Epidemiological study on long-term effects of pesticides on human health*

Bull. Org. mond. Sante 1971, 45, 181-199 Bull. Wld Hlth Org. Epidemiological Study of Long-Term Effects of Pesticides on Human Health* B. PACCAGNELLA,' F. GHEZZO,' L. PRATI,1 U. FEDRAZZONI1 & G. BELLONI2 Two rural areas in Italy with different rates of pesticide usage were subjected to a cross-sectional investigation, based on the identification of organochlorine compounds in the environment and in human fatty tissues and on the determination of their concentration and a prospective investigation, based on the collection ofdata ofpast andpresentpathology and symptomatology from two stratified random groups of inhabitants of both areas. The cross-sectional investigation showed only small differences in environmental pollu- tion and no differences in the storage of organochlorine compounds in fatty tissues in the two areas. The prospective investigation showed a higher prevalence of pathology and symptomatology in the population of the area where pesticides were used in larger quantities. Several haematochemical tests were used in an attempt to differentiate between populations at different levels of risk of toxic damage, but without success. Information on the effects of pesticides on human health may be obtained directly from epidemiological or clinical studies and indirectly from experimental studies with laboratory animals. The results of many studies on mammalian systems were presented at a conference held in New York in 1967 (Kraybill, 1969). The clinical studies have been mainly con- cerned with acute or subacute toxicology; the epi- demiological studies, on the other hand, have dealt mainly with the symptomatology and pathology of people subjected to different levels of exposure to different kinds of pesticide, especially organo- phosphorus compounds. Most epidemiological studies of organochlorine compounds have been concerned with concentrations in the environment, in food, and in human adipose and other tissues. There is a lack of epidemiological information about the long-term effects of pesticides on exposed populations, and such studies are difficult because of the difficulty of finding a suitable unexposed or less- exposed population to use as a control (Durham, 1965). This paper gives the first results of an epidemio- logical study on the long-term effects of pesticides on human health. * This work was carried out with financial assistance from the World Health Organization. I Istituto di Igiene, Universiti, Ferrara, Italy. 2 Ospedale Civile, Lendinara, Rovigo, Italy. MATERUILS AND METHODS The investigation was based on a cross-sectional study and on a prospective study, both carried out in the Provinces of Ferrara and Rovigo in Italy. The populations chosen for these studies live in the following rural areas: (1) the village of St Martino (District of Ferrara, Province of Ferrara-referred to as area A) with 6 123 residents (December 1964), where fruit cultivation is highly developed and pesticides have been widely used for about 15 years; and (2) the village of St Bellino, near Landinara (Province of Rovigo-referred to as area B), with 1 538 residents (December 1964), as a control population, where the agriculture is still traditional and where pesticides have been used in much smaller quantities for only the last 10 years. The mean amount of pesticides used annually for agricultural purposes in 1965 in area A was 100 kg and in area B was 30 kg. The modal values were much higher, 255.7 kg in area A and 53.1 kg in area B, because of an extremely skewed distribution. In area A the average amount of pesticides used annually from 1964 to 1967 was about 13-14 kg/ hectare; of this amount, 60% were carbamates, 25% were organophosphorus compounds, and 15% were organochlorine compounds. In area B, these compounds were used in about the same proportions as in area A, although the total amounts used 2716 - 181- B. PACCAGNELLA AND OTHERS 0) -i w- ID 0 C' (N I ' (N (N C') (N~ ' 0 ID0 N UOD U')D ID CD) N UN D 0D 0 C' C'4 ) (N Y) U') N- CD CD N C') C') (N )'-0 (0 ) C Y) C') (N- '-'-0 C1 ID* 0 co 0 co co (N (N C' 0 ID'- CO ID (N~1- 0 IDW co 0) C 0 0 x0) PI D '- ID 4 (N (Nm C) (N C' C') 0 (N) (N (N (Nt (N C') C') c ID4 T- ID4 N- ID 1- o') C') '( C')o( C') UO 01 r-N C') (N I 0 ID C') 00 (N-'- ID C) ID 0 6 6 0 ID ID) 0'-'- 0 (N 0 0) ID ID) 6'i4 o 11 ID 0 0) 0) (N (N(N0 (N c0 0 A C 0 C 0 C 0 C,) NN NO) '-'-0 (N (Nj C.) 0 0) 04 U-') IDC CV)~~~~~~~' 0V) 04 (N ID (N C') U) in 0)C')N (N VN C') '- 0 0~ (N ID, %' 0)ID ID ID N CD I- (N CD O C') (N (N '- 0 C'C) 0ID U-D 0V) N D ') ID I) -0 c') C')(N (N '- ID r- ID ID6C')'- 0) ,) 0 r, C') 0 IDt I C') U) C' C) co 0 '- r; A ID C 0 x 0 U U,) C' '- C' 0 (N P0 (N (N C' coI 0 0 ' 6C' 3 ( o-i6oi6o (' I U') ID I (N -(N ' N C, A 182 0 6 z E 0 6 z 0) 6 z 0 (A C~ 0 z 0 I) CA 0(0 r-o 0 CL 4-o x) a) *0 C I-0 C 0 x 0) 0 6 z E 0 LL - 6 z 6 z 0 6 z 0 6 z 0.n <00 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH annually from 1964 to 1967 were only about 4-5 kg/ hectare. No information is available on the use of pesticides in the two areas during the years preceding our study. The cross-sectional study was based on gas chromatographic identification and determinations of organochlorine compounds in the environment of both rural areas and on determinations of these compounds in the fatty tissues of subjects resident in the areas. In spite of the widespread use of organophospho- rus compounds and carbamates, the levels of these pesticides were not determined because of their short persistence in the environment. Throughout the year, samples of vegetables, fruits, animal fats, hay, soil, well-water and drinking- water, and raw and pasteurised milk were taken from the environments of some of the selected families or from the markets of the two areas; samples of surface water were drawn from the Po river and from the smaller rivers and canals of the Province of Ferrara; wild birds were captured in area A. For the prospective study, two random stratified samples (about 10%) of the populations living in the two rural areas were selected, in January 1965, from the General Register: for the purposes of stratifica- tion the following categories were recognized: (1) families of farmers and of non-farmers and (2) families that had resided in the village for 1-5, 6-10, 11-15, and more than 16 years. In the village of St Martino 469 subjects (245 males and 