Four-angle radiographic assessment of cement mantle thickness in cemented total hip arthroplasty

Four-angle radiographic assessment of cement mantle thickness in cemented total hip arthroplasty

The Journal of Arthroplasty Vol. 18 No. 7 2003 Four-Angle Radiographic Assessment of Cement Mantle Thickness in Cemented Total Hip Arthroplasty Kenji...

183KB Sizes 0 Downloads 69 Views

The Journal of Arthroplasty Vol. 18 No. 7 2003

Four-Angle Radiographic Assessment of Cement Mantle Thickness in Cemented Total Hip Arthroplasty Kenji Kawate, MD, Hiroshi Yajima, MD, Yasuharu Tomita, MD, Kazuya Sugimoto, MD, Tetsuji Ohmura, MD, Nobuyuki Hiyoshi, MD, and Yoshinori Takakura, MD

Abstract: In this study, the cement mantle thickness of 57 hips undergoing primary cemented total hip arthroplasty (THA) were investigated with 4 angle radiographs. All surgeries were performed with a standardized posterolateral surgical approach. In addition to conventional anteroposterior and lateral radiographs, obturator oblique and iliac oblique radiographs were taken. In the evaluation of the cementing grade on the anteroposterior radiograph, the incidence of hips with thin cement mantles was 25%. In evaluating the cementing grade on the anteroposterior and lateral radiographs, we found an incidence of hips with thin cement mantles of 37%. From the evaluations with 4 angle radiographs, 10 hips were additionally assigned as being of a poor cementing grade (Mulroy’s grade C-2). Thin cement mantles were mainly seen on the iliac oblique radiographs. Key words: cemented total hip arthroplasty, radiographic assessment, oblique radiograph, cement mantle, cementing grade. © 2003 Elsevier Inc. All rights reserved.

Aseptic loosening of cemented femoral components in hip arthroplasty is considered a serious clinical problem. It is partly caused by malalignment of femoral components, and the thin cement mantles themselves are considered a cause of osteolysis [1–7]. Ebramzadeh and Sarmiento et al. [1] and Star and Colwell et al. [2] reported the adverse effects of a thin cement mantle on stem loosening. Huddleston [3] and Maloney and Jasty et al. [4] reported femoral lysis after cemented hip arthroplasty around the femoral components. However, cement

mantle thickness in past reports has only been assessed with conventional anteroposterior and lateral radiographs. Since 1995, in addition to the anteroposterior and lateral radiographs, obturator oblique and iliac oblique radiographs have been taken after cemented total hip arthroplasty in our institution. In the present study, the cement mantle thickness was assessed with conventional anteroposterior and lateral radiographs and with 4 angle radiographs, including obturator oblique and iliac oblique radiographs. The results of the 2 kinds of assessment were compared.

From the Department of Orthopaedic Surgery, Nara Medical University, Kashihara, Nara, Japan. Recevied May 23, 2002; accepted May 12, 2003. No benefits or funds were received in support of this study. Reprint requests: Kenji Kawate, MD, Department of Orthopaedic Surgery, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan. © 2003 Elsevier Inc. All rights reserved. 0883-5403/03/1807-0017$30.00/0 doi:10.1054/S0883-5403(03)00275-4

Patients and Methods Of the 307 primary total hip arthroplasties performed at our institution from July 1995 to May 2001, 57 hips treated with cemented total hip arthroplasty were investigated. All surgeries were performed by a single surgeon. There were 54 hips

914

Four-Angle Assessment of Cement Mantle in THA • Kawate et al.

915

mantle. Determination of femoral stem size was based on preoperative templating and intraoperative acceptance of size-specific femoral rasps in the intramedullary canal. Each femoral rasp was designed to be larger than the corresponding stem so that the femoral rasp formed a 2-mm-thick cement mantle throughout. The distal centralizer geometry was triangularly shaped, made of PMMA, and varied from 10 to 17 mm in diameter in 1 mm increments. Radiographic Evaluation

Fig. 1. Natural Hip stem. (A) Anterior view. (B) Lateral view.

