This case series includes six consecutive cases of thanatophoric dysplasia (TD) diagnosed at ultrasound and confirmed by autopsy, which presented during the 1-year period from February 2007 to January 2008 at the National Center for Fetal Medicine (NCFM) in Trondheim, Norway. We recorded anatomical ultrasound findings, including evaluation of the brain, biometry, including biparietal diameter (BPD) and mean abdominal diameter (MAD, mean of anteroposterior and transverse diameters), karyotype, X-ray findings, observations at delivery and autopsy reports. The body weight and brain weight of the specimens at autopsy were compared with the corresponding weights in normal fetuses1. For the MAD value of each of the six TD individuals, we computed the mean and SD of the corresponding BPD values from a population-based registry at the NCFM, which comprised approximately 31 000 measurements in the relevant age range2. These values were used to compute Z-scores for each of the observed BPD/MAD ratios in the sample of TD fetuses. In the general population, the average BPD/MAD ratio was 1.05.
In all six cases, short limb dwarfism with TD type 1 was diagnosed (Table 1), at a mean gestational age of 19 + 4 (range, 19 + 0 weeks to 20 + 0) weeks. All had short long bones in all limbs, bowed femora and short ribs with bell-shaped thorax. There were no associated major anomalies.
Table 1. Pre- and postnatal findings in six cases of thanatophoric dysplasia
Year of first US
LMP-based age (days)
Femur length (mm)
Body weight (g)
Total lung weight (g)
Brain weight (g)
BPD, biparietal diameter; LMP, last menstrual period; MAD, mean abdominal diameter; TD 1, thanatophoric dysplasia type 1; US, ultrasound.
In contrast to the normal brain at 20 weeks' gestation (Figure 1), abnormal and deep transverse sulci in the temporal lobes could be imaged in all cases; these features were confirmed at autopsy (Figure 2 and Table 1). In three of the fetuses the brain was too autolytic for appropriate histological description. The others showed migrational disturbances, with disorganization of the cortex, heterotopia and polymicrogyria. Two autopsy reports described dysplastic hippocampus.
In the six TD fetuses, the mean BPD was 51.5 (range, 49–54) mm and mean MAD was 45 (range, 41–47) mm. These MAD values corresponded to MAD values in normal fetuses of the same age. The BPD/MAD ratio was 1.15 (range, 1.09–1.20). The average Z-score of the TD fetuses was 2.44 (range, 1.05–3.39). Using the very robust sign test to determine whether the ratios of TD fetuses were significantly higher than those of the general population, we obtained a P-value of 0.016.
At autopsy, the mean ratio of brain weight to body weight was 20.6% (range, 15.4–24.1%). The weights of the lungs were low.
TD is recognized as one of the more common sporadic lethal skeletal dysplasias (usually caused by dominant de-novo mutations), with reports of prevalence at birth ranging between 1:20 0003 and 1:60 0004. Autosomal recessive occurrence of TD is unusual but has been described5.
Specific sonographic signs can help to distinguish between lethal skeletal disorders and specify the diagnosis, for example identifying polydactyly in short-rib polydactyly syndromes. In TD, various brain anomalies have been described in postmortem examinations, especially a very typical form of medial temporal lobe dysplasia with significant abnormal gyration resulting in deep and transverse temporal sulci6, 7. Previously we presented data from the present ultrasound study, demonstrating the abnormal temporal lobe gyration in TD fetuses8. Two other case reports have shown similar brain features using ultrasound and magnetic resonance imaging9, 10. In contrast, the cortex of the temporal lobes is always smooth in normal fetuses at 20 weeks (Figure 1).
In our experience, the abnormal gyration typical of TD is expressed mainly at the inferior temporal lobe (Figure 2). Because the plane of the BPD cuts through the cavum septi pellucidi and the cerebral peduncles, and lies above the cerebellum, it lies relatively high above the inferior part of the temporal lobe, and abnormal gyration may be overlooked in this plane. The abnormal gyration is best imaged by the parasagittal or oblique plane (Figure 1a) that depicts the temporal lobe11, 12. Oblique sections that are more axially insonated to the inferior temporal lobe may also show this striking uneven pattern (Figures 2c and e).
The size of the head compared with the body was significantly larger in our TD fetuses than in normal fetuses. Among TD fetuses, we interpreted the increased size of the BPD relative to abdominal size as a sign of megalencephaly. This was confirmed at autopsy: normative weight data of fetuses and their organs can be used to establish and quantify the normal relationship of brain size to body size1; according to these data, the mean ratio of brain weight to body weight in normal fetuses at 19–20 weeks is 14.9%. The corresponding mean value of our TD cases was 20.6%.
Since 197113, the neuropathology of TD has been the subject of several case series and case reports, as summarized in a review on cerebral cortex malformation in TD by Hevner6. Megalencephaly and macroscopic abnormalities of the temporal lobe in the form of polymicrogyria, enlarged temporal lobe and deep and transverse temporal sulci are repeated autopsy findings14–16. In his review on TD, Hevner concluded that 100% of TDs present megalencephaly, hippocampal dysplasia and rudimentary dentate gyrus. Temporal lobe polymicrogyria is found in 97% of cases, and it is likely that 100% have enlarged temporal lobes and deep and transverse temporal sulci which are already present in the second trimester, by 18 weeks6.
Megalencephaly and the specific skeletal anomalous development in TD both involve a mutation in the fibroblast growth factor receptor 3 (FGFR3) gene, which has been mapped to chromosome band 4p16.37. FGFR3 belongs to a family of four genes (FGFR1–4), encoding receptors with tyrosine kinase activity. Dominant mutations in three members of the FGFR family (FGFR1–3) have been shown to account for two groups of skeletal disorders, namely short-limb dwarfisms and craniosynostoses17–19. TD can be divided into two subtypes, depending primarily upon whether the bone of the femur is curved (TD 1) or straight (TD 2)20, 21. Typically, TD 2 shows cloverleaf skull, but TD 1 can also present this skull anomaly. In addition to TD, anomalies in FGFR3 are involved in achondroplasia, hypochondroplasia, hypochondroplasia-like dysplasia and SADDAN (severe achondroplasia-developmental delay-acanthosis nigricans)4, 7.
It is now known that FGFR3 shows highly localized but changing patterns of expression throughout central nervous system (CNS) development22. Activation of FGFR3 selectively promotes growth of the caudolateral (occipitotemporal) cortex. This may explain the premature growth and sulcation of the occipitotemporal cortex in TD23. Thus, megalencephaly and the skeletal disorders in TD 1 and TD 2 are caused by the same gene defect. Recently, medial temporal lobe dysgenesis was also described in hypochondrodysplasia24. Such temporal lobe malformations have not been documented in achondroplasia or SADDAN.
In conclusion, abnormal deep transverse gyration of the temporal lobes is a well-known CNS anomaly that can be visualized by ultrasound and may be of additional aid in the diagnostic work-up of fetuses with dwarfism. In combination with short limbs, narrow chest and macrocephaly, temporal lobe dysplasia may be a pathognomonic sonographic sign for TD in mid-trimester fetuses with short-limb dysplasia. It is likely that temporal lobe dysplasia can be found in all fetuses with TD. In addition, the BPD/MAD ratio demonstrates that fetuses with TD have significantly different body proportions, with a relatively large BPD, than do normal fetuses.
Håkon Gjessing performed the statistical analyses and Nancy Lea Eik-Nes revised the text.