- Email: [email protected]
Description of the potential of an arthrometer for standard and reduced radiographs suitable to measurement of angles and segments of hip, knee, foot and joint space widths
ORIGINAL ARTICLE
Joint Bone Spine 2002 ; 69 : 282-92 © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved S1297319X0200372X/FLA
Description of the potential of an arthrometer for standard and reduced radiographs suitable to measurement of angles and segments of hip, knee, foot and joint space widths Michel Lequesne1*, Gérard Morvan2 1 Rheumatology department, hôpital Léopold Bellan, 75014 Paris, France; 2cabinet de radiologie, 5, rue Alfred Bruneau, 75116 Paris, France
(Submitted for publication May 10, 2001; accepted in revised form September 5, 2001)
Summary – Background. Primary osteoarthritis is usually selected in either epidemiological or therapeutic studies. This implies exclusions. Among cases of secondary osteoarthritis considered for either stratification or exclusion – or for prognosis and treatment in daily practice – are those due to architectural defects. Parameters of the latter should be measured to ascertain diagnosis. At present, measurements have to be performed either on digitized reduced films or standard radiographs. Objective. To finalize an instrument capable of measuring the main angles and segments characteristic of the main dysmorphisms of the hip, knee and foot on different sizes of films. Methods. An arthrometer drawn on transparent material to be placed on radiographs was designed, involving several appropriate protractors and millimetric scales; it was tested on 60 hip, 35 knee and 17 foot radiographs with various architectural defects. Angles and segments most often used according to literature were measured. Reduction rates of films were various, reflecting the range of radiograph sizes currently used in everyday practice. Results. Measurements were easily performed on radiographs from standard (100%) up to 50% of reduction rate. So the arthrometer allows the recognition, especially in moderate, not obvious forms, of the following developmental or acquired dysmorphisms: hip congenital dysplasia and subluxation, including coxa valga and neck excess of anteversion; acetabular protrusion and coxa vara; tilt deformity; knee: patellar height abnormalities, patellar maltracking, trochlear depth insufficiency; foot: pes cavus, flat foot. Angle and segment ratios do not change in reduced film, whereas segments (absolute lengths) obviously should be converted according to the reduction rate for retrieving the classical values established for decades. Conclusion. The arthrometer allows us to measure the relevant parameters of various dysmorphisms without drawing lines over the films themselves. It is suitable to reduced as well to standard radiographs. However, only the instrument and its ability to be used in various films sizes are here presented. Study of reproducibility of measurements – especially concerning the joint space width – remain to be performed. Joint Bone Spine 2002 ; 69 : 282-92. © 2002 Éditions scientifiques et médicales Elsevier SAS arthrometer / dysmorphisms / foot / hip / joint space width / knee / radiographic measurements
* Correspondence and reprints: 33 rue Guilleminot 75014 Paris, France. E-mail address: [email protected] (M. Lequesne).
Radioarthrometry: hip, knee, foot, joint space widths
Diagnosis and evaluation of the main developmental or acquired architectural defects in joints of lower limbs are of importance in daily practice as well as in scientific works. Measurements of dysmorphisms are interesting for orthopaedic surgeons, radiologists and rheumatologists. In many studies, notably epidemiologic surveys and therapeutic trials, the distinction between primary and secondary osteoarthritis (OA) of the lower limb joints are mandatory as prognosis and treatment are different. Among secondary OA, those related to a developmental dysplasia, as acetabular insufficiency or protrusion, coxa valga, patellar subluxation, patella alta, trochlear dysplasia, major varus or valgus deformity of the knee, for example, should be carefully considered, especially with regard to prognosis [1, 2] and treatment, in daily practice, in hip [2] as well in knee [3] OA. A faint developmental defect is often doubtful; it should be measured to be ascertained. In its frank forms, measurements are needed for planning surgical corrections [4-6]. In a recent editorial on hip dysplasia, about ten studies involving measurement of certain angles are cited [5]. Nowadays, we have to deal most often with digitized radiographs, reduced according to variable proportions: 75%, 66%, 50% or certain intermediate values. However, standard plain films (‘100%’) are far from having vanished and are claimed by many orthopaedic surgeons. Thus, a tool capable of measuring angles and segments and joint space widths either on standard or on reduced films appears necessary. MATERIAL AND METHOD We designed an instrument printed on a transparent material, suitable for both standard and digitized films reduced by up to 50%. A first model was made especially for all the measurements related to hip malformations, congenital or acquired [6]. Now, we are presenting an arthrometer1 appropriate for assessing main architectural radiographic abnormalities of hip, knee and foot and for appraising joint space (JS) width. It involves several designs of protractors and of millimetric rulers drawn on transparent material. This instrument is put on the X-ray in the appropriate location. Every measure is read directly on it, therefore marking lines on the radiograph is avoided. 1
The arthrometer (© M. Lequesne, 2001) is distributed by Transfert International, 15 rue du Louvre, 75001 Paris, FR. Tel (33, 1) 44 88 20 44 ; Fax (33, 1) 44 88 20 40.
283
The design was carried out by trying different drafts of the arthrometer on radiographs of standard size (100%) and of various coefficients of reduction reflecting the range of radiographic sizes currently used in daily practice down to 50%. Instrument drafts were tested on 60 hip, 35 knee and 17 foot radiographies to achieve the final design. The latter comprises a protractor with concentric circles for measuring VCE and VCA angles of the hip (figure 1, left) and a larger protractor for measuring all the other angles of hip, knee and foot (figure 2, right). Points C and C’ are punched for allowing marks with a pencil tip on the center C of the femoral head and the middle C’ of the femoral neck. Three millimetric rulers are provided at 100%, 75% and 66%, so segments lengths and JS widths may be read directly on film of analogous size; in case of 50% reduction, measurement is performed with the 100% ruler and divided by two. If the reduction rate is neither 75%, 66% nor 50%, the lengths found should be converted according to the intermediate reduction rate. Radiologists usually mention it. However, in numerous cases, mention of the reduction rate is replaced, on a side of the film, by a scale over which each division (to be measured) stand for 1 cm (i.e, if one division length is 0.66, it means that the reduction rate is 66%). Value ‘100%’ is the classical one, coming from decades of studies. Normal values (and certain borderline values) will be indicated. Let us recall that angles to be measured and length ratios do not change on digitized reduced films. Protractors of the arthrometer were designed for the right hip. For the left hip, either the instrument or the radiograph has to be flipped over RESULTS: MEASUREMENT OF THE MAIN DYSMORPHIES The hip Congenital dysplasia and subluxation of the hip These two degrees of the developmental dysmorphism are found as a single or associated cause of hip OA in 20–40% of patients [1, 2, 4, 5, 7-12]. In well-chosen cases, it is possible to correct the architectural defect by osteotomy and/or shelf operation. Dysplasia and subluxation involve three dysmorphic components: acetabular insufficiency, excess of verticality, excess of anteversion of the femoral neck. However, often only one or two of these three flaws are present. Combinations are different from one patient to another and should be evaluated systematically.
284
M. Lequesne, G. Morvan
Figure 1. The arthrometer (©M. Lequesne 2001) drawn on transparent film.
Figure 2. Sketch of an anteroposterior view radiograph of a tilting pelvis in the standing position. Right hip: acetabular insufficiency: VCE 18°, HTE 15°. Left hip: normal.
