Differential diagnosis of perinatal hypophosphatasia: radiologic perspectives
Perinatal hypophosphatasia (HPP) is a rare, potentially life-threatening, inherited, systemic metabolic bone disease that can be difficult to recognize in utero and postnatally. Diagnosis is challenging because of the large number of skeletal dysplasias with overlapping clinical features. This review focuses on the role of fetal and neonatal imaging modalities in the differential diagnosis of perinatal HPP from other skeletal dysplasias (e.g., osteogenesis imperfecta, campomelic dysplasia, achondrogenesis subtypes, hypochondrogenesis, cleidocranial dysplasia). Perinatal HPP is associated with a broad spectrum of imaging findings that are characteristic of but do not occur in all cases of HPP and are not unique to HPP, such as shortening, bowing and angulation of the long bones, and slender, poorly ossified ribs and metaphyseal lucencies. Conversely, absent ossification of whole bones is characteristic of severe lethal HPP and is associated with very few other conditions. Certain features may help distinguish HPP from other skeletal dysplasias, such as sites of angulation of long bones, patterns of hypomineralization, and metaphyseal characteristics. In utero recognition of HPP allows for the assembly and preparation of a multidisciplinary care team before delivery and provides additional time to devise treatment strategies.
KeywordsComputed tomography Hypophosphatasia Metabolic bone disease Perinatal Radiography Skeletal dysplasia Ultrasound
Hypophosphatasia (HPP) is a rare, inherited, systemic, metabolic bone disease caused by low tissue-nonspecific alkaline phosphatase activity [1, 2, 3]. In patients with HPP, low alkaline phosphatase activity results in the accumulation of phosphorylated substrates, specifically inorganic pyrophosphate, pyridoxal 5′-phosphate and phosphoethanolamine [1, 3, 4, 5]. Elevated inorganic pyrophosphate levels inhibit mineralization of the bone matrix, leading to hypomineralization of the skeleton [2, 6, 7, 8]. The inability of pyridoxal 5′-phosphate, the circulating form of vitamin B6, to cross the blood-brain barrier likely contributes to the seizures observed in some infants with HPP [2, 8].
HPP is a clinically heterogeneous disease traditionally categorized by the age of onset of the first signs and symptoms as perinatal onset (in utero and at birth), infantile onset (age < 6 months), childhood onset (age ≥ 6 months to <18 years), and adult onset (age ≥ 18 years) or, in patients with only dental manifestations, as odonto-HPP [2, 3, 9]. Characteristic signs, symptoms and complications of perinatal HPP include skeletal manifestations (e.g., hypomineralization, chest deformity, bowing, craniosynostosis) [9, 10], vitamin B–responsive seizures [9, 11, 12, 13] and respiratory failure [11, 14]. Before the availability of enzyme replacement therapy, mortality among patients with perinatal/infantile HPP was high, ranging from 58% to 100% within the first year of life [15, 16, 17]. The incidence of HPP has been estimated to be 1:100,000 in Ontario, Canada, based on the local birth rate in 1957 . The prevalence of perinatal and infantile HPP in Europe has been estimated to be 1:538,000, based on molecular diagnoses made from 2000 to 2009 . Local populations with a higher incidence of HPP include the Mennonite communities in Canada  and an endogamous village in Hungary [20, 21]. Because of the rarity of HPP, its true incidence and prevalence remain unknown .
HPP is confirmed with consistently low age- and gender-adjusted alkaline phosphatase activity in conjunction with medical history and physical findings, radiologic findings, elevated levels of tissue-nonspecific alkaline phosphatase substrates or sequencing of the ALPL gene [3, 22, 23]. In utero and postnatal recognition and diagnosis of perinatal HPP based on radiologic findings can be challenging because of features that overlap with many of the more than 400 other skeletal dysplasias, the phenotypic variability and a lack of information about the in utero natural history of HPP [24, 25, 26]. The skeletal abnormalities and the gestational and postnatal ages at which they manifest vary across skeletal dysplasias, including HPP. Many sonographic and radiographic findings are not pathognomonic for a specific disorder, as obtaining reliable information regarding skeletal mineralization is difficult with prenatal sonography and computed tomography (CT). These difficulties are confounded by a general lack of familiarity with HPP among the health care providers who perform prenatal ultrasound (US) and neonatal imaging. In addition, abnormalities can be detected at earlier gestational ages than they have in the past , underscoring the need for obstetricians, ultrasonographers and radiologists to possess in-depth knowledge of the appearance of the fetal skeleton at all gestational ages . This review focuses on the role of fetal and neonatal imaging modalities in the differential diagnosis of perinatal HPP.
