T2 Vertebral Bone Marrow Changes After Space Flight

A. LeBlanc,1* C. Lin,1 H. Evans,1 L. Shackelford,2 C. Martin,1 and T. Hedrick3

Bone biopsies indicate that during immobilization bone marrowadipose tissue increases while the functional cellular fractiondecreases. One objective of our Spacelab flight experiment wasto determine, using in vivo volume-localized magnetic reso-nance spectroscopy (VLMRS), whether bone marrow composi-tion was altered by space flight. Four crew members of a 17 daySpacelab mission participated in the experiment. The apparentcellular fraction and transverse relaxation time (T 2) were deter-mined twice before launch and at several times after flight.Immediately after flight, no significant change in the cellularfraction was found. However, the T 2 of the cellular, but not the fatcomponent increased following flight, although to a variableextent, in all crew members with a time course for return tobaseline lasting several months. The T 2 of seven control sub-jects showed no significant change. Although these observa-tions may have several explanations, it is speculated that theobserved T 2 changes might reflect increased marrow osteoblas-tic activity during recovery from space flight. Magn Reson Med41:495–498, 1999. r 1999 Wiley-Liss, Inc.

Key words: magnetic resonance imaging; volume-localized mag-netic resonance spectroscopy; immobilization; microgravity

Iliac crest biopsies in patients with cervical cord lesionshave documented increased marrow fat and decreasedfunctional cell volume (1). Bone marrow samples fromtransiliac bone biopsies in the same patient at 3 and 15weeks after traumatic spinal cord injury showed thatmarrow adipose tissue volume increased 54%, from aninitial value of 33% to 52% of marrow volume (2). Adecrease in the functional cell population from about 53%to 40% showed that the increase in marrow fat was notsimply due to replacement of lost bone, but represents achange in the fat vs functional cell fraction. In these samepatients, trabecular bone loss was on the order of 30% after25 weeks of immobilization.

Rats flown aboard Cosmos biosatellites (690, 782, 936)have shown increased bone marrow triglycerides, imply-ing increased fatty marrow during flight (3). Centrifugationof a group of rats on Cosmos 936 prevented this increase.Growing rats flown on the 18.5 day Cosmos flight 1129showed increases in marrow fat in the tibia and humerus(4). The fractional marrow area occupied by fat in the flightanimals was about 18% compared with 5% in the non-flight controls. Although not measured, these increases inpercentage fat imply a corresponding loss in functional cellvolume. Even larger changes in marrow fat were observedin rear limb-suspended rats, a ground-based model used tosimulate the effects of microgravity on bone (5). As in

human immobilization, these marrow changes were accom-panied by decreases in trabecular bone volume and forma-tion. We were interested in determining whther similarchanges occur in astronauts exposed to microgravity. Sev-eral MRI experiments were performed on four crew mem-bers of a 17 day shuttle flight whose primary purpose wasto explore life and microgravity science. One objective ofour flight experiment was to determine, using volume-localized proton spectroscopy (VLMRS), whether micro-gravity would cause changes in the relative proportions offat and functional cells of the spinal bone marrow.

MATERIALS AND METHODS

According to the experimental plan, MRI imaging of fourmale crew members (age 6 SD, 43 6 4 years) of the 17 daySpacelab mission was performed approximately 30 and 50days before launch and at 2, 10, and 30 days after landing.Based on the initial postflight results, additional postflightVLMRS sessions were obtained at various times dependingon crew availability. In addition, multiple measurementswere made in seven normal controls (six male and onefemale; age 6 SD, 43 6 11 years) over a 2 year periodspanning the time of the flight experiment.

The MRI unit was a 1.5 T Siemens machine located atThe Methodist Hospital in Houston, TX. VLMRS wereobtained using a technique previously published (6). Briefly,a cubic volume, 15 3 15 x 15 mm, located in the center ofthe L3 vertebral body, was selected based on sagittal,transverse, and coronal scout images of the lumbar spine.A gradient inversion spectroscopy technique with a sur-face coil was used to acquire spectra at TEs of 12, 18, 24,and 30 msec, with a TR of 2 sec. Figure 1 gives a typical setof spectral data. The relative amount of lipid signal, whichincludes both methyl and methylene resonances, andwater signal at each TE was calculated by integrating theareas under the lipid and water peaks after baselinecorrection. Although the signal from olefinic proton oflipids cannot be separated from that of water, its contribu-tion to the water peak is small. The integrated lipid andwater values were plotted against TE to obtain lipid andwater signal intensity at time zero and to calculate T2. Thesignal intensity of lipid and water at time zero representsthe amount of cellular and fat components in the bonemarrow. The apparent cellular fraction was calculatedfrom the ratio of the water signal to the sum of the lipid andwater signal intensities at time zero and multiplied by 100,to be expressed as a percent.

