Abstract
Background
Most cervical cancers are directly linked to oncogenic or high-risk human papillomavirus (HR-HPV) infection. This study evaluates associations between diet quality and genital HPV infection in women.
Methods
This study included 10 543 women from the 2003–2016 National Health and Nutrition Examination Survey. The outcome was the genital HPV infection status (HPV-negative, low-risk [LR] HPV, and HR-HPV). Dietary quality was evaluated using the Healthy Eating Index (HEI), in which a higher score indicates a better diet quality.
Results
Women who did not consume total fruits (15.8%), whole fruits (27.5%), or green vegetables and beans (43%) had a significantly higher risk of HR-HPV infection than women who complied with the Dietary Guidelines for Americans (HR-HPV odds ratio = 1.76, 1.63, and 1.48 for a HEI score of 0 vs 5, respectively) after adjusting confounding factors. Similar results of these food components on LR-HPV infection were found. In addition, intake of whole grains and dairy was inversely associated with LR-HPV infection.
Conclusions
This study showed that women who did not eat fruits, dark-green vegetables, and beans had a higher risk of genital HR-HPV infection. Intake of these food components is suggested for women to prevent HPV carcinogenesis.
Keywords: antioxidants, cervical cancer, diet, human papillomavirus, infection, nutrition, women
This study evaluates associations between diet quality using the Healthy Eating Index and genital HPV infection in women. The results showed women who did not eat fruits, dark-green vegetables, and beans had an elevated risk of genital oncogenic HPV infection.
Cervical cancer is the second leading cause of cancer death for women aged 20–39 years in the United States [1]. The primary cause of cervical cancer and precancerous lesions of the cervix is human papillomavirus (HPV) infection [2–4]. Almost all cervical cancers (99.7%) are directly linked to the previous infection of oncogenic or high-risk HPVs (HR-HPVs) [5]. With this in mind, HPV infection is common in US women, as approximately 80% are estimated to have at least 1 type of HPV infection in their lifetime [6]. Specifically, the prevalence of any HPV genital infection is 40.3%, and HR-HPV genital infection is 19% in US adult women [7]. Although most HPV infections are asymptotic and around 80% of HPV infections are typically cleared up within 2 years, some persist and progress to cervical cancer [8]. Even though HPV vaccination is available to prevent HPV infection, HPV infection remains a significant threat to women's health due to the low vaccination rate for recommended women and many women not in the recommended age range. Based on 2019 data, the HPV vaccination completion rate among female adolescents is only 57% in the United States [9]. As for women already infected with HPV, routine monitoring tests (the Papanicolaou [Pap] test, HPV test, and HPV/Pap cotest) are suggested, but no effective antiviral drug is available to treat or clear HPV infection [10]. As a result, it is crucial to identify modifiable factors to prevent HPV persistence.
Increasing evidence supports the theory that certain foods (such as vegetables and fruits) and food-derived antioxidants (such as vitamins A, B2, C, and D, and folate) may clear HPV infection and prevent cervical cancer [7, 11–14]. Studies have shown that antioxidant deficiency or oxidative stress affects the host's immunity system and is a potential cofactor of HPV carcinogenesis [11]. For further clarification, oxidative stress, which is the excessive production of reactive oxygen species (ROS), is an imbalance between oxidants and antioxidants in favor of the former [15]. Studies have shown that oxidative stress is involved in cancer initiation, promotion, and progression and may interact with HPV infections [16]. For cervical carcinogenesis, HR-HPV oncoproteins (E6 and E7) are necessary but insufficient for cervical cancer development [16] as cofactors are needed to induce cervical tumor transformation [17]. During HPV infection, immune cells produce ROS to eliminate the intruding virus, but excessive ROS, also known as oxidative stress, can potentially harm host tissues and weaken the immune response. To avoid this damage, antioxidant defense mechanisms are engaged to remove the extra ROS and maintain the oxidation-reduction (redox) status, the balance between the levels of oxidants and antioxidants [18, 19]. Besides antioxidants, dietary patterns also play a role in the immune response against viruses [20]. For instance, the protein found in animal and plant food sources (such as eggs and nuts) is important for maintaining a strong immune system by generating antibodies and other immune cells against viruses. However, most studies are based on a small sample size and a few selected nutrients or dietary factors. Thus, to thoroughly evaluate the quality of a diet, the objective of this study is to assess a wide range of dietary components associated with genital HPV infection in women using the Healthy Eating Index (HEI), which is the tool used to assess diet quality based on how closely it aligns with the Dietary Guidelines for Americans [21].
METHODS
Study Population
The study was based on the 14-year (2003–2016) National Health and Nutrition Examination Survey (NHANES), the cross-sectional nationwide US survey. The NHANES only performed the HPV tests for women aged 18–59 years. Thus, the following criteria were used for study population selection: women, 18–59 years old, with valid genital HPV infection data and HEI diet information. In addition, women with any dose of HPV vaccination or cancer history were excluded. The NHANES data were deidentified and publicly accessible, so no informed consent is needed.
