ORIGINAL

ARTICLES
Natural History of Perinatal and Infantile Hypophosphatasia:
A Retrospective Study
Michael P. Whyte, MD1,2, Edward Leung, MD3, William R. Wilcox, MD, PhD4, Johannes Liese, MD5, Jes
us Argente, MD, PhD6,
Gabriel

A. Martos-Moreno, MD, PhD6, Amy Reeves, MS1, Kenji P. Fujita, MD7, Scott Moseley, MS, MS8,
and Christine Hofmann, MD5, on behalf of the Study 011-10 Investigators*

Objective To report clinical characteristics and medical history data obtained retrospectively for a large cohort of
pediatric patients with perinatal and infantile hypophosphatasia.
Study design Medical records from academic medical centers known to diagnose and/or treat hypophosphata-
sia were reviewed. Patients born between 1970 and 2011 with hypophosphatasia and any of the following signs/
symptoms at age <6 months were eligible: vitamin B6–dependent seizures, respiratory compromise, or rachitic
chest deformity (NCT01419028). Patient demographics and characteristics, respiratory support requirements,
invasive ventilator–free survival, and further complications of hypophosphatasia were followed for up to the first
5 years of life.
Results Forty-eight patients represented 12 study sites in 7 countries; 13 patients were alive, and 35 were dead
(including 1 stillborn). Chest deformity, respiratory distress, respiratory failure (as conditioned by the eligibility
criteria), failure to thrive, and elevated calcium levels were present in >70% of patients between birth and age
5 years. Vitamin B6–dependent seizures and respiratory distress and failure were associated significantly
(P < .05) with the risk of early death. Serum alkaline phosphatase activity in all 41 patients tested (mean [SD]:
18.1 [15.4] U/L) was below the mean lower limit of normal of the reference ranges of the various laboratories
(88.2 U/L). Among the 45 patients with relevant data, 29 had received respiratory support, of whom 26 had died
at the time of data collection. The likelihood of invasive ventilator–free survival for this cohort decreased to 63%
at 3 months, 54% at 6 months, 31% at 12 months, and 25% at 5 years.
Conclusions Patients with perinatal or infantile hypophosphatasia and vitamin B6–dependent seizures, with or
without significant respiratory distress or chest deformities, have high morbidity and mortality in the first 5 years
of life. (J Pediatr 2019;209:116-24).
Trial registration ClinicalTrials.gov: NCT01419028.

H
ypophosphatasia is the rare inborn-error-of-metabolism characterized enzymatically by low activity of the tissue-
nonspecific isoenzyme of alkaline phosphatase (TNSALP).1-3 In hypophosphatasia, deficient activity of TNSALP on
cell surfaces leads to extracellular accumulation of its natural substrates: inorganic pyrophosphate (PPi), a potent in-
hibitor of hydroxyapatite crystal growth; pyridoxal 50 -phosphate (PLP), the major circulating form of vitamin B6; and phos-
phoethanolamine (PEA), a degradation product of cell-surface phosphatidylinositol–glycan anchors for various ectoproteins
like TNSALP.1,2,4 Because PPi inhibits hydroxyapatite crystal propagation, sufficiently high extracellular PPi levels in hypo-
phosphatasia can impair mineralization of the growing skeleton, causing rickets,
and can lead to osteomalacia in adults.1,2,5 TNSALP dephosphorylates PLP to
pyridoxal to cross the blood–brain barrier and enter neurons. Thus, profound
TNSALP deficiency in the most severely affected patients with hypophosphatasia 1
From the Center for Metabolic Bone Disease and
Molecular Research, Shriners Hospital for Children, St
also can compromise neurotransmitter synthesis and cause vitamin B6–depen- 2
Louis, MO; Division of Bone and Mineral Diseases,
dent seizures.1,6 Department of Internal Medicine, Washington University
School of Medicine at Barnes-Jewish Hospital, St Louis,
3
Largely owing to genetic heterogeneity, hypophosphatasia severity is broad- MO; Children’s Hospital Research Institute of Manitoba,
University of Manitoba, Winnipeg, Manitoba, Canada;
ranging, featuring a continuum from essentially no skeletal mineralization at birth 4
Department of Human Genetics, Emory University
School of Medicine, Atlanta, GA; 5University Children’s
Hospital, University of Wu €rzburg, Wu €rzburg, Germany;
6
~o Jesu
Hospital Infantil Universitario Nin
s, Universidad
Auto
noma de Madrid, CIBERobn, ISCIII, IMDEA Food
Institute, CEIUAM+CSIC, Madrid, Spain; and 7Clinical
ALP Alkaline phosphatase Research and 8Biostatistics, Alexion Pharmaceuticals,
IVFST Invasive ventilator–free survival time Inc, Boston, MA
PEA Phosphoethanolamine *List of additional members of the Study 011-10
PLP Pyridoxal 50 -phosphate Investigators is available at www.jpeds.com (Appendix 1).

PPi Inorganic pyrophosphate The conflicts of interest and prior presentation is avail-
able at www.jpeds.com.
TNSALP Tissue-nonspecific isoenzyme of alkaline phosphatase
ULN Upper limit of normal 0022-3476/$ - see front matter. ª 2019 Elsevier Inc. All rights reserved.
https://doi.org/10.1016/j.jpeds.2019.01.049

