Cerebral Function Monitoring: A New Scoring System for the Evaluation of Brain Maturation in Neonates Vladimir F. Burdjalov, MD; Stephen Baumgart, MD; and Alan R. Spitzer, MD ABSTRACT. Objective. Cerebral function monitoring (CFM), using compressed single-channel amplitude-integrated electroencephalogram recorded from 2 biparietal electrodes, has been shown previously to be a simple bedside tool for monitoring neonatal central nervous system (CNS) status. As the pattern of the CFM changes with gestational age, the technique can be used to assess brain maturation in premature infants. We have developed a new scoring system for the interpretation of neonatal CFM recordings. The objective of this study was to evaluate CFM tracings at increasing gestational and postnatal ages to develop a scoring system to quantify CFM pattern changes. Methods. Term and preterm neonates were studied with CFM at 12 to 24 hours of life, 48 to 72 hours of life, and then weekly or biweekly until hospital discharge. Each study comprised 8 to 24 hours of continuous CFM recording. CFM recordings were evaluated using the scoring system for record continuity, presence of cyclic changes in electrical activity, degree of voltage amplitude depression, and bandwidth. Each variable was scored for each recording. All variables were summed to yield a total score (minimum 0, maximum 13). Total scores were correlated with gestational and postconceptional ages. Results. Thirty infants were studied with gestational ages at birth that ranged from 24 to 39 weeks and birth weights that varied between 450 and 3850 g. A total of 146 CFM tracings were analyzed. With advancing gestational and postconceptional age, scores for each variable as well as total scores progressively increased with CNS maturation. The highest scores were attained at 35 to 36 weeks’ postconceptional age, which corresponded to previously reported subjective observations performed by visual description of CFM patterns. Of the 4 component variables that we analyzed, the most sensitive indicators of CNS maturity were 1) the presence of a cycling pattern, 2) the continuity of the record pattern, and 3) the CFM recording bandwidth. Conclusions. Our proposed scoring system may be a valuable tool to quantify changes during CFM more objectively, reflecting variations in CNS activity in newborn infants and allowing for better statistical comparisons between amplitude-integrated electroencephalogram tracings from different patients as well as from the same patient at different points of time. Pediatrics 2003; 112:855– 861; amplitude-integrated EEG, brain maturation, cerebral function monitoring, premature neonate.
From the Division of Neonatology, Department of Pediatrics, State University of New York at Stony Brook, Stony Brook, New York. Received for publication Sep 27, 2002; accepted Feb 18, 2003. Reprint requests to (A.R.S.) SUNY-Stony Brook, HSC T11-060, Stony Brook, NY 11794-8111. E-mail:
[email protected] PEDIATRICS (ISSN 0031 4005). Copyright © 2003 by the American Academy of Pediatrics.
ABBREVIATIONS. IVH, intraventricular hemorrhage; HIE, hypoxic-ischemic encephalopathy; CFM, cerebral function monitoring; aEEG, amplitude-integrated electroencephalogram; CNS, central nervous system.
D
uring the past decade, the survival of premature neonates has improved dramatically.1 The neonatal population, especially the extremely low birth weight infant, however, is still at great risk for many complications that occur during the newborn period that may result in clinically significant neurologic injury, including intraventricular hemorrhage (IVH), periventricular leukomalacia, hypoxic-ischemic encephalopathy (HIE), seizures, and meningitis. Ultimately, such problems may, in turn, lead to developmental delay and cerebral palsy.2 These circumstances emphasize the need for improved surveillance of cerebral function during this critical period of time. Cerebral function monitoring (CFM) is a bedside, readily available, user-friendly device for continuous recording of amplitude-integrated electroencephalogram (aEEG) data. The compressed form of this recording allows evaluation of baseline brain wave activity and detection of seizures. CFM recordings are sampled from two biparietal electrodes that integrate electrical activity in the underlying brain regions that receive the bulk of cerebral blood flow. Since it was first introduced for continuous cerebral activity monitoring in adults,3 CFM has been used increasingly in some neonatal centers to follow hypoxic-ischemic brain injury, detect seizures, monitor the effects of different interventions and events on neonatal cerebral activity, and predict future outcome.4 –18 Neonatal aEEG recorded by CFM has also been correlated with gestational age and maturity of preterm and term infants’ electrical brain activity.19 –23 One of the current problems with evaluating aEEG tracings, however, is that most investigators have used subjective visual pattern recognition to assess the record’s continuity, bandwidth, presence of a cycling pattern, and the occurrence of epileptiform activity.13,19 –21,23 Various investigators have used different combinations of these descriptive patterns, and there has been no uniform agreement on how best to analyze the aEEG, particularly in very immature neonates. Furthermore, none of the existing techniques for interpreting CFM data seems to be suitable for statistical comparisons between different PEDIATRICS Vol. 112 No. 4 October 2003
855
tracings, either from different infants or from the same infant at different postconceptional ages. We therefore have developed a scoring system for the interpretation of neonatal CFM recordings. Our system incorporates all variables of the neonatal aEEG tracing previously described and grades them according to measured changes in the signal pattern that can be readily scored. Our hypothesis was that numerical scoring of aEEG tracings at increasing postconceptional ages would accurately quantify the maturational changes previously described with more traditional pattern recognition analysis.
