A direct referencing method of the tibial plateau for the posterior tibial slope in medial unicompartmental knee arthroplasty

The knee MRI in Studies I and II was performed using a 1.5-Tesla imager (Signa HDxT, General Electric Medical Systems, Milwaukee, WI) with an eight-channel knee coil (HD TR knee phase array). The patients were positioned supine on the MRI scanning table with the knee extended and the patella facing upward. Coronal, sagittal and axial images were obtained in a standardized manner. Images were acquired according to a common standard knee MRI protocol of the Department of Radiology, which included coronal and sagittal multiple echo recalled gradient echo images for the present study (TR: Cor/Sag 800/600, each with 4 echoes). A 3.5-mm slice thickness with a 0.5-mm slice space was used.

The intraoperative lateral knee radiograph was taken using a portable X-ray machine (MobileDaRt Evolution MX8 Version, SHIMADZU, Kyoto, Japan), which has wireless flat panel detectors (FPDs, DR-ID800, FUJIFILM, Tokyo, Japan) and can show reference X-ray images on an installed monitor immediately after radiography. After arthrotomy, stripping off of the medial capsular attachment from the proximal tibia, and resection of the anterior half of the medial meniscus, an arthroscopic hook probe was placed on the medial second quarter of the MTP (between the medial 1/4 and 1/2 MTP), parallel to the AP tibial line [27, 28] with the knee flexed at approximately 90°. Radiography was repeated until the correct laterality was confirmed on the monitor of the portable machine.

Using sagittal MRI, we evaluated the height of the anterior/posterior osteophyte and anterior/posterior expansion of subchondral bone depression (attrition) [29] on the medial second quarter of the MTP to determine whether they can affect the MTP indicated by the arthroscopic probe placed on it. Relative to a line tangent to the most prominent aspects of the anterior and posterior cortices of the MTP including cartilage remnants (line t in Fig. 1A), osteophyte height was classified into three categories as follows: none (N), there is no osteophyte; low (L), same as or lower than the line t; and high (H), higher than line t. Based on location, expansion of the subchondral depression was classified into three categories as follows: none (N), there is no depression; moderate (M), the depression expands to the anterior or the posterior 1/3 of the MTP but does not reach the most prominent aspect of the anterior and posterior cortices; and severe (S), the depression expands to or over the most prominent aspects of the anterior or posterior cortices (Fig. 1B).

The preoperative bony MTP in sagittal MRI was defined as a line tangent to the most prominent aspects of the anterior and posterior cortices (subchondral bone plate) according to methods previously described [24, 30] (line m in Fig. 2A and C). The cartilage slope of the MTP was defined as a line tangent to the most prominent aspects of the anterior and posterior cartilage remnants of the MTP (line l in Fig. 2A and C). Frontal and sagittal views of the knee MRI were simultaneously displayed on an image analysis monitor. While the frontal view showed a middle slice of the knee (Fig. 2B), a slice of the medial second quarter of the MTP (Fig. 2B) was displayed on the sagittal view. The angle formed by the two tangential lines (lines m and l) was measured on both slices using image analysis software (PACS system FABRICA Ver. 1.0.0.23, Cure Hope Corp., Osaka, Japan). A positive value was given to the angle measurement when the cartilage PTS (line l) was larger than the preoperative bony PTS (line m).

All surgical procedures in Study II were performed by one senior surgeon (M.A.), and a fixed-bearing UKA implant (Tribrid UKA system, Kyocera Corp., Kyoto, Japan) was implanted through the medial mini-parapatellar skin incision. The operation was performed using the so-called tibia-cut first and spacer block technique. First, the substitute tibial AP line [28] was drawn to pass through the medial tibial eminence and the medial edge of the patellar tendon at the joint level. Then, an arthroscopic hook probe was placed in the AP direction on the medial second quarter of the MTP. An individual implant PTS was determined by setting a gauge inserted into the cut slot of the extramedullary guide parallel to the probe (see Fig. 7E).

