Retinal Manifestations of Incontinentia Pigmenti: A Case Series of 14 Patients Highlighting the Importance of Intravenous Fluorescein Angiography and the Benefits of Early Laser Photocoagulation

Abstract

Purpose:

This case series describes the nature and frequency of retinal manifestations in patients with incontinentia pigmenti (IP).

Methods:

This is a retrospective single-center case series of all known patients with IP who presented to Associated Retina Consultants (Phoenix, AZ) between May 2016 and April 2019. Twenty-eight eyes of 14 patients with a dermatologic diagnosis of IP were included (n = 28). Most patients underwent examination under anesthesia with fundus photographs and intravenous fluorescein angiography (IVFA).

Results:

Of the 28 eyes, 8 (28.6%) had abnormal retinal findings on fundus examination. Of the 26 eyes that had IVFA, 10 (38.5%) had abnormal findings: Seven eyes (26.9%) had peripheral ischemia, 2 (7.7%) had previous peripheral laser scarring, and 2 (7.7%) had active peripheral neovascularization. Three eyes with normal examination results were found to have mild ischemia by IVFA. Patients with ischemia confirmed by IVFA were treated with laser photocoagulation. During follow-up, 4 previously treated eyes received additional laser photocoagulation. No patients showed vision loss, vitreous hemorrhage, retinal detachment, or adverse effects of treatment. No patients required vitreoretinal surgery.

Conclusions:

IP is a potentially blinding disease. Our case series demonstrates the efficacy of early treatment and the importance of ancillary testing with IVFA and fundus photography.

Introduction

Incontinentia pigmenti (IP) is a rare X-linked dominant syndrome, first described by Garrod in 1906 as a syndrome with peculiar pigmentation of the skin. ¹ The incidence of this condition is noted to be between 0.2 in 100 000 and 1 in 50 000. ² The gene IKBKG (inhibitor of the kappa light polypeptide gene enhancer in B cells, kinase gamma), which is localized on chromosome Xq28, is deleted in 80% of IP cases and is the primary causative gene for this condition. ^3,4 The disease is usually lethal in men, but has a higher female survival rate owing to lyonization, with female incidence of IP between 90% and 97%. ⁵ The condition is usually diagnosed clinically based on skin findings with confirmatory genetic testing and/or skin biopsy. IKBKG plays an important role in the nuclear factor (NF)–κB pathway, protecting cells against apoptosis from tumor necrosis factor α. An IKBKG gene mutation results in a truncated NF-κB protein that is unable to protect against cell death, causing several dermatologic, ophthalmic, and neurological clinical manifestations.

Dermatologic findings of IP are characterized by distinct skin lesions that follow the lines of Blaschko, reflecting embryonic neuroectoderm migration patterns; the rash evolves into hypopigmented and hyperpigmented macules. ⁶ Retinal findings can be explained by examining the role of NF-κB protein in protection from capillary endothelial cell apoptosis. IKBKG gene mutations result in subsequent cell death and lead to microvascular complications including retinal ischemia and neovascularization. ⁷ The reported retinal findings range from simple retinal hemorrhages to potentially blinding complications from retinal ischemia, vascular occlusion, neovascularization, and retinal detachment (RD). ⁵ The purpose of this study is to describe our retinal findings and experiences in a large private-practice setting of patients diagnosed with IP.

Methods

This was a retrospective case series of patients diagnosed with IP who were referred to physicians at Associated Retina Consultants (ARC) in Phoenix, Arizona, for ophthalmic examination. Patients were evaluated between May 2016 and April 2019. All children who were examined were diagnosed by a dermatologist prior to retina examination. Fourteen patients were evaluated in this study, which included all patients referred to ARC with a diagnosis of IP. Twelve of the patients had examinations under anesthesia (EUA). Of the 2 who did not have an EUA, 1 was a 15-year-old child who was examined adequately in clinic, and the other was an infant who was lost to follow-up after the initial clinic visit. Fundus examination of the infant was deemed adequate, but intravenous fluorescein angiogram (IVFA) was not performed. Each EUA was performed after informed consent was obtained from the parents or legal guardians. The examination consisted of indirect ophthalmoscopy using 20- and 28-diopter lenses, color fundus photographs, and IVFA using the RetCam (Natus Newborn Care).

