Immune deficiencies, infection, and systemic immune disorders

Clinical outcome in children with chronic granulomatous

disease managed conservatively or with hematopoietic
stem cell transplantation
Theresa Cole, BM,a Mark S. Pearce, PhD,b Andrew J. Cant, MD,a Catherine M. Cale, PhD,c David Goldblatt, PhD,d and
Andrew R. Gennery, MDa Newcastle upon Tyne and London, United Kingdom

Background: Chronic granulomatous disease (CGD) is a surgery, and admissions compared with post-HSCT children.
primary immunodeficiency characterized by serious infections Children undergoing transplantation have better height for age.
and inflammation. It can be managed conservatively with Survival is good at the end of the pediatric age range and also
prophylactic antimicrobial agents or curatively with after HSCT. (J Allergy Clin Immunol 2013;132:1150-5.)
hematopoietic stem cell transplantation (HSCT). In the United
Kingdom and Ireland there are cohorts of children managed Key words: Chronic granulomatous disease, hematopoietic stem
both conservatively and curatively. cell transplant, children
Objectives: This study aimed to compare clinical outcomes
(mortality and morbidity) in children managed conservatively
and curatively. Chronic granulomatous disease (CGD) is characterized by
Methods: Children were identified from specialist centers and serious, life-threatening bacterial and fungal infections associated
advertising through special interest groups. Clinical data were with abnormal inflammatory responses, leading to colitis, restric-
collected from medical records regarding infections, tive lung disease, chorioretinitis, and granulomas in the liver,
inflammatory complications and growth, other admissions, and lung, skin, and lymphoid tissue. United Kingdom (UK) and
curative treatment. Comparisons were made for patients not European data have shown that despite improvement in manage-
undergoing HSCT and patients after HSCT. ment of CGD over recent years, there is considerable mortality by
Results: Seventy-three living children were identified, 59 (80%) the third decade of life.1-3 Hematopoietic stem cell transplanta-
of whom were recruited. Five deceased children were also tion (HSCT) can cure CGD and improve growth.4 CGD has
identified. Clinical information was available for 62 children (4 also been treated with gene therapy, although with limited success
deceased). Thirty (48%) children had undergone HSCT. thus far.5 There remains much discussion about the optimum tim-
Children who did not undergo transplantation had 0.71 episodes ing of HSCT. Recently published data demonstrate that children
of infection/admission/surgery per CGD life year (95% CI, 0.69- with CGD have poor quality of life compared with those who
0.75 events per year). Post-HSCT children had 0.15 episodes of have undergone HSCT.6
infection/admission/surgery per transplant year (95% CI, 0.09- The UK currently has 2 similarly sized cohorts of children with
0.21 events per year). The mean z score for height and body CGD: those managed conservatively and those who have under-
mass index (BMI) for age was significantly better in post-HSCT gone HSCT. This provides a unique opportunity to examine the 2
children. Survival in the non-HSCT group was 90% at age 15 groups and contribute further information regarding the appro-
years. Survival in the post-HSCT group was 90%. priate timing of HSCT.
Conclusions: Children with CGD not undergoing This study compares clinical features of children managed
transplantation have more serious infections, episodes of conservatively and those who have undergone transplantation
with regard to the following: numbers of infections, inflammatory
From the Institutes of aCellular Medicine and bHealth and Society, Newcastle University, complications, growth, surgical procedures, and admissions to the
Newcastle upon Tyne; cthe Department of Clinical Immunology, Great Ormond Street hospital.
Hospital, London; and dthe Institute of Child Health, University College London.
This article presents independent research funded by the National Institute for Health
Research (NIHR) under its Research for Patient Benefit (RfPB) Programme (grant ref-
erence no. PB-PG-0909-19060). The views expressed are those of the author(s) and not
METHODS
necessarily those of the National Health Service, the NIHR, or the Department of Children aged 16 years and less were eligible for this study because they are
Health. transitioned to adult care after this point. Potential participants were identified
Disclosure of potential conflict of interest: T. Cole has received a grant from the National from the UK and Ireland CGD Registry, which was initially compiled in 2000
Institute for Health Research. The rest of the authors declare that they have no relevant and was designed to identify all patients with CGD in the UK and Ireland.1 It
conflicts of interest. was compiled by contacting consultants across the UK and Ireland and asking
Received for publication February 18, 2013; revised May 12, 2013; accepted for publi- whether they had any patients with CGD. Once patients were enrolled, infor-
cation May 20, 2013. mation was gathered by researchers from clinical records. The last information
Available online July 16, 2013.
was added to the registry on enrolled patients in 2004; no new patients were
Corresponding author: Theresa Cole, BM, Department of Paediatric Immunology and In-
added after this time. New patients receiving a diagnosis since the registry
fectious Diseases, Old Children’s Outpatients, Royal Victoria Infirmary, Newcastle
upon Tyne, NE1 4LP United Kingdom. E-mail: [email protected]. was compiled were identified through advertising by special interest groups.
0091-6749/$36.00 These groups included the British Paediatric Allergy, Immunology and Infec-
Ó 2013 American Academy of Allergy, Asthma & Immunology tion Group; the Travellers (adult immunologists); the British Society for Pae-
http://dx.doi.org/10.1016/j.jaci.2013.05.031 diatric Gastroenterology, Hepatology & Nutrition; the British Society for