116
 Volume 209  June 2019

to arthropathies and/or dental complications without skeletal local laws of the appropriate regulatory authorities. The study
abnormalities in adult life.1,2 The clinical nosology of pediatric protocol and subsequent amendments were approved at each
hypophosphatasia, aiming to organize this continuum, now institution by the institutional review board, independent
includes (by increasing severity) odonto-hypophosphatasia, ethics committee, or research ethics board. Details regarding
mild childhood hypophosphatasia, severe childhood hypo- patient informed consent are provided in Appendix 2,
phosphatasia, infantile hypophosphatasia, and perinatal Methods (available at www.jpeds.com).
hypophosphatasia.1,7,8 Benign prenatal hypophosphatasia
also has been described; it is characterized by skeletal
Study Design and Patients
manifestations detected in utero or at birth that improve
Physicians with access to records of patients with hypo-
spontaneously ex utero, with outcomes ranging from
phosphatasia were identified and contacted via e-mail by
odonto-hypophosphatasia to infantile hypophosphatasia.9
the study sponsor. Examples of recruitment efforts included
Information concerning the prevalence and nature of hy-
outreach to physicians with experience in metabolic bone
pophosphatasia complications and survival rates with
diseases and genetic testing laboratories to obtain lists of
supportive care alone is incomplete for perinatal hypophos-
physicians who had referred DNA samples that were posi-
phatasia, which manifests in utero and is apparent at birth,
tive for ALPL mutation(s). Medical records from patients
and infantile hypophosphatasia, which presents before
(living or dead) were reviewed if the diagnosis of hypophos-
6 months of age.1,2 Death from these 2 life-threatening forms
phatasia had included at least 1 of the following: (1) ALPL
of hypophosphatasia seems to result most often from respira-
mutation(s); (2) serum alkaline phosphatase (ALP) activity
tory complications caused by poor mineralization of the
below the normal range and either plasma PLP or urinary
bones of the thorax.1,10,11 Perinatal hypophosphatasia is
PEA levels above the laboratory’s upper limit of normal
considered almost always fatal within a few days after birth,
(ULN); or (3) serum ALP activity below the normal range
and mortality in infantile hypophosphatasia has been esti-
and report of hypophosphatasia-related radiographic ab-
mated as approximately 50% during the first year of life.1,2
normalities. In addition, patients were required to have
However, no published comprehensive retrospective or pro-
signs and symptoms characteristic of hypophosphatasia
spective studies include a global compilation of the features
before 6 months of age and at least 1 of the following hypo-
of life-threatening perinatal and infantile hypophosphatasia.
phosphatasia characteristics associated with significant clin-
Herein we describe the natural history of patients with
ical compromise and poor outcomes: (1) respiratory
perinatal or infantile hypophosphatasia receiving supportive
complications that required respiratory support measures
care born between 1970 and 2011. They were selected for this
or medication(s) and/or that were associated with other
study because they displayed the severe manifestations of hy-
pulmonary problems (eg, pneumonia, respiratory tract
pophosphatasia (eg, respiratory compromise, seizures, chest
infection, respiratory failure); (2) vitamin B6–dependent
deformity) of a cohort of patients who began a clinical study
seizures (ie, responsive to supplemental pyridoxine); or
of enzyme replacement therapy in 2008 (asfotase alfa; Stren-
(3) rachitic chest deformity. These represent features of “se-
siq, Alexion Pharmaceuticals, Inc, Boston, Massachusetts).12
vere” pediatric hypophosphatasia and therefore would likely
A portion of the data from that study regarding overall sur-
exclude benign prenatal hypophosphatasia9 and less severe
vival and ventilatory support requirements was published
instances of infantile hypophosphatasia. Thus, the medical
in 2016,13 and the findings after an average of 7 years of treat-
records evaluated in our study represented patients with
ment were published in 2019.14 Now we report unpublished
perinatal or infantile hypophosphatasia likely to show
data concerning clinical characteristics including signs,
high morbidity and mortality. Exclusion criteria included
symptoms, and complications; medical history (including
any other clinically significant disease or other treatment
respiratory support, medications, and hospitalizations);
with asfotase alfa at any time before data abstraction.
and routine clinical laboratory values and ALPL mutations
in this natural history cohort of patients with perinatal or in-
fantile hypophosphatasia. Data Collection
Abstraction of the medical record onto case report forms by
trained personnel occurred at a single time point between
Methods September 2012 and April 2013. Data from up to the patient’s
fifth birthday were collected. This included demographics;
This multinational, noninterventional, retrospective chart diagnostic history, including age at hypophosphatasia diag-
review of the natural history of perinatal and infantile nosis; clinical laboratory results (serum ALP, urine PEA,
hypophosphatasia (ClinicalTrials.gov: NCT01419028) was and plasma PLP) obtained closest to the date of diagnosis;
performed at academic medical centers that had diagnosed ALPL gene mutation analysis results; and clinical course
and/or managed severe pediatric hypophosphatasia. The study including comorbidities, mention of developmental delays,
was conducted in accordance with guidelines set by the World hospitalizations, medications, therapies, and procedures.
Medical Association Declaration of Helsinki and the Interna- Additional data collected included date, cause, and age of
tional Conference on Harmonisation Guideline for Good death. Details regarding data collection are provided in
Clinical Practice and in compliance with national, state, and Appendix 2, Methods.
117
THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume 209

Outcome Measures [n = 1], and Taiwan [n = 1]). The first patient was born in 1970
Information regarding the primary outcome measure of and the last in 2011. Among the 48 patients, 13 (27%) were
overall survival and the tertiary outcome measure of ventila- diagnosed with hypophosphatasia before 1990, 14 (29%) be-
tory support status was published in 2016.13 Secondary and tween 1990 and 1999, and 21 (44%) during or after 2000. Thir-
tertiary assessments, reported herein, include the proportion teen (27%) were alive (median [min, max] age: 7.7 [2, 20]
of patients requiring invasive ventilation (mechanical via years), and 35 (73%) had died (Kaplan–Meier median [95%
intubation or tracheostomy) or noninvasive respiratory sup- CI] age at death: 0.7 [0.4, 1.2] years) (Figure 1; available at
port (continuous or bilevel positive airway pressure or sup- www.jpeds.com). Patients who had died comprised 100%
plemental oxygen) stratified by mortality, and by invasive (13/13) of those diagnosed before 1990, 71.4% (10/14) of
ventilator–free survival time (IVFST). IVFST is henceforth those diagnosed between 1990 and 1999, and 57.1% (12/21)
defined as time to first invasive ventilation or death. Respira- of those diagnosed in 2000 or later, suggesting some
tory “distress” and “failure” were defined as transient vs improvement in survival with better availability and/or
chronic respiratory difficulty, respectively. Additional assess- advances in supportive care. Their demographic and disease
ments included radiographic findings reported at the time of characteristics are summarized in Table I. As required for
hypophosphatasia diagnosis, medication history, hospitaliza- study inclusion, all presented with at least 1 of the specified
tions, prevalent and clinically important signs, symptoms, key characteristics of hypophosphatasia before 6 months of
and complications of hypophosphatasia (stratified by mor- age (median [min, max] age: 1 [0, 179] days). Median (min,
tality) and association with death. max) age of hypophosphatasia diagnosis was 0.2 (0, 3) years.
Consistent with the range of severity of their
Statistical Analyses hypophosphatasia, median (min, max) age of
All available data were included in the statistical analyses. hypophosphatasia diagnosis among the 13 patients who were
Because data collection was retrospective, not all patients alive at the time of chart review was 0.4 (0, 3) years, and
had complete information. No imputation was performed among the 34 patients who were dead and had this
for missing data except when only a partial birthdate was pro- information recorded was 0.04 (0, 0.8) years. Among the
vided (ie, if the day was missing, it was set to 15, and if the total patient group, 14 (29%) were reported to have prenatal
month was missing, it was set to June). Data were censored, ultrasound scan manifestations of hypophosphatasia, 15
if applicable, at date of data collection for living patients and (31%) were not reported to have evidence of
at date of last contact for patients whose status was unknown hypophosphatasia on prenatal ultrasound scan, and 19
at data collection. All statistical analyses were conducted using (40%) had no available ultrasound scan data. In the 29
SAS release 9.2 (SAS Institute, Inc, Cary, North Carolina). patients whose charts had sufficient information (23/35 dead
IVFST estimates were quantified using the Kaplan–Meier and 6/13 alive), signs of hypophosphatasia reported on
method. Survival and invasive respiratory support data (ie, prenatal ultrasound scan were similar between those who
mechanical via intubation or tracheostomy) were considered would die and those who would live (12/23 vs 2/6,
simultaneously and used to calculate the time to first invasive respectively).
ventilation or death. CIs were generated using the Greenwood One patient was stillborn, and 13 patients died within the
formula for variance, and CIs for proportions were generated first month of life; 9 on the first day, and 4 on days 5, 11, 27,
using the normal approximation to the binomial distribution. and 31. In total, 10 of these 14 patients received respiratory
All CIs were 2 sided, with a significance level set at alpha = .05. support before death. Of the 14, 7 (50%) were diagnosed
Any P values, if generated, were nominal. The association of before 1990 (and only 1 was diagnosed before 1980), 3
the relative risk of death with select hypophosphatasia compli- (21%) were diagnosed between 1990 and 2000, and 4
cations was also individually and separately examined post hoc (29%) were diagnosed after 2000. The infant who was still-
without considering mutual correlation and confounding ef- born could not be assessed for the key inclusion characteris-
fects. Similarly, CIs for relative risks were obtained by tics but was enrolled based on his profoundly low serum ALP
assuming the normal approximation to the binomial distribu- activity, radiographic findings, and respiratory compromise
tion. Percentages of patients were calculated and are provided that was attributed to pulmonary hypoplasia (cause of
below only when the denominator was 48 (representing the death). Among the 35 patients who would die, the most prev-
entire enrolled group). All other data are presented as n/N. alent complication, as anticipated, was respiratory compro-
Further details regarding the statistical methods are provided mise (33/35).
in Appendix 2, Methods. At hypophosphatasia diagnosis, serum ALP was available
for 85% (41/48) of patients (mean  SD [min, max]:
18.1  15.4 [0, 55.0] U/L). ALP reference ranges were pro-
Results vided for 26 of these patients; all 26 patient values were below
the mean lower limit of normal of the reference ranges of the
Patient Demographics and Characteristics various laboratories (88.2 U/L). Among the 13 patients who
Of the 65 patients screened, 48 were eligible and enrolled from were alive at the time of chart review, 12 (92%) had recorded
12 sites (6 in the US [n = 26] and 1 each in Australia [n = 2], serum ALP activity (mean  SD [min, max]: 34.0  14.5
Canada [n = 11], Germany [n = 6], Spain [n = 1], Switzerland [2.0, 55.0] U/L). Among the 35 patients who were dead at
118 Whyte et al
June 2019 ORIGINAL ARTICLES

