Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989 11

A rational approach to the nutritionalcare of patients with cystic fibrosis

P R Durie MD FRCP(C) P B Pencharz MB ChB FRCP(C) Divisions of Gastroenterology and Nutrition,Department of Pediatrics, and the Research Institute, The Hospital for Sick Children; and the Department ofPediatrics, University of Toronto, Ontario, Canada

Keywords: cystic fibrosis; growth; malnutrition; survival; energy

IntroductionFor many years, chronic undernutrition withsignificant weight retardation and linear growthfailure, has been recognized as generalized problemof most large patient populations with cystic fibrosis(CF). Despite acknowledgment of its seriousness,chronic malnutrition was felt to be an inherentconsequence of the disease; others argued thatmalnutrition was a natural consequence of physio-logical adaptation to advanced pulmonary disease. Anumber of early studies of CF patients, however,showed a good correlation between the severity ofmalnutrition and an accelerated decline in pulmonaryfunction which in turn adversely affected overallsurvival"2. Many authors have speculated that thesetwo factors are causally associated, but it is not clearwhether prevention of malnutrition and growthfailure will ameliorate the rate ofprogression oflungdisease and improve survival. In recent years therehas been renewed interest in evaluating the multipleinterdependent variables giving rise to chronicmalnutrition and growth failure. In a number of CFcentres, nutritional support is now viewed as anintegral part of the care of the patients with cysticfibrosis and aggressive programmes have beeninstituted to prevent malnutrition.

It is now recognized that most growth problems inpatients with CF are due to unfavourable energybalance rather than to an inherent factor of thedisease itself. Roy et al.3 drew attention to a numberof reports from the Cystic Fibrosis Clinic in Torontoindicating that by far the majority of CF patientsattending the Toronto clinic closely conformed to thenormal distribution for growth45. Cross-sectionalgrowth data from the Toronto clinic showed a normaldistribution of height percentiles in males andfemales. In females, however, particularly afteradolescence, weight distribution was skewed towardsthe lower centiles, but this was less evident thansimilar data obtained from other centres. In acomparative study of two CF clinic populations ofsimilar size and age distribution (Toronto and Boston),Corey et al.6 found a marked difference in medianage of survival between the two centres (Table 1).Median survival in Boston (21 years) was markedlyworse than in Toronto (30 years); furthermore, after10 years of age there was a dramatic separation insurvival curves between the two centres. Pulmonaryfunction was no different in the two clinic populations.Patients in the Toronto clinic, however, were tallerthan those in the Boston clinic, and males in Torontowere heavier. It is noteworthy that, with the exceptionof nutritional management of their patients, the

Table 1. Characteristics of cystic fibrosis clinic populationsin Boston and Toronto (1982)0

Boston Toronto

Number of patients 499 534Male/female (%) 57/43 58/42Age: mean±SD (years) 15.9+9.6 15.2+8.3

range (years) 0-45 0-43Median survival 21 3050% (years)

*Adapted from Corey et al.5

general approach to patient care, particularly pul-monary care, was similar in the two clinics. It wassuggested that the improved survival in the TorontoCF population could be attributed to superior nutri-tional status.A review of existing dietary practices in the two

clinics revealed a striking difference in philosophy.Standard practice in Boston7 closely resembled thetraditional approach in most centres throughout theworld. This approach involved prescribing a low-fat,carbohydrate-rich diet. It was reasoned that reductionin dietary fat would improve bowel symptoms andreduce stool bulk. Recognizing the problem ofmaldigestion and absorption of long-chain triglyceridesmany centres advocated use of artificial diets withprotein hydrolysates and substitution oflong-chain fatwith medium-chain triglycerides (MCT)8. However, aseries of reports showed no long term benefits togrowth when protein hydrolysates and MCT wereused as supplements or substitutes9. Fortunately,these supplements are now hardly ever used throughreasons of cost, poor compliance and unpalatability.Too often, the net effect of a restrictive, unpalatablediet was to present young CF patients with tastelesschoices and to exclude them from the many energy-rich foods that are part of a 'normal' Western diet.Chronic malnutrition from reduced energy intakeappears to have been an unfortunate, yet deliberateiatrongenic effect in most CF programmes worldwide.Since the early 1970s, the Toronto group advocated

a calorically enriched diet by encouraging notrestricting fat intake and encouraging additionalenzyme supplements to enhance digestionl'0"'.Because fat is the most energy rich, economical andappetising energy source, patients were encouragedto eat larger portions than their peers; to add fat inthe form of butter or untrimmed meat, and to havehigh calorie snacks between meals and before bed. Fat

12 Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989

malabsorption occurred, but with additional pan-creatic enzyme supplements, net absorbed energyimproved, resulting in better growth. It is gratifyingthat in recent years increasing numbers of clinicsworldwide have accepted a similar philosophy for thenutritional care oftheir patients. Thus, it appears thatthe potential for growth among patients with CFclosely resembles that of the normal population.

Pathogenesis of malnutritionA variety of complex related and unrelated factorsmay give rise to the progression ofenergy imbalancein patients with CF. The net effect on growth potentialvaries considerably, according to marked differencesin disease severity from patient to patient, and withdisease progression. Specifically, energy imbalanceresults from three interrelating factors: increasedenergy losses, reduced energy intake and increasedenergy expenditure (Figure 1).

Energy lossesFecal nutrient losses from maldigestion/malabsorptionare known to contribute to energy imbalance. Only1-2% of total pancreatic capacity for secreting enzymeis required to prevent maldigestion12, and yet in themajority ofCF patients a degree ofpancreatic failureis present at diagnosis. In those who exhibit maldi-gestion, very good correlations exist between residualpancreatic function (colipase secretion) and fat mal-absorption up to daily fat losses of approximately 40%of intake (Figure 2). Patients with documentedsteatorrhoea, therefore, appear to have small- butvarying degrees of residual pancreatic function andthis observation at least partially explains why somepatients with pancreatic insufficiency appear to digestnutrients better than others when given pancreaticenzyme supplements with meals. Other factorsmay contribute to the problem of maldigestion.' Inthe absence of adequate pancreatic bicarbonatesecretion13, gastric acid entering the duodenum maylower intestinal pH until well into the jejunum.Pancreatic lipase is readily denatured below pH 2and even if not denatured, enzymatic activity isconsiderably reduced at a low pH. Bile acids arereadily precipitated in an acid milieu'4, and duodenalbile acid concentration may fall below the criticalmicellar concentration, thereby exacerbating fatmaldigestion. Precipitated bile salts also appear to belost from the enterohepatic circulation in greaterquantities, which reduces the total bile salt pool andalters the glycocholate : taurocholate ratio. Oraltaurine supplements have been reported to benefitsome patients'5. This tendency is exacerbated by the

ENERGY-,..I ;* )

,,,Bd,. .e <'n ';- _i'"- .'* .'.v;f -.;:,*:et,ion.r>si:= r

| ; i xr4 4|41tWpE

* ,,, S tt i ¢~t 4 0 t-/| XiA

e ,/ 4 * ~~~~~jsiFf )+s

Figure 1. Factorsproducing energy imbalance in patients withcystic fibrosis (adapted from Roy et al.3)

100 -

80 -

60-

40 -

20-

0

C

LA.

