Skeletal health in long-duration astronauts: Nature, assessment, and management recommendations from the NASA bone summit


Address correspondence to: Jean D Sibonga, PhD, Bone & Mineral Laboratory SK/272, NASA Johnson Space Center, 2101 NASA Parkway, Houston, TX 77058. E-mail:


Concern about the risk of bone loss in astronauts as a result of prolonged exposure to microgravity prompted the National Aeronautics and Space Administration to convene a Bone Summit with a panel of experts at the Johnson Space Center to review the medical data and research evidence from astronauts who have had prolonged exposure to spaceflight. Data were reviewed from 35 astronauts who had served on spaceflight missions lasting between 120 and 180 days with attention focused on astronauts who (1) were repeat fliers on long-duration missions, (2) were users of an advanced resistive exercise device (ARED), (3) were scanned by quantitative computed tomography (QCT) at the hip, (4) had hip bone strength estimated by finite element modeling, or (5) had lost >10% of areal bone mineral density (aBMD) at the hip or lumbar spine as measured by dual-energy X-ray absorptiometry (DXA). Because of the limitations of DXA in describing the effects of spaceflight on bone strength, the panel recommended that the U.S. space program use QCT and finite element modeling to further study the unique effects of spaceflight (and recovery) on bone health in order to better inform clinical decisions.


Since the early days of manned spaceflight, the U.S. National Aeronautics and Space Administration (NASA) has been concerned about potential adverse effects of prolonged weightlessness, including bone atrophy due to the lack of forces (muscular and gravitational) on the skeleton[1-4] and general deconditioning under spaceflight conditions.[2, 5-8] Consequently, the major thrust of research has focused on better understanding the pathophysiology of increased urinary excretion of calcium[9, 10] and loss of areal bone mineral density (aBMD) at weight-bearing skeletal sites.[11] With the construction of the International Space Station (ISS), these concerns reached a higher level of importance, because astronauts were now capable of living and working in space for extended durations and potentially suffering adverse skeletal consequences. The risk of fractures is of particular concern upon re-exposure to mechanical loading, such as during the exploration of unknown planetary terrain or return to Earth's full gravity field. Moreover, cumulative skeletal deconditioning could increase the risks for premature osteoporosis and for fractures in later life.

Limited data exist for understanding the bone loss and fracture risk in astronauts. Typically, NASA astronauts are young to middle-aged (ranging from 25 to 55 years), and are predominantly male (male:female ratio ∼5:1). In contrast to patient populations with recognized risk factors for bone loss and fracture, astronauts are physically fit and healthy. Despite this, it is plausible that long-duration spaceflight (defined by NASA as longer than 30 days) would have measurable detrimental short-term and long-term impacts on bone health. Generally, long-duration astronauts have lived aboard a spacecraft, such as Mir or the current ISS, for 120 to 180 days and the skeletal effects of these missions are not easily modeled on Earth. Equivalent to a rare syndrome, space-induced bone loss is expected to affect only a small number of individuals, the total of which will probably not exceed 100 by the end of the ISS program in 2020. Table 1 outlines specific characteristics of the long-duration astronaut cohort as of 2010.

Table 1. Characteristics of the Long-Duration Astronaut Cohort at Time of Bone Summit (2010)
 Astronauts on long-duration missions
Males (n = 29)Females (n = 6)
  1. All values in table are means ± SD (with range in parentheses). In 2010, relative to the NASA Astronaut Corps (total 331 astronauts), the cohort of long-duration astronauts is 33 astronauts and predominantly male (ratio of males to females 29:6 or 4.8:1). Of this group, 4 astronauts have served on two separate long-duration spaceflights out of a total of 39 separate spaceflight missions. The traits of the corps have been consistent with time; over the 2 years since the Bone Summit, the total number of long-duration crewmembers has increased to 55, the number of repeat fliers has increased by 1 and the total number of flights has increased by 12. For computed averages, the astronaut data (eg, age, duration, body composition data) for the second flight was treated as if there was another separate astronaut.
Typical space mission duration (days)161 ± 36 (58–215)165 ± 39 (90–194)
Average age at time of flight (years)47 ± 5 (37–55)45 ± 4 (41–53)
Body mass index (kg2/m)25.8 ± 2.0 (21.2–30.7)23.4 ± 2.4 (20.4–25.9)
Body weight (kg)80 ± 6 (63–97)67 ± 8 (57–82)
Height (cm)176 ± 6 (163–185)170 ± 4 (165–178)
Total body lean mass (kg)61 ± 5 (45–69)47 ± 5 (39–54)
Total body fat mass (kg)16 ± 4 (6–27)19 ± 7 (13–33)
% Body fat20 ± 4 (9–27)27 ± 8 (19–41)

The complexities of spaceflight-induced bone loss in a small understudied population raise the question of whether current assessments of skeletal health in astronauts are sufficient. NASA is legally and ethically responsible for providing a safe and healthy work environment.[12] Bone strength must be considered in the selection of candidates for the corps of astronauts that will participate in future flights. In addition, skeletal deficits due to spaceflight that persist after return to Earth's gravity should be monitored because they might predispose astronauts to premature fragility fractures soon after return to Earth or later in life.

It is a current medical requirement to assess the skeletal integrity of all astronaut candidates by dual-energy X-ray absorptiometry (DXA), and to monitor bone health of all active and retired NASA astronauts likewise by triennial measurements of bone density. Moreover, DXA scans are required before and after missions in all ISS astronauts, ie, those with prolonged habitation of space (>30 days), to evaluate the skeletal effects of spaceflight and to monitor the restoration to preflight status. However, many other tests for evaluating the skeletal effects of spaceflight, including assays for gonadal hormones and quantitative computed tomography (QCT), are by astronaut consent only. Regardless of the technique used, it is difficult to investigate the effects of space on bone, or of the different interventions for bone loss in the flight environment, when the number of long-duration astronauts is so small.

