Color
Doppler Imaging
Technique
of the
and Normal Vascular
Eye
and Orbit
Anatomy
Wolfgang E. Lieb, MD; Steven M. Cohen, MD; Daniel A. Merton; Jerry A. Shields, MD; Donald G. Mitchell, MD; Barry B. Goldberg, MD \s=b\ Color Doppler imaging is a recent advance in ultrasonography that allows simultaneous two-dimensional imaging of structure and blood flow to be performed. Doppler information is superimposed in color over a conventional grayscale ultrasound image. Using this technique, we examined 40 normal eyes. The central retinal artery, posterior ciliary arteries, ophthalmic artery, and central retinal vein could be located in all patients. Using the color image as a guide, Doppler spectral analysis allows quantitative assessment of blood flow velocity in these vessels. Color Doppler imaging is a new modality for the study of ocular and orbital hemodynamics.
(Arch Ophthalmol. 1991;109:527-531) "D eal time A-mode and B-mode ultrasonography has been used for the diagnostic evaluation of ophthalmic disorders since the early 1960s.M Mod¬ ern
digital high-resolution equipment
has improved diagnostic imaging and made it an essential part of ophthalmologic evaluations. Ultrasonography is routinely used today for display, dif¬ ferentiation, and measurement of in¬ traocular and orbital tumors and tu¬ morlike lesions, detection of retinal detachments in eyes with opaque me¬ dia, axial-length measurements for calAccepted for publication August 21, 1990. From the Oncology Service, Wills Eye Hospital (Drs Lieb and Shields), Philadelphia, Pa, and the Department of Radiology, Division of Ultrasound, Jefferson Medical College, Thomas Jefferson University, Philadelphia (Drs Cohen, Mitchell, and Goldberg and Mr Merton). Reprint requests to Oncology Service, Wills Eye Hospital, Ninth and Walnut Streets, Phila-
delphia, PA 19107 (Dr Shields).
culation of intraocular lens power, and evaluation of traumatic injuries to the globe and orbit. In addition, ultrasono¬ graphy is useful to detect and diagnose optic nerve lesions such as optic nerve sheath meningiomas, inflammatory flu¬ id collection in the nerve sheath, and orbital tumors or inflammatory con¬ ditions. See also pp 522, 532, 537, and 582.
Doppler ultrasound is a method of detecting changes in the frequency of sound reflected from flowing blood, allowing estimation of flow velocity. Conventional continuous-wave Dopp¬ ler ultrasonography of the carotid ar¬ teries and the periorbital vessels is frequently employed in patients with ischemie ocular disease, and has also been used to evaluate the ophthalmic artery.' The technology of Duplex scanning allows simultaneous B-mode imaging and Doppler evaluation to be performed. Since the lumina of the
vessels in the eye and orbit are too small to be imaged with conventional Duplex scans, Doppler spectra are ob¬ tained without precise localization and without knowledge of the Doppler an¬ gle. To facilitate localization and as¬ sessment of the Doppler angle, the two-dimensional flow information in color Doppler imaging (CDD is en¬ coded in color and superimposed on the gray-scale image of the structure.M Color Doppler imaging is currently being used in echocardiography,8 for evaluation of peripheral arterial and venous disease, for evaluation of the vascular pattern of the genitourinary system,910 and in the normal and patho-
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
state of abdominal organ perfu¬ sion. One major application is the ex¬ amination of the extracranial vessels supplying the central nervous system, ie, the carotid and vertebral arter¬ ies,1112 and the qualitative and quanti¬ tative analysis of neonatal intracranial vasculature.13"1' This modality provides the advantage of rapid orientation and provides an "anatomic" display by de¬ picting Doppler shifts throughout the entire image. There are only limited reports in the literature of applications of CDI in the study of the eye and orbit. Canning and Restori1" applied this technique to quantify rectus-muscle movement dur¬ ing voluntary saccades. Laser Doppler studies were performed for the retinal vessels by Riva et al" and duplexDoppler has also been used to demon¬ strate flow in intraocular tumors18 and to evaluate the central retinal artery (CRA), central retinal vein (CRV), and ophthalmic artery. 19,2° We have report¬ ed the value of CDI in the noninvasive diagnosis of an orbital varix21 and in the diagnosis of cavernous sinus fístu¬ las and intraocular tumors (W.E.L. and P. M. Flaharty, MD, unpublished data, 1990). Recently, Erickson et al22 described their experience using CDI in the normal orbit and in the evalua¬ tion of seven patients with pathologic conditions in the orbit. We herein outline our technique of using CDI to visualize the normal vas¬ cular system of the eye and orbit, and discuss our initial experience examin¬ ing 40 normal eyes, the preliminary results, and indications for future po¬ tential applications of this new imaging mode.
logic
PATIENTS AND METHODS
imaged to facilitate determination of Doppler flow angle for estimation of velocity. The scan images were digitally recorded on videotape and later reviewed with the benefit of cine-loop and frame-byframe analysis of selected segments. Im¬ ages were photographed during cine-loop replay of the recording by using a 35-mm camera that photographed directly from an were
Maximum Systolic Blood Flow Velocities in Orbital Vessels (N 40)
the
All examinations were performed with a color Doppler unit (QAD I, Quantum Medi¬ cal Systems Ine, Issaquah, Wash) using a 7.5-MHz linear phased array transducer. The estimated in situ peak temporal aver¬ age (SPTA) intensity at 4.2-cm depth in the color imaging mode (provided by Quantum Medical Systems) is 7 mW/cm for the 7.5MHz transducer. When spectrum analysis is performed, the SPTA intensity is approx¬ imately 71 mW/cm". In the spectrum analy¬ sis mode, the SPTA intensity exceeds the limits suggested by current Food and Drug Administration guidelines for ophthalmic application of 17 mW/cm2, but is lower than those given in the ultrasound safety guide¬ lines of the American Institute for Ultra¬ sound in Medicine. The study was ap¬ proved by the institutional review board of the Thomas Jefferson University Hospital (Philadelphia, Pa) and standardized in¬ formed consent was obtained from all sub¬ jects examined. Doppler spectral analysis lasted approximately 45 seconds per eye. The ultrasound transducer was applied to the closed eyelids using sterile ophthalmic
methylcellulose
as a coupling gel. During examination, the patient was in a supine position and care was taken not to apply
the
pressure to the eye to avoid artifacts. Hori¬ zontal and vertical scans through the eye and orbit were performed. Depending on the direction of flow with respect to the transducer, the blood flow data are dis¬ played in either red or blue. The colors can be arbitrarily assigned but, in this study, flow toward the transducer is depicted as red and away from the transducer as blue. The color saturation in the image repre¬ sents the average frequency (first moment average) from a spectral analysis performed at each sample site. These frequencies can be turned into velocities by solving the Doppler equation for velocity. When examining the eye and orbit through the eyelids, the ultrasound beam is essentially parallel to the orbital and ocular vessels; thus, most arterial flow is depicted in red. Arteries can usually be distin¬ guished from veins by noting their pulsatility. Pulsed Doppler spectral analysis also helped to distinguish between pulsatile ar¬ terial flow and the usually more continuous or minimally pulsatile venous flow, and allowed for the quantification of data. When the ultrasound beam was at an angle of 90° to a vascular structure or if a vessel con¬ tained only stagnant blood, no Doppler flow information was obtained and the structure was shown in gray-scale display only. All examinations were performed in a "low" or "medium" flow setting to allow for optimal detection of low to medium Doppler frequency shifts of the slow-flowing blood in the small orbital vessels. For the ophthal¬ mic artery, medium to high flow settings were applied since flow in this vessel is faster. Color threshold levels were adjusted to minimize artifacts by lid and involuntary eye movements. To obtain Doppler spectra, a sample volume of approximately 0.2 0.2 mm was chosen within the vessel. The proximal and distal portions of the vessel
isolated on-board color monitor. The dura¬ tion of the examination ranged from 10 to 20 minutes. Using the above-described technique, we examined 10 healthy volunteers on two separate occasions and the normal eyes of 30 patients with intraocular tumors in their other eyes. Our study population was as follows: 18 men and 12 women; mean age (±SD), 45.34 ±16.3 years (range, 20 to 76 years). None of the patients or the volun¬ teers had a history of ocular or systemic vascular disease or any clinically evident vascular anomaly in the examined eye and orbit. RESULTS
A horizontal scan through the globe and orbit at the level of the optic nerve allows the depiction of the CRA, and CRV (Fig 1). The CRA and accompa¬ nying CRV can be seen within the anterior 2 mm of the optic nerve shad¬ ow. The CRA can be traced up to where it enters the optic nerve approx¬ imately 13 mm behind the optic disc
(Fig 2).
