Mortality rate of people exposed to Mustard Gas during Iran-Iraq war in Sardasht, Iran: a 32 years retrospective cohort study

Mustard agents are vesicant chemicals agent, including two important types: Sulfur mustard and Nitrogen mustard. Sulfur mustard, also known as Mustard gas (MG), is the most well-known and hazardous vesicant chemicals agent which was synthesized for the first time by Despretz in 1822 [1, 2]. This fast-absorbing substance can enter the body through various routes including the respiratory tract, skin, and anterior surface of the eye, as well as gastrointestinal tract, which consequently can lead to recurrent complications in various organs and death in the short and long term [3‐5].

The use of MG as a chemical weapon in war has so far caused hundreds of thousands of people to be exposed to small or large amounts of this agent in acute form. In other words, since the first use of MG as a chemical weapon by the German army in 1917 in the war against Canadian and French troops, this agent has been repeatedly used in various wars against different countries such as Iran in the 8-years war between Iran and Iraq which has caused injury, complications and death to military and civilian people [6, 7].

Sardasht is one of the cities in Iran that its civilians were exposed to MG during the Iran-Iraq war from 1981 to 1989. The city, located in northwestern Iran with latitude of 36º 9′ 4.48″ and a longitude of 45º 29′ 10.46″, was bombarded by Iraq on June 28, 1987. The bombing dropped seven 250 kg MG bombs, four of which exploded in a densely populated downtown, affecting civilians [3]. People exposed to MG in the city of Sardasht in 1987 during the Iran-Iraq war were civilians and had similar characteristics to the general population. In addition, these people were exposed to this agent without any protective clothing and mask.

So far, a number of studies have investigated the long-term mortality rate of exposed people to MG in war; however, in one hand, their number is limited and, on the other hand, they are all conducted on a military population and there is no information on long-term mortality in a civil population such as Sardasht [4, 8]. Meanwhile, after 32 years of exposure of these people to this agent in Sardasht, no study has measured the impact of exposure to MG on mortality rate in these people. Therefore, we investigated mortality rate among people exposed to MG during Iran-Iraq war in Sardasht, Iran after 32 years. It should be noted that the paper does not address other long-term medical consequences of exposure to the agent.

This study aimed at investigating the mortality rate of people exposed to MG in the city of Sardasht, which was bombarded by seven 250 kg MG bombs during the Iran-Iraq war. The results of this study are important because, unlike other similar studies, the exposed individuals in Sardasht are civilians and therefore have similar characteristics to the general population. Moreover, they had no protective clothing and mask to protect themselves from this agent.

The results of this study on all-cause mortality and its comparison with the general Iranian population showed that mortality rate in people exposed to MG in Sardasht did not increase after 32 years. Indeed, observed death was remarkably lower than expected death (SMR: 0.687). Most of the similar studies have focused on the impact of chronic exposure or on the impact of acute exposure on cause-specific mortality. However, in a similar study concerning impact of acute MG exposure on all-cause mortality, Norman et al. selected a sample of 2718 men admitted to American special gas hospitals in August and October of 1918 at the end of the World War I with clear evidence of MG exposure. Then, they compared mortality of the cohort to the general population (US white male population) from 1918 to 1965 and indicated that observed death was very close to expected death after 47 years (SMR: 0.99) [4]. However, in the present study, although the subject were civilians who had no protective clothing and mask to deal with this agent, the number of deaths observed was remarkably lower than expected. As an explanation, we believe that the result of the current study can be partly due to the provision of immediate and special care under the national care management scheme. Another study by Mokida et al. in Japan compared the mortality of people exposed to MG in factory workers to the general Japanese population after 57 years, and they also showed lower number of observed deaths than the number of expected deaths. They attributed the reason to the receipt of special health care beside existence of a healthy worker effect [5]. However, in our study, the role of receiving special health care is much more obvious; the study mentioned in Japan was based on factory workers population, while the current study is conducted on civilians and here we do not have a healthy worker, where we can attribute this reduction in mortality to the role of receiving special health care with more certainty. As a further explanation, other studies have highlighted the provision of good special services for war survivors in Iran. For example, the study by Mousavi et al., indicated that health care utilization (outpatient and inpatient) are more frequent among war survivors than the Iranian general population [14].

Besides, all-cause standardized mortality ratio according to the severity of late complications showed that in individuals with severe lung complications and severe skin complications, the observed death was much higher than expected. Of course, it should be noted that the severe skin complications were slightly more than expected, while the severe lung complications were more than twice the expected amount. Moreover, cox regression indicated that only severe lung complications increased risk of death in the cohort. There has been a lack of similar studies addressing long term mortality according to severity of late complications. However, this result indicated that when subjects were categorized according to severity of late complications, only severe complications, especially severe lung complications, affected all-cause mortality. Various studies have shown that the severity of the complications, especially lung ones in people exposed to MG, can become more severe over time [15, 16]. Even the evidence has shown that people who were exposed to low doses of MG and did not have acute clinical symptoms such as blisters also developed pulmonary bronchiolitis complications over time [17]. Therefore, preventing the progression of respiratory complications to severe complications in people affected by MG using early detection is very important and can improve the survival of the injured individuals.

