Top Banner
ELSEVIER l Original Contribution Ultrasound in Med. & Biol., Vol. 23, No. 9. pp. 3413-1420. 1997 Copyright 0 1997 World Federation for Ultrasound in Medicme & Biology Printed in the USA. All righls rrserved 03Ol-5629/97 $ I7.00 + .OO PI1 SO301-5629(97)00143-9 LIVER HEMOSTASIS USING HIGH-INTENSITY FOCUSED ULTRASOUND SHAHRAM VAEZY,* ROY MARTIN, +* UDO SCHMIEDL,* MICHAEL CAPS,~ SHARI TAYLOR, M.D., II KIRK BEACH, §* STEVE CARTER,* PETER KACZKOWSKI,¶ GEORGE KEILMAN,** SCOTT HELTON,~ WAYNE CHANDLER,# PIERRE MOURAD,~ MATTHEW RICE,+? RONALD ROY** AND LAWRENCE CRUM¶* *Bioengineering, *Anesthesiology, *Radiology, “Surgery, IIPathology, ‘Applied Physics Laboratory and #Laboratory Medicine, University of Washington, Seattle, WA; **Sonic Concepts, Woodinville, WA; ‘+Madigan Hospital, Fort Lewis. WA; and **Aerospace and Mechanical Engineering, Boston University, Boston, MA (Received 3 March 1997; injnal form 7 July 1997) Abstract-Liver hemorrhage, the major cause of death in hepatic trauma, is notoriously difficult to control. We report on the use of high-intensity focused ultrasound (HIFU) to arrest the bleeding from incisions made in rabbit livers. A HIFU transducer, with a spherically curved aperture of 6.34 cm’ area, a focal length of 4 cm and a frequency of 3.3 MHz was used. In approximately 94% of the incisions, the hemorrhage was reduced to a slow oozing of blood in less than 2 min. The maximum temperature of liver tissue around the incision area, during HIFU application, was measured to be 86°C. The mechanism of hemostasis, confirmed by histological examina- tion, appears to be coagulative necrosis of a volume of liver tissue around the incision. We believe that acoustic hemostasis, with the unique characteristic of “volume cauterization,” offers a novel method for the management of liver hemorrhage and, thus, has major clinical implications. 0 1997 World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasound, HIFU, Liver, Trauma, Surgery, Temperature, Coagulative necrosis, Rabbit, Hemor- rhage, Acoustic hemostasis. INTRODUCTION Uncontrolled hemorrhage of the liver is the primary cause of death in hepatic trauma, with a mortality rate of lo-15% (Beal 1996). Major blood lossresulting in com- plicating hemorrhagic shock often occurs prior to the patient’s arrival at surgery. After arrival, the surgeon faces a major challenge in arresting severe bleeding due to several factors. First, the liver is the largest internal organ in the body, by weight, consisting on average of 2.7% of the total adult body weight, about 1500 g (Sny- der et al. 1975). Second, the liver is highly vascularized, with an average blood flow of about 100 mL/lOO mg/ min, or about 25% of the resting cardiac output (Rowe11 1986). Third, the liver is soft and tears easily, with the ultimate tensile strength of 2.4 g/mm* and the ultimate percentage of elongation of 46%. In perspective, these values are 22% and 72% less than those of the cardiac muscle, respectively (Yamada 1973). As a consequence Address correspondence to: Roy W. Martin, Ph.D., Departments of Anesthesiology and Bioengineering. University of Washington, Box 356540, Seattle, WA 98195 USA. E-mail: rmartinu.washington.edu of thesefactors, arresting rapid liver hemorrhage or even slow oozing of blood with conventional methods is a time-consuming and difficult task. In addition, liver is particularly susceptible to injury in trauma. The large mass undergoes high deceleration forces on impact (e.g., car accidents)and, because of its low tensile strength, the liver ruptures easily under such stress.Also, the large surface profile of the liver gives it a high probability of being struck by penetrating objects (e.g., gun shots). Consequently, hepatic injury occurs in 30% of penetrat- ing and 15-20% of blunt abdominal trauma (Beitsch 1994). Clearly, better and more rapid ways of treating patients after a traumatic hepatic injury are needed in both civilian and military environments, where rapid assessment and stabilizing treatment is imperative if de- creases in morbidity and mortality are to be realized. We have been interested in the possibility of using high-intensity focused ultrasound (HIFU) to arrest liver hemorrhage. The therapeutic use of focused ultrasound was first proposed in the 1940s (Lynn et al. 1943) and then pursuedby Fry and colleaguesstarting in the 1950s (WJ Fry et al. 1954; WJ Fry et al. 1955). The research 1413
8

