Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica

Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica Downloaded from www.sjweh.fi on July 08, 2021 Original article Scand J Work Environ Health 2007;33(4):293-303 doi:10.5271/sjweh.1146 Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica by Monge P, Wesseling C, Guardado J, Lundberg I, Ahlbom A, Cantor KP, Weiderpass E, Partanen T Affiliation: Central American Institute for Studies on Toxic Substances (IRET), Universidad Nacional, PO Box 863000, Heredia, Costa Rica. pmonge@una.ac.cr Key terms: cancer epidemiology; casecontrol study; child; childhood cancer; childhood leukemia; Costa Rica; developing country; fetal exposure; parental occupational exposure; pesticide; pregnancy; reproductive effect; risk; tropics This article in PubMed: www.ncbi.nlm.nih.gov/pubmed/17717622 This work is licensed under a Creative Commons Attribution 4.0 International License. Print ISSN: 0355-3140 Electronic ISSN: 1795-990X Original article Scand J Work Environ Health 2007;33(4):293–303 Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica 1, 2 1, 2 1 by Patricia Monge, PhD, Catharina Wesseling, PhD, Jorge Guardado, LicComp, Ingvar Lundberg, 2, 3 4 5 6, 7 1 PhD, Anders Ahlbom, PhD, Kenneth P Cantor, PhD, Elisabete Weiderpass, PhD, Timo Partanen, PhD Monge P, Wesseling C, Guardado J, Lundberg I, Ahlbom A, Cantor KP, Weiderpass E, Partanen T. Parental oc- cupational exposure to pesticides and risk of childhood leukemia in Costa Rica. Scand J Work Environ Health 2007;33(4):293–303. Objectives Parental exposure to pesticides and the risk of leukemia in offspring were examined in a popula- tion-based case–control study in Costa Rica. Methods All cases of childhood leukemia (N=334), in 1995–2000, were identified at the Cancer Registry and the Children’s Hospital. Population controls (N=579) were drawn from the National Birth Registry. Interviews of parents were conducted using conventional and icon-based calendar forms. An exposure model was constructed for 25 pesticides in five time periods. Results Mothers’ exposures to any pesticides during the year before conception and during the r fi st and second trimesters were associated with the risk [odds ratio (OR) 2.4, 95% cond fi ence interval (95% CI) 1.0–5.9; OR 22, 95% CI 2.8–171.5; OR 4.5, 95% CI 1.4–14.7, respectively] and during anytime (OR 2.2, 95% CI 1.0–4.8). An as- sociation was found for fathers’ exposures to any pesticides during the second trimester (OR 1.5, 95% CI 1.0–2.3). An increased risk with respect to organophosphates was found for mothers during the r fi st trimester (OR 3.5, 95% CI 1.0–12.2) and for fathers during the year before conception and the r fi st trimester (OR 1.5, 95% CI 1.0–2.2 and OR 1.6, 95% CI 1.0–2.6, respectively), and benzimidazoles during the r fi st, second, and third trimesters of pregnancy (OR 2.2, 95% CI 1.0–4.4; OR 2.2, 95% CI 1.0–5.0; OR 2.2, 95% CI 1.0–5.2, respectively). There was a suggestion of an exposure–response gradient for fathers as regards picloram, benomyl, and paraquat. Age at diagnosis was positively associated with fathers’ exposures and inversely associated with mothers’ exposures. Conclusions The results suggest that parental exposure to certain pesticides may increase the risk of leukemia in offspring. Key terms cancer epidemiology; case–control study; childhood cancer; children; developing country; fetal exposure; pregnancy; reproductive effect; tropics. Leukemias are the most common childhood cancers (1, than adults who receive such exposures later in life, due 2), and their etiology is not well understood. Known or to different physical dimensions, less mature immune suspected general risk factors include male gender, age systems, unique diets, and metabolic characteristics 1–4 years, particular genetic polymorphisms, ethnicity (3, 4). Parental exposures to occupational hazards may (Caucasian), high socioeconomic status, small families, contribute to the risk of leukemia or other cancers in low birthweight, and chemical and physical risk factors offspring through damage to germ cells of either parent such as ionizing radiation, electromagnetic e fi lds, chemical prior to pregnancy; intrauterine and early extrauterine agents, dusts, fumes, spores, drugs and infections (1, 2). exposure via maternal or paternal exposure to toxic Environmental exposures during development and compounds and their metabolites through transplacental childhood may place children at a higher leukemia risk transmission during gestation; or directly through breast Central American Institute for Studies on Toxic Substances, Universidad Nacional, Heredia, Costa Rica. Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden. National Institute for Working Life, Stockholm, Sweden. National Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden. Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD, USA. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. Cancer Registry of Norway, Oslo, Norway. Reprint requests to: Dr P Monge, Central American Institute for Studies on Toxic Substances (IRET), Universidad Nacional, PO Box 86–3000, Heredia, Costa Rica. [E-mail: pmonge@una.ac.cr] Scand J Work Environ Health 2007, vol 33, no 4 293 Pesticides exposure and childhood leukemia feeding, take-home exposure through contaminated Diseases (ICD) 0–1]. Population controls (N=579), person(s) or workclothes during the postnatal period frequency matched to the cases by birth year, were (5–11). Bone marrow originates from the mesoderm, and drawn from the National Birth Registry, with the use of its development occurs at the end of the first trimester of computerized random selection. The number of controls pregnancy. Exposure to teratogens during this trimester was chosen using cost-efficiency considerations (30). could sensitize embryos to problems in development, As Costa Rica uses landmarks to indicate addresses including childhood leukemia. Radiation during the first instead of street names and house numbers, we estab- trimester has been found to be associated with higher lished a procedure to increase the likelihood of locating risks of leukemia in children (12, 13). families while maintaining randomness and allowing for An increased risk of childhood leukemia has been neighborhood replacement of nontraceable controls. Ad- found to be associated with occupational paternal exposure dresses and sometimes telephone numbers of the cases to pesticides prior to and during pregnancy (8, 14–16) and were obtained from files of the National Children’s Hos - parents’ pesticide exposures at home or in gardens (8, pital. In most cases, this information was adequate for 17–19). Pesticides associated with childhood leukemia in- tracing the family. When information was inadequate, clude chlordane, dichlorvos, monomethyldithiocarbamate we used databases of the local social security clinics. (metam sodium), propoxur, and dicofol (20–23). For the controls, the Birth Registry provided an address The median incidence of childhood leukemias for of the mother at the time of birth, which was sometimes 58 populations with cancer registry-based data for the complete but often restricted to a neighborhood. Using 1990s was 45 (range 25–64) per million person-years. national electoral databases and local social security Costa Rica ranks third, with a rate of 63 per million, clinics, we were able to ascertain exact addresses for close to the 64 per million for Latin populations in Los 62% of the controls. If the family was not located but Angeles (24). To date, no etiologic studies have been information on a new address was available, we at- conducted in Costa Rica. tempted to locate them at the new address. Otherwise, Agriculture, an important economic activity in Costa the control was replaced with a child of the same age Rica, is associated with excessive and inappropriate use living in the same neighborhood. If the address was un- of pesticides (25). An annual average of 2.5 kg of active traceable, a control child of the same age was randomly pesticide ingredient per inhabitant was estimated for the selected from the archives of the health center in the year 1996, compared with 0.7 kg in the Netherlands, for neighborhood mentioned on the birth certificate. The example, which represents a high European rate (26). overall response rate was 90% for the cases (19 refused Several pesticides that have been linked to childhood to participate and 15 were not located) and 90.5% for the leukemia are in widespread use in Costa Rica. Propoxur controls (55 refused to participate). Refusals were also is used in commercial domestic pest control products, substituted with controls in the same neighborhood. The and metam sodium and dicofol in ornamental plants and effective number of cases was 300; that of controls was fruit growing, among other crops. Dichlorvos is used for 579 (figure 1). Case diagnoses were extracted directly crops, as well as for livestock. from Cancer Registry data and confirmed from files of Most agricultural workers experience complex ex- the Children’s Hospital. The study was approved by the posure patterns to pesticides over time, the result being Institutional Review Boards (IRB) of the Children’s serious methodological limitations in many studies Hospital, Ministry of Health of Costa Rica, and the addressing pesticide exposures and childhood cancer, Karolinska Institutet in Sweden. All of the participants calling for improved exposure assessment with regard to gave informed consent prior to the interviews. chemical type and timing (7, 23, 27–29). With this need We conducted face-to-face interviews with parents in mind, we conducted a population-based case–control in 2001–2003 using an interview form to collect demo- study to evaluate associations between parental occu- graphic data and data on known and suspected risk fac- pational exposure to pesticides and the risk of leukemia tors of childhood leukemia. Parents who were active in in offspring. agriculture or livestock production during the assessment period completed an additional interview that utilized an icon-calendar form (ICF) (31). The response rate of ICF completion, based on sufficient and reasonably good Study population and methods quality data, was 90%, with interviews of 83 cases and 139 control parents. After the data were cleaned and Ascertainment of cases and controls some parental questionnaires with incomplete data were excluded, data for 876 mothers and 762 fathers remained All cases of childhood leukemia (ages 0–14 years at di- for the analyses (figure 1). agnosis, N=334) diagnosed in Costa Rica in 1995–2000 Out of the parents of the cases, 16.9% were active in were identified at the Cancer Registry and the Children’s agriculture; of the parents of the controls, 15.6% were Hospital of Costa Rica [International Classification 294 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al Figure 1. active. With consideration for the small proportion of Cases Controls parents outside the active labor force, these figures are (334) (579) comparable with the 18.2% of the total labor force in Refusals Refusals agriculture in the national census of 2000. (19) (55) Not Not located located (215) Interview data on exposures and their determinants (15 *) A conventional interview format was offered to both par- ents. It had three forms, one each for the mother, father Neighborhood control and child (completed by either parent). The interview replacement included the location of the family residence; education (270) of the father and mother; smoking and substance abuse of the father and mother and diet of the mother; medical Cases Controls history of the mother, including vaccinations, the moth- Both parents (267) Both parents (492) Mother only (32) Mother only (85) er’s X-rays and medications; pregnancy history; birth Father only (1) Father only (2) parameters; occupational, environmental and home pes- ticide exposures of both parents; occupational histories of the parents; and exposures of the father and mother to In agriculture: nonpesticide toxic agents. For those in agriculture, data Quantitative and qualitative ICF on the parental use of pesticides, agricultural tasks, fre- (83) (139) Qualitative ICF only quency of exposure (number of applications per month (6) (22) and hours per day), major determinants of pesticide ex- Missing data posure (task technology, personal protective equipment, (5) (6) field reentry, storing of pesticides, personal hygiene) * For 8 cases, addresses were not located, 6 were living in other For 8 case countries and 1 entire family had died. s addresses were not located, 6 were living in other countries and 1 entire were extracted with the ICF for the etiologically relevant family had died. period on a month-to-month basis (32). The relevant Figure 1. Study structure. (ICF = icon-based calendar form) period was taken as the period from 12 months before conception until the diagnosis of cancer for the cases and oxamyl, quintozene, and aldrin (33). Cyproconazole, until either the interview date or the age of 15 years for foxim, and fenamiphos were also included because the controls, whichever occurred first. their use was cited frequently by the participants. We The tasks were classified according to their estimated constructed two exposure assessment models. A quan- hazard (0 = no exposure; 4 = very high exposure). titative model combined ICF data with external data on Tasks with no exposure (0) included coffee picking the 25 prioritized pesticides for 14 crops, 21 calendar and organic agriculture. Tasks with a hazard value of years, and 14 geographic regions. Since no measurement 1 included tasks such as the watering of plants, tractor data were available, we used the application rates of driving, and garden maintenance; those with a hazard pesticides from the external data as a basic component value of 2 included fertilizing and defoliating; those of the estimated exposure intensity. The application rates with a hazard value of 3 included backpack application were obtained from a database of the Central American and the mixing of pesticides; and those with a hazard Institute for Studies in Toxic Substances-Universidad value of 4 included the application of pesticides with Nacional (IRET-UNA) that draws from