Differentiation of the CRV is quite easy, not only through color coding as a blue structure but also by its Doppler spectral characteristics, ie, its continu¬ ous flow throughout systole and diasto¬ le. Color information alone can some¬ times be misleading since vessel loops may simulate flow away from the probe and thus be coded as blue. In this case, the pulsatility of the arteriole clearly distinguishes it from the usually nonpulsatile vein. On either side of the optic nerve slightly posteri¬ or to the CRA, the short and long posterior ciliary arteries can be identi¬ fied (Fig 1). These vessels and the CRA have pulsatility similar to the
ophthalmic artery, differing mainly by
the maximum of the systolic velocity (Fig 3). In some sections continuous with the central retinal vessels, the flow in the retinal arteriolar vessels could be demonstrated; however, sepa¬ rate branches of the retinal arteriolar system could not be resolved. At the level of the retinal vessels, the venous flow is so low that it usually cannot be detected. However, we were able to depict some of the vortex veins in several cases. The blood flow in the choroid and retina could not always be distinguished unless a retinal detach¬ ment was present, allowing the retinal vessels to be identified in the normally
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
=
Mean ± SD Blood
cm/s Central retinal
artery
Flow, (Range)
10.3 ± 2.1
(6.4-17.2)
Central retinal vein
-2.9 ± 0.73
Ophthalmic
31.4 ± 4.2
artery Posterior
artery
ciliary
(1.9-5.4) (23.5-39.8) 12.4 ± 4.8
(1.4-22.7)
echo-free vitreous cavity. The Doppler spectrum of the posterior ciliary arter¬ ies showed velocity-time spectra very similar to those of the CRA. The end diastolic flow in the posterior ciliary
Fig 1.—A horizontal scan through the globe shows the central retinal artery (CRA) and accompanying vein within the anterior portion of the optic nerve shadow. In the same plane on both sides of the optic nerve (N), Doppler shifts within the short posterior ciliary arteries (PCA) can be demonstrated. Fig 2.—The central retinal artery (CRA) can be traced posteriorly where it enters the optic nerve (N) approximately 13 mm behind the disc (arrow). Fig 3.—Doppler spectrum (velocity-time curve) of the central retinal artery (CRA). The Doppler angle "A" can be corrected by align¬ ment of a cursor in the vessel. The angle correction allows assessment of the velocity
of blood flow in these vessels. The central retinal vein (CRV) runs parallel to the CRA in this section.
Fig 4.-The ophthalmic artery (OPHT ART) be traced deeper into the orbit and can be localized in the majority of patients slightly lateral to the optic nerve (N) as it enters the orbit. Further anteriorly it crosses over the optic nerve to divide into its nasal branches. Fig 5.—Spectrum analysis of the ophthalmic artery (OA) in a medium flow setting. Note the high, steep maximum systolic peak. In the two-dimensional image the portions of the ophthalmic artery nasal to the optic nerve (N) coming toward the transducer could be easily identified. The horizontal portion between the two segments of the ophthalmic artery was not imaged. Since in the horizontal segment the Doppler angle is 90°, no Doppler shift was can
seen.
Fig 6. —Superior ophthalmic vein (SOV) with flow away from the transducer and therefore encoded in blue. In addition to these big ves¬ sels, pulsatile arterial and continuous venous flow inside the eye was seen, most likely rep¬ resenting venous flow in the vortex veins draining into the superior ophthalmic vein. Fig 7. —Spectrum analysis of the superior ophthalmic vein demonstrates the character¬ istic continuous, nonpulsatile venous flow.
