588

9. Sloand EM, Pitt E, Chiarello RJ, Nemo GJ. HIV testing: state of the art. JAMA 1991; 266: 2861-66. 10. Palasanthiran P, Ziegler JB, Stewart GJ, et al. Breast feeding during primary maternal human immunodeficiency virus infection and the risk of transmission from mother-to-infant. J Infect Dis (in press). 11. Colebunders R, Kapita B, Nekwei W, et al. Breastfeeding and transmission of HIV. Lancet 1988; ii: 1487. 12. van de Perre P, Simonon A, Msellati P, et al. Postnatal transmission of human immunodeficiency virus type 1 from mother to infant: a prospective cohort study in Kigali, Rwanda. N Engl J Med 1991; 325: 593-98. 13. van de Perre P,Hitimana DG, Simonon A, et al. Postnatal transmission of HIV-1 associated with breast abscess. Lancet 1992; 339: 1490-91. 14. Hira SK, Mangrola UG, Mwale C, et al. Apparent vertical transmission of human immunodeficiency virus type 1 by breast-feeding in Zambia. J Pediatr 1990; 117: 421-24. 15. Krivine A, Firtion G, Cao L, Francoual C, Herion R, Lebon P. HIV replication during the first few weeks of life. Lancet 1992; 339: 1187-89. 16. Newell ML, Peckham CS, Lepage P. HIV-1 infection in pregnancy: implications for women and children. AIDS 1990; 4 (suppl 1):

S111-17. Multicentre Study. Epidemiology, clinical features, and prognostic factors of paediatric HIV infection. Lancet 1988; ii: 1043-45. 18. European Collaborative Study. Risk factors for mother-to-child transmission of HIV-1. Lancet 1992; 339: 1007-12. 19. Hutto C, Parks WP, Lai S, et al. A hospital-based prospective study of perinatal infection with human immunodeficiency virus type 1. J Pediatr 1991; 118: 347-53. 20. Blanche S, Rouzioux C, Guihard Moscato ML, et al. A prospective study of infants born to women seropositive for human immunodeficiency 17. Italian

virus type 1. N Engl J Med 1989; 320: 1643-48. 21. Kind C, Brandle B, Wyler CA, et al. Epidemiology of vertically transmitted HIV-1 infection in Switzerland: results of a nationwide prospective study. Eur J Pediatr 1992; 151: 442-48. 22. Ryder RW, Manzila T, Baende E, et al. Evidence from Zaire that breast-feeding by HIV-1-seropositive mothers is not a major route for perinatal HIV-1 transmission but does decrease morbidity. AIDS

1991; 5: 709-14. 23. Daar ES, Moudgil T, Meyer RD, Ho DD. Transient high levels of viremia in patients with primary human immunodeficiency virus type 1 infection. N Engl J Med 1991; 324: 961-64. 24. Holmberg SD, Horsburgh CR, Ward JW, Jaffe HW. Biologic factors in the sexual transmission of human immunodeficiency virus. J Infect Dis 1989; 160: 116-25. 25. Centers for Disease Control. Recommendations for assisting in the prevention of perinatal transmission of human T-lymphotropic virus virus and type-III/lymphadenopathy-associated acquired immunodeficiency syndrome. MMWR 1985; 34: 721-32. 26. Rogers MF. Breast-feeding and HIV infection. Lancet 1987; ii: 1278. 27. Global Programme on AIDS. Consensus statement from the WHO/ UNICEF Constitution on HIV Transmission and Breast-Feeding. Weekly Epidemiol Rec 1992; 67: 177-84. 28. Victora CG, Smith PG, Vaughan JP, et al. Evidence for protection by breast-feeding against infant deaths from infectious diseases in Brazil. Lancet 1987; ii: 319-22. 29. Ades AE, Parker S, Berry T, et al. Prevalence of maternal HIV-1 infection in Thames Regions: results from anonymous unlinked neonatal testing. Lancet 1991; 337: 1562-64. 30. Heymann SJ. Modeling the impact of breast-feeding by HIV-infected women on child survival. Am J Public Health 1990; 80: 1305-09.

