Dr Cheng's new affiliation is with the National University Health System, Singapore.
Propranolol in the management of periorbital infantile haemangioma
Article first published online: 11 JUN 2010
© 2010 The Authors. Journal compilation © 2010 Royal Australian and New Zealand College of Ophthalmologists
Clinical & Experimental Ophthalmology
Volume 38, Issue 6, pages 547–553, August 2010
How to Cite
Cheng, J. F., Gole, G. A. and Sullivan, T. J. (2010), Propranolol in the management of periorbital infantile haemangioma. Clinical & Experimental Ophthalmology, 38: 547–553. doi: 10.1111/j.1442-9071.2010.02344.x
- Issue published online: 5 AUG 2010
- Article first published online: 11 JUN 2010
- Received 26 January 2010; accepted 25 April 2010.
- capillary haemangioma;
Background: Infantile haemangiomas are the commonest tumours of the orbit in children. Treatment is usually expectant, unless they are visually threatening. Although steroids, other pharmacological and surgical treatment modalities have their place, there are risks involved. A previous case series reported the successful use of propranolol for infantile haemangioma. The safety and efficacy of propranolol in the treatment of periorbital haemangioma was reviewed in a serious of our patients.
Methods: We performed a retrospective review of patients seen by two ophthalmologists (TJS and GAG), collecting data on colour, size of lesion, duration of treatment and side-effects of treatment. Our main outcome measures were colour and size of infantile haemangioma before and after treatment, the change in astigmatism of our patients and the incidence of complications from propranolol.
Results: We reviewed 10 patients with infantile haemangioma. They were treated with propranolol oral syrup 2 mg/kg/day in divided doses for a mean duration of 32.8 (range 12–42) weeks. All our patients had a reduction in colour and size of the lesions. The mean lesion size decreased from 756.7 to 543.2 mm2 after treatment (P = 0.075). Five patients had significant astigmatism and 60% had successful reduction of astigmatism after treatment. None of our patients suffered significant side-effects of propranolol.
Conclusion: Propranolol appears to be a safe and effective treatment in the management of infantile haemangioma.
Infantile haemangiomas are the commonest tumours of the orbit in children. They are benign vascular lesions, classically presenting within the first month of life and progressively increasing in size until the child reaches about 6 months of age, remain stable for a few years, after which the lesions usually undergo some degree of spontaneous involution.1,2
In the periorbital region, infantile haemangioma can cause vision deprivation, astigmatism and strabismus.3–5 Treatment is usually expectant, unless vision is threatened. Interventions in the past included oral or intralesional steroids, α-interferon, vincristine, cyclophosphamide, lasers, embolization or surgical excision.5–11 Although these treatment modalities have their place, there are risks in using steroids and other pharmacological agents in young children. Surgery and other interventions also have risks. There have been recent reports of the successful use of propranolol as a relatively safe modality in the management of haemangiomas.12
We present a case series of 10 children who had been treated with propranolol at a dose of 2 mg/kg/day.
This is a retrospective chart review of patients with potentially vision-threatening infantile haemangiomas seen by two ophthalmologists and who commenced treatment between December 2008 and May 2009. Institutional Review Board and Ethics Committee approval was obtained (HREC/09/QRCH/100).
We collected data on vision, refractive error, location, size and colour of the lesions. The history, duration of lesions and presence of other lesions were also noted. Patients were assessed by a paediatrician for any contraindications to propranolol before treatment. If fit, they were started on propranolol oral syrup 2 mg/kg/day in divided doses. Throughout treatment, patients were monitored by their paediatrician for complications of propranolol especially hypotension, hypoglycaemia and asthma.
Main outcome measures were lesion size, colour and astigmatism. Mean lesion size was determined by measuring the surface area clinically or from imaging in the plane of largest dimension. A t-test was performed for lesion size and astigmatism.
There were 10 patients with infantile haemangioma (Table 1). There were seven female and three male patients. The mean duration of treatment was 32.8 (range 12–42) weeks. Seven out of 10 patients are still having ongoing treatment with propranolol.