Prenatal diagnosis of skeletal dysplasia relies on cross-sectional imaging modalities (US, CT and magnetic resonance imaging [MRI]), whereas postnatal diagnosis relies more heavily on radiography . The International Society of Ultrasound in Obstetrics and Gynecology  and the United Kingdom’s National Institute for Health and Care Excellence  recommend that all pregnant women undergo US scanning at 10 to 14 weeks to establish gestational age and at 18 to 22 weeks to screen for structural anomalies. Thereafter, the frequency of fetal monitoring depends on the severity of findings, the mother’s health and the family’s wishes. High-resolution US is required to clearly identify the skeletal abnormalities of HPP. Two- or three-dimensional (2-D or 3-D) US may be used to visualize the skeleton by gestational week 12 . Although radiologists are usually trained with 2-D images and generally prefer 2-D to 3-D US when reviewing image slices, 3-D US may allow the radiologist to more clearly visualize characteristic dysmorphic findings of the face, hands, feet, vertebrae, ribs and skull sutures in skeletal dysplasias .
Prenatal CT is a useful modality when skeletal dysplasia is suspected after sonographic examination [33, 34] and is best performed from 30 weeks’ gestation; image quality is poor at earlier gestational ages because of relatively poor skeletal mineralization and artefacts caused by fetal movement [29, 34]. Although the added diagnostic value of CT over US has not been formally assessed in a large study, CT allows detailed visualization of the fetal skeleton and is less dependent on amniotic fluid volume and fetal position than US [33, 34, 35]. The risk associated with fetal radiation is a common concern with prenatal CT; however, the risk to benefit ratio can be relatively low if the radiation dose is kept to a minimum by selecting appropriate technical parameters [34, 36, 37, 38, 39]. A recommended threshold of radiation that would have negligible risk to the fetus is 50 milligray . Advances in model-based iterative reconstruction methods for ultra-low-dose fetal CT yield fetal radiation exposures as low as 0.5 milligray while maintaining excellent image quality for the diagnosis of skeletal dysplasias .
MRI has shown only limited utility in prenatal diagnosis of skeletal dysplasias and is not routinely used in HPP diagnosis [29, 41]; however, fetal MRI can provide valuable details when targeted US is unable to clarify the diagnosis [42, 43]. Fetal “black bone” MRI, compared with standard MRI sequences, may improve visualization of the mineralized skeleton .
Cross-sectional imaging modalities (e.g., US, CT, MRI) are generally reserved for specific skeletal and systemic abnormalities. Whole-body MRI findings associated with HPP have been described in children  but may be less useful/practical in a perinatal setting.
Long bones: shortening, bowing, angulation
Small/narrow thorax (chest size smaller than abdominal circumference)
Osteochondral spurs (Bowdler spur)
Lucencies (“cupping” or “tongues”)
Short and beadeda
Deficient/absent ossification of bones
Tubular bones, skull vault, ribs, vertebrae
Abnormal sonolucency of bony structures
Increased nuchal translucency
Wide sutures and fontanellesb
The gestational age at which skeletal abnormalities are apparent in perinatal HPP varies widely, with some cases detected on prenatal US as early as 13 weeks’ gestation . However, HPP diagnosis based on US findings cannot be definitive until halfway through the second trimester, when characteristics of in utero HPP become more evident. In the second trimester, all bony features become more apparent as the fetus grows and increase in visibility as ossification progresses. If the diaphysis of a tubular bone is not ossified in the second trimester, it is likely abnormal rather than physiologically related to gestational age.
Prenatal US findings may help predict lethality of a skeletal dysplasia . The three most accurate predictors of fatality when evaluated in conjunction with fetal amniotic fluid volume are 3-D fetal lung volume [67, 68], the ratio of femur length to abdominal circumference and the ratio of chest circumference to abdominal circumference . In general, the risk of fatality is greater if chest size is small and/or multiple rib fractures are present because this will lead to breathing difficulties ex utero [14, 67, 68]. Lethal perinatal HPP is characterized by diffuse hypomineralization of the fetal skeleton with the absence of many bones and a lack of posterior acoustic shadowing from bones that are sonographically visible [54, 67]. In particular, the neural arches and the thoracic spine may be poorly ossified or absent [56, 67]. In general, lethal perinatal HPP is clearly more severe than other forms of HPP at first detection, with a lack of improvement in skeletal signs with increasing gestational age. The severity of hypomineralization may also predict fatality. A diagnosis of lethal HPP can usually be made by the late second or early third trimester. Lack of mineralization of bones in the hands is considered an important feature. However, no correlation is apparent between the gestational age when skeletal disease is first observed and the severity of HPP after birth .
