A, Patients with high-grade dysplasia on high-resolution anoscopy presented with anal pain more frequently than did patients with low-grade or no dysplasia, and patients with either high- or low-grade dysplasia on high-resolution anoscopy were more likely to present with anal lesions on physical examination compared with patients with no dysplasia. B, Symptoms stratified by anal Papanicolaou smear findings showed no significant trends: patients with high-grade dysplasia on anal Papanicolaou smear reported anal pain, bleeding, pruritus, overall symptoms, and anal lesions with similar frequency as those with low-grade dysplasia, atypical squamous cells of undetermined significance (ASCUS), or no dysplasia.
Hicks CW, Wick EC, Leeds IL, Efron JE, Gearhart SL, Safar B, Fang SH. Patient Symptomatology in Anal Dysplasia. JAMA Surg. 2015;150(6):563-569. doi:10.1001/jamasurg.2015.28
High-resolution anoscopy (HRA) is becoming increasingly advocated as a method of screening for anal dysplasia in high-risk patients.
To describe, through HRA findings, the association between patient symptomatology and anal dysplasia among patients at high risk for anal dysplasia.
Design, Setting, and Participants
Univariable and multivariable analyses were conducted of data from a prospectively maintained HRA database on all patients undergoing HRA with biopsy from November 1, 2011, to March 13, 2014, at a tertiary care HRA clinic. Data included demographics, medical history and comorbidities, HIV status and related measures (CD4 cell counts, HIV viral load, and use of highly active antiretroviral therapy), sexual orientation (when available), patient symptoms at initial presentation, physical examination findings, anal Papanicolaou (Pap) smear findings.
Main Outcomes and Measures
High-resolution anoscopy diagnosis of high- vs low-grade dysplasia or no dysplasia.
One hundred sixty-one HRA biopsy specimens (mean [SEM], 1.77 [0.11] biopsy specimens per patient) were obtained from 91 patients (mean [SEM] age, 45.7 [1.2] years; 61 men [67%]; 47 black patients [52%]; and 70 human immunodeficiency virus–positive patients [77%]). Twenty-seven patients (30%) had high-grade dysplasia, 26 had low-grade dysplasia (29%), and 38 had no dysplasia (42%). The majority of patients (63 [69%]) were asymptomatic (anal pain, 11 [12%]; bleeding, 14 [15%]; and pruritus, 10 [11%]). Forty-one patients (45%) presented with anal pain (odds ratio, 5.25; 95% CI, 1.44-21.82; P = .02), and patients with either high- or low-grade dysplasia were more likely to present with anal lesions on physical examination compared with patients without dysplasia (odds ratio, 4.34; 95% CI, 1.78-11.20; P = .002). Multivariable analysis suggested that anal pain was independently associated with high-grade dysplasia (odds ratio, 6.42; 95% CI, 1.18-43.3; P = .03).
Conclusions and Relevance
Anal dysplasia is a silent disease that is frequently asymptomatic. However, patients with anal pain, anal lesions, and other high-risk factors are at increased risk of having high-grade anal dysplasia. These patients may benefit from routine screening with HRA.
The prevalence of anal cancer has been increasing during the past few decades, especially among the human immunodeficiency virus (HIV)–positive population.1 It is estimated that anal cancer affects approximately 46 in 100 000 HIV-positive men and 10 in 100 000 HIV-positive women compared with 2 in 100 000 HIV-negative men and negligible rates for HIV-negative women.2 The incidence of anal cancer is even higher in HIV-positive men who have sex with men (MSM) (approximately 131 in 100 000).1
There is well-established evidence that anal intraepithelial neoplasia (AIN) progresses to cancer with time, particularly within the immunosuppressed population.3 For HIV-positive MSM with high-grade AIN, progression to anal cancer has been reported to occur in 10% to 50% of cases within 2 years.4- 6 Although no formal screening guidelines currently exist, health care professionals experienced in the care of high-risk populations have therefore started screening their patients for anal dysplasia.7 In 2012, the American Society of Colon and Rectal Surgeons published recommended practice parameters for anal squamous neoplasms, including surveillance examinations at 4- to 6-month intervals for both high-risk individuals and as surveillance after treatment of AIN.3 However, to our knowledge, the association between patient symptomatology and anal dysplasia findings has never been reported, so the utility of clinical presentation findings as an indication of anal cancer risk in high-risk patients is unknown. The objective of the current study was to describe the association between patient symptomatology and high-resolution anoscopy (HRA) findings by analyzing the breakdown of symptoms in all patients at high risk for anal dyplasia who present for HRA, as well as symptoms in patients grouped by HRA findings of high-grade (AIN grades 2 and 3) vs low-grade (AIN grade 1) dysplasia.
