Measuring the X-Ray: Problems and Solutions

T. Derek V. Cooke, M.D., MB, BChir, FRCSC
Allan Scudamore, PhD
Kingston, ON

The value of radiographs (X-rays) in musculoskeletal care is beyond dispute, yet the time-honoured technology is cumbersome in terms of film processing, handling, storage and retrieval. Fortunately, digital imaging methods now promise greater convenience and lower costs in the long-term. Currently there are two options: Computer Radiography (CR) using radiosensitive phosphor screens read by laser scanners, and Direct Radiography (DR) where image capture is immediate via sensors hard-wired to the computer. These systems, despite their high capital costs for conversion, are being adopted by many institutions. A central feature of them is PACS (Picture Archiving Computer Systems) which manage the data storage and retrieval networks. Apart from the better logistics afforded by these systems, digital images have a range of clinical advantages including better image quality (sharpness, contrast, magnification etc.) and facilitation of metric imaging.

Despite these advantages, digital image capture does not necessarily provide a more reliable radiograph. Imprecision arises from other sources like poor positioning of the patient, and image distortion due to parallax. Standardized Imaging is a concept designed to limit, or correct for, these sources of error. An example of such a method is Standardized Knee Imaging (SKI**)^1-3.

The patient is set up in a standardized position in a frame (Figure 1), which also presents one or more registration grids into the X-ray field. The grid image(s), superimposed on the target image, provide a means of correction, by digital processing, to compensate for the distortion error. Either film or digital image capture is appropriate⁴. Using SKI we can expect reproducibility of angular measurements to be within 1.5 and dimensional measures within 1.4mm1. This leads to better clinical evaluations^5-7.

For example, Figure 2 shows results of a realignment osteotomy, and Figure 3 shows how the change in limb alignment can reflect focal attrition of plastic in a TKA, even before symptoms have occurred⁸. Likewise, in knee osteoarthritis (where deterioration of the joints has been correlated with alignment change)^9,10, progression of the disease may be monitored by standardized imaging. Also in the clinic, the method offers standardized metric imaging for preoperative sizing of implants of all types.

In the area of research, SKI has proved to be an invaluable measurement tool. For example, in population studies of varus-aligned osteoarthritis (OA), an unexpected femoral abnormality (reduced femoral valgus angle) was discovered to be a significant feature of the disease^11,12. The findings have alerted us to the option of femoral corrective osteotomies in may cases of varus OA (Figure 4)¹³. Priorities in the further development of standardized imaging are the adoption of digital image capture (where not already applied), and development of software for automated analysis of images, based on recognition of registration markers and anatomical landmarks.

Further, the capability for standardized metric imaging is obviously relevant to surgical planning via manipulative software, and there is room for development here. In some cases, this approach could circumvent the need for more costly procedures like CT.

References

1. Siu D., Cooke T.D.V., Broekhoven L.D., Lam M., Fisher B., Saunders G., and Challis T.W. A standardized technique for lower limb radiography. Practice, applications, and error analysis. Invest Radiol 26:71-77, 1991

2. Cooke, T.D.V., Sorbie C., Scudamore R.A., Bryant J.T., Siu D., and Fisher B. A quantitative approach to radiography of the lower limb. J Bone Joint Surg 73B: 715-720, 1991

3. Sanfridsson J., Ryd L., Eklund K., Kouvaras Y., and Jonsson K. Angular configuration of the knee. Comparison of the conventional measurements and the QUESTOR precision radiography system. Acta Radiologica 37:633-638, 1996.

4. Sanfridsson J., Svahn G., Jonsson K., and Ryd L. Computed Radiography for characterisation of the weight-bearing knee. Acta Radiologica 38:514-519, 1997.

5. Cooke, T.D.V., Price, N., Fischer, B., and Hedden, D.. The inwardly pointing knee. An unrecognized problem of external rotational mal-alignment Clin Orthop 260:56-60, 1990.

6. Harrison M.M., Cooke T.D.V., Fisher S.B., and Griffin M.P. Patterns of knee arthrosis and patellar subluxation. Clin Orthop 309:56-63, 1994.

7. Cooke T.D.V., Scudamore R.A., Greer W. Axial alignment of the lower limb and its association with disorders of the knee. In, Guest Ed. Minas T, Essentials of Alignment and Cartilage Repair of the Knee. In Eds Drez, D and DeLee J: Operative Techniques in Sports Medicine WB Saunders. 2000; 8. pp98-107.

8. Cornwall G.B., Bryant J.T., Hansson C.M., Rudan J., Kennedy L.A., and Cooke T.D.V. A quantitative technique for reporting the surface degradation of UHMWPE components of retrieved total knee replacement. J Appl Biomater 6:9-18, 1995.

9. Cerejo R., Dunlop D., Cahue S., Channin D., Song J., and Sharma L. The influence of alignment on the risk of knee osteoarthritis progression according to baseline stage of disease. Arth Rheum 46:2632-2636, 2002.

10. Cooke T.D.V., Kelly B.P., Harrison L., Mohamed G., and Khan B. Radiographic grading for knee osteoarthritis: a revised scheme that relates to alignment and deformity. J Rheumatol :26:641-644, 1999.

11. Cooke, T.D.V., Li J., Scudamore A., Wyss U., Bryant T., and Costigan P. Axial lower-limb alignment: comparison of knee geometry in normal volunteers and osteoarthritis patients. Osteoarthr Cart 5:39-47, 1997.

12. Cooke T.D.V., Harrison L., Khan B., Scudamore R.A., and Chaudhary M.A. Analysis of limb alignment in the pathogenesis of osteoarthritis: a comparison of Saudi Arabian and Canadian cases. Rheumatol Int 22:160-164, 2002.

13. Cooke T.D.V., Scudamore R.A., and Greer W. Varus knee osteoarthritis: whence the Varus? J. Reumatol. 30:2521-2523, 2003.

**Footnote. SKI or Standardized Knee Imaging describes the proprietary techniques developed by OAISYS Inc to obtain Metric Images of the lower limb and their analysis. See www.oaisysmedical.com for details.

Figure Legends

Figure 1: SKI frame with turntable to position patient. Registration grid is mounted to the frame. Turntable is rotated for frontal and orthogonal views.

Figure 2ab: SKI radiographs and digital analysis (frontal plane) of realignment osteotomy to right leg. Note the change of Hip-Knee-Ankle (HKA) angle from varus 13.5o to valgus +3o. The dots shown on the images are the registration markers used for parallax correction.

Figure 3: SKI radiograph and analysis (frontal plane) of failed Rt. TKA medial collapse. The progressive change in HKA (now at -21) indicates bearing wear.

Figure 4: SKI radiographs (frontal plane) of femoral valgus osteotomy in a case of varus knee OA before and after femoral valgus Osteotomy. Note that varus was not corrected by the previous tibial re-alignment.