Electroretinogram And Cue-Elicited Craving In Withdrawn Cocaine-Dependent Patients: A Replication

American Journal of Drug and Alcohol Abuse, May, 2001 by David A. Smelson, Alec Roy, Monique Roy, Dora Tershakovec, Charles Engelhart, Miklos F. Losonczy

INTRODUCTION

In an effort to improve treatment efficacy, researchers have begun to examine the usefulness of subgrouping cocaine-dependent patients to match them to the most appropriate treatment modalities. One subgrouping variable may be the extent to which cocaine addicts may demonstrate biologic dysregulation. For example, changes in the dopaminergic system have been reported with cocaine administration (1). In the eye, dopamine appears to be a critical neuromodulator in retinal processing (2, 3). Furthermore, various dopamine receptors have been identified in several levels of the mammalian retina (4, 5), including the [D.sub.1] and [D.sub.2] types (6), the subtypes thought to be involved in the rewarding effects of cocaine (7, 8). The electroretinogram (ERG) is a noninvasive physiologic technique for recording electrical potentials from the retina in response to flashes of light. Previous research has suggested that dopamine in the retina is involved with ERG responses, particularly blue cone b wave amplitude (9-11).

Because of the presence of dopamine in the retina, our initial studies examined whether withdrawn cocaine-dependent patients would show abnormal blue cone ERG amplitudes. Our findings showed that about 50% of cocaine-dependent patients displayed a reduced blue-cone b wave ERG amplitude (12, 13). Furthermore, this subgroup had more self-reported cocaine craving (14). We then wished to examine the relationship between the ERG blue-cone b wave amplitude and reactivity to laboratory-based cocaine cues designed to stimulate a craving state. Our preliminary data from 14 patients suggested that the subgroup of cocaine-dependent patients with a decreased ERG blue-cone b wave amplitude display more cue-elicited cocaine craving (15). Therefore, we were interested in trying to replicate these preliminary findings with a new series of withdrawn cocaine-dependent patients.

METHODS

In our first study, we recruited a consecutive series of 14 withdrawn cocaine-dependent patients from a locked inpatient drug dependence unit at the Department of Veterans Affairs New Jersey Health Care System, East Orange Campus. For the present study, an additional 21 withdrawn cocaine-dependent patients (17 African-American, 2 Hispanic, and 2 Caucasian) were recruited from the same substance abuse treatment program as the original sample; the patients had similar backgrounds and drug use histories. We recruited subjects who met DSM-IV criteria (16) for cocaine dependence, who reported using at least 12 g of cocaine during the 3 months prior to their treatment, and who were drug, alcohol, and medication free at the time of the ERG and cue-exposure procedure. Subjects were excluded if they (a) had a history of opiate, barbiturate, benzodiazapine, alcohol, or marijuana dependence or whose principal diagnosis was not cocaine dependence; (b) met DSM-IV criteria for another Axis I disorder; (c) were currently taking prescription medication that could effect the central nervous system; (d) had a preexisting eye condition; and (e) had clinical evidence of any cognitive impairment. Random urine toxicology screens were done at admission and several times throughout the inpatient treatment program to confirm the absence of any cocaine, opiates, barbiturates, or benzodiazapines.

After giving written informed consent, all subjects completed a demographic cocaine history questionnaire. At baseline, subjects completed the Voris Cocaine Craving Questionnaire (VCCQ), which has good reliability and validity (15, 17). During the cue-exposure procedure, subjects were seated in a comfortable chair in our laboratory and were shown a 2-minute videotape of people administering intravenous cocaine, followed by a 2-minute videotape of people smoking cocaine. Previous research has shown that craving for cocaine is elicited by these same videotape cues (15, 18). After viewing the videotapes, subjects were again asked to complete the VCCQ.

Subjects were then taken to the ERG laboratory at the University of Medicine and Dentistry--New Jersey Medical School. The ERG was recorded using standard methods (19) and bipolar Burian-Allen contact lens electrodes (Hansen Ophthalmic, Iowa City, IA). The pupil was dilated widely to at least 8 mm, and pupil diameter was measured before and after each recording session.

The ERG was detected with a signal averager (model CA-1000, Nicolet Spirit, Madison, WI) coupled to a Ganzfeld stimulator, which provides a relatively strong adapting field (17,000 trolands) at maximum brightness. Superimposed on this steady white-light field were flashes obtained from a strobe (model PS2, Grass Instruments, Quincy, MA). The flash could be modified by inserting filters into the receptacle in the strobe housing. The following filters (Kodak Wratten, Rochester, NY) were used: 98 (450 nm), 48 (471 nm), 71 (488 nm), 61 (534 nm), 21 (593 nm), and 29 (633 nm). The numbers in parentheses indicate the nominal wavelength of maximum transmission of each filter at the approximate color temperature of the strobe. The flash rate for cone ERG recording was 5.1 Hz. At least 100 responses were averaged. For the blue cone ERG, 1000 responses were averaged.