Article

PDF
Access to the PDF text
Service d'aide à la décision clinique
Advertising


Free Article !

Journal Français d'Ophtalmologie
Vol 23, N° 8  - octobre 2000
p. 756
Doi : JFO-10-2000-23-8-0181-5512-101019-ART3
Articles originaux

An ICG angiogram-based clinical method for characterizing the choroidal circulation used to assess the hemorrheologic effects of pentoxifylline
 
ARTICLES SCIENTIFIQUES

ARTICLES ORIGINAUX

Journal Français d'Ophtalmologie2000; 23: 756-762
© Masson, Paris, 2000

R.W. Flower (1), J.I. Lim (2)

(1)Department of Ophthalmology, University of Maryland School of Medicine (Baltimore, Maryland).
(2)Department of Ophthalmology, University of Southern California School of Medicine (Los Angeles, California).

J. Fr. Ophtalmol., 2000 23, 8 : 756-762.

RÉSUMÉ

Évaluation de la circulation choroïdienne par angiographie au vert d'indocyanine et des effets hémorréologiques de la Pentoxifylline

R.W. Flower, J.I. Lim

Introduction : L'angiographie au vert d'indocyanine, méthode utilisée en clinique, permet d'étudier le flux sanguin choroïdien, de détecter les modifications de la circulation choroïdienne et ainsi d'évaluer l'effet de thérapeutiques visant à modifier le flux sanguin oculaire.

Une acquisition rapide des images est nécessaire en raison de la grande vitesse circulatoire au niveau de la choroïde. Ce progrès est possible grâce à la mise au point d'ophtalmoscopes à balayage laser, qui permettent d'acquérir des images à une vitesse de 20 images par seconde composées de 256 × 256 pixels.

L'évaluation du flux choroïdien est d'autant plus importante et intéressante dans la mesure où l'on tente actuellement de localiser des vaisseaux nourriciers des vaisseaux choroïdiens néoformés et de les différencier des vaisseaux choroïdiens normaux. Enfin, la Pentoxifylline a été identifiée dans plusieurs études comme entraînant une modification du flux oculaire chez l'homme et l'évaluation de son effet thérapeutique peut servir de test sur l'intérêt de ces mesures cliniques du flux choroïdien.

Matériel et méthode : L'angiographie en ICG a été effectuée sous anesthésie générale chez des primates après dilatation avec un rétinographe de Zeiss modifié comportant un laser diode dont l'émission était à 805 nm. L'acquisition de 15 images/seconde a été digitalisée. Afin d'obtenir des images toujours du même champ, le réticule de la caméra a été aligné avec le centre de la fovéa de l'animal. La séquence d'enregistrement de l'ordre de 2 secondes a été répétée à deux reprises. L'examen angiographique à l'ICG était réalisé par une injection intraveineuse de 0,1 ml de 25 ml/ml de colorant suivie d'un flash de sérum salé. Après son obtention, une solution de Pentoxifylline de 0,1 ml à 25 mg/ml dont la concentration était confirmée par une chromatographie gazeuse a été injectée. Des examens angiographiques ont été obtenus à 5,7, 20 et 40 minutes après l'injection.

L'étude a été effectuée 21 fois chez 3 singes adultes (macaques mulata). Ces expériences ont été répétées six fois sur une période de dix mois pour l'un des animaux et quatre fois sur une période d'un mois pour les deux autres animaux. Lors des examens de contrôles, les animaux ont reçu un volume identique de sérum salé.

L'analyse de l'angiogramme a reposé sur le principe de la soustraction qui met en évidence uniquement la différence entre les fluorescences à 1/15 de seconde d'intervalle.

La représentation de la distribution du vert d'indocyanine dans la choriocapillaire au cours d'un cycle cardiaque est infiniment plus évidente lorsqu'on représente en 3 dimensions l'intensité de la brillance de chaque point comme une hauteur au-dessus du plan du fond d'oeil. De façon conventionnelle, la représentation en 3 dimensions du remplissage de la choriocapillaire au pôle postérieur juste avant le pic systolique du pouls intraoculaire correspond à la distribution instantanée relative du flux sanguin choroïdien et permet une comparaison entre les différents angiogrammes.

