Total parenteral nutrition in pediatrics: the Borden award address.

Research in parenteral nutrition in infants has proceeded rapidly over the past few years, thanks in large part to the perfection of safe central venous delivery of hypertonic nutritive infusates. At present, there are clear definitions of indications and expectations of results for this method of therapy in two well-defined groups of patients--i.e., selected surgical neonates and infants with chronic intractable diarrhea. In addition, we have suggestive evidence of another potentially valuable application in the nutritional management of very low birthweight infants. However, in this group, a controlled study will be necessary before the role of total parenteral nutrition (TPN) in neonatal care of such infants can be determined precisely. Results obtained with TPN in adults with inflammatory bowel disease or acute renal failure suggest that trials of this technique in pediatric patients with these disorders should be carried out. As a result of the research in TPN carried out thus far, we have learned how to minimize or to treat many of the complications of the technique and we have identified at least the ways by which still others can be prevented. The future holds many new advances not only in the refinement of existing parenteral nutritional solutions but also, and perhaps of even greater importance; in the perfection of individualized total nutritional therapy for specific patients using the enteral route for those discrete components of intake for which digestive and/or absorptive mechanisms are unimpaired and using the parenteral route for the remainder.

Afferents to the central nucleus of the amygdala and functional subdivisions of the periaqueductal gray: neuroanatomical substrates for affective behavior.

Evidence suggests the periaqueductal gray (PAG) is involved in the integration of behavioral and autonomic components of affective behavior. Our laboratory has shown that electrical stimulation of the ventrolateral periaqueductal gray (vl PAG) versus the dorsolateral periaqueductal gray (dl PAG), in the rabbit, elicits two distinct behavioral/cardiorespiratory response patterns. Furthermore, evidence suggests that the amygdaloid central nucleus (ACe) may influence cardiovascular activity during emotional states. The purpose of this study was to delineate the topography and determine the origin of forebrain projections to the PAG and the ACe, as well as commonalties and differences in the pattern of afferents. Examination of common afferents may lend insights into their function as components of a forebrain system regulating autonomic activity during emotional states. Separate retrograde tracers were injected into functional subdivisions of the PAG and the ACe in rabbits. PAG injections led to neuronal labeling in numerous cortical regions including the ipsilateral medial prefrontal and insular cortices. Additionally, bilateral labeling was observed in several hypothalamic nuclei including the paraventricular nucleus, the dorsomedial nucleus and the ventromedial nucleus as well as the region lateral to the descending column of the fornix. Sparse labeling was also seen in various basal forebrain regions, thalamic nuclei and amygdaloid nuclei. Many of these regions were also labeled following injections in the ACe. Although double-labeled cells were never observed, afferents to the ACe were often proximal to PAG afferents. Implications of these findings are discussed in terms of two functionally distinct behavioral/cardiovascular response patterns.

Electrophysiological evidence for hypothalamic defense area input to cells in the lateral tegmental field of the medulla of rabbits.

Single cell recordings were made from neurons in the lateral tegmental field of the medulla (LTFM) during electrical stimulation of the hypothalamic defense area (HDA) of the rabbit. Fifty-four cells were inhibited by HDA stimulation; 23 of these cells received barosensory input. Twenty-two LTFM cells showed an increase in firing rate during HDA stimulation; 10 of these cells received barosensory input. The results of this study provide evidence that the hypothalamic defense area makes functional connections with cardiovascular-influenced neurons in the LTFM.

Cardiovascular responses elicited by electrical and chemical stimulation of the rostral medullary raphe of the rabbit.

Electrical stimulation of the rostral medullary raphe (RMR) of the rabbit elicited pressor responses that were accompanied by tachycardia or bradycardia. Stimulation of dorsal sites (the dorsal raphe obscurus) evoked a pressor/tachycardia response and stimulation of ventral sites (the ventral raphe obscurus, raphe magnus and raphe pallidus) produced a pressor/bradycardia response. Electrical stimulation of the RMR after sinoaortic denervation led to an increase in the magnitude of the pressor response elicited from all stimulation sites, a decrease in the magnitude of the bradycardia produced by stimulation at the ventral sites, but had no effect upon the magnitude of the tachycardia observed from stimulation of the dorsal sites. These findings suggest that electrical stimulation of the dorsal sites leads to inhibition of the cardiomotor component of the baroreceptor reflex. The results of vagal blockade experiments demonstrated that baroreceptor attenuation of the pressor responses at ventral sites was mediated primarily by parasympathetic input to the heart. Chemical stimulation of the RMR with L-glutamate also led to a pressor/tachycardia response at the dorsal sites and a pressor/brachycardia response at the ventral sites. This finding provides evidence that neuronal cell bodies, not axon of passage, mediated the responses elicited by electrical stimulation.

