Elsevier

Reproductive Toxicology

Volume 12, Issue 4, July–August 1998, Pages 469-476
Reproductive Toxicology

Original Contributions
Evaluation of reproductive outcomes in women inadvertently exposed to hexachlorobenzene in southeastern Turkey in the 1950s

https://doi.org/10.1016/S0890-6238(98)00019-7Get rights and content

Abstract

In southeastern Turkey during the period of 1955–57, women were accidentally exposed to the fungicide hexachlorobenzene (HCB) after eating contaminated seed grain and developed porphyria cutanea tarda (PCT). While HCB has been shown to be a potent oocyte toxicant in primates and has been identified as an ovarian follicular fluid contaminant in women, its effect on human reproduction is poorly understood. This study was undertaken to evaluate the effects of HCB on women with a known high dose exposure. A retrospective controlled cohort comparison study of three groups was conducted. Group 1, those with confirmed PCT; Group 2, controls for the region and Group 3, controls for the country of Turkey, were followed-up after approximately 40 years (n = 42/group). Blood samples were taken for analysis of serum HCB, estradiol, follicle-stimulating hormone (FSH) and inhibin. Frequency of HCB detection was greatest in Group 3, while number of cases with HCB values exceeding 1 ng/mL was significantly greater in Groups 1 and 2. There were no differences in the other biochemical measures. Interviews were completed for each patient regarding reproductive history (number of pregnancies, live births, spontaneous abortions, still births and sex of live babies). Multiple comparisons of the three groups, based on Fisher’s test found the groups were inhomogeneous. When serum HCB was analyzed using correlated response logistic regression, there was a strong relationship between serum HCB levels and risk for spontaneous abortion but not sex ratio of children. The following findings were made. HCB is detectable and ubiquitous in serum samples from women in the country of Turkey with identified and unidentified exposure events. Spontaneous abortion risk is not restricted to women with identifiable exposure to HCB but to a surrogate marker of exposure (serum HCB sample). The risk of spontaneous abortion with HCB exposure requires further investigation.

Introduction

Between 1955 and 1959, cases of mixed porphyria resembling porphyria cutanea tarda were identified in the region of southeastern Turkey 1, 2, 3, 4, 5, 6. Doctors in the region thought this outbreak was related to inadvertent ingestion of seed grain contaminated with the fungicide hexachlorobenzene. Seed grain treated with Chlorable and Surmesan (active ingredient is HCB; 20% concentration) to combat Tilletia tritici was used as food when it arrived after the planting season was over. Symptoms of toxicity took approximately 6 months to develop. It was not until 1961–63 that experimental evidence demonstrated HCB could induce porphyria in rats 7, 8, 9, 10. As a result of the latency of identification of the causative factor in the outbreak of PCT, an estimated 3,000 to 4,000 patients developed PCT with an annual mortality rate of 10%. Children between the ages of 6 and 15 years of age made up more than 80% of the affected population. It was also estimated that 1,000 to 2,000 children under the age of one year died. As a result, in many villages there were no children between the ages of 2 and 5 in the mid- to late 1950s 11, 12, 13, 14.

Table 1 summarizes patient symptoms associated with the acute and chronic phases of PCT from follow-up studies on 225 of the original patients 30 years after the inadvertent exposure 11, 12, 13, 14. At 30 years follow-up, the average age of subjects was 32.8 years and the average age of onset of symptoms had been 10.3 years. Laboratory findings confirmed that 25% of the patients had abnormally high levels of urinary porphyrin excretion compared to controls for the country 25 to 30 years after exposure. Furthermore, there were significantly elevated levels of HCB in breast milk of lactating mothers that had PCT (0.51 ± 0.75 ppm mothers with PCT; 0.16 ± 0.23 ppm unaffected siblings; 0.07 ± 0.07 ppm controls for the country of Turkey) 25 to 30 years after the initial exposure. Reproductive outcomes of 225 original patients reported 15 abortions in 188 pregnancies and 31 children born to porphyric mothers died within the first several years of life (13).

The reproductive effects of HCB have been studied more extensively in animal models. Direct administration of high doses of HCB by gavage (64 mg/kg) to Rhesus monkeys reported hepatic disease consistent with porphyria and degenerative changes in ovarian structure including widespread primordial germ cell destruction (15). Studies in cynomolgus monkeys demonstrated initiation of primordial germ cell loss at low doses of HCB (0.1, 1.0, and 10.0 mg/kg) without evidence of induction of porphyria (16). The primordial germ cell appears to be exquisitely sensitive to HCB levels before the onset of more clinically recognized conditions (porphyria). Despite evidence of germ cell destruction, the cynomolgus monkeys were capable of producing embryos after ovulation induction, oocyte retrieval and spermatozoon insemination in vitro (16).