On the anteroposterior (AP) view On the coronal view taken in standing position, the lower limbs being in medial rotation 15° ± 5, the angles VCE, CC’D, HTE (figure 2 right hip) and the acetabular depth are measured. The VCE angle measures the lateral covering of the femoral head by the acetabular roof in the functional position (normal value [NV]: ≥ 25 ), V being the vertical line, parallel to the lateral edge of the film, drawn
Figure 3. The arthrometer in place for measuring VCE angle.
through C, which is the center of the femoral head. The center C correct location is determined by placing the appropriate circle or a between-circles interval on the femoral head contour. VCE value is read directly on the protractor (figure 3). Before moving the instrument on the X-ray for the next measurements, the center C
Radioarthrometry: hip, knee, foot, joint space widths
285
should be marked on the film by inserting the tip of a pencil into the punch at point C of the arthrometer. Likewise, the center C’ of the femoral neck should be marked by sliding the instrument’s millimetric segment C’ so as to place it over the narrowest portion of the femoral neck. The middle C’ of the latter is marked on the film, as it was done for C, with a pencil tip. The CC’D neck-shaft angle (NSA) measures the incline of the neck on the shaft of the femur. Possible dysmorphisms are coxa valga: CC’D ≥ 140 , and coxa vara: CC’D ≤ 120 ; intermediate values are normal or borderline (table I).CC’D is measured by placing the arthrometer vertically, with the line TD centered on the femoral shaft axis. The arthrometer is moved gradually until point C and C’(axis of the neck) previously marked on the film coincide with a line of the large protractor or falls between two lines (figure 4). Then the value of the CC’D angle is read directly. This measurement is correct only if the lower limb was positioned in about 15° internal rotation, so the neck will be approximately in the coronal plan. The HTE angle measures acetabular roof acclivity (NV: ≤ 10°). It is formed by (i) a ‘true’ horizontal line (parallel to the upper edge of the film) through point T, and (ii) the TE line, T and E being medial and lateral limits of the acetabular roof (condensed thick line) Table I. Values of the main angles and segments characteristic of the architecture of normal and dysmorphic hip, knee, foot on radiographs. Hip VCE NSA (CC’D) HTE VCA Neck anteversion Acetabul. anteversion Acetabul. depth When a c medial to ii (fig. 8) Knee Tibiofemoral angle Trochlear angle Foot Djian-Annonier angle
Normal*
Dysmorphic*
≥ 25° 120°–137° ≤ 10° ≥ 25° 7–15°
< 20° > 140° > 12° < 20° > 20°
20°–25° ≥ 9 mm F: 0–5 mm M: 0–2 mm
> 25° < 9 mm ≥ 6 mm ≥ 3 mm
0° ± 3° 125°–143°
≥ 4° ≥ 145°
125° ± 5°
≤ 115° ≥ 135°
* In this table, the gap between normal and dysmorphic values corresponds to ‘borderline’ architectures, of which the harmful role is doubtful. Abbreviations: F: females; M: males; NSA: neck-shaft angle.
Figure 4. Measurement of the neck-shaft angle CC’D: when the line TD of the arthrometer is superimposed on the diaphysis axis, the instrument should be moved vertically until one of the rays of the large protractor will be on or close to the CC’D axis of the femoral neck. In this example, point T of the arthrometer is on the top of the neck-shaft angle by chance.
respectively. It is measured by the left part of the large protractor of the arthrometer staying vertical after having placed the T correctly on the film (figure 5). Acetabular depth (NV ≥ 9 mm) is the perpendicular distance from the deepest point of the acetabulum to the EP line, where P is the ipsilateral upper corner of the symphysis pubis [4]. It is measured using the millimitergraduated segment perpendicular to TD, the line TD being placed over the dotted line EP on the film (figure 6). On the false profile view This view is a partially oblique view of the hip taken in standing position with the pelvis rotated backward 25° from profile, the lower limb close to the radiologic table staying stable with the axis of the foot remaining paral-
286
M. Lequesne, G. Morvan
Figure 5. Measurement of the HTE angle.
Figure 7. Sketch of false profile of the right hip [13], VCA: normal. VCA’ (dotted line): anterior insufficiency of coverage (VCA’ = 17°).
On the CT scan view This view is performed when an excess or a lack of anteversion of the neck is likely according to certain clinical and/or radiological features. The femoral neck anteversion angle (NV: 7–20°) and the acetabular anteversion angle (NV: 20–25°) are usually measured and noted by radiologists.
Figure 6. Acetabular depth and its measurement by segment a d. An abnormal shallow acetabulum was drawn here.
lel to the table [13]. The VCA angle (NV: ≥ 25°) measures the anterior covering of the femoral head by the acetabular roof, of which the anterior radiological edge is named A. V is the vertical line drawn through C. Measurement is made as for VCE, point C having been determined with the help of the concentric circles applied on the femoral head (figure 7).
Acetabular protrusion Found in about 5% of patients with hip OA [7, 8]. It consists of an excess of depth of the acetabulum: VCE and VCA angles are usually ≥ 45°, HTE angle between 0 and –10°, with a frequent coxa vara: CC’D ≤ 120°. However, the main criterion is the location of the acetabular line ac medial to the ilio-ischiatic ii line beyond the following distances: ac–ii (figure 8) is ≥ 6 mm in women, and ≥ 3 mm in men with protrusio [14]. This ac–ii measurement is valid exclusively on full pelvis radiograph, not on a single hip view, which is able to create by itself an excessive inner projection of ac line. Tilt deformity The sequela of a mild slipping of the femoral epiphysis during adolescence, usually painless at the time of occurrence and often resulting from athletic activity [15]. The severity of this acquired abnormality can be
Radioarthrometry: hip, knee, foot, joint space widths
Figure 8. (a) Acetabular protrusion is measured by the gap between the ilio ischiatic line ii and the acetabular line ac when the latter is medial to ii. (b) Tilt deformity: center C of the femoral head is below the neck axis C’C” (≥ 2 mm) (these two independent dysmorphisms are drawn here together in only one sketch).
evaluated as indicated in figure 8. Axis of the neck drawn from C’ and C” (middle of the neck) and extended proximally normally traverses the femoral head close to its center C. Tilt deformity, also called caput varum, is testified by a point C below the neck axis of at least 2 mm. THE KNEE Panoramic anteroposterior hip-ankle weightbearing view Angulation deformity Angulation deformity of the tibiofemoral knee joint (genu varum, genu valgum) is the most frequent dysmorphism. Normally, the axis of the tibia is the continuation of the mechanical axis of the femur which has to be drawn on the radiograph (figure 9). However, an angle of ± 3° is considered as tolerable [16]. Measurement is made on a set of three radiographs taken in the standing position: pelvis, femurs, and tibias, included the ankles. Measurement is performed with the help of the upper part of the large protractor, as shown in figure
287
Figure 9. The arthrometer in place for measuring the tibiofemoral angle (here: genu varum 7).
9, point T being placed in the middle of the tibial interspinous feature. Lateral view Patella height Patella height (figure 10) is measured according to one of the following ratios: (i) the Insall-Salvati index modified Grelsamer [17] is the ratio A-T1/A-B, where A and B are the inferior and superior limit (respectively) of the articular surface of the patella, and T1 the tibial tuberosity. NV: 1.5; (ii) the Blakburne-Peel index [18] takes as inferior basis the tangential line to tibial plateaus. The perpendicular from A to this tangent (A-P) is the numerator. The ratio A-P/A-B is normally 0.8; (iii) the Caton index A-T2/A-B, where T2 is the anterosuperior angle of the tibia, is normally 1 ± 0.2. It is ≥ 1.2 in patella alta and between 0.6 and 0.8 in patella infera. This index is valuable whatever the knee flexion between 10–80°, whereas other indices are valid only on the knee flexion 30° [19]. Depth of the trochlear groove This may be evaluated and measured on the lateral view. Normally, the line marking the median trochlear
288
M. Lequesne, G. Morvan
Figure 10. Lateral view of the knee. Parameters of the patella heights (see text). A-B: outline (length) of the articular surface of the patella. T1: tibial tuberosity; T2: anterosuperior edge of the tibia.
