Total parenteral nutrition The state of the art
William C. Heird, M.D.,* and Robert W. Winters, M.D.,** New York, N. Y.
THE FIRST d o c u m e n t e d r e p o r t of success o f total p a r e n t e r a l n u t r i t i o n in a n infant was p u b l i s h e d by Helfrick and Abelson 1 in 1944, Their patient was a 5too-old i n f a n t with e x t r e m e m a r a s m u s who had reached the point at which "all observers agreed that he would not survive more than a day or two at most." Using infusions of a mixture of 50% glucose and 10% casein hydrolysate in saline followed by a homogenized olive o i l - l e c i t h i n e m u l s i o n , these i n v e s t i g a t o r s were able to deliver 130 calories in 150 ml/kg/day via an ankle vein.-Despite thrombophlebitis, total parenteral nutrition was provided for 5 days. Toward the end of this period "the fat pads of the cheeks had returned, the ribs were less prominent, the general nutritional status was much improved, and the (patient's) former expression of dire misery was gone." This early d e m o n s t r a t i o n of the efficacy of total p a r e n t e r a l n u t r i t i o n offered great p r o m i s e but also pointed out some important problems. During the next 20 or so years many tried to duplicate this impressive p e r f o r m a n c e , u s u a l l y w i t h o u t success. T h e failures could be ascribed to one }3f two general reasons: either
From the Department of Pediatrics, Columbia University College of Physicians andSurgeons, and the Babies Hospital, Columbia-Presbyterian Medical Center. This paper is based upon an invited "State of the Art Address '" which was delivered on May 3, 1974, at the meeting of the Society for Pediatric Research, Washington, D. C. and sponsored by TIlEJOURNALOF PEDlATRlcsEducational Foundation. The original research of the authors is supported by grants from the National lnstitutes of Health (HD-08434 and 5 MO1RR-00645). *Reprint address:Department OfPediatrics, Columbia University College of Physicians and Surgeons, 630 IV. 168thSt., New York, N. Y. 10032. **CareerScientist of the Health Research Councilof the City of New York(1-309).
VoL 86, No. 2, pp. 2-16
the strenuous performance of frequent changing of peripheral sites proved too much or insufficient calories were provided to allow the administered amino acids to be anabolized. In 1966 Dudrick and his co-workers 2 provided a major breakthrough which made total parenteral nutrition a practical reality. Dudrick devised a method by which a c a t h e t e r could be i m p l a n t e d and m a i n t a i n e d in t h e superior vena cava for long periods of time. Since the blood flow in this central vein is high, Dudrick reasoned Abbreviations Used lean body mass TBW: total body water PUFA: polyunsaturated fatty acids EFA: essential fatty acid LBM:
t h a t t h e n e c e s s a r i l y h y p e r o s m o l a r n u t r i t i v e fluid, if given by a slow continuous infusion, would be diluted, thus avoiding damage to the vein. Using this delivery system and a hypertonic glucose-protein hydrolysate fluid to Which electrolytes, minerals, and vitamins were added, Dudrick and associates 3 showed that adequate growth and development could be achieved, first in Beagle puppies and subsequently in an infant with a shortened gastrointestinal tract. 4 These dramatic demOnstrations p r o v i d e d the s t i m u l u s for t h e c u r r e n t widespread use of total parenteral nutrition in pediatric patient s . T h e general s u b j e c t of p a r e n t e r a l n u t r i t i o n bears upon almost all subspecialties of pediatrics. Because of these complexities we have chosen to limit this paper to a critical and selected survey of results of clinical studies, representing our data as well as that of others. F u r t h e r m o r e , we will concentrate only on total parenteral nutrition, a term which we define as provision of all nutrients solely by the parenteral route.
Volume 86 Number 1 C O M P O S I T I O N OF F L U I D S FOR TOTAL PARENTERAL NUTRITION The first requirement for total parenteral nutrition is a source of nitrogen containing both essential and nonessential ~imino acids. Two general types of nitrogen sources are available: hydrolysates of either fibrin or casein or a mixture of pure crystalline amino acids. In passing, it should be noted that only 40 to 50% of the total nitrogen of the hydrolysates consists of amino acids in the free form, the remainder being peptides, s the precise metabolic fate of which is unknown. This latter fact, coupled with the well-known variation in composition of the hydrolysates, led to the advent of crystalline amino acid mixtures. Currently only one of t h e s e , F r e A m i n e , is available c o m m e r c i a l l y in this country. Others are in various phases of experimental development. The general range of intake from either of these nitrogen sources varies from 2.5 to 4.0 gmlkg/ day in infants. We prefer the lower figure since we have shown it to give positive nitrogen balance and weight gain equivalent to the higher figure but with less risk of azotemia.6. 7 Although both the hydrolysates and the current crystalline amino acid mixtures produce growth and development under appropriate conditions, neither can be considered ideal for parenteral nutrition in infants (see below). A second requirement is the provision of sufficient nonprotein calories to meet full caloric expenditure. In the United States, glucose or a mixture of glucose and fructose is used most commonly. In other countries, various intravenous fat preparations, such as Intralipid or Lipofundin-S, have been used along with monosaccharities and polyols. Ethanol has also been used in conjunction with other caloric sources, 8 but in general most investigators have shied away from it, particularly in view of the relative unpredictability of blood alcohol levels in infants 9 and the likelihood of hepatotoxicity? ~ We use glucose exclusively. The amount required in an infant is 25 to 30 gm/kg/day. ~1 Other n u t r i t i v e r e q u i r e m e n t s s u c h as e l e c t r o l y t e s , minerals, and vitamins are provided by various additives to the infusate. We generally give usual maintenance requirements for electrolytes. 12 In the case of minerals and vitamins there is no precise knowledge of parenteral requirements. In some of these instances, the parenteral requirement is likely to be substantially different from the oral one. Clearly, much additional research must be done in these areas. Finally, note should be taken of the need for essential fatty acids and for trace minerals. Again, there is no precise information on parenteral requirements, but based upon a large body of background evidence, both are
Total parenteral nutrition
Table I. Usual composition of infusate
Constituent Nitrogen source Glucose NaCI KH2PO 4 Ca gluconate MgSO4 MV1 Vitamin B12 Folic acid Vitamin K 1 Total volume
A mount 2.5 gm/kg/day 25-30 gm/kg/day 3-4 mM/kg/day 2-3 mM/kg/day* 0.25 mM/kg/day (0.5 mEq/kg/day) 0.125 mM/kg/day (0.25 mEq/kg/day) 1 ml/day 50/.tg/day 50-75/xg/day 250-500/xg/day 130 ml/kg/day
*KH2PO4 should be limited to 2 mM/kg/day; additional potassium should be providedas KCI. likely to be required. 13,14Presently, in the United States, conventional total parenteral nutrition mixtures contain neither essential fatty acids nor trace minerals. Periodic blood or plasma transfusions have been suggested as a means of meeting these requirements, but these methods are probably grossly inadequate, at least as regards essential fatty acids, ts Trace minerals may be present as contaminants of the infusate but in amounts which are likely to be far below any reasonable estimate of the requirements. ~6 The usual composition of the nutritive fluid which we use is listed in Table I. It is obvious that such a fluid is a chemically complex mixture and that many types of interactions among Constituents may occur (see Stegink and associatest7). Fig. 1 shows the approximate separate o s m o l a r c o n t r i b u t i o n of the e l e c t r o l y t e s and minerals, the amino acids, and glucose of a typical fatfree total parenteral nutrition fluid. The total osmolality (about 1,740 mOsm/kg of water) is nearly six times the isotonic value of 300 mOsm/kg of water and therefore slow, continuous infusion in a central vein is mandatory if damage to the vein is to be avoided. These fluids are a l s o e x c e l l e n t c u l t u r e m e d i a for c e r t a i n m i c r o o r g a nisms, 18-2~and for this reason meticulous aseptic technique in mixing the fluid is a most important precaution in minimizing the risk of sepsis. It is well to bear in mind that the composition of the infusate can by no means be uniform from patient to patient. Glucose must be started at low concentrations and slowly increased; electrolytes and/or minerals may have to be altered frequently in accordance with clinical and/or chemical feedback information. Achievement of such flexibility requires a team approach to total parenteral nutrition (see below).