Table I. Patient demographics and diagnostic history

Characteristics Overall Alive Dead
Demographics
Gestational age at birth n = 36 n = 12 n = 24
Median (min, max), wk 39 (30, 41) 38.5 (35, 41) 39.0 (30, 41)
Age at data collection for living patients n = 13 n = 13
Median (min, max), y 7.7 (2, 20) 7.7 (2, 20) —
Sex, n (%) n = 48 n = 13 n = 35
Male 26 (54) 6 (46) 20 (57)
Female 22 (46) 7 (54) 15 (43)
Region of birth, n (%) n = 48 n = 13 n = 35
North America 33 (69) 5 (39) 28 (80)
Europe 8 (17) 7 (54) 1 (3)
Australia 2 (4) 0 2 (6)
Middle East 1 (2) 1 (8) 0
Asia 1 (2) 0 1 (3)
Unknown* 3 (6) 0 3 (9)
Race, n (%) n = 48 n = 13 n = 35
White 40 (83) 11 (85) 29 (83)
Black or African American 3 (6) 0 3 (9)
Asian 2 (4) 1 (8) 1 (3)
American Indian or Alaska Native 1 (2) 1 (8) 0
Unknown 2 (4) 0 2 (6)
Ethnicity, n (%) n = 48 n = 13 n = 35
Not Hispanic or Latino 41 (85) 12 (92) 29 (83)
Hispanic or Latino 1 (2) 0 1 (3)
Unknown 6 (13) 1 (8) 5 (14)
Diagnostic history
Signs of in utero skeletal disease on ultrasound scan, n (%) n = 48 n = 13 n = 35
Yes 14 (29) 2 (15) 12 (34)
No 15 (31) 4 (31) 11 (31)
Unknown 19 (40) 7 (54) 12 (34)
Age at onset of signs/symptoms, mo n = 47 n = 13 n = 34
Mean (SD) 1.1 (1.7) 2.3 (2.1) 0.7 (1.2)
Median (min, max) 0.03 (0, 5.9) 2.4 (0, 5.9) 0.03 (0, 5.2)
Age at hypophosphatasia diagnosis, mo n = 47 n = 13 n = 34
Mean (SD) 5.2 (9.3) 13.7 (14.3) 1.9 (2.5)
Median (min, max) 2.0 (0, 40.9) 4.4 (0.2, 40.9) 0.5 (0, 10.1)
Criteria that formed the basis for diagnosis of hypophosphatasia,† n (%) n = 48 n = 13 n = 35
Clinical laboratory results
Serum ALP 41 (85) 11 (85) 30 (86)
Plasma PLP 6 (13) 7 (54) 6 (17)
Urinary PEA 12 (25) 7 (54) 5 (14)
Other: serum calcium 3 (6) 0 3 (9)
Radiographic findings 38 (79) 11 (85) 27 (77)
Signs and symptoms 25 (52) 11 (85) 14 (40)
ALPL gene mutation 17 (35) 9 (69) 8 (23)
Laboratory test results
Serum ALP activity, U/L n = 41 n = 12 n = 29
Mean (SD) 18.1 (15.4) 34.0 (14.5) 11.6 (10.3)
Median (min, max) 15.0 (0, 55.0) 34.0 (2.0, 55.0) 10.0 (0, 42.0)
Plasma PLP, ng/mL n=6 n=5 n=1
Mean (SD) 623 (1154) 154 (93) 2972 (NA)
Median (min, max) 150 (43, 2972) 150 (43, 300) 2972 (2972, 2972)
Serum calcium, mmol/L n = 32 n = 11 n = 21
Mean (SD) 2.8 (0.4) 2.7 (0.3) 2.8 (0.4)
Median (min, max) 2.6 (2.3, 4.0) 2.6 (2.4, 3.5) 2.7 (2.3, 4.0)

NA, not available.