Figure 12. Axial view, knee flexed 30°. The arthrometer is in place for measuring the sulcus angle; its point T should be on the trochlear groove.
groove is largely and consistently behind the contour of the lateral trochlear border (figure 11). When it runs closer to this trochlear border in its upper part (dotted line), it is the sign of a depth insufficiency in the superior part of the trochlea. In most dysmorphic cases, this line crosses the anterior contour of condyles. More generally, the measurement of the trochlear depth on the lateral view was proposed as the distance between trochlear groove line G and condylotrochlear margins M, the lateral and medial ones being superimposed. Measured 1 cm below the upper border of the trochlea, the distance MG is normally of 5.94 ± 1.74 mm. It is < 5 mm in most of the pathological cases [20]. Radiographic axial view (knee flexed 30°)
Figure 11. Lateral view. The trochlear groove line visible through condyle and trochlea; continuous line: normal location; dotted line: depth insufficiency. M: margin; G: groove of trochlea.
Trochlear dysmorphism Trochlear dysmorphism with insufficiency of depth is measured by the trochlear or sulcus angle L G M, where L and M are the lateral and the medial trochlear edges respectively and G the trochlear groove. The measurement is formed as shown in figure 12: the film being rotated 90°, patella on the right, the point T of the arthrometer is placed on the trochlear groove G, its line TD staying on the trochlear wall placed inferiorly. NV: 125–143°; dysmorphism: > 143°.
Radioarthrometry: hip, knee, foot, joint space widths
289
Patella malpositions On the axial view (knee flexed 30°) may be seen patellar tilting (figure 13B), and patellar subluxation (to be measured as the gap between the two dotted lines, figure 13C). Sometimes, these two components are associated together (figure 13D) and also with patella alta. (Normal feature seen in figure 13A.) THE FOOT Weight-bearing profile
Figure 13. Axial view, knee flexed 30°. Patella malpositions. A: normal; B: lateral tilt (maltracking); C: lateral subluxation; D: tilt and subluxation associated.
This radiograph permits the longitudinal arch of the foot to be shown and possible pes cavus and flat foot to be measured [21]. The Djian-Annonier angle (DA) is so formed: its peak is the lowest point of the talonavicular interspace; its two extremities are the downward sloping point of the sesamoid under the first metatarsal head at the front and this of the calcaneus at the rear. NV: 125° ± 5°. In frank pes cavus, it is ≤ 115° and in frank flat foot it is ≥ 135°. The line TD of the arthrometer is placed along the posterior branch of the Djian-Annonier angle on the radiograph and the value of this angle is read directly on the protractor (figure 14). The Meary-Tomeno line (MT) is formed by two segments: the posterior one is the axis of the talus; the anterior one is a line drawn from the center of the first
Figure 14. Weight-bearing lateral view of the foot. The arthrometer is in place for measuring the Djian-Annonier angle DA (here: 123°).Otherwise, MT: Meary-Tomeno line; S: Schnepp angle.
290
M. Lequesne, G. Morvan
Lelievre parabole segment (HL), a smooth curve that joins the fronts of the metatarsal heads. This curve loses its smooth form when the second metatarsal is too long or the first metatarsal is too short. JOINT SPACE WIDTH, PROSTHETIC COMPONENTS AND OTHER MEASUREMENTS JS width may be measured with the help of one of the millimetric rulers chosen according to the radiograph reduction scale. Standard of 100% is obviously better. The appropriate JS subchondral bone limits, usually at the narrowest JS site, are marked with a fine pencil point before placing the ruler. In proceeding carefully under a common lens (magnification × 2 for example), it is possible to measure to within about 0.25 or 0.50 mm according to our experience, though with an error margin of reproducibility which is discussed thereafter Migration and wearing of prosthetic components may be measured according to analogous process. The Cobb angle of scoliosis and other angles and segments are also measurable with the arthrometer. Figure 15. Weight-bearing superoinferior view of the foot. HL: Hofmann-Lelievre curve.
metatarsal head parallel to the upper border of the first metatarsal; this segment is normally in line with the posterior one. Should these two segments form an angle open downward, it testifies pes cavus; an angle open upward testifies flat foot. The Schnepp’s angle (S) is the angle of incidence to ground of the first metatarsal in weight-bearing position. Its superior branch is the axis of the first metatarsus. NV: 17° to 22°. It represents another mean to assess pes cavus (> 22°) and flat foot. It is measured in using the left part of the large protractor, its 0 -T axis on the horizontal line, T point on the top of S angle. Weight-bearing supero-inferior view The same areas of the arthrometer allow us to measure: (i) the angle between the axis of the first and fifth metatarsal (NV: 32°); (ii) the varus of the first metatarsal (angle between the first and the second metatarsal axis) (NV: 10°); (iii) the angle between the first metatarsal and the first phalanx (NV: 10°); and (iv) the angle between the first and second phalanx is on average 5° (figure 15). One of the main criteria of the foot architecture is to be appraised, not exactly measured. This is the Hofman-
DISCUSSION The instrument here proposed involves means for measuring the main angles and segments specific to hip, knee and foot morphology. Certain classical tools, such as either a pocked articulated goniometer or the femorocoxometer already allowed some of these measurements. The arthrometer appears to have the following advantages: it combines in only one tool several means of measurements, appropriate protractors and millimetric rulers; it allows drawing lines on the radiographs to be avoided (except with regard to the tibio-femoral angle), allows verification if they have already been traced. Above all, this arthrometer was designed in such proportions that it suits reduced (up to 50%) as well as standard (‘100%’) radiographs. It involves not only a millimetric ruler of normal size, but also two others at 75% and 66% reduction rates, rates often used by radiologists. The use of the arthrometer comprises certain limits. On the one hand, in rare patients with small stature, the 50% reduction rate yields an image that is too narrow, beneath the protractor rays. In such cases, the answer is to draw on the film the lines (sufficiently extended) figuring the appropriate angles and to apply the arthrometer on them. On the contrary, on non-reduced films of certain persons with large stature, it may
Radioarthrometry: hip, knee, foot, joint space widths
occur – in fact rarely – that the femoral head contour passes over the larger arthrometer circle. However, in placing that circle concentrically to the ‘oversized’ femoral head, it is easy to fix and mark the femoral head center. On the other hand, the degree of correctness of measurement has not been evaluated by intra- and interobserver reproducibility. Perhaps this is not very important in matter of angles; to our knowledge, such validation was not performed in studies on either hip dysplasia or knee dysmorphisms [1-3, 9], probably because corrective surgery, difficult to adjust with an accuracy of 2–3 degrees, does not imply an extreme correctness in baseline scores which lead to the surgical decision. However, the linear measurements should be more accurate, specially those of the JS width and the possible rapid chondrolysis. The reproducibility evaluation of the millimetric rulers, especially the reduced ones, remains to be performed. Indeed, with a ruler of normal size, reproducibility test were done in the 90’ in patients with medial tibiofemoral OA. The SD of the measurement disparities in intraobserver evaluation were from 0.23 to 0.37 mm [22, 23] and the intra-class correlation coefficient from 0.92 to 0.77 [23, 24]: results of measurers were unequal. At the physician’s office, the less capable measurer has to account for a possible error margin of 2 SD, i.e., 0.37 × 2 = 0.74mm. Above this threshold, he will be allowed to register as true difference from one side to the other one in a JS width (incipiens OA) or from one time to another (example: rapid or half a rapid chondrolysis when 1 mm of JS width is lost within 3 to 6 months). Nevertheless, new reproducibility checking should be implemented to ensure the validity of the arthrometer in hip and knee OA. In the frame of possible structural-modulator drug trials, measurement must be done with very more accurate material and methods. Table I shows values of the main angles and segments. Angles – and ratios – do not change with reduction of radiographs. Pelvis inclination on the right or left in anteroposterior X-ray view taken in the standing position, usually resulting from lower limb inequality, includes an issue: the reference basis of angles to be measured may be either the ‘absolute’ horizontal/vertical axis or the horizontal/vertical axis of the pelvis itself. The original Wiberg’s CE angle took as axis 0 the line passing through C and perpendicular to the horizontal axis of the pelvis, i.e., the line joining the two femoral head centers [25]. However, we consider that measurement of the VCE angle from the vertical line and of the HTE