Heird and Winters
The Journal of Pediatrics January 1975
50o 3 o o - -.
~Rx'~yv J'~' '~'' ~
Electrolytes Fig. 1. Osmolality of the usual infusate whose composition is shown in Table I. Such a fluid provides about 110 Cal/kg/day in a volume of 125 ml/kg/day.
Duration d 60 -
Avg.Wt. Gain g/kg/d 20 5
"7 0.20 O, IO
Fig. 2, Duration of total parenteral nutrition, average daily gain in body weight, and average daily nitrogen balance in surgical patients on total parenteral nutrition receiving at least 100 Cal/ kg/day.
TOTAL PARENTERAL NUTRITION IN SURGICAL PATIENTS T h e widest use o f total p a r e n t e r a l n u t r i t i o n in pediatric patients has been in patients, particularly infants, who h a v e surgical d i s o r d e r s o f t h e gastrointestinal tract. Clinical and metabolic results. From our r a t h e r large experience with these patients, we have chosen a group of 21 on whom we have shOrt-term metabolic data as well as long-term follow-up information. 21 These patients were principally newborn infants with anomalies of the gastrointestinal tract, the most dramatic of which were those requiring major resections of small bowel. In all of these patients, t h e c o m m o n denominator was an inability to tolerate full enteral findings. A summary of data obtained during 21 periods of total parenteral nutrition in these infants after achievem e n t of a daily caloric intake of greater than 100 Cal/kg is shown in Fig. 2. As we have pointed out earlier, ]1 it is likely that the wide variation is both weight gain and ni-
trogen balance is accounted for by two factors, v&., the previous nutritional status of the infant and the degree of ongoing stress. Clinical follow-up of these 21 patients, at least 1 yr after total parenteral nutrition, revealed that t5 were well and had normal gastrointestinal function as indicated by growth performance, ability to ingest a normal diet of age, and normal bowel habits. Two subjects continued to require special diets. Four of the infants died, all of causes unrelated to total parenteral nutrition. These results are typical of those being obtained by others, all of which show a dramatic positive outcome in a group of patients who, prior to the advent o f total parenteral nutrition, would certainly have had a very high m o r t a l i t y rate. Thus, given a p p r o p r i a t e p a t i e n t s e l e c t i o n , total p a r e n t e r a l n u t r i t i o n can t r u l y be regarded as life-saving in surgical patients. T h e r e are two a d d i t i o n a l c o n s i d e r a t i o n s which, although they derive specifically from published data in surgical patients, have general relevance to total parenteral nutrition in infants. Importance of caloric intake. A s u m m a r y of data from 17 surgical infants reported by Coran22is shown in Fig. 3. Intralipid and glucose were used as the caloric sources, with the daily caloric intake averaging 88 Cal/ kg, all given by peripheral vein. These data are compared to data from 14 of our surgical patients 21 who received 110 to 120 Cal/kg/day from fat-free infusates delivered by central vein. The two groups were matched as closely as possible for age, duration of total parenteral nutrition, initial body weight, and diagnosis. Despite these similarities, there was a highly significant difference in daily weight gain between the two groups in that Coran's infants averaged only 8.7 gm/kg, whereas ours averaged 12.6 gm/kg. It seems likely that increased Caloric intake is responsible for this difference. One i m p l i c a t i o n c o n c e r n i n g I n t r a l i p i d as a caloric source has been that, with central venous delivery, its attendant risks can be avoided. This conclusion m a y be premature. If a caloric intake (and therefore a weight gain) comparable to the conventional glucose fat-free regimen is provided, the final infusate is quite hypertonic. This point is illustrated in Fig. 4, which compares t h e final o s m o l a l i t y o f isocaloric i n f u s a t e s w i t h a n d without fat, the former delivering the highest recomm e n d e d level of fat, i.e., 4 gm/kg/day. At a daily caloric intake of 110 Cal/kg in a volume of 125 ml/kg, the fatcontaining infusate would have an osmolality of about 1250 mOsm/kg of water (about four times the isotonic value) and probably would damage peripheral veins. Thus, use of this mixture, which should give better growth than that observed by Coran, 22 would require
Volume 86 Number 1
Total parenteral nutrition
DURATION For 0
(doYsl t 80
WEIGHT GAIN (g/kg/d)
- - - ~0- -
Fig. 3. Comparison of duration Of total parenteral nutrition, average gain in body weight, and average daily nitrogen balance in a group of surgical patients22receiving intravenous fat and glucose totalling 88 Cal/kg/day (open circles) with a comparable group 21 receiving more than 100 Cal/kg/day from glucose alone.
% GLUCOSE + FAT
1500 A~,x~xxxx /v-vx,,x~ r -.-v-~--~v x•
Fig. 4. Comparison of osmolality of equicaloric nutrient mixtures providing 110 Cal/kg/day in a volume of 125 ml/kg/day. The left bar shows the osmolality of the mixture with glucose as the sole source of calories; the right bar shows the final osmolality delivered from infusates containing glucose and an intravenous fat emulsion (Intralipid) in which fat is administered at 4 g/kg/day.
either an athletically inclined house staff, willing and able to change peripheral veins frequently, or continued use of the central venous route. To be sure, less highly c o n c e n t r a t e d f a t - g l u c o s e - a m i n o acid i n f u s a t e s are give n peripherally in conjunction with some oral caloric intake. But when long-term, total parenteral nutrition is necessary, we feel that central venous delivery is required if optimal results are to be obtained. Interpretation of weight gain and nitrogen balance. The second relevant point has to do directly with the interpretation of weight gain and of nitrogen balance data in infants receiving total parenteral nutrition and indirectly with the quality of growth which occurs in infants being fed under these conditions. Fig. 5 is presented as background for this point, It shows a highly simplified view of the major components of body composition, i.e., fat, water, and protein. LBM is regarded as the TBW p l u s b o d y protein. It is im-
Fig. 5. Highly simplified view of body composition showing fat, water, and protein as the sole components. LBM in this diagram is defined as the sum of TBW and body protein. portant to note that body fat is essentially anhydrous; thus there is a reciprocal relationship between total body water and fat. This fact is especially important when considering the body composition of infants, as shown in Fig. 6. It is clear from this figure that, during the latter part of gestation, body fat increases considerably at the expense of TBW while body protein remains reasonably constant. Fig. 7 is based upon this background. As shown in the figure, LBM is related approximately to the nitrogen content of the body through a factor, f. The curve, adapted from Widdowson's 23data on the composition of human fetuses, shows the approximate variation of the value for this factor in infants of various weights and gestational ages, after an assumed 10% loss of TBW as an immediate postnatal diuresis. Thus, for 1.0 kg infants, f has a value of about 62 gm of LBM per gram of n i t r o g e n , whereas, for i n f a n t s weighing 2.5 kg or greater, the corresponding value is about 42 gm of LBM per gram of nitrogen. This fall occurs, of course, because of increasing fat deposition and the reciprocal
Heird and Winters
The Journal of Pediatrics January 1975
,i i i i i i i i i i i i i i i i i liiiiiiiiiiiiii (%l
Fat free wet weight
. . . . . . . . . . . . . . . . . . . . . . .
