DOI: 10.1002/pd.4396
ORIGINAL ARTICLE
Prenatal diagnosis of congenital femoral deficiency and fibular hemimelia† Christof Radler1*, Abigail K. Myers2, Renee J. Hunter2, Pedro P. Arrabal3 and John E. Herzenberg2 1
Department of Pediatric Orthopaedics, Orthopaedic Hospital Speising–Vienna, Vienna, Austria International Center for Limb Lengthening, Rubin Institute for Advanced Orthopedics, Sinai Hospital of Baltimore, Baltimore, MD, USA 3 Louis and Henrietta Blaustein Women’s Health Center, Department of Obstetrics and Gynecology, Sinai Hospital of Baltimore, Baltimore, MD, USA *Correspondence to: Christof Radler. E-mail: [email protected] † Study conducted at the Rubin Institute for Advanced Orthopedics at Sinai Hospital of Baltimore in Baltimore, MD, USA. 2
ABSTRACT Objectives Routine ultrasonography can detect congenital femoral deficiency (CFD) and fibular hemimelia (FH), but prenatal detection rate and its relation to deformity severity have never been reported. Whether mothers prefer prenatal diagnosis is also unknown. We aimed to determine whether mothers prefer prenatal diagnosis, to report detection rates for CFD and/or FH, and to correlate detection rates to severity of limb shortening. Methods Surveys were mailed to 171 mothers who gave birth to children with CFD/FH between 2000 and 2008. Bilateral femoral and tibial lengths were measured on postnatal radiographs. We calculated corresponding femoral/ tibial lengths at gestational weeks 20 and 30. Results Sixty-five surveys were returned, and 56 radiographs were reviewed. Most mothers (63%) preferred prenatal diagnosis as it enables prenatal counseling. Congenital limb shortening was detected in 24 cases (37%) and was not detected in 41 cases (63%). Detection rate was 52% (12 of 23) in CFD cases, 23% (three of 13) in FH cases, and 30% (six of 20) in combined cases. CFD cases with severe shortening had a higher detection rate. Conclusions Ultrasonographers should measure both femoral and tibial lengths. Unilateral shortening should result in pediatric orthopedic consultation to estimate limb-length discrepancy at maturity and discuss treatment. © 2014 John Wiley & Sons, Ltd.
Funding sources: None Conflicts of interest: None declared
INTRODUCTION Congenital femoral deficiency (CFD) and fibular hemimelia (FH) are rare longitudinal defects of the lower limb. CFD presents with mild femoral shortening to severe dysplasia with absence or pseudarthrosis of the femoral head and/or neck area and hypoplasia of the lateral femoral condyle.1–4 FH presents with hypoplasia or aplasia of the fibula with shortening and malalignment of the tibia, distinct joint deformities at the knee and ankle joint, and ray deficiencies.1,3,5 In the majority of cases, both conditions are unilateral.1,2,4,5 Prenatal diagnosis of CFD and FH was first reported in the late 1980s and early 1990s.6–8 Whereas some studies have evaluated the prenatal diagnosis of skeletal dysplasias in large case series,9–11 the detection rate for congenital longitudinal deficiencies of the lower limb has not been reported. Early consultation with an expert would help promote acceptance of the deformity and allow discussion of options.12– 14 Although amputation and/or prosthetic fitting has traditionally been recommended for most cases,15–17 reconstructive treatment Prenatal Diagnosis 2014, 34, 940–945
can be performed with good functional results.3,18,19 This may involve multiple surgical procedures starting at approximately age 2 years to the end of growth depending on the severity of the shortening.3 We aimed to determine whether mothers prefer prenatal diagnosis, report the detection rates for CFD and/or FH, and correlate detection rates to the severity and shortening of the involved limb.
METHODS The hospital’s institutional review board approved this study, and the study was conducted in accordance with the Helsinki Declaration of 1975 (revised in 1983). Patients were identified retrospectively by diagnosis of CFD and FH using the hospital’s computer database. Surveys were mailed to 171 mothers who gave birth to a child with CFD and/or FH between January 2000 and December 2008 and presented for consultation and/or treatment. The mothers were sent the survey, a letter explaining the study, © 2014 John Wiley & Sons, Ltd.
Prenatal diagnosis
and a consent form. After obtaining consent, the medical records were reviewed to assess exclusion criteria. Exclusion criteria were underlying syndrome or genetic abnormality other than CFD and FH, multiple gestations, a family history of a limb defect, and birth outside of the USA. The survey consisted of three questions: (1) Did you have any ultrasounds (sonograms) before your baby was born? (2) Did any of your ultrasound tests show a short leg? (3) When would you have preferred to find out about your baby’s congenital leg shortening? Question 3 was answered by selecting either ‘before birth’ or ‘after birth’. Another section on the survey asked participants to share general comments. Responses and comments were analyzed on the basis of prenatal or postnatal diagnosis and year of birth. The rates of true-positive and false-negative diagnoses were also determined. Postnatal radiographs were analyzed of the patients whose surveys were returned. The severity of the deformity as found in the most recent preoperative bilateral long leg radiographs was graded using the classification systems of Paley3 and Pappas4 for CFD and the classification of Achterman and Kalamchi5 for FH cases. The Pappas classification4 for CFD describes types I to IX, with type I being the most severe. The Paley classification3 is more treatment based (i.e., surgical necessities and options) and is organized into types 1 to 4, with 4 being the most severe. The Achterman and Kalamchi classification5 is a descriptive and morphologic system that consists of types I, IIa, and IIb, with IIb being the most severe. Cases of FH-associated ray deficiencies and/or foot deformities were recorded. The femoral length, tibial length, and overall limb length (i.e., femur and tibia) of both limbs were measured. From these measurements, the limb-length discrepancy (LLD) and discrepancies of femoral length and tibial length at the time of the radiograph were determined. In the next step, the LLD and femoral and tibial length at birth were calculated using the lower-limb multiplier20 in a reversed way. In a second step, the prenatal multiplier21 was used on this data to calculate the LLD and femoral and tibial length at the time when secondtrimester (20 weeks) and third-trimester (30 weeks) ultrasounds are usually obtained. These times were chosen according to recommendations for standard second- and third-trimester ultrasonographic examinations from the American College of Radiology, the American College of Obstetricians and Gynecologists, and the American Institute of Ultrasound in Medicine.22–24 Additionally, LLD at maturity was predicted using the lower-limb multiplier.20 The detection rate was analyzed according to the amount of LLD and femoral and tibial shortening. For descriptive statistics, categorical variables are reported in percentages and numeric variables are given as mean with minimum, maximum, and standard deviations reported. Subgroups were defined and compared using a twotailed Fisher exact test with p < 0.05 considered significant (IBM SPSS Statistics 19.0, Somers, NY, USA).