Human Papillomavirus Infection
In this study, HPV infection status reflects HPV infection prevalence, which is the presence of infection at examination time. The measures of genital HPV infection were based on HPV genotyping using DNA samples extracted from self-collected vaginal swabs. A total of 37 genotypes of HPV were measured using the Roche Linear Array. Based on the International Agency for Research on Cancer, these 37 types of HPV are divided into 2 categories: low-risk (LR) HPV and HR or oncogenic HPV [22, 23]. HR-HPV includes 12 types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, and 59), whereas LR-HPV includes the other 25 types (6, 11, 26, 40, 42, 53, 54, 55, 61, 62, 64, 66, 67, 68, 69, 70, 71, 72, 73, 81, 82, 83, 84, IS39, and CP6108). The details of the HPV measures are retained on the NHANES website. This study classified participants into 3 groups: no HPV, LR-HPV, and HR-HPV. The no-HPV group was for those without any HPV infection. The HR-HPV group was defined as those with positive results for any of the 12 HR-HPV types. Similarly, those with any of the 25 LR-HPV types and without HR-HPV infections were included in the LR-HPV group.
Healthy Eating Index Score
Dietary quality is broadly defined as a dietary pattern with the combination of foods and beverages based on a variety of key food groups’ compliance with dietary guidelines [24]. In this study, the HEI-2015 score was applied to evaluate diet quality [25, 26]. These scores were calculated based on two 24-hour dietary intake reports. The 13 HEI components and a total score can be calculated to evaluate adherence to a set of food and critical recommendations of the Dietary Guidelines for Americans [21]. A higher HEI score means better diet quality. Each component has a score range of 0–5 or 0–10; a total score is the sum of these 13 components with a range of 0–100 (Supplementary Table 1). The HEI can be classified into 2 major types: adequacy components and moderation components. For the adequacy components, a higher score refers to a higher intake of the related food. For the moderation components, a higher score indicates a lower related food intake. Total fruits include whole fruits and 100% fruit juice, but whole fruits include all forms of fruits except juice. For the HEI-2015 scoring standards, a full score means dietary quality in the specific component complies with the Dietary Guidelines for Americans [21], and a score of 0 means no intake for 8 out of the 9 adequacy categories: total fruits, whole fruits, total vegetables, greens and beans, whole grains, dairy, total protein foods, and seafood and plant proteins. Other HEI-2015 scoring standards are available in a previous publication [26].
Statistical Analyses
The participant's demographic factors, health-related behaviors, and HEI scores were summarized overall and by genital HPV status using descriptive statistics. The outcome is the grouped genital HPV status: no HPV, LR-HPV, and HR-HPV. Both numeric and categorical HEI component scores associated with HPV status were evaluated. For categorical score analyses, each HEI component score was categorized into 4 levels (Tables 1 and Supplementary Table 4). Most of the cut points were primarily selected based on quartiles. For evaluating associations between categorical factors and the 3 HPV statuses, the Rao-Scott χ2 test was applied. In addition, we tested 1 HEI component at a time using the weighted multinomial logistic models with the HPV status (no HPV, LR-HPV, and HR-HPV) as the outcome. Both unadjusted and adjusted models were applied to evaluate associations between the HEI scores and HPV infection status. For adjusted models, the 9 factors listed in Table 2 were used for adjustment. The details of these 9 factors are listed in the Supplementary Material. We obtained both adjusted and unadjusted ORs (odds ratios) and 95% confidence intervals (CIs) for the HEI components and total score. All analyses were weighted to account for the complex sampling design applied in NHANES using SAS 9.4 (SAS Institute). The weighting analytical procedures using SAS followed the NHANES suggested instruction.
Table 1.
Distributions of Selected Healthy Eating Index (HEI) Scores by Vaginal Human Papillomavirus (HPV) Infection Status in Women
| Component (Range) HEI Score | Total, No. (%) | No HPV, No. (%)a | Low-risk HPV, No. (%)a | High-risk HPV, No. (%)a | P Valueb |
|---|---|---|---|---|---|
| Greens and beans (0–5) | <.001 | ||||
| 0 | 4560 (43.0) | 2266 (53.4) | 1120 (23.4) | 1174 (23.2) | |
| 0.1–2.4 | 1759 (16.9) | 1041 (62.0) | 370 (19.8) | 348 (18.1) | |
| 2.5–4.9 | 1490 (14.2) | 905 (65.0) | 319 (20.1) | 266 (15.0) | |
| 5 | 2734 (25.9) | 1680 (64.2) | 600 (20.3) | 454 (15.6) | |
| Total fruits (0–5) | <.001 | ||||
| 0 | 1592 (15.8) | 717 (46.1) | 450 (26.8) | 425 (27.1) | |
| 0.1–2.4 | 3931 (38.0) | 2135 (58.5) | 908 (21.9) | 888 (19.5) | |
| 2.5–4.9 | 2313 (22.1) | 1355 (62.1) | 516 (21.2) | 442 (16.7) | |
| 5 | 2707 (24.2) | 1685 (66.6) | 535 (17.6) | 487 (15.8) | |
| Whole fruits (0–5) | <.001 | ||||
| 0 | 3060 (27.5) | 1394 (47.5) | 832 (26.1) | 834 (26.3) | |
| 0.1–2.4 | 2399 (23.8) | 1331 (59.9) | 534 (20.7) | 534 (19.4) | |
| 2.5–4.9 | 1600 (15.1) | 951 (63.9) | 350 (20.6) | 299 (15.4) | |
| 5 | 3484 (33.5) | 2216 (66.5) | 693 (18.7) | 575 (14.9) | |
| Whole grains (0–10) | <.001 | ||||
| 0 | 3660 (31.2) | 1853 (54.0) | 939 (24.0) | 868 (22.0) | |
| 0.1–0.9 | 1369 (12.5) | 704 (52.6) | 334 (26.4) | 331 (21.0) | |
| 1.0 −3.4 | 2743 (26.6) | 1619 (63.7) | 577 (18.5) | 547 (17.8) | |
| 3.5–10 | 2771 (29.7) | 1716 (63.8) | 559 (19.5) | 496 (16.7) | |
| Dairy (0–10) | .002 | ||||
| 0–2.4 | 2327 (19.7) | 1220 (54.8) | 587 (25.0) | 520 (20.1) | |
| 2.5–4.9 | 2970 (26.7) | 1632 (57.9) | 699 (22.3) | 639 (19.8) | |
| 5–7.4 | 2329 (23.2) | 1325 (60.6) | 509 (20.2) | 495 (19.2) | |
| 7.5–10 | 2917 (30.4) | 1715 (62.4) | 614 (19.5) | 588 (18.1) | |
| Total score (0–100) | <.001 | ||||
| 0–39.9 | 1867 (16.9) | 849 (48.1) | 509 (26.7) | 509 (25.2) | |
| 40–59.9 | 5688 (52.8) | 3122 (59.3) | 1304 (21.4) | 1262 (19.2) | |
| 60–69.9 | 1870 (18.5) | 1176 (63.1) | 391 (19.3) | 303 (17.6) | |
| 70–79.9 | 836 ( 8.6) | 543 (69.1) | 160 (17.0) | 133 (14.0) | |
| 80–100 | 282 ( 3.2) | 202 (69.8) | 45 (19.9) | 35 (10.2) | |
aRaw frequency (weighted percentage).
