Direct MR arthrography of the wrist- value in detecting complete and partial defects of intrinsic ligaments and the TFCC in comparison with arthroscopy

Abstract

To assess the value of two-compartment magnetic resonance (MR) wrist arthrography in comparison with diagnostic arthroscopy for the evaluation of defects of the triangular fibrocartilage complex (TFCC) and intrinsic ligaments in patients with refractory wrist pain. The direct MR arthrographies were evaluated following arthroscopic classification with consideration of complete and partial defects. The distinction between these types of lesions has clinical implications for treatment procedures such as cast immobilization, arthroscopic debridement, surgical repair or partial intercarpal arthrodesis. Seventy-five patients (25 female, 50 males, mean age 38.3 years) who suffered from refractory wrist pain without radiography evidence of carpal instability underwent two-compartment wrist MR arthrography. Under aseptic conditions a solution of gadopentate dimeglumine and iodinated contrast agent (concentration 2.5 mmol/l) was injected into the radiocarpal and midcarpal joints under fluoroscopy guidance. Using a scanner of 1.5 T field strength and a wrist-coil following sequences were acquired: coronary and sagittal T (1)-weighted spin-echo (SE) sequences (TR 500 ms, TE 25 ms, matrix 512 x 512.3 mm) and coronary fast low angle shot (FLASH) 3D sequences (TR 24 ms, TE 11 ms, matrix 256 x 256, 1.5 mm, flip angle 50 degrees). All patients underwent subsequent arthroscopy of the wrist. The direct MR arthrographies were evaluated retrospectively by two observers experienced in the diagnosis of wrist pathology. They were not aware of the clinicial, arthrographic and arthroscopic findings. Pathology of the scapholunate ligament was classified according to the guidelines of the German Society of Hand Surgery (DGH), lesions of the lunotriquetral ligament according to Hempfling and lesions of the TFCC according to Palmer. Twenty-five complete and 47 partial defects were detected arthroscopically (TFCC: 21/20, scapholunate ligament: 3/18, lunotriquetral ligament: 1/9). The TFCC showed a higher prevalence for degenerative lesions (11 type 2C-lesions and 20 type 2A/B lesions) than for traumatic lesions (5 type 1A lesions, 5 type 1D lesions). For direct MR arthrography, the obtained sensitivities and specificities in assessing complete defects were 96 % and 99.6 % (T (1)-weighted SE) and 92 % and 100 % (FLASH 3D), respectively. For all partial defects, sensitivities and specificities were 68.1 % and 93.3 % (T (1)-weighted SE) and 63 % and 96.1 % (FLASH 3D), respectively. For depicting partial defects of the scapholunate ligament the T (1)-weighted SE sequence (83.3/95.5 %) was superior to the FLASH 3D sequence (64.7/96.6 %), p < 0.05. For the evaluation of the TFCC (T (1)-weighted SE: 65/94.4 %, FLASH 3D: 70/94.6 %) and the lunotriquetral ligament (T (1)-weighted SE: 44/89.4 %, FLASH 3D: 44 /96.7 %), direct MR arthrography showed an insufficient correlation with arthroscopy. Direct MR arthrography proved to be of equal value compared with diagnostic arthroscopy in detecting complete defects of the intrinsic ligaments and the TFCC. The method has the potential of replacing diagnostic arthroscopy for the evaluation of the intrinsic ligaments and the TFCC. The T (1)-weighted SE sequence appeared to be superior to the FLASH 3D sequence in evaluating partial defects of the scapholunate ligament. Direct MR arthrography did not reliably detect partial defects of the TFCC and the lunotriquetral ligament.