Adulthood consumption of preserved and nonpreserved vegetables and the risk of nasopharyngeal carcinoma: A systematic review

Authors

  • Lisa Gallicchio,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Genevieve Matanoski,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Xuguang (Grant) Tao,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    2. Division of Occupational and Environmental Medicine, Johns Hopkins School of Medicine, Baltimore, MD, USA
    Search for more papers by this author
  • Liwei Chen,

    1. Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    2. Center for Human Nutrition, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Tram K. Lam,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Kristina Boyd,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Karen A. Robinson,

    1. Department of General Internal Medicine, Johns Hopkins School of Medicine, Baltimore, MD, USA
    Search for more papers by this author
  • Lyssa Balick,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Stephanie Mickelson,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Laura E. Caulfield,

    1. Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    2. Center for Human Nutrition, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • James G. Herman,

    1. Department of Oncology, Johns Hopkins School of Medicine, Baltimore, MD, USA
    Search for more papers by this author
  • Eliseo Guallar,

    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    Search for more papers by this author
  • Anthony J. Alberg

    Corresponding author
    1. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
    2. Department of Oncology, Johns Hopkins School of Medicine, Baltimore, MD, USA
    • Hollings Cancer Center, Medical University of South Carolina, 86 Jonathan Lucas Street, P.O. Box 250955, Charleston, SC 29425 USA
    Search for more papers by this author
    • Fax: +843-792-9588.


Abstract

The incidence rates of nasopharyngeal carcinoma (NPC) are dramatically higher in certain regions of Asia compared to the rest of the world. Few risk factors for NPC are known; however, in contrast to the hypothesized health benefits of nonpreserved vegetables, it is thought that preserved vegetable intake may play a role in contributing to the higher incidence of NPC in high-risk regions. Therefore, the purpose of this study was to systematically review the epidemiologic evidence on the associations between adulthood intake of preserved and nonpreserved vegetables and NPC risk. A search of the epidemiological literature from 1966 to 2004 was performed using several bibliographic databases, including PubMed and the Chinese Biomedical Literature Database System. There were no language restrictions. Meta-analysis was conducted to obtain pooled odds ratios (ORs) for the highest-versus-lowest categories of preserved and nonpreserved vegetable intake. A total of 16 case-control studies were identified in the search. Results showed that highest-versus-lowest preserved vegetable intake was associated with a 2-fold increase in the risk of NPC (Random Effects Odds Ratio (RE OR) 2.04; 95% Confidence Limits (CL) 1.43, 2.92). Conversely, high nonpreserved vegetable intake was associated with 36% decrease in the risk of NPC (RE OR 0.64; 95% CL 0.48, 0.85). Findings for both preserved and nonpreserved vegetables were consistent across vegetable type and by country of study. Further research in high-risk areas to gain insight into the risk associated with preserved vegetables and protection associated with nonpreserved vegetables may advance understanding of NPC and yield clues for prevention. © 2006 Wiley-Liss, Inc.

Nasopharyngeal carcinoma (NPC) is a rare malignancy in most parts of the world, with age-adjusted incidence rates under 1 per 100,000 persons per year for both men and women.1 In certain geographical regions, however, the incidence of NPC is dramatically higher. The highest rates of NPC have been documented among Cantonese residing in the central region of the Guangdong Province in Southern China and in Hong Kong (25–30 per 100,000 persons per year).2 Intermediate incidence rates (2–6 per 100,000 persons per year) have been noted in other areas of Southeast Asia such as Taiwan, Malaysia, Vietnam, Micronesia and Guam.1

Risk factors for NPC include Epstein-Barr Virus (EBV) infection3, 4, 5, 6 and specific HLA antigen haplotypes.7 In addition, certain dietary factors may partially explain the variation in NPC rates across geographic regions. The consumption of Cantonese-style salted fish has consistently been found to be associated with an increased risk of NPC.8, 9, 10, 11, 12, 13, 14 Cantonese-style salted fish contains high levels of volatile nitrosamines produced during the preservation process,15, 16 which have been shown in animal and laboratory studies to be both mutagenic and carcinogenic.17 Cantonese-style salted fish, however, is not frequently consumed in many communities of Southern China.2 Hence, it has been hypothesized that consumption of other preserved foods may contribute to the burden of NPC in high-risk regions.2

Preserved vegetables, or vegetables that are salted, dried, canned, fermented or pickled, are commonly consumed in high-risk NPC areas. The preservation processes used to prepare these types of vegetables are also thought to increase the levels of carcinogenic nitrosamines,2 such as N-nitrosodimethylamine and other volatile N-nitroso compounds.18 On the other hand, nonpreserved vegetables contain many components that may prevent cancer, such as fiber, carotenoids, vitamin C, folate and other vitamins.19

The few published reviews that have mentioned the relationship between preserved and nonpreserved vegetable intake and the risk of developing NPC were part of larger reviews on either the epidemiology of NPC or the effects of dietary intake on cancer risk,2, 6, 19 and did not characterize the magnitude of the associations between nonpreserved and preserved vegetable intake and NPC. Therefore, the purpose of this report was to systematically review the epidemiologic evidence on the associations between preserved and nonpreserved vegetable intake and the risk of developing NPC.

Abbreviations:

AICR, American Institute for Cancer Research; EBV, Epstein-Barr Virus; FFQ, food frequency questionnaire; NPC, nasopharyngeal carcinoma; OR, odds ratio; RE OR, random effects odds ratio; WCRF, World Cancer Research Fund.

Material and methods

This work was carried out as a part of a project funded by the World Cancer Research Fund (WCRF) and the American Institute for Cancer Research (AICR) to develop a report entitled “Food, Nutrition, Physical Activity and the Prevention of Cancer: A Global Perspective.” All of the work funded under this project was conducted using a standardized protocol developed by WCRF (http://www.wcrf.org/research/second_wcrf_aicr_report.lasso). While all work contained in this study was performed using the WCRF protocol, the conclusions of this study may differ from those in the WCRF report, as the WCRF report includes other data and uses different criteria for judgment.