Fig. 6. Changes in body composition as a function of fetal weight and gestational age (data of Widdowson23). Table II. Inconsistencies between measured gain in body weight and nitrogen balance and calculated deposition of LBM during total parenterai nutrition
Glucose-ethanol-amino acidst Glucose-fat-amino acids]" Glucose-amino acids]" Glucose-amino acids$
Avg. caloric intake (Cal/kg/day)
Avg. A body weight
75 96 104 115
Avg. A LBM* (gm/kg/day)
Avg. A body weight minus ALBM (gm/kg/day)
+3.8 +7.5 +8.3 +11.8
+9.7 +9.2 +12.6 +8.0
--5.9 --1.7 -4.3 +3.8
*Computed from nitrogen balance using a value for f o f 42 gm LBM/gm N. t Data of Asch and associates.24 $ Data of Heird and associates.21
decrease in TBW in the face of a relatively constant proportion of body protein (Fig. 6). These background considerations can be used to a p proximate the amount of LBM expected for a given degree of positive nitrogen balance. Thus, in a 2.5 kg infant, assuming steady-state conditions, 1 gm of positive nitrogen balance would be roughly equivalent to the deposition of 42 gm of LBM (at this age, f = 42). Table II shows how this relationship can be used in critical evaluation of data on nitrogen balance and growth. Data from a recently published abstract of a study 24 designed to test the difference between three total parenteral nutrition regimens in surgical infants are compared with data from our surgical patients. 21 With the first regimen, total daily weight gain was 3.8 gm/kg, but the expected daily LBM deposition, calculated from the nitrogen balance (using a value for f o f 42 gm of LBM per gram of positive nitrogen balance) was 9.7 gm/kg. Thus, the observed total weight gain is far short of the expected increase in LBM resulting in a physiologically inexplicable negative value for the weight gain not accounted for by LBM. Similar discrepancies, although to a lesser extent, appear with the second and third sets of data. It seems most likely that
methodological errors in nitrogen balance are operative here, especially in view of the fact that our patients (see the last line of Table II) who received a caloric intake nearly comparable to the third regimen had an observed daily weight gain of 11.8 gm/kg with a deposition of 8.0 gm/kg/day of LBM, leaving a positive weight gain unaccounted for by LBM of 3.8 gm/kg/day. This additional weight is likely to be partly or completely fat. Certainly all of it cannot be extracellular fluid. If it were, the infants would have been grossly edematous at the end of the average 27 day period of total parenteral nutrition; in fact, little or no edema was observed. It is widely recognized that the technique of nitrogen balance is filled with errors, nearly all in the positive direction. 25 Thus in the absence of significant changes in hydration, sustained weight gain is probably as good as, or better than, nitrogen balance as an index of growth. It is certainly easier to measure. TOTAL PARENTERAL NUTRITION IN PATIENTS WITH CHRONIC DIARRHEA Another group of patients in whom total parenteral nutrition appears to be useful are those infants with chronic intractable diarrhea. This problem, because of
Volume 86 Number 1
its high mortality rate, 26 has vexed every experienced pediatrician. Such infants usually present with moderate or severe malnutrition secondary to the protracted d i a r r h e a . A d m i s s i o n weight is often less t h a n birth weight. Extensive diagnostic studies to reveal the cause of the diarrhea (i.e., stool and urine cultures, roentg e n o g r a p h i c studies, sweat test, i m m u n o g l o b u l i n s , cathecholamines) are negative. Usually the diarrhea is unresponsive to any type of formula feeding. Clinical results. Keating has had extensive experi, e n c e in t r e a t i n g t h e s e infants with total p a r e n t e r a l nutrition.27, 28 He studied 16 such patients, all of whom fulfilled the above criteria. The average age of the patients was 3 wk (range, 2.5-12 wk) and the average period of total parenteral nutrition was 28 days (range, 14-42 days). Fifteen of the 16 infants survived; after at least 1 year, all had normal gastrointestinal function, with the exception of six (all Black) who remained lactose i n t o l e r a n t . Our e x p e r i e n c e as well as that o f others29, 30 is in a g r e e m e n t with this over-all result, which represents a dramatic improvement in the outcome of infants with this_disorder compared with the expected outcome prior to availability of total parenteral nutrition. 26 These observations raise some interesting questions about the nature of the disease or diseases which masquerade under the name of chronic intractable diarrhea. Avery and associates 26 discussed the evidence and adv a n c e d t h e h y p o t h e s i s , d e p i c t e d in a h i g h l y s c h e matized form in Fig. 8, that chronic diarrhea leads to a state of malnutrition through malabsorption of n u t r i e n t s as well as failure to r e p l a c e t h e s e losses through oral intake. Once malnutrition is established, because of deficiency of either protein, calories, vitamins, and/or ions, it in turn effects adverse changes in gastrointestinal mucosa, flora, reflexes, and/or motility such as to perpetuate the diarrheal state. In such infants, total parenteral nutrition produces "bowel rest" and seems to break this cycle, presumably by altering the gastrointestinal structure and function so that nutrients given by mouth can again be absorbed and utilized. In this connection, Greene and associates 3~ have shown that total parenteral nutrition produces a marked change in intestinal morphology in that the crypt:villus ratio increases from about 1:1 before total parenteral nutrition to about 1!4 after total parenteral n u t r i t i o n . This finding, as well as o t h e r histologic changes, suggests that the mucosal absorptive surface is greatly increased by total parenteral nutrition. Regardless of mechanism, the over-all clinical results of total parenteral nutrition in these infants rank with those obt a i n e d in Selected surgical patients as a l i f e - s a v i n g measures.
Total parenteral nutrition
~ f x NITROGEN
1.0 1.5 2.0 2.5 3.O 3.5 L I I I 26 31 33 35
f___, 70 (kg) /
Fig. 7. Expected relationship between the factor (f) relating nitrogen balance to body weight and gestational age. The curve was derived from the data shown in Fig. 6 and corrected for an assumed 10% loss of TBW by early postnatal diuresis (see text for usefulness off). Table III. Comparison of conventional nutritional management with TPN in very low-birth weight infants* Con yen tional
No. of infants 23 14 Birth weight (gm • 943 • 217 863 +143 Initial weight loss (% body weight +SD) 13.5 _+ 5.3 11.2 • 4.8 Average daily weight gain (gm _+ SD) 20.0 • 3.2 21.1 • 2.6 Time to regain body weight (days • SD) 18.2 • 8.4 8.9 • 5.1(10)~ Time to achieve discharge weight 76.4 • 19.9 79.8 • 11.1 Age of nonsurvivors (days • SD) 7.3 • 4.5 27.3 • 23.85 Mortality rate (%) 56.5 42.9t * Data of Heird and Winters. 6 t p=<0.01. Sp = < 0.1 b u t > 0.05.
TOTAL PARENTERAL NUTRITION IN VERY LOW-BIRTH-WEIGHT INFANTS The third category of patients in whom total parenteral nutrition has been studied are the infants with very low birth weight. The problems and risks of establishing a completely adequate enteral intake in these infants are well known. In addition, there is the added c o n c e r n , based p r i n c i p a l l y on a n i m a l studies, that malnutrition at some critical point of brain growth may lead to nonrecoupable losses of brain cells. 31
Heird and Winters
The Journal of Pediao'ics January 1975
I MALN UTR ITION I I(deficiencies of I I protein,colories,,I [ vitamins ~ ions) I Fig. 8. Schematized view of the interrelationships between chronic diarrhea and malnutrition (see text).
Time to achieve initial weight (doys)
Time to achieve >lOOCal/kg/d (doys)
Fig. 9. Relationship between the time to regain initial body weight and the time to achieve an intake of greater than 100 Cal/kg/day in low-birth-weight infants on total parenteral nutrition.
Avg. N Bol. g/kg/d
as[ ,t 151=---=----
Fig. 10. Duration of total parenteral nutrition, average daily gain in body weight, andaverage daily nitrogen balance in very low-birth-weight infants on total parenteral nutrition receiving at least 100 Cal/kg/day.
