PRENATAL DIAGNOSIS
Prenat Diagn 2004; 24: 287–289. Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/pd.853
Fetal open spina bifida: a natural history of disease progression in utero Joseph R. Biggio*, Katharine D. Wenstrom and John Owen University of Alabama at Birmingham, Department of Obstetrics and Gynecology, Division of Maternal Fetal Medicine and Reproductive Genetics, Birmingham, Alabama Objective To determine the natural history of the prenatal development of ventriculomegaly and talipes in fetuses with open spina bifida. Study design All fetuses with isolated open spina bifida and managed at our center between January 1996 and March 2000 were retrospectively evaluated. Ultrasonographic images and reports were reviewed from examinations performed every 3 to 4 weeks from the time of diagnosis to delivery for lesion level and type, ventriculomegaly (defined as an atrial width of ≥10 mm), and lower extremity appearance. Results Of the 53 pregnancies identified, 20 (38%) were electively terminated. In the 33 ongoing gestations, the lesions ranged from lower thoracic to sacral; 79% were characterized as meningomyeloceles and 21% as myeloschises. Fifty-five percent (n = 18) had ventriculomegaly at diagnosis (early onset, mean gestational age at diagnosis 22 ± 5 weeks), 33% (n = 11) subsequently developed ventriculomegaly (late onset, mean 29 ± 6 weeks), and 12% (n = 4) had normal ventricle size at the last sonogram before birth (mean 38 ± 1 weeks). The ventricular size prior to delivery was significantly smaller with late-onset ventriculomegaly than with early-onset: 15 ± 4 mm versus 28 ± 10 mm, (p = 0.001). Only 6% (n = 2) had talipes at the initial sonogram, and 18% (n = 6) were subsequently determined to have talipes (mean 30 ± 6 weeks). Conclusion Most fetuses with open spina bifida develop ventriculomegaly, and the majority do so by 21 weeks’ gestation. Fetuses that develop ventriculomegaly later in gestation have less severe ventricular dilation at birth. In contrast, a minority of fetuses have congenital talipes, and because most cases develop after 20 weeks, they are not predicted by early midtrimester sonographic evaluation. Copyright 2004 John Wiley & Sons, Ltd. KEY WORDS:
neural tube defect; natural history; ultrasonography; ventriculomegaly; talipes
Neural tube defects are a major cause of childhood morbidity and mortality and complicate 1 to 2 per 1000 births (Botto et al., 1999; Milunsky, 1998). Anencephaly accounts for nearly half of all the cases and is invariably lethal. The severity of open spina bifida varies from large lesions that encompass much of the vertebral axis to small sacral lesions that involve only the meninges (Botto et al., 1999). The combination of maternal serum screening programs and sonographic findings have enabled the prenatal diagnosis of spina bifida in more than 90% of the cases (Milunsky, 1998). Although the prenatal diagnosis of spina bifida maximizes the reproductive options for women with affected fetuses, they must make difficult decisions with incomplete information. Fetuses with spina bifida frequently have ventriculomegaly at the time of diagnosis, but, if not present, the likelihood of developing this or other conditions, such as talipes, later in gestation is indeterminate. Prenatal surgical closure of the spinal defect is being investigated to prevent or retard the development of ventriculomegaly and hence reduce the need for postnatal shunt placement (Bruner et al., 1999; Tulipan et al., *Correspondence to: Joseph R. Biggio, Jr, University of Alabama at Birmingham, 619, 19th Street South, OHB 450, Birmingham, AL 35249-7333. E-mail:
[email protected]
Copyright 2004 John Wiley & Sons, Ltd.
1999). Several authors have also suggested that the intrauterine milieu can exacerbate spinal cord neurologic damage, and that in utero closure of the spinal defect might minimize the neurologic damage (Tulipan et al., 1999). However, it is currently unclear which fetuses would benefit from prenatal therapy, because the natural history and the sonographic signs of progressive neurologic damage (if any) have not been well characterized. Until it is possible to identify, early in pregnancy, those fetuses at highest risk of developing ventriculomegaly or talipes, optimal selection of candidates for fetal surgery will remain problematic. Such information would be valuable not only to women contemplating their reproductive options but also to investigators conducting clinical trials of in utero surgery. We therefore sought to describe the natural history of the prenatal development of ventriculomegaly and talipes in fetuses with open spina bifida. MATERIAL AND METHODS With the approval of the institutional review board, we identified all pregnancies complicated by fetal open spina bifida managed prenatally and postnatally at our institution from January 1996 through March 2000. Fetuses with other organ system malformations or aneuploidy were excluded. We also excluded three fetuses Received: 6 November 2003 Revised: 17 January 2003 Accepted: 27 January 2004