databases of the hands (32). ministries, specialized crop offices, and agricultural Checklists of pesticides enhanced the interview as bank loan records (34). Our experts, knowledgeable in prompt lists. The pesticides selected for the checklists pesticide use, checked and corrected these reports for were based on a toxicity-based prioritization system de- any obvious inconsistencies. We converted the calendar signed for this study (33). The interview procedure has time into individual time that had birth of the child as been described earlier in detail by Monge et al (31). the zero time reference (30). Exposures were quasiquan- titatively assessed for five time periods in the model Exposure assessment models that integrated personal data with the external data on pesticide, crop, and geographic region. The periods were The following 22 pesticides were identified as represent - the year before conception, the first trimester, the second ing high priority on the basis of historical data on use trimester, the third trimester, and the first year of life of and toxicity data on 2,4-dichlorophenoxyacetic acid the child (32). (2,4-D), picloram, glyphosate, benomyl, chlorothalonil, The qualitative model used binary (yes, no) indi- paraquat, carbofuran, mancozeb, terbufos, methamido- cators of occupational exposure in agricultural areas, phos, deltamethrin, methomyl, triadimefon, fluazifop, derived from the ICF interview for 14 selected chemical captafol, lead arsenate, malathion, dichlorvos, diuron, Scand J Work Environ Health 2007, vol 33, no 4 295 Pesticides exposure and childhood leukemia groups of pesticides for all time period, in which any their 95% confidence intervals (95% CI) as estimates exposure to one of the compounds of the chemical group of relative risk, separately for the five time periods. was sufficient to trigger “exposure”: phenoxyacetic For preterm birth children, adjustment was applied acids, organophosphates, organochlorines, carbamates, for the conception date. Exposures were expressed dithiocarbamates, pyrethroids, triazines, benzimidazoles, as qualitative (yes, no), semiquantitative (unexposed, chlorinated phthalides, conazoles, copper compounds, low exposure, high exposure) and quantitative metrics arsenic compounds, chlorinated urea derivatives, and for specific pesticides and groups of pesticides. Stick - “other pesticides”. The “other pesticides” group in- ing to the basic concept for confounding, we applied cluded pesticides not in any of the specific pesticide correlation-based selection of confounders instead of groups, the latter group consisting mainly of paraquat, more-refined methods, expecting little difference in the chlorothalonil, and glyphosate. results. Low correlations among the controls between All of the exposure assessment was done blinded as variables of pesticide exposure and maternal age at con- to the case–control status of the children. ception, infectious disease of the child during the first year, the mother’s and child’s exposure to X-rays during pregnancy and first year of life, respectively, mother’s Statistical analysis tobacco and alcohol consumption during pregnancy, Unconditional crude and adjusted logistic regression father’s smoking, and history of newborn jaundice and models were used to estimate the odds ratios (OR) and vaccination of the child resulted in the inclusion only Table 1. Odds ratios (OR) and 95% confidence intervals (95% CI) for parents’ exposure to prioritized pesticides (ie, those to which >3 cases had been exposed)—entries with a ≥1 excess with a lower confidence limit of <1. (N = number of exposed cases) Pesticide Exposed versus unexposed High versus low exposure Total leukemia Acute lymphocytic leukemia Total leukemia Acute lymphocytic leukemia a a N OR 95% CI N OR 95% CI Cases OR 95% CI Cases OR 95% CI Low High Low High Insecticides Foxim Fathers b b b b 1st year of life 5 5.1 1.0–26.5 5 6.1 1.2–31.6 1 4 ∞ 1 4 ∞ Herbicides Picloram Fathers Year before conception 11 1.3 0.6–2.9 11 1.6 0.7–3.4 3 8 6.3 1.0–38.6 3 8 6.3 1.0–38.6 1st trimester 8 2.8 0.9–8.1 8 3.3 1.1–9.8 3 5 3.1 0.3–29.4 3 5 3.1 0.3–29.4 c c c c 1st year of life 10 1.3 0.6–2.8 10 1.5 0.7–3.4 2 8 12.4 1.6–98.3 2 8 12.4 1.6–98.3 Paraquat Fathers Year before conception 39 1.1 0.7–1.7 30 1.0 0.6–1.5 14 25 2.3 1.1–5.2 12 18 1.67 0.7–4.1 1st year of life 45 1.5 1.0–2.3 36 1.4 0.9–2.2 18 27 1.9 0.9–4.1 15 21 1.7 0.7–4.0 Mothers d d d d Year before conception 7 3.4 1.0–11.8 6 3.5 1.0–12.7 2 5 7.5 0.5–122.7 2 4 6.0 0.4–101.6 e e b b b b 2nd trimester 4 7.8 0.9–70.6 4 9.5 1.1–85.5 1 3 ∞ 1 3 ∞ Fungicides Benomyl Fathers 1st year of life 14 1.8 0.9–3.8 12 1.8 0.9–4.0 4 10 5.5 1.1–26.4 3 9 6.6 1.2–35.4 Mancozeb Fathers 1st trimester 11 1.9 0.8–4.3 11 2.3 1.0–5.2 4 7 2.5 0.5–13.2 4 7 2.5 0.5–13.2 Adjusted for residence (urban or rural). No controls with high exposure. Crude P-value from a hypergeometric distribution of >0.10. The unadjusted figures for exposure to picloram during the first year of life and low versus high exposures are OR 8.8 (95% CI 1.34–57).4 for total leuke - mia and OR 7.65 (95% CI 1.35–57.4) for acute lymphocytic leukemia. Unadjusted figures. The unadjusted figures are OR 7.65 (95% CI 0.9–69.0) for total leukemias and OR 9.15 (95% CI 1.0–82.3) for acute lymphocytic leukemia. 296 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al of the urban or rural residence in all of the logistic Results models controlling for unmeasured urban or rural risk factors. The mother’s X-ray exposure during pregnancy The cases and controls were similar with respect to was included because of its moderate correlation with maternal age at conception, the proportion of parents phthalide exposure. The calculations were done with exposed to pesticides, maternal smoking and alcohol STATA Release 8.0 (Stata Corporation, College Station, consumption history during pregnancy, and maternal TX, USA). The unexposed formed the reference group education. The cases were more often boys, and their in all of the analyses, except for those between the high families represented a slightly lower socioeconomic and low exposure groups (table 1 on page 296). The status. Acute lymphocytic leukemia represented 83.7% intratime-period correlation coefficients between pesti - of the cases (table 2). cides were low (median absolute |r|=0.003). Altogether 13.9% of the parents of the cases (24.6% of the fathers and 4.3% of the mothers) was ever ex- posed to any pesticides in work during the exposure Table 2. Characteristics of the cases (total leukemia) and controls. (NOS = not otherwise specified) assessment, while the percentage of the control parents was 11.5% (22.2% for the fathers and 2.2% for the Characteristics Cases Controls mothers). (N=300) (N=579) (%) (%) Table 3 shows the odds ratios for the time windows for all of the pesticides and according to the following Age at diagnosis biocide category: insecticides, herbicides, and fungi- 0 years 4.7 . cides for the fathers and mothers. All of the odds ratios 1–4 years 43.3 . 5–9 years 34.0 . 10–14 years 18.0 . Table 3. Odds ratios (OR) and 95% confidence intervals (95% CI) Gender for parents’ exposure to pesticides to which >3 cases had been Female 44.3 50.8 exposed through biocide action—exposed versus unexposed. Male 55.7 49.2 (N = number of exposed cases) Agricultural exposure to pesticides Maternal 4.3 2.2 Pesticide group Fathers Mothers Paternal 24.6 22.2 a a Leukemia type N OR 95% CI N OR 95% CI Acute lymphocytic leukemia 84.0 . All pesticides Acute nonlymphocytic leukemia 13.7 . Other leukemias 2.3 . 1 year before conception 64 1.2 0.9–1.8 11 2.4 1.0–5.9 NOS 0.3 . b 1st trimester of pregnancy 45 1.3 0.9–2.0 11 22.0 2.8–171.5 2nd trimester of pregnancy 45 1.5 1.0–2.3 9 4.5 1.4–14.7 Maternal age at conception 3rd trimester of pregnancy 36 1.2 0.8–1.9 9 2.2 0.8–5.8 <20 years 11.0 11.9 1st year of life 60 1.2 0.8–1.8 10 2.0 0.8–4.8 20–24 years 33.0 30.1 Anytime 66 1.4 0.9–2.0 13 2.2 1.0–4.8 25–29 years 28.7 26.3 Insecticides 30–34 years 14.7 18.3 ≥35 years 11.7 13.0 1 year before conception 41 1.4 0.9–2.1 7 4.6 1.2–17.8 Missing 1.0 0.5 1st trimester of pregnancy 20 1.2 0.7–2.1 7 6.9 1.4–33.2 Maternal smoking during pregnancy 2nd trimester of pregnancy 21 1.2 0.7–2.0 3rd trimester of pregnancy 19 2.2 1.2–4.1 7 3.4 1.0–11.8 No 92.3 94.6 1st year of life 35 1.3 0.8–2.0 6 2.9 0.8–10.5 Yes 5.0 3.8 Anytime 44 1.4 0.9–2.1 9 3.0 1.0–8.4 Missing 2.7 1.6 Herbicides Maternal alcohol intake during pregnancy 1 year before conception 53 1.2 0.8–1.7 9 2.0 0.8–5.0 No 95.1 94.8 1st trimester of pregnancy 35 1.4 0.9–2.1 Yes 4.1 5.2 8 5.3 1.4–20 2nd trimester of pregnancy 37 1.6 1.0–2.5 Missing 1.0 .. 3rd trimester of pregnancy 31 1.3 0.8–2.1 7 2.3 0.8–6.8 Maternal education 1st year of life 53 1.3 0.8–1.9 7 1.5 0.6–4.1 ≤Primary school 51.9 54.2 Anytime 60 1.4 0.9–2.0 11 1.4 0.9–2.0 >Primary to secondary school 34.9 32.7 Fungicides Technical or professional 13.1 13.0 1 year before conception 30 1.6 1.0–2.6 4 7.7 0.9–69.7 Missing 3 2.1 1st trimester of pregnancy 21 1.7 0.9–3.0 a 4 7.8 0.9–70.6 Socioeconomic status of family 2nd trimester of pregnancy 16 1.2 0.6–2.3 Low 29.7 22.3 3rd trimester of pregnancy 16 1.7 0.9–3.4 4 7.8 0.9–70.6 Middle 46.3 58.7 1st year of life 28 1.5 0.9–2.5 4 2.6 0.6–11.8 High 10.3 10.0 Anytime 36 1.9 1.1–3.0 6 1.9 1.1–3.0 Missing 13.7 9.0 Adjusted for residence (urban or rural). a b Assessed by interviewers with a form using various components—scored This OR is based on 11 exposed case mothers and 1 exposed control as low, medium, or high. mother. The unadjusted OR was 21.6 (95% CI 2.8–168.0). Scand J Work Environ Health 2007, vol 33, no 4 297 Pesticides exposure and childhood leukemia were above 1.0. The fathers’ exposures showed an excess trimester for acute lymphocytic leukemia. Exposure to risk for all of the pesticides during the second trimester, “other” pesticides showed elevations for most of the insecticides during the third trimester, herbicides dur- time periods for total leukemias with respect to the year ing second trimester, and fungicides in most of the time before conception and to the first, second, and third periods and anytime. trimester and for all of the time windows for acute lym- The odds ratios were, on the average, higher among phocytic leukemia in respect to the year before concep- the mothers than among the fathers; this finding indi - tion, the first, second and third trimesters, and the first cated higher relative, but not necessarily higher absolute, year of life. Other groups of pesticides had few, if any, excesses. The numbers of exposed mothers were smaller exposed mothers. than those of the fathers. For the mothers, increased Table 5 shows the associations, by chemical group, odds ratios were found for all exposure to pesticides for for the father’s time period exposures with respect to the year prior to conception, during the first and second total leukemias. Increased odds ratios were found for trimester, and anytime; for exposure to insecticides exposure to organophosphates during the year before during the year before conception, during the third conception and the first trimester and for benzimid - trimester, and anytime; exposure to herbicides during azoles during the first, second, and third trimesters of the first and second trimesters (high OR 13.8 and 4.6, pregnancy. The results for acute lymphocytic leukemia respectively); and exposures to fungicides anytime. Six were similar (not tabulated). mothers of cases were exposed to insecticides, and three Table 1 (on page 296) shows the associations of spe- were exposed fungicides, during the first trimester, with cific pesticides to total leukemia and acute lymphocytic no exposed controls. These data were not quantifiable in leukemia. In the exposed versus unexposed analysis, the the models, the odds ratio being infinite. father’s exposures to picloram and mancozeb during the The results for acute lymphocytic leukemia were first trimester of pregnancy and to foxim exposure dur - similar with the exception of exposure to insecticides for ing the first year of life of the child showed an associa - the fathers during the third trimester of pregnancy, which tion with acute lymphocytic leukemia. Foxim exposure was not significant for acute lymphocytic leukemia. of the father during the first year of life of the child was Table 4 shows the odds ratios for exposure to pesti- associated with total leukemia. The mother’s exposures cides according to the chemical group for the mothers to paraquat during the year before conception and the with respect to total leukemia and acute lymphocytic second trimester of pregnancy were associated with leukemia. Increases were found for exposure to organo- acute lymphocytic leukemia. In the exposure–response phosphates in the first trimester for total leukemias and analyses contrasting high and low exposure, excluding for exposure to organophosphates in the first and third the unexposed, excesses were found for total leukemia Table 4. Odds ratios (OR) and 95% confidence intervals (95% CI) for total leukemias and acute lymphocytic leukemia with respect to the mothers’ exposure to pesticides to which >3 cases had been exposed, according to chemical group—exposed versus unexposed. (N = number of exposed cases) Time period Total leukemia Acute lymphocytic leukemia a a N OR 95% CI N OR 95% CI Year before conception Phenoxyacetic acids 4 1.3 0.4–4.8 . . .. Organophosphates 9 2.3 0.9–6.0 8 2.5 0.9–6.7 Paraquat, chlorothalonil, glyphosate and others 11 2.8 1.1–7.2 10 3.1 1.2–8.1 1st trimester Organophosphates 7 3.5 1.0–12.2 6 3.7 1.0–13.1 Paraquat, chlorothalonil, glyphosate and others 9 3.7 1.2–11.1 8 4.0 1.8–12.3 2nd trimester Organophosphates 5 2.5 0.7–9.5 5 3.0 0.8–11.5 Paraquat, chlorothalonil, glyphosate and others 8 3.3 1.1–10.2 8 4.0 1.3–12.5 3rd trimester Organophosphates 8 2.7 0.9–7.9 8 3.3 1.1–9.6 Paraquat, chlorothalonil, glyphosate and others 9 3.7 1.2–11.2 9 4.5 1.5–13.6 1st year of life Organophosphates 7 1.8 0.6–4.9 6 1.8 0.6–5.4 Paraquat, chlorothalonil, glyphosate and others 9 2.1 0.8–5.3 9 2.5 1.0–6.5 Adjusted for residence (urban or rural). 