Study reference

Study characteristics

Patient characteristics

Index test

(test of interest)

Reference test

Follow-up

Outcome measures and effect size

Comments

Caras, 2011

Type of study^[1]:

double-blind, randomized, placebocontrolled,

2-period crossover trial

Setting and country:

At 10 centers across the United States between

June 13, 2008 and December 1, 2008.

Funding and conflicts of interest:

funded by Abbott Products Inc,

à S.C., D.B., and L.Z. were employees of Abbott, Marietta, GA, at the time of

this analysis and writing of the manuscript. S.S.S. is an employee of

Abbott, Hannover, Germany.

Inclusion criteria:

- Subjects ages 7 to 11 years who had a confirmed diagnosis of

CF and EPI;

- Patients must have been receiving treatment with a

commercially available PERT product at a stable dose for >3 months. They had to be in clinically stable condition, without evidence of acute respiratory disease, for at least 1 month before enrollment.

- only patients with a stable body weight (defined as a decline of no more than 5% within 3 months of enrollment)

-Patients had to be able to swallow capsules and to consume a standardized diet designed to provide sufficient fat to require a minimum of 12,000 lipase units of pancreatic enzyme

supplementation per meal.

Exclusion criteria:

- if they had severe medical

conditions that might limit participation in or completion of the study or if they had recently undergone major surgery (excluding appendectomy).

- a body mass index percentile for age of <10%; ileus or acute abdomen; malignancy of the digestive tract (excluding pancreatic cancer); HIV; celiac disease; Crohn’s disease; known allergy to pancrelipase (pancreatin) or the inactive ingredients in pancrelipase

delayed-release capsules; or exposure to an experimental drug within 30 days of the start of the study.

N= 17 subjects (8 received placebo then pancrelipase, 9 received pancrelipase then

placebo)

median age (range):

8,0 (7–11) years

Boys: n= 12 (70,6%)

mean daily lipase dose: 4472 U/g

fat consumed.

Other important characteristics:

12 subjects from 8 of the 10 centers provided samples

for this analysis (10 in both treatment periods and 2 in only 1

period); the median age was 9.0 years, all subjects were of white

race, and 9 (75,0%) were boys.

Describe index test:

Sparse stool sampling for percentage fat (PF) analysis

Bowel movements were collected individually to facilitate determination of the PF in each sample.

Individual stool samples were analyzed for fat content by removing a 5-g aliquot from the remaining 50% of each homogenized bowel movement. Therefore, PF sampling did not affect the CFA determination. Stool fat was determined by nuclear magnetic resonance (NMR) methods

PF was calculated based on the dry weight of each bowel movement

Comparator test^[2]:

PF values for each subject were derived using 4 different

sparse-sample methods (3 multiple-sample methods and 1 singlesample method):

- mean PF from the first bowel movement on each day;

- mean PF from the last bowel movement on each day;

- mean PF from 3 randomly selected bowel movements, each on a different

day;

- PF from a single random sample selected irrespective of day

Describe reference test^[3]:

72-hour CFA

The CFA (coefficient of fat absorption) is calculated from fat intake and stool fat output data collected during a 72-hour period.

Subjects were administered 2 doses of blue food dye (250mg of FD and C Blue #2 indigo carmine) 72 hours apart to mark the beginning and end of the stool collections, which were performed during both crossover periods. Stool collection began after passage of the first blue-stained bowel movement and ended with passage of

the second

Stool fat was determined by nuclear magnetic resonance (NMR) methods

Cut-off point: 80%

Time between the index test en reference test:

For how many participants were no complete outcome data available?

N=1 (5,9%)

Reasons for incomplete outcome data described? Not reported

Outcome measures

The comparison of PF with CFA as a dichotomous variable (CFA <80% vs CFA ≥80%) included evaluation of sensitivity, specificity, and positive predictive value (PPV).

PF values ≤30% were greatly predictive for CFA values ≥80%, as shown by PPV,

sensitivity, and specificity values ≥0,89.

ROC curves of sensitivity

versus 1–specificity showed high accuracy (AUCs ≥0.93) for the multiple-sample methods.

PPV= 0,88 sensitivity = 0,78, à were lower at the 30% PF threshold for the single-sample method compared with the multiplesample

methods; an AUC of 0,91 in the ROC plot indicated slightly lower accuracy

à See table 1 below

the correlation between PF and CFA as a continuous variable based on 3 groups of samples: pancrelipase,

placebo, and both combined. à See table 2 below

72-hour CFA à hospitalization is required to ensure complete stool collection and a complete record of dietary fat intake

The subjects received 12,000-lipase unit pancrelipase

delayed-release capsules. The correct number of capsules to be consumed was calculated to provide a target dose of 4000 lipase units/gram of dietary fat intake according to the upper limit of the

recommendation of CF consensus statements (30–32). Each subject received an individualized, prospectively designed diet to maintain normal nutrition, containing at least 40% of energy derived from fat.

Caras, 2011

Kent, 2018

Type of study:

Accuracy study

Setting and country: The Children’s Hospital at Westmead, Sydney, Australia.

Funding and conflicts of interest:

This research project was supported by grants from the James Fairfax

Institute of Paediatric Nutrition and the Johno Johnston Scholarship

(DSK).

The authors report no conflicts of interest.

Inclusion criteria:

-identified as having CF by the

New South Wales State Program.

Exclusion criteria:

-

N= 24 infants (12 breast-fed and 12 formula-fed)

age: ranging in age from 1 to 13 months (20, <6 months of age)

Sex: 10/24 baby girls

Other important characteristics:

Mean height: 59,9 cm (± 8.0 [48,0–76,7])

mean weight: 5,8 kg (± 2,0 [3,1–10,3])

Describe index test:

surrogate IPF testing with ¹³C-mixed triglyceride breath testing (¹³C-MTG breath test)

Oral pancreatic enzyme substitution (PERT; if already commenced) was stopped at least 8 hours before the start of the ¹³C-MTG breath test.

- Infants <6 months of age received 200-mg MTG test substrate mixed in 5 ml expressed breast milk, infant formula or apple gel.

- babies (>6 months of age) MTG was mixed in apple-gel or

yoghurt.

Duplicate breath samples (150 ml) were collected before

administration of the test meal and at 30-minute intervals thereafter over the study period of 5 hours. Expired infant’s breath was collected by using an anaesthetic mask, connected via a one-way

flap valve to an aluminized plastic bag. Breath samples were analyzed promptly on the same day of the test. ¹³C enrichment was measured using NDIRS. Breath test results were analyzed and expressed as cumulative percentage dose of ¹³C recovered (cPDR) over the 5-hour test period (% 13C cPDR).

Cut-off point(s): 13

Option 1

to determine pancreatic function status is based on the %¹³C cumulative dose recovered (cPDR) value after 5 hours, with a minimum reference %¹³C cPDR value of 13, therefore, chosen as cut-off point for PS.

Option 2

comparison of the individual results of the ¹³C cPDR values over the whole 5-hour study period à CF babies are classified to be PS if ≥80% of ¹³C cumulative dose results are above the minimum values provided by the group of healthy individuals.

Comparator test:

fecal elastase-1 testing (FE1 measurements)

Oral pancreatic enzyme substitution (PERT; if already commenced) was stopped at least 1 day before the FE1 assessment.

Monoclonal FE1 levels were determined in all CF infants

using a commercially available enzyme-linked immunosorbent

assay (ELISA) kit (ScheBo, Biotech, Giessen, Germany).

Polyclonal FE1 levels were determined using a commercially available polyclonal ELISA kit (Bioserv Diagnostics, Rostock,

Germany).

Cut-off point(s):

Monoclonal FE1 levels FE1 According to manufacturer’s instructions à concentrations <200mg/g stool reflect pancreatic insufficiency

According to Beharry et al (2002) and Loser et al (1996)

a cut-off point of 100mg/g stool

Polyclonal FE1 levels

According to manufacturer’s instructions, values >200mg elastase/g faeces indicate normal exocrine pancreatic function; values between 100 and 200mg elastase/g faeces indicate moderate exocrine pancreatic insufficiency and values <100mg elastase/g faeces indicate severe exocrine pancreatic insufficiency.

Describe reference test:

72-h fecal fat balance test

Oral pancreatic enzyme substitution (PERT; if already commenced) was stopped at least 1 day before the fecal fat assessment.

Stool samples were collected for all CF babies over a 72-hour period. Additionally, for formula-fed infants, a record of the volume and kind of formula ingested over this period were kept. The

effectiveness of fat absorption was measured by the coefficient of fat malabsorption (CFA)

([fecal fat (g)/fat intake (g)] [100]).

