T1ρ and T2 Relaxation Times are Sensitive to Ischemic Injury in Femoral Head Specimens from a Piglet Model of Avascular Necrosis Independent of a Freeze/Thaw Cycle
Casey P. Johnson1,2, Ferenc Toth3, Cathy S. Carlson4, Stefan Zbyn1,2, Kai D. Ludwig1,2, Harry K. W. Kim5,6, and Jutta M. Ellermann1,2

1Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States, 2Department of Radiology, University of Minnesota, Minneapolis, MN, United States, 3Department of Veterinary Population Medicine, University of Minnesota, St. Paul, MN, United States, 4Department of Veterinary Clinical Sciences, University of Minnesota, St. Paul, MN, United States, 5Texas Scottish Rite Hospital, Dallas, TX, United States, 6Department of Orthopaedic Surgery, UT Southwestern Medical Center, Dallas, TX, United States


We investigated whether T1, T2, and T1ρ mapping can detect early ischemic injury to bone, marrow, and cartilage in an animal model of femoral head avascular necrosis. We imaged and compared six pairs of freshly-harvested ischemic and control femoral head specimens. We then imaged the specimens a second time after a freeze/thaw cycle. We found that T1ρ and T2 mapping were sensitive to ischemic injury to the femoral heads 48 hours after onset of ischemia. Furthermore, this sensitivity was maintained after the freeze/thaw cycle. This work indicates that T1ρ and T2 mapping may help assess ischemic bone and joint disorders.


Avascular necrosis (AVN) of the femoral head is a serious condition affecting children and young adults that can lead to severe femoral head deformity and osteoarthritis [1,2]. It is important to detect and treat AVN early to prevent deformity [1,2]. Contrast-enhanced MRI is the leading approach to detect femoral head ischemia [3], but its clinical use is limited due to safety concerns and its inability to directly detect and quantify ischemic injury to tissue. A recent ex vivo MRI study comparing quantitative mapping of T1, T2, and T1ρ relaxation times in a piglet model of AVN demonstrated that T1ρ mapping may be a particularly sensitive technique to detect ischemic injury to the femoral head as early as 48 hours after onset of ischemia [4]. A limitation of this prior study was that specimens underwent a freeze/thaw cycle before imaging. The purpose of our current study was to further investigate the sensitivity of T1, T2, and T1ρ mapping to early ischemic injury to the femoral head in the piglet model by: (i) imaging freshly-harvested femoral head specimens 48 hours after onset of ischemia; and (ii) measuring the effect of a freeze/thaw cycle on the findings. We hypothesized that: (i) T1ρ mapping would be most sensitive in detecting ischemic injury to the freshly-harvested femoral heads; and (ii) the sensitivity of the relaxation times would not be affected by a freeze/thaw cycle.


Animals. Whole right femoral head ischemia was surgically induced in N=6 six-week-old male piglets by placing a tight ligature around the femoral neck and transecting the ligamentum teres [5]. The piglets were euthanized 48 hours after surgery and the operated (right) and control (left) femoral heads were harvested and refrigerated for <12 hours at 4°C prior to imaging.

Imaging. The freshly-harvested specimens were individually imaged using a preclinical 9.4T MRI scanner. The specimens were mounted to a holder and immersed in Fomblin. T1, T2, and T1ρ maps were acquired using a magnetization-prepared 2D fast spin echo (FSE) sequence (Table 1). Immediately after imaging, the specimens were wrapped in saline-soaked gauze and frozen at -20°C for 3-41 days. The specimens were then thawed at 4°C and imaged a second time using the same protocol and slice positioning.

Histology. After imaging, the thawed specimens were processed for histological analysis with H&E staining in the imaging plane. The slides were qualitatively assessed by a blinded pathologist.

Data Analysis. Median T1, T2, and T1ρ relaxation times were calculated in four regions of interest (ROIs) (see labels in Figure 1): (i) secondary ossification center (SOC) (i.e., femoral head bone and marrow); (ii) epiphyseal cartilage (EC); (iii) articular cartilage (AC); and (iv) metaphyseal bone and marrow (Met). We tested whether the ROI measures differed between the pairs of: (i) operated and control femoral heads (both before and after freezing); and (ii) fresh and thawed specimens. Paired t-tests were used with p<0.0014 considered significant after Bonferroni correction for 36 comparisons (4 ROIs × 3 relaxation times × 3 comparisons).