CFM Tracing Interpretation
METHODS Study Population This study was conducted in the Regional Perinatal Center of the neonatal intensive care unit of the State University of New York at Stony Brook during the period from January 2000 to February 2002. Thirty patients were enrolled with birth weights that ranged from 450 g to 3850 g (median: 862 g) and gestational ages at birth from 24 weeks to 39 weeks (median: 27 weeks). Gestational age was assessed by the last menstrual period and/or by prenatal ultrasonography and was confirmed by physical examination, using the Ballard Scoring System. Inclusion criteria for study participation were 1) Absence of a major chromosomal or/and congenital malformation; 2) absence of HIE, IVH, or periventricular leukomalacia; and 3) absence of any sedative medication. Cranial ultrasonography was initially performed on all premature infants ⬍34 weeks of gestation in the first 3 days of life. Thereafter, studies were done 1 week later according to clinical indications and were repeated at the discretion of the attending neonatologist. Infants who demonstrated any of the previously mentioned abnormalities on any study were excluded from aEEG evaluation. This study was approved by the IRB of the State University of New York at Stony Brook. Informed consent was obtained from the parents of each patient enrolled in the study.
CFM Recordings Compressed aEEG recordings were obtained using the Cerebral Function Monitor Multi-Trace 2 (Lectromed, Olympic Medical Inc, Seattle, WA). A pair of standard gold-disk EEG electrodes was attached to the scalp parietal areas in the P3 to P4 positions (International 10-20 System) using Elefix EEG-electrode paste (Nihon Kohden, Tokyo, Japan). The electrodes were covered with a small wad of sterile cotton and secured with Transpore (3M, St Paul, MN) nonocclusive tape to prevent drying of the paste. A reference electrode was similarly placed over the frontal midline region of the scalp. The EEG signal obtained from these electrodes was processed by amplification, a special filtration algorithm to attenuate signals below 2 Hz and above 16 Hz, amplitude and time compression, and rectification. Finally, that compressed aEEG signal was recorded on heat-sensitive paper, using a semilogarithmic scale, and run at a speed of 6 cm/h. The CFM was calibrated before each
TABLE 1.
Continuity
0
Discontinuous
None
1
Somewhat continuous Continuous
Waves first appear
3 4 5
856
From each recording, the most stable uninterrupted period of at least 3 to 4 hours’ duration were chosen for analysis. The following 4 component variables of the aEEG record were evaluated, categorized, and graded according to the proposed scoring system (Table 1): 1. Record “continuity”: Continuity was assessed by observing the overall density of the sample tracing. Continuity refers to the appearance of frequent variations in the aEEG electrical activity response. High levels of continuity meant that there was constant and frequently alternating electrical activity (pen peaks and troughs), so that the recording either appeared very tightly compressed on the tracing or not. Low levels of continuity had a much reduced number of electrical variations, with greater separation of the recording signal peaks and troughs. 2. Presence or absence of “cycling”: Cycling refers to the emergence and progression of periods during the CFM epoch analyzed where the peak-to-trough width of the recording would expand and subsequently contract. Cycling was observed as variations in both amplitude and continuity of electrical activity on aEEG tracings. 3. Amplitude (in V) of the lower border: The magnitude of the CFM tracing’s lower border (voltage troughs) was estimated as the average lower microvolt level during the recording epoch. A line drawn though the lower margin of the aEEG band appeared with half of the microvolt troughs below the line and half above. With the emergence of cycling (see above), the narrowest part of the recording was evaluated. 4. Bandwidth of the aEEG: Bandwidth reflects a combination of the voltage span (peak-to-trough) of the tracing and the magnitude of the aEEG depression (amplitude of the lower border). The span was calculated as the difference between the upper and lower voltage margins of the tracing’s narrowest part. Each variable was scored, and the individual component scores were summed to determine the total score for each recording. The minimum possible total score was 0, and the maximum was 13. Individual component variable scores and the total scores from all CFM recordings were subsequently evaluated in relationship to each infant’s postconceptional age. The weighted R value, or correlation coefficient by least-squares method, was calculated as a measure of correlation between scores and postconceptional age.24 Intercoder reliability for total scores was estimated by ex-