Liver hemostasis using high-intensity focused ultrasound

Mar 08, 2023

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Liver hemostasis using high-intensity focused ultrasound

ELSEVIER

l Original Contribution

Ultrasound in Med. & Biol., Vol. 23, No. 9. pp. 3413-1420. 1997 Copyright 0 1997 World Federation for Ultrasound in Medicme & Biology

Printed in the USA. All righls rrserved 03Ol-5629/97 $ I7.00 + .OO

PI1 SO301-5629(97)00143-9

LIVER HEMOSTASIS USING HIGH-INTENSITY FOCUSED ULTRASOUND

SHAHRAM VAEZY,* ROY MARTIN, +* UDO SCHMIEDL,* MICHAEL CAPS,~

SHARI TAYLOR, M.D., II KIRK BEACH, §* STEVE CARTER,* PETER KACZKOWSKI,¶

GEORGE KEILMAN,** SCOTT HELTON,~ WAYNE CHANDLER,# PIERRE MOURAD,~

MATTHEW RICE,+? RONALD ROY** AND LAWRENCE CRUM¶* *Bioengineering, *Anesthesiology, *Radiology, “Surgery, II Pathology, ‘Applied Physics Laboratory and #Laboratory Medicine, University of Washington, Seattle, WA; **Sonic Concepts, Woodinville, WA; ‘+Madigan Hospital, Fort

Lewis. WA; and **Aerospace and Mechanical Engineering, Boston University, Boston, MA

(Received 3 March 1997; injnal form 7 July 1997)

Abstract-Liver hemorrhage, the major cause of death in hepatic trauma, is notoriously difficult to control. We report on the use of high-intensity focused ultrasound (HIFU) to arrest the bleeding from incisions made in rabbit livers. A HIFU transducer, with a spherically curved aperture of 6.34 cm’ area, a focal length of 4 cm and a frequency of 3.3 MHz was used. In approximately 94% of the incisions, the hemorrhage was reduced to a slow oozing of blood in less than 2 min. The maximum temperature of liver tissue around the incision area, during HIFU application, was measured to be 86°C. The mechanism of hemostasis, confirmed by histological examina- tion, appears to be coagulative necrosis of a volume of liver tissue around the incision. We believe that acoustic hemostasis, with the unique characteristic of “volume cauterization,” offers a novel method for the management of liver hemorrhage and, thus, has major clinical implications. 0 1997 World Federation for Ultrasound in Medicine & Biology.

Key Words: Ultrasound, HIFU, Liver, Trauma, Surgery, Temperature, Coagulative necrosis, Rabbit, Hemor- rhage, Acoustic hemostasis.

INTRODUCTION

Uncontrolled hemorrhage of the liver is the primary cause of death in hepatic trauma, with a mortality rate of lo-15% (Beal 1996). Major blood loss resulting in com- plicating hemorrhagic shock often occurs prior to the patient’s arrival at surgery. After arrival, the surgeon faces a major challenge in arresting severe bleeding due to several factors. First, the liver is the largest internal organ in the body, by weight, consisting on average of 2.7% of the total adult body weight, about 1500 g (Sny- der et al. 1975). Second, the liver is highly vascularized, with an average blood flow of about 100 mL/lOO mg/ min, or about 25% of the resting cardiac output (Rowe11 1986). Third, the liver is soft and tears easily, with the ultimate tensile strength of 2.4 g/mm* and the ultimate percentage of elongation of 46%. In perspective, these values are 22% and 72% less than those of the cardiac muscle, respectively (Yamada 1973). As a consequence

Address correspondence to: Roy W. Martin, Ph.D., Departments of Anesthesiology and Bioengineering. University of Washington, Box 356540, Seattle, WA 98195 USA. E-mail: rmartinu.washington.edu

of these factors, arresting rapid liver hemorrhage or even slow oozing of blood with conventional methods is a time-consuming and difficult task. In addition, liver is particularly susceptible to injury in trauma. The large mass undergoes high deceleration forces on impact (e.g., car accidents) and, because of its low tensile strength, the liver ruptures easily under such stress. Also, the large surface profile of the liver gives it a high probability of being struck by penetrating objects (e.g., gun shots). Consequently, hepatic injury occurs in 30% of penetrat- ing and 15-20% of blunt abdominal trauma (Beitsch 1994). Clearly, better and more rapid ways of treating patients after a traumatic hepatic injury are needed in both civilian and military environments, where rapid assessment and stabilizing treatment is imperative if de- creases in morbidity and mortality are to be realized.

