Study reference

Study characteristics

Patient characteristics¹

Clinical presentation

Follow-up

Outcome measures and effect size ³

Comments

Biswas, 2017

to identify a biomarker candidate expressed during hemangioma growth, but not during involution

Type of study:

Prospective longitudinal study

Setting: Hemangioma and Vascular Malformation Clinic and the Primary Care Clinic at Nationwide Children’s Hospital

Country: Ohio

Source of funding:

Inclusion criteria: the presence of a IH on the skin ≥1 cm diameter with diagnosis confirmed by Doppler ultrasound, and age ≤ 5 months

Exclusion criteria: treatment of IH prior to or during the study, taking steroids or propranolol for other co-morbidities, hemoglobinopathies, PHACES syndrome, rapidly involuting or non-involuting congenital hemangiomas, and parents/guardian unable to provide informed consent

N total at baseline:

IH group:34

Control group: 34

Length of follow-up:

Data collection was done at 2, 4, 6, 9, 12, 18 and 24 months of age +/− 2 weeks.

miR-126 abundance was significantly elevated in the IH group at age 4 months and 6 months of age, but was not significantly different from healthy controls at age 9 months (Fig. 1).

At 4 months of age there was a moderately strong correlation between IH size and urinary miR abundance (Fig. 3a). Increased urinary miR-126 abundance was detectable for all IH. At 6 months there was a moderately strong correlation between IH size and miR-126 abundance. All IH had increased miR-126 abundance compared to controls.

Conclusion:

These are the first reported results to identify a potential miR biomarker in the urine of children with hemangiomas. Measurement of urinary levels of miR-126 could potentially be used to monitor hemangioma response to therapies.

Bracken, 2011

PHACE syndrome, to report the spectrum of clinical and imaging abnormalities

Type of study:

Retrospective review

Setting:

Country: Ireland

Source of funding:

Inclusion criteria:

Exclusion criteria:

N total at baseline:

12 patients with PHACE syndrome

Important prognostic factors¹:

Age:

N=9, 75% presented within the neonatal

period (range 2–28 days, mean 12 days, median 10 days).

N=2 presented at 6 weeks of age, N=1 presented at 5 months

Sex:

N=10, 83% were female

The majority of the patients (n=10, 83%) presented initially

to the dermatology service for evaluation of a facial

haemangioma. The remaining 2 patients, both female, presented early in the neonatal period with cardiovascular compromise. One of these patients presented on

day 5 of life with cardio respiratory distress secondary to coarctation of the aorta, requiring cardiothoracic surgical

intervention on day 8 of life. The second patient presented on day 10 of life with congestive cardiac

failure due to a large intracranial arteriovenous fistula, requiring endovascular treatment in the neonatal period.

All of the patients had a segmental craniofacial haemangioma.

Three of 5 patients with involvement of S3 segment (60%) had cardiac anomalies, only 2 of 7 patients without S3 involvement (29%) had cardiac anomalies. The single patient with S4 involvement did not have any cardiac anomaly.

Ten of the 12 patients had cerebral arterial anomalies. These cerebral arterial anomalies were multiple in 8 of 10 cases (range = 2–6, median = 2.5, mean = 3).

Length of follow-up:

ranged from 11 months to 13 years (median 5 years, mean 4 years and 7 months).

One patient in had a posterior fossa abnormality The neuroimaging findings in this case included hypoplasia of the left cerebellar

hemisphere with a retro cerebellar cyst (felt to be secondary

to the hypoplasia), with a cleft between the vermis and the left cerebellar hemisphere. This patient, a dizygotic twin,

also had hypoplasia of the A1 segment of the left anterior

cerebral artery and a left-sided facial haemangioma involving S1, S2 and S4 segments.

Conclusion:

Investigation of

infants, in particular female infants, with segmental

craniofacial haemangiomas should include MR imaging of the brain and MR angiography of the cerebral and cervical arteries. As awareness of PHACE syndrome increases, so too does knowledge of its clinical and imaging features.

[…]

As part of the multidisciplinary team, the

radiologist has an important contribution to make both at the

stage of initial diagnosis of PHACE syndrome and in the

assessment of potential complications.

Chen, 2010

to evaluate the role of 3D

computed tomography angiography (3D-CTA) in differential diagnosis of

deep

infantile hemangiomas (DIHs) and early venous malformations (EVMs)

Type of study:

Setting:

Country: China

Source of funding:

Exclusion criteria: Uncommon vascular

lesions or other soft tissue tumors were excluded clinically.

N total at baseline:

18

Important prognostic factors¹:

Age:

range, 4–18 months; mean 9 months

Sex:

6 male and 12 female

The anatomic sites

affected including the head and neck (n= 13), trunk (n= 3), and limbs (n= 2).

Length of follow-up:

-

A total of 11 patients were diagnosed with DIHs by 3D-CTA

and pathology. In plan scanning, the lesions showed isodensity (9/11) or slightly lower density (2/11), while bright, homogenous, and well-defined enhancement could be observed in contrast images

(11/11). MIP images demonstrated clearly a high density area with

feeding vessels. VR images showed an unique appearance of a

compact solid mass, with feeding vessels entering it peripherally.