Pregnancy in cases of perinatal HPP may be complicated by polyhydramnios . Whether perinatal HPP is associated with other maternal complications, small size for gestational age or premature birth has not been systematically studied. A retrospective review of 15 Manitoban Mennonite patients with perinatal HPP reported during an 80-year period (1927–2007) found that most of the infants (73.3% [11/15]) were born at full term, 13.3% (2/15) were born early at 36 weeks’ gestation, and 13.3% (2/15) were born prematurely at 30 and 33 weeks’ gestation . Birth weights (n=6) ranged from below the 5th percentile (2.3 kg) to within the 25th–50th percentile (3.3 kg). As technology advances, we may learn more about maternal complications.
Metaphyseal abnormalities similar to those observed in HPP are also observed in rickets and osteopathy of prematurity . Active rickets may present with widened zones of provisional calcification and wide costochondral junctions, including widening along the anterior ends of the ribs (i.e. rachitic rosary). Osteopathy of prematurity is associated with radiologic changes characteristic of rickets, and fractures may be seen in infants with very low birth weights [69, 70].
Differential diagnoses for perinatal hypophosphatasia (HPP)
Osteogenesis imperfecta type II
Absent ossification of whole bones
Diaphyseal spurs (tibial dimple)
Metaphyseal lucencies (“tongues”)
Wrinkled appearance on sonograms
Shortening of the long bones
Angulation (tibiae and short fibula)
Mild to moderate bowing
Diaphyseal spurs (tibial dimple)
Femoral bowing (type I)
Small thoracic circumference
Thick/broad irregular ribs
Continuous beading of ribs
Flaring at anterior rib ends
Small chest circumference
11 pairs of ribs
Absent or hypoplastic scapula
Small, bell-shaped thorax
Narrow barrel-shaped thorax
Hypoplastic or absent clavicles
Absent ossification of skull base
Wide sutures and fontanelles
Prominent falx cerebri in sonograms
Hypomineralization of skull
Abnormal skull shape
Flattened facial profile
Severe hypomineralization of skull
Flattened facial profile
Hypomineralization of the skull
Wide sutures and fontanelles
Parietal bones absent in extreme cases
Craniosynostosis (type II)
Narrowed or closed cranial sutures
Underossification of vertebrae and neural arches
Absent pedicles and bodies of the vertebrae
Increased nuchal translucency
Absent pedicles of the T spine
Severe hypomineralization of vertebral bodies
Large nuchal lucencies
Deficient ossification of the pubic rami
Wide symphysis pubis
Hands and feet
Underossification and sonolucency of hands
Hands appear grossly normal and are echogenic
Short fingers and toes
Very small hands and feet
Decreased echogenicity of bones
Small for gestational age
Normal bone density
Flat nasal bridge
Short crown–rump length
Polyhydramnios (type II)
Normal trunk length
Depressed nasal bridge
In achondrogenesis and hypochondrogenesis, deficient ossification of the vertebral bodies is usually most severe in the lumbosacral and cervical spine . In achondrogenesis, the whole spine may be unossified, with complete absence of vertebral bodies. In contrast, HPP typically presents as deficient spine ossification in the thoracic region, with a sharp demarcation between almost normal ossification in the lumbar spine and complete absence of ossification in the thoracic spine.
Suspicion of hypophosphatasia: Next steps
If perinatal HPP is suspected, alkaline phosphatase activity may be assessed in umbilical cord blood or chorionic villus samples [86, 87, 88, 89]. Chorionic villus sampling for assay of alkaline phosphatase activity has been used to diagnose HPP as early as 11–12 weeks’ gestation [89, 90]. In some cases, however, alkaline phosphatase activity may fail to identify affected fetuses, as alkaline phosphatase activity varies with gestational age [89, 90]. After 13 weeks’ gestation, the placenta may produce alkaline phosphatase, affecting the interpretation of the analysis.
Measurement of parental serum alkaline phosphatase activity can be useful for prenatal diagnosis of HPP . A retrospective analysis of 77 cases of fetal skeletal dysplasia (including 17 of HPP) in Japan from 2007 to 2016 showed that the presence of at least one abnormally low maternal (<123 IU/L) or paternal alkaline phosphatase value (<165 IU/L) any time during pregnancy had high sensitivity (82%) specificity (75%), and positive predictive value (80%) for HPP .