*Recorded as unknown.
†Diagnosis was based on a combination of laboratory findings, a combination of laboratory and radiographic findings, or genetic analysis (see inclusion criteria). ALPL gene mutation analysis was not
required for diagnosis. Reference ranges for each laboratory varied.

the time of chart review, 29 (83%) had substantially lower re- (84.8  13.5 [73.9, 99.8] nmol/L; n = 3) was within the
corded serum ALP activity (mean  SD [min, max]: normal reference ranges of the various laboratories (mean
11.6  10.3 [0, 42.0] U/L), suggesting correlation with hypo- lower limit of normal: 53.2 nmol/L; mean ULN:
phosphatasia severity. Plasma PLP level was elevated above 172.0 nmol/L). Hypercalcemia had been documented in 14
the mean ULN of the various laboratories (31.3 ng/mL) in of 21 patients, and hyperphosphatemia had been reported
all of the few patients with such data (n = 6), with a in 10 of 21 patients with available data. Mean  SD (min,
mean  SD (min, max) of 623  1154 (43, 2972) ng/mL. max) serum calcium concentration (2.8  0.4 [2.3, 4.0]
Mean  SD (min, max) serum 25-hydroxyvitamin D mmol/L [11.2 mg/dL]; n = 32) was above the mean ULN
Natural History of Perinatal and Infantile Hypophosphatasia: A Retrospective Study 119
THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume 209

of the various laboratories (2.6 mmol/L; 10.4 mg/dL), with a as well as the study inclusion criterion of rachitic chest defor-
mean 1.1-fold elevation. Mean  SD (min, max) serum mity (32/37) (Table II; available at www.jpeds.com).
phosphate concentration (2.0  0.4 [1.1, 3.1] mmol/L Radiographic absence of some or all bones was described in
[6.19 mg/dL]; n = 29) matched the mean ULN of the various 7 of 30 patients. Of these 7 patients, 5 had signs and
laboratories (2.0 mmol/L; 6.19 mg/dL). Mean  SD (min, symptoms of hypophosphatasia at birth or reported within
max) serum creatinine concentration (136.6  325.5 [0.1, 1 day of birth.
1246.4] mmol/L [1.55 mg/dL]; n = 23) was elevated at the
time of hypophosphatasia diagnosis compared with the Respiratory Support Requirements and
mean ULN of the various laboratories (68.3 mmol/L; Invasive Ventilation-Free Survival
0.77 mg/dL) because of 2 especially high values. The values Forty-five (94%) of the 48 patients had documented respira-
were proportionately distributed as “low” (5/16), “normal” tory status. Of the 45, 29 had received either “noninvasive” or
(7/16), and “high” (4/16) within the reference ranges. “invasive” respiratory support. At the time we collected their
ALPL mutation analysis results obtained using early data, 26 of the 29 patients had died (Figure 2). Respiratory
sequencing techniques at a variety of commercial and support was necessary within the first 6 days of life for
research laboratories between March 1995 and November most of these patients (17/29). Invasive ventilation was
2011 were reported for 21 patients; mutations were detected necessary for 19 of the 29 patients, supplemental oxygen
in 19. Most patients were compound heterozygotes (12/19), was necessary in 9, and continuous positive airway pressure
and the remainder carried homozygous mutations (5/19) was necessary in 1. Death occurred for 18 of the 19 who
or seemed to carry heterozygous mutations (2/19). For the received invasive ventilation (Figure 2).
2 patients without an identified ALPL mutation, copy num- Among the 48 patients, the median IVFST was 7.8 months
ber variation or reverse transcriptase polymerase chain reac- (95% CI 2.6-9.9; Figure 3). The Kaplan–Meier estimate of
tion had not been performed. Because these 2 patients had probability of being alive and not invasively ventilated
laboratory and radiographic findings characteristic of severe during the first year of life was 63% at 3 months, 54% at
hypophosphatasia, they were included in the study. 6 months, and 31% at 12 months. Only 25% of patients
At the time of hypophosphatasia diagnosis, radiographic were alive at 5 years of age (Figure 3). Rates of respiratory
findings had been reported in the medical record for 77% support–free survival (data not shown) were comparable
of patients (37/48). Multiple skeletal abnormalities were with rates of IVFST, as respiratory support was invasive in
common and often included mention of osteopenia (33/37) most patients. Additional information concerning

A B
Dead
30
Alive
26/32
25

20 18/26
Patients (n)

15

10
7/26

5 3/13
2/3
1/3 1/26
0/3 0/26 0/3
0
Requiring Invasive Supplemental O2 CPAP BiPAP
respiratory ventilation (9/29) (1/29) (0/29)
support (19/29)
(29/45)
Most Intensive Support Administered

Figure 2. Respiratory support administration. A, Distribution of living and dead patients who had required respiratory support.
B, Greatest required support stratified by type and number of patients alive vs dead at data collection. Values along the x-axis
represent the total number of patients (dead and alive combined) who received the specific type of support compared with the
total number of patients for whom data were available. BiPAP, bilevel positive airway pressure; CPAP, continuous positive airway
pressure.
120 Whyte et al
June 2019 ORIGINAL ARTICLES

100
Year Survival Rate (95% CI)
0.25 0.625 (0.473, 0.745)
90
0.5 0.542 (0.392, 0.670)
1 0.313 (0.189, 0.444)
80
1.5 0.250 (0.139, 0.378)
Surviving Patients (%) 70
2 0.250 (0.139, 0.378)
3 0.250 (0.139, 0.378)
60 4 0.250 (0.139, 0.378)
5 0.250 (0.139, 0.378)
50

40

30

20

10 Patients at Risk
48 26 15 12 11 11 11 11 10 9 8
0
0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Time From Birth to Invasive Ventilation or Death (Years)

Figure 3. Kaplan–Meier plot of time from birth during which patients were not mechanically ventilated by intubation or tra-
cheostomy; 36 patients had an event of invasive ventilation or death, and 12 were censored. Inset: cumulative probability of
invasive ventilation-free survival.

individual patient ventilation status is in Table III (available seizures were associated significantly (P < .05) with the risk
at www.jpeds.com). of early death independent of other confounding risk
factors (Table IV). In addition, 24 of 27 patients with
Common Signs, Symptoms, and Complications appropriate data for age reportedly had developmental
Associated with Hypophosphatasia delays, including gross motor skills (23/24), walking (10/
Table IV summarizes the incidence/prevalence of the 24), and fine motor skills (2/24). Cognitive delays were
most common complications of perinatal and infantile documented for 3 of 19 of patients with available data.
hypophosphatasia noted in this study population, as well
as their association with mortality. The eligibility criteria Medication Histories and Hospitalizations
conditioned the high prevalence of chest deformity (39/43), Treatment histories were positive for medications or other
respiratory distress (30/39), and respiratory failure (26/36). therapies for hypophosphatasia signs and symptoms for
Respiratory distress and failure and vitamin B6–dependent most patients (65% [31/48]). Medications included adrenergic