Study search

For this specific report, we sought all evidence on the association between vegetable intake in adults and NPC reported in randomized clinical trials, cohort studies and case-control studies. Using the search strategy listed in the Appendix, the following electronic databases were searched: PubMed, Embase, Pascal, ISI Web of Science, NIAAA Alcohol and Alcohol Problems Science, The Cochrane Library, Biological Abstracts, Cumulative Index to Nursing and Allied Health Literature, Index Medicus for WHO Eastern Mediterranean Region, Index Medicus for South East Asian Region, African Index Medicus and Latin American and Caribbean Center on Health Sciences Information. Because of the high NPC burden in China, we also searched the Chinese Biomedical Literature Database System. The search period was 1966 through October 2004. There were no language restrictions. In addition, the study team hand-searched the references cited in the 1997 WCRF report,19 in the articles chosen for data abstraction, and in the relevant review articles or meta-analyses identified in the PubMed search.

Study selection

The following exclusion criteria were applied to the abstracts identified in the literature search: (i) no original data (reviews, editorials, meta-analyses); (ii) studies not addressing the association between vegetable intake and NPC; (iii) studies not in humans; and (iv) case reports and case series. The full-text articles of all references selected after applying these criteria were reviewed. The following exclusion criteria were applied to full-text articles: (i) no measure of vegetable intake at the individual level (ecological study); (ii) no assessment of vegetable intake in adulthood; and, (iii) measures of association or variability not reported or could not be calculated using the data provided. If separate reports from the same study were published, the report with the most updated data was selected for inclusion or, in the case of duplicate publication, only 1 publication was included. The eligibility of each abstract or full-text article was assessed independently in a standardized manner by 2 reviewers.

Data abstraction

Data abstraction for selected articles was performed serially by 2 reviewers using an electronic abstraction database created by WCRF. Disagreements between reviewers were resolved by consensus.

In the studies included in this systematic review, many categories of vegetable consumption were examined. During the data abstraction, a specific vegetable was noted and each vegetable was classified into 1 of the following specific subgroups defined a priori: nonstarchy root vegetables and tubers (e.g. carrots), cruciferous vegetables (e.g., cabbage, mustard greens, cauliflower), allium vegetables (e.g., onions, garlic), green leafy vegetables (not including cruciferous vegetables; e.g., spinach, seaweed) and “other nonstarchy vegetables,” which was a default category for vegetables not included in 1 of the subgroups listed above. To select the vegetable subgroup, each reviewer followed a standardized protocol developed by WCRF (http://www.wcrf.org/research/second_wcrf_aicr_report.lasso). If a specific vegetable was not listed in the standardized protocol, a nutritionist assigned the appropriate subgroup. Data on the preservation status of each vegetable were also recorded. Vegetables were considered “preserved” if they were described as preserved, fermented, salted, canned, pickled or dried. Vegetables were considered “nonpreserved” if they contained no adjectives in their names or contained the words “fresh” or “raw.”

To assess study quality, we adapted the criteria used by Longnecker et al20 for observational studies. The criteria used were whether (i) data were collected in a similar manner for cases and controls, (ii) all cases were interviewed within 6 months of diagnosis, (iii) the same interview schedule was used for cases and controls, (iv) the interviewers were blinded with respect to the case-control status of the person interviewed, (v) the time period over which the cases and controls were interviewed was the same, (vi) the same exclusion criteria was applied to cases and controls, (vii) the response rate among controls was at least 70%, (viii) the diagnosis of cancer was histologically confirmed in at least 90% of the cases, (ix) the controls were people who, had they developed the disease, would have been cases, (x) the dietary questionnaire used was validated, (xi) if the study was a matched case-control study, the authors either conducted a matched analysis or showed that an unmatched analysis was equivalent to a matched analysis and presented an unmatched analysis and, (xii) the authors controlled for potential confounding or examined potential effect modification by the NPC risk factors, EBV and salted fish intake.

Statistical analysis

All analyses were stratified according to whether the vegetables were preserved or nonpreserved. For descriptive purposes, the range of odds ratio estimates and the unweighted means are reported for both preserved and nonpreserved vegetable intake. The initial quantitative analyses focused on the broadest categories of preserved and nonpreserved vegetable intake, which served as estimates of “total vegetable” intake. The relevant categories of preserved vegetable intake were: “all preserved vegetables,” “salted leafy vegetables,” “canned, salted, or pickled vegetables,” and “vegetables in brine as seasoning.” The applicable exposure categories of nonpreserved vegetable intake were “all vegetables,” “all fresh vegetables,” and “vegetables.” More refined analyses were then performed using the specific vegetable subgroups of cruciferous vegetables, green-leafy vegetables, nonstarchy vegetables and nonstarchy roots and tubers. No study reported separate results for allium vegetables.

We examined the random effects odds ratio (RE OR) and 95% confidence limits (95% CL) for NPC comparing the highest-versus-lowest categories of vegetable intake. Pooled OR estimates were obtained using inverse-variance weights in random effects models. Statistical heterogeneity was assessed using the DerSimonian and Laird's Q statistic and the I2 statistic. Subgroup analyses were conducted by restricting the analysis to studies conducted in China. Sensitivity analyses to examine the influence of each individual study were conducted by excluding each study from the meta-analysis and comparing the point estimates including and excluding the study. Publication bias was examined using funnel plots.