Résultats : Les aspects initiaux de la cartographie en 3 dimensions pour chaque animal sont reproductibles pour un même oeil. La comparaison des cartographies en 3 dimensions a constamment mis en évidence une augmentation du flux sanguin choroïdien rétromaculaire dans les 5 à 10 minutes après l'injection de Pentoxifylline, avec un retour progressif au niveau initial en 20 à 40 minutes. L'injection de mêmes volumes de sérum salé n'a pas entraîné de modification du flux sanguin choroïdien.

Discussion : La circulation choroïdienne peut être représentée par une surface en 3 dimensions tout en conservant la distribution spatiale de la dynamique sanguine dans toute l'aire d'observation du fond d'oeil. La distribution du flux sanguin dans la choriocapillaire est relativement stable au cours du temps mais varie en fonction des cycles cardiaques dans ce plexus vasculaire fin. La perfusion de la choriocapillaire est unique et différente dans chaque oeil, tout comme son angio-architecture.

L'augmentation du flux choroïdien survenait malgré une fréquence cardiaque et une pression sanguine stable. De plus, cette modification du flux sanguin choroïdien n'était pas due à une augmentation du volume intravasculaire puisque l'on ne la retrouvait pas après injection de sérum salé. De même, dans la mesure où l'anesthésie a été réalisée par trois techniques différentes, son éventuelle influence était éliminée. La méthode qualitative d'évaluation utilisée dans cette étude démontre des modifications certaines de la circulation choroïdienne au pôle postérieur après injection intraveineuse de Pentoxifylline. La Pentoxifylline provoque, du moins transitoirement, une augmentation de la circulation choroïdienne sous-maculaire. Le bénéfice thérapeutique de cette médication doit être déterminé par quantification.

Mots clés : ICG. , angiographie. , Pentoxifylline. , analyses.

SUMMARY

An ICG angiogram-based clinical method for characterizing the choroidal circulation used to assess the hemorrheologic effects of pentoxifylline.

Purpose: To demonstrate an indocyanine green (ICG) angiography-based clinical method for characterizing choroidal blood flow and for detecting changes in choroidal circulation patterns, and by use of that method, to demonstrate that pentoxifylline affects choroidal blood flow.

Methods: High-speed ICG angiography was performed in rhesus monkeys before and after intravenous administration of pentoxifylline or saline (which served as a control) while monitoring blood pressure and heart rate. From these data, three-dimensional surface maps indicating the instantaneous relative distribution of choroidal blood flow during the peak of intra-ocular pressure pulse systole in a 30° field, centered on the macula, were generated to characterize the state of the choroidal circulation at various times during the experiments.

Results: Comparisons of the 3-dimentional surface maps consistently indicated an increase in sub-macular choroidal blood flow occurring within 5 to 10 minutes post-pentoxifylline injection, with a gradual return to baseline level 20-40 minutes later. Injection of equal volumes of saline produced no changes in choroidal blood flow.

Conclusions: Posterior-pole choroidal blood flow can be characterized as by a three-dimensional surface representing the instantaneous relative distribution of choroidal blood flow during the peak of intra-ocular pressure pulse systole. Pentoxifylline does, at least transiently, increase sub-macular choroidal blood flow.

Key words : ICG. , angiography. , pentoxifylline. , analysis.


INTRODUCTION

When ICG fluorescence angiography first was introduced as a method for routine visualization of choroidal blood circulation, emphasis was placed on high image acquisition rates because of the well-recognized high speed of choroidal blood flow [1]. The first widely distributed commercial systems for ICG angiography, however, sacrificed image temporal resolution to achieve high image spatial resolution; fundus images composed of 1024 × 1024 pixels were obtained, but only at a rate of about one per second. This meant that virtually all information obtained from choroidal angiography related to the venous and late phases of dye circulation; only occasional images of the arterial and choriocapillaris filling phases were recorded. Under those conditions, the first generally recognized clinical application of ICG angiography to be developed was ICG-guided photocoagulation of occult AMD-related CNV [2], a method that depended on CNV staining that is visible in very late phase angiogram images.