In contrast to the two groups Of patients discussed above, we regard total parenteral nutrition in this group as a strictly investigational procedure--certainly not an accepted part of routine neonatal care. C l i n i c a l r e s u l t s . Our group has a s c e r t a i n e d the feasibility and safety of this method of feeding in a preliminary uncontrolled study of total parenteral nutrition in 14 consecutive low-birth-weight infants. The mean birth weight of these infants was 863 gm (range, 720-1,150 gm). In eight infants, total parenteral nutrition was started within the first 48 hr of life using an umbilical venous catheter (the tip of which was above the diaphragm) for 2 to 3 days and then changing to a conventional superior vena cava catheter. In the six others total parenteral nutrition (via a superior vena cava catheter) was started later, after unsuccessful attempts at conventional feedings. The average period of total parenteral nutrition was 17.7 days (range, 5-24 days). In 10 of these infants, a caloric intake of greater than 100 Cal/kg/day was achieved over an average period of 5.6 days. The average time required to regain initial body weight was 7.8 days, the principal determinant of this time being the time needed to achieve a full caloric intake (Fig. 9). The data in Fig. 9 also demonstrate that there is considerable variability in the time required to achieve-a~full caloric intake. This variability is due to the inability of these infants to adapt to the high glucose loads. More recently we have given small amounts of insulin (0.25-0.50 U), which seems to be a useful maneuver in speeding this adaptive response, provided very close monitoring is carried out. Data from eight infants who received total parenteral nutrition providing a full caloric intake for an,average of 15.5 days are shown in Fig. 10: During this time the average daily weight gain was 15.2 gm/kg. The average daily nitrogen balance was 0.22 gm/kg and was consistent with the weight gain. The weight gain unaccounted for by LBM, presumably fat, varied from 3 to 4 gm/kg/ day, depending upon the precise figure chosen for the factor relating LBM to nitrogen balance (see above). Six of the infants died: five of respiratory insufficiency and one of sepsis. Eight infants survived; after 17 to 39 mo, three are normal, neurologically, two are "suspect," and two are definitely abnormal (in one of these the abnormality is almost certainly secondary to a s u b sequent unrelated illness). Comment. The results of this study are conclusive in demonstrating that total parenteral nutrition can produce satisfactory growth and positive nitrogen balance without undue risk. They are certainly inconclusive in respect to whether the short- or long-term outcome was
Volume 86 Number 1
G-I TRACT I
Total parenteral nutrition
Urea Essential Nitrogen Protein Synthesis
....._ Non-essential Nitrogen
-~...,.N e w C e l l Growth
Fig. 11. Schematized basis of nutritional therapy in renal failure. (From Winters, R. W., editor: The body fluids in pediatrics, Boston, 1973, Little, Brown & Company.)
favorably affected. One fact which emerged from the study is that experience with a large n u m b e r of babies will be needed if one wishes to test accurately the superiority of total parenteral nutrition over conventional management. Table III shows data which support this statement, It compares obvious short-term variables in the 14 infants receiving total parenteral nutrition with those of 23 infants of similar size who were managed conventionally over the 2 yr period immediately prior to use of total parenteral nutrition. The mortality rate among those who received total parenteral nutrition was 43%, whereas in the conventionally managed infants it was 57%. If these differences were to prevail in a controlled study, at least 75 infants would be needed to show 5% significance and twice that many to show 1% significance. Clearly, if any such study is ever mounted, it will require the participation of a n u m b e r of Neonatal Centers. The short-term and the long-term outlook for very low-birth-weight infants is obviously a multifactorial problem of which early adequate nutrition is only one factor. Until this factor can be isolated and studied in a controlled fashion, total parenteral nutrition cannot be regarded as having anY proved place in the managem e n t of these infants. On the other hand, we have shown that, with adequate precautions, it can be used without u n d u e risk; thus, further cautious exploration seems justified. TOTAL PARENTERAL NUTRITION ACUTE RENAL FAILURE
Other patients who may be benefited by a special variation of total parenteral nutrition are those with acute renal failure. To our knowledge, there has been no s y s t e m a t i c study o f this type o f total p a r e n t e r a l
nutrition in pediatric patients. The technique has been applied to adults and is worthy of discussion because of its potential applicability to pediatrics. Rationale and clinical results. The theoretical basis for n u t r i t i o n a l t h e r a p y in a c u t e r e n a l f a i l u r e is schematized in Fig. 1 1. If the essential amino acids plus a full caloric intake are supplied to an azotemic patient, the nonessential nitrogen needed for growth can be derived from urea through the transintestinal conversion of urea to ammonia. Under these conditions, net synthesis* of new tissue could occur and should be accompanied by a concomitant uptake of potassium and phosphate from the extracellular fluid. Thus the entire process can be e n v i s i o n e d as a r e c y c l i n g of urea, potassium, and phosphate from the extracellular to the intracellular fluid. A recycling of the "uremic toxins" might also occur provided there are no irreversible metabolic steps in their production.* Abel and associates 33 have published a well-designed, double-blind controlled clinical study comparing the effects of providing e s s e n t i a l a m i n o acids and full caloric intake from glucose with provision of glucose alone in adults with acute renal failure, usually of i s c h e m i c origin. Survival from the episode of acute renal failure as well as over-all survival was significantly b e t t e r in those who r e c e i v e d essential a m i n o acids, despite the greater p r e d o m i n a n c e of males in this group, and therefore the higher risk of death. Furthermore, there were differences in the patterns of change of blood urea nitrogen and plasma creatinine. In the study group, both rose for the first 3 days and then fell; in the control group, both rose steadily over the first week. The changes in blood urea concentration were expected, owing to the recycling of urea. The creatinine changes, however, were unexpected, since this substance probably cannot be recycled to any major extent. This finding strongly suggests that recovery of the necrotic renal tubular epithelium was more rapid in the study group than in the control group, an exciting possibility that should be examined more closely, probably in an animal model. Comment. Based upon these studies the probable u s e f u l n e s s of this m o d i f i e d f o r m o f total p a r e n t e r a l nutrition in acute renal failure can be defined: (1) The evidence of Abel and co-workers 33,34suggests that dialysis can be reduced in frequency or avoided entirely. *The s c h e m e s h o w n in Fig. 11 depicts the n e t result (i.e., the difference between anabolism a n d catabolism) of the over-all process. It is important to know w h e t h e r total parenteral nutrition u n d e r these conditions effects a reduction in the catabolic process or accelerates the anabolic process, since some of the steps of catabolism may be irreversib l e - e . g . , the production of sulfuric acid from catabolism of sulfur-containing amino acids.
