Propranolol Reduces Infantile Hemangioma Volume and Vessel Density

Abstract

Objective

To evaluate changes in infantile hemangioma tissue before and after propranolol therapy, using gray-scale and color Doppler ultrasound imaging.

Study Design

Case series with chart review.

Setting

Tertiary pediatric hospital.

Subjects and Methods

Medical records and image studies of head and neck infantile hemangioma patients treated with propranolol, identified in a quality improvement database, were reviewed. Patients with imaging before and at least 4 weeks following the initiation of treatment were included. Data collected included sex, age, location, and concurrent treatment. Student t tests were used to evaluate change in cutaneous lesion area, volume, and vessel density. Logistic regression was used to compare lesion area, volume, and vessel density.

Results

Of the 177 patients identified, 19 met inclusion criteria. Fourteen of 19 were female, and 5 of 19 were older than 1 year. Mean lesion area change with treatment was 13.0 cm² (range, −2.8 to 28.9 cm², P = .05). Measured volume change was a mean of 10.3 cm³ (range, 1.5-19.2 cm³, P = .01). Mean vessel density change was 4.4 vessels per cm² (range, 2.5-6.3 vessels per cm², P < .01). Treatment decreased clinically determined hemangioma area proportionately less than gray-scale and color Doppler ultrasound measured lesion volume. Gray-scale and color Doppler ultrasound measured treatment response did not differ with sex, lesion location, or age at propranolol initiation.

Conclusion

Gray-scale and color Doppler ultrasound imaging of propranolol-treated infantile hemangiomas detected a significant reduction in lesion volume and vessel density. Patient age at propranolol treatment and concomitant corticosteroid use did not affect lesion volume change.

Keywords

infantile hemangioma propranolol gray-scale and color Doppler ultrasound imaging

Infantile hemangiomas (IHs) have been treated with a variety of agents. In 2008, propranolol was described as a dramatically effective IH treatment.¹ Currently, propranolol is widely used and is recommended as the first-line treatment for IH in many locations.^2-5 However, IH response to propranolol treatment is sometimes inconsistent.³ In some cases, IHs have regrown after cessation of propranolol therapy.⁶ Propranolol’s mechanism for shrinking IH is unknown.⁷ It has been hypothesized that propranolol could induce IH shrinkage through vasoconstriction, inhibition of angiogenesis, induction of apoptosis, or dysregulation of the renin-angiotensin system.⁸ In support of this hypothesis, propranolol has been shown to decrease angiogenesis in vitro.⁹

As with any new treatment, there are many unanswered questions. Optimal duration of propranolol therapy for IH has not been established, nor has consensus about why some IHs recur after propranolol cessation. Infantile hemangioma recurrence is especially troublesome in function-threatening IH (ie, airway, periocular), and IH treatment efficacy is typically measured clinically through change in IH appearance and area.^3,5 Refining treatment duration will require methods other than clinical examination to determine optimal length of therapy. Identifying additional methods to accurately quantify IH’s treatment response is important because it is still unclear whether specific subgroups are more likely to be successfully treated with propranolol.¹⁰ Imaging of IH pre- and posttreatment may be a helpful adjunct to quantify therapeutic response and determine duration of therapy. Magnetic resonance imaging (MRI), computerized tomography (CT), and ultrasound imaging are all useful in IH diagnosis.^11-14 Both MRI and CT often require general anesthesia in young children and are expensive to repeat. Given these considerations, ultrasound imaging is appealing for use in this population as a rapid and noninvasive means of evaluating IH size and monitoring propranolol therapy.

Beyond characterizing the dimensions of the IH, color Doppler ultrasound can be used to evaluate vessel density in pediatric vascular lesions and in some inflammatory conditions predicts disease activity.¹⁵ It is possible that serial IH volume and vessel density measurements might aid in determining candidacy for propranolol therapy or monitoring treatment outcome. In this manner, ultrasound would be used for more than diagnosis.¹⁶ Prior reports describe ultrasound imaging as a useful means of monitoring treatment response in patients receiving medical/surgical intervention for hemangiomas.¹¹ The purpose of this report is to describe an initial experience with combined gray-scale and color Doppler ultrasound (GCDU) imaging use to monitor changes in IH volume and blood vessel density in response to propranolol treatment.