RESULTS Surveys were sent to 171 mothers. Sixty-five surveys were completed and returned (38% response rate). All mothers had at least one prenatal ultrasound. A majority of respondents (41 mothers, 63%) preferred prenatal diagnosis, fewer respondents Prenatal Diagnosis 2014, 34, 940–945
941
(21 mothers, 32%) preferred to know after birth, and a small minority (three mothers, 5%) were undecided. Of the patients who had a positive ultrasound, 88% of the respondents (21 mothers) wanted a prenatal diagnosis and 13% (three mothers) wanted a postnatal diagnosis. Of the 41 patients who had a negative ultrasound, 49% of the respondents (20 mothers) would have wanted a prenatal diagnosis, 44% (18 mothers) did not want a prenatal diagnosis, and 7% (three mothers) were undecided. Overall, more mothers (21 of 24 mothers, 88%) who received a positive prenatal diagnosis reported wanting to know before birth than those who did not receive a prenatal diagnosis (20 of 41 mothers, 49%). We statistically tested the hypothesis that more mothers who had a positive ultrasound wanted the prenatal diagnosis than those who had a false-negative ultrasound using a Fisher exact test and found a p-value of 0.006 supporting this hypothesis. Comments on the survey form showed that 20 mothers who had or wanted to have the prenatal diagnosis appreciated the time to prepare, to research the condition and different treatment options, and to become emotionally accustomed. Some mothers wished for more information at the time of prenatal diagnosis (seven mothers). Mothers who preferred a postnatal diagnosis feared that a prenatal diagnosis would have affected their experience during pregnancy (11 mothers). During routine ultrasonography, congenital limb shortening was detected (true positive) in 24 cases (37%) and was not detected (false negative) in 41 cases (63%). For infants born between the years 2000 and 2003, the overall detection rate was 38% (12 of 32 cases) versus 36% (12 of 33 cases) for those born between 2004 and 2008. The detection rate for those two time periods was not significantly different (p = 1 using a two-tailed Fisher exact test). We were able to review the preoperative radiographs of 56 of the 65 patients whose mothers returned the survey. The mean age of the patients when the radiographs were obtained was 2.5 ± 1.6 years (range, 6 months to 7 years 6 months). We analyzed the radiographs of 31 male and 25 female patients with the left side being affected in 27 cases and the right in 29 cases. There was no significant difference regarding the detection rate for left versus right side [eight of 27 (30%) vs 13 of 29 (45%); p = 0.279 using a two-tailed Fisher exact test]. Of the 56 cases, CFD was present in 23 cases (41%), FH in 13 cases (23%), and a combination of CFD and FH in 20 cases (36%). Table 1 shows the classification of the CFD cases according to the Pappas4 and Paley3 classification and of the FH cases according to the Achterman and Kalamchi5 classification. We found a detection rate of 52% (12 of 23 cases) in CFD cases, a detection rate of 23% (three of 13 cases) in FH cases, and a detection rate of 30% (six of 20 cases) in combined cases. The detection rate between CFD and FH (p = 0.159), between CFD and combined cases (p = 0.216), and between FH and combined cases (p = 1) was not significantly different using a two-tailed Fisher exact test. A ray deficiency (three-ray or four-ray foot) was found in 11 of the 13 FH cases and in five of the 20 cases presenting with combined FH and CFD. Of four FH cases with a foot deformity, 50% (two of four cases) were diagnosed prenatally. For all FH and combined CFD and FH cases with ray deficiencies, the detection rate was only 25% (four of 16 cases). © 2014 John Wiley & Sons, Ltd.
C. Radler et al.
942
Table 1 Classification of the patients whose radiographs were available for analysis (n = 56) Pappas classification of CFD4 Classification
Paley classification of CFD3
Classification of FH according to Achterman and Kalamchi5
Number of cases
Classification
Number of cases
Classification
Number of cases
II
3
1a
28
1a
14
III
6
1b
2
1b
2
IV
4
1c
3
2
17
V
1
2a
6
VI
1
2b
2
VII
7
3a
2
VIII
3
3b
0
IX
18
CFD, congenital femoral deficiency; FH, fibular hemimelia.
Cases of CFD and cases of combined CFD and FH that had a Pappas classification4 of VII to IX showed a detection rate of 32% (nine of 28 cases), whereas cases classified Pappas II to VI had a detection rate of 60% (nine of 15), which was not a significant difference (p = 0.109 using a two-tailed Fisher exact test). Cases classified as Paley3 type 1a had a detection rate of 36% (ten of 28 cases), whereas types 1b, 1c, 2a, 2b, and 3a had a detection rate of 53% (eight of 15 cases), a difference which again was not significant (p = 0.338). Cases with FH type 1a or 1b according to Achterman and Kalamchi5 had a detection rate of 13% (one of eight cases) versus 40% (two of five cases) for type 2, a difference which was not significant (p = 0.550 using a two-tailed Fisher exact test). Table 2 presents the predicted limb, femoral, and tibial length discrepancies at birth, gestational week 30, and gestational week 20.20,21 Table 3 shows the detection rates for the three diagnostic groups (i.e., CFD, FH, and combined CFD and FH) on the basis of limb, femoral, and tibial discrepancies at gestational week 30. The predicted lengths of the short and normal femora, tibiae, and limbs (i.e., femur plus tibia) at birth, gestational week 20, and gestational week 30 as well as the detection rate according to LLD at maturity are shown in Tables 4 and 5, which is available online as supporting information.
DISCUSSION Although rare, CFD and FH can lead to significant morbidity postnatally. Prenatal diagnosis of these conditions can allow time for mothers to learn about treatment options, such as limb reconstruction or amputation. Most mothers preferred prenatal diagnosis, and some mothers wanted to receive more information at the time of diagnosis. We found a detection rate of 52% in CFD cases (12 of 23), 23% in FH cases (three of 13), and 30% in combined cases (six of 20). Our detection rate for CFD cases was more than double the detection rate in FH cases. This might be because the femoral length is usually measured and used to detect skeletal dysplasia, intrauterine growth restriction, or Down syndrome.9,25–30 These are higher rates of detection than those reported in published literature. The RADIUS31 study included 15 530 low-risk pregnant women who underwent screening ultrasound and found five cases of limb-reduction deficiencies with two Prenatal Diagnosis 2014, 34, 940–945
being detected (40% detection rate). However, there is no information as to whether these deficiencies were severe or were transverse or longitudinal. The authors of the EUROSCAN study group13 reported a detection rate for limb-reduction deficiencies of 36% (89 of 250) dropping to 25% (34 of 138) for isolated cases without associated deformities. In their case series with 138 isolated limb-reduction deficiencies, a longitudinal defect with absence or severe hypoplasia of a lateral part of the limb was found in 12 cases with only one being diagnosed prenatally (8% detection rate). In a further study, the sensitivity for isolated limb-reduction deficiencies was only 4% (two of 49),32 whereas an extended study of the same group showed an increase of the detection rate for isolated cases from 2% (one of 43) to 13% (two of 15) from the period 1979 to 1988 versus 1989 to 1992.33 Our study found an overall detection rate for infants born between 2000 and 2003 of 38% (12 of 32 cases) versus 36% (12 of 33 cases) for those born between 2004 and 2008, indicating no increase of the detection rate between those two recent time periods. Kevern and coworkers12 evaluated prenatal diagnostic detection rates in a cohort of 26 203 infants. Of 60 patients with a limb abnormality, only 15 were detected prenatally resulting in a 25% detection rate. Subdividing the cases into major and minor abnormalities, the detection rate was 33% (13 of 39 cases) and 10% (two of 21 cases), respectively. In our series, the detection rate in combined CFD with FH cases with more severe shortening (LLD > 2 cm) was double (40%, four of ten) compared with less severe shortening (20%, two of ten, LLD < 2 cm). For isolated CFD, we found a higher detection rate when femoral discrepancy was more than 2 cm (58%, five of 12) versus less than 2 cm (46%, six of 11). However, the rate of detection for FH cases in our series did not increase as severity increased. Goncalves and Jeanty27 found that fetuses with mild femoral shortness were difficult to differentiate from mild shortening owing to skeletal dysplasia. In their series, only two of 139 cases had CFD, and two had femur–fibular–ulna complex. The two CFD cases were within the group showing mild shortening defined as a length greater than 80% of the mean for gestational age. Using this delineator, mild shortening was found in 12 of the 20 cases of combined CFD and FH with a detection rate of 25.0% (three of 12 cases) in our series. Mild cases of CFD and FH show only mild shortening prenatally © 2014 John Wiley & Sons, Ltd.