in 40 women and 3 hips in 3 men. Fourteen patients had bilateral THA and 29 patients had unilateral THA. The average age at surgery was 73 years (range, 43– 85 years). The average height and weight of the patients were 1.49 m (range, 1.39 – 1.62 m) and 54 kg (range, 40 –78 kg). The etiology of the hip lesions were osteoarthrosis secondary to hip dysplasia in 33, rapidly destructive coxarthrosis in 12, primary osteoarthrosis in 7, rheumatoid arthritis in 4, and osteonecrosis of the femoral head in 1. This high rate of dysplastia is close to the national average in our country, and the gender of most of the patients reflects the high rate of hip dysplasia in this particular population. Fifty-seven femoral components made of cobaltchrome alloy (Natural Hip stem, Sulzer Medica, Winterthur, Switzerland) (Fig. 1) were implanted with a standardized posterolateral surgical approach without trochanteric osteotomy, with a modular triangular distal centralizer made of PMMA, using a third-generation cementing technique [8 –10], broaching, canal brushing, canal lavage, canal drying, vacuum mixing of Simplex-P cement (Howmedica, Rutherford, NJ), canal plugging, and cement pressurization with a cement gun with a proximal femoral seal. Eight stem sizes were available, with stem lengths varying from 105 to 170 mm. Diameter at the stem tip of the 8 stems varied from 7 to 10.5 mm. A 2.225-mm-high PMMA button was provided at the inner side of the proximal stem designed to obtain a proximal medial cement

In addition to the conventional anteroposterior and lateral radiographs, obturator oblique and iliac oblique radiographs were taken with the pelvis and leg settled on a holder at a 45° angle within 3 months after surgery. Radiographs were evaluated twice by 2 authors to assess the cementing grade. The following parameters were measured by a digital caliper (Digimatic, Mitutoyo, Kawasaki, Japan) and a protractor. The medullary axis was defined by the line of best fit passing through midpoints of the medullary canal in the anteroposterior view, measured at the level of the isthmus and at points 20 mm proximal and 20 mm distal to the isthmus. The prosthesis axis was defined by the axis of symmetry of the distal half of the femoral stem on the anteroposterior radiograph. The orientation of the prosthesis within the femur was calculated as the angle between the prosthesis and medullary axes. Cement mantle thickness and canal width at the stem tip was evaluated in 4 angles on the radiographs, including those taken on the inclined holder. The thickness of the cement mantle was assessed for each anteroposterior and lateral zone, as described by Gruen, McNeice, and Amstutz [11] and Johnston and Fitzgerald et al. [12] (Fig. 2). For the 2 radiographs taken on the inclined holder, the figure was divided into 7 zones and cement mantle thickness was assessed for each zone, except zone 4, as shown in Fig. 2. The thickness of the cement mantle was determined for each zone by measuring the minimum thickness with a digital caliper. The diameter of the acetabular shell was measured and compared with its actual size to correct radiographic magnification. This is because the diameter of the femoral ball was not measurable because of metalon-metal articulation. Cement mantles were ranked as grades A, B, C-1, C-2, and D according to the cementing grades described by Mulroy, Estok, and Harris [13] with the following criteria: grade A if the cementing technique resulted in a complete filling of the medullary canal around the stem, so that a distinction be-

916 The Journal of Arthroplasty Vol. 18 No. 7 October 2003

Fig. 2. (A) Zones of the proximal femur on anteroposterior radiographs. (B) Zones of the proximal femur on lateral radiographs. (C) Zones of the proximal femur on obturator oblique radiographs. (D) Zones of the proximal femur on iliac oblique radiographs.

tween cortical bone and cement was not evident; grade B if the distribution of cement was nearly complete, but it was possible to distinguish cortical bone from cement in some areas; grade C-1 if an extensive radiolucent line (along more than 50 % of the cement-bone interface) or voids in the cement were present; grade C-2 if either a thin (⬍1 mm) mantle of cement at any site or a defect in the mantle of cement, with the metal in direct contact with cortical bone, were present; and grade D gross deficiencies in the mantle of cement, such as no cement distal to the tip of the stem, major defects in the mantle of cement, or multiple large voids were seen. Although the radiographs taken at our institution were usually enlarged about 1.1-fold, the diameter of the acetabular shell was measured and compared with its actual size to correct radiographic magnification. Because thin cement mantle was defined as less than 1 mm in size by Mulroy, Estok, and Harris [13], the corrected values below 0.91 mm obtained from the radiographs were defined as thin cement mantle with respect to actual measurements. Statistical Analysis Statistical analysis was performed using Spearman’s correlation coefficient by rank test to evaluate the cementing grade. The Mann-Whitney’s U test was used to evaluate the width of the intramedullary canal. Significance was determined at the conventional P⬍.05 level.