To explore the sources of statistical heterogeneity in the meta-analyses, meta-regression analyses were performed. Variables included in the analysis were: country of study (China/other), number of cases (<150/≥150), source of controls (population/hospital), statistical adjustment for foods other than vegetables (yes/no), histological confirmation of cases (yes/no) and method of dietary assessment (food frequency questionnaire/other).

Results

Search results

The search on dietary intake and NPC yielded 3,225 references; 1,862 were identified in PubMed (Fig. 1). Of these references, 3,196 were excluded after abstract review. Of the 29 articles that were obtained for full-text review, we excluded 2 that did not measure vegetable intake at the individual level,21, 22 1 that did not assess vegetable intake in adulthood,10 5 that were duplicate publications,23, 24, 25, 26, 27 and 5 that did not report a measure of association or variability for the examined vegetable/NPC association.9, 11, 28, 29, 30 Sixteen studies met the inclusion criteria and were included in the meta-analysis. Six of these studies (37.5%) were published in Chinese language journals.

Figure 1.

Study selection process. aNumber of references are from search for diet and NPC; searches in other databases (ISI Web of Science, Cochrane Library, Biological Abstracts, Index Medicus for WHO Eastern Mediterranean Region, Index Medicus for South East Asian Region, African Index Medicus and Latin American and Caribbean Center on Health Sciences Information) did not identify any additional references.

Study characteristics

All 16 of the included studies were case-control studies (Table I). Eleven of the studies enrolled 150 cases or less,12, 32, 35, 36, 37, 38, 39, 40, 41, 42 whereas the largest study was comprised of 935 cases.34 Twelve studies stated that the NPC cases under study were histologically confirmed.14, 31, 32, 33, 34, 37, 38, 39, 40, 41, 42 The source of controls was population-based in 11 studies,12, 14, 32, 33, 34, 35, 36, 37, 40, 42, 43 hospital-based in 4 studies13, 31, 38, 39 and not stated in 1 study.41

Table I. Studies Reporting Odds Ratios (OR) and 95% Confidence Limits (95% CL) for the Association between Adult Consumption of Preserved and Nonpreserved Vegetables and NPC
ReferencesStudy countryNo of cases/controlsHistological confirmation of casesSource of controlsAge range (years)Type of dietary questionnaireType of preserved vegetables examined with OR and 95% CL reported (WCRF classification)1Type of nonpreserved vegetables examined with OR and 95% CL reported (WCRF classification)1
  • 1

    WCRF classification of vegetables: NS, nonstarchy vegetables; GL, green leafy vegetables; ONS, other nonstarchy vegetables; CR, cruciferous vegetables; R + T, root and tubers.

  • 2

    FFQ, food frequency questionnaire.

Huang et al.,31 2002China175/350YesHospital12–77FFQ2 Tomatoes (ONS)
Laouamri et al.,32 2001Algeria72/72YesPopulation9–70UnknownVegetables in brine as seasoning (ONS), salted dried tomatoes as seasoning (ONS) 
Ward et al.,33 2000Taiwan371/321YesPopulation<75FFQ All vegetables (NS)
Yuan et al.,34 2000China935/1032YesPopulation15–74FFQPreserved leafy vegetables (ONS), preserved stem vegetables (ONS), preserved root vegetables (ONS), all preserved vegetables (NS)Bok choi (Chinese white cabbage, CR), other dark green leafy vegetables (GL), light green/orange leafy vegetables (ONS), gourds (ONS), green/red peppers (ONS), tomatoes (ONS), all fresh vegetables (NS)
Zou et al.,35 1999China102/202NoPopulation14–82FFQSalted other nonstarchy vegetables (ONS), salted vegetables (ONS)Tomatoes (ONS)
Armstrong et al.,14 1998Malaysia282/282YesPopulationMean, 45FFQSalted leafy vegetables (GL), salted root (ONS)Chinese flowering cabbage (CR), other green vegetables (NS), carrots, sweet potatoes (NS, R + T), tomatoes (ONS), water spinach (GL), Chinese kale (CR)
Farrow et al.,36 1998United States133/212YesPopulation18–74FFQ Dark green vegetables (NS), yellow vegetables (ONS)
Huang et al.,37 1997China104/104YesPopulationNAFFQPreserved green leafy vegetables (GL)Green pepper and celery (NS), cabbage (CR), cauliflower (CR)
Cai et al.,38 1996China115/115YesHospital16–68Dietary historyPickled leaf mustard (CR)Green leafy vegetables (GL), yellow color vegetables (ONS)
Ye et al.,39 1995China135/135YesHospital21–75Dietary historyPickled leaf mustard (CR), pickled radish (CR)All vegetables (NS), white color vegetables (ONS), green leafy vegetables (GL)
Lee et al.,13 1994Singapore200/406NoHospital<45FFQMei chye (preserved green leafy vegetable; GL, ONS), tang chye (preserved cabbage; CR), chye poh (preserved Chinese radish; CR), canned, salted, or pickled vegetables (NS), sze chuan chye (salted Chinese tuber; ONS), kiam chye (salted mustard greens; CR), preserved turnip root 
Zheng et al.,40 1994China88/176YesPopulation<30 to 50+FFQ Leafy vegetables (GL)
Wang et al.,41 1993  China122/122YesUnknown13–70FFQ All vegetables (NS), colored vegetables (ONS)
Ning et al.,12 1990  China100/292No  Population<64FFQ Carrots (NS, RT)
Jeannel et al.,42 1990  Tunisia80/160Yes  PopulationMean (cases only), 41FFQFungus on pickles (ONS) 
Armstrong et al.,43 1978Malaysia60/150No  Population25–81Dietary recallPickled chilli (ONS)Vegetables (NS), sliced raw chilli (ONS)

Thirteen studies were conducted among Chinese residents of Asian countries; of these, 9 were conducted in China12, 31, 34, 35, 37, 38, 39, 40, 41 and the others were carried out in Taiwan,33 Malaysia,14, 43 and Singapore.13 The remaining 3 studies were conducted in the United States,36 Algeria,32 and Tunisia.42

Dietary intake information was collected using a food-frequency questionnaire in 12 studies,12, 13, 14, 31, 33, 34, 35, 36, 37, 40, 41, 42 a dietary history questionnaire in 2 studies,38, 39 and dietary recall in 1 study.43 In 1 study, the type of instrument used to assess food intake was not specified.32 Food frequency data were reported in all studies; none of the studies detailed the amount of food intake.

