| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Section of Epidemiology and Biostatistics and the Division of Maternal-Fetal Medicine, Department of Obstetrics, Gynecology and Reproductive Sciences, and Department of Environmental and Community Medicine, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, New Brunswick, New Jersey.
Address reprint requests to: Cande V. Ananth, PhD, MPH, Department of Obstetrics, Gynecology and Reproductive Sciences, University of Medicine and Dentistry of New Jersey, Robert Wood Johnson Medical School, 125 Paterson Street, New Brunswick, NJ 08901-1977; E-mail: ananthcv{at}epi.umdnj.edu.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT PATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: A population-based, retrospective cohort study of singleton live births in New Jersey (1989–93) was performed. Mother-infant pairs (n = 544,734) were identified from linked birth certificate and maternal and infant hospital discharge summary data. Women diagnosed with previa were included only if they were delivered by cesarean. Fetal growth, defined as gestational age-specific observed-to-expected mean birth weight, and preterm delivery (before 37 completed weeks) were examined in relation to previa. Severe and moderate categories of fetal smallness and large for gestational age were defined as observed-to-expected birth weight ratios below 0.75, 0.75–0.85, and over 1.15, respectively, all of which were compared with appropriately grown infants (observed-to-expected birth weight ratio 0.86–1.15).
RESULTS: Placenta previa was recorded in 5.0 per 1000 pregnancies (n = 2744). After controlling for maternal age, education, parity, smoking, alcohol and illicit drug use, adequacy of prenatal care, maternal race, as well as obstetric complications, previa was associated with severe (odds ratio [OR] 1.37, 95% confidence interval [CI] 1.25, 1.50) and moderate fetal smallness (OR 1.24, 95% CI 1.17, 1.32) births. Preterm delivery was also more common among women with previa. Adjusted OR of delivery between 20–23 weeks was 1.81 (95% CI 1.24, 2.63), and 2.90 (95% CI 2.46, 3.42) for delivery between 24–27 weeks. OR for delivery by each week between 28 and 36 weeks ranged between 2.7 and 4.0. Approximately 12% of preterm delivery and 3.7% of growth restriction were attributable to placenta previa.
CONCLUSION: The association between low birth weight and placenta previa is chiefly due to preterm delivery and to a lesser extent with fetal growth restriction. The risk of fetal smallness is increased slightly among women with previa, but this association may be of little clinical significance.
Placenta previa is a significant contributor to preterm delivery, low birth weight, and perinatal mortality.1–3 However, the contribution of fetal growth restriction in low birth weight (under 2500 g) infants seen among women with placenta previa remains largely unexplored. Indeed, the relationship between placenta previa and fetal growth restriction is poorly understood. Earlier studies have reported conflicting results. Studies from Britain have found an increased number of growth-restricted infants among women diagnosed with placenta previa in the 1970s and 1980s,4,5 whereas others conducted during this approximate time period failed to corroborate these findings.6,7 Almost a decade ago, Wolf et al8 reported a lack of association between ultrasound-diagnosed placenta previa and intrauterine growth restriction based on a small case-control study.
The major drawbacks of all these studies are small sample size, with lack of statistical power to detect differences, and the lack of control for important confounding factors. To overcome these limitations, we examined the relationship between placenta previa, fetal growth restriction, and preterm delivery in a large population-based setting comprised of over half a million pregnancies, with extensive control for confounding variables.