Results Alignment of the Femoral Stem The femoral stems averaged 0.7° of valgus, with a standard deviation (SD) of 0.7° on the anteroposterior radiographs. Angles ranged from 2.5° of valgus to 2.25° of varus. Overall, 81% of femoral stems were implanted within 1°of neutral, whereas 95% of stems were implanted within 2° of neutral. Cement Mantle Thickness and Cementing Grade The average cement mantle thicknesses of hips evaluated on anteroposterior, lateral, obturator, and iliac oblique radiographs are shown in Table 1. The grades assigned to the cementing technique, when evaluated on just the anteroposterior radiograph described by Mulroy, Estok, and Harris [13], were grade A for 6 cases (11%), B for 7 cases (12%), C-1 for 30 cases (53%), C-2 for 13 cases (23%), and D for 1 case (2%). Thin cement mantle was mainly seen at zone 5 (9 hips) on anteroposterior radiographs. A defect was also seen at zone 2 (1 hip) on anteroposterior radiographs. The incidence of cases assigned as grades C-2 and D, which have often been reported to be associated with loosening and osteolysis [4,13], was 25%. The grades assigned to the cementing technique, when evaluated on the anteroposterior and lateral radiographs, were grade A for 6 cases (11%), B for 7

Four-Angle Assessment of Cement Mantle in THA • Kawate et al.

917

Table 1. Cement Mantle Thickness Measured With 4 Angle Radiographs Zones*

Anteroposterior (mm) Lateral (mm) Obturator (mm) Iliac (mm)

1

2

3

5

6

7

3.07 ⫾ 1.37 1.17 ⫾ 0.57 4.23 ⫾ 2.02 2.14 ⫾ 0.89

1.97 ⫾ 1.14 1.28 ⫾ 0.53 3.29 ⫾ 1.30 1.48 ⫾ 0.73

1.74 ⫾ 0.76 2.43 ⫾ 1.06 2.51 ⫾ 1.05 1.77 ⫾ 0.94

1.45 ⫾ 0.50 2.18 ⫾ 0.88 2.25 ⫾ 0.97 1.24 ⫾ 0.66

1.70 ⫾ 0.60 2.78 ⫾ 1.18 2.35 ⫾ 0.93 1.40 ⫾ 0.64

2.67 ⫾ 1.05 3.56 ⫾ 1.65 2.60 ⫾ 1.13 2.67 ⫾ 1.23

*Values are average plus standard deviation.

cases (12%), C-1 for 23 cases (40%), C-2 for 20 cases (35%), and D for 1 case (2%). Thin cement mantle was mainly seen at zones 8 (16 hips) and 9 (14 hips) on lateral radiographs. In this evaluation, 7 hips were additionally assigned as grades C-2, downgraded from C-1. When the cementing grade was evaluated in 4 angles using the 2 oblique radiographs combined with anteroposterior and lateral radiographs, the grades assigned were grade A for 4 cases (7%), B for 6 cases (11%), C-1 for 16 cases (28%), C-2 for 29 cases (52%), and D for 2 cases (4%), indicating an incidence of cases assigned as grades C-2 and D as high as 56%. Thin cement mantles were mainly seen at zones 2 (10 hips), 3 (11 hips), 5 (13 hips), and 6 (12 hips) on the iliac oblique radiographs. Defects were seen at zones 5 (1 hip) and 7 (1 hip) on the iliac oblique radiographs. In this evaluation, 10 hips were additionally assigned as grade C-2,

downgraded from A (2 hips), B (1 hip), and C-1 (7 hips) because of thin cement mantle less than 0.91 mm on iliac oblique radiographs, indicating an inconsistency between the 2 angle and 4 angle evaluations (Fig. 3). The etiologies of the 10 hips were osteoarthrosis secondary to hip dysplasia in 8 and rheumatoid arthritis in 2. One hip was downgraded from C-2 to D because of a defect of the cement mantle on iliac oblique radiograph. A significant difference was seen in cases assigned as grades C-1 or C-2 among the 3 evaluations (P⫽.000003) (Table 2). Shape of the Intramedullary Canal The average widths of the intramedullary canal at the stem tip level and at the distal one third level of the stem are shown in Table 3. Canal shapes were seen to be a little oval. When the 10 hips (group B)

Fig. 3. Postoperative radiographs of a 74-year-old woman whose cementing grade was changed after evaluation of the iliac radiograph. Thin cement mantle was detected at zones 3, 5, and 6 on iliac oblique radiograph. (A) Anteroposterior radiograph. (B) Lateral radiograph. (C) Obturator oblique radiograph. (D) Iliac oblique radiograph.