We have been interested in the possibility of using high-intensity focused ultrasound (HIFU) to arrest liver hemorrhage. The therapeutic use of focused ultrasound was first proposed in the 1940s (Lynn et al. 1943) and then pursued by Fry and colleagues starting in the 1950s (WJ Fry et al. 1954; WJ Fry et al. 1955). The research

1413

Page 2: Liver hemostasis using high-intensity focused ultrasound

Ultrasound in Medicine and Biology Volume 23. Numher 9, 1997

was aimed at applying focused ultrasound from outside the body to attack tumors deep within the brain without destroying intervening tissue (Barnard et al. 1956). Since this early work, investigation of noninvasive acoustic surgery has continued, as reviewed by a number of authors (FJ Fry 1993; Hill and ter Haar 1995; Sanghvi and Hawes 1994; Crum and Hynynen 1996; Holm and Skjoldbye 1996). Specifically, the use of HIFU to treat liver cancer and benign prostatic hypertrophy has been studied (ter Haar 1995; Gelet et al. 1993; Yang et al. 1992; Damianou and Hynynen 1994; Prat et al. 1995; Fan and Hynynen, 1996a,b). Currently, we are aware of clinical trials in several countries, including England, for treating liver cancer (Hill and ter Haar 1995), and Japan and Austria for treating benign prostatic hypertrophy (Uchida et al. 1995; Madersbacher et al. 1993). In regard to applications in treating liver and prostate diseases, the vision of acoustic surgery, born earlier this century, is at the forefront of clinical practice.

A characteristic of the acoustic lesions created by HIFU in livers is that they are effectively self-cauterized. Lesion tissue shows coagulative necrosis, with little or no flow of blood from the healthy normal region to the affected tissue (Chen et al. 1993; ter Haar and Robertson 1993). This result suggests the use of focused ultrasound for cauterizing. Three published papers have described the successful use of HIFU to occlude blood flow in intact arteries in viva (Delon-Martin et al. 1995; Hynynen et al. 1996a,b). However, there is no docu- mented successful application of HIFU to cauterize a bleeding fresh wound.

We became interested in using ultrasound to pro- duce hemostasis in the hope that such methods would ultimately provide a lifesaving treatment during the so- called “golden hour,” a brief grace period after a severe injury in which proper therapy can save the life of the injured. An acoustic hemostasis method, capable of ex- tracorporeal application, will provide a valuable method for military and civilian emergency medicine. During the course of this investigation, we have discovered the potential of ultrasound to treat bleeding liver in a surgical environment, a step toward our long-term goal of extra- corporeal acoustic hemostasis. The purpose of this paper is to report the first successful application of HIFU in producing hemostasis in hemorrhagic liver.

METHOD

The HIFU transducer had an aperture of 6.34 cm* that was spherically curved to produce a focus at 4 cm. It was operated at 3.3 MHz, continuous mode, producing an acoustic power of 65 watts, measured with an ab- sorber-type radiation force balance (Sonic Concepts, Woodinville, WA). This power was produced by apply-

ing 300 peak-to-peak volts to the transducer matching network. The half-maximum beam width and length. I mm and 10 mm, respectively, were measured using a target reflection measurement in a water tank. The spatial peak intensity at the focus, defined as the acoustic power divided by the cross-sectional area of the focal spot, was approximately 3000 W/cm’ at the focus.

We performed acute studies on 10 New Zealand rabbits (3.5-4.5 kg weight): 8 for the purpose of study- ing the ability of HIFU to stop the bleeding, and 2 for the purpose of performing thermal measurements during HIFU application. The procedures were carried out ac-