|
PATIENTS AND METHODS |
|---|
|
TOP
ABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Fetal growth, one of the outcomes evaluated, was expressed as the ratio of the observed birth weight to the expected mean birth weight for each gestational age; this ratio was defined as the fetal growth ratio. The mean birth weight for each week of gestation was derived from the 1989–93 New Jersey singleton live births. Several degrees of fetal smallness, based on the fetal growth ratio, were defined as follows: severe when the fetal growth ratio was below 0.75, moderate when the fetal growth ratio was between 0.75 and 0.85, and large when the ratio was over 1.15. These indices of fetal growth were compared with appropriately grown infants (fetal growth ratio 0.86–1.15). The cut-points for these indices of fetal growth have been validated by investigators in other populations.9 We also examined the risks of delivering small for gestational age births, defined using birth weight centiles, in women with and without placenta previa. Birth weight centiles were defined as infants with birth weight below the 1st centile, below the 3rd centile, below the 5th centile, and those below the 10th centile, all of which were compared with infants at or above the 10th centile. These birth weight centiles were derived from the 1989–93 New Jersey singleton births.
Gestational age categories for defining preterm delivery included 20–23 weeks, 24–27 weeks, 28–30 weeks, and thereafter in 1-week intervals between 31 and 36 weeks, each of which was compared with pregnancies that ended after 36 weeks. To minimize errors in the assignment of gestational age, we restricted the analysis to pregnancies that ended between 20 and 44 weeks, a restriction shown to reduce the degree of error in gestational age.10 Gestational age and birth weight were obtained from birth certificate data.
Women whose medical records data contained an International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) diagnosis code 641.0 or 641.1 were considered to have had placenta previa. However, the severity and grades of previa were not recorded. Furthermore, we restricted this group only to women with a diagnosis of previa who had a cesarean delivery. This restriction was necessary to eliminate the probable and suspected cases of previa, as well as cases diagnosed as previa but which resolve later in pregnancy.
Sociodemographic and clinical variables that were considered as potential confounding variables included maternal age, parity (primipara versus multipara), race-ethnicity (white non-Hispanic, black non-Hispanic, black or white Hispanic), and infant gender. Socioeconomic and lifestyle factors included marital status (married or unmarried), level of maternal education (below 12, 12, 13–16, and over 16 years of completed schooling), cigarette smoking, alcohol and illicit drug use, and prenatal care utilization (inadequate, intermediate, adequate, and adequate plus). Categories of prenatal care utilization were defined using the Adequacy of Prenatal Care Utilization Index developed by Kotelchuck.11 These variables were obtained from the birth certificate. Information on payer status (type of medical insurance) was obtained from the mother’s primary medical insurance that was coded on the hospital Unified Billing Patient Summary at the time of delivery. Insurance type was grouped as Medicaid, Medicaid HealthStart, Self-pay, Managed Care, and Indemnity for the purposes of this study.
Information on medical complications of pregnancy that were considered as potential confounders included previous cesarean delivery (ICD-9-CM diagnosis code 654.2), diabetes (ICD-9-CM 250), preeclampsia (ICD-9-CM 642.4 to 642.5), chronic hypertension (ICD-9-CM 401 to 405 and 642.0 to 642.2), chronic renal disease (ICD-9-CM 646.2), asthma during pregnancy (ICD-9-CM 493), syphilis (ICD-9-CM 647), and abruptio placentae (ICD-9-CM 641.2). Information on congenital anomalies (ICD-9-CM 740 to 759) was also assessed.
Risks of fetal growth restriction and preterm delivery were compared among women with and without a diagnosis of placenta previa. Unadjusted odds ratios were computed as measures of effect. Odds ratios were derived from multivariable logistic regression models after adjusting for confounding variables. Potential confounders were retained in the models for adjustment if their presence changed the unadjusted relative risks by at least 10%. Because the two outcomes, degrees of fetal growth restriction and preterm delivery, were categorical with more than two levels, adjusted odds ratios (OR) with 95% confidence intervals (CI) were derived from fitting multinomial polytomous logistic regression models.12 We calculated the population attributable risk for degrees of fetal growth restriction and preterm delivery in relation to placenta previa.13 Finally, we also modeled birth weight, gestational age, and fetal growth ratio as continuous variables in a multivariable linear regression model. From these models, we derived the mean differences in these outcomes for women with and without placenta previa, after controlling for confounders. All statistical analyses were performed using the SAS 8.0 software, SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOP
ABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Women with placenta previa were more likely to be older and multiparous. The frequencies of maternal complications and congenital anomalies were more common among women with previa than those without this condition. The proportion of male fetuses and the proportion of smokers were also increased among the previa than the nonprevia group. Women diagnosed with placenta previa were more likely to be delivered earlier than those without placenta previa (Figure 1). By 36 weeks, 8% of women without placenta previa had delivered compared with a rate of almost 40% among those with placenta previa.
