During the proliferative phase, the most common complication was ulceration of the hemangioma (14.7% of patients). Median age of ulceration was 6 weeks (range, 2–48 weeks). Fourteen of the 15 cases were treated with conservative wound care; 1 case required urgent operative intervention due to bleeding from the superior labial artery. Four patients with ulceration (26.7%) subsequently went on to require debulking surgery at a later stage. Hemangiomas with a cutaneous component (superficial and mixed types) had a significantly higher rate of ulceration as compared to lesions without a cutaneous component (deep type; 15/81 versus 0/21, respectively). Of the 15 cases that were complicated by ulceration, 14 (93.3%) involved mixed depth lesions (Fig. 2). Lesions that involved the vermillion border also had significantly higher rates of ulceration (11/40 versus 4/62, respectively).
Ten patients (9.8%) reported feeding difficulties (with breast or bottle feeding), and 12 patients (11.8%) reported bleeding from the lesion; all of these cases involved ulcerated lesions. Two patients with feeding difficulties developed failure to thrive and required admission to hospital for nasogastric tube feeding while the ulcerated lesion healed. No other cases required hospital admission or nasogastric feeding. No patients had documented airway involvement or reported issues with speech delay.
During the involutional phase, the most common complication was residual lip contour deformity as identified by the attending surgeon during clinic follow-up visits (occurring in 26.5% of patients). All of these cases involved noticeable excess tissue bulk and residual fibrofatty tissue at the site of the involuted hemangioma. Residual lip contour deformity was associated with: hemangiomas having a deep component1 (either mixed depth or exclusively deep) and hemangiomas that involved the vermillion border2 (Table 3). No other lesion or patient characteristics were associated with residual lip contour deformity.
Hemangiomas that involved the vermillion border had significantly higher rates of residual lip contour deformity as compared with those that did not cross the vermillion border (40.0% versus 17.7%; χ2 = 6.19, P = 0.013). All patients with residual lip contour deformity had hemangiomas classified as having a deep component (mixed or exclusively deep); none of the patients with exclusively superficial lesions demonstrated this complication. A total of 29 patients had hemangiomas with both vermillion border involvement and a deep component; the majority (55%) of these patients experienced residual lip contour deformity.
Medical Treatment (Proliferative Phase)
A total of 45.1% of patients were treated expectantly with close observation by clinic pediatricians (Table 4). The most common medical treatment was administration of an oral beta-blocker (propranolol or atenolol), used in 37.2% of all cases (used in 51% of cases since 2008, when beta-blockers first became available for treatment of hemangiomas at our institution). Indications for beta-blocker treatment in the proliferative phase included: anatomically sensitive location (100% of cases), concern for growth (55% of cases), age of child (47% of cases), and evidence of ulceration (19% of cases). The median age of beta-blocker initiation was 3.8 months (range, 1–21 months). The average propranolol dose was 1.6 mg/kg/d for a mean duration of 13 months. There were no reported side effects to beta-blocker therapy.
Although it was not possible to objectively determine the effect of oral beta-blockers on hemangioma size, resolution of ulceration, or impact on cosmetic appearance, the prescribing pediatricians noted improvement in these domains for most patients. However, there was no difference in surgical rates between those treated and not treated with oral beta-blockers (32.1% versus 23.0%; χ2 = 2.27, P = 0.13).
Surgical Intervention (Involutional Phase)
All 27 patients with identified residual lip contour deformity were offered debulking/corrective surgery; 20 patients underwent surgery at some point during the involutional phase. Seven of the 27 patients (25.9%) who were offered surgical intervention for excess tissue bulk declined intervention. Specific reasons for declining surgery were not identified.
Lesion depth was the primary factor associated with the need for surgical intervention. None of the patients with exclusively superficial lesions were taken to the operating room (Fig. 3); however, 28.6% of patients with a deep component to their hemangiomas underwent surgical treatment. All of the patients who declined surgery also had a deep component to their lesion (6 mixed depth hemangiomas, and 1 exclusively deep lesion).
All operations occurred during the involutional phase; the median age of first surgery was 52 months (range, 18–150 months). In all but one case, the primary indication for surgery was residual lip contour deformity requiring debulking of excess tissue (the single outlier was an ulcerated lesion that eroded into the labial artery, resulting in arterial bleeding). All surgical patients underwent debulking of the lesion with primary closure through advancement flaps as part of their operation (Fig. 4). Other than the single patient with a bleeding lesion requiring intervention, all operative cases were elective outpatient procedures. Six patients required more than 1 operation. Two patients had very large lesions that required multiple debulking procedures; 2 patients required scar revisions; and 2 patients required dehiscence repairs. The remaining 14 patients underwent a single operation.
