Chondro-osseous border in baby hip ultrasonography for developmental dysplasia of the hip: an indispensable litmus paper for the accuracy of scientific publications

Abstract

Purpose

The aim of our study was to use the chondro-osseous border (COB) as an indispensable assessment criterion to evaluate the quality of baby hip ultrasonography (US) images in the literature pertaining to the application of Graf’s technique.

Materials and methods

Our literature review search yielded 144 articles. Of these, 41 contained images that were stated to be based on the application of Graf’s technique. Two reviewers, a radiologist and an orthopaedic surgeon, both course instructors for the use of baby hip US for developmental dysplasia of the hip (DDH), independently evaluated the articles to assess the diagnostic validity of the published images on the basis of a single criterion: the identification of the COB.

Results

Of the 41 articles which were analyzed, 15 contained images without a COB, which corresponds to 36% (roughly, one out of three). Articles from countries where universal screening is performed (Germany, Italy, Switzerland, Mongolia) were all correct. All the articles from Turkey, a country which has made significant contribution with published material about Graf’s technique over many years, and most of the articles from China (80%), where Graf’s technique has become popular in recent years, contained correct images.

Conclusion

Published literature lacks strict criteria for the publication of correct images corresponding to Graf’s technique in baby hip US. This fact raises concerns about the quality of published material (and, as a corollary, the quality of clinical application of the technique) and should have ramifications on scientific journal policies regarding the publication of such manuscripts.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40477-023-00821-4.

Introduction

Originally described by Dr. Reinhard Graf in 1980, Graf’s technique in baby hip ultrasonography (US) has been in use for more than 40 years [1]. Universal screening for developmental dysplasia of the hip (DDH) with Graf’s technique is being performed in Central European countries, including Germany [2], Austria [3], parts of Northern Italy and Switzerland [4]. Ultrasonographic population screening has been exported to non-European countries, Mongolia being a recent example [5].

One of the main strengths of Graf’s US technique is the fact that it consists of clearly discrete steps, which have been thoroughly described by Graf [6, 7]. In brief, the main steps of the diagnostic procedure are: (1) baby placement and examination using the cradle and probe handle; (2) anatomical identification (Checklist 1); (3) usability check (Checklist 2); (4) morphological classification; and (5) final classification. The first three steps aim to safeguard the quality of the scan. Anatomical identification and usability check are the quality control checklists which ensure whether the produced image may be used for diagnosis or not.

Anatomical identification (Checklist 1) is a straightforward procedure which confirms the presence of specific anatomical structures on the image (Fig. 1). It is mandatory that all those structures are identified on the image (with only a few exceptions which are beyond the scope of this publication and refer to decentered/dislocated hips). The first structure which should be identified in a scan is the chondro-osseous border (COB). The COB is the border between the cartilaginous proximal femur and the bony diaphysis, anatomically corresponding to the transitional zone of the metaphysis, the area which is responsible for the longitudinal growth of the bone (Fig. 2). The ultrasonographic form of the COB depends on the age of the baby, and has been thoroughly described [6, 7](Fig. 3).

Fig. 1.

Anatomical identification (Checklist 1) refers to the recognition of specific anatomic entities on the sonographic image of the hip, which must be identified on every US image, to confirm that it may be used for diagnosis (COB chondro-osseous border, FH femoral head, SF synovial fold, JC joint capsule, L labrum, HC hyaline cartilage of the roof of the acetabulum, BR bony roof, cc concavity, cv convexity, TP turning point)

Fig. 2.

Histological section of the right hip of a neonate. Arrows demonstrate the exact position of the chondro-osseous border (FH femoral head, SF synovial fold, JC joint capsule, L labrum, HC hyaline cartilage of the roof of the acetabulum). (Used with permission from Graf R, Baumgartner F, Lercher K. Ultraschalldiagnostik der Säuglingshüfte. Ein Atlas. 2nd Edition. 2012. Springer-Verlag Berlin Heidelberg)

Fig. 3.

The ultrasonographic form of the chondro-osseous border depends on the age of the baby and has been thoroughly described [6]: In neonates it is curved (a), in young infants medially intermittent (described as “palisades”) (b) and in older infants the medial part is obscured by the echo shadow of the lateral part (c)

The importance of the correct characterization of the COB is two-fold: (1) It is the very first anatomical structure which must be identified in every scan (Checklist 1), regardless of the hip type according to Graf, and (2) it may appear distorted in the case of “tilting errors”, which form a significant source of misdiagnosis, the most important of which is due to a caudo-cranial tilt of the probe (Fig. 4). No image without a clearly defined COB may be used for diagnosis, so it is very important to document the COB in any US image of the baby hip for DDH, otherwise the US examination has to be repeated (Fig. 5).

Fig. 4.

Absence or distortion of the chondro-osseous border is the characteristic feature of a caudo-cranial probe tilt. Figures 4a and 4b both belong to the same patient; however, a is a standard plane image, while b is a tilted image (caudo-cranial). Obviously, there is no discernible chondro-osseous border in (b)

Fig. 5.

