Gastroesophageal Reflux Disease in Children with Neurological Impairment Raj Srivastava, MD, FRCP(C), MPH Center for Pediatric Clinical Effectiveness,

Gastroesophageal Reflux Disease in Children

with Neurological Impairment

Raj Srivastava, MD, FRCP(C), MPH Center for Pediatric Clinical Effectiveness,

Children’s Hospital of Philadelphia,Dec 5th, 2008

Overview

• Children with neurologic impairment– Why this population?– High resource utilization

• Nutritional and respiratory conditions

• Current Studies

• Future Steps – Care process model – Multi-center studies

Why This Population?

• Children with NI are living longer and use increasing resources of the health care system

• NI results from many different conditions but they share several common clinical issues (e.g. nutritional and respiratory conditions)

• Lack of sufficient evidence base due to small numbers• Idiosyncratic practices within institutions• Lack of a clear medical group that studies outcomes in

this population• Opportunity for further study (generalists and specialists,

multidisciplinary)

High Resource Utilization

High Resource Utilization

• Agency for Healthcare Research Quality– Kid’s Inpatient Database– 1997, 2000, 2003

• National estimates of hospitalizations– Children ages 0-18 years

– Clustered, stratified, weighted sample• 2.9 million hospital discharges• 3,400 hospitals• 38 states

Total Number of HospitalizationsChildren with Neurologic Impairment

Year

0

50000

100000

150000

200000

250000

300000

350000

1997 2000 2003

33% increase, p <.01

Year

Hospital Type 1997 2000 2003

All Hospitals* 3.9% 4.6% 5.1%

Children’s Hospitals* 9.0% 11.9% 12.6%

Percentage of All Hospitalizations Children with Neurologic Impairment

*p < 0.01

Nutritional and Respiratory Conditions

• Children with neurological impairment (NI) and severe functional limitations have dysfunctional swallowing and gastroesophageal reflux disease (GERD) placing them at risk for aspiration pneumonia (AP)

• AP leads to repeated hospitalization, respiratory failure, compromised quality of life, and death.

• Initial GERD management consists of medications

Nutritional and Respiratory Conditions

• Some children with NI and GERD fail medical management

• There are competing management approaches for treating the GERD in order to prevent AP and subsequent respiratory failure

• Few published studies comparing treatment efficacy and quality of life outcomes with long-term follow-up for treating these conditions in this population

Fundoplication• Fundoplication is an anti-reflux

procedure used to treat GERD in children who have failed medical management.

• Fundoplication is the third most common procedure performed by pediatric surgeons in the U.S.

• Half of these procedures are performed on children with NI

Gastrojejunal feeding tubes• GJ tubes are an anti-reflux

procedure used to treat GERD in children who have failed medical management.

• GJ tubes are frequently used in children with NI

Background and Rationale

• Multi-center randomized control trial (RCT) is the best approach to answering questions about GERD management in this population:– Equipoise– Outcomes – Sufficient number of patients– Expertise to conduct study

Clinical Question

• In children with neurodevelopmental disabilities who have GERD and have failed medical management, what is the next best management option?– Compare time to develop AP and survival – Treatment with first fundoplication compared to first

gastrojejunal feeding tube (change from gastrostomy to GJ feeding tube in radiology)

Study Design and Timeline

Enrollment Period for Cohort

Pre-

Enroll

men

t

Period

Jan

1997

Oct 20

06

Born

Outcomes

Dec 2

005

Post-

Enroll

men

t

and

F/U

Period

• Retrospective cohort followed for outcomes • Excluded medical management only patients

First Fundo or First GJT

GERDNI

AP Death

Inclusion criteria

Enterprise Data WarehouseIntegrated Reporting and Analysis

A single source for complex data analysis and reporting

EDWEDWFinancial

Data

FinancialData

Clinical Data

Clinical Data

Claims&Eligibility

Claims&Eligibility

Slide Courtesy of Brent James, Intermountain Healthcare

Study GroupsFundoplication

N = 323Gastrojejunal Feeding Tube

N = 43p-value

Age at time of procedure (mean)

16 months (S.D. 16 months)

24 months (S.D. 20 months)