REVIEW ARTICLE Cyclical parenteral nutrition

In most instances, parenteral nutrition has been given as a continuous infusion over 24 h. In 1976, a cyclical protocol for parenteral nutrition was developed. Whilst aminoacids were provided by the usual 24 h infusion, dextrose was infused discontinuously.1 The same discontinuous regimen had been applied previously to patients on home parenteral nutrition; between infusions, venous access was maintained by means of a heparin lock.2 Since then, nocturnal cyclical parenteral nutrition (CPN) with diurnal catheter locking has become common practice among home patients.2.3 Even though studies point to the value of CPN in hospital, it has yet to be widely taken up?°° The advantages of nocturnal CPN are both metabolic and clinical: it closely simulates the normal hormonal and metabolic rhythm of the human feeding pattern,3and it allows patients to enjoy unhindered daytime mobility. This review will focus on the metabolic consequences, indications, complications, clinical benefits, and costeffectiveness of CPN.

Metabolic effects of

cyclical

parenteral

vs

continuous

nutrition

Energy expenditure and nitrogen balance In rats, sequential infusion of aminoacids and energy (fat plus glucose) over 24 h results in higher energy expenditure, lower nitrogen use, and lower weight gain than with continuous infusion. Feeding-induced thermogenesis

is a more important factor in overall energy expenditure than physical activity. These and other observations have led to the conclusion that in laboratory animals, optimum gain in lean body mass results from continuous provision of both aminoacids and energy substrates. Studies in human beings have yielded less definite results. In undernourished non-stressed or mildly hypermetabolic adults and children, CPN led to similar levels of energy expenditure, together with gains in lean body mass and visceral protein similar to those found with continuous parenteral nutrition, even though water, fat, and glycogen were halved. Although not as effective as standard parenteral nutrition in increasing lean body mass in hypermetabolic adults receiving steroids or radiotherapy, CPN does lead to an equivalent gain of visceral protein, but with lower water and fat deposits.6 These effects are

deposits

Gastroenterology and Nutritional Support Unit (C Matuchansky, MD, B Messing, MD), and Centre for Home Parenteral Nutrition (M. Beliah), Saint-Lazare Hospital, Paris, France; Gastroenterology and Nutrition Unit, and Center for Home Parenteral Nutrition (K N. Jeejeebhoy, PhD), SaintMichael Hospital, Toronto, Canada; Gastroenterology Unit and Centre for Home Parenteral Nutrition (P. Beau, MD), Jean Bernard Hospital, Poitiers, France; and Department of Medicine (J. P. Allard, MD), Toronto General Hospital, Toronto, Canada. Correspondence to Prof C Matuchansky, Gastroenterology and Nutritional Support Unit, Saint-Lazare Hospital, 107 bis rue du Faubourg Saint Denis, 75010 Paris, France.

ADDRESSES

589

comparable with those found in stressed patients receiving cyclical (nocturnal) rather than continuous enteral feeding. The differences between CPN and continuous parenteral nutrition are not due to differences in physical activity, but are caused directly by the sequence and duration of nutrient delivery.3 Metabolic profile

Early work showed that CPN was the optimum technique simulating the postabsorptive state, in which lipolysis was favoured and visceral protein was preserved.1 However, patients in this study received a continuous infusion of aminoacids, together with cyclical hypertonic dextrose. Thus, a convincing parallel with patients on nocturnal CPN for