288
J. R. BIGGIO ET AL.
that underwent in utero closure of their spinal defects because this intervention may have altered the natural history (Sutton et al., 1999; Tulipan et al., 1999). All sonographic examinations were performed by one of four maternal–fetal medicine physicians with expertise in prenatal ultrasonography or a maternal–fetal medicine fellow under the direct supervision of one of these four physicians. All examinations were conducted using either an Accuson 128 XP with a 5-MHz transducer or a GE Logiq 700 with a multifrequency transducer. Prior to August 1997, static images were recorded on a thermal photo paper and stored in the patient chart with the ultrasound text report. After August 1997, images and text reports were archived in a digital storage system. In every second-trimester examination, static images routinely included the intracranial anatomy, including a measurement of the maximum width of the atria of the lateral ventricle, the spine in the sagittal and the transverse planes, and the lower extremities. All ongoing pregnancies were assessed every 3 to 4 weeks until delivery. The static images and text reports of all sonographic examinations were reviewed. The lesion level was defined by the highest vertebral level at which dysraphism was visualized. The lesion type was classified as meningocele, myelomeningocele, or myeloschisis. Myeloschisis was defined as a dysraphism in which the spinal cord and vertebral arches and pedicles were widely splayed dorsally and were part of the myelomeningocele wall (Romero et al., 1988). Ventriculomegaly was defined as a maximum atrial width of at least 10 mm. Talipes was diagnosed if images of the lower extremities demonstrated a transverse view of the metatarsals in the same plane as a longitudinal view of the tibia. Operative reports of closure of the spinal lesion and postnatal radiographs were reviewed to confirm the lesion type and the location and the presence or the absence of talipes. Statistical analysis was performed using SAS (SAS Institute, Cary, NC). Intergroup comparisons used chisquare and Fisher’s exact tests, as appropriate, for categorical variables. The Kruskal–Wallis test was used to compare continuous variables. P ≤ 0.05 was considered statistically significant. RESULTS We identified 53 fetuses with isolated open fetal spina bifida managed at our referral center from January 1996 to March 2000. Twenty women (38%) opted for a pregnancy termination. The lesion level and type are
Table 1—Lesion level and type of fetuses with open spina bifida All Terminated Continuing pregnancies pregnancies pregnancies (n = 53) (n = 20) (n = 33) Thoracic Myeloschisis Myelomeningocele L1–L3 Myeloschisis Myelomeningocele L4–L5 Myelomeningocele Sacral Myelomeningocele Meningocele Myeloschisis Myelomeningocele Meningocele
29% (16) 69% (11) 31% (5) 49% (26) 8% (2) 92% (24) 14% (7) 100% (7) 8% (4) 75% (3) 25% (1) 24% (13) 73% (39) 2% (1)
25% (5) 100% (5) — 70% (14) 7% (1) 93% (13) 0 — 5% (1) — 100% (1) 30% (6) 65% (13) 5% (1)
33% (11) 54% (6) 45% (5) 36% (12) 8% (1) 92% (11) 21% (7) 100% (7) 9% (3) 100% (3) — 21% (7) 79% (26)
displayed in Table 1. Characteristics of the sonographic findings at diagnosis are listed in Table 2. Ventriculomegaly was present on the initial sonographic evaluation in 65% (13 of 20) of pregnancies that were terminated and in 55% (18 of 33) of the continuing pregnancies (mean GA 22 ± 5 weeks, p = 0.45). In the continuing pregnancy group, an additional 33% (11 of 33) developed ventriculomegaly later in gestation, at a mean gestational age at diagnosis of 28 ± 6 weeks. Fetuses with evidence of ventriculomegaly at the first sonographic examination had significantly larger ventricles prior to delivery than those fetuses diagnosed with ventriculomegaly later (29 ± 10 mm vs 15 ± 4 mm, p < 0.001). Only 12% (4) of fetuses did not develop ventriculomegaly prenatally, as determined by their final sonographic evaluation prior to birth. In a subanalysis of the 23 fetuses with an initial ultrasound prior to 25 weeks, similar findings were observed. The ventricular size just before delivery was 26 ± 11 mm in fetuses with ventriculomegaly diagnosed prior to 25 weeks versus 15 ± 4 mm in those diagnosed later (p = 0.005). The higher spinal lesion levels were associated with larger ventricular widths (p = 0.003). At the initial sonographic examination, talipes was diagnosed in 6% (2 of 33) of continuing pregnancies and was diagnosed later in gestation in an additional 18% (6 of 33). The mean gestational age at diagnosis was 30 ± 6 weeks. Seventy-six percent (76%) of fetuses had lower extremities that appeared normal at all prenatal sonographic examinations. At birth, only two fetuses thought to have normal extremities prenatally were found to have talipes; all fetuses diagnosed with talipes
Table 2—Characteristics at the ultrasound which diagnosed open spina bifida
Median (range) gestational age at diagnosis Mean ventricular width at diagnosis Ventriculomegaly present at diagnosis Clubfoot present at diagnosis Copyright 2004 John Wiley & Sons, Ltd.
All pregnancies (n = 53)
Terminated pregnancies (n = 20)
Continuing pregnancies (n = 33)
20 weeks (16–35) 12 ± 2 mm 59% (31) 13% (7)
19 weeks (17–21) 11 ± 3 mm 65% (13) 25% (5)
20 weeks (16–35) 13 ± 2 mm 55% (18) 6% (2) Prenat Diagn 2004; 24: 287–289.