298 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al Table 5. Odds ratios (OR) and 95% confidence intervals (95% CI) for total leukemia with respect to the fathers’ exposure to pesticides to which >3 cases had been exposed, according to chemical group—exposed versus unexposed. (N = number of exposed cases) Chemical group Total leukemia Year before conception 1st trimester 2nd trimester 3rd trimester First year of life a a a a a N OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI Phenoxyacetic acids 28 1.0 0.6–1.6 13 0.9 0.4–1.7 15 0.9 0.5–1.7 19 1.1 0.6–1.9 29 1.1 0.7–1.7 Organophosphates 48 1.5 1.0–2.2 33 1.6 1.0–2.6 31 1.4 0.8–2.3 30 1.5 0.9–2.5 43 1.3 0.9–2.1 Organochlorines 7 1.0 0.4–2.5 6 1.7 0.6–5.2 5 1.2 0.4–3.8 6 1.5 0.5–4.4 7 1.3 0.5–3.3 Carbamates 14 1.3 0.6–2.5 6 1.2 0.4–3.3 8 1.6 0.6–4.1 7 2.0 0.7–5.8 11 1.1 0.5–2.5 Dithiocarbamates 16 1.4 0.7–2.7 14 1.7 0.8–3.5 11 1.2 0.6–2.7 11 1.2 0.6–2.7 15 1.5 0.7–2.9 Pyrethroids 10 1.2 0.5–2.6 7 1.6 0.6–4.4 7 1.4 0.5–3.8 7 1.4 0.5–3.8 8 0.9 0.4–2.0 Triazines 7 0.8 0.3–1.9 – . . – . . 4 1.0 0.3–3.4 5 0.8 0.3–2.4 Benzimidazoles 19 1.8 0.9–3.4 14 2.1 1.0–4.4 13 2.2 1.0–5.0 12 2.2 1.0–5.2 16 1.7 0.9–3.5 Chlorinated phthalides 8 1.6 0.6–4.1 5 2.5 0.7–9.4 5 3.4 0.8–14 4 2.6 0.6–11.8 7 2.9 0.8–11 Conazoles 8 1.5 0.6–3.7 – . . 4 2.0 0.5–8.1 4 2.0 0.5–7.9 9 1.5 0.6–3.6 Copper 11 1.0 0.5–2.1 6 1.1 0.4–3.0 4 0.7 0.2–2.1 5 0.8 0.3–2.4 8 0.7 0.3–1.7 Others 56 1.1 0.8–1.6 38 1.1 0.7–1.7 37 1.1 0.7–1.8 36 1.1 0.7–1.8 56 1.1 0.8–1.7 Adjusted for residence (urban or rural). Adjusted for exposure to X-rays during pregnancy, in the 2nd trimester, and in the first year of life. Mainly paraquat, chlorothalonil, and glyphosate. among the fathers with respect to exposure to picloram Table 6. Significant (lower 95% confidence limit ≥1.0) excesses for either gender (unexposed versus exposed). (N = number of during the year before conception and during the first exposed cases) year of life of the child (table 1 on page 296). Elevations for total leukemia and acute lymphocytic leukemia were Pesticide Boys Girls also found for the fathers exposed to benomyl during the a a N OR 95% CI N OR 95% CI first year of life of the child, and an increased risk for total leukemia was found for those exposed to paraquat Total leukemia during the year before conception. Insecticides Associations according to the gender of the child Malathion, fathers Year before conception 5 8.5 1.1–74.1 2 0.9 0.2–4.9 were found for the fathers’ exposures, to benomyl, Herbicides mancozeb, and malathion for the boys and to picloram Picloram, fathers for the girls (table 6). 1st trimester 3 1.6 0.3–8.0 5 4.5 1.1–19.2 The analysis of age at diagnosis showed a tendency Fungicides towards early diagnosis (1–5 years) for children whose Benomyl, fathers mothers were exposed to any pesticides or groups of Year before conception 13 2.5 1.1–6.0 4 0.9 0.3–2.8 pesticides. The father’s exposures appeared to be associ- 1st year 12 3.0 1.1–7.7 2 0.5 0.1–2.6 ated with later diagnosis in the life of the child, with the Acute lymphocytic leukemia exception of exposure to fungicides (table 7). For the 14 Insecticides cases diagnosed before their first birthday, the data were Malathion, fathers Year before conception 5 10.4 1.2–91.1 1 0.5 0.1–4.8 too few for analysis. Herbicides Picloram, fathers 1st trimester 3 1.9 0.38–9.8 5 5.2 1.2–22.5 1st year of life 3 0.7 0.18–2.7 7 2.7 1.1–7.8 Discussion Fungicides Benomyl, fathers We found an elevated risk of childhood leukemia in Year before conception 12 2.8 1.1–6.9 3 0.9 0.3–2.8 1st year of life 11 3.3 1.2–8.8 1 0.3 0.1–2.5 association with parents’ occupational exposures to Mancozeb, fathers pesticides prior to and during pregnancy and over the 1st trimester 9 3.1 1.1–9.0 3 1.0 0.3–3.3 first year of life. This finding is in accordance with the results of several studies (8, 14, 16, 17, 19, 22, Adjusted for residence (urban or rural). 23, 35–38). The relative excesses in our study were foxim, paraquat, mancozeb, and malathion, compounds strongest for the mothers and were found for the dif- for which we did not find earlier data on humans with ferent types of biocides and several chemical groups. regard to childhood cancer. We found excesses specifically for picloram, benomyl, Scand J Work Environ Health 2007, vol 33, no 4 299 Pesticides exposure and childhood leukemia chemical groups of pesticides, organophosphates (for the The odds ratios for maternal exposure tended to be first and third trimesters) and “other” pesticides (chloro - higher than for paternal exposure. This difference was thalonil, paraquat, and glyphosate) showed consistently found for all pesticides pooled and exposure at anytime, higher odds ratios for the mothers than for the fathers. before conception, and during the first and second tri - Benzimidazoles had a higher odds ratio for the fathers. mesters of pregnancy; for exposure to insecticides any For specific pesticides, the small number of exposed time, in the year before conception, and during the first mothers precluded comparison. A stronger association and second trimesters; and for exposure to herbicides between childhood leukemia and maternal exposure to during the first and second trimesters of pregnancy. pesticides compared with paternal exposure was also The odds ratios were similar between the mothers and reported by Meinert et al (8). Our data are unclear as to fathers for exposure to fungicides any time, but, for the whether exposures before and during pregnancy, as com- time periods, were considerably higher for the moth- pared with exposures after birth, are more important, ers, although with fewer exposed cases. With regard to whereas Meinert et al (8) found that earlier exposures confer higher risk. Our data suggest that maternal ex- Table 7. Odds ratios (OR) and 95% confidence intervals (95% posures were associated with leukemias diagnosed early CI) for total leukemia and pesticide exposure by age at diagnosis. in life, while paternal exposures tended to be associated (N = number of exposed cases) with later diagnoses (table 7). Parent Age at diagnosis Fathers’ exposures showed an association with child- hood leukemia for insecticides (third trimester), herbi- 1–5 years 6–15 years cides (second trimester), fungicides (anytime and the b b N OR 95% CI N OR 95% CI year before conception), organophosphates (before conception and during the first trimester of pregnancy), Mothers and benzimidazoles (anytime and during all three tri- Total pesticides 1 year before conception 7 3.0 1.1–8.3 4 2.0 0.6–6.6 mesters). The risk excess for paraquat exposure after c c c c 1st trimester of pregnancy 7 27.6 3.4–226.1 4 18.0 2.0–162.2 birth was significant, as was that for mancozeb exposure 2nd trimester of pregnancy 6 5.8 1.6–21.1 3 3.3 0.7–15.2 during the first trimester of pregnancy (acute lympho - 3rd trimester of pregnancy 6 2.9 1.0–8.5 3 1.6 0.4–6.4 cytic leukemia) and foxim exposure after birth and any- 1st year of life 6 2.3 0.8–6.5 4 1.8 0.5–5.8 Fathers time. There was a suggestion of an exposure–response Total pesticides association for paraquat, picloram, and benomyl before 1 year before conception 26 0.8 0.5–1.3 37 1.8 1.1–2.9 conception and after birth (table 1). These results sup- 1st trimester of pregnancy 20 0.9 0.5–1.6 24 1.4 0.8–2.4 port the hypothesis that pesticides are etiologic factors 2nd trimester of pregnancy 21 1.1 0.6–1.8 24 1.6 0.9–2.6 for leukemia, but the findings were not informative 3rd trimester of pregnancy 18 0.9 0.5–1.7 18 1.2 0.7–2.1 1st year of life 26 0.9 0.5–1.4 33 1.6 1.0–2.5 about the exposure routes or mechanisms (39, 40). With Herbicides regard to the gender of the child, benomyl exposure dur- 1 year before conception 23 0.9 0.5–1.5 31 1.7 1.0–2.7 ing the year before conception and during the first year 1st trimester of pregnancy 16 1.0 0.6–1.9 22 1.9 1.1–3.3 of life and malathion exposure during the year before 2nd trimester of pregnancy 19 1.4 0.8–2.5 22 2.1 1.2–3.6 3rd trimester of pregnancy 16 1.1 0.6–2.0 20 1.7 1.0–3.1 conception showed elevations among the boys, whereas 1st year of life 23 0.9 0.5–1.6 31 1.8 1.1–3.0 picloram exposures during the first trimester and after Fungicides birth were associated with acute lymphocytic leukemia 1 year before conception 20 2.0 1.1–3.6 12 1.3 0.6–2.6 1st trimester of pregnancy 13 1.9 1.0–3.9 9 1.5 0.7–3.3 among the girls. The reasons for these differences are 2nd trimester of pregnancy 12 1.7 0.8–3.5 5 0.8 0.3–2.1 not apparent. 3rd trimester of pregnancy 11 2.2 1.0–4.8 6 1.3 0.5–3.4 All of the six pesticides (picloram, paraquat, manco- 1st year of life 16 1.6 0.9–3.1 12 1.4 0.7–2.8 zeb, benomyl, malathion, and foxim) that were associ- Organophosphates 1 year before conception 23 1.3 0.8–2.2 24 1.7 1.0–2.9 ated with a risk of childhood leukemia in our data have 1st trimester of pregnancy 12 1.0 0.5–2.0 20 2.3 1.3–4.2 been previously found to be associated with chromo- 2nd trimester of pregnancy 14 1.2 0.6–2.2 16 1.7 0.9–3.1 some aberrations or mutagenicity [41–45; unpublished 3rd trimester of pregnancy 14 1.3 0.7–2.5 15 1.7 0.9–3.2 1st year of life 19 1.1 0.6–1.9 23 1.7 1.0–2.9 data: US Environmental Protection Agency, Office of Paraquat Drinking Water. Paraquat Health Advisory (draft)]. 1 year before conception 14 0.7 0.4–1.3 24 1.6 1.0–2.8 However, with the exception of malathion, for which 1st year of life 12 1.1 0.6–2.0 25 2.0 1.2–3.5 detectable mutations were associated with its exposure Data for first-year diagnosis were insufficient. One mother and one father in three different types of human culture cells, including of the 14 cases diagnosed during this year were exposed to any pesticides white blood cells and lymph cells (42, 43), the data are during the exposure assessment period. Adjusted for residence (urban or rural). inconsistent and inconclusive (41–45). Benomyl and Unstable estimate due to the small number (1) of controls in each cat- mancozeb have been found to be associated with cancer egory of age at diagnosis. Only sets with at least one P(OR)≤0.05. in humans (46, 47), but not with leukemia. 300 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al The mothers’ and fathers’ exposures were correlated assessment involved a “prompt list” of pesticides, which because they often shared the same agricultural setting, should have increased the sensitivity of the exposure and possibly due to similarity in the pattern of respond- assessment but may have decreased the specificity. ing during the interview. The correlation coefficient These effects were probably nondifferential between between the mothers and fathers, calculated for odds the cases and controls and would have tended to bias ratios for insecticides, herbicides and fungicides in the the odds ratios towards unity. The only data we have on five time periods (table 3), was moderately low (0.32). It the original control children who were replaced because remains uncertain whether the case mothers and fathers of incomplete addresses were geographic location and would recall their job details (which were the basis for age, both of which were retained in the replacement of the assessment of their exposures) better than the con- the control children. trols did. This uncertainty leaves room for a possible We did not adjust for multiple comparisons, as we positive information bias. Exposure assessment was were generally not dealing with joint or overall null blinded to the case–control status of the children. hypotheses (48). Exposures were correlated between the time periods After three decades of research, the role of parental (ie, those exposed during one time period were likely occupational exposures in the development of childhood to be exposed also during other time periods). Thus the leukemia remains unclear (11). Pediatric leukemias have correlation coefficients between the father’s exposure multifactorial etiologies involving both human genetics (binary yes, no) to organophosphate pesticides during and interactions between environmental factors (39). the five time periods ranged from 0.66 to 0.85; those for Previous studies with other populations have reported the father’s exposure to phenoxyacetic herbicides were associations of childhood leukemia with environmental 0.57–0.84; those for the father’s exposure to paraquat exposure to pesticides. Children of farmers and farm were 0.55–0.72; and those for the mother’s exposure to workers are exposed to agricultural chemicals in utero organophosphate pesticide were 0.60–0.82. through transplacental transmission during gestation and As the causes of childhood leukemia are poorly also during the postnatal period through diverse path- known, there is little a priori basis for the selection and ways. They live or work on farms with their parents and inclusion of confounders in analytic models. Using a list come into contact with agricultural chemicals through of preselected possible confounders, we calculated cor- direct contact with plants, soil, water, air, and stored relation coefficients for the controls between exposure pesticides through their parents’ take-home exposures, to these possible confounders and pesticide exposures. via direct physical contact with their parents through This procedure allowed us to prioritize the factors to be inhalation of the parents’ breath, workclothes, and skin controlled in the logistic models (urban or rural resi- and via other routes (10). dence in most of the models) and discard variables with Epidemiologic studies of pesticides and childhood low correlations (absolute maximum value |r|=0.17) with leukemia have been reviewed earlier (7, 11, 27). Some the pesticide variables. We do believe that using other studies have investigated pesticide exposures of the child means of selecting confounders would have changed the at home, in gardens, or on farms (8, 19, 35–37). A vari- results to a meaningful extent. ety of exposure assessment methods was used in these Both the cases and controls were selected from the studies. Some used parental occupation as a surrogate general population. The procedure for selecting and lo- for pesticide exposure (14, 17, 29, 35), others included cating the controls probably ensured the representation interviews or self-reports (8, 19, 36, 49), and others of the general infant and child population from which used the inclusion in registers to represent exposure the cases were identified. The controls were replaced (23, 50, 51). Some differentiated