Data were obtained from our prospectively maintained institutional review board–approved HRA database. The study was conducted according to accepted ethical standards for human investigation and had approval from the Johns Hopkins Hospital Institutional Review Board prior to its initiation. All medical records of patients 18 years or older undergoing HRA with biopsy between November 1, 2011, and March 13, 2014, were reviewed for inclusion in the study. Patients who underwent HRA without biopsies, did not have biopsy results available in the records, or were undergoing repeat HRA were excluded (n = 94).
Data on demographics, medical history and comorbidities, HIV status and related measures (CD4 cell counts, HIV viral load, and use of highly active antiretroviral therapy), sexual orientation (when available), patient symptoms at initial presentation, physical examination findings, anal Papanicolaou (Pap) smear findings, and HRA findings were analyzed for all eligible patients.
Anal Pap smears and HRAs were performed in our weekly HRA clinic. Patients were referred to the clinic from their primary care physician or our institution’s HIV clinic and/or if they had a history of abnormal Pap smear results. After written informed consent was obtained, the patient was positioned in the left lateral decubitus position and an anal Pap smear was performed. The perianal skin was inspected on gross examination and with the colposcope. Lidocaine cream, 4%, was placed at the anal verge and within the anal canal during the digital anorectal examination. Gauze soaked in acetic acid, 5%, was then placed within the anal canal for 3 minutes. After the gauze was removed, the dentate line was identified and visualized with the colposcope. Lugol iodine solution was applied. Cold forceps were used to obtain biopsy specimens of areas of acetowhitening and Lugol negativity, in addition to vascular abnormalities or findings consistent with dysplasia. Anal Pap smear and all HRA biopsy specimens were submitted to the pathology department for further analysis.
Cytologic findings on the anal Pap smear were categorized as no dysplasia, atypical squamous cells of undetermined significance, low-grade dysplasia (low-grade squamous intraepithelial lesions and AIN grade 1), or high-grade dysplasia (high-grade squamous intraepithelial lesions [HSIL], AIN grade 2, AIN grade 3, low-grade squamous intraepithelial lesions suspicious for HSIL, and atypical squamous cells suspicious for HSIL). Patients were classified as having high-grade anal dysplasia based on HRA pathologic findings of AIN grade 2 or 3, low-grade anal dysplasia based on HRA pathologic findings of AIN grade 1, or no dysplasia based on HRA pathologic findings of no dysplasia.
Baseline and descriptive statistics were summarized as mean (SEM) or number (percentage), as appropriate. Descriptive statistics and univariable analyses, including t tests (continuous variables) or Pearson χ2 tests or Fisher exact tests (categorical variables), were used to describe the breakdown of symptoms in all patients presenting for HRA and in patients grouped by findings of high-grade (AIN grades 2 and 3) vs low-grade (AIN grade 1) dysplasia. Multivariable logistic regression was performed to determine independent predictors of high-grade dysplasia. Model covariates were chosen based on factors found to be significant on univariable analysis (age, sex, anal pain, anal lesions on examination, HIV viral load >20 copies/mL, and high-grade dysplasia on anal Pap smear).
Two-tailed tests were used for all analyses, and P ≤ .05 was considered significant. Statistical analyses were performed using JMP, version 9.0 (SAS Institute).
A total of 161 HRA biopsies (mean [SEM], 1.77 [0.11] biopsies per patient) were obtained from 91 patients (mean [SEM] age, 45.7 [1.2] years; 61 men [67%]; 47 black patients [52%]; and 70 HIV-positive patients [77%]) during the 2½-year study period. A summary of the demographic characteristics of included patients is in Table 1. Of those who were asked, 73% of patients (22 of 30) were MSM and 78% of patients (14 of 18) practiced anoreceptive sex. A history of anal condyloma was common (30 [33%]), and more than half of the patients (46 [51%]) had previously had some form of anorectal surgical intervention.
The majority of patients (63 [69%]) were asymptomatic at the time of screening. Among those presenting with symptoms, bleeding was most common (14 [15%]), followed by anal pain (11 [12%]) and pruritus (10 [11%]). Forty-one patients (45%) had anal lesions on physical examination. Anal Pap smear cytologic results demonstrated no dysplasia in 19 patients (22%), atypical cells of uncertain significance in 23 (27%), low-grade dysplasia in 19 (22%), and high-grade dysplasia in 25 (29%) (Table 1). Pathologic results found on HRA demonstrated the presence of high-grade (AIN grades 2 and 3) dysplasia in 27 patients (30%), low-grade (AIN grade 1) dysplasia in 26 (29%), and no dysplasia in 38 (42%). The anal Pap smear and pathologic findings on HRA were consistent in 60 patients (66%).