He#'d and Winters
(2) The chemical composition of the body fluids can be improved, Indeed, in some of the patients given essential a m i n o acids plus glucose, a d m i n i s t r a t i o n o f exogenous potassium a n d phosphate was necessary to prevent or correct hypokalernia and hypophosphatemia. (3) If renal function returns more rapidly, as seems likely, m a n y of the c o m p l i c a t i o n s s e c o n d a r y to the uremic state can be minimized or avoided. TOTAL PARENTERAL NUTRITION IN INFLAMMATORY BOWEL DISEASE Still another group of pediatric patients in whom total parenteral nutrition may be useful are those with inflammatory bowel disease. Cohen and associates 35 have treated 13 such patients, p r e d o m i n a n t l y in the t e e n a g e group, to a s c e r t a i n w h e t h e r c o m p l e t e " b o w e l r e s t " p r o d u c e d by total parenteral nutrition would improve the underlying disease. T h e s e p a t i e n t s had had p r e v i o u s u n s u c c e s s f u l trials of surgery, corticosteroids, and/or azulfidine. They had a high incidence of perianal disease, notably fistulas and/or abscesses. Nutritional debilitation was moderate or severe. Total parenteral nutrition was administered without any other therapy to 1 1 patients for an average of 5 weeks. Of these, six had a complete remission for 12 to 18 montfis; in another three, partial remission for up to 12 months was induced. The nutritional status of these patients was much improved, and their perianal fistulas closed spontaneously. The two remaining patients died of their disease. These investigators feel that total.parenteral nutrition has a definite role in the m a n a g e m e n t of these patients, and their results would seem to bear out this assumption. Similar results have been obtained in adults) 6, 37 Although not a panacea, total parenteral nutrition certainly seems deserving of a trial in this group of troub l e s o m e disorders, p a r t i c u l a r l y if surgery, corticosteroids, and/or azulfidine can be avoided, or if these latter agents could be used in smaller doses for briefer periods of time to manage mild relapses after a course of total parenteral nutrition. METABOLIC COMPLICATIONS OF TOTAL PARENTERAL NUTRITION General. M e t a b o l i c c o m p l i c a t i o n s of a l m o s t e v e r y conceivable type have been associated with total parenteral nutrition. Hyperglycemia, with attendant osmotic diuresis and secondary changes in water and electrolyte metabolism, is always a threat and may, if undetected, lead to e x t r e m e d e g r e e s o f h y p e r o s m o l a r i t y . 38 Hypoglycemia may also occur with sudden cessation of the glucose l o a d - - e . g . , w h e n the c a t h e t e r b e c o m e s dis-
The Journal of Pediatrics January 1975
lodged or when total parenteral nutrition is abruptly stopped for some other reason. Disorders of electrolyte and mineral metabolism occur when either too much or too little of these substances is given with respect to the needs of the individual patient. Hypophosphatemia may be a particularly important complication. Travis and associates 39 have shown that it is accompanied by changes in the organic p h o s p h a t e c o m p o u n d s o f the e r y t h r o c y t e which, in turn, affect the oxygen dissociation curve. F u r t h e r m o r e , Jacob and associates 4~ have shown that hypophosphatemia severely depresses adenine nucleotides in erythrocytes, platelets, and leukocytes, resulting in their shortened survival, diminished function (including an impairment of chemotaxis), and/or abnormal hydration. In formulating the electrolyte and mineral intake, the presence of any of these substances in the nitrogen source must be taken into account. For example, casein hydrolysate contains appreciable amounts of phosphate (casein is a phosphoprotein). Generally, the crystalline amino acid mixtures have little added electrolyte or minerals; however, an amino acid mixture ( F r e A m i n e II, McGaw Laboratories) which contains some added phosphate (although probably not enough for infants) will be marketed soon. Disorders of acid-base equilibrium are u n c o m m o n in patients who receive hydrolysates; when they occur they can usually be explained by other obvious causes. In our experience, however, hyperchloremic metabolic acidosis occurs consistently when certain of the crystalline amino acid mixtures are used. Azotemia occurs consistently in small infants who receive 4 gm/kg/day of nitrogen; in our experience it is u n c o m m o n with the recommended intake of 2.5 gm/ kg/day (see above).' H y p e r a m m o n e m i a is a n o t h e r c o m p l i c a t i o n which we al have studied. B e c a u s e the p r o t e i n h y d r o l y s a t e s contain appreciable amounts of preformed ammonia a n d b e c a u s e J o h n s o n and associates 42 d o c u m e n t e d moderate but asymptomatic h y p e r a m m o n e m i a in infants given hydrolysates, we switched to a crystalline amino acid mixture ( F r e A m i n e ) , which contains no preformed a m m o n i a , as the nitrogen source for total parenteral nutrition. Nonetheless, four cases of severe symptomatic hyperammonemia were encountered; all were surgical patients who had received total parenteral nutrition for more than 3 weeks. The hyperammonernia rapidly responded t o arginine and/or ornithine infusions and could be prevented by supplementation of F r e A m i n e with 0.5-1.0 mM/kg/day of arginine, al On the basis of these observations, which have been confirmed.
Volume86 Number 1
Total parenteral nutrition
PLASMA rci-] (mEq/I)
lie II _,IL__~
tO5 ~ -
IOC 9 95
OAMINOSOL ,& FREAMINE 0 NEOAMINOSOL
Fig. 12. Blood acid-base changes seen in infants receiving crystalline amino acid mixtures (NeoAminosol or FreAmine) or fibrin hydrolysate (Aminosol). by others, 43,44we believe that this particular amino acid solution is relatively deficient in arginine, particularly for long-term total parenteral nutrition. Abnormal plasma aminograms have been noted by others 45-48as well as by US. 49 In general, the pattern of the plasma aminograms tends to reflect the amino acid composition of the infusate. Other aspects of this problem are discussed below. Hypovitaminoses as well as hypervitaminoses, es~ pecially involving vitamins A and D, have been recorded. Decimal errors in the addition of the vitamins have been implicated as causes. However, since we have no precise knowledge of parenteral requirements of the two fat-soluble, potentially toxic vitamins, A and D, it may be, as hinted by a recent report, 5~ that subclinical hypervitaminosis A and D is being produced with the usual vitamin dosage. Essentially fatty acid deficiency has been commented on recently by a n u m b e r of investigators. 13, 51-53 Likewise, deficiencies of trace minerals, such as zinc and copper, have been reported 16, 54 and probably will be seen more often when they are looked for in a systematic fashion. Finally, disorders of hepatic function and/or structure are often seen. TM55, 56 There are rather consistent modest rises in levels of serum glutamic oxalacetic transaminase and serum glutamic pyruvic transaminase when total parenteral nutrition is initiated, and these enzyme changes are sometimes associated with hyperbilirubinemia and hepatomegaly. In our experience, these abnormalities are transient despite continuation of the total parenteral nutrition regihaen, u If detected early, many of the metabolic complications can be corrected by changes in the composition of the infusage. To enable early detection, the patients must be monitored extensively, especially in the face of metabolic instability. Using a stringent monitoring schedule, coupled with the ability to rapidly adjust the infusate composition, we have been able to maintain
plasma electrolyte, acid-base, mineral, glucose, and urea values within normal limits in the majority of cases. W h e n abnormalities in these variables do occur, they are usually not severe and are readily corrected by alteration of the composition of the infusate. M e t a b o l i c acidosis. The hyperchloremic metabolic acidosis which occurs with certain of the crystalline amino acid mixtures deserves more detailed discussion because there seems to be considerable confusion as to its precise cause. We showed 57 that infants receiving the fibrin hydrolysate, Aminosol, showed no significant deviation of acid-base status, whereas, those receiving F r e A m i n e developed a mild hyperchloremic metabolic acidosis and those receiving the experimental mixture, NeoAminosol, developed a more severe hyperchloremic metabolic acidosis (see Fig. 12). A variety of sometimes confusing explanations have been offered as causes of this acidosis. All total parenteral nutrition solutions have acid pH values and positive values for titratable acidity. Nonetheless, these have been implicated by some 58 as the cause of the acidosis. In fact, neither of these variables is relevant in the absence of a precise knowledge of the chemical Composition of the specific fluid. Thus, we showed that the protein hydrolysates, which do not produce acidosis, have titratable acidities which are three to four times those of the crystalline mixtures, which do. 57 Even this finding is not necessarily relevant, since the chemical composition and the metabolic fate of the substances being titrated is not precisely known. This point can be illustrated by considering two solutions of acid; 0.1M lactic acid and 0.1M hydrochloric acid. Both have acid pH values and both have identical values for titratable acidity. Infusion of lactic acid into an otherwise normal individual would not cuase a sustained metabolic acidosis, however, since lactic acid is metabolized to carbon dioxide and water. Infusion of an equal amount of hydrochloric acid, on the other hand, would cause a
Heird and Winters
