Ultrasonic measurement of acetabular alpha angle to evaluate developmental dysplasia of the hip: differences between AIUM guideline and European consensus

Abstract

Purpose

To clarify the difference between the alpha angles measured by European method and by AIUM method.

Materials and methods

In this prospective study, infants less than 6-month-old with clinical suspicion of DDH were enrolled from the pediatric outpatient service in a large provincial teaching hospital from October 2022 to October 2023. The hips were examined according to the protocols described by AIUM guideline and by European consensus. The acetabular alpha angle was measured on the standard sections. Statistically significant difference between groups was determined by using a paired-sample t test.

Results

A total of 225 infants (median age, 4.0 months [interquartile range, 3.3–5.2 months]; 104 boys) were enrolled into the study. The mean value of alpha angle measured by AIUM method (62.1°; 95% CI 61.7, 62.4) was smaller than that by European method (63.4°; 95% CI 63.1, 63.7) (P < 0.0001). The 95% limits of agreement between alpha angles measured by the two methods were − 6.6 to 9.2 degree. Alpha angle measured by AIUM method could diagnose DDH with low sensitivity (56% [9 of 16 hips; 95% CI 31, 79]) and specificity (78% [300 of 384 hips; 95% CI 74, 82]) with Graf’s technique as a reference standard. The position of transducer arranged on the skin of infant by AIUM method was 1.6 ± 0.3 cm back to the position by European method.

Conclusion

Alpha angle measured by AIUM method is smaller than that by European method and should not be applied to Graf typing of DDH.

Introduction

Developmental dysplasia of the hip (DDH) is a common pediatric hip disorder with an incidence of about 1% in the newborn population [1, 2]. DDH encompasses a spectrum of abnormal hip, ranging from mild dysplasia to irreducible dislocation [3]. Dislocation of the hip is relatively rare with an incidence of 0.1% [2]. Left untreated, DDH may lead to chronic pain, limping, and degenerative arthritis of hip in early adulthood [4]. After early diagnosis, DDH can usually be treated successfully with non-invasive methods (e.g., by Pavlik harness) in the first months after birth [5]. Ultrasound is now widely used in diagnosis and screening of DDH [5, 6]. There are different methods to evaluate hips with ultrasound. Graf established a static method to evaluate morphology of acetabulum quantitatively with the alpha angle [7]. Terjesen et al. [8] measured “femoral head coverage” to assess the degree of lateralization of the femoral head to the acetabular fossa. Harcke et al. [9] evaluated the positional relationship between the femoral head and the acetabulum using a dynamic approach.

Among all the reported metrics, the acetabular alpha angle is the only index recommended by both the European consensus [10] and the American Institute of Ultrasound in Medicine (AIUM) guideline [11, 12] for assessment of DDH. But, the standard planes to measure the alpha angle are different. The standard plane recommended by AIUM is defined by identifying a straight iliac line, the tip of the acetabular labrum, and the transition from the os ilium to the triradiate cartilage [11]. The standard plane recommended by European consensus [10] should include the hyperechoic tip of the lower limb of the os ilium, chondro-osseous junction, and the straight mid portion of the bony part of acetabular roof (wing of the os ilium), the isoechoic labrum, synovial fold, and joint capsule, and the hypoechoic acetabular cartilage, femoral head, and bony rim [7]. Currently, in some countries, sonographers may use different methods (i.e., AIUM or European) to assess alpha angle, which may bring potential risks arising from different scanning protocols for the evaluation of DDH.

Based on such differences (Table 1), we hypothesize that there is a difference between the value of alpha angle measured with the methods recommended by AIUM guideline and European consensus. Currently, in some countries, sonographers may use different methods (i.e., AIUM or European) to assess alpha angle, which may bring potential risks arising from different scanning protocols for the evaluation of DDH. To clarify the difference between the alpha angles measured by European method and by AIUM method, we will use both of the two methods, i.e., AIUM method and European method, to measure alpha angle in the same group of infants. The differences between the value of alpha angles will be confirmed. The impact of the differences on the diagnosis of DDH will be clarified. The position of the transducer arranged on the skin of infants during ultrasound scanning by the two methods will also be described.