Cut-off point(s):

infants <6 months of age à fecal fat excretion <10% of fat intake in formula-fed infants and based on fecal fat excretion < 2 g/day in breast-fed infants

babies >6 months of age à fat excretion ≤7% of oral fat intake.

Reasons for incomplete outcome data described?

Not reported

For how many participants were no complete outcome data available?

Not reported

Outcome

Mean total stool weight (72-hour stool collection): 73,8 ± 45,6 g.

Mean fat excretion in faeces: 3,1±2,9 g per day (ranged from <0.1–9.4 g/day).

13/ 24 CF babies were determined PS (fecal fat excretion ranged from <0.1–1.9 g/day)

11/24 infant’s PI (fecal fat excretion ranged from 2.9 to 9.4 g/day).

Mixed Triglyceride Breath Test

Mean basal delta results of the 24 CF babies were

-26.7±3.8%.

Option 1

17/24 CF babies were determined as PI

7/24 CF babies as PS

Option 2

20/24 babies were determined PI

4/24 CF babies PS.

Comparison of the 13C-Mixed Triglyceride

Breath Test Versus the Fecal Fat Balance Test

Option 1

Sensitivity rates

PI: 82%

PS: 38%

specificity rates

PI: 38%

PS: 82%

Option 2

Sensitivity rates

PI: 100%

PS: 31%

specificity rates

PI: 31%

PS: 100%

Fecal Elastase-1 Results According to Monoclonal ELISA Assessments

manufacturer’s instructions:

18/ 24 CF babies PI

6/24 PS

Beharry et al (2002) and Loser et al. (1996):

17/24 PI

7/ 24 CF PS

Fecal Elastase-1 Results According to Polyclonal ELISA Assessments

16/24 CF babies PI,

3/24 CF babies PS

5/24 CF babies had FE1 concentrations between 100–200mg elastase/g stool, indicating moderate pancreatic insufficiency = PS.

Comparison of the Fecal Fat Balance Test and

FE1 Analysis With Monoclonal Kit

manufacturer’s cut-off values

Sensitivity

PI: 100%,

PS: 46%

Specificity

PI: 46%

PS: 100%

cut-off value used by Beharry et al (2002) and Loser et al (1996)

sensitivity

PI: 100%

PS: 54%

specificity

PI: 54%

PS: 100%

Comparison of the Fecal Fat Balance Test and

FE1 Analysis With Polyclonal Kit

Sensitivity

PI: 92%,

PS: 45%

Specificity

PI: 45%

PS: 92%

PI à pancreatic-insufficient require oral pancreatic enzyme replacement

therapy (PERT)

PS à pancreatic-

sufficient

Walkowiak, 2010

Type of study:

Accuracy study

Setting and country: Poland

Funding and conflicts of interest:

Polish Ministry of Science and Higher Education

Inclusion criteria:

-Preselected on the basis of previously performed FFB studies

-CF diagnosis

-no or mild steathorrea

Exclusion criteria:

-

N= 55

age: 7-18 years

Sex: 25 females

Other important characteristics:

All were pancreas insufficient

All had body weight >3^rd percentile

All received 2,000-10,000 U of lipase/kg of body weight/day

AS values lower than 10% were considered to be

normal and higher than 20% as abnormal.

Reasons for incomplete outcome data described?

Not reported

For how many participants were no complete outcome data available?

For 5 subjects 3 day stool collection was not available

Outcome

Correlations between FFE/FFC and AS based upon 1-day stool collection

FFE: r=0.208,

P<0.011;

FFC: r=0.362, P<0.000006,

Correlations between FFE/FFC and AS based upon 3-day stool collection

FFE: r=0.394, P<0.005 FFC: r=0.454, P<0.001

Accuracy AS for determination of abnormal FFC

1 day stool collection for AS cut off levels:

Accuracy AS for determination of abnormal FFC

3 day stool collection for AS cut off levels:

Accuracy AS for determination of abnormal FFE

1 day stool collection for AS cut off levels:

Accuracy AS for determination of abnormal FFE

3 day stool collection for AS cut off levels:

Citation

Level (NHMRC) and

quality (ADA)

Study design and sample

Intervention and outcomes

Results & Conclusions

Comments

Proesmans, 2003

NHMRC Level II

ADA Quality:

Positive

Randomised cross-over design

N=15

(CF subjects; median age 8.7

years)

Inclusion criteria:

· Patients with persistent

symptomatic steatorrhea

despite a daily dose of at

least 10,000U lipase/kg

per day were candidates

for the study

Persistent steatorrhea was

defined as >7g faecal fat/day

(when fat intake is at least

100g/day) or fat absorption

<93%

Exclusion criteria:

· Severe lung disease with

FEV1 <30%

· Pulmonary exacerbation

in the previous month or

liver cirrhosis with portal

hypertension

· Median daily lipase intake

13,500U/kg per day

Intervention:

· Patients started at random

with ‘no omeprazole’ or

‘omeprazole’ for a 1 month

period and then crossed over to the other group

· Patients < 20kg were treated with 10mg daily, patients

>20kg received 20mg

· At the end of each month, a 3 day stool collection and a 3 day weighed food record were obtained allowing calculations of caloric intake, daily intake of fat, protein and carbohydrate

Outcomes:

· Daily stool fat loss

· Fat absorption

Results:

· Daily intake of calories, fat,

carbohydrate and protein was not significantly different between the two

treatment periods

· Median daily faecal fat loss decreased significantly (p>0.01) during omeprazole

treatment, 13g (11.5 – 16.5g/day) to 5.5g (4.9 – 8.1g/day) respectively.

· The same improvement was noted when fat reabsorption was calculated, 87% (81 – 89%) versus 94% (90 – 96%) with omeprazole.

· In all but one patient fat absorption improved

- For this patient treated with 16,000U lipase/kg, fat loss was 10.7g/day (reabsorption 86.3%)

without versus 14.3g/day (87.9%) with omeprazole – there was no obvious explanation for his lack of

improvement

Conclusions:

· Omeprazole can improve fat digestion in CF children with residual steatorrhea

despite high – dose lipase treatment

Author limitations:

· Small sample size

· Should have included all 24

patients and potentially

analysed separately to give a

‘real life’ example of

improvements when patients

may not be compliant with

enzyme supplementation

Appraiser limitations:

· No blinding

Francisco, 2002

NHMRC Level II

ADA Quality:

Neutral

Double blind randomised

placebo-controlled crossover study

N = 22

CF children & adults

Inclusion criteria:

· CF & PI

Exclusion criteria:

· Pregnancy

· Cholestasis

Intervention:

· Measure of baseline PERT

· Adjuvant therapy was

commenced 3 days prior to

admission.

- Paeds <40kg had

ranitidine 5mg/kg or 10mg/kg daily divided

into equal doses 30mins

prior to BF & D.

- Paeds >40kg & adults

received 150mg or 300mg

bd 30 mins before BF.

· Each group was tested whilst having placebo.

· The order of Rx was

randomised.

· Adults were also tested whilst receiving omeprazole 20mg/d 30 mins before BF.

· Fat absorption was calculated & diet controlled.

Outcomes:

· Effect of gastric acid

suppressant therapy with

either ranitidine or

omeprazole on CF pts

receiving ph sensitive enteric coated microtablet.

Results:

· The linear model for all subjects showed no overall adjuvant drug effect on fat absorption p=0.32.

· In adults only drug treatments showed no difference in fat absorption p=0.15

· Paired t test subgroup analysis of adults showed a 4.97% p= 0.003 in mean fat

absorption all other t tests showed no significance comparing low dose

ranitidine to placebo.

· There was marked inter-subject & intrasubject variability in fat absorption.

· No overall sig improvement in fat absorption could be demonstrated

Conclusions:

· No overall sig improvement in fat absorption could be demonstrated

Author limitations:

Adherence with therapy

· Timing of PERT

· Gastric emptying

· Bacterial overgrowth

· Bile acid abnormalities

· Small bowel disease

Appraiser limitations:

· Small sample size

Ng, 2014

NHMRC Level I

ADA Quality:

Positive

Systematic Review

N= 17 trials (273 participants)

Inclusion criteria:

· All randomised and quasi

randomised trials

involving agents that

reduce gastric acidity compared to placebo or

comparator treatment.

· Children and adults with

defined CF, diagnosed

clinically by sweat or gene

testing and including all

degrees of severity.

· All doses and routes of

administration were

considered.

Exclusion criteria:

· Not RCT, examining agents

which are not used to

reduce gastric acidity eg

prokinetic agents vs PPI,

outcomes irrelevant to the

study question.

