Magnetic Resonance Imaging 32 (2014) 1085–1090

Contents lists available at ScienceDirect

Magnetic Resonance Imaging

j ourna l homepage: www.mr i journa l .com

Assessment of T1, T1ρ, and T2 values of the ulnocarpal disc in healthy

subjects at 3 tesla

Isabel Rauscher a, Benjamin Bender a,e, Gerd Grözinger a, Oliver Luz a, Rolf Pohmann b, Michael Erb c,Fritz Schick d, Petros Martirosian d,⁎a Department of Diagnostic and Interventional Radiology, Eberhard-Karls University Tübingen, Tübingen, Germanyb Max Planck Institute for Biological Cybernetics, Magnetic Resonance Center, Tübingen, Germanyc Department of Biomedical Magnetic Resonance, Eberhard-Karls University Tübingen, Tübingen, Germanyd Section on Experimental Radiology, Eberhard-Karls University Tübingen, Tübingen, Germanye Department of Diagnostic and Interventional Neuroradiology, Eberhard-Karls University Tübingen, Tübingen, Germany

⁎ Corresponding author at: Eberhard-Karls UniverExperimental Radiology,Hoppe-Seyler-Str. 3, 72076 Tübin29 87751; fax: +49 7071 29 5392.

E-mail address: petros.martirosian@med.uni-tuebing

http://dx.doi.org/10.1016/j.mri.2014.05.0100730-725X/© 2014 Elsevier Inc. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 28 October 2013Revised 3 April 2014Accepted 26 May 2014

Keywords:Triangular fibrocartilage complex (TFCC)Ulnocarpal discWrist MRIMR relaxometry

Objective: The purpose of this study was to implement clinically feasible imaging techniques fordetermination of T1, T1ρ, and T2 values of the ulnocarpal disc and to assess those values in a cohort ofasymptomatic subjects at 3 tesla. Resulting values were compared between different age groups, sinceformer histological findings of the ulnocarpal disc indicated frequent early degenerative changes of thistissue starting in the third decade of life, even in asymptomatic subjects.Materials and methods: Twenty-seven healthy subjects were included in this study. T1 measurements wereperformed using 3D spoiled gradient-echo (GRE) sequence with variable flip angle. A series of T1ρ and T2-weightedimages was acquired by a 3D GRE sequence after suitable magnetization preparation. T1,T1ρ, and T2 maps of theulnocarpal disc were calculated pixel-wise. Representative mean values from extended regions were analysed.

Results:Mean T1 values of the ulnocarpal disc ranged from722 ms in a 39 year-old subject to 1264 ms in a 65 year-old subject, T1ρ ranged from 9.2 ms (26 year-old subject) to 25.9 ms (65 year-old subject). Calculated T2 valuesshowed a large range from 4.1 ms to 22.3 ms. T1ρ and T1 values tended to increase with age (p b 0.05), whereas T2did not.Conclusions: MR relaxometry of the ulnocarpal disc is feasible, and T1,T1ρ, and T2 values show modest variance inasymptomatic subjects. The potential of relaxation mapping to reveal relevant structural changes in patients has tobe investigated in further studies.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Osteoarthritis of the wrist is one of the most common conditionsencountered in clinical orthopaedic practice. Apart from the secondarytype of arthritis for example after fracture or ligament disruption, thedegenerative primary type of arthritis shows high prevalence. Pain atthe ulnar aspect of the wrist is a diagnostic challenge for radiologistsbecause of the small and complex anatomical structures involved. Thetriangular fibrocartilage complex (TFCC) consists of the ulnocarpal disc,dorsal and palmar radioulnar ligaments, the meniscus homologue, andsheath of the extensor carpi ulna [1]. The TFCC transmits load from thewrist to the ulna and stabilizes the distal radioulnar joint. As themeniscus for the knee, the ulnocarpal disc of thewrist serves as a shockabsorber during movement. The central fibrocartilaginous portionabsorbs 20% of the axial or compressive load transmitted through the

sity Tuebingen, Section ongen,Germany. Tel.:+497071

en.de (P. Martirosian).

wrist [2,3]. Damage to the TFCC is a major cause of wrist pain and iscommonly reported, particularly in elderly individuals [4], and thepresence of nociceptive fibers within the disc [5] suggests that it isimplicated in wrist pain.

