Childhood physical abnormalities following paternal exposure to sulfur mustard gas in Iran: a case-control study

Subjects and setting

We began this study in 2004, about 17 years after the chemical attack in Sardasht. We followed children up to March 2009. Sardasht is a western Iranian city that was chemically bombarded with sulfur mustard gas (HD) by Iraqi forces on June 28, 1987. Documents maintained by military and civil authorities in Sardasht region confirm that approximately 8025 inhabitants were exposed to mustard gas and approximately 4500 people received medical treatment. Of these, records confirming early clinical manifestations of exposure and subsequent medical complications were available for 735 male survivors [9].

All Sardasht men that had a confirmed history of mustard gas exposure were selected and their progenies, who were born at least nine months after the exposure were considered as the exposed group. However, children whose both parents had confirmed history of exposure to mustard gas were not included in this study.

Since Iraqi forces attacked Iranian soldiers and civilians with various chemical agents several times, a special commission was assigned by Veteran Affairs Organization to confirm the chemical exposure and to determine the severity of injury. This commission consisted of a pulmonologist, a dermatologist, a neurologist, and an ophthalmologist that based their decision on hospital documents, mustard gas exposure stigmata, and clinical signs and symptoms of chronic complications of exposure, and appropriate paraclinical studies as required for diagnostic confirmation. All Sardasht men studied had official confirmation of chemical exposure from this commission. These patients were contacted using the Veteran Affairs Organization's database and by placing signposts and billboards across the city.

Control population

Rabat is a nearby city that has ethnic, cultural, and geographical characteristics similar to Sardasht but was not exposed to chemical attack. Rabat is 20 Km from Sardasht, and according to governmental documents, no proven traces of chemical agents were found in the Rabat region. There is no difference in access to medical care between the two regions. In this study, 164 couples from Sardasht were frequency matched by age with 136 couples from an existing cohort from Rabat. The initial groups were larger due to differences in the age distribution, we used a random number table to remove randomly several cases and controls to obtain a similar distribution of individuals in each age group.

Data Collection

We developed an interview sheet and a checklist as data-gathering tools. Our data collection included demographic data of all chemically injured males and their spouses in the exposed (Sardasht) group, age, occupation, educational level, date of marriage; close relative marriage, and any history of chemical injury. The same data was also gathered from the non-exposed (Rabat) group.

Trained general practitioners (GPs) evaluated medical histories from all progenies born at least nine months after the date of the chemical exposure in both exposed and non-exposed groups. In addition, the GPs performed a complete physical examination. If any physical abnormality or special disorder was suspected, the progeny was referred to a pediatrician to confirm the diagnosis. If the pediatrician considered additional diagnostic tools necessary for confirmation, paraclinical studies (e.g. echocardiography, spirometry, various imaging techniques) were performed.

Based on these examinations and confirmatory paraclinical studies, various abnormalities were identified and recorded. All abnormality entities were coded based on International Classification of Diseases, revision 10 (ICD-10) and/or disorders were classified as follows: diseases of the blood; endocrine, nutritional and metabolic diseases; mental and behavioral disorders; diseases of the nervous system and epilepsy; diseases of the eye, adnexa, ear and mastoid process; diseases of the circulatory system; diseases of the respiratory system; diseases of the digestive system; diseases of the musculoskeletal system; diseases of the genitourinary system; and congenital malformations. Whenever any disagreement occurred between the diagnostic opinion of GPs and the pediatrician, the pediatric diagnosis was accepted. In addition, if a progeny had a history of physical abnormality and disorder based on medical records but had normal physical exam at the time of the present study, due to previous medical treatment or surgical management, they were included as an event. Physical abnormalities or disorders were defined as any abnormality that can be detected by medical history, physical exam, or paraclinical studies.