Intervention:

· Agents that reduce gastric

acidity- PPI or H2 Receptor

antagonists.

· Other drug therapies such as prostaglandin E2 analogues & sodium bicarbonate.

· Compared to placebo mostly

Primary outcome:

· Measures of nutritional status as assessed by weight, height and indices of growth.

· Symptoms related to increased gastric acidity such as epigastric pain or heartburn

· Complications of increased

gastric acidity such as gastric

or duodenal ulcers.

Secondary outcome:

· Faecal fat, faecal nitrogen

excretion and other measures of fat malabsorption

· Measures of lung function

· Measures of quality or life,

mortality, any adverse effects reported.

Results:

Primary outcomes:

· Drug therapies that reduce gastric acidity improve gastrointestinal symptoms such as abdominal pain

· Seven trials reported significant improvement in measures of fat

malabsorption

· Two trials report no significant improvement in nutritional status.

Secondary outcomes:

· Insufficient evidence to indicate weather there is an improvement in nutritional

status, lung function, quality of life or survival.

Conclusions:

· Trials have shown limited evidence that agents that reduce gastric acidity are

associated with improvements in

gastrointestinal symptoms and fat absorption.

· Overall not able to draw firm conclusions from the evidence available.

Limitations (author noted)

Author limitations:

· 14 trials were of a crossover design and did not have the appropriate info to conduct comprehensive meta- analysis.

· The number of trials assessing each of the different agents was small.

· The included trials were generally not reported

adequately enough to allow

judgement of risk of bias.

· Unable to group outcome data in to 1, 3, 6 and 12 month time points as planned.

· Due to lack of trials it was not possible to investigate

heterogeneity between trials using chi2 test and I2 statistic.

· Unable to identify any

reporting biases. No trials

described the methods of

randomization

· Several studies did not

adequately blind or discuss

blinding risk of bias from

incomplete data was unclear in 4 trials.

Appraiser limitations:

· 7 trials limited to children and 3 in adults only.

· Large variation between the trials in terms of design,

duration, treatment and

outcome measures.

Citation

Level (NHMRC) and

quality (ADA)

Study design and sample

Intervention and outcomes

Results & Conclusions

Comments

Littlewood, 2011

NHMRC Level

IV

ADA Quality:

NEUTRAL

Observational, noninterventional

study, single arm

study

N=64

CF patients with PI

(mean age 20.7)

Inclusion criteria:

· PEI patients who had previously taken high doses of pancreatic enzymes (>40000IU lipase/meal).

· Also, patients who require

at least 40000IU lipase/meal (but were previously taking less than this)

Exclusion:

· Hypersensitivity to porcine proteins

Intervention:

· Change to pancreatin 40,000

capsules from lower dose

preparation

Outcomes:

· Serious suspected adverse

drug reactions including

fibrosing colonopathy

· Maldigestion symptoms, body

weight

Results:

· No serious suspected adverse drug reactions related to pancreatin 40000

· No cases of fibrosing colonopathy.

· Mean number of capsules taken per individual per day was notably reduced from 43.8 lower strength capsules to 24.6 of 40000IU capsules (as expected).

· 3 subjects discontinued Pancreatin 40000 due to ‘mild stomach ache/ abdominal pain’ or ‘bowel did not settle.’

· Overall 36% patients had an increase in average daily lipase dose at last visit vs pain reduced- 24.2% at baseline to

10.7% at the end of the study.

· Percentage of patients with fatty stools reduced from 30.2% at baseline to 8.9% at the end of the study

· Mean body weight increased for patients <18 years of age, but remained stable in >18 years of age.

Conclusions:

· No safety concerns arose during this study in which most patients received doses of >10,000 lipase units per kg/day

baseline & 55% had a decrease in average daily lipase dose.

· Percentage of patients with abdominal for 18 months

Author limitations:

· Due to observational study

design with no control or

comparison group it is not

possible to determine what

proportion of improvement in

mal-digestion symptoms was

due to pancreatin 40000 vs

other factors.

· Symptom improvement may be due to improved compliance

due to increased monitoring

during the study period.

· Not possible to determine how

much weight gain was due to

growth in subjects <18years of

age, rather than the pancreatin

40000. Conversely a portion of

patients had a reduction in

body weight, and the reasons

for this cannot be determined

as this study did not collect info

such as pulmonary fx.

· Limited sample size

Appraiser limitations:

· Non-serious drug reactions

were not recorded or reported

as part of this study.

· Outcome measures of

malabsorption and weight are

poorly described, subjective measures and prone to bias/

error etc

· This study only looked at

subjects who required ‘high

doses’ of lipase (>40000IU/meal).

Trapnell, 2009

NHMRC Level

II

ADA Quality:

POSITIVE

Double blind randomised

placebo controlled two period

crossover trial

N=31 Males & females with CF

Inclusion criteria:

· >12 years of age with

confirmed CF & EPI

· Clinically stable, including

stable weight

· On commercially available

PERT product stable dose

for last 3 months.

Exclusion criteria:

· Hx of ileus/ DIOS in last 6

months

· GI malignancy in last 5

years

· Hx of pancreatitis

· Fibrosing colonopathy or HIV.

· Subjects < 18yrs also excluded if BMI centile less than 10%.

Primary objective of the study was to demonstrate superior efficacy of Creon over placebo in improving fat absorption as measured by the CFA.

Intervention:

· An individualised diet

providing at least 100g/day

was provided during days 1 to

5 of the crossover period.

· Subjects were randomised 1:1

to one of two crossover

treatment sequences: Creon

then placebo or placebo then

Creon. Creon 24000 capsules

were administered to achieve

a dose of 4000 IU lipase/ g fat.

A “washout” period of 3 to 14

days separated the study

periods when subjects had

usual diet & enzymes. Blinding

was maintained by provision of

identical capsules and

packaging for placebo and

Creon.

Outcome:

· Efficacy and safety of Creon

24000

Results:

· Mean CFA was significantly greater with Creon 88.6% than placebo 49.6% p<0.001.

Conclusions:

· The data in this study provide strong evidence for the effectiveness of Creon 24000 capsules at a dose of approx.

4000 lipase units/ fat in treatment of EPI in CF.

Author limitations:

· The relatively high dose of

4000IU lipase/ g fat chosen

together with dosing per gram

of fat may not be easily

compared with dosing practices

commonly used in the clinical

setting.

· Lack of dose response data for

PERT

· CFA is not routinely etermined

in clinical practice

· Short term duration of the

study does not allow conclusions regarding long

term tolerability or

symptomatology

Appraiser limitations:

· Creon 24000 not a product

available in Aust or NZ.

· Placebo was not clearly

defined, was difficult to tell if it

was no enzyme or usual enzyme.

· Funding & design of study is an

obvious conflict of interest.

Konstan, 2010

NHMRC Level

II

ADA Quality:

Neutral

Randomised, placebo

controlled study with crossover

design

N=31 subjects with CF >7yrs

Inclusion criteria:

· PI confirmed by faecal

Elastase

· Clinically stable at study

entry

· Taking optimal doses of

PERT product

· Adequate nutrition status

per BMI cut-offs specified

· Able to swallow, able to

eat high fat diet

· On birth control

· Ok to be on PPI or H2

therapy

· Informed consent.

Exclusion criteria:

· Known hypersensitivity to

Ultrase MT or porcine proteins.

· Allergy to stool marker.

· Use of narcotics

· Regularly taking bowel

stimulants

· Hx of bowel resection or

portal HT.

· Acute pulmonary

exacerbation, pancreatitis,

GI conditions, uncontrolled DM, DIOS or any other conditions

known to increase faecal fat loss.

Intervention:

· Dietitian developed,

individualised high fat diet (2g fat/kg body wt) taken for entire comparison phase.

· Pancrealipase (6-7days)-

crossover (up to 10 days)-

placebo (6-7 days) OR vice versa.

· Pancrealipase dose is at a stabilised, individualised dose

as Assessed by the dietitian,

administered under supervision with each meal or snack and recorded.

Primary Outcomes:

· Stools collected- frequency

and characteristics recorded

by study personnel.

· 72hr faecal fat completed- The

CFA%, CAN% calculated using

fat and nitrogen content of

stool & food.

· Adverse events (AE) assessed

and recorded using MedDRA

codification.

Secondary outcomes:

· Medical, physical, biochemical

Ax- pre, during and post study.

Results:

· Patients treated with pancrealipase had

significantly higher mean CFA% than those treated with placebo. P<0.00001

· 76% of patients achieved a CFA% of >85% when treated with pancrealipase compared with 19% of placebo treated

patients.