Table 1 Differences between scanning techniques recommended by AIUM guideline and European Consensus to measure acetabular alpha angle

Methods

Patients

Our prospective study was approved by the Ethics Committee of our hospital on 2020–10–28 and registered on the Chinese clinical trial registry website (http://www.chictr.org.cn, number ChiCTR2000040953). The patients consisted of a consecutive sample of infants who met predetermined inclusion criteria from the pediatric outpatient service in our hospital from October 2022 to October 2023. Written informed consent was obtained from the parent(s) of the infants. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Inclusion criteria were infants less than 6-month-old presenting to our department for an ultrasound scan with suspicion of DDH. Infants were excluded if standard coronal plane described by the AIUM guideline or European consensus to evaluate alpha angle was not obtained in ultrasound examination (Fig. 1).

Fig. 1

Flowchart of the study

Ultrasound examination

Ultrasound imaging of hip was performed with a Philips EPIQ 7 system (Philips) equipped with a 10-MHz transducer by one operator (C.Z., with 11 years of experience in DDH evaluation). The hips were examined according to the protocols described by AIUM guideline and by European consensus. All the infants were examined in a lateral decubitus position with the hip in neutral position. According to the protocols by AIUM guideline, the transducer was placed parallel to the back of the infant and then rotated into an oblique coronal plane until the ilium appear straight. Next, adjusted to ensure that the imaging plane was through the deepest part of the acetabulum. The relationship of the femoral head to the posterior acetabulum was tested both at rest and with gentle stress in the transverse flexion view [11]. According to the Graf technique [7] recommended by European consensus [10], the transducer was positioned vertical on the greater trochanter of leg and parallel to the bolsters of the cradle or couch. Finally, moved the probe forwards, backwards, or rotate slightly until the correct sectional plane was acquired. The anatomical structures including chondro-osseous border, femoral head, synovial fold, joint capsule, labrum, cartilage, bony roof, and bony rim were identified on ultrasound [7]. After the standard section was confirmed by the presence of the lower limb of the Ilium, straight echo of ilium, and the acetabular labrum, the acetabular alpha angle was measured by the same operator (C.Z., with 10 years of experience in DDH detection). To evaluate the impact of operator-dependent variability, imaging data for a subset of cases were independently acquired and analyzed by a second operator (Y.Z., with 3 years of experience in DDH detection) (Fig. 2).

Fig. 2

Ultrasound imaging of hip of a 5-month-old girl by European method (i.e., Graf technique) (A) and AIUM method (B). The footprints of transducer left on the skin showed the location relationship of the transducer (C), EU, European, AIUM, American Institute of Ultrasound in Medicine. Schematic (D) showed where the ultrasound beam was projected onto the acetabular surface when detected using European method (full line) and AIUM method (dotted line). Alpha angle measured on the AIUM standard coronal plane (F, 56°) was smaller than that on the European standard coronal plane (E, 64°)

The footprint of the transducer left on the skin of infants after ultrasound examination by different protocols were recorded with a camera to study the spatial location relationship of the transducer. The distance between the central points of the footprint of transducers and the angle between the arrangement of transducers were measured (Fig. 2A–C).

Statistical analyses

In order to avoid the interaction between the protocols, a random number generator was used to determine the order in which the protocols were used first for each infant. The differences between the acetabular alpha angle acquired from AIUM standard plane and the Graf standard plane were analyzed by a paired samples t test. The agreements of the alpha angles between the two methods were analyzed by Bland–Altman analysis. The coefficient of variation of alpha angle was obtained by dividing the standard deviation by the mean and multiplying by 100%. The difference was considered significant at a P value < 0.05. The spatial location relationship of the transducers was measured in ImageJ 1.47v (National Institutes of Health, USA). The statistical analyses were performed in SPSS 22.0 (IBM) and Prism 8.0 (GraphPad).