bComparison of scores among the 3 HPV groups.
Table 2.
Demographic and Behavioral Factors Associated With Vaginal HPV Status in Women, NHANES Year 2003–2016
| Variable | Total, No. (%)a | Vaginal HPV, No. (%)a | P Valueb | ||
|---|---|---|---|---|---|
| None | Low-risk | High-risk | |||
| Total | 10 543 (100) | 5892 (59.3) | 2409 (21.5) | 2242 (19.2) | |
| Age, y | |||||
| 18–26 | 2410 (17.7) | 1164 (50.9) | 472 (18.2) | 774 (30.9) | <.001 |
| 27–35 | 2230 (20.7) | 1233 (57.2) | 492 (21.7) | 505 (21.1) | |
| 36–45 | 2681 (26.3) | 1577 (59.6) | 631 (22.7) | 473 (17.7) | |
| 46–59 | 3222 (35.3) | 1918 (64.6) | 814 (22.1) | 490 (13.3) | |
| Race | |||||
| Non-Hispanic white | 4042 (63.5) | 2433 (62.7) | 817 (19.2) | 792 (18.0) | <.001 |
| Non-Hispanic black | 2453 (13.5) | 1004 (40.2) | 765 (33.0) | 684 (26.8) | |
| Mexican American and other Hispanic | 3099 (16.0) | 1792 (58.6) | 692 (22.2) | 615 (19.2) | |
| Other | 949 (7.1) | 663 (66.7) | 135 (18.4) | 151 (15.0) | |
| Education | |||||
| <High school | 745 (4.1) | 449 (59.9) | 64 (20.6) | 132 (19.5) | <.001 |
| High school | 3436 (32.1) | 1725 (52.5) | 891 (24.4) | 820 (23.1) | |
| >High school | 5514 (63.8) | 3302 (63.0) | 1196 (20.2) | 1016 (16.8) | |
| Marital status | |||||
| Married/living with a partner | 5959 (63.2) | 3837 (66.9) | 1203 (18.9) | 919 (14.2) | <.001 |
| Never married | 2485 (19.9) | 1144 (49.2) | 560 (20.7) | 781 (30.1) | |
| Widowed/divorced/separated | 1740 (16.9) | 728 (43.5) | 573 (32.0) | 439 (24.5) | |
| Income-to-poverty ratio | |||||
| ≤1 | 2549 (17.8) | 1199 (47.4) | 624 (25.1) | 726 (27.6) | <.001 |
| 1.1–2 | 2411 (19.6) | 1270 (53.3) | 613 (25.2) | 528 (21.5) | |
| 2.1–4 | 2451 (27.5) | 1474 (62.3) | 521 (20.0) | 456 (17.7) | |
| >4 | 2431 (35.2) | 1591 (67.0) | 470 (18.8) | 370 (14.2) | |
| Smoking | |||||
| Never | 6346 (61.0) | 3910 (64.8) | 1274 (18.4) | 1162 (16.8) | <.001 |
| Former | 1438 (16.6) | 821 (61.7) | 370 (23.2) | 247 (15.1) | |
| Current | 2047 (22.4) | 830 (43.6) | 630 (28.8) | 587 (27.6) | |
| Number of days used alcohol in past year | |||||
| 0 | 3279 (30.0) | 2074 (66.7) | 664 (19.3) | 541 (14.0) | <.001 |
| 1–25 | 2968 (35.5) | 1635 (58.7) | 701 (21.9) | 632 (19.4) | |
| >25 | 2369 (34.5) | 1177 (54.5) | 606 (22.9) | 586 (22.6) | |
| Lifetime illegal substance use | |||||
| No | 5177 (49.8) | 3237 (66.1) | 1038 (18.5) | 902 (15.3) | <.001 |
| Yes | 3954 (50.2) | 1904 (52.9) | 1083 (24.7) | 967 (22.4) | |
| Number of sex partners past year | |||||
| 0 | 1404 (14.9) | 873 (65.3) | 334 (22.4) | 197 (12.3) | <.001 |
| 1 | 6289 (72.2) | 3787 (63.1) | 1359 (20.2) | 1143 (16.7) | |
| ≥2 | 1216 (12.9) | 336 (30.4) | 380 (29.5) | 500 (40.2) | |
Abbreviations: HPV, human papillomavirus; NHANES, National Health and Nutrition Examination Survey.
aWeighted percent based on the NHANES sampling weights.
bTested using the Rao-Scott χ2 test.