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
arteries was, however, higher, reflect¬ the low-resistance vascular chan¬ nels of the choroid. Deeper into the orbit, sections of the main ophthalmic artery could be seen in cross section. The ophthalmic artery was traced temporally to the nerve posteriorly, then traveling anteriorly and nasally where several branches may be depicted (Fig 4). Hayreh and Dass21 and Hayreh2'' have shown in a large series of cadaver dissections that there is considerable variability in the course of the ophthalmic artery and its branches. In their series, the crossing of the ophthalmic artery over the optic nerve in the midorbit was a common finding (80% of cases).242'' The crossing of the ophthalmic artery over the optic nerve could be demonstrated by CDI in all our subjects. The flow-velocity waveform is similar to that of the internal carotid artery, showing a steep maximum systolic peak, often a dicrotic notch, and low diastolic flow velocity (Fig 5). The supraorbital ar¬ tery could be identified when angling the ultrasound transducer more supe¬ riorly and anteriorly. In the superior nasal orbit, we identified the superior ophthalmic vein (Fig 6) fed by the vortex veins and running posteriorly, crossing the optic nerve. The Doppler spectrum with the continuous, nonpul¬ satile flow pattern is characteristic of venous flow (Fig 7). Flow velocity measurements of the CRA (Fig 3) and posterior ciliary ar¬ teries were obtained as a baseline for further evaluations in patients with ocular and orbital vascular disease. In the 10 healthy young individuals, the peak systolic velocities were measured in the CRAs and ophthalmic artery in both eyes after angle correction on two different occasions. The CRA was identified within the anterior optic nerve portion slightly posterior to it entering the globe. The ophthalmic artery was identified during its course parallel to the optic nerve. In both vessels a preset sample volume was placed after the Doppler angle was estimated using the angle-correction menu of the unit and placing a cursor along the course of the vessel of inter¬ est. Mean peak systolic velocities in the CRA ranged from 6.4 cm/s to 17.2 cm/s, with a mean (±SD) of 10.3±2.1 cm/s. The mean ( ± SD) peak systolic velocity in the ophthalmic artery was 31.4 ±4.2 cm/s and in the posterior ciliary arteries was 12.4 ±4.8 cm/s (Table). No significant difference was noted between right and left eyes or between male and female subjects. In the group of normal subjects with paired observations, on two occasions
ing
paired t test was performed for all variables and no statistically signifi¬ cant difference was seen between the first and second measurements. a
COMMENT
Color Doppler imaging combines both anatomic and velocity data and displays the information in a format that can be easily comprehended. Blood-flow velocity information can be obtained quantitatively within the en¬ tire two-dimensional gray-scale image. This has revolutionized noninvasive vascular imaging techniques, and in many areas of medicine, the role of this new technique is well established.2''"3 Color Doppler imaging is being used in other medical specialties to examine the heart," the large abdominal and thoracic vessels,81 the vasculature in the male genitalia,1'1" peripheral limb vessels, and the extracranial and intracranial vasculature.1112 Recently, the importance of this examination for ana¬ tomic delineation and functional disor¬ ders in the cerebrovascular system in newborns has been stressed.131'12 A recent report demonstrated CDIs of two presumed orbital varices, one su¬ perior ophthalmic vein thrombosis, and an orbital arteriovenous malforma¬ tion.23 The authors concentrated on the topographic display of the vascular structures in 26 normal orbits and demonstrated CRA, CRV, the superi¬ or ophthalmic vein, and ophthalmic
artery.
In the present study, CDI was able to localize small vessels in the eye, the orbit, and the optic nerve (Fig 1), expanding on the recent observations
by Erickson and associates.22 Using low-flow settings and appropriate
we were able to detect flow in the retinal vessels as well. Although this technique cannot distinguish the separate portions of the arteriolar bed, we could see what represents a sum¬ mation of these vessels. It can be clearly distinguished from flow in the posterior ciliary arteries and the large vascular channels of the choroid that they supply. Differentiation between the short and long posterior ciliary arteries and distinction among them has so far not been possible in our hands. Color Doppler imaging allows, for the first time, reproducible noninva¬ sive imaging and selective Doppler in¬ formation of orbital vessels. At pre¬ sent, it is not likely that CDI will allow blood volume measurements, since di¬ ameters of the vessels cannot be accu¬ rately assessed. The analysis of selec¬ tively obtained Doppler spectra from the different vessels, ie, CRA, CRV,
threshold,
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
and the posterior ciliary arteries and branches of the ophthalmic artery, might allow some insight into the hemodynamics of the eye and its supply¬ ing vessels. We demonstrated that this technique allows accurate measure¬ ment of flow velocities in the CRA and ophthalmic artery. Our values corre¬ spond to the values found by Berger et
al20 using duplex scanning. They are higher than the flow velocities found by Riva et al17 in the branch retinal arteries and veins using laser Doppler velocimetry. However, our findings for the maximum systolic flow velocity
in the CRA are similar to the results of Feke et al,33 who also used laser Dopp¬ ler velocimetry to assess blood flow in the retinal arterioles. In addition to measuring the velocities, the analysis of the waveforms provides information about the vessel and the vascular sys¬ tem proximal and distal to it. Com¬ pared with the CRA, the spectrum of the posterior ciliary vessels showed a slightly higher systolic velocity and also a higher end diastolic velocity indicating a low-resistance vascular system distal to the sample area. Inaccuracies in our quantitative measurements can arise from errors in interpreting our Doppler-shifted fre¬ quency spectrum and in measuring the Doppler angle. Feke et al33 have re¬ ported their estimate of uncertainty associated with their individual mea¬ surements of the maximum systolic velocity to be about 18% using laser velocimetry. This is the range of error seen in most studies reporting Doppler flow data. As in any new imaging technique, the safety of pulsed Doppler ultra¬ sound is of concern. Color Doppler imaging involves only slightly higher ultrasound exposure than conventional B-mode examination. However, when spectral analysis is performed, higher pulsed Doppler energy is used. In con¬ trast to conventional Doppler, howev¬ er, the more accurate localization of the structure to be analyzed with CDI significantly reduces the exposure time. At the SPTA intensity levels used for our examinations, to our knowledge, no known adverse effects have been seen.23 Although at the in¬ tensity levels used in our study, no bioeffects have been found, the expo¬ sure time to higher levels of pulsed
Doppler during spectrum analysis should be kept to a minimum.
Lizzi et al34,8' studied the bioeffects of ultrasound in the eyes of rabbits. Due to their interest in therapeutic applica¬ tions of ultrasound, they looked at the effects of very high ultrasound ener¬ gies to the eye and they were able to
produce power
minimal choroidal lesions at
outputs of approximately 100
W/cm2. This is several orders of magni¬ tude more than the power output used in
our
study.
Our results suggest that CDI might have several promising applications in ophthalmology. Further studies are under way to investigate its usefulness in evaluating arteriolar and venous oc¬ clusive disease of the eye, vascularity patterns in intraocular and orbital tu-
mors, and orbital vascular malforma¬ tions. If quantitative assessment of blood-flow velocities and waveform analysis in the CRA, some of the pos¬ terior ciliary arteries, and the ophthal¬ mic artery prove to be accurate, sensi¬ tive, and sufficiently reproducible, this noninvasive and clinically easy-to-perform technique may give physiologic and pathophysiologic information in a number of ocular diseases, thereby suggesting the possibility of CDI being
used
as a
diagnostic tool in low-tension
and ischemie vascular disor¬ ders of the eye and orbit. This study was supported in part by the Ocular Oncology Fund and the Oncology Research Fund, Wills Eye Hospital; by the Eye Tumor Research Foundation, Inc, Gladwyne, Pa; and by the Deutsche Forschungsgemeinschaft, Bonn, Fed-
glaucoma
eral Republic of Germany (Dr Lieb). The authors do not have any commercial or proprietary interest in the instruments and devices used in this study. They do not receive
payments
as
consultants,
reviewers,
or
evaluators.