224 females) were randomly selected; in the village of St Bellino (control area), 152 subjects (74 males and 78 females) were selected from the 1 538 residents. On the first occasion, all the selected people of the two rural areas were visited. During the second stage of the prospective investigation in area A, 119 of the original 469 subjects were not seen (42 refused, 6 had died, 47 had changed their place of residence, and 24 were temporarily absent); in area B, on the second occasion, 12 of the original 152 subjects were not seen (1 refused and 11 were temporarily absent). The two populations were classified on each occasion according to sex and age and according to their occupational exposure to pesticides (handlers and non-handlers). These data show a prevalence of male and adult subjects among the handlers and a prevalence of female and younger subjects among the non-handlers. There were no systematic differ- ences between the sex and age distributions of the handlers and of the non-handlers in the two areas or between the total populations of the two areas (Table 1). The randomly selected people from the two villages were examined at 2-year intervals after 1965, clinical and biochemical tests being made on each occasion. The clinical assessment, which was made on a standard questionnaire, was based on clinical find- ings and on past and present pathology and recurrent symptoms (e.g., frequent respiratory difficulty, frequent anorexia) related to disorders of the auto- nomic nervous system (Grob, 1963; WHO Expert Committee on Insecticides, 1967). The examiners did not know the level of exposure of each person examined. Blood samples were drawn from subjects 20-60 years of age and the following constituents were determined: plasma cholinesterase and red blood cell acetylcholinesterase by the method of Michel (1949); plasma arylesterase by the method of Zeller (1956); endogenous serum lipoproteinlipase by the method of Hoerlein & Pilz (1962); haemoglobin by the cyanmethaemoglobin method; total serum proteins by the method of Waddell & Hill (1956); and total serum cholesterol by the method of Leffler (1959). Serum alkaline and acid phosphatase levels were determined by the method of Bessey et al. (1946), red and white blood cell counts were performed, and an ECG tracing was made to evaluate alterations in cardiac rhythm. For comparisons between and within the two populations, the following tests were used: x2 test for frequencies; Student's t test and analysis of variance for haematochemical data. For comparisons of the levels of organochlorine compounds in the fatty and non-fatty tissues from biopsies and necropsies, evaluations were made after logarithmic transformation of the original data (ppm x 1 000). RESULTS Cross-sectional study The results of the gas chromatography determina- tions are summarized in Tables 2-11. The following observations may be made. (1) Heptachlor epoxide was always present in milk and occurred at similar levels in the two areas (A, 0.022 ppm; B, 0.016 ppm); it was also present in 183 184 B. PACCAGNELLA AND OTHERS Table 2. Concentration of heptachlor epoxide Area A Area B No. No. No. of Mean No. of Mean sf- otsiv concen. S.E. Range of posie concen- S.E. Range pies sam- tration pies same tration ples pies soil (ppm) 4 1 - - 0-0.02 7 0 - - - hay (ppm) - - - - - 6 0 - - - well-water (ppm x 103) 5 0 - - - 7 0 - - - drinking-water (ppm x 103) 10 0 - - - 3 0 - - - surface water (ppm x 103) 33 0 - - - 3 0 - - - fruits (ppm) 18 1 - - 0-0.02 9 0 - - - vegetables (ppm) 25 2 - - 0-0.04 25 1 - - 0-0.03 milk (ppm) 38 37 0.022 0.002 0-0.06 5 4 0.016 0.005 0-0.03 animal fats (ppm) 8 4 0.179 0.097 0-0.77 9 8 0.023 0.095 0-0.80 wild birds (ppm) 17 11 0.167 0.053 0-0.84 - - - - - animal fats, at a higher level in area B (0.323 ppm) ppm) than in those from area A (0.343 ppm), and than in area A (0.179 ppm) (Table 2). at similar levels in milk from the two areas (A, (2) Dieldrin was present at similar mean levels in 0.084 ppm; B, 0.056 ppm) (Table 3). soil from the two areas (A, 0.028 ppm; B, 0.023 ppm), (3) TDE was present in only some samples of at higher levels in animal fats from area B (1.372 milk and in wild birds from area A (Table 4). Table 3. Concentration of dieldrin Area A Area B No. No. No. of Mean No. of Mean sofas-ti concen- S.E. Range sof poive concen S.E. Range pies sam- tration pies Sam- trio ples ples soil (ppm) 4 2 0.028 0.016 0-0.06 7 3 0.023 0.012 0-0.08 hay (ppm) - - - - - 6 0 - - - well-water (ppm x 103) 5 0 - - - 7 1 - - 0-4.25 drinking-water (ppm x 103) 10 0 - - - 3 0 - - - surface water (ppm x 103) 33 5 0.059 0.038 0-1.24 3 0 - - fruits (ppm) 18 2 - - 0-0.07 9 0 - - - vegetables (ppm) 25 1 - - 0-0.38 25 3 - - 0-1.31 milk (ppm) 38 38 0.084 0.006 0.01-0.17 5 4 0.050 0.015 0-0.07 animal fats (ppm) 8 6 0.343 0.106 0-0.82 9 8 1.372 0.457 0-4.15 wild birds (ppm) 17 8 0.224 0.115 0-2.00 - - - - - LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH 185 Table 4. Concentration of TDE Area A Area B No. No. No. of Mean INo. of Mean of posi- concen- S.E. Range sam titeconcen- S.E. Rangesam- tive trationsa-te trationpies sam- ples sam- pies ples soil (ppm) 4 0 - - - 7 0 hay (ppm) - - - - - 6 0 well-water (ppm x 103) 5 0 - - - 7 0 - - - drinking-water (ppm x 103) 10 0 - - - 3 0 - - - surface water (ppm x 103) 33 0 - - - 3 0 - - - fruits (ppm) 18 0 - - - 9 0 - - - vegetables (ppm) 25 0 - - - 25 0 - - - milk (ppm) 38 14 0.019 0.005 0-0.17 5 0 - - - animals fats (ppm) 8 0 - - - 9 0 - - - wild birds (ppm) 17 5 0.143 0.081 0-1.35 - - - - (4) In area A, DDE was always present in soil and in milk from the two areas (A, 0.058 ppm; B, in animal fats, was frequently present in milk, in 0.032 ppm) were not very different, while the wild birds, and in fruits, and was sometimes present levels in animal fats from area A were higher in water samples: in area B it was often present in (1.621 ppm) than in those from area B (0.722 ppm) milk and in animal fats. The mean levels of DDE (Table 5). Table 5. Concentration of DDE Area A Area B No. No. No. of Mean No. of Maof posi- concen- S.E. Range of poie concen- S.E. Range sam- tive tration sam- tive trtone- SE agples sam- ~~~~~~pies sam- ples ples soil (ppm) 4 4 0.535 0.239 0.14-1.23 7 1 - - 0-0.04 hay (ppm) - - - - - 6 1 - - 0-0.17 well water (ppm x 103) 5 1 - - 0-0.48 7 0 - - - drinking-water (ppm x 103) 10 2 0.073 0.049 0-0.37 3 0 - - - surface water (ppm x 103) 33 1 - - 0-0.20 3 