| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Departments of Obstetrics & Gynecology and Pediatric Research, Riks-hospitalet University Clinic, University of Oslo, Oslo, Norway.
Address reprint requests to: J. F. Frøen, University of Oslo, Rikshospitalet University Clinic, Department of Pediatric Research, N-0027 Oslo, Norway; E-mail: j.f.froen{at}klinmed.uio.no.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: All 76 cases of sudden intrauterine unexplained death and 165 explained stillbirths among singletons in Oslo, Norway (1986–1995) were compared with 582 controls. Information on maternal health was registered from antenatal health cards and the Medical Birth Registry of Norway. The effect of pregnancy duration was studied before logistic regression analysis corrected for pregnancy duration was performed.
RESULTS: Urinary tract colonization or infection in pregnancy was associated with a reduced risk for subsequent sudden intrauterine unexplained death (odds ratio [OR] 0.29 [0.12–0.74]), whereas insufficient physiologic hemodilution during pregnancy (lowest hemoglobin greater than 13 g/mL) increased the risk for sudden intrauterine unexplained death (OR 9.50 [1.30–69.3]). However, the risk for sudden intrauterine unexplained death remained unaffected by the total number of indicators of impaired maternal health during pregnancy, in contrast to significant impact on other stillbirths.
CONCLUSION: Urinary tract colonization or infection may offer protection against sudden intrauterine unexplained death. We hypothesize that the subsequent maternal immune response offers transplacental protection against lethal fetal infections by common pathogens of urinary tract infections. Other health indicators have little impact on sudden intrauterine unexplained death.
Unexplained antepartum stillbirths constitute one of the largest groups in current classifications of the developed world’s perinatal mortality. Nonetheless, as indicated by the term "unexplained," very little is known about the mechanisms involved in these tragic deaths that typically occur close to term.1,2 Although a multitude of causes of death can be envisaged, previous studies suggest that these deaths constitute an epidemiological entity different from other causes of stillbirth in terms of maternal characteristics.1,3 One condition applies, however, and that is the exclusive inclusion of truly unexplained antepartum stillbirths validated by adequate postmortem examinations—defined as sudden intrauterine unexplained death in Table 1
.1,3
|
We have revisited the hypothesis of low incidence of maternal health problems and obstetric complications with a fully validated cohort of sudden intrauterine unexplained deaths corrected for the duration of pregnancy.
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). As controls for pregnancies terminating in sudden intrauterine unexplained death we used 1) all pregnancies terminating in stillbirths in which an adequate cause of death was demonstrated (explained cases, n = 165) and 2) all singleton pregnancies leading to the live birth of an infant in Oslo identified through the Medical Birth Registry of Norway. The randomization function in Perl 5.6 (O’Reilly & Associates Inc., Sebastopol, CA) was used to select a random sample among them (control cases, n = 582). A detailed description of the validation process and details on the classification of cases have been published,1 but the following details are of interest in this context. Only a full external and internal autopsy including cranial contents and x-ray was accepted as an adequately performed autopsy. Descriptions of postmortem signs consistent with the common final death mechanisms of most pathologies (eg, "cerebral hypoxiaischemia," "hypoxia," "anoxia," or "asphyxia") at autopsy were not accepted as a cause of death if no pathology was found in the fetus, placenta, or umbilical cord to explain why it occurred. Nor was fetal growth restriction accepted as a cause of death if no cause for this condition was found.
Satisfactory documentation on performed antibody testing for toxoplasmosis, rubella, cytomegalovirus infection, and herpes simplex infection, and samples from the fetus and mother for microbiological cultures and diagnosis of other infections was found in 93% of sudden intrauterine unexplained death cases. In 98% of cases, we found the doctors’ and midwives’ standardized records for antepartum care (maternity health cards), including the mothers’ reported health during pregnancy. The registration of the selected indicators of impaired health from maternity health cards was performed blinded for pregnancy outcome and classification of cause of death (by RAM). In addition, all registrations made in the compulsory notification to the Medical Birth Registry of Norway on maternal health before and during pregnancy were collected for both stillbirths and controls. To evaluate the completeness of registrations in stillbirths versus live births, the registrations of their siblings born within ±5 years were also collected.