It was not until many years later that commercial scanning laser ophthalmoscope (SLO) systems began to appear in ophthalmology clinics, making possible wide spread routine visualization of the entire transit phase of ICG angiography. SLOs typically acquire images at the rate of 20 per second - albeit with images composed of only 256 × 256 pixels. Ability to acquire high temporal resolution images of the choroidalcirculation did not lead immediately to additional generally accepted clinical applications of ICG angiography, but it did lead eventually to revisiting the possibility of identifying and photocoagulating the feeder vessels supplying sub- and juxta-foveal AMD-associated CNV [3], [4]. The preliminary results achieved by this treatment approach are promising, but a key element of it, ability to locate feeder vessels, depends on high-speed acquisition of angiogram images. This is especially true in the case of occult CNV, wherein it appears that reduction of CNV blood flow may be accomplished by modulating blood flow through the underlying choriocapillaris [5]. The exciting potential of such a novel approach to treatment of exudative AMD notwithstanding, the act of examining transit phase angiogram image sequences to locate feeder vessels underscores the complexity of choroidal blood flow patterns.

It is intuitive that if the role of the choroid in various diseases of the eye is to be better understood, a fundamental step is development of a way to adequately characterize choroidal blood flow, so that abnormal circulation patterns can be differentiated from normal ones. Moreover, given that the choriocapillaris is a true vascular plexus, the complex choroidal angioarchitecture makes it entirely possible that the changes in blood flow patterns producing clinically significant results may be subtle ones, most likely associated with the choriocapillaris.

This study explores a method for characterizing the choroidal circulation in a way that permits detection of subtle changes in choroidal blood flow. The method is based on an earlier studies that demonstrated the choroidal circulation can be represented by an unique three-dimensional surface, while conserving the spatial distribution of blood flow dynamics across the entire observed fundus area [6], and that showed choriocapillaris blood flow patterns are relatively stable over extended periods [7]. Utility of the method is demonstrated by showing changes in choroidal blood flow induced by intravenous injection of pentoxifylline. Pentoxifylline was used since investigators already have demonstrated alteration of human ocular blood by treatment with pentoxifylline. In diabetic patients, Ferrari and coworkers showed improvement in retinopathy and nephropathy, as well as improvement in the macro-vasculature, in a four-year clinical trial with pentoxifylline [8]. Others have shown choroidal blood flow improvement in nonproliferative diabetic retinopathy patients following pentoxifylline treatment [9]. (Sebag et al.)

METHODS

Characterization of the choroidal circulation

The choriocapillaris is a thin, planer vascular plexus, through which the circulation of blood is determined by a network of perfusion pressure gradients established amongst the interspersed arterial and venous vessels connected to its posterior aspect. Since the magnitudes of those perfusion pressure gradients vary cyclically during each heartbeat, the instantaneous distribution of blood flow through the choriocapillaris, likewise, varies cyclically. Moreover, the pattern of the network of perfusion pressure gradients, as well as its cyclical variance, is as uniquely different for each eye as are the details of each eye's choroidal angioarchitecture. The possibility to systematically characterize the choroidal circulation lies in the observation that for each eye, the cyclically varying pattern of choriocapillaris blood flow distribution has been shown to be consistent from heartbeat to heartbeat over protracted time periods.