0 - - - fruits (ppm) 18 9 0.027 0.008 0-0.13 9 0 - - - vegetables (ppm) 25 4 - - 0-0.04 25 0 - - - milk (ppm) 38 27 0.058 0.008 0-0.17 5 3 0.032 0.013 0-0.06 animal fats (ppm) 8 8 1.621 0.872 0.21-7.60 9 4 0.722 0.417 0-3.70 wild birds (ppm) 17 13 0.485 0.118 0-1.44 - - - - - 186 B. PACCAGNELLA AND OTHERS Table 6. Concentration of o,p'-DDT Area A Area B No. No. No. of Mean No. of Mean of posi- cocn- S.E. Rne of posi concen- S.E. Range amtie tration sam- tive trationpies sam- pies sam- ples pies soil (ppm) 4 4 0.555 0.092 0.44-0.83 7 1 - - 0-0.39 hay (ppm) - - - - - 6 0 - - - well-water (ppm x 103) 5 0 - - - 7 1 - - 0-0.15 drinking-water (ppm x 103) 10 1 - - 0-0.08 3 0 - - - surface water (ppm x 103) 33 0 - - - 3 0 - - - fruits (ppm) 18 8 0.094 0.055 0-0.95 9 1 - - 0-0.12 vegetables (ppm) 25 0 - - - 25 0 - - - milk (ppm) 38 0 - - 5 0 - - animal fats (ppm) 8 1 - - 0-0.59 9 0 - - wild birds (ppm) 17 1 - - 0-045 - - - - - (5) o,p'-DDT was found in some samples of soil fats from area A (2.690 ppm and 1.499 ppm, respec- and of fruit from area A, but otherwise it was very tively) and in animal fats (1.322 ppm), soil (1.314 seldom found (Table 6). ppm), and hay (0.277 ppm) from area B. Except for milk (A, 0.036 ppm; B, 0.036 ppm) and animal fats (6) p,p'-DDT was present in all the samples from (A, 1.499 ppm; B, 1.322 ppm), p,p'-DDT levels both areas (except the drinking-water of area B); were higher in all the samples from area A than in the highest mean levels were in soil and in animal those from area B (Table 7). Table 7. Concentration of p,p'--DDT Area A Area B No. No. Nof posi Mean No. of Mean sam- psie concen S.E. Range om sim- concen- S.E. Range pIes pIes pies sm ples soil (ppm) 4 4 2.690 0.291 2.30-3.55 7 2 0.314 0.282 0-2.00 hay (ppm) - - - - - 6 3 0.277 0.138 0-0.78 well-water (ppm x 103) 5 4 0.808 0.661 0-3.45 7 2 0.207 0.163 0-1.15 drinking-water (ppm x 103) 10 5 0.365 0.171 0-1.58 3 0 - - - surface water (ppm x 103) 33 4 0.035 0.020 0-0.54 3 1 0.002 - - fruits (ppm) 18 15 0.731 0.389 0-6.38 9 4 0.136 0.077 0-0.63 vegetables (ppm) 25 5 0.074 0.064 0-1.62 25 0 - - - milk (ppm) 38 25 0.036 0.004 0-0.11 5 3 0.036 0.015 0-0.08 animal fats (ppm) 8 8 1.499 0.529 0.30-4.88 9 5 1.322 0.753 0-7.00 wild birds (ppm) 17 9 0.181 0.073 0-1.00 - - - - - LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH 187 Table 8. Concentration of total DDT equivalent (DDT and DDT derivatives) Area A Area B No. No. No. of Mean No. of Mean of posi. concen- S.E. Range of posi- concen- S.E. Rangesam- tive agaionsam- tive crtone- SE agples sam- taonpies sam- rgo pies pies soil (ppm) 4 4 3.843 0.645 2.96-5.75 7 2 0.375 0.343 0-2.43 hay (ppm) - - - - - 6 3 0.308 0.1t63 0-0.97 well-water (ppm x 103) 5 4 0.914 0.767 0-3.98 7 2 0.228 0.183 0-1.30 drinking-water (ppm x 103) 10 6 0.454 0.198 0-1.98 3 0 - - - surface water (ppm x 103) 33 4 0.041 0.022 0-0.54 3 1 0.002 - - fruits (ppm) 18 15 0.851 0.452 0-7.47 9 4 0.149 0.008 0-0.75 vegetables (ppm) 25 8 0.079 0.064 0-1.62 25 0 - - - milk (ppm) 38 30 0.122 0.019 0-0.39 5 3 0.074 0.031 0-0.15 animal fats (ppm) 8 8 3.376 1.463 0.89-13.32 9 5 2.124 1.209 0-11.11 wild birds (ppm) 17 16 0.905 0.247 0-3.89 - - - - - (7) The findings for total DDT were similar to those for p,p'-DDT (Table 8). Data on the levels of several organochlorine com- pounds in samples of fat obtained by biopsy from (8) The total level of organochlorine compounds residents of areas A and B are summarized in was similar in the two areas, if the values for soil Table 10. Statistical analysis showed that the only and fruit were excluded (Table 9). significant (P <0.05) differences were the higher Table 9. Concentration of total organochlorine compounds Area A Area B No. No. No. of MenNo. of Ma s- tisve concen- S.E. Range sam- ptive concen- S.E. Range pies sam- tainpies sam- trio ples ples soil (ppm) 4 4 3.875 0.648 3.04-5.80 7 3 0.398 0.349 0-2.48 hay (ppm) _ _ _ - - 6 3 0.308 0.163 0-0.97 well-water (ppm x 103) 5 4 0.914 0.767 0-3.98 7 2 0.835 0.787 0-5.55 drinking-water (ppm x 103) 10 6 0.454 0.198 0-1.98 3 0 - - - surface water (ppm x 103) 33 8 0.121 0.050 0-1.24 3 1 0.002 - - fruit (ppm) 18 16 0.861 0.451 0-3.93 9 4 0.149 0.088 0-0.75 vegetables (ppm) 25 10 0.096 0.079 0-2.00 25 4 0.056 0.052 0-1.31 milk (ppm) 38 38 0.229 0.024 0.02-0.53 5 4 0.144 0.046 0-0.25 animal fats (ppm) 8 8 3.897 1.500 1.13-13.92 9 9 3.820 1.222 1.68-11.48 wild birds (ppm) 17 16 1.296 0.310 0-3.69 - - - - B. PACCAGNELLA AND OTHERS Table 10. Concentration of organochlorine compounds (ppm) in samples of fat obtained by biopsy Area A (33 samples) Area B (11 samples) Mean Geometric MeanGemticoncen- S.E. Range mean concen- S.E. Range meantration men tration ma heptachlor epoxide 0.239 0.053 0-1.02 0.045 1.039 0.155 0.35-2.15 0.914 dieldrin 0.312 0.130 0-3.15 0.012 1.945 0.529 0-5.70 0.860 DDE 7.044 1.332 0.73-41.17 5.052 6.528 0.741 4.42-9.57 6.029 p,p'-DDT 2.603 0.567 0-16.50 0.838 1.856 0.420 0-4.23 0.942 o,p'-DDT - - 0-0.66 - - - - - total DDT equivalent 10.443 2.008 0.81-62.20 7.483 9.103 1.128 2.81-14.85 8.293 total organochlorine 10.994 2.012 0.81-62.20 7.940 12.087 1.693 4.01-21.44 10.895 levels of heptachlor epoxide and of dieldrin in area B. The total levels of organochlorine com- pounds stored in fat were not statistically different in the two populations (10.994 ppm for area A and 12.087 ppm for area B). Table 11 summarizes the data on the concentration of organochlorine compounds in fatty and non-fatty tissues obtained post mortem from residents of area A. The levels of heptachlor epoxide (P <0.05) and of dieldrin (P <0.001) were higher in fat obtained at autopsy than in that obtained by biopsy: the mean values for heptachlor epoxide were 0.513 ppm and 0.053 ppm, respectively, while the comparable values for dieldrin were 0.908 ppm and 0.130 ppm, respec- tively. For the other pesticides and for the total storage of organochlorine compounds no significant differences were observed. Higher levels were always found in fatty tissues than in non-fatty tissues (for all kinds of pesticides). MoreDDE was found in the liver than in the kidneys, brain, and spleen, but this was not true of DDT. The amounts stored in the kidneys, brain, and spleen were similar. The levels of organochlorine compounds in non-fatty tissues were quite high, especially in compa- rison with those found in other similar investiga- tions (De Vlieger et al., 1968; Casarett et al., 1968). At present it is difficult