To solicit clinical guidance, experts in osteoporosis, endocrinology, rheumatology, gerontology, and physical medicine and rehabilitation, with subspecialties in bone densitometry, bone epidemiology, male osteoporosis, and nutrition, were convened by NASA in 2010 as a Bone Summit Panel. The panel was to evaluate what NASA is currently doing to manage an occupational risk for fractures that may occur later in life. The panel was asked to recommend the skeletal measures that would be useful for the selection of astronauts and the surveillance of premature osteoporosis, the measured outcome that would serve as a trigger for possible medical intervention, and the measures that should be used to evaluate the efficacy of countermeasures being studied during spaceflight. As noted, evaluating trends in astronaut data is limited by the measures available, the delayed accumulation of data and the small sample size of astronauts. The Bone Summit Panel reviewed all available bone-relevant data accumulated from long-duration astronauts who served on the Mir spacecraft and/or the ISS. All data were from astronauts who are employed by NASA; no biomedical data from cosmonauts or astronauts employed by the international space agencies were released for chart review. Tools for risk surveillance and recommendations for future research developed at this conference are presented here.

Charge to Bone Summit Panel

The panel was asked to review data relevant to mineral metabolism and to bone structure, density, turnover, and strength that were collected from long-duration astronauts from 1994 to 2010. Over this period, data were acquired from 35 long-duration astronauts. All measures conducted immediately postflight were within 1 month of landing. These data included (1) the effects of resistive exercise during flight on aBMD (g/cm2) as measured immediately postflight by DXA, (2) levels of endocrine regulators and biochemical markers of bone remodeling, (3) changes immediately postflight in DXA measures, (4) changes immediately postflight in compartmental volumetric BMD (vBMD, g/cm3) of cortical and trabecular bone as assessed by QCT, and (5) serial changes in both DXA and QCT measurements after return to Earth.

During a closed session, medical records of individual long-duration astronauts were reviewed. The 25 cases were selected (with some duplication) as a result of the presence of a unique trait or an extreme skeletal response to space, as follows: (1) female (n = 6); (2) repeat fliers on long-duration mission (n = 4); (3) users of new (since 2009) advanced resistive exercise device (ARED) for the weightless environment (n = 5); (4) those scanned by QCT (n = 10); (5) those with hip bone strength estimated by finite element modeling (FEM)—a computational tool used to estimate failure loads of complex structures—(n = 7); and (6) those with a loss of aBMD >10% in either the hip or spine (n = 6). In addition, relevant published data studying long-duration astronauts were reviewed. Some preliminary data from research studies currently in progress had been available for review but are not included here. This report summarizes the data presented and the panel's conclusions and recommendations.

Long-Duration Astronaut Data: Bone Loss and Recovery

Densitometry and biochemistry


Since 1998, NASA medically required measurements of aBMD by DXA to assess skeletal integrity of astronauts. NASA's medical standards for bone health in astronauts include T-score cut-points (NASA-STD-3001 NASA Space Flight Human System Standard, Crew Health); a T-score of –1.0 or less at the hip or lumbar spine has disqualified an applicant from the astronaut corps, partially because of the expectation that an astronaut might develop at T-score of –2.0 or less after spaceflight, based on a calculated average monthly loss of 1.0% to 1.5% aBMD during long-duration spaceflight.[11] Among astronauts selected, this criterion also serves as a reason for disqualification from long-duration missions. When considering the use of any countermeasure to prevent bone loss, it is expected that the intervention will ensure that an astronaut returns from a mission with a T-score of −2.0 or better.

Figure 1 displays astronaut medical data that were available at the time of the Bone Summit, which includes the first 23 ISS expeditions. Changes in aBMD due to spaceflight have been quantified by serial DXA whole-body measures or regional scanning of the hip (total hip, femoral neck, and trochanter), lumbar spine, and forearm (Fig. 2). Notably, after spaceflight all long-duration astronauts showed a loss in aBMD exceeding the least significant change (LSC) in at least one of these skeletal regions,[4] and some astronauts had a >10% aBMD loss at both the hip and lumbar spine. (Note: Hologic QDR 4500 and QDR 2000 were used for measurement of astronaut BMD. For QDR 4500, LSC is 0.019 [trochanter], 0.035 [femoral neck], and 0.025 g/cm2 [lumbar spine], and the LSC for the Hologic QDR 2000 was 0.024 [trochanter], 0.050 [femoral neck], and 0.035 [lumbar spine]. All values g/cm2.) However, no long-duration astronaut returned from spaceflight with a hip or lumbar spine T-score less than or equal to −2.5. The panel was asked how NASA should interpret and utilize these data to assess fracture risk in astronauts.

Figure 1.

Areal BMD T-scores for skeletal regions of hip and spine in astronauts before and after ISS missions. T-scores were calculated according to procedures of the ISCD from preflight and postflight measurements of areal BMD by DXA. Thirty sets of preflight to postflight measurements were from 27 different U.S. astronauts (23 ISS spaceflight expeditions, some with multiple or previously flown U.S. crewmembers). Changes in T-scores are superimposed on the ISCD diagnostic guidelines for osteoporosis (T-scores ≤ –2.5) developed for perimenopausal and postmenopausal women, and men over the age of 50. T-scores between –1.0 and –2.5 represent the low bone mass range (osteopenia). These diagnostic criteria were the only evidence-based guidelines available for evaluating skeletal integrity when BMD measurements by DXA became the medically-required test for long-duration astronaut crew health (internal document, NASA Med Vol. B). Clinical decisions and medical standards for human health and performance at NASA are based upon these diagnostic guidelines. ISCD = International Society for Clinical Densitometry.

Figure 2.