Anatomical and functional connections of neurons of the rostral medullary raphe of the rabbit.

Single cell recordings were made from neurons in the rostral medullary raphe (RMR) of the rabbit. The recording sites were ones that had been shown to yield pressor responses from electrical stimulation and by pressure injections of glutamate. Electrical stimulation of the intermediolateral (IML) region of the spinal cord led to antidromic activation of 12 of the 100 cells studied. Eleven of these cells were located in raphe pallidus or raphe magnus, and one cell was located in raphe obscurus. These findings were consistent with the results of horseradish peroxidase (HRP) histochemistry experiments. Injections of HRP into the IML led to heavy cell body labeling in raphe pallidus and raphe magnus, but sparse labeling in raphe obscurus. Cells in the RMR could be orthodromically activated by electrical stimulation of the putative defense area of the periaqueductal (PAG) but not by stimulation of putative defense areas in the hypothalamus. Most of these cells were located in raphe pallidus or raphe magnus. Similarly, HRP injections into raphe pallidus and raphe magnus led to heavy cell body labeling in the PAG but not the hypothalamus; no cell body labeling was found in the PAG when injections were made into raphe obscurus.

Auditory cortex lesions prevent the extinction of Pavlovian differential heart rate conditioning to tonal stimuli in rabbits.

The present study examined the role of the auditory cortex in the extinction of differentially conditioned heart rate (HR) responses in rabbits. Lesions were placed bilaterally in either the auditory cortex or the visual cortex. Three days after recovery from surgery, the auditory cortex lesion group and the visual cortex lesion control group were habituated to the tonal conditioned stimuli (CSs), and then given 2 days of Pavlovian differential conditioning (60 trials per day) in which one tone (CS+) was always paired with the unconditioned stimulus and another tone (CS-) was never paired with the unconditioned stimulus. Animals that had demonstrated reliable differential conditioning (CS+ response at least 5 beats greater than the CS- response) were placed on an extinction schedule for 7 days. The extinction schedule was identical to the differential conditioning schedule with the exception that shock never followed the CS+. The results of the study indicate that auditory cortex lesions prevent the extinction of differential bradycardia conditioned responses (CRs) to tonal CSs. Whereas the bradycardia responses to the CS+ quickly extinguished in the group that had control lesions in the visual cortex, the auditory cortex lesion group continued to exhibit significantly larger bradycardiac HR CRs to the CS+ relative to the CS- during all 7 days of extinction. These results suggest that the animals in the auditory cortex lesioned group did not inhibit responses to a previously reinforced stimulus (i.e., CS+) as well as animals with control lesions in the visual cortex.

Basal and reactive plasma catecholamine levels under stress and anesthesia in rabbits.

The present study was undertaken to determine if electrical stimulation of the hypothalamic defense area (HDA) and baroreflex activation elicited by head up body tilt produced changes in plasma catecholamine (CA) levels in anesthetized rabbits. We also compared the effects of two anesthetics, isoflurane and sodium pentobarbital, upon basal and reactive CA levels, and upon autonomic reactivity. HDA stimulation was found to produce significant increases in plasma norepinephrine (NE) levels but not epinephrine (E) levels. Passive tilt was found to produce statistically significant increases in NE levels for both anesthetics used and a significant increase in E levels for animals anesthetized with isoflurane. Basal and reactive measurements provided evidence that pentobarbital has a more suppressive effect upon the autonomic nervous system than isoflurane: (a) Basal NE levels were significantly lower in pentobarbital anesthetized animals than in isoflurane-anesthetized animals; and (b) Baroreceptor sensitivity to a passive tilt stressor was significantly higher for animals anesthetized with isoflurane than for animals anesthetized with pentobarbital.

Comparison of peripheral blood flow patterns associated with the defense reaction and the vigilance reaction in rabbits.

