Effects of repetitive freeze–thawing cycles on T2 and T2* of the Achilles tendon

European Journal of Radiology 83 (2014) 349–353

Contents lists available at ScienceDirect

European Journal of Radiology journal homepage: www.elsevier.com/locate/ejrad

Effects of repetitive freeze–thawing cycles on T2 and T2* of the Achilles tendon Eric Y. Chang a,b,∗ , Won C. Bae b , Sheronda Statum b , Jiang Du b , Christine B. Chung b,a a b

Department of Radiology, VA San Diego Healthcare System, 3350 La Jolla Village Drive, San Diego, CA 92161, United States Department of Radiology, University of California, 200 West Arbor St., San Diego, CA 92103, United States

a r t i c l e

i n f o

Article history: Received 18 April 2013 Received in revised form 24 August 2013 Accepted 12 October 2013 Keywords: Achilles tendon Freeze–thaw T2 T2* UTE

a b s t r a c t Introduction: In this study we sought to evaluate the effects of multiple freezing and thawing cycles on two MR parameters to study Achilles tendon, T2 and T2*. Materials and methods: Four fresh Achilles tendons were imaged on a 3T clinical scanner and again after 1, 2, 4, and 5 freeze–thaw cycles with spin-echo (SE) and ultrashort echo time (UTE) sequences. Regions of interest were manually drawn over the entire Achilles tendon and mono-exponential curves were used to determine T2 and T2* relaxation times. Results: There was no statistically significant difference in mean T2 or T2* values between the fresh specimens and after subsequent cycles of freeze–thaw treatment (p > 0.1). Linear regression between SE T2 values at baseline and after successive freeze–thaw cycles demonstrated moderate agreement (r = 0.60) whereas UTE T2* values at baseline and after successive-freeze thaw cycles demonstrated strong agreement (r = 0.92). Conclusion: These findings suggest that changes between specimens seen in vitro are due to factors other than frozen storage. Furthermore, our results suggest that there is stronger agreement between baseline (fresh) and successive freeze–thaw T2* values of tendon obtained with the UTE technique in comparison to T2 values obtained with a conventional clinical CPMG technique. Published by Elsevier Ireland Ltd.

1. Introduction For clinical and research use, cadaveric tendons often undergo multiple freeze–thaw cycles. Clinically, this is most often seen with tendon allografts, such as those used for anterior cruciate ligament reconstruction. In the research arena, it is often difficult to obtain and process material immediately after death and in vitro tissue is frequently subject to at least one freeze–thaw cycle for convenience. The effects of successive freeze–thaw cycles on material properties of tendon have been studied, and most investigators have found negligible or no decline in biomechanical parameters with one [1,2] or two cycles [3]. This is not invariable, however, with some authors noting biomechanical change after a single freeze thaw cycle [4–6]. As the number of cycles approaches four [7] or five [8], the general consensus is that detrimental biomechanical changes will be seen.

∗ Corresponding author at: VA San Diego Healthcare System, 3350 La Jolla Village Drive, San Diego, CA 92161, United States. Tel.: +1 858 552 8585x7656; fax: +1 619 471 0503. E-mail addresses: [email protected] (E.Y. Chang), [email protected] (W.C. Bae), [email protected] (S. Statum), [email protected] (J. Du), [email protected] (C.B. Chung). 0720-048X/$ – see front matter. Published by Elsevier Ireland Ltd. http://dx.doi.org/10.1016/j.ejrad.2013.10.014

Multi-parametric quantitative magnetic resonance imaging (qMRI) has shown promise as a surrogate for the structural and material properties of biologic tissue [9], including tendon [10]. The effects of freeze–thawing on cartilage has been studied by multiple authors and it has been demonstrated that a single freeze–thaw cycle increases mono-exponential T2 values [11,12] with a corresponding increase with the total number of cycles [12]. Studies on other tissues have been performed, such as with fish muscle, and it has been shown that T2 is less sensitive to freeze–thaw changes than T1 and magnetization transfer rate [13]. Discrepancies between qMRI values obtained on tendon in vivo and in vitro have been noted and some have suggested these changes are, in part, due to freezing and thawing [10,14–16]. However, we know of no studies that systematically evaluate the effects of successive freeze–thawing on qMRI values of human tendon. In this study we sought to evaluate the effects of multiple freezing and thawing cycles on two MR parameters used to study Achilles tendon, T2 and T2*. 2. Methods 2.1. Sample preparation This anonymized cadaveric study was exempted by the Institutional review board. Four fresh human ankles from three donors (3