The ulnocarpal disc of the distal radioulnar joint anatomy wasstudied in 109 wrist joints from 64 fresh adult cadavers and wasfound to be a strong fibrocartilaginous semi-circular biconcavestructure [6]. A study of Mikic et al. [7] showed that degenerativechanges in the ulnocarpal disc begin in the third decade and becomeprogressively more frequent and more severe with advancing age.The changes comprise reduced cellularity, loss of elastic fibers,mucoid degeneration of the ground substance, exposure of collagenfibers, fibrillation, erosion, ulceration, abnormal thinning, and,ultimately, disc perforation. An immunohistochemical study of theTFCC [8] in nine elderly individuals demonstrated that the ulnocarpaldisc is a fibrocartilaginous structure which labels for type II collagenand aggregan, a proteoglycan-typical marker of a cartilaginousphenotype, whereas the meniscal homologue of the TFCC is a morefibrous structure.

1086 I. Rauscher et al. / Magnetic Resonance Imaging 32 (2014) 1085–1090

Characterization of the TFCC by non-invasive imaging ischallenging due to its small dimensions. Traditionally, imagingtools for investigation of wrist pain include x-ray, wrist arthrography,CT, and magnetic resonance imaging (MRI). Non-invasive diagnosingof tears in vivo in the ulnocarpal disc is often difficult and manyclinicians favor arthroscopy over MRI, because of its low sensitivity fordetecting smaller tears. However, as the quality of MRI improvessteadily and new sequence techniques are emerging,MRI is likely to beincreasingly useful for diagnosing TFCC damage.

Degeneration of the TFCC is a progressive and irreversiblecondition which finally might result in perforation of the ulnocarpaldisc. Therefore, a sensitive method for detecting early structuralchanges in the biochemical composition of the ulnocarpal disc wouldbe valuable not only for early diagnosis, but also for monitoring theprogression of disease.

Several studies have shown the potential of new MRI techniquessuch as dGEMRIC [9,10], diffusion-weighted imaging, T1ρ mapping[11,12], and T2 mapping [12–14] to reveal early degenerativechanges in the biochemical composition of hyaline cartilage of theknee. T1ρ and T2 relaxation times reflect changes associated withproteoglycan and collagen content and tissue hydration. In the kneejoint, in vivo studies have shown increased T1ρ and T2 values incartilage for patients with mild osteoarthritis [12,15].

Further studies have been conducted in the meniscus of younghealthy subjects with resulting T1ρ values ranging from approx. 10 msto 20 ms, and T2 values ranging from approx. 8 ms to 16 ms [16].Recent studies focused on effects of acute [17] and long lasting [18]load on the meniscus: In both studies a slight but significantprolongation of T1ρ and T2 values (up to approx. 3 ms prolongationaftermarathon run) has been reported. Subjects with higher age in thefifties and sixties tended to show relatively long relaxation times T1ρand T2 values of the meniscus (approx. 3 ms to 5 ms longer than inyoung subjects), but it should be mentioned that subjects with higherage had symptoms of osteoarthritis in the respective study [16].

To our knowledge, there has been no report on quantitativerelaxometric characterization of the ulnocarpal disc. The purpose ofthis study was to develop and apply MR techniques with clinicallyfeasible examination time for reliable assessment of T1, T1ρ, and T2 of theulnocarpal disc in thehumanwrist at 3.0 T. Derived T1, T1ρ, and T2 valuesof asymptomatic subjects were compared for different age groups.