Methods

With Seattle Children’s Hospital (SCH) Institutional Review Board (IRB) approval, the SCH Vascular Anomalies Clinic quality improvement database was used to identify patients with IH treated with propranolol. Data collected included patient age at treatment, propranolol dosage, clinical IH dimensions and location, ultrasound-measured IH volume, and flow Doppler vessel density. Clinical measurements were obtained in a gently held child by one author (J.A.P.) using a tape measure to determine the largest length and width of each IH. Surface protrusion of the hemangioma was measured but not included in this data set. Children were included in the study if they had a diagnosis of IH and had GCDU with images taken within 25 weeks prior to and at least 4 weeks following the initiation of propranolol treatment. Participants whose imaging was of insufficient quality to interpret IH volume or vessel density were excluded from the study. Propranolol therapy, when clinically indicated or desired, was initiated and administered in a standardized manner as we have described previously, with a goal of maintaining propranolol at 2 mg/kg/d for 6 to 8 months.¹⁷ Clinical IH dimensions, length and width, were used to calculate the cutaneous IH area. Infantile hemangioma volume was calculated by using IH dimensions from GCDU, MRI, or CT in the prolate ellipsoid formula (height * width * length * (π/6)).¹⁸ Using GCDU and a high-frequency transducer (ranging from 5-15 mHz), a 1-cm² area within the IH was localized where the highest vessel density occurred (Philips iU22, Bothell, Washington). By GCDU, the vessels in the 1-cm² box were reported as vessels per square centimeter. For consistency, all measurements were made and reviewed by one radiologist (N.-J.V.).

Data were analyzed with descriptive statistics and, as measurements were normally distributed, Student t test. The t tests were used to compare the proportional volume changes using STATA 10.0 (StataCorp, College Station, Texas). Linear regressions were done to evaluate the association between vessel density changes and 2-dimensional clinical area measurements against the log-transformed proportional volume changes. Proportional volume changes were log-transformed so that we could detect a linear relationship between exposure and outcome. In this way, a 50% reduction in size is perceived as the equal but opposite change as a doubling of size. The t tests were also used to evaluate whether change in IH volume, vessel density, and clinical area was different than zero.

Results

A total of 177 IH patients were identified who had been treated with propranolol. Nineteen patients met inclusion criteria and were included in the analyses ( Figure 1 ). Sixteen patients had GCDU imaging, which was used to determine IH volume and vessel density. Three patients had imaging (either MR or CT) before propranolol therapy that only allowed volume determination at that time point. One patient was excluded because imaging was performed more than 25 weeks prior to treatment, and 3 patients were excluded because of the difficulty in interpreting hepatic and intraspinal IH volume and vessel density. The average time between image acquisition and propranolol initiation was 2 weeks (range, 0-14 weeks), with some having the imaging on the day propranolol therapy began. The patient with imaging 14 weeks prior to propranolol initiation was younger than 12 months when this therapy was begun. Of the included patients, 14 were female ( Table 1 ). Prior to propranolol therapy, 9 patients had received single-therapy systemic steroids, 1 had a single pulsed-dye laser treatment, and 1 received vincristine and systemic steroids. In all 11 cases that had prior treatment, propranolol was used because the previous therapy had not changed the IH or reduced its size, as determined by clinical measurements, photographs, and diagnostic imaging. The total oral propranolol dose of 2 mg/kg/d was given in 3 doses. Duration of therapy varied, but among children whose propranolol treatment was initiated before 12 months of age, treatment was continued at least until 12 months of age. In this group, mean duration of propranolol therapy was 33.7 weeks (median, 33.8 weeks; range, 18.6-50.4 weeks). Those children whose treatment was initiated after 12 months of age were treated for at least 6 months. In this group, mean duration of propranolol therapy was 42.6 weeks (median, 42.9 weeks; range, 24.7-60 weeks). Two patients consistently missed 1 dose of propranolol per day, and dosage was decreased in 1 patient due to possible sleep disruption. In 1 subject, the propranolol dose was increased to 3 mg/kg/d in an unsuccessful attempt to improve the clinical response.