between maternal and by age group and the place of residence of the original paternal exposures (8, 14, 35, 36, 52). Recommenda- control. Thirty-four cases were not included in the study tions included improving exposure assessment methods, (19 refused to participate and 15 could not be located). particularly for maternal exposure. Our exposure assess- The distribution of the year of diagnosis of the 34 cases ment included a sophisticated interview form to improve not included in the study (19 refusals and 15 not located) recall and a model for specific pesticides and chemical was similar to that of the cases in the analysis. The ad- groups that incorporated aspects of time (frequency of dresses of the unincluded cases were urban for 58% use, length of application), application technique, and and rural or semirural for 42%; these percentages are protective practices. We assessed exposures for fathers similar to those of the addresses of the case population. and mothers separately and integrated interview infor- There was no information available on whether the 14 mation and external data on time-, locality-, and crop- rural or semirural cases were exposed differently than specific pesticide application practices (31, 32). the cases included in the analysis. Misclassification of A prioritization system for specific pesticides based exposure due to interview-based assessment was pos- on toxicity and frequency of use was also used by sible in our study. Nevertheless, our model for exposure Reynolds et al (23) in a case–control study of childhood Scand J Work Environ Health 2007, vol 33, no 4 301 Pesticides exposure and childhood leukemia leukemia using a geographic information system in the Keifer, Viria Bravo, Lawrence Engels, Clemens Ruepert, exposure assessment of the mothers during pregnancy Igor Burstyn, Moniek Zuurbier, Ana Cecilia Rodríguez, on the basis of the distance of the home from exposure and Patricia Stewart who helped during the different sources. They found excess risks for areas with high stages of this project. We thank also Tatiana Sánchez for application volumes of metam sodium and dicofol. Ex- her help with the drawings, Fabio Chaverri and Fernando posures to these two pesticides were not assessed. Ramírez for their review of the external data on pesticide This report addresses sources of occupational expo- application used in the model, and Margarita Mena for sure. The results concerning environmental and house- her administrative support. hold exposures to pesticides will be reported separately, as those for other reported risk factors, such as radiation, organic solvents, medications, early infectious diseases References in childhood, immunologic aspects, and genetic suscep- tibility. In conclusion, we found associations between 1. Little J. Epidemiology of childhood cancer. Lyon: International occupational parental pesticide exposures and childhood Agency for Research on Cancer (IARC); 1999. IARC scientic fi leukemia in a developing country with heavy use of pes- publications 149. ticides and a high rate of childhood leukemia. 2. Parkin DM, Kramárová E, Draper GJ, Masuyer E, Michaelis We recommend the prevention of pesticide hazards, J, Neglia J, et al. International incidence of childhood cancer, especially in the population stratum reproductively ac- vol II. Lyon: International Agency for Research on Cancer tive and among children. The C139 convention of the (IARC); 1998. IARC scientific publications 144. 3. Committee on Pesticides in the Diets of Infants and Children, International Labour Organization (ILO) concerning the National Research Council. Pesticides in the diets of infants prevention and control of occupational hazards caused and children. Washington (DC): National Academy Press; by carcinogenic substances and agents was established in 1974, taking into account the relevant work of the 4. Van Larebeke N, Birnbaum L, Boogaerts M, Bradke M, Da- World Health Organization (WHO) and the Interna- vis DL, Demarini DM, et al. Unrecognized or potential risk tional Agency for Research on Cancer (IARC) regarding factors for childhood cancer. Int J Occup Environ Health. 2005;11:199–201. cancer research. The convention has been ratified by 5. Shu XO, Stewart P, Wan-Quing W, Dehui H, Potter JD, Buck- several countries, but not by Costa Rica. It establishes ley JD, et at. Parental occupational exposure to hydrocarbons the following issues (i) periodic listing of carcinogenic and risk of acute lymphocytic leukemia in offspring. Cancer substances that would be prohibited or made subject to Epidemiol Biomarkers Prev. 1999;8:783–91. authorization and control, (ii) replacement of carcino- 6. Savitz DA, Chen JH. Parental occupation and childhood genic substances, (iii) exposure minimization, (iv) ap- cancer: review of epidemiological studies. Environ Health Perspect. 1990;88:325–37 propriate system of records, (v) provision of information 7. Zahm SH, Ward MH. Pesticides and childhood cancer. Environ on dangers and protection to workers, (vi) appropriate Health Perspect. 1998;106 Suppl 3: 909–25. medical examinations or biological or other tests for 8. Meinert R, Schuz J, Keltsch U, Kaatsch P, Michaelis J. Leuke- investigations, and (vii) codes of industrial or agricul- mia and non-Hodgkin’s lymphoma in childhood and exposure tural conduct. All of these measures are relevant in the to pesticides: results of a register-based case control study in prevention of childhood cancer through parental occupa- Germany. Am J Epidemiol. 2000;151:639–46. 9. Ries LAG, Smith MA, Gurney JG, Linet M, Tamra T, Young tional exposures, such as carcinogenic pesticides. JL, et al, editors. Cancer incidence and survival among children and adolescents: United States SEER Program 1975–1995. Bethesda (MD): National Cancer Institute, SEER Program; 1999. NIH publication 99–4649. Acknowledgments 10. Curl CL, Fenske R, Kissel JC, Shirai J, Moate T, Griffith W, et al. Evaluation of take-home organophosphorus pesticide ex- posure among agricultural workers and their children. Environ This work was partially supported by the Research Health Perspect. 2002;110:787–92. Department of the Swedish International Cooperation 11. Bufe fl r PA, Kwan ML, Reynolds P, Urayama K. Environmental Agency SAREC/Sida, a grant from the International and genetic risk factors for childhood leukemia: appraising the Scholar in Occupational and Environmental Health, Fog- evidence. Cancer Invest. 2005;21:60–75. arty grant 5 D43 TW00642–07 from the University of 12. John EM, Savitz DA, Sandler DP. Prenatal exposure to Washington (Seattle, WA, USA), and the Intramural Pro- parents’ smoking and childhood cancer. Am J Epidemiol. 1991;133(2):123–32 gram of the United States National Cancer Institute. 13. Doll R, Wakaford R. Risk of childhood cancer from fetal ir- Special thanks go to Rocío Loría-Bolaños, Mari- radiation. Br J Radiol. 1997;70:130–9. anela Rojas, Rebeca Alvarado, Heidi Morales, Carolina 14. Shu XO, Gao YT, Brinton LA, Linet MS, Tu JT, Zheng W, Castillo Abdalla, Jenny Umaña, Gina Solano, Emilia Fraumeni JF Jr. A population-based case-control study of González, Gabriela Rodríguez, and Emma Chacón who childhood leukemia in Shanghai. Cancer. 1988;62:635–44. were responsible for the data collection and to Matthew 15. Infante-Rivard C, Labuda K, Krajinovic M, Sinnett D. Risk of 302 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al childhood leukemia associated with exposure to pesticide and 35. Buckley JD, Robison LL, Swotinsky R, Garabrant DH, with gene polymorphism. Epidemiology. 1999;10:481–7. LeBeau M, Manchester P, et al. Occupational exposures of 16. Petridou E, Dessypris N. Maternal pesticide exposure and parents of children with acute nonlymphocytic leukemia: a childhood leukemia. Epidemiol. 2000;11:230. report from the Children’s Cancer Study Group. Cancer Res. 17. Lowengart RA, Peters JM, Cicioni C, Buckey J, Bernstein L, 1989;49:4030–7. Preston-Martin S, et al. Childhood leukemia and parent’s oc- 36. Meinert R, Kaatasch P, Kaletsch U, Krummenauer F, Miesner cupation an home exposures. J Natl Cancer Inst. 1987;79:39– A, Michaelis J. Childhood leukemia and exposure to pesti- 46. cides: results of a case-control study in Northern Germany. Eur 18. Leiss JK, Savitz DA. Home Home pesticide pesticide use use and and childhood childhood can can- - J Cancer. 1996;32A:1943–8. cer: a case-control study. Am J Public Health. 1995;85(2):249– 37. Abadi-Korek, Stark B, Zaizov R, Shaham J. Parental occupa- 52. tional exposure and the risk of acute lymphoblastic leukemia in 19. Menegaux F, Baruchel A, Bertrand Y, Lescoeur B, Leverger G, offspring in Israel. J Occup Environ Med. 2006;48:165–74. Nelken B, et al. Household exposure to pesticides and risk of 38. Infante-Rivard C, Sinnet D. Preconceptional paternal exposure childhood leukaemia. Occup Environ Med. 2006;63:131–4. to pesticides and increased risk of childhood leukemia. Lancet. 20. Reeves JD. Household insecticide-associated blood dyscrasias 1999;354:1819. in children. Am J Hematol Oncol. 1982;4:438–9 39. Greaves M. Childhood leukaemia. Br Med J. 2002;324:283– 21. Infante PF, Epstein SS, Newton WA Jr. Blood dyscrasias and 87. childhood tumors and exposure to chlordane and heptachlor. 40. Alexander FE, Patheal SL, Biondi A, Brandalise S, Cabrera Scand J Work Environ Health. 1978;4:137–50. ME, Chan LC, et al. Transplacental chemical exposure and 22. Reeves JD. Household insecticide associated aplastic anaemia risk of infant leukemia with MLL gene fusion. Cancer Res. and acute leukaemia in children. Lancet. 1981;8;2:300–1. 2001;61:2542–6. 23. Reynolds P, Von Behren J, Gunier R, Goldberg D, Harnly M, 41. National Research Council. Drinking water and health, vol. 5. Hertz A. Agricultural pesticide use and childhood cancer in Washington (DC): Board on Toxicology and Environmental California. Epidemiology. 2005;16:93–100. Health Hazards, Commission on Life Sciences, Safe Drinking 24. Monge P, Wesseling C, Rodríguez AC, Cantor K, Weider- Water Committee, National Academy Press; 1983. pass E, Reutfors, et al. Childhood leukemia in Costa Rica, 42. US Environmental Protection Agency. Memorandum from 1981–1996. Paediatr Perinatal Epidemiol. 2002;16:210–8. the Office of Pesticides and Toxic Substances to Office of 25. Wesseling C, Aragón A, Castillo L, Corriols M, Chaverri F, de Pesticide Programs Division Director. Washington (DC): US la Cruz E, et al. Hazardous pesticides in Central America. Int Environmental Protection Agency; 1991. J Occup Environ Health. 2001;7:287–94 43. US Public Health Service. Hazardous substance data bank. 26. Chaverri F, Blanco J. Importaciones, formulación y uso de Washington (DC): US Public Health Service; 1995. plaguicidas en Costa Rica [Imports, formulation and use of 44. Edwards IR, Ferry DG, Temple WA. Fungicides and related pesticidas in Costa Rica]. Heredia (Costa Rica): Universidad compounds. In: Hayes WJ, Laws ER. Handbook of pesticide Nacional, Instituto Regional de Estudios en Sustancias Tóxi- toxicology. New York (NY): Academic Press; 1991. cas: Editorial, Universidad Nacional; 2002. 45. Jablonicka A, Polakova H, Karelova J, Vargova M. Analysis 27. Daniels JL, Olshan AF, Savitz DA. Pesticides and childhood of chromosome aberrations and sister-chromatid exchanges cancers. Environ Health Perspect. 1997;105:1068–77. in peripheral blood lymphocytes of workers with occupational 28. Colt JS, Blair A. Parental occupational exposures and risk of exposure to the mancozeb-containing fungicides Novozir childhood cancer. Environ Health Perspect. 1998;106 suppl Mn80. Mutat Res. 1989;224:143–6. 3:909–25. 46. US Environmental Protection Agency (US EPA). List of 29. Feychting M, Plato N, Nise G, Ahlbom A. Paternal occupation- chemicals evaluated for carcinogenic potential. Washington al exposures and childhood cancer. Environ Health Perspect. (DC): US EPA, Office of Pesticide Programs; 1996. 2001;109:193–6. 47. US Environmental Protection Agency. Pesticide fact sheet 30. Miettinen OS. Theoretical epidemiology: principles of occur- number 125: mancozeb. Washington (DC): Ofc fi e of Pesticides rence research in medicine. New York (NY): Wiley; 1985. and Toxic Substances; 1987. 31. Monge P, Wesseling C, Engel L, Keifer M, Zuurbier M, Rojas 48. Rothman KJ. No adjustments are needed for multiple compari- M, et al. An icon-based interview for the assessment of occupa- sons. Epidemiology. 1990;1:43–6. tional pesticide exposure in a case-control study of childhood 49. Flower KB, Hoppin JA, Lynch CF, Blair A, Knott C, Shore leukemia. Int J Occup Environ Health. 2004;10:72–8. DL, Sandler DP. Cancer risk and parental pesticide application 32. Monge P, Partanen T, Wesseling C, Bravo V, Ruepert C, in children of agricultural health study participants. Environ Burstyn I. Assessment of pesticide exposure in the agricultural Health Perspect. 2004;112:631–5. population of Costa Rica. Ann Occup Hyg. 2005;49(5):375– 50. Kristensen P, Andersen A, Irgens LM, Bye AS, Sundeheim L. 84. Cancer in offspring of parents engaged in agricultural activities 33. Valcke M, Chaverri F, Monge P, Bravo V, Partanen T, Wessel- in Norway: incidence and risk factors in the farm environment. ing C. Pesticide prioritization for a case-control study on child- Int J Cancer. 1996;65:39–50. hood leukaemia in Costa Rica: a simple stepwise approach. 51. Reynolds P, Von Behren J, Gunier R, Goldberg D, Hertz A. Environ Res. 2005;97:335–47. Childhood cancer and agricultural pesticide use: an ecologic 34. Wesseling C, Bravo V. Pesticide use on the main crops of Costa study in California. Environ Health Perspect. 2002;9:319–24. Rica 1970–2000: data for retrospective exposure assessment 52. VanSteensel-Mol HA, Valkenburg HA, Van Zanen GE. Child- in epidemiologic studies: report to the US National Cancer hood leukemia and parental occupation: a register-based case- Institute. Heredia (Costa Rica): Central American Institute for control study. Am J Epidemiol. 1985;121:216–24. Studies on Toxic Substances (IRET), Universidad Nacional; 2002. Received for publication: 7 September 2006 Scand J Work Environ Health 2007, vol 33, no 4 303 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Scandinavian Journal of Work, Environment & Health Unpaywall

Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica

Loading next page...
