September 2010

Effect of a Single Application of Pulsed Dye Laser Treatment of Port-wine Birthmarks on Intraocular Pressure

Author Affiliations

Author Affiliations: Departments of Internal Medicine (Dr Quan), Neurology (Dr Comi and Ms Irving), Pediatrics (Drs Comi and Krakowski), and Dermatology (Dr Cohen) and The Wilmer Ophthalmological Institute (Dr Parsa), Johns Hopkins University School of Medicine, Baltimore, Maryland; and Division of Pediatric Dermatology, Rady Children's Hospital, San Diego, and Department of Dermatology, University of California, San Diego (Dr Krakowski).


Copyright 2010 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2010

Arch Dermatol. 2010;146(9):1015-1018. doi:10.1001/archdermatol.2010.223

Background  A new pathophysiologic mechanism has been proposed that indicates that periorbital port-wine birthmarks (PWBs) serve as alternate collateral blood passageways when orbital venous drainage is impaired. The occlusion of such collateral venous channels could, therefore, potentially exacerbate impaired ocular venous flow and trigger the development or worsening of glaucoma in patients with Sturge-Weber syndrome. We investigated to what extent a single application of laser therapy, which occludes only the most superficial portions of a facial PWB, might affect intraocular pressure. Pressures before and after laser treatment were measured to determine pressure difference in 15 patients receiving laser treatment.

Observations  The greatest pressure differences were observed in patients with a PWB closest to the eye (P = .02). Posttreatment pressures were significantly decreased, relative to pretreatment pressures, only in patients with a PWB on the eyelid compared with patients with a facial PWB not near the eyes (2.33 vs 0.75 mm Hg; P = .004). No correlation was found between change in pressure and patient age, PWB size, or number of previous treatments.

Conclusions  A single laser application to a PWB does not appear to show a clinically relevant change in intraocular pressure. Further study is needed longitudinally in a broad range of patients.

Yellow flashlamp-pumped pulsed tunable dye laser (PDL) therapy has become the standard of care for port-wine birthmarks (PWBs) (also known as port-wine stains) in the United States and internationally since the approval of this device by the Food and Drug Administration in 1984.1 Laser therapy of PWBs and related vascular malformations may reduce the risk of complications, which include increasing vascular ectasia, blood-filled blebs, bleeding, ulceration, and infection.2 Eradication of PWBs early in life has also been shown to improve body image and associated psychosocial problems.3

Although the most effective age for initiating therapy has not been established, many children begin laser treatment during the first 6 months of life, particularly for facial PWBs.1,4,5 A new hypothesis proposes a role for periocular PWBs serving as a collateral blood passageway for orbital venous flow that is impaired to the cavernous sinus.6 We report the results of a retrospective single-center study designed to determine to what extent a single application of laser treatment to facial PWBs might increase intraocular pressure.


This was a single-center, retrospective study involving patients with PWBs who were seen for PDL treatment at the Johns Hopkins Outpatient Center, or later at the Johns Hopkins Harriet Lane Clinic, Baltimore, Maryland, from July 31, 2006, to August 28, 2006, and July 2, 2007, to July 30, 2007. The same PDL was used (V-Star Pulsed Dye Laser; Cynosure Inc, Westford, Massachusetts) for all subjects in this study. Patients were eligible for the study if they had a facial PWB and had their eye pressures checked before and after treatment as part of patient care. Patients were included regardless of age, history of treatments, or whether they had a diagnosis of Sturge-Weber syndrome (SWS). Institutional review board approval was given for this study.

