- SIDS =
- sudden infant death syndrome •
- ECG =
Peter Schwartz and colleagues1 have recently published the results of a 19-year study relating prolongation of the QTc to the sudden infant death syndrome (SIDS). The authors performed electrocardiograms (ECGs) on 3- to 4-day-old infants and obtained information on whether they had survived to their first birthday for 33 034 infants. The authors report the QTc for the 24 SIDS deaths, the other 10 deaths, and a group of 9725 surviving infants. Half of the SIDS deaths had a QTc of >440 msec (97.5 centile), while the QTc in none of the other deaths was >420 msec. They cautiously propose use of a prolonged QTc in the newborn period as a marker for risk for SIDS and suggest treatment with β-blockers for infants in epidemiologic risk groups.
The first consideration in commenting on this study needs to be the adequacy of the methodology. The study is particularly impressive because of the very large numbers of infants involved. This is important when considering risk for SIDS because of its low incidence of <1 per 1000 infants. Because the diagnosis of SIDS is made by exclusion, the basis for the diagnosis is critical. The criteria used to determine SIDS deaths seems appropriate as only autopsied cases were included. The method of evaluation of the ECGs and the statistics used also appear to be acceptable. We conclude that this is a significant paper whose findings must be taken seriously.
The authors' recommendations are another matter. Before a prolonged QTc can be accepted as a marker of risk for SIDS, this finding must be confirmed. It is disturbing that the large study (7254 infants with 15 SIDS deaths) reported by Southall et al2 did not demonstrate a difference in QTc when surviving infants were compared with the deaths. The explanation by Schwartz et al that the QTc in these deaths showed a tendency toward prolongation does not suggest a very robust finding. The recommendation that the QT interval needs to be measured beyond the first 48 hours because of variability in the immediate newborn period needs to be further explored to determine the optimal time for testing. The logistics of routine testing beyond nursery discharge complicate the issue. The question of the infant's inability to shorten the QT interval with increases in heart rate seems a potentially fruitful area for further study. The suggestion that a prolonged QTc in a healthy infant be treated with β-blockers is particularly disturbing. Although this treatment is effective for patients with the long QT syndrome, a well-established clinical entity, there is no assurance that it will be effective in preventing SIDS. Association does not mean cause.
Not only is the cause of SIDS still unknown, the mechanism of death whether primarily cardiac or respiratory failure is not known either. The sleep apnea theory as the mechanism for SIDS, proposed over 25 years ago, although never proven has spawned an entire industry of home monitoring. Despite the fact that in the intervening years, the sleep apnea theory essentially has been discredited and monitoring has had no effect on SIDS rates, the practice is still widespread. Monitoring, although not having the consequences of irradiation of the thymus, an earlier recommended treatment for prevention of SIDS,3 is an expensive procedure with an emotional toll for the families involved.
Research into causes of SIDS is difficult because of the low incidence of the deaths and our inability to identify the individual infant at risk. The current rate of SIDS is <1 per 1000 live births. Epidemiologic characteristics can identify groups of infants at higher risk, but in even those with the highest risk such as very low birth weight infants, the risk is about 1 in 100. This means that 99 high-risk infants will survive for each infant who dies, and we are unable to predict which of the 100 will actually die. In the present case, a prolonged QTc of >440 msec occurs in 2.5% of all infants born, or 25 in 1000 live births. Assuming that half of the SIDS deaths would have a prolonged QTc, then >2000 infants would need to be screened and at least 50 infants would need to be treated to potentially prevent one death. If we were sure that treatment would prevent the death and that it was safe for the normal infant, this degree of overdiagnosis and treatment might be justified. At present we have no such assurance. If we use screening and treatment generally at this stage in our knowledge, we would never find out if the treatment were effective. Almost all of the treated infants would not be expected to die. One physician or even one large clinic would not have the number of patients necessary to establish the treatment's effectiveness. If we accept the recommendation to screen and treat infants based on a prolonged QTc, we run the real risk of adding yet another chapter of futile, sometimes dangerous, management of risk for SIDS to the depressing story that already exists. The question of a prolonged QTc should be taken seriously as a matter for confirmation and study. It is too soon to recommend it as a basis for clinical management.
- Received August 3, 1998.
- Accepted September 12, 1998.
- Address correspondence to Joan E. Hodgman, MD, University of Southern California School of Medicine, Women's and Children's Hospital, L919, 1240 Misson Dr, Los Angeles, CA 90033.
- Southall DP,
- Arrowsmith WA,
- Stebbens V,
- Alexander JR
- Hagler S,
- Rosenblum P,
- Rosenblum A
- Copyright © 1999 American Academy of Pediatrics