-J

LL

10 -

8 -

6-

4 -

2 -

1 -

0

A0

0

00 \\2

0

0

S

0I I Ir I -

0 40 80 120 160

COLIPASE units/kg/h

Figure 2. Comparison offecal fat excretion (% offat intake)with pancreatic colipase secretion in 28 patients with severepancreatic dysfunction. Open circles; control subjects. Closedcircles; patients with cystic fibrosis. Closed triangles; patientswith Shwachman syndrome- Open square; patient withcongenital pancreatic hypoplasia. (r=-0.92, P <0.001)Reprinted with permission from Gaskin et al.'2Gastroenterology 86: 1-7, copyright 1984 by the AmericanGastroenterological Association

binding of bile salts to unabsorbed protein or neutrallipid. Excessive mucus, with altered physicalproperties, may have a deleterious effect on thethickness of intestinal unstirred layer, furtherlimiting nutrient absorption. Despite improvementsin the enzymatic activity and intestinal delivery ofingested pancreatic enzymes, many patients continueto have severe steatorrhoea and azotorrhoea.

Energy intakeThere has been very little documentation of actualenergy intakes in patients with;CF. It is widelyaccepted that energy intake should exceed normalrequirements, and crude estimates have suggestedthat patients require 120-150% of the recommendeddaily allowance (RDA)3. However,, when we accu-rately evaluated nutrient intakes in a group ofhealthy adolescents, we were surprised to learn thatenergy intakes were frequently in the range of80-100% of RDA for age, body weight and sex16.Those with normal growth percentiles for height andweight did show higher energy intakes than thepatients with growth retardation. Furthermore,other CF centres who have since encouraged moreliberal attitudes to dietary fat intake have noted acorresponding improvement in energy intake andgrowth'7"8Some patients with cystic fibrosis are particularly

prone to complications that limit appetite andoral intake. Oesophagitis is quite common in

- MddmwptionInuM

400 we I Wo DIWON wM dole cii

- LungdhmM-ohmicifttft lm-work of brooMp

x it ohltdw?

Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989 13

patients with advanced pulmonary disease and isfrequently associated with pain, anorexia and vomitingfollowing bouts of coughing'9'20. The distal intestinalobstruction syndrome (meconium ileus equivalent), anunusual form of subacute obstruction within the distalileum and proximal colon2l, is particularly prevalentin adolescents and adults with pancreatic failure; itfrequently causes recurrent, crampy abdominal painand patients often find that the symptoms areexacerbated by eating. Other abdominal symptoma-tology, including extrahepatic biliary obstruction,cholangitis, advanced liver disease and severeconstipation are less likely to be associated withdecreased dietary intake.Acute respiratory exacerbations are common causes

of restricted oral intake, often resulting in acuteweight loss. In patients with mild pulmonary disease,there is rapid catch-up in weight following improve-ment in respiratory symptoms, but in the terminalstages of pulmonary disease, chronic anorexia is aconsistent feature. Further, patients with a severechronic disease may be prone to bouts of clinicaldepression, which in the adolescent or adult mayinduce severe anorexia.

Energy expenditureConsiderable data are emerging to suggest thatpatients with CF have increased rates of energyexpenditure, but there is disagreement regarding thefundamental mechanisms. In 1984, Pencharz et aL22,evaluated the relationship between heart rate andenergy expenditure using an exercise cycle withgraded workloads and simultaneous measurementsofoxygen consumption and carbon-dioxide productionusing a closed-circuit indirect calorimeter and heart-rate telemetry. Subjects were malnourished and werereceiving nutritional rehabilitation by continuousnasogastric tube feeding with a semi-elemental diet.Absorbed energy intake was calculated by subtactingstool energy content from the energy content of thefeed. The energy needs ofthe patients were shown tobe 25-80% higher than healthy individuals of thesame age, sex and size. It was hypothesized thatenergy expenditure increased due to increased workofbreathing in patients with advanced lung disease.Consequently, a patient with advanced lung diseasemight not be able to ingest sufficient calories to meetenergy needs, resulting in energy imbalance andweight loss. In a subsequent study by the samegroup, resting energy expenditure was measured bycontinuous computerized open-circuit indirect calori-metry in 71 patients (8.9 to 35.5 years) who were notsuffering from an acute respiratory infection2s. Nutri-tional status and pulmonary function were studiedsimultaneously. Resting energy expenditure wasfound to be above normal (range 95-153%) ofpredictedvalues for age, sex and weight, and was negativelycorrelated with pulmonary function and nutritionalstatus (percentage of body fat). In addition, inagreement with previous observations of other2,pulmonary function was positively correlated withnutritional status. These findings have since beenconfirmed by Buchdahl et aL24 who demonstratedthat patients with CF had resting energy expenditureof 9%o above body weight and 7% above lean body mass,respectively, in comparison with healthy controls.Although patients with severe pulmonary diseasetended to have higher energy expenditure, increasedlevels of energy expenditure were also seen in those

with normal pulmonary function and in 4/5 patientswith above average weight for expected age.This observation is of great interest because

laboratory studies have suggested that the primaryCF defect might be associated with an energy-requiring mechanism at the intracellular level,possibly within mitochondria. Feigal and Shapiro25measured 02 consumption in cultured fibroblastsfrom CF homozygotes and heterozygotes. The rate of02 consumption exceeded control fibroblasts by 2times in homozygotes and 1-1/2 times in hetero-zygotes. Subsequent studies in CF nasal epitheliumshowed oxygen consumption exceeded control tissueby 2-3 fold26. In this regard, Shepherd et al.27 investi-gated total energy expenditure using the doubly-labelled water method in clinically well, appropriatelynourished CF infants without clinical evidence oflungdisease and data was compared with studies in healthyinfants. This methodology measures total energyexpenditure in unrestricted subjects. CF infantsevaluated by this method had 25% higher rates ofenergy expenditure in comparison with data obtainedby the same method in 16 healthy infants matchedfor age and body weight. These exciting preliminaryobservations require fuirther confirmation and detailedanalysis of potential confounding factors.