 
/lp/unpaywall/parental-occupational-exposure-to-pesticides-and-the-risk-of-childhood-BqdpMi460m

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0355-3140
DOI
10.5271/sjweh.1146
Publisher site
See Article on Publisher Site

Abstract

Downloaded from www.sjweh.fi on July 08, 2021 Original article Scand J Work Environ Health 2007;33(4):293-303 doi:10.5271/sjweh.1146 Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica by Monge P, Wesseling C, Guardado J, Lundberg I, Ahlbom A, Cantor KP, Weiderpass E, Partanen T Affiliation: Central American Institute for Studies on Toxic Substances (IRET), Universidad Nacional, PO Box 863000, Heredia, Costa Rica. pmonge@una.ac.cr Key terms: cancer epidemiology; casecontrol study; child; childhood cancer; childhood leukemia; Costa Rica; developing country; fetal exposure; parental occupational exposure; pesticide; pregnancy; reproductive effect; risk; tropics This article in PubMed: www.ncbi.nlm.nih.gov/pubmed/17717622 This work is licensed under a Creative Commons Attribution 4.0 International License. Print ISSN: 0355-3140 Electronic ISSN: 1795-990X Original article Scand J Work Environ Health 2007;33(4):293–303 Parental occupational exposure to pesticides and the risk of childhood leukemia in Costa Rica 1, 2 1, 2 1 by Patricia Monge, PhD, Catharina Wesseling, PhD, Jorge Guardado, LicComp, Ingvar Lundberg, 2, 3 4 5 6, 7 1 PhD, Anders Ahlbom, PhD, Kenneth P Cantor, PhD, Elisabete Weiderpass, PhD, Timo Partanen, PhD Monge P, Wesseling C, Guardado J, Lundberg I, Ahlbom A, Cantor KP, Weiderpass E, Partanen T. Parental oc- cupational exposure to pesticides and risk of childhood leukemia in Costa Rica. Scand J Work Environ Health 2007;33(4):293–303. Objectives Parental exposure to pesticides and the risk of leukemia in offspring were examined in a popula- tion-based case–control study in Costa Rica. Methods All cases of childhood leukemia (N=334), in 1995–2000, were identified at the Cancer Registry and the Children’s Hospital. Population controls (N=579) were drawn from the National Birth Registry. Interviews of parents were conducted using conventional and icon-based calendar forms. An exposure model was constructed for 25 pesticides in five time periods. Results Mothers’ exposures to any pesticides during the year before conception and during the r fi st and second trimesters were associated with the risk [odds ratio (OR) 2.4, 95% cond fi ence interval (95% CI) 1.0–5.9; OR 22, 95% CI 2.8–171.5; OR 4.5, 95% CI 1.4–14.7, respectively] and during anytime (OR 2.2, 95% CI 1.0–4.8). An as- sociation was found for fathers’ exposures to any pesticides during the second trimester (OR 1.5, 95% CI 1.0–2.3). An increased risk with respect to organophosphates was found for mothers during the r fi st trimester (OR 3.5, 95% CI 1.0–12.2) and for fathers during the year before conception and the r fi st trimester (OR 1.5, 95% CI 1.0–2.2 and OR 1.6, 95% CI 1.0–2.6, respectively), and benzimidazoles during the r fi st, second, and third trimesters of pregnancy (OR 2.2, 95% CI 1.0–4.4; OR 2.2, 95% CI 1.0–5.0; OR 2.2, 95% CI 1.0–5.2, respectively). There was a suggestion of an exposure–response gradient for fathers as regards picloram, benomyl, and paraquat. Age at diagnosis was positively associated with fathers’ exposures and inversely associated with mothers’ exposures. Conclusions The results suggest that parental exposure to certain pesticides may increase the risk of leukemia in offspring. Key terms cancer epidemiology; case–control study; childhood cancer; children; developing country; fetal exposure; pregnancy; reproductive effect; tropics. Leukemias are the most common childhood cancers (1, than adults who receive such exposures later in life, due 2), and their etiology is not well understood. Known or to different physical dimensions, less mature immune suspected general risk factors include male gender, age systems, unique diets, and metabolic characteristics 1–4 years, particular genetic polymorphisms, ethnicity (3, 4). Parental exposures to occupational hazards may (Caucasian), high socioeconomic status, small families, contribute to the risk of leukemia or other cancers in low birthweight, and chemical and physical risk factors offspring through damage to germ cells of either parent such as ionizing radiation, electromagnetic e fi lds, chemical prior to pregnancy; intrauterine and early extrauterine agents, dusts, fumes, spores, drugs and infections (1, 2). exposure via maternal or paternal exposure to toxic Environmental exposures during development and compounds and their metabolites through transplacental childhood may place children at a higher leukemia risk transmission during gestation; or directly through breast Central American Institute for Studies on Toxic Substances, Universidad Nacional, Heredia, Costa Rica. Department of Public Health Sciences, Karolinska Institutet, Stockholm, Sweden. National Institute for Working Life, Stockholm, Sweden. National Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden. Division of Cancer Epidemiology and Genetics, National Cancer Institute, Bethesda, MD, USA. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden. Cancer Registry of Norway, Oslo, Norway. Reprint requests to: Dr P Monge, Central American Institute for Studies on Toxic Substances (IRET), Universidad Nacional, PO Box 86–3000, Heredia, Costa Rica. [E-mail: pmonge@una.ac.cr] Scand J Work Environ Health 2007, vol 33, no 4 293 Pesticides exposure and childhood leukemia feeding, take-home exposure through contaminated Diseases (ICD) 0–1]. Population controls (N=579), person(s) or workclothes during the postnatal period frequency matched to the cases by birth year, were (5–11). Bone marrow originates from the mesoderm, and drawn from the National Birth Registry, with the use of its development occurs at the end of the first trimester of computerized random selection. The number of controls pregnancy. Exposure to teratogens during this trimester was chosen using cost-efficiency considerations (30). could sensitize embryos to problems in development, As Costa Rica uses landmarks to indicate addresses including childhood leukemia. Radiation during the first instead of street names and house numbers, we estab- trimester has been found to be associated with higher lished a procedure to increase the likelihood of locating risks of leukemia in children (12, 13). families while maintaining randomness and allowing for An increased risk of childhood leukemia has been neighborhood replacement of nontraceable controls. Ad- found to be associated with occupational paternal exposure dresses and sometimes telephone numbers of the cases to pesticides prior to and during pregnancy (8, 14–16) and were obtained from files of the National Children’s Hos - parents’ pesticide exposures at home or in gardens (8, pital. In most cases, this information was adequate for 17–19). Pesticides associated with childhood leukemia in- tracing the family. When information was inadequate, clude chlordane, dichlorvos, monomethyldithiocarbamate we used databases of the local social security clinics. (metam sodium), propoxur, and dicofol (20–23). For the controls, the Birth Registry provided an address The median incidence of childhood leukemias for of the mother at the time of birth, which was sometimes 58 populations with cancer registry-based data for the complete but often restricted to a neighborhood. Using 1990s was 45 (range 25–64) per million person-years. national electoral databases and local social security Costa Rica ranks third, with a rate of 63 per million, clinics, we were able to ascertain exact addresses for close to the 64 per million for Latin populations in Los 62% of the controls. If the family was not located but Angeles (24). To date, no etiologic studies have been information on a new address was available, we at- conducted in Costa Rica. tempted to locate them at the new address. Otherwise, Agriculture, an important economic activity in Costa the control was replaced with a child of the same age Rica, is associated with excessive and inappropriate use living in the same neighborhood. If the address was un- of pesticides (25). An annual average of 2.5 kg of active traceable, a control child of the same age was randomly pesticide ingredient per inhabitant was estimated for the selected from the archives of the health center in the year 1996, compared with 0.7 kg in the Netherlands, for neighborhood mentioned on the birth certificate. The example, which represents a high European rate (26). overall response rate was 90% for the cases (19 refused Several pesticides that have been linked to childhood to participate and 15 were not located) and 90.5% for the leukemia are in widespread use in Costa Rica. Propoxur controls (55 refused to participate). Refusals were also is used in commercial domestic pest control products, substituted with controls in the same neighborhood. The and metam sodium and dicofol in ornamental plants and effective number of cases was 300; that of controls was fruit growing, among other crops. Dichlorvos is used for 579 (figure 1). Case diagnoses were extracted directly crops, as well as for livestock. from Cancer Registry data and confirmed from files of Most agricultural workers experience complex ex- the Children’s Hospital. The study was approved by the posure patterns to pesticides over time, the result being Institutional Review Boards (IRB) of the Children’s serious methodological limitations in many studies Hospital, Ministry of Health of Costa Rica, and the addressing pesticide exposures and childhood cancer, Karolinska Institutet in Sweden. All of the participants calling for improved exposure assessment with regard to gave informed consent prior to the interviews. chemical type and timing (7, 23, 27–29). With this need We conducted face-to-face interviews with parents in mind, we conducted a population-based case–control in 2001–2003 using an interview form to collect demo- study to evaluate associations between parental occu- graphic data and data on known and suspected risk fac- pational exposure to pesticides and the risk of leukemia tors of childhood leukemia. Parents who were active in in offspring. agriculture or livestock production during the assessment period completed an additional interview that utilized an icon-calendar form (ICF) (31). The response rate of ICF completion, based on sufficient and reasonably good Study population and methods quality data, was 90%, with interviews of 83 cases and 139 control parents. After the data were cleaned and Ascertainment of cases and controls some parental questionnaires with incomplete data were excluded, data for 876 mothers and 762 fathers remained All cases of childhood leukemia (ages 0–14 years at di- for the analyses (figure 1). agnosis, N=334) diagnosed in Costa Rica in 1995–2000 Out of the parents of the cases, 16.9% were active in were identified at the Cancer Registry and the Children’s agriculture; of the parents of the controls, 15.6% were Hospital of Costa Rica [International Classification 294 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al Figure 1. active. With consideration for the small proportion of Cases Controls parents outside the active labor force, these figures are (334) (579) comparable with the 18.2% of the total labor force in Refusals Refusals agriculture in the national census of 2000. (19) (55) Not Not located located (215) Interview data on exposures and their determinants (15 *) A conventional interview format was offered to both par- ents. It had three forms, one each for the mother, father Neighborhood control and child (completed by either parent). The interview replacement included the location of the family residence; education (270) of the father and mother; smoking and substance abuse of the father and mother and diet of the mother; medical Cases Controls history of the mother, including vaccinations, the moth- Both parents (267) Both parents (492) Mother only (32) Mother only (85) er’s X-rays and medications; pregnancy history; birth Father only (1) Father only (2) parameters; occupational, environmental and home pes- ticide exposures of both parents; occupational histories of the parents; and exposures of the father and mother to In agriculture: nonpesticide toxic agents. For those in agriculture, data Quantitative and qualitative ICF on the parental use of pesticides, agricultural tasks, fre- (83) (139) Qualitative ICF only quency of exposure (number of applications per month (6) (22) and hours per day), major determinants of pesticide ex- Missing data posure (task technology, personal protective equipment, (5) (6) field reentry, storing of pesticides, personal hygiene) * For 8 cases, addresses were not located, 6 were living in other For 8 case countries and 1 entire family had died. s addresses were not located, 6 were living in other countries and 1 entire were extracted with the ICF for the etiologically relevant family had died. period on a month-to-month basis (32). The relevant Figure 1. Study structure. (ICF = icon-based calendar form) period was taken as the period from 12 months before conception until the diagnosis of cancer for the cases and oxamyl, quintozene, and aldrin (33). Cyproconazole, until either the interview date or the age of 15 years for foxim, and fenamiphos were also included because the controls, whichever occurred first. their use was cited frequently by the participants. We The tasks were classified according to their estimated constructed two exposure assessment models. A quan- hazard (0 = no exposure; 4 = very high exposure). titative model combined ICF data with external data on Tasks with no exposure (0) included coffee picking the 25 prioritized pesticides for 14 crops, 21 calendar and organic agriculture. Tasks with a hazard value of years, and 14 geographic regions. Since no measurement 1 included tasks such as the watering of plants, tractor data were available, we used the application rates of driving, and garden maintenance; those with a hazard pesticides from the external data as a basic component value of 2 included fertilizing and defoliating; those of the estimated exposure intensity. The application rates with a hazard value of 3 included backpack application were obtained from a database of the Central American and the mixing of