All of the studies matched or adjusted for age and gender, 10 matched or adjusted for area of residence,12, 14, 31, 32, 33, 37, 38, 39, 40, 42 and 5 adjusted for intake of foods other than salted fish.33, 36, 37, 38, 41 Only 2 of the studies adjusted for salted fish intake38, 41; none adjusted for EBV. Other adjustment factors included education, cigarette smoking status, occupational and environmental exposures, history of chronic ear and nose conditions, ethnicity and socioeconomic status.

Quality assessment

Most of the 16 studies failed to fulfill minimal quality criteria. Many of the studies lacked information on the specifics of data collection for cases and controls, including whether data were collected in a similar manner for all study participants and whether interviewers were blinded to the case status or exposure status of the participants. Although all but one32 of the studies analyzed included incident cases of NPC, only 2 explicitly stated that cases were interviewed within 6 months of diagnosis.13, 33 Further, only 8 studies stated that the response rate among the controls was at least 70%.12, 13, 14, 33, 34, 35, 36, 43

Only 1 study explicitly stated that the instrument used to assess dietary intake was validated.36 This study was conducted in the United States and did not contribute to the preserved vegetable intake analyses. In addition, only 2 studies took into account salted fish intake, a known risk factor for NPC.38, 41 Both of these studies showed no change in the significance of the ORs after adjustment for salted fish. No study considered EBV infection.

Preserved vegetables

Ten studies reported results for at least 1 preserved vegetable (Table I).13, 14, 32, 34, 35, 37, 38, 39, 42 Within these 10 studies, the association between the intakes of 22 preserved vegetables and NPC were reported. In all but 1 study, the OR for the highest category of preserved vegetable intake was associated with an increased risk of NPC compared to the lowest category of intake, with the ORs ranging from 0.3 to 6.1 (unweighted median, 1.8). The pooled ORs were statistically significant for total preserved vegetable intake (Table II; Fig. 2; RE OR 2.04; 95% CL: 1.43, 2.92), green leafy vegetable intake (RE OR 2.50; 95% CL: 1.67, 3.74), and all nonstarchy vegetable intake (RE OR 1.92; 95% CL 1.10, 3.35), and was of borderline statistical significance for cruciferous vegetable intake (RE OR 1.81; 95% CL: 0.98, 3.65).

Figure 2.

Forest plot of OR and 95% confidence intervals for highest-versus-lowest category of preserved vegetable intake and NPC. Results are from a random-effects model. Box size is proportional to weight of study. Matching/adjustment variable abbreviations are as follows: G, gender; A, participant age; R, area of residence; S, smoking.

Table II. Meta-Analysis of NPC Odds Ratio (OR) Estimates (95% CL) by Vegetable Exposure
Vegetable exposureNo. of studiesReferencesRandom-effects ORI2 statistic %Q test (p-value)
  • 1

    Values in parentheses indicate 95% confidence limits.

All preserved vegetables613, 14, 32, 34, 35, 372.04 (1.43, 2.92)163 (9, 85)10.02
 Cruciferous313, 38, 391.81 (0.98, 3.65)52 (0, 86)0.1
 Green leafy313, 14, 372.50 (1.67, 3.74)5 (0, 90)0.3
 All nonstarchy613, 14, 34, 35, 431.92 (1.10, 3.35)78 (52, 90)0.0001
All nonpreserved vegetables533, 34, 39, 41, 430.64 (0.48, 0.85)50 (0, 82)0.09
 Cruciferous314, 34, 370.88 (0.67, 1.15)26 (0, 92)0.3
 Green leafy514, 34, 38, 39, 400.55 (0.32, 0.96)73 (33, 89)0.005
 Nonstarchy roots and tubers212, 140.55 (0.34, 0.88)0.4
 All nonstarchy1012, 14, 34, 35, 36, 37, 38, 39, 41, 430.62 (0.47, 0.81)56 (11, 78)0.02

There was evidence of statistical heterogeneity for the total preserved vegetables analysis [Q(5) = 13.4, p = 0.02]. Meta-regression results showed that none of the variables examined significantly explained the observed heterogeneity.

Four studies reported on 3 or more categories of preserved vegetable and NPC to assess the presence of a dose-response trend between vegetable intake and NPC (Table III).13, 14, 34, 39 All 4 studies showed that the risk of NPC increased with greater intake of preserved vegetables.

Table III. Results of Studies Examining the Association between Nonpreserved and Preserved Vegetable Intake and NPC, using more than 2 Categories of Vegetable Intake (Dose–Response Analyses)
ReferencesVegetable1Intake frequencyORp-for-trendAdjustment/matching variables
  • 1

    One vegetable analysis displayed per study; vegetable chosen represents the broadest preserved and nonpreserved vegetable category reported in the study.

  • 2

    Values in parentheses indicate 95% CI or p-value.

  • 3

    NR, Not reported.