The data (T2 of cellular component, T2 fat component,and apparent cellular fraction) were analyzed for twogroups of subjects (four astronauts and seven normalcontrols). The data available for the astronauts includedtwo preflight measurements and six to seven postflightmeasurements. In the normal controls, 4–11 observationswere noted over an 85 week period. The data were ana-

1Department of Medicine, Baylor College of Medicine.2Space and Life Sciences, Johnson Space Center.3Department of Radiology, Baylor College of Medicine, Houston, Texas.Grant sponsor: NASA; Grant numbers: NAS 9–18952 and NAGW 4437.*Correspondence to: Adrian LeBlanc, 6501 Fannin Mail Code NB1–004,Houston TX, 77030. E-mail: aleblanc@bcm.tmc.eduReceived 25 May 1998; revised 16 September 1998; accepted 18 September1998.

Magnetic Resonance in Medicine 41:495–498 (1999)

495r 1999 Wiley-Liss, Inc.

lyzed using non-linear regression models, e.g., a Far-azdaghi and Harris growth curve model as well as standardlinear regression. Significance was set at P, 0.05.

These studies were approved by the Johnson SpaceCenter and Baylor College of Medicine Institutional Re-view Boards for Human Research.

RESULTS

Figures 2 and 3 show the apparent cellular fraction (%) inthe control and flight groups as a function of time. Immedi-ately after flight no significant change compared withbaseline was found. Although the postflight measurementsmay indicate some change in the apparent cellular fraction,standard linear regression indicated, in general, no statisti-cally significant deviation from a non-time-related flatresponse for either the flight or control groups. Figures 4and 5 show the T2 of the functional cellular fraction in thecontrols and flight crew. There appeared to be a smallnonsignificant decrease in the T2 immediately after flight.These data appeared to show a general pattern afterflight—an increase followed by a return toward baseline.The increase in T2 was apparent in all crew membersseveral weeks after landing. In three of the four crewmembers, the T2 remained elevated above baseline formore than 4 months after landing. The exponential growth

curve model fit to the data indicated a statistically signifi-cant change over time for the astronauts, but not for thecontrols. There was no change in the T2 of the fat of theflight crew or controls.

DISCUSSION

T2 is a measure of signal decay that results from the loss ofphase coherence of the hydrogen nuclei. Differences in T2

between tissues, e.g., liver, marrow, etc., provide hightissue contrast for radiologic imaging. Although the patho-physiologic meaning of T2 is still not fully understood, thein vivo T2 of most tissues is dependent predominantly onthe type and concentration of the ionic and macromolecu-lar (predominantly proteins) composition of the intracellu-lar fluid. In vivo measurements with VLMRS have beenshown to measure the functional cellular and fat fractionsaccurately when compared with biopsy specimens fromthe same anatomical region (7). A modified version ofVLMRS, which corrects for eddy currents, was used for thepresent studies (6).

The present investigation measured two aspects of thebone marrow: apparent cellular and fat fractions and the T2

of these components. The values for the apparent cellularfraction that we obtained in this study demonstrated awide inter-individual range, from 45% to 75%. However,they were quite stable over time. The reason for this widerange in values is not clear, but it is similar to the values

FIG. 1. Typical set of spectral data.

FIG. 2. Functional cellular fraction (%) in L3 bone marrow in fourastronauts before and after flight.

FIG. 3. Functional cellular fraction (%) in L3 bone marrow in sevencontrol subjects as a function of time.

FIG. 4. T2 of functional cellular fraction in L3 in four astronauts beforeand after flight.

496 LeBlanc et al.

and range of values measured from human biopsy speci-mens (8,9). The size of the functional cellular compartmenthas been correlated with aging, osteoporosis, and boneremodeling activity (1,8–13). While we did not find asignificant change in the functional cellular fraction, wedid observe T2 changes. The lack of change in the size ofthe compartment may not be surprising considering thatthe flight only lasted 17 days. The implication is that whilethe size of the functional cellular compartment did notchange during flight, this time is sufficient to result in T2

changes within the marrow after flight, presumably reflect-ing changes that occurred or were initiated during flight. Itis known, for example, that loss of bone and red cell massbegins within the first few weeks of space flight or bed rest(14–16).