291
angle from the horizontal line is most appropriate for the following reasons. Let us consider the right hip in figure 2: inclination of the pelvis results in a functional uncovering of the femoral head, adding to the effect of architectural uncovering of dysplasia, if any exists. VCE angle measures these two uncovering together. The hyper-pressure in the superolateral part of the hip joint is logically the result of this global uncovering, which plays its role at the beginning of each step, before muscle constriction returns the pelvis to horizontal position. In fact, two arguments suggest that a greater than 1.5 cm leg length inequality is clinically relevant and harmful: (i) superolateral joint space narrowing due to hip OA is more marked in patients with 1.5 cm leg length inequality than in those with legs of about the same length [26], and (ii) hip OA occurs earlier or is more marked on the side of the longer leg (‘long-leg syndrome’) [27]. Conversely, on the side of the shorter leg, a functional ‘overcoverage’ of the femoral head occurs, affording some degree of protection. The exact situation of the E and/or A points can be difficult to determine. Following the anterior edge of the acetabulum through the femoral head on the AP view can help to find point E. Likewise, on the false profile view, following the anterior edge of the acetabulum from bottom to top helps in finding point A. When measuring the VCE and VCA angles of a hip OA with joint space narrowing implying secondary lateral or anterosuperior femoral head migration, the circle of the instrument corresponding to the femoral head should be moved to its original position under the acetabular roof, in order to simulate a joint space of ideal width. This allows the estimation of the VCE and VCA angles that existed before OA occurred. When measuring the NSA angle (CC’D), it is important to check whether a fixed lateral rotation deformity or an omission has prevented placement of the femoral neck within or near the coronal plane. Lateral rotation results in the greater trochanter being superimposed in part on the base of the femoral neck and the lesser trochanter being too prominent medially. This increases the apparent NSA angle on the film. With regard to knee joints, trochlear dysplasia and patella alta are not the only factors of patellar instability. Vastus medialis insufficiency acts in the same way, which results in patella tilting, that quadriceps contraction is able to reduce. More important is the too-lateral situation of anterior tibial tuberosity (AT) in relation to trochlear groove (TG). This gap AT-TG, called tibial tubercle distance, is measured by radiologists on special
292
M. Lequesne, G. Morvan
X-ray or TDM views; NV varies with the flexion angle of the knee and with authors [28]. REFERENCES 1 Cooperman DR, Wallenstein R, Stulberg SD. Acetabular dysplasia in the adult. Clin. Orthop 1983 ; 175 : 79-85. 2 Murphy SB, Ganz R, Muller ME. The prognosis of untreated dysplasia of the hip. J. Bone Joint Surg 1995 ; 77A : 985-9. 3 Cooke TD, Kelly BP, Harrison L, Mohamed G, Khan B. Radiographic grading for knee osteoarthritis. A revised scheme that relates to alignment and deformity. J Rheumatol 1999 ; 26 : 641-4. 4 Murray RO. The aetiology of primary osteoarthritis of the hip. Br J Radiol 1965 ; 38 : 810-24. 5 Lequesne M. Congenital dysplasia of the hip in adults. Is there still room for surgical treatment? Rev Rhum [Engl Ed] 1999 ; 66 : 4-8. 6 Lequesne M. Femorocoxometry: angles and segments characteristic of dysplastic and dysmorphic hip conditions in adults. Rev Rhum [Engl Ed] 1999 ; 66 : 136-42. 7 Falconnet M, Vignon G. Étude statistique, étiologique et clinique de 200 observations de coxarthrose. Lyon Med 1968 ; 17 : 1171-86. 8 Lequesne M. La coxarthrose: critères de diagnostic. Etiologie sur 200 cas. Rôle de la dysplasie congénitale. Peyron J, Ed. Epidemiology of osteoarthritis, Rueil-Malmaison, Geigy 1981 : 198-210. 9 Trousdale RT, Ekkernkamp A, Ganz R, Wallrichs SL. Periacetabular and intertrochanteric osteotomy for the treatment of osteoarthritis in dysplastic hips. J. Bone Joint Surg 1995 ; 77 A : 73-85. 10 Wilson MG, Poss R. Osteoarthritis of the hip. Moskowitz R, Howell D, Goldberg V, Mankin HJ, Eds. Osteoarthritis. Diagnosis and medical/surgical management, 2nd Ed., Philadelphia: W B Saunders 1992 : 621-49. 11 Aronson J. Osteoarthritis of the young adult hip: etiology and treatment. In Instructional course lectures. Am Acad Orthop Surgeons Vol 35. St Louis; CV Mosby 1986: 119-28. 12 Harris WH. Etiology of osteoarthritis of the hip. Clin. Orthop 1986 ; 213 : 20-33. 13 Lequesne M, Larédo JD. The faux profil (oblique view) of the hip in standing position. Contribution to the evaluation of osteoarthritis of the adult hip. Ann Rheum Dis 1998 ; 57 : 676-81.
14 Armburster TG, Guerra J, Resnick D, Goergen TG, Feingold M, Niwayama G, et al. The adult hip: an anatomic study. Part I: the body landmarks. Radiology 1978 ; 128 : 1-10. 15 Murray RO, Duncan C. Athletic activity in adolescence as an etiological factor in degenerative hip disease. J Bone Joint Surg 1971 ; 53 B : 406-19. 16 Massé JP, Glimet T, Alperovitch A, Kuntz D. Valeur de l’angle fémoro-tibial de 244 sujets exempts de gonarthrose. Rev Rhum 1985 ; 542 : 91-4. 17 Grelsamer RP, Meachow S. The modified Insall-Salvati ratio for assessment of patellar height. Clin. Orthop 1992 ; 282 : 170-6. 18 Blackburne JS, Peel TE. A new method for measuring patellar height. J Bone Joint Surg 1977 ; 59 B : 241-2. 19 Caton J, Deschamps G. Étude critique des différentes méthodes de mesure de la hauteur de la rotule et mise au point d’une technique originale. In: Bard H, Drapé JL, Goutallier D, Larédo JD, Eds. Le genou traumatique et dégénératif, GETROA opus XXIV. Montpellier: Sauramps; 1997. p. 209-14. 20 Malghem J, Maldague B. Depth insufficiency of the proximal trochlear groove on lateral radiographs of the knee : relation to patellar dislocation. Radiology 1989 ; 170 : 507-10. 21 Braun S. Pied creux pathologique de l’adulte. Rev Prat 1997 ; 47 : 26-31. 22 Massé JP, Glimet T, Nguyen M, Lequesne M. Précision de la mesure de l’interligne dans la gonarthrose fémoro-tibiale interne (GFTI). Reproductibilité. Rev Rhum 1990 ; 57 : 680. 23 Mazières B, Levourc’h P, Boichut D. Mesure des interlignes du genou arthrosique. Reproductibilité: coefficient de variation intra et inter observateur. Rev Rhum 1990 ; 5 : 680. 24 Dougados M, Gueguen, Nguyen M, Thiesa, Listrat V, Jacob L, et al. Longitudinal radiologic evaluation of osteoarthritis of the knee. J Rheumatol 1992 ; 19 : 378-84. 25 Wiberg G. Studies on dysplastic acetabula and congenital subluxation of the hip joint, with special reference to the complication of osteoarthritis. Acta Chir Scandin 1939 Suppl 58. 26 Danielsson LG. Incidence and prognosis of coxarthrosis.. Acta Orthop Scand 1964 Suppl 66. 27 Gofton JP. Studies in osteoarthritis of the hip. Part IV. Biomechanics and clinical considerations. Can Med Assoc J 1971 ; 104 : 1007-11. 28 Goutallier D, Bernageau J. Intérêt pratique de la TA – GT. In: Bard H, Drapé JL, Goutallier D, Larédo JD, Eds. Le genou traumatique et dégénératif. GETROA, opus XXIV. Montpellier: Sauramps; 1997. p. 259-70.