0.008

Gender (female) 146 (45%) 13 (30%) 0.07

Previous AP 50 (15%) 9 (21%) 0.36

Tracheostomy 21(7%) 9 (21%) <0.001

Cerebral Spinal Fluid Shunt 38 (12%) 12 (28%) 0.004

Chronic Lung Disease 50 (15%) 7 (16%) 0.89

Seizures 117 (36%) 21 (49%) 0.11

Relative surgical contraindications¶

28 (9%) 12 (28%) <0.001

Complex Chronic Condition

CCC* - Cardiovascular 27 (63%) 139 (43%) 0.014

CCC* – Other congenital or genetic defect

124 (38%) 24 (56%) 0.028

Reason for Neurological Impairment▲

Cerebral Palsy 165 (42%) 20 (47%) 0.55

Brain or Spinal Cord Anomaly 122 (38%) 20 (47%) 0.26

Chromosomal Anomalies 50 (15%) 11 (26%) 0.09

SurvivalFundoplication vs. Gastrojejunal Tube

Aspiration Pneumonia FreeFundoplication vs. Gastrojejunal Tube

Comparative Studies

Fundoplication GJ tube p-value

Albanese et al 1993 (n =112)

Mortality 4 (8.8%) Mortality 2 (5.9%) Non-significant

Wales et al

2001 (n =111)

Mortality 11 (17.5%)

AP 23 (36.5%)

Mortality 6 (12.5%)

AP 15 (31.3%)

Non-significant

Srivastava et al

in press (n=366)

Mortality 40 (12%)

AP 48 (15%)

Mortality 9 (21%)

AP 7 (16%)

Non-significant

Objectives

• To examine child and caregiver quality of life for children with NI who received a first fundoplication for treatment of GERD

Eligible: NI or at risk GERD0 – 21 years old

At time of procedure

Screening +

Entry

Fundoplication

Gastrojejunal Feeding Tube

1) Baseline Functional Status – (WeeFIM®)2) Child QoL – PedQL®/CHQ3) Caregiver QoL – PSI/SF364) Nutrition Outcomes

Repeat all assessments

Primary Outcome = Child QoL, Caregiver QoL (1 year post-procedure)Secondary Outcome = Nutrition, Mortality, Adverse Events, Costs, Long-Term Outcomes

1 month post

procedure

6 month post

procedure

Repeat all assessments

Prospective Study

Study EnrollmentChildren with a

Fundoplication at PCMC Jan 2005 – Feb 2007

N = 236

Eligible Patients with FIRST Fundoplication

N =95

Excluded Patients N = 137

-Procedure Not Done-G Tube Only-No NI Diagnoses -Redo Fundoplication-Spanish speaking-Unknown

Excluded Patients N = 42

-Declined-Unable to Consent-Materials Not Retrieved-DiedStudy Cohort

N = 53Baseline Data

Reason for NI

Genetic Syndrome 34 (57%)

Developmental Delay 29 (48%)

Seizures 20 (33%)

Hypotonia 18 (30%)

Microcephaly 10 (17%)

Cerebral Palsy 10 (17%)

Brain Injury 8 (13%)

Hydrocephalus 7 (12%)

Anoxic Brain injury 6 (10%)

Neurodegenerative Disease 5 (8%)

Brain Abnormality 5 (8%)

Spinal Muscular Atrophy 3 (5%)

Myotonic Dystrophy 2 (3%)

CNS Infection 2 (3%)

Characteristics

VariablesStudy Patients

N = 60

Age (years) 2.2

Complete Dependence on mobility, communication and self-care

54 (90%)

Indications for Fundoplication  

Vomiting 32 (52%)

Feeding related 32 (52%)

Failure to thrive 25 (42%)

Failure of medical therapy 59 (98%)

6 month OutcomesVariables Study Patients

N = 60

Ongoing Symptoms – vomiting, gagging or aspiration 26 (43%)

ED visits* post-fundoplication 14 children; 18 visits

Admissions* post-fundoplication 15 children; 25 admits

Revision of fundoplication 4 (6.7%)

Weight (increase) 1.8 kg

Weight for z score (increase) -2.15 to -1.44, p <0.001

Death – none related to complication of surgery 5 (8%)