|Patient||Gender||Age haemangioma first noted||Age at presentation (months)||Age at start of treatment||Laterality||Location of lesion||Visual compromise||Duration of treatment (weeks)||Other comments|
|1||Female||6 months||11||11||Right||Inferior medial orbit with deep orbital component||Nil||38|
|2||Female||2 weeks||4||10||Left||Medial upper eyelid||Nil||42||Stopped treatment at 42 weeks|
|3||Male||2 months||3||23||Left||Lower lid with orbital extension||Nil||42||Had previous surgery but further proliferation of lesion. Stopped treatment at 42 weeks|
|4||Female||3 weeks||1||2||Left||Upper eyelid||Nil||36|
|5||Female||3 weeks||1||1||Right||Lower lid extending to nose and cheek||Astigmatism||34||Had oral steroid before treatment with propranolol|
|6||Male||3 months||7||7||Left||Medial upper eyelid||Astigmatism||36||Had oral steroid before treatment but developed cushingoid symptoms|
|7||Female||6 weeks||2||2||Right||Lower lid extending to cheek||Amblyopia and astigmatism||36|
|8||Female||2 weeks||3||12||Right||Medial canthal region||Nil||24|
|9||Female||3 months||7||7||Right||Sub-brow region||Astigmatism||12||Stopped treatment at 12 weeks after surgery|
|10||Male||3 days||3||3||Right||Right upper lid||Nil||28|
Seven patients had propranolol as the initial treatment, three had prior treatments of surgery or steroids. Patients 6 and 7 had oral steroids and patient 3 had surgery before treatment with propranolol.
Steroids were discontinued for patient 6 because the lesion had continued to increase in size. Patient 7 developed hypertension and left ventricular hypertrophy after 4 months treatment with prednisolone, which had to be discontinued. Both patients showed improvement after starting oral propranolol.
Patient 3 had initial partial excision because the lesion was adjacent to vital structures. The residual lesion had rapidly re-grown and propranolol was later instituted.
The pre- and post-treatment lesion size, colour and refraction are summarized in Table 2.
|Patient||Size of lesion (mm2)||Refraction||Visual compromise||Final outcome|
|Before treatment||After treatment||Before treatment||After treatment|
|1||148||0||Significant astigmatism but unable to record||R +0.75/−3.00 × 25 L +0.75||Astigmatism||Decrease in size and colour|
|2||170||55||R +1.50 L +1.50||Unchanged||Nil||Decrease in size and colour|
|3||1625||518||R +0.25 L +0.25||Unchanged||Nil||Decrease in size and colour|
|4||392||126||R +5.75 L +5.75||R +3.75/1.0 × 90 L +3.75/+1.00 × 90||Nil||Decrease in size and colour|
|5||3561||3203||R −1.75/+2.50 × 90 L +0.25||R +0.75/−1.50 × 120 L +0.75/−1.50 × 30||Astigmatism||Decrease in size and colour and reduce astigmatism|
|6||113||28||R +4.50/−1.00 × 90 L +6.75/−3.50 × 55||R +2.25 L +3.25/−1.00 × 70||Astigmatism||Decrease in size and colour and astigmatism|
|7||942||940||R +3.50/−2.00 × 50 L +1.25||R +2.75/−1.00 × 90 L +2.75/−1.00 × 90||Amblyopia and astigmatism||Reduced in size and colour and astigmatism|
|8||209||179||R +1.00 L +1.00||R +1.00/−0.25 × 180 L +1.00/−0.25 × 180||Nil||Initial reduction in size and increased after stopping therapy; decreased in colour|
|9||289||301||R −1.75/+3.50 × 120 L +0.75||R +1.75/−3.50 × 30 L +0.75||Astigmatism||Initial decrease in size but increased after 3 months; no change in astigmatism; complete surgical excision|
|10||118||82||R +2.25 L +2.25||R +3.25 L +3.25||Nil||Decrease in size and colour|
All patients demonstrated a change in colour of the overlying skin from deep purple to a less purple colour or normal skin colour, often commencing within 24 h after commencement of treatment and a decrease in size of the lesions. The area of the lesions based on their two largest dimensions was calculated. We were unable to measure the volume of the lesions as there was inadequate data for some patients. The mean pre-treatment lesion size was 756.7 mm2, mean post-treatment lesion size was 543.2 mm2. This difference was not statistically significant (P = 0.075). Two patients, 8 and 9, developed compliance issues after commencing treatment their haemangiomas returned to original size but did not continue to enlarge.
Five patients demonstrated significant astigmatism (>1.5 dioptres [D]). The mean pre-treatment astigmatism was 2.8 D and post-treatment astigmatism was 1.9 D. This difference was not statistically significant. Three patients (5, 6 and 7) had significant reduction in the degree of astigmatism. Patient 1 required spectacle correction for her residual astigmatism. Because of significant astigmatism (3.5 D), patient 9 underwent surgery to excise the mass.
Patient 1 presented to the paediatric eye clinic at 10 months of age with a haemangioma in the right medial lower lid, which was gradually getting larger. On examination, there was a 27 mm × 12 mm purple lesion in the right medial lower lid (Fig. 1). Magnetic resonance imaging scans showed a large vascular lesion, 30 mm in diameter (Fig. 2). She was started on propranolol at 11 months of age and after treatment, the periorbital infantile haemangioma was no longer visible and there was a marked change in the size of the lesion as seen on the magnetic resonance imaging.
