It has been shown that the incidence of ALL is higher among boys than girls [2
]. The male preponderance is particularly significant in T cell ALL with an increasing male to female ratio of 4/1. Young et al [2
] have demonstrated that the incidence of ALL was 1.4 times higher in boys than girls among children of age 14 years. Consistent with the established data the risk for development of ALL was found to be 1.6 times higher among boys than girls in this study.
Genetic factors are presumed to play a significant role in development of acute leukemia including ALL. Some inborn chromosomal defects have been shown to be associated with high risk of ALL such as trisomy 21 Down Syndrome [16-18]. When a sibling of a child develops ALL, the risk of the disease in that child has been found to be 2-4 times higher than the risk in normal population . The concordance of acute leukemia in identical twins has been estimated to be 20-25%. These findings suggest that leukemogenesis originates in utero and blood born metastases may also occur from one twin to the other. On the other hand, shared environmental exposures to leukomogens may play a role in the increased incidence of leukemia between siblings and other family members. Families with constitutional p53 abnormalities have been noted to have children with both AML and ALL . Based on the above data, we investigated whether the incidence of the malignant disorders among close relatives of children with ALL was higher than those in controls and showed that it was 5.36 fold increased in ALL group.
There have been conflicting results in the literature about the influence of increased maternal and paternal age in the development of ALL . In the present study, although no relationship was detected between the maternal age and the risk of ALL, increased paternal age (over 40 years old) has been shown to be associated with 3.5 fold increased risk.
The association of higher birth weight and the risk of the development of ALL is an area of controversy as well [5,7]. Daling et al  have reported an increased incidence of ALL among children between the ages of 2-4 years whose birthweight was higher than normal, however, beyond the age of 4 years such an association disappeared. In contrast, Mc Mohan et al  have found no association between the risk of ALL and the birth weight of children. In this study, the risk of ALL was found to be 3.9 fold increased in children born with a weight over 4 kg. However, no significant effect of birth weight on different age groups was detected as shown by Daling et al.
Spontaneous fetal loss particularly prior to the birth of a child has been shown by a number of investigators to be with 2 fold increased leukemia risk in that child . No such relationship was demonstrated in this study. Furthermore, history of maternal smoking during pregnancy also did not seem to influence the development of ALL. In contrast to our results, Stjernfeldt et al  have reported two fold increased risk of leukemia in children with history of maternal smoking during gestation; however, some other investigators have not confirmed this finding [21,22].
Large epidemiological studies have documented that ionizing radiation under certain conditions of exposure induces human cancers including leukemia [11,15]. In a study by the American National Academy of Science  prenatal exposure to ionizing irradiation during the first trimester was found to be associated with five fold increased risk of childhood cancers and when the exposure occurred beyond the first trimester, the risk has been shown to be increased 1.5 folds.
Several investigators have reported that intrauterine exposure to 0.3 - 0.8 cGy radiation was associated with significantly increased risk of ALL [10,12]. Since the effect of ionizing radiation on development of cancer is well-established, large measures are taken to avoid exposure to radiation, and nowadays ionizing radiation is underrepresented as a risk factor among leukemia cases. Thus, in this study a history of diagnostic X-ray during gestation was present in only two cases. Consequently, no conclusive statement could be yielded regarding the relationship between in-utero exposure to radiation and the risk of ALL.
The role of viral infections in the pathogenesis of leukemia has frequently been brought to the attention. Although it has not been confirmed in humans, many viruses have been shown to induce leukemogenesis in animals. The Ebstein Barr virus (EBV) infection has been shown to contribute, at least partially to development of African Burkitt Lymphoma and ALL with L3 morphology. In immunocompromised individuals this virus has been shown to cause lymphoproliferative disorders, and in its most aggressive form, EBV infections can result in chromosomal breaks, translocation and malignant transformation to a monoclonal proliferation . In our study group, we tried to identify if there was an association between in-utero exposure to viral agents and risk of ALL. Indeed, we have found significantly higher incidence of history of acute febrile infections with rash during gestation in the ALL group than in controls. Not only maternal febrile infections but also the incidence of physical trauma and medication usage during gestation were found to be increased in the ALL group when compared to controls. However, this finding may represent a recall bias due to the possibility of mothers of leukemic children to remember such gestational events much more than those mothers in the control group. When the effect of breast-feeding was analysed, we did not find a difference in the incidence of being devoid of breast-feeding between the groups with ALL and controls as shown by Shu et al  in 1995.
The effects of paternal occupation and exposure to some chemicals including pesticides on the development of childhood cancers have also been studied. Contaminations in drinking water, food additives, pesticides in environment can expose children particularly living in rural area . In the present study, the finding of higher incidence of living in rural area and having paternal occupation of farming in the ALL group may suggest an association between pesticide exposure and risk of ALL. However, a selection bias may also be considered.
Exposure to electromagnetic fields has been shown to induce chromosomal breaks and lymphoblastic transformation in experimental studies [8,14,24]. However, the results of clinical studies are conflicting. Measurement of exposure is difficult, given that electromagnetic field not only exists near high voltage power lines, but is also generated by several equipments including computers, television monitors and cellular phones. We compared the rates of living proximity to high voltage power lines and habit of watching television too close among ALL and control groups, but did not show a significant difference. Although a similar method was also used in several previous studies, we believe that more sensitive and reliable methods would be more informative.
In conclusion, our present study suggests that sex, paternal age, existence of malignant disorder among close relatives, higher birth weight, and exposure to pesticides is associated with increased risk of ALL in children.