RESULTS
From the NHANES 2003–2016 survey, 10 543 eligible women were included in this study. The prevalences of HR-HPV, LR-HPV, and HPV negative status were 19.2%, 21.5%, and 59.3%, respectively. The participants’ demographic and behavioral factors overall and by HPV status are shown in Table 2. Among the participating women, 35.3% were aged 46–59 years, 63.5% were non-Hispanic white, and 63.2% were married or lived with a partner. For behavioral factors, 12.9% of women had 2 or more sexual partners during the past year. All demographic and behavioral factors listed in Table 2 showed significant associations with HPV infection status (all P values < .001). While HR-HPV infection was 19.2% overall, the HP-HPV prevalence was high for women aged 18–26 years (30.9%), non-Hispanic black (26.8%), received high school education (23.1%), never married (30.1%), and low income (poverty income ratio ≤ 1, 27.6%). For health-related behaviors, high genital HR-HPV rates were observed in women who were current smokers (27.6%) and had 2 or more sex partners in the past year (40.2%).
As shown in Figure 1 and Supplementary Table 1, the overall mean HEI total score for US women was 53.2 (standard error = 0.27). The HEI total scores were significantly lower for women with HR-HPV and LR-HPV than those without HPV (P < .001 for both; Supplementary Table 2). The means of the HEI total score were 54.6, 51.6, and 50.8 for women with no HPV, LR-HPV, and HR-HPV (Supplementary Table 1). The means of the 13 HEI components and total score by HPV status are shown in Figure 1 and Supplementary Table 1. For the HEI total score treated as a numeric variable, there was a significant inverse relationship between the HEI total score versus HR-HPV (OR = 0.92 per 10-increment score) after adjusting for other confounding factors (Supplementary Table 2). While the overall HR-HPV prevalence was 19.2%, the HR-HPV prevalence was the highest (25.2%) for women with a low-quality diet (HEI total score < 40) and only 10.2% for women with a high-quality diet (HEI total score ≥ 80; Table 1).
Figure 1.
Mean of healthy eating index (HEI) by human papillomavirus (HPV) infection status in women. The vertical axis refers to the average percentage of the HEI scores. The outer edge indicates a score of 100%, the maximum score (5 or 10 for the HEI individual components and 100 for the total score), and the center indicates 0%, the minimum score. The long-dashed line indicates no HPV infection, the short-dashed line low-risk HPV infection, and the solid line high-risk HPV infection. The components that were significantly (P <.05) associated with HPV based on the unadjusted models with categorical HEI scores are indicated by “U.” The significant results based on the adjusted models for HR-HPV and LR-HPV infection are indicated by “AH” and “AL,” respectively. This plot was created based on the tool posted in this website: https://epi.grants.cancer.gov/hei/interpret-visualize-hei-scores.html.
Based on the unadjusted results for the categorical HEI components (Table 2 and Supplementary Table 3), there were 8 HEI components significantly associated with HPV status. After adjusting for the 9 factors listed in Table 2, the 3 components (total fruits, whole fruits, and greens and beans) remained significant for both HR-HPV and LR-HPV. As shown in Supplementary Table 1, the HEI scores of these 3 food categories were lower for women with HR-HPV (1.66–2.06) compared with women without HPV infection (score = 2.19–2.74, P < .001). As shown in Table 1, the 5 HEI components (greens and beans, total fruits, whole fruits, whole grains, and total dairy) and total score were significantly associated with genital HPV infection based on the unadjusted results. HR-HPV was more prevalent (23.0%–27.1%) in women that did not consume any total fruits, whole fruits, or greens and beans than women who consumed sufficient amounts of the same food categories (14.9%–15.8%) based on the dietary guideline: total fruits (≥0.8 cup equivalent per 1000 kcal), whole fruits (≥0.4 cup equivalent per 1000 kcal), and greens and beans (≥0.2 cup equivalent per 1000 kcal). For reference, women with no consumption of the 3 food categories mentioned above had the lowest score of 0, while women that consumed sufficient amounts had the highest score of 5. The results of other components associated with HPV status are listed in Supplementary Table 3.
After adjusting for the 9 factors listed in Table 2, the 3 HEI components (greens and beans, total fruits, and whole fruits) remained significantly inversely associated with both LR-HPV and HR-HPV, and 2 components (whole grains and total dairy) were inversely associated with LR-HPV but not HR-HPV (Table 3). Women who ate greens and beans (score > 0) had a significantly lower risk of HPV infection compared with women who did not eat these categories of food (HR-HPV OR = 0.68, LR-HPV OR = 0.8, for score 5 vs 0). In addition, women who consumed total fruits had a significantly lower risk of HPV infection compared with women who did not eat total fruits (HR-HPV OR = 0.57, LR-HPV OR = 0.54, for score 5 vs 0). Similar inverse associations were observed for whole fruits. Moreover, we were interested in testing whether there was a dosage effect of the intake of greens and beans, total fruits, and whole fruits regarding HPV infection risk. Thus, we performed extra analyses with the highest score level (score = 5) as the reference. As shown in Supplementary Table 5, women with a moderate score (0.1–2.4 and 2.5–4.9) for these 3 food categories had a similar risk of HR-HPV and LR-HPV compared with those with the full score (score = 5).