|
|
). Similarly, the mean fetal growth ratio was lower among previa cases compared with infants in the nonprevia group, indicating smaller fetal size of infants born to women with previa. Likewise, women with previa were delivered approximately 2.1 weeks earlier than those without this condition. The distribution of mean birth weight in the previa and nonprevia cohorts were, however, similar (Figure 2).
|
|
). Similarly, OR 1.24 (95% CI 1.17, 1.32) was noted for the analysis pertaining to moderately small infants (fetal growth ratio 0.75–0.84). Small for gestational age births defined using birth weight centiles (below 1, below 3, below 5, and below 10 centile) were also examined in relation to previa (Table 3). Among women diagnosed with placenta previa, the rate of small for gestational age (below 10 centile) births was 11.55%, compared with a rate of 9.9% in the nonprevia group, yielding an adjusted OR of 1.12 (95% CI 1.06, 1.18). The population attributable risk was 3.7% (fetal growth ratio below 0.75 and 0.75–0.85 combined) in relation to placenta previa.
|
). The risks of delivering before 37 completed weeks for women with and without placenta previa were 37.68% and 6.90%, respectively. The adjusted OR for delivering between 20 and 32 weeks in relation to previa ranged between 1.8 and 4.0. Thereafter, the relative risks progressively declined. Finally, the population attributable risk for delivering before 37 completed weeks in relation to placenta previa was 12%, implying that approximately one in eight preterm deliveries in this cohort were attributable to placenta previa.
|
|
DISCUSSION |
|---|
|
TOP
ABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The association between placenta previa and low birth weight has been fairly well established. However, the association between previa and fetal growth is, at best, equivocal. Prior studies have noted strong,21 weak,22 or an absence of association6,8 between placenta previa and fetal growth restriction. Low birth weight is an extremely heterogeneous index to assess fetal well-being because it combines prematurity (delivery before 37 weeks), fetal growth restriction, as well as other intrinsic fetal disorders such as chromosomal abnormalities and congenital malformations. Efforts to isolate these intrinsic pathways that lead to low birth weight would benefit from studying the individual components, namely, prematurity and growth restriction. In that regard, our study clearly demonstrates that the association between placenta previa and low birth weight is chiefly due to the influences of prematurity, and virtually very little due to impaired fetal growth. This might help explain the conflicting results from prior studies on placenta previa and growth restriction. The small attributable risk for growth restriction (3.7%) relative to that for preterm delivery (12%) in relation to placenta previa further confirms our observation that the association between placenta previa and low birth weight is chiefly due to prematurity. Despite the presence of a statistical association between placenta previa and fetal smallness observed in our study, the magnitude of these associations, from a clinical perspective, is likely to be of little significance.
Barr et al21 performed a continuous-wave Doppler study of umbilical arteries on 100 women with placenta previa who were matched with an equal number of controls (women with normally implanted placentas) on gestational age. Their results showed elevated placental vascular resistance among patients with previa compared with controls. Their study also showed an increased rate of fetal growth restriction in cases than controls. It is generally believed that the increased incidence of small for gestational age births among women with placenta previa is largely due to placental insufficiency, thought to arise because of decreased placental perfusion caused by suboptimal implantation site. Consequently, this leads to decreased nutrient transfer from the maternal to fetal circulation.2 In a small hospital-based case-control study of placenta previa, Wolf et al8 reported an absence of association between previa and small for gestational age births (4.1% among previa patients and 5.8% among controls). They concluded that the lack of association between placenta previa and small for gestational age births was the result of conservative management with maternal transfusions to maintain hematocrit levels, and bed rest to optimize placental perfusion.8 One could argue that if similar management protocols such as the one documented by Wolf et al8 were applied to all patients in our study, then the risk of small for gestational age births among women diagnosed with placenta previa would be expected to be even lower than those reported here.