Pathogenegis of an energy deficitA model is proposed to explain the etiology of theenergy deficit in the CF patient (Figure 3), and definesthe web of interdependent variables which may giverise to chronic malnutrition, and growth failure inpatients with CF. It must be re-emphasized, however,that in the majority of patients with minimal ormoderate pulmonary disease normal growth velocitycan be achieved by consuming adequate diets, thuscompensating for the three fundamental factorsresponsible for an energy deficit (Figure 1). Expresseddifferently, the majority of patients with CF canmaintain normal growth velocity and nutritionalstatus by voluntary intake ofcalories until severe lungdisease supervenes6. We and others have speculatedthat malnutrition and decline in pulmonary functionare closely interrelated, but the cause-effect relation-ships remain to be proven. As lung disease worsens,most commonly in older adolescents and young adults,a number of factors come into play which mightpredisopose to an energy deficit and malnutrition.There may be an increase in frequency and severityof pulmonary infections which in turn induceanorexia. Chest infections often given rise to vomiting

CELLULAR PANCREATIC IATROGENICDEFECT? INTSTINAL PSYCHOGENIC

BILIARY GASTROINTESTINAL

+i It LOSSE III-. - IEANOREXIA

|ENERGYDEFcrT1VOMITINGi t

I WEIHT LOSS PU -RYINFECTIONS

RESPIRATORY LUNG IMMUNEMSCLES PARENCHYMA DYSFUNCTION

H DETERIORATING LUNG FUNCTKON rFigure 3. Interdependent factors which may give rise toprogressive energy deficit as lung function deteriorates

14 Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989

which may further reduce intake. These factors, demineralization may be quite common. Mineralin combination with additional energy needs from metabolism was assessed in patients. with CF byincreased work of breathing.; may, induce an energy quantifying active metabolites of vitamin D anddeficit. Weight loss will result, initially producing a radiographic bone mass in adolescents and youngsignificant loss of adipose tissue, but with time there adults with CF and age-matched controls, living inis a marked loss of lean tissue, with muscle wasting. two geographic regions with low and high sunlightRespiratory muscle wasting would adversely affect; exposurem. The levels -of 25-hydroxycholecalciferolrespiratory motion and coughing, resulting in further were lower in patients with.CF than in the controls,deterioration of lung function. Malnutrition is known and serum concentrations ofthe active hormone wereto adversely affect lung elasticity28 and a variety of -.significantly lower in the area of low sunlightaspects of immune function4A All these factors would, exposure. Mean bone density in patients with CF wastherefore, contribute to progressive deterioration of significantly below the normal American standards.lung function. In essence, a vicious cyele is esta bihed, It was concluded that seasonal sunlight exposureleading inevitably to end-stage pulmonary failure and influenced the levels ofvitamin D metabolites. Olderdeath. patients with CF appeared to be at increased risk for,

developing abnormalities ofmineral metabolism. ThisDeficits of essential nutrients study suggests subtle deficiencies of vitamin D inDeficits of essential micronutrients can arise as a patients who are deprived of sunlight, despiteresult of primary malnutrition, but are commonly due seemingly adequate ingestion of vitamin D. Frankto secondary features of the diseae*30, By way of osteomalacia has been reported in a black patientexample, CF patients with pancreatic insufficiency with CF and maltilobular cirrhosis39.frequently malabsorb. fat soluble vitamins and this -Protein-calorie malnutrition might-well be a factorplaces them at risk of developing clinical Signs and in these observations. Gibbens et aL40 determinedsymptoms of deficiency. the prevalence of osteoporosis among patients with

CF by evaluating vertebral bone density using aWater soluble vitamins highly sensitive computed tomography technique. InAll water-soluble vitamins except vitamin B12 appear comparison with age, race and sex-matched controls,to be well absorbed, and there is no good evidence of the patients with CF had lower bone density,clinically significant deficiencies in.well nourished approximately 10% less than the controls. Thepatients. Vitamin B12 absorption should be normal- decrease in bone density was not-related to age, butized with adequate pancreatic enzyme replacement a strong correlation existed between bone density andtherapy. Vitamin B12 supplementation is not disease severity; poor bone mineralization was morenecessary, apart from those patients with meconium common in-patients with poor nutritional status.ileus who have undergone extensive ileal resectionand show biochemical evidence of vitamin B12 Vitamin -E: This powerful antioxidant protectsdeficiency. nutrients. such as polyunsaturated fatty acids and

vitamin A.from oxidation and has been shown to beFat soluble vitamins extremely .important in mediating prostaglandinDeficiencies of vitamins A, D, E and K have been synthesisO.. The inportance of. bile. salts. in therepeatedly demonstrated at diagnosis30'31. In most absorption of vitamin E have beenemphasized41andinstances, supplementation offat-soluble vitamins is children with chronic cholestasis from a variety ofinstituted following diagnosis, but the recommended causes rapidly develop clinical signs of neurologicaldose for each vitamin has not been adequately dysfunction. Signs of vitamin E deficiency are quiteestablished and current routines vary considerably3l. fiequently noted in newly diagnosed infants with CF,Optimal quantities and formulation of each prep- who may have evidence of haemolytic anaemia42.aration have been established on arbitrary grounds. Serum vitamin E levels are also frequently low in theThere is little doubt, however, that supplementation absence of symptoms. Supplemental doses of alpha-with fat soluble vitamins.is a necessary part of the tocopherol acetate appear to result in a biochemicalnutritional care in CF patients with pancreatic response31, -but vitamin E tolerance tests. in CFinsufflciency or severe liver disease. patients with pancreatic insufficiency do suggest poor

absorptionsofthis vitamin even with pancreatic enzymeVitamin A: Clinically overt signs of vitamin A supplementation. Clinical features: of vitamin! Edeficiency are rare. Increased intracranial pressure deficiency have been reported in children and adultsis an uncommon complication33. Treated patients with CF; these include ophthalmoplegia, diminishedappear to have high hepatic concentrations ofvitamin reflexes, decreased vibratory and proprioceptive sense,A34. In fact, serum concentrations of vitamin A are ataxia and muscle weakness4344. Axonal dystrophyfrequently low in spite ofsupplementation, but levels and degenerative changes within the posteriorusually correlate with serum concentrations of retinol- columns have been demonstrated postmortem45. Itbinding protein suggesting that malnutrition contri- should be noted, however, that the majority ofbutes to the deficiency3l.. It has been suggested patients with severe neurological disease appeared tothat zinc deficiency may inhibit mobilization and have advanced hepatic disease, with focal. biliarytransport of vitamin A, buit biochemical evidence of cirrhosis4344. Since the prevalence of -liver diseasetrue zinc deficiency is not commonly recognized36. increases with age, this problem is likely to become

more prominent as increasing numbers of patients-Vitamin D: Vitamin D deficiency rickets and other survive into adulthood.clinically significant abnormalities of mineral andbone metabolism were infrequently reported inearly studies of patients with CF30'37- Nevertheless,more recent studies have demonstrated that bone

Vitamin K: Overt haemorrhagic manifestations ofvitamin K deficiency may be seen in untreatedindividuals with CF46. Unexplained purpura,

Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989 15

intestinal blood loss, bleeding from an injection siteor from a minor surgical procedure may be presentin the newborn period or in infancy. Catastrophic,sometimes fatal intracranial haemorrhage can occur,

but it is extremely unusual for treated patients withCF to have coagulation defects attributable to a

deficiency of vitamin K. It must be remembered,however, that older children, particularly those on

antibiotics or with advanced liver disease may besusceptible to secondary coagulation abnormalitieseven when supplemented with vitamin K.