289
SPINA BIFIDA NATURAL HISTORY
prenatally had such a lesion at birth. Lesion level did not affect the likelihood of developing talipes (p = 0.15). COMMENT In this series, the majority of fetuses with open spina bifida had developed ventriculomegaly by 21 weeks’ gestation. With serial sonographic examinations, an additional one-third of fetuses developed ventriculomegaly prior to delivery. There was no gestational age beyond which ventriculomegaly did not develop, as it was identified as late as 38 weeks. However, in the fraction of fetuses diagnosed with ventriculomegaly after 24 weeks, the ventricular enlargement tended to be less severe than in those found to have ventriculomegaly at the first evaluation. In contrast, only a small percentage of fetuses (13%) had talipes at the time of the initial ultrasound. Although an additional 18% of fetuses in continuing pregnancies developed talipes prior to delivery, ultimately only a quarter of all fetuses had congenital clubfoot. Clubfoot was not an inevitable outcome for every fetus with open spina bifida. This study is the largest series reported to date in which fetuses with open spina bifida were followed with serial sonographic examination. Bannister et al. examined 21 fetuses and found that 18 had ventriculomegaly at diagnosis (16–23 weeks), but reported serial measurements on only 5 (Bannister et al., 1998). In these five, the ventricular measurements progressively enlarged. Although Babcook et al. did not study fetuses serially, these authors obtained cross-sectional data that suggested that the prevalence of ventriculomegaly increases progressively during gestation from 44% at 24 weeks or earlier to 94% later in gestation (Babcook et al., 1994). Neither report commented on the development of talipes. Our study confirms these earlier reports and adds information that may be useful in counseling patients. Although fetuses without ventriculomegaly at 24 weeks will more than likely develop ventriculomegaly at some time during gestation, patients should be counseled that it tends to be less severe than if the ventriculomegaly had been diagnosed earlier in the gestation. Because more than half of fetuses had evidence of ventriculomegaly at the diagnosis of open spina bifida, the impact of fetal surgery on the ultimate need for shunt placement may be limited and should be addressed in future clinical trials of fetal surgery.
Copyright 2004 John Wiley & Sons, Ltd.
Patients can be informed that only a minority of fetuses develop talipes prenatally. We did not find a significant association between the lesion level and the development of talipes, but the low number of fetuses with this deformation limited our power to detect a clinically significant relationship. Whether the development of talipes correlates with the neurologic damage and to what extent this can be prevented by fetal surgery is a topic for further research. Because our series of ongoing pregnancies included severely affected cases, we were able to study disease evolution in fetuses affected with these types of lesions, and the inability to monitor anatomic changes in the terminated cases did not prevent us from studying the natural history in severely affected fetuses. Although this study describes the evolution of ventriculomegaly and talipes in fetuses with open spina bifida, it is not clear why progressive changes occur in some fetuses but not in others. Unfortunately, that is precisely the information needed to optimize the utility of fetal surgery or other prenatal therapeutic procedures. A prospective study with serial sonographic examinations of a large number of fetuses may help in addressing these issues. In the interim, information from this study can be used to assist patients in making reproductive decisions. REFERENCES Babcook C, Goldstein R, Barth R, Damato N, Callen P, Filly R. 1994. Prevalence of ventriculomegaly in association with myelomeningocele: correlation with gestational age and severity of posterior fossa deformity. Radiology 190: 703–707. Bannister C, Russell S, Rimmer S. 1998. Pre-natal brain development of fetuses with a myelomeningocele. Eur J Pediatr Surg 8(Suppl. 1): 15–17. Botto L, Moore C, Khoury M, Erickson J. 1999. Neural-tube defects. N Engl J Med 341: 1509–1519. Bruner J, Tulipan N, Paschall R, et al. 1999. Fetal surgery for myelomeningocele and the incidence of shunt-dependent hydrocephalus. JAMA 282: 1819–1825. Milunsky A. 1998. Maternal serum screening for neural tube and other defects. In Genetic Disorders and the Fetus: Diagnosis, Prevention, and Treatment, Milunsky A (ed.). The Johns Hopkins University Press: Baltimore; 635–701. Romero R, Pilu G, Jeanty P, Ghidini A, Hobbins J. 1988. Neural tube defects. In Prenatal Diagnosis of Congenital Anomalies, Appleton & Lange: Norwalk; 36–43. Sutton L, Adzick N, Bilaniuk L, Johnson M, Crombleholme T, Flake A. 1999. Improvement in hindbrain herniation demonstrated by serial fetal magnetic resonance imaging following fetal surgery for myelomeningocele. JAMA 282: 1826–1831. Tulipan N, Bruner J, Hernanz-Schulman M, et al. 1999. Effect of intrauterine myelomeningocele repair on central nervous system structure and function. Pediatr Neurosurg 31: 183–188.
Prenat Diagn 2004; 24: 287–289.