Table IV. Hypophosphatasia presentation stratified by age interval and association with death
Observed incidence, % (n/N)* Death rate, % (n/N)
By age at occurrence
Relative risk (95% CI) of
<6 mo 6 mo–3 y 3–5 y death in patients with vs without With Without
Signs, symptoms, or complications Overall (n = 48) (n = 27) (n = 13) the manifestation manifestation† manifestation‡
Respiratory failure 72 (26/36) 47 (18/38) 42 (8/19) 0 (0/11) 10 (1.56-64.20)§ 100 (26/26) 10 (1/10)
Respiratory distress 77 (30/39) 64 (25/39) 50 (10/20) 18 (2/11) 2.6 (1.02-6.62)§ 87 (26/30) 33 (3/9)
Decreased oxygen saturation 67 (22/33) 53 (16/30) 53 (10/19) 0 (0/10) 1.8 (0.92-3.54) 82 (18/22) 45 (5/11)
Tachypnea 63 (20/32) 53 (17/32) 47 (8/17) 18 (2/11) 1.7 (0.94-3.08) 85 (17/20) 50 (6/12)
Vitamin B6–dependent seizures 26 (10/38){ 17 (6/36) 14 (3/21) 0 (0/11) 1.6 (1.18-2.06)§ 100 (10/10) 64 (18/28)
Elevated serum or urine calcium 72 (28/39) 68 (26/38) 74 (14/19) 18 (2/11) 1.2 (0.72-1.94) 75 (21/28) 64 (7/11)
Pneumonia 42 (15/36) 18 (6/34) 50 (10/20) 18 (2/11) 1.1 (0.72-1.69) 73 (11/15) 67 (14/21)
Nephrocalcinosis 52 (16/31) 36 (11/31) 63 (10/16) 40 (4/10) 1.0 (0.63-1.68) 69 (11/16) 67 (10/15)
Chest deformity 91 (39/43) 80 (28/35) 96 (22/23) 83 (10/12) 0.7 (0.54-0.84)§ 67 (26/39) 100 (4/4)
Respiratory tract infection 64 (21/33) 42 (13/31) 79 (15/19) 46 (5/11) 0.7 (0.49-1.13) 62 (13/21) 83 (10/12)
Craniosynostosis 61 (19/31) 50 (15/30) 74 (14/19) 64 (7/11) 0.7 (0.41-1.20) 53 (10/19) 75 (9/12)
Failure to thrive 76 (28/37) 67 (24/36) 91 (19/21) 64 (7/11) 0.6 (0.45-0.82)§ 61 (17/28) 100 (9/9)
Early tooth loss 44 (10/23) 0 (0/24) 71 (10/14) 78 (7/9) 0 (–, –) 0 (0/10) 92 (12/13)
*Calculated as the number of patients with disease history (n) divided by the number of patients with available data (N)  100. Patients for whom status was unknown or missing were not included.
†Number of patients who were dead (n) divided by the total number of patients (both alive and dead) with the complication (N).
‡Number of patients who were dead (n) divided by the total number of patients (both alive and dead) without the complication (N).
§Relative risk significantly different from 1 (P < .05).
{Seven patients had documented vitamin B6–dependent seizures; an additional 3 patients had a history of vitamin B6–dependent seizures based upon hospitalization and medication records (age
was not available for these 3 patients).

Natural History of Perinatal and Infantile Hypophosphatasia: A Retrospective Study 121

THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume 209

inhalants, nonsteroidal anti-inflammatory drugs, systemic sample collection (eg, sites, time periods, assays). Delays in
corticosteroids, loop diuretics, anticonvulsants, and vitamin diagnosis for some patients that occurred despite the pres-
B6. Two patients, both alive at the time of chart review, had ence of the most severe characteristics of hypophosphatasia
been our research subjects and received a bone marrow trans- underscore the need for better appreciation of the signs,
plantation during infancy,15,16 followed soon after by stromal symptoms, and biochemical and radiographic hallmarks of
cell transplantation in one.15 Thirty-nine (81%) of the 48 pa- hypophosphatasia, especially now that an effective treatment
tients had hospitalizations during their first 5 years of life, 22 is available for patients with hypophosphatasia, including for
from respiratory compromise complications and 7 from severely affected newborns and infants.14,18
seizure-related complications. Multiple hospitalizations, The 13 surviving patients in our study nevertheless suf-
some prolonged, had occurred for most living and dead pa- fered significant morbidity during their first 5 years of life.
tients. Approximately one-half of the patients who had died Those with available data had failure to thrive and chest
(18/35) and one-third of the living patients (4/13) had hospi- deformity, but reportedly none had vitamin B6–dependent
talization(s) for respiratory compromise. (Additional infor- seizures, which in hypophosphatasia is considered a sign of
mation concerning the hospitalizations is shown in Table V impending death.6,19 At least one-half of the 13 surviving
[available at www.jpeds.com].) patients went on to manifest early deciduous tooth loss, res-
piratory tract infection, respiratory compromise, craniosy-
nostosis, hypercalcemia, and/or nephrocalcinosis.
Discussion Respiratory complications that troubled these survivors
included decreased oxygen saturation, distress, pneumonia,
This chart review collected natural history data from pa- and tachypnea, although none progressed to chronic respi-
tients with perinatal or infantile hypophosphatasia (ie, ratory failure during the time frame of data collection and
hypophosphatasia-related findings apparent at birth or only 1 transiently was given ventilation. All patients with
manifesting in the first 6 months of life, respectively). vitamin B6–dependent seizures and respiratory failure
Data were collected up to 5 years of life. In earlier reports, died, as did most requiring any respiratory support (26/
we described the natural history of hypophosphatasia 32). Finally, within the time frame of this study, all but 1
selectively in 15 Manitoban patients with perinatal hypo- surviving patient reportedly had developmental (motor)
phosphatasia10 and documented the natural history of hy- delays and 2 reportedly had cognitive delays. The motor de-
pophosphatasia spanning an average of 6.5 years in lays were readily attributed to their skeletal abnormalities
children with odonto-, mild childhood, severe childhood, and muscle weakness.12,14,20
and infantile hypophosphatasia.17 Our current study is a We focused in 201613 on the overall mortality rate of 73%
multinational systematic assessment of the natural history by 5 years of age in this current study population, with the
of specifically perinatal and infantile hypophosphatasia greatest relative risks for death noted among patients with
caused by a wide variety of ALPL mutations. Patients the inclusion criteria of respiratory failure, respiratory
selected for this study were required to have a history of distress, and vitamin B6–dependent seizures. Most of the
rachitic chest deformity, respiratory compromise, or 48 patients (58%) died within 12 months of birth, and me-
vitamin B6–dependent seizures to represent the extreme dian time to death was 8.9 months.13 Thus, our current anal-
end of the hypophosphatasia disease spectrum and, there- ysis adds to what is known about mortality among these
fore, to reflect the perinatal and infantile hypophosphata- patients. Although mortality was greatest among patients
sia patient population involved in a clinical study of diagnosed before 1990 (100%), the rate remained high for
asfotase alfa treatment.12,14 Patients with benign prenatal those diagnosed between 1990 and 1999 (71%) and those
hypophosphatasia9 were excluded, as shown by the severe diagnosed in 2000 and later (57%), suggesting that mortality
clinical course of the study subjects. rates for patients with perinatal and infantile hypophospha-
Our data from 48 patients manifesting complications of tasia with the aforementioned inclusion criteria remain
hypophosphatasia before 6 months of life, and frequently high even with modern neonatal and pediatric intensive
within the first month of life, showed that skeletal disease care units. The proportion of patients surviving without
in utero was suspected based on prenatal ultrasound scan invasive ventilation decreased during the first year, from
in only one-third. Perhaps this has changed with more 63% at 3 months to 31% at 12 months. The median time
advanced and prevalent use of this technology, or it indicates from birth to invasive ventilation or death was 7.8 months.
that postnatal presentation of the disorder is common. Thus, these observations confirm published reports of high
Although the median age at diagnosis of hypophosphatasia mortality in general among patients with perinatal and infan-
was 0.2 years, diagnosis also was made as late as 3 years of tile hypophosphatasia receiving supportive care.1,2
age. Patients who had died at the time of chart review had As first published in 2012, asfotase alfa treatment in the
been diagnosed earlier than those who were alive at chart re- similar population of patients improved survival,12 with
view and had numerically lower mean serum ALP activity. 84% (31/37) of patients alive at 5 years of age13 vs 27%
However, these results should be interpreted with caution, (13/48) in the current study. This emphasizes the importance
as statistical comparisons of ALP activity between alive and of prompt diagnosis of perinatal and infantile hypophospha-
dead patients seemed problematic because of differences in tasia so that supportive measures and effective treatment can
122 Whyte et al
June 2019 ORIGINAL ARTICLES