Trace metal deficienciesPlasma zinc levels appear to be low only in patientswith moderate to severe malnutrition and levelscorrelate directly with plasma proteins, retinol-binding protein and vitamin A48. There is no obviousdefect of zinc absorption or metabolism. Plasma levelsof copper and ceruloplasmin may be elevated inpatients with cystic fibrosis, but usually in proportionto the severity ofpulmonary disease, possibly becauseceruloplasmin is an acute phase reactant48. It hasbeen proposed that cystic fibrosis is caused by an

acquired deficiency of selenium and that a diet highin selenium and vitamin E could be of great clinicalbenefit. However, there is no reliable evidence tosupport the concept that selenium is of any clinicalsignificance49. Symptomatic hypomagnesemia, withevidence of a positive Trousseau sign, tremulousness,muscle cramps and weakness, may develop in patientsreceiving aminoglycosides50, and is reported to be a

secondary complication in patients treated for distalintestinal obstruction syndrome with repeated oraldoses of n-acetylcysteine51.Iron deficiency anaemia with low serum ferritin is

frequently seen in patients with advanced pulmonarydisease30, but may also be seen in the stable patient52.In patients with pulmonary insufficiency, poly-cythemia seems to occur less commonly than in otherpulmonary disorders of comparable severity,suggesting that these patients have a relativeanaemia even though haemoglobin levels are withinthe normal range. The precise mechanism of iron-deficiency anaemia is poorly understood, since thereis no evidence of a defect of iron absorption or

metabolism. Earlier reports showed increased ironabsorption in children with CF not receivingpancreatic extracts, but these studies were probablycarried out in children with depleted iron stores53.

Essential fatty acid deficiencyIn infancy, particularly prior to diagnosis, clinicalfeatures of essential fatty acid deficiency (EFAD) can

occur with desquamating skin lesions, increasedsusceptibility to infection, poor wound healing,thrombocytopenia and growth retardation30. In olderpatients, who are adequately treated, overt clinicalevidence of EFAD is extremely rare. Most patientswith pancreatic insufficiency, nevertheless, haveabnormal blood and tissue lipids54. Changes includedecreased lineoleic and increased palmitoleic, oleicand eicosatetraenoic acids. Some reports suggest thatthese biochemical abnormalities reflect an underlyingdefect of fatty acid metabolism55, while others haveargued that the low plasma and tissue levels are dueto increased metabolic usage and malabsorption inundernourished patients56. In a survey of 32 patientswe found that low plasma essential fatty acid levelswere confined to patients with less than 5% of

-2 SD

20.000

1000

-E

w

I00

Q

cpo

..

*

MEAN

(Io.*4t

00 00*

1W _._l

* 0

10 0OCFLNONCF

m * oCONTROLS1.0 , *

0.1 __ _ __

20 40 0o 6o 100EFA (mg/di)

120 140

Figure 4. Comparison ofpancreatic function (expressed aspancreatic colipase output (units/kg body weight/h) withplasma essential fatty acids in control subjects, and patientswith cystic fibrosis (CW) and Shwachman Syndrome. Log coli-pase secretion reflects the range ofpancreatic disease. Patientswith colipase output between 100 units/kg body weight/h havepancreatic insufficiency with steatorroea. Only CF patientswith pancreatic colipase output less than 5% ofmean normalvalues showed evidence of essential fatty acid deficiency

pancreatic function57 (Figure 4). The occasional reportof patients with low levels and normal fat digestionmay be due to the fact that some patients with 2 to5% of pancreatic function do not have steatorrhoea.At present, there is no compelling evidence to suggestthat supplementation with essential fatty acids haveany major clinical benefits, but evaluation of subtledefects at the cellular level, attributable to membranedysfunction or free-radical injury, remain to beadequately evaluated.

Nutritional evaluation and therapy -a rational approach

ClinicalThe nutritional support of patients with CF shouldbe an integral part of overall care, and requires closeclinical evaluation, monitoring of growth rates andappropriate dietary counselling (Table 2). At diagnosis,height, weight (percentiles), and anthropometry (skinfolds, mid-arm circumference) should be carefullymeasured. During follow-up visits, careful monitoring

Table 2. Nutritional assessment

At diagnosis Subsequent visits

Growth Height/weight Height/weight(percentiles) (percentiles)

Arthropometry ArthropometryDiet Caloric intake by Caloric intake by

observation recallDigestiorn *Fecal fat (% of Stool microscopyabsorption intake) (Steatocrit)

Bentiromide test Serum caroteneHaemogram, iron Repeat initial

stores measurementsPT/PTT as indicatedVitamins A, E,250H-D

Plasma proteinsPlasma EFA

*If normal, pancreatic stimulation test

*

16 Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989

of growth should be instituted, preferably every3-6 months. When patients receive an adequatediet, normal growth can be expected until advancedrespiratory disease supervenes. Individuals failingto achieve normal growth velocity deserve carefulevaluation - particularly young children with littlepulmonary disease. Close involvement of a qualified,experienced dietitian is invaluable. Assessmentshould include careful evaluation of caloric intake andcompliance with pancreatic enzyme supplements;additionally, assessment should include determiningthe adequacy of stool fat absorption (72-h fecal fat)and documenting fat and energy intake whilecontinuing regular pancreatic enzyme supplements.Adjustments in the dose of enzymes, with or withoutthe use of agents to improve digestion (histamineantagonists, bicarbonate, etc), may be usedjudiciously. Patients with mild pulmonary disease willoften lose weight following an acute respiratoryinfection, but generally catch-up growth follows afterrecovery. Individuals suffering from recurrentabdominal pain due to distal intestinal obstructionsyndrome often reduce caloric intake to control theirsymptoms; aggressive treatment may be necessary.In our experience, this is best achieved by intestinallavage with a balanced electrolyte solution containingpolyethylene glycol58. Similarly, signs of gastro-oesophageal reflux and oesophagitisl9,20 must besought, and aggressive treatment instituted becausesevere symptoms will reduce caloric intake.Generally, patients with hepatic disease will grownormally, except in rare instances when hepaticdecompensation supervenes.The diet must be calorically adequate for individual

needs. Since fat is the most economic and appetizingenergy source it should be encouraged rather thanrestricted. The diet should be as normal for age andpeer group as possible. Actual energy requirementsof patients with CF are unknown but are estimatedto be between 120 and 150% of the recommendeddaily allowance for age3. In this regard, althoughCF patients have been characterized as havingvoracious appetites, accurate records have revealedthat energy intake is often very similar to that fornormal children ofthe same age, sex, and stature andin some cases may even be reducedl6-'8. Dietaryintake is closely related to an individual's self-esteemand general feeling of well being. It is, therefore,essential that patients who have nutritionaldifficulties receive psychological support; this isparticularly important in adolescence and adulthood.Exercise programs aimed at improving physicalcapacity are considered to be important. Improvedmuscle mass may lead to a sense of accomplishment,and stimulate an interest in providing nutritionalsupport for physical goals.As a general rule, protein intakes of children with