The incidence of placenta previa is generally reported to range between three and seven per 1000 singleton pregnancies.14,17,19,23 We found an incidence of 5.0 per 1000 in this singleton population when cases were restricted to those delivered by cesarean. The incidence of previa in this cohort was 6.0 per 1000 births without the restriction. Inclusion of women with placenta previa but who delivered vaginally did not alter our conclusions (data not shown).
A few limitations of our study merit attention. The completeness and validity in the reporting of a diagnosis of previa on hospital discharge summaries using the ICD codes remain unknown. Although there is potential for some degree of misclassification of previa cases, this bias would have been nondifferential with respect to the outcomes examined. The reported associations in the presence of such a misclassification, if any, are likely to drive the OR toward the null value.24 Second, a small fraction of women may have had more than one pregnancy during the 5 years studied (1989–93), which violates the statistical assumption of independence in the regression models. This, however, would have very little impact on the CI although the effect measure (OR) would be unaffected.25 We were unable to correct for this nonindependence during statistical analysis because the linked vital statistics and hospital discharge summary data do not identify subsequent pregnancies to the same woman.
This large population-based cohort study comprising over half million singleton births enabled us to carefully examine the risks of fetal growth restriction in relation to placenta previa. Not only were we able to evaluate associations after adjustment for potential confounders, but we also examined degrees of fetal growth ratio, as well as risks of small for gestational age births based on birth weight centiles. The concordance in the association between placenta previa and growth restriction defined using two different indices (fetal growth ratio and birth weight centiles) further strengthens our conclusions. In summary, our study offers two important conclusions: Most of the association between placenta previa and low birth weight is due to prematurity, and very little to fetal smallness, and the statistical, but small differences in fetal smallness between women with placenta previa and nonprevia explain the conflicting results of prior studies.
|
Footnotes |
|---|
Received December 8, 2000. Received in revised form February 13, 2001. Accepted March 14, 2001.
|
REFERENCES |
|---|
|
TOP
ABSTRACTPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Green JR. Placenta previa and abruptio placentae. In: Creasy RK, Resnik R, eds. Maternal-fetal medicine. 3rd ed. Philadelphia: WB Saunders, 1994:602–19.
3. Naeye RL. Abruptio placentae and placenta previa: Frequency, perinatal mortality, and cigarette smoking. Obstet Gynecol 1980;55:701–4.
4. Varma TR. Fetal growth and placental function in patients with placenta previa. J Obstet Gynaecol Br Commonw 1973;80:311–5.[Medline]
5. Chapman MG, Furness ET, Jones WR, Sheat JH. Significance of ultrasound location of placental site in early pregnancy. Br J Obstet Gynaecol 1979;86:846–8.[Medline]
6. Gabert HA. Placenta previa and fetal growth. Obstet Gynecol 1971;38:403–6.
7. Comeau J, Shaw L, Marcell CC, Lavery PJ. Early placenta previa and delivery outcome. Obstet Gynecol 1983;61: 577–80.
8. Wolf EJ, Mallozzi A, Rodis JF, Egan JFX, Vintzileos AM, Campbell WA. Placenta previa is not an independent risk factor for a small for gestational age infant. Obstet Gynecol 1991;77:707–9.
9. Kramer MS, McLean FH, Olivier H, Willis DH, Usher RH. Body proportionality and head and length "sparring" in growth retarded neonates: A critical reappraisal. Pediatrics 1989;84:717–23.