For newborns, alkaline phosphatase activity must be compared with age- and gender-adjusted reference ranges for the testing laboratory; alkaline phosphatase reference ranges vary widely depending on patient age and gender and the laboratory and methods used . If alkaline phosphatase activity is low or suspicion for HPP is high based on images, additional testing may be necessary. In newborns, these tests should include urine concentrations of phosphoethanolamine, and serum concentrations of pyridoxal 5′-phosphate (i.e. vitamin B6), calcium, vitamin D and parathyroid hormone .
Genetic testing for ALPL mutations can be confirmatory in cases of diagnostic uncertainty [91, 92]. However, clinicians should be aware of the depth of coverage with whole-exome sequencing and next-generation sequencing technology, as some pathogenic variants may not be detected [92, 93].
Medical genetic evaluation and genetic counseling
A medical genetic evaluation should be obtained when a diagnosis of HPP is being considered. In many centers, this will be through a prenatal genetics clinic. A medical genetics physician or genetics counselor can help obtain gene testing and interpret results, especially if variants of unknown significance are found. If gene testing results are normal, alternative genetic causes for the apparent bony abnormalities can be pursued in consultation with the radiologist and obstetrician. Family genetic counseling should also begin when HPP is suspected, and a detailed pedigree should be obtained. Genetic counseling in cases of suspected HPP may be particularly difficult due to the autosomal-dominant and autosomal-recessive patterns of inheritance and the phenotypic heterogeneity of the disease . The family should receive counseling by a clinician familiar with the treatment of children with HPP so that they are informed on available treatment options before making decisions about terminating the pregnancy.
Prompt diagnosis of HPP is important, as the health care provider must evaluate treatment strategies for newborns with HPP as early as possible. If the decision is made to treat, treatment should begin as early as possible postnatally. All decisions must be made in consultation with the parents. It is important to assemble a multidisciplinary team for care in the perinatal period. The team should include the following specialists: neonatologist, pediatrician, geneticist, endocrinologist, radiologist, nephrologist, specialist nurses, genetic counselor, social worker, physical therapist, occupational therapist, respiratory physician (especially if long-term ventilation is needed), and an ears, nose, and throat specialist, neurologist and craniofacial surgeon (if craniosynostosis is present). Neonatologists should be prepared to provide invasive respiratory support, as many babies born with skeletal dysplasias are likely to require ventilation.
Perinatal HPP is associated with a broad spectrum of imaging findings that overlap with other perinatal skeletal dysplasias. Certain features (e.g., sites of angulation and hypomineralization, spurs, metaphyseal characteristics) may help distinguish HPP from other skeletal dysplasias. ALPL gene mutation testing during pregnancy can confirm the diagnosis before delivery. Alkaline phosphatase results are essential for confirming a suspected diagnosis of perinatal HPP. Early recognition of the disease provides more opportunity for education and counseling to prepare the parents, allows for the assembly and preparedness of a multidisciplinary care team upon delivery, and provides additional time to consider and discuss treatment options, with the goal of improving duration and quality of life or minimizing unnecessary suffering for the affected child and the family.
Dr. Gen Nishimura of Tokyo Metropolitan Children’s Medical Center, Japan, contributed images that appear in this article.
Compliance with ethical standards
Conflicts of interest
This review was sponsored by Alexion Pharmaceuticals Inc. Writing and editorial support was provided by Lela Creutz, PhD, and Bina J. Patel, PharmD, CMPP, of Peloton Advantage, LLC, and funded by Alexion Pharmaceuticals Inc. Amaka C. Offiah has received consulting fees, travel support and research grant support from Alexion Pharmaceuticals Inc. for participating on advisory boards and to support her research. Jerry Vockley has received consulting fees, travel support and research grant support from Alexion Pharmaceuticals Inc. for consulting and participating on advisory boards. Craig F. Munns has received consulting fees from Alexion Pharmaceuticals, Inc. Jun Murotsuki has received honoraria, travel support and consulting fees from Alexion Pharmaceuticals, Inc.