Sequences of images showing the changes in instantaneous distribution of ICG dye in the choriocapillaris that occur during a single heartbeat have been obtained by sequential subtraction of sequences of high-speed ICG angiogram images. However, these changes become much more apparent when such image data are rendered as sequences of 3-dimensional surface mappings in which the brightness of each point on the original image is proportional to the height of the surface above the plane of the observed fundus area. The rendering of a 2-dimensional image into a 3-dimensional surface mapping is demonstrated schematically in figure 1.

figure 2B shows a sequence of the 3-dimensional surface mappings that demonstrates the changes that occur in the instantaneous ICG dye distribution in the choriocapillaris at various stages of a single intraocular pressure pulse pulse. The family of intraocular pressure pulse curves in , figure 2A was derived by calculating the first time-derivative of curves representing the time-varying brightness of a cluster of point locations on the observed fundus area, as previously reported. Since the same sequence of 3-dimentional surface mappings occurs for each heartbeat, in theory any one of them could be selected to characterize the choroidal blood flow of the particular eye from which they were derived. However, as a matter of convention and because it always appears to have the most distinctive shape, the 3-dimensional mapping that occurs just prior to the peak of systole of the intraocular pressure pulse is chosen to characterize the choroidal circulation of a given eye. This convention ensures that all data bear the same phase relationship with the cardiac cycle, making comparisons from different angiogram studies possible and consistent.

figure 3 shows the characteristic 3-dimensional surfaces from 21 different eyes. As mentioned above, because of the variability in choroidal angioarchitecture from eye to eye, each is unique to the particular eye from whose high-speed angiogram data it was derived. In this figure, the characteristic surfaces are arranged in four arbitrary groups approximately according to major topographical similarities.

Thus, the characteristic 3-dimensional surface for any eye can be constructed directly from its high-speed digital ICG angiogram, obviously it is not necessary to go through the re-rendering steps indicated in figure 1. Comparisons of the characteristic surfaces derived under conditions can then be used to determine if significant changes have occurred in the choroidal circulation as a result.

Preparation of experimental subjects

Three healthy adult rhesus (macaque mulatta) monkeys (a 2-year-old female, an 8-year-old male, and a 4-year-old female) were the subjects for this study. After fasting for at least 6 hours, one monkey was sedated with ketamine (15mg/kg intramuscular), intubated, and maintained anesthetized with halothane. The other 2 monkeys were sedated and anesthetized with ketamine (100mg) or telazol (30-50mg) and maintained anesthetized with pentobarbital (32.5mg as needed). An intravenous catheter was inserted into a peripheral vein and connected to a stopcock injection port for delivery of drugs.

Each monkey's eyes were dilated with 1% tropicamide and 2.5% phenylephrine hydrochloride, and the retinas were examined with by indirect ophthalmoscopy to ensure no abnormalities were present. Fifteen minutes later, one after another, each monkey was positioned prone in front of a modified Zeiss fundus camera for ICG angiography, its head was stabilized with a chin rest, and a lid speculum was inserted into one eye.

ICG angiography

ICG angiography was performed using a modified Zeiss fundus camera. The fundus camera's usual xenon flash tube light source was replaced by an 805nm wavelength laser diode, coupled to the fundus camera's illumination optics via a small integrating sphere whose exit port was located at the position normally occupied by the flash tube arc. The fundus camera's photographic film camera was replaced with an infrared-sensitive vidicon tube (Model 4532URI Ultracon, Burle Industries), in front of which an 807nm wavelength cut-on filter, used to exclude the excitation laser light while admitting ICG dye fluorescence light. Images were digitally recorded at a rate of 15 per second and later analyzed using custom software.

To assure that the same fundus field of view was obtained for each angiogram, the crosshairs of the fundus camera eyepiece reticule were aligned with the monkey's foveal center. (This alignment was repeated prior to each image series and had to be maintained only during image acquisition, which lasted less than 2 seconds.) Each angiographic sequence included in the data set was repeated at least twice to assure consistency of the observed blood flow information.