Preserved vegetable intake
 Yuan et al.,34 2000All preserved vegetables<41 times/year1.0 (reference)0.001Age, gender, level of education, cigarette smoking, cooking exposures, occupational exposures, history of chronic ear and nose condition
41–102 times/year0.91 (0.69, 1.19)2
103–200 times/year1.18 (0.91, 1.54)
≥335 times/year1.43 (1.11, 1.81)
 Armstrong et al.,14 1998Salted leafy vegetablesMonthly1.0 (reference)NR3Age, gender, residence history
Weekly1.72 (1.03, 2.87)
≥Weekly3.33 (1.84, 6.01)
 Ye et al.,39 1995Pickled leaf mustard<50 times/year1.0 (reference)0.01 < p < 0.025Age, gender, area of residence
50–900 times/year1.52 (NR)
≥900 times/year6.11 (NR)
 Lee et al.,13 1994Canned, salted, or pickled vegetablesNone1.0 (reference)0.004Age, gender, education, dialect group
<1/month1.1 (0.7, 1.6)
1–3/month1.5 (0.8, 2.5)
1–3/week4.9 (1.8, 12.9)
Nonpreserved vegetable intake
 Yuan et al.,34 2000All fresh vegetables<539 times/year1.0 (reference)0.13Age, gender, level of education, cigarette smoking, cooking exposures, occupational exposures, history of chronic ear and nose condition
539–694 times/year1.12 (0.87, 1.44)
695–865 times/year0.95 (0.74, 1.24)
≥866 times/year0.85 (0.65, 1.10)
 Zou et al.,35 1999TomatoesNone1.0 (reference)NRAge, gender, date of death
Low intake1.13 (0.67, 1.88)
High Intake2.59 (0.69, 9.72)
 Farrow et al.,36 1998Dark green vegetablesFirst quartile1.0 (reference)0.60Age, alcohol consumption, cigarette smoking, total caloric intake
Second quartile0.47 (0.22, 1.00)
Third quartile1.14 (0.58, 2.22)
Fourth quartile0.99 (0.51, 1.94)
 Armstrong et al.,14 1998Other green vegetables (green onions, broccoli, chives, watercress, Chinese lettuce)Monthly1.0 (reference)NRAge, gender, residence history
Weekly0.70 (0.32, 1.53)
≥weekly0.50 (0.23, 1.07)
 Ye et al.,39 1995Green leafy vegetables<150 times/year1.0 (reference)0.01 < p< 0.025Age, gender, area of residence
150–910 times/year0.51 (NR)
≥910 times/year0.29 (NR)
 Zheng et al.,40 1994Leafy vegetablesMonthly1.0 (reference)NRAge, gender, area of residence
Weekly0.2 (0.014)
Daily0.1 (0.008)
 Ning et al.,12 1990CarrotsNone(reference)NRAge, gender, area of residence, race
Yearly0.5 (0.3, 1.0)
Monthly0.5 (0.3, 1.0)
Weekly0.8 (0.3, 1.9)

Nonpreserved vegetables

Thirteen studies reported results for at least 1 nonpreserved vegetable.12, 14, 31, 33, 34, 35, 36, 37, 38, 39, 40, 41, 43 Among the 13 studies, the association between the intakes of 32 nonpreserved vegetables and NPC were reported (Table I). The highest category of nonpreserved vegetable intake was almost always associated with a lower risk of NPC compared with the lowest category of intake. For these 32 unique analyses, ORs ranged from 0.1 to 2.59 (unweighted median, 0.63), with all but 2 ORs under 1.0. The pooled ORs were statistically significant for total nonpreserved vegetable intake (Table II; Fig. 3; RE OR 0.64; 95% CL: 0.48, 0.85) green leafy vegetable intake (RE OR 0.55; 95% CL: 0.32, 0.96), nonstarchy roots and tubers (RE OR 0.55; 95% CL 0.34, 0.88) and all nonstarchy vegetables (RE OR 0.62; 95% CL 0.47, 0.81).

Figure 3.

Forest plot of OR and 95% confidence intervals for highest versus lowest category of nonpreserved vegetable intake and NPC. Results are from a random-effects model. Box size is proportional to weight of study. Matching/adjustment variable abbreviations are as follows: G, gender; A, participant age; S, smoking; SES, socioeconomic status.

There was evidence of statistical heterogeneity for the green-leafy and all nonstarchy vegetable analyses [green-leafy, Q(4) = 14.8, p = 0.01; all nonstarchy, Q(9) = 20.6, p = 0.02]. Meta-regression results showed that none of the variables examined significantly explained the observed heterogeneity.

Seven studies reported on 3 or more categories of nonpreserved vegetable and NPC to assess for the presence of a dose-response trend between vegetable intake and NPC (Table III). Three of these studies showed evidence of a decrease in the risk of NPC with greater intake of nonpreserved vegetables.14, 39, 40 In the other 4 studies, 3 showed no evidence of a dose-response relationship between the amount of nonpreserved vegetable intake and the risk of NPC12, 34, 36 and 1 reported that greater intake of nonpreserved vegetables was associated with increased NPC risk.35

Publication bias

We judged the funnel plots for all preserved vegetable intake estimates and all nonpreserved vegetable intake estimates to be symmetrical, although the small number of studies limited our ability to identify systematic patterns indicative of publication bias.

Discussion

Summary of findings

This systematic review was carried out to assess the associations between adult intake of preserved and nonpreserved vegetables and NPC risk. The epidemiologic evidence identified was limited to case-control studies. Most of the studies were small and did not meet minimal quality criteria. Thus, the nature of the evidence does not justify drawing strong inferences. Nevertheless, the results, based on 16 case-control studies, were consistent in showing that the risk of NPC increased with consumption of preserved vegetables, but decreased with consumption of nonpreserved vegetables.

Preserved vegetable intake and NPC risk

For preserved vegetables, highest-versus-lowest category of consumption in adulthood was associated with a 2-fold increase in the risk of NPC. Dose-response trends were consistently observed in the few studies that reported these findings. The association between preserved vegetable intake and NPC was consistent across preserved vegetable subgroups and did not differ when considering the country in which the study was conducted.