Histological evaluation identified extensive necrosis of the bone marrow and deep epiphyseal cartilage in 4/6 operated femoral heads, while 1/6 had subtle necrosis and 1/6 had no necrosis due to surgical failure. All femoral heads were included in the data analyses. Relaxation time maps from a piglet with extensive necrosis are shown in Figure 1, and corresponding photomicrographs are shown in Figure 2. Results of the ROI analyses for all six piglets are tabulated in Table 2. The freeze/thaw cycle resulted in a significant decrease in T1ρ relaxation times in the SOC, EC, and AC. T2 relaxation times were also significantly decreased in the EC and AC. Despite these changes in absolute relaxation times, the percent increase in relaxation times between the operated and control femoral heads were remarkably similar before and after the freeze/thaw cycle (Figure 3). T1ρ had the greatest percent increase in the SOC due to ischemic injury (~40%), which was maintained after freezing.


Our findings support that: (i) T1ρ and T2 mapping can detect early ischemic injury to the femoral head; and (ii) their sensitivities are maintained after a freeze/thaw cycle. This work motivates investigation of T1ρ and T2 mapping as non-contrast-enhanced methods for early detection of AVN in the femoral head and other regions. Furthermore, the techniques can be effectively used to detect ischemic injury after a freeze/thaw cycle, which corroborates the initial report [4] and will facilitate future investigation of T1ρ and T2 mapping sensitivities to AVN, their contrast mechanisms, and the pathophysiology of and potential treatments for AVN.


This study was supported by the NIH (K01AR070894, K01OD021293, and P41EB015894), the W. M. Keck Foundation, and Texas Scottish Rite Hospital for Children.


1. Moya-Angeler J, Gianakos AL, Villa JC, Ni A, Lane JM. Current concepts on osteonecrosis of the femoral head. World J Orthop. 2015; 6(8):590-601.

2. Larson E, Jones LC, Goodman SB, Koo KH, Cui Q. Early-stage osteonecrosis of the femoral head: where are we and where are we going in year 2018? Int Orthop. 2018; 42(7):1723-1728.

3. Jaramillo D, Villegas-Medina OL, Doty DK, Dwek JR, Ransil BJ, Mulkern RV, Shapiro F. Gadolinium-enhanced MR imaging demonstrates abduction-caused hip ischemia and its reversal in piglets. AJR Am J Roentgenol 1996; 166(4):879-87.

4. Johnson CP, Wang L, Tóth F, Aruwajoye O, Carlson CS, Kim HK, Ellermann JM. Quantitative MRI helps to detect hip ischemia: preclinical model of Legg-Calvé-Perthes disease. Radiology 2018; 289(2):386-395.

5. Kim HK, Su PH. Development of flattening and apparent fragmentation following ischemic necrosis of the capital femoral epiphysis in a piglet model. J Bone Joint Surg Am 2002; 84-A(8):1329-34.


Table 1. Magnetization-prepared 2D FSE imaging parameters for quantitative T1, T2, and T1ρ mapping.

Table 2. ROI analysis of T1, T2, and T1ρ relaxation times in the operated vs. control and fresh vs. thawed femoral heads. Group ROI measures are given as mean ± standard deviation. Statistically significant (p<0.0014) paired differences are indicated with an asterisk (*). While the freeze/thaw cycle resulted in significant changes in absolute relaxation times in the ROIs, it had little impact on the relaxation time differences between the operated and control specimens.

Figure 1. T1, T2, and T1ρ relaxation time maps for a representative pair of operated and control femoral heads imaged before (fresh) and after (thawed) a freeze/thaw cycle. In both the fresh and thawed specimens, T1ρ relaxation times were prominently increased in the SOC of the operated vs. control femoral head. Although absolute T1ρ relaxation times were, in general, decreased after the freeze/thaw cycle, the sensitivity to ischemic injury was maintained.

Figure 2. Photomicrographs of H&E-stained slides of the pair of femoral head specimens shown in Figure 1. (a) In the SOC, the operated femoral head had evidence of marrow necrosis (arrows). (b) There was also evidence of necrosis of chondrocytes in the deep layer of the epiphyseal cartilage in the operated femoral head (arrowheads).

Figure 3. Average percent increase in relaxation times in the operated versus control femoral heads. The percent increase values were similar between the fresh (solid bars) and thawed (hashed bars) specimens across all relaxation times and ROIs. T1ρ had the greatest percent increase in the SOC (~40%), followed by T2 (~20%). T2 and T1ρ had a similar percent increase in the EC (~12%) and AC (~8%). T1 had relatively little change in the SOC, EC, or AC. The metaphysis is not affected by the surgery, thus little change in relaxation times was observed between the operated and control femoral heads, as expected.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)