Distribution of DXA aBMD percentage changes after long-duration Mir or ISS spaceflight. Percentage change of preflight aBMD per month was calculated by subtracting the first postflight DXA BMD measurement from the preflight measurement and normalizing by the mission duration (typically 4–6 months). Data are plotted for groups of crewmembers who served on Mir (n = 28 cosmonauts, and n = 7 U.S. astronauts), on ISS with access to only the iRED for weight-bearing exercises (n = 24 U.S. astronauts), and on ISS after the ARED became available (n = 8 crewmembers). BMD changes are reported for LSpine, Fem.Neck, trochanter, total hip, and wrist. (Wrist is 1/3 radius+ ulna, and is for ISS crewmembers only.) All mean aBMD changes from preflight to postflight, for all groups, were significant (p < 0.05), as determined by Student's t test (one-tailed, paired), except for the effect of ARED on lumbar spine and wrist (p < 0.06 for trochanter). ARED users include crewmembers who were concurrently taking a bisphosphonate (n = 3), crewmembers with access to ARED for <4 months (n = 3), in addition to crewmembers with ARED access for the entire spaceflight (∼6 months) and not taking bisphosphonates (n = 2). Astronaut measurements had to be combined to ensure nonidentifiable data. LSpine = lumbar spine; Fem.Neck = femoral neck.

According to NASA medical health standards,[13] the preflight aBMD requirement for flight certification is based on an estimated total spaceflight-related decline in aBMD originally reported in 18 cosmonauts.[11] An average monthly rate of aBMD loss was calculated from data from Mir missions (conducted from 1995 to 1998) ranging in duration from 4 to 14 months[11] and was used to predict likely aBMD loss for other mission designs. Although there were insufficient measurements to assess whether the loss during flight was linear, there was an overall average 1.0% to 1.5% aBMD decline per month for hip and lumbar spine. This finding highlighted the accelerated rate of aBMD loss at weight-bearing skeletal sites during spaceflight, contrasting starkly with the typical age-related rate of bone loss of 0.5% to 1.0% per year for comparable sites in older individuals on Earth. A similar rate of flight-related bone loss was found in U.S. crewmembers on ISS expeditions flown from 2000 to 2009. Since 2009, the availability of the ARED on the ISS (Fig. 3A) may have attenuated the aBMD decline in the astronauts[14] by providing load-bearing exercise up to 600 pound force (lbf). This exercise capability contrasts with that of the previously-used interim resistive exercise device (iRED) (Fig. 3B), which provided only one-half of the resistance loading of the ARED. Coincident with the change from using iRED to ARED, the average monthly loss in aBMD decreased from roughly 1.0% (n = 24 iRED users) to 0.3% to 0.5% per month (n = 11 ARED users to-date, unpublished NASA data). Likewise, there is a consistent trend, observed in data from 45 different long-duration crewmembers,[15] for aBMD to increase in the postflight period (Fig. 4). However, there is considerable heterogeneity in the extent to which aBMD is regained after flight with some astronauts appearing to have a persistent deficit. Notably, DXA measurement of aBMD is often the only index considered when evaluating the efficacy of in-flight bone loss countermeasures and the return of bone health following flight. Overall, there is concern that DXA may underestimate skeletal risks due to spaceflight and reambulation on Earth, highlighting the potential utility of expanding measurements of bone beyond DXA aBMD to obtain enhanced estimations of bone strength and fracture risk.[16]

Figure 3.

(A) ARED. A crewmember performs a deadlift exercise on the ARED, which delivers up to 600 pound force (lbf) resistance exercise. Installed in December 2009, the ARED is used by crewmembers 3 to 6 days per week. It can provide high loads to the lower back, hip, knee, and ankle joints. The group mean declines in aBMD during flight appear reduced since the ARED was made available on the ISS, although in some of these astronauts, the beneficial effects of ARED on aBMD may be in combination with bisphosphonate use. In 2 cases, postflight aBMD has increased for lumbar spine relative to preflight BMD. Photo courtesy of NASA. (B) iRED. A crewmember performs a squat exercise on iRED. The iRED hardware was the primary mode of resistance exercise on the ISS from mid-2001 to 2010. The iRED delivered up to 300 lbf of resistive force through two canister assemblies (∼150 lb each). BMD losses were reported at ∼1.1% per month when the iRED and treadmill only were available for crew use. Photo courtesy of NASA.

Figure 4.

Representative trends in aBMD recovery in the lumbar spine (A) and femoral neck (B) after a long-duration mission. The percentage change in aBMD from the preflight value is plotted as a function of days after landing (when DXA scans were performed). The scatter-plot displays both longitudinal and cross-sectional data from U.S. and Russian long-duration crewmembers (n = 45) from 56 different long-duration missions (“pluses” denote data from a repeat flier). Data were fitted to a mathematical equation (solid line) with 95% confidence limits (dashed lines). The intersection of the horizontal dotted line with the y-axis = 50% of the average bone loss due to spaceflight and the intersection of the vertical dotted line with the x-axis = the number of days needed to recover 50% of lost bone (“half life”). In general, the recovery of aBMD for all sites measured was prolonged, with the time needed to regain mass being greater than the time for loss of mass to occur. Recovery of aBMD does not necessarily reflect a recovery pattern for bone strength (Adapted from Sibonga and colleagues[15]).

