Figure 1. Clinical features of conjunctival papilloma by age. A, Small sessile tarsal papilloma in a 9-year-old child. B, Multiple confluent papillomas of the fornix, caruncle, and tarsal conjunctiva in an 11-year-old child. C, Small sessile papilloma on the bulbar conjunctiva with a prominent episcleral feeder vessel in an adult (45 years old). D, Multiple forniceal papillomas in an adult (43 years old).
Figure 2. Conjunctival papillomas before and after treatment. Conjunctival papilloma involving the marginal palpebral conjunctiva (arrow in A) resolved completely (B) after treatment with double freeze-thaw cryotherapy. Caruncle papilloma (C) resolved completely (D) after treatment with oral cimetidine (300 mg 3 times a day) for 3 months. Caruncle papilloma (E) resolved completely (F) after a single session of photodynamic therapy. Caruncle and inferior fornix papilloma (G) was completely excised (H) with adjunctive cryotherapy; topical interferon alfa-2b was given for 3 months after excisional biopsy to prevent recurrence.
Kaliki S, Arepalli S, Shields CL, Klein K, Sun H, Hysenj E, Lally SE, Shields JA. Conjunctival PapillomaFeatures and Outcomes Based on Age at Initial Examination. JAMA Ophthalmol. 2013;131(5):585-593. doi:10.1001/jamaophthalmol.2013.83
Author Affiliations: Ocular Oncology Service, Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania (Drs Kaliki, C. L. Shields, Klein, Sun, Hysenj, Lally, and J. A. Shields and Ms Arepalli). Dr Kaliki is now with the Ocular Oncology Service, L. V. Prasad Eye Institute, Hyderabad, India, and Dr Sun is now with the Department of Ophthalmology, Shandong Provincial Hospital, Jinan, China.
Importance Conjunctival papilloma is a benign epithelial tumor occurring in both children and adults with varying clinical features and outcomes. In this article, we describe our experience regarding the difference in the clinical features and outcomes of conjunctival papilloma based on age at initial examination.
Objective To evaluate the clinical features, treatment, and outcomes in patients with conjunctival papilloma based on age at initial examination.
Design Retrospective study.
Setting Ocular Oncology Service, Wills Eye Hospital, Philadelphia, Pennsylvania.
Participants Ten children and adolescents (aged ≤20 years) and 63 adults (aged >20 years) with conjunctival papilloma.
Interventions Excisional biopsy, cryotherapy, oral cimetidine, topical or injection interferon alfa-2b, and photodynamic therapy.
Main Outcome Measure Tumor response.
Results A comparison of conjunctival papillomas between age groups revealed significant differences in the mean number of tumors per eye (children and adolescents vs adults, 2 vs 1; P = .05), tumor basal dimension (8 vs 6 mm; P = .05), and associated feeder vessels (20% vs 47%; P = .05). Primary treatment included sole treatment with oral cimetidine (15% vs 5%), topical interferon alfa-2b (0% vs 1%), cryotherapy (0% vs 3%), photodynamic therapy (0% vs 1%), excisional biopsy and cryotherapy (38% vs 65%), excisional biopsy and cryotherapy with adjuvant oral cimetidine (8% vs 9%), and excisional biopsy and cryotherapy with adjuvant topical or injection interferon alfa-2b (38% vs 15%). Significant differences in age groups in treatment outcome during the follow-up period (mean, 24 vs 38 months) included complete regression with single treatment (38% vs 95%; P < .01) and tumor recurrence (15% vs 1%; P = .05).
Conclusions and Relevance Conjunctival papillomas are larger and more likely to be multiple in children and adolescents than in adults. Excisional biopsy and cryotherapy with or without adjuvant oral cimetidine and/or topical interferon alfa-2b provide satisfactory tumor control. Papilloma recurrence is more common in children and adolescents than in adults.