Prenatal diagnosis
943
Table 2 Limb, femoral, and tibial length discrepancies are summarized for the different diagnostic subgroups at birth, gestational week 30, and gestational week 20 Diagnosis Combined CFD and FH
CFD
FH
Parameter
N
Minimum (cm)
Maximum (cm)
Mean (cm)
SD (cm)
LLD at birth
20
1.03
8.18
3.55
2.11
Femoral discrepancy at birth
20
0.27
6.10
2.06
1.84
Tibial discrepancy at birth
20
0.18
3.48
1.51
0.92
LLD at week 30
20
0.63
5.00
2.17
1.29
Femoral discrepancy at week 30
20
0.17
3.72
1.26
1.12
Tibial discrepancy at week 30
20
0.11
2.13
0.92
0.56
LLD at week 20
20
0.36
2.85
1.24
0.73
Femoral discrepancy at week 20
20
0.09
2.12
0.72
0.64
Tibial discrepancy at week 20
20
0.06
1.21
0.53
0.32
LLD at birth
23
0.91
6.50
3.87
1.76
Femoral discrepancy at birth
23
0.97
6.31
3.73
1.70
0.66
0.16
0.20
3.97
2.36
1.08
Tibial discrepancy at birth
23
0.05
LLD at week 30
23
0.55
a
Femoral discrepancy at week 30
23
0.59
3.86
2.28
1.04
Tibial discrepancy at week 30
23
0.03a
0.41
0.09
0.13
LLD at week 20
23
0.32
2.26
1.35
0.61
Femoral discrepancy at week 20
23
0.34
2.20
1.30
0.59
0.23
0.05
0.07
3.48
1.84
1.11
0.91
0.25
0.30
3.20
1.61
2.00
Tibial discrepancy at week 20
23
0.02
LLD at birth
13
0.39
a
a
Femoral discrepancy at birth
13
0.22
Tibial discrepancy at birth
13
0.42
LLD at week 30
13
0.24
2.13
1.13
0.68
Femoral discrepancy at week 30
13
0.13a
0.56
0.15
0.18
Tibial discrepancy at week 30
13
0.26
1.95
0.98
0.61
LLD at week 20
13
0.14
1.21
0.64
0.39
Femoral discrepancy at week 20
13
0.08a
0.32
0.09
0.10
Tibial discrepancy at week 20
13
0.15
1.11
0.56
0.35
20
21
The LLD and femoral and tibial lengths at birth were calculated using the lower-limb multiplier. The prenatal multiplier was used with these data to calculate the LLD and femoral and tibial length at the time when second-trimester (20 weeks) and third-trimester (30 weeks) ultrasounds are usually obtained. CFD, congenital femoral deficiency; FH, fibular hemimelia; LLD, limb-length discrepancy; SD, standard deviation. a Negative values indicate that the specific bone was longer on the side with the short extremity.
and fall into the group of cases with a length greater than 80% of the mean for gestational age. This group can be especially hard to detect and to delineate from normal and other differential diagnoses, such as intrauterine growth restriction or skeletal dysplasia. Eight cases were in the group with femoral shortness between 30% and 80% of the mean with a detection rate of 38% (three of eight cases). In the subgroup of isolated CFD, we had six cases in the more than 80% of the mean group (detection rate 50%; three of six cases), 16 cases in the 30% to 80% group (detection rate 50%; eight of 16 cases), and one had a femoral length of less than 30% of the mean with a detection rate of 100% (one of one case). The cases with isolated FH (n = 13) had six cases in the above 80% group (detection rate 0%; zero of six) and seven in the 30% to 80% of the mean group (detection rate 43%; three of seven). Taking the percentiles as a threshold and using the multiplier formulae,20,21 we found that 22 of 23 femoral lengths at week 30 were below the fifth percentile in the CFD group. In the CFD Prenatal Diagnosis 2014, 34, 940–945
with FH group (n = 20), 11 femoral lengths and 17 tibial lengths were below the fifth percentile, whereas all the FH cases showed tibial lengths under the fifth percentile. Associated foot deformities in the FH group increased the detection rate from 23% to 50% (two of four). This is not surprising as the detection rate for clubfoot is reported to be as high as 81%.34,35 However, our cases with associated ray deficiencies (n = 16) had a detection rate of only 25%. According to Jeanty and Kleinman,7 CFD should be differentiated immediately from thanatophoric dysplasia, achondroplasia, achondrogenesis, short limb polydactyly, chondroectodermal dysplasia, campomelic dysplasia, and osteogenesis imperfecta as in all these disorders the findings are bilateral. They conclude that a normal fetus with only one short femur has virtually no other differential diagnosis than CFD. There are several biases in our study. Owing to the low incidence of these conditions, we were unable to report a greater number of cases. As the study was conducted at a © 2014 John Wiley & Sons, Ltd.
C. Radler et al.
944
Table 3 Length discrepancies at week 30 and the detection rate according to the severity of the shortening for the different diagnostic groups Diagnosis Combined CFD and FH
CFD
FH
Parameter LLD < 2 cm
N
True positive (number of cases)
10
False negative (number of cases)
2
8
Detection rate (%) 20.0
LLD > 2 cm (up to 5 cm)
10
4
6
40.0
Femoral discrepancy < 1 cm
13
3
10
23.1
Femoral discrepancy > 1 cm (up to 3.72 cm)
7
3
4
42.9
Tibial discrepancy < 1 cm
11
4
7
36.4
Tibial discrepancy > 1 cm (up to 2.13 cm)
9
2
7
22.2
Femoral discrepancy < 2 cm
11
5
6
45.5
Femoral discrepancy > 2 cm (up to 3.9 cm)
12
7
5
58.3
Tibial discrepancy < 1 cm
8
2
6
25.0
Tibial discrepancy > 1 cm (up to 1.95 cm)
5
1
4
20.0
p-value 0.628
0.613
0.642
0.684
1
CFD, congenital femoral deficiency; FH, fibular hemimelia; LLD, limb-length discrepancy.
tertiary referral center specializing in limb lengthening and reconstruction, cases in which the family decided on prosthetic management or amputation may not present to the center, resulting in a referral bias. Although the survey response rate of 38% (65 of 171) seems satisfactory, a nonresponse bias exists. In addition, secondary to the retrospective nature of data collection, we cannot account for cases in which fetal death may have occurred or mothers may have chosen pregnancy termination. This may have resulted in significant bias because mothers who chose to terminate pregnancy were not included in this study, which means that the most severe cases of prenatal CFD and FH may have been excluded. Only a few cases of termination of pregnancy in an isolated case of CFD have been reported in the literature.10,36 However, the true frequency of termination of pregnancy for isolated CFD or FH might be higher. The survey questions themselves introduce some bias especially as they are asked retrospectively. We analyzed only radiographs of the patients whose surveys were completed and returned, which resulted in a low sample size. Measuring the radiographs introduces additional bias. The child has to stand or lie still, and the knee joint must be fully extended to allow accurate length measurements on the radiographs. When full knee extension was doubtful, lateral view radiographs were additionally analyzed. All radiographs were measured by the first author to minimize interobserver variance. Lastly, the use of the prenatal multiplier has some inherent error rate. That being said, the mean values for the long (normal side) femora and tibiae at the 30th and 20th weeks of gestation that were calculated using the multiplier and the prenatal multiplier20,21 correspond very well with the values reported in literature.37,38 For the normal tibiae, we calculated a mean tibial length at week 30 of 4.73 cm (SD 0.34) and at week 20 of 2.69 cm (SD 0.20). Values in literature have been reported to be 4.8 cm (SD 0.5)37 and 5.0 cm38 for week 30 and 2.7 cm (SD 0.2)37 and 2.7 cm38 for week 20. For the normal femora, we found a mean Prenatal Diagnosis 2014, 34, 940–945
length at week 30 of 5.88 cm (SD 0.39) and at week 20 of 3.35 cm (SD 0.22) with values in literature of 5.6 cm (SD 0.3) for week 30 and 3.1 cm (SD 0.3) for week 20.37,38 The calculated tibial lengths seem to be extremely accurate, which supports the quality of our data and measurements.
CONCLUSION In order to improve the detection rate of CFD and FH, the standard prenatal sonographic protocol for the required extremity views would need to be modified; the current standard fetal extremity views are described as ‘extremities: legs and arms: presence or absence’. The standard views, were they to include documentation and measurement of the long bones of the extremities on all fetuses prenatally, could improve detection rates. Unilateral femoral and/or tibial shortening in a prenatal ultrasonograph should point towards CFD and/or FH. Proper prenatal counseling is necessary to help ensure that the patient is able to make an informed decision about continuing or terminating the pregnancy. An accurate prenatal diagnosis with measurements of femoral and tibial lengths allows for prediction of LLD at birth and at maturity. From these data, surgeons who are experienced in the field of limb lengthening and reconstruction can estimate the type and number of surgical procedures that would be necessary to equalize limb length and can provide a treatment outline. This can be helpful to couples who are seeking further information about postnatal outcomes and sequelae. Additionally, these patients should be offered genetic counseling, consultations with a maternal fetal medicine specialist, and psychological counseling.