Recent studies have reported contamination of reproductive and nonreproductive sites with HCB 17, 18, 19, 20, 21. Adverse reproductive outcomes such as decreased neonatal survival, placental transfer in the rabbit (22), reduced egg volume and hatchability in quail (23), and embryo morbidity in herring gulls have been reported in association with HCB contamination (24).

A recent annotated bibliography discusses the grave concern of HCB as an environmental contaminant (21). HCB is an ubiquitous environmental contaminant which is lipophilic and tends to remain in fat stores, serum, and follicular fluid (21). Low concentrations have been isolated in follicular fluid during in vitro fertilization in Germany, Austria 17, 20 and Canada (18). These studies have reported significant individual and geographical variations in concentrations. In Canada, one study found the rate of detection of HCB in the serum and the follicular fluid was 83.3% and 65.3%, respectively. Mean concentrations of HCB in the serum and follicular fluid were 0.276 and 0.159 ng/mL, respectively (18).

The demonstration that HCB is a potent ovotoxicant, prevalent as a contaminant in the environment with evidence of sequestration in the follicular fluid of women raises the issue of what effects HCB has on the reproductive health of women. The women in Turkey who were exposed to HCB as children have now completed their reproductive lifetime in terms of childbearing years and provide critical information in our understanding of measurable reproductive effects. The first objective of this retrospective cohort analysis was to establish whether the diagnosis of PCT made in childhood, (representing clinically significant levels of exposure to HCB) was associated with significant levels of serum HCB contamination in adulthood. The second objective was to determine if clinically significant HCB exposure was associated with adverse reproductive outcomes. If PCT was not associated with adverse reproductive outcomes, then analysis of serum HCB levels with reproductive outcomes would be conducted. Association of HCB levels and outcomes that were not significantly different between the PCT group and the regional controls would argue for a nonporphyrinogenic mechanism of toxicity.

Section snippets

Experimental design

This study was a retrospective controlled cohort study of reproductive outcomes of three groups living in Turkey (n = 42/group). Group 1 consisted of individuals with clinically confirmed PCT who had been studied in the previous follow-up reports at 25 and 30 years 11, 12, 13, 14. Group 2 subjects were selected by age to match Group 1 and lived in the same geographic area (Diyarbakir) during the original episode. Group 3 consisted of age matched subjects selected from Ankara, the capital of

Results

The demographics of the study populations are summarized in Table 2. There was no significant difference in the overall ages of the subjects of the three groups. The median age at time of exposure to HCB for Group 1 (PCT) subjects was 6 years. Groups 2 and 3 had no known exposures to HCB.

The biochemical measurements for the three groups are summarized in Table 2. Median levels and quartiles are provided based on the skewness that predominated. The number of cases with HCB values exceeding 1

Discussion

This study used a retrospective cohort analysis to examine the effects of an accidental exposure to a putative reproductive toxicant on reproductive outcomes at approximately 40 years follow-up. The index group was a pediatric population diagnosed with PCT (Group 1) during the fundamental event (inadvertent exposure by ingestion of seed grain in southeastern Turkey in 1955–1957) previously followed up at 25 and 30 years 11, 12, 13, 14. This group was compared to cohorts that lived in the same

Acknowledgements

The authors acknowledge Professor Ishan Dogramaci, President of Bilkent University Foundation, for his assistance in establishing this study. Further, the authors acknowledge the contributions of the midwives who participated in this study.

References (38)

  • C Cam

    Une nouvelle dermatose epidemique des enfants

    Am J Derm Syph

    (1960)
  • I Dogramaci et al.

    A survey of 592 cases of cutaneous porphyria seen in southeastern Turkey

    Turk J Pediatr

    (1962)
  • S Cam et al.

    Acquired toxic porphyria cutanea tarda due to hexachlorobenzene. Report of 348 cases caused by this fungicide

    JAMA

    (1963)
  • R.K Ockner et al.

    Acquired porphyria in man and rat due to hexachlorobenzene intoxication

    Nature (London)

    (1961)
  • J.D Wray et al.

    Hexachlorobenzene as a cause of porphyria turcica

    Turk J Pediatr

    (1962)
  • R Enriquez de Salamanca et al.

    Experimental porphyriashexachlorobenzene-induced experimental porphyria

    Semin Dermatol

    (1986)
  • H.A Peters et al.

    Epidemiology of hexachlorobenzene-induced porphyria in Turkey. Clinical and laboratory follow-up after 25 years

    Arch Neurol

    (1982)
  • D.J Cripps et al.

    Porphyria turcica due to hexachlorobenzenea 20 to 30 year follow-up study on 204 patients

    Br J Derm

    (1984)
  • H Peters et al.

    Turkish epidemic hexachlorobenzene porphyria. A 30 year study

    Ann NY Acad Sci

    (1987)
  • Cited by (0)

    This work was supported by Environmental Health Directorate, Health Canada, and The Medical Research Council of Canada.

    View full text