could not be drawn. Subsequent studies have shown a circadian rhythm in glycogen and fat use with CPN but not with continuous parenteral nutrition. With the latter technique there is a predominance of carbohydrate oxidation and lipogenesis.3,8,9Fat oxidation during a 3 h lipid infusion is more pronounced with CPN.8 This finding is consistent with the experimental observation that the liver of continuously fed rats contains 2-5 times more total fat and triglycerides than that of rats receiving cyclical nutrition. Compared with continuous parenteral nutrition, CPN increases insulin secretion in proportion to the rate of infusion without apparent changes in counter-regulatory hormones. During CPN, the observed decrease in highaffmity receptors to insulin has been interpreted as reflecting receptor downregulation by hyperinsulinaemia. Halving the rate of infusion for 30 min results in a rapid decline in plasma insulin, which reaches pre-infusion values 30 min after cessation.1O 24 h insulin secretion in adults, estimated by urinary C-peptide excretion, is similar to that with continuous parenteral nutrition.8 In patients with insulin and non-insulin-dependent diabetes who are on CPN, continuous subcutaneous insulin infusion by a portable pump has been used successfully," preventing both the hypoglycaemia following morning interruption of parenteral nutrition and the hyperglycaemia or ketosis that

develop during daytime fasting. Finally, CPN mimics, at least partly, the normal circadian rhythm of feeding. CPN results in higher rates of lipid oxidation and lower rates of lipid storage than continuous parenteral nutrition because it produces lower plasma insulin concentrations and insulin/glucagon ratios, which are the major signals for lipolysis.3 Patients on CPN whose caloric and nitrogen needs are not met during feeding can be given additional substrates, such as aminoacids and/or fat, which do not disturb the postabsorptive hormonal state. In practice, about half the patients on home CPN are entirely dependent on parenteral nutrition and fail to receive any nutrients when not infused. Other patients eat meals when not being infused; this allows them to reduce the amount of nutrients from parenteral nutrition.9 A possible drawback of nocturnal CPN is that calories and nitrogen are administered when the patient is least active, resulting in decreased nitrogen incorporation into muscle tissue and a suboptimum nitrogen balance. However, diurnal infusion restricts physical activity and its potential benefits. Since the long-term consequences of high nocturnal msulin concentrations are not known, they can be avoided by reducing the supply of glucose, adding lipid, and tapering the infusion before cessation of parenteral nutrition. These measures also minimise the energy cost of dietary thermogenesis.

may

Lipoproteins As compared

with

eating

CPN does not significantly circadian rhythms of total,

three meals a day, nocturnal change the characteristics of

high-density lipoprotein, and low-density lipoprotein (LDL) cholesterol, apolipoproteins, and total serum proteins. However, peak triglyceride and free-fatty-acid concentrations shift from daytime to night;12 the usual morning blood sample may not be reliable, and should be delayed for up to 2 h after the cessation of infusion. During CPN, there is a correlation between the rate of phospholipid infusion and the concentration of serum lipoprotein X (an abnormal lipoprotein of similar density to LDL)." Lipoprotein X may be involved in lipid overloading of the reticuloendothelial system and hepatocytes, thereby leading to hepatic phospholipidosis. These findings suggest that the rate of phospholipid infusion should not exceed 15

mg/kg per h.13 Water and minerals In adults on CPN, 2500-3000 ml may be infused overnight without substantial haemodilution, although up to 50% of subjects have increased nocturnal diuresis. CPN does not significantly increase the mean glomerular filtration rate nor the urinary sodium excretion,’4 but does cause hypouricaemia because of an increased urate

clearance. 15 The faster the rate of CPN infusion, the greater the loss of urinary calcium. Since the increased loss occurs mainly during infusion it is likely that this increases the filtered calcium load. 16 Daily fractional calcium excretion is unchanged. The urinary calcium losses decrease with time and there is no difference, compared with continuous parenteral nutrition, in losses of calcium, magnesium, or phosphorus in patients on long-term CPN.14 Because patients receiving both CPN and oral nutrients have a lower urinary excretion of calcium and magnesium than patients receiving only CPN, the role of gastrointestinal factors in mineral excretion, including adaptation, may be important. Other factors possibly increasing calciuresis in patients on CPN are: increased serum insulin concentrations, volume expansion, sulphate burden, and proton load. The role of the proton load in increasing calciuresis is countered by acetate infusionP