Table 1.

Selected Characteristics of Participants (n = 19) in This Study

Variable	No. (%)
Sex
Female	14 (73.7)
Male	5 (26.3)
Patient’s age at initiation of propranolol, mo
<12	14 (73.7)
12-28	5 (26.3)
Tumor location
Frontonasal (zone 1)	2 (10.5)
Maxillary (zone 2)	1 (5.3)
Parotid (zone 3)	11 (57.9)
Other	5 (26.3)
Pretreatment area (clinically assessed), cm²
2.25-12	6 (35.3)
12.25-18	5 (29.4)
20.25-121	6 (35.3)
Missing	2 (10.5)
Pretreatment volume (radiologically assessed), cm³
0.5-4.4	5 (26.3)
4.5-19.7	7 (36.8)
19.8-72.5	7 (36.8)
Total propranolol dose, mg/kg/d
1-2	3 (15.8)
2-<3	15 (78.9)
3+	1 (5.3)
Prior steroid treatment
Yes	12 (63.2)
No	7 (36.8)
Patient still taking propranolol
Yes	5 (26.3)
No	14 (73.7)

No subjects stopped propranolol because of side effects.

We observed a significant mean reduction in IH volume, area, and vessel density as determined with log transformation of proportional IH measurements ( Table 2 ). The mean proportional IH volume change did not differ between measured time points in all variables evaluated ( Table 3 ). Note that initiation of propranolol in children younger than 1 year was associated with a mean reduction of IH volume of 70%. If propranolol was initiated in children older than 1 year, the IH volume was reduced by 30%. There was no statistical difference between these variables. The mean proportional volume change, for all subjects, was a reduction of 41%. The mean proportional area change was not significantly changed ( Table 2 ). The mean vessel density as determined with GCDU imaging was significantly decreased ( Table 2 and Figures 2B & D , 3B & D , 4B & D ).

Table 2.

Change in Hemangioma Volume, Area, and Vessel Density over Time, as Analyzed with Log-Transformed Proportional Changes

Hemangioma Measurements	No. Observations	Time A	Time B	Mean Difference	P Value	Linear Association between Measured Differences and Log of Proportional Change^a
Mean volume change, cm³, mean (min-max)	19	21.5 (10.7 to 32.4)	11.2 (0.8 to 21.6)	10.3 (1.5 to 19.2)	.01	0.03 (−0.03 to 0.09)
Clinical mean area change, cm², mean (min-max)	15^b	26.1 (9.4 to 42.7)	13.1 (7.6 to 18.5)	13.0 (−2.8 to 28.9)	.05	0.03 (−0.08 to 0.02)
Change in mean vessel density per cm², mean (min-max)	15^c	13.4 (12.1 to 14.7)	9.0 (7.6 to 10.4)	4.4 (2.5 to 6.3)	<.01	0.12 (−0.08 to 0.33)

Proportional change = volume measurement 2 (time B)/volume measurement 1 (time A).

Two patients missing clinical area measurements at time A, 2 at time B.

Four patients missing vessel density measurements at time A and 3 at time B.

Table 3.

Mean Proportional Volume Change by Groups Defined by Sex, Age, Location of Infantile Hemangioma (IH), and Concurrent Treatment

	No.	Mean Proportional Volume Change	P Value^a
All IH	19	0.6
Sex
Female	14	0.5
Male	5	0.7
Difference of means		0.2	.29
Age at propranolol treatment
<1 y	14	0.7
12-28 mo	5	0.3
Difference of means		0.4	.54
IH location
Frontonasal (zone 1)	2	0.1
Maxillary (zone 2)	1	<0.1
Parotid (zone 3)	11	0.8
Other	5	0.5
Concurrent steroid treatment
Propranolol alone	13	0.6
Propranolol with steroids	6	0.5
Difference of means		0.2	.31

Obtained by performing a t test comparing the means of the log-transformed proportional volume changes.

Specific case examples are provided ( Figures 2 - 4 ). These demonstrate the changes seen clinically and with GCDU imaging.