Experimental protocol

Blood pressure and heart rate monitoring at three-minute intervals was begun and continued throughout each experimental session using an automated blood pressure monitoring device (Omega 1400 RP-110 non-invasive blood pressure monitor, In Vivo Research Lab Incorporated, Tulsa, Oklahoma). Baseline ICG angiograms were acquired at the start of each experiment. Following intravenous injection of 0.1ml of 25-mg/ml ICG dye, followed immediately by a 1.0-ml saline flush, transit of the dye through the choroid was recorded in thirty-two consecutive video angiographic images. After acquiring the baseline ICG angiographic sequence, 1.0ml of a 25-mg/ml solution of pentoxifylline (Trental, Hoescht-Roussel Pharmaceuticals, Inc., Somerville, New Jersey) was injected intravenously as a bolus, again followed by a 1.0-ml saline flush. (Pentoxifylline powder was mixed with sterile water to a concentration of 25-mg/ml one-hour prior to use. Representative samples of the pentoxifylline solutions were maintained on dry ice until concentration was confirmed by the capillary gas chromatography method of Burrows [10]).

Additional pairs of angiograms were acquired five to seven minutes later, and at 20 to 40 minutes post-drug injection. Immediately thereafter, identical sets of angiograms were obtained at identical time intervals, but following injection of 1.0-ml saline instead of pentoxifylline solution. (The order of pentoxifylline and saline solution administration was alternated from experiment to experiment in order to avoid a false positive result due to volumetric changes alone.)

All experimental data were obtained in a series of six such experimental sessions conducted over a period of ten months for the first monkey (the 2-year-old) and in four sessions each over a 1-month period for the second and third monkeys. At least 2 days elapsed between any consecutive sessions on the same animal to allow for recovery.

(This protocol complied with The Association for Research in Vision and Ophthalmology guidelines and policies for the care and use of laboratory animals, National Institutes of Health.)

Angiogram analysis

Each digital angiogram sequence of 32 images was analyzed, as described above, to generate a 3-dimentional surface characterizing choriocapillaris blood flow at the posterior pole, centered on the fovea. Since the primary data manipulation is subtraction of one angiogram image from the preceding one to yield the difference in fluorescence that occurred in a 1/15-second-time interval, the technique is independent of the degree of background fluorescence. Misalignment between successive images greater than one pixel results in an image shift and causes an "edge effect"; presence of these "edge effects" in this study invalidated the angiogram series in which they occurred. Since the 3-dimensional surfaces ultimately represent instantaneous change in the degree of fluorescence, there is no direct method by which to quantify these data. This form of data representation is intended to be qualitative.

The contours of the characteristic surfaces corresponding to the relative distributions of blood flow before and after drug delivery were compared to each other, as well as to those from derived from the saline injection experiments, to determine if any changes occurred. A change in the contour of the 3-dimensional surfaces was interpreted as indicating a change in the distribution of ICG dye-filling rate at the corresponding area in the fundus over which the contour change lies.

RESULTS

During these experiments, the heart rates and blood pressures of the monkeys varied less than 10% from baseline levels, and there were no significant responses of the heart rates and/or blood pressures to intravenous injections of either pentoxifylline or saline. The baseline characteristic 3-dimensional surfaces for each animal were reproducible over time as reported earlier; compare, for example, the left-hand surfaces in the top rows of figure 4, figure 5, which are from the same eye, but from two different experimental sessions.

A comparison of each monkey's pre-drug-injection to post-drug-injection characteristic surfaces in all the experiments revealed significant changes in their topographies, corresponding to changes in the distribution of choriocapillaris blood flow as a result of the pentoxifylline injections. This is demonstrated in figure 4 where the characteristic 3-dimensional surfaces from representative experiments for each of the three monkeys are shown. Five minutes after pentoxifylline injection, blood flow in the posterior pole region increase relative to that which was present pre-injection. However, within 20-40 minutes post-injection, the blood flow levels returned to the pre-injection levels.

After intravenous injections of saline instead of pentoxifylline, however, no differences were seen from the baseline levels, as demonstrated for one of the monkeys in figure 5. This demonstrated that the increases in blood flow seen following pentoxifylline injection were not due to the volume of fluid injected alone. The order in which the saline and pentoxifylline injections were made during a particular experimental session did not affect the outcome with respect to changes in the posterior pole blood flows.

