Correspondence

The Interval between Pregnancies and Preeclampsia

N Engl J Med 2002; 346:1831-1832June 6, 2002

Article

To the Editor:

Skjærven et al. (Jan. 3 issue)1 analyzed data from the Medical Birth Registry of Norway and concluded that an extended interval between births is a previously unrecognized risk factor for preeclampsia. After adjustment for the interval between births, a change in partner was no longer a risk factor for preeclampsia. The authors did not consider two potential sources of error that may have biased their findings. The first is confounding by unregistered induced abortions, because only outcomes of pregnancies at least 16 weeks in duration were included in the birth registry. Induced abortions protect against preeclampsia2,3 and are obtained more frequently by unmarried, separated, or divorced women than by women in stable unions.4 The failure to account for terminations of pregnancies that occurred between registered births would result in erroneously long interpregnancy intervals attributed to women who changed partners. Adjustment for induced abortions would decrease the relative risk associated with the interbirth interval and increase the relative risk associated with changing partners.

A second potential source of error is the validity of the diagnosis of preeclampsia, which is notoriously prone to error. Hospital-discharge diagnoses of preeclampsia often do not capture milder cases5 and are commonly unsubstantiated by careful audits of medical charts for the presence of strictly defined, newly diagnosed hypertension and proteinuria in pregnancy. Moreover, such diagnostic error seems unlikely to be random: women with preexisting and gestational hypertension would be more likely than normotensive women to be misclassified as having preeclampsia. Was the diagnosis of preeclampsia in the birth registry audited against medical charts, or were subjects classified solely on the basis of summary diagnoses?

Audrey F. Saftlas, Ph.D., M.P.H.
University of Iowa College of Public Health, Iowa City, IA 52242
audrey-saftlas@uiowa.edu

Richard J. Levine, M.D.
National Institute of Child Health and Human Development, Bethesda, MD 20892

5 References

1. 1

   Skjaerven R, Wilcox AJ, Lie RT. The interval between pregnancies and the risk of preeclampsia. N Engl J Med 2002;346:33-38
   Full Text | Web of Science | Medline

2. 2

   Eras JL, Saftlas AF, Triche E, Hsu C-D, Risch HA, Bracken MB. Abortion and its effect on risk of preeclampsia and transient hypertension. Epidemiology 2000;11:36-43
   CrossRef | Web of Science | Medline

3. 3

   Sibai BM, Gordon T, Thom E, et al. Risk factors for preeclampsia in healthy nulliparous women: a prospective multicenter study. Am J Obstet Gynecol 1995;172:642-648
   CrossRef | Web of Science | Medline

4. 4

   Skjeldestad FE, Borgan JK. Trends in induced abortion during the 12 years since legalization in Norway. Fam Plann Perspect 1994;26:73-76
   CrossRef | Medline

5. 5

   Ales KL, Charlson ME. In search of the true inception cohort. J Chronic Dis 1987;40:881-885
   CrossRef | Medline

Author/Editor Response

The authors reply:

To the Editor: Drs. Saftlas and Levine raise a valid concern that a history of induced abortions may have influenced our results. Induced abortions have been reported to protect against preeclampsia in the next pregnancy.1 Confounding would occur if women who changed partners between two registered births were also more likely to have had an unrecorded induced abortion during this interval. Such abortions would reduce the apparent risk of preeclampsia in their later pregnancies. (It is worth noting that this protection could actually be the result of a shorter interval between pregnancies, if the induced abortion had been obtained because of fetal abnormality and the woman soon became pregnant again.)

We are unable to address this problem directly because there are no data on induced abortions in the Medical Birth Registry of Norway. We therefore carried out a simulation in which we assumed that there was an unrecorded abortion midway between the two recorded pregnancies for half of all women who changed partners and for none of the women who kept the same partner. Even with these extreme assumptions, we did not find a higher risk of preeclampsia among the women who changed partners (odds ratio, 0.9; 95 percent confidence interval, 0.8 to 1.0, with adjustment for the interpregnancy interval). This result reinforces our conclusion that the higher risk of preeclampsia after a change in partner (as reported in many previous studies2-4) is an artifact of the interpregnancy interval.

The second possible “bias” raised by Saftlas and Levine is not actually a bias but a built-in imprecision. They suggest that the diagnosis of preeclampsia is less accurate in our population-based registry than in more clinically based studies. This is unquestionably true. However, this imprecision would not explain our findings of an increased risk of preeclampsia associated with a longer interbirth interval or the disappearance of an association between a change of partner and preeclampsia when adjustment is made for the interbirth interval. The fact remains that the strong association between the interbirth interval and the risk of preeclampsia was not identified through a clinical study but in a population-based registry. Despite the lack of clinical detail inherent in such registries, they can be the source of new clues for clinical research.

Other investigators have recently published findings consistent with our own.5,6 One of these reports6 was based on an analysis by another group of investigators of the same data base we used.

Rolv Skjærven, Ph.D.
University of Bergen, 5021 Bergen, Norway
rolv.skjaerven@smis.uib.no

Allen J. Wilcox, M.D., Ph.D.
National Institute of Environmental Health Sciences, Research Triangle Park, NC 27709

Rolv T. Lie, Ph.D.
University of Bergen, 5021 Bergen, Norway

6 References

1. 1

   Eras JL, Saftlas AF, Triche E, Hsu C-D, Risch HA, Bracken MB. Abortion and its effect on risk of preeclampsia and transient hypertension. Epidemiology 2000;11:36-43
   CrossRef | Web of Science | Medline

2. 2

   Robillard PY, Hulsey TC, Alexander GR, Keenan A, de Caunes F, Papiernik E. Paternity patterns and risk of preeclampsia in the last pregnancy in multiparae. J Reprod Immunol 1993;24:1-12
   CrossRef | Web of Science | Medline

3. 3

   Trupin LS, Simon LP, Eskenazi B. Change in paternity: a risk factor for preeclampsia in multiparas. Epidemiology 1996;7:240-244
   CrossRef | Web of Science | Medline

4. 4

   Dekker GA, Robillard PY, Hulsey TC. Immune maladaptation in the etiology of preeclampsia: a review of corroborative epidemiologic studies. Obstet Gynecol Surv 1998;53:377-382
   CrossRef | Medline

5. 5

   Basso O, Christensen K, Olsen J. Higher risk of pre-eclampsia after change of partner: an effect of longer interpregnancy intervals? Epidemiology 2001;12:624-629
   CrossRef | Web of Science | Medline

6. 6

   Trogstad LI, Eskild A, Magnus P, Samuelsen SO, Nesheim BI. Changing paternity and time since last pregnancy: the impact on pre-eclampsia risk: a study of 547 238 women with and without previous pre-eclampsia. Int J Epidemiol 2001;30:1317-1322
   CrossRef | Web of Science | Medline

Citing Articles (1)

Citing Articles

1. 1

   Clive J. Petry, Ken K. Ong, David B. Dunger. (2007) Does the fetal genotype affect maternal physiology during pregnancy?. Trends in Molecular Medicine 13:10, 414-421
   CrossRef