References 1. Coleman
DJ, Lizzi FL, Jack RL. Ultrasono-
graphy of the Eye and Orbit. Philadelphia, Pa: Lea & Febiger; 1977:42-47. 2. Ossoinig KC. The role of clinical echography in modern diagnosis of periorbital and orbital lesions. In: Bleeker G, ed. Proceedings of the Third International Symposium on Orbital Disorders. Amsterdam, the Netherlands: Dr W. Junk Publishers; 1978:496-537.
3. Rochels R. Ultraschalldiagnostik in der Augenheilkunde. In: Lehrbuch und Atlas. Munich, West Germany: Ecomed; 1986. 4. Guthoff R. Ultraschall in der Ophthalmologischen Diagnostik: ein Leitfaden f\l=u"\rdie Praxis. Stuttgart, West Germany: Enke, B\l=u"\chereides Au-
genarztes; 1988;116:106-147.
5. Marmion VJ. Strategies in Doppler ultrasound. Trans Ophthalmol Soc UK. 1986;105:562\x=req-\ 567. 6. Taylor KJW, Holland S. Doppler ultrasound, I: basic principles, instrumentation, and pitfalls.
Radiology. 1990;174:297-307.
7. Scoutt LM, Zawin ML, Taylor KJW. Doppler ultrasound, II: clinical applications. Radiology.
1990;174:309-319.
8. Duncan WJ. Color Doppler in Clinical Cardiology. Philadelphia, Pa: WB Saunders Co; 1988:1-5. 9. Middleton WD, Thorne DA, Melson GL. Color Doppler ultrasound of the normal testis. AJR Am J Roentgenol. 1989;152:293-297. 10. Middleton WD, Melson GL. Testicular ischemia: color Doppler sonographic findings in five patients. AJR Am J Roentgenol. 1989;152:1237\x=req-\ 1239. 11. Erickson SJ, Mewissen MW, Foley WD, et al. Stenosis of the internal carotid artery: assessment using color Doppler imaging compared with angiography. AJR Am J Roentgenol. 1989; 152:1299-1305. 12. Erickson SJ, Mewissen MW, Foley WD, et
al. Color Doppler evaluation of arterial stenoses and occlusions involving the neck and thoracic inlet.
Radiographics. 1989;9:389-406.
13. Mitchell DG, Merton D, Needleman L, et al. Neonatal brain: color Doppler imaging, I: technique and vascular anatomy. Radiology. 1988;167:303\x=req-\ 306. 14. Mitchell DG, Merton D, Desai H, et al. Neonatal brain: color Doppler imaging, II: altered flow patterns from extracorporeal membrane oxygenation. Radiology. 1988;167:307-310. 15. Mitchell DG, Merton D, Mirsky PJ, Needleman L. Circle of Willis in newborns of 53 healthy full-term infants. Radiology. 1989;172:201-205. 16. Canning CR, Restori M. Doppler velocity mapping of extra-ocular muscles: a preliminary re-
port. Eye. 1989;3:409-414.
17. Riva CE, Grunwald JE, Sinclair SH, Petrig BL. Blood velocity and volumetric flow rate in human retinal vessels. Invest Ophthalmol Vis Sci.
1985;26:1124-1132. 18. Guthoff R, Berger RW, Helmke K, Winkler B. Doppler sonographische Befunde bei intraokularen tumoren. Fortschr Ophthalmol. 1989;86:239\x=req-\
241. 19. Canning CR, Restori M. Doppler ultrasound studies of the ophthalmic artery. Eye. 1988;2:92-95. 20. Berger RW, Guthoff R, Helmke K, Winkler P, Draeger J. Doppler sonographische Befunde der arteria und vena centralis retinae. Fortschr Ophthalmol. 1989;86:334-336. 21. Lieb WE, Merton DA, Shields JA, Cohen SM, Mitchell DG, Goldberg BB. Color Doppler imaging in the demonstration of an orbital varix. Br J
Ophthalmol. 1990;74:305-308.
22. Erickson SJ, Hendrix LE, Massaro BM, et al. Color Doppler flow imaging of the normal and abnormal orbit. Radiology. 1989;173:511-516. 23. Lizzi FL, Mortimer AJ. Bioeffects considerations for the safety of diagnostic ultrasound. J
Ultrasound Med. 1988;7(suppl):1-38. 24. Hayreh SS, Dass R. The ophthalmic artery, II: intra-orbital course. Br J Ophthalmol. 1962; 46:165-185. 25. Hayreh SS. Arteries of the orbit in the human being. Br J Surg. 1963;50:938-953. 26. Nelson R, Pretorius DH. The Doppler signal: where does it come from and what does it mean? AJR Am J Roentgenol. 1988;151:439-447. 27. Schwaighofer B, H\l=u"\bschP, Barton P, Seidl G, Braunsteiner A, Kainberger F. Angiodynographie: Technische Grundlagen und klinische Anwendungen. Ultraschall. 1988;9:176-179. 28. Powis RL. Color flow imaging: understanding its science and technology. J Diagnos Med Sonogr. 1988;4:234-245. 29. Merritt CR. Doppler flow imaging. J Clin Ultrasound. 1987;15:591-597. 30. Grant EG, Tessler FN, Perrella RR. Clinical Doppler imaging. AJR Am J Roentgenol. 1989; 152:707-717. 31. Bezzi M, Mitchell DG, Needleman L, Goldberg, BB. Iatrogenic aneurysmal portal-hepatic venous fistula. J Ultrasound Med. 1988;7:457-461. 32. Wong WS, Tsuruda JS, Liberman RL, Chirino A, Vogt JF, Gangitano E. Color Doppler imaging of intracranial vessels in neonate. AJNR.
1989;10:425-430. 33. Feke GT, Tagawa H, Deupree DM, Goger DG, Sebag J, Weiter JJ. Blood flow in the normal human
retina.
Invest
Vis
Sci.
mental cataract production by high frequency ultrasound. Ann Ophthalmol. 1978;10:934-942. 35. Lizzi FL, Coleman DJ, Driller J, Fanzen LA, Leopold M. Effects of pulsed ultrasound on ocular tissue. Ultrasound Med Biol. 1981;7:245-252.