The Journal of Pediatrics January 1975
lion I ;ap I :.:,,....:,:.|
McJ* ~ ::::::::::::::"::::::::: "//'/~,
AMINO ACID MIXTURES
Fig. 13. General types of cation-anion patterns seen in crystalline amino acid mixtures which produce metabolic acidosis and in protein hydrolysates which do not. s u s t a i n e d m e t a b o l i c acidoisis b e c a u s e it is not metabolizable and can be excreted only by the kidney, a process which requires an appreciable interval of time. A n o t h e r possibility, excessi.ve e n d o g e n o u s production of organic and sulfuric acids, was proposed by Chan, 59 who reported equivalently high values in infants on total parenteral nutrition regimens which contain either type of nitrogen source compared to infants who received only glucose, an inappropriate comparison. Furthermore, his study can be faulted on serious methodologic and theoretic grounds. First, Chan s9 found substantially more sulfate excretion than could p o s s i b l y be e x p l a i n e d e v e n if all of the methionine (both D and L forms), the only source of sulfur-containing intake, were catabolized. If the additional sulfate were derived from tissue catabolism, an unlikely anomalous metabolic state must be a s s u m e d - i.e., the patient would be anabolic for nitrogen, potassium, and phosphate but catabolic with respect to sulfur-containing amino acids. An error in the urinary sulfate estimate is a more likely explanation and we 6 have confirmatory evidence of the occurrence of such errors with the method used by Chan. Second, Chan 59ignored the intake of potential acid in the infusate (see b e l o w ) , a serious theoretical error when conducting a balance of net acid where balance is defined as intake minus output. For these reasons, we can dismiss this explanation of the origin of the metabolic acidosis. Other possibilities deal with abnormal losses of base, either in the stool or in the urine. In our studies, loss of base in stool was low and renal net acid excretion was normal or supernormal, 57 thereby excluding both of these possibilities. T h e remaining possibility is that infusates containing crystalline amino acids deliver a load of potential acid not present in the hydrolysates. This possibility was studied by measuring the inorganic cation-anion pat-
tern of each type of infusate. 57Fig. 1 3 shows the general pattern of results obtained. In both fibrin and casein hydrolysates, inorganic cations exceed inorganic anions, resulting in an anion gap which is probably accounted for by glutamate, aspartate, and negatively charged peptides. In effect, all of these substances, upon metabolism, yield bicarbonate. The crystalline amino acid mixtures, especially NeoAminosol, however, have cation gaps which equal their content of arginine and lysine, both of which are present as the hydrochloride salts. These substances are known to generate hydrochloric acid upon metabolism (regardless or whether they are catabolized and anabolized). Furthermore, t h e y are present in sufficient amounts in these mixtures to cause the observed acidosis. The acidosis is not a serious problem. Once it is recognized, it can be prevented or treated by adding lactate instead of chloride to the infusate. Ultimately, it can be prevented by incorporating metabolizeable salts of arginine and lysine (e.g., acetate) rather than hydrochloride salts in amino acid mixtures. F r e A m i n e II, shortly to appear on the market, substantially incorporates these suggestions and should not produce acidosis. Essential fatty acid deficiency. Another metabolic complication that deserves c o m m e n t concerns the increasing n u m b e r of reports of chemical and clinical E F A deficiency during prolonged fat-free total parenteral nutrition. 13, 51-53 Holman]3, 60 has comprehensively reviewed this general subject; the interested reader is referred to these authorative sources for relevant details. Based upon Holman's 13, 60 reviews, certain potentially important effects of E F A deficiency which may be of clinical relevance in the context of total parenteral nutrition can be identified. These are dep i c t e d in Fig. 14. Some of t h e r e l a t i v e l y gross pathophysiologic changes which E F A deficiency may produce include growth failure, lesions of the skin and hair, and a substantial increase in metabolic rate. Characteristic changes occur in the pattern of P U F A in plasma lipids and in the lipid fractions of various tissues, including the brain. 16 These changes presumably underlie t h e pathophysiologic effects. Specifically, E F A ' s and their metabolit es a r e known r o b e i m p o r t a n t components of cell membranes, and abnormalit!es in the red cell m e m b r a n e have been directly documented in E F A deficiency, including cases in which h u m a n subjects received fat-free total parenteral nutrition. 62 Similar changes may underlie the structural and functional abnormalities noted in EFA-deficient mitochondria, which seem to behave in an over-all fashion as if they are oxidatively uncoupled. M e m b r a n e changes c o u l d also be involved irl the abnormal electrocardiogrphic patterns as
Volume 86 Number 1
well as in the abnormalities noted in small bowel and hepatic structure and function. Finally, there is the tantalizing question of the role of E F A deficiency in altering the synthesis of prostaglandins. 63 It seems quite clear that there is a great deal to learn about the physiology and biochemistry of E F A deficiency in conventional fat-free total parenteral nutrition as well as in assessing the effects of providing what is certainly a vast excess (of the order of 10 fold) of P U F A to patients receiving fat emulsions such as I n tralipid. A recent report 64 suggests that both clinical and biochemical correction of E F A deficiency can be effected in patients who receive total parenteral nutrition by the cutaneous application of sunflowerseed oil, a rich source of linoleic acid. These provocative results deserve confirmation. CATHETER-RELATED COMPLICATIONS IN TOTAL PARENTERAL NUTRITION The second category of complications during total parenteral nutrition are those related to the presence of the central venous catheter and to the delivery system. Of these, sepsis is the most important. Although thrombosis a n d d i s l o d g e m e n t of t h e c a t h e t e r o c c u r occasionally, catheter malposition and/or perforation are avoidable with good technique, including roentgenographic confirmation of the site of the catheter tip. In a total of 1025 patient days of total parenteral nutrition, representing 35 patients and 50 catheters, we e n c o u n t e r e d o n l y t h r e e i n s t a n c e s of sepsis. One of these patients died; the other two responded to withdrawal of the catheter and to chemotherapy. In our total experience, we have encountered an occasional thrombosis with s u b s e q u e n t r e c a n a l i z a t i o n of the caval system and occasional dislodgement of the catheter. We have also had one instance of malposition resulting in infusion of the fluid into the pleural space. Our low r e c o r d o f septic c o m p l i c a t i o n s has b e e n equalled or bettered by others.65 Such results certainly do not justify the extensive publicity given to a recent report o f high sepsis rates 66 in the form of accompanying damning editorial comments 67as well as republication in the lay press. 68 Such lay ptiblicity hardly gives confidence to the parents of patients requiring total parenteral nutrition and in our opinion is unjustified. Some investigators have suggested that amphotericin, pushed periodically through the catheter, may reduce septic c o m p l i c a t i o n s from C a n d i d a - - t h e so-called "amphotericin flush. ''69 This practice, in our opinion, is u n n e c e s s a r y . C e r t a i n l y , it is no s u b s t i t u t e for rigid
Total parenteral nutrition
Fig. 14. Some important effects of essential fatty acid deficiency in animals and/or in man (see text). adherence to Listerian surgical principles in the insertion and care of the catheter, in the frequent dressing changes, or in the mixing of the nutritive infusates. 7~ TEAM APPROACH TO TOTAL PARENTERAL NUTRITION As has a l r e a d y b e e n d o c u m e n t e d , the s e e m i n g simplicity of the technique of total parenteral nutrition is deceptive. Its proper execution requires a well-trained and d e d i c a t e d team. T h e t e a m m u s t i n c l u d e both pediatricians and pediatric surgeons, one of whom must be the primary physician. A pharmacist trained in additive techniques is another absolute necessity. Responsive microbiologic and especially microchemical laboratory support is also crucial, as is a nurse specifically assigned to t h e team. At t h e Babies Hospital, total parenteral nutrition is administered only in a setting of an intensive-care unit. The patient is seen in the morning and again in the late afternoon. At the latter time, the monitoring data for the day are reviewed and the primary physician writes the prescription for the next day's infusate. The pharmacist mixes the fluid in the evening, and it is delivered to the ward to begin then or on th e following m o r n i n g . T h e t r i - w e e k l y d r e s s i n g changes are done by the nurse assigned to the team. This approach minimizes complications and maximizes good results. In our opinion institutions unable or unwilling to mount such a team should refer patients to other institutions of proved competence in total parenteral nutrition for infants. UNANSWERED QUESTIONS FUTURE PROSPECTS
General comments. There are a n u m b e r of major u n a n s w e r e d questions c o n c e r n i n g total p a r e n t e r a l
Heird and Winters
The Journal of Pediatrics January 1975
"the greotest" L
"yes or no
Fig. 15. The sine curve followed by many significant medical discoveries.