Quiz Ref IDConjunctival papilloma is a benign epithelial tumor with prominent intrinsic vascularity that displays characteristic “hairpin” vascular loops.1- 3 These lesions most often show an exophytic (sessile or pedunculated) growth pattern but can exhibit a mixed or, rarely, an inverted growth pattern.4,5 A causal relationship between human papillomavirus (HPV) and conjunctival papilloma has been documented in 5% to 45% of cases.6- 9 Based on the association with cervical carcinogenesis, HPV has been divided into 2 types: high risk (types 16, 18, 33, and 31) and low risk (types 6 and 11).10 High-risk types cause uterine cervical cancer and a significant proportion of carcinomas in other sites, including the vulva, vagina, penis, anus, larynx, and tonsils.11,12 Low-risk types cause genital warts (condyloma acuminata), low-grade intraepithelial lesions of the uterine cervix, and, rarely, recurrent respiratory papillomatosis.10,11 Conjunctival papillomas are associated most commonly with low-risk types 6 and 11 (especially in children) and rarely with high-risk types 16, 18, and 33 (especially in adults).6- 9,13 Conjunctival papilloma in infants whose mother had documented genital HPV infection has led to the speculation that the virus could be inoculated into the newborn's conjunctiva during vaginal birth.14,15 In adults, the mechanism of infection is believed to be self-inoculation.16
In a pathology-based review of 1120 epibulbar tumors at the Armed Forces Institute of Pathology, conjunctival papilloma accounted for 1% of cases in children and adolescents (aged <20 years) and 11% in adults (aged >20 years).17 In a clinic-based study of 1643 conjunctival tumors, less than 1% of cases were classified as papilloma in children and adolescents and 1% as papilloma in adults.18 The incidence of conjunctival papilloma is highest among patients aged 20 to 39 years.17,19 In this article, we describe our experience with 73 consecutive patients with 108 conjunctival papillomas, exploring the features and outcomes of these cases based on age at initial examination.
A computerized database search was conducted for cases of conjunctival papilloma from April 1, 1970, through September 1, 2012, at the Ocular Oncology Service, Wills Eye Hospital, Philadelphia, Pennsylvania. Institutional review board approval was obtained. The medical records of all patients with a diagnosis of conjunctival papilloma were reviewed retrospectively, and those patients with a definitive clinical and/or histopathologic diagnosis of conjunctival papilloma were included in the study. The patients with conflicting clinical and histopathologic diagnosis were excluded. Patients were divided into 2 groups, including children and adolescents (aged ≤20 years) and adults (aged >20 years).
The demographic data obtained included age (in years), race (white, African American, Asian, or Hispanic), and sex. The presence and/or history of systemic or ocular disease, drug history, and presence of concomitant cutaneous papilloma were noted. Details of prior treatment were recorded.
Ocular details included laterality of conjunctival papilloma (unilateral or bilateral), eye affected (right or left), and number of tumors per eye. The following tumor details were recorded based on external and slitlamp biomicroscopic examination: tumor epicenter location (limbus, bulbar conjunctiva, fornix, tarsal conjunctiva, eyelid margin, punctum, plica semilunaris, or caruncle), clockface position of tumor epicenter (1- to 12-o’clock position), quadrantic epicenter location (superior, nasal, inferior, or temporal), tumor basal dimension and thickness (both in millimeters), tumor configuration (exophytic [sessile or pedunculated], inverted, or mixed), associated intrinsic pigment, intrinsic vascularity, and feeder vessels. All anterior segment findings were recorded on a large anterior segment drawing and anterior segment photographs. The details shown by anterior segment fluorescein angiography, anterior segment optical coherence tomography (Visante), and ultrasound biomicroscopy were recorded when available. Posterior segment examination was performed with indirect ophthalmoscopy, and related fundus findings were noted.
Details were recorded for the primary treatment of conjunctival papilloma, including cryotherapy, excisional biopsy and cryotherapy, topical and/or injection interferon alfa-2b, photodynamic therapy, oral cimetidine, or combinations of therapy. The outcome of each treatment modality (no tumor regression, partial regression, complete regression, or recurrence) was also recorded, as were the number of recurrences, the duration (in months) between treatment and recurrence, and the subsequent treatment of recurrent lesions.