Quantitative computed tomography

In 1998, a flight experiment was conducted to evaluate the effects of spaceflight on QCT parameters at the hip and lumbar spine in ISS astronauts.[17, 18] Whole-bone geometry and vBMD of cortical, trabecular, and integral bone (cortical + trabecular) were measured 30 to 60 days before spaceflight, 7 to 10 days after landing, and 1 year after landing.[17, 18] In ISS astronauts (n = 16), vBMD declined at variable rates (Table 2A) at lumbar spine and hip trabecular and integral bone measurement sites.[17] Spaceflight resulted in a reduction in total hip bone mineral content. As reported, the reduced volume of cortical tissue (cm3), in combination with a stable total tissue volume (cm3) suggested that cortical thickness was reduced because of endocortical resorption.[17] In the spine, accelerated vBMD losses that occurred in integral bone were comparable to losses in trabecular bone.[17] QCT measures of the lumbar spine did not provide additional information about spaceflight-induced changes above and beyond DXA aBMD.[17] In contrast, as further described in Table 2B, QCT can provide additional information at the hip regarding spaceflight-induced changes, and recovery after return. QCT revealed that after 12 months of reambulation on Earth, total hip bone volume increased at both the proximal femur (total hip) and femoral neck whereas the vBMD was still decreased (ratio postflight vBMD/preflight vBMD <1; Table 2B).[18] This readaptation to Earth's 1g field in middle-aged astronauts is reminiscent of the expansion of bone's cross-sectional area observed with aging in terrestrial populations and with weight loss.[19, 20] Therefore, as part of a study extension, a fourth QCT hip scan was obtained 2 to 4 years after flight in 8 of the original 16 ISS crewmembers. Figure 5 displays different patterns of recovery in the hip and spine as assessed by DXA and QCT scans. In Fig. 5A, there is a tendency for aBMD to recover in the lumbar spine (L1–L4) after return, whereas trabecular vBMD, in QCT scans of L1 and L2 (Fig. 5B), declines after the first year in all but 1 astronaut. A similar discordant pattern is noted in Fig. 5C, D, where femoral neck aBMD in most individuals increased over the first year after return to Earth; however, 1 astronaut whose femoral neck aBMD exceeded preflight measurements 4 years after return, showed a reduction in trabecular vBMD over the same period.[21]

Table 2. (A) Changes in the Hip (Femoral Neck and Trochanter) and Lumbar Spine Measured in Astronauts After Return From Missions on the ISSa
Index DXA aBMD%/Month (n = 14)Index QCT vBMD% Change/month (n = 14)
Lumbar spine−0.8 ± 0.5Integral lumbar spine−0.9 ± 0.5
  Trabecular lumbar spine−0.7 ± 0.6
Femoral neck−1.1 ± 0.5Integral femoral neck−1.2 ± 0.7
  Trabecular femoral neck−2.7 ± 1.9
Trochanter−1.2 ± 0.9Integral trochanter−1.5 ± 0.9
  Trabecular trochanter−2.2 ± 0.9
(B) Changes at the Hip (Femoral Neck and Proximal Femur) With Reambulation on Earth by QCT Measurements in 16 ISS Astronautsb
 % Spaceflight change (means ± SD)% Reambulation change (means ± SD)Ratio of 1 year postflight to preflight
  • aAll values are means ± SD of aBMD by DXA and of vBMD by QCT performed in identical ISS astronauts (n = 14). Integral vBMD is measurement of combined cortical and trabecular bone.[17] At the lumbar spine, QCT measurement of vBMD did not generate additional information beyond DXA aBMD. Measurements were conducted as soon after spaceflight as possible (ie, within days).[17]
  • b

    All values for vBMD, BMC, cortical and total bone volumes are group means ± SD for ISS astronauts (n = 16). Data for 2 additional crewmembers were obtained since the earlier report in 2004.[17] Absolute BMD values were used to generate the ratio of postflight to preflight measurements.

    ISS = International Space Station; aBMD = areal bone mineral density; DXA = dual-energy X-ray absorptiometry; QCT = quantitative computed tomography; vBMD = volumetric BMD; BMC = bone mineral content; BMD = bone mineral density.

  • *Significant (p < 0.05) difference from preflight values.
  • **Significant difference between 1 year postflight and preflight measurements.[18]
Femoral neck
Total vBMD (g/cm3)−9.4 ± 6.4*0.9 ± 5.90.91**
Total BMC (g)−10.8 ± 10.6*8.1 ± 11.8*0.96
Cortical volume (cm3)−8.0 ± 11.2*8.0 ± 11.8*0.99
Total volume (cm3)−1.4 ± 10.97.2 ± 9.9*1.05
Total proximal femur
Total vBMD (g/cm3)−10.4 ± −9.7*4.4 ± 4.7*0.93**
Total BMC (g)−11.1 ± 11.2*12.2 ± 11.8*0.99
Cortical volume (cm3)−9.2 ± 10.8*11.5 ± 12.2*1.01
Total volume (cm3)−0.7 ± 10.1*7.2 ± 7.3*1.06
Figure 5.

Different individual trends in QCT and DXA data of the lumbar spine and femoral neck of ISS astronauts after return to earth. Eight crewmembers of the original 16 who participated in the QCT flight study[17] consented to an additional scan to evaluate recovery beyond 1 year after return to Earth. The fourth QCT scans were performed between 2 and 4 years after landing because of the staggered dates of return. The postflight vBMD and aBMD measurements were normalized to the preflight BMD (the y-axis) and plotted as a function of days after landing (the x-axis) when the scans were performed. The postflight DXA measurement of aBMD of the lumbar spine (L1–L4) increased in all 8 astronauts and continued to increase or stabilize over the next 2 to 4 years (A), whereas the QCT measurement of vBMD in the trabecular compartment of the lumbar spine (L1 and L2) increased in 2 of 8 astronauts during the first year of reambulation on Earth (B). After the first year, trabecular vBMD declined or stabilized (some at a level below baseline measurement) in 6 of 8 astronauts. For the femoral neck (C), the DXA measurement of aBMD for 2 astronauts exceeded or returned to the preflight measurement during the postflight monitoring period, whereas in 1 of these same astronauts, the QCT measurement of vBMD in the trabecular compartment (D) declined over the same time period (Reprinted from Carpenter RD, LeBlanc AD, Evans H, Sibonga JD, Lang TF. Long-term changes in the density and structure of the human hip and spine after long-duration spaceflight. Acta Astronautica. 2010;67:71–81, with permission from Elsevier[21]).