Joint Bone Spine 2002 ; 69 : 282-92 © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved S1297319X0200372X/FLA
Description of the potential of an arthrometer for standard and reduced radiographs suitable to measurement of angles and segments of hip, knee, foot and joint space widths Michel Lequesne1*, Gérard Morvan2 1 Rheumatology department, hôpital Léopold Bellan, 75014 Paris, France; 2cabinet de radiologie, 5, rue Alfred Bruneau, 75116 Paris, France
(Submitted for publication May 10, 2001; accepted in revised form September 5, 2001)
Summary – Background. Primary osteoarthritis is usually selected in either epidemiological or therapeutic studies. This implies exclusions. Among cases of secondary osteoarthritis considered for either stratification or exclusion – or for prognosis and treatment in daily practice – are those due to architectural defects. Parameters of the latter should be measured to ascertain diagnosis. At present, measurements have to be performed either on digitized reduced films or standard radiographs. Objective. To finalize an instrument capable of measuring the main angles and segments characteristic of the main dysmorphisms of the hip, knee and foot on different sizes of films. Methods. An arthrometer drawn on transparent material to be placed on radiographs was designed, involving several appropriate protractors and millimetric scales; it was tested on 60 hip, 35 knee and 17 foot radiographs with various architectural defects. Angles and segments most often used according to literature were measured. Reduction rates of films were various, reflecting the range of radiograph sizes currently used in everyday practice. Results. Measurements were easily performed on radiographs from standard (100%) up to 50% of reduction rate. So the arthrometer allows the recognition, especially in moderate, not obvious forms, of the following developmental or acquired dysmorphisms: hip congenital dysplasia and subluxation, including coxa valga and neck excess of anteversion; acetabular protrusion and coxa vara; tilt deformity; knee: patellar height abnormalities, patellar maltracking, trochlear depth insufficiency; foot: pes cavus, flat foot. Angle and segment ratios do not change in reduced film, whereas segments (absolute lengths) obviously should be converted according to the reduction rate for retrieving the classical values established for decades. Conclusion. The arthrometer allows us to measure the relevant parameters of various dysmorphisms without drawing lines over the films themselves. It is suitable to reduced as well to standard radiographs. However, only the instrument and its ability to be used in various films sizes are here presented. Study of reproducibility of measurements – especially concerning the joint space width – remain to be performed. Joint Bone Spine 2002 ; 69 : 282-92. © 2002 Éditions scientifiques et médicales Elsevier SAS arthrometer / dysmorphisms / foot / hip / joint space width / knee / radiographic measurements
* Correspondence and reprints: 33 rue Guilleminot 75014 Paris, France. E-mail address: [email protected] (M. Lequesne).
Radioarthrometry: hip, knee, foot, joint space widths
Diagnosis and evaluation of the main developmental or acquired architectural defects in joints of lower limbs are of importance in daily practice as well as in scientific works. Measurements of dysmorphisms are interesting for orthopaedic surgeons, radiologists and rheumatologists. In many studies, notably epidemiologic surveys and therapeutic trials, the distinction between primary and secondary osteoarthritis (OA) of the lower limb joints are mandatory as prognosis and treatment are different. Among secondary OA, those related to a developmental dysplasia, as acetabular insufficiency or protrusion, coxa valga, patellar subluxation, patella alta, trochlear dysplasia, major varus or valgus deformity of the knee, for example, should be carefully considered, especially with regard to prognosis [1, 2] and treatment, in daily practice, in hip [2] as well in knee [3] OA. A faint developmental defect is often doubtful; it should be measured to be ascertained. In its frank forms, measurements are needed for planning surgical corrections [4-6]. In a recent editorial on hip dysplasia, about ten studies involving measurement of certain angles are cited [5]. Nowadays, we have to deal most often with digitized radiographs, reduced according to variable proportions: 75%, 66%, 50% or certain intermediate values. However, standard plain films (‘100%’) are far from having vanished and are claimed by many orthopaedic surgeons. Thus, a tool capable of measuring angles and segments and joint space widths either on standard or on reduced films appears necessary. MATERIAL AND METHOD We designed an instrument printed on a transparent material, suitable for both standard and digitized films reduced by up to 50%. A first model was made especially for all the measurements related to hip malformations, congenital or acquired [6]. Now, we are presenting an arthrometer1 appropriate for assessing main architectural radiographic abnormalities of hip, knee and foot and for appraising joint space (JS) width. It involves several designs of protractors and of millimetric rulers drawn on transparent material. This instrument is put on the X-ray in the appropriate location. Every measure is read directly on it, therefore marking lines on the radiograph is avoided. 1
The arthrometer (© M. Lequesne, 2001) is distributed by Transfert International, 15 rue du Louvre, 75001 Paris, FR. Tel (33, 1) 44 88 20 44 ; Fax (33, 1) 44 88 20 40.
283
The design was carried out by trying different drafts of the arthrometer on radiographs of standard size (100%) and of various coefficients of reduction reflecting the range of radiographic sizes currently used in daily practice down to 50%. Instrument drafts were tested on 60 hip, 35 knee and 17 foot radiographies to achieve the final design. The latter comprises a protractor with concentric circles for measuring VCE and VCA angles of the hip (figure 1, left) and a larger protractor for measuring all the other angles of hip, knee and foot (figure 2, right). Points C and C’ are punched for allowing marks with a pencil tip on the center C of the femoral head and the middle C’ of the femoral neck. Three millimetric rulers are provided at 100%, 75% and 66%, so segments lengths and JS widths may be read directly on film of analogous size; in case of 50% reduction, measurement is performed with the 100% ruler and divided by two. If the reduction rate is neither 75%, 66% nor 50%, the lengths found should be converted according to the intermediate reduction rate. Radiologists usually mention it. However, in numerous cases, mention of the reduction rate is replaced, on a side of the film, by a scale over which each division (to be measured) stand for 1 cm (i.e, if one division length is 0.66, it means that the reduction rate is 66%). Value ‘100%’ is the classical one, coming from decades of studies. Normal values (and certain borderline values) will be indicated. Let us recall that angles to be measured and length ratios do not change on digitized reduced films. Protractors of the arthrometer were designed for the right hip. For the left hip, either the instrument or the radiograph has to be flipped over RESULTS: MEASUREMENT OF THE MAIN DYSMORPHIES The hip Congenital dysplasia and subluxation of the hip These two degrees of the developmental dysmorphism are found as a single or associated cause of hip OA in 20–40% of patients [1, 2, 4, 5, 7-12]. In well-chosen cases, it is possible to correct the architectural defect by osteotomy and/or shelf operation. Dysplasia and subluxation involve three dysmorphic components: acetabular insufficiency, excess of verticality, excess of anteversion of the femoral neck. However, often only one or two of these three flaws are present. Combinations are different from one patient to another and should be evaluated systematically.
284
M. Lequesne, G. Morvan
Figure 1. The arthrometer (©M. Lequesne 2001) drawn on transparent film.
Figure 2. Sketch of an anteroposterior view radiograph of a tilting pelvis in the standing position. Right hip: acetabular insufficiency: VCE 18°, HTE 15°. Left hip: normal.
