- Email: [email protected]
Intravenous infusions— solutions and emulsions
P.I. Folb
36 PROBLEMS OF NUTRITION
Intravenous infusions solutions and emulsions TOTAL PARENTERAL
Gallbladder disease (SED-1 O, 637) In SEDA-8 (p. 315) the association of prolonged total parenteral nutrition (TPN) with cholestasis and jaundice was reviewed. A high incidence of cholelithiasis and cholecystitis has been noted in patients receiving TPN for a number of months or longer. A recent study by Messing et al (1 c) has contributed towards an understanding of the pathogenesis of these complications of TPN. Using biliary ultrasonography, they reported that nearly half of 23 patients developed biliary 'sludge'. Gallstones were demonstrated in 6 of 14 sludge-forming patients, but they were not observed in patients who were sludge-negative. Analysis of bile from these patients revealed cholesterol c r y s t a l s and small stones of mixed bilirubin-cholesterol type. In every instance these findings were noted a minimum of 3 weeks after the initiation of parenteral nutrition. In 5 of 7 patients who developed sludging this was reversed after a short period of oral refeeding. These observations need confirmation in a larger study, but they appear to indicate that gallbladder stasis may be linked with the development of cholelithiasis in prolonged TPN. In a related paper, Roslyn et al (2 c) have reported a higher-than-expected incidence of gallbladder disease during the course of prolonged TPN in 25 of 109 patients (23%) with ileal disorders who had been treated with TPN for a mean duration of Side Effects of Drugs .4 nnual 9 M.N.G. Dukes, editor 9 Elsevier Science Publishers B.V., 1985 ISBN 0 444 90394 1 $0.85 per article per page (transactional system) $0.20 per article per page (licensing system)
13.8 months. Comparison of the incidence of gallbladder disease in age- and sex-matched patients with ileal disorders not receiving TPN revealed a significantly higher incidence of cholecystitis and cholelithiasis in the TPN patients. The administration of TPN to high-risk patients such as those with ileal disorders may increase their inherent liability to gallbladder and :biliary tract disease. A limitation of this study is that it is a retrospective analysis, and no account was taken of the extent of the underlying ileal disease or ileal resection. It remains to be determined whether a particular element of the TPN solution is in itself cholelithogenic, or whether factors associated with TPN administration and bowel rest (including inadequate food stimulation of the gallbladder) promote gallbladder stasis and alter physiological bile acid metabolism. In an ultrasonographic study of 21 children receiving long-term TPN, 9 (43%) were found to have cholelithiasis and calculous cholecystitis, and 5 had undergone cholecystectomy (3c). Only children with ileal disorders or previous ileal resection developed stones. The prevalence of stones was nearly twice that observed earlier in similarly affected adults not receiving TPN. In an age group that very seldom develops cholelithiasis these data provide further evidence that the prolonged administration of parenteral nutrition induces cholelithiasis and cholecystitis, and they highlight the necessity for the consideration of cholecystiffs as a diagnosis in a child receiving TPN who experiences suggestive abdominal symptomatology. Micronutrient deficiencies
(SED-1 O, 635;
SEDA-8, 317) A patient has been described with multiple enteric fistulae, treated for several months with TPN, who developed severe
Intravenous infusions - solutions and emulsions
Chapter 36
depression associated with delirium, dermatitis, pallor, paresthesiae, nausea, vomiting, anorexia and headaches. The symptoms improved within a few days of the initiation of treatment with parenteral biotin (4r Biotin levels were not measured prior to replacement so that the association was speculative. Nevertheless, there are reasons to believe that biotin deficiency was responsible at least in part for his various symptoms. There may be a larger number of patients at risk of developing biotin deficiency with TPN than is generally appreciated. Nutritional copper deficiency is well documented in patients maintained on TPN for long periods of time without copper supplementation. Anemia and neutropenia are the most striking hematological abnormalities associated with copper deficiency. Sideroblastic changes and nuclear maturation defects are observed in erythroid precursors (sR). in a study of 25 patients in whom the serum copper levels during TPN (mean duration 6 weeks) were measured, an association of hypocupremia with the underlying disease was investigated. Hypocupremia occurred in 3 patients with intestinal fistulae and 1 patient with radiation-induced enteritis. No patient with cancer cachexia developed hypocupremia. Hypocupremia was not associated in any instance with clinical or metabolic symptoms (6c). A 5 89 girl with intestinal pseudoobstruction became severely depleted of zinc during prolonged TPN and she developed the characteristic bullous rash on the face, hands and feet of acrodermatitis enteropathica. Light and electron microscopy of a lesion revealed pronounced extracellular edema with cysts, and cleft formation in the deep part of the epidermis. A few acantholytic cells were seen. This development was linked to zinc deficiency (7c). Phosphorus deficiency with clinical rickets has been reported as a consequence of intravenous alimentation in a pre-term baby (8c). Laboratory investigation revealed hypophosphatemia as the main cause of this picture. Recovery was achieved b y providing phosphorus supplementation. The case demonstrates that conventional infusates do not necessarily meet the phosphorus requirement of rapidly growing infants. In the patient under consideration the intakes of phosphorus, calcium and vitamin D respectively were 1 mmol/kg/d, 1.2 mmol/kg/d
297
