Page 365 - HIVMED_v21_i1.indb

Page 3 of 8 Original Research

variables. We assessed all potential risk factors in univariate Of the 80 women with a positive hrHPV test, 6 (7.5%) had a
and multivariable logistic regression models to determine concurrent VIA positive result. Of these women, two were
independent associations. Our final logistic regression model positive for HPV type 16 only, one for HPV 16 and other
did not include VIA data as HPV infection is known to types (channel 5) and three tested positive for other hrHPV
precede VIA status. Incidence rates (IR) and IR ratios for types (2 were channel 3 positive and the other had missing
cervical disease defined as VIA positivity, high-grade channel data).
squamous intraepithelial neoplasia (HGSIL) and cervical
cancer were assessed with the follow-up data. All statistical In the univariate analysis shown in Table 2, women aged
tests were two-tailed with p < 0.05 considered significant 22–29 years (OR 3.0, 95% CI: 1.2–7.9, p = 0.02), women with
and 95% confidence intervals [CI] used. In summary, our a positive VIA within 2 years prior to HPV screening
descriptive, univariate and multivariate analysis were based (OR 6.1, 95% CI: 2.9–12.8, p < 0.001), women on second-line
on all data 100% (n = 321). Follow-up data was based on 96% ART (OR 2.38, 95% CI: 1.3–4.4) and those with a positive
(n = 308) women with hrHPV test results and without VIA at time of HPV screen (OR 6.6, 95% CI: 1.6–27.6,
incident cervical disease. p < 0.001) were more likely to test positive for hrHPV.
Having a detectable viral load at the time of screening and
Ethical consideration sexual debut between the age of 13 and 16 years were also
significantly associated with hrHPV positivity (OR: 2.3, 95%
The study was approved by the Newlands Clinic Research CI: 1.01–5.3, p = 0.04, OR:3.2, CI: 1.3–7.9, p < 0.01). However,
team and the Medical Research Council of Zimbabwe no statistical significance for hrHPV positivity was observed
(Approval MRCZ/A/1980). for CD4 count < 200 cells/mm (OR 1.2, CI: 0.4–4.0). In the
3
multivariate analysis, only younger age (22–29 years),
Results (adjusted odds ratio [aOR] 2.9; 95% CI: 1.0–8.8, p = 0.05),
ART regimen (aOR 2.1; 95% CI: 1.1–4.0, p < 0.05) and early
We recruited 321 women with a median age of 44 years (IQR:
38–50). One hundred and thirty-eight (43.0%) women were sexual debut at ages 13–16 years (aOR 4.0, 95% CI: 1.4–11.5,
p < 0.05) showed evidence of association with hrHPV
married and 29.6% (n = 95) were widowed. Two-thirds of the positivity (Table 2).
women (63.9%, n = 205) were educated to high-school level
(at least 8 years of formal education). Most women (64.5%, Visual inspection with acetic acid follow-up
n = 207) reported sexual debut between 17 and 21 years, and outcomes after high-risk human papillomavirus
13.4% (n = 43) were 16 years or below at sexual debut. testing
Regarding duration on ART, 233 (70.8%) had been on ART for
more than 5 years (median ART duration = 7.8 years, IQR: Three hundred and eight (96%) participants with negative
4.4–10.8 years). In total, 252 (78.5%) were on a first-line ART VIA at the time of hrHPV screen were followed up for a
regimen (1 Non-nucleoside Reverse Transcriptase Inhibitor median of 2 years (645 total person-years of follow-up).
[NNRTI] + 2 Nucleoside/Nucleotide Reverse Transcriptase Follow-up was made to determine incident VIA positivity,
Inhibitor [NRTI]) and had a median CD4 count of 525 cells/ incident HGSIL and incident cervical cancer (confirmed by
3
3
mm (IQR 366–687 cells/mm . One-third (33.7%, n = 108) LEEP biopsy) by HPV infection status at baseline. The VIA
reported that they were unaware of their partner’s HIV status positivity IR in hrHPV-negative women was 0.79 per 100
and 15.6% (n = 50) were in HIV sero-discordant relationships. person-years (95% CI: 0.30–2.11). In hrHPV-positive women,
the VIA positivity IR was 9.97 per 100 person-years (95% CI:
Amongst the 310 women who had HIV viral load results, 5.90–16.83). Incident rate ratio between the two groups of
91.2% (n = 283) had undetectable viral loads (below 50 women was 12.57 (95% CI: 4.14–38.19). There was no evidence
copies/mL) and 2.9% (n = 9) had a viral load above 1000 of incident cervical disease (HGSIL) in women who tested
copies/mL (Table 1).
negative for hrHPV infection. Amongst women with hrHPV
infection, the IR of HGSIL was 6.41 per 100 person-years
All the 321 women had a successful HPV test, 80 (24.9%; 95% (95% CI: 3.33–12.32). During the follow-up period, there was
CI: 20.2% – 29.7%) were positive for any hrHPV. Human no incident of cervical cancer regardless of hrHPV infection
papillomavirus 16 positivity without other coinfections was status (Figure 2).
in 9 (11.2%) women, 10 (12.5%) were positive for HPV 18/45
alone, 52 (65%) for other hrHPV. A small proportion, four (5%) Of the 14 incident VIA-positive women at follow-up, eight
had HPV 16 and other hrHPV coinfections, and five (6.3%) (57%) at baseline were positive for other hrHPV subtypes
had HPV 18 and other hrHPV. There were no co-infections (four being channel 3 subtypes and the others having
with HPV 16 and HPV 18/45. Data for specific channels missing channel data), two were positive for HPV16, one
positive for other hrHPV-positive samples were available for was positive for HPV 16 and other hrHPV subtypes (channel
56/61 samples (91.8%). Amongst these 56 samples, 40 (71.4%) 5). Two were positive for hrHPV 18/45 only and one for
were positive in channel 3 (HPV31/33/35/52/58), 5 (7.1%) in hrHPV 18/45 and other (channel 3). None of the women had
channel 4 (HPV51/59) and 15 (26.8%) in channel 5 (HPV 16/18/45 coinfections at baseline. None of the women with
39/56/66/68). One was positive in both channels 3 and 4, positive channel 4 subtypes had incident cervical disease at
whilst two were positive in both channels 3 and 5 (Figure 1). follow-up.

http://www.sajhivmed.org.za 357 Open Access