For centuries, the traditional diet of the Chinese and those of other Asian peoples has included the consumption of preserved foods, including preserved vegetables. In these countries, the reasons that foods are preserved include to extend the shelf-life and to improve the nutritional quality and taste.44 For example, fermentation of vegetables using bacteria or molds can result in the removal of toxins and anti-nutritional compounds.45 Further, fermentation followed by drying can improve shelf-life by removing the biological components in foods responsible for deterioration.45 Preserved foods are eaten worldwide, but the highest levels of intake occur in Asia, and, specifically, in China.19 From a public health perspective, Asia is the region of greatest concern with respect to any adverse health consequences that may be linked to preservation methods.

The increase in NPC risk with the intake of preserved vegetables is biologically plausible based on the higher concentrations of nitrates and nitrosamines associated with the preservation processes. Nitrosamines are known mutagens and animal carcinogens that induce the formation of DNA adducts, which, if not repaired, may lead to point mutations and a greater likelihood of cancer.17 The N-nitrosamines are a large group of nitrosamine compounds with a common carcinogenic mechanism.46 Preserved vegetables contain N-nitrosodimethylamine and other volatile N-nitroso compounds18 having both mutagenic and carcinogenic effects in laboratory animals.47, 48, 49 If nitrosamines do contribute to the etiology of NPC, the variation in nitrosamine content in preserved foods due to regional differences in preservation methods may contribute to the heterogeneity in assigning risk to preserved vegetable consumption in different geographical locations.

The link between preserved vegetable intake and the risk of malignancy may not be limited to the nasopharynx, as preserved vegetable intake has also been observed to be associated with cancers of the colon,50 stomach,51 and esophagus.52, 53 Similar to results of the present systematic review, the results of these studies suggest that preserved vegetable intake may be associated with increased risk of other malignancies, with evidence of a dose-response relationship.50, 51

Considered in combination with the epidemiologic evidence linking preserved vegetable consumption to increased NPC risk, the biological plausibility of the association between preserved vegetable intake and NPC supports the possibility that the relationship between preserved vegetable intake and NPC could be real. Further research is warranted to determine if this is the case. If so, this finding is most relevant to Asian countries, where preserved vegetables are a major dietary constituent.

Non-preserved vegetable intake and NPC risk

For nonpreserved vegetables, highest-versus-lowest categories of consumption in adulthood were associated with an ˜40% decrease in the risk of NPC. Dose-response trends were not consistently observed, but were present in the majority of studies that reported these findings. The inverse association was consistent across nonpreserved vegetable subgroups and did not differ when considering the country in which the study was conducted.

The inverse association between nonpreserved vegetable intake and NPC risk is consistent with what is known about the protective effect of vegetable intake on the development of cancer of other types. In 1997, WCRF rated the evidence as convincing or probable that vegetable intake was associated with a decreased risk of several malignancies, including cancer of the mouth and pharynx, esophagus, lung, stomach, colon and rectum.19 Further, several scientific reviews have concluded that the consumption of higher levels of vegetables is associated with a reduced risk of cancer at most sites.54, 55, 56, 57, 58, 59, 60 In fact, it was reported in 1997 that, when studies of all cancer sites were taken together, 78% showed a significant decrease for higher intake of at least 1 vegetable and/or fruit category examined.19

The protective effects of vegetables are thought to be mediated by multiple components, including beta-carotene, alpha-tocopherol, retinoids, phytoestrogens and folate. These components are involved in numerous biological processes that may alter cancer risk, including the inhibition of cell growth, the normal synthesis and methylation of DNA, and protection against oxidative stress and DNA damage. To date, no study has examined the association between these specific components of vegetables and NPC risk. If nonpreserved vegetable intake is truly associated with decreased NPC risk, the mechanisms are likely the same as those by which vegetables are proposed to decrease the risk of other cancer types. It could be that consumption of nonpreserved vegetables is not a true protective factor in the development of NPC, but merely serves as a marker of lower consumption of preserved vegetables or, perhaps, an otherwise healthy lifestyle.

Quality of evidence and limitations

The methodological and analytical limitations of the studies in this systematic review make it impossible to establish firm conclusions regarding the associations between preserved and nonpreserved vegetable intake and the risk of NPC. Most importantly, the case-control design of each of the 16 studies is associated with a number of potential biases,61 including selection and recall bias, which should be considered in interpreting the results of each study individually and the overall pooled estimates reported in this systematic review. Further, only 1 study included in this review explicitly stated that the instrument used to assess dietary intake was validated36; hence, it is unknown how accurate the questionnaires were in assessing food intake, especially in a case-control design, where recall of intake over many years is important. Importantly, only 2 studies considered salted fish intake and no study examined EBV infection, both potentially important confounders or effect modifiers.38, 41 Significant associations observed may be due to these other factors and, therefore, these factors should be taken into account when assessing the association between vegetable intake and NPC. In addition, this review included a number of different types of preserved vegetables and, therefore, it cannot be stated which specific types of preservation methods may underlie the observed association between preserved vegetable intake and NPC. In fact, the significant heterogeneity we observed in the meta-analyses could have been due to the inherent variation in how the intake of “preserved vegetables” and “nonpreserved vegetables” were measured and classified and the actual variation in preservation techniques among study populations. However, given the consistency of the associations within the sub-categories under preserved and nonpreserved vegetables, it seems unlikely that this variability led to spurious conclusions.