All statistics were performed with SPSS 10.0.7 (SPSS Inc., Chicago, IL). Logistic regression with one covariate was used to find crude (unadjusted) odds ratios (ORs) with 95% confidence intervals, and to decide whether observed differences were statistically significant. As all analyses of nonchronic markers of impaired health had to be corrected for pregnancy duration, true crude ORs are not reported for these, but logistic regression adjusting only for pregnancy duration is referred to as "crude" ORs in Table 3. Cross-tables were also used to detect possible associations between covariates. Multiple logistic regressions were used to obtain adjusted ORs and to test for possible interaction terms. The Hosmer-Lemeshow test was used to test goodness of fit of the different models.5 Univariate analysis of variance was used to compare pregnancy duration between groups.
|
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, in 71% of sudden intrauterine unexplained deaths, 77% of explained stillbirths, and 74% of live-born controls. Thus, only the registration from maternity health cards was submitted to further analysis.
|
), and all subsequent analyses of nonchronic disorders were therefore corrected for the duration of pregnancy. The only registered condition that significantly affected pregnancy duration in controls was preeclampsia, whereas infections other than urinary tract infections in explained stillbirths mainly occurred in pregnancies with a longer duration than mean. In the sudden intrauterine unexplained death group it was noted that the intrauterine death occurred later if the woman was given sick leave during pregnancy. However, as sick leave was regarded as a consequence of impaired health (the other variables), it was not entered into the multiple regressions.
In regression analysis the "crude" values, only corrected for the duration of pregnancy, reinforced the impression of reduced occurrence of bacteriuria or urinary tract infection in sudden intrauterine unexplained death (Table 3). This was confirmed in the multiple logistic regression analysis, which found little confounding effect of the other variables or interaction terms, and the occurrence of bacteriuria or urinary tract infection in pregnancy remained strongly associated with a decreased risk for sudden intrauterine unexplained death, with an adjusted OR of 0.3. Because of the uncertainties regarding the specificity associated with retrospective discrimination between bacteriuria and urinary tract infection, these two variables were combined in the multiple models. Further, no difference was made between cases treated with antibiotics and nontreated cases to preserve statistical strength in the analysis. However, all of these subgroups showed similar "crude" ORs when only adjusted for pregnancy duration (Table 4).
|
). This was even more so for explained stillbirths in which maternal infections other than urinary tract infection also were a significant determinant for future fetal death. The impression of little impact of maternal health on the risk for sudden intrauterine unexplained death obtained from the crude percentiles was confirmed in regression analysis. The risk for other antepartum stillbirths was observed to have an increasing tendency with increasing numbers of indicators of impaired maternal health during pregnancy, and a significant correlation to the total number of diagnoses or indicators. This was not observed in sudden intrauterine unexplained death (Table 5).
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Thus, any protection provided to the fetus against these strains of bacteria found in intrauterine infections would be an important defensive mechanism in intrauterine life. Both the fact that such gram-negative bacteria are weak stimulators of immunity and the fact that more than 80 strains of E coli have been identified explain why maternal protection of the fetus against, for example, E coli is generally low. However, these strains also constitute common pathogens in urinary tract infections, which lead to highly increased levels of maternal antibodies against the pathogen from her intestinal flora and may offer transplacental protection to the fetus against this strain. We therefore speculate that ascending infections or bacterial colonization of the maternal urinary tract provide fetal protection against subsequent ascending infections to fetal tissues.
These findings may shed new light on our understanding of mechanisms involved in sudden intrauterine unexplained death: If protection against infection gives protection against sudden intrauterine unexplained death, undetected subclinical infections could be involved. Subclinical infections are also frequently present in the unexplained deaths of sudden infant death syndrome, and experience from this field of research may be useful in research on sudden intrauterine unexplained death despite the significantly different epidemiology.10 Among other candidates, extraintestinal E coli endotoxin has been found elevated in sudden infant death syndrome,11,12 and there was a consumption of endotoxin immunoglobulin G (IgG) and triggering of endotoxin IgM in cases of sudden infant death syndrome.13 Our hypothesis that high levels of maternal immunoglobulins will protect the fetus could be reflected by the finding that the toxicity of serum from victims of sudden infant death syndrome in a chick embryo model was completely reversed by commercial immunoglobulins.14 Also, in perinatal brain injury the role of inflammation and infection has lately become a subject of great interest, as there is now convincing evidence that intrauterine infections (also by the same strains of bacteria) are associated with brain white matter lesions and subsequent cerebral palsy.15,16 We have demonstrated in our laboratory that low-dosage endotoxin from E coli has significant detrimental effects on brain metabolism and survival during perinatal hypoxia in the piglet.17,18 Together, these results may suggest that subclinical infections and the fetal inflammatory response can not only cause hypoxia but also reduce perinatal hypoxia tolerance and survival.