Physical activity and CPN Physical activity while receiving CPN should benefit body protein metabolism, although the amounts and patterns of exercise required, and their influence during long-term intravenous feeding, are not yet known. A multicentre survey of 194 patients on home parenteral nutrition from nine European countries showed that 87% received CPN at night and 2% by day, while 11% had continuous infusion.18 In the USA questionnaire, patients on home parenteral nutrition preferred the big-bag to the multiple-bottle technique.19 Exercise has variable effects on protein metabolism,

depending on its duration and on the nutritional status of the patient. Oxidation of branched-chain aminoacids is enhanced. In rats on continuous parenteral nutrition, exercise has no effect on nitrogen balance, but increased aminoacid turnover in skeletal muscle is found. In human beings in the fed state, exercise increases protein degradation and decreases protein synthesis;2° the initial increase in nitrogen excretion is followed by adaptation over three weeks, which may indicate that repeated regular exercise limits the effect on aminoacid catabolism. Glucose lowers

590

INDICATIONS FOR CYCLICAL HOME PARENTERAL NUTRITION IN ADULTS

*Excludlng Crohn’s and radiation ententls, tlncludmg systemic sclerosis OASIS, Oley-ASPEN Information System, HPN, Home parenteral nutntlon

whole-body protein degradation and may suppress the exercise-induced nitrogen loss.2° In children with proteinenergy malnutrition, moderate but continuous exercise enhances the metabolism of dietary protein and growth. retention and improvement of nutritional status related to physical activity has been reported in patients on enteral nutrition, astronauts after space flights, and immobilised patients after limb injury.

Nitrogen

CPN is also associated with a low rate of catheter infection and occlusion: 0-45 and 0-11 per year, respectively, in hospital patients (unpublished data). Mixtures of lipids with other nutrients often result in catheter obstruction, which can be avoided by a separate lipid infusion.26 Cyclical feeding does not seem to contribute significantly to catheter obstruction. Similarly, bacterial endocarditis and superior vena caval thrombosis do not seem to be related to cyclical

feeding. Indications be instituted in all cases where prolonged nutrition in hospital and at home is required. The parenteral indications for home CPN are shown in the table.18,19,21-23 In adult patients in hospital, CPN has also been given in the treatment of severe gastrointestinal diseases. The indications in 200 consecutive cases (unpublished data of the authors) were: Crohn’s disease 19%, short-bowel syndrome 10%, radiation enteritis 17%, malabsorption and pancreatitis 25%, malignant disease 17%, and others 12%. In paediatric patients, CPN has been given for: short-bowel chronic intestinal syndrome, pseudo-obstruction, intractable diarrhoea because of resistant villous atrophy or and disease. microvillous Crohn’s atrophy, Contraindications to CPN include patients requiring over 3500 ml per day, and those with congestive heart failure or other conditions in which a fluid load cannot be tolerated.

CPN

can

Complications Catheter-related and metabolic

complications in home CPN patients have been publishedbut connections between the frequency of these complications and the process of cyclical feeding have not been established. The importance of a team-approach to the management of CPN patients has been widely discussed.2,24 Catheter and implanted reservoir-related complications In CPN patients, the mean yearly frequency of catheterrelated infection was 0-38 and 0-45 in two multicentre surveys from Europe18 and the UK and Ireland, respectively.21 There is an inverse relation between the frequency of catheter-related sepsis and the experience of the team managing the patient. In a review of the national status of home parenteral nutrition in the USA, the annual frequency of catheter-related sepsis was 0-37." The yearly frequency of catheter occlusion, in the two European multicentre surveys, was 0-18 and 032.Z1 Catheter migration was reported Oil to 0-12 times per year. The frequency of infectious complications was similar when comparing the implanted reservoir with the external catheter, although clotting is more common with the latter and skin erosions are seen only with the former.25

If there is no infection at the site of catheter entry, of catheter-related sepsis by a local antibiotic-lock technique avoids catheter removal in 93% of cases.27 This technique is feasible only with CPN, since the antibioticlock is applied during the hours when the patient is not being fed.