Currently in Other AMA Journals ARCHIVES OF GENERAL PSYCHIATRY The Cholinergic Rapid Eye Movement Induction Test With Arecoline in Depression J. Christian Gillin, MD; Laura Sutton, RN; Caroline Ruiz; John Kelsoe, MD; Renée M. Dupont, MD; Denis Darko, MD; S. Craig Risch, MD; Shahrokh Golshan, PhD; David Janowsky, MD (Arch Gen Psychiatry. 1991;48:264-270)
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
Ophthalmol
1989;30:58-65. 34. Lizzi FL, Packer AJ, Coleman DJ. Experi-
Doppler Imaging
Technique
of the
and Normal Vascular
Eye
and Orbit
Anatomy
Wolfgang E. Lieb, MD; Steven M. Cohen, MD; Daniel A. Merton; Jerry A. Shields, MD; Donald G. Mitchell, MD; Barry B. Goldberg, MD \s=b\ Color Doppler imaging is a recent advance in ultrasonography that allows simultaneous two-dimensional imaging of structure and blood flow to be performed. Doppler information is superimposed in color over a conventional grayscale ultrasound image. Using this technique, we examined 40 normal eyes. The central retinal artery, posterior ciliary arteries, ophthalmic artery, and central retinal vein could be located in all patients. Using the color image as a guide, Doppler spectral analysis allows quantitative assessment of blood flow velocity in these vessels. Color Doppler imaging is a new modality for the study of ocular and orbital hemodynamics.
(Arch Ophthalmol. 1991;109:527-531) "D eal time A-mode and B-mode ultrasonography has been used for the diagnostic evaluation of ophthalmic disorders since the early 1960s.M Mod¬ ern
digital high-resolution equipment
has improved diagnostic imaging and made it an essential part of ophthalmologic evaluations. Ultrasonography is routinely used today for display, dif¬ ferentiation, and measurement of in¬ traocular and orbital tumors and tu¬ morlike lesions, detection of retinal detachments in eyes with opaque me¬ dia, axial-length measurements for calAccepted for publication August 21, 1990. From the Oncology Service, Wills Eye Hospital (Drs Lieb and Shields), Philadelphia, Pa, and the Department of Radiology, Division of Ultrasound, Jefferson Medical College, Thomas Jefferson University, Philadelphia (Drs Cohen, Mitchell, and Goldberg and Mr Merton). Reprint requests to Oncology Service, Wills Eye Hospital, Ninth and Walnut Streets, Phila-
delphia, PA 19107 (Dr Shields).
culation of intraocular lens power, and evaluation of traumatic injuries to the globe and orbit. In addition, ultrasono¬ graphy is useful to detect and diagnose optic nerve lesions such as optic nerve sheath meningiomas, inflammatory flu¬ id collection in the nerve sheath, and orbital tumors or inflammatory con¬ ditions. See also pp 522, 532, 537, and 582.
Doppler ultrasound is a method of detecting changes in the frequency of sound reflected from flowing blood, allowing estimation of flow velocity. Conventional continuous-wave Dopp¬ ler ultrasonography of the carotid ar¬ teries and the periorbital vessels is frequently employed in patients with ischemie ocular disease, and has also been used to evaluate the ophthalmic artery.' The technology of Duplex scanning allows simultaneous B-mode imaging and Doppler evaluation to be performed. Since the lumina of the
vessels in the eye and orbit are too small to be imaged with conventional Duplex scans, Doppler spectra are ob¬ tained without precise localization and without knowledge of the Doppler an¬ gle. To facilitate localization and as¬ sessment of the Doppler angle, the two-dimensional flow information in color Doppler imaging (CDD is en¬ coded in color and superimposed on the gray-scale image of the structure.M Color Doppler imaging is currently being used in echocardiography,8 for evaluation of peripheral arterial and venous disease, for evaluation of the vascular pattern of the genitourinary system,910 and in the normal and patho-
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
state of abdominal organ perfu¬ sion. One major application is the ex¬ amination of the extracranial vessels supplying the central nervous system, ie, the carotid and vertebral arter¬ ies,1112 and the qualitative and quanti¬ tative analysis of neonatal intracranial vasculature.13"1' This modality provides the advantage of rapid orientation and provides an "anatomic" display by de¬ picting Doppler shifts throughout the entire image. There are only limited reports in the literature of applications of CDI in the study of the eye and orbit. Canning and Restori1" applied this technique to quantify rectus-muscle movement dur¬ ing voluntary saccades. Laser Doppler studies were performed for the retinal vessels by Riva et al" and duplexDoppler has also been used to demon¬ strate flow in intraocular tumors18 and to evaluate the central retinal artery (CRA), central retinal vein (CRV), and ophthalmic artery. 19,2° We have report¬ ed the value of CDI in the noninvasive diagnosis of an orbital varix21 and in the diagnosis of cavernous sinus fístu¬ las and intraocular tumors (W.E.L. and P. M. Flaharty, MD, unpublished data, 1990). Recently, Erickson et al22 described their experience using CDI in the normal orbit and in the evalua¬ tion of seven patients with pathologic conditions in the orbit. We herein outline our technique of using CDI to visualize the normal vas¬ cular system of the eye and orbit, and discuss our initial experience examin¬ ing 40 normal eyes, the preliminary results, and indications for future po¬ tential applications of this new imaging mode.
logic
PATIENTS AND METHODS
imaged to facilitate determination of Doppler flow angle for estimation of velocity. The scan images were digitally recorded on videotape and later reviewed with the benefit of cine-loop and frame-byframe analysis of selected segments. Im¬ ages were photographed during cine-loop replay of the recording by using a 35-mm camera that photographed directly from an were
Maximum Systolic Blood Flow Velocities in Orbital Vessels (N 40)
the
All examinations were performed with a color Doppler unit (QAD I, Quantum Medi¬ cal Systems Ine, Issaquah, Wash) using a 7.5-MHz linear phased array transducer. The estimated in situ peak temporal aver¬ age (SPTA) intensity at 4.2-cm depth in the color imaging mode (provided by Quantum Medical Systems) is 7 mW/cm for the 7.5MHz transducer. When spectrum analysis is performed, the SPTA intensity is approx¬ imately 71 mW/cm". In the spectrum analy¬ sis mode, the SPTA intensity exceeds the limits suggested by current Food and Drug Administration guidelines for ophthalmic application of 17 mW/cm2, but is lower than those given in the ultrasound safety guide¬ lines of the American Institute for Ultra¬ sound in Medicine. The study was ap¬ proved by the institutional review board of the Thomas Jefferson University Hospital (Philadelphia, Pa) and standardized in¬ formed consent was obtained from all sub¬ jects examined. Doppler spectral analysis lasted approximately 45 seconds per eye. The ultrasound transducer was applied to the closed eyelids using sterile ophthalmic
methylcellulose
as a coupling gel. During examination, the patient was in a supine position and care was taken not to apply
the
pressure to the eye to avoid artifacts. Hori¬ zontal and vertical scans through the eye and orbit were performed. Depending on the direction of flow with respect to the transducer, the blood flow data are dis¬ played in either red or blue. The colors can be arbitrarily assigned but, in this study, flow toward the transducer is depicted as red and away from the transducer as blue. The color saturation in the image repre¬ sents the average frequency (first moment average) from a spectral analysis performed at each sample site. These frequencies can be turned into velocities by solving the Doppler equation for velocity. When examining the eye and orbit through the eyelids, the ultrasound beam is essentially parallel to the orbital and ocular vessels; thus, most arterial flow is depicted in red. Arteries can usually be distin¬ guished from veins by noting their pulsatility. Pulsed Doppler spectral analysis also helped to distinguish between pulsatile ar¬ terial flow and the usually more continuous or minimally pulsatile venous flow, and allowed for the quantification of data. When the ultrasound beam was at an angle of 90° to a vascular structure or if a vessel con¬ tained only stagnant blood, no Doppler flow information was obtained and the structure was shown in gray-scale display only. All examinations were performed in a "low" or "medium" flow setting to allow for optimal detection of low to medium Doppler frequency shifts of the slow-flowing blood in the small orbital vessels. For the ophthal¬ mic artery, medium to high flow settings were applied since flow in this vessel is faster. Color threshold levels were adjusted to minimize artifacts by lid and involuntary eye movements. To obtain Doppler spectra, a sample volume of approximately 0.2 0.2 mm was chosen within the vessel. The proximal and distal portions of the vessel
isolated on-board color monitor. The dura¬ tion of the examination ranged from 10 to 20 minutes. Using the above-described technique, we examined 10 healthy volunteers on two separate occasions and the normal eyes of 30 patients with intraocular tumors in their other eyes. Our study population was as follows: 18 men and 12 women; mean age (±SD), 45.34 ±16.3 years (range, 20 to 76 years). None of the patients or the volun¬ teers had a history of ocular or systemic vascular disease or any clinically evident vascular anomaly in the examined eye and orbit. RESULTS
A horizontal scan through the globe and orbit at the level of the optic nerve allows the depiction of the CRA, and CRV (Fig 1). The CRA and accompa¬ nying CRV can be seen within the anterior 2 mm of the optic nerve shad¬ ow. The CRA can be traced up to where it enters the optic nerve approx¬ imately 13 mm behind the optic disc
(Fig 2).