CFM Scoring System Summarized
Score
2
recording using the manufacturer’s recommended procedure. Periodic recalibration was performed during CFM epochs to be certain that there was no significant baseline drift during the course of the study period. Continuous CFM recording was performed on each patient for an 8- to 24-hour period, adhering to the following schedule as closely as possible: at 12 to 24 hours of life, at 48 to 72 hours of life, and then weekly or biweekly until hospital discharge. On occasion, the clinical status of an individual infant made recording impossible and the study was then conducted as soon as it was again feasible. For most infants, there was only 1 recording during the first week of life. In some infants who had ⬎1 study performed, the best technical CFM tracing that had the least artifact and fewest interruptions because of intensive care needs was selected.
Cycling
Not definite, somewhat cycling Definite cycling, but interrupted Definite cycling, noninterrupted Regular and mature cycling
Amplitude of Lower Border
Bandwidth Span and Amplitude of Lower Border
Severely depressed (⬍3 V) Somewhat depressed (3–5 V) Elevated (⬎5 V)
Very depressed: low span (ⱕ15 V) and low voltage (5 V) Very immature: high span (⬎20 V) or moderate span (15–20 V) and low voltage (5 V) Immature: high span (⬎20 V) and high voltage (⬎5 V) Maturing: moderate span (15–20 V) and high voltage (⬎5 V) Mature: low span (⬍15 V) and high voltage (⬎5 V)
CEREBRAL FUNCTION MONITORING IN PREMATURE NEONATES
amining the percentages of scores within ⫾1 and ⫾2 points for at least 2 readers. One author’s total scores were considered for statistical analysis (V.F.B.).
RESULTS
During this study, we analyzed a total of 146 recordings in 30 qualified infants. Seven infants who had severe IVH or HIE were excluded. The number of recordings for each patient varied from 1 to 10 (median: 4.5). Examples of the tracing analysis are shown in Fig 1. We observed in the tracing pattern certain maturational changes that seemed to correspond to scores specific for particular gestational ages. Recordings done on younger gestational age infants (24 –26 weeks) were characterized by discontinuous CFM background continuity (continuity scores 0 –1); the complete absence of or only rudimentary cyclic changes (cycling scores 0 –2), with the first such cyclical changes being noted at approximately 26 weeks of postconceptional age; slightly depressed lower border amplitude (amplitude scores 1–2); and a predominantly broad bandwidth (scores 0 –2). With advancing postconceptional age (27–28 weeks), the CFM pattern showed a more continuous tracing (the majority of scores were 1–2, with only 3 patients showing scores of 0). Cyclical 20- to 30minute periods of wider amplitude were intermixed with periods of narrower bandwidth and began to emerge after 27 weeks of postconceptional age (cy-
cling scores ⱖ2). Cycling became more clearly recognizable after 29 weeks (scores ⱖ3) and were fully established by 34 weeks of postconceptional age (scores ⱖ4). There was also a progressive elevation of the minimum level of electrical amplitude (majority of lower border amplitude scores were at 2) and narrowing of the amplitude’s bandwidth (scores were 2–3, with 4 recordings showing a score of 1). The cyclical periods reached a completely mature pattern after 36 weeks’ postconceptional age (score of 5). Also with advancing gestational age, continuity increased progressively, reaching its maximum by 30 to 31 weeks’ postconceptional age. After 27 weeks’ postconceptional age, the lower border amplitude of the aEEG band remained elevated and bandwidth became progressively narrower, with scores of 3 between postconceptional ages of 29 to 34 weeks. The height of the aEEG band reached its maximum level by 35 to 36 weeks’ postconceptional age. The total score calculated for each recording also progressively increased from 1 to 2 at the earliest stages of maturity and reached its maximum (a score of 13) at approximately 39 weeks’ postconceptional age. The weighted regression of the mean total CFM score for all variables compared with mean postconceptional age is demonstrated in Fig 2. The progressions in individual components of the CFM score are seen in Figs 3 to 5. Lower border amplitude, although an integral component of the maturational
Fig 1. The left side of the figure demonstrates a progressive series of CFM monitor recordings; the right side shows the component score values for the respective studies. There is a maturation of the tracings from A through F in these recordings. Postconceptional age ranges are indicated. Co, continuity of the recording; Cy, presence of cycling; LB, lower border amplitude score; B, bandwidth; T, total score.
