One of the ironies in modern technology for arthroplasty is the stress shielding in cementless stems. The aim of the development of cementless stems had been reduction of stress shielding which cement stems are not free from. In healthy femur, trabecula start form the femoral head and reach at both medial and lateral cortex in rather narrow area around lesser trochanter. So the load from the femoral head is transferred at the level on both medial and lateral side. Cement stems should have binding to the cortical bone from collar to the tip of the stem where the cement interlays, and then the load is transferred gradually from the tip to the collar, which means mild stress shielding. When distal bonding is removed, the load could be transferred as normal femur. This should have been one of the biggest requests for cementless stem. But in realty many cementless stems have difficulty to obtain a load transfer at the level like normal femur.

Since 1990, we have been mainly using lateral flare stems to obtain contact on both medial and lateral side at proximal level. In the present study, different types and length of the designs were compared by 3-Dimensional fill, 3-Dimensional fit and Finite Element Analysis.

Dislocation following total hip arthroplasty is one of the most common complications, occurring in 1% to 5% of all cases. Several causes for dislocation have been suggested that

Mismatching of cup positioning and stem anteversion

Most often positioning of the stem is anatomically predetermined, while the orientation of the cup is much more flexible. Since July 2005, stem first method has been applied for all cases. During this method, canal preparation and stem trial was done first, and then cup orientation was determined according to the stem direction and impingement. For the bigger cups 34mm or 38mm heads were applied in this series. In the present study dislocation ratio was compared to cup first method.

In the stem fist group (SF), the following procedures were done consequently.

Canal was prepared for the stem. Revelation lateral flare high proximal load transfer stem (DJO) was mainly selected. But for the case with high anteversion over 50 degrees, Modulas; conical distal load transfer stem with modular neck (Lima) was selected.

From October 2002 to July 2008, there were 191 THA cases. There were 81 hips in Standard group and 109 hips in SF group. There were 63 females and 18 males in Standard group and 90 females and 19 males in SF group (p=0.41). Average age was 61.0(22–81) in Standard group and 60.2(29–89) in SF group (p=0.53). In Standard group, 64 were replaced for osteoarthritis, 15 for rheumatoid arthritis and two for avascular necrosis. In SF group, 86 were replaced for osteoarthritis, 17 for rheumatoid arthritis and six for avascular necrosis (p=0.53). As for Crowe’s classification, 61 type I, 18 type II and 2 type III were included in Standard group. And 88 type I, 15type II, 4 type III and 2 type IV were included in SF group (p=0.29). Average anteversion of femoral neck were 23.1(−2 to 70) degree in Standard group and 26.2(−4 to 65) degree in SF group measured with CAT scan (p=0.274). MoM bearing surfaces were used with 71 hips (87.7%) in Standard group and 100 hips (91.7%) in SF group (p=0.35). Only in SF group, big metal head were used in 24hips(22%) with 34mm and in 12hips(11%) as 38mm diameter. Average leg length difference between pre and post operation was 11.5mm(0 to 36) in Standard group and 8.0mm(−18 to 30) in SF group (p< 0.05). Average cup inclination was 43.2(25 to 84) degree in Standard group and 40.9 (22 to 66) degree in SF group (p< 0.05). Average cup anteversion was 8.2 degree (0 to 22.8) in Standard group and 7.1 degree (−12 to 30.5) in SF group (p< 0.05). Average operating time was 111.9min (67–150) in Standard group and 97.5min(60–162) in SF group (p< 0.05). Average intra operative hemorrhage was 744ml(10–2757) in Standard group and 487ml(10–1374) in SF group (p< 0.05). The dislocation rate was decreased from 3.7% (3/81 cases) in Standard group to 0.0% (0/109) in SF group.

In conclusion our study suggested that Stem first method and utilization of big metal head would decrease the dislocation rate in primary cases. More bleeding from canal during accetabular reaming was expected. However less bleeding was observed in SF group.