2. Materials and methods

2.1. Subjects

For this study 27 subjects, all asymptomatic volunteers, wererecruited. Written informed consent was obtained from all subjects.Inclusion criteria were good health according to medical history,physical examination as well as the absence of contraindications toMR imaging. Subjects were included, if they had no clinical evidenceof wrist osteoarthritis; they had to have intact joint functionwith fullstrength and no history of chronic or frequent wrist pain. Furtherexclusion criteria were inflammatory arthritis, wrist affectionssecondary to other causes (acute or chronic infection, metabolicabnormalities, previous surgery or previous fracture) and chronic ortraumatic lesions of the TFCC. The age of the healthy subjects rangedfrom 23 to 65 years (n = 18 male, n = 9 female). Three subjectshad to be excluded from the study after undergoing MRI examina-tion because of strong motion related artifacts in MR images.

2.2. Imaging techniques

All 27 subjects were examined using a 3.0 T whole-body MRIscanner (Tim-Trio, Siemens Healthcare, Erlangen, Germany). Mea-surements of the dominant wrist were performed with a dedicated

circular–polarized transmit–receive wrist coil (USA Instruments,Aurora, OH, USA). The dominant hand was identified by asking thesubject which hand he or shewould use to write (right wrist n = 25,left wrist n = 2). The subjects were positioned with the upperextremity extended overhead, and the wrist was imaged inpronation. The total examination time for recording anatomicalimages and raw data necessary for reliable T1, T1ρ, and T2 mapping(including B1 mapping) was 33 min.

Morphology of the TFCC and other structures of the wrist wasassessed by using a coronal fat-saturated fast-spin-echo sequencewith TR/TE = 3700/28 ms, echo train length of 11, acquisitionbandwidth of 160 Hz/pixel, field-of-view of 68 × 68 mm2, slicethickness of 1.5 mm, 20 slices, 240 × 320 matrix, 2 acquisitions, andscan time of 5 min.

2.3. T1 measurements

Measurements of T1 were performed using 3D spoiledgradient echo (GRE) sequence with two different flip angles of8° and 42°, respectively (according to [19,20]). The imagingparameters were: TR/TE = 80/2.96 ms, field-of-view =68 × 68 mm2, slice thickness = 3 mm, matrix = 96 × 192, nom-inal voxel size = 0.7 × 0.35 × 3 mm3, number of slices = 10,acquisition bandwidth = 260 Hz/pixel, and acquisition timeof 2 min.

2.4. B1 measurements

Three-dimensional B1 maps were calculated using the actual flipangle (AFI) imaging technique [21] with the following parameters:TR1/TR2/TE = 20/100/5 ms, flip angle = 60°, field-of-view =68 × 68 mm2, slice thickness = 3 mm, matrix = 48 × 96, nominalvoxel size = 1.4 × 0.7 × 3 mm3, number of slices = 20, acquisitionbandwidth = 260 Hz/pixel, acquisition time = 2:58 min.

2.5. T1ρ and T2 measurements

A series of T1ρ and T2-weighted images was acquired with a 3DGRE sequence with T1ρ or T2 magnetization preparation, respectively[22–25]. The imaging parameters were: TR/TE = 80/2.96 ms,flip angle = 20°, field-of-view = 68 × 68 mm2, slice thickness =3 mm,matrix = 96 × 192, nominal voxel size = 0.7 × 0.35 × 3 mm3,number of slices = 10, acquisition bandwidth = 260 Hz/pixel, andacquisition time of 2 min. T1ρmagnetization preparationwas performedusing a spin-lock techniquewith a pulse series of 90°y–SLx–90°-y, whereSL is the spin-lock pulse, according to [22]. For T2 magnetizationpreparation, a pulse sequence of 90°y–T2prep/2–180°x–T2prep/2–90°−y

was applied, where T2prep is the T2-sensitive preparation time.This approach was initially described by Mugler et al. [24]and applied for T2 mapping in tissue [26]. In order tocalculate T1ρ and T2 maps, measurements were performedwith spin-lock time (TSL) and T2prep time of 2, 4, 8, 16 and 32 ms,respectively.