pesticides; and those with a hazard Institute for Studies in Toxic Substances-Universidad value of 4 included the application of pesticides with Nacional (IRET-UNA) that draws from databases of the hands (32). ministries, specialized crop offices, and agricultural Checklists of pesticides enhanced the interview as bank loan records (34). Our experts, knowledgeable in prompt lists. The pesticides selected for the checklists pesticide use, checked and corrected these reports for were based on a toxicity-based prioritization system de- any obvious inconsistencies. We converted the calendar signed for this study (33). The interview procedure has time into individual time that had birth of the child as been described earlier in detail by Monge et al (31). the zero time reference (30). Exposures were quasiquan- titatively assessed for five time periods in the model Exposure assessment models that integrated personal data with the external data on pesticide, crop, and geographic region. The periods were The following 22 pesticides were identified as represent - the year before conception, the first trimester, the second ing high priority on the basis of historical data on use trimester, the third trimester, and the first year of life of and toxicity data on 2,4-dichlorophenoxyacetic acid the child (32). (2,4-D), picloram, glyphosate, benomyl, chlorothalonil, The qualitative model used binary (yes, no) indi- paraquat, carbofuran, mancozeb, terbufos, methamido- cators of occupational exposure in agricultural areas, phos, deltamethrin, methomyl, triadimefon, fluazifop, derived from the ICF interview for 14 selected chemical captafol, lead arsenate, malathion, dichlorvos, diuron, Scand J Work Environ Health 2007, vol 33, no 4 295 Pesticides exposure and childhood leukemia groups of pesticides for all time period, in which any their 95% confidence intervals (95% CI) as estimates exposure to one of the compounds of the chemical group of relative risk, separately for the five time periods. was sufficient to trigger “exposure”: phenoxyacetic For preterm birth children, adjustment was applied acids, organophosphates, organochlorines, carbamates, for the conception date. Exposures were expressed dithiocarbamates, pyrethroids, triazines, benzimidazoles, as qualitative (yes, no), semiquantitative (unexposed, chlorinated phthalides, conazoles, copper compounds, low exposure, high exposure) and quantitative metrics arsenic compounds, chlorinated urea derivatives, and for specific pesticides and groups of pesticides. Stick - “other pesticides”. The “other pesticides” group in- ing to the basic concept for confounding, we applied cluded pesticides not in any of the specific pesticide correlation-based selection of confounders instead of groups, the latter group consisting mainly of paraquat, more-refined methods, expecting little difference in the chlorothalonil, and glyphosate. results. Low correlations among the controls between All of the exposure assessment was done blinded as variables of pesticide exposure and maternal age at con- to the case–control status of the children. ception, infectious disease of the child during the first year, the mother’s and child’s exposure to X-rays during pregnancy and first year of life, respectively, mother’s Statistical analysis tobacco and alcohol consumption during pregnancy, Unconditional crude and adjusted logistic regression father’s smoking, and history of newborn jaundice and models were used to estimate the odds ratios (OR) and vaccination of the child resulted in the inclusion only Table 1. Odds ratios (OR) and 95% confidence intervals (95% CI) for parents’ exposure to prioritized pesticides (ie, those to which >3 cases had been exposed)—entries with a ≥1 excess with a lower confidence limit of <1. (N = number of exposed cases) Pesticide Exposed versus unexposed High versus low exposure Total leukemia Acute lymphocytic leukemia Total leukemia Acute lymphocytic leukemia a a N OR 95% CI N OR 95% CI Cases OR 95% CI Cases OR 95% CI Low High Low High Insecticides Foxim Fathers b b b b 1st year of life 5 5.1 1.0–26.5 5 6.1 1.2–31.6 1 4 ∞ 1 4 ∞ Herbicides Picloram Fathers Year before conception 11 1.3 0.6–2.9 11 1.6 0.7–3.4 3 8 6.3 1.0–38.6 3 8 6.3 1.0–38.6 1st trimester 8 2.8 0.9–8.1 8 3.3 1.1–9.8 3 5 3.1 0.3–29.4 3 5 3.1 0.3–29.4 c c c c 1st year of life 10 1.3 0.6–2.8 10 1.5 0.7–3.4 2 8 12.4 1.6–98.3 2 8 12.4 1.6–98.3 Paraquat Fathers Year before conception 39 1.1 0.7–1.7 30 1.0 0.6–1.5 14 25 2.3 1.1–5.2 12 18 1.67 0.7–4.1 1st year of life 45 1.5 1.0–2.3 36 1.4 0.9–2.2 18 27 1.9 0.9–4.1 15 21 1.7 0.7–4.0 Mothers d d d d Year before conception 7 3.4 1.0–11.8 6 3.5 1.0–12.7 2 5 7.5 0.5–122.7 2 4 6.0 0.4–101.6 e e b b b b 2nd trimester 4 7.8 0.9–70.6 4 9.5 1.1–85.5 1 3 ∞ 1 3 ∞ Fungicides Benomyl Fathers 1st year of life 14 1.8 0.9–3.8 12 1.8 0.9–4.0 4 10 5.5 1.1–26.4 3 9 6.6 1.2–35.4 Mancozeb Fathers 1st trimester 11 1.9 0.8–4.3 11 2.3 1.0–5.2 4 7 2.5 0.5–13.2 4 7 2.5 0.5–13.2 Adjusted for residence (urban or rural). No controls with high exposure. Crude P-value from a hypergeometric distribution of >0.10. The unadjusted figures for exposure to picloram during the first year of life and low versus high exposures are OR 8.8 (95% CI 1.34–57).4 for total leuke - mia and OR 7.65 (95% CI 1.35–57.4) for acute lymphocytic leukemia. Unadjusted figures. The unadjusted figures are OR 7.65 (95% CI 0.9–69.0) for total leukemias and OR 9.15 (95% CI 1.0–82.3) for acute lymphocytic leukemia. 296 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al of the urban or rural residence in all of the logistic Results models controlling for unmeasured urban or rural risk factors. The mother’s X-ray exposure during pregnancy The cases and controls were similar with respect to was included because of its moderate correlation with maternal age at conception, the proportion of parents phthalide exposure. The calculations were done with exposed to pesticides, maternal smoking and alcohol STATA Release 8.0 (Stata Corporation, College Station, consumption history during pregnancy, and maternal TX, USA). The unexposed formed the reference group education. The cases were more often boys, and their in all of the analyses, except for those between the high families represented a slightly lower socioeconomic and low exposure groups (table 1 on page 296). The status. Acute lymphocytic leukemia represented 83.7% intratime-period correlation coefficients between pesti - of the cases (table 2). cides were low (median absolute |r|=0.003). Altogether 13.9% of the parents of the cases (24.6% of the fathers and 4.3% of the mothers) was ever ex- posed to any pesticides in work during the exposure Table 2. Characteristics of the cases (total leukemia) and controls. (NOS = not otherwise specified) assessment, while the percentage of the control parents was 11.5% (22.2% for the fathers and 2.2% for the Characteristics Cases Controls mothers). (N=300) (N=579) (%) (%) Table 3 shows the odds ratios for the time windows for all of the pesticides and according to the following Age at diagnosis biocide category: insecticides, herbicides, and fungi- 0 years 4.7 . cides for the fathers and mothers. All of the odds ratios 1–4 years 43.3 . 5–9 years 34.0 . 10–14 years 18.0 . Table 3. Odds ratios (OR) and 95% confidence intervals (95% CI) Gender for parents’ exposure to pesticides to which >3 cases had been Female 44.3 50.8 exposed through biocide action—exposed versus unexposed. Male 55.7 49.2 (N = number of exposed cases) Agricultural exposure to pesticides Maternal 4.3 2.2 Pesticide group Fathers Mothers Paternal 24.6 22.2 a a Leukemia type N OR 95% CI N OR 95% CI Acute lymphocytic leukemia 84.0 . All pesticides Acute nonlymphocytic leukemia 13.7 . Other leukemias 2.3 . 1 year before conception 64 1.2 0.9–1.8 11 2.4 1.0–5.9 NOS 0.3 . b 1st trimester of pregnancy 45 1.3 0.9–2.0 11 22.0 2.8–171.5 2nd trimester of pregnancy 45 1.5 1.0–2.3 9 4.5 1.4–14.7 Maternal age at conception 3rd trimester of pregnancy 36 1.2 0.8–1.9 9 2.2 0.8–5.8 <20 years 11.0 11.9 1st year of life 60 1.2 0.8–1.8 10 2.0 0.8–4.8 20–24 years 33.0 30.1 Anytime 66 1.4 0.9–2.0 13 2.2 1.0–4.8 25–29 years 28.7 26.3 Insecticides 30–34 years 14.7 18.3 ≥35 years 11.7 13.0 1 year before conception 41 1.4 0.9–2.1 7 4.6 1.2–17.8 Missing 1.0 0.5 1st trimester of pregnancy 20 1.2 0.7–2.1 7 6.9 1.4–33.2 Maternal smoking during pregnancy 2nd trimester of pregnancy 21 1.2 0.7–2.0 3rd trimester of pregnancy 19 2.2 1.2–4.1 7 3.4 1.0–11.8 No 92.3 94.6 1st year of life 35 1.3 0.8–2.0 6 2.9 0.8–10.5 Yes 5.0 3.8 Anytime 44 1.4 0.9–2.1 9 3.0 1.0–8.4 Missing 2.7 1.6 Herbicides Maternal alcohol intake during pregnancy 1 year before conception 53 1.2 0.8–1.7 9 2.0 0.8–5.0 No 95.1 94.8 1st trimester of pregnancy 35 1.4 0.9–2.1 Yes 4.1 5.2 8 5.3 1.4–20 2nd trimester of pregnancy 37 1.6 1.0–2.5 Missing 1.0 .. 3rd trimester of pregnancy 31 1.3 0.8–2.1 7 2.3 0.8–6.8 Maternal education 1st year of life 53 1.3 0.8–1.9 7 1.5 0.6–4.1 ≤Primary school 51.9 54.2 Anytime 60 1.4 0.9–2.0 11 1.4 0.9–2.0 >Primary to secondary school 34.9 32.7 Fungicides Technical or professional 13.1 13.0 1 year before conception 30 1.6 1.0–2.6 4 7.7 0.9–69.7 Missing 3 2.1 1st trimester of pregnancy 21 1.7 0.9–3.0 a 4 7.8 0.9–70.6 Socioeconomic status of family 2nd trimester of pregnancy 16 1.2 0.6–2.3 Low 29.7 22.3 3rd trimester of pregnancy 16 1.7 0.9–3.4 4 7.8 0.9–70.6 Middle 46.3 58.7 1st year of life 28 1.5 0.9–2.5 4 2.6 0.6–11.8 High 10.3 10.0 Anytime 36 1.9 1.1–3.0 6 1.9 1.1–3.0 Missing 13.7 9.0 Adjusted for residence (urban or rural). a b Assessed by interviewers with a form using various components—scored This OR is based on 11 exposed case mothers and 1 exposed control as low, medium, or high. mother. The unadjusted OR was 21.6 (95% CI 2.8–168.0). Scand J Work Environ Health 2007, vol 33, no 4 297 Pesticides exposure and childhood leukemia were above 1.0. The fathers’ exposures showed an excess trimester for acute lymphocytic leukemia. Exposure to risk for all of the pesticides during the second trimester, “other” pesticides showed elevations for most of the insecticides during the third trimester, herbicides dur- time periods for total leukemias with respect to the year ing second trimester, and fungicides in most of the time before conception and to the first, second, and third periods and anytime. trimester and for all of the time windows for acute lym- The odds ratios were, on the average, higher among phocytic leukemia in respect to the year before concep- the mothers than among the fathers; this finding indi - tion, the first, second and third trimesters, and the first cated higher relative, but not necessarily higher absolute, year of life. Other groups of pesticides had few, if any, excesses. The numbers of exposed mothers were smaller exposed mothers. than those of the fathers. For the mothers, increased Table 5 shows the associations, by chemical group, odds ratios were found for all exposure to pesticides for for the father’s time period exposures with respect to the year prior to conception, during the first and second total leukemias. Increased odds ratios were found for trimester, and anytime; for exposure to insecticides exposure to organophosphates during the year before during the year before conception, during the third conception and the first trimester and for benzimid - trimester, and anytime; exposure to herbicides during azoles during the first, second, and third trimesters of the first and second trimesters (high OR 13.8 and 4.6, pregnancy. The results for acute lymphocytic leukemia respectively); and exposures to fungicides anytime. Six were similar (not tabulated). mothers of cases were exposed to insecticides, and three Table 1 (on page 296) shows the associations of spe- were exposed fungicides, during the first trimester, with cific pesticides to total leukemia and acute lymphocytic no exposed controls. These data were not quantifiable in leukemia. In the exposed versus unexposed analysis, the the models, the odds ratio being infinite. father’s exposures to picloram and mancozeb during the The results for acute lymphocytic leukemia were first trimester of pregnancy and to foxim exposure dur - similar with the exception of exposure to insecticides for ing the first year of life of the child showed an associa - the fathers during the third trimester of pregnancy, which tion with acute lymphocytic leukemia. Foxim exposure was not significant for acute lymphocytic leukemia. of the father during the first year of life of the child was Table 4 shows the odds ratios for exposure to pesti- associated with total leukemia. The mother’s exposures cides according to the chemical group for the mothers to paraquat during the year before conception and the with respect to total leukemia and acute lymphocytic second trimester of pregnancy were associated with leukemia. Increases were found for exposure to organo- acute lymphocytic leukemia. In the exposure–response phosphates in the first trimester for total leukemias and analyses contrasting high and low exposure, excluding for exposure to organophosphates in the first and third the unexposed, excesses were found for total leukemia Table 4. Odds ratios (OR) and 95% confidence intervals (95% CI) for total leukemias and acute lymphocytic leukemia with respect to the mothers’ exposure to pesticides to which >3 cases had been exposed, according to chemical group—exposed versus unexposed. (N = number of exposed cases) Time period Total leukemia Acute lymphocytic leukemia a a N OR 95% CI N OR 95% CI Year before conception Phenoxyacetic acids 4 1.3 0.4–4.8 . . .. Organophosphates 9 2.3 0.9–6.0 8 2.5 0.9–6.7 Paraquat, chlorothalonil, glyphosate and others 11 2.8 1.1–7.2 10 3.1 1.2–8.1 1st trimester Organophosphates 7 3.5 1.0–12.2 6 3.7 1.0–13.1 Paraquat, chlorothalonil, glyphosate and others 9 3.7 1.2–11.1 8 4.0 1.8–12.3 2nd trimester Organophosphates 5 2.5 0.7–9.5 5 3.0 0.8–11.5 Paraquat, chlorothalonil, glyphosate and others 8 3.3 1.1–10.2 8 4.0 1.3–12.5 3rd trimester Organophosphates 8 2.7 0.9–7.9 8 3.3 1.1–9.6 Paraquat, chlorothalonil, glyphosate and others 9 3.7 1.2–11.2 9 4.5 1.5–13.6 1st year of life Organophosphates 7 1.8 0.6–4.9 6 1.8 0.6–5.4 Paraquat, chlorothalonil, glyphosate and others 9 2.1 0.8–5.3 9 2.5 1.0–6.5 Adjusted for residence (urban or rural). 