Differentiation of the CRV is quite easy, not only through color coding as a blue structure but also by its Doppler spectral characteristics, ie, its continu¬ ous flow throughout systole and diasto¬ le. Color information alone can some¬ times be misleading since vessel loops may simulate flow away from the probe and thus be coded as blue. In this case, the pulsatility of the arteriole clearly distinguishes it from the usually nonpulsatile vein. On either side of the optic nerve slightly posteri¬ or to the CRA, the short and long posterior ciliary arteries can be identi¬ fied (Fig 1). These vessels and the CRA have pulsatility similar to the
ophthalmic artery, differing mainly by
the maximum of the systolic velocity (Fig 3). In some sections continuous with the central retinal vessels, the flow in the retinal arteriolar vessels could be demonstrated; however, sepa¬ rate branches of the retinal arteriolar system could not be resolved. At the level of the retinal vessels, the venous flow is so low that it usually cannot be detected. However, we were able to depict some of the vortex veins in several cases. The blood flow in the choroid and retina could not always be distinguished unless a retinal detach¬ ment was present, allowing the retinal vessels to be identified in the normally
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
=
Mean ± SD Blood
cm/s Central retinal
artery
Flow, (Range)
10.3 ± 2.1
(6.4-17.2)
Central retinal vein
-2.9 ± 0.73
Ophthalmic
31.4 ± 4.2
artery Posterior
artery
ciliary
(1.9-5.4) (23.5-39.8) 12.4 ± 4.8
(1.4-22.7)
echo-free vitreous cavity. The Doppler spectrum of the posterior ciliary arter¬ ies showed velocity-time spectra very similar to those of the CRA. The end diastolic flow in the posterior ciliary
Fig 1.—A horizontal scan through the globe shows the central retinal artery (CRA) and accompanying vein within the anterior portion of the optic nerve shadow. In the same plane on both sides of the optic nerve (N), Doppler shifts within the short posterior ciliary arteries (PCA) can be demonstrated. Fig 2.—The central retinal artery (CRA) can be traced posteriorly where it enters the optic nerve (N) approximately 13 mm behind the disc (arrow). Fig 3.—Doppler spectrum (velocity-time curve) of the central retinal artery (CRA). The Doppler angle "A" can be corrected by align¬ ment of a cursor in the vessel. The angle correction allows assessment of the velocity
of blood flow in these vessels. The central retinal vein (CRV) runs parallel to the CRA in this section.
Fig 4.-The ophthalmic artery (OPHT ART) be traced deeper into the orbit and can be localized in the majority of patients slightly lateral to the optic nerve (N) as it enters the orbit. Further anteriorly it crosses over the optic nerve to divide into its nasal branches. Fig 5.—Spectrum analysis of the ophthalmic artery (OA) in a medium flow setting. Note the high, steep maximum systolic peak. In the two-dimensional image the portions of the ophthalmic artery nasal to the optic nerve (N) coming toward the transducer could be easily identified. The horizontal portion between the two segments of the ophthalmic artery was not imaged. Since in the horizontal segment the Doppler angle is 90°, no Doppler shift was can
seen.
Fig 6. —Superior ophthalmic vein (SOV) with flow away from the transducer and therefore encoded in blue. In addition to these big ves¬ sels, pulsatile arterial and continuous venous flow inside the eye was seen, most likely rep¬ resenting venous flow in the vortex veins draining into the superior ophthalmic vein. Fig 7. —Spectrum analysis of the superior ophthalmic vein demonstrates the character¬ istic continuous, nonpulsatile venous flow.