ACKNOWLEDGEMENTS The authors thank Amanda E. Chase, MA, and Stacy C. Specht, MPA, for their invaluable assistance with the manuscript. © 2014 John Wiley & Sons, Ltd.
Prenatal diagnosis
945
WHAT’S ALREADY KNOWN ABOUT THIS TOPIC?
WHAT DOES THIS STUDY ADD?
• Ultrasonography can detect congenital femoral deficiency
• We found that the detection rate for CFD cases (52%) was more
(CFD) and fibular hemimelia (FH); however, prenatal detection rate, relationship between detection rate and deformity severity, and whether mothers prefer prenatal diagnosis were unknown.
than double the rate of FH cases (23%). Most mothers (63%) preferred prenatal diagnosis. • Measure long bones of both lower extremities during ultrasonography to allow prenatal and orthopedic consultation.
REFERENCES 1. Manner HM, Radler C, Ganger R, Grill F. Knee deformity in congenital longitudinal deficiencies of the lower extremity. Clin Orthop Relat Res 2006;448:185–92. 2. Manner HM, Radler C, Ganger R, Grill F. Dysplasia of the cruciate ligaments: radiographic assessment and classification. J Bone Joint Surg Am 2006;88(1):130–7. 3. Paley D, Standard SC. Lengthening reconstruction surgery for congenital femoral deficiency. In Limb Lengthening and Reconstruction Surgery, Rozbruch RS, Ilizarov S (eds). Informa Healthcare USA: New York, 2007;393–428. 4. Pappas AM. Congenital abnormalities of the femur and related lower extremity malformations: classification and treatment. J Pediatr Orthop 1983;3(1):45–60. 5. Achterman C, Kalamchi A. Congenital deficiency of the fibula. J Bone Joint Surg Br 1979;61(2):133–7. 6. Keret D, Timor IE. Pediatric update #4. Familial congenital short femur: intrauterine detection and follow-up by ultrasound. A case report. Orthop Rev 1988;17(5):500–4. 7. Jeanty P, Kleinman G. Proximal femoral focal deficiency. J Ultrasound Med 1989;8(11):639–42. 8. Sepulveda W, Weiner E, Bridger JE, Fisk NM. Prenatal diagnosis of congenital absence of the fibula. J Ultrasound Med 1994;13:655–7. 9. Kurtz AB, Needleman L, Wapner RJ, et al. Usefulness of a short femur in the in utero detection of skeletal dysplasias. Radiology 1990;177 (1):197–200. 10. Kurtz AB, Wapner RJ. Ultrasonographic diagnosis of second-trimester skeletal dysplasias: a prospective analysis in a high-risk population. J Ultrasound Med 1983;2(3):99–106. 11. Parilla BV, Leeth EA, Kambich MP, et al. Antenatal detection of skeletal dysplasias. J Ultrasound Med 2003;22(3):255–8. 12. Kevern L, Warwick D, Wellesley D, et al. Prenatal ultrasound: detection and diagnosis of limb abnormalities. J Pediatr Orthop 2003;23(2):251–3. 13. Stoll C, Wiesel A, Queisser-Luft A, et al. Evaluation of the prenatal diagnosis of limb reduction deficiencies. EUROSCAN Study Group. Prenat Diagn 2000;20(10):811–8. 14. Uffelman J, Woo R, Richards DS. Prenatal diagnosis of bilateral fibular hemimelia. J Ultrasound Med 2000;19(5):341–4. 15. Alman BA, Krajbich JI, Hubbard S. Proximal femoral focal deficiency: results of rotationplasty and Syme amputation. J Bone Joint Surg Am 1995;77(12):1876–82. 16. Birch JG, Walsh SJ, Small JM, et al. Syme amputation for the treatment of fibular deficiency. An evaluation of long-term physical and psychological functional status. J Bone Joint Surg Am 1999;81 (11):1511–8. 17. Westberry DE, Davids JR. Proximal focal femoral deficiency (PFFD): management options and controversies. Hip Int 2009;19(Suppl 6): S18–25. 18. Patel M, Paley D, Herzenberg JE. Limb-lengthening versus amputation for fibular hemimelia. J Bone Joint Surg Am 2002;84(2):317–9. 19. Stanitski D. Fibular hemimelia. In Limb Lengthening and Reconstruction Surgery, Rozbruch RS, Ilizarov S (eds). Informa Healthcare USA: New York, 2007; 449–59. 20. Paley D, Bhave A, Herzenberg JE, Bowen JR. Multiplier method for predicting limb-length discrepancy. J Bone Joint Surg Am 2000; 82(10):1432–46.
Prenatal Diagnosis 2014, 34, 940–945
21. Paley J, Gelman A, Paley D, Herzenberg JE. The prenatal multiplier method for prediction of limb length discrepancy. Prenat Diagn 2005; 25(6):435–8. 22. American College of Radiology. Practice guideline for the performance of obstetrical ultrasound. URL http://www.acr.org/~/media/ACR/ Documents/PGTS/guidelines/US_Obstetrical.pdf [accessed on 4 December 2013]. 23. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 101: Ultrasonography in pregnancy. Obstet Gynecol 2009;113:451–61. 24. The American Institute of Ultrasound in Medicine. AIUM practice guideline for the performance of obstetric ultrasound examinations. URL http://www. jultrasoundmed.org/content/29/1/157 [accessed on 4 December 2013]. 25. Armstrong R, McCann E, Fryer A, et al. A review of prenatally detected femoral abnormalities. Clin Dysmorphol 2009;18(3):127–30. 26. Bromley B, Brown DL, Benacerraf BR. Short femur length associated with severe intrauterine growth retardation. Prenat Diagn 1993;13(6):449–52. 27. Goncalves L, Jeanty P. Fetal biometry of skeletal dysplasias: a multicentric study. J Ultrasound Med 1994;13(12):977–85. 28. Nyberg DA. May all your femurs be long! Ultrasound Obstet Gynecol 2008;31(5):489–92. 29. Papageorghiou AT, Fratelli N, Leslie K, et al. Outcome of fetuses with antenatally diagnosed short femur. Ultrasound Obstet Gynecol 2008; 31(5):507–11. 30. Todros T, Massarenti I, Gaglioti P, et al. Fetal short femur length in the second trimester and the outcome of pregnancy. BJOG 2004;111(1):83–5. 31. Crane JP, LeFevre ML, Winborn RC, et al. A randomized trial of prenatal ultrasonographic screening: impact on the detection, management, and outcome of anomalous fetuses. The RADIUS Study Group. Am J Obstet Gynecol 1994;171(2):392–9. 32. Stoll C, Alembik Y, Dott B, Roth MP. Evaluation of prenatal diagnosis of limb reduction defects by a registry of congenital anomalies. Prenat Diagn 1994;14(9):781–6. 33. Stoll C, Dott B, Alembik Y, Roth MP. Evaluation of routine prenatal diagnosis by a registry of congenital anomalies. Prenat Diagn 1995;15(9):791–800. 34. Offerdal K, Jebens N, Blaas HG, Eik-Nes SH. Prenatal ultrasound detection of talipes equinovarus in a non-selected population of 49 314 deliveries in Norway. Ultrasound Obstet Gynecol 2007;30(6):838–44. 35. Radler C, Myers AK, Burghardt RD, Arrabal PP, Herzenberg JE, Grill F. Maternal attitudes towards prenatal diagnosis of idiopathic clubfoot. Ultrasound Obstet Gynecol 2011;37(6):658–62. 36. Magriples U, Copel JA. Accurate detection of anomalies by routine ultrasonography in an indigent clinic population. Am J Obstet Gynecol 1998;179(4):978–81. 37. Merz E, Kim-Kern MS, Pehl S. Ultrasonic mensuration of fetal limb bones in the second and third trimesters. J Clin Ultrasound 1987; 15(3):175–83. 38. Romero R, Athanassiadis AP, Jeanty P. Fetal skeletal anomalies. Radiol Clin North Am 1990;28(1):75–99.
SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s web site.
© 2014 John Wiley & Sons, Ltd.