To determine the association between patient symptomatology and HRA dysplasia findings, presenting anal symptoms and other clinical factors were analyzed for patients with high-grade vs low-grade vs no dysplasia. Patients with high- and low-grade dysplasia were younger (P = .04) and more frequently male (P = .01) compared with those with no dysplasia (Table 2). Race, sexual orientation (when available), HIV status, comorbidities, and smoking status were not statistically different between groups. There was a trend for more patients with high- and low-grade dysplasia to have HIV viral loads greater than 20 copies/mL compared with those with no dysplasia, but this trend was not statistically significant (P = .07) (Table 2). Patients with high-grade dysplasia on HRA had a higher incidence of high-grade dysplasia on anal Pap smear (54%) than patients with low-grade (19%) or no dysplasia (18%) (P = .02).
Patients with high-grade dysplasia presented with anal pain more frequently than did patients with low-grade or no dysplasia (odds ratio, 5.25; 95% CI, 1.44-21.82), and patients with either high- or low-grade dysplasia were more likely to present with anal lesions on physical examination compared with patients with no dysplasia (odds ratio, 4.34; 95% CI, 1.78-11.20). There were no significant differences when comparing anal bleeding, pruritus, or overall symptomatology between groups (Figure, A). In contrast, symptomatology stratified by findings on anal Pap smears showed no significant trends: patients with high-grade dysplasia on anal Pap smears reported similar frequencies of anal pain, bleeding, pruritus, overall symptomatology, and anal lesions compared with those with low-grade dysplasia, atypical squamous cells of undetermined significance, or no dysplasia (Figure, B).
Multivariable logistic regression analysis accounting for variables found to be significant on univariable analysis suggested that anal pain (P = .03), younger patient age (P = .04), and high-grade dysplasia (P = .05) on anal Pap smears were each independently associated with the presence of high-grade dysplasia on HRA (Table 3). Male sex (P = .13) and the presence of anal lesions on physical examination (P = .15) showed a trend toward increasing risk of high-grade dysplasia on HRA, but this trend was not statistically significant. An HIV viral load greater than 20 copies/mL was also not significantly associated with high-grade dysplasia after risk adjustment (Table 3).
Anal cancer is an increasingly prevalent disease that affects nearly 8000 people per year in the United States.8 Although patients who are HIV positive and those who are immunosuppressed (ie, after a transplant) have the highest incidence of disease, rates of invasive anal cancer are increasing among all men and women at an estimated rate of 2.0% to 5.1% per year.1,8 As a result, routine screening surveillance protocols for anal dysplasia that are similar to those for cervical dysplasia have been a topic of discussion since 1997.9 The peak incidence of cervical cancer was estimated to be 115 per 100 000 women in the United States when routine cytologic screening by Pap smear was introduced in the late 1940s.10 Given that anal cancer is now more prominent in certain populations than cervical cancer was at that time, there is significant debate regarding the appropriateness of introducing routine screening guidelines for anal cancer as well.
Despite substantial discussion on the topic, there are still no official recommendations for the screening and surveillance of anal cancer to date. Risk factors for HSILs of the anus have previously been studied, but there is a paucity of data regarding the association between patient symptoms and anal pathologic findings. In the current study, we describe the distribution of symptoms, both overall and based on HRA pathologic findings, among a group of patients who were deemed to be high-risk for anal dysplasia by their primary care physician. We demonstrate that although symptoms are reported infrequently among patients presenting for HRA screening, anal pain is an independent predictor of high-grade dysplasia. Anal lesions also appear to be precursors for more advanced disease, although the presence of an anal lesion was not significantly associated with high-grade dysplasia after risk adjustment.
Given the rapid increase in anal cancer rates that has been observed during the past 2 decades,1,8 several studies have reported on risk factors that are thought to be predictive of anal cancer and its precursor—HSIL. In a study of more than 1000 histologic specimens by Swedish et al,11 the authors found that HIV status, human papillomavirus infection, and abnormal anal cytologic results were independently predictive of high-grade AIN. These findings are consistent with our results demonstrating that patients with high-grade dysplasia on HRA had a higher incidence of high-grade dysplasia on anal Pap smear than did patients with low-grade or no dysplasia. Human immunodeficiency virus status was not an independent predictor of dysplasia in our study, likely because many patients who were referred to our clinic were HIV positive (70 [77%]). In addition, we did not assess human papillomavirus status as a predictor because we do not routinely screen for human papillomavirus status in presenting patients based on its high sensitivity (90%) but low specificity (40%) for HSIL.12 Other studies have reported time from HIV diagnosis, use of highly active antiretroviral therapy, CD4 cell count, sexual practices (anoreceptive sex or MSM), and smoking as potential predictors of high-grade dysplasia.11,13,14 However, these factors have not been routinely supported by larger, more recent series,11,15- 20 and were also not found to be predictive of high-grade dysplasia in our study.