cording to the guidelines of the United States National Institutes of Health (NIH) for the use of laboratory ani- mals. The animals were preanesthetized by a SC injec- tion of 0.25 mL Acepromizine@ (10 mg/mL) and 1.5 mL of Ketaset@ (100 mg/mL), and subsequently adminis- tered an IV injection of 0.2 to 0.4 mL of an 8:l mixture of Ketaset@ (100 mg/mL) and Rompin@ (20 mg/mL). Additional injections of the same mixture were used, as needed, to maintain effective anesthesia. The abdomen was opened with a midline incision from the xiphoid process to the level of the kidneys. The animal was submerged in warm (approximately 40°C) tap water from the chest down. A portion of the liver was exposed for study by pulling on the distal tip slightly with padded forceps. An incision was made in a region of the liver with a scalpel. Profuse bleeding began immediately after the incision was made (Fig. la). The HIFU probe was held under the water and swept at a distance of approx- imately 4 cm (the transducer’s focal length) from the surface of the incised liver (Fig. lb). The sweep speed, and the number of passes along the incision varied, depending on the amount of bleeding. Incisions with severe bleeding required slower sweeps with more passes. On the average, the sweep speed was approxi- mately 1 mm/s. The incision and the probe were in direct view. The probe was aimed manually and visually at the site to be treated. The targeting and duration of exposure was controlled while visually monitoring the results. Two time-points in the experiment were measured: I. The time for reducing the profuse bleeding to a slow oozing of blood was defined as the major hemostasis time; 2. The time for arresting the bleeding entirely was defined as the complete hemostasis time. In most exper- iments, after hemostasis was achieved at one incision site, another lobe of the liver was exposed and the incision and treatment processes were repeated. The depth and length of the incisions were measured after the HIFU treatment.

A total of 27 incisions and HIFU treatments were conducted in the animals. Control incisions were made in 4 instances, in 4 different rabbits. The 4 control incisions were allowed to bleed for 2, 3,4 and 10 min. after which

Page 3: Liver hemostasis using high-intensity focused ultrasound

Liver hemostasis using HIFU 0 S. VAEZY et al. 1415

Fig. 1. The sequence of the experiment: (a) The incision and its profuse bleeding; (b) the HIFU transducer in action; and (c) the arrest of hemorrhage and completion of the HIFU treatment.

HIFU was applied to arrest the hemorrhage, except in the last case. In the last case, the animal was euthanized at the IO-min time-point, and the untreated liver incision region was used for histological examination. Represen- tative tissue samples were obtained from treated and untreated regions for histological examination. The sam-

ples were fixed in formalin, embedded in paraffin, and stained with hematoxylin and eosin.

Thermal measurements were performed in the HIFU treatment of 14 incisions, made in the livers of 2 rabbits. A Chromega/Constantan thermocouple with 0.005inch diameter wires was used for temperature measurements. The linear response of the thermocouple was verified in a water bath from 30°C to 100°C. The response times were determined by rapid immersion in 80°C and 0°C water (i.e., temperature elevation and drop from the room temperature, 25°C). The response times, defined as the time between the immersion and the pla- teau of the final temperature, were 50 ms and 100 ms for the temperature elevation and drop, respectively. These response times are much faster than all of thermal re- sponses to treatment recorded in the liver. The thermo- couple was inserted in the liver parallel to the surface, so that the thermal junction was approximately 5 mm deep. The position of the thermocouple was fixed using super- ficial sutures in the liver and the skin. An incision was made at a distance of l-2 mm from the junction of the thermocouple, and HIFU was applied to stop the bleed- ing. The thermal measurements continued for the dura- tion of the HIFU treatment.

All animals were alive at the end of experiment, when they were euthanized with an overdose of the same anesthetic mixture, followed by a 2-mL KC1 saturated solution within 2 min.

RESULTS

Figure 1 shows the sequence of the acoustic hemo- stasis experiment. Profuse bleeding of the liver in the water tank occurred immediately after the incision was made (Fig. la). The intense acoustic field induced acous- tic streaming in the water surrounding the incision area (Fig. lb). The streaming often moved the blood away from the bleeding site, thus permitting a clear visualiza- tion of the area to be treated. The HIFU treatment was stopped when the hemorrhage was completely arrested (Fig. lc). The treatment produced a discoloration of the tissue around the incision from its natural maroon to a gray-tan color, appearing necrotic or “cooked”. The mean + standard deviations of the length and depth of the incisions were 16.1 + 4.87 mm and 4.2 ? 1.94 mm, respectively.