Both the statistical strength of the association and the existing knowledge of possible biological effects strengthen the assumption that there is a causal association between maternal urinary tract infections or colonization of the urinary tract and the reduced risk for sudden intrauterine unexplained death. These hypotheses should be tested in larger multicenter studies, as is currently being performed in the International Interactive Inquiry on Sudden Intrauterine Unexplained Death.19
Of course, we do not argue that urinary tract infection is beneficial in pregnancy in general. Sudden intrauterine unexplained death occurs in only 0.1% of singleton pregnancies in our population, whereas the association between urinary tract infection and cases of intrauterine growth restriction is well known,20 although a review of current evidence concluded that it was unlikely that bacteriuria was a prominent factor in the genesis of low birth weight and preterm infants.21 Unfortunately, Naeye’s extensive and well-known study on increased perinatal mortality from maternal urinary tract infection22 did not include data on unexplained stillbirths. Stillbirths and the effect of bacteriuria alone were not reported separately, and the study reported only univariate analyses. However, it is still interesting to note that all of the excess mortality took place when birth followed the combined pyuriabacteriuria within 15 days, and mostly in pregnancies with other complications like hypertension or acetonuria. No previous studies have associated unexplained stillbirths with urinary tract infection.
Our finding of an insufficient hemodilution as a risk factor for sudden intrauterine unexplained death adds to recent reports of high hemoglobin concentration at first measurement during antenatal care as a risk factor for stillbirth, especially preterm and growth-restricted antepartum stillbirths.23 However, the mechanisms are unknown, with the exception of the fact that high viscosity increases the risk for thrombosis in the uteroplacental circulation.24 This was not found in sudden intrauterine unexplained deaths that were without significant infarctions of the placenta, but a reduced flow in the maternal intervillous space in general has also been suggested as a mechanism that may contribute to sudden intrauterine unexplained death.23,25
Maternal urinary tract infection during pregnancy protects the fetus against sudden intrauterine unexplained death. We hypothesize that the following immune response or elevated levels of maternal antibodies may be protective against subsequent lethal fetal infections by common pathogens of urinary tract infections. Other health indicators have little impact on sudden intrauterine unexplained death, and the risk for sudden intrauterine unexplained death remained unaffected by increasing numbers of indicators of impaired maternal health, in contrast to a significant impact on other stillbirths. However, insufficient physiologic hemodilution during pregnancy increases the risk for sudden intrauterine unexplained death.
|
Footnotes |
|---|
Received February 5, 2002. Received in revised form April 22, 2002. Accepted May 16, 2002.
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Yudkin PL, Wood L, Redman CW. Risk of unexplained stillbirth at different gestational ages. Lancet 1987;1(8543): 1192–4.[Medline]
3. Huang DY, Usher RH, Kramer MS, Yang H, Morin L, Fretts RC. Determinants of unexplained antepartum fetal deaths. Obstet Gynecol 2000;95:215–21.
4. Alessandri LM, Stanley FJ, Garner JB, Newnham J, Walters BN. A case-control study of unexplained antepartum stillbirths. Br J Obstet Gynaecol 1992;99:711–8.[Medline]
5. Hosmer DW, Lemeshow S. A goodness-of-fit test for the multiple logistic regression model. Communications Stat 1980;A10:1043–69.