and 100 IU/kg/d. The authors recommend that more phosphorus and calcium should be supplied to rapidly growing pre-term infants treated with prolonged TPN. They may need a phosphorus intake as high as 2 mmol/kg/d and a calcium intake of 2.5 mmol/kg/d (8c). L o w levels o f plasma carnitine and reduced urinary carnitine excretion with persistently elevated plasma bilirubin levels, reactive hypoglycemia and generalized skeletal muscle weakness have been described in a patient requiring long-term TPN (9c). Subsequent administration of L-carnitine corrected the plasma carnitine level and the reactive hypoglycemia and this was associated with a return of the plasma bilirubin levels to normal and with recovery of skeletal muscle strength. L-Carnitine is required for the transport of long-chain fatty acids into mitochondria for subsequent oxidation. In carnitine deficiency, fatty acid oxidation is reduced and fatty acids may be diverted into triglyceride synthesis, particularly in the liver. There is also a re-adjustment of the balance between carbohydrate and fat metabolism, and giycolysis is increased. As far as is known, there is no synthesis of carnitine b y intestinal microflora, so that patients receiving TPN have their dietary intake of carnitine reduced to nothing. If they have hepatic dysfunction in addition, they may be unable to make available sufficient carnitine for normal fatty acid metabolism, and hepatic dysfunction may be exacerbated by excessive triglyceride deposition in the hepatocytes. Metabolic bone disease (SED-1 O, 635; SEDA-8, 318)
In a comparison of 22 bone biopsy specimens from 16 patients treated with prolonged TPN, there was significantly greater osteoid and a reduced total bone area and bone formation rate, measured by double tetracycline labeling, as age- and sex-matched normal controls. Aluminium was found in specimens from 14 patients receiving casein hydrolysate but not in the patients receiving amino acids as their nitrogen source. The aluminium was localized to the surface of mineralized bone. Reduced bone formation correlated inversely with the logarithm of the aluminium level (10c). The findings described in this study are similar to those reported in aluminiumintoxicated patients receiving chronic hemo-
Chapter 36 P.L Folb
298 dialysis. Both groups also have low circulating parathyroid hormone levels. Aluminium may be an important pathogenic factor in the low bone turnover and in the osteodystrophy of patients receiving prolonged TPN, possibly acting as a toxin with an inhibitory effect at the mineralization front. No direct proof of a role of aluminium in the pathogenesis of this TPN-related bone disease has been provided by these associations nor is it clear whether aluminium accumulation in bone is the factor responsible for osteomalacia in TPN patients. Aluminium can accumulate in the parathyroid glands and in-vitro studies have shown that large amounts of aluminium impair parathyroid secretion. Low or lownormal serum parathyroid hormone levels and hypercalciuria and hyperphosphatemia are consistent with a state of physiological hypoparathyroidism. This might arise, in part, from aluminium accumulation in the parathyroids or it may be due to parenteral loading with calcium ( 11 c). Calcium balance was studied in 5 patients receiving TPN with otherwise constant nutrient infusions who were randomized i n t o 2 levels of amino acid administration. One group received 1 g/kg ideal body weight of amino acids while the other received 2 g/kg ideal body weight. All the patients entered a state of negative calcium balance, the mean urinary calcium excretion being increased from 287 to 455 mg/d. There was no effect of the amino acid regimens on serum calcium, ionized calcium, parathyroid hormone or 25-hydroxyvitamin D. However, calcium excretion by the kidney corrected for glomerular filtration rate was increased at the higher amino acid dose, indicating a reduction in renal reabsorption of calcium. Daily urinary excretion of sulfate, ammonia and titratable acidity were all increased with the higher amino acid infusion (12c). This study demonstrates the calciuretic effect of high amino acid infusion during TPN. The importance of monitoring calcium balance during TPN is obvious when consideration is given to the serious potential complications of hypercalciuria and skeletal calcium loss.
Miscellaneous complications A 35-year-old female developed progressive acute renal failure after surgical drainage of a pancreatic abscess and subsequent
TPN. Renal biopsy revealed severe obstructive renal oxalosis with concomitant interstitial nephritis. Primary oxalosis was excluded by determination of glyeeric acid and glycolic acid in the urine. Since other preconditions for increased oxalate formation were not present, an adverse reaction to xylitol was considered the likely cause. During 4 weeks of TPN the patient had received 3.0 g/kg/d of this sugar alcohol (total dose 4480 g) (13c).
Infections (SED-1 O, 639) Baley and co-workers have reported an apparent increase in disseminated fungal infections in very-low-birth-weight infants who required prolonged TPN, central or venous in-line catheters, and multiple courses of broad-spectrum antibiotics for documented or suspected bacterial sepsis. (14c). Polymicrobial bacteremia associated with lipid emulsion in a neonatal intensive care unit developed in 5 infants during a 48hour period. Two infants died. KlebsieUa pneumoniae serotypes 21 and 24 were isolated from 4 of the infants, and both K. pneumoniae serotype 24 and Enterobacter cloacae were isolated from the fifth child. The evidence suggested that extraneous contamination of the lipid emulsion bottles had occurred when the hands of a nurse became transiently colonized with these organisms while she was caring for an infant with K. pneumoniae and E. cloacae. Repeated entry into the lipid emulsion bottle which was used for multidose medication probably resulted in contamination (15c). Between 1979 and 1981 Fleer et al (16 c) experienced an increasing incidence of septicemia caused by coagulase-negative staphylococci in a neonatal intensive care unit. A detailed analysis of the neonatal intensive care unit population in 1981 revealed that more than 90% of the cases occurred in premature infants of low birth weight (less than 2500 g). All had received TPN solutions. A case-control study did not uncover differences in underlying diseases or modes of treatment between patients and controls. Random bacteriological checks of TPN fluids revealed that nearly 20% of these solutions had been contaminated, mainly with coagulase-negative staphylococci. The incidence of staphylococcal
Intravenous infusions - solutions and emulsions
Chapter36
299