· Absorption of proteins measured by CAN% showed similar results. P=<0.00001 for the pancrealipase group.

· Patients treated with pancrealipase had reduced number of bowel movements

per day compared with placebo- 1.7 vs 2.9 normal stool consistency movements

respectively (statistical analysis not performed)

· 6 patients reported at least 1 treatment related AE on pancrealipase compared to 18 patients while on placebo. Most

AEs were GI and consistent with CF. No clinically significant effects of treatment were noted physically, on labs or vital

signs.

Conclusions:

· This pancrealipase (Ultrase MT with HP- 55 coating, rather than Eudragit) is a safe and effective treatment for malabsorption associated with PEI.

Author limitations:

· Only limitation vaguely stated

was short treatment period

Appraiser limitations:

· N=31 but only 24 patients

provided analysable data.

· The study does not describe the reason for drop-outs of poor

data collection.

· Small sample size

· Adverse events only vaguely

described as described as ‘GI’

· The study reports measuring

compliance to the study drug

but does not report any data

around how the compliance

was with the drug or the diet

prescription.

· Method of randomisation not

described.

Kashirskaya, 2015

NHMRC Level

IV

Prospective open label multicentre study

N=40 children with CF

Inclusion criteria:

· 1 month to <4 years with

EPI

· Accessing eight centres in

Russia between June and Dec 2012.

· Exclusion criteria:

Intestinal/bowel resection/solid organ transplant impacting

bowel

· DIOS, GI malignancy, hx of

fibrosing colonopathy

Intervention:

· Creon micro administered at a

dose of 5000 lipase units per 120mL of formula or

breastfeed, or 1000 lipase

units/Kg body weight/meal,

according to the CF

Foundation Guidelines.

· Creon micro given with liquid

with pH <5.5 or by adding it to

acidic soft food.

Outcomes:

· Safety- measured by AE

· Efficacy Ax- height and body

weight at baseline, month 1

and month 3 and compared

with standard population using

percentiles/z-scores. Stool

frequency and consistency

measures were also measures.

· Acceptance of treatment and

ease of use- using Likert scale

for subjects and ease of use as

per caregiver.

Results:

· Adverse events occurred in 40% of subjects- none were serious or led to cessation of treatment

· At 3 month mark, mean +/- SD increases from baseline z-scores were height for age 0.13 +/- 0.48, weight for age 0.2 +/-

0.39 and BMI for age 0.29 +/- 0.65.

· Treatment was rated easy by 95% of caregivers and acceptance by subjects was good/very good by 90%.

Conclusions:

· Creon micro was well tolerated . Growth parameters increased over the 3 month treatment period- owning to good

intervention from CF Foundation guidelines.

Limitations (author noted)

Author limitations

· Difficulties measuring height

· Open –label study design

· Lack of placebo/control group

· Short study duration

· Inability to measure clinical

symptoms at baseline

· Lack of direct measurement of

Efficacy

Appraiser limitations:

· Unclear if the time of day

weight was measured and

consistent scales used

· Test of caregivers accuracy of

dosage/food + PERT diary etc

· Lung function was not

acknowledged – how severe a

disease did this patients

have/admissions and this may

effect growth

· Small numbers

Borowitz, 2006a

NHMRC Level

IV

ADA Quality:

Neutral

Open label multicentre dose

ranging study

N=23 subjects from 11

different centres

(ages 13-45 years)

Inclusion criteria:

· Pancreatic insufficient based on fecal elastase

<100ug/g.

Exclusion criteria:

· Pregnancy

· Breast feeding

· DIOS in last 6 months

· Taking any proton pump

inhibitors, H2 receptor

antagonists, or antacids.

· Hx of fibrosing colonopathy, aspergilosis

or liver disease

Intervention:

· Subjects admitted to hospital

discontinued usual PERT.

Subjects received TCT

containing lipase dose of 100,

500, 1000, 2500 or 5000 USP

U/kg with each meal and snack

for 3 days. 72hr 100g fat/ day

diet, nil PERT, stool collected

and measured. Within 14 days

admitted to hospital routine.

CFA was measured.

Primary outcomes:

· To determine the acute safety

and tolerability of 5 dose levels

of TCT (TheraCLEC-Total), a proprietary formulation of

microbial-derived lipase,

protease and amylase.

Secondary outcomes:

· To determine the effect of TCT

on fat & protein absorption, GI

symptoms and stool weight &

frequency

Results:

· TCT increased the CFA of fat and nitrogen absorption in all groups except in the low dose group. At other dosing levels, the mean CFA and nitrogen

absorption increases were 19.1% ± 24.9% and 17.8% ± 13.6% respectively whereas the mean stool weight decreased by 517 ± 362g.

Conclusions:

· TCT was well tolerated in this short term exposure study. The preliminary efficacy data demonstrate lipase and protease activity with little difference seen with lipase doses greater than 500USP U/kg

per meal.

Limitations (author noted)

Author limitations

· Treatment CFA was lower than

previous reports in the

literature, authors cited precise

methodology may have

contributed to this.

· Subjects were instructed to

take their complete dose of TCT

enzymes within 5 minutes of

the start of each meal – this

may have resulted in less than

optimal mixing of enzymes with

food.

Appraiser limitations:

· Small numbers, n = 23 not

powered for formal statistical

testing.

· No control group.

· Did not describe weight (kg).

· Didn’t report significance of

change in CFA

Graff, 2010

NHMRC Level

IV

ADA Quality:

Neutral

Open label multi centre single

treatment arm study

N = 18

Inclusion criteria:

· Children with CF and EPI

· 1 month - <7 years

· Accessing 9 centres across

the US from April 2009 –

June 2009 with weight

>3.75Kg

· Already on a commercially

available PERT with stable

clinical condition

Exclusion criteria:

· At risk of non-adherence

by caregiver or inability to

finish study.

· Fibrosing colonopathy,

DIOS.

Intervention:

· Study drug given as 3000-,

6000-, and 12000- lipase unit

capsules and the correct

number of capsules to be

consumed was calculated to

provide 8000 lipase unit/Kg

body weight/day without

exceeding the max lipase dose

of 10000 lipase units /Kg body

weight/day. Patients on a

higher dose before study,

continued on higher dose (n?).

· According to CF consensus CF

Guidelines

Outcomes:

· Safety and tolerability measured by AEs and weight

at screening/baseline at end of

treatment

· Ease of accurate dosing by

caregivers

· Clinical symptom assessmentsstool frequency and

consistency

Results:

· Patients received a mean +/- SD dose in lipase units/Kg bodyweight/day of 7542

+/- 1335 vs 6966 +/- 3392 on standard therapy.

· 50% patients has AEs but were mild but did not require cessation

· Clinical symptom assessment results were similar between treatments

· Slight preference for study drug over standard therapy with ease of accurate dosing- 33.3% caregivers thoughts study

drug was easier, 5.6% thought study drug was harder

Conclusions:

· Children aged <7 years with EPI due to CF, the new formulation pancrealipase

delayed-release capsules were clinically comparable with standard with standard

therapy in terms of safety, tolerability, efficacy, but not clinically meaningful.

Author limitations

· Small sample size

· Lack of placebo/control group

Appraiser limitations:

· Bias in choosing patients who

would finish study etc.

· Lung function was not

acknowledged – how severe a

disease did this patients

have/admissions and this may

effect growth

· Short study duration

Borowitz, 2006b

NHMRC Level

II

ADA Quality:

Positive

Randomised, double blind,

parallel dose-ranging study

N=117

Inclusion criteria:

· CF & PI

· Had baseline coefficient of

fat (CFA) and coefficient of

nitrogen absorption (CAN)

determined in inpatient

setting while not receiving

PERT

Exclusion criteria:

· Nil stated

Interventions:

· Subjects had a baseline CFA

and CNA while not receiving

PERT. They were then

randomised to treatment with

ALTU-135 continuing 5000

(low), 25,000 (mid) or 100,000

(highest) units of lipase per

meal or snack for 28 days.

After 14 days CFA and CNA

were re-measured.

Outcomes:

· The primary outcomes were

change from baseline in CFA

and CAN treatments.

Results:

· Treatment coefficient of fat absorption was significantly greater in the 25000 and 100000 units of lipase groups

compared to 5000 group.

· Findings for coefficient of nitrogen absorption were similar.