Result

A total of 225 infants (median age,4.0 months [interquartile range, 3.3–5.2 months]; 104 boys) were enrolled into the study with complaints of asymmetric folds (n = 205), breech presentation (n = 15), a hip click (n = 3), and family history of developmental dysplasia (n = 2). In our study, asymmetric skin folds was the most common reason for referral although it is not a risk factor for DDH [10,11,12,13,14]. Two infants were excluded because the large ossification center of femoral head blocked the ultrasonic beam and prevented visualization of the lower limb of ilium and measurement of alpha angle. Finally, 450 hips were included in our study, 400 hips were examined by the senior operator (C.Z.), 50 hips were examined by the junior operator (Y.Z.). According to Graf's criteria, the hips were diagnosed as normal hips (n = 432) and DDH (n = 18) in our study. According to AIUM’s method, no instable hip was diagnosed.

The mean value of alpha angle measured on the AIUM standard coronal plane (62.1°; 95% CI 61.7, 62.4) was smaller than that on the European standard coronal plane (63.4°; 95% CI 63.1, 63.7) (P < 0.0001, Fig. 2E, F). The standard deviation of alpha angle measured by AIUM method (3.8°) was greater than that by European method (3.0°). The coefficient of variation of alpha angle measured by AIUM method (6.2%) was larger than that by European method (4.7%). These indicated the relatively large dispersion of the alpha angle measured by AIUM method in our study (Fig. 3). The 95% limits of agreement between alpha angles measured by AIUM method and by European method were − 6.6 to 9.2 degree (Fig. 4). Similar results were observed for the cases imaged by the junior operator (Y.Z.) (Figs. 3, 4).

Fig. 3

The mean value of alpha angle measured with AIUM method was smaller than that with European method. A Measured by senior operator (n = 400), B measured by junior operator (n = 50). **< 0.0001, AIUM, American Institute of Ultrasound in Medicine

Fig. 4

Bland–Altman graph of alpha angle difference versus average angle between alpha angle measured with European method and AIUM method. A Measured by senior operator (n = 400), B measured by junior operator (n = 50). SD = standard deviation, AIUM, American Institute of Ultrasound in Medicine

There were 23.7% (101/425) hips with alpha angle smaller than 60° measured with AIUM method, while with European method, there were only 4.2% (18/425) hips with alpha angle smaller than 60°. Our analysis revealed 19.5% (83/425) hips were inaccurately classified as DDH using alpha angle measured with AIUM method. With Graf’s technique as a reference standard, alpha angle measured by AIUM method as a solely index could diagnose DDH with low sensitivity (55% [10 of 18 hips; 95% CI 31, 77]) and specificity (77% [313 of 407 hips; 95% CI 73, 81]).

The position of transducer arranged on the skin of infant by AIUM method was 1.6 ± 0.3 cm back to the position of the transducer arranged by European method, with the inferior edge of the transducer apart in an angle of 16.4° ± 6.6° (Fig. 2C). The transducer and the beam of ultrasound should be vertical and do not tilt by European method, while by the AIUM method, the transducer was tilted anteriorly with the ultrasound beam in a postero-anterior direction (Fig. 2A–D).

Discussion

Acetabular alpha angle has been widely used to evaluate morphology of acetabulum for assessment of DDH by ultrasound and recommended by several guideline or consensus [10,11,12]. Different standard section planes to measure alpha angle have been described by European consensus [10] and AIUM guideline [11], which may potentially lead to inconsistencies of the value of alpha angle and impact the diagnosis of DDH and choice of strategy. In this study, hips of infants were imaged according to techniques recommended by European consensus and AIUM guideline. The mean value of alpha angle measured on the AIUM standard coronal plane (62.1°; 95% CI 61.7, 62.4) was smaller than that on the European standard coronal plane (63.4°; 95% CI 63.1, 63.7) (P < 0.0001). The 95% limits of agreement between alpha angles measured by European method and by AIUM method were -6.6 to 9.2 degree. With Graf’s technique as a reference standard, alpha angle measured by AIUM method as a solely index could diagnose DDH with low sensitivity (55% [10 of 18 hips; 95% CI 31, 77]) and specificity (77% [313 of 407 hips; 95% CI 73, 81]).