Patient 9 had persistent astigmatism. She presented at 7 months with a lesion in the right sub-brow region that had gradually increased in size since 3 months of age. This lesion measured 23 mm by 16 mm. There was also evidence of astigmatism with cycloplegic refraction of R +1.75/−3.50 × 120, L +0.75 D. There was an initial response with a decrease in size and colour of the lesion to 19 mm by 12 mm by 8 weeks of treatment. However, in a later review at 12 weeks, the lesion had started to increase in size again to 24 mm by 16 mm. There was also no reduction in the astigmatism. She had subsequent successful surgical excision of the infantile haemangioma (Fig. 3). The astigmatism 2 weeks after surgery had reduced to R +1.25/−1.50 × 40, L +0.75 D.
All of our patients were initially assessed by a paediatrician for contraindications to propranolol and monitored by their paediatrician for any side-effects throughout treatment. None of the patients were reported to have developed side-effect especially those of asthma, hypotension or hypoglycaemia.
The successful use of propranolol by Leaute-Labreze et al. has revolutionized the management of infantile haemangioma.12 They reported two children who were on propranolol for cardiac problems, but who coincidentally also had facial infantile haemangiomas. These had significantly reduced in size after the propranolol was started. They then successfully treated nine other children with facial infantile haemangiomas using propranolol.
In our study, 10 children with visually significant infantile haemangiomas were treated with propranolol with reduction in size of the lesions. However, we were unable to demonstrate statistical significance as our numbers were small. All our patients had decrease in colour of the lesions. This dramatically improved the aesthetic appearance for our patients. There was also a reduction in the astigmatism in 60% of our patients with astigmatism. This is comparable with the 63% reduction when using intralesional corticosteroid injection by Weiss et al.4 Our data confirms the French experience that propranolol is an effective treatment for infantile haemangioma with reduction in size, colour and astigmatism in our patients.
Haemangioma may present in other cutaneous structures besides the periocular area. They may be life-threatening when they occur within the airway presenting as subglottic or laryngotracheal haemangiomas. There have been case reports of resolution of stridulous respiratory obstruction with the use of propranolol in these patients.13–15
Despite these successes, there have been some concerns about the side-effects of propranolol on these infants.16 Lawley et al. reported two very young children, an 8-week-old and a 36-day female, who were started on propranolol at a dose of 2.5 and 2 mg/kg/day, and who developed hypotension and hypoglycaemia, respectively. These normalized without any specific intervention.17 Propranolol has been used to treat cardiac conditions in children safely for over 40 years. Doses as high as 1200 mg a day, accidentally ingested in a 3-year-old child, had been reported with no adverse outcomes.18 All our patients was started on a dose of 2 mg/kg/day in two or three divided doses. None of our patients demonstrated hypotension, hypoglycaemia or asthma.
Factors that may influence the safe use of propranolol in children include patient's age, history of prematurity and adequate nutrition and hydration.17,18 A regimen of starting propranolol with 0.5 mg/kg/day and doubling the dose every 3 days until the dose reaches 2 mg/kg/day have also been suggested.16,17 However, we feel that with appropriate paediatric assessment and selected monitoring of patients, starting propranolol at a dose of 2 mg/kg/day is safe. Only one of our patients was younger than 2 months of age when propranolol was commenced and none had reported side-effects when propranolol was started at 2 mg/kg/day. We feel that patients who were premature, having problems with feeding and hydration and patients who were under 2 months of age may require inpatient monitoring on commencement of propranolol and an incremental dosing regimen may be used.
Further studies regarding the minimal effective dose, duration of use and long-term side-effects will enable us to better define its role in infantile haemangioma management. Investigation into the mechanisms of action of propranolol may also lead to a better understanding of the pathogenesis of these lesions and in turn the development of more effective treatments.
Propranolol appears to be effective and safe in treating infantile haemangiomas. Periorbital haemangiomas show a good response to propranolol with no evidence of severe side-effects in our patients.
We would like to thank the staff at the ophthalmology clinic of the Royal Children's Hospital and Kris Kamusinski at Herston Multimedia for all their help and support.
- 1Vascular tumors, malformations and degeneration. In: TasmanW, JaegerEA, eds. Duane's Clinical Ophthalmology, Vol. 2, revised edition. Philadelphia, PA: J.B. Lippincott, 1992; 1–6., .
- 2Vascular disease. In: TaylorD, HoytCS, eds. Pediatric Ophthalmology and Strabismus, 3rd edn. Philadelphia, PA: Elsevier Saunders, 2004; 385–90., .