On the anteroposterior (AP) view On the coronal view taken in standing position, the lower limbs being in medial rotation 15° ± 5, the angles VCE, CC’D, HTE (figure 2 right hip) and the acetabular depth are measured. The VCE angle measures the lateral covering of the femoral head by the acetabular roof in the functional position (normal value [NV]: ≥ 25 ), V being the vertical line, parallel to the lateral edge of the film, drawn
Figure 3. The arthrometer in place for measuring VCE angle.
through C, which is the center of the femoral head. The center C correct location is determined by placing the appropriate circle or a between-circles interval on the femoral head contour. VCE value is read directly on the protractor (figure 3). Before moving the instrument on the X-ray for the next measurements, the center C
Radioarthrometry: hip, knee, foot, joint space widths
285
should be marked on the film by inserting the tip of a pencil into the punch at point C of the arthrometer. Likewise, the center C’ of the femoral neck should be marked by sliding the instrument’s millimetric segment C’ so as to place it over the narrowest portion of the femoral neck. The middle C’ of the latter is marked on the film, as it was done for C, with a pencil tip. The CC’D neck-shaft angle (NSA) measures the incline of the neck on the shaft of the femur. Possible dysmorphisms are coxa valga: CC’D ≥ 140 , and coxa vara: CC’D ≤ 120 ; intermediate values are normal or borderline (table I).CC’D is measured by placing the arthrometer vertically, with the line TD centered on the femoral shaft axis. The arthrometer is moved gradually until point C and C’(axis of the neck) previously marked on the film coincide with a line of the large protractor or falls between two lines (figure 4). Then the value of the CC’D angle is read directly. This measurement is correct only if the lower limb was positioned in about 15° internal rotation, so the neck will be approximately in the coronal plan. The HTE angle measures acetabular roof acclivity (NV: ≤ 10°). It is formed by (i) a ‘true’ horizontal line (parallel to the upper edge of the film) through point T, and (ii) the TE line, T and E being medial and lateral limits of the acetabular roof (condensed thick line) Table I. Values of the main angles and segments characteristic of the architecture of normal and dysmorphic hip, knee, foot on radiographs. Hip VCE NSA (CC’D) HTE VCA Neck anteversion Acetabul. anteversion Acetabul. depth When a c medial to ii (fig. 8) Knee Tibiofemoral angle Trochlear angle Foot Djian-Annonier angle
Normal*
Dysmorphic*
≥ 25° 120°–137° ≤ 10° ≥ 25° 7–15°
< 20° > 140° > 12° < 20° > 20°
20°–25° ≥ 9 mm F: 0–5 mm M: 0–2 mm
> 25° < 9 mm ≥ 6 mm ≥ 3 mm
0° ± 3° 125°–143°
≥ 4° ≥ 145°
125° ± 5°
≤ 115° ≥ 135°
* In this table, the gap between normal and dysmorphic values corresponds to ‘borderline’ architectures, of which the harmful role is doubtful. Abbreviations: F: females; M: males; NSA: neck-shaft angle.
Figure 4. Measurement of the neck-shaft angle CC’D: when the line TD of the arthrometer is superimposed on the diaphysis axis, the instrument should be moved vertically until one of the rays of the large protractor will be on or close to the CC’D axis of the femoral neck. In this example, point T of the arthrometer is on the top of the neck-shaft angle by chance.
respectively. It is measured by the left part of the large protractor of the arthrometer staying vertical after having placed the T correctly on the film (figure 5). Acetabular depth (NV ≥ 9 mm) is the perpendicular distance from the deepest point of the acetabulum to the EP line, where P is the ipsilateral upper corner of the symphysis pubis [4]. It is measured using the millimitergraduated segment perpendicular to TD, the line TD being placed over the dotted line EP on the film (figure 6). On the false profile view This view is a partially oblique view of the hip taken in standing position with the pelvis rotated backward 25° from profile, the lower limb close to the radiologic table staying stable with the axis of the foot remaining paral-
286
M. Lequesne, G. Morvan
Figure 5. Measurement of the HTE angle.
Figure 7. Sketch of false profile of the right hip [13], VCA: normal. VCA’ (dotted line): anterior insufficiency of coverage (VCA’ = 17°).
On the CT scan view This view is performed when an excess or a lack of anteversion of the neck is likely according to certain clinical and/or radiological features. The femoral neck anteversion angle (NV: 7–20°) and the acetabular anteversion angle (NV: 20–25°) are usually measured and noted by radiologists.
Figure 6. Acetabular depth and its measurement by segment a d. An abnormal shallow acetabulum was drawn here.
lel to the table [13]. The VCA angle (NV: ≥ 25°) measures the anterior covering of the femoral head by the acetabular roof, of which the anterior radiological edge is named A. V is the vertical line drawn through C. Measurement is made as for VCE, point C having been determined with the help of the concentric circles applied on the femoral head (figure 7).
Acetabular protrusion Found in about 5% of patients with hip OA [7, 8]. It consists of an excess of depth of the acetabulum: VCE and VCA angles are usually ≥ 45°, HTE angle between 0 and –10°, with a frequent coxa vara: CC’D ≤ 120°. However, the main criterion is the location of the acetabular line ac medial to the ilio-ischiatic ii line beyond the following distances: ac–ii (figure 8) is ≥ 6 mm in women, and ≥ 3 mm in men with protrusio [14]. This ac–ii measurement is valid exclusively on full pelvis radiograph, not on a single hip view, which is able to create by itself an excessive inner projection of ac line. Tilt deformity The sequela of a mild slipping of the femoral epiphysis during adolescence, usually painless at the time of occurrence and often resulting from athletic activity [15]. The severity of this acquired abnormality can be
Radioarthrometry: hip, knee, foot, joint space widths
Figure 8. (a) Acetabular protrusion is measured by the gap between the ilio ischiatic line ii and the acetabular line ac when the latter is medial to ii. (b) Tilt deformity: center C of the femoral head is below the neck axis C’C” (≥ 2 mm) (these two independent dysmorphisms are drawn here together in only one sketch).
evaluated as indicated in figure 8. Axis of the neck drawn from C’ and C” (middle of the neck) and extended proximally normally traverses the femoral head close to its center C. Tilt deformity, also called caput varum, is testified by a point C below the neck axis of at least 2 mm. THE KNEE Panoramic anteroposterior hip-ankle weightbearing view Angulation deformity Angulation deformity of the tibiofemoral knee joint (genu varum, genu valgum) is the most frequent dysmorphism. Normally, the axis of the tibia is the continuation of the mechanical axis of the femur which has to be drawn on the radiograph (figure 9). However, an angle of ± 3° is considered as tolerable [16]. Measurement is made on a set of three radiographs taken in the standing position: pelvis, femurs, and tibias, included the ankles. Measurement is performed with the help of the upper part of the large protractor, as shown in figure
287
Figure 9. The arthrometer in place for measuring the tibiofemoral angle (here: genu varum 7).
9, point T being placed in the middle of the tibial interspinous feature. Lateral view Patella height Patella height (figure 10) is measured according to one of the following ratios: (i) the Insall-Salvati index modified Grelsamer [17] is the ratio A-T1/A-B, where A and B are the inferior and superior limit (respectively) of the articular surface of the patella, and T1 the tibial tuberosity. NV: 1.5; (ii) the Blakburne-Peel index [18] takes as inferior basis the tangential line to tibial plateaus. The perpendicular from A to this tangent (A-P) is the numerator. The ratio A-P/A-B is normally 0.8; (iii) the Caton index A-T2/A-B, where T2 is the anterosuperior angle of the tibia, is normally 1 ± 0.2. It is ≥ 1.2 in patella alta and between 0.6 and 0.8 in patella infera. This index is valuable whatever the knee flexion between 10–80°, whereas other indices are valid only on the knee flexion 30° [19]. Depth of the trochlear groove This may be evaluated and measured on the lateral view. Normally, the line marking the median trochlear
288
M. Lequesne, G. Morvan
Figure 10. Lateral view of the knee. Parameters of the patella heights (see text). A-B: outline (length) of the articular surface of the patella. T1: tibial tuberosity; T2: anterosuperior edge of the tibia.
