Pap test screening is an important component of primary care for women with human immunodeficiency virus (HIV) infection. In women without HIV infection, routine cytology screening for vaginal intraepithelial neoplasia (VAIN) and vaginal cancer after hysterectomy is not recommended if they have no history of cervical dysplasia.1–3
Two guidelines have been published on Pap test screening in HIV-infected women after hysterectomy. The guideline by Department of Health and Human Services published in 2013 recommends routine screening for HIV-infected women after a hysterectomy for high-grade precancer or cervical cancer; however, it does not recommend routine screening for those who underwent a hysterectomy for benign disease without a history of high-grade precancer or cervical cancer.4 A guideline published by the Infectious Disease Society of America suggests that HIV-infected women who have had a hysterectomy, especially those with a history of abnormal cervical cytology, should undergo regular screening with Pap tests at intervals similar to those for women who have not undergone a hysterectomy.5 However, it remains unclear whether Pap test screening is not necessary for HIV-infected women after hysterectomy without a history of high-grade precancer or cancer.6,7
Our research aims to assess the prevalence of abnormal vaginal cytology and VAIN in HIV-infected women who had a hysterectomy for conditions other than cervical dysplasia or cancer. We also aim to explore the risk factors associated with vaginal dysplasia in women who had no cervical dysplasia before hysterectomy and the time to develop the abnormal Pap test result.
MATERIALS AND METHODS
This is a retrospective cohort study of HIV-infected women from two Harris Health System clinics, Thomas Street Health Center and Northwest Community Health Center, in Houston, Texas. These two sites have 68 health care providers, four of whom do Pap test screening; health care providers saw 1,827 HIV-infected women in 2015. The rate of routine Pap test screening is 47%. Pap test screening was either done by physicians or nurse practitioners. The populations included in this study are HIV-infected women with no history of cervical dysplasia or cancer who have a history of hysterectomy and received routine gynecologic follow-up within the Harris Health System from January 2000 to January 2015. Institutional review board approval was obtained through the Baylor College of Medicine institutional review board and Harris Health System institutional review board. Median follow-up time for the Pap test was 16 years.
Inclusion criteria included HIV-infected women with a history of hysterectomy for benign reasons without any history of cervical dysplasia or cervical cancer. Women may have had HIV diagnosed after hysterectomy. Cervical dysplasia is defined as having a record from a personal report or from a documented Pap test report in the medical record of atypical cells of undetermined significance (ASC-US) and human papillomavirus (HPV) positive, low-grade squamous intraepithelial lesions (LSIL), or high-grade squamous intraepithelial lesions (HSIL) before hysterectomy or an abnormal pathology report from hysterectomy. Exclusion criteria included HIV-uninfected women, HIV-infected women with a history of hysterectomy for cervical dysplasia or cervical cancer or with a pathology report from hysterectomy showing cervical dysplasia or cancer, and women with no Pap test information available after hysterectomy.
The electronic medical record for each patient was reviewed for date of birth, race and ethnicity, gravidity and parity, current or past smoking history, history of illicit drug use, history of alcohol abuse, date and age at hysterectomy, reason for hysterectomy, history of genital warts, cytology and pathology reports before hysterectomy, cytology results with date of each Pap test after hysterectomy, CD4 count (the number of CD4 T lymphocytes) at hysterectomy if available, viral load (the number of HIV particles per milliliter of blood) at hysterectomy if available, CD4 count within 6 months of each Pap test if available, viral load within 6 months of each Pap test if available, if on antiretroviral therapy at the time of last Pap test, and vaginal biopsy results if available. Data were collected and coded on a data collection form.
Pap test results of ASC-US and HPV-positive, LSIL, or HSIL after hysterectomy were defined as abnormal; Pap test results in the normal or ASC-US and HPV– categories were defined as normal. Women were considered to be cases and have an abnormal vaginal Pap test results if any Pap test result was positive after hysterectomy. The proportion of cases and the asymptotic 95% confidence interval (CI) were used to estimate the prevalence of abnormal vaginal cytology.
Descriptive statistics such as median, interquartile range (25th, 75th percentile) for the continuous variables, frequency, and proportion for the categorical variables were calculated for all patients and by groups. Viral load values were logarithm-transformed for further data analysis. Demographic and clinical characteristics were compared between cases and noncases using Wilcoxon rank-sum test for the continuous variables and χ2 test or Fisher exact test for categorical variables.