Table 3.
Selected HEI Scores Associated With Vaginal HPV Infection Status in Women.
| Component (range) HEI Score | Unadjusteda | Adjusteda | ||
|---|---|---|---|---|
| Low-risk HPV vs No HPV, OR (95% CI) | High-risk HPV vs No HPV, OR (95% CI) | Low-risk HPV vs No HPV, OR (95% CI) | High-risk HPV vs No HPV, OR (95% CI) | |
| Greens and beans (0–5) | ||||
| 0 | 1 | 1 | 1 | 1 |
| 0.1–2.4 | 0.73 (.60–.89)** | 0.68 (.56–.81)*** | 0.81 (.63–1.03) | 0.72 (.57–.92)** |
| 2.5–4.9 | 0.71 (.57–.87)*** | 0.53 (.43–.66)*** | 0.78 (.61–.99)* | 0.61 (.47–.80)*** |
| 5 | 0.72 (.61–.85)*** | 0.56 (.48–.66)*** | 0.80 (.66–.98)* | 0.68 (.55–.83)*** |
| Total fruits (0–5) | ||||
| 0 | 1 | 1 | 1 | 1 |
| 0.1–2.4 | 0.64 (.52–.8)*** | 0.57 (.46–.70)*** | 0.76 (.60–.96)* | 0.68 (.51–.89)** |
| 2.5–4.9 | 0.59 (.47–.73)*** | 0.46 (.36–.58)*** | 0.67 (.52–.87)** | 0.61 (.45–.85)** |
| 5 | 0.46 (.36–.57)*** | 0.40 (.32–.51)*** | 0.54 (.41–.72)*** | 0.57 (.42–.78)*** |
| Whole fruits (0–5) | ||||
| 0 | 1 | 1 | 1 | 1 |
| 0.1–2.4 | 0.63 (.52–.76)*** | 0.58 (.47–.72)*** | 0.68 (.56–.84)*** | 0.73 (.56–.95)* |
| 2.5–4.9 | 0.59 (.46–.76)*** | 0.44 (.34–.55)*** | 0.67 (.51–.88)** | 0.57 (.42–.78)*** |
| 5 | 0.51 (.42–.63)*** | 0.40 (.32–.50)*** | 0.60 (.48–.76)*** | 0.62 (.47–.81)*** |
| Whole grains (0–10) | ||||
| 0 | 1 | 1 | 1 | 1 |
| 0.1–0.9 | 1.13 (.92–1.40) | 0.98 (.81–1.19) | 1.16 (.89–1.50) | 1.12 (.86–1.45) |
| 1.0–3.4 | 0.65 (.55–.77)*** | 0.69 (.58–.81)*** | 0.67 (.54–.83)*** | 0.82 (.66–1.03) |
| 3.5–10 | 0.69 (.59–.80)*** | 0.65 (.53–.78)*** | 0.82 (.69–.98)* | 0.87 (.67–1.12) |
| Dairy (0–10) | ||||
| 0–2.4 | 1 | 1 | 1 | 1 |
| 2.5–4.9 | 0.85 (.72–1.00)* | 0.93 (.77–1.13) | 0.86 (.70–1.06) | 1.07 (.83–1.38) |
| 5.0–7.4 | 0.73 (.60–.89)** | 0.86 (.70–1.06) | 0.76 (.61–.93)** | 0.96 (.74–1.25) |
| 7.5–10 | 0.68 (.57–.82)*** | 0.79 (.64–.98)* | 0.74 (.60–.90)** | 0.89 (.70–1.15) |
| Total score (0–100) | ||||
| 0–39.9 | 1.54 (1.30–1.82)*** | 1.62 (1.35–1.94)*** | 1.36 (1.10–1.69)** | 1.27 (1.01–1.60)* |
| 40–59.9 | 1 | 1 | 1 | 1 |
| 60–69.9 | 0.85 (.69–1.04) | 0.86 (.71–1.04) | 0.89 (.70–1.13) | 1.01 (.80–1.29) |
| 70–79.9 | 0.68 (.52–.89)** | 0.62 (.46–.84)** | 0.77 (.55–1.06) | 0.87 (.62–1.21) |
| 80–100 | 0.79 (.46–1.34) | 0.45 (.26–.77)** | 1.06 (.58–1.94) | 0.70 (.38–1.31) |
Abbreviations: CI, confidence interval; HEI, Healthy Eating Index; HPV, human papillomavirus; OR, odds ratio.
aOdds ratio (95% CI), multinomial logistic models adjusted for 9 factors (age, race, education, income-to-poverty ratio, marital status, smoking status, lifetime illegal substance use, past 12-months alcohol intake, number of sexual partners in past 12 months).
* P < .05, ** P < .01; *** P < .001 based on multinomial logistic models, bold: adjusted P < .05 for high-risk HPV .
After adjusting the 9 factors, 2 HEI components (whole grains and total dairy) had an inverse association with LR-HPV but not HR-HPV. As shown in Table 3, women with a higher whole grain intake had a lower risk of LR-HPV infection (score = 3.5–10, OR = 0.82, P < .05) compared with women who did not eat whole grains (score = 0) after adjusting for the 9 factors. Similarly, women with a higher intake of total dairy had a lower risk of LR-HPV infection (score = 7.5–10, OR = 0.74, P < .001) than those with a lower intake of the total dairy (score = 0–2.5). For the HEI total score, we used the group with a HEI total score of 40–60 as the reference because the average HEI score (53.2) overall was in this range. After adjusting for other factors (Table 3), women with a low-diet quality (HEI total score < 40) had a higher risk of both HR-HPV and LR-HPV infection compared with those with a score of 40–60 (HR-HPV OR = 1.27, LR-HPV OR = 1.36, both P < .001). The results of other HEI components, which were not significantly associated with genital HPV infection, are listed in Supplementary Table 4. When testing numeric HEI scores associated with HPV infection, similar results were found (Supplementary Table 2).