If treatment was deemed necessary, laser photocoagulation was performed during the same anesthesia session. Laser photocoagulation was performed in a dense scatter pattern to areas of ischemic retina, identified by IVFA, regardless of the presence or absence of retinal neovascularization. Laser was performed using either an argon (532 nm) or diode laser (810 nm), depending on physician preference. Retinoscopy was not performed during the EUA. Four of the patients had additional EUAs after a range of 3 to 12 months from the initial EUA; 2 patients had a total of 3 EUAs with a follow-up range of up to 3 years. Follow-up interval was determined by findings at the time of the EUA: If no pathology was discovered, the recommended follow-up interval was 1 year; if vascular pathology was discovered and treated, the recommended follow-up interval was 2 to 6 months depending on disease severity. Criteria for re-treatment with laser during the follow-up EUA was persistent/additional regions of ischemia that were not previously treated with laser and/or persistent or new active neovascularization.

Data collected included visual acuity (age-dependent), intraocular pressure, anterior examination findings noted with a slitlamp or penlight examination, fundus examination findings, FA results, treatment performed (if any), laterality and symmetry of the condition, and the number of EUAs needed for each patient.

Dermatologic information regarding the patients in the case series was collected through a retrospective review of medical records. Patients were diagnosed by the characteristic skin rash, usually combined with a skin biopsy to confirm the IP diagnosis. Genetic counseling was recommended for all patients, but the majority refused genetic testing because of financial issues.

Results

We evaluated the ophthalmic findings of 14 patients and 28 eyes examined at ARC. Thirteen patients were female and 1 was male. The man’s survival was suspected to be due to somatic mosaicism. Demographic information is summarized in Table 1. The average age of the patients was 1.88 ± 3.8 years. The patients were followed for an average of 7.8 ± 11.6 months.

Table 1.

Summary of Patient Information.

	No. of patients (%) (n = 14)
Sex
Female	13 (92.9)
Male	1 (7.1)
Birth history
Normal	11 (78.5)
Premature	3 (21.5)
Medical history
Seizures	0 (0)
Intraocular pressure, mm Hg
Normal (range, 9-22)	14 (100)
Visual acuity
Normal	13 (92.9)
Abnormal	1 (7.1)
Anterior examination
Normal	14 (100)

Three of the 14 patients were born prematurely. Of the patients born prematurely, 1 was born at 36 weeks’ gestation with a birth weight of 2835 grams, 1 was born at 36 weeks’ gestation with an unknown birth weight, and 1 was born at 34 weeks’ gestation with a birth weight of 2381 grams. The diagnosis of IP was made by skin biopsy in 2 of these 3 patients, and there was a strong family history with suggestive skin findings in the last; therefore, retinopathy of prematurity (ROP) was not suspected in these patients.

None of the patients had other medical problems, and none experienced seizures. Intraocular pressure was normal in all patients. Visual-acuity examination was limited because of the age of the patients. One 5-year-old patient had subnormal vision (20/70 OD and 20/30 OS). The anterior segment examination findings in all eyes were normal. Six of the 14 patients had at least 1 affected eye with fundus and/or IVFA findings. Of the 28 eyes, 20 (71.4%) had normal retina examination results, and 8 (28.6%) had abnormal retinal findings. Of the abnormal fundoscopic findings, 2 had scars indicative of previous peripheral laser photocoagulation, 1 had visible peripheral neovascularization, 1 had nonspecific peripheral retinal pigment epithelium disruption, and 4 had peripheral ischemia (with 3 having retinal hemorrhages). By IVFA, 10 of the 26 eyes (38.4%) had abnormal findings. Seven eyes (26.9%) had peripheral retinal nonperfusion, 2 (7.7%) had active neovascularization, and 2 (7.7%) had evidence of previous laser photocoagulation but no active disease. Of note, 3 eyes with normal fundus examination results were found to have mild peripheral ischemia by IVFA, and conversely, 1 patient with peripheral intraretinal hemorrhages on examination did not show any peripheral ischemia by IVFA. Two of the 14 patients had normal retinal examination findings in 1 eye and peripheral retinal ischemia or neovascularization requiring treatment in the other eye. Four of the 14 patients had bilateral disease, and 2 of these patients had highly asymmetric disease. Images of the more severe retinal abnormalities are shown in Figures 1 to 3.

Figure 1.

Severe bilateral disease.

Figure 2.

Asymmetric disease.

Figure 3.

Quiescent disease post laser treatment.

Patients with peripheral ischemia confirmed by IVFA were treated with laser photocoagulation. During the course of follow-up, 4 previously treated eyes were treated with additional laser treatment, and 2 required a third session. No patients showed evidence of vision loss, vitreous hemorrhage, RD, or adverse effects of treatment. No patients required surgical intervention with pars plana vitrectomy. The results are summarized in Table 2.

Table 2.

Summary of Retinal Findings.