Patients typically applied topical anesthetic ointment (lidocaine, 15%; prilocaine, 5%; and phenylephrine, 0.25%) to their PWB 2 hours before laser treatment. Once the patient was in the laser treatment room, 2 procainamide anesthetic eye drops were placed in each eye. With the patient sitting upright (so as to simulate pressure readings as taken in an ophthalmologist's office), a handheld automated applanation tonometer (Tonopen; Mentor Ophthalmics Inc, Norwell, Massachusetts) was used to take 3 pretreatment intraocular pressure readings for each eye. Measurements were recorded, and all had a less than 5% error variance as indicated by the device's digital readout. Protective shielding was then placed over the patient's eyes. All 3 groups received similar protective shielding during the laser treatments. An episcleral eye shield or metal contact lens was not used on any of the patients. In the case of an eyelid PWB, an external eye shield was used for a facial PWB, and laser therapy was performed by shifting the eye shield out of the field of laser treatment. The patient underwent PDL treatment of the PWB, with all laser specifications recorded. Our usual pulse duration is set at 2 milliseconds (ms) for patients with type 1 or 2 skin and slightly longer with increasing Fitzpatrick skin types but not usually longer than 6 or 10 ms. Since most of our treated patients have type 1 or 2 skin, the majority of our subjects were treated at 2 ms. Immediately following laser treatment, with the patient again sitting upright and the protective shielding removed, the applanation tonometer was used to measure another 3 posttreatment pressure readings in each eye, all with less than 5% error variance. Depending on the size of the PWB and the length of time that had passed between the first pressure check and the second one, anesthetic eye drops may have been reapplied. Pressure measurements were taken by the same individual (S.Y.Q.) both before and after treatment.

We created a deidentified database for review by collecting data from the medical records of each visit. Data collected included the following: age of patient at time of visit; characteristics of the PWB including size, location, color, and appearance; the total energy applied during laser treatment; and number of previous laser treatments. Because the data sets were not normally distributed, nonparametric analyses were used and the data was analyzed with SPSS 15.0 software (SPSS Inc, Chicago, Illinois). A total of 15 subjects (or 30 eyes) were studied. A database was set up with both mean and lowest pressure difference measured (posttreatment minus pretreatment pressure; a negative number means that the pressure was decreased after laser treatment relative to the pretreatment pressure). The location of the PWB was designated into the following 3 groups in terms of location in relation to each eye: (1) involving the eyelid of the eye being measured; (2) near the eyes but not involving the eyelid of the eye being measured (defined as the border of the PWB closest to the eye being within 2 cm of the lid margin but not involving the eyelid itself); and (3) on the face but not near the eyes.


Data from 15 subjects (10 female; median age, 16 years [range, 5-73 years]) were collected (Table). The mean standard deviation of the pressure measurements (3 measurements per eye) was 1.3 mm Hg (range, 0-2.5 mm Hg) before treatment and 1.5 mm Hg (range, 0-4 mm Hg) after treatment. The results did not differ whether the mean pressures or lowest pressures were used for the calculations; the difference in mean pressures is reported herein.

Image not available
Data Recorded From Each Patient Visit

Kruskal-Wallis analysis of all 30 eyes showed a significant effect of PWB location on the change in mean intraocular pressures, post–laser treatment minus pre–laser treatment measurements (P =.02; Figure 1). The closer the eye being measured is to the treated PWB, the greater the impact on the difference in intraocular pressure.

Figure 1.
Image not available

Median intraocular pressure difference (post–laser treatment minus pre–laser treatment measurements). Error bars show range. The eyes with a port-wine birthmark (PWB) on the eyelid showed a median drop in intraocular pressure that was significant compared with the eyes with a PWB not near them (P < .01).

In the eyes with a PWB on the eyelid (n = 8), the posttreatment intraocular pressure was decreased relative to the pre–laser treatment pressure by a median amount of 2.33 mm Hg (range, 0.67 to –7.0 mm Hg). In eyes where the PWB was located on the face but not near eyes (n = 14), the median pressure difference was only 0.75 mm Hg (range, 1.67 to –0.67 mm Hg), and this was significantly different compared with the eyes having the PWB on the eyelid (P = .004; Figure 1). For eyes where the PWB was located near the eyes but did not involve the eyelid of the eye being measured (n = 8), there was a median post–laser treatment pressure difference of −1.17 mm Hg (range, 4.0 to –5.33 mm Hg). This group was not significantly different from either of the other 2 groups (vs the group of eyes with the PWB involving the eyelid [P = .32] and vs the group of eyes with the PWB on face but not near eyes [P = .051]).