Infantile hemangiomas located in anatomically sensitive areas of the body pose a particular challenge for clinicians. Lesions involving the lip exemplify this issue, as there are both functional and cosmetic concerns to address. The 2 most common complications arising from these lesions are ulceration in the proliferative phase and residual lip contour deformity in the involutional phase. The morphology of individual hemangiomas seems to predispose lesions to particular complications. Consequently, the specific physical characteristics of lip hemangiomas may be used to predict which lesions will be at highest risk for ulceration or contour deformity.
During the proliferative phase, development of an ulceration is the most common and concerning complication.6 In our dataset, the age of initial ulceration was quite young (median age, 6 weeks), indicating that these infants are often presenting to care with an ulcerated hemangioma. Unfortunately, early development of an ulceration often precludes preventative treatment of this complication. Consequently, the majority of ulcerated hemangiomas are treated in a reactive manner. Only 1 patient required urgent surgical intervention (due to acute labial artery bleeding); the remaining 14 patients were managed nonoperatively with oral medication (steroids, beta-blockers, or both), wound care, analgesia, and nutritional support as needed.
Infants with ulcerated hemangiomas had high rates of feeding difficulties, and 2 patients developed clinically significant failure to thrive. These patients eventually required urgent admission to hospital for supplemental nasogastric tube feeding to regain weight and allow the lesion to heal. This highlights that ulcerated lip hemangiomas are poorly tolerated in young infants, and clinicians must be aware of the possibility of feeding difficulty. Given the high risk for failure to thrive, all patients with ulcerated hemangiomas were monitored closely by our clinic pediatricians for weight gain while the lesion healed.
All 15 hemangiomas that ulcerated were either superficial or mixed depth (having both superficial and deep components); none of the exclusively deep hemangiomas were complicated by ulceration. This seems to indicate that development of ulceration requires the presence a cutaneous component to the hemangioma. Mixed depth lesions were at the highest risk for ulceration, perhaps due to the combination of cutaneous involvement and displacement of deeper lip architecture, contributing to stress on the overlying skin and predisposing the lesion to incidental trauma.
The primary physical characteristic associated with residual lip contour deformity and need for surgery was the depth of the hemangioma. None of the exclusively superficial hemangiomas required surgical debulking; however, among patients with mixed depth or deep lesions, over one-quarter (28.6%) underwent debulking surgery. Furthermore, all 7 patients who were offered surgery (and subsequently declined) also had a deep component to their lesion. This stark difference highlights the impact of hemangioma depth on the risk of developing a contour deformity. Although hemangiomas with a deep component do involute significantly over time, there is often residual bulky tissue or excess skin/fibrofatty tissue.1 Thus, despite involution, deep lesions often require operative intervention to restore adequate lip contour.
Involvement of the vermillion border was also associated with high rates of lip contour deformity. Lesions that cross the vermillion are more likely to distort lip architecture and symmetry. This is significant, as even very small degrees of asymmetry are noticeable at conversational distance. Patients with hemangiomas, who had a deep component in conjunction with vermillion border involvement, were at the highest risk of developing significant contour deformity. Given their association with surgical intervention, these 2 lesion characteristics may be of considerable use to surgeons in providing parents with a prognosis with respect to the possibility of requiring surgery.
Optimal timing for surgical intervention for problematic lip hemangiomas is controversial. Several authors have suggested that early resection during the proliferative phase is beneficial.3,7,8 However, this approach may result in higher rates of unnecessary procedures and may increase the degree of lip resection required. Early intervention ameliorates the opportunity for further involution of the hemangioma and forces parents to make an earlier decision than necessary regarding surgery. Our data indicate that there is a subset of families who decline surgery despite the identification of lip contour distortion by their surgeon. Furthermore, ongoing research continues to raise concerns about the neurodevelopmental status of children below 3 years old undergoing general anesthesia for elective procedures.9,10 Delaying intervention until 4–6 years old reduces this risk in addition to allowing further involution (potentially reducing the size of required resection). Consequently, our practice is to delay surgery to allow involution of the lesion and potentially avoid the need to intervene completely.
Importantly, the primary indication for debulking surgery is aesthetic. There does not seem to be a functional limitation to excess lip bulk or tissue (none of the patients reported functional issues related to these findings). All parents of patients who underwent surgical debulking reported concerns with lip appearance. Although none of the patients reported mental health issues, the potential for teasing and social ridicule during the school years was a frequently cited concern among parents.
During the past decade, the nonselective beta-blocker propranolol has become standard treatment to induce involution in infantile hemangiomas.11–13 In this dataset, there were no differences in ulceration rate or surgical intervention rate based on treatment with propranolol. As noted above, ulceration seems to occur very early with lip hemangiomas, reducing the role for propranolol in preventing this complication. However, this drug may still have a role in limiting the extent of the ulceration (and was often prescribed with this indication). Regardless of ulceration, treatment with propranolol did not affect the rate of developing lip contour deformity. Although propranolol is postulated to initiate involution, the presence of a deep component often results in significant residual bulk after involution, necessitating surgical intervention.