Smaller values of alpha angle were obtained by AIUM method in our study. The 95% limits of agreement between alpha angles measured by AIUM method and by European method (− 6.6 to 9.2 degree) was larger than the intra-reader differences of the alpha angle for European method (− 5.1 to 5.6 degree) by the study of Peterlein and colleagues [15]. This indicates that the difference between the alpha angles measured by these two different methods is not solely due to measurement bias. Such differences resulted in misdiagnosis of DDH in 19.5% (83/425) hips by using alpha angles measured with AIUM method in our study.

Blake and colleagues [16] compared alpha angles between coronal ultrasound views with hips in flexion versus neutral positions. Alpha angle decreased on coronal flexion views (mean decrease 3.3 degrees). They supposed the slightly different probe orientation relative to the acetabulum when switching from flexion to neutral position resulted in the differences of alpha angles. However, they could not explain why the differences were always in 1 direction (i.e., alpha angle always decreased on coronal flexion views). In our study, the infants were examined with the hips in neutral position. We confirm that the position of transducer arranged on the skin of infant by AIUM method is different from that by European method with hips. The transducer is tilted anteriorly with the ultrasound beam in a postero-anterior direction by AIUM method. This section is similar to the nonstandard section produced by tilting the transducer ventrally and cephalally as described by Graf [7], which may result in a smaller alpha angle.

A relatively large dispersion of the alpha angle measured by AIUM method was found in our study. AIUM method does not describe where the transducer should be arranged on the surface of skin at the beginning of the examination (European method emphasized the transducer placed on the greater trochanter of femur), this may lead to variation of standard section planes during image acquisition by operators. What’s more, the anatomical marks defined in the standard coronal section described by AIUM method are relatively less strict than European method, which may lead to more than one standard sections be obtained by AIUM method and cause the greater variability of the value of alpha angle measured in different sections.

In our study, hips in a consecutive sample of infants with a median age about 4 months were examined with ultrasound imaging. The incidence of DDH was 4.2% (18/425) diagnosed with European method. This was approximate to the incidence reported by Dorien et al. that 2.5% patients showed DDH by ultrasound imaging of hips between three and five months with Graf technique [2], while, if the alpha angle measured by AIUM method was used for Graf typing, the incidence of DDH increased fabulously up to 23.7% (101/425) in our study. The sensitivity (55% [10 of 18 hips; 95% CI 31, 77]) and specificity (77% [313 of 407 hips; 95% CI 73, 81]) for alpha angle measured by AIUM method as an only index to diagnose DDH was low. These indicated that alpha angle measured by AIUM method is different from European method and cannot be used for Graf typing of DDH.

Our study has limitations. Firstly, most of the data were obtained from one operator. There may be mutual influence in evaluating the alpha angle between the two methods. To counteract this interaction, we used randomization to determine with which method the infants were examined first. Secondly, we cannot determine which method is better or more conducive to the diagnosis and treatment of DDH based on data in our study. On the one hand, there is still controversy about the need for treatment of mild dysplasia of hip such as Graf type IIb diagnosed by European method [17,18,19]. On the other hand, although the AIUM method does not rely solely on the alpha angle and does refer to acetabular morphology and the effect of passively applied stress, the data from the American payer database showed that ultrasound (by AIUM method we speculated) may not effectively increase DDH detection [20]. What could we confirm was that the alpha angle measured by the two methods was different. Thirdly, variations in hip positioning, infant movement, or transducer placement may introduce significant sources of variability in ultrasound assessments. To mitigate the confounding effects of hip positioning on study outcomes, the infants were examined in the lateral decubitus position with the hips in a neutral position. Parents or guardians were asked to soothe the baby before and during ultrasound imaging. The scan was only done when the baby was calm and not crying or squirming. During data acquisition, the transducer was placed strictly following the recommended guidelines. When a standard coronal view was obtained, the transducer's position was recorded for later analysis to reduce the impact of different handling variations on the results.

In conclusion, alpha angle measured by AIUM method is smaller than that by European method and should not be applied to Graf typing of DDH. The method recommended by AIUM should be avoided when evaluating alpha angle.