CF are more than adequate16, but nitrogen balancemay be particularly sensitive to insufficient totalenergy intake. Provided total energy intake isadequate, we recommend that protein intake equalthe recommended daily allowance for age, sex andweight. The use of fat as a source of energy, whichhas been previously discussed in some detail, providesan excellent source of palatable, energy rich calories.The limited reserves of essential fatty acids in CFpatients and the particular vulnerability of mal-nourished patients to essential fatty acid deficiencyrequires some specific attention. Although, there is

no existing evidence to suggest that biochemicalessential fatty acid deficiency has any major clinicalimpact, we do recommend a diet that containsadequate quantities of linoleic acid to maintainnormal or close to normal fatty acid profiles. Thereis no known defect in the transport of mono-saccharides and some investigators have evensuggested enhanced glucose absorption. Despite theabsence of pancreatic amylase in most patients,complex carbohydrates are generally well toleratedand are good sources of energy. Supplemented dosesof fat soluble vitamins are indicated in patients withpancreatic insufficiency or severe hepatic-biliarydisease; generally, two to three times normal intakeis recommended.

BiochemicalBiochemical evaluation at diagnosis should includea careful assessment of nutritional status andpancreatic function (Table 3). We still recommendquantitative evaluation of fecal fat losses whileaccurately measuring fat intake, for documentationof steatorrhoea. Alternatively, recent modificationsto the oral bentiromide test (N-benzyl-tyrosyl-aminobenzoic acid), provides a less costly and time-consuming approach for evaluating pancreaticfunction59. Poor substitutes include documentation offat on stool microscopy, stool trypsin or chymotrypsinactivity, serum carotene and vitamin A and Elevels. Many of these tests are useful, however, formonitoring response to treatment on return visits.Serum levels ofimmunoreactive trypsinogen may bereduced in the older patient with pancreaticinsufficiency, but only after the age of 7-8 years6O.The most accurate method of assessing pancreaticfunction is the direct pancreatic stimulation test'2,which has been described separately (see p 2). Thisextremely invasive, time-consuming and difficult testshould be reserved for the evaluation ofpatients withpancreatic sufficiency, or in those where the diagnosisof CF remains in doubt.Routine biochemical measures ofnutritional status,

such as serum protein or albumin may be misleading.In the newly diagnosed infant, however, hypo--proteinaemia and hypoalbuminaemia are a truereflection of the severity of malnutrition, andvalues usually revert to normal with nutritionalrehabilitation. Plasma proteins with a short half-lifesuch as retinol-binding protein and prealbumin, havebeen incompletely evaluated as markers of nutritionalstatus in patients with CF. With advancing age,particularly in patients with severe pulmonarydisease, albumin levels may be depressed, althoughelevated globulin may produce an increase in totalprotein concentration. The increase in globulin hasbeen attributed to chronic pulmonary disease. Thecause of decreased albumin levels in these patientsis incompletely understood. Studies have shownnormal albumin turnover rates and expandedplasma volume, suggesting that haemodilution maycontribute. There appears to be an overall correlationbetween the severity of pulmonary disease andalbumin concentration.

Specific nutritional considerationsThe standard guidelines for the nutritional evaluationand support of patients with CF must be modifiedaccording to individual needs, the age of the patientand specific complications of the disease.

Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989 17

Table 3. Long-term enteral feeding of malnourished patients with cystic fibrosis

Toronto' Ottawa65 Brisbane'

Study designPatients (male/female) 14 (5/9) 10 (5/5) 10 (5/5)Age - mean (years) 12.9 13.6 8.9

- range (years) 5-22 6-21 3-13Enteral route Gastrostomy Jejunostomy Nasogastric/gastrostomySupplement type Semi-elemental Intact Semi-elementalDuration (years) 1.1 1.6 1.0Controls Concurrent Retrospective Prospective

Patient characteristicsWeight as % of height 82±10 80+9 No data@FEV1 (%) 47+15 No data 66+16FVC (%) 53+13 64+18 84+12

0 Expressed as Z-score

InfancyNewly diagnosed infants may be profoundly anorexicand indifferent to food; hypoalbuminaemia, oedemaand anaemia of infancy, particularly those under3 months of age deserve special attention30. Inaddition, following surgery during the neonatal periodfor meconium ileus, active nutritional managementis imperative. In these cases, a short course ofparenteral or total enteral nutrition may be the onlyway to ensure adequate nutrition in the first fewweeks of care. Albumin infusions may be indicatedto reduce oedema. In general, however, patientsimprove rapidly with adequate attention to caloricrequirements, vitamin needs and pancreatic enzymesupplementation; routine oral feeding with a standardage-appropriate formula becomes possible very quickly.Long-term use ofprotein hydrolysate, medium-chaintriglycerides and polysaccharide supplements arerarely indicated. Infants who cannot maintainadequate growth with high-calorie standard nutrientshardly ever do better when given these supplements,unless they are artificially delivered by enteric tubes.

Adolescence and adulthoodWith close attention to energy needs and foodintake, it is possible to maintain adequate energybalance for optimal growth in the majority ofpatientswith CF, especially when lung function is notimpaired. However, a number of affected adolescentsand adults, especially females, will develop weightloss in association with advanced pulmonary disease(Figure 5). By definition, these patients have anenergy imbalance since they seem not to be able tomaintain adequate energy intake by voluntarymeans. A variety of short-term and long-termapproaches to artificial nutritional supplementationhave been undertaken in the hope that restorationof nutritional status might result in easier control ofchest infections, ameliorate the rate of decline inrespiratory function, and extend survival. Highenergy liquid dietary supplements are frequentlyadvocated, and although their use may be successfulin the very short term, no reliable information isavailable regarding their long-term efflcacy. It isour impression that many of these energy-richsupplements are at best substitutes for normal dietaryhabits and do not result in long-term improvementin nutritional status. Patients with growth failure orweight loss, particularly those with deteriorating

TYPICAL GROWTH CHARACTERISTICS OF ANADOLESCENT WITH CF

AGE(years)

Figure 5. With adequate energy intake, most children withcystic fibrosis will grow normally to adolescence or adulthood,until severe lung disease develops. At this stage they appearto be unable to maintain adequate energy intake by voluntarymeans

pulmonary function, therefore, may be considered ascandidates for more invasive, artificial forms ofsupplementary nutrition. Existing short and long-term studies are critically reviewed.