treatment

Thrombosis and haematological disorders Venous thrombosis, largely subclinical, takes place in up 60% of infants of low-birth weight on total parenteral nutrition. Only 10% of these patients are clinically recognised.28 Complications associated with thrombosis range from catheter occlusion to combined superior and inferior vena caval occlusion. Adult patients on parenteral nutrition may express higher spontaneous monocyte macrophage procoagulant activity than normal controls 28 In-vitro studies have demonstrated the potential thrombogenicity of parenteral solutions: dextrose and full-strength aminoacid solutions increased monocyte macrophage procoagulant activity, which is inhibited by lipid emulsion.28 In children receiving CPN, infusion of lipid emulsion may be associated not only with the bleeding tendency noted in the fat-overload syndrome, but also with reduced thrombocytopenia, platelet life-span, and medullary sea-blue histiocytes,29 which are all features associated with activation of the monocyte/macrophage system. No connection has been established between such findings and cyclical feeding. to

Liver and biliary tract While hepatic steatosis may be reduced by infusing lipids energy source, the suggestion that CPN compared with continuous parenteral nutrition is better tolerated by the liver1 has yet to be proven, even if cholestasis was noted less frequently. Although liver function test abnormalities during CPN may range from 15--40% of adult patients treated for over 3 weeks, half of these decrease either partly or totally by the end of treatment. The incidence of hepatic dysfunction can be reduced by infusing less energy, by separating lipids from other nutrients, and by administration of oral antibiotics. 30 as an

591

Long-term CPN that supplements oral intake does not reduce appetite in patients with short-bowel syndrome.3o Since patients with a short bowel may absorb up to 60% of their oral intake, it is reasonable to reduce the amount of energy infused during home CPN.31 An increased incidence of hepatic cholestasis is found in premature infants, newborns, and young children on parenteral nutrition, and there is a risk of developing fulminant hepatic failure. Presence of liver disease does not seem to be related to cyclical feeding in either adults or children. 30 In combination with total bowel rest, parenteral nutrition seems to be the main factor in the accelerated formation of gallbladder sludge and stones.32 But neither the technique of infusing parenteral nutrition (continuous or cyclical) nor nutrient composition is important. Stimulating gallbladder contraction through oral feeding, with subsequent endogenous cholecystokinin release, seems to be the best way of precluding sludge formation. Since CPN was initiated 20 years ago, the frequency of gallbladder complications has declined from over 40% to under 15%.32 Today, the frequency of symptomatic gallstone disease during home CPN does not exceed that observed in patients with resected or diseased ileum. Bone Metabolic bone

disease, whose features may include calciuresis, low-to-normal

increased

hypercalcaemia, plasma 25 and 1,25 dihydroxyvitamin D concentrations and low-to-normal parathyroid hormone values has been reported in patients receiving home CPN,33,34 though the remain unclear. Possible factors include: discontinuous delivery of aminoacids, aluminium toxicity, and hypersensitivity to vitamin D. During longer term CPN, calcium loss from bone may be limited by oral feeding (unpublished data of the authors).

exact causes

Clinical benefits Whatever the benefits of increased physical activity in CPN patients, one clear advantage of cyclical feeding is an improved quality of life .2,4,19,22 Diurnal interruption of parenteral nutrition may also favour oral intake.9 Five-year life expectancy of non-cancer adult patients on home CPN is at least 65 %22 and may be even higher. About 65% of patients on home CPN are fully rehabilitated. 1821 The length of hospital stay for recovery from the underlying illness, as well as training, ranges from 3 weeks to 3 months.20 Finally, there is a clear link between the healing process and patients’ ability to participate in their own care.