Preoperative bony PTS (α) and postoperative implant PTS (β) on a lateral radiograph were measured according to a method previously reported [20]. A line passing through the center of 2 circles located over the AP width of the tibia was drawn (line t in Fig. 3A and B). Then, a line perpendicular to line t was drawn (line l₁). The angle between line l₁ and a line tangential to the subchondral plate of the MTP (line m₁) was measured as the preoperative bony PTS (angle α in Fig. 3A). Similarly, implant PTS was defined as an angle between a line l₂ and a line m₂ tangential to the upper surface of the tibial tray (angle β in Fig. 3B). On the intraoperative lateral knee radiograph, the native bony MTP was defined as a line tangent to the subchondral plate of the MTP (line m₃). The cartilage slope of the MTP was defined as a line of the probe axis placed on the medial second quarter of the MTP with cartilage remnants (line l₃). The angle formed by these two lines was measured (angle γ in Fig. 3C).

Eighty-four and 66% of total knees had an anterior “low” and a posterior “low” osteophyte, respectively. In 14 knees with Grade 4 OA, one knee (7%) had a “high” posterior osteophyte. Although the number of knees with anterior and/or posterior osteophytes increased with the K-L grade, the number of knees with anterior osteophytes increased from the lower grade compared with the number of knees with posterior osteophytes.

Fifty-six and 37% of the total knees had anterior “moderate” and posterior “moderate” depression, respectively. Although knees with anterior and/or posterior depression increased with the K-L grade, the number of knees with anterior depression increased from the lower grade compared with the number of knees with posterior depression. There was no knee with anterior and/or posterior “severe” depression in these 73 knees.

Angular measurements were performed for all knees except for one knee with a posterior “high” osteophyte. The mean angle of the total knees between the two tangential lines (line m and line l in Fig. 2) on the sagittal view of MRI was −0.8° ± 0.7° (−2.6°–1.0°, n = 72). There was no significant difference between the mean angle in Grade 2 (−0.6° ± 0.5°, −1.9°–0.0°, n = 13), Grade 3 (−0.7° ± 0.5°, −1.6°–0.5°, n = 46), and Grade 4 (−0.9° ± 1.1°, −2.5°–1.0°, n = 13) (Fig. 5). The angle between the two tangential lines in one knee with the posterior “high” osteophyte was 3.2 degrees.

There was no knee with anterior and/or posterior “high” osteophytes in Study II. Furthermore, there was no knee with anterior and/or posterior “severe” depression. The mean angle between the two tangential lines (lines m and l in Fig. 2) on the preoperative sagittal MRI was −1.0° ± 0.7° (−2.8–0.3, n = 36). The mean angle between the probe axis and the bony MTP on the intraoperative lateral knee radiograph (lines m₃ and l₃ in Fig. 3C) was −0.6° ± 0.4° (−1.7–0.0, n = 36). The mean angle on the intraoperative lateral knee radiograph was smaller than that on the sagittal MRI (p < 0.001, paired t test, n = 36, Fig. 6A). A weak but significant correlation between the angles on the sagittal MRI and on the lateral knee radiograph was observed (r = 0.38, p < 0.05, Fig. 6B). The mean angles of the preoperative PTS and the implant PTS were 8.5° ± 2.4 (3.3–14.1) and 7.9° ± 2.2° (3.4–11.6), respectively, and there was a significant difference between these mean angles (p < 0.05, paired t test, n = 36, Fig. 6C). The mean angle difference between the preoperative and implant PTS was −0.5° ± 1.5° (−2.8–1.8) (Fig. 6D). The RMS error of the tibial slope was 1.6° with this method. There was a strong correlation between the preoperative and implant PTS (r = 0.83, p < 0.001, Fig. 6E).

An 88-year-old male patient presented with grade 4 knee OA (Fig. 7A). Preoperative MRI showed anterior and posterior “low” osteophytes (*) and anterior and posterior “mild” depression on the MTP (Fig. 7B). A hook probe (Fig. 7C) was placed on the medial second quarter of the MTP in an AP direction. Then, an intraoperative lateral knee radiograph was taken (Fig. 7D). A gauge (g in Fig. 7E) inserted into the slot of the tibial cut block was set parallel to the probe (p in Fig. 7E). A postoperative lateral knee radiograph demonstrated that the native bony PTS was recreated (Fig. 7F).