According to the Graf technique, no image without a clearly defined chondro-osseous border may be used for diagnosis in baby hip US for DDH. It is crucial to document the chondro-osseous border in any scan image, otherwise the examination needs to be repeated. In a, b and c, there is no chondro-osseous border, rendering all of them non-diagnostic

The aim of our study was to use the COB as an important assessment criterion to evaluate the quality of the images in the published literature, which claim to have demonstrated the application of Graf’s technique.

Materials and methods

On May 5, 2023, we searched PubMed using the terms “hip ultrasound Graf”. Our search retrieved 144 articles. The search results were saved. We accessed the articles which were retrievable and ended up with 41 articles which contained images stated to be based on application of Graf’s technique.

Of the studies that we analyzed, two were published in 2017, seven in 2018, six in 2019, five in 2020, six in 2021, thirteen in 2022 and two in 2023. The countries of origin of the articles were Brazil, Canada, China, France, Germany, Switzerland, India, Italy, Japan, Mexico, Mongolia, Netherlands, Poland, Romania, Saudi Arabia, Singapore, Spain, Turkey, United Kingdom, and USA. Most of the 41 articles were from China (n = 10) and Turkey (n = 7).

Two reviewers, a radiologist, and an orthopaedic surgeon, both baby hip US course instructors with 20 and 10 years of hands-on experience, respectively, in the Graf technique, independently evaluated the articles to assess the diagnostic validity of the published images on the basis of a single criterion: the identification of the COB. In 36 out of the 41 articles, both readers agreed on the first read. After a second read of the 5 articles where the observers differed in their opinion, they ended up in consensus in all.

Results

Due to legal restrictions (copyright) and our concern for collegiality, examples from the contents of the articles are not presented here. However, a table including the digital object identifiers (DOIs) or PubMed identifiers (PMIDs) of the articles and our judgement about the presence or absence of the COBs is available online (Online Resource).

Out of the 41 articles which were analyzed 15 contained images without a COB, which corresponds to 36% of the articles. Articles which came from countries where nationally or locally generalized screening is performed (Germany, Italy, Switzerland, Mongolia) were all correct. There was no publication from Austria and only two publications came from Germany, a fact which may be explained by the long history of application of Graf’s technique (so data from these countries have already been published in the past). All the articles from Turkey (100%), a country which has made significant contribution to literature on hip US for DDH with published material about Graf’s technique for many years, and most of the articles from China (80%), a country where Graf’s technique has become very popular in recent years, contained correct images.

Discussion

Our study shows that published literature referring to the application of the Graf technique, which has been thoroughly described and is firmly safeguarded by quality criteria (i.e., anatomical identification and usability check), or commenting on therapeutic decisions and results based on this technique, is still far from satisfactory. This implies that a significant number of published studies, which do not fulfill the diagnostic criteria mentioned above, as evidenced by the published scan images, are of questionable value. This is particularly true in the case of articles which claim to evaluate reliability and reproducibility of the Graf technique. The same may apply to the clinical conclusions (application and evaluation of treatment).

It has been documented that the performance of the examiners who practice Graf’s technique heavily depends on the application of precisely defined examination steps, which should be followed in a ritual way [6, 7]. Hands-on training schemes are organized nowadays all over the world, with proven effectiveness [8]. A longstanding training scheme is being organized in Germany and Austria by the Ultrasound Academy of the German Society for Ultrasound in Medicine and the Austrian Society for Ultrasound in Medicine, respectively [9, 10]. Inadequately trained examiners are by far the most common reason for the erroneous application of the technique [11].

It should be mandatory that the images which are published in the literature fulfill the diagnostic criteria which have been already clearly described. The fact that published literature has an educational function and also helps set standards for quality of practice makes the need for strict reviewer and editorial vetting of the submitted manuscripts imperative.

Our study has several limitations. Firstly, only the presented images in the articles that were retrieved after the PubMed search were evaluated (not all the articles contained a large number of images). However, it is reasonable to assume that the authors published their most representative images. Secondly, we used only two observers and their eventual consensus. However, both observers in this study have not only undergone structured courses for baby hip US, but also have been course instructors.

Conclusion

Despite the long-established use of Graf’s technique for US screening and US diagnosis-based treatment of DDH (nearing half a century), published literature still lacks strict criteria for the publication of correct US images corresponding to this technique. This fact raises concerns about the quality of the published material, and subsequently, the validity of clinical application of the technique. It should be mandatory that clinical examiners receive appropriate training by well-organized training courses, which are held in many countries, and that they strictly apply the principles of the technique without exemptions or compromise. Journal reviewers should also have due awareness of the Graf technique for best critical appraisal of submitted articles featuring this method.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by KC and JO’B. The first draft of the manuscript was written by KC and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

Data availability statement

All data generated or analysed in the present study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose. The authors have no competing interests to declare that are relevant to the content of this article.

Ethical approval

No ethical approval is required for this study. No Institutional Review Board approval was applicable for this study, which entailed a literature review.