Economic and Programmatic Aspects of Congenital Syphilis Prevention
Schmid, George, Bulletin of the World Health Organization
In 2001, an estimated 720 000 infants, almost all in developing countries, were born with HIV (1) and few will survive until their tenth birthday. This high mortality in children who will never survive and contribute to society has captured the world's attention. Numerous approaches to intervention based on screening of pregnant women for human immunodeficiency virus (HIV) and prophylaxis or therapy of sero-positive women have been investigated and, in industrialized countries, adopted. In developing countries, identification of all infected mothers and institution of prevention of mother-to-child-transmission (PMTCT) programmes, currently using (principally) single-dose antiretroviral prophylaxis for mother and baby, might prevent about one-third to one-half of the 720 000 cases (240 000-360 000 cases) (2).
In sub-Saharan Africa, an estimated two million or more women with active syphilis become pregnant each year; in an estimated 1 640 000 of them, infection remains undetected during pregnancy (3). Although estimates vary, adverse pregnancy outcomes occur in up to 80% of women with acute syphilis, including stillbirth (40%), perinatal death (20%) and serious neonatal infection (20%) (4). In African countries, syphilis is the leading cause of perinatal mortality (21%) (5). Women with early syphilis are most likely to infect their fetus (6). To identify these women, titres of reaginic antibody are often measured: titres of 1:8 or greater ("high titre") indicate greatest risk (7-9). It is estimated that about half of pregnant women with active infection have "high-titre infection" (7-10). If this estimate is correct, the numbers of fetal deaths in Africa each year from untreated maternal syphilis rival those from HIV infection. Screening pregnant women for syphilis, treating those who are seropositive, and preventing reinfection will be effective if screening is performed early in pregnancy (11, 12). Universal institution of an effective programme to prevent congenital syphilis (CS) could avoid 492 000 deaths in Africa alone (1 640 000 (number of infected women) x 0.50 (rate of high-titres) x 0.60 (probability of fetal death) = 492 000). However, the prevention of infant deaths due to syphilis receives lower priority than prevention of those due to HIV infection.
This indifference is not because it is not known how to prevent CS. Prevention of CS is simpler than PMTCT of HIV, and highly cost-effective (13), yet often not carried out in middle-income or low-income countries. This paper reviews the economic rationale for screening pregnant women for syphilis, the reasons why screening is often unsuccessful, and suggests ways forward.
As syphilis incidence has fallen in the developed world, questions of whether screening of pregnant women for syphilis should continue to be recommended have resulted in full economic evaluations in the United Kingdom (14) and Norway (15), and partial evaluations in the United Kingdom and Australia (16-18). Despite varied modelling approaches, these analyses all concluded that screening is cost-effective (all evaluations) and cost-saving (full evaluations) at threshold prevalences considerably less than 1%, and should therefore continue.
Syphilis screening in developed countries is cost-effective at very low prevalences because treating CS is expensive. A child with CS is hospitalized, on average, 7.5 days longer than a child without CS (mean cost (2001 dollars) US$ 5 253) (I9). Although no study has followed children with CS to determine ultimate outcome, even an occasional case of neurological damage, as modelled in two analyses (14, 15), raises the mean health-care cost of a case of CS dramatically. When indirect costs (14, 15) are included, the cost of a case of CS escalates sharply. Thus, serological screening of pregnant women, which is inexpensive, is cost-effective even at very low prevalences of maternal infection. …