to correlate this storage of pesticides in fatty and in not-fatty tissues with pathology (Durham, 1969), although it is reasonable to suppose that it may cause some toxic effects. Prospective study Between 1965 and 1968 the subjects selected from the two populations were examined twice and the investigations are being continued. From data collected from both groups (handlers and non-handlers) in the two areas, the past and present pathology and the recurrent symptomatology were calculated. The pathology and symptomato- logy that occurred more than 2 years before the first examination are considered as past pathology and symptomatology. Past pathology. The distribution of diseases that occurred more than 2 years before the study is shown in Table 12 for the whole population and separately for males aged 21-60 years of age. The difference between the proportions of people with diseases (74.0% in area A and 72.4% in area B) was not significant (P> 0.05); however, in both populations the handlers were more affected than the non-handlers, especially in area B (P <0.05). The total number of past diseases per person decreased in the following order: handlers B> handlers A> non-handlers A> non-handlers B. The frequencies of diseases in the two areas, grouped for organs or systems (Table 13), confirm the results mentioned above. Since there was a preponderance of adult males among the handlers and young people and of females among the non-handlers in both populations, the incidence of past morbidity was studied in males in the 21-60-year age group in order to give a better comparison between and within the areas. 188 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH 189 Table 11. Concentration of organochlorine compounds (ppm) in samples of fatty and not-fatty tissues obtained at autopsy Materil Fat Liver Kidney Brain Spleen(52 samples) (15 samples) (15 samples) (15 samples) (15 samples) heptachlor epoxide mean conc. 0.513 0.072 - - - S.E. 0.060 0.024 - - - range 0-1.58 0-0.34 - - 0-0.19 geometric mean 0.140 0.015 - - - dieldrin mean conc. 0.908 0.088 - - S.E. 0.115 0.028 - - - range 0-3.55 0-0.40 - - 0-0.27 geometric mean 0.221 0.016 - - - DDE mean conc. 9.939 0.554 0.023 0.022 0.099 S. E. 1.131 0.146 0.014 0.010 0.057 range 0.41-40.50 0.03-2.83 0-0.20 0-0.13 0-0.69 geometric mean 6.888 0.293 0.008 0.008 0.019 p,p'-DDT mean conc. 4.037 0.627 0.429 0.224 0.041 S.E. 0.615 0.344 0.191 0.140 0.029 range 0-22.16 0-8.00 0-2.60 0-2.00 0-0.34 geometric mean 2.136 0.039 0.053 0.015 0.007 o,p'-DDT mean conc. - 0.065 0.082 0.036 0.023 S.E. - 0.049 0.037 0.029 0.016 range - 0-1.10 0-0.54 0-0.42 0-0.19 geometric mean - 0.007 0.018 0.007 0.007 total DDT equivalent mean conc. 15.070 1.313 0.535 0.285 0.188 S.E. 1.822 0.449 0.221 0.161 0.083 range 0.63-65.62 0.03-11.32 0-2.92 0-2.09 0-0.95 geometric mean 10.600 0.476 0.098 0.022 0.033 total mean conc. 16.453 1.472 0.535 0.285 0.223 S.E. 1.924 0.506 0.221 0.161 0.112 range 0.77-68.16 0.03-11.32 0-2.92 0-2.09 0-1.41 geometric mean 11.120 0.566 0.098 0.022 0.034 B. PACCAGNELLA AND OTHERS Table 12. Distribution of diseases that occurred more than 2 years before the study Area A Area B Group No. in No. of Percentage No in No. of Percentage groupi diseases/ of persons No. diseases/ of personsperson affected group person affected Whole sample population handlers 112 2.17 80.4 24 2.62 95.8 non-handlers 357 1.58 72.0 128 1.03 68.0 total 469 1.73 74.0 152 1.29 72.4 Males, 21-60 years of age handlers 70 4.30 91.5 21 3.86 93.2 non-handlers 77 3.67 83.1 26 3.77 100.0 total 147 3.97 87.1 47 3.81 97.8 Table 13. Percentage frequency of diseases that occurred more than 2 years before the study Area A Area B Type of disease Nond Non- Hnleshandlers Hnders handlers Whole sample population cardiovascular 9.8 9.2 8.3 0.8 respiratory 55.4 53.5 91.7 64.1 digestive 28.6 27.2 37.5 13.3 liver and bile ducts 24.1 11.2 4.2 0.8 kidney and urinary tract 2.7 1.1 0 0 locomotor 56.3 29.4 54.2 17.2 allergic 12.5 5.6 19.2 1.6 sensory 6.3 3.9 20.8 3.1 skin 2.7 3.1 8.3 2.3 nervous 18.8 14.9 8.3 0.8 Males, 21-60 years of age cardiovascular 18.6 10.4 9.5 0 respiratory 107.1 98.7 142.9 188.5 digestive 92.9 96.1 104.8 84.6 liver and bile ducts 41.4 39.0 4.8 0 kidney and urinary tract 12.9 6.5 4.8 0 locomotor 61.4 44.2 67.1 19.2 allergic 22.9 16.9 14.3 3.8 sensory 41.4 26.0 23.8 57.7 skin 11.4 10.4 9.5 0 nervous 20.0 19.5 14.3 23.1 Apart from some small differences within the areas, the four groups of men showed similar amounts of past disease. Thus the differences mentioned above must have been due to the different proportions of young people and females in the samples. Present pathology. The distribution of present morbidity among subjects examined twice is sum- marized in Table 14, both for the whole population and for males aged 21-60 years. In general, the number of persons with current disease was smaller at the time of the second examin- ation (except for the non-handlers in area A), so that the differences between handlers and non- handlers were much smaller at the second examina- tion than at the first. As regards the total number of diseases per person, the populations of areas A and B were not dis- tinguishable at the first examination; at the second examination the relative morbidity rates were as follows: handlers A = non-handlers A = handlers B> non-handlers B. This morbidity situation was observed at both the first and the second examina- tions and also in male subjects 21-60 years of age (Table 14). It may be concluded that population A seems more homogeneous than population B in regard to present pathology. This tendency will be further examined in the next phase of the prospective study. The diseases recorded at the first and second examinations are listed in greater detail in Table 15, from which the following conclusions can be drawn. 