Canal Flare Index, defined as the ratio of the intracortical width of the femur at a point 20mm proximal to the lesser trochanter and at the canal isthmus by Noble et al,; is considered to express the proximal femoral geometory, but it is usually measured by a plain A-P X-ray. Then it is thought the index is influenced by rotational position of the femur, so we made 3-D femoral model based on CAT scans and measured the canal flare index three dimensionally. Then the effect of observation from rotated direction was evaluated.

CAT scans of 49 femurs (18 male, 31 female) were obtained from the pelvis to the feet. The average age was 60.4 years old ranging from 25 to 82. Forty nine femurs contained 22 osteoarthritis of hip joint, 12 trauma, 9 knee arthritis, 3 avascular necrosis of femoral head, 3 normal candetes. From those data, 3-D models of normal side were individually made for measuring the parameters. 3-D models were made using CAD software. We measured the canal flare index at which the femur posterior condyles were parallel to the plane, reproducing the situation to take A-P X-ray. After that, those 3-D models were rotated and investigated the difference of the value to study the effect of femur position.

The canal flare index was between 2.8 and 6.6 with the average value at 4.65. The stovepipe (canal flare index< 3), the normal range (3~canal flare index< 4.7), the champagne flute (4.7~canal flare index), included 2%(1 femur), 61.2%(30 femurs), 36.7%(18 femurs), respectively. About the effect of rotation, we found the value of canal flare index was more sensitive to proximal femur rotation than the canal isthmus. The results of the canal flare index at the plane parallel to the posterior condyle line varied widely compared with the results at the position considering the anteversion. So it was suggested that the canal flare index at the patella front position does not represent the canal characteristics. It should be argued in 3-D space.

Introduction: For longer lasting and bone conserving cementless stem fixation, stable and physiological proximal load transfer from the stem to the canal should be one of the most essential factors. According to this understanding, we have been developing a custom stem system with lateral flare and an off-the-shelf (OTS) lateral flare stem system was added to the series. On the other hand, dysplastic hips are often understood that they have larger neck shaft angle as well as larger anteversion. In other words they are in the status called “coxa valga.” From this point of view we had been mainly using custom stems for the dysplastic cases before. After off-the-shelf lateral flare stem system; which is designed to have very high proximal fit and fill to normal femora; was added, we have been using 3D preoperative planning system to determine custom or OTS. Then in most of the cases, OTS stem were suitably selected. Our pilot study of virtual insertion of OTS lateral flare stem into 38 dysplastic femora has shown very tight fit in all 38 cases. The reason was analyzed that the excessive anteversion is twist of proximal part over the distal part and the proximal part has almost normal geometry. In the present study, 59 femora were examined by the 3D preoperative planning system how the excessive anteversion effect to the coxa valga status.

Materials and Methods: Fifty-nine femoral geometry data were examined by the 3D preoperative planning system. Thirty-three hip arithritis, 3 RA, 2 metastatic bone tumours, 5 AVN, 1 knee arthritis, 12 injuries, and 3 normal candidates were included. Among them one arthritic Caucasian and one AVN South American were included. The direction of the femoral landmarks; centre of femoral head (CFH), lesser trochanter (LTR), and asperas in 3 levels (just below LTR, upper 1/3, mid femur; A1-3); were assessed as the angle from knee posterior condylar (PC) line. Neck shaft angle of each case was assessed from the view perpendicular to PC line and neck shaft angle form the view perpendicular to CFH and femoral shaft (i.e. actual neck shaft angle).

Results: Average anteversion was 34.4 +/−9.9 degree. CFH and LTR correlated well (i.e. they rotate together). A1, A2, A3 correlated well (i.e. they rotate together). LTR and A1 correlate just a little, LTR and A2 were independent each other. So the twist existed around A1. Neck shaft angle was 138.7+/−6.6 in PC line view and in actual view 130.3+/−4.4. No excessive neck shaft angle was observed in actual view. Even the case that has the largest actual neck shaft angle (140.4), the virtual insertion showed good fit and fill with the lateral flare stem.