It should be noted that, while we found no evidence of publication bias, selective reporting of results among the studies included in this review, especially those results pertaining to the association between preserved vegetable consumption and NPC, must be considered when interpreting our findings. Among the 16 studies included in the review, only 6 were included in the meta-analysis for the broad category of “preserved vegetable intake.” Of the 10 studies not included, 3 conducted analyses examining the association between broad categories of preserved vegetable intake and NPC and reported that the association was not statistically significant without reporting the point estimate and a measure of variability. Inclusion of these nonsignificant results in the meta-analysis for the broad category of preserved vegetable intake would likely attenuate the association we observed between preserved vegetable consumption and NPC. This appears to have been less problematic with respect to nonpreserved vegetable intake. Of the eleven studies that were not included in the meta-analysis of the broad category of “nonpreserved vegetable intake,” none reported an analysis of an association between a broad category of nonpreserved vegetable intake and NPC. Therefore, the association we observed between nonpreserved vegetable intake and NPC was less likely to be inflated by selective reporting. In general, however, without reviewing the original protocols for the studies, it is impossible to know for certain whether the investigators examined other exposures and only reported the ones for which significant associations were found. Hence, the extent to which selective reporting occurred and how it affected the estimates remains uncertain.

Conclusions

The evidence included in this review was relatively uniform in pointing toward higher risk of NPC associated with preserved vegetable intake and lower risk of NPC associated with consumption of nonpreserved vegetables. The evidence supporting these associations was robust, based on the strength of the associations, including evidence of dose-response trends and the consistency of the findings across studies, categories of vegetable type and geographic region.

However, inferences need to be drawn cautiously because the body of evidence on this topic has limitations. Even if these associations have been accurately measured, the potential role of confounding factors remains to be thoroughly addressed. For example, it is unclear whether or not the link between preserved vegetable intake and increased NPC risk is direct, or reflects a correlation between consumption of preserved vegetables and other preserved foods, such as salted fish, that may be harmful. This is important because, for example, it could be that nonpreserved vegetables are not actually protective but merely serving as a marker of lower consumption of preserved foods. Alternatively, consumption of preserved vegetables may not be a true risk factor, but a measure of lower consumption of nonpreserved vegetables. Furthermore, the interrelationship of preserved and nonpreserved vegetable intake has not been addressed. Many other explanations are possible, and the findings to date indicate the conducting of well-designed studies in high-risk areas is warranted for the insights they may provide to the etiology of NPC, with implications for developing prevention strategies. Since data to address this topic are unlikely to be generated from cohort studies due to practical constraints, future case-control studies that assess the inter-relationship of preserved and nonpreserved vegetable intake while taking more complete account of the range of potential confounders will be particularly valuable.

Acknowledgements

Dr. Gallicchio is supported by an Institutional National Research Service Award from the National Cancer Institute (T32 CA09314).

Appendix–WCRF PubMed search strategy

A–Outcome

Nasopharyngeal[tiab] OR nasal[tiab] OR nasal sinus[tiab] OR pharynx[tiab] OR head and neck[tiab] OR aerodigestive[tiab]

AND Cancer[tiab] OR carcinoma[tiab] OR neoplasms[tiab] OR tumor[tiab]