1150
J ALLERGY CLIN IMMUNOL COLE ET AL 1151
VOLUME 132, NUMBER 5

measurements available, z scores were calculated for age-related weight (up to

Abbreviations used age 10 years), age-related height, and age-related BMI by using World Health
AR: Autosomal recessive Organization (WHO) growth standards with WHO Anthro software version
BMI: Body mass index 3.2.27 for children aged 0 to 5 years and WHO Anthro plus software for chil-
CGD: Chronic granulomatous disease dren age 5 years and older.8 The z score expresses the data as a number of SDs
HSCT: Hematopoietic stem cell transplant greater than or less than the mean. A score of less than 22 indicates a measure
UK: United Kingdom 2 SDs less than the mean and is defined as low height/weight/BMI for age.
WHO: World Health Organization Comparisons were made for children who did not undergo transplantation
with those who were at least 2 years after HSCT because growth significantly
improves in the post-HSCT group by 2 years after transplantation.4
Statistical analysis was performed with SPSS version 17.0 software (SPSS,
Gastroenterology; the British Society for Haematology; the British Paediatric Chicago, Ill). Comparisons were made for non-HSCT and post-HSCT serious
Respiratory Society; the British Thoracic Society; and the British Infection infections, surgical procedures, and other admissions. Comparisons of
Association. Two major centers care for many pediatric patients with CGD, categorical data were compared by using the x2 test and the Fisher exact
and other centers with cohorts of pediatric patients with CGD were identified test. Nonnormally distributed continuous data were analyzed by using the
and approached through pediatric immunology and HSCT specialists. These Mann-Whitney U test. Continuous data were compared by using
were identified from the original registry, knowledge of teams at the 2 major independent-sample t tests.
national centers, and the CGD nurse specialist.
Children with a confirmed diagnosis of CGD on nitroblue tetrazolium
testing or flow cytometry with or without protein expression abnormalities or
identified genetic mutations were eligible for the study. Children were defined RESULTS
as having X-linked or autosomal recessive (AR) inheritance according to the Seventy-three living children were eligible for inclusion, of
results of flow cytometry and demonstration of absent nicotinamide adenine whom 59 (80%) were recruited. Five children who died before
dinucleotide phosphate oxidase components. commencement of the study were also identified. Sixty-four
A favorable opinion was obtained from the local regional ethics committee, (82%) patients were seen at the 2 national centers, and 14 children
and approval was obtained from the National Information Governance Board were identified from 5 other centers. Clinical information was
Ethics and Confidentiality Committee for access to records of deceased abstracted for 62 children, 58 of the living children and 4 who died
patients in England and Wales. Written consent was obtained from parents for
before commencement of the study. No records were available for
participating subjects.
Once children were identified, cross-checking of names and dates of birth
data abstraction for 1 living and 1 deceased patient. See Table I for
from different centers was performed to prevent duplication of cases. If children baseline demographics.
were cared for at more than 1 hospital (eg, specialist transplant care or annual X-linked children received diagnoses when significantly youn-
specialist review at 1 center) along with their local hospital, these were ger than children with AR inheritance (P 5 .017). There was no
identified. Notes were reviewed at the specialist centers and also local hospitals, significant difference in age at diagnosis for male and female pa-
where possible. Data were gathered by a researcher manually on a paper case tients with AR disease: 5.21 years compared with 4.06 years, re-