2.6. T1 quantification

In this work, T1 was obtained on the basis of the variable flip angletechnique [19,20] allowing for fast three-dimensional T1 mapping.The regular Ernst equation [27] describes signal yield in spoiled GREimaging with the excitation flip angle β and repetition time TR andcan be reordered in the following linear form:

SGREsinβ

¼ E1SGREtanβ

þM0 1−E1ð Þ; ð1Þ

1087I. Rauscher et al. / Magnetic Resonance Imaging 32 (2014) 1085–1090

the slope E1 = exp(−TR/T1) can be estimated by linear regression,allowing for T1 calculation according to:

T1 ¼ − TRlnE1

: ð2Þ

2.7. B1 quantification

Accurate T1 quantification using the variable flip angle techniqueand also calculation of T1ρ and T2 values (reported in the nextsection) need knowledge of the effective flip angle. For assessment ofthe B1 field distribution the 3D AFI method was applied which usestwo interleaved FLASH acquisitions with repetition times TR1 andTR2. If longitudinal relaxation of the magnetization between therepetitions can be linearized (TR1 b TR2 b b T1), flip angle (θ) mapscan be extracted from the two resulting images according to [21]:

θ≈ arccosrn−1n−r

; ð3Þ

in which n = TR2/TR1, r = S2/S1 and S1, S2 are the signals acquiredduring TR1 and TR2, respectively.

2.8. T1ρ and T2 quantification

Due to the use of relatively short repetition times, the signalintensity in T1ρ-weighted measurements depends also on T1relaxation time, because of incomplete recovery of the longitudinalmagnetization [28] in the TR cycle. In the series of measurements forT1ρ, TR is fixed, whereas the TSL period varies in duration. Accordingto the statements in the appendix of [22] the fact has to beconsidered that free longitudinal relaxation in the steady stateis restricted to the time interval (TR − TSL). As a consequence,signal intensity in the T1ρ-weighted GRE experiment can be writtenas follows:

ST1ρ¼ S0sinα

e−TSL=T1ρ 1−e− TR−TSLð Þ=T1

� �

1−e−TSL=T1ρe− TR−TSLð Þ=T1 cosαþ Const; ð4Þ

with α as flip angle of the excitation pulse of the GRE sequence, S0 asthe amplitude of the signal and Const introduced to account forbackground noise. To determine the T1ρmap, the signal intensities inthe T1ρ-weighted images were fitted using Eq. (4).

Those effects of T1 relaxation time must be considered as well forthe T2 measurements. The timing of the sequences and the effects ofvariable preparation times on the course of longitudinalmagnetizationare equal to the above mentioned T1ρ measurements. For this reasonlongitudinal relaxation effects in the relevant but variable time period(TR- T2prep) must be considered accordingly using Eq. (5):

ST2prep¼ S0sinα

e−T2prep=TE 1−e− TR−T2prepð Þ=T1

� �

1−e−T2prep=TEe− TR−T2prepð Þ=T1 cosαþ Const ð5Þ

2.9. Image analysis

First, all images were co-registered using Statistical ParametricMapping (SPM5, Wellcome Trust Centre for Neuroimaging, London,UK) software. Calculation of B1 maps were performed on the scannerusing the image calculation environment (ICE) software from themanufacturer. T1 maps were calculated using Eqs. (1) and (2) andsubsequently, T1ρ and T2 maps of the ulnocarpal disc were generatedbased on Eqs. (4) and (5). For segmentation, a region of interest was

defined by using in-house software developed with Matlab (Math-works, Natick, MA).

Due to its small size, the ulnocarpal disc was segmentedmanuallyusing the coronal three-dimensional T1ρimages. The best slice forquantification was determined under consideration of potentialpartial volume effects and best visibility. The ulnocarpal disc of theTFCC was defined as the central–radial hypointense structureconnecting with the dorsal and palmar radioulnar ligaments and tothe meniscus homologue. Segmentation and quantification wereperformed for each MRI scan separately by one radiologist (I.R.).

2.10. Statistical analysis

All statistical testing was performed with the Statistics Toolbox(Version 7.4) of Matlab. Descriptive statistical evaluation wasperformed for the correlation between T1, T1ρ, and T2 of theulnocarpal disc and age. A linear correlation was assumed and thePearson product–moment correlation coefficient was calculated.After classifying the subjects into age groups, the differencesbetween mean T1, T1ρ, and T2 values of the ulnocarpal disc (withmean and standard deviations) were evaluated using t-test.

