Incidence and Prevalence of Recurrent Respiratory Papillomatosis among Children in Atlanta and Seattle

Abstract

The incidence and prevalence of recurrent respiratory papillomatosis (RRP) for children aged <18 years were estimated in 2 US cities, Atlanta and Seattle, in 1996. All otolaryngologists in a 24-county area in metropolitan Atlanta (101 physicians) and an 8-county area in metropolitan Seattle (139 physicians) agreed to participate in the study. Medical record chart abstraction was performed only for children with documented current residence in the study area (21 patients in Atlanta and 14 patients in Seattle). The incidence rate for juvenile RRP was 1.11/100,000 population in Atlanta and 0.36/100,000 in Seattle. The prevalence rate was 2.59/100,000 population in Atlanta and 1.69/100,000 in Seattle. In neither city did prevalences differ significantly when stratified by sex or race. Extrapolation of these estimates to the US population suggests that 80–1500 incident cases and 700–3000 prevalent cases of juvenile RRP will occur in the United States during 1999.

Recurrent respiratory papillomatosis (RRP) is caused by human papillomavirus (HPV) infection and is characterized by recurring papillomas in the respiratory tract. Pediatric RRP is most often diagnosed in children aged 2–3 years [1] and is presumed to occur by transmission from infected mothers to their infants [2]. Frequent recurrence of disease and resultant surgery cause chronic inflammation, permanent damage to the vocal cords, and significant childhood morbidity. Children with RRP have an average of 4 surgical treatments per year to remove lesions, and some children require surgery every 2–3 weeks [1]. Deaths occur when papillomas obstruct the airway or when they spread to the lung parenchyma. From 1% to 3% of children with RRP die as a result of papilloma spread to the lung parenchyma.

No accurate data on the incidence and prevalence of pediatric RRP are available. The earliest estimate was determined on the basis of the results of a 1976 survey of 4200 otolaryngologists around the United States [3]. Fifty-one percent responded and reported 1500 new cases of RRP, which reflected an estimated incidence of 0.7/100,000. Fifty-six percent of RRP patients were aged ≤16 years, and the estimated incidence for this population was 0.4/100,000. A 1995 study queried 1346 otolaryngologists, with a 23% response rate [4]. The authors extrapolated patient numbers to 100% physician participation and reported an RRP incidence rate of 4.3/100,000 in children aged <14 years, a 10-fold difference from the 1976 estimate. The only other available estimates of pediatric RRP are from 2 Danish studies that reported annual incidence rates of 0.4/100,000 for children aged <20 years [5] and 0.6/100,000 for children aged <14 years [6].

We conducted this population-based study to estimate the prevalence and incidence of pediatric RRP. We studied 2 geographically defined populations in Atlanta and Seattle by contacting all otolaryngologists in those areas and reviewing the medical records of their pediatric patients who had RRP during 1996.

Patients and Methods

We chose 24 counties that encompass metropolitan and suburban Atlanta. The counties included Barrow, Bartow, Butts, Carroll, Cherokee, Clayton, Cobb, Coweta, DeKalb, Douglas, Fayette, Forsyth, Fulton, Gwinnett, Hall, Haralson, Henry, Lamar, Newton, Paulding, Polk, Rockdale, Spaulding, and Walton. US Census Bureau Population Statistics for 1990 [7] enumerated 810,902 persons aged <18 years in this area. Around Seattle, we chose 8 counties: King, Kitsap, Mason, Pierce, Skagit, Snohomish, Thurston, and Whatcom, which include the cities of Seattle and Tacoma. US Census Bureau Population Statistics for 1990 identified 827,164 persons aged <18 years in this area. Persons in the Atlanta and Seattle study areas were unlikely to seek medical care elsewhere.

We identified practicing otolaryngologists in the 24-county Atlanta and 8-county Seattle areas through the following: (1) lists of registered otolaryngologists from state licensing boards; (2) Who's Who of Certified Physicians; (3) local telephone directory yellow pages; and (4) membership lists from the American Society of Otolaryngology, Head and Neck Surgery. Otolaryngologists were initially contacted by letter and 2 weeks later by telephone and asked to identify all patients aged <18 years treated for RRP during the 1996 calendar year. All physicians were polled a second time from December 1996 through February 1997 to determine whether they had seen any new patients or had performed new procedures on existing patients.

We reviewed the medical records of eligible patients to obtain data on demographic characteristics, the progression of illness, insurance coverage, city of current residence, and the residence at the time of initial diagnosis. Only patients with documented current (1996) residence in the Atlanta or Seattle areas were enrolled in the study. Children who lived outside the study area at the time of their diagnosis but were treated within the study area were excluded from the incidence count. Children who lived within the study area during their treatment in 1996 were included in the prevalence estimate.

Statistical analyses were performed using SAS Software (SAS, Cary, NC). The Wilcoxon rank sum test was used for comparing the age at diagnosis and years of illness in populations that were not normally distributed. Because of small numbers of incident and prevalent cases in the target populations, we used Poisson regression to calculate estimates and 95% CIs and to test for differences between populations.

Results

We contacted 146 otolaryngologists in the Atlanta area and 183 in Seattle, and 101 (69%) and 139 (76%), respectively, agreed to participate in the study. All those who did not participate were retired, deceased, or had left the area. It is of interest that only 10 of the practicing otolaryngologists in Atlanta and 4 in Seattle were currently caring for pediatric RRP patients in the year studied, 1996. A total of 21 children in the Atlanta area and 18 in the Seattle area were patients of the otolaryngologists contacted. Of these, 21 (100%; 9 were diagnosed in 1996) of the Atlanta children and 14 (78%; 3 newly diagnosed) of the Seattle children were living in the study area (table 1). In both areas, the majority of patients were seen at University teaching hospitals. In Atlanta, 52% of all RRP cases identified attended Egleston Children's Hospital, and in Seattle, 93% attended the Children's Hospital and Regional Medical Center, University of Washington. The remaining cases were identified from private practices.