For the food components related to genital HPV infection, we observed that women who ate healthily tended to practice other health behaviors. For example, as shown in Supplementary Tables 6–8, women with a maximum score of 5 in total fruits had a lower chance of being current smokers (11.1% vs 43.6%, P < .001), frequent alcohol drinkers (>25 days used alcohol during the past year, 31.7% vs 34.6%, P < .005), and illegal substances user in their lifetime (42.0% vs 60.4%, P < .001) than those who did not eat fruits. In addition, women who ate more total fruits were less likely to perform risky sexual behavior than those who did not eat fruits (≥2 sex partners during the past year, 9.1% vs 18.0% for a score of 5 vs 0, P < .001; Supplementary Table 9).
DISCUSSION
This study shows that the intake of fruits, dark-green vegetables, and beans is associated with a lower risk of LR-HPV and HR-HPV infection for US women based on a nationwide US survey. For US women aged 15–59 years, the prevalence of any HPV infection was 40.7%, and HR-HPV infection was 19.2%, which is consistent with a previous study [7]. The intake of whole grains and dairy was inversely associated with LR-HPV but not HR-HPV. While HR-HPV prevalence for US women was 19.2%, women who did not eat greens and beans or fruit (HEI score = 0 for these components) had a significantly higher HR-HPV prevalence (23.2%–27.1%) than those who ate food in these categories. It should also be noted that these associations remain significant even after adjusting for the 9 confounding factors. Generally, US women have low HEI scores, with less than half of the optimal score of 5, in greens and beans and fruits (greens and beans, 2.02; total fruits, 2.41; and whole fruits, 2.48). In fact, many US women do not eat any greens and beans and fruits (HEI component score = 0). As shown in Table 1, around 43% of women did not eat any greens and beans, 27.5% did not eat any whole fruits, and 15.8% did not have any total fruits (whole fruits and juice). Moreover, these HEI scores were even lower for women with HR-HPV (greens and beans, 1.66; total fruits, 2.06; and whole fruits, 1.98).
Our results showing that an increase in fruits and greens and beans leads to a decrease in HPV infection are consistent with our previous antioxidant study, which indicated the 4 dietary antioxidants (vitamins A, B2, E, and folate) were inversely associated with HR-HPV infection [7]. The 4 dietary antioxidants can be found in dark-green vegetables (such as spinach, kale, and broccoli), beans (such as lima beans, peas, soybeans, and black beans), and fruits (such as oranges, grapes, blueberry, and mango). In addition, the Mediterranean diet, characterized by the intake of vegetables, legumes, fruits, whole grains, fish, and olive oil, was associated with a lower risk of HR-HPV [27]. Additionally, studies showed that fruits and greens and beans contain vitamin C, which can potentially reduce genital HPV infection and cervical cancer incidences in women [28, 29]. For example, a longitudinal study showed that total and whole fruit intake based on the HEI-2010 scores could enhance HR-HPV clearance [30]. Furthermore, in an in vitro study, the protective effect of Indian fig fruit extract was shown to reduce the HPV-18 viral load and to show effective cytotoxicity in an HPV-18–containing cervical cancer cell line [31]. Whereas most previous related studies did not directly test the dosage effect of fruits and vegetables on HPV infection [30, 32, 33], this study showed that varying amounts of fruits and vegetables consumed did not have a significant impact on HR-HPV infection for women who ate fruits or greens and beans based on the multivariable models. However, women who did not eat fruits or greens and beans had a higher risk of HR-HPV. Thus, this result indicates that any intake of fruits and greens and beans has similar benefits in lowering the risk of HR-HPV infection.
The potential biological mechanism of fruits, dark-green vegetables, and beans on inhibiting HPV infection may be through enhancing immune response and decreasing inflammation. In the immune system response, nutrition is a critical factor because malnutrition (especially low intake of fruits and vegetables) can cause a weakened immune system, which leads to a failure to clear HPV infection. As a result, it leads to HPV persistence and an increased potential of progression to cervical intraepithelial neoplasia and cervical cancer [34]. Furthermore, dietary intake can modify the physiological responses to chronic inflammation, which plays an essential role in cervical cancer development [34]. For example, Mediterranean dietary patterns, including a high intake of fruits, vegetables, and whole grains, benefit from decreased inflammation [35].
In addition to the dietary quality of fruits and greens and beans, this study shows that a higher intake of dairy products (such as milk, yogurt, and cheese) and whole grains are associated with a lower risk of LR-HPV, which can cause genital warts. In addition, LR-HPV types often have a high recurrence rate and no antiviral drug treatment [36]. Furthermore, vitamin K, which can be found in dairy, green vegetables, and fruits, was shown to be inversely associated with HPV infection (both LR-HPV and HR-HPV) in women [37]. Similarly, it also showed that the intake of whole grains could reduce C-reactive protein levels, an inflammatory biomarker [38]. Additionally, whole grains containing numerous antioxidants and nutrients (such as folate) are a key part of the Mediterranean diet, associated with a lower risk of HR-HPV [27].