Short-term studiesA number of short-term studies have used a varietyof parenteral and enteral feeding techniques in anattempt to renourish malnourished CF patients. Anearly study by Shepherd et al.61 evaluated 12 mal-nourished CF patients (mean age 5.43 years), 6 monthsbefore and 6 months after a 3-week period ofparenteralnutrition. During the pre-treatment period, while

18 Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989

receiving 'conventional' dietary management, thepatients had inadequate growth velocity, but 6 monthsafter the short period of intravenous nutrition theyappeared to exhibit continuing catch-up growth, fewerpulmonary infections, and a significant improvementin clinical score.Unfortunately, other investigators have failed to

show lasting improvement following short-termnutritional support, and the improved nutritionalstatus in the patients included in Shepherd's studycould best be explained by aggressive pulmonarymanagement while patients were hospitalized. Inaddition, the very young age of the patients inquestion suggest that closer attention to voluntarynutrition may well have prevented the problem at theoutset. Mansell et al.62 who evaluated older mal-nourished CF patients (aged 10-17 years) alsodemonstrated improvement in nutritional statusfollowing a one-month period of supplementalparenteral nutrition when patients were given 120%of their energy needs. Immediately -followingsupplementation bodyweight, triceps skin-fold-thick-ness, and mid-arm muscle circumference increasedsignificantly. Maximum inspiratory airway pressurealso increased, suggesting improvement in respiratorymuscle strength. However, none ofthe indices of lungfunction improved. One month following parenteralnutrition the patients were once again malnourishedto levels similar to those before commencement ofparenteral nutrition. In a study from Montreal63,supplemental feeding by nasogastric tube wasinstituted while in hospital and continued at homefor 4 weeks. Patients showed considerable weightgain, attributable to increased caloric intake, but thenutritional changes were transient and not accom-panied by long-term improvement in growth orfunction. In a study from our centre, Pencharz et aL22evaluated body composition, nutritional status andenergy needs of 6 undernourished adolescents andadults with CF. Lean body mass was preserved butthere was significant wasting of adipose tissue.Following a brief period of nasogastric feeding witha semi-elemental diet, the effects ofrefeeding on bodycomposition was reassessed. After refeeding, bodyweight, body fat and total body potassium allincreased signiflcantly but fat-free body mass andtotal body nitrogen did not change. All subjects wereencouraged to continue nasogastric tube feeding athome, but none were able to continue for longer than2-3 months. The majority ofpatients discontinued thesupplemental feeding because ofnasal irritation andcoughing up the tube. Thus, based on the precedingstudies it can be concluded that any nutritionalbenefits derived by brief periods of supplementalnutrition are short lived and not accompaniedby any long-term improvement in growth orfunction.

As shown in Table 3, three major studies haveapproached the problem by using various forms ofnocturnal enteral supplementation 66. In a studyfrom Toronto", patients were given nocturnalsupplemental feeding of a semi-elemental formulagiven by gastrostomy tube for an average period ofone year. The adolescent and adult patients weresuffering from moderate to severe lung disease andall were markedly wasted or stunted. Gastrostomytubes were placed endoscopically without incision andunder local anaesthesia. A contemporary groupof patients with CF (age, sex, nutritional, andpulmonary function-matched) derived from the cliniccomputerized data bank were pair-matched to thestudy group. In a second Canadian study, fromOttawa65, 10 malnourished CF patients (mean age13.6 years) with moderate to severe lung disease wereprovided with nocturnal supplemental feeding of anintact formula by a needle jejunostomy tube forperiods of 10-36 months. Pancreatic enzymesupplements were added to the formula. Jejunostomyfeedings appeared to pose no technical or medicaldifficulties. In the third study from Australia,Shepherd et al.66 evaluated 10 undernourished- CFpatients (mean age 8.9 years) who are unable tomaintain normal growth by oral means. Thesepatients were followed during a one-year course ofnutritional supplementation with a balanced-peptideor a semi-elemental formula delivered overnight bynasogastric or gastrostomy feeding. The patientswere compared concurrently with height, sex, andpulmonary function-matched patients receivingconventional nutritional therapy. In all 3 studies,normal activity and regular meals were permittedduring daytime hours.Evaluation of these studies revealed that chronic

enteral supplemental feeding resulted in a significantimprovement in catch-up growth and positive changesin body composition (Table 4). Long-term benefitsparticularly the effects on pulmonary function andsurvival remain somewhat more controversial. In thetwo Canadian studies"465, the.rate of deterioration ofpulmonary function appeared to have been slowedin the supplemented group. In Shepherd's study",respiratory function deteriorated in the control groupbut appeared to actually improve inthe patient group.However, Shepherd's patients were considerablyyounger and were suffering firom -less-compromised

Table 4. Effects oflong-term enteral feeding in malnourished patientswith cystic fibrosis

Toronto& OttawaM Brisbane6

Nutritional statusAWeight (kg) tAleight (cm) tAWeight as ta- rt IMus a

Long-term studies 7 Of neagnht ()Since briefperiods of nutritional supplementation of Tpotassium(l )chronically malnourished CF patients did lead to a Body fat (%) ttransient improvement in nutritional status, it Mid-arm mule No databecame evident that long-term approaches to artificial circumferencesupplemental feeding might achieve and maintain Protein synthesisNo datanormal nutrition in patients who were unable to meet Rpiratory fuhctiontheir energy needs. Further to this, it was felt thatt Patients Unchangedreversal of malnutrition may have a favourable Controls Deterioratedinfluence on the course of pulmonary disease, andconsequently upon survival. *Compared with year before in

ttt

No data

ttt

No data

No data No datat No data

No data t

Unchanged ImprovedDeteriorated@ Deteriorated

ntervention

Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989 19

pulmonary function, in comparison with those in thetwo Canadian studies. It could be argued, that at earlystages in the natural history ofthe disease, patientswith mild pulmonary disease should be able tomaintain normal growth velocity with conventionaldietary therapy. On the other hand, it is possible thatearly aggressive treatment of the nutritional deficitmight have a more beneficial effect on pulmonarydisease.

ConclusionsThere is no convincing proof that aggressivesupplemental nutrition actually improves survival.Furthermore, although we are convinced thatnutritional rehabilitation results in an improvementin patient well-being, the psychological impact ofaggressive invasive means of supplemental nutritionhave not been adequately evaluated. Other questionsremain to be answered before invasive enteraltechniques can be universally accepted for the routinemanagement of malnourished patients with CF.These include appropriate clinical and psychologicalcriteria for patient selection, and evaluation of thecost-effectiveness and complication rate of variousapproaches. No consensus has been reached regardingthe appropriate formula type, or the optimal methodof administration. Careful evaluation ofthe long-termbenefits, particularly survival are necessary, butthese are likely to be extremely difficult, if notimpossible to conduct with conclusive results.Prevention of growth failure should be a primary

aim in the management ofCF patients; obviously thiswould be best achieved without the necessity ofinstituting invasive approaches. In our experience,normal growth can be achieved in the vast majorityof patients with rational use of a normal-high energy(high fat) diet from the time of diagnosis and withcareful monitoring of progress (Table 5). With thisapproach, most patients will grow normally untiladvanced lung disease supervenes. It is usually at thispoint that severe malnutrition becomes a problem andone is forced into considering artificial, often invasivemeans of intervention. It is extremely doubtful,however, that any form of aggressive nutritionaltherapy has any impact on outcome during theterminal stages when the patient is either permanentlyhospitalized or in a practical sense bed ridden fromcardiopulmonary failure67. Unfortunately, this isoften the time when demands for nutritional inter-vention become most intense. Perhaps the mostimportant role ofthe care-giver during this period, is

Table 5. Approach to nutritional treatment ofpatients withcystic fibrosis

Encourage good feeding habits earlyIn patients with growth failureassess energy intakeassess absorptive functionassess for gastrointestinal complications which wouldreduce intake

evaluate psychological/family dysfunctionencourage voluntary supplements and/or modify enzymetherapy

Invasive methods of nutritional supplementation should beconsidered as a last resort, but before malnutrition andsevere lung disease supervenes

In patients with end-stage pulmonary failure invasivemethods will only prolong the agony of dying

to discourage aggressive artificial means of nutri-tional support and other examples of our technicalversatility, which only add to the discomfort andfrustration of the last days of life.