10. Balcazar H. Mexican American, intrauterine growth retardation and maternal risk factors. Ethnicity Dis 1993;3: 169–75.[Medline]
11. Kotelchuck M. The adequacy of prenatal care utilization index: Its US distribution and association with low birth weight. Am J Pub Health 1994;84:1486–9.
12. Ananth CV, Kleinbaum DG. Regression models for ordinal responses: A review of methods and applications. Int J Epidemiol 1997;26:1323–33.
13. Eide GE, Gefeller O. Sequential and average attributable fractions as aids in the selection of preventive strategies. J Clin Epidemiol 1995;48:645–55.[Medline]
14. Williams MA, Mittendorf R. Increasing maternal age as a determinant of placenta previa: More important than increasing parity? J Reprod Med 1993;38:425–8.[Medline]
15. Handler AS, Mason ED, Rosenberg DL, Davis FG. The relationship between exposure during pregnancy to cigarette smoking and cocaine use and placenta previa. Am J Obstet Gynecol 1994;170:884–9.[Medline]
16. Ananth CV, Savitz DA, Luther ER. Maternal cigarette smoking as a risk factor for placental abruption, placenta previa, and uterine bleeding in pregnancy. Am J Epidemiol 1996;144:881–9.
17. Ananth CV, Smulian JC, Vintzileos AM. The association of placenta previa with history of cesarean delivery and abortion: A meta analysis. Am J Obstet Gynecol 1997;177: 1071–8.[Medline]
18. Demissie K, Breckenridge MB, Joseph L, Rhoads GG. Placenta previa: Preponderance of male sex at birth. Am J Epidemiol 1999;149:824–30.
19. Monica G, Lilja C. Placenta previa, maternal smoking and recurrence risk. Acta Obstet Gynecol Scand 1995;74: 341–5.[Medline]
20. Rasmussen S, Albrechtsen S, Dalkker K. Obstetric history and risk of placenta previa. Acta Obstet Gynecol Scand 2000;79:502–7.[Medline]
21. Barr HS, Platt LD, DeVore GR, Horenstein J. Fetal umbilical velocimetry for the surveillance of pregnancies complicated by placenta previa. J Reprod Med 1988;33:741–4.[Medline]
22. Neri A, Gorodesky I, Bahary C, Ovadia Y. Impact of placenta previa on intrauterine fetal growth. Isr J Med Sci 1980;16:429–32.[Medline]
23. Iyasu S, Saftlas AK, Rowley DL, Koonin LM, Lawson HW, Atrash HK. The epidemiology of placenta previa in the United States, 1979 through 1987. Am J Obstet Gynecol 1993;168:1424–9.[Medline]
24. Rothman K, Greenland S. Modern epidemiology. 2nd ed. Philadelphia: Lippincott Williams and Wilkins, 1998.
25. Liang K-Y, Zeger SL. Longitudinal data analysis using generalised estimating equations. Biometrika 1986;73: 13–22.
This article has been cited by other articles:
|
|
 |
|
  L. E. G. Kalanithi, J. L. Illuzzi, V. B. Nossov, Y. Frisbaek, S. Abdel-Razeq, J. A. Copel, and E. R. Norwitz Intrauterine Growth Restriction and Placental Location J. Ultrasound Med., November 1, 2007; 26(11): 1481 - 1489. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Getahun, Y. Oyelese, H. M. Salihu, and C. V. Ananth Previous Cesarean Delivery and Risks of Placenta Previa and Placental Abruption. Obstet. Gynecol., April 1, 2006; 107(4): 771 - 778. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Baboonian, D. A Smith, D. Shapland, G. Arno, B. Zal, J. Akiyu, and J. C. Kaski Placental infection with Chlamydia pneumoniae and intrauterine growth restriction Cardiovasc Res, October 15, 2003; 60(1): 165 - 169. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and vaginally (
).