ORIGINAL ARTICLE
Prenatal diagnosis of congenital femoral deficiency and fibular hemimelia† Christof Radler1*, Abigail K. Myers2, Renee J. Hunter2, Pedro P. Arrabal3 and John E. Herzenberg2 1
Department of Pediatric Orthopaedics, Orthopaedic Hospital Speising–Vienna, Vienna, Austria International Center for Limb Lengthening, Rubin Institute for Advanced Orthopedics, Sinai Hospital of Baltimore, Baltimore, MD, USA 3 Louis and Henrietta Blaustein Women’s Health Center, Department of Obstetrics and Gynecology, Sinai Hospital of Baltimore, Baltimore, MD, USA *Correspondence to: Christof Radler. E-mail: [email protected] † Study conducted at the Rubin Institute for Advanced Orthopedics at Sinai Hospital of Baltimore in Baltimore, MD, USA. 2
ABSTRACT Objectives Routine ultrasonography can detect congenital femoral deficiency (CFD) and fibular hemimelia (FH), but prenatal detection rate and its relation to deformity severity have never been reported. Whether mothers prefer prenatal diagnosis is also unknown. We aimed to determine whether mothers prefer prenatal diagnosis, to report detection rates for CFD and/or FH, and to correlate detection rates to severity of limb shortening. Methods Surveys were mailed to 171 mothers who gave birth to children with CFD/FH between 2000 and 2008. Bilateral femoral and tibial lengths were measured on postnatal radiographs. We calculated corresponding femoral/ tibial lengths at gestational weeks 20 and 30. Results Sixty-five surveys were returned, and 56 radiographs were reviewed. Most mothers (63%) preferred prenatal diagnosis as it enables prenatal counseling. Congenital limb shortening was detected in 24 cases (37%) and was not detected in 41 cases (63%). Detection rate was 52% (12 of 23) in CFD cases, 23% (three of 13) in FH cases, and 30% (six of 20) in combined cases. CFD cases with severe shortening had a higher detection rate. Conclusions Ultrasonographers should measure both femoral and tibial lengths. Unilateral shortening should result in pediatric orthopedic consultation to estimate limb-length discrepancy at maturity and discuss treatment. © 2014 John Wiley & Sons, Ltd.
Funding sources: None Conflicts of interest: None declared
INTRODUCTION Congenital femoral deficiency (CFD) and fibular hemimelia (FH) are rare longitudinal defects of the lower limb. CFD presents with mild femoral shortening to severe dysplasia with absence or pseudarthrosis of the femoral head and/or neck area and hypoplasia of the lateral femoral condyle.1–4 FH presents with hypoplasia or aplasia of the fibula with shortening and malalignment of the tibia, distinct joint deformities at the knee and ankle joint, and ray deficiencies.1,3,5 In the majority of cases, both conditions are unilateral.1,2,4,5 Prenatal diagnosis of CFD and FH was first reported in the late 1980s and early 1990s.6–8 Whereas some studies have evaluated the prenatal diagnosis of skeletal dysplasias in large case series,9–11 the detection rate for congenital longitudinal deficiencies of the lower limb has not been reported. Early consultation with an expert would help promote acceptance of the deformity and allow discussion of options.12– 14 Although amputation and/or prosthetic fitting has traditionally been recommended for most cases,15–17 reconstructive treatment Prenatal Diagnosis 2014, 34, 940–945
can be performed with good functional results.3,18,19 This may involve multiple surgical procedures starting at approximately age 2 years to the end of growth depending on the severity of the shortening.3 We aimed to determine whether mothers prefer prenatal diagnosis, report the detection rates for CFD and/or FH, and correlate detection rates to the severity and shortening of the involved limb.
METHODS The hospital’s institutional review board approved this study, and the study was conducted in accordance with the Helsinki Declaration of 1975 (revised in 1983). Patients were identified retrospectively by diagnosis of CFD and FH using the hospital’s computer database. Surveys were mailed to 171 mothers who gave birth to a child with CFD and/or FH between January 2000 and December 2008 and presented for consultation and/or treatment. The mothers were sent the survey, a letter explaining the study, © 2014 John Wiley & Sons, Ltd.
Prenatal diagnosis
and a consent form. After obtaining consent, the medical records were reviewed to assess exclusion criteria. Exclusion criteria were underlying syndrome or genetic abnormality other than CFD and FH, multiple gestations, a family history of a limb defect, and birth outside of the USA. The survey consisted of three questions: (1) Did you have any ultrasounds (sonograms) before your baby was born? (2) Did any of your ultrasound tests show a short leg? (3) When would you have preferred to find out about your baby’s congenital leg shortening? Question 3 was answered by selecting either ‘before birth’ or ‘after birth’. Another section on the survey asked participants to share general comments. Responses and comments were analyzed on the basis of prenatal or postnatal diagnosis and year of birth. The rates of true-positive and false-negative diagnoses were also determined. Postnatal radiographs were analyzed of the patients whose surveys were returned. The severity of the deformity as found in the most recent preoperative bilateral long leg radiographs was graded using the classification systems of Paley3 and Pappas4 for CFD and the classification of Achterman and Kalamchi5 for FH cases. The Pappas classification4 for CFD describes types I to IX, with type I being the most severe. The Paley classification3 is more treatment based (i.e., surgical necessities and options) and is organized into types 1 to 4, with 4 being the most severe. The Achterman and Kalamchi classification5 is a descriptive and morphologic system that consists of types I, IIa, and IIb, with IIb being the most severe. Cases of FH-associated ray deficiencies and/or foot deformities were recorded. The femoral length, tibial length, and overall limb length (i.e., femur and tibia) of both limbs were measured. From these measurements, the limb-length discrepancy (LLD) and discrepancies of femoral length and tibial length at the time of the radiograph were determined. In the next step, the LLD and femoral and tibial length at birth were calculated using the lower-limb multiplier20 in a reversed way. In a second step, the prenatal multiplier21 was used on this data to calculate the LLD and femoral and tibial length at the time when secondtrimester (20 weeks) and third-trimester (30 weeks) ultrasounds are usually obtained. These times were chosen according to recommendations for standard second- and third-trimester ultrasonographic examinations from the American College of Radiology, the American College of Obstetricians and Gynecologists, and the American Institute of Ultrasound in Medicine.22–24 Additionally, LLD at maturity was predicted using the lower-limb multiplier.20 The detection rate was analyzed according to the amount of LLD and femoral and tibial shortening. For descriptive statistics, categorical variables are reported in percentages and numeric variables are given as mean with minimum, maximum, and standard deviations reported. Subgroups were defined and compared using a twotailed Fisher exact test with p < 0.05 considered significant (IBM SPSS Statistics 19.0, Somers, NY, USA).