190 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH Table 14. Distribution of present morbidity among the subjects examined twice Area A Area B No. of diseases/ Percentage of No. of diseases/ Percentage of Group No. person persons affected No. person persons affected in First Second First Second in First Second First Secondgroup exami- exami- exami- exami- group exami- exami- exami- exami- nation nation nation nation nation nation nation nation Whole sample population handlers 94 3.14 2.18 93.6 74.3 29 2.83 1.65 96.5 82.8 non-handlers 256 2.26 2.25 73.0 70.6 111 1.57 1.16 74.8 64.0 total 350 2.50 2.23 78.6 71.6 140 1.83 1.26 79.3 67.9 Males, 21-60 years of age handlers 55 2.35 1.74 I 78.2 78.2 22 3.04 1.50 95.4 72.7 non-handlers 53 2.53 2.17 77.3 77.3 I 18 1.22 1.22 61.1 38.8 total 108 2.44 1.95 77.8 77.8 40 2.22 1.37 80.0 57.4 (1) Diseases of the cardiovascular and sensory systems and of the skin occurred with equal fre- quency in the populations of both areas at the first and the second examinations. (2) Diseases of the respiratory system were more frequent at the first examination in area B than in area A, but at the second examination this type of disease occurred with the same frequency in both populations; at the second examination in area A there was a higher frequency among the non-handlers than among the handlers. (3) Diseases of the digestive system were more frequent in population A than in population B, but the difference was smaller at the second examination than at the first. (4) The differences in the frequency of liver and bile duct diseases between and within the two populations were the same at both examinations, and such diseases were more frequent in area A than in area B. (5) Diseases of the locomotor system decreased among the handlers of area A and increased among handlers of area B, while among the non-handlers of both populations the frequency did not change. (6) The frequency of allergic diseases was approx- imately the same in both populations; it decreased among the handlers of areas A and B but did not change among the non-handlers. (7) Diseases of the nervous system were more frequent in area A than in area B at both examina- tions but their prevalence decreased in all the groups, especially the population of area A. An epidemiological investigation of recurrent herpes, undertaken at the same time among the people of the two areas, showed a higher incidence of recurrences among handlers (43.4%) than among non-handlers (23.9%) and among the handlers there was an increase in recurrences with age over 20 years (Stigliano et al., 1967). Table 15 also shows the prevalence of diseases among males 21-60 years of age of the two popula- tions at the first and second examinations. From these figures it may be seen that there was a higher incidence of digestive, liver, and locomotor diseases in area A than in area B, and that there were no differences between people occupationally exposed to pesticides and those not exposed. Past symptomatology. No cases of acute intoxica- tion by anticholinesterase pesticides have been reported in recent years in area A or in area B. The distribution of past symptoms in the two groups from each of the two areas is shown in Tables 16 and 17. There was no significant difference (P> 0.05) in the occurrence of symptoms among handlers in the two areas. Symptoms occurred more frequently among 191 B. PACCAGNELLA AND OTHERS Table 15. Percentage frequency of diseases detected (present pathology) at the first and at the second examinations Area A Area B Type of disease Handlers Non-handlers Handlers Non-handlers First Second First Second First Second First Second examination examination examination examination examination examination examination examination Whole sample population cardiovascular 11.7 17.0 9.4 10.9 13.8 6.9 9.9 18.0 respiratory 46.8 11.7 43.0 26.6 79.3 24.1 55.9 17.1 digestive 30.8 28.7 30.1 32.0 31.0 34.5 9.9 17.1 liver and bile ducts 25.5 26.6 10.2 16.4 3.4 3.5 0.9 2.7 kidney and urinary tract 3.0 4.2 1.2 3.5 0 0 0 0 locomotor 51.1 43.6 29.7 30.9 37.9 48.3 18.9 18.0 allergic 14.9 10.6 4.3 4.7 10.3 3.4 2.7 2.7 sensory 16.0 21.3 10.2 18.7 20.7 13.8 14.4 18.9 skin 2.1 3.2 2.0 3.1 6.9 0 3.6 1.8 nervous 18.1 6.4 11.7 9.0 6.9 3.5 0 0 Males, 21-60 years of age cardiovascular 10.9 10.5 5.7 9.4 13.6 9.1 0 5.6 respiratory 41.8 9.1 50.9 24.5 86.4 27.3 55.6 5.6 digestive 29.1 25.5 37.7 35.9 40.9 27.3 5.6 11.1 liver and bile ducts 29.1 30.9 22.6 24.5 0 0 0 5.6 kidney and urinary tract 1.8 1.8 5.7 5.7 0 0 0 0 locomotor 49.1 34.5 32.1 32.1 50.0 40.9 16.7 16.7 allergic 9.1 5.5 5.7 1.9 13.6 4.6 5.6 5.6 sensory 10.9 18.2 9.4 17.0 18.2 13.6 11.1 16.7 skin 0 1.8 3.8 5.7 9.1 0 0 0 nervous 16.4 5.5 17.0 17.0 4.5 0 0 0 handlers than among non-handlers, both in area A (P<0.001) and in area B (P<0.001); non-handlers in area A exhibited more symptoms than the non- handlers in area B (P<0.05). Thus the population of area A had more symptoms than the population of area B (P <0.05). As regards the total number of symptoms per person, the population of area A had significantly more than that of area B (P <0.001): there was a particularly low frequency of symptoms among the handlers of area B, in relation to that of the other three groups: past symptoms were thus distributed among the groups in the following order: handlers A=handlers B>non-handlers A> non-handlers B. The same results are observed if the data for male subjects 21-60 years of age (Table 16) are analysed. Present symptomatology. The distribution of present symptoms in subjects of the two areas who were examined twice is summarized in Tables 18 and 19. As regards the number of people with symptoms, the situation at the first examination (handlers A= handlers B> non-handlers A = non-handlers B) had changed at the second examination, when symptoms had become less prevalent in area B and the four groups could then be arranged in decreasing order of prevalence of symptoms as follows: handlers A> handlers B=non-handlers A>non-handlers B. 192 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH Table 16. Distribution of past symptoms Area A Area B Group No. of Percent- No. of Percent- symp- age of symp- age of toms/ persons toms/ persons person affected person affected Whole sample population handlers 1.78 53.6 1.33 58.3 non-handlers 0.36 20.4 0.18 10.2 total 0.70 28.4 0.36 17.8 Males, 21-60 years of age handlers 2.44 64.3 1.42 57.1 non-handlers 0.27 15.5 0.03 3.8 total 1.30 38.7 0.65 27.6 A significant change occurred in the number of present symptoms in the four groups between the first and the second examinations. Statistical ana- lysis shows that the groups were arranged in decreas- ing order of prevalence at the time of the first examination as follows: handlers A=handlers B> non-handlers A>non-handlers B; by the second examination the order had become: handlers A> handlers B= non-handlers A > non-handlers B. The distribution of the data concerning present symptoms in male subjects aged 21-60 years is also shown in Table 18, and