· Subjects with baseline coefficient of fat absorption <40% and >40% in the 25000

and 100000 groups had a mean increase of 31 and 8 percentage points in coefficient in fat

Conclusions:

· ALTU-135 was efficacious during the 1 month period at the dose of 25000 units of lipase

Appraiser limitations:

· Short duration - only for 28

Days

· Unclear how adherence was

measured

· Timing of PERT not defined

Van de Vijver, 2011

NHMRC Level

III-3

ADA Quality:

Positive

Phase 2 randomised,

investigator – blinded, parallel

group pilot study

N=16 paediatric CF patients

(6-30months of age)

Patients were recruited from 5

university medical centres in

Belgium and the Netherlands

Inclusion criteria:

· CF confirmed by genetic

testing or abnormal sweat

test combined with clinical

signs.

· All subjects had a

documented history of

abnormal coefficient of

faecal fat absorption (CFA)

or lower than 15μg faecal

elastase per gram,

confirming a diagnosis of

CF-related PI.

Exclusion criteria:

· Nil stated

Intervention:

· Use of pancrease MT

(pancrelipase microtablets,

2mm, enteric coated) orally

· Run in period (Day 1 – 5)

where participants received

500U lipase/kg/meal at a

maximum of 5 times per day

· Randomisation period (days 6

– 11) where participants were

randomised to receive 500,

1000, 1500, 2000U

lipase/kg/meal

· Study visits were conducted on

day 1, day 6 and day 11.

· A carton was dispensed to the

parent or guardian at visits 1

and 2 and contained 6 bottles

of study medication. Each

bottle contained 5 capsules.

Each capsule contained the

meal – specific dose required

for each dose group and the

appropriate number of microtablets for the subjects

weight in kilograms with a

maximum of 10,000U

lipase/kg/day.

· Parents were instructed to

record in a diary the date, time

and amounts of all foods and

formula consumed and date

and time of study medications

· Patients had to maintain a

defined diet containing 125 –

150% of the recommended

daily allowance appropriate

for patients with CF.

Primary outcome:

· Medication efficacy, assessed

by 72 hour faecal fat exertion,

expressed as a coefficient of

faecal fat absorption (CFA) and

13C mixed triglyceride breath

test.

Secondary outcome:

· Safety and palatability

Results:

Primary Outcome

· Compliance, defined as use of the study medication, was 89 – 99%

· The changes in median weight from screening to the end of the study were 0.05kg (-1 to 0.2) in the 500U group, 0.0kg (-0.1 – 0.7) in the 1000U group, - 0.05kg (-0.2 – 0.1) in the 1500U group and 0.15kg (-0.3 – 0.5) in the 2000U group.

· At the time of randomisation, the median CFA of fat was 93% (91-96) in the 500U group, 90% (85 – 95) in the 1000U group, 83% (67-93%) in the 1500U group and 92% (90 – 96) in the

2000U group.

· None of the 4 doses influenced the CFA relative to the baseline period.

· During the run in period the median cumulative % 13C was 11 (-8 – 59). After randomisation the median cumulative % 13C was 18 (14 – 23) in the 500U group, 14 (-1 – 17) in the 1000U group, 10 (10 – 27) in the 1500U group, and 3 (1 – 49) in the 2000U group.

Secondary outcome

· The mean palatability score was 2.8 during the run in period (range 0 -3) and 2.6 (range 0.3 – 2.8) during the randomisation period. Mean scores >2 were observed in the 500U, 1000U and 1500U group. The 2000U group scored 1.8.

· Gastrointestinal symptoms were reported in some patients – 3 adverse events were reported in 18 enrolled subjects (17%) during the run in period

and 4 events during randomisation. During run in complaints were diarrhoea

(1), vomiting (1), and rhinitis (1). In the 500U group complaints were abdominal pain 91), abnormal stools (1) and

increased bowel movements (1). One patient randomised to 1000U group had constipation. In the 2000U group there was one event of vomiting and 1 of

rhinitis.

Conclusions:

· Pancrease MT at a dose of 500IU lipase/kg/meal resulted in a CFA of approximately 89% in the population studied

· Doses were well tolerated and

palatability was rated fair to good

· Increasing the dose higher than 500IU lipase/kg/meal does not increase the coefficient of fat absorption

· In this study, there was a large variation in cumulative % 13C with the median cumulative 13C in the 500U and 1000U groups in the low – normal range and

the median cumulative 13C in the 1500U and 2000U groups below the normal

range.

Author limitations:

· Large variation in results could

be caused by a wide range of

underlying residual

endogenous lipase activity and

perhaps insufficient breath

sampling.

Appraiser limitations:

· Small sample size

· Inadequate recording of oral

intake/ fat intake

Konstan, 2004

Level II

ADA Quality:

Positive

Randomized double blind

placebo controlled crossover

study

12,000

N=26 (4 withdrawals)

20,000

N= 31 (6 withdrawals)

Inclusion criteria:

· Subjects had CF with

clinically apparent PI with

two or more bowel

movements per day.

· >7 years of age

· Clinically stable

· Established on PERT

· Good general CF health

· Birth control (as required)

Exclusion criteria:

· A known contraindication,

sensitivity or hypersensitivity to porcine pancreatic enzymes

· History of bowel resection

· Pregnancy or lactating

· Present or recent (7 days)

treatment with H2-

receptor antagonists,

antacids, atidiarrhoeal

agents, sucralfate

Intervention:

· The study compared two doses

of one formulation of entericcoated

pancreatic enzymes:

Ultrase MT12 (12 000 lipase

units per capsule) and Ultrase

MT20 (20 000 lipase units per

capsule), to placebo in two

separate safety and efficacy

studies.

· The Pt took a high fat diet, the

enzyme dose did not exceed

2500 IU of lipase per KG per

meal.

· Primary outcome - was

percent absorption of dietary

fat,

· Secondary outcome -

absorption of protein,

· Safety was assessed by

laboratory values, vital signs

and adverse events.

Outcome:

· Safety & efficacy

Results:

· A significant difference in both fat and

protein absorption occurred with the enzyme therapy groups.

· The Ultrase MT12 and Ultrase MT20 groups experienced a mean fat and protein absorption 79.4% and 83.8%,

and 87.3% and 88.6%, respectively.

there was a significant difference in both fat (P = 0.0001) and protein (P = 0.0001)

absorption between enzyme and placebo groups, with more fat and protein being absorbed in the enzyme group

· No adverse events related to study drug were reported.

Conclusion

· The enzymes significantly increased fat and protein absorption. There were no

major adverse events as a result of enzyme therapy.

Author limitations:

· Short time period. A longer

period of reporting could

potentially provide long term

evidence of safety and efficacy.

Appraiser limitations:

· Nil

Trapnell, 2011

NHMRC Level

II

ADA Quality: postive

Randomised, placebo controlled PERT withdrawal study

n=49

Inclusion criteria:

· Males & females aged 7 to

60 years with CF & PI

Exclusion criteria:

· Exacerbation of CF

· DIOS

· Clinically significant GI

symptoms

Intervention:

· Usual PERT was discontinued, a high fat diet initiated (100g fat/day) and Pancrease administered based on

patients usual lipase requirements, to a maximum of 10000 IU lipase/kg/day.

The average daily dose of lipase

during open label & double blind

phases (6274 & 6403 iu/kg/d).

Primary outcomes:

· 72hr stool collection under supervision in clinical research unit and day 3 & 6 of

treatment.

- Weighed and analysed for fat & nitrogen using NMR spectroscopy. CFA & CAN calculated.

Secondary outcomes:

· Drug taken, stool characteristics recorded in subject diary

· Adverse event information collected during each study visit.

Results:

· Pancrease improved fat absorption as shown by significantly lower mean + SD change in CFA between open label and double blind phases for Pancrease (-1.5 + 5.88%; p<0.001) compared to placebo (-34.1 + 23.03%).

Conclusions:

· This study demonstrated Pancrease was effective in treating PI due to CF and was safe and well tolerated.

Author limitations:

· The dose used in the study

limits generalisation of the

recommendations made with regard other lesser doses used

by other CF pts

Appraiser limitations:

· PANCREAZE not a product

available in Aus & NZ

· Funding & design of study is an

obvious conflict of interest.

Graff, 2010

NHMRC Level

II

ADA Quality:

Positive

Multicentre, randomised, double-blind, placebocontrolled,

2-period crossover, superiority study of the new formulation of pancrelipase delated release 12,000-lipase unit capsuled with placebo.

Children with CF and PI aged 7-

11 years

Inclusion criteria:

· On a PERT product at a stable dose for >3/12,

· Clinical stable without acute respiratory disease for at least 1/12 preenrolment.

· Stable body weight.

Exclusion criteria:

· BMI<10%

· Ileus, malignancy of the

digestive tract

· HIV, Coeliac disease, Chron’s disease

· Known allergy to pancrelipase, or exposure to an experimental drug

within 30 days of the start of the study.

