| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
University of California, San Francisco San Francisco, CA
94117
Ambroise Paré Clinic Neuilly-sur-Seine, France
University of California, San Francisco San Francisco, CA
94117
We read with interest the article by Ecklund and Jaramillo [1] in the April 2002 issue of the American Journal of Roentgenology. The aim of the authors was to determine the value of MR imaging, especially fat-suppressed three-dimensional spoiled gradient-recalled echo sequences, in identifying patterns of growth arrest after physeal insult in children. They present a comprehensive explanation with a systematic approach to analyzing bone bridges with regard to the needs of the surgeon. Nonetheless, the body of the article contains statements that are unclear and should be further discussed.
The most important point concerns the distribution of physeal bridges by anatomic site and cause, reported in table 1 [1]. The authors report that six proximal tibias and—surprisingly—one distal femur were affected by Blount disease. A review of the most comprehensive articles published on this subject [2,3,4,5] makes it clear that in Blount disease, also called idiopathic tibia vara or osteochondrosis deformans tibiae, the delay in ossification and bone growth affects only the medial tibial epiphysis. By definition, Blount disease, as mentioned by Erlacher in 1922 and described by Blount in 1937, is characterized by a local disturbance of the growth of the medial aspect of the proximal tibial epiphysis [3].
We recently reviewed the case of an 18-year woman that supports these findings. This patient has a long history of Blount disease, diagnosed at the age of 10 and never treated. A photo of her lower limbs (Fig. 1A) shows a substantial and bilateral varus deformity. Although she has apparent late-stage Blount disease, the anteroposterior radiographs of her knees (Figs. 1B and 1C) show bilateral marked irregularity of the medial tibia consistent with Langenskiöld type VI changes [4]. They also show marked varus deformity of both lower legs and lateral subluxation of both tibias. In this patient, as in all previously published cases to our knowledge, femur abnormalities are not found in association with Blount disease. Thus, we are greatly puzzled by the attribution of distal femur abnormality to Blount disease in the article [1]. Such a surprising statement requires further explanation and documentation.
|
|
|
The second point regards the authors' interesting approach to quantifying bony bridges by means of maximum-intensity-projection MR imaging. The inner tracing on figure 3D [1], which indicates the bony bridge, seems too small. Unlike the authors, we find it difficult to differentiate between the bony bridge and the normal small mammillary physeal undulations on this image, especially in the adjacent dorsomedial low-signal area. It is clear that the bony bridge, as described by the authors, extends into the dorsomedial quadrant. The inner tracing in this case must be more irregular, as it is on figure 1A [1]. On the other hand, it is not at all clear why two independent reviewers were needed to analyze the images if interobserver differences were not quantified. The interobserver difference for this method would be interesting to know in regard to the problem of defining the extent of bony bridges on maximum intensity projection.
Finally, in the results section [1], the authors report a significant correlation (p < 0.008) between T1 signal intensity and the size of the bone bridges. Yet they conclude the section with the apparently contradictory statement, "There were no statistically significant associations between the bony bridge characteristics, such as size, location, T1 signal intensity, Salter-Harris fracture type, or patient sex." Clarification is needed.
To conclude, this articleis interesting and informative, but we believe the points we have raised tend to weaken the authors' argument and require further explanation.
References
Harvard Medical School Children's Hospital Boston, MA
02115
Harvard Medical School Massachusetts General Hospital Boston, MA
02114
We thank Guermazi et al. for their comments and for their meticulous review of our work [1]. The following is a point-by-point response to their letter.
First, we agree that Blount disease is primarily a disease of the proximal tibia. The importance of distal femoral changes, however, is well known [2, 3]. These changes have become more apparent with MR imaging, which often displays changes in the distal femoral physis [4]. Most of the recent reviews of Blount disease suggest that the disease, particularly in its infant form, is related to medial compression forces across the medial aspect of the knee [5]. In this light, it is not surprising that abnormalities of the femoral physis and other medial structures in the knee, such as the medial meniscus, can accompany tibia vara. In our experience of more than 20 cases of Blount disease studied on MR imaging, the most common femoral abnormality is a focal widening of the distal femoral physis (Fig. 2). We agree that a bony bridge in the distal femur is unusual, but we suspect that as MR imaging becomes more prevalent, more femoral changes in Blount disease will be reported.
|
Second, the use of maximal intensity projection of spoiled gradient-echo MR images for bony bridge quantification was pioneered by Borsa [6]. We agree that discrepancies in quantification can be found, although it is unlikely, even following the tracing described by Dr. Guermazi, that there would be differences of more than 25%, which was the threshold established in our article. Smaller differences are usually not clinically significant. The tracing in the illustration was obtained at a workstation by manipulating the image window and level; the inevitable loss of information between the digital and the printed images may account for some of the discrepancies mentioned by Guermazi et al.
Third, we agree that it would have been ideal to do an evaluation of the interobserver variability in this technique. The purpose of having the two observers, however, was not to evaluate the bridge size quantification technique but to obtain a consensus regarding location of the lesion, size (in 25% intervals), and MR imaging characteristics of the bony bridges.
In conclusion, we thank Guermazi et al. for pointing out a mistake in the last paragraph of the results section. In the revised manuscript we submitted, the text read, "There were no statistically significant associations between the bridge size, location, or T1 signal intensity and the Salter-Harris fracture type or the patient gender." This agrees with the statement in the abstract that reads "Signal characteristics and bridge size did not vary with the cause" and is what we meant to say.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