Although incidence and prevalence estimates were somewhat higher in Atlanta, the differences were not statistically significant. Age-specific prevalence seemed to decrease sharply among children aged >5 years in Atlanta, and aged >9 years in Seattle (table 2). The numbers of cases were small, and estimates did not differ significantly either between or within the cities. The RRP prevalence was similar among boys and girls in Atlanta, but differed sharply in Seattle, where RRP was more prevalent among boys. Because of small numbers, these prevalence estimates did not differ significantly. Finally, in both cities, non-white children were at greater risk for RRP, but the prevalence estimates were not significantly different in either city (Atlanta prevalence rate ratio, 1.86; 95% CI, 0.79–4.37; P = .16; Seattle prevalence rate ratio, 1.73; 95% CI 0.47–6.39; P = .41). The average age at diagnosis was 4.6 years in Atlanta and 3.0 years in Seattle. These differences were not statistically significant (P = .32), and there were no significant differences in age at diagnosis when stratified according to sex or race within each population. In Atlanta, children with RRP were ill for an average of 2.2 years, a significant difference from Seattle, where the average was 4.8 years (P = .03). No significant differences were observed between the average number of years with RRP by sex or race within either population.

Discussion

This is the first population-based survey to estimate incidence and prevalence of pediatric RRP. Clinical RRP requires multiple surgical procedures (i.e., patients must see a physician). We obtained cooperation from all practicing otolaryngologists in the Atlanta and Seattle areas. Furthermore, the nature of Atlanta and Seattle populations is that residents do not usually seek medical care in other cities. Therefore, we believe our study identified essentially all pediatric RRP cases in the 2 populations. As expected, RRP was a rare disease: only 35 patients (21 in Atlanta and 14 in Seattle) were identified. The differences in incidence we observed (1.11 in Atlanta and 0.36 in Seattle) and prevalence (2.59 in Atlanta and 1.69 in Seattle) probably reflect imprecision due to small numbers rather than true differences in the 2 populations. Therefore, we believe incidence and prevalence are accurately summarized by the lower and upper 95% CIs in the 2 areas. We estimated that 0.12–2.13 new cases of RRP occurred per 100,000 children aged <18 years in Atlanta and Seattle. We similarly estimated the prevalence of RRP as 1.00–3.97 per 100,000 children aged <18 years in these populations.

Our estimated range of RRP incidence is similar to the incidence of 0.4 per 100,000 children aged <16 years that was estimated in 1976 [3], but is much lower than the 4.3 per 100,000 incidence rate estimated by the 1995 study of children aged <14 years with RRP [4]. It is interesting to note that our estimated prevalence range is similar to the incidence rate reported by the 1995 study. Both of these earlier studies estimated RRP incidence rates by query-ing otolaryngologists nationally. They made the assumption, as we did, that RRP patients require surgical intervention; therefore, identified cases represent the population served by reporting physicians (i.e., the United States). Unfortunately, both studies had poor response rates (51% [3] and 23% [4]) and estimated the number of RRP cases by extrapolating according to the fraction of responses of surveyed physicians. In our study, a small geographically defined area was used to contact all practicing otolaryngologists, thereby eliminating the need for extrapolation and possible overestimation of the rates.

Our study estimated pediatric RRP incidence and prevalence in 2 cities and confirmed that it is a rare disease, with severe consequences for its young victims. To estimate the burden of pediatric RRP nationally, we applied our 1996 crude incidence and prevalence estimates to the United States. Based on April 1999 Census Bureau population estimates, 80–1500 children aged <18 years will be diagnosed with RRP this year, and 700–3000 prevalent cases are now receiving treatment. It is obvious that these estimates have severe limitations, since they were derived from data for 2 cities that may not represent the nation. Nonetheless, they provide a crude estimate of the magnitude of the problem.

We obtained outstanding cooperation from all otolaryngolo-gists who practice in the study areas, and presumably we identified all children who underwent surgery for recurring papillomas of the respiratory tract. However, these severe cases probably represent only a fraction of patients with pediatric respiratory HPV infection; subclinical respiratory papillomatosis does not require surgical intervention.

Acknowledgments

We thank the following physicians and clinic workers for their cooperation in allowing us to look at medical records: Ramon Franco, Andrew Golds, Terry Good, Gerald Gussack, Howard Herman, H. Chapman Jones, Rose Kerr, Michael Koriwchak, Norman Novelly, Michael Pickford, Craig Rone, James Thomsen, Ann White, and Judy Young. We also thank James Dobbins and Michele Reyes for technical assistance and John P. O'Connor for editorial assistance.

Figures and Tables

Table 1

Incidence and prevalence estimates of juvenile recurrent respiratory papillomatosis for Atlanta and Seattle, 1996.

Table 2

Age-, sex-, and race-specific incidence and prevalence estimates of juvenile recurrent respiratory papillomatosis for Atlanta and Seattle, 1996.

Research protocols were in compliance with human subjects experimentation guidelines of the US Department of Health and Human Services and were followed in the conduct of this study.

a

Present affiliations: Division of Cancer Prevention and Control, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, Georgia (L.R.A.); Children's Hospital of Philadelphia, Pennsylvania (I.N.J.)