report form at a single time point. This date was used as the date of abstraction spectively (P 5 .451). Genetic mutation was documented in the
for calculation of CGD life years per transplant years. Data collected from notes of 24 (39%) patients. Other than for siblings, the mutations
medical records with a standard case report form included the following: growth were different in each patient.
(height and weight at outpatient visit), respiratory and gastrointestinal symp-
There were a total of 470 life years for CGD; this was made up
toms, infections, hospitalizations, medications, surgery, and treatments re-
ceived, including HSCT or gene therapy. Hospital admissions were classified
of 286 life years for patients who did not undergo transplantation
according to the diagnosis given by the clinical care team at the time. and 184 pretransplant years for patients who underwent
For patients who underwent transplantation, data were collected regarding transplantation.
pretransplantation health and growth, conditioning for transplantation, At the time of data collection, 30 (48%) children had under-
posttransplantation course, and complications. Further information was gone 34 HSCT procedures. Thirty were first transplantations.
gathered for time since transplantation, including growth, number of hospi- Twenty-eight (93%) were male. The median age at first trans-
talizations, infections, and medication. plantation was 5.25 years (range, 0.7-15.3 years). Median
For children receiving a diagnosis since the registry was developed, posttransplantation follow-up for living patients was 3.84 years
information was gathered in line with the original registry, covering sex, (range, 0.2-9.9 years). Total transplantation follow-up years were
diagnostic tests, and mode of inheritance. For deceased patients, information
124. Twenty-one transplants were conditioned with busulfan and
from the medical records was collected regarding progress up until the point of
death and cause of death.
cyclophosphamide, 8 with treosulfan and fludarabine, 3 with
Number of CGD life years were calculated for children managed conser- busulfan and fludarabine, and 1 with treosulfan with
vatively (date of birth until date of data abstraction or death) plus children cyclophosphamide.
managed with HSCT (date of birth until HSCT). Total number of transplant There were 138 infection episodes (0.29 per CGD life year;
years was calculated (date of transplantation until date of data abstraction or 95% CI, 0.24-0.33 infections per life year; Table II). There
death). If children received more than 1 transplant, the latest transplantation were a total of 20 proved or presumed fungal infections. There
date was taken for calculating the number of transplant years. were 8 proved fungal pneumonias (5 Aspergillus fumigatus, 2
Number of events was calculated for commonly occurring infections (as Aspergillus nidulans, and 1 Scopulariopsis species), 7 pre-
previously identified1): suppurative adenitis, pneumonia, perianal abscess, sumed fungal pneumonias, 2 cases of osteomyelitis, and 2
liver abscess, osteomyelitis, septicemia, brain abscesses, and splenic abscess.
brain abscesses.
The median age at diagnosis of infection was calculated. Numbers of infec-
tions, as described above, were calculated for the posttransplantation period.
One child had 2 episodes of suppurative adenitis after HSCT
Number of admissions for other reasons and surgical procedures were calcu- 2.04 and 3.17 years after transplantation. Two children had
lated for pretransplantation and posttransplantation time. pneumonia diagnosed during their transplantation admission and
Height and weight measurements were collected from patient records at 1 had pneumonia a week after discharge after HSCT. One child
least once per year when available. Using the most recent height and weight was given a diagnosis of a fungal brain abscess during his
1152 COLE ET AL J ALLERGY CLIN IMMUNOL
NOVEMBER 2013

TABLE I. Baseline demographics posttransplantation group compared with the nontransplanted