When available, histopathologic confirmation of conjunctival papilloma was recorded; the histopathologic features included papillomatous configuration, intralesional vascular tufts, presence and severity of cellular dysplasia and/or atypia, acanthotic conjunctival epithelium, koilocytosis, and associated inflammation. The HPV test results and HPV type (type 6, 11, 13, 16, 18, 31, 33, 45, or 51) were recorded when available, as were the clinical features, outcome at date last seen, and total follow-up duration (in months).
Data were summarized as number and percentage based on age at initial examination. Continuous variables were analyzed with paired t tests and categorical data with Fisher exact tests. The cutoff for statistical significance was P < .05. Statistical analysis was performed with SPSS software, version 13.0 (SPSS).
Among 73 patients with conjunctival papilloma, there were 10 children and adolescents (≤20 years) and 63 adults (>20 years). Their mean age at initial examination was 43 years (median, 43 years; range, 4-85 years) (Table 1). Tumors were more common in male patients (46 [63%]) than in female patients (27 [37%]). A comparison of demographic features in children and adolescents vs adults showed nonsignificant differences in previous herpes simplex virus keratitis (0% vs 6%), cutaneous papilloma (10% vs 19%), genital warts (0% vs 5%), and tumor laterality (unilateral, 100% vs 97%; bilateral, 0% vs 3%).
Conjunctival papillomas were evaluated on the basis of sex and decade of initial examination (Table 2). The incidence of conjunctival papilloma was highest in the age group 31 to 40 years (23%). An evaluation by decade of appearance revealed no significant differences (P < .05) by sex.
The tumor features are described in Table 3 and displayed in Figure 1. A comparison of these features by age group (children and adolescents vs adults) showed significant differences (P < .05) in the mean number of tumors per eye (2 vs 1; P = .05), mean tumor size (8 vs 6 mm; P = .05), and associated feeder vessels (20% vs 47%; P = .05). The most common location of the tumor epicenter was the fornix in children and adolescents (27%) and the caruncle in adults (24%).
The results of treatment are listed in Table 4 and displayed in Figure 2. Primary treatment of 99 tumors (9 patients were unavailable for follow-up) included cryotherapy in 3 (3%), excisional biopsy and cryotherapy in 61 (62%), topical interferon alfa-2b in 1 (1%), photodynamic therapy in 1 (1%), oral cimetidine in 6 (6%), excisional biopsy and cryotherapy with adjuvant topical or injection interferon alfa-2b in 18 (18%), and adjuvant oral cimetidine in 9 (9%). There was no significant difference in method of primary treatment between age groups.
Significant differences in treatment outcome by age group (children and adolescents vs adults) during the follow-up period (mean, 24 vs 38 months; median, 12 vs 12 months) were seen, including complete regression with a single administration of any form of treatment (38% vs 95%; P < .01) and tumor recurrence (15% vs 1%;P = .05). Overall, complete regression with primary treatment was achieved in 87 patients (88%), and 12 (12%) had residual and/or recurrent tumor, requiring additional treatment. Overall, tumor recurrence was noted in 3 patients (3%) during a mean follow-up of 36 months. One patient (aged 9 years) with advanced papilloma had 4 recurrences after excisional biopsy, cryotherapy, and adjuvant injection of interferon alfa-2b (10 MU) and at this writing has been receiving oral cimetidine (300 mg 3 times a day) and topical interferon alfa-2b (1 MU 3 times a day) for more than 1 year, with minimal response. A second patient (aged 11 years; findings published elsewhere20) had 3 recurrences after excisional biopsy and cryotherapy, with failure to respond to 3 cycles of topical mitomycin C, but showed a dramatic response to oral cimetidine (300 mg 3 times a day), with complete regression within 4 months. A third patient (an adult) with 1 recurrence after excisional biopsy, cryotherapy, and adjuvant oral cimetidine responded well to 1 session of triple freeze-thaw cryotherapy and topical interferon alfa-2b (1 MU 3 times a day) for 1 year. The treatment details for each recurrence are listed in Table 4.