Seven patients were diagnosed with EVMs by 3D-CTA and pathology. In plan scanning, the lesions showed mixed density.

Only slight or nonspecific enhancement could be observed in contrast images (7/7). MIP images demonstrated implicate vascular area

without definite feeding vessels. VR images also showed an ill-defined, obscure vessels area rather than solid masses

Conclusion:

DIHs can be distinguished

from EVMs on 3D-CTA imaging. DIHs are solid tumors of

proliferating endothelial cells with feeding blood vessels and compact,

well-defined masses are shown in 3D-CTA images. While EVMs are spongy masses composed of abnormal veins and only poorly defined patches appear in 3D-CTA images.

Chern, 2012

to correlate presenting cutaneous findings with lumbar ultrasonography (LUS) results based on a large number of LUS tests obtained by regional primary care providers

Type of study:

Retrospective chart review

Setting:

Country: Alabama

Source of funding:

Inclusion criteria:

Exclusion criteria:

N total at baseline:

1273

157 patients

were excluded because of incomplete records.

1116 patients were included in the study

Important prognostic factors¹:

For example

age ± SD:

I:

C:

Sex:

562 boys and 554 girls

Groups comparable at baseline?

Lumbar ultrasonography (LUS) was performed in 943 (84%) of 1116 infants because of the presence of cutaneous stigmata and in 173 infants (16%) because of the presence

of congenital anomaly. Among the 943 infants who presented with cutaneous stigmata, the 5 most commonly reported stigmata were sacral dimple (638 patients, 68%), hairy patch (96

patients, 10%), hemangioma (82 patients, 8.7%), deviated gluteal cleft (37 patients, 3.9%), and skin tag (24 patients, 2.5%). Thirty-one patients (3.3%) had various presenting

cutaneous findings that were categorized as “others” for analysis purposes.

An abnormal lumbar ultrasonography result was reported in 58 (16,2%) of 943 infants with cutaneaous anomaly. Reported abnormalities included low

conus (defined as below the L2–3 interspace), fat in the filum, thick filum, cyst in the filum, vertebral anomalies,

and soft-tissue abnormalities. All 58 patients were referred

for MR imaging study, and abnormal results were

confirmed on 30 MR imaging studies. Findings on the

other 28 imaging studies were unremarkable.

Conclusion:

Cutaneous markers as currently defined by general practitioners are not useful markers for predicting

OSD. The vast majority of findings on lumbar ultrasonography studies performed under these circumstances will

be negative.

The great majority of infants who were referred for lumbar ultrasonography

because of cutaneous findings did not have OSD.

Approximately 5% of these infants will have abnormal lumbar ultrasonography results, and surgical exploration

would be deemed necessary in less than 1% of these infants.

While one may argue that lumbar ultrasonography

is a cheap, nonharmful, and minimally invasive screening tool, the high false-positive rate means that some infants

will be subjected to more involved testing, such as MR imaging under general anesthesia with its accompanying

risks.

Drolet 2010

to determine the positive predictive value of infantile hemangioma for occult spinal

anomalies and to make evidence-based recommendations for screening.

Type of study: multicentre prospective cohort study

Setting:

Country:

Source of funding:

Inclusion criteria:

Exclusion criteria: Patients with IHs isolated to the skin of the perineum or coccygeal region or

with lesions not involving the midline were excluded from

the study.

N total at baseline:

Data from 48 patients were analysed;

A total of 55 patients were offered enrollment at 9 sites in 48 months; 4 families declined enrollment, and 3 patients were excluded from study analysis because they were lost to follow-up.

Important prognostic factors¹:

Age:

the mean age at evaluation was 34.1 weeks (8

months; range, 1-277 weeks)

Sex:

I: % M

C: % M

The average size of IH

was 46.1 cm2 (range, 2.3-132 cm2). Overall, 21 of 48 of the infants (43.8%) had spinal abnormalities confirmed with MRI.

When the infants who did not undergo MRI were excluded (7 patients underwent ultrasound

scanning only), 21 of 41 had spinal abnormalities,

and the positive predictive value of a lumbosacral IH for spinal abnormality was 51.2%.

Most infants found to have a spinal abnormality had no symptoms at enrollment; however, 3 of 21 infants who were found to have a spinal abnormality (14.2%) had at least one neurologic sign: lower extremity muscle weakness, abnormal gait, or limb length discrepancy.

Twenty-six patients underwent spinal ultrasound scanning.

The average age at the time of the examination was 2.0

months (8.5 weeks). A spinal abnormality was identified

with ultrasound scanning alone in 7of 26 study participants

(26.9%), a much lower percentage than in the group who underwent

MRI (51.2%). The clinical characteristics in these

two groups were not significantly different.