ARTICLES
857
Fig 2. Progressive maturation of the mean total CFM scores measured between 24 and 39 weeks’ postconceptional age. The total number of studies included is 146. There are multiple overlapping points at intersections of mean CFM score and postconceptional age, with the number of studies at each point indicated, and the regression was accordingly weighted.24
Fig 3. The regression of the cycling component for the CFM score is shown with postconceptional age (N ⫽ 146, 30 overlapping weighted means).24
score, had a lower correlation coefficient (R ⫽ 0.46, P ⬍ .2) and is omitted for brevity. Intercoder reliability was 82% (14 of 17) for independently scored recordings that came within 1 point of each other, whereas 100% of recordings that were independently scored were within 2 points. DISCUSSION
This study defines a new scoring system for aEEG evaluation that was devised to assess objectively the developmental maturation of the neurologically unimpaired premature infant. There was a progressive increase in both the overall score on the aEEG and 858
the 4 individual component scores that correlated closely to the chronologic and neurodevelopmental maturation of these infants. Our own data, as well as more subjective interpretations from other studies, have shown that the pattern of the neonatal aEEG changes with advancing gestational and postconceptional ages.19 –23 The components of the aEEG tracing that we quantified (record continuity, presence and stage of a cycling pattern, magnitude of tracing depression, and bandwidth) have been verified previously. Of the component variables described on the aEEG, the emergence of a cycling pattern seemed to
CEREBRAL FUNCTION MONITORING IN PREMATURE NEONATES
Fig 4. The regression of the continuity component for the CFM score is shown with postconceptional age (N ⫽ 146, 30 overlapping weighted means).24
Fig 5. The regression of the bandwidth component for the CFM score is shown with postconceptional age (N ⫽ 146, 30 overlapping weighted means).24
have the highest correlation with postconceptional age and could be considered the single best determinant of cerebral maturity. This finding might be explained by the emergence and establishment of sleep-wake cycles as determined by the level of integration of higher central nervous system (CNS) functions.25 aEEG “cycling” has not been correlated, however, with behavioral or sleep states in our study and may not be analogous to standard EEG interpretations of arousal. In contrast, aEEG “continuity” may depend more on the general status of CNS electrical activity than on maturity, and a continuous aEEG pattern is typically established during earlier stages of development in our study. It should be noted that this finding also differs from standard EEG interpretation of discontinuous versus continu-
ous background continuity with premature CNS development and may not be analogous. Inspection of the component variables that we have described permits the clinician to summarize the overall integration of an individual infant’s CNS electrical activity in greater detail. This approach may facilitate the evaluation of patients for seizures and other potentially abnormal patterns that may become more important in a less healthy premature infant population. However, no attempt was made in the present study to evaluate such abnormal patterns. The EEG is the current standard that reflects the state of CNS electrophysiologic activity. The EEG, however, is an impractical technology for use as a repetitive, continuous monitoring device in this popARTICLES
859
ulation of infants. During the past 2 decades, CFM has gained increasing attention in clinical neonatal research, because the aEEG is an easily applicable, readily available, and inexpensive device for continuous bedside evaluation of brain activity.3,26 The aEEG has been shown to have a good correlation with the standard multichannel EEG in term and critically ill neonates,14,20,27–29 and it can overcome some of the disadvantages of the latter as a monitoring device. Additional conventional EEG work is required, however, to validate observations made in premature infants with aEEG recordings. One of the great advantages of CFM is its simplicity and the possibility of quick on-line interpretation and analysis of overall brain function. To date, however, there has been no uniform agreement