dextran-induced anaphylactoid/anaphylactic reactions was investigated after hapten inhibition employing 2 modes of administration of dextran 1 (20c). The trial was an open prospective study of 12,060 patients; they were randomized into one group receiving a mixture of 20 ml of 15% dextran 1 and 500 ml of either dextran 70 or dextran 40 while the other group received an intravenous injection of 20 ml of 15% dextran 1 before the infusion of dextran 70 or dextran 40. The admixture of dextran PLASMA SUBSTITUTES 1 with dextran 70 or 40 did not prevent severe dextran-induced adverse reactions. Dextrans (SED-1 O, 642; SEDA-8, 319) Of 34,955 patients administered an intravenous injection of 20 ml of 15% dextran Two case-reports of anaphylaxis follow- 1 - 2 minutes before the start of an infusion ing intraperitoneal administration of dextran of dextran 70 or dextran 40, only 1 patient at the conclusion of gynecological opera- developed a severe dextran-induced adverse tions to prevent the formation of adhesions reaction. The incidence of severe dextran(17 c, 18 r have drawn attention to this induced adverse reactions after pre-injection adverse effect during the immediate post- of 20 ml dextran 1 was significantly lower operative period. In one of the patients than after pre-injection of 10 ml of 15% (17 c) the onset of anaphylaxis was 75 min- dextran 1 (P=0.01), and it was also markedutes after completion of the operation. After ly lower than that of a large Swedish historan initial response to standard treatment ical control series. The incidence of mild there was a recurrence 20 minutes later. dextran reactions was not affected by preThe patient was again treated and she re- injection with dextran 1 (21c). The authors covered over 12 hours. The second reaction recommended that patients due to receive is explained by the slow absorption of dex- dextran 70 or 40 should be given a protran from the peritoneal cavity which phylactic intravenous injection of 20 ml acts as a reservoir of the offending agent. of 15% dextran 1 before the first unit to The authors (17 c) recommended removal reduce the risk of severe adverse reactions. Over a 13-day period of infusion of of as much of the intraperitoneal dextran as possible in order to prevent recurrence dextran 40 500 ml/d to 5 diabetics and 2 of anaphylaxis. If culdocentesis is not ef- non-diabetics, all 7 patients developed acute fective, it was recommended that consider- renal failure. Diabetes meUitus was thought to be a special risk factor. Diabetic giomeration should be given to peritoneal dialysis. In an open prospective study the preven- ulopathy with increased glomerular permetion of dextran-induced anaphylactic reac- ability might play a role in the pathogenesis tions by hapten inhibition was investigated of the renal failure. According to these in 29,252 patients in various hospitals in authors, diabetic patients suffering from Sweden, Norway and Finland. Ten milli- peripheral and cerebral occlusive arterial liters of 15% dextran 1 with an average insufficiency should not receive prolonged molecular weight of 1 dalton was injected dextran infusion therapy (21 c). intravenously 2 minutes before infusion of In a retrospective Swedish investigation dextran 70 (Macrodex) or dextran 40 between 1970 and 1979, 478 reports of dex(Rheomacrodex). An incidence of severe tran-induced anaphylactoid/anaphylactic rereactions of 0.024% was recorded. Com- actions (DIAR) were obtained (22c). The pared with a large historical control series, number of units of dextran 40 and 70 pre-injection of 10 ml dextran 1 exerted a used during this decade was 31,598 and partial protective effect, but no statis- 1,051,668, respectively. During the latter tical proof of this was found. Mild, short- 5 years of the study, when reporting of lasting adverse reactions to a pre-injection of severe adverse drug reactions had become dextran 1 were observed in 0.072% of pa- mandatory, the incidence of severe DIAR tients (19 c). per unit administered was 0.013% for In another study the incidence of dextran 40 and 0.025% for dextran 70; septicemia in infants who had received contaminated TPN fluids was 10-fold higher than in infants who had received sterile solutions. The majority of the staphylococci that were isolated from the blood cultures were multiply resistant to antibiotics. Removal or replacement of the intravenous line and the administration of cefalotin and fresh plasma was universally successful in the management of these patients.
300 the incidence of fatal DIAR was 0.003% and 0.004%, respectively. A male/female ratio of 1.5:1 was found for all DIARs while for the most severe DIAR this was 3:1. The mean age of the patients was higher in the severe than in the mild reactions. A previous allergic disease did not apparently predispose to severe DIAR. Diabetes mellitus, cardiovascular and pulmonary diseases, and chronic inflammation were more common among the patients who had suffered the most severe reactions. Fatal DIAR during anesthesia occurred almost exclusively in patients under spinal or epidural blockade. A favorable outcome of resuscitation was possibly promoted by infusion of large volumes of crystalloid fluids. Steroids, noradrenaline or isoprenaline did not influence the outcome. Fatal DIAR occurred in 1 case after infusion of a minute amount of dextran, and the authors strongly advised against 'biological test doses'. Modified fluid gelatins [SED-1 O, 643)
Urticaria, edema and anaphylactic shock occur occasionally as a result of giving modified fluid gelatins used as colloid plasma substitutes. The mechanisms of these reactions are not understood. In a report on 3 cases of anaphylactic shock studied by skin tests and in vitro leukocyte histamine release (LHR), the intradermal skin tests were positive for concentrations of 1:1000 to 1:10 of the commercial preparation concerned. Thirty control subjects were negative for the undiluted preparation. Positive LHR was obtained from patient leukocytes washed and incubated in trisalbumin-calcium buffer with serial dilutions of the fluid gelatins. Controls were negative. The conclusion of the authors was that skin tests and LHR might be useful in the diagnosis and prediction of patients who are reactive to gelatins. The findings further suggested that the mechanism involved may be a release of mediators from mast
Chapter 36 P.L Folb cells or basophils, but the precise immunological effect remains undefined (23c).