nutrition, especially in pediatrics. (1) It is clear that better amino acid solutions are needed, a problem to be d i s c u s s e d in m o r e detail below. (2) Better caloric sources are needed. Fat emulsions hold promise, but there are important unanswered questions relative to their use: Does competitive binding between fatty acids and bilirubin occur71? What are the effects of fat emulsions on pulmonary diffusion capacity72? W h a t is the upper metabolic limit for clearance of the infused fat? These and other questions must be answered in infants. Maltose 73 and oligosaccharides deserve further evaluation as possible intravenous caloric sources. (3) Precise definitions of p a r e n t e r a l r e q u i r e m e n t s for E F A ' s , vitamins, and trace minerals are needed, as are safe ways of delivering these substances. (4) The vast array of potential endocrine and metabolic responses under the conditions of total parenteral nutrition in which the g a s t r o i n t e s t i n a l tract and liver are b y p a s s e d n e e d s thorough exploration. (5) There is great need for a more accurate definition of indications, nonindications, and c o n t r a i n d i c a t i o n s for total p a r e n t e r a l n u t r i t i o n in various categories of pediatric patients. Given this list, there are plenty of research problems to go around. Our plea would be that every pediatrician who performs total parenteral nutrition should try not only to improve his patient but also to attempt to answer at least one of these questions in a systematic fashion. "Ideal" amino acid mixtures. The most pressing problem, in our opinion, is that of defining the "ideal" amino acid solution for infants. Such a fluid, obviously, should provide the essential amino acids for infants; these should include histidine, cystine, and probably tryosine 74, 7s An essential:nonessential amino acid ratio of about 1 : 1 is probably desirable, since tissue proteins generally have approximately this ratio. Furthermore, this ratio is generally r e c o m m e n d e d for the oral amino acid r e q u i r e m e n t s for n o r m a l growing i n f a n t s . T h e composition of the nonessential amino acids is also im-
portant. Certainly, the practice of "topping off" the mixture with excessive glycine with attendant marked hyperglycinemia 49 should be avoided. Alanine, proline, serine, and arginine should be provided. There seems to b e an u n w a r r a n t e d fear of i n c l u d i n g g l u t a m a t e and aspartate in the crystalline amino acid mixtures, 76 particularly in the light of the fact that these amino acids have been given for years in the hydrolysates. 5 The a r g i n i n e r e q u i r e m e n t and its role in p r e v e n t i n g hyperammonemia as well as the acidogenic or alkalogenic potential of the mixture have already been touched upon. On the basis of present information, neither of these problems is insurmountable. Finally, it should be obvious that only the L forms of amino acids should be used. One c o m m e r c i a l l y available m i x t u r e p r o v i d e s methionine as a mixture of the D and L forms. It should be no surprise, therefore, that levels of plasma methionine from infants who receive this mixture are quite high.a7, 49 Investigators as well as pharmaceutical companies have an interdependent relationship in the d e v e l o p m e n t and clinical testing of new total parenteral nutrition mixtures. In addition, both parties have a further complex relationship and responsibility with the Food and D r u g A d m i n i s t r a t i o n . Since we as i n v e s t i g a t o r s speak ultimately for the patient, we must assume independent responsibility in finding ways of accelerating F D A approval of new mixtures without lowering standards. CONCLUSION The sine curve that m a n y significant new medical discoveries seem to follow is shown in Fig. 15. Total parenteral nutrition has certainly followed this course. Initially there was uncritical acceptance--total parenteral nutrition was the greatest thing that ever happened! But as increasing septic and metabolic complications were seen, the curve went from an uncritical acc e p t a n c e to an u n c r i t i c a l r e j e c t i o n - - t o t a l p a r e n t e r a l nutrition ~as the worst thing that ever happened!* Currently, as systemati c studies of risks, indications, and nonindications as well as metabolic and clinical results accumulate, we are emerging slowly from this depression. We do not know precisely where the curve will come to rest. Our opinion is that, when the above factors have been accurately and carefully defined, total parenteral nutrition will find a proved and even life-saving place in the nutritional m a n a g e m e n t of selected infants and children. *In t r u t h w e k n o w o f n o o n e w h o actually c u t o f f h i s ear, as is implied in t h e figure. W e are, h o w e v e r , a w a r e o f s o m e n e a r m i s s e s !
Volume 86 Number 1
T o t a l p a r e n t e r a l nutrition
REFERENCES 1. Helfrick FW, and Abelson NM: Intravenous feeding of a complete diet in a child: report of a case, J PEDIATR 25:400, 1944. 2. Dudrick SJ, Vars HM, and Rhoades JE: Growth of puppies receiving all nutritional requirements by vein, Fortschritte der parenteralen Ern~hrung, Symposion der International Society of Parenteral Nutrition, Munich, 1967, Pallas Verlag, p 2. 3. Dudrick SJ, Wilmore DW, Vars HM, and Rhoades JE: Long-term total parenteral nutrition with growth, development, and positive nitrogen balanc e , Surgery 64:134, 1968. 4, Wilmore DW, and Dudrick SJ: Growth and development of an infant receiving all nutrients by vein, J A M A 203:860, 1968. 5. Winters RW, and Hasselmeyer EG, editors: Intravenous nutrition in high risk infants, New York, 1974, John Wiley & Sons, Inc. 6. Heird WC, and Winters RW: Unpublished data. 7. Driscoli JM Jr, Heird WC, Schullinger 1N, Gongaware RD, and Winters RW: Total intravenous alimentation in low-birth-weight infants: A preliminary report, J PEDIATR 81:145, 1972. 8. Benda GI, and Babson SG: Peripheral intravenous alimentation of the small premature infant, J PEDIATR 79:494, 1971. 9. Peden VH, Sammon TJ, and Downey DA: Intravenously induced infantile intoxication with ethanol, J. PEDIATR 83:490, 1973. 10. Rubin E, and Lieber CS: Fatty liver, alcoholic cirrhosis produced by alcohol in primates, N Engl J Med 290:128, 1974. 11. Heird WC, Driscoll JM Jr, Schullinger JN, Grebin B, and Winters RW: Intravenous alimentation in pediatric patients, J PEDIATR80:351, 1972. 12. Winters RW, editor: The body fluids in pediatrics, Boston, 1973, Little, Brown & Company. 13. Holman RT: Essential fatty acid deficiency in humans, in Galli C, Jacini G, and Pecile A, editors: Dietary lipids and postnatal development, New York, 1973, Raven Press, p 127. 14. WHO Expert Committee: Trace elements in human nutrition, World Health Organization, Technical RePort Series No 532, Geneva, 1973. 15. Caldwell MD, Meng HC, and Jonsson HT: Essential fatty acid deficiency (EFAD)--now a human disease, Fed Proc 33:915, 1974. 16. Greene HL: Trace metals and vitamins, in Winters RW, and Hasse!meyer EG, editors: Intravenous nutrition in high risk infants, New York, 1974, John Wiley & Sons, Inc. 17. Stegink LD, Shepard JA, Fry LK,~and Filer LJ Jr: Sugaramino acid complexes in parenteral alimentation, Pediatr Res 8:386, 1974. 18. Boeckman CR, and Krill CE Jr: Bacterial and fungal infections complicating parenteral alimentation in infants and children, J Pediatr Surg 5:1!7, 1970. 19. Brennan MF, O'Connell R C, Rosol J, and Kundsin RB: The growth of Candida albicans in nutritive solutions given parenterally, Arch Surg 103:705, 1971. 20. Goldmann DA, Martin WT, and Worthington JW:
Growth of bacteria and fungi in total parenteral nutrition solutions, Am J Surg 126:314, 1973. Heird WC, MacMillan R, and Winters RW: Total parenteral nutrition in pediatrics, in Barness L, editor: Infant feeding, in press. Coran AG: The lung-term total intravenous feeding of infants using peripheral veins, J Pediatr Surg 8:801, 1973. Widdowson E: Changes in body proportions and composition during growth, in Davis JA, and Dobbing J, editors: Scientific foundations of paediatrics, London, 1974, Wm. Heinemann Medical Books, Ltd., Chap 12, p 153. Asch MJ, Sperling M,Fiser R, Leake R, OhW, and Moore TC: Metabolic and hormonal studies comparing three parenteral nutrition regimens in infants, V Annual meeting of Am Pediatr Surg Assoc, New Orleans, April 6, 1974. Wallace WM: Nitroger~ content of the body and its relation to retention and loss of nitrogen, Fed Proc 18:1125, 1959. Avery GB, Villavicencio O, Lilly JR, and Randolph JG: Intractable diarrhea in early infancy, Pediatrics 41:712, 1968. Keating JP, and Ternberg JL: Amino acid-hypertonic glucose treatment for intractable diarrhea in infants, Am J Dis Child 123:336, 1971. Keating JP: Parenteral nutrition in infants with malabsorption, in Winters RW, and Hasselmeyer EG, editors: Intravenous nutrition in high risk infants, New York, 1974, John Wiley & Sons, Inc. Hyman CJ, Reiter J, Rodman J, and Drash AL: Parenteral and oral alimentation in the treatment of the nonspecific protracted diarrheal syndrome of infancy, J PEDIATR78:17, 1971. Greene HL, McCabe DR, Stifel FB, Merenstein GB, and Herman RH: Intractable diarrhea and malnutrition in infancy: Changes in intestinal morphology and disaccharidase activities during treatment with total parenteral nutrition or oral elemental diets, in press. Winick M: Malnutrition and brain development, J PEDIATR74:667, 1969. Heird WC, Driscoll JM Jr, and Winters RW: Total parenteral nutrition in very low birth weight infants, in Size at birth, Ciba Symposium, 1974. Abel RM, Beck CH Jr, Abbott WM, Ryan JA Jr, Barnett GO, and Fischer JE: Acute renal failure: Treatment with intravenous amino acids and glucose, N Engl J Med 288:695, i973. Abel RM: Parenteral nutrition in the treatment of renal failure, in Fischer JE, editor: Total parenteral nutrition, Boston, 1974, Little, Brown & Company. Cohen MI, Daum F, and Boley F: Personal communication. Fischer JE, Foster GS, Abel RM, Abbott WM, and Ryan JA: Hyperalimentation as primary therapy for inflammatory bowel disease, Am J Surg !25:165, 1973. MacFayden BV, Dudrick sJ, and Ruberg RL: Management of gastrointestinal fistulas with parenteral hyperalimentation, Surgery 74:100, 1973. Chance GW: Results in very low birth weight infants ((1300 gm birth w e i g h t ) , in W i n t e r s RW, and Hasselmeyer EG, editors: Intravenous nutrition in high risk infants, New York, 1974, John Wiley & Sons, Inc.