with respect to the assessment and interpretation of CFM recordings. On the basis of pattern recognition analysis, various authors have used different assessments of the aEEG tracing with some agreement, but there has not been a consistent approach to the various components of the recording that seem to change with increasing gestational age.13,19 –21,23 We believe that the current scoring system provides such an approach. Last, we urge some caution in using this scoring system in the first few days of life. In some infants, a definitive pattern of aEEG was not established (and could not be scored) until a few days after birth. This finding may be related to perinatal events and/or prenatal/perinatal interventions, and it requires additional investigation with respect to its significance. Moreover, our proposed scoring system has some inherent subjectivity within the component variables. Such a detailed scoring system may nevertheless allow anyone who is interested in the continuous monitoring of neonatal cerebral status to apply this score successfully in the assessment of maturation and CNS integrity. With additional technologic advances, such as digitized aEEG recording and computerized calculations of the various measures that we have outlined, the problem of subjectivity of scoring could potentially be overcome. CONCLUSIONS
CFM seems to be useful for following the normal maturation of the neonatal brain. Our proposed scoring system (although admittedly arbitrary) may become a valuable tool to quantify changes during maturation, more objectively reflect variations in CNS activity in newborn infants, and allow for better statistical comparisons between aEEG tracings from different patients, as well as from the same patient at different points of time. Future work should define the aEEG as seen in brain-injured premature neonates to better define pattern abnormalities in these infants. A different balance or even new component factors may prove to be more appropriate when using aEEG to assess acute HIE, when screening for subtle or cystic white matter injury, or when evaluating recovery from injury, as compared with evaluating maturation. 860
ACKNOWLEDGMENTS This investigation was supported with an equipment grant from Olympic Medical, Inc. We are grateful to all neonatal nursing staff in the neonatal intensive care unit at SUNY at Stony Brook for invaluable help and great patience. We also thank Dr Joseph DeCristofaro and Dr Anatoliy Ilizarov for advice and encouragement.
REFERENCES 1. Guyer B, Hoyert DL, Ventura SJ, MacDorman MF, Strobino DM. Annual summary of vital statistics—1998. Pediatrics. 1999;104:1229 –1246 2. Volpe JJ. Neurology of the Newborn. Philadelphia, PA: WB Saunders; 2001:217– 497 3. Maynard DE, Prior P, Scott DF. Device for continuous monitoring of cerebral activity in resuscitated patients. BMJ. 1969;11:545–546 4. Viniker DA, Maynard DE, Scott DF. Cerebral function monitor studies in neonates. Clin Electroencephalogr. 1984;14:185–192 5. Archibald F, Verma UL, Tejam NA, Handworker SM. Cerebral function monitor in the neonate. II. Birth asphyxia. Dev Med Child Neurol. 1984; 26:162–168 6. Hellstrom-Westas L, Rosen I, Svenningsen NW. Silent seizures in sick infants in early life. Diagnosis by continuous cerebral function monitoring. Acta Paediatr Scand. 1985;74:741–748 7. Hellstrom-Westas L, Rosen I, Svenningsen NW. Predictive value of early continuous amplitude integrated EEG recordings on outcome after severe birth asphyxia in full term infants. Arch Dis Child Fetal Neonatal Ed. 1995;72:F34 –F38 8. Murdoch-Eaton D, Toet M, Livingston J, Smith I, Levene M. Evaluation of the Cerebro Trac 2500 for monitoring of cerebral function in neonatal intensive care. Neuropediatrics. 1994;25:122–128 9. Hellstrom-Westas L, Bell AH, Skov L, Greisen G, Svenningsen NW. Cerebroelectrical depression following surfactant treatment in preterm neonates. Pediatrics. 1992;89:643– 647 10. Thornberg E, Ekstrom-Jodal B. Cerebral function monitoring: a method of predicting outcome in term neonates after severe perinatal asphyxia. Acta Paediatr. 1994;83:596 – 601 11. Eaton DG, Wertheim D, Oozeer R, Dubowitz LM, Dubowitz V. Reversible changes in cerebral activity associated with acidosis in preterm neonates. Acta Paediatr. 