Cardioplegic solutions; particulate debris (SED-I O, 642, SEDA-8, 324) The potential hazard of particulate debris (SED-10, 642) in unfiltered cardioplegic solutions has been assessed using the isolated rat heart preparation (24c). Five different intravenous solutions were evaluated. Particles were c o u n t e d in each solution, and each fell within the limits for particulate contamination defined by both the United States and the British pharmacopeias. Continuous intracoronary infusion of each solution over 20 minutes at constant temperature and pressure resulted in a progressive reduction of coronary flow due to a rise in coronary vascular resistance. Filtration of these same solutions through a 0.8 /~m filter immediately prior to entry in the heart largely prevented these coronary vascular changes. The filter was examined by scanning electron microscopy and showed amorphous and crystalline debris (24c). In studies of hypothermic ischemia for 180 minutes using filtered cardioplegic solutions, 90% recovered their pre-ischemic functional capacity. Hearts receiving unfiltered solutions failed to recover and had significantly higher levels of creatine kinase leakage. These results suggested that even commercially produced solutions conforming to limits of particulate contamination acceptable for intravenous administration may prove harmful if given unfiltered directly into the coronary arteries. The likely mechanism of damage is particleinduced coronary vasoconstriction (24c). These facts should be considered in the light of evidence advanced in previous volumes in this series that even a degree of particulate contamination normally considered acceptable may produce problems, e.g. in ophthalmic preparations, radiological contrast media and of course intravenous solutions.
REFERENCES 1. Messing B, Bories C, Kunstlinger F et al (1983) Does total parenteral nutrition induce gallbladder sludge formation and lithiasis? Gastroenterology, 84, 1012. 2. Roslyn JJ, Pitt HA, Mann LL et al (1983) Gallbladder disease in patients on long-term parenteral nutrition. Gastroenterology, 84, 1486.
3. Roslyn JJ, Berquist WE, Pitt HA et al (1983) Increased risk of gallstones in children receiving total parenterat nutrition. Pediatrics, 71,784. 4. Levenson JL (1983) Biotin-responsive depression during hyperalimentation. J. Parent. Ent. Nutr., 7, 181. 5. Williams DM (1983) Copper deficiency in
Intravenous infusions - solutions and emulsions
Chapter 36
humans. Sem. HaematoL, 20, 118. 6. Bozzetti F, Inglese MG, Terno G e t al (1983) Hypocupremia in patients receiving total parenteral nutrition. J. Parent. Ent. Nutr., 7, 563. 7. Welsmannn K, Kvist N, Kobayasi T (1983) BuUous acrodermatitis due to zinc deficiency during total parenteral nutrition: an ultrastructural study of the epidermal changes. Acta Derrrt Venereol., 63, 143. 8. The TS, Koll6e LA, Boon JM et al (1983) Rickets in a preterm infant during intravenous alimentation. Acta Paediatr. Scand., 72, 769. 9. Worthley LIG, Fishlock RC, Snoswell AM (1983) Carnitine deficiency with hyperbilirubinemia, generalized skeletal muscle weakness and reactive hypoglycemia in a patient on long-term total parenteral nutrition: treatment with intravenous L-carnitine. J. Parent. Ent. Nutr., 7, 176. 10. Ott SM, Maloney NA, Klein GL et al (1983) Aluminium is associated with low bone formation in patients receiving chronic parenteral nutrition. Ann. Intern. Med., 98, 910. 11. Klein G L Ott SM, Alfrey AC et al (1982) Aluminium as a factor in the bone disease of long-term parenteral nutrition. Trans. Assoc. Ar~ Physicians, 95, 155. 12. Bengoa JM, Sitrin MD, Wood RJ et al (1983) Amino acid-induced hypercalciuria in patients on total parenteral nutrition. Am. J. Clin. Nutr., 38, 264. 13. Schultze G, Pommer W, Offermann G e t al (1983) Acute kidney failure in secondary renal oxalosis: additional indications for a causal connection to xylitol infusions, lnfusionsther. Klin. Ernahr., 10, 322. 14. Baley JE, Kliegman RM, Fanaroff AA (1984) Disseminated fungal infections in very lowbirth-weight infants: clinical manifestations and epidemiology. Pediatrics, 73, 144. 15. Jarvis WR, Highsmith AK, Allen JR et al (1983) Polymicrobial bacteremia associated with lipid emulsion in a neonatal intensive care unit.
301
Pediatr. Infect. Dis., 2, 203. 16. Fleer A, Senders RC, Visser MR et al (1983) Septicemia due to coagulase-negative staphylococci in a neonatal intensive care unit: clinical and bacteriological features and contaminated parenterat fluids as a source of sepsis. Pediatr. Infect. Dis., 2, 426. 17. Borten M, Seibert CP, Taymor ML (1983) Recurrent anaphylactic reaction to intraperitoneal dextran 75 used for prevention of postsurgical adhesions. Obstet. Gyn.., 61,755. 18. Reisner LS (1984) Anaphylaxis to intraperitoneal dextran. Anaesthesia, 60, 259. 19. LjungstriSm KG, Renck H, Hedin H e t al ( 1983) Prevent ion of dextran-induced anaphylactic reactions by hapten inhibition. I. A Scandinavian multicenter study on the effects of 10 ml of dextran 1, 15%, administered before dextran 70 or dextran 40. Acta Chit. Scand., 149, 341. 20. Renck H, Ljungstr6m KG, Rosberg B e t al (1983) Prevention of dextran-induced anaphylactic reactions by hapten inhibition. II. A comparison of the effects of 20 ml dextran 1, 15%, administered either admixed to or before dextran 70 or dextran 40. Acta Chit. Scand., 149, 349. 21. Renck H, Ljungstr6m KG, Hedin H e t al ( 1983) Prevention of dextranqndu ced anaphylactic reactions by hapten inhibition. III. A Scandinavian multicenter study on the effects of 20 ml dextran 1, 15%, administered before dextran 70 or dextran 40. Acta Chir. Scand., 149, 355. 22. Ljungstr6m KG, Renck H, Strandberg K et al (1983) Adverse reactions to dextran in Sweden 1970-1979.Acta Chir. Scand., 149, 253. 23. Vervloet D, Senft M, Dugue P e t al (1983) Anaphylactic reactions to modified fluid gelatins. J. Allergy Clin. lmmunol., 71,535. 24. Robinson LA, Braimbridge MV, Hearse DJ (1984) The potential hazard of particulate contamination of cardioplegic solutions. J. Thorac. Cardiovasc. Surg., 87, 48.