Male sex 52 (92%) group (Table V). Weight-for-age z scores were calculated for 18
X-linked inheritance 53 (85%) Median age at diagnosis non-HSCT and 7 post-HSCT children at less than 10 years.
(y [range]): 1.81 (0-11.9) Four children had low weight-for-age scores (z score < 22) in
AR inheritance 9 (15%) Median age at diagnosis the non-HSCT group. No post-HSCT child had low weight for
(y [range]): 4.2 (0.31-14.8) age.
P47Phox 3 Pretransplantation and posttransplantation weights and heights
P40Phox 1 were available for 14 children who had undergone HSCT with at
AR defect not 5 least 1 year of posttransplantation follow-up. Median time after
documented HSCT was 4.3 years (range, 1.3-9.0 years). Seven children did not
Post-HSCT 30 (48%) Median age at 1st HSCT
have z scores for weight for age calculated because they were
(y [range]): 5.25 (0.7-15.3)
older than 10 years at the most recent measurement. Therefore
Median age at data collection
(y [range]): 10.57 (3.90-15.22) weight-for-age comparisons before and after transplantation
were not performed. The mean height-for-age z score before
Non-HSCT 32 Median age at data collection
transplantation was 21.41 (SD, 1.03). The mean height-for-
(y [range]): 8.95 (0.51-16.4)
age z score after transplantation was 20.54 (SD, 1.07). Three
(21%) children had height-for-age z scores of less than 22, which
was deemed low height for age before HSCT. One child continued
transplantation admission after his second HSCT. No child had to have low height for age after HSCT. Mean BMI for age before
any other significant infection after discharge after HSCT. The transplantation was 0.2 (SD, 0.77). Mean BMI for age after trans-
number of episodes of infection in the non-HSCT and post-HSCT plantation was 0.49 (SD, 1.04). No child had low BMI for age be-
patients is described in Table III. fore or after HSCT. There was a significant improvement in height
Twenty-five (40%) children had CGD colitis (24 X-linked) at a for age but no significant difference in BMI for age when compar-
median age of 3.99 years (range, 0.28-12.24 years). Thirteen ing pretransplantation and posttransplantation data (P 5 .011 and
children with colitis went on to transplantation. Six children had P 5 .42, respectively).
11 admissions for gastrointestinal symptoms related to colitis. All children with CGD received antibiotic prophylaxis with
There were 52 admissions for other reasons in 30 children (0.11 cotrimoxazole, except 1 patient who died during his presenting
admissions per CGD life year; 95% CI, 0.08-0.14). Median age at acute illness. Fifty-four (87%) had antifungal prophylaxis with
admission was 3.5 years (range, birth to 12.1 years; Table IV). itraconazole. Six were treated for fungal infections at diagnosis
Thirty-seven children underwent 97 surgical procedures (0.21 and therefore continued to receive antifungal agents other than
surgical procedures per CGD life year; 95% CI, 0.17-0.24 events itraconazole at the time of data collection. Three children were
per year). Median age at surgery was 3.2 years (range, 0.28-13.6 switched to voriconazole prophylaxis because of difficulties with
years; Table IV). itraconazole. No child received prophylactic IFN-g; 13 children
There were a total of 15 admissions in 10 children after did receive it as part of treatment for severe infection.
discharge after HSCT (0.12 admissions per transplant year; 95% At the time of data collection, 12 (40%) children in the post-
CI, 0.06-0.18). Median age at admission was 6.4 years (range, HSCT group were still receiving prophylactic medication
1.4-12.4 years). Median time after HSCT was 0.75 years (range, (cotrimoxazole, aciclovir, antifungal agents, immunoglobulin
0.08-4.83 years). Ten (67%) admissions were within the first year. replacement, or penicillin). For those more than 1 year from
The admissions at greater than 1 year after HSCT were for 2 upper transplantation, 9 (39%) of 23 were still receiving some form of
respiratory tract infections, headache and confusion, a seizure, prophylaxis.
and new-onset type 1 insulin-dependent diabetes mellitus. Of the 5 deceased children identified at the start of the study, 2
All children who underwent HSCT had at least 1 central venous underwent transplantation and 3 did not undergo transplantation.
line for the procedure. One child required evacuation of an Two further children died after recruitment to the study, 1 after
intracranial hematoma after a fungal brain abscess during his HSCT and the other while awaiting a second round of gene
transplantation admission. Two children underwent surgical therapy.
procedures after discharge after HSCT. One was a roux-en-Y Causes of death were as follows: respiratory failure caused by
enterostomy for a bile duct stricture associated with previous mulch pneumonitis in a previously undiagnosed patient; Burkhol-
hepatobiliary surgery before HSCT. The other was excision of a deria cepacia sepsis and cerebrovascular event after central ve-
lower lip lesion. nous catheter insertion in the non-HSCT children; with fungal
Height and weight measures were available for 30 children not infection, multiorgan failure after conditioning, and severe influ-
undergoing transplantation and 15 children with greater than 2 enza in the post-HSCT children. Information was unavailable for
years post-HSCT follow-up. Most recent height and weight 1 deceased non-HSCT patient. Median age at death in the non-
measures were used for analysis. Median age for the non-HSCT HSCT group was 6.84 years (range, 0.92-16.41 years). Median
group was 9.0 years (range, 0.83-15.25 years). Median age for the age at death in the post-HSCT group was 9.18 years (range,
post-HSCT group was 10.1 years (range, 4.17-13.8 years). 8.55-12.58 years). There was no statistically significant difference
Median time after HSCT was 4.42 years (range, 2.25-9.00 years). in age at death for the 2 groups (P 5 .513).
Six (20%) children in the non-HSCT group had low height for age
(z score < 22). No child in the post-HSCT group had low height
for age. Five children had low BMI for age in the non-HSCT DISCUSSION
group compared with none in the posttransplantation group. This study reviewed the clinical course in children with CGD
Height for age and BMI for age were significantly better in the and those who have undergone transplantation. It demonstrates
J ALLERGY CLIN IMMUNOL COLE ET AL 1153
VOLUME 132, NUMBER 5