the same relationship between the groups is seen as was evident with the total population. Thus it can be concluded that symptoms were less prevalent in area B, where anticholinesterase agents were used in smaller quantities. Blood chemistry. Of the total of 621 subjects, chemical constituents of the blood were determined in 400 (64.4%) at the first examination; 277 (59.1 %) Table 17. Percentage frequency of selected symptoms that occurred more than 2 years before the study (past symptomatology) Area A Area B Symptoms Handlers Non-handlers Total Handlers Non-handlers Total respiratory difficulty 22.3 7.0 10.7 29.2 0.8 5.3 anorexia 22.3 7.3 10.9 12.5 0 2.0 vomiting and abdominal tenderness 13.4 1.4 4.3 0 0.8 0.7 excessive salivation and perspiration 8.0 0.6 2.4 4.2 0.8 1.3 decreased salivation 8.0 1.4 3.0 8.3 3.9 4.6 headache and hemicrania 22.3 6.4 10.2 25.0 0 4.0 muscular tremors 4.5 0.3 1.3 0 0 0 reduced deep perception 1.8 0.6 0.9 0 0.8 0.7 frequent and involuntary micturition 1.8 1.1 1.3 4.2 0 0.7 anxiety, restlessness, etc. 17.0 1.4 5.1 12.5 2.3 4.0 drowsiness and difficulty in concentration 5.4 0.8 1.9 16.7 0 2.6 bilateral ankle and wrist weakness 6.3 0.6 1.9 0 0.8 0.7 asthenia 12.5 0.8 3.6 8.3 2.3 3.3 vertigo 9.8 1.7 3.6 0 3.1 2.6 difficulty in focusing, etc. 6.3 0.3 1.7 0 0.8 0.7 increased lacrimation 11.6 2.8 4.9 12.5 0.8 2.6 tinnitus 4.5 1.4 2.1 0 0.8 0.7 193 B. PACCAGNELLA AND OTHERS Table 18. Distribution of present symptoms Area A Area B No. of symptoms/person Percentage of No. of symptoms/person Percentsge of persons Group First Second First Second First Second First Second examination examination examination examination examination examination examination examination Whole sample population 51.1 54.3 21.9 15.2 29.7 25.7 Males, 21-60 years of age 63.6 80.0 15.1 37.7 39.8 59.2 Table 19. Percentage frequency of selected symptoms at the first and at the second examinations (present symptomatology) Area A Area B Handlers Non-handlers Handlers Non-handlers Symptoms First Second First Second First Second First Second examina- examina- examina- examina- examina- examina- examina- examina- tion tion tion tion tion tion tion tion respiratory difficulty 36.2 18.1 6.3 6.3 24.1 3.5 0.9 0 anorexia 14.9 6.4 0.8 0.4 3.4 0 1.8 0 vomiting and abdominal tenderness 16.0 17.0 9.4 8.2 13.8 3.5 0 0 excessive salivation and perspiration 6.4 6.4 0.8 0.4 3.4 0 0.9 0 decreased salivation 9.6 8.5 1.6 0.8 6.9 6.9 0 0 headache and hemicrania 26.6 18.1 8.6 4.7 27.6 6.9 3.6 0 muscular tremors 4.3 2.1 0 0 0 0 0 0 reduced deep perception 2.1 0 0.4 0.4 3.4 0 0 0 frequent and involuntary micturition 4.3 1.1 0.4 0 3.4 0 0 0 anxiety, restlessness, etc. 18.1 8.5 0.8 0 13.8 0 1.8 0 drowsiness and difficulty in concentration 6.4 5.3 0.8 0.4 13.8 0 0.9 0 bilateral ankle and wrist weakness 7.4 3.2 0 0 3.4 0 0.9 0 asthenia 16.0 17.0 2.0 0 6.9 0 3.6 0 vertigo 10.6 8.5 1.2 0.8 6.9 0 3.6 0 difficulty in focusing, etc. 8.5 4.3 0.4 0.4 0 0 0.9 0 increased lacrimation 17.0 9.6 1.2 0.8 13.8 6.9 0.9 0 tinnitus 5.3 2.1 0.8 0 3.4 0 2.7 0 - 194 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH Table 20. Chemical constituents of the blood in the populations at the first examination Area A Area B Constituent Handlers Non-handlers Handlers Non-handlers Mean Mean Mean MeanNo. (±S.D.) No. (S.D.) No. (-S.D.) No. (±S.D.) total plasma cholinesterase 103 0.773 174 0.789 24 0.752 99 0.830 4 pH/h (0.221) (0.187) (0.225) (0.205) thermostable plasma 101 0.122 172 0.118 24 0.141 99 0.128 cholinesterase (4 pH/h) (0.046) (0.053) (0.077) (0.033) RBC acetylcholinesterase 103 0.723 174 0.680 24 0.712 99 0.661 (4 pH/h) (0.203) (0.178) (0.197) (0.105) total plasma arylesterase 103 265.91 174 276.54 24 234.79 99 245.95 (,&mol/ml/h) (79.54) (56.28) (50.09) (62.21) thermostable plasma 103 36.55 174 38.96 24 38.28 99 36.39 arylesterase (,umol/ml/h) (15.50) (11.31) (17.67) (14.41) haemoglobin (g/100 ml) 94 14.30 166 13.47 21 13.86 87 13.66(1.57) (1.80) (1.97) (2.14) total serum protein 103 7.38 174 7.68 24 7.34 99 7.26 (g/100 ml) (0.779) (0.721) (0.915) (0.743) total serum cholesterol 101 211.36 174 209.55 23 207.37 97 214.56(mg/i 00 ml) (68.60) (55.13) (72.62) (58.52) serum endogenous lipo- 102 41.93 171 45.94 23 28.25 97 41.63 protein lipase (26.49) (23.54) (18.27) (25.49) (t&mol/ml/24 h) of the 469 subjects in area A and 123 (80.9%) of the 152 subjects in area B were thus examined. The results are summarized in Table 20. An analysis of variance was carried out on these data using two criteria (professions and areas), with the following results: (1) In area A, the mean values for total plasma arylesterase (272.6 ,mol/ml/h), serum endogenous lipoprotein lipase (44.4 umol/ml/24h), and total serum protein (7.6 g/100 ml) were higher (P <0.001, P<0.05, and P<0.001, respectively) than the mean values in area B (243.8 ,umol/ml/h, 39.1 ,umol/ml/ 24 h, and 7.3 g/100 ml, respectively). These higher values in the population of area A may be related to different nutritional habits in the two areas, especially in regard to fat intake. Serum endogenous lipo- protein lipase and plasma arylesterase are concerned with fat metabolism: the former hydrolyses the ester linkages of long-chain fatty acids at the level of lipoproteins and the latter, positively correlated with the free fatty acids in the blood (Augustinsson & Henricson, 1965), is supposed to interfere with the transesterification of fatty acids (Pilz & Hoerlein, 1964). (2) The mean acetylcholinesterase (AChE) and haemoglobin levels were higher in the handlers of both areas than in the non-handlers. The mean values for AChE were 0.723zpH/h in handlers and 0.680,4pH/h in non-handlers in area A and 0.712 zpH/h in handlers and 0.661 4pH/h in non-handlers in area B. The haemoglobin levels were 14.30 g/ 100 ml in handlers and 13.47 g/100 ml in non- handlers in area A and 13.86 g/100 ml in handlers and 13.66 g/100 ml in non-handlers in area B. The mean AChE and haemoglobin levels in the handlers might be due to the presence, in this group, of a greater proportion of male subjects. The chemical determinations were repeated on 189 subjects of area A (84 handlers and 105 non- handlers) and on 99 of area B (26 handlers and 73 non-handlers) at the second examination. The data from the two examinations are summarized in Table 21 and the following observations may be made: (1) Total plasma arylesterase activity was greatly reduced in area A at the second examination, so