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
arteries was, however, higher, reflect¬ the low-resistance vascular chan¬ nels of the choroid. Deeper into the orbit, sections of the main ophthalmic artery could be seen in cross section. The ophthalmic artery was traced temporally to the nerve posteriorly, then traveling anteriorly and nasally where several branches may be depicted (Fig 4). Hayreh and Dass21 and Hayreh2'' have shown in a large series of cadaver dissections that there is considerable variability in the course of the ophthalmic artery and its branches. In their series, the crossing of the ophthalmic artery over the optic nerve in the midorbit was a common finding (80% of cases).242'' The crossing of the ophthalmic artery over the optic nerve could be demonstrated by CDI in all our subjects. The flow-velocity waveform is similar to that of the internal carotid artery, showing a steep maximum systolic peak, often a dicrotic notch, and low diastolic flow velocity (Fig 5). The supraorbital ar¬ tery could be identified when angling the ultrasound transducer more supe¬ riorly and anteriorly. In the superior nasal orbit, we identified the superior ophthalmic vein (Fig 6) fed by the vortex veins and running posteriorly, crossing the optic nerve. The Doppler spectrum with the continuous, nonpul¬ satile flow pattern is characteristic of venous flow (Fig 7). Flow velocity measurements of the CRA (Fig 3) and posterior ciliary ar¬ teries were obtained as a baseline for further evaluations in patients with ocular and orbital vascular disease. In the 10 healthy young individuals, the peak systolic velocities were measured in the CRAs and ophthalmic artery in both eyes after angle correction on two different occasions. The CRA was identified within the anterior optic nerve portion slightly posterior to it entering the globe. The ophthalmic artery was identified during its course parallel to the optic nerve. In both vessels a preset sample volume was placed after the Doppler angle was estimated using the angle-correction menu of the unit and placing a cursor along the course of the vessel of inter¬ est. Mean peak systolic velocities in the CRA ranged from 6.4 cm/s to 17.2 cm/s, with a mean (±SD) of 10.3±2.1 cm/s. The mean ( ± SD) peak systolic velocity in the ophthalmic artery was 31.4 ±4.2 cm/s and in the posterior ciliary arteries was 12.4 ±4.8 cm/s (Table). No significant difference was noted between right and left eyes or between male and female subjects. In the group of normal subjects with paired observations, on two occasions
ing
paired t test was performed for all variables and no statistically signifi¬ cant difference was seen between the first and second measurements. a
COMMENT
Color Doppler imaging combines both anatomic and velocity data and displays the information in a format that can be easily comprehended. Blood-flow velocity information can be obtained quantitatively within the en¬ tire two-dimensional gray-scale image. This has revolutionized noninvasive vascular imaging techniques, and in many areas of medicine, the role of this new technique is well established.2''"3 Color Doppler imaging is being used in other medical specialties to examine the heart," the large abdominal and thoracic vessels,81 the vasculature in the male genitalia,1'1" peripheral limb vessels, and the extracranial and intracranial vasculature.1112 Recently, the importance of this examination for ana¬ tomic delineation and functional disor¬ ders in the cerebrovascular system in newborns has been stressed.131'12 A recent report demonstrated CDIs of two presumed orbital varices, one su¬ perior ophthalmic vein thrombosis, and an orbital arteriovenous malforma¬ tion.23 The authors concentrated on the topographic display of the vascular structures in 26 normal orbits and demonstrated CRA, CRV, the superi¬ or ophthalmic vein, and ophthalmic
artery.
In the present study, CDI was able to localize small vessels in the eye, the orbit, and the optic nerve (Fig 1), expanding on the recent observations
by Erickson and associates.22 Using low-flow settings and appropriate
we were able to detect flow in the retinal vessels as well. Although this technique cannot distinguish the separate portions of the arteriolar bed, we could see what represents a sum¬ mation of these vessels. It can be clearly distinguished from flow in the posterior ciliary arteries and the large vascular channels of the choroid that they supply. Differentiation between the short and long posterior ciliary arteries and distinction among them has so far not been possible in our hands. Color Doppler imaging allows, for the first time, reproducible noninva¬ sive imaging and selective Doppler in¬ formation of orbital vessels. At pre¬ sent, it is not likely that CDI will allow blood volume measurements, since di¬ ameters of the vessels cannot be accu¬ rately assessed. The analysis of selec¬ tively obtained Doppler spectra from the different vessels, ie, CRA, CRV,
threshold,
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
and the posterior ciliary arteries and branches of the ophthalmic artery, might allow some insight into the hemodynamics of the eye and its supply¬ ing vessels. We demonstrated that this technique allows accurate measure¬ ment of flow velocities in the CRA and ophthalmic artery. Our values corre¬ spond to the values found by Berger et
al20 using duplex scanning. They are higher than the flow velocities found by Riva et al17 in the branch retinal arteries and veins using laser Doppler velocimetry. However, our findings for the maximum systolic flow velocity
in the CRA are similar to the results of Feke et al,33 who also used laser Dopp¬ ler velocimetry to assess blood flow in the retinal arterioles. In addition to measuring the velocities, the analysis of the waveforms provides information about the vessel and the vascular sys¬ tem proximal and distal to it. Com¬ pared with the CRA, the spectrum of the posterior ciliary vessels showed a slightly higher systolic velocity and also a higher end diastolic velocity indicating a low-resistance vascular system distal to the sample area. Inaccuracies in our quantitative measurements can arise from errors in interpreting our Doppler-shifted fre¬ quency spectrum and in measuring the Doppler angle. Feke et al33 have re¬ ported their estimate of uncertainty associated with their individual mea¬ surements of the maximum systolic velocity to be about 18% using laser velocimetry. This is the range of error seen in most studies reporting Doppler flow data. As in any new imaging technique, the safety of pulsed Doppler ultra¬ sound is of concern. Color Doppler imaging involves only slightly higher ultrasound exposure than conventional B-mode examination. However, when spectral analysis is performed, higher pulsed Doppler energy is used. In con¬ trast to conventional Doppler, howev¬ er, the more accurate localization of the structure to be analyzed with CDI significantly reduces the exposure time. At the SPTA intensity levels used for our examinations, to our knowledge, no known adverse effects have been seen.23 Although at the in¬ tensity levels used in our study, no bioeffects have been found, the expo¬ sure time to higher levels of pulsed
Doppler during spectrum analysis should be kept to a minimum.
Lizzi et al34,8' studied the bioeffects of ultrasound in the eyes of rabbits. Due to their interest in therapeutic applica¬ tions of ultrasound, they looked at the effects of very high ultrasound ener¬ gies to the eye and they were able to
produce power
minimal choroidal lesions at
outputs of approximately 100
W/cm2. This is several orders of magni¬ tude more than the power output used in
our
study.
Our results suggest that CDI might have several promising applications in ophthalmology. Further studies are under way to investigate its usefulness in evaluating arteriolar and venous oc¬ clusive disease of the eye, vascularity patterns in intraocular and orbital tu-
mors, and orbital vascular malforma¬ tions. If quantitative assessment of blood-flow velocities and waveform analysis in the CRA, some of the pos¬ terior ciliary arteries, and the ophthal¬ mic artery prove to be accurate, sensi¬ tive, and sufficiently reproducible, this noninvasive and clinically easy-to-perform technique may give physiologic and pathophysiologic information in a number of ocular diseases, thereby suggesting the possibility of CDI being
used
as a
diagnostic tool in low-tension
and ischemie vascular disor¬ ders of the eye and orbit. This study was supported in part by the Ocular Oncology Fund and the Oncology Research Fund, Wills Eye Hospital; by the Eye Tumor Research Foundation, Inc, Gladwyne, Pa; and by the Deutsche Forschungsgemeinschaft, Bonn, Fed-
glaucoma
eral Republic of Germany (Dr Lieb). The authors do not have any commercial or proprietary interest in the instruments and devices used in this study. They do not receive
payments
as
consultants,
reviewers,
or
evaluators.