350

E.Y. Chang et al. / European Journal of Radiology 83 (2014) 349–353

females, mean age 78 years old, range 69–82) were harvested below the knee within 6 h of death. The Achilles tendons were imaged within the intact ankles in three specimens and in one specimen, the tendon was imaged after it was carefully dissected free from surrounding tissues. In particular, the tendon was not mechanically damaged during removal.

to determine T2 and T2* relaxation times. T2 was derived through exponential fitting of the equation: S(TE) ∝ exp(−TE/T2) + constant. T2* was derived through exponential fitting of the equation: S(TE) ∝ exp(−TE/T2*) + constant. Constant refers to background noise, a separate fitting parameter. 2.4. Statistical analysis

2.2. MR imaging acquisition at baseline and after successive freeze–thaw cycles Specimens were placed parallel to the B0 field and imaging in the anatomically axial plane was performed on a clinical 3T MR scanner (Signa HDx, GE Healthcare Technologies, Milwaukee, WI) which had gradients capable of a slew rate of 150 T/m/s and amplitude of 40 mT/m on each axis. Hardware modifications included the addition of a custom transmit-receive switch to the receiver preamplifiers for rapid switching at the end of a radiofrequency excitation pulse to allow for ultrashort echo time (UTE) imaging with a nominal TE of 8 ␮s. A 2 in. receive-only surface coil was used for the whole ankles and a 1-in. diameter birdcage transmit-receive coil was used for the dissected tendon. The quantitative imaging protocol is shown in Table 1. T2 and T2* was measured for each tendon. Carr–Purcell–Meiboom–Gill (CPMG) acquisitions were acquired for T2 quantification, where spin echo (SE) signal from eight echoes (TE = 10, 20, 30, 40, 50, 60, 70 and 80 ms, TR = 2000 ms) was subject to a simple exponential signal decay model to calculate T2, similar to what has been previously used on Achilles tendon [16]. 2D UTE acquisitions were used for T2* quantification, where UTE signal acquired after a series of TEs (TE = 0.01, 0.1, 0.2, 0.4, 0.6, 0.8, 2, 4, 10, 15, 20, and 30 ms, TR = 100 ms) was subject to a simple exponential signal decay model to calculate T2*. For each of the sequences used in this study, a field of view (FOV) of 5 cm and a slice thickness of 3 mm were prescribed. Imaging was performed at the same location within the midtensile portion of the tendon in the fresh specimens and after 1, 2, 4, and 5 freeze–thaw cycles. Each cycle consisted of a 24–36 h freeze period at −70 ◦ C followed by a 24 h thaw period at room temperature. Each specimen was wrapped in moist gauze during freezing to prevent dehydration. 2.3. Image analysis T2 and T2* values were obtained using Levenberg–Marquardt fitting algorithm developed in-house. The analysis algorithm was written in MATLAB (The Mathworks Inc., Natick, MA, USA) and was executed offline on axial images obtained with the protocols described above. Regions of interest were manually drawn over the entire Achilles tendon and a mono-exponential curve was used