RESULTS Surveys were sent to 171 mothers. Sixty-five surveys were completed and returned (38% response rate). All mothers had at least one prenatal ultrasound. A majority of respondents (41 mothers, 63%) preferred prenatal diagnosis, fewer respondents Prenatal Diagnosis 2014, 34, 940–945
941
(21 mothers, 32%) preferred to know after birth, and a small minority (three mothers, 5%) were undecided. Of the patients who had a positive ultrasound, 88% of the respondents (21 mothers) wanted a prenatal diagnosis and 13% (three mothers) wanted a postnatal diagnosis. Of the 41 patients who had a negative ultrasound, 49% of the respondents (20 mothers) would have wanted a prenatal diagnosis, 44% (18 mothers) did not want a prenatal diagnosis, and 7% (three mothers) were undecided. Overall, more mothers (21 of 24 mothers, 88%) who received a positive prenatal diagnosis reported wanting to know before birth than those who did not receive a prenatal diagnosis (20 of 41 mothers, 49%). We statistically tested the hypothesis that more mothers who had a positive ultrasound wanted the prenatal diagnosis than those who had a false-negative ultrasound using a Fisher exact test and found a p-value of 0.006 supporting this hypothesis. Comments on the survey form showed that 20 mothers who had or wanted to have the prenatal diagnosis appreciated the time to prepare, to research the condition and different treatment options, and to become emotionally accustomed. Some mothers wished for more information at the time of prenatal diagnosis (seven mothers). Mothers who preferred a postnatal diagnosis feared that a prenatal diagnosis would have affected their experience during pregnancy (11 mothers). During routine ultrasonography, congenital limb shortening was detected (true positive) in 24 cases (37%) and was not detected (false negative) in 41 cases (63%). For infants born between the years 2000 and 2003, the overall detection rate was 38% (12 of 32 cases) versus 36% (12 of 33 cases) for those born between 2004 and 2008. The detection rate for those two time periods was not significantly different (p = 1 using a two-tailed Fisher exact test). We were able to review the preoperative radiographs of 56 of the 65 patients whose mothers returned the survey. The mean age of the patients when the radiographs were obtained was 2.5 ± 1.6 years (range, 6 months to 7 years 6 months). We analyzed the radiographs of 31 male and 25 female patients with the left side being affected in 27 cases and the right in 29 cases. There was no significant difference regarding the detection rate for left versus right side [eight of 27 (30%) vs 13 of 29 (45%); p = 0.279 using a two-tailed Fisher exact test]. Of the 56 cases, CFD was present in 23 cases (41%), FH in 13 cases (23%), and a combination of CFD and FH in 20 cases (36%). Table 1 shows the classification of the CFD cases according to the Pappas4 and Paley3 classification and of the FH cases according to the Achterman and Kalamchi5 classification. We found a detection rate of 52% (12 of 23 cases) in CFD cases, a detection rate of 23% (three of 13 cases) in FH cases, and a detection rate of 30% (six of 20 cases) in combined cases. The detection rate between CFD and FH (p = 0.159), between CFD and combined cases (p = 0.216), and between FH and combined cases (p = 1) was not significantly different using a two-tailed Fisher exact test. A ray deficiency (three-ray or four-ray foot) was found in 11 of the 13 FH cases and in five of the 20 cases presenting with combined FH and CFD. Of four FH cases with a foot deformity, 50% (two of four cases) were diagnosed prenatally. For all FH and combined CFD and FH cases with ray deficiencies, the detection rate was only 25% (four of 16 cases). © 2014 John Wiley & Sons, Ltd.
C. Radler et al.
942
Table 1 Classification of the patients whose radiographs were available for analysis (n = 56) Pappas classification of CFD4 Classification
Paley classification of CFD3
Classification of FH according to Achterman and Kalamchi5
Number of cases
Classification
Number of cases
Classification
Number of cases
II
3
1a
28
1a
14
III
6
1b
2
1b
2
IV
4
1c
3
2
17
V
1
2a
6
VI
1
2b
2
VII
7
3a
2
VIII
3
3b
0
IX
18
CFD, congenital femoral deficiency; FH, fibular hemimelia.
Cases of CFD and cases of combined CFD and FH that had a Pappas classification4 of VII to IX showed a detection rate of 32% (nine of 28 cases), whereas cases classified Pappas II to VI had a detection rate of 60% (nine of 15), which was not a significant difference (p = 0.109 using a two-tailed Fisher exact test). Cases classified as Paley3 type 1a had a detection rate of 36% (ten of 28 cases), whereas types 1b, 1c, 2a, 2b, and 3a had a detection rate of 53% (eight of 15 cases), a difference which again was not significant (p = 0.338). Cases with FH type 1a or 1b according to Achterman and Kalamchi5 had a detection rate of 13% (one of eight cases) versus 40% (two of five cases) for type 2, a difference which was not significant (p = 0.550 using a two-tailed Fisher exact test). Table 2 presents the predicted limb, femoral, and tibial length discrepancies at birth, gestational week 30, and gestational week 20.20,21 Table 3 shows the detection rates for the three diagnostic groups (i.e., CFD, FH, and combined CFD and FH) on the basis of limb, femoral, and tibial discrepancies at gestational week 30. The predicted lengths of the short and normal femora, tibiae, and limbs (i.e., femur plus tibia) at birth, gestational week 20, and gestational week 30 as well as the detection rate according to LLD at maturity are shown in Tables 4 and 5, which is available online as supporting information.
DISCUSSION Although rare, CFD and FH can lead to significant morbidity postnatally. Prenatal diagnosis of these conditions can allow time for mothers to learn about treatment options, such as limb reconstruction or amputation. Most mothers preferred prenatal diagnosis, and some mothers wanted to receive more information at the time of diagnosis. We found a detection rate of 52% in CFD cases (12 of 23), 23% in FH cases (three of 13), and 30% in combined cases (six of 20). Our detection rate for CFD cases was more than double the detection rate in FH cases. This might be because the femoral length is usually measured and used to detect skeletal dysplasia, intrauterine growth restriction, or Down syndrome.9,25–30 These are higher rates of detection than those reported in published literature. The RADIUS31 study included 15 530 low-risk pregnant women who underwent screening ultrasound and found five cases of limb-reduction deficiencies with two Prenatal Diagnosis 2014, 34, 940–945
being detected (40% detection rate). However, there is no information as to whether these deficiencies were severe or were transverse or longitudinal. The authors of the EUROSCAN study group13 reported a detection rate for limb-reduction deficiencies of 36% (89 of 250) dropping to 25% (34 of 138) for isolated cases without associated deformities. In their case series with 138 isolated limb-reduction deficiencies, a longitudinal defect with absence or severe hypoplasia of a lateral part of the limb was found in 12 cases with only one being diagnosed prenatally (8% detection rate). In a further study, the sensitivity for isolated limb-reduction deficiencies was only 4% (two of 49),32 whereas an extended study of the same group showed an increase of the detection rate for isolated cases from 2% (one of 43) to 13% (two of 15) from the period 1979 to 1988 versus 1989 to 1992.33 Our study found an overall detection rate for infants born between 2000 and 2003 of 38% (12 of 32 cases) versus 36% (12 of 33 cases) for those born between 2004 and 2008, indicating no increase of the detection rate between those two recent time periods. Kevern and coworkers12 evaluated prenatal diagnostic detection rates in a cohort of 26 203 infants. Of 60 patients with a limb abnormality, only 15 were detected prenatally resulting in a 25% detection rate. Subdividing the cases into major and minor abnormalities, the detection rate was 33% (13 of 39 cases) and 10% (two of 21 cases), respectively. In our series, the detection rate in combined CFD with FH cases with more severe shortening (LLD > 2 cm) was double (40%, four of ten) compared with less severe shortening (20%, two of ten, LLD < 2 cm). For isolated CFD, we found a higher detection rate when femoral discrepancy was more than 2 cm (58%, five of 12) versus less than 2 cm (46%, six of 11). However, the rate of detection for FH cases in our series did not increase as severity increased. Goncalves and Jeanty27 found that fetuses with mild femoral shortness were difficult to differentiate from mild shortening owing to skeletal dysplasia. In their series, only two of 139 cases had CFD, and two had femur–fibular–ulna complex. The two CFD cases were within the group showing mild shortening defined as a length greater than 80% of the mean for gestational age. Using this delineator, mild shortening was found in 12 of the 20 cases of combined CFD and FH with a detection rate of 25.0% (three of 12 cases) in our series. Mild cases of CFD and FH show only mild shortening prenatally © 2014 John Wiley & Sons, Ltd.