Bleeding from surface capillaries (< 0.5 mm in diameter) of the incision was arrested almost immedi- ately (a few s) after the application of HIFU. More time was required to stop the bleeding from larger vessels (0.5-2.5 mm in diameter) in the incision. In approxi- mately 94% of the incisions, major hemostasis was achieved in a time less than or equal to 2 min of contin- uous application of HEW (Fig. 2). The average -+- stan-

Page 4: Liver hemostasis using high-intensity focused ultrasound

Lltrasound in Medicine and Biology Volume 23, Number Y, IYY7

3% 3%

t t

0.5 1 1.5 2 2.5 3

Major Hemostasis Time (minutes)

Fig. 2. Histogram of the HIFU treatment time required to produce major hemostasis in 27 liver incisions.

dard deviation of the major hemostasis time was 1.4 + 0.69 min. We found that, in 7 cases, after achieving major hemostasis additional application of HIFU was necessary for complete hemostasis. The additional HIFU application appeared to be required by those incisions that included a large cut vessel (i.e., larger than 2.5 mm in diameter). In approximately 80% of the incisions, complete hemostasis was achieved in a time less than or equal to 3 min of continuous HIFU application (Fig. 3).

Figure 4 shows a representative plot of temperature vs. time in an incision during HIFU application to stop the bleeding. The plot shows 3 major temperature eleva- tions, each followed by a cooling period. Each peak occurred when the HIFU focal point passed by the ther- mocouple. The maximum temperature was reached within 1-2 s of HIFU application. The range of the maximum temperature elevation, in 14 thermal measure- ments, was 59-86°C. The range of the cooling time from the peak temperature to the baseline of 40°C was 40- 60 s.

12 T--- I

Fig. 3. Histogram of the HIFU treatment time required for achieving complete hemostasis in 27 liver incisions.

30 A-- ~~~~~~~- --- 0 20 40 60 80 100

Time (seconds)

Fig. 4. A representative plot of liver tissue temperature as a function of time. The temperature measurements were obtained in a typical incision during HIFU application to arrest the

bleeding.

Figure 5 is a photograph of the liver sample used for histological examination. Region 1, from a control inci- sion made 10 min before euthanasia, was used as a normal liver sample, and region 2, from a HIFU-treated incision, was used as a treated liver sample. On light microscopic examination, the untreated liver exhibited normal architecture with no evidence of cell injury or necrosis (Fig. 6A). The HIFU-treated region was non- uniform in appearance (Fig. 6B). There were scattered, well-demarcated zones of coagulative necrosis, alternat- ing with small foci of cavitation and tissues containing only minimal histologic alteration. The zones of coagu- lative necrosis had distorted hepatic architecture with detachment of individual hepatocytes, although the cy- toplasmic borders were preserved (Fig. 6C). The nuclei were dark and condensed (i.e., pyknotic), indicative of coagulative necrosis, and the cytoplasm was deeply stained. Portal tracts were similarly affected by the ther-

Fig. 5. Photograph of the rabbit liver. The scale shown is in cm. The bright, shiny areas are reflections of the camera flash. Regions used for histological studies: 1. Untreated region; and

2. HIFU-treated region.

Page 5: Liver hemostasis using high-intensity focused ultrasound

Liver hemostasis using HIFU 0 S. VAEZY et al. 1417

Fig. 6. Photographs of histological sections (magnification bars equal 200 (pm): A. Untreated liver, showing intact architecture and no evidence of necrosis. A portal tract in the center of the field shows normal architecture of vein, artery, and the bile duct; B. treated liver showing coagulative necrosis (left half of field, arrows) with disruption of hepatic architecture, detachment of individual hepatocytes, and smudged, pyknotic nuclei. Also seen are foci of cavitation that contain erythrocytes (arrowhead 1) or are empty (arrowhead 2); C. coagulative necrosis in the treated liver involving a portal tract (center of the field) and adjacent hepatocytes; D. liver tissue adjacent to the HEW-treated region, showing

intact architecture with sinusoidal congestion (arrowheads). A central vein is in the center of the field (arrow).

ma1 injury, with distortion and necrosis of the artery, vein and bile duct (compare Fig. 6A with 6C). The areas of cavitation were either empty or contained amorphous, eosinophilic debris or erythrocytes. The areas adjacent to the treated region contained only minimal alteration, with normal architecture and congestion of the hepatic sinusoids by erythrocytes (Fig. 6D). The congestion was thought to be due to blockage of blood vessels in the adjacent treated areas. Vascular structures were also af- fected by the HIFU treatment, although not preferen- tially. Larger portal vein branches within the treated region displayed cellular discohesion and pyknosis, par- ticularly in the tunica media, with disruption of the endothelial lining (compare Fig. 7A with 7C). Similarly, coagulative necrosis also involved the endothelial lining and thin layer of investing connective tissue of the cen- tral veins (compare Fig. 7B with 7D). No thrombi were

identified in either the central veins or portal vein branches.