To translate bone density data to an estimation of bone strength, FE models were developed from hip QCT scans of 11 ISS astronauts.[17, 22] FEM is a computational tool that estimates hip strength (in Newtons, N, of force) for specific loading orientations (Table 3). FEM detected a significant decline in hip strength after spaceflight (group means ± SD) for two modeled orientations of loading (axial loading in one-legged stance and posterolateral loading that assumes falling backward to the side).[22] When declines in hip strength were divided by total months in space, as similarly reported for loss of DXA aBMD, the monthly decrease in FEM estimates of hip strength are approximately double the monthly rate of decrease in aBMD (Table 3).[22] Figure 6 shows the correlation between the spaceflight-related (preflight to postflight) changes in FEM strength and aBMD. There was little correlation between DXA and either of the two loading models (R2 = 0.23 for one-legged stance and R2 = 0.05 for posterolateral fall).[22] These data do not indicate whether DXA or FEM is superior in predicting bone health but do suggest that FEM may capture changes in bone that DXA does not.[22]

Table 3. Hip Bone Strength from Finite Element Models of QCT Data From 11 Long-Duration Astronauts Before (Preflight) and Immediately After (Postflight) ISS Missions
Loading conditionPreflightPostflightpLoss in hip strength (%/month)
  1. Hip strength (force to failure, in Newtons [N]) is estimated for two loading orientations from finite element models of QCT scans. All values are group means ± SD for ISS astronauts (n = 11). Average calculated monthly loss in hip strength was calculated from total percentage loss over mission divided by total months of spaceflight.[22] Immediate postflight QCT scans after ISS missions were performed within 1 month of landing due to travel to site of preflight QCT scanner.
  2. QCT = quantitative computed tomography; ISS = International Space Station.
Stance13,200 ± 300 N11,200 ± 2400 N<0.0012.2
Fall2,580 ± 560 N2,280 ± 590 N<0.0031.9
Figure 6.

Spaceflight-induced changes of hip bone strength in ISS astronauts. Hip bone strength of ISS astronauts before and after spaceflight was assessed by preflight and postflight scans of hips by both DXA and QCT. FE models of QCT data estimated hip bone strength for two loading scenarios (single-legged stance and posterolateral fall) whereas hip aBMD represents the widely-applied surrogate for bone strength. The y-axis is the change in hip strength estimated by FEM and the x-axis is change in aBMD as a DXA-measured surrogate for hip bone strength. There was poor correlation (stance: R2 = 0.23; fall: R2 = 0.05) between these two methods for assessing changes, suggesting that QCT and FEM detect changes in hip bone strength due to spaceflight that are not captured by DXA aBMD (Adapted from Keyak and colleagues[22]).

Bone turnover markers

Figure 7 displays the group trends in biochemical markers of bone turnover from U.S. crewmembers aboard both the ISS and Russian Mir spacecraft.[23] The data for N-telopeptide (NTX) measures from 24-hour non-fasted urine specimens suggest that bone resorption increased early during long-duration missions and remained elevated throughout the period of weightlessness but was restored to baseline status upon return to Earth.[23] In contrast, the concentration of bone-specific alkaline phosphatase (BAP), a marker of bone formation (Fig. 7), was reduced or unchanged during spaceflight,[23] suggesting that an uncoupling of bone resorption and formation occurs. The increase in bone resorption, without an increase in bone formation, could be expected to yield net loss of bone mass.

Figure 7.

Bone remodeling during spaceflight assessed by biomarkers of bone turnover. Changes in bone turnover biomarkers, measured in urine and blood obtained before, during, and after spaceflight, are used to reflect cumulative changes in bone cell activities. N-telopeptide (NTX) (n = 11) in urine (% change from preflight measure) and bone-specific alkaline phosphatase (BSAP, U/L) (n = 6) produced in circulation by osteoblasts were measured in crewmembers on the Mir spacecraft and the ISS. Biomarkers highlight the uncoupling of bone remodeling during spaceflight. Bone resorption was elevated early and persisted throughout spaceflight, whereas bone formation was unresponsive to the stimulation of bone resorption, resulting in net bone loss. The vertical lines denote the temporal separation of in-flight from postflight samples (Adapted from Smith and colleagues[23]). FD# = number of flight days into the mission; R + # = number of days after return to Earth.

In-flight countermeasure evaluations

Resistive exercise is the only countermeasure routinely used to mitigate bone loss in all long-duration astronauts. The ARED (Fig. 3A) was flown to the ISS in December 2008 to supplement the iRED (Fig. 3B), along with the cycle ergometer and treadmill exercise hardware that have also been used in flight since 2000. In the few ARED users in which data have been obtained (Fig. 2), the average monthly loss in aBMD (n = 8) was reduced to 0.47% ± 0.53% (group mean ± SD) at the total hip with an average monthly gain of 0.30% ± 0.94% at the lumbar spine—compared with monthly losses of 1.1% ± 0.45% (total hip) and 0.71% ± 0.51% (lumbar spine) in aBMD with use of the iRED (n = 24). In next paragraph, the interpretation of this apparent change is confounded by the use of bisphosphonates in 3 of the astronauts during the period of ARED availability.

Pharmacologic measures are also being considered to counter flight-related bone loss. To this end, bisphosphonate therapy during spaceflight is being studied in a joint experiment by NASA and the Japan Aerospace Exploration Agency (JAXA). The study is testing bisphosphonates (alendronate or zoledronic acid) to prevent loss of bone mass, structure, and strength at the hip, quantified by DXA, QCT, and FEM, in long-duration astronauts. The study was approved for testing in astronauts, with a planned statistical comparison to 16 historical controls.[17, 18] However, only alendronate has been offered to U.S. astronauts because of NASA's concern about the limited safety data for zoledronic acid relative to alendronate. Participants in the bisphosphonate flight study also had access to ARED for resistive exercise, whereas the historical controls did not. Thus, the attenuated postflight aBMD loss detected in ARED users will be confounded by the antiresorptive effects of alendronate in the exercising astronauts. The data from this bisphosphonate trial are not available for presentation here as the study is ongoing.