Discussion

This study demonstrates that GCDU imaging can measure changes in IH volume and vessel density in conjunction with propranolol therapy. These parameters, volume and vessel density, were significantly reduced in all subjects. Interestingly, there was no difference in proportional IH volume change when controlling for patient sex, age at initiation of propranolol, IH location, or when steroid therapy was initiated with propranolol; however, the absence of a significant effect may be a reflection of the small numbers of patients in each subgroup. This is especially apparent when propranolol was initiated in children younger than 1 year, in whom there was a 70% IH volume reduction, compared to a 30% IH volume reduction in subjects having propranolol initiated after age 1 year. The lack of significant statistical difference between these groups is probably a reflection of a small sample size, not the absence of a clinically meaningful difference between the 2 groups. Usual clinical measures were not as sensitive to IH change in response to propranolol when compared with change measured by GCDU imaging.

Using clinical parameters, it is difficult to objectively measure noncutaneous IH treatment response. Focusing only on the cutaneous portion of an IH does not always indicate how the deeper portions of IH are changing with medical therapy. Gray-scale color Doppler ultrasound imaging offers clinicians a means to noninvasively “see” below the surface to better characterize and measure potential changes in a treated IH. Our findings show greater proportional changes in GCDU imaging measures of vessel density change and IH volume change as compared with the proportional change shown in clinical cutaneous area measurements. These parameters could potentially be used to determine when to stop propranolol therapy since—with maximal reduction in lesion volume and vessel density—the benefits of continued empiric propranolol therapy may be minimal. This information is potentially important in understanding how propranolol affects IH tissue and could allow us to better determine optimal dosing and therapeutic duration for this medication.

Currently, there is no direct evidence to explain propranolol’s effect on IH.⁸ Findings that propranolol therapy reduces IH volume and vessel density could be explained by a simple reduction in blood flow to the IH through vasoconstriction. It is also possible that propranolol therapy somehow triggers small-vessel pruning that reduces angiogenesis and blood flow through the lesion, ultimately causing larger caliber vessels to shrink. It may also be that propranolol simply accelerates the natural tendency of IH to shrink and “involute” through a mechanism unrelated to angiogenesis. Whatever the mechanism of propranolol’s effect on IH, the use of GCDU imaging opens an avenue to critically monitor and hopefully refine this therapy.

Using GCDU imaging in a systematic manner sheds light on controversies surrounding IH therapeutic response to propranolol. When evaluating proportional IH volume changes, as expected, there were no differences between male and female response. It is possible that the traditional IH treatment, corticosteroids, could potentiate propranolol’s effect in a synergistic manner. Our data do not support this idea. Our initial findings in this small series show no difference in proportional volume reduction when therapy was initiated after 1 year of age compared with patients treated under a year of age. This is different than what would be expected with corticosteroid treatment of IH.

This study is limited by a small sample size, which may have affected our ability to detect clinically meaningful differences between subgroups of IH patients. We were also limited by the absence of controls (patients with IH who did not receive propranolol treatment). This limited our ability to evaluate whether the changes we saw among our cohort were the direct result of propranolol treatment or merely reflected spontaneous shrinkage of the IH measured. However, IHs not treated with propranolol do not demonstrate the rapid shrinkage seen with propranolol treatment, so it is hard to believe that the changes we demonstrate would not be due to propranolol treatment. We are continuing to use GCDU imaging to evaluate our IH patients and to help us determine and measure treatment response. With this clinical practice, we will be better able to address this study’s limitations in the future.

Conclusion

Gray-scale color Doppler ultrasound imaging can be used to measure both the volume and vessel density of IHs. These parameters have the potential of objectively identifying optimal IH treatment duration and optimizing medical therapy while helping us understand how propranolol causes IH to shrink.

Author Contributions

Matthew M. Bingham, conception and design, acquisition of data, analysis and interpretation of data, drafting of article, final approval of article; Babette Saltzman, analysis and interpretation of data, critical revision of article, final approval of article; Nghia-Jack Vo, conception and design, acquisition of data, analysis and interpretation of data, critical revision of article, final approval of article; Jonathan A. Perkins, conception and design, acquisition of data, analysis and interpretation of data, drafting and critical revision of article, final approval of article.