We then used the Kaplan-Meier method to analyze the association between these factors and event-free survival where the event is defined by the development of the first abnormal Pap test result after hysterectomy. The survival time is calculated by the duration in years between time of hysterectomy and date of the first abnormal Pap test result for the patients with any abnormal Pap test results after hysterectomy and date of last clinical record for the patients with all normal Pap test results after hysterectomy. The Kaplan-Meier curves were generated by levels of the factors and log-rank tests were used to test the difference in the Kaplan-Meier curves among strata. Cox proportional hazard models were used to test the effect of risk factors on the survival using first category as the reference group. Unadjusted and adjusted hazard ratios and their 95% CIs were reported. A P value <.05 was considered statistically significant. SAS 9.4 was used for all analyses.
The prevalence in this study of abnormal Pap test results after hysterectomy was 31% (95% CI 25–37%). Among 238 women, Pap test results showed 12 (5%) with ASC-US and HPV-positive, 55 (23.1%) with LSIL, and seven (2.9%) with HSIL. No specific demographic risk factor was identified as being associated with progression to vaginal dysplasia after hysterectomy (Table 1).
Of those who underwent vaginal biopsies for abnormal Pap test results, 15 (28%) had normal biopsy results, 23 (43%) had VAIN 1, nine (16%) had VAIN 2, and seven (13%) had VAIN 3. No patients had invasive vaginal cancer.
Race, smoking history, alcohol history, and illegal drug use were not associated with development of abnormal Pap test results (Table 1). No specific reason (bleeding, pelvic pain, uterine leiomyomas) for hysterectomy was associated with development of abnormal Pap test results. Additionally, no historic obstetric or gynecologic demographic variable was associated with development abnormal Pap test results (Table 2).
We calculated the effect size, and statistical power for the variables did not achieve the statistical significance in the tests. For example, the effect size of race, smoking, alcohol, and drug use variables varies from 0.013 to 0.164 by the χ2 tests. Per the general guideline suggested by Cohen, these effect sizes are small (small: 0.1, medium: 0.3, large: 0.5).8 With a current sample size of 238, the statistical power to detect the difference between two groups varies from 5.3% to 61.6%. Larger sample sizes are needed to detect the significant difference for these small effect sizes.
The most recent CD4 count or viral load was not significantly different between abnormal and normal Pap test results; however, the nearest CD4 count and viral load were significantly different between the two groups (Table 3). Of note, patients with an increased CD4 count or with a decreased log of viral load were less likely to develop an abnormal Pap test result; the P values are .02 and .001, respectively (Figs. 1 and 2). Patients who were younger at hysterectomy had a longer period of time to develop an abnormal Pap test result (Fig. 3).
Adjusting other covariates, age and nearest viral load had a significant effect on the abnormal Pap test result–free survival, whereas the effect of nearest CD4 count was not significant (P=.14). More specifically, older patients and higher viral load significantly increased the risk of development of an abnormal Pap test result. For example, the patients with viral load values greater than 400 had approximately two times the risk of developing an abnormal Pap test result than those lower than 400; the adjusted hazard ratio is 2.1 (95% CI 1.2–3.5) (Table 4). No difference in time to abnormal Pap test result was noted with antiretroviral use (P=.22).
Our findings suggest that a significant portion of HIV-infected women who have had a hysterectomy for benign reasons will have abnormal vaginal cytology and abnormal vaginal biopsy results. More than 30% of HIV-infected women who had no prehysterectomy history of abnormal Pap test results had abnormal vaginal Pap test results. Among those who had vaginal biopsies, 29% had VAIN 2 or 3, suggesting that Pap testing posthysterectomy in the HIV population may be indicated. No specific demographic risk factors were identified in women with abnormal cytology or VAIN. Pap test screening guidelines for HIV-infected women may need to be modified in the future in response to the large number of abnormal vaginal cytologic screening tests found.
Two other studies have similarly evaluated the incidence of abnormal Pap test results in HIV-infected women. A previous study by Paramsothy et al6 in 2004 demonstrated that abnormal Pap test results were found among the HIV-infected population, but only when no known Pap tests were performed or abnormal Pap test results had been documented previously. Massad9 investigated the Pap test results posthysterectomy of 3,700 HIV-seropositive women and concluded that, although Pap test results were more often abnormal in HIV-seropositive women than in HIV-seronegative women, VAIN 2 or worse was infrequent. Of note, the Massad study did not differentiate between the populations of women with abnormal cytology before hysterectomy and those with no history of an abnormal Pap test result before hysterectomy. Our study specifically looked at those women who had no known history of abnormal cytology before surgery, a group that we might have presumed to be at low risk.