Intervention:

· Patients received pancrelipase

12,000-lipase unit capsules or

identical placebo capsules. Target dose of 4000 lipase units/g dietary fat.

· Each patient received an individualised prospectively designed diet containing >40% of calories from fat.

· No minimum daily dietary fat

requirement was implemented.

· Patients remained on usual

PERT day 0-14.

· At visit 2 (day 1 of crossover),

patients were randomised to 1

of 2 treatment sequences –

pancrelipase followed by

placebo or vice versa.

· Each treatment was taken for 5 days.

· Visit 3 took place at the end of the first crossover period (day 6 or 7).

· There was a washout period of 3-14 days during which patients received their previous PERT.

· Patients entered second crossover period at visit 4.

· Dietary recording took place during each treatment period. GI symptoms were assessed.

Primary outcomes:

· Primary – CFA – patients were

administered 2 x 250mg doses

blue food diet 72 hours apary

marking beinning and end of

each stool collection period.

Stool colletions were

performed during each

crossover period.

Secondary outcomes

· CAN

· Clinical symptoms – Clinical

Global Impression of disease

symptoms.

Results:

· During pancrelipase treatment, the

mean (SD) daily lipase dose was 4472

(743) units/g fat consumed (16,941

(3602) IU/kg body weight. Adherence

was 100%.

· Mean (SE) CFA values for pancrelipase and placebo were 82.8% (2.7%) and 47.4% (2.7%) respectively, for a treatment difference of 35.4%

(p<0.001).

· The mean (SA) CAN values for

pancrelipase and placebo were 80.3% (3.2%) and 45.0% (3.2%) (p,0.001)

respectively for a treatment difference of 35.3% (p<0.001).

· Mean (SD) fat intake for pancrelipase

and placebo was 338.3g (59.8) and 346.6 (53.1)g respectively and mean nitrogen intake was 44.6 (20.2) and 45.3 (18.6)g.

· All stool characteristics were

significantly improved and stool

frequency was significantly reduced with pancrelipase compared wth placebo (all, p<0.001). Smyptoms of abdo pain, flatulence and stool consistency were less severe during receipt of pancrelipase compared with placebo. Symptoms remained stable on pancrelipase but worsened slightly on placebo.

· TAEs were reported in 5 patients during pancrelipase treatment and 9 patients during placebo. No TEAEs were considered related to pancrelipase treatment whereas 4 patients had TEAEs considered related to placebo (GI symtpoms).

Conclusions:

· Pancrelipase delayed release capsules were associated with improvements in CFA, CAN, stool properties and EPI symptoms compared with placebo.

· Pancrelipase delayed release capsules appeared to be well tolerated.

Author limitations:

· Limited age range investigated.

Small sample size.

Appraiser limitations:

· 12,000IU preparation n/a in Australia.

Wooldrigde, 2009

NHMRC Level

II

ADA Quality:

Positive

Randomised double blind

placebo controlled two

treatment crossover design

multicentre phase III trial

n=19

Inclusion criteria:

· CF & PI

· >7 years of age

Exclusion criteria:

· Nil stated

Intervention:

· Patients in the randomised trial could receive an of the four dosage formulations of EUR 1008 (5000, 10000, 15000 and 20000 units of lipase/

capsule) with a starting dose of

1000 lipase /kg/meal and a targeted maximum dose of 2500 units of lipase/kg/meal and 4000 lipase / gram fat/

day.

· Patients in the supplemental

study were given the 5000 IU

lipase/ capsule and were given

2000Iu lipase/kg/meal.

Maximum daily dose was

limited to 10000 lipase units/

kg.

Outcomes:

· Safety & efficacy of PERT

Results:

· EUR 1008 treatment compared to placebo resulted in a significantly higher meal CFA 88.3% vs 62.8% p<0.001.

Conclusions:

· EUR 1008 led to clinically and

statistically significant improvements in CFA in the randomised study, and

control of malabsorption and clinical

symptoms in both studies.

Author limitations:

· Nil

Appraiser limitations:

· EUR 1008 not a product

available in Aust & NZ.

· Funding & design of study is an

obvious conflict of interest.

Borowtiz, 2012

NHMRC Level

IV

ADA Quality:

Neutral

N=215 (36 withdrew)

Inclusion criteria:

· PI

· CF >7 yrs age

· Able to take PERT capsules

Exclusion criteria:

· Pregnant/breastfeeding

· Unwilling to use birth

control

· Hx fibrosing colonopathy,

organ transplant,

significant bowel

resection, acute/chronic

diarrhoeal illness

unrelated to PI

· Deranged LFT – liver

transaminases >5 x upper

limit of normal, bilirubin >

1.5 x upper limit of normal

· Inability to discontinue

enteral feeds

· Known hypersensitivity to

food additives

· Pancreatic sufficiency

Intervention:

· Liprotamase 32,500 USP, 1 capsule mid meal x3 daily, 1 capsule with snacks twice daily.

· The main meal dose was allowed to be increased from 1 – 2 capsules, based on strict criteria in study protocol, (steatorrhea, wt loss or lack of wt gain in paeds).

Primary Outcomes:

· Evaluation of long term safety and tolerability of Liprotamase treatment in subjects with CF related EPI

Secondary Outcomes:

· Nutritional status through serial measures of height, weight, BMI z scores at each time point, referenced to year 2000 CDC growth charts, as well as lung function FEV1.

Results:

· Overall 211 (98.6%) of the 214 subjects experienced at least one treatment emergent adverse event (TEAE).

· 36 (16.8%) of the subjects discontinued due to a TEAE.

· There were another 33 withdrawals

unrelated to study treatment.

· 6 subjects (3%) had ALT values >5 times ULN, mostly transient.

· Ht, wt & BMI z scores were relatively

stable over the remaining treatment

period.

· FEV1 remained stable.

Conclusions:

· Treatment with a mean 5.5 capsules

daily of Liprotamase with meals &snacks for 12 months was safe & well tolerated in CF subjects with EPI. Liver

enzymes were not elevated beyond

Author limitations:

· Possible selection bias as

investigators may have invited

participants with GI symptoms.

Appraiser limitations:

· Didn’t measure fat absorption

in this study, only indirectly.

· Possible Bias due to drug

company involved in study

design analysis and

interpretation of data.

· Almost 20% of participants

withdrawing due to a TEAE is

possibly an indication that

some people don’t tolerate this

enzyme

· Author mentioned vitamin expected. Subjects maintained or gained weight & BMI z scores. Pulmonary function was maintained. levels mainly stable as evidence of absorption, but patients were on a ADEK vitamin supplement (supplied in the USA)

Wooldridge, 2014

NHMRC Level

IV

ADA Quality:

Neutral

Prospective cohort study

N=15 (3 withdrew)

Multi-site (6) study of patients

who had completed previous

study and continued treatment

with ZENPEP. 12 month open

label follow up.

Inclusion criteria:

· Between ages of 1 and 12

months diagnosed with CF

· Needing PERT

· Mmonoclonal fecal

elastase >200 ug/g stool

· Weight to length ratio

>10th %ile, and clinically Stable

Exclusion criteria:

· Weight to length ratio

<10th %ile

· Clinically unwell

- Concomitant illness, acute URTI, or LRTI.

Intervention:

· Ongoing clinical care

- ZENPEP 3000 lipase

(PERT)

Primary outcomes:

· Safety - frequency, duration, and severity on treatment emergent adverse events

(TEAE).

Secondary outcomes:

· Growth parameters

- weight, length, weight for age %ile, length for age %ile, weight for length

%ile, & Z scores

Results:

Primary

· 5 Patients experienced 15 TEAEs judged possibly (n = 13) or probably (n = 2) related to medication.

Secondary

· Median body wt increased 7.8 kg to 11 kg.

· Median body lgth was 68.5 cm increased to 81 cm.

· Median wt for age %ile increased from 22nd to the 49th %ile. Median lgth for age %ile increased from the 36.5th %ile to

42nd.

· Median wt for length %ile increased

from the 41.5th to the 55.5th %ile.

· Growth Z scores were variable.

· Mean wt for lgth z scores positive at all treatment visits.

· Median wt for lgth z scores positive at 6 and 12 months.

· Although positive overall the

improvement was not linear over time.

Conclusions:

· Infants with CF safely tolerated long

term Zenpep for up to 12/12.