Statistical analyses were performed with Excel (version 2011, Microsoft Corporation, Redmond, Washington) and R software, version 2.10.1 (2009) (R Foundation for Statistical Computing, Vienna, Austria). First, data was summarized for each solution and each imaging parameter. To determine effects of freeze–thaw cycles on SE T2 and UTE T2* values, repeated measures ANOVA was used. Next, linear regression between T2 and T2* values of baseline and successive freeze–thaw cycles was performed to determine strength of correlation as well as intraclass correlation coefficients [17]. To determine the agreement between the baseline and successive freeze–thaw T2 and T2* values, an analysis similar to Bland–Altman analysis [18] was performed and the bias and limits of agreement were calculated. For all statistical analyses, p-values less than 0.05 were considered significant. 3. Results The Achilles tendon demonstrated very low signal intensity on all CPMG source images whereas source images acquired with the UTE technique showed abundant intratendinous signal on many images (Fig. 1). No tendon demonstrated tendinosis or tendon tearing on CPMG sources images. However, excellent curve fitting was able to be performed on all data. Mean SE T2 and UTE T2* values for all four specimens at baseline (fresh) and after five freeze–thaw cycles are shown in Table 2. Mean SE T2 value at baseline for all four specimens was 17.23 ± 6.49 ms and after five freeze–thaw cycles was 17.09 ± 1.93 ms. Mean UTE T2* value at baseline for all four specimens was 1.18 ± 0.45 ms and after five freeze–thaw cycles was 1.11 ± 0.37 ms. Repeated measures ANOVA did not find a statistically significant difference in mean SE T2 values (p = 0.59) or UTE T2* values (p = 0.14) (Fig. 2) after freeze–thaw cycles. Linear regression between SE T2 values at baseline and after successive freeze–thaw cycles demonstrated moderate agreement (r = 0.60, Fig. 3A) whereas UTE T2* values at baseline and after successive-freeze thaw cycles demonstrated a strong agreement (r = 0.92, (Fig. 3C). Comparison between SE T2 values at baseline and after successive freeze–thaw cycles showed a small bias of 1.5 ms with a moderate limit of agreement (LoA) of ±4.6 ms (Fig. 3B) whereas UTE T2* values at baseline and after successive freeze–thaw cycles showed a negligible bias of 0.09 with a small LoA of ±0.18 ms (Fig. 3D). These findings suggest stronger

Fig. 1. UTE source images (A) show abundant signal in the Achilles tendon on many images. Mono-exponential decay curve fitting for determination of T2* (B) show accurate fitting. Echo times are in milliseconds (ms).

E.Y. Chang et al. / European Journal of Radiology 83 (2014) 349–353

351

Table 1 Imaging parameters for CPMG T2 and UTE T2* sequences. Sequence

TR [ms]

TE [ms]

FOV [cm]

Matrix

T2 UTE T2*

2000 100

10, 20, 30, 40, 50, 60, 70, 80 0.01, 0.1, 0.2, 0.4, 0.6, 0.8, 2, 4, 10, 15, 20, 30