Histopathologic results were available in 74 of 99 treated tumors (9 in children and adolescents and 65 in adults). The diagnosis of conjunctival papilloma was confirmed in all cases. A comparison of histopathologic features by age (children and adolescents vs adults) showed no significant difference in focal epithelial atypia or dysplasia (4 [44%] vs 32 [49%]; P = >.99), koilocytosis (0 vs 2 [3%]; P = >.99), acanthotic epithelium (4 [44%] vs 25 [38%]; P = .73), or stromal inflammation (5 [56%] vs 26 [40%]; P = .49). The epithelial atypia was categorized as mild (3 [75%] vs 26 [81%]), moderate (1 [25%] vs 4 [12%]), or severe (0 vs 2 [6%]). In situ hybridization results were available in 3 patients (1 child [aged 11 years] and 3 adults), of whom 1 was positive for HPV 6 and 11.
In this study, Quiz Ref IDconjunctival papilloma was more common in adults than in children or adolescents at a ratio of 6:1. The incidence of conjunctival papilloma was highest among patients aged 21 to 40 years (34%), similar to findings in other studies.17,19,21 This is analogous to the peak age at onset of genital HPV infection in sexually active men and women.22,23 In our analysis, coexistent genital warts were reported in 3 patients, 2 of whom were aged 21 to 40 years. Furthermore, coexistent cutaneous papilloma was noted in 18%, a sign that HPV infection can be multifocal. Conjunctival papilloma is caused by HPV infection with type 6, 11, 16, 33, and 45, similar to types causing genital warts (types 2, 3, 6, 11, 16, 18, and 30-32) and dissimilar to those causing cutaneous papilloma (types 1-4 and 26-29).6- 9,24,25
Findings in other studies indicate that conjunctival papilloma has a male predominance (51%-60%).19,21 Similarly in our study, we found a male predominance in conjunctival papillomas in both children and adolescents (70%) and adults (62%). Papillomatosis in other sites of the body show a female predominance for genital (33% vs 67%) and cutaneous (49% vs 51%) papillomas and a male predominance for upper respiratory involvement (77% vs 23%).24,26- 28
Multicentric conjunctival papillomas were more common in children and adolescents than in adults, and the numerous lesions coalesced to form massive papillomatosis in 1 patient (Figure 1B). Ash17 studied 245 conjunctival papillomas in patients of all ages (range, 1-97 years) and found that the most common location of conjunctival papilloma was the bulbar conjunctiva (42%). Sjö et al19 studied 245 papillomas in patients aged 17 to 84 years and found the most common site to be the palpebral conjunctiva (38%). In our study of patients of all ages (range, 4-85 years), the most common site was the caruncle (23%), but we noted a difference based on age at initial examination. Childhood papillomas were more commonly located in the inferior fornix (27%) and adult papillomas in the caruncle (24%). Factors influencing this pattern of distribution could include eye rubbing and tear flow movement from the superotemporal fornix to the lacrimal lake in the medial and inferior conjunctiva, causing autoinoculation of HPV.19 Furthermore, the lesions were larger in children and adolescents (8 mm) than in adults (6 mm), presumably because tumors in the inferior fornix were hidden, leading to more delayed diagnoses in younger patients.
Conjunctival papilloma most commonly has an exophytic growth pattern.4 Inverted papillomas of the conjunctiva are rare.4 In a pathology-based study of 245 lesions, all papillomas were of the exophytic type with no inverted papillomas.19 In our study, 99% of cases were exophytic, with 1 inverted papilloma. Quiz Ref IDAlthough inverted papillomas are benign, this variant has greater tendency for malignant transformation into transitional cell carcinoma, squamous cell carcinoma, or mucoepidermoid carcinoma.5,29
Quiz Ref IDThe various treatment options for conjunctival papilloma include surgical excision, cryotherapy, carbon dioxide laser, oral dinitrochlorobenzene, oral cimetidine, topical mitomycin C, and topical and/or intralesional interferon alfa-2b.20,30- 35 No-touch surgical excision and adjunctive double freeze-thaw cryotherapy are the preferred methods of treatment.3 In our series, the most common treatment modality was surgical excision with adjunctive cryotherapy. Photodynamic therapy has been described elsewhere for treatment of conjunctival squamous cell carcinoma and vascular tumors.36,37 In our series, photodynamic therapy was used in 1 case of conjunctival papilloma, and dramatic tumor regression was noted (Figure 2E and 2F).