An explanation for the observed T2 changes might beincreased osteoblastic activity following flight, reflecting acoherent increase in bone formation. During microgravity,rat studies have documented reduced osteoblastic activity(4,5). Bone marrow removed from rat bones previouslyunloaded for 11 days showed reduced number of adherentmarrow stromal cells and reduced osteogenic potential insubsequent in vitro cultures as well as decreased bonemass (17). Similarly, the expression of mRNA level forgrowth factors and peptide production and the prolifera-tion and osteogenic differentiation of marrow stromal cellsare reduced in rats during tail suspension (18). The differ-entiation of osteoblasts in culture, in response to systemichormones, is reduced during microgravity (19). It wouldseem reasonable, therefore, that if unloading decreasesosteoblast number, loading after return to 1 G mightstimulate their formation. In support of this, our 17 weekbed rest (microgravity simulation) studies demonstrated asignificant increase in bone formation markers after ream-bulation, i.e., alkaline phosphatase was increased by 50%and osteocalcin by 33% (20). Although the data (n5 3)were not sufficient for statistical analysis, formation mark-ers appear elevated during recovery from 4 months ofspace flight (21). Limited follow-up data in five cosmo-nauts who spent 6 months in space suggest that recovery oflost bone mass occurs after flight in most but not all crewmembers (22). The long time course of the observed T2 timeresponse is consistent with the remodeling process—about4–6 months.

Another possibility is that the observed T2 change in thecellular component represents an increase in the cellularproliferation in response to gravity to replace lost red cellsfollowing flight (increased hematopoiesis). The loss of redcell mass during weightlessness has been documented(16). However, the time frame of the T2 response is muchlonger than needed for replacement of lost red cells, whichshould be complete within 1 month after flight and istherefore not likely to be the sole explanation for ourobservations (23). While it is tantalizing to speculate thatthe T2 changes reflect changes in marrow composition ofthe cellular component, there are other possibilities. Forexample, a change in the amount of unbound water, forinstance, edema or venous pooling, could change T2.However, edema or venous pooling in the bone marrow forseveral months after flight would be difficult to understandand seems unlikely. A change in marrow blood flow mightalso cause a shift in T2, although it is difficult to know howand to what degree this would affect T2.

Bone loss is a known consequence of long-durationspace flight (24,25) and preventing this loss is an importantNASA goal. If these T2 changes represent a response of thetrabecular skeleton to mechanical loading, monitoring T2

changes in the marrow might be useful in assessing theresponse to techniques designed to prevent bone loss fromspace flight. These findings may also have significance formedical research on earth as well as in microgravity.Assuming that bone marrow is an essential element in theprocess that leads to changes in bone mass, these findingscould provide a better understanding of the basic physiol-ogy of the remodeling process, e.g., the early phase of boneformation.

CONCLUSIONS

We found no significant change in the apparent cellularfraction of the bone marrow of L3. However, the T2 of thecellular, but not the fat, component increased followingflight in all crew members. These increases in T2 returnedto baseline with a time course that is long compared withthe time in microgravity; in three of the four crew mem-bers, the T2 remained elevated above baseline for morethan 4 months after landing. Although the interpretation ofthese findings needs further investigation, these resultssuggest that quantitative VLMRS measurements of thebone marrow may provide in vivo information on boneremodeling.

ACKNOWLEDGMENTS

The cooperation and interest of the crew in obtaining themeasurements for this experiment is appreciated. In particu-lar, our research group thanks the crew for their willing-ness to undergo the additional MRI scans to elucidate theevolution of the bone marrow T2 phenomena—we areaware of their busy postflight schedule and the inconve-nience involved in fighting Houston traffic to reach theMedical Center.

REFERENCES1. Minaire P, Meunier P, Edouard C, Bernard J, Courpron P, Bourret J.

Quantitative histological data on disuse osteoporosis. Calcif Tissue Res1974;17:57–73.

FIG. 5. T2 of functional cellular fraction in L3 in seven controlsubjects as a function of time.

T2 Vertebral Bone Marrow Changes 497

2. Minaire P, Edouard C, Arlot M, Meunier PJ. Marrow changes inparaplegic patients. Calcif Tissue Int 1984;36:338–340.

3. Ahlers I, Tigranyan RA, Ahlersova E, Paulikova E, Datelinka I, SmajdaB, Toropila M, Praslicka M. Space flight effects upon plasma and tissuelipids in rats (abstract). Physiologist 1980;23:S131–S132.

4. Jee WSS, Wronski TJ, Morey ER, Kimmel DB. Effects of spaceflight ontrabecular bone in rats. Am J Physiol 1983;244:R310–314.