Heird and Winters
Travis SF, Sugerman H J, Ruberg RL, Dudrick SJ, Delivoria-Papadopoulos M, Miller LD, and Oski FA: Alterations of red-cell glycolytic intermediates and oxygen transport as a consequence of hypophosphatemia in patients receiving intravenous hyperalimentation, N Engl J Med 285:763, 1971. Jacob HS, Yawata Y, Craddock P, Hebbel R, Howe R, and Silvis S: H y p e r a l i m e n t a t i o n h y p o p h o s p h a t e m i a : Hematologic-neurologic dysfunction due to ATP depletion, Trans Assoc Am Phys 86:143, 1973. Heird WC, Nicholson JF, Driscoll JM Jr, Schullinger JN, and Winters RW: Hyperammonemia resulting from intravenous a!imentation using a mixture of synthetic Lamino acids: A preliminary report, J PEDIATR81:162, 1972. Johnson JD, Albritton WL, and Sunshine P: Hyperammonemia accompanying parenteral nutrition in newborn infants, J PEDIATR81:154, 1972. Greene HL: Personal communication. Dudrick SJ, MacFadyen BV Jr, Van Buren CT, Ruberg RL, and Maynard AT: Parenteral hyperalimentation: Metabolic problems and solutions, Ann Surg 176:259, 1972. Stegink LD, and Baker GL: Infusion of protein hydrolysates in the newborn infant: Plasma amino acids concentrations, J PEDIATR78:595, 1971. Ghadimi H, Abaci F, Kumar S, and Rathi M: Biochemical aspects of intravenous alimentation, Pediatrics 48:955, 1971. Pildes RS~ Ramamurthy RS, Cordero GV, and Wong PWK: Intravenous supplementation of L-amino acids and dextrose in low-birth-weight infants, J PEDIATR 82:945, 1973. Das JB, and Filler RM: Amino acid utilization dur!ng total parenteral nutrition in the surgical neonate, J Pediatr Surg 8:793, 1973. Heird WC, and Nicholson JF: Unpublished data. O'Tuama LA, Kirkman HN, and James PN: Raised intracranial pressure after hyperalimentation, Lancet 2:1101, 1973. Paulsrud JR, Pensler L, Whitten CF, Stewart S, and Holman RT: Essential fatty acid deficiency in infants induced by fat-free intravenous feedings, Am J Clin Nutr 25:897, 1972. Caldwell MD, Johnson HT, and Othersen HB: Essential fatty acid deficiency in the infant: A real danger during prolonged parenteral alimentation, J PEDIATR 25:897, 1972. White HB Jr, Turner MD, Turner AC, and Miller RC: Blood lipid alterations in infants receiving intravenous fat-free alimentation, J PEDIATR83:305, 1973. Karpel JT, and Peden VH: Copper deficiency with parenteral nutrition, J PEDIATR80:32, 1972. Cohen MI: Changes in hepatic function, in Winters RW, and Hasselmeyer EG, editors: Intravenous nutrition in high risk infants, New York, 1974, John Wiley & Sons, Inc. Wigger JH: Cholestasis in premature infants receiving total intravenous alimentation, Pediatric Pathology Club, Toronto, 1971. Heird WC, Dell RB, Driscoll JM Jr, Grejbin B, and Winters RW: Metabolic acidosis resulting from intravenous
The Journal of Pediatrics January 1975
67. 68. 69.
alimentation mixtures containing synthetic amino acids, N Engl J Med 287:943, 1972. Chan JCM, Malekzadeh M, and Hurley J: pH and titratable acidity of amino acid mixtures used in hyperalimentation, JAMA 220:1119, 1972. Chan JCM: The influence of synthetic amino acid and casein hydrolysate on the endogenous production and urinary excretion of acid in total intravenous alimentation, Pediatr Res 6:789, 1972. Holman RT: Essential fatty acid deficiency, t'n Holman RT, editor: Progress in the chemistry of fats and other lipids, Oxford, 1968, Pergamon Press, v01 IX, part 2, p. 275. Galli C, White HB Jr, and Parletti R: Brain lipid modifications induced by essential fatty acid deficiency in growing male and female rats, J Neurochem 17:347, 1970. Wilmore DW, Moylan JA, Helmkamp GM, and Priutt BA Jr:Clinical evaluation of a 10% intravenous fat emulsion for parenteral nutrition in thermally injured patients, Ann Surg 178:503, 1973. Ranwell PW, Shaw JE, Clarke GB, Grostic MF, Kaiser DG, and Pike JE: Prostaglandins, in Holman RT, editor: Progress in the chemistry of fats and other lipids, Oxford, 1968, Pergamon Press, vol IX, part 2, p. 233. Press M, Hartop PJ, and Prottey C: Correction of essential fatty acid deficiency in man by the cutaneous application of sunflower-seed oil, Lancet 1:597, 1974. Goldmann DA: Microbiological safety of solutions and delivery to the patient: Delivery to the patient, Symposium on total parenteral nutrition, Jan. 17-19, 1972, Council on Foods and Nutrition of American Medical Association, p. 132. Curry CR, and Quie, PF: Fungal septicemia in patients receiving parenteral hyperalimentation, N Engl J Med 285:1221, 1971. Duma RS: First of all do no harm, editorial, N Engl J Med 185:1258,1971. Altman LK: Doctors warned on infections in intravenous therapy, NY Times Nov. 27, 1971. Brennan MF, Goldman MH, O'Connell RC, Knudsin RB, and Moore FD: Prolonged parenteral alimentation: Candida growth and the prevention of candidemia by amphoterician instillation, Ann Surg 176:265, 1972. Goldman DA, and Maki DG: Infection control in total parenteral nutrition, JAMA 223:1360, 1973. Thiessen H, Jacobsen J, and Brodersen R: Displacement of albumin-bound bilirubin by fatty acids, Acta Pediatr Scand 61:285, 1972. Greene HL: Effects of intralipid on the lung, & Winters RW, and Hasselmeyer EG, editors: Intravenous nutrition in high risk infants, New York, 1974, John Wiley & Sons, inc. (In press.) Young JM, and Weser E: The metabolism of circulating maltose in man, J Clin Invest 50:986, 1971. Synderman SE: The protein and amino acid requirements of the premature infant, in Jonxis JHP, Visser HKA, and Troelstra JA, editors: Nutricia symposium, metabolic processes in the foetus and newborn infant, Leiden, 1971, Stenfert Kroese, p. 128. Sturman JA, Gaul G, and Raiha NCR: Absence of cystathionase in human fetal liver: Is cystine essential? Science 169:74, !970.