The present study compared the skeletal muscle and visceral blood flow patterns elicited by electrical stimulation of the hypothalamic defense area (HDA) and the hypothalamic vigilance area (HVA) of the rabbit. Electrical stimulation of the HDA evoked a pressor response, tachycardia, hyperventilation, an increase in blood flow to the skeletal muscles and decreased blood flow to visceral organs. Stimulation of the HVA yielded a pressor response, bradycardia, inspiratory apnea and decreased blood flow to both the skeletal muscles and the viscera. Intravenous injections of atropine methyl nitrate significantly reduced the HVA-elicited bradycardia and the HDA-elicited increase in blood flow to the skeletal muscles, thereby providing evidence that the bradycardia was mediated by vagal efferents and that the rabbit has an atropine-sensitive cholinergic vasodilation system. The decrease of blood flow to the visceral organs associated with the defense reaction and vigilance reaction was reversed by intravenous injections of the alpha-1 receptor blocker prazosin.

Modulation of the baroreceptor reflex by stimulation of the hypothalamic defense and vigilance areas.

Electrical stimulation of the hypothalamic defense area (HDA) elicits a pressor/tachycardia response that is believed to prepare an animal for fight or flight. In contrast, electrical stimulation of the hypothalamic vigilance area (HVA) evokes a pressor/bradycardia response that is associated with the inhibition of movement. The differences in the behavioral components of these two affective response patterns suggest differential modulation of the baroreceptor reflex. The present study tested this idea by assessing the effects of electrical stimulation of the HDA and the HVA upon the bradycardia/depressor response elicited by stimulation of the aortic nerve (AN) in rabbits. Concurrent HDA and AN stimulation was observed to attenuate the AN-elicited bradycardia but enhanced the depressor response elicited by AN stimulation. In contrast, concurrent stimulation of the HVA and AN enhanced the bradycardia elicited by AN stimulation but reduced the magnitude of the AN-elicited depressor response. These results provide evidence for differential modulation of the baroreceptor reflex during the defense and vigilance reactions.

Hypothalamic, midbrain and bulbar areas involved in the defense reaction in rabbits.

The present study mapped neuroanatomical sites in the hypothalamus and periaqueductal gray (PAG) of the rabbit which, when stimulated electrically, evoked the cardiorespiratory components of the defense reaction (CRDR). This included increases in heart rate, blood pressure, hindlimb blood flow and respiration rate. All of the components of the CRDR were elicited by electrical stimulation of the posterior hypothalamus, at sites dorsal and medial to the fornix. Although there were regions throughout the PAG in which electrical stimulation elicited concomitant increases in blood pressure, hindlimb blood flow and respiration rate, only stimulation of the dorsal PAG evoked tachycardia. Injection of horseradish peroxidase into the rostral ventrolateral medulla (RVLM) led to heavy retrograde and anterograde labeling in the region of the hypothalamus that yielded the CRDR when stimulated electrically. Heavy labeling was also observed in the dorsal and ventral PAG. The results of this study provide evidence that the posterior hypothalamus and the dorsal PAG are nodal structures in the mediation of the CRDR and that cells in posterior hypothalamus, dorsal PAG and ventral PAG make monosynaptic connections with the RVLM.

Role of auditory cortex in the acquisition of differential heart rate conditioning.

Previous findings from our laboratory indicate that lesions of the auditory cortex disrupt the retention of differentially conditioned bradycardiac responses to tonal stimuli in rabbits. In the present experiment, the effect of lesions of the auditory cortex on the acquisition of differential bradycardiac conditioning was examined. The effect of lesions in the auditory cortex were compared to the effect produced by control lesions in the visual cortex. After 7 days of recovery, animals received 7 days of differential Pavlovian bradycardiac conditioning in which one tone (CS+) was paired with the unconditioned stimulus, and another tone (CS-) was never paired with the unconditioned stimulus. All animals demonstrated differential conditioning during the first 3 days of conditioning. On days 4-7, however, auditory cortex lesioned animals did not exhibit significant differential heart rate (HR) conditioning, whereas control animals with lesions in the visual cortex showed no loss of conditioning during this period. The loss of differential conditioning in animals with lesions in the auditory cortex appears to be due to an increase in the magnitude of the response to the CS-. These data support the hypothesis that the auditory cortex serves to inhibit the response to the CS- in differential conditioning of bradycardia to acoustic stimuli, and that the inhibition may be mediated by a descending corticothalamic or corticolimbic pathway.