It is important to note that there are several strengths of this study. Large-scale population-based NHANES data were used so that the study findings are representative of the US female population. In addition, the large sample size and abundance of HPV health-related information in this survey allowed for the control of potential confounders in the modeling and categorizing of HEI scores. Thus, the dosage effect of the HEI scores could be tested. Nevertheless, this study still has potential limitations. First, the HEI scores were calculated based on the self-reported two 24-hour dietary intake reports, so recall bias may exist, and the 2 interview days may not be representative of the general diet pattern for the interviewees. Second, this study cannot infer a causal association between diet quality and HPV infection because of the nature of the cross-sectional survey. Finally, this cross-sectional study, with an HPV test per person, can only measure HPV prevalence, and so it cannot distinguish between transient and persistent infection with HPV, which are usually measured using ≥2 consecutive HPV tests.
In summary, this study highlights that the insufficient consumption of fruits, dark-green vegetables, and beans can significantly increase the risk of genital HR-HPV infection in women. Therefore, health care providers should suggest eating fruits, dark-green vegetables, and beans, based on the dietary guidelines for women, to prevent HPV infection or persistence. In the Dietary Guidelines for Americans [21], a range of 1600 to 2400 calories per day is suggested for adult women. Using the average of 2000 calories per day, 0.8 cups (1 cup = 8 ounces) of whole fruits, 1.6 cups of total fruits, and 0.4 cups of greens and beans are suggested to be consumed daily for US women. Moreover, we also identified the clustering of an unhealthy diet, smoking, illegal substance use, and risky sexual behavior in women, meaning that women participating in 1 risky behavior are more likely to have others. Thus, health care providers should also consider integrated interventions targeting multiple risky behaviors. Together, our findings can contribute to understanding the impact of dietary quality on HPV infection and provide valuable insight into developing treatments and interventions for preventing HPV persistence and cervical cancer. Further prospective longitudinal studies will be needed to confirm the causal effects of dietary quality on HPV infection in women.
Supplementary Data
Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Supplementary Material
Contributor Information
Hui-Yi Lin, Biostatistics Program, School of Public Health, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
Qiufan Fu, Biostatistics Program, School of Public Health, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
Tung-sung Tseng, Behavior and Community Health Sciences Program, School of Public Health, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
Xiaodan Zhu, Biostatistics Program, School of Public Health, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
Krzysztof Reiss, Department of Hematology and Oncology, School of Medicine, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
L Joseph Su, School of Public Health, University of Texas Southwestern Medical Center, Dallas, Texas, USA.
Michael E Hagensee, Section of Infection Diseases, Department of Medicine, School of Medicine, Louisiana State University Health Sciences Center, New Orleans, Louisiana, USA.
Notes
Financial support. This study was supported by the Center for Translational Viral Oncology, funded by the National Institutes of Health, National Institute of General Medical Sciences (grant number P20 GM121288 to K. R. and H. L.).
References
- 1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin 2022; 72:7–33. [DOI] [PubMed] [Google Scholar]
- 2. Koshiol J, Lindsay L, Pimenta JM, Poole C, Jenkins D, Smith JS. Persistent human papillomavirus infection and cervical neoplasia: a systematic review and meta-analysis. Am J Epidemiol 2008; 168:123–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Baseman JG, Koutsky LA. The epidemiology of human papillomavirus infections. J Clin Virol 2005; 32(Suppl 1):S16–24. [DOI] [PubMed] [Google Scholar]
- 4. de Sanjose S, Brotons M, Pavon MA. The natural history of human papillomavirus infection. Best Pract Res Clin Obstet Gynaecol 2018; 47:2–13. [DOI] [PubMed] [Google Scholar]
- 5. Yim EK, Park JS. The role of HPV E6 and E7 oncoproteins in HPV-associated cervical carcinogenesis. Cancer Res Treat 2005; 37:319–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Myers ER, McCrory DC, Nanda K, Bastian L, Matchar DB. Mathematical model for the natural history of human papillomavirus infection and cervical carcinogenesis. Am J Epidemiol 2000; 151:1158–71. [DOI] [PubMed] [Google Scholar]
- 7. Lin HY, Fu Q, Kao YH, et al. Antioxidants associated with oncogenic HPV infection in women. J Infect Dis 2021; 224:1520–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jaisamrarn U, Castellsague X, Garland SM, et al. Natural history of progression of HPV infection to cervical lesion or clearance: analysis of the control arm of the large, randomised PATRICIA study. PLoS One 2013; 8:e79260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Elam-Evans LD, Yankey D, Singleton JA, et al. National, regional, state, and selected local area vaccination coverage among adolescents aged 13–17 years—United States, 2019. MMWR Morb Mortal Wkly Rep 2020; 69:1109–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Liu Y, Li H, Pi R, Yang Y, Zhao X, Qi X. Current strategies against persistent human papillomavirus infection (review). Int J Oncol 2019; 55:570–84. [DOI] [PubMed] [Google Scholar]