begin. A rapidly worsening clinical course often occurs in We acknowledge the International Skeletal Dysplasia Registry (http://
perinatal or infantile hypophosphatasia, and therefore either ortho.ucla.edu/isdr), from which a number of cases were identified.
form of hypophosphatasia requires close monitoring. Chest
Submitted for publication Aug 30, 2018; last revision received Jan 16, 2019;
deformities and failure to thrive are common acquired fea-
accepted Jan 29, 2019.
tures of infantile hypophosphatasia12 but, in the current
Reprint requests: Michael P. Whyte, MD, Shriners Hospital for Children, 4400
study, were associated with a somewhat-lower risk of death Clayton Ave, St Louis, MO 63110. E-mail: mwhyte@shrinenet.org
compared with respiratory failure or compromise or, as pub-
lished in 2007, with vitamin B6–dependent seizures.6 These Data Statement
latter 2 characteristics of severe hypophosphatasia were asso-
ciated with greater mortality in this study, likely because pul- Data sharing statement available at www.jpeds.com.
monary compromise is understandably life-threatening and
vitamin B6–dependent seizures likely signal the most pro-
found deficiency of TNSALP activity. In fact, a similar References
finding was recently reported among Japanese patients with 1. Whyte MP. Hypophosphatasia—aetiology, nosology, pathogen-
perinatal hypophosphatasia, where failure to thrive did not esis, diagnosis and treatment. Nat Rev Endocrinol 2016;12:
correlate with a poorer prognosis.20 However, a rachitic chest 233-46.
and failure to thrive are typically diagnosed in infants who 2. Whyte MP. Hypophosphatasia and how alkaline phosphatase promotes
have survived at least several weeks, permitting these compli- mineralization. In: Thakker RV, Whyte MP, Eisman J, Igarashi T, eds.
Genetics of bone biology and skeletal disease. 2nd ed. San Diego (CA):
cations to manifest.10,12,21 Possibly, chest deformity or failure Elsevier (Academic Press); 2018. p. 481-504.
to thrive affected some patients in our study but went unre- 3. Rockman-Greenberg C. Hypophosphatasia. Pediatr Endocrinol Rev
ported. The relatively small number of patients in the multi- 2013;10(suppl 2):380-8.
national study reported herein may have been insufficient 4. Whyte MP, Mahuren JD, Fedde KN, Cole FS, McCabe ER, Coburn SP.
to identify further signs and symptoms that influenced Perinatal hypophosphatasia: tissue levels of vitamin B6 are unremarkable
despite markedly increased circulating concentrations of pyridoxal-50 -
mortality. phosphate. Evidence for an ectoenzyme role for tissue-nonspecific alka-
Our study was retrospective and often limited by the avail- line phosphatase. J Clin Invest 1988;81:1234-9.
able data in case files. Data concerning motor and cognitive 5. Millan JL, Whyte MP. Alkaline phosphatase and hypophosphatasia.
delays were collected from chart review and had not been as- Calcif Tissue Int 2016;98:398-416.
sessed with standardized tools. Furthermore, the eligibility 6. Baumgartner-Sigl S, Haberlandt E, Mumm S, Scholl-B€ urgi S, Sergi C,
Ryan L, et al. Pyridoxine-responsive seizures as the first symptom of in-
criteria were such that patients would be excluded if their fantile hypophosphatasia caused by two novel missense mutations
signs and symptoms represented relatively mild infantile hy- (c.677T>C, p.M226T; c.1112C>T, p.T371I) of the tissue-nonspecific
pophosphatasia, which therefore likely increased the overall alkaline phosphatase gene. Bone 2007;40:1655-61.
relative risks. In addition, the study included data reflecting 7. Whyte MP, Zhang F, Wenkert D, McAlister WH, Mack KE,
a timeframe of perhaps 41 years (the first patient was born Benigno MC, et al. Hypophosphatasia: validation and expansion of the
clinical nosology for children from 25 years experience with 173 pediat-
in 1970, and the last in 2011) and enrolled 48 patients. This ric patients. Bone 2015;75:229-39.
small number of study subjects representing many years of 8. Whyte MP, Coburn SP, Ryan LM, Ericson KL, Zhang F. Hypophospha-
follow-up may not have provided sufficient statistical power tasia: biochemical hallmarks validate the expanded pediatric clinical
to identify correlations between outcomes and additional nosology. Bone 2018;110:96-106.
insidious features of hypophosphatasia. Furthermore, the 9. Wenkert D, McAlister WH, Coburn SP, Zerega JA, Ryan LM,
Ericson KL, et al. Hypophosphatasia: nonlethal disease despite skeletal
broad time span for data acquisition did not fully reflect presentation in utero (17 new cases and literature review). J Bone Miner
recent advances in pediatric intensive care, including respira- Res 2011;26:2389-98.
tory support and ventilation. Lastly, the associations between 10. Leung EC, Mhanni AA, Reed M, Whyte MP, Landy H, Greenberg CR.
risk factors and mortality were analyzed individually without Outcome of perinatal hypophosphatasia in Manitoba Mennonites: a
considering mutual correlation and potentially confounding retrospective cohort analysis. JIMD Rep 2013;11:73-8.
11. Rodriguez E, Bober MB, Davey L, Zamora A, Li Puma AB, Chidekel A,
aspects among risk factors. Importantly, the most common et al. Respiratory mechanics in an infant with perinatal lethal hypophos-
complications of hypophosphatasia are subject to survivor- phatasia treated with human recombinant enzyme replacement therapy.
ship bias, such as tooth loss and, therefore, might take longer Pediatr Pulmonol 2012;47:917-22.
to develop. 12. Whyte MP, Greenberg CR, Salman NJ, Bober MB, McAlister WH,
In conclusion, this retrospective natural history study Wenkert D, et al. Enzyme-replacement therapy in life-threatening hypo-
phosphatasia [with Supplementary Appendix]. N Engl J Med 2012;1366:
confirmed a high rate of morbidity and mortality among 904-13.
patients with perinatal and infantile hypophosphatasia 13. Whyte MP, Rockman-Greenberg C, Ozono K, Riese R, Moseley S,
within the first 5 years of life. Respiratory failure and vitamin Melian A, et al. Asfotase alfa treatment improves survival for perinatal
B6-dependent seizures are comorbidities highly predictive of and infantile hypophosphatasia. J Clin Endocrinol Metab 2016;101:334-
early death. Our observations are of particular significance 42.
14. Whyte MP, Simmons JH, Moseley S, Fujita KP, Bishop N, Salman NJ,
because with the availability of an effective enzyme et al. Asfotase alfa for infants and young children with hypophosphata-
replacement therapy (asfotase alfa)14,18 future natural history sia: 7 year outcomes of a single-arm, open-label, phase 2 extension trial.
studies in this population of young children are unlikely. n Lancet Diabetes Endocrinol 2019;7:93-105.