1994;83:486 – 492 12. Wertheim D, Mercuri E, Faundez JC, Rutherford M, Acolet D, Dubowitz L. Prognostic value of continuous electroencephalographic recording in full term infants with hypoxic ischemic encephalopathy. Arch Dis Child. 1994;71:F97–F102 13. AI-Naqeeb N, Edwards AD, Cowan FM, Azzopardi D. Assessment of neonatal encephalopathy by amplitude-integrated electroencephalography. Pediatrics. 1999;103:1263–1271 14. Toet MC, Hellstrom-Westas L, Groenendaal F, Eken P, de Vries LS. Amplitude integrated EEC 3 and 6 hours after birth in full term neonates with hypoxic-ischaemic encephalopathy. Arch Dis Child Fetal Neonatal Ed. 1999;81:F19 –F23 15. Stinger E, Eriksson E, Stiffer A, Schollin J, Aman J. Monitoring of early postnatal glucose homeostasis and cerebral function in newborn infants of diabetic mothers. A pilot study. Early Hum Dev. 2001;62:23–32 16. Romeo MG, Tina LG, Cilauro S, et al. The importance of using the cerebral function monitor (CFM) in the neurological prognosis of neonates in intensive care. Pediatr Med Chir. 1998;20:197–199 17. Hellstrom-Westas L, Westgren U, Rosen I, Svenningsen NW. Lidocaine for treatment of severe seizures in newborn infants. I. Clinical effects and cerebral electrical activity monitoring. Acta Paediatr Scand. 1988;77: 79 –78 18. Benders MJ, Meinesz JH, van Bel F, van de Bor M. Changes in electrocortical brain activity during exchange transfusions in newborn infants. Biol Neonate. 2000;78:17–21 19. Verma UL, Archibald F, Tejam NA, Handworker SM. Cerebral function monitor in the neonate. I: normal patterns. Dev Med Child Neurol. 1984;26:154 –161 20. Thornberg E, Thiringer K. Normal patterns of the cerebral function monitor trace in term and preterm neonates. Acta Paediatr Scand 1991; 79:20 –25 21. Hellstrom-Westas L, Rosen I, Svenningsen NW. Cerebral function monitoring during the first week of life in extremely small low birthweight (ELSBW) infants. Neuropediatrics. 1991;22:27–32 22. Kuhle S, Klebermass K, Olischar M, et al. Sleep-wake cycles in preterm infants below 30 weeks of gestational age. Preliminary results of a prospective amplitude-integrated EEG study. Wien Klin Wochenschr. 2001;113:219 –223
CEREBRAL FUNCTION MONITORING IN PREMATURE NEONATES
23. Klebermass K, Kuhle S, Kohlhauser-Vollmuth C, Pollak A, Weninger M. Evaluation of the Cerebral Function Monitor as a tool for neurophysiological surveillance in neonatal intensive care patients. Childs Nerv Syst. 2001;17:544 –550 24. Bland J, Altman D. Calculating correlation coefficients with repeated observations: part 2— correlation between subjects. BMJ. 1995;310: 633 25. Thoman EB, Denenberg VH, Sieve J, Zeidner LP, Becker P. State organization in neonates: developmental inconsistency indicates risk for developmental dysfunction. Neuropediatrics. 1981;12:45–54 26. Prior PF, Maynard DE: Monitoring Cerebral Function. Long-Term Record-
ings of Cerebral Electric Activity and Evoked Potentials. Amsterdam, the Netherlands: Elsevier; 1986 27. Hellstro¨ m-Westas L. Comparison between tape-recorded and amplitude-integrated EEG monitoring in sick newborn infants. Acta Paediatr Scand. 1992;81:812– 819 28. Greisen G. Tape-recorded EEG and the cerebral function monitor: amplitude integrated, time compressed EEG. J Perinat Med. 1994;22:541–546 29. Toet MC, Van Der Meij W, De Vries LS, Uiterwaal CS, Van Huffelen KC. Comparison between simultaneously recorded amplitude integrated electroencephalogram (cerebral function monitor) and standard electroencephalogram in neonates. Pediatrics. 2002;109:772–779
LAWSUIT SAYS DOCTORS WERE PAID ENDORSERS
“Documents released yesterday (May 29, 2003) in the case of a drug company whistle-blower shed light on how extensively doctors were involved in promoting unapproved uses of a Warner-Lambert drug, Neurontin. Warner-Lambert paid dozens of doctors tens of thousands of dollars each to speak to other physicians about how Neurontin, an epilepsy drug, could be prescribed for more than a dozen other medical uses that had not been approved by the Food and Drug Administration (FDA). . . It is illegal for a drug company to market a medicine for use the FDA has not approved, but doctors can prescribe a drug in any manner that they think is best for their patients. . . Corporate documents made public in the case. . . state that a . . . strategy was to focus on respected doctors in the major teaching hospitals who would serve as ‘Neurontin champions.’”
Petersen M. New York Times. May 30, 2003
Noted by JFL, MD
ARTICLES
861