- 11. Ono A, Koshiyama M, Nakagawa M, et al. The preventive effect of dietary antioxidants on cervical cancer development. Medicina (Kaunas) 2020; 56:604. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Barchitta M, Maugeri A, La Mastra C, et al. Dietary antioxidant intake and human papillomavirus infection: evidence from a cross-sectional study in Italy. Nutrients 2020; 12:1384. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Siegel EM, Salemi JL, Villa LL, Ferenczy A, Franco EL, Giuliano AR. Dietary consumption of antioxidant nutrients and risk of incident cervical intraepithelial neoplasia. Gynecol Oncol 2010; 118:289–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Feng CY, Lin M, Lakhaney D, et al. The association between dietary intake and cervical intraepithelial neoplasia grade 2 or higher among women in a high-risk rural area of China. Arch Gynecol Obstet 2011; 284:973–80. [DOI] [PubMed] [Google Scholar]
- 15. Georgescu SR, Mitran CI, Mitran MI, et al. New insights in the pathogenesis of HPV infection and the associated carcinogenic processes: the role of chronic inflammation and oxidative stress. J Immunol Res 2018; 2018:5315816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Calaf GM, Urzua U, Termini L, Aguayo F. Oxidative stress in female cancers. Oncotarget 2018; 9:23824–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. De Marco F. Oxidative stress and HPV carcinogenesis. Viruses 2013; 5:708–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Checconi P, De Angelis M, Marcocci ME, et al. Redox-Modulating agents in the treatment of viral infections. Int J Mol Sci 2020; 21:4084. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Tavassolifar MJ, Vodjgani M, Salehi Z, Izad M. The influence of reactive oxygen species in the immune system and pathogenesis of multiple sclerosis. Autoimmune Dis 2020; 2020:5793817. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Thirumdas R, Kothakota A, Pandiselvam R, Bahrami A, Barba FJ. Role of food nutrients and supplementation in fighting against viral infections and boosting immunity: a review. Trends Food Sci Technol 2021; 110:66–77. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. US Department of Agriculture and US Department of Health and Human Services . Dietary guidelines for Americans, 2020–2025, 9th edn. https://www.dietaryguidelines.gov/. Accessed 10 January 2022.
- 22. Munoz N, Castellsague X, de Gonzalez AB, Gissmann L. Chapter 1: HPV in the etiology of human cancer. Vaccine 2006; 24(Suppl 3):S3/1–10. [DOI] [PubMed] [Google Scholar]
- 23. World Health Organization . IARC Monograph on the evaluation of carcinogenic risks to humans: human papillomaviruses, Vol 90. Lyon: International Agency for Research on Cancer, 2007. [Google Scholar]
- 24. Wirt A, Collins CE. Diet quality—what is it and does it matter? Public Health Nutr 2009; 12:2473–92. [DOI] [PubMed] [Google Scholar]
- 25. Kennedy ET, Ohls J, Carlson S, Fleming K. The healthy eating index: design and applications. J Am Diet Assoc 1995; 95:1103–8. [DOI] [PubMed] [Google Scholar]
- 26. Krebs-Smith SM, Pannucci TE, Subar AF, et al. Update of the healthy eating index: HEI-2015. J Acad Nutr Diet 2018; 118:1591–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Barchitta M, Maugeri A, Quattrocchi A, Agrifoglio O, Scalisi A, Agodi A. The association of dietary patterns with high-risk human papillomavirus infection and cervical cancer: a cross-sectional study in Italy. Nutrients 2018; 10:469. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Koshiyama M. The effects of the dietary and nutrient intake on gynecologic cancers. Healthcare (Basel) 2019; 7:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Giuliano AR, Siegel EM, Roe DJ, et al. Dietary intake and risk of persistent human papillomavirus (HPV) infection: the Ludwig-McGill HPV Natural History Study. J Infect Dis 2003; 188:1508–16. [DOI] [PubMed] [Google Scholar]
- 30. Naresh A, Hagensee M, Myers L, Cameron J. Association of diet quality and dietary components with clinical resolution of HPV. Nutr Cancer 2021; 73:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Grace VMB BL, Wilson DD. The effect of Indian fig fruit extract on human papilloma virus containing cervical cancer cells (HeLa) by decreasing the HPV18 L1 gene load. Asian Pac J Cancer Prev 2021; 22:785–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Tomita LY, Horta BL, da Silva LLS, Malta MB, Franco EL, Cardoso MA. Fruits and vegetables and cervical cancer: a systematic review and meta-analysis. Nutr Cancer 2021; 73:62–74. [DOI] [PubMed] [Google Scholar]
- 33. Hwang JH, Lee JK, Kim TJ, Kim MK. The association between fruit and vegetable consumption and HPV viral load in high-risk HPV-positive women with cervical intraepithelial neoplasia. Cancer Causes Control 2010; 21:51–9. [DOI] [PubMed] [Google Scholar]
- 34. Medina-Contreras O, Luvian-Morales J, Valdez-Palomares F, et al. Immunonutrition in cervical cancer: immune response modulation by diet. Rev Invest Clin 2020; 72:219–30. [DOI] [PubMed] [Google Scholar]
- 35. Silva AR, Moraes BPT, Goncalves-de-Albuquerque CF. Mediterranean diet: lipids, inflammation, and malaria infection. Int J Mol Sci 2020; 21:4489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Soheili M, Keyvani H, Soheili M, Nasseri S. Human papilloma virus: a review study of epidemiology, carcinogenesis, diagnostic methods, and treatment of all HPV-related cancers. Med J Islam Repub Iran 2021; 35:65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Jiang Y, Xu S, Lan J, Zhang J, Chen T. Dietary vitamin K intake and HPV-infection status among American women: a secondary analysis from national health and nutrition examination survey data from 2003 to 2016. Int J Public Health 2022; 67:1604616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Taskinen RE, Hantunen S, Tuomainen TP, Virtanen JK. The associations between whole grain and refined grain intakes and serum C-reactive protein. Eur J Clin Nutr 2022; 76:544–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.