Conclusion: In many high anteversion cases, coxa valga is a product of the observation from non perpendicular direction to CFH-shaft plane. Selection or designation of the stem for high anteversion cases should be carefully determined by 3D observation.

One of the most important characteristic of the developmental dysplastic hip (DDH) is high anteversion in femoral neck. Neck-shaft angle is also understood to be higher (i.e. coxa-valga) in DDH femora. From this understanding many DDH intended stems were designed having larger neck shaft angle.

According to the result of our prior study; reported in ISTA 2005 etc.; using computer 3-D virtual surgery of high fit-and-fill lateral flare stem into high anteversion patients, it was revealed that the geometry of proximal femur itself does not have big difference from normal femora but they are only rotated blow lessertrochanter.

It is very important to know what anteversion is, and where anteversion is located, to design a better stem and to decide more proper surgical procedures for DDH cases with high anteversion.

In the present study, the geometry of 57 femora was assessed in detail to reveal the geometry of anteversion and its location in the DDH femora.

Fifty seven CAT scan data with many causes were analyzed. Thirty-two DDH, 3 Rheumatic Arthritis (RA), 2 metastatic bone tumors, 4 avascular necrosis (AVN), 1 knee arthritis, 12 injuries, and 3 normal candidates were included. Whole femoral geometries were obtained from CAT scan DICOM data and transferred to CAD geometry data format. All the following landmarks were measured its direction by the angle from posterior condylar line. The assessed landmarks were

anteversion,

All the directions were measured by the angle from the medial of the femur.

The direction of anteversion and lesser trochanter were well correlated, (R=0.55, Y=0.56X−35) i.e. femoral head and lesser trochanter were rotated together.

The direction of lesser trochanter and aspera in upper 1/6 section had no relation even they are located very close with only several cm distance, (R=−0.03, Y=−0.02X−88) i.e. however the lesser trochanter was rotated, the upper most aspera was located almost at the same direction (−87.5+/−7.58 degree).

The direction of aspera at upper 1/6 and middle femur were strongly correlated. (R=0.63, Y=0.81X-22) i.e. they stay at the same direction.

The results mean that the anteversion is a twist between normal proximal femur (from femoral head and lesser trochanter) and normal distal femur. The twist was located just blow lesser trochanter within several centimeter.

The anteversion has been understood as the abnormal mutual position between femoral neck and femoral shaft. In high anteversion hips the neck shaft angle was also believed to be higher, so several DDH oriented stems have higher neck shaft angle i.e. coxa-valga geometries. It has been believed that the location of the anteversion was around neck part. This study revealed that the deformity was located in the very narrow part just below lesser trochanter. It has been discussed that DDH oriented stems should have fit to different canal geometries, but understanding the biomechanics of abnormal anteversion and its treatment should be more important.

Since 1993, we have been developing preoperative planning system based on CAT scan data. In early period it was used to decide cup diameter and orientation for Total Hip Arthroplasty (THA). It was done using hemisphere object locating proper position and orientation. According to our progress, we have started using it for custom stem designing, stem selection and stem size planning too since 1995. Since 2001, we have been using it for almost all THA cases. We also have started use it for any case we have question about 3D geometries. Since 2005 we started computer planed 2 staged THA after leg elongation for high riding hips and reported at ISTA 2007 too. Now our policy became that every tiny question we have, we shall analyze and plan preoperatively.

In our population, the incidence of the developmental dysplastic hips is higher. The necks often have bigger anteversion, and less acetabular coverage. So we often use screws for cup fixation. The screw direction allowed in thin shell thickness is limited and less bone coverage makes good cup fixation difficult. With highly defected cases and with revision cases the situation is more difficult.

In the present study, we have developed acetabular 3D preoperative planning method with screw direction, length, and for the cases with defect, cup supporter pre-shaping with models and prediction of the allograft volume.