3. Results

Fig. 1 depicts mean T1, T1ρ and T2 values of the individualulnocarpal disc. Furthermore, mean values and standard deviationswere calculated for five age groups: 20–29 years (n = 8; mean26 years), 30–39 years (n = 5; mean 34 years), 40–49 years (n = 4;mean 46 years), 50–59 years (n = 5; mean 56 years), and 60–69years (n = 5; mean 64 years). T1, T1ρ, and T2 values of each age groupare summarized in Table 1.

Mean T1 values in representative ROIs of the ulnocarpal discranged from 722 ms (39 year-old subject) to 1264 ms (65 year-oldsubject). Variability of T1 values in younger subjects was modest andless pronounced than in the groups with higher age. T1 valuesincreased significantly with age (r = 0.552, p = 0.003). There was alow mean T1 level in subjects of 25–35 years with an increase of T1values in subjects between 35 and 55 years, and a stagnation of T1values from 55 years onwards (Fig. 1a).

Mean T1ρ values of the ulnocarpal disc ranged from 9.2 ms for a26-year old subject to 25.9 ms for a 65 year-old subject. There was atendency of increasing T1ρ with subject age (r = 0.531, p = 0.004).However, strong prolongation of T1ρ mainly occurred in ourasymptomatic subjects aged 55 years or more (Fig. 1b).

Mean T2 values (Fig. 1c) of the disc were more variable than T1and T1ρ. Mean values varied between 4.1 ms and 22.3 ms. Nosignificant linear correlation of T2 parameters with age could beshown (r = 0.250, p = 0.209).

Figs. 2 and 3 depict representative proton-density-weightedcoronal MR images of the wrist (a) and corresponding T1 (b), T1ρ (c)and T2 (d) maps within a segmented ulnocarpal disc of a 31 year-oldand a 65 year-old subject, respectively.

4. Discussion

As the meniscus for the knee, the ulnocarpal disc of thewrist serves as a shock absorber during movement. The centralfibrocartilaginous portion absorbs 20% of the axial or compressiveload transmitted through the wrist. The radiocarpal joint absorbs80% [2,3].

Degenerative changes occur with aging in all components ofsynovial joints, including intra-articular-fibrocartilaginous struc-tures [29]. Age changes in the menisci of the knee joint as well asin the TFCC [7,32] have been studied [30,31]. In a histological study of180 cadaver wrist joints (ranging from premature infants to persons

Fig. 1. Calculated mean T1 (a), T1ρ (b) and T2 (c) of the articular disc vs. age for eachsubject. Individual results are marked with 'x'. The horizontal line connects the meanT1, T1ρ and T2 times in each age group (20–29, 30–39, 40–49, 50–59, 60–69 years)while the vertical line depicts the standard deviation and the mean age of the group

Table 1Mean and standard deviation of T1, T1ρ and T2 values of the ulnocarpal disc in each agegroup (20–29, 30–39, 40–49, 50–59, 60–69 years).

Age group T1 (ms) T1ρ (ms) T2 (ms)

20–29 years (n = 8) 866 ± 58 14.3 ± 3.2 11.8 ± 4.430–39 years (n = 5) 867 ± 99 16.5 ± 3.4 8.7 ± 3.640–49 years (n = 4) 944 ± 92 14.5 ± 2.3 8.1 ± 3.450–59 years (n = 5) 1005 ± 114 18.3 ± 5.4 14.0 ± 7.360–69 years (n = 5) 1021 ± 175 20.6 ± 5.0 13.1 ± 3.4

1088 I. Rauscher et al. / Magnetic Resonance Imaging 32 (2014) 1085–1090

,.

94 years old) [7], no completely normal structure of the ulnocarpaldisc was observed in the TFCC from the fifth decade onward, whileno signs of degeneration were seen in fetuses, and in subjects whodied in the first and second decade of life. In the third decade

degenerative changes began to appear, with the ulnocarpal discbecoming less cellular. Additionally, the incidence of perforationincreased with age.