918 The Journal of Arthroplasty Vol. 18 No. 7 October 2003 Table 2. Number of Grade C-1 or C-2 Diagnosed With 3 Evaluations Cementing Grade

Anteroposterior radiograph 2 angle evaluations 4 angle evaluations

C-1*

C-2*

30 23 16

13 20 29

*Values are the number of hips.

additionally assigned as grade C-2 in 4 angle evaluations were compared with the other 47 hips (group A), the widths of the intramedullary canal on iliac oblique radiographs of the 10 hips were significantly narrower than those of the others at the distal one third of the stem (P⫽.049).

Discussion Regarding the relationship between thin cement mantle and loosening, Ebramzadeh and Sarmiento et al. [1] assessed 836 cemented femoral components, using survival analysis, during a 21-year follow-up period. They reported the adverse effects of thick or thin cement mantles on stem loosening. Star and Colwell et al. [2] evaluated 100 cemented THAs and reported the adverse effects of suboptimal (thin) cement mantles at the medial diaphysis on stem loosening. Regarding the relationship between thin cement mantle and osteolysis, Huddleston [3] reported femoral osteolysis after cemented hip arthroplasty around femoral components in 51 hips. Maloney and Jasty et al. [4] investigated focal femoral osteolysis in well-fixed cemented femoral components and indicated the adverse effects of thin cement mantles on osteolysis. In 15 (60%) of 25 hips that they reviewed, the area of osteolysis cor-

responded with either a defect in the cement mantle or an area of very thin cement that was less than 1 mm thick. Similarly, osteolysis developed in 7 (26%) of 27 hips graded C-2 or D in cases of Harris Precoat stems [6]. The osteolysis also developed in cases graded A or B. However, cement mantle thickness in past reports has only been assessed with conventional anteroposterior and lateral radiographs. There is the possibility that some cases graded A or B with 2-angle evaluations in the past reports may have been downgraded to grades C-2 or D with 4-angle evaluations. In the present study, obturator oblique and iliac oblique radiographs were assessed in addition to anteroposterior and lateral radiographs for the evaluation of the cement mantle. In the evaluation, 10 hips were additionally assigned as grade C-2, indicating that evaluations on iliac oblique radiographs could cause gradings inconsistent with the grades of the usual 2-angle evaluations. When the 10 hips were compared with the other 47 hips, the average width of the intramedullary canal at the distal third of the stem on iliac oblique radiographs of the 10 hips was significantly narrower than that of the other 47 hips. This probably resulted from the high rate of dysplastic cases in the patients studied here. The relatively thin cement mantle seen on the iliac oblique radiographs appeared to be caused by the stem being rotated toward the orientation of rasping on insertion. This is because of the small amount of space in the iliac oblique plane in the oval intramedullary canal. Although the cement mantle thickness depends on stem design, size, and preparation of the femoral intramedullary canal, the restricted space in the iliac oblique plane is disadvantageous compared with the other planes. The relatively high rate of grade C-2 with evaluation on just the anteroposterior and lateral radiographs in the present study was because of zones 8

Table 3. Width of the Intramedullary Canal at the Stem Tip and Distal Third Level of the Stem on 4 Angle Radiographs

At distal 1/3 level

At stem tip level

Anteroposterior Lateral Obturator Iliac Anteroposterior Lateral Obturator Iliac

*Values are average plus standard deviation.

Group A*

Group B*

P value

15.0 ⫾ 1.5 16.3 ⫾ 1.7 17.0 ⫾ 2.0 14.7 ⫾ 1.6 12.5 ⫾ 1.8 15.0 ⫾ 1.5 14.7 ⫾ 2.3 13.2 ⫾ 2.2

14.4 ⫾ 1.1 15.4 ⫾ 1.4 17.0 ⫾ 1.7 13.7 ⫾ 1.1 12.1 ⫾ 0.7 14.4 ⫾ 1.1 14.6 ⫾ 1.7 12.5 ⫾ 1.0

.16 .193 .975 .049 .801 .571 .930 .549

Four-Angle Assessment of Cement Mantle in THA • Kawate et al.