- 2.Whyte MP (2018) Hypophosphatasia and how alkaline phosphatase promotes mineralization. In: Thakker RV, Whyte MP, Eisman J, Igarashi T (eds) Genetics of bone biology and skeletal disease, 2nd edn. Elsevier (Academic Press, London), San Diego, CA, pp 481–504Google Scholar
- 3.Rockman-Greenberg C (2013) Hypophosphatasia. Pediatr Endocrinol Rev 10(Suppl 2):380–388Google Scholar
- 11.Baumgartner-Sigl S, Haberlandt E, Mumm S et al (2007) Pyridoxine-responsive seizures as the first symptom of infantile hypophosphatasia caused by two novel missense mutations (c.677T>C, p.M226T; c.1112C>T, p.T371I) of the tissue-nonspecific alkaline phosphatase gene. Bone 40:1655–1661CrossRefGoogle Scholar
- 13.Balasubramaniam S, Bowling F, Carpenter K et al (2010) Perinatal hypophosphatasia presenting as neonatal epileptic encephalopathy with abnormal neurotransmitter metabolism secondary to reduced co-factor pyridoxal-5′-phosphate availability. J Inherit Metab Dis 33(Suppl 3):S25–S33CrossRefGoogle Scholar
- 19.Greenberg CR, Evans JA, McKendry-Smith S et al (1990) Infantile hypophosphatasia: localization within chromosome region 1p36.1-34 and prenatal diagnosis using linked DNA markers. Am J Hum Genet 46:286–292Google Scholar
- 29.Offiah AC (2015) Skeletal dysplasias: an overview. In: Allgrove J, Shaw NJ (eds) Calcium and bone disorders in children and adolescents, Vol. 28, 2nd edn, revised. Karger, Basel, Switzerland, pp 259–276Google Scholar
- 31.Antenatal care for uncomplicated pregnancies [clinical guideline CG62] (2016, Updated 2017) National Institute for Health and Care Excellence, London. Available via https://www.nice.org.uk/guidance/cg62 (accessed August 3, 2018)
- 38.Hall CM, Offiah A, Forzano F et al (2012) Diagnosis of fetal skeletal dysplasias. In: Fetal and perinatal skeletal dysplasias: an atlas of multimodality imaging. Radcliffe Publishing, LondonGoogle Scholar
- 39.Sadro CT, Dubinsky TJ (2013) CT in pregnancy: risks and benefits. Appl Radiol 42:6–16Google Scholar
- 41.Noel AE, Brown RN (2014) Advances in evaluating the fetal skeleton. Intl J Womens Health 6:489–500Google Scholar
- 45.Offiah AC, Hall CM (2003) Radiological diagnosis of the constitutional disorders of bone. As easy as A, B, C? Pediatr Radiol 33:153–161Google Scholar
- 47.Kritsaneepaiboon S, Jaruratanasirikul S, Dissaneevate S (2006) Clinics in diagnostic imaging (112). Perinatal lethal hypophosphatasia (PLH). Singap Med J 47:987–992Google Scholar
- 51.Gortzak-Uzan L, Sheiner E, Gohar J (2000) Prenatal diagnosis of congenital hypophosphatasia in a consanguineous Bedouin couple. A case report. J Reprod Med 45:588–590Google Scholar
- 55.Sergi C, Mornet E, Troeger J, Voigtlaender T (2001) Perinatal hypophosphatasia: radiology, pathology and molecular biology studies in a family harboring a splicing mutation (648+1A) and a novel missense mutation (N400S) in the tissue-nonspecific alkaline phosphatase (TNSALP) gene. Am J Med Genet 103:235–240CrossRefGoogle Scholar
- 61.Wiebe S, Suchet I, Lemire EG (2007) Radiographic and prenatal ultrasound features of perinatal lethal hypophosphatasia - differentiation from osteogenesis imperfecta type II. S Afr J Radiol 11:32–35Google Scholar
- 68.Barros CA, Rezende Gde C, Araujo Junior E et al (2016) Prediction of lethal pulmonary hypoplasia by means fetal lung volume in skeletal dysplasias: a three-dimensional ultrasound assessment. J Matern Fetal Neonatal Med 29:1725–1730Google Scholar
- 75.Mornet E, Nunes ME (2016) Hypophosphatasia. In: Pagon RA, Adam MP, Ardinger HH et al (eds) GeneReviews. University of Washington, SeattleGoogle Scholar
- 77.Mundlos S (1999) Cleidocranial dysplasia: clinical and molecular genetics. J Med Genet 36:177–182Google Scholar
- 85.Benacerraf BR, Lister JE, DuPonte BL (1988) First-trimester diagnosis of fetal abnormalities. A report of three cases. J Reprod Med 33:777–780Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.