Our study revealed some congruent results with respect to thehistological findings: There was a significant difference (p b 0.05) inT1 and T1ρ values between subjects aged less than 30 years (n = 8)and subjects with more than 50 years of age (n = 10). Mean T1values increased from 867 ± 57.5 ms (age b 30 years) to 1013 ±139.5 ms (age N 50 years) and T1ρ values from 14.3 ± 3.2 ms(age b 30 years) to 19.5 ± 5.0 ms (age N 50 years). No significantdifference in T1, T1ρ, and T2 values was observed between youngsubjects under 30 years with middle-aged subjects of 30–50 years,although we found a slightly increasing T1 and T1ρ in subjects of theelder group. The comparison of subjects aged 30–50 years versussubjects aged over 50 years revealed a significant difference(p b 0.05) only for T2 values. An onset of changes in the thirddecade of life as revealed by histological studies was not backed up inour study. However, microscopic and our studies were performed indifferent cohorts, and therefore, we would not necessarily expectthat MR properties must show significant changes starting in thetwenties. Furthermore, it is not yet clarified, which grade ofdegenerative changes in histology is necessary to generate corre-sponding effects in MR relaxation maps.

To our knowledge, there have been no quantitative magneticresonance measurements of the ulnocarpal disc before. Akella et al.were the first to report on T1ρ relatedMR imaging of the wrist in vivoat 1.5 T: T1ρ dispersion of wrist cartilage has been described in [33].In this study, T1ρ values of cartilage ranged from 40.5 to 56.6 ms andT2 values from 28.1 to 34.5 ms for various intercarpal locations. T1ρvalues were 32%–63% higher than T2 values. In our study, mean T1ρvalues of the ulnocarpal disc were 47% higher than T2 values.However, it should be mentioned that wrist cartilage and theulnocarpal disc are not directly comparable.

Generally, the TFCC is a relatively small structure in the humanwrist with a thickness of a few millimeters. The length of the discvaries in most cases between 14 and 16 mm and the width between9 and 11 mm [6]. Due to its small size and a slice thickness of 3 mmin our MRI protocol, the ulnocarpal disc could only be covered in twoto three slices. The slice with the best delination of the disc waschosen for analysis. We analyzed the entire ulnocarpal disc withoutconsidering zonal variations. This may be a limitation, and futurestudies with even higher spatial resolution will be required toanalyze T1, T1ρ and T2 values in anatomic subregions of theulnocarpal disc, especially in areas where the disc merges in morefibrous structures, such as the meniscus homologue. MRI wasperformed using a section thickness of 3 mm which was chosen inconsideration of measurement time and SNR. Thinner slices wouldfurther reduce partial volume effects, but would unfortunately leadto unacceptable long examination times at 3 T.

Our study with limited number of participants indicated asignificant dependence of T1ρ on age, most likely reflectingdegenerative processes within the ulnocarpal disc of the TFCC, ashistological cadaver studies of the TFCC indicate [34]. Additionally, T1values of the ulnocarpal disc showed similar significant differencesfor different age groups (p b 0.05). One possible explanation ofincreased T1 in some cases may be the loss of cellularity and theconsecutive increase in tissue hydration.

Relaxation values of meniscus and TFCC are often thought to becomparable due to their fibrocartilage structure. Histological andMR-related changes in degenerative disease can be expected to besimilar. However, we would not necessarily expect very similar agedependence, because the knee joint is mechanically stressed in allsubjects during walking and standing. In contrast, mechanical stressof the wrist is much more dependent on the individual behavior(especially on the regular occupation and sportive activities).

Fig. 2. Anatomical coronal MR image of thewrist of a 31-year oldmale subject (a) and corresponding T1 (b), T1ρ (c) and T2 (d) maps in the area of the ulnocarpal disc. Mean values:T1 = 820 ms, T1ρ = 10.6 ms, T2 = 2.8 ms.

1089I. Rauscher et al. / Magnetic Resonance Imaging 32 (2014) 1085–1090

Lacking stress of the TFCC inmost participants of our study could be agood reason for the lack of marked increase of relaxation time withage in the TFCC in those cases.