Cost- utility analysis An analysis of the home CPN programme at the Toronto General Hospital has shown that it resulted in net savings of nearly$20 000 per patient in 1986. Survival, adjusted for quality of life, increased by 3-3 years when compared with the alternative of treating these patients in hospital with intermittent

nutritional

support

when

needed.22

Additionally, the cost-utility of long-term CPN compared favourably with other health-care programmes, especially when CPN was used in patients with gastrointestinal failure associated with non-malignant conditions. Reassessment of the needs for parenteral nutrition because of changes in the patient’s underlying condition, the adaptation of their bowel, and technological advances have a positive impact on quality of life and cost.35

patients in hospital, the value of short-term (1 to 2 weeks) perioperative continuous parenteral nutrition was recently questioned. However, this caveat does not preclude the recommendation of CPN for patients with gastroentrological disease who are obliged to undergo longer For

of inadequate oral intake support is impossible.36

phases

at

times when enteral

We thank Dr J. Barbier, Dr M. Carretier, Mrs J. Chassin, Dr P. Courtois, Dr P. Hautefeuille, Dr M. Morichau-Beauchant, Dr J. C. Rambaud, Dr J. L. Terrier, Dr P. Valleur, and all the physicians, nurses, and pharmacists who work in our nutritional support teams.

REFERENCES 1. Maini B, Blackburn GL, Bistrian BR, et al. Cyclic hyperalimentation: an optimal technique for preservation of visceral protein. J Surg Res 1976; 20: 515-25. 2. Kennedy G, Jeejeebhoy KN. Home parenteral nutrition. In: Jeejeebhoy KN, ed. Total parenteral nutrition in the hospital and at home. Boca Raton: CRS Press, 1983: 87-122. 3. Putet G, Bresson JL, Ricour C. Nutrition parentérale exclusive chez l’enfant. Influence de l’apport continu ou cyclique sur l’utilisation des nutriments. Arch Fr Pediatr 1984; 41: 111-15. 4. Matuchansky C, Morichau-Beauchant M, Druart F, Tapin J. Cyclic (nocturnal) total parenteral nutrition in hospitalized adult patients with severe digestive diseases. Gastroenterology 1981; 81: 433-37. 5. Martins FM, Sandberg G, Ekman L, Lindmark L. Metabolic response of simultaneous versus sequential intravenous administration of amino acids and energy substrates to rats. Am J Clin Nutr 1985; 42: 61-68. 6. Messing B, Pontal PJ, Bernier JJ. Metabolic study during cyclic total parenteral nutrition in adult patients with and without corticosteroidinduced hypercatabolism: comparison with standard total parenteral nutrition. J Parenter Enter Nutr 1983; 7: 21-25. 7. Lerebours E, Rimbert A, Hecketsweiler B, Hellot MF, Denis P, Colin R. Comparison of the effects of continuous and cyclic nocturnal parenteral nutrition on energy expenditure and protein metabolism. J Parenter Enter Nutr 1988; 12: 360-64. 8. Just B, Messing B, Darmaun D, Rongier M, Camillo E. Comparison of substrate utilization by indirect calorimetry during cyclic and continuous total parenteral nutrition. Am J Clin Nutr 1990; 51: 107-11. 9. Just B, Messing B, Darmaun D. Oral nutrition in patients receiving home cyclic parenteral nutrition: pattern of substrate utilization. Am J Clin Nutr 1991; 54: 560-64. 10. Byrne WJ, Lippe BM, Strobel CT, Levin SR, Ament ME, Kaplan SA. Adaptation to increasing loads of total parenteral nutrition: metabolic, endocrine, and insulin receptor responses. Gastroenterology 1981; 80: 947-56. 11. Beau P, Marechaud R, Matuchansky C. Cyclic total parenteral nutrition, diabetes mellitus, and subcutaneous insulin pump. Lancet 1986; 327: 1272-73. 12. Matuchansky C, Fabre J, Guillard O, Morichau-Beauchant M, Reinberg A. Effects of cyclic (nocturnal) total parenteral nutrition and continuous enteral nutrition on circadian rhythms of blood lipids, lipoproteins and apolipoproteins in humans. Am J Clin Nutr 1985; 41: 727-34. 13. Messing B, Peynet J, Poupon J, et al. Effect of fat-emulsion phospholipids on serum lipoprotein profile during one month of cyclic total parenteral nutrition. Am J Clin Nutr 1990; 52: 1094-100. 14. Lipkin EW, Ott SM, Chesnut CH, Chait A. Mineral loss in the parenteral nutrition patient. Am J Clin Nutr 1988; 47: 515-23. 15. Morichau-Beauchant M, Beau P, Druart F, Matuchansky C. Effects of prolonged, purine-free total parenteral and enteral nutrition on urate homeostatis in man. Am J Clin Nutr 1982; 35: 997-1002. 16. Wood RJ, Bengoa JM, Sitrin MD, Rosenberg IH. Calciuretic effect of cyclic versus continuous total parenteral nutrition. Am J Clin Nutr 1985; 41: 614-19. 17. Berkelhammer CH, Wood RJ, Sitrin MD. Acetate and hypercalciuria during total parenteral nutrition. Am J Clin Nutr 1988; 48: 1482-89. 18. Messing B, Landais P, Goldfarb B, Irving M. Home parenteral nutrition in adults: a multicentre survey in Europe. Clin Nutr 1989; 8: 3-9. 19. Howard L, Heaphey LL, Timchalk M. A review of the current national status of home parenteral and enteral nutrition from the provider and consumer perspective. J Parenter Enter Nutr 1986; 10: 416-24. 20. Millward DJ, Davies CTM, Halliday D, Wolman SL, Matthews D, Rennie M. Effects of exercise on protein metabolism in humans as explored with stable isotopes. Fed Proc 1982; 41: 2686-91. 21. Mughal M, Irving M. Home parenteral nutrition in the United Kingdom and Ireland. Lancet 1986; 328: 383-86. 22. Detsky AS, McLaughlin JR, Abrams HB, et al. A cost-utility analysis of the home parenteral nutrition program at Toronto General Hospital: 1970-1982. J Parenter Enter Nutr 1986; 10: 49-57.