Analysis of the three samples of pentoxifylline solutions tested indicated concentrations of 24.8mg/ml, 26.5mg/ml and 30.3mg/ml.

DISCUSSION

Our experimental results consistently demonstrated increased posterior pole choroidal blood flow in rhesusmonkeys following intravenous injection of pentoxifylline. Documenting heart rate and blood pressure stability ensured that the observed increases in blood flow were not due to increased heart rates or blood pressures. In fact, during each angiogram, the heart rate and blood pressure tended to slightly decrease in the monkeys following drug; if anything, these trends would be likely to lead to reduced blood flows, the opposite of the effect caused by the pentoxifylline.

The post-drug-injection blood flow increases, likewise, were not due to increased intravascular volume change. The absence of changes in the topographies of the characteristic 3-dimensional surfaces after saline injections of the same volumes as the pentoxifylline injections indicates that the volume changes alone were not responsible for the changes induced by pentoxifylline. The pentoxifylline-induced changes in choroidal blood flow appear also to have been independent of the form of anesthesia used to sedate the monkeys, as the same effects were observed with both inhalation and intravenous anesthesia.

Previous studies, based on indirect measurements, reported increased ocular blood flow in response to pentoxifylline. Brancato, Menchini, and Michelone reported a double blind study in which 30 diabetic patients received pentoxifylline (Trental) orally; pentoxifylline-treated patients had reduced erythrocyte aggregation in their conjunctival circulations compared to placebo-treated patients [11]. Iwafune and Yoshimoto reported earlier absorption of hemorrhage and less ocular neovascularization, as well as a decreased arm-to-retina circulation time, in 20 patients treated with pentoxifylline versus 22 untreated patients [12]. More recently, investigators utilized entoptic phenomenon to show increased retinal blood flow in patients following pentoxifylline therapy [13], [14]. However, the effect of pentoxifylline on the choroidal circulation has not heretofore been determined.

The results of the qualitative method of evaluation used in this study demonstrate definite changes in the choroidal posterior pole circulation after the intravenous injection of pentoxifylline. The changes occurred immediately because the intravenous injection route used assured rapid delivery of a large drug dose, bypassing the parenteral system. But even so, the results are consistent with the findings of other investigators who have measured in-patients retinal blood flow changes following oral pentoxifylline treatment [13], [14].

Whether or not this pentoxifylline effect of increasing choroidal circulation has therapeutic benefit remains to be determined and is beyond the scope of this study, whose primary purpose was to demonstrate the utility of a quantitative measurement of choroidal blood flow changes.

There is no inherent benefit to assigning numerical values to the z-axes of the 3-dimensional surface graphs, since the axes of the curves can be scaled arbitrarily. It is not the height of the curves that is important, but the topography of the characteristic 3-dimensional surfaces. Indeed, the graphs can and should be arbitrarily scaled so that the topographies of different graphs can be more easily compared to each other for the purpose of detecting changes.

This method is imminently applicable to human use, since it is based on high-speed ICG angiograms, which are routinely obtainable, now that commercial SLO systems are readily available. Certainly it opens the door to routinely investigating the effect of certain therapeutic drugs on the choroidal circulation, and it may potentially provide a way to quickly titrate doses of drugs known to affect systemic blood circulation.


Figure 1.
The rendering of a 2-dimensional image into a 3-dimensional surface mapping.
 

Figure 2.

(Les tableaux sont exclusivement disponibles en format PDF).


Figure 3.
A series of the characteristic 3-dimentional surfaces derived from high-speed ICG angiogram images for 21 different eyes, demonstrating that each characteristic surface is unique to the eye from which it was derived. These surfaces are organized into four arbitrary groups approximately according to major topographical similarities.


Figure 4.
The characteristic 3-dimensional surfaces from representative experiments for each of the three monkeys injected with pentoxifylline.