Bone Summit Panel Recommendations

Overall assessment

Long-duration astronauts can experience 10% to 15% aBMD decline during a spaceflight mission and still meet NASA standards for bone health. In the panel's opinion, interventions to mitigate bone loss during spaceflight are advisable for long-duration astronauts, in order to minimize risk for adverse skeletal effects later in life. However, because the long-term adaptation of bone to space has only been evaluated in a small number of astronauts, the panel considers the available data insufficient to confidently recommend a specific therapeutic course of action. Additional well-designed studies are warranted.

Measures of bone structure (such as trabecular microarchitecture, whole-bone geometry, and 3D mass distributions in cortical and trabecular compartments) are key determinants of bone strength[24] that could be uniquely affected by weightlessness and may not be sufficiently accounted for by DXA technology. The data (reviewed at the Summit) suggest QCT-based measures may provide distinct information about bone health in astronauts. The specifics of these changes and how they influence the fracture risk of an astronaut with reloading are not well understood. Moreover, if spaceflight-induced changes persist, then they could combine with later age-related changes and prematurely increase fracture risk later in life. Thus, the impact of space-induced skeletal changes on hip strength will remain an open issue for NASA until structural indices of bone in long-duration astronauts, such as those acquired with QCT, can be serially evaluated.

A resolution of the relative merits of aBMD versus QCT measures in this situation is hampered by novel constraints, including understudied exposure variables (spaceflight), small numbers of affected individuals, and variation in subject characteristics (age, training, flight duration, and flight conditions). Nevertheless, the potential value of additional measures should not be ignored. Hence, the panel recommends that all long-duration astronauts be evaluated for spaceflight-induced changes in separate compartments of the hip (total, femoral neck, and trochanter) by conducting preflight and postflight QCT-based measures. Also, the panel recommends that QCT measures be incorporated in a surveillance program and the evaluation of astronaut eligibility and that efforts to better understand the appropriate role of QCT measures should be ongoing. The heterogeneous changes in the spine with aging, however, are problematic and a surveillance regimen for the spine cannot be recommended at this time beyond monitoring BMD and vertebral fracture assessments with DXA. However, new technologies and analyses should be explored in research studies to generate data for the panel's review in the near future.

Based upon its review of trends in astronaut data and individual case reports, the Bone Summit Panel offered the following recommendations for bone health management in long-duration astronauts: (1) expand the characterization of spaceflight changes to include QCT of the hip for risk surveillance, (2) investigate the impact of spaceflight on hip bone strength with the use of FEM, and (3) modify current risk mitigation approaches before, during, and after spaceflight to optimize effectiveness.

QCT for risk surveillance

The sole reliance on DXA to assess spaceflight effects and countermeasure efficacy may provide inadequate information to NASA decision-makers. Although aBMD as measured by DXA may be a robust predictor for osteoporosis-related fractures in the general population, it may be insufficient for understanding fracture risk in astronauts. First, the astronaut population is physically fit but exposed to novel environmental conditions and mission activities in microgravity that may have unique effects on bone. Standard aBMD-based clinical guidelines for monitoring fracture risk and assessing therapeutic efficacy are derived from terrestrial populations that are generally older and with conventional clinical risk factors (eg, menopause, rheumatoid arthritis, glucocorticoid treatment, falls, etc.).

Second, aBMD measures may not adequately detect the unique changes induced by spaceflight and postflight reloading. DXA, for example, cannot distinguish cortical restructuring (ie, modeling) that has occurred as the result of spaceflight, that could occur with exercise loading of bones during spaceflight, or could occur after return to Earth's gravity.[18] An increase in cortical bone will likely overwhelm the aBMD measure and mask quantitatively less impressive but biomechanically important effects in trabecular bone. Although not fully understood, other indices of bone structure (such as trabecular microarchitecture, whole-bone geometry and cross-sectional dimensions, and cortical bone width) are recognized determinants of bone strength, but are poorly measured with DXA.[24, 25] The available astronaut data suggest that spaceflight may result in a unique form of bone loss with structural implications not adequately assessed by aBMD measures. Data from both QCT and DXA indicate that skeletal changes caused by spaceflight are specific to skeletal regions and to bone compartments.[4, 11, 26] Moreover, there is a precipitous rate of aBMD decline in crewmembers[4, 11] that may be due in part to the aggressive osteoclast activity inferred from iliac crest biopsies of skeletally-unloaded bed rest subjects.[27-29] Bone turnover markers suggest uncoupled bone remodeling during spaceflight,[23] a condition that could result in the formation of stress risers in trabecular bone.[30] Similarly accelerated losses in postmenopausal women can result in microarchitectural disruptions and increased fractures, especially at central skeletal sites.[31] The failure of hip trabecular vBMD to return to preflight values by 2 years after return is of concern,[21] especially since deficits in trabecular vBMD of the femoral neck were reported to predict hip fracture independent of aBMD in elderly men.[25] Thus, the space-induced changes in bone quality could put an astronaut at a higher fracture risk than the average Earth-based person with the same aBMD. The persistent structural deficits after spaceflight also could be further worsened by changes induced by a second flight as well as by aging or other risk factors.[21, 32]

Although the exact contribution of trabecular bone loss or structural change on hip strength requires further definition, the absence of recovery-to-baseline measurements in trabecular bone indicates irreversible changes in strength have probably occurred. Therefore, it could be clinically meaningful to implement QCT hip scans to assess the efficacy of new in-flight countermeasures and for surveillance postflight, in which an absence of recovery could be a trigger to review bone health risk factors and/or to consider intervention to prevent expected age-related bone loss. The specific changes to bone microarchitecture (eg, trabecular thickness, trabecular separation, connectivity, etc.) and their impact on strength and fracture risk will require further research since QCT does not detect trabecular microarchitectural changes.