References 1. Coleman
DJ, Lizzi FL, Jack RL. Ultrasono-
graphy of the Eye and Orbit. Philadelphia, Pa: Lea & Febiger; 1977:42-47. 2. Ossoinig KC. The role of clinical echography in modern diagnosis of periorbital and orbital lesions. In: Bleeker G, ed. Proceedings of the Third International Symposium on Orbital Disorders. Amsterdam, the Netherlands: Dr W. Junk Publishers; 1978:496-537.
3. Rochels R. Ultraschalldiagnostik in der Augenheilkunde. In: Lehrbuch und Atlas. Munich, West Germany: Ecomed; 1986. 4. Guthoff R. Ultraschall in der Ophthalmologischen Diagnostik: ein Leitfaden f\l=u"\rdie Praxis. Stuttgart, West Germany: Enke, B\l=u"\chereides Au-
genarztes; 1988;116:106-147.
5. Marmion VJ. Strategies in Doppler ultrasound. Trans Ophthalmol Soc UK. 1986;105:562\x=req-\ 567. 6. Taylor KJW, Holland S. Doppler ultrasound, I: basic principles, instrumentation, and pitfalls.
Radiology. 1990;174:297-307.
7. Scoutt LM, Zawin ML, Taylor KJW. Doppler ultrasound, II: clinical applications. Radiology.
1990;174:309-319.
8. Duncan WJ. Color Doppler in Clinical Cardiology. Philadelphia, Pa: WB Saunders Co; 1988:1-5. 9. Middleton WD, Thorne DA, Melson GL. Color Doppler ultrasound of the normal testis. AJR Am J Roentgenol. 1989;152:293-297. 10. Middleton WD, Melson GL. Testicular ischemia: color Doppler sonographic findings in five patients. AJR Am J Roentgenol. 1989;152:1237\x=req-\ 1239. 11. Erickson SJ, Mewissen MW, Foley WD, et al. Stenosis of the internal carotid artery: assessment using color Doppler imaging compared with angiography. AJR Am J Roentgenol. 1989; 152:1299-1305. 12. Erickson SJ, Mewissen MW, Foley WD, et
al. Color Doppler evaluation of arterial stenoses and occlusions involving the neck and thoracic inlet.
Radiographics. 1989;9:389-406.
13. Mitchell DG, Merton D, Needleman L, et al. Neonatal brain: color Doppler imaging, I: technique and vascular anatomy. Radiology. 1988;167:303\x=req-\ 306. 14. Mitchell DG, Merton D, Desai H, et al. Neonatal brain: color Doppler imaging, II: altered flow patterns from extracorporeal membrane oxygenation. Radiology. 1988;167:307-310. 15. Mitchell DG, Merton D, Mirsky PJ, Needleman L. Circle of Willis in newborns of 53 healthy full-term infants. Radiology. 1989;172:201-205. 16. Canning CR, Restori M. Doppler velocity mapping of extra-ocular muscles: a preliminary re-
port. Eye. 1989;3:409-414.
17. Riva CE, Grunwald JE, Sinclair SH, Petrig BL. Blood velocity and volumetric flow rate in human retinal vessels. Invest Ophthalmol Vis Sci.
1985;26:1124-1132. 18. Guthoff R, Berger RW, Helmke K, Winkler B. Doppler sonographische Befunde bei intraokularen tumoren. Fortschr Ophthalmol. 1989;86:239\x=req-\
241. 19. Canning CR, Restori M. Doppler ultrasound studies of the ophthalmic artery. Eye. 1988;2:92-95. 20. Berger RW, Guthoff R, Helmke K, Winkler P, Draeger J. Doppler sonographische Befunde der arteria und vena centralis retinae. Fortschr Ophthalmol. 1989;86:334-336. 21. Lieb WE, Merton DA, Shields JA, Cohen SM, Mitchell DG, Goldberg BB. Color Doppler imaging in the demonstration of an orbital varix. Br J
Ophthalmol. 1990;74:305-308.
22. Erickson SJ, Hendrix LE, Massaro BM, et al. Color Doppler flow imaging of the normal and abnormal orbit. Radiology. 1989;173:511-516. 23. Lizzi FL, Mortimer AJ. Bioeffects considerations for the safety of diagnostic ultrasound. J
Ultrasound Med. 1988;7(suppl):1-38. 24. Hayreh SS, Dass R. The ophthalmic artery, II: intra-orbital course. Br J Ophthalmol. 1962; 46:165-185. 25. Hayreh SS. Arteries of the orbit in the human being. Br J Surg. 1963;50:938-953. 26. Nelson R, Pretorius DH. The Doppler signal: where does it come from and what does it mean? AJR Am J Roentgenol. 1988;151:439-447. 27. Schwaighofer B, H\l=u"\bschP, Barton P, Seidl G, Braunsteiner A, Kainberger F. Angiodynographie: Technische Grundlagen und klinische Anwendungen. Ultraschall. 1988;9:176-179. 28. Powis RL. Color flow imaging: understanding its science and technology. J Diagnos Med Sonogr. 1988;4:234-245. 29. Merritt CR. Doppler flow imaging. J Clin Ultrasound. 1987;15:591-597. 30. Grant EG, Tessler FN, Perrella RR. Clinical Doppler imaging. AJR Am J Roentgenol. 1989; 152:707-717. 31. Bezzi M, Mitchell DG, Needleman L, Goldberg, BB. Iatrogenic aneurysmal portal-hepatic venous fistula. J Ultrasound Med. 1988;7:457-461. 32. Wong WS, Tsuruda JS, Liberman RL, Chirino A, Vogt JF, Gangitano E. Color Doppler imaging of intracranial vessels in neonate. AJNR.
1989;10:425-430. 33. Feke GT, Tagawa H, Deupree DM, Goger DG, Sebag J, Weiter JJ. Blood flow in the normal human
retina.
Invest
Vis
Sci.
mental cataract production by high frequency ultrasound. Ann Ophthalmol. 1978;10:934-942. 35. Lizzi FL, Coleman DJ, Driller J, Fanzen LA, Leopold M. Effects of pulsed ultrasound on ocular tissue. Ultrasound Med Biol. 1981;7:245-252.
Currently in Other AMA Journals ARCHIVES OF GENERAL PSYCHIATRY The Cholinergic Rapid Eye Movement Induction Test With Arecoline in Depression J. Christian Gillin, MD; Laura Sutton, RN; Caroline Ruiz; John Kelsoe, MD; Renée M. Dupont, MD; Denis Darko, MD; S. Craig Risch, MD; Shahrokh Golshan, PhD; David Janowsky, MD (Arch Gen Psychiatry. 1991;48:264-270)
Downloaded From: http://archopht.jamanetwork.com/ by a New York University User on 05/22/2015
Ophthalmol
1989;30:58-65. 34. Lizzi FL, Packer AJ, Coleman DJ. Experi-
