Acceptability of High-Resolution Anoscopy for Anal Cancer Screening in HIV-infected Patients

JO Lam; GM Barnell; M Merchant; CG Ellis; MJ Silverberg


HIV Medicine. 2018;19(10):716-723. 

In This Article


This study found that HRA screening for anal cancer was generally well tolerated among HIV-infected patients participating in an anal cancer screening programme in a large integrated health care system. Although we had anticipated that the lack of tolerability of a first-time KPNC HRA may increase patients' hesitancy to participate in future screening, we found that this was not the case. Neither the lack of tolerability of initial HRA nor the presence of a pre-existing anal condition (which could increase discomfort during and after the procedure) affected the likelihood of patients returning for a repeat procedure. In examination of EHR-based measures, younger age appeared to be associated with patients not returning for a repeat HRA, but 100% of patients who completed our survey stated that they would return for a follow-up HRA if it was recommended. Despite overall high tolerability of HRA as a screening method, several findings should be noted. Patients with a pre-existing anal condition at screening, women, and ever smokers were more likely to need prescription opioid pain medication. Additionally, younger patients (< 50 years old) reported greater HRA-associated pain. These factors are important considerations for practitioners caring for these patients, particularly during HRA pre-procedure counselling.

Our results are consistent with previous studies which also reported high tolerability of HRA among MSM and men with HIV infection.[14,22,23] Our study contributes data on the acceptability of HRA among women, and provides more comprehensive measures of lack of tolerability of HRA through evaluation of both objective clinical measures and subjective patient-reported measures in a greater number of patients. In addition, 70.8% of survey respondents had not heard of HRA before, suggesting that our study may have included a largely HRA-naïve population. Pain and bleeding during and after HRA were the primary areas of concern in all studies, and both were reported to be acceptable to patients. Schofield et al.[14] found that pain was generally ranked low and 92.4% of MSM would attend anal screening in the future. Hillman et al. reported that, although HRA acceptability was strongly correlated with pain and bleeding, over half of patients had minimal or no pain and that pain lasted only a few hours and resulted in few complications. Unlike our study, however, Hillman et al.[23] found that most patients (92.9%) required intervention for pain relief, although it is unclear what proportion of these patients required prescription pain medication. In our study, dispensation of opioid analgaesia was relatively low (~ 5%) even though the majority of patients had HRA with biopsy (as opposed to HRA with visualization only), and some also received treatment for HSIL at the initial HRA visit, which we expected to be less well tolerated given the lengthier and more intensive clinical procedure. Davis et al.[22] reported high HRA acceptance, with or without biopsy, among a cohort that included 3% women but did not report results for women separately and did not distinguish patients having a first-time HRA. Other studies of anal examinations for cancer screening have reported similarly high tolerability and willingness of patients to comply with screening, including studies of anal Pap smears,[27] self-collected anal swabs[28] and annual anorectal digital examinations to detect anal cancer in HIV-infected MSM.[29,30]

A novel finding in our study is that patients < 50 years old reported greater HRA-associated pain. The reason for this finding is unclear. Younger patients may have had less prior experience with anorectal screening procedures, such as anal Pap smears or digital rectal examinations, and so, despite provider counselling regarding HRA, they may have been less prepared for the discomforts associated with HRA. However, we were not able to assess the non-HRA anorectal screening history of our patients to confirm this hypothesis. Younger age was not associated with a greater likelihood of needing pain medication, so it is also possible that younger patients reported greater pain because they decided to forgo the pain medication prescription offered by their provider. Younger patients may be more sexually active and therefore more likely to be frustrated by any pain which could limit their sexual activity. In general, we found that patients were well informed about the discomforts of the HRA procedure, which reflects the great attention to this topic in provider-patient discussions.

Although we did not identify significant barriers to patients returning for a repeat HRA, a substantial proportion (50%) of patients who received an initial HRA did not return for a repeat procedure within 2 years. While pain at initial HRA did not influence compliance with a repeat HRA, mild discomfort, bleeding or some other unmeasured qualitative factor related to the initial HRA experience may have dissuaded patients from obtaining a repeat procedure, particularly those who had no dysplasia or a nonworrisome biopsy, and therefore had reduced concern about their anal cancer risk and need for future screening. Both Oakland and San Francisco HRA clinics had periods of long waitlists and follow-up backlogs during the study period, and so scheduling difficulties may have had some effect on the percentage of repeat HRAs. The relatively low repeat HRA rate could also have been provider-driven. As there are no standardized guidelines for HRA screening intervals, there may have been some variation in how patients were scheduled for follow-up care. While KPNC is an integrated health care system, and there is some coordination across sites to determine best anal cancer screening practices, providers may also make patient-specific recommendations for HRA follow-up that differ from the overall practice across KPNC. Lastly, it is possible that patients obtained a repeat HRA from a non-KPNC facility which was not captured in our EHR, or that they returned for a repeat HRA at KPNC after our study's follow-up period.