Our sample size estimation was based on 95% power, a two-sided alpha (0.01), and an expected difference of 10% with a continuity correction. We provide descriptive statistics on the populations as groups. To take into account the correlation between families, we used SAS GENMOD procedure (SAS 9.2 SAS, Gary, NC), assumed a binomial distribution, logit link function and an exchangeable correlation matrix type.. We provide point estimates as Odds Ratios (ORs) with 95% Confidence Intervals (CIs). We used 2-sided P-values.

Possible explanation of findings

Sulfur mustard is a cytotoxic agent with mutagenic and carcinogenic effects [15]. Its active intermediate, sulfonium ion, reacts rapidly with proteins and nucleic acids, alters chemical functional groups such as amines, carboxyls, phosphates, S-H, and O-H groups, and produces alkylation products. This process may result in cross-linking between adjacent strands of DNA, which has been shown to be extremely lethal to cells [6];

Confirmed effects of sulfur mustard gas on spermatogenesis may explain the observed overall increase in physical abnormalities among the progenies of chemical victims [2, 12, 25, 26]. However, few studies, previously performed, were conclusive in determining a causal relationship. Pour-Jafari et al. [27] studied the rate of congenital malformations among progenies and their parents of Iranian victims before and after chemical warfare exposure and found that the rate of major malformations has increased from 33 per 1000 to 258 per 1000. Although they had used his cases as their own controls, and thus reduced selection bias, they did not adjust for the effect of parents' increasing age on malformation occurrence. Taher et al[28] claimed that the use of mustard gas in the Iran-Iraq conflict might have increased the number of cleft lip and cleft palate in children, however, they were not able to establish any causal effect between these two events, nor could they exclude the effects of other possible causes.

Strengths & Limitations

Strengths of our study include its sample size and locally relevant controls. Sardasht is one of the rare instances in the world with a large population of mustard gas victims. Despite initial resistance to studying the effects of exposure, overtime, initial politico-ethical resistance has faded. No similar study has yet been performed with this population. Furthermore, presence of an unexposed population in a nearby city (Rabat) with characteristics similar to the exposed group provided a unique opportunity to further strengthen this analysis. An inherent limitation of studies like this, where the participants may consider probable benefits by over-reporting adverse outcomes, and also where a long time is passed from the date of the event under investigation, is the possibility of recall bias. We aimed to minimize these biases by an inclusive physical exam and appropriate paraclinical studies performed by GPs and further confirmation by a pediatrician. However, our exposed group included only the progenies of those exposed males who were present in the city at the time of study. Moreover, we do not know the number of men exposed that died or moved since exposure. We used clinical examination and paraclinical tests of live birth children to determine disorders and malformations. It is possible that kariotyping and other genetic studies could have revealed more problems [25, 29]. It is also possible that miscarriages or abortions would have yielded differing effects. Finally, it is possible that our control population differs importantly from the exposed population that we have not recognized. This issue exists with any non-randomized comparison and we are unable to overcome this concern.

Our study found a significant association between exposure to mustard gas and common disorders and malformations. We did not find that any specific disorder or malformation was associated with exposure. We expected this as, with anencephaly, for example, there is a prevalence rate of about one in 1000 live births [1], thus, even with a doubling of risk rates, a much greater population would need to be studied to reveal strong association with chemical exposure. Considering the high number of chemical victims in Iran, this study may catalyze further comprehensive assessments with larger study populations. Using a post hoc sample size calculation, we find that our study had greater than 95% power to detect malformations and greater than 99% power to detect malformations/disorders.

We evaluated all clinical disorders and malformations, regardless of hypothesis driven associations with chemical exposure. Our reasoning for this is that the link between genetic disorders and resulting illnesses is not yet completely understood. So, for example, while rheumatic fever with congenital involvement may be most often associated with communicable disease genesis, we cannot rule out that parental chemical exposure may be associated with predispositions to certain illnesses [30].

Interpretation

Our study found a significant association between overall frequency of physical abnormalities and disorders and paternal exposure to mustard gas. Given the considerable victims of mustard gas among Iranian civilians and military personnel, as well as civilians and military in other conflicts, the effects of war may have a lasting and important effect on generations to come.