DISCUSSION

The mechanism of acoustic hemostasis appears to be coagulative necrosis due to an elevation in the tem- perature of the liver tissue in the focal region of HIFU. The coagulative necrosis appeared to cause a collapse of blood vessels (sinusoids and vessels up to 2 mm in diameter), leading to hemostasis. Our thermal results show clearly that HIFU produces a temperature elevation in tissue, confirming the well-documented thermal effect of HIFU (FJ Fry 1993; ter Haar 1995; Lele 1977). The histological results also confirm that the HIFU treatment of the liver was predominantly due to thermal injury, as evidenced by the presence of coagulative necrosis. These findings are comparable to those of the other investiga-

Page 6: Liver hemostasis using high-intensity focused ultrasound

Volme 73. Nwnbrr 9. 1997

Fig. 7. Photographs of histological sections showing vessel walls (delineated by arrowheads) in the normal and trcarecl liver (magnifications bars equal 100 (pm). A. The wall of a normal larger portal vein branch: R. normal central vein: C. the wall of a HIFU-treated portal vein branch. The cells ofthe tunica media are discohesive and have pyknotic nuclei: D. a central vein in the HIFU-treated region, displaying coagulative necrosis of the vessel wall and the wrrounding

hepatocytes.

tors (Yang et al. 1992: Sibille et al. 1993; Susani et al. 1993), and are similar to the histological changes seen with other types of thermal injury, such as electrocau- tery. These results suggest that HIFU-induced hemosta- sis may be applicable when the normal hemostatic mech- anisms are either too slow (major liver trauma or during liver resection) or are not functioning properly due to platelet or coagulation factor deficiencies.

The in viva measurements of the thermal behavior of the liver tissue as a function of HIFU treatment provided several interesting observations. First, the HIFU beam had to be very close to the thermocouple (l-2 mm) to obtain a thermal response. The temperature elevation at the focus seemed to be highly localized, and to dissipate rapidly spatially. This effect may have been due, in part, to the water around the liver, although we have observed a similar effect in recent studies where water did not surround the liver. The thermocouple was located well within the coagulated region by the end of a treatment period, representing a sampling of a point in

the tissue that would undergo acoustic hemostasis. The maximum measured temperature during HIFU was found to be 86°C. We believe that this temperature represents the temperature of the liver tissue when the HIFU beam was at or very near the focus. for l-2 s exposure. Peak temperature recordings below 86°C may have been due to a larger distance between the focus and the thermocouple. There are a number of other factors for consideration (Malcolm and ter Har. 1996), including the relationship between the temperature elevations and the acoustic absorption, the applied intensity at the site, the specific heat of the tissue, and the rate of heat dissipation in tissue and as a result of blood perfusion. After coag- ulative necrosis starts, all of the above parameters may change, and affect the cooling times observed. Also. the cooling time may vary depending on whether or not the HIFU beam was still in a close distance to the thermo- couple, treating adjacent bleeding tissue. Further studies will be needed in the future to elucidate all the thermal

Page 7: Liver hemostasis using high-intensity focused ultrasound

Liver hemostasis using HIFU 0 S. VAEZ~ er al. I419

variables involved with in viva HIFU treatment of the liver.

Although thermal injury appears to be the mecha- nism responsible for acoustic hemostasis, a mechanical effect of HIFU must also be considered. The mechanical effect of HIFU usually stems from acoustically-induced cavitation activity. Cavitation results from the generation of vapor and gas-filled voids during the negative portions of the acoustic pressure field. The collapse of these voids during the subsequent positive portions of the field can result in enormous concentrations of energy. The inten- sity threshold for cavitation in excised liver, at 3 MHz, has been reported to be 1300 W/cm2 (Chan and Frizzell, 1977), below our estimated spatial peak intensity of 3000 W/cm* at the focus. Except in certain restricted cases (FJ Fry et al. 1995; Crum and Hynynen 1996), excessive cavitation is to be avoided in the use of HIFU for therapeutic applications. At this stage, we are uncertain of the role of cavitation in acoustic hemostasis, and whether its presence is desirable or undesirable.

In the course of HIFU treatment, and as a result of hand movement and deep focusing for optimal hemosta- sis, the acoustic beam was sometimes focused in tissues underlying the liver. consisting of bowels, skin, and stomach. The HIFU produced a thermal effect (slight discoloration) in the underlying tissues. The effect ap- peared to be minimal, because the beam had traveled through a layer of liver, to reach the underlying tissue with a reduced intensity. We are currently investigating methods to absorb the acoustic beam as it leaves the other side of the liver to prevent any damage to the underlying tissue. Also, a careful investigation of the side effects and precise, controlled application of HIFU is required (Sibille et al. 1993: Chapelon et al. 1990; Watkin et al, 1996).