6. Maleckiene L, Nadisauskiene R, Stankeviciene I, Cizauskas A, Bergstrom S. A case-referent study on fetal bacteremia and late fetal death of unknown etiology in Lithuania. Acta Obstet Gynecol Scand 2000;79:1069–74.[Medline]
7. Tolockiene E, Morsing E, Holst E, Herbst A, Ljungh A, Svenningsen N, et al. Intrauterine infection may be a major cause of stillbirth in Sweden. Acta Obstet Gynecol Scand 2001;80:511–8.[Medline]
8. Mathews JE, Mathai M, Peedicayil A, Mathews KP, Ponnaiya J, Jasper MP. Subclinical chorioamnionitis as a causal factor in unexplained stillbirths. Int J Gynaecol Obstet 2001;74:195–7.[Medline]
9. Naeye R. Disorder of placenta, fetus and neonate. Diagnosis and clinical significance. St. Louis: Mosby Year Book, 1992:257–61.
10. Frøen JF, Arnestad M, Vege A, Rognum TO, Saugstad OD, Stray-Pedersen B. Comparative epidemiology of sudden infant death syndrome and sudden intrauterine unexplained death. Arch Dis Child. In press.
11. Pearce JL, Luke RK, Bettelheim KA. Extraintestinal Escherichia coli isolations from SIDS cases and other cases of sudden death in Victoria, Australia. FEMS Immunol Med Microbiol 1999;25:137–44.[Medline]
12. Crawley BA, Morris JA, Drucker DB, Barson AJ, Morris J, Knox WF, et al. Endotoxin in blood and tissue in the sudden infant death syndrome. FEMS Immunol Med Microbiol 1999;25:131–5.[Medline]
13. Oppenheim BA, Barclay GR, Morris J, Knox F, Barson A, Drucker DB, et al. Antibodies to endotoxin core in sudden infant death syndrome. Arch Dis Child 1994;70:95–8.[Abstract]
14. Sayers NM, Drucker DB, Hutchinson IV, Barson AJ. Preliminary investigation of lethally toxic sera of sudden infant death syndrome victims and neutralisation by commercially available immunoglobulins and adult sera. FEMS Immunol Med Microbiol 1999;25:193–8.[Medline]
15. Yoon BH, Jun JK, Romero R, Park KH, Gomez R, Choi JH, et al. Amniotic fluid inflammatory cytokines (interleukin-6, interleukin-1beta, and tumor necrosis factor-alpha), neonatal brain white matter lesions, and cerebral palsy. Am J Obstet Gynecol 1997;177:19–26.[Medline]
16. Dammann O, Leviton A. Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 1997;42:1–8.[Medline]
17. Frøen JF, Munkeby BH, Stray-Pedersen B, Saugstad OD. Interleukin-10 reverses acute detrimental effects of endotoxin on perinatal cerebral hypoxiaischemia. Brain Res 2002;942:87–94.[Medline]
18. Frøen JF, Amerio G, Stray-Pedersen B, Saugstad OD. Detrimental effects of nicotine and endotoxin in the newborn piglet brain during severe hypoxemia. Biol Neonate. In press.
19. INSIDE Study-group. International Interactive Inquiry on Sudden Intrauterine Unexplained Death. Available at: http://www.med.uio.no/inside. Accessed 2002 Jul 10.
20. McGrady GA, Daling JR, Peterson DR. Maternal urinary tract infection and adverse fetal outcomes. Am J Epidemiol 1985;121:377–81.
21. Cunningham FG, MacDonald PC, Gant NF, Leveno KJ, Gilstrap LC, Hankins GD, et al. Williams obstetrics. Stamford, CT: Appleton & Lange, 1997:1125–44.
22. Naeye RL. Causes of the excessive rates of perinatal mortality and prematurity in pregnancies complicated by maternal urinary-tract infections. N Engl J Med 1979;300: 819–23.[Abstract]
23. Stephansson O, Dickman PW, Johansson A, Cnattingius S. Maternal hemoglobin concentration during pregnancy and risk of stillbirth. JAMA 2000;284:2611–7.
24. Naeye RL. Placental infarction leading to fetal or neonatal death. A prospective study. Obstet Gynecol 1977;50: 583–8.
25. Steer P, Alam MA, Wadsworth J, Welch A. Relation between maternal haemoglobin concentration and birth weight in different ethnic groups. BMJ 1995;310(6978): 489–91.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