that the values were not significantly higher than those of the people in area B. 195 B. PACCAGNELLA AND OTHERS Table 21. Chemical constituents of the blood of the persons examined twice Handlers Non-handlers Constituent First examination Second examination First examination Second examination Mean S.D. Mean S.D. Mean S.D. Mean S.D. Area A total plasma cholinesterase (a pH/h) 0.777 0.220 0.831 0.213 0.776 0.180 0.809 0.196 thermostable plasma cholinesterase (a pH/h) 0.118 0.044 0.142 0.055 0.115 0.051 0.129 0.056 RBC acetylcholinesterase (a pH/h) 0.737 0.194 0.725 0.178 0.695 0.192 0.670 0.173 total plasma arylesterase (tM mol/ml/h) 263.48 68.48 258.83 69.19 274.93 51.50 262.88 74.66 thermostable plasma arylesterase ( s mol/ml/h) 35.42 13.14 26.12 10.71 38.52 11.27 25.36 11.48 haemoglobin (g/100 ml) 14.30 1.58 14.05 1.67 13.49 1.72 13.79 1.80 serum total proteinaemia (g/1 00 ml) 7.45 0.77 7.70 0.76 7.59 0.77 7.61 0.78 total serum cholesterol (g/100 ml) 217.67 70.57 216.33 53.97 200.82 56.47 207.85 52.42 endogenous serum lipoprotein lipase (is mol/ml/24 h) 47.15 27.47 38.83 22.27 46.14 18.38 38.38 19.19 Area B total plasma cholinesterase (J pH/h) 0.759 0.224 0.844 0.149 0.821 0.211 0.854 0.163 thermostable plasma cholinesterase (a pH/h) 0.145 0.067 0.119 0.028 0.118 0.035 0.146 0.070 RBC acetylcholinesterase(4a pH/h) 0.697 0.196 0.858 0.190 0.667 0.205 0.862 0.149 total plasma arylesterase (g mol/ml/h) 238.01 50.32 284.63 54.16 250.47 61.81 268.68 69.36 thermostable plasma arylesterase (is mol/ml/h) 36.61 21.41 33.13 19.28 36.44 13.90 28.99 16.81 haemoglobin (g/100 ml) 13.80 1.89 13.64 1.27 13.64 2.32 13.73 1.77 serum total proteinaemia (g/100 ml) 7.37 0.88 7.13 1.05 7.27 0.76 7.40 0.88 total serum cholesterol (g/100 ml) 206.11 69.69 187.16 46.76 214.38 61.28 196.58 47.46 endogenous serum lipoprotein lipase (Is mol/ml/24 h) 28.41 19.07 33.45 12.45 43.29 26.33 55.67 27.56 (2) Thermostable plasma arylesterase activity de- creased more in area A than in area B, so that, at the second examination, its mean level was higher in area B than in area A. (3) AChE levels decreased in area A and in- creased in area B, so that the mean level was higher in area B than in area A at the second examination. (4) Considerable changes occurred in the levels of endogenous serum lipoprotein lipase, the mean level of which was higher in area B at the second examina- tion, whereas at the first examination it was higher in area A. (5) In area B, the mean level of total serum cholesterol decreased so that at the second examina- tion it was lower than in area A. 196 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH Table 22. Serum phosphatase levels and leucocytes counts in the blood of the two populations at the second examination Area A Area B Characteristic Handlers Non-handlers Handlers Non-handlers No. Mean No. Mean No. Mean No. Mean __ (S.D.) (S.D.) __ (S.D.) 01 (S.D.) alkaline phosphatase 35 22.57 75 20.44 26 32.06 79 29.92 (mU/ml) (13.53) (10.82) (10.39) (11.97) acid phosphatase 35 11.91 75 10.66 26 18.17 79 16.81 (mU/ml) (8.37) (6.58) (7.18) (8.02) loucocyte count 61 5138.93 94 5 271.22 26 4 813.46 79 5 496.52 (per mm3) (1 432.22) (1 592.20) (1 376.85) (1 750.46) (6) Total serum protein was much higher in area A than in area B at the second examination than it had been at the first. These changes made it difficult to differentiate between the people of the two areas. They may be largely ascribed to intraindividual variations and to the fact that in field studies it is impossible properly to standardize the populations selected for blood sampling. At the second examination serum alkaline and acid phosphate determinations were made on blood samples from 110 subjects of area A (35 handlers and 75 non-handlers) and from 105 subjects of area B (26 handlers and 79 non-handlers). Leucocyte counts were also made for 155 subjects of area A (61 handlers and 94 non-handlers) and for 105 sub- jects of area B (26 handlers and 79 non-handlers). The results are summarized in Table 22 and show higher mean levels of serum phosphatase in area B than in area A (P <0.01). The mean number of leucocytes did not differ in the people of the two areas, although slightly lower levels were found in handlers than in non-handlers (0.05 <P <0.10). This last observation agrees with the results of Davignon et al. (1965) and may be related to some long-term effects on the haemopoietic system. The results of the phosphatase determinations, on the other hand, are difficult to interpret and further investigation is required. Electrocardiograms. At the second examination, electrocardiograms were made for 228 subjects from area A (89 handlers and 139 non-handlers) and 105 from area B (29 handlers and 76 non-handlers). The ECG tracings were evaluated according to the Minnesota code, and the results are summarized in Table 23. This table shows (1) a higher frequency Table 23. Percentage frequency of electrocardiogram abnormalities Area A Area B Abnormality Minnesota Non- T Handlers Non- Totalcode H andlers handlers Totalhnlr (%) %) !%) (%) (% Q-wave and QS items I1-3 5.6 1.4 3.1 6.9 0 1.9 electric axis 111-2 3.4 4.3 3.9 0 1.3 0.9 ST depression IV1-4 9.0 7.9 8.3 0 1.3 0.9 T-wave items V1-3 1.1 4.3 3.1 3.4 2.6 2.9 AV-conduction Vl -4 3.4 1.4 2.2 0 2.6 1.9 ventricular conduction V 11-4 6.7 1.4 3.5 3.4 1.3 1.9 arrhythmias VIIII-9 3.4 4.3 3.9 0 1.3 0.9 197 B. PACCAGNELLA AND OTHERS of Q-wave and QS items and of ventricular conduc- tion abnormalities among the handlers of both populations than among the non-handlers, and (2) a higher frequency of ST-depression among the handlers of area A than among those of area B. The apparent relationship between exposure to pesticides and certain ECG findings (e.g., the ventricular conduction abnormalities that were more frequent among handlers than among non-handlers) will be further studied. DISCUSSION AND CONCLUSIONS The cross-sectional study showed that environ- mental pollution by organochlorine compounds was a little higher in area A than in the control area B but this difference was not significant. The level of storage of these compounds in fat of the people of the two areas did not differ. The prospective study showed consistently higher morbidity in the population of area A, where pesticides had been used for a longer period and at higher rates, than in the control population of area B. Since the levels of organochlorine compounds in human fat were