592

23. Howard L, Heaphey L, Fleming CR, Lininger L, Steiger E. Four years of North Amencan registry home parenteral nutrition outcome data and their implications for patent management. J Parenter Enter Nutr

1991; 15: 384-93. MC, Saint-Aubert B, Jonquet O, Astre C, Joyeux H. Ambulatory home total parenteral nutrition. J Parenter Enter Nutr 1987; 11: 475-79. 25. Howard L, Claunch C, McDowell R, Timchalk M. Five years of experience in patients receiving home nutrition support with the implanted reservoir: a comparison with the external catheter. J Parenter Enter Nutr 1989; 13: 478-83. 26. Beau P, Matuchansky C. Lipid delivery and catheter obstruction during cyclic total parenteral nutrition. Lancet 1987; 330: 1095-96. 27. Messing B, Man F, Colimon R, Thuillier F, Beliah M. Antibiotic-lock technique is an effective treatment of bacterial catheter-related sepsis during parenteral nutrition. Clin Nutr 1990; 9: 220-25. 28. Wakefield A, Cohen Z, Craig M, et al. Thrombogenicity of total

24. Gouttebel

29.

parenteral nutrition solutions: I. Effect on induction of monocyte/ macrophage procoagulant activity. Gastroenterology 1989; 97: 1210-19. Goulet O, Girot R, Maier-Redelsperger M, Bougle D, Virelizier JL, Ricour C. Hematologic disorders following prolonged use of

intravenous fat emulsions in children. J Parenter Enter Nutr 1986; 10: 284-88. 30. Messing B, Colombel JF, Heresbach D, Chazouilleres O, Galian A. Chronic cholestasis and macronutrient excess in patients treated with prolonged parenteral nutrition. Nutrition 1992; 8: 30-36. 31. Messing B, Pigot F, Rongier M, Morin MC, Ndeindoum U, Rambaud JC. Intestinal absorption of free oral hyperalimentation in the very short bowel syndrome. Gastroenterology 1991; 100: 1502-08. 32. Messing B. Gallbladder sludge and lithiasis: a complication of bowel rest. Nutrition 1990; 6: 190-91. 33. Vernejoul de C, Messing B, Modrowski D, Bielakoff J, Buisine H, Miravet L. Multifactorial low remodeling bone disease during cyclic total parenteral nutrition. J Clin Endocrinol Metab 1985; 60: 109-13. 34. Shike M, Shils ME, Heller A, et al. Bone disease in prolonged parenteral nutrition: osteopenia without mineralization defect. Am J Clin Nutr 1986; 44: 89-98. 35. Baptista RJ, Lahey MA, Bistnan BR, et al. Periodic reassessment for improved, cost-effective care in home total parenteral nutrition: a case report. J Parenter Enter Nutr 1984; 8: 708-10. 36. Detsky AS. Parenteral nutrition: is it helpful? N Engl J Med 1991; 325: 573-75.