Figure 5.
Comparison of the effects of pentoxifylline injection and injection of an identical volume of saline in the same eye. The characteristic 3-dimensional curves in the top row demonstrate the effects of pentoxifylline injection, whereas those in the bottom row demonstrate that saline injection has no appreciable effect. These data are from the same monkey eye shown in the top row of Figure 4. Note the consistency of pentoxifylline-induced changes in the same eye demonstrated by comparing the top row of this figure with that of Figure 5.


REFERENCE(S)

[1] Flower RW, Hochheimer BF. A clinical technique and apparatus for simultaneous angiography of the separate retinal and choroidal circulations. Invest Ophthal, 1973;12:248-261.

[2] Yannuzzi LA, Slakter JS, Sorenson JA, et al. Digital indocyanine green videoangiography and choroidal neovascularization. Retina, 1992;12:191-223.

[3] Shariga F, Ojima Y, Matsuo T, Takasu I, Matsuo N. Feeder vessel photocoagulation of subfoveal choroidal neovascularization secondary to age-related macular degeneration. Ophthalmology, 1998; 105:662-669.

[4] Staurenghi G, Orzalesi N, La Capria A, Aschero M. Laser treatment of feeder vessels in subfoveal choroidal neovascular membranes: A revisitation using dynamic indocyanine green angiography. Ophthalmology, 1998;105:2297-2305.

[5] Flower RW. Experimental studies of indocyanine green dye-enhanced photocoagulation of choroidal neovascularization feeder vessels. Am J Ophthalmol, (In press: April, 2000).

[6] Klein GJ, Baumgartner RH, Flower RW. An image processing approach to characterizing choroidal blood flow. Invest Ophthalmol Vis Sci, 1990;31:629-37.

[7] Flower RW, Fryczkowski AW, McLeod DS. Variability in choriocapillaris blood flow distribution. Invest Ophthalmol Vis Sci, 1995; 36:1247-1258.

[8] Ferrari E, Fioravanti M, Patti AL, Viola C, Solerte SB. Effects of long-term treatment (4 years) with pentoxifylline on haemorheological changes and vascular complications in diabetic patients. Pharmatherapeutica, 1987;5:26-39.

[9] Sebag J, Tang M, Brown S, Sadun AA, Charles MA. Effects of pentoxifylline on choroidal blood flow in nonproliferative diabetic retinopathy. Angiology, 1994;45:429 33.

[10] Burrows JL. Determination of Oxpentifylline and three metabolites in plasma by automated capillary gas chromatography using nitrogen-selective detection. J Chromatogr, 1987;423:139-46.

[11] Brancato R, Menchini U, Michelone C. Evaluation of the effect of pentoxifylline on the conjunctival microcirculation of diabetic subjects. Ric Clin Lab, 1981;11 Suppl. 1: 327-31.

[12] Iwafune Y, Yoshimoto H. Clinical use of pentoxifylline in haemorrhagic disorders of the retina. Pharmatherapeutica, 1980;2:429-38.

[13] Sonkin PL, Kelly LW, Sinclair SH, Hatchell DL. Pentoxifylline increases retinal capillary blood flow velocity in patients with diabetes. Arch Ophthalmol, 1993;111:1647-52.

[14] Sonkin PL, Sinclair SH, Hatchell DL. The effect of pentoxifylline on retinal capillary blood flow velocity and whole blood viscosity. Am J Ophthalmol, 1993;115:775-80.


© 2000 Elsevier Masson SAS. Tous droits réservés.
EM-CONSULTE.COM is registrered at the CNIL, déclaration n° 1286925.
As per the Law relating to information storage and personal integrity, you have the right to oppose (art 26 of that law), access (art 34 of that law) and rectify (art 36 of that law) your personal data. You may thus request that your data, should it be inaccurate, incomplete, unclear, outdated, not be used or stored, be corrected, clarified, updated or deleted.
Personal information regarding our website's visitors, including their identity, is confidential.
The owners of this website hereby guarantee to respect the legal confidentiality conditions, applicable in France, and not to disclose this data to third parties.
Close
Article Outline
You can move this window by clicking on the headline
@@#110903@@