Natural History of Perinatal and Infantile Hypophosphatasia: A Retrospective Study 123

THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume 209

15. Whyte MP, Kurtzberg J, McAlister WH, Mumm S, Podgornik MN, 19. Hofmann C, Liese J, Schwarz T, Kunzmann S, Wirbelauer J, Nowak J,
Coburn SP, et al. Marrow cell transplantation for infantile hypophos- et al. Compound heterozygosity of two functional null mutations in
phatasia. J Bone Miner Res 2003;18:624-36. the ALPL gene associated with deleterious neurological outcome in an
16. Cahill RA, Wenkert D, Perlman SA, Steele A, Coburn SP, McAlister WH, et al. infant with hypophosphatasia. Bone 2013;55:150-7.
Infantile hypophosphatasia: transplantation therapy trial using bone frag- 20. Taketani T, Onigata K, Kobayashi H, Mushimoto Y, Fukuda S,
ments and cultured osteoblasts. J Clin Endocrinol Metab 2007;92:2923-30. Yamaguchi S. Clinical and genetic aspects of hypophosphatasia in Japa-
17. Whyte MP, Wenkert D, Zhang F. Hypophosphatasia: natural history nese patients. Arch Dis Child 2014;99:211-5.
study of 101 affected children investigated at one research center. Bone 21. Hofmann C, Girschick HJ, Mentrup B, Graser S, Seefried L, Liese J, et al.
2016;93:125-38. Clinical aspects of hypophosphatasia: an update. Clin Rev Bone Miner
18. Strensiq [package insert]. Boston (MA): Alexion Pharmaceuticals, Inc; 2018. Metab 2013;11:60-70.

50 Years Ago in THE JOURNAL OF PEDIATRICS

Funduscopic Photography and Fluorescein Angioretinography in
Evaluation of Children with Neurologic Handicaps
Baird HW, Pileggi AJ, Harley RD. J Pediatr 1969;6:937-45.

B aird et al obtained funduscopic photographs with fluorescein angioretinography in 80 of 300 neurologically

handicapped children referred for ophthalmoscopic examination. Macular degenerative disease was present in
20. The authors further described 6 cases with different diagnoses and concluded that fundoscopic photography is a use-
ful tool for retinal examination of handicapped children. In the following years, however; a restricted field of view, reflec-
tion artefacts, universal need for general anesthesia, and specific positioning, cost, and availability were recognized as
major limitations to regular and widespread use of fundus photography and fluorescein angiography in children.
Fifty years later, the world of retinal imaging has surpassed most barriers. With the use of wide field and ultra-
widefield (UWF) imaging systems, it is possible to obtain ora-to-ora view of the retina.1 UWF has allowed to expand
the field of view of imaging from 30 to 60 to up to 200 in a single image, capturing more of the periphery at retina.2
Most pediatric retinal diseases manifest in peripheral retina and UWF angiography has proved to be extremely useful
to diagnose the proliferative and exudative changes. Although most cases still require general anesthesia or light
sedation, UWF fundus photography and angiography can be done as an outpatient procedure without the need for
anesthesia, using “flying-baby” technique to position the child primarily for retinopathy of prematurity screening.2
Confocal scanning laser ophthalmoscopy, allows for elimination of artefact reflections. Portable wide-angle cameras
are now available for fundus photography, which allow direct visualization and capture in pediatric patients unable to
position themselves.3 These, and the advent of smart phone fundus photography, have opened up new gateways for
telemedicine consulting and screening programs.3
Fundus imaging now plays an important role in retinopathy of prematurity, familial exudative vitreoretinopathy,
Coats disease, Incontentia pigmenti, X-linked retinoschisis, Stargardt disease, Best disease, toxoplasmosis chorioreti-
nitis, juvenile sarcoidosis, choroidal melanoma, juvenile idiopathic arthritis, pars planitis, and traumatic retinal
detachment.2 Recent imaging tools have not only allowed improved diagnosis, screening, documentation,
monitoring, and image-guided treatment of these diseases. They have also widened the scope of research.

Payal Gupta, DNB

Department of Pediatric Ophthalmology
Sadguru Netra Chikitsalaya
Chitrakoot, Madhya Pradesh, India

Piyush Gupta, MD, FAMS

Department of Pediatrics
University College of Medical Sciences
Delhi, India
References

1. Patel CK, Buckle M. Ultra-widefield imaging for pediatric retinal disease. Asia Pac J Ophthalmol 2018;7:208-14.
2. Calvo CM, Hartnett ME. The utility of ultra-widefield fluorescein angiography in pediatric retinal diseases. Int J Retin Vitr 2018;4:21.
3. Khanamiri HN, Nakatsuka A, El-Annan J. Smartphone fundus photography. J Vis Exp 2017;125:e55958.