	No. of eyes (%)
Fundus examination
Normal	20/28 (71.4)
Abnormal	8/28 (28.6)
Bilateral disease	4/28 (14.2)
Asymmetric disease	2/28 (7.1)
Unilateral disease	4/28 (14.2)
IVFA
Normal	16/26a (61.5)
Abnormal	10/26 (38.5)
Bilateral disease	8/26 (30.1)
Asymmetric disease	4/26 (15.4)
Unilateral disease	2/26 (7.7)
Peripheral ischemia ± intraretinal hemorrhages	7/26 (26.9)
Previous peripheral laser photocoagulation	2/26 (7.7)
Peripheral neovascularization	2/26 (7.7)
Peripheral nonspecific RPE disruption	1/26 (3.8)

Abbreviations: IVFA, intravenous fluorescein angiography; RPE, retinal pigment epithelium.

^a One patient did not have an IVFA.

From a dermatologic standpoint, 85.7% had characteristic vesicular or hyperpigmented rashes following Blaschko lines consistent with IP. All patients were diagnosed with IP by a dermatologist prior to retinal examination. Photographs of dermatologic abnormalities are shown in Figure 4A and 4B. The male patient had negative results from genetic testing but was referred out for mosaic chromosomal analysis and somatic IKBKG gene testing from the skin. Most patients had confirmatory biopsy results, 2 had a family history of IP, and the majority refused genetic testing because of financial issues.

Figure 4.

(A) Lower extremity and (B) upper extremity incontinentia pigmenti rash.

Conclusions

IP, also known as Bloch-Sulzberger syndrome, is a rare genetic disease with multiple organ findings, most commonly skin changes. Reported retinal manifestations of IP include RD, retinal pigment epithelium abnormalities, and various retinal vascular abnormalities (eg, hemorrhages, neovascularization, or avascular peripheral retina). To our knowledge, few other studies have examined the differences in retinal findings by using fundus examination alone or combined with IVFA. In our series, 3 eyes with a normal fundus examination were found to have mild peripheral retinal ischemia by IVFA, and conversely, 1 patient with peripheral intraretinal hemorrhages on examination did not show any peripheral ischemia by IVFA. We believed that IVFA was needed to visualize the full extent of disease and determine the appropriateness of laser photocoagulation, and our approach was to treat all peripheral nonperfused retinas with laser ablation regardless of the presence or absence of neovascularization to reduce the long-term risk of vision loss from RD.

Our series of 14 patients showed abnormal findings in 28% of eyes by examination and 38% of eyes by IVFA. The differences in abnormal findings seen with fundus examinations and IVFA highlighted the need for IVFA to truly characterize disease activity and determine the need for laser treatment. Some patients had unilateral disease, some bilateral, and some asymmetric disease, indicating there was no specific laterality pattern like that seen in familial exudative vitreoretinopathy. Some eyes required more than 1 session of laser photocoagulation, indicating the need for more than 1 EUA and IVFA to monitor disease progression and treatment response. In our series, 2 eyes required a total of 2 sessions of laser, and 2 eyes had a total of 3 sessions. The decision for re-treatment was based on finding new (or previously missed) areas of peripheral nonperfusion and/or active retinal neovascularization.

Of note, in contrast to other case series in the literature, none of our patients developed advanced disease manifestations such as vitreous hemorrhage or RD. The reason for this finding was uncertain but it could be related to the early diagnosis of IP and prompt referral by the dermatology department as well as our aggressive treatment approach. Additionally, none of the patients in our series had seizures or other serious neurological problems, which might explain why the retinal findings were mild in our series. It is unknown whether the presence of neurological disease increases the incidence of more advanced retinal disease, but Chen et al had discovered a trend toward RD in their series of patients with neurologic disease. ⁸

Peng and colleagues ⁹ published the largest IP case series in the known literature of a sample of 61 children referred to Xin Hua Hospital in China. Forty-seven of their 61 patients (77%) demonstrated abnormal retinal findings—28 with bilateral disease and 19 with unilateral anomalies. They found a range of abnormalities that were classified into several stages: stage 1 for eyes with retinal pigment epithelial changes only; stage 2 for eyes with retinal vascular abnormalities; stage 3 for epiretinal membranes or fibrotic hyperplasia with avascular zones, retinal neovascularization, and/or vitreous hemorrhage; stage 4a for partial RD; stage 4b for total RD, and stage 5 for phthisis bulbi and secondary glaucoma. Thirty-three of 122 studied eyes (27%) were found to have stage 4 or 5 disease, unlike our case series. The authors did admit that they might have had a referral bias, primarily receiving referrals for the most challenging cases. Rates of retinal involvement in other case series of IP previously have ranged from 36% to 77%. ^{10 -12}