298 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al Table 5. Odds ratios (OR) and 95% confidence intervals (95% CI) for total leukemia with respect to the fathers’ exposure to pesticides to which >3 cases had been exposed, according to chemical group—exposed versus unexposed. (N = number of exposed cases) Chemical group Total leukemia Year before conception 1st trimester 2nd trimester 3rd trimester First year of life a a a a a N OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI N OR 95% CI Phenoxyacetic acids 28 1.0 0.6–1.6 13 0.9 0.4–1.7 15 0.9 0.5–1.7 19 1.1 0.6–1.9 29 1.1 0.7–1.7 Organophosphates 48 1.5 1.0–2.2 33 1.6 1.0–2.6 31 1.4 0.8–2.3 30 1.5 0.9–2.5 43 1.3 0.9–2.1 Organochlorines 7 1.0 0.4–2.5 6 1.7 0.6–5.2 5 1.2 0.4–3.8 6 1.5 0.5–4.4 7 1.3 0.5–3.3 Carbamates 14 1.3 0.6–2.5 6 1.2 0.4–3.3 8 1.6 0.6–4.1 7 2.0 0.7–5.8 11 1.1 0.5–2.5 Dithiocarbamates 16 1.4 0.7–2.7 14 1.7 0.8–3.5 11 1.2 0.6–2.7 11 1.2 0.6–2.7 15 1.5 0.7–2.9 Pyrethroids 10 1.2 0.5–2.6 7 1.6 0.6–4.4 7 1.4 0.5–3.8 7 1.4 0.5–3.8 8 0.9 0.4–2.0 Triazines 7 0.8 0.3–1.9 – . . – . . 4 1.0 0.3–3.4 5 0.8 0.3–2.4 Benzimidazoles 19 1.8 0.9–3.4 14 2.1 1.0–4.4 13 2.2 1.0–5.0 12 2.2 1.0–5.2 16 1.7 0.9–3.5 Chlorinated phthalides 8 1.6 0.6–4.1 5 2.5 0.7–9.4 5 3.4 0.8–14 4 2.6 0.6–11.8 7 2.9 0.8–11 Conazoles 8 1.5 0.6–3.7 – . . 4 2.0 0.5–8.1 4 2.0 0.5–7.9 9 1.5 0.6–3.6 Copper 11 1.0 0.5–2.1 6 1.1 0.4–3.0 4 0.7 0.2–2.1 5 0.8 0.3–2.4 8 0.7 0.3–1.7 Others 56 1.1 0.8–1.6 38 1.1 0.7–1.7 37 1.1 0.7–1.8 36 1.1 0.7–1.8 56 1.1 0.8–1.7 Adjusted for residence (urban or rural). Adjusted for exposure to X-rays during pregnancy, in the 2nd trimester, and in the first year of life. Mainly paraquat, chlorothalonil, and glyphosate. among the fathers with respect to exposure to picloram Table 6. Significant (lower 95% confidence limit ≥1.0) excesses for either gender (unexposed versus exposed). (N = number of during the year before conception and during the first exposed cases) year of life of the child (table 1 on page 296). Elevations for total leukemia and acute lymphocytic leukemia were Pesticide Boys Girls also found for the fathers exposed to benomyl during the a a N OR 95% CI N OR 95% CI first year of life of the child, and an increased risk for total leukemia was found for those exposed to paraquat Total leukemia during the year before conception. Insecticides Associations according to the gender of the child Malathion, fathers Year before conception 5 8.5 1.1–74.1 2 0.9 0.2–4.9 were found for the fathers’ exposures, to benomyl, Herbicides mancozeb, and malathion for the boys and to picloram Picloram, fathers for the girls (table 6). 1st trimester 3 1.6 0.3–8.0 5 4.5 1.1–19.2 The analysis of age at diagnosis showed a tendency Fungicides towards early diagnosis (1–5 years) for children whose Benomyl, fathers mothers were exposed to any pesticides or groups of Year before conception 13 2.5 1.1–6.0 4 0.9 0.3–2.8 pesticides. The father’s exposures appeared to be associ- 1st year 12 3.0 1.1–7.7 2 0.5 0.1–2.6 ated with later diagnosis in the life of the child, with the Acute lymphocytic leukemia exception of exposure to fungicides (table 7). For the 14 Insecticides cases diagnosed before their first birthday, the data were Malathion, fathers Year before conception 5 10.4 1.2–91.1 1 0.5 0.1–4.8 too few for analysis. Herbicides Picloram, fathers 1st trimester 3 1.9 0.38–9.8 5 5.2 1.2–22.5 1st year of life 3 0.7 0.18–2.7 7 2.7 1.1–7.8 Discussion Fungicides Benomyl, fathers We found an elevated risk of childhood leukemia in Year before conception 12 2.8 1.1–6.9 3 0.9 0.3–2.8 1st year of life 11 3.3 1.2–8.8 1 0.3 0.1–2.5 association with parents’ occupational exposures to Mancozeb, fathers pesticides prior to and during pregnancy and over the 1st trimester 9 3.1 1.1–9.0 3 1.0 0.3–3.3 first year of life. This finding is in accordance with the results of several studies (8, 14, 16, 17, 19, 22, Adjusted for residence (urban or rural). 23, 35–38). The relative excesses in our study were foxim, paraquat, mancozeb, and malathion, compounds strongest for the mothers and were found for the dif- for which we did not find earlier data on humans with ferent types of biocides and several chemical groups. regard to childhood cancer. We found excesses specifically for picloram, benomyl, Scand J Work Environ Health 2007, vol 33, no 4 299 Pesticides exposure and childhood leukemia chemical groups of pesticides, organophosphates (for the The odds ratios for maternal exposure tended to be first and third trimesters) and “other” pesticides (chloro - higher than for paternal exposure. This difference was thalonil, paraquat, and glyphosate) showed consistently found for all pesticides pooled and exposure at anytime, higher odds ratios for the mothers than for the fathers. before conception, and during the first and second tri - Benzimidazoles had a higher odds ratio for the fathers. mesters of pregnancy; for exposure to insecticides any For specific pesticides, the small number of exposed time, in the year before conception, and during the first mothers precluded comparison. A stronger association and second trimesters; and for exposure to herbicides between childhood leukemia and maternal exposure to during the first and second trimesters of pregnancy. pesticides compared with paternal exposure was also The odds ratios were similar between the mothers and reported by Meinert et al (8). Our data are unclear as to fathers for exposure to fungicides any time, but, for the whether exposures before and during pregnancy, as com- time periods, were considerably higher for the moth- pared with exposures after birth, are more important, ers, although with fewer exposed cases. With regard to whereas Meinert et al (8) found that earlier exposures confer higher risk. Our data suggest that maternal ex- Table 7. Odds ratios (OR) and 95% confidence intervals (95% posures were associated with leukemias diagnosed early CI) for total leukemia and pesticide exposure by age at diagnosis. in life, while paternal exposures tended to be associated (N = number of exposed cases) with later diagnoses (table 7). Parent Age at diagnosis Fathers’ exposures showed an association with child- hood leukemia for insecticides (third trimester), herbi- 1–5 years 6–15 years cides (second trimester), fungicides (anytime and the b b N OR 95% CI N OR 95% CI year before conception), organophosphates (before conception and during the first trimester of pregnancy), Mothers and benzimidazoles (anytime and during all three tri- Total pesticides 1 year before conception 7 3.0 1.1–8.3 4 2.0 0.6–6.6 mesters). The risk excess for paraquat exposure after c c c c 1st trimester of pregnancy 7 27.6 3.4–226.1 4 18.0 2.0–162.2 birth was significant, as was that for mancozeb exposure 2nd trimester of pregnancy 6 5.8 1.6–21.1 3 3.3 0.7–15.2 during the first trimester of pregnancy (acute lympho - 3rd trimester of pregnancy 6 2.9 1.0–8.5 3 1.6 0.4–6.4 cytic leukemia) and foxim exposure after birth and any- 1st year of life 6 2.3 0.8–6.5 4 1.8 0.5–5.8 Fathers time. There was a suggestion of an exposure–response Total pesticides association for paraquat, picloram, and benomyl before 1 year before conception 26 0.8 0.5–1.3 37 1.8 1.1–2.9 conception and after birth (table 1). These results sup- 1st trimester of pregnancy 20 0.9 0.5–1.6 24 1.4 0.8–2.4 port the hypothesis that pesticides are etiologic factors 2nd trimester of pregnancy 21 1.1 0.6–1.8 24 1.6 0.9–2.6 for leukemia, but the findings were not informative 3rd trimester of pregnancy 18 0.9 0.5–1.7 18 1.2 0.7–2.1 1st year of life 26 0.9 0.5–1.4 33 1.6 1.0–2.5 about the exposure routes or mechanisms (39, 40). With Herbicides regard to the gender of the child, benomyl exposure dur- 1 year before conception 23 0.9 0.5–1.5 31 1.7 1.0–2.7 ing the year before conception and during the first year 1st trimester of pregnancy 16 1.0 0.6–1.9 22 1.9 1.1–3.3 of life and malathion exposure during the year before 2nd trimester of pregnancy 19 1.4 0.8–2.5 22 2.1 1.2–3.6 3rd trimester of pregnancy 16 1.1 0.6–2.0 20 1.7 1.0–3.1 conception showed elevations among the boys, whereas 1st year of life 23 0.9 0.5–1.6 31 1.8 1.1–3.0 picloram exposures during the first trimester and after Fungicides birth were associated with acute lymphocytic leukemia 1 year before conception 20 2.0 1.1–3.6 12 1.3 0.6–2.6 1st trimester of pregnancy 13 1.9 1.0–3.9 9 1.5 0.7–3.3 among the girls. The reasons for these differences are 2nd trimester of pregnancy 12 1.7 0.8–3.5 5 0.8 0.3–2.1 not apparent. 3rd trimester of pregnancy 11 2.2 1.0–4.8 6 1.3 0.5–3.4 All of the six pesticides (picloram, paraquat, manco- 1st year of life 16 1.6 0.9–3.1 12 1.4 0.7–2.8 zeb, benomyl, malathion, and foxim) that were associ- Organophosphates 1 year before conception 23 1.3 0.8–2.2 24 1.7 1.0–2.9 ated with a risk of childhood leukemia in our data have 1st trimester of pregnancy 12 1.0 0.5–2.0 20 2.3 1.3–4.2 been previously found to be associated with chromo- 2nd trimester of pregnancy 14 1.2 0.6–2.2 16 1.7 0.9–3.1 some aberrations or mutagenicity [41–45; unpublished 3rd trimester of pregnancy 14 1.3 0.7–2.5 15 1.7 0.9–3.2 1st year of life 19 1.1 0.6–1.9 23 1.7 1.0–2.9 data: US Environmental Protection Agency, Office of Paraquat Drinking Water. Paraquat Health Advisory (draft)]. 1 year before conception 14 0.7 0.4–1.3 24 1.6 1.0–2.8 However, with the exception of malathion, for which 1st year of life 12 1.1 0.6–2.0 25 2.0 1.2–3.5 detectable mutations were associated with its exposure Data for first-year diagnosis were insufficient. One mother and one father in three different types of human culture cells, including of the 14 cases diagnosed during this year were exposed to any pesticides white blood cells and lymph cells (42, 43), the data are during the exposure assessment period. Adjusted for residence (urban or rural). inconsistent and inconclusive (41–45). Benomyl and Unstable estimate due to the small number (1) of controls in each cat- mancozeb have been found to be associated with cancer egory of age at diagnosis. Only sets with at least one P(OR)≤0.05. in humans (46, 47), but not with leukemia. 300 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al The mothers’ and fathers’ exposures were correlated assessment involved a “prompt list” of pesticides, which because they often shared the same agricultural setting, should have increased the sensitivity of the exposure and possibly due to similarity in the pattern of respond- assessment but may have decreased the specificity. ing during the interview. The correlation coefficient These effects were probably nondifferential between between the mothers and fathers, calculated for odds the cases and controls and would have tended to bias ratios for insecticides, herbicides and fungicides in the the odds ratios towards unity. The only data we have on five time periods (table 3), was moderately low (0.32). It the original control children who were replaced because remains uncertain whether the case mothers and fathers of incomplete addresses were geographic location and would recall their job details (which were the basis for age, both of which were retained in the replacement of the assessment of their exposures) better than the con- the control children. trols did. This uncertainty leaves room for a possible We did not adjust for multiple comparisons, as we positive information bias. Exposure assessment was were generally not dealing with joint or overall null blinded to the case–control status of the children. hypotheses (48). Exposures were correlated between the time periods After three decades of research, the role of parental (ie, those exposed during one time period were likely occupational exposures in the development of childhood to be exposed also during other time periods). Thus the leukemia remains unclear (11). Pediatric leukemias have correlation coefficients between the father’s exposure multifactorial etiologies involving both human genetics (binary yes, no) to organophosphate pesticides during and interactions between environmental factors (39). the five time periods ranged from 0.66 to 0.85; those for Previous studies with other populations have reported the father’s exposure to phenoxyacetic herbicides were associations of childhood leukemia with environmental 0.57–0.84; those for the father’s exposure to paraquat exposure to pesticides. Children of farmers and farm were 0.55–0.72; and those for the mother’s exposure to workers are exposed to agricultural chemicals in utero organophosphate pesticide were 0.60–0.82. through transplacental transmission during gestation and As the causes of childhood leukemia are poorly also during the postnatal period through diverse path- known, there is little a priori basis for the selection and ways. They live or work on farms with their parents and inclusion of confounders in analytic models. Using a list come into contact with agricultural chemicals through of preselected possible confounders, we calculated cor- direct contact with plants, soil, water, air, and stored relation coefficients for the controls between exposure pesticides through their parents’ take-home exposures, to these possible confounders and pesticide exposures. via direct physical contact with their parents through This procedure allowed us to prioritize the factors to be inhalation of the parents’ breath, workclothes, and skin controlled in the logistic models (urban or rural resi- and via other routes (10). dence in most of the models) and discard variables with Epidemiologic studies of pesticides and childhood low correlations (absolute maximum value |r|=0.17) with leukemia have been reviewed earlier (7, 11, 27). Some the pesticide variables. We do believe that using other studies have investigated pesticide exposures of the child means of selecting confounders would have changed the at home, in gardens, or on farms (8, 19, 35–37). A vari- results to a meaningful extent. ety of exposure assessment methods was used in these Both the cases and controls were selected from the studies. Some used parental occupation as a surrogate general population. The procedure for selecting and lo- for pesticide exposure (14, 17, 29, 35), others included cating the controls probably ensured the representation interviews or self-reports (8, 19, 36, 49), and others of the general infant and child population from which used the inclusion in registers to