36 PROBLEMS OF NUTRITION
Intravenous infusions solutions and emulsions TOTAL PARENTERAL
Gallbladder disease (SED-1 O, 637) In SEDA-8 (p. 315) the association of prolonged total parenteral nutrition (TPN) with cholestasis and jaundice was reviewed. A high incidence of cholelithiasis and cholecystitis has been noted in patients receiving TPN for a number of months or longer. A recent study by Messing et al (1 c) has contributed towards an understanding of the pathogenesis of these complications of TPN. Using biliary ultrasonography, they reported that nearly half of 23 patients developed biliary 'sludge'. Gallstones were demonstrated in 6 of 14 sludge-forming patients, but they were not observed in patients who were sludge-negative. Analysis of bile from these patients revealed cholesterol c r y s t a l s and small stones of mixed bilirubin-cholesterol type. In every instance these findings were noted a minimum of 3 weeks after the initiation of parenteral nutrition. In 5 of 7 patients who developed sludging this was reversed after a short period of oral refeeding. These observations need confirmation in a larger study, but they appear to indicate that gallbladder stasis may be linked with the development of cholelithiasis in prolonged TPN. In a related paper, Roslyn et al (2 c) have reported a higher-than-expected incidence of gallbladder disease during the course of prolonged TPN in 25 of 109 patients (23%) with ileal disorders who had been treated with TPN for a mean duration of Side Effects of Drugs .4 nnual 9 M.N.G. Dukes, editor 9 Elsevier Science Publishers B.V., 1985 ISBN 0 444 90394 1 $0.85 per article per page (transactional system) $0.20 per article per page (licensing system)
13.8 months. Comparison of the incidence of gallbladder disease in age- and sex-matched patients with ileal disorders not receiving TPN revealed a significantly higher incidence of cholecystitis and cholelithiasis in the TPN patients. The administration of TPN to high-risk patients such as those with ileal disorders may increase their inherent liability to gallbladder and :biliary tract disease. A limitation of this study is that it is a retrospective analysis, and no account was taken of the extent of the underlying ileal disease or ileal resection. It remains to be determined whether a particular element of the TPN solution is in itself cholelithogenic, or whether factors associated with TPN administration and bowel rest (including inadequate food stimulation of the gallbladder) promote gallbladder stasis and alter physiological bile acid metabolism. In an ultrasonographic study of 21 children receiving long-term TPN, 9 (43%) were found to have cholelithiasis and calculous cholecystitis, and 5 had undergone cholecystectomy (3c). Only children with ileal disorders or previous ileal resection developed stones. The prevalence of stones was nearly twice that observed earlier in similarly affected adults not receiving TPN. In an age group that very seldom develops cholelithiasis these data provide further evidence that the prolonged administration of parenteral nutrition induces cholelithiasis and cholecystitis, and they highlight the necessity for the consideration of cholecystiffs as a diagnosis in a child receiving TPN who experiences suggestive abdominal symptomatology. Micronutrient deficiencies
(SED-1 O, 635;
SEDA-8, 317) A patient has been described with multiple enteric fistulae, treated for several months with TPN, who developed severe
Intravenous infusions - solutions and emulsions
Chapter 36
depression associated with delirium, dermatitis, pallor, paresthesiae, nausea, vomiting, anorexia and headaches. The symptoms improved within a few days of the initiation of treatment with parenteral biotin (4r Biotin levels were not measured prior to replacement so that the association was speculative. Nevertheless, there are reasons to believe that biotin deficiency was responsible at least in part for his various symptoms. There may be a larger number of patients at risk of developing biotin deficiency with TPN than is generally appreciated. Nutritional copper deficiency is well documented in patients maintained on TPN for long periods of time without copper supplementation. Anemia and neutropenia are the most striking hematological abnormalities associated with copper deficiency. Sideroblastic changes and nuclear maturation defects are observed in erythroid precursors (sR). in a study of 25 patients in whom the serum copper levels during TPN (mean duration 6 weeks) were measured, an association of hypocupremia with the underlying disease was investigated. Hypocupremia occurred in 3 patients with intestinal fistulae and 1 patient with radiation-induced enteritis. No patient with cancer cachexia developed hypocupremia. Hypocupremia was not associated in any instance with clinical or metabolic symptoms (6c). A 5 89 girl with intestinal pseudoobstruction became severely depleted of zinc during prolonged TPN and she developed the characteristic bullous rash on the face, hands and feet of acrodermatitis enteropathica. Light and electron microscopy of a lesion revealed pronounced extracellular edema with cysts, and cleft formation in the deep part of the epidermis. A few acantholytic cells were seen. This development was linked to zinc deficiency (7c). Phosphorus deficiency with clinical rickets has been reported as a consequence of intravenous alimentation in a pre-term baby (8c). Laboratory investigation revealed hypophosphatemia as the main cause of this picture. Recovery was achieved b y providing phosphorus supplementation. The case demonstrates that conventional infusates do not necessarily meet the phosphorus requirement of rapidly growing infants. In the patient under consideration the intakes of phosphorus, calcium and vitamin D respectively were 1 mmol/kg/d, 1.2 mmol/kg/d
297