5 5

320 × 256 256 × 256

Table 2 Results. Specimen 1 (whole) mean ± SD

Specimen 2 (whole) mean ± SD

Specimen 3 (whole) mean ± SD

Specimen 4 (dissected tendon) mean ± SD

Overall (4 specimens) mean ± SD

T2 (ME-SE) Fresh Post 1 cycle Post 2 cycles Post 4 cycles Post 5 cycles

21.04 18.69 22.91 20.98 18.92

± ± ± ± ±

3.72 2.01 5.68 4.95 3.51

10.09 11.52 14.10 14.36 14.38

± ± ± ± ±

1.95 1.55 2.38 1.37 2.36

13.64 19.12 14.21 18.57 17.76

± ± ± ± ±

1.50 3.35 0.65 1.50 1.42

24.15 11.02 16.71 22.10 17.31

± ± ± ± ±

6.64 5.83 1.79 3.45 2.05

17.23 15.09 16.98 19.00 17.09

± ± ± ± ±

6.49 4.42 4.13 3.43 1.93

T2* (UTE) Fresh Post 1 cycle Post 2 cycles Post 4 cycles Post 5 cycles

1.04 0.98 1.05 0.94 0.88

± ± ± ± ±

0.07 0.08 0.08 0.06 0.08

1.78 1.41 1.67 1.69 1.57

± ± ± ± ±

0.17 0.16 0.17 0.16 0.12

1.17 1.12 1.58 1.15 1.23

± ± ± ± ±

0.11 0.14 0.20 0.14 0.16

0.71 0.69 0.74 0.64 0.76

± ± ± ± ±

0.04 0.05 0.05 0.06 0.05

1.18 1.05 1.26 1.11 1.11

± ± ± ± ±

0.45 0.30 0.44 0.44 0.37

agreement and higher reproducibility of UTE T2* values after multiple freeze–thaw cycles in comparison to SE T2 values. 4. Discussion The purpose of this study was to evaluate the change of successive freezing and thawing cycles on qMRI parameters used to study tendon. In this pilot study, we have found that MR imaging of cadaveric tendon in the fresh state and again after five successive freeze–thaw cycles does not significantly alter T2 and T2* measurements. These findings are important because they suggest that frozen storage can be continued to be used in the study of the MR imaging characteristics of tendon ex vivo. Histologic studies have shown that freezing tissue results in changes at the cellular level, including dysfunction of cell metabolism, ice crystal formation and cell death [19,20]. In fact, authors have noted that a single freeze–thaw cycle completely

Fig. 2. Variations in (A) conventional spin echo T2 and (B) UTE T2* values in tendon samples with freeze/thaw cycles. Mean ± Standard Error of the Mean, n = 4.

kills the scant tendon cells that are present [21]. On electron microscopy, changes in the extra-cellular matrix have also been noted with an increase in the mean diameter of collagen fibrils and a decrease in the mean number of fibrils [6]. However, authors have found that there are no gross macroscopic changes even after eight freeze–thaw cycles [22]. Despite presumed microscopic changes, our findings are in keeping with the lack of macroscopic change in that T2 and T2* were not sensitive up to five cycles of freeze–thaw treatment. A possible explanation for this is that the total water content and orientation of collagen fibrils remains the same. Our results also show stronger agreement and higher reproducibility between baseline (fresh) and successive freeze–thaw T2* values of tendon obtained with the UTE technique in comparison to T2 values obtained with a conventional clinical CPMG technique. One possible explanation for this is that tendon demonstrates a majority of “short” T2/T2* components [15] and the signal from the majority of the tendon has decayed by the time the first echo is acquired on the clinical CPMG technique. Hence there are more uncertainties in plotting only the “long” T2 components as there is far less signal compared with UTE techniques. Quantitative T2 values obtained in our in vitro study are comparable to those recently reported by Juras et al. who utilized a similar multi-echo spin-echo technique, which were reported to range from 8.3 to 25.6 at 7 T [16]. Our mono-exponential T2* values are comparable to a prior in vivo study performed by Gold et al. which averaged 1.2 ± 0.2 ms [23,24]. In a prior study, Juras et al. also measured T2* of in vivo tendon at both 3 T and 7 T, utilizing biexponential fitting [15]. Linear combination of the short and long components which were reported in their paper results in monoexponential T2* values of 2.7–9.6 ms. Previously we have shown that histologically normal areas of in vitro Achilles tendon demonstrated a global mean T2* value of 2.18 ms (range 1.76–2.6 ms) [10,25]. The fact that the T2* values obtained in this current study are shorter than those by Juras et al. [15] and our previously reported results [10,25] are likely due to specimen variation. Our study has a number of limitations. First, the sample size is small, limiting the generalizability of these results. However, studies that involve repeated measurements are lengthy, and our study involved 20 successive scanning sessions. Regardless, future studies that include a larger sample of normal and abnormal tendons are necessary. Second, these results should not be generalized to other MR parameters which have been used to study tendon, including T1 and T1 rho. In fact, other authors have found T1 and magnetization

352

E.Y. Chang et al. / European Journal of Radiology 83 (2014) 349–353

Fig. 3. Linear regression of post-freeze/thaw vs. baseline (A) SE T2 and (C) UTE T2* values show a moderate agreement for SE T2 with r = 0.60 and strong agreement for UTE T2* with r = 0.92. Plots of post-pre difference vs. baseline (B) SE T2 and (D) UTE T2* values. (B) SE T2 values showed a small bias of 1.5 ms (9% of the mean) with a moderate limit of agreement (LoA) of ±4.6 ms (27% of the mean). (D) UTE T2* values showed a negligible bias of 0.09 ms (8% of the mean) with a small LoA of ±0.18 ms (15% of the mean).