5. Morey-Holton E, Wronski TJ. Animal models for simulating weightless-ness. Physiologist 1981;24:S45–S48.

6. Lin C, Wendt RE, Evans HJ, Rowe RM, Hedrick TD, LeBlanc AD. Eddycurrent correction in volume-localized MR spectroscopy. J Magn ResonImaging 1994;4:823–827.

7. Ballon D, Jakubowski A, Gabrilove J, Graham MC, Zakowski M,Sheridan C, Koutcher JA. In vivo measurements of bone marrowcellularity using volume-localized proton NMR spectroscopy. MagnReson Med 1991;19:85–95.

8. Hartsock RJ, Smith EB, Petty CS. Normal variations with aging of theamount of hematopoietic tissue in bone marrow from the anterior iliaccrest. Am J Clin Pathol 1965;43:326–331.

9. Meunier P, Aaron J, Edouard C, Vignon G. Osteoporosis and thereplacement of cell populations of the marrow by adipose tissue. ClinOrthop 1971;80:147–154.

10. Han ZH, Palnitkar S, Dillon P, Parfitt AM, Effect of bone marrowcomposition on bone remodeling rate in beagles (abstract). In: Proceed-ings of the ASBMR 18th Annual Meeting, Washington, DC, 1996. Vol.11. p S148.

11. Lips P, Van Ginkel FC, Netelenbos JC. Bone marrow and bone remodel-ing. Bone 1985;6:343–344.

12. Martin RB, Chow BD, Lucas PA. Bone marrow fat content in relation tobone remodeling and serum chemistry in intact and ovariectomizeddogs. Calcif Tissue Int 1990;46:189–194.

13. Wronski TJ, Smith JM, Jee WSS. Variations in mineral apposition rate oftrabecular bone within the beagle skeleton. Calcif Tissue Int 1981;33:583–586.

14. Rambaut PC, Johnston RS. Prolonged weightlessness and calcium lossin man. Acta Astronaut 1979;6:1113–1122.

15. Schneider VS, McDonald J. Skeletal calcium homeostasis and counter-measures to prevent disuse osteoporosis. Calcif Tissue Int 1984;36:S151–S154.

16. Udden MM, Driscoll TB, Pickett MH, Leach-Huntoon CS, Alfrey CP.Decreased production of red blood cells in human subjects exposed tomicrogravity. J Lab Clin Med 1995;125:442–449.

17. Keila S, Pitaru S, Grosskopf A, Weinreb M. Bone marrow frommechanically unloaded rat bones expresses reduced osteogenic capac-ity in vitro. J Bone Miner Res 1994;9:321–327.

18. Zhang R, Supowit SC, Klein GL, Lu Z, Christensen MD, Lozano R,Simmons DJ. Rat tail suspension reduces messenger RNA level forgrowth factors and osteopontin and decrease the osteoblastic differentia-tion of bone marrow stromal cells. J Bone Miner Res 1995;10:415–423.

19. Carmeliet G, Nys G, Bouillon R. Microgravity reduces the differentia-tion of human osteoblastic MG-63 cells. J Bone Miner Res 1997;12:797–793.

20. LeBlanc A, Schneider V, Spector E, Evans H, Rowe R, Lane H. Calciumabsorption, endogenous secretion and endocrine changes during andafter long-term bed rest. Bone 1995;16:301S–304S.

21. Smith SM, Oganov VS, Morukov BV, Larina IM, Nillen JL, Lane HW.Calcium homeostasis during extended-duration space flight (abstract).Federated Am Soc Exp Biol 1996;10:2702.

22. LeBlanc A, Shackelford L, Schneider V. Future human bone research inspace. Bone 1998;22:113S–116S.

23. Legen’kov VI, Kiselev RK, Gudim VI, Moskaleva GP. Changes inperipheral blood of crew members of the Salyut-4 orbital station.Kosmicheskaya Biol I Aviakosmicheskaya Med 1977;6:1–12.

24. LeBlanc A, Schneider V, Shackelford L, West S, Oganov V, Bakulin A,Veronin L. Bone mineral and lean tissue loss after long durationspaceflight (abstract). In: Proceedings of the ASBMR 18th AnnualMeeting, Washington, DC, 1996. Vol. 11. p S323.

25. Oganov VS, Grigoriev AI, Voronin LI, Rakmanov AS, Bakulin AV,Schneider V, LeBlanc A. Bone mineral density in cosmonauts after4.5–6 month long flights aboard orbital station MIR. Aerosp EnvironMed 1992;26:20–24.

498 LeBlanc et al.