No existing cytology screening guidelines specify whether cytologic screening of the vagina is indicated in women with HIV who have had hysterectomies for benign reasons. To date there has been little evidence to guide clinicians as to whether to continue Pap screening in this population. It is significant that, in this study, 31% of patients had abnormal Pap test results and 29% of women with vaginal biopsies had abnormal biopsy results.
There was a statistically significant increase in the number of Pap tests in patients with a history of normal Pap test results (normal or ASC-US and HPV-negative) compared with those with a history of abnormal Pap test results (ASC-US and HPV-positive, LSIL, or HSIL) (P=.001). No significant difference in the most recent CD4 count or viral load was found between those with normal compared with abnormal Pap test results. However, before the first abnormal Pap test result, there was a statistically significant difference in the CD4 count (P<.001) and in the viral load (P=.001) between those with normal or abnormal Pap test results (Table 3).
We examined the number of CD4 count and viral load available according to each Pap test. Approximately 94% of Pap tests had a corresponding CD4 count and viral load. The CD4 count and viral load chosen were those that were closest in time to the Pap test within a duration of 6 months; however, the exact date was not recorded. This same interval of CD4 count and viral load to Pap test was also used for the Pap test closest to hysterectomy, although, as a result of lack of documentation at the time of hysterectomy before the patient entered the Harris Health System, this information was not always available.
More patients on antiretroviral therapy did have abnormal Pap test results (P=.01), but no significant difference was found in the grade of the Pap test result if the patient was on antiretrovirals by the time of the abnormal Pap test result (P=.33) (Table 5). We hypothesize that more abnormal Pap test results were found in the population of patients being followed in the clinic on antiretroviral therapy because most of such patients undergoing Pap tests were also being prescribed antiretroviral therapy.
Limitations of this study include sample size and retrospective study design, which restrict the ability to control for variables and are dependent on the information documented in the patient's chart. The numbers were smaller than some previous studies because we purposely excluded women with prior abnormal Pap test results (guidelines do exist for the latter population).9 This study included patients from a single health system in a single demographic area and thus the findings may not be generalizable to other populations. Strengths of the study are the specific focus on HIV-infected women with no known prehysterectomy cytologic abnormality and the resulting addition to the previously scarce literature on this subject. Whether to focus on cytologic abnormalities (31% in this study) or high-grade vaginal dysplasia (29% of those biopsied resulted in an abnormal Pap test result and 6.7% of the total number of patients) remains to be determined.
1. Saslow D, Solomon D, Lawson HW, Killackey M, Kulasingam SL, Cain J, et al. American Cancer Society, American Society for Colposcopy and Cervical Pathology, and American Society for Clinical Pathology screening guidelines for the prevention and early detection of cervical cancer. Am J Clin Pathol 2012;137:516–42.
2. Fontaine PL, Saslow D, King VJ. ACS/ASCCP/ASCP guidelines for the early detection of cervical cancer. Am Fam Physician 2012;86:501, 506–7.
3. Cervical cancer screening and prevention. Practice Bulletin No. 157. American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;127:e1–20.
4. Masur H, Brooks JT, Benson CA, Holmes KK, Pau AK, Kaplan JE, et al. Prevention and treatment of opportunistic infections in HIV-infected adults and adolescents: Updated Guidelines from the Centers for Disease Control and Prevention, National Institutes of Health, and HIV Medicine Association of the Infectious Diseases Society of America. Clin Infect Dis 2014;58:1308–11.
5. Aberg JA, Gallant JE, Ghanem KG, Emmanuel P, Zingman BS, Horberg MA, et al. Primary care guidelines for the management of persons infected with HIV: 2013 update by the HIV Medicine Association of the Infectious Diseases Society of America. Clin Infect Dis 2014;58:1–10.
6. Paramsothy P, Duerr A, Heilig CM, Cu-Uvin S, Anderson JR, Schuman P, et al. Abnormal vaginal cytology in HIV-infected and at-risk women after hysterectomy. J Acquir Immune Defic Syndr 2004;35:484–91.
7. Massad LS, Evans C, Weber K, Cejtin HE, Golub ET, DiGilio K, et al. Hysterectomy among women with HIV: indications and incidence. J Acquir Immune Defic Syndr 2007;44:566–8.
8. Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988.
9. Massad LS, Xie X, Greenblatt RM, Minkoff H, Sanchez-Keeland L, Watts DH, et al. Effect of human immunodeficiency virus infection on the prevalence and incidence of vaginal intraepithelial neoplasia. Obstet Gynecol 2012;119:582–9.