This study has several limitations related to its retrospective nature. Lip hemangiomas are relatively rare, requiring a long period of study to collect a significant sample size. Consequently, several significant changes in management practice have occurred during this period that may have influenced the results, most importantly the introduction of propranolol in 2008. Loss to follow-up rate is unclear, as patients were not specifically contacted if they were not continuing to be seen in our clinic, leaving us to assume that they did not have unrecognized complications or require further intervention. Given that some of the younger patients in the study may yet go on to have surgery, we may be underestimating the true rate of surgical treatment for lip hemangiomas at our institution.
Although we were unable to quantitatively assess tumor size, we suspect that the size and volume of lip hemangiomas have an impact on prognosis. Hemangiomas with larger volumes may be at higher risk of having residual contour deformities following the involutional phase as such lesions likely have a greater degree of residual fibrofatty tissue. It is possible this may confound our results due to the inability to include this factor in our analysis. Further work may focus on elucidating the relationship between tumor volume and aesthetic outcomes in lip hemangiomas.
Based on our experience and data regarding infantile hemangiomas of the lips, we have made the following recommendations:
- (1) Given the high risk of ulceration in the proliferative phase, all patients with hemangiomas of the lips should have early assessment in a specialized, multidisciplinary clinic with expertise in vascular birthmarks.
- (2) Individualized treatment of lip hemangiomas using medical therapy should include early treatment of lip hemangiomas with deep components, feeding difficulties, or other interventions based on team/family preferences.
- (3) All patients with ulcerated hemangiomas are at high risk for feeding difficulties and should be monitored closely for failure to thrive by a pediatrician.
- (4) During the involutional phase, hemangiomas with a deep component are at the highest risk of persistent lip contour deformities. The surgeon should have iterative discussions with the patient and their family regarding the need for debulking procedures.
The authors would like to thank the Department of Surgery, University of Calgary, for funding support through the Calgary Surgical Research Development Fund.
Ethical approval for this study was obtained by the University of Calgary Conjoint Health Research Ethics Board.
1. Chang LC, Haggstrom AN, Drolet BA, et al.; Hemangioma Investigator Group. Growth characteristics of infantile hemangiomas: implications for management. Pediatrics. 2008;122:360–367.
2. O TM, Scheuermann-Poley C, Tan M, et al. Distribution, clinical characteristics, and surgical treatment of lip infantile hemangiomas. JAMA Facial Plast Surg. 2013;15:292–304.
3. Li WY, Chaudhry O, Reinisch JF. Guide to early surgical management of lip hemangiomas based on our experience of 214 cases. Plast Reconstr Surg. 2011;128:1117–1124.
4. Sharma VK, Fraulin FO, Harrop AR, et al. The opportunities and obstacles in developing a vascular birthmark database for clinical and research use. Can J Plast Surg. 2011;19:122–124.
5. Fraulin FO, Flannigan RK, Sharma VK, et al. The epidemiological profile of the Vascular Birthmark Clinic at the Alberta Children’s Hospital. Can J Plast Surg. 2012;20:67–70.
6. Kim HJ, Colombo M, Frieden IJ. Ulcerated hemangiomas: clinical characteristics and response to therapy. J Am Acad Dermatol. 2001;44:962–972.
7. Hynes S, Narasimhan K, Courtemanche DJ, et al. Complicated infantile hemangioma of the lip: outcomes of early versus late resection. Plast Reconstr Surg. 2013;131:373e–379e.
8. Goldenberg DC, Hiraki PY, Marques TM, et al. Surgical treatment of facial infantile hemangiomas: an analysis based on tumor characteristics and outcomes. Plast Reconstr Surg. 2016;137:1221–1231.
9. Andropoulos DB, Greene MF. Anesthesia and developing brains—implications of the FDA warning. N Engl J Med. 2017;376:905–907.
10. Ho AM, Fleming ML, Mizubuti GB. Anesthetic neurotoxicity and the developing brain. CMAJ. 2017;189:E1028–E1029.
11. Léauté-Labrèze C, Dumas de la Roque E, Hubiche T, et al. Propranolol for severe hemangiomas of infancy. N Engl J Med. 2008;358:2649–2651.
12. Léauté-Labrèze C, Hoeger P, Mazereeuw-Hautier J, et al. A randomized, controlled trial of oral propranolol in infantile hemangioma. N Engl J Med. 2015;372:735–746.
Copyright © 2019 The Authors. Published by Wolters Kluwer Health, Inc. on behalf of The American Society of Plastic Surgeons.
13. Chinnadurai S, Fonnesbeck C, Snyder KM, et al. Pharmacologic interventions for infantile hemangioma: a meta-analysis. Pediatrics. 2016;137:e20153896.