We used an approach, in applying HIFU, that con- sisted of focusing the ultrasound deep in the tissue that is adjacent to the bleeding site and, thus, inducing hemo- stasis over an extended volume. This is in contrast to conventional cauterization techniques where cauteriza- tion is effective just at the surface, usually resulting in concomitant damage of the tissue surface (Tranberg et al. 1986). We believe that “volume cauterization” is the key to our success in this application and is unique to acous- tic hemostasis.

SUMMARY

We used HIFU to produce hemostasis in surgical incisions made in rabbit livers. The transducer was spherically curved, with an aperture of 6.34 cm*, and a focal length of 4 cm. It was operated at 3.3 MHz, producing a spatial average intensity of about 3000 WI cm*. The average time of major hemostasis was 1.4 min.

In more than 80% of the incisions, complete hemostasis was achieved in 3 min or less. The mechanism of hemo- stasis appears to be coagulative necrosis of the liver tissue. Thermal measurements and histological results confirmed the coagulative necrosis of the treated tissue due to thermal effects of HIFU. The acoustic hemostasis may have significant clinical implications for both trauma and elective liver surgeries.

Acknowledgement-We acknowledge the support of the Defense Ad- vanced Research Programs Administration under its MURI program. This work was supported by a grant from DARPA. Number NOOOIJ- 96-0630.

REFERENCES

Barnard JW. Fry WJ, Brennan JF. Small localized lesions in the white and grey matter of the cat brain. Arch Neural Psychiatry 1956:75: 15-35.

Beal SL. Liver. In: Ivatury RR, Cayten CG. eds. The textbook of pene- trating trauma, Baltimore, MD: William & Wilkins. 1996:571-S8S.

Beitsch P. Liver and biliary tract trauma. In: Lopez-Viego MA, ed. The Parkland trauma handbook. St. Louis, MO: Mosby; 1994:279-296.

Chan SK, Frizzell LA. Ultrasonic threshold for structural changes in the mammalian liver. Ultrasonic Symp. Proc 1977. IEEE Cat No 77CH1264-USA.

Chapelon J, Margonari J. Cathignol D, Gelet A, Guers C. Theillere Y. Thresholds for tissue ablation by focused ultrasound. IEEE Ultra- son Symp 1990: 3:1653-1656.

Chen L, Rivens I, ter Haar G, Riddler S. Hill CR. Bensted JPM. Histological changes in rat liver tumours treated with high-intensity focused ultrasound. Ultrasound Med Biol 1993; 19:67--74.

Crum L, Hynynen K. Sound therapy. Physics World 1996:9:2X-33. Damianou C. Hynynen K. Focal spacing and near-field heating during

pulsed high temperature ultrasound therapy. Ultrasound Med Biol 1993:19:777-787,

Damianou C. Hynynen K. The effect of various physical parameters on the size and shape of necrosed tissue volume during ultrasound surgery. J Acoust Sot Am 1994;95: 1641-l 649.

Delon-Martin C. Vogt C. Chignier E. Guers C. Chapelon 1. Cathignol D. Venous thrombosis generation by mean\ of high-intensity fo- cused ultrasound. Ultrasound Med Biol 1995:2 1: I 1% I 19.

Fan X, Hynynen K. A study of various parameters ot spherically curved phased arrays for noninvasive ultrasound surgery. Phys Med Biol 1996a;41:591-608.

Fan X. Hynynen K. Ultrasound surgery using multiple somcations- treatment time considerations. Ultrasound Med Biol 19Y6h:22:47 I- 482.

Fry FJ. Intense focused ultrasound in medicine. Eur l.Jrol 1993: 23(suppl 1):2-7.

Fry FJ, Sanghvi NT, Foster RS, Bihrle R. Hennige C. Ultrasound and microbubbles: Their generation. detection and potential utilization in tissue and organ therapy-experimental. LJltraqound Med Biol 1995;21:1227-1238.

Fry WJ. Barnard JW. Fry FJ. Krumins RF. Brennan JI‘. Ultrasonic lesions in the mammalian central nervous system. Science 1955: 122517-518.

Fry WJ. Mosberg WH. Barnard, JW, Fry FJ. Production of focal destructive lesions in the central nervous system with uitrdsound. J Neurosurg 1954;l l:471-478.