Acknowledgment: Much of the work from our units referredto in this paper has been made possible through the generoussupport of the Canadian Cystic Fibrosis Foundation.

References1 Sproul A, Huang N. Growth patterns in children with

cystic fibrosis. J Pediatr 1964;65:664-762 Kraemer R, Rudeberg A, Hadorn B, Rossi E. Relative

underweight in cystic fibrosis and its prognostic value.Acta Paediatr Scand 1978;67:33-7

3 Roy CC, Darling P, Weber AM. A rational approach tomeeting macro- and micronutrient needs in cysticfibrosis. J Pediatr Gastroenterol Nutr 1984;3(Suppl 1):S154-62

4 Gurwitz D, Corey M, Francis PWJ, Crozier D, LevisonH. Perspectives in cystic fibrosis. Pediatr Clin N Am1979;26:603-15

5 Corey M. Longitudinal studies in cystic fibrosis. In:Sturgess JM, ed. Proceedings of the 8th InternationalCongress on Cystic Fibrosis (1980). Mississauga, Canada:Imperial Press, 246-55

6 Corey M, McLaughlin FJ, Williams M, Levison H. Acomparison of survival, growth and pulmonary functionin patients with cystic fibrosis in Boston and Toronto.J Clin Epidemiol 1988;41:588-91

7 Shwachman H. Therapy ofcystic fibrosis ofthe pancreas.Pediatrics 1960;25:155-63

8 Allan JD, Mason A, Moss AD. Nutritional supplemen-tation in treatment ofcystic fibrosis ofthe pancreas.AmJ Dis Child 1973;125:22-6

9 Gracey M, Burke V, Anderson C. Assessment ofmedium-chain triglycerides in infants with cysticfibrosis. Arch Dis Child 1969;44:401-3

10 Crozier DN. Cystic fibrosis: a not so fatal disease. PediatrClin North Am 1974;21:935-48

11 Pencharz PB. Energy intakes and low fat diets inchildren with cystic fibrosis. J Pediatr GastroenterolNutr 1983;2:400-1

12 Gaskin KJ, Durie P, Lee L, Hill R, Forstner GG. Colipaseand lipase secretion in childhood onset of pancreaticinsufficiency. Gastroenterology 1984;86:1-7

13 Gaskin KJ, Durie P, Corey M, Wei P, Forstner G.Evidence for a primary defect of bicarbonate secretionin cystic fibrosis. Pediatr Res 1982;16:554-7

14 Zentler-Munro PL, Fine DR, Batten JC, Northfield TC.Effect of cimetidine on enzyme inactivation, bile acidprecipitation and lipid solubilisation in pancreaticsteatorrhea due to cystic fibrosis. Gut 1985;26:892-901

15 Belli DC, Levy E, Darling P, et aL Taurine improves theabsorption ofa fat meal in patients with cystic fibrosis.Pediatrics 1987;80:517-23

16 Bell L, Linton W, Corey ML, Durie P, Forstner GG.Nutrient intakes of adolescents with cystic fibrosis. JCan Dietet Assoc 1981;42:62-71

17 Parsons JG, Beaudry P, Dumas A, Pencharz P. Energyneeds and growth in children with cystic fibrosis. JPediatr Gastroenterol Nutr 1983;2:214-9

18 Daniels L, Davidson GP, Martin AJ. Comparison ofmacronutrient intake of healthy controls and childrenwith cystic fibrosis on low fat or non-restricted fat diets.J Pediatr Gastroenterol Nutr 1987;7:381-6

19 Scott RB, O'Laughlin EV. Gall DG. Gastroesophagealreflux in patients with cystic fibrosis. J Pediatr1985;108:223-7

20 Feigelson J, Girault F, Pecau Y. Gastosophageal refluxand esophagitis in cystic fibrosis. Acta Paedzatr Scand1987;76:989-90

21 Rosenstein BJ, Lengbaum TS. Incidence of distalintestinal obstruction syndrome in cystic fibrosis. JPedzatr Gastroenterol Nutr 1983;2:294-301

20 Journal of the Royal Society of Medicine Supplement No. 16 Volume 82 1989

22 Pencharz P, Hill R, Archibald E, Levy L, Newth C.Energy needs and nutritional rehabilitation in under-nourished adolescents and adults with cystic fibrosis.J. Pediatr Gastroenterol Nutr 1984;3(Suppl 1):S147-53

23 Vaisman N, Pencharz PB, Corey M, Canny GJ, HahnE. Energy expenditure of patients with cystic fibrosis.J Pediatr 1987;111:496-500

24 Buchdahl RM, Cox M, Fulleylove C, et aL Increasedresting energy expenditure in cystic fibrosis. J ApplPhysiol 1988;64(5):1810-6

25 Feigel RJ, Shapiro BL. Mitochondrial calcium uptakeand oxygen consumption in cystic fibrosis. Nature1979;178:276-7

26 Stutts MJ, Knowles MR, Gatzy JT, Boucher RC. Oxygenconsumption, and ouabain binding sites in cystic fibrosisnasal epithelium. Pediatr Res 1986;20:1316-20

27 Shepherd RH, Vasques-Velasquez L, Prentice A, et al.Increased energy expenditure in children with cysticfibrosis. Lancet 1988;i:1300-3

28 Thet LA, Alvarez H. Effect of hyperventilation andstarvation on rat lung mechanics and surfactantAm RevRespir Dis 1982;126:286-90

29 Chandra RK, Newberne PM, eds. Nutrition, immunityand infection. New York: Plenum Press, 1977

30 Chase PM, Long MA, Lavin MH. Cystic fibrosis andmalnutrition. J Pediatr 1979;95:337-47

31 Congden PJ, Bruce G, Rothburn MM, et aL Vitaminstatus in treated patients with cystic fibrosis. Arch Dis-Child 1981;81:708-14

32 Deren JJ, Avora B, Tuskes PP, Hansell J, Sibing MS.Malabsorption of crystalline vitamin B12 in cysticfibrosis. N Engl J Med 1973;288:949-50

33 Abernathy RS. Bulging fontanelle as presenting signin cystic fibrosis. Am J Dis Child 1976;130:1360-3

34 Underwood BA, Denning CR. Blood and liver concen-trations of Vitamin A and E in children with cysticfibrosis of the pancreas. Pediatr Res 1972;6:26-31

35 Rasmussen M, Michalsen H, Lie SO, Nilsson A, PetersenLB, Norum KR. Intestinal retinol esterification andserum retinol in children with cystic fibrosis. J. PediatrGastroenterol Nutr 1986;5:397-403

36 Jacob R. A., Sandstead HH, Solomons NH, Rieger C,Rothberg R. Zinc status and vitamin A transport incystic fibrosis. Am J Clin Nutr 1978;31:638-44

37 Hubbard VS, Farrel PM, di Sant' Agnese PA. 25-hydroxycholecalcyern levels in patients with cysticfibrosis. J Pediatr 1979;94:84-6.