and 100 IU/kg/d. The authors recommend that more phosphorus and calcium should be supplied to rapidly growing pre-term infants treated with prolonged TPN. They may need a phosphorus intake as high as 2 mmol/kg/d and a calcium intake of 2.5 mmol/kg/d (8c). L o w levels o f plasma carnitine and reduced urinary carnitine excretion with persistently elevated plasma bilirubin levels, reactive hypoglycemia and generalized skeletal muscle weakness have been described in a patient requiring long-term TPN (9c). Subsequent administration of L-carnitine corrected the plasma carnitine level and the reactive hypoglycemia and this was associated with a return of the plasma bilirubin levels to normal and with recovery of skeletal muscle strength. L-Carnitine is required for the transport of long-chain fatty acids into mitochondria for subsequent oxidation. In carnitine deficiency, fatty acid oxidation is reduced and fatty acids may be diverted into triglyceride synthesis, particularly in the liver. There is also a re-adjustment of the balance between carbohydrate and fat metabolism, and giycolysis is increased. As far as is known, there is no synthesis of carnitine b y intestinal microflora, so that patients receiving TPN have their dietary intake of carnitine reduced to nothing. If they have hepatic dysfunction in addition, they may be unable to make available sufficient carnitine for normal fatty acid metabolism, and hepatic dysfunction may be exacerbated by excessive triglyceride deposition in the hepatocytes. Metabolic bone disease (SED-1 O, 635; SEDA-8, 318)
In a comparison of 22 bone biopsy specimens from 16 patients treated with prolonged TPN, there was significantly greater osteoid and a reduced total bone area and bone formation rate, measured by double tetracycline labeling, as age- and sex-matched normal controls. Aluminium was found in specimens from 14 patients receiving casein hydrolysate but not in the patients receiving amino acids as their nitrogen source. The aluminium was localized to the surface of mineralized bone. Reduced bone formation correlated inversely with the logarithm of the aluminium level (10c). The findings described in this study are similar to those reported in aluminiumintoxicated patients receiving chronic hemo-
Chapter 36 P.L Folb
298 dialysis. Both groups also have low circulating parathyroid hormone levels. Aluminium may be an important pathogenic factor in the low bone turnover and in the osteodystrophy of patients receiving prolonged TPN, possibly acting as a toxin with an inhibitory effect at the mineralization front. No direct proof of a role of aluminium in the pathogenesis of this TPN-related bone disease has been provided by these associations nor is it clear whether aluminium accumulation in bone is the factor responsible for osteomalacia in TPN patients. Aluminium can accumulate in the parathyroid glands and in-vitro studies have shown that large amounts of aluminium impair parathyroid secretion. Low or lownormal serum parathyroid hormone levels and hypercalciuria and hyperphosphatemia are consistent with a state of physiological hypoparathyroidism. This might arise, in part, from aluminium accumulation in the parathyroids or it may be due to parenteral loading with calcium ( 11 c). Calcium balance was studied in 5 patients receiving TPN with otherwise constant nutrient infusions who were randomized i n t o 2 levels of amino acid administration. One group received 1 g/kg ideal body weight of amino acids while the other received 2 g/kg ideal body weight. All the patients entered a state of negative calcium balance, the mean urinary calcium excretion being increased from 287 to 455 mg/d. There was no effect of the amino acid regimens on serum calcium, ionized calcium, parathyroid hormone or 25-hydroxyvitamin D. However, calcium excretion by the kidney corrected for glomerular filtration rate was increased at the higher amino acid dose, indicating a reduction in renal reabsorption of calcium. Daily urinary excretion of sulfate, ammonia and titratable acidity were all increased with the higher amino acid infusion (12c). This study demonstrates the calciuretic effect of high amino acid infusion during TPN. The importance of monitoring calcium balance during TPN is obvious when consideration is given to the serious potential complications of hypercalciuria and skeletal calcium loss.
Miscellaneous complications A 35-year-old female developed progressive acute renal failure after surgical drainage of a pancreatic abscess and subsequent
TPN. Renal biopsy revealed severe obstructive renal oxalosis with concomitant interstitial nephritis. Primary oxalosis was excluded by determination of glyeeric acid and glycolic acid in the urine. Since other preconditions for increased oxalate formation were not present, an adverse reaction to xylitol was considered the likely cause. During 4 weeks of TPN the patient had received 3.0 g/kg/d of this sugar alcohol (total dose 4480 g) (13c).
Infections (SED-1 O, 639) Baley and co-workers have reported an apparent increase in disseminated fungal infections in very-low-birth-weight infants who required prolonged TPN, central or venous in-line catheters, and multiple courses of broad-spectrum antibiotics for documented or suspected bacterial sepsis. (14c). Polymicrobial bacteremia associated with lipid emulsion in a neonatal intensive care unit developed in 5 infants during a 48hour period. Two infants died. KlebsieUa pneumoniae serotypes 21 and 24 were isolated from 4 of the infants, and both K. pneumoniae serotype 24 and Enterobacter cloacae were isolated from the fifth child. The evidence suggested that extraneous contamination of the lipid emulsion bottles had occurred when the hands of a nurse became transiently colonized with these organisms while she was caring for an infant with K. pneumoniae and E. cloacae. Repeated entry into the lipid emulsion bottle which was used for multidose medication probably resulted in contamination (15c). Between 1979 and 1981 Fleer et al (16 c) experienced an increasing incidence of septicemia caused by coagulase-negative staphylococci in a neonatal intensive care unit. A detailed analysis of the neonatal intensive care unit population in 1981 revealed that more than 90% of the cases occurred in premature infants of low birth weight (less than 2500 g). All had received TPN solutions. A case-control study did not uncover differences in underlying diseases or modes of treatment between patients and controls. Random bacteriological checks of TPN fluids revealed that nearly 20% of these solutions had been contaminated, mainly with coagulase-negative staphylococci. The incidence of staphylococcal
Intravenous infusions - solutions and emulsions
Chapter36
299