Gelet A. Chapelon JY. Margonari J, Theillere Y. Gerry-F. Cathignol D. Blanc E. Prostatic tissue destruction by high-intensity focused ultrasound: experimentation on canine prostate. J Endourol 1993: 71249-53.

Hill CR. ter Haar GR. Review article: High intensity focused ultra- sound-potential for cancer treatment. Br J Radio1 1995:68: 1296- 1303.

Holm HH. Skjoldbye B. Interventional ultrasound. Ultl-nsound Med Biol 1996:22:773-789.

Page 8: Liver hemostasis using high-intensity focused ultrasound

Ultrasound in Medicine and Biology Volume 23. Number 0. 1997

Hynynen K, Chung A. Colucci V. Jolesz F. Potential adverse effects of high-intensity focused ultrasound exposure on blood vessels irr ,~i~~~~. Ultrasound Med Biol 1996a;22: 193-201.

Hynynen K. Colussi V. Chung A, Jolesz F. Noninvasive arterial oc- clusion using MRI-guided focused ultrasound. Ultrasound Med Biol lYY6b;22:1071-1077.

Lele PP. Thresholds and mechanisms of ultrasonic damage to orga- nized animal tissues. In: Hazzard DWG, Litz ML, eds. Symposium on biological effects and characterization of ultrasound sources. HEW Publication (FDA) 787-8048. Washington, DC:US GPO: 1977:234-237.

Lynn JG. Zwemer RL, Chick AJ, Miller AF. A new method for the generation and use of focused ultrasound in experimental biology. J Gen Physiol 1942;26: 179-193.

Madersbacher S, Kratzik C, Szabo N, Susani M. Vingers L, Marberger M. Tissue ablation in benign prostatic hyperplasia with high-inten- sity focused ultrasound. Eur Ural 1993;23 Suppl 1:39-43.

Malcolm A. ter Haar G. Ablation of tissue volumes using high intensity focused ultrasound. Ultrasound Med Biol 1996;22:659-669.

Prat F. Centnrti M, Sibille A, Abou El Fadil F, Henry L. Chapelon JY, Cathignol D. Extracorporeal high-intensity focused ultrasound for VX2 Liver tumors in the rabbit. Hepatology 1995;21:832-836.

Rowell LB. Human circulation regulation during physical stress. New York: Oxford University Press: 1986:79-83.

Sanghvi NT. Hawes RH. High-intensity focused ultrasound. Gastroin- test Endosc Clin N Am 1994;4:383-95.

Sibille A, Prat F. Chapelon JY. El-Fadil FA, Henry L, Theillere Y.

Ponshon T. Cathignol D. Extracorporeal ablation of libcr tissue by high intensity focused ultrasound. Oncology 1993:50:375-379.

Snyder WS, Nassrt ES, Karhausen LR, Howells GP, Tipton IH. Report of the task group on reference man. Oxford: Pergamon Presh: 1975.

Susani M, Madersbacher S. Kratzik C. Vingers 1.. Marbergrr M. Morphology of tissue destruction induced by focused ultrasound. Eur Ural 1993:23 Suppl 1:34--3X.

ter Haar G, Robertson D. Tissue destruction with focused ultrasound I?! ~~ivo. Eur Ural 1993;23 Suppl 1:8-l].

ter Haar G. Ultrasound focal beam surgery. Ultrasound Med Biol 1995;21:1089-1100.

Tranberg KG. Rigotti P, Brackett KA, Bjomson HS, Fischer JE. Joffe SN. Liver resection. A comparison using the Nd-YAG laser. an ultrasonic surgical aspirator, or blunt dissection. Am J Surg 1986: 151:368-73.

Uchida T. Yokoyama E. Iwamura M, Koshiba K. Terai A. Terachi T. Ohishi K, Yoshida 0. High intensity focused ultrasound for benign prostatic hyperplasia. Int J Ural 1995;2: 181-5.

Watkin N, ter Haar G, Rivens 1. The intensity dependence of the site of maxima1 energy deposition in focused ultrasound surgery. Ultra- sound Med Biol 1996;22:483-9 I.

Yamada H. Strength of biological materials. Huntington, NY: Robert E. Krieger Publishing Co. 1973: 106-109, 203-204.

Yang R. Sanghvi N, Rescorla F, Galliani C, Fry F, Griffith S, Grosfeld J. Extracorporeal liver ablation using sonography-guided high- intensity focused ultrasound. Invest Radio1 1992:27:796-803.