Results of thepresented study showamodestvarianceof relaxationtimes T1 and T1ρ in the ulnocarpal disc of asymptomatic subjects,whereas T2 values showed a relatively high less age-dependentvariability. Further studies including symptomatic individuals are

Fig. 3. Anatomical MR image of thewrist of a 65-year old female subject (a) and correspond1130 ms, T1ρ = 21.5 ms, T2 = 10.2 ms.

necessary to find out whether MR relaxometry can help in diagnosisand/or therapy monitoring of degenerative painful affections of theTFCC. Clinical symptoms such as pressure pain or reduced range ofmovement should be evaluated carefully in symptomatic subjects.Correlation of MR features with histology is desired in order to betterunderstand which biochemical changes within the ulnocarpal discmight lead to effects in proton density and relaxation characteristics.

ing T1 (b), T1ρ (c) and T2 (d) maps in the area of the ulnocarpal disc. Mean values: T1 =

1090 I. Rauscher et al. / Magnetic Resonance Imaging 32 (2014) 1085–1090

References

[1] Palmer AK, Werner FW. The triangular fibrocartilage complex of the wrist-anatomy and function. J Hand Surg [Am] 1981;6:153–62.

[2] Palmer AK, Werner FW, Glisson RR, Murphy DJ. Partial excision of the triangularfibrocartilage complex. J Hand Surg [Am] 1988;13:391–4.

[3] Linscheid RL. Biomechanics of the distal radioulnar joint. Clin Orthop Relat Res1992;275:46–55.

[4] Lee DH, Dickson KF, Bradley EL. The incidence of wrist interosseous ligament andtriangular fibrocartilage articular disc disruptions: a cadaveric study. J Hand Surg[Am] 2004;29:676–84.

[5] Cavalcante ML, Rodrigues CJ, Mattar Jr R. Mechanoreceptors and nerve endingsof the triangular fibrocartilage in the human wrist. J Hand Surg [Am]2004;29:432–5 [discussion 436–38].

[6] Mikic ZD. Detailed anatomy of the articular disc of the distal radioulnar joint. ClinOrthop Relat Res 1989;245:123–32.

[7] Mikic ZD. Age changes in the triangular fibrocartilage of the wrist joint. J Anat1978;126:367–84.

[8] Milz S, Sicking B, Sprecher CM, Putz R, Benjamin M. An immunohistochemicalstudy of the triangular fibrocartilage complex of the wrist: regional variations incartilage phenotype. J Anat 2007;211:1–7.

[9] Burstein D, Gray M. New MRI techniques for imaging cartilage. J Bone Joint SurgAm 2003;85-A(Suppl. 2):70–7.

[10] Bashir A, Gray ML, Boutin RD, Burstein D. Glycosaminoglycan in articularcartilage: in vivo assessment with delayed Gd(DTPA)(2-)-enhanced MRimaging. Radiology 1997;205:551–8.

[11] Regatte RR, Akella SV, Wheaton AJ, Lech G, Borthakur A, Kneeland JB, et al. 3D-T1rho-relaxation mapping of articular cartilage: in vivo assessment of earlydegenerative changes in symptomatic osteoarthritic subjects. Acad Radiol2004;11:741–9.

[12] Li X, Benjamin Ma C, Link TM, Castillo DD, Blumenkrantz G, Lozanzo J, et al. Invivo T(1rho) and T(2) mapping of articular cartilage in osteoarthritis of the kneeusing 3 T MRI. Osteoarthritis Cartilage 2007;15:789–97.

[13] Dunn TC, Lu Y, Jin H, Ries MD, Majumdar S. T2 relaxation time of cartilage at MRimaging: comparisonwith severity of knee osteoarthritis. Radiology 2004;232:592–8.

[14] Mosher TJ, Dardzinski BJ, Cartilage MRI. T2 relaxation time mapping: overviewand applications. Semin Musculoskelet Radiol 2004;8:355–68.