124 Whyte et al
June 2019 ORIGINAL ARTICLES

and manuscript development. Medical writing and editorial

Appendix 1
support were provided by Bina J. Patel, PharmD, CMPP, of
Peloton Advantage, LLC (Parsippany, NJ), an OPEN Health
Additional Members of the Study 011-10 company, and was funded by Alexion. M.P.W. was the
Investigators principal study investigator and received honoraria, travel
support, and institutional research funding and/or grant
The authors gratefully acknowledge the help of the following support from Alexion Pharmaceuticals, Inc. C.H. was a
individuals: Michael Beck, MD, Children’s Hospital, clinical study investigator and received consulting fees and
University of Mainz, Mainz, Germany; Linda DiMeglio, institutional research funding and/or grant support from
MD, Indiana University School of Medicine, Indianapolis, Alexion Pharmaceuticals, Inc. W.W., E.L., G.M.-M., J.A.,
IN; Paul Wuh-Liang Hwu, MD, PhD, National Taiwan and J.L. were clinical study investigators and received
University Hospital, Taipei, Taiwan; Katherine L. Madson, institutional research funding and/or grant support from
MD, PhD, and Upasana Nanda, MPH, Shriners Hospital Alexion Pharmaceuticals, Inc. S.M. and K.F. are employees
for Children, St. Louis, MO, USA; Peter Simm, MD, Royal of and may own stock/options in Alexion Pharmaceuticals,
Children’s Hospital, Parkville, Victoria, Australia; Jill Inc, which sponsored the study. A.R. declares no conflicts
Simmons, MD, Vanderbilt University Medical Center, of interest.
Nashville, TN; Joel Steelman, MD, Cook Children’s Health Portions of this study were presented at the Pediatric
Care System, Fort Worth, TX; Robert D. Steiner, MD, Academic Societies and Asian Society for Pediatric Research
Oregon Health & Science University, Portland, OR; and Joint Meeting, Mary 3-6, 2014, Vancouver, British Colombia,
Andrea Superti-Furga, MD, Centre Hospitalier Universitaire, Canada; the European Calcified Tissue Society Congress,
Lausanne, Switzerland. May 17-20, 2014, Prague, Czech Republic; the Society for
the Study of Inborn Errors of Metabolism Annual
Disclosures Symposium, September 2-5, 2014, Innsbruck, Austria; and
at the annual meeting of the German Society for Pediatric
This study was sponsored by Alexion Pharmaceuticals, Inc, and Adolescent Medicine (DGKJ), September 11-14, 2014,
Boston, MA, which was involved in all stages of the study Leipzig, Germany.

Natural History of Perinatal and Infantile Hypophosphatasia: A Retrospective Study 124.e1

THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume 209

Screened (N=65) Screen failure (n=17 [26%])

• Lack of documented diagnosis of
hypophosphatasia only (n=13)
• Onset of signs and symptoms of
hypophosphatasia at age >6 mo
only (n=3)
Enrolled • Both (n=1)
(N=48 [74%])

Alive at data Dead at data

abstraction (n=13 [27%]) abstraction (n=35 [73%])

Hypophosphatasia characteristics Hypophosphatasia characteristics

(key eligibility criteria) (key eligibility criteria)
• Respiratory compromise and rachitic chest (n=7) • Respiratory compromise and rachitic chest (n=18)
• Rachitic chest only (n=6) • Respiratory compromise, rachitic chest, and
vitamin B6–dependent seizures (n=8)
• Respiratory compromise and vitamin
B6-dependent seizures (n=2)
• Respiratory compromise only (n=5)
• Rachitic chest only (n=1)
• Stillborn, pulmonary hypoplasia (n=1)

Figure 1. Patient disposition flowchart.

Table II. Radiographic findings reported at

hypophosphatasia diagnosis
Radiographic findings %* (n/N)
Osteopenia 89 (33/37)
Rachitic chest/deformed ribs 86 (32/37)†
Metaphyseal fraying 68 (19/28)
Metaphyseal flaring 67 (20/30)
Metaphyseal radiolucencies (tongues) 63 (15/24)
Metaphyseal widening 55 (12/22)
Long bone bowing 48 (14/29)
Thin, gracile bones 44 (12/27)
Sclerosis 30 (6/20)
Bony spurs 26 (6/23)
Fracture (nonunion) 23 (7/30)
Absence of some/all bones 23 (7/30)
*Percent calculated as the number of patients with the finding (n) divided by the number of pa-
tients with available data (N)  100. Patients for whom status was “unknown” or missing were
not included.
†A total of 39 patients had chest deformity documented in disease history (Table IV in the main
text) and 32 patients had radiographic evidence of rachitic chest at the time of diagnosis.

124.e2 Whyte et al
June 2019 ORIGINAL ARTICLES

Table III. Respiratory support type* for each patient

BiPAP, bilevel positive airway pressure; CPAP, continuous positive airway pressure; IV, invasive ventilation.
Types of respiratory support are color-coded, with each type of support appearing in the same color.
*Respiratory support types from greatest to least: IV, CPAP, BiPAP, and/or supplemental oxygen.
†
Age (years/months/days) at which respiratory support was initiated, if available, is indicated in parentheses.
‡
Last respiratory support is the type of support that was needed up to age 5 years for patients who were alive or the final support applied before death.
§
Age (years/months/days) at death, if available, is indicated in parentheses. If patient was alive, requirement for respiratory support at age 5 years, if needed, is indicated.
{
Patient was stillborn.

Natural History of Perinatal and Infantile Hypophosphatasia: A Retrospective Study 124.e3

THE JOURNAL OF PEDIATRICS  www.jpeds.com Volume 209

Table V. Hospitalizations for reasons consistent with a diagnosis of hypophosphatasia

All patients (N = 48) Deceased patients (n = 35) Living patients (n = 13)
†
Reasons Hospitalizations, n Patients, n (%)* Patients, n (%) Patients, n (%)†
Hospitalizations for any reason 133 39 (81) 27 (77) 12 (92)
All respiratory compromise 46 22 (46) 18 (51) 4 (31)
Respiratory failure 5 2 (4) 2 (6) 0 (0)
Respiratory distress 21 12 (25) 11 (31) 1 (8)
Other‡ 20 11 (23) 7 (20) 4 (31)
Craniosynostosis 16 9 (19) 4 (11) 5 (39)
Seizure related 8 7 (15) 6 (17) 1 (8)§
Renal related 3 2 (4) 1 (3) 1 (8)
Others{ 62 28 (58) 16 (46) 12 (92)
*Percentage is calculated as n divided by 48, the number of enrolled patients,  100.
†Percentage is calculated using the number of deceased or alive patients.
‡Includes pneumonia, asthma, tachypnea, dyspnea, tachydyspnea, asthma, ventilator use, labored respirations, ventilatory care.
§Reason was to eliminate possibility of seizures.
{Reasons included hypercalcemia, infections, vomiting, failure to thrive/developmental delays, hypophosphatasia evaluation/diagnostics, orthopedic treatment (surgery, casts, fractures), exper-
imental therapies (stem cell or bone marrow transplants, enzyme replacement using Paget’s bone disease plasma infusions), abdominal pain, rehabilitation, icterus neonatorum, choking, head
injury/concussion, nasal gastric tube placement, prematurity, cardiac failure, patent ductus arteriosus ligation. Patients may have had multiple hospitalizations. Patients and hospitalizations
may be counted in multiple “reasons” categories.

124.e4 Whyte et al