Disclosures

Competing interests: None.

Sponsorships: None.

Funding source: This project was funded in part by the State of Washington ESHB 2687, section 606(28). The funding organization had no role in study design and conduct; collection, analysis, and interpretation of the data; and writing or approval of the manuscript.

Footnotes

Acknowledgements

Eden Palmer, figure preparation; Stacy Russ, manuscript preparation.

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

This article was presented as a poster at the 19th Annual Workshop on Vascular Anomalies, International Society for the Study of Vascular Anomalies (ISSVA); June 2012; Malmo, Sweden.

References

Leaute-Labreze

Dumas de la Roque

Hubiche

Boralevi

Thambo

Taieb

. Propranolol for severe hemangiomas of infancy. N Engl J Med. 2008;358:2649-2651.

Baetz

Eigelshoven

Marquard

Bruch-Gerharz

Homey

Meissner

. Infantile hemangioma: successful treatment with propranolol. Hautarzt. 2010;61:290-292.

Buckmiller

Munson

Dyamenahalli

Dai

Richter

. Propranolol for infantile hemangiomas: early experience at a tertiary vascular anomalies center. Laryngoscope. 2010;120:676-681.

Guldbakke

Rordam

Huldt-Nystrom

Hanssen

Hoivik

. Propanolol used in treatment of infantile hemangioma. Tidsskr Nor Laegeforen. 2010;130:1822-1824.

Sans

Roque

Berge

. Propranolol for severe infantile hemangiomas: follow-up report. Pediatrics. 2009;124:e423-e431.

Zaher

Rasheed

Hegazy

Abdelhalim

Gawdat

. Oral propranolol: an effective, safe treatment for infantile hemangiomas. Eur J Dermatol. 2011;21:558-563.

Tan

Itinteang

Leadbitter

. Low-dose propranolol for infantile haemangioma. J Plast Reconstr Aesthetic Surg. 2011;64:292-299.

Storch

Hoeger

. Propranolol for infantile haemangiomas: insights into the molecular mechanisms of action. Br J Dermatol. 2010;163:269-274.

Lamy

Lachambre

Lord-Dufour

Beliveau

. Propranolol suppresses angiogenesis in vitro: inhibition of proliferation, migration, and differentiation of endothelial cells. Vascul Pharmacol. 2010;53:200-208.

10.

Starkey

Shahidullah

. Propranolol for infantile haemangiomas: a review. Arch Dis Child. 2011;96:890-893.

11.

Rasalkar

Chu

Cheng

Lee

. An institutional review of paediatric haemangiomas: prevalence, imaging features, and outcomes. Hong Kong Med J. 2010;16:334-340.

12.

Gutzeit

Stuckmann

Tosoni

Erdin

Binkert

. A cavernous hemangioma of the thyroid gland: first documentation by ultrasound of a rare pathology. J Clin Ultrasound. 2011;39:172-174.

13.

Dubois

Alison

. Vascular anomalies: what a radiologist needs to know. Pediatr Radiol. 2010;40:895-905.

14.

Wortsman

. Clinical usefulness of variable-frequency ultrasound in localized lesions of the skin. J Am Acad Dermatol. 2010;62:247-256.

15.

Spalinger

Patriquin

Miron

. Doppler US in patients with Crohn disease: vessel density in the diseased bowel reflects disease activity. Radiology. 2000;217:787-791.

16.

Dubois

Patriquin

Garel

. Soft-tissue hemangiomas in infants and children: diagnosis using Doppler sonography. AJR Am J Roentgenol. 1998;171:247-252.

17.

Cushing

Boucek

Manning

Sidbury

Perkins

. Initial experience with a multidisciplinary strategy for initiation of propranolol therapy for infantile hemangiomas. Otolaryngol Head Neck Surg. 2011;144:78-84.

18.

Yetter

Acosta

Olson

Blundell

. Estimating splenic volume: sonographic measurements correlated with helical CT determination. AJR Am J Roentgenol. 2003;181:1615-1620.