Length of follow-up:

Loss-to-follow-up:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Incomplete outcome data:

Intervention:

N (%)

Reasons (describe)

Control:

N (%)

Reasons (describe)

Seven patients

underwent ultrasound scanning only, and 19 patients underwent

both ultrasound scanning and MRI. For those infants that underwent both studies, agreement of results in the

two imaging modalities was evaluated. In 5 cases, the ultrasound

scanning report did not detect a spinal anomaly,

whereas MRI was diagnostic of OSD or intraspinal hemangioma,

resulting in a false-negative rate of 50%. Cohen’s kappa

coefficient for agreement between ultrasound scanning and

MRI was 0.27 (95% confidence interval [CI], –0.15-0.7; P =

.21), showing that the level of agreement was consistent

with chance. Our results show that ultrasound scanning has a sensitivity rate of 50% (95% CI, 18.7%-81.3%) and a specificity

rate of 77.8% (95% CI, 40%-97.2%).

Conclusion:

In conclusion, infants with IH overlying

the lumbosacral spine measuring ≥2.5 cm in diameter

are at high risk for underlying spinal anomalies and should undergo MRI as part of their evaluation. When no evidence of a patent sinus tract, ulceration of skin, or neurologic signs or symptoms is evident at presentation, MRI can be deferred

until 4 to 6 months of age. Infants found to have a tethered

cord or other spinal abnormalities should be referred to appropriate

pediatric specialists (neurology or neurosurgery)

for further evaluation and management.

Titel (jaar)

Exclusie reden

Wright and Wycoco (2017)

potential marker of PHACES

Albanese, Mohandas et al. (2018)

N<10

Allen, Sandquist et al. (2003)

geen full tekst

Bayer, Frommelt et al. (2013)

PHACE syndroom beschrijving --> UV 1 of 3?

Bittles, Sidhu et al. (2005)

NICH (n=2)

Canty, Horii et al. (2014)

kans/epidemiologie hepatische hemangiomen bij cutane hemangiomen

Coats, Paysse et al. (1999)

N<10, case report

De Graaf, Knol et al. (2014)

geen primaire diagnostiek, signalering complicaties

Dêbska-Rutkowska, Kwast et al. (2017)

slechts 3 hemangiomen

Eivazi, Wiegand et al. (2010)

geen diagnostische gegevens

Furuta, Shimose et al. (2017)

geen full tekst

Garcia-Romero, Chakkittakandiyil et al. (2014)

follow-up propranolol, beloop

Gorincour, Kokta et al. (2005)

NICH/RICH

Grantzow, Schmittenbecher et al. (2008)

guideline management, geen diagnostiek

He and Huang (2017)

follow-up lokaal timolol

Horii, Drolet et al. (2010)

risico van grote hemangiomen op hepatische hemangiomen

Hughes, Hill et al. (2004)

prevalentie hepatische hemangiomen --> UV 1 of 3?

Judd, Chapman et al. (2007)

N<10

Kutz, Aranibar et al. (2015)

follow-up propranolol

Leng, Wang et al. (2008)

slechts 5 hemangiomen

Lope, Hutcheson et al. (2010)

slechts 4 hemangiomen

Lotfinia, Sayahmeli et al. (2014)

slecht enkele hemangiomen, dysraphina

Miller and Frieden (2005)

geen full tekst

North, Waner et al. (2000)

reeds in review van Vught

O'Neill, Gallegos et al. (2017)

MRI voor diagnostiek occult spinal dysraphisme

Prada, Mortera et al. (2010)

N<10

Robertson, Kimble et al. (2016)

follow-up propranolol

Rossler, Rothoeft et al. (2011)

N<10

Rozylo-Kalinowska, Brodzisz et al. (2002)

slechts 7 hemangiomen, doppler ultrasonography

Sardana, Gupta et al. (2009)

slechts 5 hemangiomen, dysraphisme

Schicchi, Tagliati et al. (2018)

slechts 5 hemangiomen, MRI

Shi, Song et al. (2014)

follow-up propranolol

Siegel, Tefft et al. (2012)

risico op stroke bij PHACES --> UV 1 of 3?

Sleep, Fairhurst et al. (2002)

N<10

Sodhi, Bartlett et al. (2011)

beschrijvende studie

Steiner, McCoy et al. (2018)

beschrijvende studie, PHACES spectrum evt. uv 1 of 3?

Stockman, Boralevi et al. (2007)

complicaties bij lumbosacrale hemangiomen (ulceratie) --> UV 1 of 3?

Talaat, Elbasiouny et al. (2012)

follow-up propranolol

Tobouti, Olegário et al. (2017)

niet alleen hemangiomen

Tortora, Severino et al. (2017)

N<10

Tubbs, Wellons et al. (2004)

geen diagnostische interventie onderzocht

Verity, Restori et al. (2006)

klinisch beloop --> UV 1 of 3

Vilanova, Barceló et al. (2004)

beschrijvende studie

Visscher, Burkes et al. (2016)

beloop en behandelrespons metingen

Waelti, Rypens et al. (2018)

RICH

Wu, Liu et al. (2012)

niet alleen hemangiomen

Yang, Jiang et al. (2011)

geen full tekst

Yilmaz, Kozakewich et al. (2014)

Geen infantiele hemangiomen