represent exposure the cases were identified. The controls were replaced (23, 50, 51). Some differentiated between maternal and by age group and the place of residence of the original paternal exposures (8, 14, 35, 36, 52). Recommenda- control. Thirty-four cases were not included in the study tions included improving exposure assessment methods, (19 refused to participate and 15 could not be located). particularly for maternal exposure. Our exposure assess- The distribution of the year of diagnosis of the 34 cases ment included a sophisticated interview form to improve not included in the study (19 refusals and 15 not located) recall and a model for specific pesticides and chemical was similar to that of the cases in the analysis. The ad- groups that incorporated aspects of time (frequency of dresses of the unincluded cases were urban for 58% use, length of application), application technique, and and rural or semirural for 42%; these percentages are protective practices. We assessed exposures for fathers similar to those of the addresses of the case population. and mothers separately and integrated interview infor- There was no information available on whether the 14 mation and external data on time-, locality-, and crop- rural or semirural cases were exposed differently than specific pesticide application practices (31, 32). the cases included in the analysis. Misclassification of A prioritization system for specific pesticides based exposure due to interview-based assessment was pos- on toxicity and frequency of use was also used by sible in our study. Nevertheless, our model for exposure Reynolds et al (23) in a case–control study of childhood Scand J Work Environ Health 2007, vol 33, no 4 301 Pesticides exposure and childhood leukemia leukemia using a geographic information system in the Keifer, Viria Bravo, Lawrence Engels, Clemens Ruepert, exposure assessment of the mothers during pregnancy Igor Burstyn, Moniek Zuurbier, Ana Cecilia Rodríguez, on the basis of the distance of the home from exposure and Patricia Stewart who helped during the different sources. They found excess risks for areas with high stages of this project. We thank also Tatiana Sánchez for application volumes of metam sodium and dicofol. Ex- her help with the drawings, Fabio Chaverri and Fernando posures to these two pesticides were not assessed. Ramírez for their review of the external data on pesticide This report addresses sources of occupational expo- application used in the model, and Margarita Mena for sure. The results concerning environmental and house- her administrative support. hold exposures to pesticides will be reported separately, as those for other reported risk factors, such as radiation, organic solvents, medications, early infectious diseases References in childhood, immunologic aspects, and genetic suscep- tibility. In conclusion, we found associations between 1. Little J. Epidemiology of childhood cancer. Lyon: International occupational parental pesticide exposures and childhood Agency for Research on Cancer (IARC); 1999. IARC scientic fi leukemia in a developing country with heavy use of pes- publications 149. ticides and a high rate of childhood leukemia. 2. Parkin DM, Kramárová E, Draper GJ, Masuyer E, Michaelis We recommend the prevention of pesticide hazards, J, Neglia J, et al. International incidence of childhood cancer, especially in the population stratum reproductively ac- vol II. Lyon: International Agency for Research on Cancer tive and among children. The C139 convention of the (IARC); 1998. IARC scientific publications 144. 3. Committee on Pesticides in the Diets of Infants and Children, International Labour Organization (ILO) concerning the National Research Council. Pesticides in the diets of infants prevention and control of occupational hazards caused and children. Washington (DC): National Academy Press; by carcinogenic substances and agents was established in 1974, taking into account the relevant work of the 4. Van Larebeke N, Birnbaum L, Boogaerts M, Bradke M, Da- World Health Organization (WHO) and the Interna- vis DL, Demarini DM, et al. Unrecognized or potential risk tional Agency for Research on Cancer (IARC) regarding factors for childhood cancer. Int J Occup Environ Health. 2005;11:199–201. cancer research. The convention has been ratified by 5. Shu XO, Stewart P, Wan-Quing W, Dehui H, Potter JD, Buck- several countries, but not by Costa Rica. It establishes ley JD, et at. Parental occupational exposure to hydrocarbons the following issues (i) periodic listing of carcinogenic and risk of acute lymphocytic leukemia in offspring. Cancer substances that would be prohibited or made subject to Epidemiol Biomarkers Prev. 1999;8:783–91. authorization and control, (ii) replacement of carcino- 6. Savitz DA, Chen JH. Parental occupation and childhood genic substances, (iii) exposure minimization, (iv) ap- cancer: review of epidemiological studies. Environ Health Perspect. 1990;88:325–37 propriate system of records, (v) provision of information 7. Zahm SH, Ward MH. Pesticides and childhood cancer. Environ on dangers and protection to workers, (vi) appropriate Health Perspect. 1998;106 Suppl 3: 909–25. medical examinations or biological or other tests for 8. Meinert R, Schuz J, Keltsch U, Kaatsch P, Michaelis J. Leuke- investigations, and (vii) codes of industrial or agricul- mia and non-Hodgkin’s lymphoma in childhood and exposure tural conduct. All of these measures are relevant in the to pesticides: results of a register-based case control study in prevention of childhood cancer through parental occupa- Germany. Am J Epidemiol. 2000;151:639–46. 9. Ries LAG, Smith MA, Gurney JG, Linet M, Tamra T, Young tional exposures, such as carcinogenic pesticides. JL, et al, editors. Cancer incidence and survival among children and adolescents: United States SEER Program 1975–1995. Bethesda (MD): National Cancer Institute, SEER Program; 1999. NIH publication 99–4649. Acknowledgments 10. Curl CL, Fenske R, Kissel JC, Shirai J, Moate T, Griffith W, et al. Evaluation of take-home organophosphorus pesticide ex- posure among agricultural workers and their children. Environ This work was partially supported by the Research Health Perspect. 2002;110:787–92. Department of the Swedish International Cooperation 11. Bufe fl r PA, Kwan ML, Reynolds P, Urayama K. Environmental Agency SAREC/Sida, a grant from the International and genetic risk factors for childhood leukemia: appraising the Scholar in Occupational and Environmental Health, Fog- evidence. Cancer Invest. 2005;21:60–75. arty grant 5 D43 TW00642–07 from the University of 12. John EM, Savitz DA, Sandler DP. Prenatal exposure to Washington (Seattle, WA, USA), and the Intramural Pro- parents’ smoking and childhood cancer. Am J Epidemiol. 1991;133(2):123–32 gram of the United States National Cancer Institute. 13. Doll R, Wakaford R. Risk of childhood cancer from fetal ir- Special thanks go to Rocío Loría-Bolaños, Mari- radiation. Br J Radiol. 1997;70:130–9. anela Rojas, Rebeca Alvarado, Heidi Morales, Carolina 14. Shu XO, Gao YT, Brinton LA, Linet MS, Tu JT, Zheng W, Castillo Abdalla, Jenny Umaña, Gina Solano, Emilia Fraumeni JF Jr. A population-based case-control study of González, Gabriela Rodríguez, and Emma Chacón who childhood leukemia in Shanghai. Cancer. 1988;62:635–44. were responsible for the data collection and to Matthew 15. Infante-Rivard C, Labuda K, Krajinovic M, Sinnett D. Risk of 302 Scand J Work Environ Health 2007, vol 33, no 4 Monge et al childhood leukemia associated with exposure to pesticide and 35. Buckley JD, Robison LL, Swotinsky R, Garabrant DH, with gene polymorphism. Epidemiology. 1999;10:481–7. LeBeau M, Manchester P, et al. Occupational exposures of 16. Petridou E, Dessypris N. Maternal pesticide exposure and parents of children with acute nonlymphocytic leukemia: a childhood leukemia. Epidemiol. 2000;11:230. report from the Children’s Cancer Study Group. Cancer Res. 17. Lowengart RA, Peters JM, Cicioni C, Buckey J, Bernstein L, 1989;49:4030–7. Preston-Martin S, et al. Childhood leukemia and parent’s oc- 36. Meinert R, Kaatasch P, Kaletsch U, Krummenauer F, Miesner cupation an home exposures. J Natl Cancer Inst. 1987;79:39– A, Michaelis J. Childhood leukemia and exposure to pesti- 46. cides: results of a case-control study in Northern Germany. Eur 18. Leiss JK, Savitz DA. Home Home pesticide pesticide use use and and childhood childhood can can- - J Cancer. 1996;32A:1943–8. cer: a case-control study. Am J Public Health. 1995;85(2):249– 37. Abadi-Korek, Stark B, Zaizov R, Shaham J. Parental occupa- 52. tional exposure and the risk of acute lymphoblastic leukemia in 19. Menegaux F, Baruchel A, Bertrand Y, Lescoeur B, Leverger G, offspring in Israel. J Occup Environ Med. 2006;48:165–74. Nelken B, et al. Household exposure to pesticides and risk of 38. Infante-Rivard C, Sinnet D. Preconceptional paternal exposure childhood leukaemia. Occup Environ Med. 2006;63:131–4. to pesticides and increased risk of childhood leukemia. Lancet. 20. Reeves JD. Household insecticide-associated blood dyscrasias 1999;354:1819. in children. Am J Hematol Oncol. 1982;4:438–9 39. Greaves M. Childhood leukaemia. Br Med J. 2002;324:283– 21. Infante PF, Epstein SS, Newton WA Jr. Blood dyscrasias and 87. childhood tumors and exposure to chlordane and heptachlor. 40. Alexander FE, Patheal SL, Biondi A, Brandalise S, Cabrera Scand J Work Environ Health. 1978;4:137–50. ME, Chan LC, et al. Transplacental chemical exposure and 22. Reeves JD. Household insecticide associated aplastic anaemia risk of infant leukemia with MLL gene fusion. Cancer Res. and acute leukaemia in children. Lancet. 1981;8;2:300–1. 2001;61:2542–6. 23. Reynolds P, Von Behren J, Gunier R, Goldberg D, Harnly M, 41. National Research Council. Drinking water and health, vol. 5. Hertz A. Agricultural pesticide use and childhood cancer in Washington (DC): Board on Toxicology and Environmental California. Epidemiology. 2005;16:93–100. Health Hazards, Commission on Life Sciences, Safe Drinking 24. Monge P, Wesseling C, Rodríguez AC, Cantor K, Weider- Water Committee, National Academy Press; 1983. pass E, Reutfors, et al. Childhood leukemia in Costa Rica, 42. US Environmental Protection Agency. Memorandum from 1981–1996. Paediatr Perinatal Epidemiol. 2002;16:210–8. the Office of Pesticides and Toxic Substances to Office of 25. Wesseling C, Aragón A, Castillo L, Corriols M, Chaverri F, de Pesticide Programs Division Director. Washington (DC): US la Cruz E, et al. Hazardous pesticides in Central America. Int Environmental Protection Agency; 1991. J Occup Environ Health. 2001;7:287–94 43. US Public Health Service. Hazardous substance data bank. 26. Chaverri F, Blanco J. Importaciones, formulación y uso de Washington (DC): US Public Health Service; 1995. plaguicidas en Costa Rica [Imports, formulation and use of 44. Edwards IR, Ferry DG, Temple WA. Fungicides and related pesticidas in Costa Rica]. Heredia (Costa Rica): Universidad compounds. In: Hayes WJ, Laws ER. Handbook of pesticide Nacional, Instituto Regional de Estudios en Sustancias Tóxi- toxicology. New York (NY): Academic Press; 1991. cas: Editorial, Universidad Nacional; 2002. 45. Jablonicka A, Polakova H, Karelova J, Vargova M. Analysis 27. Daniels JL, Olshan AF, Savitz DA. Pesticides and childhood of chromosome aberrations and sister-chromatid exchanges cancers. Environ Health Perspect. 1997;105:1068–77. in peripheral blood lymphocytes of workers with occupational 28. Colt JS, Blair A. Parental occupational exposures and risk of exposure to the mancozeb-containing fungicides Novozir childhood cancer. Environ Health Perspect. 1998;106 suppl Mn80. Mutat Res. 1989;224:143–6. 3:909–25. 46. US Environmental Protection Agency (US EPA). List of 29. Feychting M, Plato N, Nise G, Ahlbom A. Paternal occupation- chemicals evaluated for carcinogenic potential. Washington al exposures and childhood cancer. Environ Health Perspect. (DC): US EPA, Office of Pesticide Programs; 1996. 2001;109:193–6. 47. US Environmental Protection Agency. Pesticide fact sheet 30. Miettinen OS. Theoretical epidemiology: principles of occur- number 125: mancozeb. Washington (DC): Ofc fi e of Pesticides rence research in medicine. New York (NY): Wiley; 1985. and Toxic Substances; 1987. 31. Monge P, Wesseling C, Engel L, Keifer M, Zuurbier M, Rojas 48. Rothman KJ. No adjustments are needed for multiple compari- M, et al. An icon-based interview for the assessment of occupa- sons. Epidemiology. 1990;1:43–6. tional pesticide exposure in a case-control study of childhood 49. Flower KB, Hoppin JA, Lynch CF, Blair A, Knott C, Shore leukemia. Int J Occup Environ Health. 2004;10:72–8. DL, Sandler DP. Cancer risk and parental pesticide application 32. Monge P, Partanen T, Wesseling C, Bravo V, Ruepert C, in children of agricultural health study participants. Environ Burstyn I. Assessment of pesticide exposure in the agricultural Health Perspect. 2004;112:631–5. population of Costa Rica. Ann Occup Hyg. 2005;49(5):375– 50. Kristensen P, Andersen A, Irgens LM, Bye AS, Sundeheim L. 84. Cancer in offspring of parents engaged in agricultural activities 33. Valcke M, Chaverri F, Monge P, Bravo V, Partanen T, Wessel- in Norway: incidence and risk factors in the farm environment. ing C. Pesticide prioritization for a case-control study on child- Int J Cancer. 1996;65:39–50. hood leukaemia in Costa Rica: a simple stepwise approach. 51. Reynolds P, Von Behren J, Gunier R, Goldberg D, Hertz A. Environ Res. 2005;97:335–47. Childhood cancer and agricultural pesticide use: an ecologic 34. Wesseling C, Bravo V. Pesticide use on the main crops of Costa study in California. Environ Health Perspect. 2002;9:319–24. Rica 1970–2000: data for retrospective exposure assessment 52. VanSteensel-Mol HA, Valkenburg HA, Van Zanen GE. Child- in epidemiologic studies: report to the US National Cancer hood leukemia and parental occupation: a register-based case- Institute. Heredia (Costa Rica): Central American Institute for control study. Am J Epidemiol. 1985;121:216–24. Studies on Toxic Substances (IRET), Universidad Nacional; 2002. Received for publication: 7 September 2006 Scand J Work Environ Health 2007, vol 33, no 4 303

Journal

Scandinavian Journal of Work, Environment & HealthUnpaywall

Published: Aug 1, 2007

There are no references for this article.