Chen et al ⁸ documented the largest cohort of patients with IP with long-term follow-up in the United States at the Wilmer Eye Institute (Baltimore, Maryland). They examined 50 eyes of 25 female participants between 1976 and 2013 who met the clinical criteria for IP. The patients had follow-up for at least 6 months. All but 2 participants had a full evaluation with IVFA following fundus examination on at least 1 visit. Of the 46 eyes studied with IVFA, 25 (54.3%) had peripheral nonperfusion on initial examination and 19 (41.3%) had evidence of retinal neovascularization on initial examination. Of the 50 eyes in the study, 10 (20%) were treated with prophylactic ablation to the nonperfused retina: One eye was treated with cryotherapy, and 9 eyes were treated with laser photocoagulation. Ablative treatment in their series was reserved for patients who had documented progression of neovascularization, development of vitreous traction, or vitreous hemorrhage. Of note, 11 eyes (22%) developed RD, with a range of follow-up from 0.5 to 22.8 years. A bimodal age pattern of RD was identified; tractional RD occurred typically in the patients younger than 2.5 years, and rhegmatogenous RD developed later in life at age 14 years or older. ⁸

Huang and colleagues ¹³ published a systematic review of 388 articles and 42 full-text peer-reviewed human studies containing 100 eyes of 60 unique patients with IP retinopathy. They found that 38 (63%) patients had bilateral eye involvement, and of the 100 affected eyes, 31 (31%) eyes developed RD. They created an algorithm based on initial examination severity: grade 0—no retinal abnormalities, grade 1—peripheral avascular retina, grade 2—presence of neovascularization or hemorrhage, grade 3—RD, and grade 4—chronic total RD and cicatricial disease. ¹³ Our findings seemed to corroborate the findings of their review, with an emphasis on the importance of IVFA to assess abnormal retinal vasculature and an association with better outcomes with early treatment of avascular retina.

Huang et al ¹³ have stated that it is paramount for all patients diagnosed with IP be evaluated by an ophthalmologist and have EUA with IVFA as soon as safely possible, regardless of the clinical examination findings, because IVFA can detect vascular changes that can otherwise be missed on fundus examination. They recommended frequent reevaluation in the first 2 years, as they have found that early detection and treatment are associated with better visual outcomes and mitigation of disease progression.

Because peripheral vascular pathology in IP is similar to that seen in ROP, laser photocoagulation and/or antivascular endothelial growth factor (anti-VEGF) treatment are logical approaches to the problem, but the optimal timing of treatment is unknown. Laser treatment, however, is not without risk; that is why Chen et al have advocated a more conservative approach, reserving treatment for cases of documented neovascularization progression, developing vitreous traction, or vitreous hemorrhage. ⁸ Our approach has been to treat any nonperfused peripheral retina with laser regardless of the presence or absence of neovascularization, and our early results have been promising so far. Similar to our methodology, Huang and colleagues erred on the side of early treatment if areas of nonperfusion were identified. ¹³ Unfortunately, there are no randomized clinical trials to our knowledge to guide the timing of laser treatment in cases of IP retinopathy.

Although we did not use anti-VEGF injections as a treatment in our series, it is reasonable to believe anti-VEGF medications could be used as monotherapy or adjunctive therapy in cases of IP retinopathy, secondary to reported efficacy of anti-VEGF treatment in similar pathologies such as ROP. The first report of anti-VEGF treatment for IP retinopathy showed good regression of neovascularization but a poor visual outcome. ¹⁴ Additional case reports have been published since then, with anti-VEGF medications used as monotherapy or adjunctive therapy with promising results. However, there is insufficient evidence in the literature and no randomized trial data to clarify the role of the anti-VEGF treatment in patients with IP retinopathy. ¹³

Limitations of our study include a lack of long-term follow-up and a sample size of only 14 patients. Another limitation is that most patients had only 1 EUA, and 1 patient had no IVFA performed. Further limitations include the young age of the patients, limiting long-term assessment of vision outcomes, and the bias resulting from a retrospective analysis. Finally, we had limited genetic test results, but every patient was examined and diagnosed with IP by the dermatology department, and most were confirmed through skin biopsies.

In conclusion, IP is a potentially blinding disease secondary to neovascularization and RD, but the disease can be managed with early diagnosis and treatment. Based on our case series, a favorable prognosis is reasonable to expect from a vitreoretinal standpoint with early detection, treatment, and good follow-up. The literature regarding IP retinopathy is limited overall, and our case series adds valuable insight from another 14 patients. Our case series demonstrates the importance of ancillary testing with IVFA and fundus photography to properly characterize retinal abnormalities, determine eyes appropriate for treatment with laser, and to evaluate treatment response. Ultimately, patients with IP retinopathy are considered at long-term increased risk of RD, and ongoing, regular-interval follow-up is recommended for patients with IP and retinal pathology.