[15] Regatte RR, Akella SV, Lonner JH, Kneeland JB, Reddy R. T1rho relaxationmapping in human osteoarthritis (OA) cartilage: comparison of T1rho with T2.J Magn Reson Imaging 2006;23:547–53.

[16] Rauscher I, Stahl R, Cheng J, Li X, HuberMB, Luke A, et al.Meniscalmeasurements ofT1rho and T2 at MR imaging in healthy subjects and patients with osteoarthritis.Radiology 2008;249:591–600.

[17] Subburaj K, Souza RB, Wyman BT, Le Graverand-Gastineau MP, Li X, Link TM,et al. Changes in MR relaxation times of the meniscus with acute loading: an invivo pilot study in knee osteoarthritis. J Magn Reson Imaging 2013. http://dx.doi.org/10.1002/jmri.24546.

[18] Stehling C, Luke A, Stahl R, Baum T, Joseph G, Pan J, et al. Meniscal T1rho and T2measured with 3.0 T MRI increases directly after running a marathon. SkeletalRadiol 2011;40:725–35.

[19] Deoni SC, Peters TM, Rutt BK. Determination of optimal angles for variablenutation proton magnetic spin-lattice, T1, and spin-spin, T2, relaxation timesmeasurement. Magn Reson Med 2004;51:194–9.

[20] Deoni SC, Rutt BK, Peters TM. Rapid combined T1 and T2mapping using gradientrecalled acquisition in the steady state. Magn Reson Med 2003;49:515–26.

[21] Yarnykh VL. Actual flip-angle imaging in the pulsed steady state: a method forrapid three-dimensional mapping of the transmitted radiofrequency field. MagnReson Med 2007;57:192–200.

[22] Aronen HJ, Ramadan UA, Peltonen TK, Markola AT, Tanttu JI, Jääskeläinen J, et al.3D spin-lock imaging of human gliomas. Magn Reson Imaging 1999;17:1001–10.

[23] Pakin SK, Xu J, Schweitzer ME, Regatte RR. Rapid 3D-T1rho mapping of the kneejoint at 3.0 T with parallel imaging. Magn Reson Med 2006;56:563–71.

[24] Mugler III JP, SpragginsTA, Brookeman JR. T2-weighted three-dimensionalMP-RAGEMR imaging. J Magn Reson Imaging 1991;1:731–7.

[25] Nugent AC, Johnson GA. T1rho imaging using magnetization-prepared projectionencoding (MaPPE). Magn Reson Med 2000;43:421–8.

[26] Giri S, Chung YC, Merchant A, Mihai G, Rajagopalan S, Raman SV, et al. T2quantification for improved detection of myocardial edema. J Cardiovasc MagnReson 2009;11:56–60.

[27] Ernst RR, AndersonWA. Application of Fourier transform to magnetic resonancespectroscopy. Rev Sci Instrum 1966;37:93–102.

[28] Li X, Han ET, Busse RF, Majumdar S. In vivo T(1rho) mapping in cartilage using3D magnetization-prepared angle-modulated partitioned k-space spoiledgradient echo snapshots (3D MAPSS). Magn Reson Med 2008;59:298–307.

[29] Barnett CH, Davies DV, McConaill MA. Synovial joints. Their structure andmechanics. Illinois: Charles C. Thomas, Springfield; 1961.

[30] Noble J, Hamblen DL. The pathology of the degenerate meniscus lesion. J BoneJoint Surg (Br) 1975;57:180–6.

[31] Refior HJ. Comparative experimental studies of the micromorphology of pre-arthrosis using rabbit knee joints. Z Orthop Ihre Grenzgeb 1974;112:706–9.

[32] Viegas SF, Ballantyne G. Attritional lesions of the wrist joint. J Hand Surg [Am]1987;12:1025–9.

[33] Akella SV, Regatte RR, Borthakur A, Kneeland JB, Leigh JS, Reddy R. T1rho MRimaging of the human wrist in vivo. Acad Radiol 2003;10:614–9.

[34] Nakamura T, Yabe Y. Histological anatomy of the triangular fibrocartilagecomplex of the human wrist. Ann Anat 2000;182:567–72.