transfer rate to be sensitive to the effects of frozen storage on fish muscle [13], but these parameters were not included in this study. Additionally, other physical properties of tendon such as elasticity and echotexture which could be measured on ultrasound would be interesting to evaluate in future studies. Third, we did not perform histological or biomechanical evaluation to correlate with the degree of cellular or material change. Although most prior studies have shown that detrimental biomechanical changes are seen as the number of freeze–thaw cycles approaches four or five [7,8], we cannot assume that the few specimens used in this study would have demonstrated this. Fourth, mono-exponential analyses were primarily utilized in this study although tendon demonstrates multi-exponential decay [15,26,27]. When we retrospectively reanalyzed our data using a bi-component fitting model [28], we were unable to reliably detect the longer component (consistently less than 5% for all samples and time points) likely due to uncorrected susceptibility causing rapid decay at longer echo times. This is to be expected since our T2* values ranged from 0.64–1.78 ms and the best fit curve was, in fact, mono-exponential. In comparison to the gradient echo UTE T2* acquisition, the longer components of tendon were able to be detected with the CPMG technique. In conclusion, in our pilot study, we have not found a significant difference in mono-exponential T2 or T2* values on Achilles tendons imaged fresh and up to five freeze–thaw cycles. These findings suggest that changes between specimens seen in vitro are due to factors other than frozen storage. Furthermore, our results suggest that there is stronger agreement and less variability between baseline (fresh) and successive freeze–thaw T2* values of tendon obtained with the UTE technique in comparison to T2 values obtained with a conventional clinical CPMG technique.

Conflict of interest None declared.

Acknowledgement The authors thank grant support from the Veterans Affairs Clinical Science Research and Development Service (Career Development Award 1IK2CX000749-01). References [1] Hirpara KM, Sullivan PJ, O’Sullivan ME. The effects of freezing on the tensile properties of repaired porcine flexor tendon. Journal of Hand Surgery 2008;33(3):353–8. [2] Matthews LS, Ellis D. Viscoelastic properties of cat tendon: effects of time after death and preservation by freezing. Journal of Biomechanics 1968;1(2):65–71. [3] Lee GH, Kumar A, Berkson E, Verma N, Bach Jr BR, Hallab N. A biomechanical analysis of bone-patellar tendon-bone grafts after repeat freeze–thaw cycles in a cyclic loading model. Journal of Knee Surgery 2009;22(2):111–3. [4] Leitschuh PH, Doherty TJ, Taylor DC, Brooks DE, Ryan JB. Effects of postmortem freezing on tensile failure properties of rabbit extensor digitorum longus muscle tendon complex. Journal of Orthopaedic Research: Official Publication of the Orthopaedic Research Society 1996;14(5):830–3. [5] Clavert P, Kempf JF, Bonnomet F, Boutemy P, Marcelin L, Kahn JL. Effects of freezing/thawing on the biomechanical properties of human tendons. Surgical and Radiologic Anatomy: SRA 2001;23(4):259–62. [6] Giannini S, Buda R, Di Caprio F, et al. Effects of freezing on the biomechanical and structural properties of human posterior tibial tendons. International Orthopaedics 2008;32(2):145–51. [7] Ren D, Sun K, Tian S, et al. Effects of gamma irradiation and repetitive freeze–thaw cycles on the biomechanical properties of human flexor digitorum superficialis tendons. Journal of Biomechanics 2012;45(2):252–6. [8] Huang H, Zhang J, Sun K, Zhang X, Tian S. Effects of repetitive multiple freeze–thaw cycles on the biomechanical properties of human flexor digitorum superficialis and flexor pollicis longus tendons. Clinical Biomechanics (Bristol, Avon) 2011;26(4):419–23. [9] Miyata S, Numano T, Homma K, Tateishi T, Ushida T. Feasibility of noninvasive evaluation of biophysical properties of tissue-engineered cartilage by using quantitative MRI. Journal of Biomechanics 2007;40(13):2990–8. [10] Filho GH, Du J, Pak BC, et al. Quantitative characterization of the Achilles tendon in cadaveric specimens: T1 and T2* measurements using ultrashort-TE MRI at 3 T. AJR American Journal of Roentgenology 2009;192(3):W117–24. [11] Reiter DA, Peacock A, Spencer RG. Effects of frozen storage and sample temperature on water compartmentation and multiexponential transverse relaxation in cartilage. Magnetic Resonance Imaging 2011;29(4):561–7. [12] Fishbein KW, Canuto HC, Bajaj P, Camacho NP, Spencer RG. Optimal methods for the preservation of cartilage samples in MRI and correlative biochemical