For the less defect cases, geometries of cup with screw holes were requested to the maker and were provided for us. Screws were attached perpendicular to each screw hole. Screw geometries have marks at every 5mm to plan proper length. The cup was located as much as closer to the original acetabular edge, keeping in the limit to avoid dislocation. Small space above the cup was accepted if anterior and posterior cup edge could be supported by original bone. Then the cup was rotated until we can obtain proper screw fixation.

For the cases with severe defects, we use cup supporters and allografts. Cup supporters are designed to be bent and fit to the pelvis during the surgery. But to shape it a properly; for good coverage and strong support; is very difficult and takes long through the limited window with fatty gloves. And mean while we get more bleeding. The geometries were obtained by CAT scan of the devices. Then proper size was determined as cup size. Chemiwood model was made and proper size supporter was opened and bent preoperatively using the model. It was scanned again and compared to the pelvic geometry again.

Using cluster cups, no dangerous screw was found as long as normal cup orientation was decided and screws were less than 30mm. Posterior screws were often too short then rotated anterior and found to have good fixation. Pre-bending could reduce surgical time remarkably.

As long as we could know, no navigation system can control the cup rotation. But acetabular preoperative planning was very useful and could reduce operative invasion. It could be done easily without using navigation system.

Objective: Establishment of the new method to evaluate fill of the hip stem.

Background: The fill of the hip stem is one of the important parameters to estimate the quality of planning or positioning of the cementless stem. It has been defined as a stem-canal width ratio on the A-P plain of X-ray images so far. However, it is quite a problem to get the correct AP images on basis so that positional difference may affect the measurement. According to our data, the fill was measured significantly different in 15, 30, 45, 60 degrees erroneous direction. First, we tried to figure out the fill of the hip stem three-dimensionally rather 2-dimensionally. Next, our new method was compared to conventional method.

Material and Methods: Leg CAT scans were performed on 32 hips of 20 patients (2 male, 18 female). Images of the canal of femora were reconstructed using CAD software. We made 2-types of canal model with or without lesser trochanter. The geometries of our lateral flare stems with different sizes were compared to each canal geometry in the CAD software and proper size was decided. Then images were observed from an accurate vertical direction of the coronal plain of the stem. We measured the 2-D fill on this plane and the 3-D fill of every 5 mm slice from the 5mm above to the 100mm below the head of lesser trochanter line (reference line). We also examined the stems 1-size smaller or larger than the appropriate ones.

Results: The mean age was 61.114 (range 24–82). The average of “3-D fill of Lateral flare stem was 51%/59% with/without lesser trochanter, and 2-D one was 74%/77%. The numerical and distributional results by these two methods to measure fill were alike but different. For example, in case without lesser trochanter, the 3-D fill showed the maximum value in the area just below the reference line. The maximum 2-D fill was recorded in 10mm caudal from the reference line. In general, this stem occupied much space in the distal area and around the lesser trochanter.

Future Plan: Extension of this evaluation method into various kinds of stems.

A dural tear is a common but troublesome complication of endoscopic spinal surgery. The limitations of space make repair difficult, and it is often necessary to proceed to an open operation to suture the dura in order to prevent leakage of cerebrospinal fluid. We describe a new patch technique in which a small piece of polyglactin 910 is fixed to the injured dura with fibrin glue. Three pieces are generally required to obtain a watertight closure after lavage with saline. We have applied this technique in seven cases. All recovered well with no adverse effects. MRI showed no sign of leakage of cerebrospinal fluid.

We studied the most superficial layer of macroscopically normal articular cartilage obtained from human femoral heads, using polarising microscopy and SEM. The most superficial layer, 4 to 8 microns thick, was acellular consisting of collagen fibrils. This layer could be peeled away as a thin film, with no broken collagen fibrils on its inferior surface or on the surface of subjacent cartilage layers. The orientation and diameter of collagen fibrils were different on these two surfaces. Our findings suggest that the most superficial layer is an independent one which is only loosely connected to the fibrous structure in the layer deep to it.