Figure 12. Axial view, knee flexed 30°. The arthrometer is in place for measuring the sulcus angle; its point T should be on the trochlear groove.
groove is largely and consistently behind the contour of the lateral trochlear border (figure 11). When it runs closer to this trochlear border in its upper part (dotted line), it is the sign of a depth insufficiency in the superior part of the trochlea. In most dysmorphic cases, this line crosses the anterior contour of condyles. More generally, the measurement of the trochlear depth on the lateral view was proposed as the distance between trochlear groove line G and condylotrochlear margins M, the lateral and medial ones being superimposed. Measured 1 cm below the upper border of the trochlea, the distance MG is normally of 5.94 ± 1.74 mm. It is < 5 mm in most of the pathological cases [20]. Radiographic axial view (knee flexed 30°)
Figure 11. Lateral view. The trochlear groove line visible through condyle and trochlea; continuous line: normal location; dotted line: depth insufficiency. M: margin; G: groove of trochlea.
Trochlear dysmorphism Trochlear dysmorphism with insufficiency of depth is measured by the trochlear or sulcus angle L G M, where L and M are the lateral and the medial trochlear edges respectively and G the trochlear groove. The measurement is formed as shown in figure 12: the film being rotated 90°, patella on the right, the point T of the arthrometer is placed on the trochlear groove G, its line TD staying on the trochlear wall placed inferiorly. NV: 125–143°; dysmorphism: > 143°.
Radioarthrometry: hip, knee, foot, joint space widths
289
Patella malpositions On the axial view (knee flexed 30°) may be seen patellar tilting (figure 13B), and patellar subluxation (to be measured as the gap between the two dotted lines, figure 13C). Sometimes, these two components are associated together (figure 13D) and also with patella alta. (Normal feature seen in figure 13A.) THE FOOT Weight-bearing profile
Figure 13. Axial view, knee flexed 30°. Patella malpositions. A: normal; B: lateral tilt (maltracking); C: lateral subluxation; D: tilt and subluxation associated.
This radiograph permits the longitudinal arch of the foot to be shown and possible pes cavus and flat foot to be measured [21]. The Djian-Annonier angle (DA) is so formed: its peak is the lowest point of the talonavicular interspace; its two extremities are the downward sloping point of the sesamoid under the first metatarsal head at the front and this of the calcaneus at the rear. NV: 125° ± 5°. In frank pes cavus, it is ≤ 115° and in frank flat foot it is ≥ 135°. The line TD of the arthrometer is placed along the posterior branch of the Djian-Annonier angle on the radiograph and the value of this angle is read directly on the protractor (figure 14). The Meary-Tomeno line (MT) is formed by two segments: the posterior one is the axis of the talus; the anterior one is a line drawn from the center of the first
Figure 14. Weight-bearing lateral view of the foot. The arthrometer is in place for measuring the Djian-Annonier angle DA (here: 123°).Otherwise, MT: Meary-Tomeno line; S: Schnepp angle.
290
M. Lequesne, G. Morvan
Lelievre parabole segment (HL), a smooth curve that joins the fronts of the metatarsal heads. This curve loses its smooth form when the second metatarsal is too long or the first metatarsal is too short. JOINT SPACE WIDTH, PROSTHETIC COMPONENTS AND OTHER MEASUREMENTS JS width may be measured with the help of one of the millimetric rulers chosen according to the radiograph reduction scale. Standard of 100% is obviously better. The appropriate JS subchondral bone limits, usually at the narrowest JS site, are marked with a fine pencil point before placing the ruler. In proceeding carefully under a common lens (magnification × 2 for example), it is possible to measure to within about 0.25 or 0.50 mm according to our experience, though with an error margin of reproducibility which is discussed thereafter Migration and wearing of prosthetic components may be measured according to analogous process. The Cobb angle of scoliosis and other angles and segments are also measurable with the arthrometer. Figure 15. Weight-bearing superoinferior view of the foot. HL: Hofmann-Lelievre curve.
metatarsal head parallel to the upper border of the first metatarsal; this segment is normally in line with the posterior one. Should these two segments form an angle open downward, it testifies pes cavus; an angle open upward testifies flat foot. The Schnepp’s angle (S) is the angle of incidence to ground of the first metatarsal in weight-bearing position. Its superior branch is the axis of the first metatarsus. NV: 17° to 22°. It represents another mean to assess pes cavus (> 22°) and flat foot. It is measured in using the left part of the large protractor, its 0 -T axis on the horizontal line, T point on the top of S angle. Weight-bearing supero-inferior view The same areas of the arthrometer allow us to measure: (i) the angle between the axis of the first and fifth metatarsal (NV: 32°); (ii) the varus of the first metatarsal (angle between the first and the second metatarsal axis) (NV: 10°); (iii) the angle between the first metatarsal and the first phalanx (NV: 10°); and (iv) the angle between the first and second phalanx is on average 5° (figure 15). One of the main criteria of the foot architecture is to be appraised, not exactly measured. This is the Hofman-
DISCUSSION The instrument here proposed involves means for measuring the main angles and segments specific to hip, knee and foot morphology. Certain classical tools, such as either a pocked articulated goniometer or the femorocoxometer already allowed some of these measurements. The arthrometer appears to have the following advantages: it combines in only one tool several means of measurements, appropriate protractors and millimetric rulers; it allows drawing lines on the radiographs to be avoided (except with regard to the tibio-femoral angle), allows verification if they have already been traced. Above all, this arthrometer was designed in such proportions that it suits reduced (up to 50%) as well as standard (‘100%’) radiographs. It involves not only a millimetric ruler of normal size, but also two others at 75% and 66% reduction rates, rates often used by radiologists. The use of the arthrometer comprises certain limits. On the one hand, in rare patients with small stature, the 50% reduction rate yields an image that is too narrow, beneath the protractor rays. In such cases, the answer is to draw on the film the lines (sufficiently extended) figuring the appropriate angles and to apply the arthrometer on them. On the contrary, on non-reduced films of certain persons with large stature, it may
Radioarthrometry: hip, knee, foot, joint space widths
occur – in fact rarely – that the femoral head contour passes over the larger arthrometer circle. However, in placing that circle concentrically to the ‘oversized’ femoral head, it is easy to fix and mark the femoral head center. On the other hand, the degree of correctness of measurement has not been evaluated by intra- and interobserver reproducibility. Perhaps this is not very important in matter of angles; to our knowledge, such validation was not performed in studies on either hip dysplasia or knee dysmorphisms [1-3, 9], probably because corrective surgery, difficult to adjust with an accuracy of 2–3 degrees, does not imply an extreme correctness in baseline scores which lead to the surgical decision. However, the linear measurements should be more accurate, specially those of the JS width and the possible rapid chondrolysis. The reproducibility evaluation of the millimetric rulers, especially the reduced ones, remains to be performed. Indeed, with a ruler of normal size, reproducibility test were done in the 90’ in patients with medial tibiofemoral OA. The SD of the measurement disparities in intraobserver evaluation were from 0.23 to 0.37 mm [22, 23] and the intra-class correlation coefficient from 0.92 to 0.77 [23, 24]: results of measurers were unequal. At the physician’s office, the less capable measurer has to account for a possible error margin of 2 SD, i.e., 0.37 × 2 = 0.74mm. Above this threshold, he will be allowed to register as true difference from one side to the other one in a JS width (incipiens OA) or from one time to another (example: rapid or half a rapid chondrolysis when 1 mm of JS width is lost within 3 to 6 months). Nevertheless, new reproducibility checking should be implemented to ensure the validity of the arthrometer in hip and knee OA. In the frame of possible structural-modulator drug trials, measurement must be done with very more accurate material and methods. Table I shows values of the main angles and segments. Angles – and ratios – do not change with reduction of radiographs. Pelvis inclination on the right or left in anteroposterior X-ray view taken in the standing position, usually resulting from lower limb inequality, includes an issue: the reference basis of angles to be measured may be either the ‘absolute’ horizontal/vertical axis or the horizontal/vertical axis of the pelvis itself. The original Wiberg’s CE angle took as axis 0 the line passing through C and perpendicular to the horizontal axis of the pelvis, i.e., the line joining the two femoral head centers [25]. However, we consider that measurement of the VCE angle from the vertical line and of the HTE