*Visits related to a complication, the gastrostomy tube, AP or GERD symptoms

Child HRQoL

Baseline 6 months p-value

General Health 30.2 30.9 0.81

Physical Functioning

19.6 22.3 0.72

Behavior 73.7 83.9 0.03

Parent Time Impact

46.6 57.6 0.09

Family Activities 44.4 54.3 0.15

Bodily Pain

Caregiver Quality of Life

Domain of Quality of Life Study Group

Mean Study Group S.D

U.S. Norm Population Mean

U.S. Norm S.D.

p-value

Physical Functioning 86.74 18.15 84.15 23.28 0.54 Role Physical 84.78 26.60 80.96 34.00 0.43 Bodily Pain 82.92 19.95 75.15 23.69 0.02 General Health 70.19 19.98 71.95 20.34 0.55 Vitality¶ 45.58 19.04 60.86 33.04 0.001 Social Functioning 74.02 26.20 83.28 22.69 0.005 Role Emotional 77.24 28.32 81.26 33.04 0.39 Mental Health 65.58 19.5 1 74.74 18.05 0.001 ¶ Comparisons of study cohort to adults with clinical depression in the domains of Vitality mean score 40 S.D 21.08, p = 0.72.

Caregiver Quality of Life

Domain of Quality of Life Study Group

Mean Study Group S.D

U.S. Norm Population Mean

U.S. Norm S.D.

p-value

Physical Functioning 86.74 18.15 84.15 23.28 0.54 Role Physical 84.78 26.60 80.96 34.00 0.43 Bodily Pain 82.92 19.95 75.15 23.69 0.02 General Health 70.19 19.98 71.95 20.34 0.55 Vitality¶ 45.58 19.04 60.86 33.04 0.001 Social Functioning 74.02 26.20 83.28 22.69 0.005 Role Emotional 77.24 28.32 81.26 33.04 0.39 Mental Health 65.58 19.5 74.74 18.05 0.001 ¶ Comparisons of study cohort to adults with clinical depression in the domains of Vitality mean score 40 S.D 21.08, p = 0.72.

Caregiver Quality of Life

Domain of Quality of Life Study Group

Mean Study Group S.D

U.S. Norm Population Mean

U.S. Norm S.D.

p-value

Physical Functioning 86.74 18.15 84.15 23.28 0.54 Role Physical 84.78 26.60 80.96 34.00 0.43 Bodily Pain 82.92 19.95 75.15 23.69 0.02 General Health 70.19 19.98 71.95 20.34 0.55 Vitality¶ 45.58 19.04 60.86 33.04 0.001 Social Functioning 74.02 26.20 83.28 22.69 0.005 Role Emotional 77.24 28.32 81.26 33.04 0.39 Mental Health 65.58 19.5 1 74.74 18.05 0.001 ¶ Comparisons of study cohort to adults with clinical depression in the domains of Vitality mean score 40 S.D 21.08, p = 0.72.

Caregiver Quality of Life

• Total Stress remains high during study period• Significantly higher stress than parental norms • 1 in 3 parents expressed clinically significant

levels of stress (scores > 90, 90th percentile)

Caregiver Quality of Life

• Total Stress remains high during study period• Significantly higher stress than parental norms • 1 in 3 parents expressed clinically significant

levels of stress (scores > 90, 90th percentile)• 1 in 3 parents of children with TBI, 1 in 5 parents

of children with CHD

Goals of Caregivers

• Symptom Reduction– Stop vomiting– Decrease gagging– No more aspiration

• Nutrition– Gain weight– Eat by mouth

• Medical – Stay healthy– Keep out of hospital

Cohort Identification

Birth cohort (2000-2005)N = 955,285

>1 Neuro CodeN = 144,749 (15.2%)

No Neuro CodeN = 810,536 (84.8%)

FundoN = 6,716 (24.3%)

No FundoN = 21,004 (75.7%)

GERDN = 27,720 (19.2%)

No GERDN = 117,029 (80.8%)

Died N = 264 (4.1%)

AliveN = 6,452 (95.9%)

Birth cohort (2000-2005)N = 955,285

>1 Neuro CodeN = 144,749 (15.2%)