Prenatal diagnosis
943
Table 2 Limb, femoral, and tibial length discrepancies are summarized for the different diagnostic subgroups at birth, gestational week 30, and gestational week 20 Diagnosis Combined CFD and FH
CFD
FH
Parameter
N
Minimum (cm)
Maximum (cm)
Mean (cm)
SD (cm)
LLD at birth
20
1.03
8.18
3.55
2.11
Femoral discrepancy at birth
20
0.27
6.10
2.06
1.84
Tibial discrepancy at birth
20
0.18
3.48
1.51
0.92
LLD at week 30
20
0.63
5.00
2.17
1.29
Femoral discrepancy at week 30
20
0.17
3.72
1.26
1.12
Tibial discrepancy at week 30
20
0.11
2.13
0.92
0.56
LLD at week 20
20
0.36
2.85
1.24
0.73
Femoral discrepancy at week 20
20
0.09
2.12
0.72
0.64
Tibial discrepancy at week 20
20
0.06
1.21
0.53
0.32
LLD at birth
23
0.91
6.50
3.87
1.76
Femoral discrepancy at birth
23
0.97
6.31
3.73
1.70
0.66
0.16
0.20
3.97
2.36
1.08
Tibial discrepancy at birth
23
0.05
LLD at week 30
23
0.55
a
Femoral discrepancy at week 30
23
0.59
3.86
2.28
1.04
Tibial discrepancy at week 30
23
0.03a
0.41
0.09
0.13
LLD at week 20
23
0.32
2.26
1.35
0.61
Femoral discrepancy at week 20
23
0.34
2.20
1.30
0.59
0.23
0.05
0.07
3.48
1.84
1.11
0.91
0.25
0.30
3.20
1.61
2.00
Tibial discrepancy at week 20
23
0.02
LLD at birth
13
0.39
a
a
Femoral discrepancy at birth
13
0.22
Tibial discrepancy at birth
13
0.42
LLD at week 30
13
0.24
2.13
1.13
0.68
Femoral discrepancy at week 30
13
0.13a
0.56
0.15
0.18
Tibial discrepancy at week 30
13
0.26
1.95
0.98
0.61
LLD at week 20
13
0.14
1.21
0.64
0.39
Femoral discrepancy at week 20
13
0.08a
0.32
0.09
0.10
Tibial discrepancy at week 20
13
0.15
1.11
0.56
0.35
20
21
The LLD and femoral and tibial lengths at birth were calculated using the lower-limb multiplier. The prenatal multiplier was used with these data to calculate the LLD and femoral and tibial length at the time when second-trimester (20 weeks) and third-trimester (30 weeks) ultrasounds are usually obtained. CFD, congenital femoral deficiency; FH, fibular hemimelia; LLD, limb-length discrepancy; SD, standard deviation. a Negative values indicate that the specific bone was longer on the side with the short extremity.
and fall into the group of cases with a length greater than 80% of the mean for gestational age. This group can be especially hard to detect and to delineate from normal and other differential diagnoses, such as intrauterine growth restriction or skeletal dysplasia. Eight cases were in the group with femoral shortness between 30% and 80% of the mean with a detection rate of 38% (three of eight cases). In the subgroup of isolated CFD, we had six cases in the more than 80% of the mean group (detection rate 50%; three of six cases), 16 cases in the 30% to 80% group (detection rate 50%; eight of 16 cases), and one had a femoral length of less than 30% of the mean with a detection rate of 100% (one of one case). The cases with isolated FH (n = 13) had six cases in the above 80% group (detection rate 0%; zero of six) and seven in the 30% to 80% of the mean group (detection rate 43%; three of seven). Taking the percentiles as a threshold and using the multiplier formulae,20,21 we found that 22 of 23 femoral lengths at week 30 were below the fifth percentile in the CFD group. In the CFD Prenatal Diagnosis 2014, 34, 940–945
with FH group (n = 20), 11 femoral lengths and 17 tibial lengths were below the fifth percentile, whereas all the FH cases showed tibial lengths under the fifth percentile. Associated foot deformities in the FH group increased the detection rate from 23% to 50% (two of four). This is not surprising as the detection rate for clubfoot is reported to be as high as 81%.34,35 However, our cases with associated ray deficiencies (n = 16) had a detection rate of only 25%. According to Jeanty and Kleinman,7 CFD should be differentiated immediately from thanatophoric dysplasia, achondroplasia, achondrogenesis, short limb polydactyly, chondroectodermal dysplasia, campomelic dysplasia, and osteogenesis imperfecta as in all these disorders the findings are bilateral. They conclude that a normal fetus with only one short femur has virtually no other differential diagnosis than CFD. There are several biases in our study. Owing to the low incidence of these conditions, we were unable to report a greater number of cases. As the study was conducted at a © 2014 John Wiley & Sons, Ltd.
C. Radler et al.
944
Table 3 Length discrepancies at week 30 and the detection rate according to the severity of the shortening for the different diagnostic groups Diagnosis Combined CFD and FH
CFD
FH
Parameter LLD < 2 cm
N
True positive (number of cases)
10
False negative (number of cases)
2
8
Detection rate (%) 20.0
LLD > 2 cm (up to 5 cm)
10
4
6
40.0
Femoral discrepancy < 1 cm
13
3
10
23.1
Femoral discrepancy > 1 cm (up to 3.72 cm)
7
3
4
42.9
Tibial discrepancy < 1 cm
11
4
7
36.4
Tibial discrepancy > 1 cm (up to 2.13 cm)
9
2
7
22.2
Femoral discrepancy < 2 cm
11
5
6
45.5
Femoral discrepancy > 2 cm (up to 3.9 cm)
12
7
5
58.3
Tibial discrepancy < 1 cm
8
2
6
25.0
Tibial discrepancy > 1 cm (up to 1.95 cm)
5
1
4
20.0
p-value 0.628
0.613
0.642
0.684
1
CFD, congenital femoral deficiency; FH, fibular hemimelia; LLD, limb-length discrepancy.
tertiary referral center specializing in limb lengthening and reconstruction, cases in which the family decided on prosthetic management or amputation may not present to the center, resulting in a referral bias. Although the survey response rate of 38% (65 of 171) seems satisfactory, a nonresponse bias exists. In addition, secondary to the retrospective nature of data collection, we cannot account for cases in which fetal death may have occurred or mothers may have chosen pregnancy termination. This may have resulted in significant bias because mothers who chose to terminate pregnancy were not included in this study, which means that the most severe cases of prenatal CFD and FH may have been excluded. Only a few cases of termination of pregnancy in an isolated case of CFD have been reported in the literature.10,36 However, the true frequency of termination of pregnancy for isolated CFD or FH might be higher. The survey questions themselves introduce some bias especially as they are asked retrospectively. We analyzed only radiographs of the patients whose surveys were completed and returned, which resulted in a low sample size. Measuring the radiographs introduces additional bias. The child has to stand or lie still, and the knee joint must be fully extended to allow accurate length measurements on the radiographs. When full knee extension was doubtful, lateral view radiographs were additionally analyzed. All radiographs were measured by the first author to minimize interobserver variance. Lastly, the use of the prenatal multiplier has some inherent error rate. That being said, the mean values for the long (normal side) femora and tibiae at the 30th and 20th weeks of gestation that were calculated using the multiplier and the prenatal multiplier20,21 correspond very well with the values reported in literature.37,38 For the normal tibiae, we calculated a mean tibial length at week 30 of 4.73 cm (SD 0.34) and at week 20 of 2.69 cm (SD 0.20). Values in literature have been reported to be 4.8 cm (SD 0.5)37 and 5.0 cm38 for week 30 and 2.7 cm (SD 0.2)37 and 2.7 cm38 for week 20. For the normal femora, we found a mean Prenatal Diagnosis 2014, 34, 940–945
length at week 30 of 5.88 cm (SD 0.39) and at week 20 of 3.35 cm (SD 0.22) with values in literature of 5.6 cm (SD 0.3) for week 30 and 3.1 cm (SD 0.3) for week 20.37,38 The calculated tibial lengths seem to be extremely accurate, which supports the quality of our data and measurements.
CONCLUSION In order to improve the detection rate of CFD and FH, the standard prenatal sonographic protocol for the required extremity views would need to be modified; the current standard fetal extremity views are described as ‘extremities: legs and arms: presence or absence’. The standard views, were they to include documentation and measurement of the long bones of the extremities on all fetuses prenatally, could improve detection rates. Unilateral femoral and/or tibial shortening in a prenatal ultrasonograph should point towards CFD and/or FH. Proper prenatal counseling is necessary to help ensure that the patient is able to make an informed decision about continuing or terminating the pregnancy. An accurate prenatal diagnosis with measurements of femoral and tibial lengths allows for prediction of LLD at birth and at maturity. From these data, surgeons who are experienced in the field of limb lengthening and reconstruction can estimate the type and number of surgical procedures that would be necessary to equalize limb length and can provide a treatment outline. This can be helpful to couples who are seeking further information about postnatal outcomes and sequelae. Additionally, these patients should be offered genetic counseling, consultations with a maternal fetal medicine specialist, and psychological counseling.