It is recognized that a recommendation based on an established relationship between QCT-based measures and fracture risk is a standard that is impossible to meet considering the limited data available on astronauts. Nevertheless, the uncertainties associated with the assessment of bone character with aBMD in astronauts suggest that more sensitive and innovative approaches may be required for longer-term surveillance. It is recommended that measurement of bone structure and of separate bone compartments using QCT should be instituted. With the end of the Space Shuttle program, launches and landings of crew transport systems would occur in Russia; thus, the panel recommended that QCT measures be conducted in the United States as close as possible to launch and landing dates.

FEM for estimations of hip bone strength

The collective evidence from spaceflights raises the possibility that prolonged space habitation affects the skeleton in such a way that declines in mechanical strength may not be detected by current clinical technologies.[17, 21, 22] Computation of QCT-FEM strength can complement the existing medical assessment tests (DXA and bone turnover markers) and the more conventional QCT structural indices. In ISS astronauts, declines in bone strength, estimated by FEM, are evident after spaceflight, in spite of the pristine medical history and extreme physical fitness of the typical astronaut before spaceflight.[22] Emerging data indicate that FEM estimates of hip strength may be related to fracture risk,[33-35] especially in combination with aBMD. FEM estimates of hip failure load quantify the ability of the hip to resist fracture for a specific load vector. This index may be the single best existing composite assessment of bone strength because of its ability to integrate applied loads with geometry and distribution of material properties (BMD, elastic modulus, and yield strength) in 3D bone structure.[36] Although there are no data to indicate that FEM estimation of strength is an improved index compared to aBMD, the use of multiple determinants of bone strength[36] by FEM, in conjunction with the single aBMD surrogate for bone strength, may enhance the assessment of fracture probability in each astronaut for individualized clinical decisions. Finally, data from QCT and FEM strength modeling can be used to optimize a probabilistic fracture model, developed by NASA, to calculate applied loads to bones and fracture probability.[37] This, and other modeling efforts, could be used for estimating fracture risk during exploration missions on planets or asteroids, monitoring risk during return to normal activities on Earth, and for understanding the combined effects of two long-duration missions or of spaceflight changes with the expected age-related changes. As surveillance data accumulate for the astronaut population, recommendations for future extended missions beyond low Earth orbit should be formulated or modified.

The possibility of low-trauma postflight fractures is a critical driver for assessment and appropriate intervention. The combination of space effects with postflight aging may predispose long-duration astronauts to fragility fractures at an earlier age on Earth. FEM has been applied to QCT scans from aging population cohorts, in which FEM-estimated hip strength is associated with incident hip fractures.[33, 34] Such modeling could also be used to direct bone rehabilitation efforts after astronauts return to Earth, to avoid the risk of overloading bones, especially with exercise. The translation of QCT data to FEM strength provides an individualized functional index at the hip for each astronaut. From the occupational risk perspective, bone medical standards based on QCT data and FEM estimation of hip fracture loads could provide NASA with the clinical practice guidelines needed for advising astronauts against physical activities with high impact loads to the hip. These data may provide the basis for developing new medical standards to supplement the current aBMD-based standards. This translation—from measured surrogates for bone strength to calculated estimates of bone strength—could add confidence when addressing important issues. Are an individual astronaut's hip bones strong enough to perform mechanically-loaded tasks on a mission, to return to preflight activities on Earth, or to accommodate age-related bone loss?

Given the constraints of low subject numbers and slow data acquisition, attempts should be made to use the most powerful research technologies and analyses available to identify astronauts who may be at risk for developing premature osteoporosis. Thus, it is recommended that NASA evaluate hip strength with QCT-derived FEM estimates as a new surrogate to assess fracture risk following long-duration spaceflight. Additional results are accumulating from ongoing terrestrial studies, which will help to inform the interpretation of astronaut data.

Risk mitigation approaches

Preflight selection and certification standards

  • The NASA Space Flight Human System Standards[13] establish ranges of crew health required to minimize risks to health and performance during spaceflight missions. Consequently, these medical standards—for selecting applicants for astronaut candidacy, for certifying an astronaut for a spaceflight mission, or for disqualifying an astronaut from a second spaceflight mission—could be used to protect those astronauts who are at greater risk for losing bone strength during spaceflight that would put them at increased fracture risk with subsequent aging. The current standards are based upon aBMD by DXA, which may not capture all parameters of bone strength and fracture risk. Therefore, NASA should expand the use of QCT-based FEM estimates of hip strength as a selection standard for bone health. The growing literature linking FEM measures of strength to fracture probability provides a basis for the practical application of these measures in astronauts.

In-flight countermeasures

  • NASA cannot validate any single in-flight mitigation strategy as long as multiple bone-loss countermeasures are applied in individual astronauts.
  • There is limited information about side effects of pharmaceutical agents under weightless conditions. Thus, NASA should continue to focus on management strategies of modifiable risk factors for bone loss (by exercise and dietary manipulations) as the standard methods for bone loss prevention during ISS missions before using Earth-based pharmaceutical interventions. This recommendation may be refined in the future as accumulated data, and research on emerging pharmaceuticals, are reviewed.
  • The limited, flight data for alendronate are encouraging and suggest that treatment may mitigate hip bone loss during flight.[38] However, the onset of the alendronate study coincides with the implementation of new ARED hardware; therefore, it is not possible to distinguish the contribution of each intervention on the prevention of aBMD loss or the underlying structural changes. Studies are required to compare intervention by combined alendronate and ARED exercise with intervention by only the ARED exercise.
  • If an on-orbit pharmaceutical intervention is deemed necessary for longer missions, then the panel suggested zoledronic acid be considered for the in-flight experiment based on two major considerations: (1) a single intravenous (IV) infusion of zoledronic acid may be preferred to weekly oral alendronate, or other oral bisphosphonates available, because of the association of oral bisphosphonates with adverse upper gastrointestinal effects and the knowledge gaps related to esophageal adhesiveness and pill dissolution in the microgravity environment, and (2) the preflight administration of an IV bisphosphonate would enable flight surgeons to address any side effects while astronauts are still on Earth.
  • Denosumab may be useful, but long-term clinical data are only now emerging and are needed to identify possible low-frequency adverse events.
  • Teriparatide might be effective for crewmembers who do not regain BMD with conventional treatments over an extended postflight period.
  • All agents being considered for flight may require appropriate testing, including in ground-based spaceflight analogs, before undertaking trials in astronauts.
  • There is some concern for in-flight stress fractures and fractures that may occur during in-flight exercise on the new ARED exercise hardware.
  • QCT testing should be part of the evaluation of the efficacy of all in-flight countermeasures to mitigate or prevent losses in hip trabecular vBMD.
  • Because the impact of changes to the trabecular microstructure of the hip is not known and may be irreversible.[39] NASA should help continue to develop technologies that can be used to understand changes in trabecular microarchitecture at sites in the hip and spine of astronauts.