· Also wt for age %ile more than doubled

Author limitations:

· Pts with wt to length ratios

>10th %ile were excluded

· Small sample size

Appraiser limitations:

· Nil

Munck, 2009

NHMRC Level I

ADA Quality:

Positive

Prospective randomised study of crossover design

N= 39

Inclusion criteria:

· CF

· PEI

Exclusion criteria:

· Meconium ileus

· Intestinal resection

· Severe concomitant diseases of other organs

· Allergy to pancreatin

· Auspected noncompliance of subject/family

Intervention:

· Comparison of Creon 10,000 preparation (10000IU lipase

per capsule) to Creon for

Children preparation (5000IU

lipase per scoop of granules)

Primary Outcome:

· Parent’s treatment preference

Secondary outcomes:

· Coefficient of fat absorption

· Clinical symptoms and safety

parameters

Results:

· 51% of parents preferred CfC, 23%

preferred C10 and 26% had no

preference.

· Mean CFA was similar for both 77.8% vs 78.7%

· Safety and tolerability were good for

both treatments

· GIT symptoms- no clinically relevant

differences.

· Mean daily energy intake was 1100

kcal/day for a mean body weight of 10.1 kg.

· Fat intake was about 40 g/day with

comparable mean daily lipase intakes

during the CfC and C10 phases (3969 vs. 4310 U/kg/day).

Conclusions:

· Those parents who had a preference

preferred CfC over C10, but both

enzyme preparations improved

malabsorption to a similar degree

· The mean faecal fat excretion was

higher than desired (9.2 and 8.3 g per

day during CfC and C10, respectively).

Author limitations:

· Applied dosages could have

been too low in some children

reflected in suboptimal CFA.

· Parents may under report

symptoms

· No attempt was made to

optimise creon dose before the

randomisation

Appraiser limitations:

· Parent preference is a very subjective primary outcome

measure and the questionnaire

used provided varying quality

of responses.

· Low clinical impact/relevance

of patient preference for one

preparation over the other

given there were no significant

differences found in actual

safety, tolerability, GIT

symptoms and CFA between

preparations. Hence this study

is just about ease of handling

CfC.

Konstan, 2013

NHMRC Level

II

ADA Quality:

Positive

Randomised, double blind,

placebo controlled cross over

N=21

CF children (>7years) and

adults with confirmed PI

Multicentre – 5 centres across

USA

Inclusion criteria:

· Children aged 7-18 yrs and

adults with confirmed CF

· PEI per faecal elastase

· Currently receiving PERT.

· ‘Adequate nutritional

status’ per BMI

Exclusion criteria:

· Major organ diseases or syndromes affecting digestion i.e. diabetes,

liver, intestinal obstruction, IBD,

pancreatitis.

· Contraindication to porcine products, acute exac on recent Abx

Intervention:

· Active study drug and identical

placebo capsules. Treatment

period (6-8days)-crossover/restabilisation

(7-10 days)- placebo (6-8 days) OR vice

versa.

· Subjects individually stabilised to an EC- bicarb buffered PERT dose at home.

· Individualised high fat diet (2g fat/kg/day) taken for entire comparison phase.

Primary outcomes:

· 72hr stool collection under supervision in clinical research unit and day 3 & 6 of

treatment.

· Weighed and analysed for fat & nitrogen using NMR spectroscopy. CFA & CAN calculated.

Secondary outcomes:

· Drug taken

· Stool characteristics recorded in subject diary

· Adverse event information collected during each study visit.

Results:

· Statistically significant improvement in fat & nitrogen absorption for active

treatment vs placebo.

· Mean co-efficient of fat absorption was 82.5% with Enteric coated bicarbonate buffered PERT, compared with 46.3% with the placebo. P=<0.001

· Overall stool frequency decreased by

40% and stool weight decreased by 50% p=<0.001.

· No safety concerns were identified.

· The most common adverse event was GI complaints and these were more common during placebo than active treatment and found not to be clinically meaningful.

Conclusions:

· This unique enteric coated-bicarbonate buffered PERT is safe and effective in treating CF associated PEI.

Author limitations:

· Bicarb buffered PERT product

compared to placebo not a

non-buffered PERT product

Appraiser limitations:

· Compliance/outcome data

· Intake at home, drug taken

stool frequency and

characteristics ‘recorded in

subject diary’ by subject at

home – potentially very

subjective

Brady, 2006

NHMRC Level

II

ADA Quality:

Neutral

Prospective randomised controlled trial with crossover

Design

N=18 CF patients with PI

Aged 12.2-27.6 years

Inclusion criteria:

· CF & PEI diagnosis

· Consumption of large EC

PERT >1000IU lipase/kg/meal

· Fat excretion >15% of dietary fat intake when consuming 50% of usual

enzyme dose

· Adequate BMI

· Mild lung disease

· No Abx

· No major organ disease that affects digestion or absorption.

Exclusion: Nil stated

Intervention:

· A unique EC high-buffered

pancrealipase formulation vs

usual brand of EC nonbuffered

PERT.

· 7 days in each treatment group- 3 days at home washout followed by 4 days at

GCRC of:

· Controlled enzyme doses (at 50% of the subjects usual lipase dose & given

immediately before meals and snacks)

· Controlled food intake, weighted and recorded with even distribution of fat.

Primary Outcome:

· Feacal fat excretion per and 72hrs of stool collection, which were counted weighed and analysed using a ravimetric method. Looked at faecal fat as a % of fat intake and g/kg/day

Secondary Outcome:

· Coefficient of fat absorption % of intake

· Nitrogen excretion

Results:

· Mean fat excretion decreased

significantly in each subject during

periods where EC high buffered

pancrealipase compared with periods of EC non buffered enzymes.

· Fat absorption 81.8% vs 75.1%

respectively p=0.01

· 72% of patients (13/18 subjects)

excreted less fat with EC high buffered

PERT. Of these, 56% (10 subjects)

decreased fat excretion by more than

5% (range 6-17%).

· NB: >5% considered clinically significant (arbitrarily chosen cutoff).

· However, 5 subjects (28%) did not

respond to the EC high buffered PERT.

Conclusions:

· EC high buffered pancrealipase

decreased fat excretion, symbolising

increased fat absorption, when compared to EC non buffered PERT

given at equivalent, reduced lipase dose in nourished CF patients with mild lung disease.

Author limitations:

· One patient did not adhere and their data was excluded.

· Relatively small sample size

· Some variability in fat intake

· Open label study design=

potential to bias subjective

data.

· Duodenal PH was not

determined, thus explanation

of the effect of bicarb in the gut

is hypothetical and can’t

explain mechanism of action.

Eg: why 5 subjects did not have

a response at all.

· Short term study period

Appraiser limitations:

· Mishandling of lab specimens

requiring a 3rd study period for

some subjects, which was

completed as an outpatient, not in the centre.

· 20 subjects was required for

90% power and only 18

analysable data sets were

received.

· Subjects usual enzyme brand

varied between 6 different

brands/capsule sizes

Baker, 2005

NHMRC Level

IV

ADA Quality:

Neutral

Retrospective cross sectional study

N= 1215

Inclusion criteria:

· >4wks of age

· CF diagnosis from accredited centres

· Willingness to fill out questionnaire and collect small random stool

specimen without stopping or changing PERT.

Exclusion criteria:

· Acute diarrhoea

· Surgically created enterostomy

· Short bowel

Intervention: nil

Outcomes:

· Primary- pancreatic sufficiency- faecal elastase.

· Enzyme dosing, growth, abdominal pain, gassiness, constipation, number of stools.

Results:

· <2yrs of age- 25% of PS patients had a weight and height <5th percentile

compared with 8.8% of PI children.

· From age 3-20 there was no difference in the distribution of BMI percentiles between PS and PI.

· >20years of age- 41% of PS patients

were of healthy BMI compared with 74% of the PI.

· There was no relationship between PERT dosing and BMI.

· There was no significant difference

between patients with PS and PI in

number of daily stools and dose of PERT had no effect on number of stools.

· Gassiness and constipation were present significantly more in PI than PS patients. However higher PERD does did not decrease these symptoms.

Conclusions:

· PERT does not correlate with growth or gastrointestinal symptoms.

Author & appraiser limitations:

· Nil

Haupt, 2011

NHMRC Level

III-2

ADA Quality:

Positive

Retrospective study

N=14482

Inclusion criteria:

· 2-20 years of age

· Taking PERT

· CF Registry from January 1

2005 – December 31 2008

Exclusion criteria

· Programs with <10 unique patients on enzyme supplementation were

excluded from centre level analyses

Intervention:

· Nil

Outcomes:

· PERT dose

- The highest reported

enzyme dose on any

encounter during the

year.

- Individual enzyme dose

was compared for all

patients at programs

within the top and

bottom BMI percentile

quartiles.

- Patient and centre level

characteristics and

median centre enzyme

dose was also compared

between top and bottom

BMI-percentile

performance quartiles.