ACKNOWLEDGEMENTS The authors thank Amanda E. Chase, MA, and Stacy C. Specht, MPA, for their invaluable assistance with the manuscript. © 2014 John Wiley & Sons, Ltd.
Prenatal diagnosis
945
WHAT’S ALREADY KNOWN ABOUT THIS TOPIC?
WHAT DOES THIS STUDY ADD?
• Ultrasonography can detect congenital femoral deficiency
• We found that the detection rate for CFD cases (52%) was more
(CFD) and fibular hemimelia (FH); however, prenatal detection rate, relationship between detection rate and deformity severity, and whether mothers prefer prenatal diagnosis were unknown.
than double the rate of FH cases (23%). Most mothers (63%) preferred prenatal diagnosis. • Measure long bones of both lower extremities during ultrasonography to allow prenatal and orthopedic consultation.
REFERENCES 1. Manner HM, Radler C, Ganger R, Grill F. Knee deformity in congenital longitudinal deficiencies of the lower extremity. Clin Orthop Relat Res 2006;448:185–92. 2. Manner HM, Radler C, Ganger R, Grill F. Dysplasia of the cruciate ligaments: radiographic assessment and classification. J Bone Joint Surg Am 2006;88(1):130–7. 3. Paley D, Standard SC. Lengthening reconstruction surgery for congenital femoral deficiency. In Limb Lengthening and Reconstruction Surgery, Rozbruch RS, Ilizarov S (eds). Informa Healthcare USA: New York, 2007;393–428. 4. Pappas AM. Congenital abnormalities of the femur and related lower extremity malformations: classification and treatment. J Pediatr Orthop 1983;3(1):45–60. 5. Achterman C, Kalamchi A. Congenital deficiency of the fibula. J Bone Joint Surg Br 1979;61(2):133–7. 6. Keret D, Timor IE. Pediatric update #4. Familial congenital short femur: intrauterine detection and follow-up by ultrasound. A case report. Orthop Rev 1988;17(5):500–4. 7. Jeanty P, Kleinman G. Proximal femoral focal deficiency. J Ultrasound Med 1989;8(11):639–42. 8. Sepulveda W, Weiner E, Bridger JE, Fisk NM. Prenatal diagnosis of congenital absence of the fibula. J Ultrasound Med 1994;13:655–7. 9. Kurtz AB, Needleman L, Wapner RJ, et al. Usefulness of a short femur in the in utero detection of skeletal dysplasias. Radiology 1990;177 (1):197–200. 10. Kurtz AB, Wapner RJ. Ultrasonographic diagnosis of second-trimester skeletal dysplasias: a prospective analysis in a high-risk population. J Ultrasound Med 1983;2(3):99–106. 11. Parilla BV, Leeth EA, Kambich MP, et al. Antenatal detection of skeletal dysplasias. J Ultrasound Med 2003;22(3):255–8. 12. Kevern L, Warwick D, Wellesley D, et al. Prenatal ultrasound: detection and diagnosis of limb abnormalities. J Pediatr Orthop 2003;23(2):251–3. 13. Stoll C, Wiesel A, Queisser-Luft A, et al. Evaluation of the prenatal diagnosis of limb reduction deficiencies. EUROSCAN Study Group. Prenat Diagn 2000;20(10):811–8. 14. Uffelman J, Woo R, Richards DS. Prenatal diagnosis of bilateral fibular hemimelia. J Ultrasound Med 2000;19(5):341–4. 15. Alman BA, Krajbich JI, Hubbard S. Proximal femoral focal deficiency: results of rotationplasty and Syme amputation. J Bone Joint Surg Am 1995;77(12):1876–82. 16. Birch JG, Walsh SJ, Small JM, et al. Syme amputation for the treatment of fibular deficiency. An evaluation of long-term physical and psychological functional status. J Bone Joint Surg Am 1999;81 (11):1511–8. 17. Westberry DE, Davids JR. Proximal focal femoral deficiency (PFFD): management options and controversies. Hip Int 2009;19(Suppl 6): S18–25. 18. Patel M, Paley D, Herzenberg JE. Limb-lengthening versus amputation for fibular hemimelia. J Bone Joint Surg Am 2002;84(2):317–9. 19. Stanitski D. Fibular hemimelia. In Limb Lengthening and Reconstruction Surgery, Rozbruch RS, Ilizarov S (eds). Informa Healthcare USA: New York, 2007; 449–59. 20. Paley D, Bhave A, Herzenberg JE, Bowen JR. Multiplier method for predicting limb-length discrepancy. J Bone Joint Surg Am 2000; 82(10):1432–46.
Prenatal Diagnosis 2014, 34, 940–945
21. Paley J, Gelman A, Paley D, Herzenberg JE. The prenatal multiplier method for prediction of limb length discrepancy. Prenat Diagn 2005; 25(6):435–8. 22. American College of Radiology. Practice guideline for the performance of obstetrical ultrasound. URL http://www.acr.org/~/media/ACR/ Documents/PGTS/guidelines/US_Obstetrical.pdf [accessed on 4 December 2013]. 23. American College of Obstetricians and Gynecologists. ACOG Practice Bulletin No. 101: Ultrasonography in pregnancy. Obstet Gynecol 2009;113:451–61. 24. The American Institute of Ultrasound in Medicine. AIUM practice guideline for the performance of obstetric ultrasound examinations. URL http://www. jultrasoundmed.org/content/29/1/157 [accessed on 4 December 2013]. 25. Armstrong R, McCann E, Fryer A, et al. A review of prenatally detected femoral abnormalities. Clin Dysmorphol 2009;18(3):127–30. 26. Bromley B, Brown DL, Benacerraf BR. Short femur length associated with severe intrauterine growth retardation. Prenat Diagn 1993;13(6):449–52. 27. Goncalves L, Jeanty P. Fetal biometry of skeletal dysplasias: a multicentric study. J Ultrasound Med 1994;13(12):977–85. 28. Nyberg DA. May all your femurs be long! Ultrasound Obstet Gynecol 2008;31(5):489–92. 29. Papageorghiou AT, Fratelli N, Leslie K, et al. Outcome of fetuses with antenatally diagnosed short femur. Ultrasound Obstet Gynecol 2008; 31(5):507–11. 30. Todros T, Massarenti I, Gaglioti P, et al. Fetal short femur length in the second trimester and the outcome of pregnancy. BJOG 2004;111(1):83–5. 31. Crane JP, LeFevre ML, Winborn RC, et al. A randomized trial of prenatal ultrasonographic screening: impact on the detection, management, and outcome of anomalous fetuses. The RADIUS Study Group. Am J Obstet Gynecol 1994;171(2):392–9. 32. Stoll C, Alembik Y, Dott B, Roth MP. Evaluation of prenatal diagnosis of limb reduction defects by a registry of congenital anomalies. Prenat Diagn 1994;14(9):781–6. 33. Stoll C, Dott B, Alembik Y, Roth MP. Evaluation of routine prenatal diagnosis by a registry of congenital anomalies. Prenat Diagn 1995;15(9):791–800. 34. Offerdal K, Jebens N, Blaas HG, Eik-Nes SH. Prenatal ultrasound detection of talipes equinovarus in a non-selected population of 49 314 deliveries in Norway. Ultrasound Obstet Gynecol 2007;30(6):838–44. 35. Radler C, Myers AK, Burghardt RD, Arrabal PP, Herzenberg JE, Grill F. Maternal attitudes towards prenatal diagnosis of idiopathic clubfoot. Ultrasound Obstet Gynecol 2011;37(6):658–62. 36. Magriples U, Copel JA. Accurate detection of anomalies by routine ultrasonography in an indigent clinic population. Am J Obstet Gynecol 1998;179(4):978–81. 37. Merz E, Kim-Kern MS, Pehl S. Ultrasonic mensuration of fetal limb bones in the second and third trimesters. J Clin Ultrasound 1987; 15(3):175–83. 38. Romero R, Athanassiadis AP, Jeanty P. Fetal skeletal anomalies. Radiol Clin North Am 1990;28(1):75–99.
SUPPORTING INFORMATION Additional supporting information may be found in the online version of this article at the publisher’s web site.
© 2014 John Wiley & Sons, Ltd.