38 Reiter EO, Brugman SM, Pike.'JW, et al. Vitamin Dmetabolities in adolescents and young adults with cysticfibrosis: Effects of sun and season. J Pediatr 1985;106:21-5

39 Friedman HZ, Longman CB, Favus MJ. Vitamin Dmetabolism and osteomalacia in cystic fibrosis. Gastro-enterology 1985;88:808-13

40 Gibbens DT, Gilsanz V. Boechat MI, Dufer D, CarlsonME, Wang CI. Osteoporosis in cystic fibrosis. JPediatr1988;113:295-300

41 Gallo-Torres H. Obligatory role of bile in absorption ofvitamin E. Lipids 1970;5:379-84

42 Farrel PM, Bieri JG, Fratantoni JF, Wood RE, di Sant'Agnese PA. The occurrence and effects of humanvitamin E deficiency. A study in patients with cysticfibrosis. J Clin Invest 1977;60:233-41

43 Elias E, Muller DPR, Scott J. Association of spino-cerebellar disorders with cystic fibrosis or chronicchildhood cholestasis and low serum vitamin E. Lancet1981;ii:1319-21

44 Sitrin MD, Lieberman F, Jensen WE, et al. Vitamin Edeficiency and neurologic disease in adults with cysticfibrosis. Ann Intern Med 1987;107:51-4

45 Sung JH. Neuraxonal dystrophy in mucoviscidosis. JNeuropathol Exp Neurol 1964;23:567-83

46 Walters TR, Koch HF. Hemorrhagic diathesis and cysticfibrosis in infancy. Am J Dis Child 1972;124:641-7

47 Komp DM, Selden RF. Coagulation abnormalities incystic fibrosis. Chest 1970;58:501-8

48 Solomons NW, Wagonfield JB, Rieger C, et al. Somebiochemical indices of nutrition in treated patients withcystic fibrosis. Am J Clin Nutr 1981;34:462-74

49 Castillo R, Landon C, Eckhardt K, Morris V. LevanderD, Lewiston N. Selenium and vitamin E status in cysticfibrosis. J Pediatr 1981;99:583-5

50 Green CG, Doershuk CF, Stern RC. Symptomatichypomagnesemia in cystic fibrosis. J Pediatr 1985;107:425-8

51 Godson C, Ryan MP, Brady HR, Bourk S, Fitzgerald MX.Acute hypomagnesemia complicating the treatment ofmeconium ileus equivalent in cystic fibrosis. Scand JGastroenterol 1988;23(Suppl 143):148-50

52 Ater JL, Herbst JJ, Landaw SA, O'Brien RT. Relativeanemia and ion deficiency in cystic fibrosis. Pediatrics1983;71:810-4

53 Heinrich HC, Bender-Gotz CH, Gabbe EE, et al.Absorption of inorganic iron in relation to iron storesin pancreatic insufficiency due to cystic fibrosis. KlinWochenschr 1977;55:587-93

54 Farrel PM, Mischler EH, Engle MJ, Brown DJ, Lau S.Fatty acid abnormalities in cystic fibrosis. Pediatr Res1985;19:104-9

55 Rogiers V, Dab I, Brokaert R, Vis HL. Long chain non-esterified fatty acid patterns in plasma of cystic fibrosispatients and their parents. Pediatr Res 1980;14:1088-91

56 Hubbard VS, Dunn GD. Fatty acid composition oferythrocyte phospholipids from patients with cysticfibrosis. Clin Chim Acta 1980;102:105-18

57 Forstner G, Durie P, Corey M. Cystic fibrosis, progressin gastroenterology and nutrition. 10th InternationalCystic Fibrosis Congress. Sydney, Australia, 1988.Excerpta Medica, Asia Pacific Congress. Series 74,pp 154-60

58 Cleghorn GJ, Stringer DA, Forstner GG, Durie P.Lavage treatment of distal intestinal obstruction inpatients with cystic fibrosis using a balanced intestinallavage solution. Lancet 1986;i:8-11

59 Weizman Z, Forstner GG, Gaskin KJ, et al. Thebentiromide test for assessing pancreatic dysfunctionusing analysis of PABA in plasma and urine. Gastro-enterology 1985;89:596-604

60 Durie P, Forstner GG, Gaskin KJ, et al. Age relatedalterations of immunoreactive pancreatic cationictrypsinogen in sera of CF patients with and withoutpancreatic insufficiency. Pediatr Res 1986;20:209-13

61 Shepherd R, Cooksley EGE, Donvill Cooke WD.Improved growth and clinical, nutritional and respiratorychanges in response to nutritional therapy in cysticfibrosis. J Pediatr 1980;97:351-7

62 Mansell AL, Andersen JC, Muttart CR, et al. Short-termpulmonary effects of total parenteral nutrition inchildren with cystic fibrosis. JPediatr 1984;104:700-5

63 Bertrand JM, Morin CL, Lasalle R, Patrick J, CoatesAL. Short-term clinical, nutritional, and functionaleffects of continuous elemental enteral alimentation inpatients with cystic fibrosis. J Pediatr 1984;104:41-6

64 Levy LD, Durie PR, Pencharz PB, Corey ML. Effects oflong-term nutritional rehabilitation on body compositionand clinical status in malnourished children andadolescents with cystic fibrosis. J Pediatr1985;107:225-30

65 Boland MP, Stocki DS, Macdonald NE, Soucy P, PatrickJ. Chronic jejunostomy feeding with a non-elementalformula in undernourished patients with cystic fibrosis.Lancet 1986;i:232-4

66 Shepherd RW, Holt TL. Thomas BJ, et al. Nutritionalrehabilitation in cystic fibrosis: Controlled studies oftheeffects on nutritional growth retardation, body proteinturnover and cause of pulmonary disease. J Pediatr1986;100:788-94

67 Levy LD), Dunie P, Pencharz PB, Corey ML. Prognosticfactors associated with patient survival during nutri-tional rehabilitation in malnourished children andadolescents with cystic fibrosis. J Pediatr GastroenterolNutr 1986;5:97-102