291
angle from the horizontal line is most appropriate for the following reasons. Let us consider the right hip in figure 2: inclination of the pelvis results in a functional uncovering of the femoral head, adding to the effect of architectural uncovering of dysplasia, if any exists. VCE angle measures these two uncovering together. The hyper-pressure in the superolateral part of the hip joint is logically the result of this global uncovering, which plays its role at the beginning of each step, before muscle constriction returns the pelvis to horizontal position. In fact, two arguments suggest that a greater than 1.5 cm leg length inequality is clinically relevant and harmful: (i) superolateral joint space narrowing due to hip OA is more marked in patients with 1.5 cm leg length inequality than in those with legs of about the same length [26], and (ii) hip OA occurs earlier or is more marked on the side of the longer leg (‘long-leg syndrome’) [27]. Conversely, on the side of the shorter leg, a functional ‘overcoverage’ of the femoral head occurs, affording some degree of protection. The exact situation of the E and/or A points can be difficult to determine. Following the anterior edge of the acetabulum through the femoral head on the AP view can help to find point E. Likewise, on the false profile view, following the anterior edge of the acetabulum from bottom to top helps in finding point A. When measuring the VCE and VCA angles of a hip OA with joint space narrowing implying secondary lateral or anterosuperior femoral head migration, the circle of the instrument corresponding to the femoral head should be moved to its original position under the acetabular roof, in order to simulate a joint space of ideal width. This allows the estimation of the VCE and VCA angles that existed before OA occurred. When measuring the NSA angle (CC’D), it is important to check whether a fixed lateral rotation deformity or an omission has prevented placement of the femoral neck within or near the coronal plane. Lateral rotation results in the greater trochanter being superimposed in part on the base of the femoral neck and the lesser trochanter being too prominent medially. This increases the apparent NSA angle on the film. With regard to knee joints, trochlear dysplasia and patella alta are not the only factors of patellar instability. Vastus medialis insufficiency acts in the same way, which results in patella tilting, that quadriceps contraction is able to reduce. More important is the too-lateral situation of anterior tibial tuberosity (AT) in relation to trochlear groove (TG). This gap AT-TG, called tibial tubercle distance, is measured by radiologists on special
292
M. Lequesne, G. Morvan
X-ray or TDM views; NV varies with the flexion angle of the knee and with authors [28]. REFERENCES 1 Cooperman DR, Wallenstein R, Stulberg SD. Acetabular dysplasia in the adult. Clin. Orthop 1983 ; 175 : 79-85. 2 Murphy SB, Ganz R, Muller ME. The prognosis of untreated dysplasia of the hip. J. Bone Joint Surg 1995 ; 77A : 985-9. 3 Cooke TD, Kelly BP, Harrison L, Mohamed G, Khan B. Radiographic grading for knee osteoarthritis. A revised scheme that relates to alignment and deformity. J Rheumatol 1999 ; 26 : 641-4. 4 Murray RO. The aetiology of primary osteoarthritis of the hip. Br J Radiol 1965 ; 38 : 810-24. 5 Lequesne M. Congenital dysplasia of the hip in adults. Is there still room for surgical treatment? Rev Rhum [Engl Ed] 1999 ; 66 : 4-8. 6 Lequesne M. Femorocoxometry: angles and segments characteristic of dysplastic and dysmorphic hip conditions in adults. Rev Rhum [Engl Ed] 1999 ; 66 : 136-42. 7 Falconnet M, Vignon G. Étude statistique, étiologique et clinique de 200 observations de coxarthrose. Lyon Med 1968 ; 17 : 1171-86. 8 Lequesne M. La coxarthrose: critères de diagnostic. Etiologie sur 200 cas. Rôle de la dysplasie congénitale. Peyron J, Ed. Epidemiology of osteoarthritis, Rueil-Malmaison, Geigy 1981 : 198-210. 9 Trousdale RT, Ekkernkamp A, Ganz R, Wallrichs SL. Periacetabular and intertrochanteric osteotomy for the treatment of osteoarthritis in dysplastic hips. J. Bone Joint Surg 1995 ; 77 A : 73-85. 10 Wilson MG, Poss R. Osteoarthritis of the hip. Moskowitz R, Howell D, Goldberg V, Mankin HJ, Eds. Osteoarthritis. Diagnosis and medical/surgical management, 2nd Ed., Philadelphia: W B Saunders 1992 : 621-49. 11 Aronson J. Osteoarthritis of the young adult hip: etiology and treatment. In Instructional course lectures. Am Acad Orthop Surgeons Vol 35. St Louis; CV Mosby 1986: 119-28. 12 Harris WH. Etiology of osteoarthritis of the hip. Clin. Orthop 1986 ; 213 : 20-33. 13 Lequesne M, Larédo JD. The faux profil (oblique view) of the hip in standing position. Contribution to the evaluation of osteoarthritis of the adult hip. Ann Rheum Dis 1998 ; 57 : 676-81.
14 Armburster TG, Guerra J, Resnick D, Goergen TG, Feingold M, Niwayama G, et al. The adult hip: an anatomic study. Part I: the body landmarks. Radiology 1978 ; 128 : 1-10. 15 Murray RO, Duncan C. Athletic activity in adolescence as an etiological factor in degenerative hip disease. J Bone Joint Surg 1971 ; 53 B : 406-19. 16 Massé JP, Glimet T, Alperovitch A, Kuntz D. Valeur de l’angle fémoro-tibial de 244 sujets exempts de gonarthrose. Rev Rhum 1985 ; 542 : 91-4. 17 Grelsamer RP, Meachow S. The modified Insall-Salvati ratio for assessment of patellar height. Clin. Orthop 1992 ; 282 : 170-6. 18 Blackburne JS, Peel TE. A new method for measuring patellar height. J Bone Joint Surg 1977 ; 59 B : 241-2. 19 Caton J, Deschamps G. Étude critique des différentes méthodes de mesure de la hauteur de la rotule et mise au point d’une technique originale. In: Bard H, Drapé JL, Goutallier D, Larédo JD, Eds. Le genou traumatique et dégénératif, GETROA opus XXIV. Montpellier: Sauramps; 1997. p. 209-14. 20 Malghem J, Maldague B. Depth insufficiency of the proximal trochlear groove on lateral radiographs of the knee : relation to patellar dislocation. Radiology 1989 ; 170 : 507-10. 21 Braun S. Pied creux pathologique de l’adulte. Rev Prat 1997 ; 47 : 26-31. 22 Massé JP, Glimet T, Nguyen M, Lequesne M. Précision de la mesure de l’interligne dans la gonarthrose fémoro-tibiale interne (GFTI). Reproductibilité. Rev Rhum 1990 ; 57 : 680. 23 Mazières B, Levourc’h P, Boichut D. Mesure des interlignes du genou arthrosique. Reproductibilité: coefficient de variation intra et inter observateur. Rev Rhum 1990 ; 5 : 680. 24 Dougados M, Gueguen, Nguyen M, Thiesa, Listrat V, Jacob L, et al. Longitudinal radiologic evaluation of osteoarthritis of the knee. J Rheumatol 1992 ; 19 : 378-84. 25 Wiberg G. Studies on dysplastic acetabula and congenital subluxation of the hip joint, with special reference to the complication of osteoarthritis. Acta Chir Scandin 1939 Suppl 58. 26 Danielsson LG. Incidence and prognosis of coxarthrosis.. Acta Orthop Scand 1964 Suppl 66. 27 Gofton JP. Studies in osteoarthritis of the hip. Part IV. Biomechanics and clinical considerations. Can Med Assoc J 1971 ; 104 : 1007-11. 28 Goutallier D, Bernageau J. Intérêt pratique de la TA – GT. In: Bard H, Drapé JL, Goutallier D, Larédo JD, Eds. Le genou traumatique et dégénératif. GETROA, opus XXIV. Montpellier: Sauramps; 1997. p. 259-70.