Postflight surveillance

  • Continued use of DXA for risk surveillance is recommended for astronauts because of the abundance of aBMD data from terrestrial populations as reference cohorts for aging effects.
  • Decisions regarding spaceflight-induced skeletal loss, for both clinical care and subsequent flight certifications, should not rely solely on measures performed using DXA. As QCT measures become available on individual astronauts, they should be incorporated into the postflight clinical evaluation of skeletal status.

An osteoporosis specialist should evaluate astronauts for treatable risk factors if his/her DXA postflight aBMD T-score is less than –2.0 or if hip QCT results do not recover to preflight values (within LSC of preflight measure) by 2 years after return. Although flight-induced loss of vertebral strength should not be ignored in future studies (for instance with QCT), the astronauts are now of an age at which some osteophytic change is conceivable, and, since extravertebral mineralization may obscure aBMD change in the spine, the primary measurement site should be the hip. Vertebral fracture assessment (VFA) is a recent addition to medically required DXA scans and should continue to be conducted in all astronauts during DXA tests.

Summary of Recommendations

Recommendation 1

Given the few numbers affected and the unique nature of the problem, the adverse effects of spaceflight on the skeleton can be considered a form of rare syndrome. In light of the current understanding of its pathophysiology and the preliminary data available, the adoption of novel assessment methods (eg, QCT, FEM) is appropriate.

Although useful, the current aBMD-based fracture standards for risk assessment (originally developed in older women) are probably not sufficient for assessing risk in astronauts. Preflight and postflight QCT hip scans should be collected on astronauts to evaluate the impact of spaceflight on structural determinants of bone strength. FEM estimates of hip strength, as performed in terrestrial population studies, should be analyzed to derive hip strength cutoffs as new medical standards for bone health in astronauts; eg, to qualify astronauts for long-duration spaceflight and to evaluate applicants for the astronaut corps.

Recommendation 2

As QCT-based assessments of bone health are refined, the current use of aBMD measures for preflight and postflight assessments should continue. DXA generates minimal cumulative radiation exposure and enables simultaneous monitoring of changes in body composition. Moreover, DXA monitoring will add to the available, historical dataset on astronauts and additional aBMD data collected in parallel with new QCT measures will considerably add to the understanding of bone change associated with spaceflight, the development of appropriate countermeasures, and the assessment of postflight bone health.

Recommendation 3

For ISS missions, NASA should focus on reducing known modifiable risk factors for bone loss, such as reduced physical activity and suboptimal nutrition.

Recommendation 4

The use of pharmacological interventions is of interest to reduce spaceflight-associated bone loss, especially for spaceflights of longer durations. The design of future studies on the utility of such interventions should be carefully considered to ensure clear outcomes.


EO has received consultancy fees, grant/research support, and/or has served on advisory boards for Amgen, Merck & Co, and Eli Lilly; EO was also a member on NASA's Research on Human Health Risks, Institute of Medicine of the National Academies. RA has received consultancy fees and grant/research support from Amgen and/or Eli Lilly. SA has received consultancy fees and serves on an advisory board for Merck & Co. and has received grant/research support from NASA Human Research Program. NB has received consultancy fees and/or received grant/research support from Amgen, Merck & Co., and Eli Lilly. EML has received grant/research support, served on advisory boards and/or served on a Speakers Bureau for Amgen, Eli Lilly, Merck & Co, and Novartis. SP has served on a Speakers Bureau for Amgen and is employed as a bone densitometry and endocrine consultant to NASA-Johnson Space Center. NW has received honoraria, consultancy fees and/or grant/research support from Amgen, Eli Lilly, Novartis, and Merck & Co. All other authors state that they have no conflicts of interest.


Appreciation is extended to Dr. Angelo A. Licata (Cleveland Clinic, Center for Space Medicine) and Dr. S.V. Reddy (Medical University of South Carolina) for their thoughtful review of this manuscript and to Dr. Jane Krauhs and Dr. Kathleen McMonigal, MD (both of Johnson Space Center) for their editorial assistance. In addition, authors wish to acknowledge the contributions of Dr. Scott M. Smith, Dr. Sara R. Zwart, Dr. Andrea M. Hanson, Dr. Harlan Evans, Ms. Elisabeth R. Spector, and Ms. Adriana Babiak-Vazquez (all of Johnson Space Center), Dr. Adrian D. LeBlanc (Universities Space Research Association), Dr. Thomas F. Lang (University of California at San Francisco), and Dr. Joyce H. Keyak (University of California at Irvine) for the presentations of astronaut data in this review. The Bone Summit activity was funded by Human Adaptation and Countermeasures Division at NASA Johnson Space Center.

Authors' roles: JS convened the Bone Summit Panel, developed the charge to the panel, drafted the manuscript and was responsible for the integrity of the data presented herein. EO provided extensive manuscript revision. EO and SA provided guidance to JS on the relevant astronaut data to review and facilitated discussion sessions at the Bone Summit. The Bone Summit Panel (EO, RA, SA, NB, ML, SP, SS, MS, and NW) provided the interpretation of the astronaut data, formulated the recommendations, revised the manuscript content and approved the final version of the manuscript.