Of note, male sex and anal lesions on physical examination were significantly associated with high-grade dysplasia on univariable analysis in our study, although they were not statistically significant in our multivariable model. The incidence of anal cancer is known to be more than 4 times higher in men compared with women.2 Similarly, suspicious lesions on physical examination are a well-recognized risk factor for anal cancer.14 As a result, digital anorectal examinations are a ubiquitous aspect of the recommended annual physical examination for HIV-positive patients.15 In our study, less than half of all patients with anal dysplasia had anal lesions and the presence of an anal lesion showed a nonsignificant association with high-grade dysplasia, indicating that the sensitivity of this clinical predictor is likely low. Therefore, although initial screening for patients considered at high risk for anal dysplasia should include a digital rectal examination, the lack of a palpable or visible lesion should not preclude a patient from undergoing more extensive screening.
One of the largest concerns regarding the implementation of a standardized screening algorithm for patients who are at risk for having anal dysplasia relates to the cost-effectiveness of this practice. High-grade dysplasia is a recognized risk for the development of anal cancer.21 However, the progression of anal dysplasia to invasive anal cancer is thought to be slower than the progression of similar dysplastic cervical lesions,1,22 and the long-term effect of routine screening on survival and other outcomes is unknown. Several screening algorithms for anal dysplasia have been proposed based on the above-mentioned published risk factor data.19,23- 27 Most of these algorithms use HIV status and findings on anal Pap smear as criteria for proceeding to HRA. However, there is currently no comprehensive approach that incorporates both symptoms and clinical data to aid in the identification of which individuals may be at the highest risk for dysplasia. This is the first study, to our knowledge, that attempts to link HRA findings with patient symptomatology, and our findings highlight the importance of maintaining a high index of suspicion for dysplasia among high-risk patients regardless of their clinical presentation and frequent lack of symptoms. Future studies investigating the utility of clinical risk calculators for anal dysplasia may be useful in identifying patients who would benefit from HRA screening.
Our study contains some limitations. The number of patients with available data on sexual practices was limited, which may have underpowered our study to identify a potential significant effect of MSM or anoreceptive sex on dysplasia risk. However, of the patients for whom we had data on this subject, nearly three-quarters practiced high-risk sexual behaviors. The high prevalence of MSM and anoreceptive sex in our patient population makes the likelihood that we would have detected a significant association between sexual practices and anal dysplasia relatively low, similar to the phenomenon that we observed with HIV status. Our sample size of 91 patients was relatively low in this preliminary study. In addition, our data represent a biased population of patients; that is, patients who were seen in the HRA clinic had already been prescreened as being at high risk because the majority were referred by our institution’s HIV care center. As such, our findings are not applicable to large populations but should be considered a preliminary description of clinical and symptomatic risk that could potentially be applied in appropriate settings such as HIV clinics. The benefit of this design is that it captures symptom and clinical data on all patients referred to a tertiary center HRA clinic, which allows us to present a comprehensive overview of the clinical presentation of a high-risk population that has been lacking from prior studies.
The development of targeted screening guidelines for anal dysplasia is becoming increasingly relevant. To date, clinical practice is guided by data from small cohort and retrospective studies, which has resulted in a range of potential risk factors associated with an increased risk of high-grade anal dysplasia. Abnormal anal Pap smear findings, HIV status, and male sex are commonly identified risk factors. We present data from a prospective high-risk HRA clinic suggesting that anal dysplasia is frequently asymptomatic, but that patients with anal pain, anal lesions, and other high-risk factors are at increased risk of having high-grade anal dysplasia. These patients may benefit from routine screening with HRA.
Accepted for Publication: November 25, 2014.
Corresponding Author: Sandy H. Fang, MD, Division of Colorectal Surgery, Department of Surgery, Johns Hopkins Hospital, 600 N Wolfe St, Blalock 618, Baltimore, MD 21287 (email@example.com).
Published Online: April 15, 2015. doi:10.1001/jamasurg.2015.28.
Author Contributions: Dr Fang had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Hicks, Efron, Gearhart, Safar, Fang.
Acquisition, analysis, or interpretation of data: Hicks, Wick, Leeds, Fang.
Drafting of the manuscript: Hicks, Gearhart, Fang.
Critical revision of the manuscript for important intellectual content: Hicks, Wick, Leeds, Efron, Safar, Fang.
Statistical analysis: Hicks.
Obtained funding: Fang.
Administrative, technical, or material support: Hicks, Wick, Efron, Fang.
Study supervision: Fang.
Conflict of Interest Disclosures: None reported.
Funding/Support: This project was completed with funding from the State of Maryland, Department of Health and Mental Hygiene, Cigarette Restitution Fund Cancer Research Grant.
Role of the Funder/Sponsor: The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.