TABLE II. Episodes of serious infection commonly associated with CGD in all patients treated conservatively and in patients
before transplantation
Total no. of No. of patients Total no. of episodes Median age Median no. of episodes
patients (%), n 5 62 who had later HSCT for all patients (y [range]) per patient (range)

Suppurative adenitis 34 (55) 18 49 2.20 (0.04-11.6) 1 (1-4)

Pneumonia/empyema 27 (44) 10 37 4.50 (0.25-13.4) 1 (1-4)
Perianal abscess 12 (20) 6 24 1.63 (0.3-12.6) 1 (1-8)
Liver abscess 10 (16) 4 11 2.38 (0.25-4.2) 1 (1-2)
Osteomyelitis 5 (8) 4 6 5.82 (0.5-10.3) 1 (1-2)
Septicemia 4 (6) 3 7 2.67 (0.5-4.1) 2 (1-2)
Brain abscess 2 (3) 2 2 7.13 (3.5-7.18) 1
Splenic abscess 2 (3) 2 2 2.66 (1.4-4.0) 1

TABLE III. Total episodes of infection, admission, and surgery in children who did not undergo HSCT and children who did
undergo HSCT
Non-HSCT and
pre-HSCT events (no. of Events per CGD Post-HSCT events Events per transplant
patients), n 5 62 year (95% CI) (no. of patients), n 5 30 year (95% CI)

Infection 138 (50) 0.29 (0.24-0.33) 2 (1) 0.02 (0.006-0.04)