similar in both areas and since these compounds form only a small part of the total amount of pesticides used, some of the differences in morbidity between the populations may be related to the greater use of other pesticides (e.g., organo- phosphorus compounds and carbamates) in area A than in area B. However, in earlier years organochlorine pesticides were used almost exclusively. No figures are avail- able on the use of organochlorine compounds in area A during the 15 years preceding our study, but as we are attempting to evaluate long-term effects we cannot ignore the possibility that previous exposure to these compounds had some effect on morbidity in the population. Signs of disorders of the autonomic nervous system, which may be caused by organophosphorus compounds, occurred more frequently in area A than in area B, and among persons occupationally exposed to pesticides more frequently than among persons not exposed. It should be noted that the organization of a trial such as the one described in this paper is bound to bring about behavioural changes among those involved, and in this way the trial has an educational value. Such behavioural changes occurred in both areas, mainly among those who handled pesticides during their work and who were not very safety conscious at the beginning of the study. Furthermore, during the course of the study, the pattern of use of pesticides in agriculture changed and the methods of spraying and the equipment used improved. In a study of the long-term effects of pesticides on human health it is important to recognize that several interrelated factors (e.g., socio-economic, nutritional, and behavioural factors) are involved. In our study the greater use of pesticides in area A was directly and indirectly related to the higher socio-economic level of the population that resulted from higher agricultural productivity; this situation is also associated with changes in several other living and nutritional habits. The influence of these factors on morbidity has been not evaluated, but it is reasonable to assume- particularly in view of the results of the blood chemistry studies-that they do have an effect. UM1 ETUDE EPIDEMIOLOGIQUE DES EFFETS A LONG TERME DES PESTICIDES SUR LA SANTP_ HUMAINE On a enquete en Italie sur les effets a long terme des pesticides sur la sante de l'homme. On a choisi a cet effet deux regions rurales distinctes: dans l'une (region A), les pesticides sont abondamment utilises depuis une quinzaine d'annees; dans l'autre (r6gion B), ils ne sont employes, en faible quantite, que depuis 10 ans. On a recherche en premier lieu la nature et la concen- tration des compos6s organochlores presents dans le milieu: l6gumes, fruits, graisses animales, fourrage, lait, sol, eaux de puits, de surface et de boisson, ainsi que dans le tissu adipeux (obtenu par biopsie ou lors d'autopsies) des habitants. On a constate que la teneur du milieu en pesticides organochlores etait legerement plus elevee dans la region A que dans la r6gion B; les taux d'accumulation de ces compos6s dans les tissus adipeux etaient en revanche tres semblables dans les deux regions. On a ensuite constitue, par 6chantillonnage aleatoire, deux groupes de population (469 habitants de la r6gion A et 152 habitants de la region B) qui a leur tour ont e divises en deux sous-groupes en fonction du degr6 198 LONG-TERM EFFECTS OF PESTICIDES ON HUMAN HEALTH 199 d'exposition aux pesticides (manipulateurs et non-mani- pulateurs). Tous les sujets ont fait l'objet d'investigations cliniques (etude des ant6cedents pathologiques et de la morbidite actuelle, recherche des symptomes 6voquant des troubles du systeme nerveux veg6tatif) et biochimi- ques (dosage des cholinesterases, de I'aryl-esterase, de la lipoproteine-lipase, de l'hemoglobine, des proteines seriques totales, du cholest6rol, des phosphatases alcaline et acide; numeration des h6maties et des leucocytes; electrocardiogramme). Sauf de legeres variations locales, les ant&edents pathologiques des membres des divers sous-groupes etaient tres semblables. En revanche, au moment de l'enquete, le taux de morbidite a e constamment plus eleve parmi les habitants de la region A que parmi ceux de la region B. Les signes d'une atteinte du systeme nerveux veg6tatif - consequence possible d'une intoxi- cation par les organophosphor6s - etaient plus fr6quents dans la r6gion A que dans la region B et plus souvent observes chez les sujets que leur profession exposait davantage aux pesticides. Les examens de laboratoire se sont aver6s peu utiles pour appr6cier le degre d'exposition aux pesticides, etant donne les changements fr6quents dans la composition des echantillons de population. Les traces electrocardiographiques ont fait ressortir une certaine correlation entre le degre d'exposition aux pesticides et la frequence des anomalies de la conduction ventriculaire. Cet aspect du probleme requiert de nou- velles recherches. REFERENCES Angustinsson, K. B. & Henricson, B. (1965) Acta physiol. scand., 64, 418 Bessey, 0. A., Lowry, 0. H. & Brock, M. J. (1946) J. biol. Chem., 164, 321 Casarett, L. J., Fryer, G. C., Yanger, W. L. & Klemmer, H. W. (1968) Arch. environ. Hlth, 17, 306 Davignon, L. F., St Pierre, J., Charest, G. & Tourangeau, F. J. (1965) Canad. med. Ass. J., 92, 597 De Vlieger, M., Robinson, J., Baldwin, M. K., Crabtree, A. N. & Van Dijk, M. C. (1968) Arch. environ. Hlth, 17, 759 Durham, W. F. (1965) Effects ofpesticides on man. In: Chichester, C. C., ed., Research in pesticides, New York, Academic Press, pp. 93-102 Durham, W. F. (1969) Ann. N.Y. Acad. Sci., vol. 160, art. 1, p. 183 Grob, D. (1963) Anticholinesterase intoxication in man and its treatment. In: Koelle, G. B., ed., Cholinesterases and anticholinesterase agents, Berlin, Springer Verlag, pp. 989-1027 Hoerlein, H. & Pilz, W. (1962) Hoppe-Seylers Z. physiol. Chem., 327, 256 Kraybill, H. F. ed. (1969) Biological effects ofpesticides in mammalian systems. In: Ann. N.Y. Academ. Sci., vol. 160, art. 1, pp. 5-422 Leffler, H. H. (1959) Amer. J. clin. path., 31, 310 Michel, H. 0. (1949) J. Lab. Clin. Med., 34, 1564 Pilz, W. & Hoerlein, H. (1964) Hoppe-Seylers Z. physiol. Chem., 335, 221 Stigliano, M., Bennetti, P. & Antonelli, A. (1967) Ann. Sclavo, 9, 667 Waddell, W. J. & Hill 0. (1956) J. Lab. Clin. Med., 48, 311 WHO Expert Committee on Insecticides (1967) Wld Hlth Org. techn. Rep. Ser., No. 356 Zeller, E. A. (1956) Arch. Biochem., 61, 231