No Neuro CodeN = 810,536 (84.8%)

FundoN = 6,716 (24.3%)

No FundoN = 21,004 (75.7%)

GERDN = 27,720 (19.2%)

No GERDN = 117,029 (80.8%)

Died N = 264 (4.1%)

AliveN = 6,452 (95.9%)

Table 3. Unadjusted rates per yearPre Post IRR* p

AP 0.12 (0.1, 0.14) 0.1 (0.08, 0.13) 0.86 (0.76, 0.97) 0.0138GERD 0.75 (0.66, 0.85) 0.67 (0.56, 0.79) 0.9 (0.85, 0.94) <.0001Esophagitis 0.01 (0.01, 0.02) 0 (0, 0.01) 0.29 (0.16, 0.53) <.0001Mechanical Vent 0.35 (0.31, 0.39) 0.27 (0.24, 0.3) 0.73 (0.68, 0.79) <.0001Pneumonia 0.19 (0.17, 0.23) 0.27 (0.24, 0.32) 1.35 (1.25, 1.47) <.0001Asthma 0.16 (0.13, 0.2) 0.25 (0.2, 0.32) 1.58 (1.45, 1.73) <.0001Any RREs 1.63 (1.45, 1.81) 1.63 (1.43, 1.86) 0.98 (0.95, 1.02) 0.3443*Incidence Rate Ratio (Post:Pre)

Table 4. Adjusted IRR using Table 1 covariates.IRR* p

AP 0.72 (0.63, 0.82) <.0001GERD 0.62 (0.59, 0.65) <.0001Esophagitis - -Mechanical Vent 0.42 (0.39, 0.46) <.0001Pneumonia 1.07 (0.98, 1.17) 0.1465Asthma 1.51 (1.37, 1.65) <.0001Any RREs 0.7 (0.68, 0.73) <.0001*Incidence Rate Ratio (Post:Pre)

Median Time to RRH

Next Steps

• Evidence-based best practice care process model for GERD and dysphagia evaluation, medical management and when to refer

Medical Management n=60 all with NI who received a Fundoplication

• Various medications had been tried and were considered to have failed in these patients– 39% had been treated with acid suppressive agents– 80% with acid blocking agents– 61% with prokinetic agents

• Duration?• Dose?• Impact on Symptoms?

Diagnosis of GERD

Child with Neurological Impairment (A) who presents with a

Feeding Concern or Nutritional Aberration (B)

Clinical Evaluation reveals cause other than

GERD or Dysphagia(C)

YES Protocol Does Not Apply

GERD Positive?

Evaluate for Dysphagia (D)Go to Page 2

YES Dysphagia Positive?

YES

NO

NO

Dysphagia Positive?

YES

NO

GERD flow diagram (July 9, 2008) Page 1

1

23

Evaluate for GERD (D)Go to Page 3

(AND)

GERD PositiveDysphagia Positive

Go to Page 4B Box 42

GERD PositiveDysphagia Negative

Go to Page 4A Box 41

GERD NegativeDysphagia Positive

Go to Page 5D Box 58

GERD Negative Dysphagia Negative

Go to Page 5C Box 57

NO

4

5

6 7 8

9

10

11

12

Next Steps

• R03 for Oct 2009 – Two hospital study– Pilot data for GJ tubes– Equipoise– Two hospitals (feasibility of multi-center study)

• Pilot study – Using evidence-based best practices guideline

Next Steps

• R01 for June 2009 – Multi-center study– Which first procedure?– Feasibility and Inclusion criteria– Outcomes– Number of sites, patients, power– DCC – Protocol– Investigator Meeting or at APSA/other

meetings

Acknowledgements

• Child Health Research Center, Primary Children’s Medical Center Foundation

• Eunice Kennedy Shriver National Institute for Child Health and Human Development

K23 HD 052553

Future Steps

• Prospective, multidisciplinary• Specifically defined patient population• Clear cut diagnostic criteria for GERD• Uniform implementation of medical therapy• Stringent documentation throughout the course

of diagnosis and therapy with objective data• Feedback to providers regarding outcomes of

children• Study comparing surgical therapies:

fundoplication/GT vs GJ tube placement