dextran-induced anaphylactoid/anaphylactic reactions was investigated after hapten inhibition employing 2 modes of administration of dextran 1 (20c). The trial was an open prospective study of 12,060 patients; they were randomized into one group receiving a mixture of 20 ml of 15% dextran 1 and 500 ml of either dextran 70 or dextran 40 while the other group received an intravenous injection of 20 ml of 15% dextran 1 before the infusion of dextran 70 or dextran 40. The admixture of dextran PLASMA SUBSTITUTES 1 with dextran 70 or 40 did not prevent severe dextran-induced adverse reactions. Dextrans (SED-1 O, 642; SEDA-8, 319) Of 34,955 patients administered an intravenous injection of 20 ml of 15% dextran Two case-reports of anaphylaxis follow- 1 - 2 minutes before the start of an infusion ing intraperitoneal administration of dextran of dextran 70 or dextran 40, only 1 patient at the conclusion of gynecological opera- developed a severe dextran-induced adverse tions to prevent the formation of adhesions reaction. The incidence of severe dextran(17 c, 18 r have drawn attention to this induced adverse reactions after pre-injection adverse effect during the immediate post- of 20 ml dextran 1 was significantly lower operative period. In one of the patients than after pre-injection of 10 ml of 15% (17 c) the onset of anaphylaxis was 75 min- dextran 1 (P=0.01), and it was also markedutes after completion of the operation. After ly lower than that of a large Swedish historan initial response to standard treatment ical control series. The incidence of mild there was a recurrence 20 minutes later. dextran reactions was not affected by preThe patient was again treated and she re- injection with dextran 1 (21c). The authors covered over 12 hours. The second reaction recommended that patients due to receive is explained by the slow absorption of dex- dextran 70 or 40 should be given a protran from the peritoneal cavity which phylactic intravenous injection of 20 ml acts as a reservoir of the offending agent. of 15% dextran 1 before the first unit to The authors (17 c) recommended removal reduce the risk of severe adverse reactions. Over a 13-day period of infusion of of as much of the intraperitoneal dextran as possible in order to prevent recurrence dextran 40 500 ml/d to 5 diabetics and 2 of anaphylaxis. If culdocentesis is not ef- non-diabetics, all 7 patients developed acute fective, it was recommended that consider- renal failure. Diabetes meUitus was thought to be a special risk factor. Diabetic giomeration should be given to peritoneal dialysis. In an open prospective study the preven- ulopathy with increased glomerular permetion of dextran-induced anaphylactic reac- ability might play a role in the pathogenesis tions by hapten inhibition was investigated of the renal failure. According to these in 29,252 patients in various hospitals in authors, diabetic patients suffering from Sweden, Norway and Finland. Ten milli- peripheral and cerebral occlusive arterial liters of 15% dextran 1 with an average insufficiency should not receive prolonged molecular weight of 1 dalton was injected dextran infusion therapy (21 c). intravenously 2 minutes before infusion of In a retrospective Swedish investigation dextran 70 (Macrodex) or dextran 40 between 1970 and 1979, 478 reports of dex(Rheomacrodex). An incidence of severe tran-induced anaphylactoid/anaphylactic rereactions of 0.024% was recorded. Com- actions (DIAR) were obtained (22c). The pared with a large historical control series, number of units of dextran 40 and 70 pre-injection of 10 ml dextran 1 exerted a used during this decade was 31,598 and partial protective effect, but no statis- 1,051,668, respectively. During the latter tical proof of this was found. Mild, short- 5 years of the study, when reporting of lasting adverse reactions to a pre-injection of severe adverse drug reactions had become dextran 1 were observed in 0.072% of pa- mandatory, the incidence of severe DIAR tients (19 c). per unit administered was 0.013% for In another study the incidence of dextran 40 and 0.025% for dextran 70; septicemia in infants who had received contaminated TPN fluids was 10-fold higher than in infants who had received sterile solutions. The majority of the staphylococci that were isolated from the blood cultures were multiply resistant to antibiotics. Removal or replacement of the intravenous line and the administration of cefalotin and fresh plasma was universally successful in the management of these patients.
300 the incidence of fatal DIAR was 0.003% and 0.004%, respectively. A male/female ratio of 1.5:1 was found for all DIARs while for the most severe DIAR this was 3:1. The mean age of the patients was higher in the severe than in the mild reactions. A previous allergic disease did not apparently predispose to severe DIAR. Diabetes mellitus, cardiovascular and pulmonary diseases, and chronic inflammation were more common among the patients who had suffered the most severe reactions. Fatal DIAR during anesthesia occurred almost exclusively in patients under spinal or epidural blockade. A favorable outcome of resuscitation was possibly promoted by infusion of large volumes of crystalloid fluids. Steroids, noradrenaline or isoprenaline did not influence the outcome. Fatal DIAR occurred in 1 case after infusion of a minute amount of dextran, and the authors strongly advised against 'biological test doses'. Modified fluid gelatins [SED-1 O, 643)
Urticaria, edema and anaphylactic shock occur occasionally as a result of giving modified fluid gelatins used as colloid plasma substitutes. The mechanisms of these reactions are not understood. In a report on 3 cases of anaphylactic shock studied by skin tests and in vitro leukocyte histamine release (LHR), the intradermal skin tests were positive for concentrations of 1:1000 to 1:10 of the commercial preparation concerned. Thirty control subjects were negative for the undiluted preparation. Positive LHR was obtained from patient leukocytes washed and incubated in trisalbumin-calcium buffer with serial dilutions of the fluid gelatins. Controls were negative. The conclusion of the authors was that skin tests and LHR might be useful in the diagnosis and prediction of patients who are reactive to gelatins. The findings further suggested that the mechanism involved may be a release of mediators from mast
Chapter 36 P.L Folb cells or basophils, but the precise immunological effect remains undefined (23c).