We then looked at only the 8 subjects with a PWB involving the eyelid. The group of eyes with a PWB involving the eyelid did not have an intraocular pressure difference (post–laser treatment minus pre–laser treatment measurements) that was significantly different from the group of contralateral eyes that did not have a PWB involving the eyelid (P = .29; Figure 2).

Figure 2.
Image not available

Intraocular pressure differences (post–laser treatment minus pre–laser treatment measurements) for the 8 subjects who had a port-wine birthmark (PWB) on the eyelid of 1 eye. The lines connect the pairs of eyes from each of the 8 subjects.

No correlation was found in the overall group between the change in intraocular pressure and either the age of the patient (Spearman ρ = 0.043; P = .82), the size of the PWB (Spearman ρ = 0.179; P = .34), or number of previous treatments (Spearman ρ = −0.076; P = .69).


The classic manifestations of SWS include a cutaneous PWB, ipsilateral ocular involvement often resulting in glaucoma, and radiologic evidence of brain abnormalities consisting of leptomeningeal enhancement and changes in venous structures.7 A new hypothesis for the pathophysiologic mechanism for SWS traces this entire constellation of findings to the absence of superficial cortical or bridging veins as the primary insult.6 Bentson and colleagues8 first demonstrated via simultaneous bilateral carotid angiography in patients with SWS that centrifugal cerebral venous flow is impeded by a relative lack of superficial cortical veins draining into the dural sinuses. These investigators also proposed that impairment of flow and pooling of blood within the leptomeninges may produce leptomeningeal thickening. The lack of a patent centrifugal venous pathway subsequently results in increased centripetal drainage. Bentson and colleagues8 also demonstrated this enhanced deep venous drainage system in the form of dilated and tortuous deep medullary and subependymal cerebral veins. Parsa6 proposed that the increase in deep venous drainage would also raise cavernous sinus pressure. This, in turn, would reduce venous drainage of the orbital veins, including the ophthalmic and facial veins, causing orbital and periorbital venous ectasia and the characteristic upper facial PWBs.6 Because cortical and orbital veins are all emissary (valveless), permitting bidirectional blood flow, it was hypothesized that obliteration of the superficial PWB by laser treatment may result in a reduction in alternative cerebral venous outflow channels including an exacerbation of ocular hypertension. Specifically, it was proposed that treatment of periocular and upper facial PWBs might obliterate cutaneous vessels draining the orbit and increase ipsilateral intraocular pressure.6

The data presented herein, however, does not show that a single laser treatment raises intraocular pressure in older children and adults who are not laser treatment naive. Somewhat unexpectedly, our data demonstrated lower, rather than higher, intraocular pressures after treatment in the group with eyelid lesions compared with the patients with a PWB not near the eye. This small but statistically significant decrease in intraocular pressure after a single laser treatment of PWB involving the eyelids could hypothetically be attributed to (1) the true treatment effect of a single laser treatment on intraocular pressure immediately after procedure, at least transiently, (2) the effect of anesthetic applied locally to the skin around the eye on intraocular pressure, (3) an effect of some other manipulation of the eye not adequately controlled for in our study, or (4) a compensatory vasodilatation at some other site transiently improving vascular outflow from the eye. We do not see any biological plausibility for a true sustained intraocular pressure-lowering effect. It is also unlikely that the topical anesthetic compound, which contains both vasoconstrictive and vasodilatory agents, applied 2 hours prior to laser treatment would account for a drop in pressure over just a few minutes between the 2 pressure measurements. There has been one study of the effects of topical anesthesia on intraocular pressure measured by applanation tonometry and the effect was negligible.9 However, the mandatory protective eye shielding, together with the digital manipulations necessary, in particular, when treating eyelid lesions, would cause aqueous egress and transiently lower intraocular pressure.