Colitis 11 (6) 0.023 (0.01-0.04) 0 —
Colonoscopy 36 (23) 0.08 (0.05-0.1) 2 (1) 0.02 (0.006-0.04)
Other admissions 52 (30) 0.11 (0.08-0.14) 15 (10) 0.12 (0.06-0.18)
Surgery 97 (37) 0.21 (0.17-0.24) 2 (2) 0.02 (0.006-0.04)
Total 334 (54) 0.71 (0.69-0.75) 19 (12) 0.15 (0.09-0.21)

fewer infections and admissions to the hospital and improved potentially life-threatening infection. Of the less serious infec-
growth in patients after transplantation compared with those who tions in the post-HSCT group, two thirds were within the first year
had not undergone transplantation. These data show that children after transplantation and were transplant related. The causes for
in the UK with CGD have significant morbidity, with a serious admission at greater than a year after transplantation were varied
infection once every 3.5 years; undergo a surgical procedure once and not specifically transplant related. This suggests that by a year
every 4.8 years; and have an admission to the hospital for any after transplantation, the children are relatively well. However, it
cause once every 1.4 years. Serious infections were only seen in is not possible to determine whether the rate of admission is
patients who had not undergone transplantation or before higher than the background rate for healthy children for all
transplantation. causes.
Suppurative adenitis and pneumonia were the most common Forty percent of children had CGD colitis. This is similar to the
infections, similar to the original UK registry, Italian registry, and findings of Jones et al,1 but it is higher than described in both the
European registry.1-3 Liver abscesses appear less common than in Italian registry (13%)2 and the European registry.3 The reasons for
the original UK registry or European registry, which reported 79 these differences are unclear and might relate to the definition and
episodes in 27 (29%) patients1 and 32% of patients,3 respectively. detail of data collection. It might be that in the UK we are more
This study shows only 16% of patients had a liver abscess, which actively asking our patients about and recording their gastrointes-
is similar to the frequency reported in the Italian registry.2 This tinal symptoms. Children might not volunteer the fact that they
apparent reduced frequency might be due to only collecting have diarrhea unless specifically asked, and not all parents will
data on children. Some of the children presented here could go be aware of their child’s bowel habits, particularly as they get
on to have liver abscesses as they get older. As expected, fungal older. Although mean z scores for height, weight, and BMI for
infections were common. Of those that were proved, A fumigatus age were within the normal range, a proportion of children with
remained the most common pathogen. However, many were pre- CGD did have scores of less than the normal range.
sumed rather than proved infections, emphasizing the difficulties Approximately half of the children in the UK and Ireland had
in identifying fungal infections. undergone HSCT at the time of data collection. This is a
Other infections were relatively common, particularly upper significant change from historical practice. This study includes
respiratory tract and skin or soft tissue infections. It is difficult to a larger group of children who underwent transplantation than any
draw conclusions about the rate of admission for upper respiratory published case series.
tract infections compared with immunocompetent children. Sixty percent of patients who underwent transplantation
There are many variations from season to season and year to discontinued prophylactic medication by 1 year after transplan-
year for the population. It is likely that children with CGD present tation. However, the 2 main centers performing transplantation in
to the hospital for evaluation more frequently because parents are patients with CGD have different policies: one stops all medica-
encouraged to seek medical help if concerned about fever or tion, and the other continues life-long pneumococcal prophylaxis.
respiratory infection in order not to miss a significant and However, this is a positive point for families considering HSCT.
1154 COLE ET AL J ALLERGY CLIN IMMUNOL
NOVEMBER 2013