Cardioplegic solutions; particulate debris (SED-I O, 642, SEDA-8, 324) The potential hazard of particulate debris (SED-10, 642) in unfiltered cardioplegic solutions has been assessed using the isolated rat heart preparation (24c). Five different intravenous solutions were evaluated. Particles were c o u n t e d in each solution, and each fell within the limits for particulate contamination defined by both the United States and the British pharmacopeias. Continuous intracoronary infusion of each solution over 20 minutes at constant temperature and pressure resulted in a progressive reduction of coronary flow due to a rise in coronary vascular resistance. Filtration of these same solutions through a 0.8 /~m filter immediately prior to entry in the heart largely prevented these coronary vascular changes. The filter was examined by scanning electron microscopy and showed amorphous and crystalline debris (24c). In studies of hypothermic ischemia for 180 minutes using filtered cardioplegic solutions, 90% recovered their pre-ischemic functional capacity. Hearts receiving unfiltered solutions failed to recover and had significantly higher levels of creatine kinase leakage. These results suggested that even commercially produced solutions conforming to limits of particulate contamination acceptable for intravenous administration may prove harmful if given unfiltered directly into the coronary arteries. The likely mechanism of damage is particleinduced coronary vasoconstriction (24c). These facts should be considered in the light of evidence advanced in previous volumes in this series that even a degree of particulate contamination normally considered acceptable may produce problems, e.g. in ophthalmic preparations, radiological contrast media and of course intravenous solutions.
REFERENCES 1. Messing B, Bories C, Kunstlinger F et al (1983) Does total parenteral nutrition induce gallbladder sludge formation and lithiasis? Gastroenterology, 84, 1012. 2. Roslyn JJ, Pitt HA, Mann LL et al (1983) Gallbladder disease in patients on long-term parenteral nutrition. Gastroenterology, 84, 1486.
3. Roslyn JJ, Berquist WE, Pitt HA et al (1983) Increased risk of gallstones in children receiving total parenterat nutrition. Pediatrics, 71,784. 4. Levenson JL (1983) Biotin-responsive depression during hyperalimentation. J. Parent. Ent. Nutr., 7, 181. 5. Williams DM (1983) Copper deficiency in
Intravenous infusions - solutions and emulsions
Chapter 36
humans. Sem. HaematoL, 20, 118. 6. Bozzetti F, Inglese MG, Terno G e t al (1983) Hypocupremia in patients receiving total parenteral nutrition. J. Parent. Ent. Nutr., 7, 563. 7. Welsmannn K, Kvist N, Kobayasi T (1983) BuUous acrodermatitis due to zinc deficiency during total parenteral nutrition: an ultrastructural study of the epidermal changes. Acta Derrrt Venereol., 63, 143. 8. The TS, Koll6e LA, Boon JM et al (1983) Rickets in a preterm infant during intravenous alimentation. Acta Paediatr. Scand., 72, 769. 9. Worthley LIG, Fishlock RC, Snoswell AM (1983) Carnitine deficiency with hyperbilirubinemia, generalized skeletal muscle weakness and reactive hypoglycemia in a patient on long-term total parenteral nutrition: treatment with intravenous L-carnitine. J. Parent. Ent. Nutr., 7, 176. 10. Ott SM, Maloney NA, Klein GL et al (1983) Aluminium is associated with low bone formation in patients receiving chronic parenteral nutrition. Ann. Intern. Med., 98, 910. 11. Klein G L Ott SM, Alfrey AC et al (1982) Aluminium as a factor in the bone disease of long-term parenteral nutrition. Trans. Assoc. Ar~ Physicians, 95, 155. 12. Bengoa JM, Sitrin MD, Wood RJ et al (1983) Amino acid-induced hypercalciuria in patients on total parenteral nutrition. Am. J. Clin. Nutr., 38, 264. 13. Schultze G, Pommer W, Offermann G e t al (1983) Acute kidney failure in secondary renal oxalosis: additional indications for a causal connection to xylitol infusions, lnfusionsther. Klin. Ernahr., 10, 322. 14. Baley JE, Kliegman RM, Fanaroff AA (1984) Disseminated fungal infections in very lowbirth-weight infants: clinical manifestations and epidemiology. Pediatrics, 73, 144. 15. Jarvis WR, Highsmith AK, Allen JR et al (1983) Polymicrobial bacteremia associated with lipid emulsion in a neonatal intensive care unit.
301
Pediatr. Infect. Dis., 2, 203. 16. Fleer A, Senders RC, Visser MR et al (1983) Septicemia due to coagulase-negative staphylococci in a neonatal intensive care unit: clinical and bacteriological features and contaminated parenterat fluids as a source of sepsis. Pediatr. Infect. Dis., 2, 426. 17. Borten M, Seibert CP, Taymor ML (1983) Recurrent anaphylactic reaction to intraperitoneal dextran 75 used for prevention of postsurgical adhesions. Obstet. Gyn.., 61,755. 18. Reisner LS (1984) Anaphylaxis to intraperitoneal dextran. Anaesthesia, 60, 259. 19. LjungstriSm KG, Renck H, Hedin H e t al ( 1983) Prevent ion of dextran-induced anaphylactic reactions by hapten inhibition. I. A Scandinavian multicenter study on the effects of 10 ml of dextran 1, 15%, administered before dextran 70 or dextran 40. Acta Chit. Scand., 149, 341. 20. Renck H, Ljungstr6m KG, Rosberg B e t al (1983) Prevention of dextran-induced anaphylactic reactions by hapten inhibition. II. A comparison of the effects of 20 ml dextran 1, 15%, administered either admixed to or before dextran 70 or dextran 40. Acta Chit. Scand., 149, 349. 21. Renck H, Ljungstr6m KG, Hedin H e t al ( 1983) Prevention of dextranqndu ced anaphylactic reactions by hapten inhibition. III. A Scandinavian multicenter study on the effects of 20 ml dextran 1, 15%, administered before dextran 70 or dextran 40. Acta Chir. Scand., 149, 355. 22. Ljungstr6m KG, Renck H, Strandberg K et al (1983) Adverse reactions to dextran in Sweden 1970-1979.Acta Chir. Scand., 149, 253. 23. Vervloet D, Senft M, Dugue P e t al (1983) Anaphylactic reactions to modified fluid gelatins. J. Allergy Clin. lmmunol., 71,535. 24. Robinson LA, Braimbridge MV, Hearse DJ (1984) The potential hazard of particulate contamination of cardioplegic solutions. J. Thorac. Cardiovasc. Surg., 87, 48.