E.Y. Chang et al. / European Journal of Radiology 83 (2014) 349–353

[13]

[14]

[15]

[16]

[17] [18]

[19] [20]

studies. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine 2007;57(5):866–73. Nott KP, Evans SD, Hall LD. The effect of freeze–thawing on the magnetic resonance imaging parameters of cod and mackerel. Food Science and Technology-Leb 1999;32(5):261–8. Wright P, Jellus V, McGonagle D, Robson M, Ridgeway J, Hodgson R. Comparison of two ultrashort echo time sequences for the quantification of T1 within phantom and human Achilles tendon at 3 T. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine 2012;68(4):1279–84. Juras V, Zbyn S, Pressl C, et al. Regional variations of T(2)* in healthy and pathologic Achilles tendon in vivo at 7 T: preliminary results. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine 2012;68(5):1607–13. Juras V, Apprich S, Pressl C, et al. Histological correlation of 7 T multi-parametric MRI performed in ex-vivo Achilles tendon. European Journal of Radiology 2013;82(5):740–4. Fisher RA. On the probable error of a coefficient of correlation deduced from a small sample. Metron 1921;1:3–32. Altman DG, Bland JM. Measurement in medicine: the analysis of method comparison studies. Journal of the Royal Statistical Society Series D (The Statistician) 1983;32(3):307–17. Rubinsky B, Lee CY, Bastacky J, Onik G. The process of freezing and the mechanism of damage during hepatic cryosurgery. Cryobiology 1990;27(1):85–97. Ashwood-Smith MJ, Farrant J. Low temperature preservation in medicine and biology. Tunbridge Wells Eng: Pitman Medical; 1980.

353

[21] Suto K, Urabe K, Naruse K, et al. Repeated freeze–thaw cycles reduce the survival rate of osteocytes in bone-tendon constructs without affecting the mechanical properties of tendons. Cell and Tissue Banking 2012;13(1): 71–80. [22] Jung HJ, Vangipuram G, Fisher MB, et al. The effects of multiple freeze–thaw cycles on the biomechanical properties of the human bone-patellar tendonbone allograft. Journal of Orthopaedic Research: Official Publication of the Orthopaedic Research Society 2011;29(8):1193–8. [23] Gold GE, Wren TAL, Nayak K, Nishimura DG, Beaupre G. In vivo short echo time imaging of Achilles tendon. International Society for Magnetic Resonance in Medicine 2001:244. [24] Gold GE, Pauly JM, Macovski A, Herfkens RJ. MR spectroscopic imaging of collagen: tendons and knee menisci. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine 1995;34(5):647–54. [25] Du J, Pak BC, Znamirowski R, et al. Magic angle effect in magnetic resonance imaging of the Achilles tendon and enthesis. Magnetic Resonance Imaging 2009;27(4):557–64. [26] Peto S, Gillis P. Fiber-to-field angle dependence of proton nuclear magnetic relaxation in collagen. Magnetic Resonance Imaging 1990;8(6):705–12. [27] Wang N, Xia Y. Anisotropic analysis of multi-component T2 and T1␳ relaxations in Achilles tendon by NMR spectroscopy and microscopic MRI. Journal of Magnetic Resonance Imaging: JMRI 2013;38(September (3)):625–33. [28] Du J, Diaz E, Carl M, Bae W, Chung CB, Bydder GM. Ultrashort echo time imaging with bicomponent analysis. Magnetic Resonance in Medicine: Official Journal of the Society of Magnetic Resonance in Medicine/Society of Magnetic Resonance in Medicine 2012;67(3):645–9.