TABLE IV. Admissions for other reasons, including surgical TABLE V. Comparison of z scores for most recent height,
procedures, in patients who did not undergo transplantation weight, and BMI for age in children who did not undergo
and patients before transplantation transplantation and children after HSCT who had height and
Reason for admission No. of events No. of patients weight documented
Non-HSCT Post-HSCT P value of
Respiratory tract infection* 14 10
group group difference
Skin and soft tissue infection 9 8
Viral illness 6 6 Height for age n 5 30 n 5 15
Central venous catheter infection 2 2 Mean (SD) 21.15 (1.15) 20.29 (0.75) .012
Other  21 14 Range 23.42 to 1.85 21.62 to 1.11
Total 52 30 BMI for age n 5 29 n 5 16
Surgical procedures Mean (SD) 20.37 (1.55) 0.58 (0.88) .012
Incision and drainage/aspirationà 59 27 Range 23.26 to 2.50 20.70 to 2.25
Laparotomy 9 6 Weight for age n 5 18 n57
Cystoscopy 2 2 Mean (SD) 0.82 (1.25) 20.24 (0.91) .141
Lymph node/soft tissue biopsy/ 11 6 Range 23.53 to 0.90 21.45 to 1.44
excision
Central venous access 3 2
Other surgery§ 13 7
Total 97 37
immunodeficiencies, including better donor-recipient matching,
*Tonsillitis, pharyngitis, croup, bronchiolitis, cough, otitis media, and exacerbation of
less toxic conditioning chemotherapy, and more availability of
asthma.
 Preseptal cellulitis, severe varicella zoster infection, seizures, acute diarrhea (not antiviral treatment, along with better supportive care. A recent
colitis), mediastinal mass, and pericardial effusion. small American case series of 11 children demonstrated 100%
àTwenty-nine cases of suppurative adenitis, 23 other soft tissue infections, and 7 survival, which provides hope for the future.13 These studies pro-
perianal abscesses. vide reassurance that HSCT is a safe and effective procedure in
§Bone marrow aspirate, pulmonary lobectomy, pulmonary artery stenting, rectal
biopsy, wound debridement, nasal space washout, transjugular portosystemic shunt,
children with CGD.
gastroenterostomy. The UK is unusual in having 2 distinct CGD cohorts of similar
ages: the nontransplantation and posttransplantarion cohorts.
Many other primary immunodeficiencies are either treated with
Without HSCT, it is absolutely necessary to continue antibiotic HSCT early in life (eg, severe combined immunodeficiency) or
and antifungal prophylaxis lifelong, whereas after HSCT, it is are managed with prophylaxis (eg, antibody deficiencies). This is
possible to discontinue all medication. This is particularly the first study to compare the outcomes of these 2 different groups
attractive for teenagers, who are known for poor compliance.9-11 in CGD.
There is currently no UK guideline for managing CGD, but cen- The major limitation to this study is the small numbers
ters adhered to current European accepted practice, with use of involved. Despite this, it does include the majority of the UK
cotrimoxazole and itraconazole but no routine use of IFN-g. pediatric CGD population. This is a result of the rarity of CGD.
Three children in each of the nontransplantation and post- Recruitment was good for face-to-face contact but less successful
transplantation groups died. Ninety percent of the children who for those approached by post. The postal approach was only used
had not undergone transplantation were still alive at age 15 years. when it was not possible to set up face-to-face meetings at clinic
This is similar to the survival rate reported by Jones et al,1 who appointments. Clinical information was available for 62 children.
found 88% survival at age 10 years. Pneumonia and septicemia Notes were unavailable for 1 deceased child, and some available
were common causes of death in the original registry. These records were limited so that some information might have been
were also the causes of 2 of the 3 deaths in the non-HSCT group missed, and rates of infection and other morbidity parameters are
in the current data. This suggests we might have become better at likely an underestimate of true values. This study is unable to
managing children with recognized CGD so that they are now draw conclusions about long-term morbidity and survival in
rarely dying in childhood. It is not possible to draw any other con- patients with CGD because it only includes pediatric data.
clusions about long-term survival because data collection stopped However, it is encouraging that identified patients were receiving
at age 16 years in this study. Further analysis of adult deaths appropriate prophylactic treatment, and survival on conservative
would be important to assess how much the survival curve has treatment is better at 15 years than previously published. It will
changed and whether long-term survival has really significantly also be necessary to evaluate adult patients with CGD to evaluate
improved. whether-long term survival has changed.
At 90%, long-term posttransplantation survival is good, and the This study shows that children with CGD in the UK still have
deaths all occurred relatively early after transplantation (the latest significant morbidity, but those who have undergone transplan-
at 2 years). Survival is similar to that demonstrated by Soncini tation have much less frequent trips to the hospital, far fewer
et al,4 who described one of the largest groups of CGD transplan- infections, and much less need for surgery. Children who under-
tation in recent years. However, this study does include some chil- went transplantation have better height for age. At a relatively
dren who were also included by Soncini et al.4 Survival in this short time of follow-up, mortality in the transplantation and
study is better than the posttransplantarion survival quoted for nontransplantation groups was similar. Longer-term studies of
the historical European series of 27 cases (85% survival) by Seger patients with other primary immunodeficiencies after HSCT show
et al12 and in the European registry, which was 80%.3 Better that HSCT-associated deaths almost all occur soon after HSCT,
survival in the current study might reflect overall improvement and therefore long-term survival might be better in the HSCT
in transplantation survival for patients with primary group, but long-term studies would be needed to fully answer this
J ALLERGY CLIN IMMUNOL COLE ET AL 1155
VOLUME 132, NUMBER 5

