and reserve troops, as noted earlier, constituted a relatively high percentage of US troops deployed to the Gulf War. Anecdotal reports of an excess of Goldenhar syndrome, a rare congenital anomaly that affects the development of facial structures, prompted another study of birth defects. The syndrome is not specifically coded in the reporting of birth defects, so the authors reviewed medical records of birth-defect categories that would have subsumed the Goldenhar syndrome. Too few cases of the syndrome were found to support definitive conclusions (Araneta et al. 1997).
A large, population-based Department of Veterans Affairs study of 15,000 US Gulf War veterans vs 15,000 non-deployed veterans found that both male and female veterans self-reported higher rates of birth defects among liveborn infants, including “moderate to severe” defects (odds ratios [ORs] 1.8–2.8). The defects were grouped into broad categories; the largest (n=151) was described as “isolated anomaly”. Male veterans self-reported a higher rate of miscarriage among their partners (OR 1.62, 95% confidence interval [CI] 1.32–1.99) than did controls. Concerned about reporting bias, the investigators suggested that those observations be confirmed by a review of medical records (Kang et al. 2001).
A population-based study in several states captured births in all hospitals, both military and civilian, and matched birth certificates with military records during the period 1989–1993 (Araneta et al. 2003). The study measured the prevalence of birth defects among infants of Gulf War veterans and non-deployed veterans in states that conducted active case ascertainment of birth defects. Military record of 684,645 Gulf War veterans and 1,587,102 non-deployed veterans were electronically linked with 2,314,908 birth certificates from Arizona, Hawaii, Iowa, and selected counties of Arkansas, Califonia, and Georgia; 11,961 Gulf War veterans’ infants and 33,052 non-deployed veterans’ infants were identified. Of those, 450 infants had mothers who served in the Gulf War and 3966 had non-deployed veteran mothers. After examining 48 specific categories of birth defects, the study found a greater prevalence of three defects—tricuspid valve insufficiency (relative risk [RR] 2.7, 95% CI 1.1–6.6), aortic valve stenosis (RR 6.0, 95% CI 1.2–31.0), and renal agenesis (RR 2.4, 95% CI 0.7–8.3)—in infants conceived after the war by Gulf War-deployed men in comparison with non-Gulf War-deployed men. Aortic valve stenosis and renal agenesis had higher RRs among infants conceived after the war by Gulf War veteran men than among infants conceived before the war. The study also found a greater prevalence of hypospadias in male infants (RR 6.3, 95% CI 1.5–26.3) conceived during or after the war and born to female Gulf War veterans (in comparison with non-Gulf War-deployed females). The study was not designed to determine whether the excess risk was caused by environmental agents, and it should be interpreted as an exploratory study that investigated 48 categories of birth defects.
A recently published population-based UK study probed the prevalence of birth defects and fetal deaths in all UK Gulf War veterans and Gulf War-era controls (that is, veterans who were not deployed to the gulf war)—a total of 105,735 veterans—studied with a validated postal questionnaire (Doyle et al. 2004). The study period covered conception after the Gulf War and before November 1997. Male Gulf War veterans reported a higher risk of miscarriages in their partners than the comparison cohort (OR 1.4, 95% CI 1.3–1.5). They also reported a higher proportion of offspring with any type of malformation (OR 1.5, 95% CI 1.3–1.7). Examination by type of malformation revealed some evidence of an increased risk of malformation but it was weakened when the analyses were restricted to clinically confirmed conditions. Female veterans did not report an excess of miscarriage, and stillbirths and malformations were too few to be usefully analyzed.