The recurrence rate in conjunctival papilloma is 6% to 27%.17,19 Incomplete tumor excision can theoretically allow liberation of virus particles into the surrounding tissues, resulting in tumor recurrence with aggressive behavior.3 In our series, the recurrence rate was 3%, but recurrence was more common in children and adolescents than in adults. However, it should be noted that 10% of patients were referred to us with recurrent papilloma after prior treatment elsewhere. In recent years, Quiz Ref IDoral cimetidine and interferon alfa-2b have been beneficial for pharmacologic control of conjunctival papillomas.20,33- 35 Currently, we use oral cimetidine (300-400 mg 3 times a day) and/or topical interferon alfa-2b (1 MU 3 times a day) for 3 months after excisional biopsy to prevent recurrence. This treatment is especially useful in patients with local or systemic immunosupression and those with multiple and/or large tumors.
Histopathology shows that conjunctival papillomas are composed of multiple fronds or fingerlike projections of conjunctival epithelium surrounding cores of fibrovascular tissue.38 Sessile papillomas (more common in adults) can exhibit various degrees of epithelial dysplasia, but pedunculated papillomas (more common in children) usually do not exhibit epithelial dysplasia.13 Foci of dysplastic epithelium are evident in 6% to 20% of cases, but carcinoma rarely develops in conjunctival papillomas.16,19 The degree of dysplasia is graded as mild (confined to the lower third of the conjunctival epithelium), moderate (involving the lower and middle thirds), or severe (involving the entire thickness of the epithelium).13 In our series, dysplastic epithelium was evident in 49% of tumors (including 44% in children and adolescents) and was mild to moderate in 98%. Koilocytosis is the morphologic landmark of HPV infection in the uterine cervix,39 but its sensitivity as an indicator of HPV in conjunctival papilloma is low. Histopathologic evaluation of 55 conjunctival papillomas revealed the presence of koilocytosis in 40% of cases.8 In our series, koilocytosis was evident in only 3% of tumors.
In summary, conjunctival papillomas are more common in adults than in children or adolescents. Childhood conjunctival papillomas are often larger and multicentric compared with those in adults. No-touch surgical excision and adjunctive double freeze-thaw cryotherapy are the preferred methods of treatment. Recurrence can be minimized by complete tumor excision, cryotherapy, and adjunctive oral cimetidine and/or interferon alfa-2b.
Submitted for Publication: October 7, 2012; final revision received December 15, 2012; accepted December 22, 2012.
Published Online: March 28, 2013. doi:10.1001/jamaophthalmol.2013.83
Correspondence: Carol L. Shields, MD, Ocular Oncology Service, 840 Walnut St, Ste 1440, Philadelphia, PA 19107 (email@example.com).
Author Contributions:Study concept and design: Kaliki, C. L. Shields, Klein, and Sun. Acquisition of data: Arepalli, Klein, Sun, and Hysenj. Analysis and interpretation off data: Kaliki, Klein, Sun, Hysenj, and J. A. Shields. Drafting of the manuscript: Kaliki, Arepalli, Klein, and Sun. Critical revision of the manuscript for important intellectual content: C. L. Shields, Sun, Hysenj, Lally, and J. A. Shields. Statistical analysis: Hysenj. Administrative, technical, and material support: Kaliki, Arepalli, C. L. Shields, Klein, Sun, and Lally. Study supervision: J. A. Shields.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported by the Eye Tumor Research Foundation, Philadelphia, Pennsylvania, Mellon Charitable Giving from the Martha W. Rogers Charitable Trust, and the Lucille Wiedman Fund for Pediatric Eye Cancer, Philadelphia (all to Drs C. L. Shields and J. A. Shields).
Previous Presentation: Presented at the Association for Research in Vision and Ophthalmology (ARVO); May 9, 2012; Ft Lauderdale, Florida.