PUBLIC HEALTH Rapid disappearance of Haemophilus influenzae type b meningitis after routine childhood immunisation with conjugate vaccines Mortality from meningitis caused by Haemophilus influenzae type b (Hib), a disease that affects mainly infants and young children, can reach 5% in industrialised countries and ten times that in nonindustrialised countries. To determine the efficacy of vaccination against Hib, we carried out a retrospective survey of the incidence of Hib meningitis over five decades in the Greater Helsinki area of Finland, where all children with bacterial meningitis are treated in one of three centres. Except for a meningococcal epidemic in the early 1970s, Hib was the leading cause of childhood bacterial meningitis until the Hib conjugate vaccines changed the picture profoundly. In 1986-87 the toxoid polysaccharide-diphtheria was (PRP-D) given experimentally

conjugate

50% of infants. In 1988-89 all infants were vaccinated, 50% with PRP-D, 50% with another conjugate vaccine, to

the

oligosaccharide-CRM197 protein conjugate (HbOC). Since 1990 a third conjugate vaccine, the polysaccharide-tetanus toxoid (PRP-T), has been

administered routinely to all infants. The vaccines were administered at age 3-6 months, with a booster dose at 14-18 months. In the first 5 years of the Hib vaccination programme the number of cases of Hib meningitis in children aged 0-4 years fell sharply, from 30 in 1986 (the first year of the programme) to none in 1991. The decline contrasts sharply with the rising trend up to the mid 1980s. Vaccination seems to be the only explanation for the observed change in the epidemiology of Hib meningitis.

Introduction The "second

generation" vaccines against Haemophilus influenzae type b (Hib), in which the hapten Hib polysaccharide or oligosaccharide is conjugated to a carrier protein, induce immunogenicity in infants more effectively than their predecessor, the polysaccharide vaccine, which protected only children aged over 18 months,1 who are already past the age of maximum risk for Hib disease. Four types of Hib conjugate vaccines are available: the first, polysaccharide-diphtheria toxoid conjugate (PRP-D, ’ProHIBIT’, Connaught Laboratories, Inc, USA)2-3 using conventional diphtheria toxoid as the carrier protein; the second, CRM197 mutant diphtheria toxoid (HbOC, ’HibTITER’, Praxis-Lederle, USA);4 the third, outer membrane protein of the group B meningococci (PRPOMP, ’PedvaxHIB’ Merck Sharp & Dohme, USA);5 and the fourth, tetanus toxoid (PRP-T, ’Act-HIB’, Pasteur Merieux Serums & Vaccins, France). All are highly immunogenic in infancy, and field trials have established the clinical efficacy of the first three.’-9 Finland has participated in those trials.7 A sixth of Finland’s population (5 million) lives in the Greater Helsinki area, where all children with bacterial meningitis (BM) are treated in three centres, Aurora City Hospital, the Children’s Hospital, University of Helsinki, and Jorvi Hospital. This concentration of cases provided an opportunity to determine the efficacy of vaccination in a long-term study of the incidence of Hib meningitis.

ADDRESSES: Children’s Hospital, University of Helsinki, Helsinki, Finland (H Peltola, MD, T. Kilpi, MD, M. Anttila, MD); National Public Health Institute, Helsinki (H. Peltola); Aurora City Hospital, Helsinki (T. Kilpi) Correspondence to Dr Heikki Peltola, Children’s Hospital, University of Helsinki, 11 Stenbackinkatu, SF-00290 Helsinki, Finland.