B–Exposure

  • 1diet therapy[MeSH Terms] OR nutrition[MeSH Terms]
  • 2diet[tiab] OR diets[tiab] OR dietetic[tiab] OR dietary[tiab] OR eating[tiab] OR intake[tiab] OR nutrient*[tiab] OR nutrition[tiab] OR vegetarian*[tiab] OR vegan*[tiab] OR “seventh day adventist”[tiab] OR macrobiotic[tiab] OR breastfeed*[tiab] OR breast feed*[tiab] OR breastfed[tiab] OR breast fed[tiab] OR breastmilk[tiab] OR breast milk[tiab]
  • 3food and beverages[MeSH Terms]
  • 4food*[tiab] OR cereal*[tiab] OR grain*[tiab] OR granary[tiab] OR wholegrain[tiab] OR wholewheat[tiab] OR roots[tiab] OR plantain*[tiab] OR tuber[tiab] OR tubers[tiab] OR vegetable*[tiab] OR fruit*[tiab] OR pulses[tiab] OR beans[tiab] OR lentils[tiab] OR chickpeas[tiab] OR legume*[tiab] OR soy[tiab] OR soya[tiab] OR nut[tiab] OR nuts[tiab] OR peanut*[tiab] OR groundnut*[tiab] OR seeds[tiab] OR meat[tiab] OR beef[tiab] OR pork[tiab] OR lamb[tiab] OR poultry[tiab] OR chicken[tiab] OR turkey[tiab] OR duck[tiab] OR fish[tiab] OR fat[tiab] OR fats[tiab] OR fatty[tiab] OR egg[tiab] OR eggs[tiab] OR bread[tiab] OR oils[tiab] OR shellfish[tiab] OR seafood[tiab] OR sugar[tiab] OR syrup[tiab] OR dairy[tiab] OR milk[tiab] OR herbs[tiab] OR spices[tiab] OR chilli[tiab] OR chillis[tiab] OR pepper*[tiab] OR condiments[tiab]
  • 5fluid intake[tiab] OR water[tiab] OR drinks[tiab] OR drinking[tiab] OR tea[tiab] OR coffee[tiab] OR caffeine[tiab] OR juice[tiab] OR beer[tiab] OR spirits[tiab] OR liquor[tiab] OR wine[tiab] OR alcohol[tiab] OR alcoholic[tiab] OR beverage*[tiab] OR ethanol[tiab] OR yerba mate[tiab] OR ilex paraguariensis[tiab]
  • 6pesticides[MeSH Terms] OR fertilizers[MeSH Terms] OR “veterinary drugs”[MeSH Terms]
  • 7pesticide*[tiab] OR herbicide*[tiab] OR DDT[tiab] OR fertiliser*[tiab] OR fertilizer*[tiab] OR organic[tiab] OR contaminants[tiab] OR contaminate*[tiab] OR veterinary drug*[tiab] OR polychlorinated dibenzofuran*[tiab] OR PCDF*[tiab] OR polychlorinated dibenzodioxin*[tiab] OR PCDD*[tiab] OR polychlorinated biphenyl*[tiab] OR PCB*[tiab] OR cadmium[tiab] OR arsenic[tiab] OR chlorinated hydrocarbon*[tiab] OR microbial contamination*[tiab]
  • 8food preservation[MeSH Terms]
  • 9mycotoxin*[tiab] OR aflatoxin*[tiab] OR pickled[tiab] OR bottled[tiab] OR bottling[tiab] OR canned[tiab] OR canning[tiab] OR vacuum pack*[tiab] OR refrigerate*[tiab] OR refrigeration[tiab] OR cured[tiab] OR smoked[tiab] OR preserved[tiab] OR preservatives[tiab] OR nitrosamine[tiab] OR hydrogenation[tiab] OR fortified[tiab] OR additive*[tiab] OR coloring*[tiab] OR coloring*[tiab] OR flavoring*[tiab] OR flavoring*[tiab] OR nitrates[tiab] OR nitrites[tiab] OR solvent[tiab] OR solvents[tiab] OR ferment*[tiab] OR processed[tiab] OR antioxidant*[tiab] OR genetic modif*[tiab] OR genetically modif*[tiab] OR vinyl chloride[tiab] OR packaging[tiab] OR labeling[tiab] OR phthalates[tiab]
  • 10cookery[MeSH Terms]
  • 11cooking[tiab] OR cooked[tiab] OR grill[tiab] OR grilled[tiab] OR fried[tiab] OR fry[tiab] OR roast[tiab] OR bake[tiab] OR baked[tiab] OR stewing[tiab] OR stewed[tiab] OR casserol*[tiab] OR broil[tiab] OR broiled[tiab] OR boiled[tiab] OR microwave[tiab] OR microwaved[tiab] OR re-heating[tiab] OR reheating[tiab] OR heating[tiab] OR re-heated[tiab] OR heated[tiab] OR poach[tiab] OR poached[tiab] OR steamed[tiab] OR barbecue*[tiab] OR chargrill*[tiab] OR heterocyclic amines[tiab] OR polycyclic aromatic hydrocarbons[tiab]
  • 12dietary carbohydrates[MeSH Terms] OR dietary proteins[MeSH Terms] OR sweetening agents[MeSH Terms]
  • 13salt[tiab] OR salting[tiab] OR salted[tiab] OR fiber[tiab] OR fiber[tiab] OR polysaccharide*[tiab] OR starch[tiab] OR starchy[tiab] OR carbohydrate*[tiab] OR lipid*[tiab] OR linoleic acid*[tiab] OR sterols[tiab] OR stanols[tiab] OR sugar*[tiab] OR sweetener*[tiab] OR saccharin*[tiab] OR aspartame[tiab] OR acesulfame[tiab] OR cyclamates[tiab] OR maltose[tiab] OR mannitol[tiab] OR sorbitol[tiab] OR sucrose[tiab] OR xylitol[tiab] OR cholesterol[tiab] OR protein[tiab] OR proteins[tiab] OR hydrogenated dietary oils[tiab] OR hydrogenated lard[tiab] OR hydrogenated oils[tiab]
  • 14vitamins[MeSH Terms]
  • 15supplements[tiab] OR supplement[tiab] OR vitamin*[tiab] OR retinol[tiab] OR carotenoid*[tiab] OR tocopherol[tiab] OR folate*[tiab] OR folic acid[tiab] OR methionine[tiab] OR riboflavin[tiab] OR thiamine[tiab] OR niacin[tiab] OR pyridoxine[tiab] OR cobalamin[tiab] OR mineral*[tiab] OR sodium[tiab] OR iron[tiab] OR calcium[tiab] OR selenium[tiab] OR iodine[tiab] OR magnesium[tiab] OR potassium[tiab] OR zinc[tiab] OR copper[tiab] OR phosphorus[tiab] OR manganese[tiab] OR chromium[tiab] OR phytochemical[tiab] OR allium[tiab] OR isothiocyanate*[tiab] OR glucosinolate*[tiab] OR indoles[tiab] OR polyphenol*[tiab] OR phytoestrogen*[tiab] OR genistein[tiab] OR saponin*[tiab] OR coumarin*[tiab]
  • 16physical fitness[MeSH Terms] OR exertion[MeSH Terms] OR physical endurance[MeSH Terms] or walking[MeSH Terms]
  • 17recreational activit*[tiab] OR household activit*[tiab] OR occupational activit*[tiab] OR physical activit*[tiab] OR physical inactivit*[tiab] OR exercise[tiab] OR exercising[tiab] OR energy intake[tiab] OR energy expenditure[tiab] OR energy balance[tiab] OR energy density[tiab]
  • 18growth[MeSH Terms] OR anthropometry[MeSH Terms]
  • 19weight loss[tiab] or weight gain[tiab] OR anthropometry[tiab] OR birth weight[tiab] OR birthweight[tiab] OR birth-weight[tiab] OR child development[tiab] OR height[tiab] OR body composition[tiab] OR body mass[tiab] OR BMI[tiab] OR obesity[tiab] OR obese[tiab] OR overweight[tiab] OR over-weight[tiab] OR over weight[tiab] OR skinfold measurement*[tiab] OR skinfold thickness[tiab] OR DEXA[tiab] OR bio-impedence[tiab] OR waist circumference[tiab] OR hip circumference[tiab] OR waist hip ratio *[tiab]
  • 201 OR 2 OR 3 OR 4 OR 5 OR 6 OR 7 OR 8 OR 9 OR 10 OR 11 OR 12 OR 13 OR 14 OR 15 OR 16 OR 17 OR 18 OR 19

Ancillary