· Demographic and clinical

factors that may be

confounders associated with

nutritional outcomes including age at diagnosis, sex, race, FEV1% predicted

Results:

· 42561 patient visits from 14482 patients from 179 programs.

· Highest quartile programs had a mean enzyme dose of 1755 lipase units/kg/meal (95% CI: 1722, 1788)

compared with 1628 lipase units/kg/meal (95% CI: 1595, 1660) for

lowest quartile ( P < .001).

· 52% of patients in top quartile programs had enzyme doses greater than 2000 lipase units/kg/meal

· The median (25th, 75th percentile)

enzyme dose from the last visit of all

patients was 1755 (1329, 2150) lipase

units/kg/meal, and the average BMI

percentile of all patients was the 45.2

percentile (SD 27.0%).

· Although the top quartile programs'

median enzyme dose was significantly

higher than that of the bottom quartile, the median enzyme dose was lower than the maximum dose recommended by the CF foundation, ranging from 1787- 1836 lipase units/kg/meal, less than the maximum recommended dose of 2500 lipase units/kg/meal.

· Note these values are less than the

enzyme doses reported in the most

recent CF Registry, reflecting the general trend of increased enzyme dose over time

Conclusions:

· CF centres that prescribe higher average pancreatic enzyme dosing have better nutritional outcomes, as measured by average BMI percentile.

· Even though the magnitude of

difference in PERT dose may not appear large, only ∼150 lipase units/kg/meal, reducing minor malabsorption at each meal could enhance weight gain over time.

Author limitations:

· Adherence to PERT and other

therapies, dietary

recommendations, or other CF

centre practices as unmeasured variables

· Several statistically significant

differences exist between the

patient characteristics of lowest

and highest quartile programs

· Patients at top quartile

programs were younger,

despite not showing a

statistically significant

difference in age at diagnosis.

Top quartile and bottom

quartile patients also

demonstrated statistically

significant differences in height

and weight for age percentile

· Selection bias, information bias, and confounding

· CF Registry is not designed to

capture all adverse events of

enzyme replacement therapy

Appraiser limitations:

nil

Citation

Level (NHMRC) and

quality (ADA)

Study design and sample

Intervention and outcomes

Results & Conclusions

Comments

Van der Haak, 2016

NHMRC level: II

ADA: Positive

Randomised, double blind

crossover trial.

n=18 CF children (5-17yrs)

Inclusion criteria:

· Confirmed CF

· PI

· Stable dose of PERT with

satisfactory symptom

control within current

recommendations (500-

4000IU/g fat ≥ 6months)

· Nil changes re dose during

Study

Exclusion criteria:

· Severe pulmonary disease

· Previous GI surgery or

meconium ileus

· Enteral nutrition

· Medication affect GI

motility

Aim to determine if PERT given after a meal, instead of before a meal, could result in improved

lipase activity in CF children.

Primary intervention:

· 13C-mixed triglyceride (13CMTG) breath test to assess pancreatic lipase activity when Creon (1333IU/g fat) given

10min prior to test meal vs.

10min after test meal.

· Breath samples taken before meal & at 30min intervals for 360min

Secondary intervention:

· 13C-octanoate breath test to assess gastric emptying

· Breath samples taken before meal & at 15min intervals for 120mins then 30min intervals for 240mins

· Light, low fat meal offered at 240min mark

Primary outcome:

· Timing of PERT on lipase

Activity

Secondary outcome:

· Effect of gastric emptying on lipase activity

Results:

Primary

· No significant difference in PCDR with PERT before (32.5±10%) or after meal

(31.2±13.7%) at 360min (p=0.68).

· When PERT was taken after a meal, PCDR was higher in normal vs. fast GE (p=0.04)

** PCDR = percentage cumulative dose recovered

Secondary

· 6/18 accelerated gastric emptying (37.3±5.4min)

· 11/8 normal gastric emptying (55.7±15.2min)

· 1/18 delayed gastric emptying (121.5min)

· No significant difference in mean PCDR of 13C-MTG with PERT before or after

meal in those with accelerated (p=0.06) or normal (0=0.31) gastric emptying

· No correlation between gastric half emptying time and PCDR when PERT taken before (r= -0.29, p=0.24) vs. After (4=0.05, p=0.85) meal

Conclusions:

· No difference in mean lipase activity when taking PERT 10 min before vs.

after meal.

· Changing timing of PERT can normalise lipase activity an and individual level.

Gastric emptying rate may influence optimal timing of PERT.

· In those with fast GE it may be beneficial to take PERT before a meal.

Author limitations:

· Small sample size (unable to do sub-group analysis)

· No concurrent control

· Unable to assess intraindividual

variability in results

(13C-MTG only taken once at each time point)

Appraiser limitations:

· Patients recruited as inpatient or outpatient

- No mention of if study was

done in inpatient or

outpatient setting

- Potential impact of being

unwell if done in inpatient

setting

· Children only (no young

children/infants or adults)

Study reference

Patient selection

Index test

Reference standard

Flow and timing

Comments with respect to applicability

Caras, 2011

Was a consecutive or random sample of patients enrolled?

Yes

Was a case-control design avoided?

Yes

Did the study avoid inappropriate exclusions?

Yes

Were the index test results interpreted without knowledge of the results of the reference standard?

Yes

If a threshold was used, was it pre-specified?

Yes

Is the reference standard likely to correctly classify the target condition?

No

Were the reference standard results interpreted without knowledge of the results of the index test?

Yes

Was there an appropriate interval between index test(s) and reference standard?

Yes

Did all patients receive a reference standard?

Yes

Did patients receive the same reference standard?

Yes

Were all patients included in the analysis?

No

Are there concerns that the included patients do not match the review question?

No

Are there concerns that the index test, its conduct, or interpretation differ from the review question?

No

Are there concerns that the target condition as defined by the reference standard does not match the review question?

No

CONCLUSION:

Could the selection of patients have introduced bias?

RISK: LOW

CONCLUSION:

Could the conduct or interpretation of the index test have introduced bias?

RISK: LOW

CONCLUSION:

Could the reference standard, its conduct, or its interpretation have introduced bias?

RISK: HIGH

CONCLUSION

Could the patient flow have introduced bias?

RISK: LOW

Kent, 2018

Was a consecutive or random sample of patients enrolled?

Yes

Was a case-control design avoided?

Yes

Did the study avoid inappropriate exclusions?

Yes

Were the index test results interpreted without knowledge of the results of the reference standard?

Unclear

If a threshold was used, was it pre-specified?

Yes

Is the reference standard likely to correctly classify the target condition?

No

Were the reference standard results interpreted without knowledge of the results of the index test?

Yes

Was there an appropriate interval between index test(s) and reference standard?

Yes

Did all patients receive a reference standard?

Yes

Did patients receive the same reference standard?

Yes

Were all patients included in the analysis?

Yes

Are there concerns that the included patients do not match the review question?

No

Are there concerns that the index test, its conduct, or interpretation differ from the review question?

No

Are there concerns that the target condition as defined by the reference standard does not match the review question?

No

CONCLUSION:

Could the selection of patients have introduced bias?

RISK: LOW

CONCLUSION:

Could the conduct or interpretation of the index test have introduced bias?

RISK: LOW

CONCLUSION:

Could the reference standard, its conduct, or its interpretation have introduced bias?

RISK: HIGH

CONCLUSION

Could the patient flow have introduced bias?

RISK: LOW

Walkiowak, 2010

Was a consecutive or random sample of patients enrolled?

unclear

Was a case-control design avoided?

Yes

Did the study avoid inappropriate exclusions?

Yes

Were the index test results interpreted without knowledge of the results of the reference standard?

Yes

If a threshold was used, was it pre-specified?

Yes

Is the reference standard likely to correctly classify the target condition?

No

Were the reference standard results interpreted without knowledge of the results of the index test?

Yes

Was there an appropriate interval between index test(s) and reference standard?

Yes

Did all patients receive a reference standard?

Yes

Did patients receive the same reference standard?

Yes

Were all patients included in the analysis?

No

Are there concerns that the included patients do not match the review question?

No

Are there concerns that the index test, its conduct, or interpretation differ from the review question?

No

Are there concerns that the target condition as defined by the reference standard does not match the review question?

No

CONCLUSION:

Could the selection of patients have introduced bias?

RISK: LOW

CONCLUSION:

Could the conduct or interpretation of the index test have introduced bias?

RISK: LOW

CONCLUSION:

Could the reference standard, its conduct, or its interpretation have introduced bias?

RISK: HIGH

CONCLUSION

Could the patient flow have introduced bias?

RISK: LOW