The results of our study should be interpreted with consideration of several limitations. We did not have results of patients' initial HRA, so we were unable to evaluate differences in HRA tolerability by extent of SIL treatment or pathological diagnosis. Nontreatment (i.e. HRA with visualization only) versus treatment (and extent of treatment) may have affected patient experience of pain, which could have influenced whether opioid analgaesia was prescribed. Extent of treatment or diagnosis may also influence returning for recommended follow-up HRA. Survey-based measures of HRA tolerability were self-reported. Also, as we did not have results of patients' initial HRA, we were unable to link survey responses to the type of treatment received. Patient expectations of pain and bleeding were assessed post-procedure (i.e. we did not have a pre-procedure measure of patients' expectations for comparison). We were unable to account for the anatomical location of the anal biopsy (i.e. proximal versus distal anal canal) which could have affected the amount of post-exposure bleeding and pain;[13] however, bleeding was a minimal concern for most patients. We were also unable to systematically assess whether patients were symptomatic at initial clinical presentation, which could have affected their reported levels of pain. Also, prescription for opioid analgaesia in patients' EHR was used as a proxy for pain and for pain medication usage. Patients who obtained over-the-counter, nonopioid medications for pain relief, or who obtained pain medications from a non-KPNC facility, would not have been captured. Thus, it is possible that we detected the patients least likely to tolerate the procedure; conversely, it is possible that patients who filled a prescription for opioid analgaesia did not need to use any of the medicine. Furthermore, as procedure codes for HRAs were not available, HRAs requiring sedation were identified in the EHR using a text string-based algorithm, which may have underestimated the number of sedations performed. In this study, all HRAs were performed by practitioners experienced with anal dysplasia screening in a setting with established screening and treatment protocols for clinical care and management of men and women at high risk of anal cancer. Given the variability in provision of HRA-directed biopsy in clinical practice, patient experiences may be different in other clinical settings or nonintegrated health care settings.[13,20,31] Additionally, in our high-volume San Francisco and Oakland clinics, our providers frequently perform HRA procedures and so patient experience may not be generalizable to lower volume clinics.

This study had several strengths. While prior research recruited men into structured studies of HRA screening, we assessed screening in the context of routine clinical practice, which may more closely reflect the engagement of high-risk groups with anal cancer screening while in usual HIV care. As KPNC has two high-volume screening clinics, we had a large sample size. The linkage of clinical, laboratory, pharmacy and administrative databases at KPNC allowed for evaluation of individual-level covariates. Furthermore, we completed a detailed survey of patient experience with HRA screening, which included quality-of-life measures in addition to measures of pain and willingness to attend follow-up screening, in a predominantly HRA-naïve population.

Our finding that high-risk patient groups may tolerate HRA screening for anal cancer contributes to ongoing discussions about the feasibility of routine screening in individuals at highest risk of anal cancer. Anal cancer is one of the most common cancers in HIV-infected individuals, with no evidence of reduced risk with effective ART. Anal cancer-related morbidity and mortality are expected to increase as life expectancy increases in highly active antiretroviral therapy (HAART)–treated populations.[4,19,21,32,33] Therefore, these findings will be increasingly relevant to future discussions on standards of care for anal cancer prevention in PWH. The finding of high tolerability of HRA screening in a largely HRA-naïve population is particularly important as primary screening using HRA-directed biopsy has been proposed as a potentially more cost-effective screening strategy in high-risk groups than triaging patients via anal cytology.[34] Also, as additional follow-up HRA may be required to monitor the progression of dysplasia, our finding that surveyed patients were willing to comply with follow-up screening is encouraging. However, the relatively low observed rate of repeat HRA warrants further investigation to ensure that patients who would benefit the most from anal cancer surveillance are obtaining the necessary care.