and 9 on lateral radiographs. The stems tended to incline toward the anterior (lateral zone 8) at the proximal part of the femur without a proximal centralizer in a standardized posterolateral surgical approach. A proximal centralizer might be useful for correct insertion of the stem, as Goldberg and Al-Habbal et al. [14] recommended. Several researchers have studied reproducibility and variability of the femoral cement mantle grading system [15,16]. Harvey, Tanzer, and Bobyn [15] reported agreement among 3 observers on the grading in only 73% of radiographs (anteroposterior view, 69%; lateral view, 77%). They also noted the difficulty of interpreting small voids. In the present study, the thickness of the cement mantle was accurately measured because the thickness was determined for each zone by measuring the minimum thickness with a digital caliper. Two observers agreed on the grading in 98% with the anteroposterior view; in 96% with 2 angle evaluations; and in 95% with 4 angle evaluations. The obvious outcome from adopting these oblique radiographs, especially the iliac oblique radiograph, was that thin cement mantles became more noticeable, allowing for more relevant evaluation. However, as a result of the hips being dysplastic, the femora have an abnormal anatomy. The present study has shown that the proximal femur of a dysplastic hip is rotated. This might account for the high number of incomplete cement mantles seen on oblique views. However, this explanation may not hold for nondysplastic hips. Considering the costs involved and the radiographic doses, evaluations using 4 angle radiographs are recommended as best suited for research study.

References 1. Ebramzadeh E, Sarmiento A, McKellop HA, et al: The cement mantle in total hip arthroplasty: analysis of long-term radiographic results. J Bone Joint Surg Am 76:77, 1994 2. Star MJ, Colwell CW, Kelman GJ, et al: Suboptimal (thin) distal cement mantle thickness as a contributory factor in total hip arthroplasty femoral component failure: a retrospective radiographic analysis fa-

3. 4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

919

voring distal stem centralization. J Arthroplasty 9:143, 1994 Huddleston HD: Femoral lysis after cemented hip arthroplasty. J Arthroplasty 3:285, 1988 Maloney WJ, Jasty M, Rosenberg A, Harris WH: Bone lysis in well-fixed cemented femoral components. J Bone Joint Surg Br 72:966, 1990 Kawate K, Maloney WJ, Bragdon CA, et al: Importance of a thin cemnt mantle: autopsy studies of eight hips. Clin Orthop 355:70, 1998 Kawate K, Ohmura T, Hiyoshi N, et al: Thin cement mantle and osteolysis with a precoated stem. Clin Orthop 365:124, 1999 Massoud SN, Hunter JB, Holdsworth BJ, et al: Early femoral loosening in one design of cemented hip replacement. J Bone Joint Surg Br 79:603, 1997 Harris WH: Is it advantageous to strengthen the cement-metal interface and use a collar for cemented femoral components of total hip replacements? Clin Orthop 285:67, 1992 Oishi CS, Walker RH, Colwell CW: The femoral component in total hip arthroplasty: six to eight-year follow-up of one hundred consecutive patients after use of a third-generation cementing technique. J Bone and Joint Surg Am 76:1130, 1994 Schmalzried TP, Harris WH: Hybrid total hip replacement: a 6.5-year follow-up study. J Bone and Joint Surg Br 75:608, 1993 Gruen TA, McNeice GM, Amstutz HC: Modes of failure of cemented stem-type femoral components: a radiographic analysis of loosening. Clin Orthop 141:17, 1979 Johnston RC, Fitzgerald RH Jr, Harris WH, Poss R: Clinical and radiographic evaluation of total hip replacement: a standard system of terminology for reporting results. J Bone Joint Surg Am 72:161, 1990 Mulroy WF, Estok DM, Harris WH: Total hip arthroplasty with use of so-called second-generation cementing techniques: a fifteen-year-average follow-up study. J Bone Joint Surg Am 77:1845, 1995 Goldberg BA, Al-Habbal G, Noble PC, et al: Proximal and distal femoral centralizers in modern cemented hip arthroplasty. Clin Orthop 349:163, 1998 Harvey EJ, Tanzer M, Bobyn JD: Femoral cement grading in total hip arthroplasty. J Arthroplasty 13: 396, 1998 Kelly AJ, Lee MB, Wong NS, et al: Poor reproducibility in radiographic grading of femoral cementing technique in total hip arthroplasty. J Arthroplasty 11:525, 1996