question. These data, along with previously reported improved cure chronic granulomatous disease with good long-term outcome and growth.
quality of life,6 do not answer the difficult question about who Br J Haematol 2009;145:73-83.
5. Holland SM. Chronic granulomatous disease. Clin Rev Allergy Immunol 2010;38:
should undergo transplantation and at what stage, but they do pro- 3-10.
vide further evidence for the importance of considering transplan- 6. Cole T, McKendrick F, Titman P, Cant AJ, Pearce MS, Cale CM, et al. Health
tation early after diagnosis for CGD and referring families to a related quality of life and emotional health in children with chronic granulom-
transplantation center for discussion. atous disease: a comparison of those managed conservatively with those that
have undergone haematopoietic stem cell transplant. J Clin Immunol 2013;33:
8-13.
Clinical implications: Children who have undergone HSCT for 7. WHO Anthro for personal computers: software for assessing growth and develop-
CGD are healthier than their counterparts conservatively man- ment of the world’s children.3.2.2 ed. Geneva: World Health Organization; 2011.
aged with prophylaxis, with few admissions to the hospital and 8. WHO AnthroPlus for personal computers: Software for assessing growth of the
world’s children and adolescents. Geneva: World Health Organization; 2009.
few infections. 9. Adeyemi AO, Rascati KL, Lawson KA, Strassels SA. Adherence to oral antidia-
betic medications in the pediatric population with type 2 diabetes: a retrospective
REFERENCES database analysis. Clin Ther 2012;34:712-9.
1. Jones LBKR, McGrogan P, Flood TJ, Gennery AR, Morton L, Thrasher A, et al. Spe- 10. Mackner LM, Crandall WV. Oral medication adherence in pediatric inflammatory
cial article: chronic granulomatous disease in the United Kingdom and Ireland: a bowel disease. Inflamm Bowel Dis 2005;11:1006-12.
comprehensive national patient based registry. Clin Exp Immunol 2008;152:211-8. 11. Dobbels F, Van Damme-Lombaert R, Vanhaecke J, De Geest S. Growing pains:
2. Martire B, Rondelli R, Soresina A, Pignata C, Broccoletti T, Finocchi A, et al. non-adherence with the immunosuppressive regimen in adolescent transplant recip-
Clinical features, long term follow up and outcome of a large cohort of patients ients. Pediatr Transplant 2005;9:381-90.
with chronic granulomatous disease: an Italian multicentre study. Clin Immunol 12. Seger RA, Gungor T, Belohradsky BH, Blanche S, Bordigoni P, Di Bartolomeo P,
2008;126:155-64. et al. Treatment of chronic granulomatous disease with myeloablative conditioning
3. van den Berg JM, van Koppen E,  Ahlin A, Belohradsky BH, Bernatowska E, Cor- and an unmodified hemopoietic allograft: a survey of the European experience,
beel L, et al. Chronic granulomatous disease: the European experience. PLoS One 1985-2000. Blood 2002;100:4344-50.
2009;4:e5234. 13. Martinez CA, Shah S, Shearer WT, Rosenblatt HM, Paul ME, Chinen J, et al. Ex-
4. Soncini E, Slatter MA, Jones LBKR, Hughes S, Hodges S, Flood TJ, et al. Unre- cellent survival after sibling or unrelated donor stem cell transplantation for
lated donor and HLA-identical sibling haematopoietic stem cell transplantation chronic granulomatous disease. J Allergy Clin Immunol 2012;129:176-83.