fields of an electrohydraulic lithotripter. Ultrasound Med. Biol. 16:798898; 1990. Child, S. Z.; Hartman, C. L.; Schery, L. A.; Carstensen, E. L. Lung damage from exposure to pulsed ultrasound. Ultrasound Med. Biol. 16:817-825; 1990. Duck, F. A.; Martin, K. Trends in diagnostic ultrasound exposure. Phys. Med. Biol. 36:1423-1432; 1991. Flint, E. B.; Suslick, K. S. The temperature of cavitation. Science 253:1397-1399; 1991. Hartman, C.; Child, S. Z.; Mayer, R.; Schenk, E.; Carstensen, E. L. Lung damage from exposure to the fields of an electrohydraulic lithotripter. Ultrasound Med. Biol. 16:675-679; 1990. Hawgood, S. Surfactant:Composition,structure and metabolism.In: Crystal, R. G.; West, J. B., eds. The lung. Scientificfoundations, Vol. 1. New York: Raven Press; 1991:247-261. Holland, C. K.; Apfel, R. E. An improved theory for the prediction of microcavitatiou thresholds. IEEE Trans. Ultrason. 36:204208; 1989. Holland, C. K.; Roy, R. A.; Apfel, R. E.; Crum, L. A. In vitro detection of cavitation induced by a diagnostic ultrasound system. IEEE Trans. Ultrason. 39:95-101 ; 1992. Hynynen, K. The threshold for thermally significant cavitation in dog's thigh muscle in FIFO.Ultrasound Med. Biol. 17:157-169; 1991. Kehrer, J. P.; Mossman, B. T.; Sevanian, A.; Trush, M. A.; Smith, M. T. Free radical mechanisms in chemical pathogenesis. Toxicol. Appl. Pharmacol. 95:349-362; 1988. Kondo, T.; Kano, E. Effect of free radicals induced by ultrasonic cavitation on cell killing. Int. J. Radiat. Biol. 54:475-486; 1988. Mercer, R. R.; Crapo, J. D. Architecture of the acinus. In: Parent, R. A., ed. Treatise on pulmonary toxicology, vol. I. Comparative biology of the normal lung. Boca Raton: CRC Press; 1992:109119. National Council on RadiationProtection and Measurements.Report #113. Exposure criteria for medical diagnostic ultrasound:I. Criteria based on thermal mechanisms.Bethesda, MD: NCRP; 1992. Output Display Standard (ODS): Standard for real-time display of thermal and mechanicalacoustic indices on diagnosticultrasound equipment. Approved by the American Institute of Ultrasound in Medicine, March 11, 1992. Penney, D. P.; Schenk, E. A.; Malthy, K.; Hartman-Raeman, C.; Child, S. Z.; Carstensen, E. L. Morphological effects of pulsed ultrasoundin the lung. UltrasoundMed. Biol. 19:127-135; 1993. Rooney, J. A. Hemolysisnear an ultrasonicallypulsatinggas bubble. Science 169:869-871; 1970.

72

Ultrasound in Medicine and Biology

1989. Tarantal, A. F.; Hendrickx, A. G. Evaluation of the bioeffects of prenatal ultrasound exposure in the cynomolgus macaque (Macaca fascicularis): I. Neonatal/Infant observations. Teratology 39:137-147; 1989a. Tarantal, A. F.; Hendrickx, A. G. Evaluation of the bioeffects of prenatal ultrasound exposure in the cynomolgus macaque (Macacafascicularis): II. Growth and behavior during the first year. Teratology 39:149-162; 1989b. Tarantal, A. F.; Chu, F.; O'Brien, W. D.; Hendrickx, A. G. Sonographic heat generation in vivo in the gravid long-tailed macaque (Macaca fascicularis). J. Ultrasound Med. 12:285-295; 1993a. Tarantal, A. F.; O'Brien, W. D.; Hendrickx, A. G. Evaluation of the

Volume 20, Number 1, 1994 bioeffects of prenatal ultrasound exposure in the cynomolgus macaque (Macaca fascicularis): III. Developmental and hematologic studies. Teratotogy 47:159-170; 1993b. Ten Have-Opbroek, A. A. W.; Plopper, C. G. Morphogenetic and functional activity of type II cells in early fetal rhesus monkey lungs: A comparison between primates and rodents. Anatomical Record 234:93-104; 1992. Tyler, W. S.; Julian, M. D. Gross and subgross anatomy of lungs, pleura, connective tissue septa, distal airways, and structural units. In: Parent, R. A., ed. Treatise on pulmonary toxicology, vol. I. Comparative biology of the normal lung. Boca Raton: CRC Press; 1992:37-48.