There are several limitations to this study. Measuring several intraocular pressures in quick succession with the attendant pressure on the cornea can lead to increased aqueous outflow and a lowered intraocular pressure. However, the repeated intraocular pressure measurements were consistent in all 3 groups and therefore would not be expected to result in the decrease in intraocular pressure seen in the group receiving laser treatment around the eye but not in the group receiving laser treatment to the face but not near the eyes. Nevertheless, no controls were included who did not receive treatment, and these would be helpful to include in future studies. Furthermore, we cannot know from this study what the impact on intraocular pressure is days or weeks after the treatment. In most cases a single laser treatment results in only partial lightening of the PWB because it only penetrates approximately 1 mm and the meaningful collateral channels may be deeper. Therefore, further studies are needed to study the impact of several laser treatments on intraocular pressure.

Ultimately, the small number of patients presented herein with a single superficial laser application does not allow us to draw decisive conclusions about the potential for more extended laser treatments to have an impact on intraocular pressure over time and ensure safety of the procedure with respect to ocular blood flow and development. We advocate the need for further study (1) in patients with SWS, (2) in infants and young children, (3) in laser treatment–naive patients, and (4) with multiple intraocular pressure measurements over time and after multiple laser treatments when more significant areas and depths of the ectatic venous channels are occluded. We are currently conducting an ongoing retrospective study evaluating the ophthalmologic records of children with SWS during courses of laser treatment that should provide additional information to guide treatment.

Back to top
Article Information

Correspondence: Bernard A. Cohen, MD, Division of Pediatric Dermatology, Johns Hopkins University School of Medicine, David Rubenstein Child Health Building, Room 2103, 200 N Wolfe St, Baltimore, MD 21287 (bcohena@jhmi.edu).

Accepted for Publication: March 1, 2010.

Author Contributions: Drs Cohen, Comi, and Parsa had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Comi, Parsa, Krakowski, and Cohen. Acquisition of data: Quan, Krakowski, and Cohen. Analysis and interpretation of data: Comi, Irving, and Cohen. Drafting of the manuscript: Quan, Irving, and Cohen. Critical revision of the manuscript for important intellectual content: Comi, Parsa, Krakowski, and Cohen. Statistical analysis: Comi. Administrative, technical, and material support: Irving. Study supervision: Krakowski and Cohen.

Financial Disclosure: None reported.

Additional Contributions: Lauren Nelson assisted in manuscript editing. We thank Medtronic Ophthalmics for donating the tonopen to the JHU pediatric dermatology laser clinic.

Tan  OTSherwood  KGilchrest  BA Treatment of children with port-wine stains using the flashlamp-pulsed tunable dye laser. N Engl J Med 1989;320 (7) 416- 421
Stier  MFGlick  SAHirsch  RJ Laser treatment of pediatric vascular lesions: port wine stains and hemangiomas. J Am Acad Dermatol 2008;58 (2) 261- 285
Troilius  AWrangsjö  BLjunggren  B Patients with port-wine stains and their psychosocial reactions after photothermolytic treatment. Dermatol Surg 2000;26 (3) 190- 196
van der Horst  CMKoster  PHde Borgie  CABossuyt  PMvan Gemert  MJ Effect of the timing of treatment of port-wine stains with the flash-lamp-pumped pulsed-dye laser. N Engl J Med 1998;338 (15) 1028- 1033
Ashinoff  RGeronemus  RG Flashlamp-pumped pulsed dye laser for port-wine stains in infancy: earlier versus later treatment. J Am Acad Dermatol 1991;24 (3) 467- 472
Parsa  CF Sturge-weber syndrome: a unified pathophysiologic mechanism. Curr Treat Options Neurol 2008;10 (1) 47- 54
Bodensteiner  JBedRoach  ESed Sturge-Weber Syndrome.  Mt Freedom, NJ Sturge-Weber Foundation1999;
Bentson  JRWilson  GHNewton  TH Cerebral venous drainage pattern of the Sturge-Weber syndrome. Radiology 1971;101 (1) 111- 118
Ehongo  ADe Maertelaer  VPourjavan  S Effect of topical corneal anaesthesia on ocular response analyzer parameters: pilot study. Int Ophthalmol 2009;29 (5) 325- 328