In this study, we also investigated the causes of deaths in this population and compared it with Iranian national population using cause-specific SMR. It is necessary to mention that no autopsy was performed to determine the cause of death because of ethical consideration. Therefore, unfortunately, in this study we did not have the cause of almost 45% of deaths. Nevertheless, the results showed that in the case of respiratory diseases, unlike other diseases, the observed death was higher than expected. It is well known that pulmonary injury is the principal cause of mortality in the first few days to weeks after intense exposure to MG. Also, the existence evidences indicate a causal relation between both acute and chronic exposure to sufficient concentrations of MG and long-term mortality due to malignant respiratory and none-malignant respiratory disease, especially bronchitis which is consistent with our result [4, 8, 18]. Of course, it is important to note that despite the evidences for a relation between exposure and long-term mortality of malignant respiratory disease, chronic exposure has been shown to be carcinogenic, while there is still doubt about carcinogenicity of acute exposure. In other words, studies which have reported an association between acute exposure and mortality from malignant respiratory disease, did not interpret this exposure as carcinogenic, but reported its association with increased risk of other lung diseases, such as bronchitis led to cancer. In the present study, we also observed excess death only due to non-malignant respiratory diseases [4, 8, 19]. Therefore, considering the above, increased mortality of respiratory diseases can be justified. However, despite increasing mortalityof respiratory disease, all-cause mortality rate has not increased. In fact, this was partly due to  the greater distribution of mild to moderate lung complications. As we have shown, only severe lung complications can affect all-cause mortality, while there were a low percent of the subjects with severe lung complications in Sardasht; therefore, all-cause mortality was not affected. It should be noted that low percentage of severe lung cases may be partly due to special care. Moreover, special care has reduced diseases and disorders in other organs such as the circulatory system which almost is the leading cause of death, thereby reducing overall mortality. As existing evidence has shown, even in people with lung involvement such as COPD and asthma, circulatory disease is the leading cause of death, and an increase in respiratory deaths without an increase in circulatory system-realted deaths could rarely affect total deaths [20]. Moreover, another reason is that deaths due to some cause in cohort group are far lower than the general population of the country, while receiving special services do not play a role in reducing them. For example, road injury death, which is almost among the top 3 causes of death in the general population of Iran, was very low in these people. Because these people travel very little given their circumstances, and therefore they have fewer deaths due to  road injury  than the general population of the country, which has definitely affected their overall mortality rate. Therefore, in this study, exposure to MG, despite involvement of the lungs and increased respiratory mortality rate, has not increased all-cause mortality rate due to the control of circulatory disease and other leading causes of death.

One of the strengths of this study is its high sample size. It can also be noted that this is the first study to investigate the effects of acute MG exposure in a population with similar characteristics to the general population. In addition, the other strength of this study is the duration of the study, which individuals were included approximately 32 years after the exposure time. The limitations of the study included the lack of information on some variables such as underlying diseases and genotypes of individuals because we used available data. Also, another limitation was the failure to acquire data of all the survivors exposed to MG in Sardasht; therefore, we used only the information of people with valid evidence about their exposure. In other words, their exposure to MG was confirmed by VMAF. We considered them as a sample of all the exposed people, and therefore expressed the results with a confidence interval. Another limitation was that we were not allowed to use autopsy to determine the cause of death; consequently, the cause of death in approximately 45% of deaths were unknown. Nevertheless, it should be noted that even if all the death causes were known, all-cause mortality rate would definitely be lower than expected, and the respiratory mortality would certainly be higher than expected.

While this study is the first one to document long-term mortality rate among the exposed Sardasht population, it is worth noting that previous studies have documented many possible long-term effects of MG exposure on fertility, quality of life, immunological parameter, mental disorders, late complications in various organs, and the incidence of cancer in Sardasht and other parts of Iran. To acquire more information, we can refer to those studies [3, 7, 16, 21‐24]. For example, based on results of two systematic reviews, late dermatological, respiratory and ophthalmological complications were the most prevalent complications caused by MG exposure in Iran. Late dermatological complications were mainly chronic severe itching, scars, dry skin, and skin discoloration. Ophthalmological complications included pinguecula, pterygium, chronic conjunctivitis, vascular tortuosity, limbal ischemia, thin corneas, diffuse corneal opacity, dry eye, and keratitis and their clinical symptoms. Respiratory complications were mainly obstructive chronic bronchitis, bronchiolitis, bronchiectasis, asthma, and fibrosis [24, 25]. In addition, some other studies have documented the effects of various treatments used for these people to improve their condition, and these studies are also available for more information [26, 27].

Overall, this study shows that people exposed to MG had an increased mortality rate due to respiratory diseases after 32 years, which is a specific consequence of the chemical injuries to the respiratory system. However, all-cause mortality rates did not increase over this period. This indicates the impact of special care measures on these individuals immediately after exposure which prevented disease and disorder specifically in circulatory system, and consequently reduced total number of death in long-term. Indeed, this result indicates that acute exposure to MG, even without wearing protective clothing and masks, has not increased all-cause mortality rate after 32 years if accompanied by special and ongoing care for those exposed. This study suggests that special and ongoing care may prevent increase of long-term mortality as a consequence of exposure to a chemical agent and  respiratory complications.