The Relationship Between Basal Metabolic Rate and Femur Bone Mineral Density in Men With Traumatic Spinal Cord Injury

Methods 

Men with chronic (time since injury, >1y) traumatic SCI aged between 18 and 45 years were recruited to this study with their informed consent. Our local hospital ethics committee approved the study design. To create homogenous groups, all motor incomplete patients were excluded from the study (ie, only patients with grades A and B on the American Spinal Injury Association Impairment Scale [AIS] were included). Exclusion criteria for this study were thyroid dysfunction, amputation, nontraumatic SCI, heterotopic ossification, respiratory dysfunction, pressure ulcer, and infections. Of 42 consecutive patients, 8 were excluded because of exclusion criteria (3 patients had pressure ulcers, 3 had heterotopic ossification, 2 had urinary tract infections). Four patients refused to participate. Thirty patients who met the inclusion criteria and agreed to participate were included in the study.

Neurologic examination was performed for all participants. Body weight (in kilograms) was measured on a digital wheelchair scale, and BMR/kg was calculated for each patient.

Measurements of BMR were determined by indirect calorimetry under standardized conditions, using a nose clip and a respiratory valve with a rubber mouthpiece held in place by a mechanical arm. The automatic gas analyzera was calibrated before each measurement and was also regularly calibrated for volume. All measurements were performed between 8:00 and 9:30 am after an overnight fast (12−14h), and room temperature was maintained at 22° to 24°C. Each patient rested for 10 minutes before the start of the measurement and was given the opportunity to get acquainted with the new environment, mouthpiece, and nose clip. During the measurement, each patient lay supine on a bed. The gas analyzing system measured continuously over a period of 45 minutes, and printouts were made at 1-minute intervals describing the mean values per minute. The first 15 minutes of each recording were discarded to ensure a steady state, because measured values in the first 15 minutes have been shown to be significantly higher compared with a subsequent period of measurement.10, 11, 12

Proximal femur BMD was measured with a bone densitometer that used dual energy x-ray absorptiometry (DXA).b Patients were allocated into groups according to criteria for osteoporosis proposed by the World Health Organization (WHO): an osteoporotic group (T score ≤ −2.5 standard deviations [SD]), an osteopenic group (T score range, −2.5 to −1 SD), and a group with normal bone density (T score ≥ −1 SD).13 T scores represent the number of SDs of BMD from the young adult mean of the healthy population.

DXA was used also to estimate lean- and fat-tissue mass (in kilograms) by standard methods.14 The radiation passed through the patient from below, and the differential absorption was measured above. The ratio of absorption between the 2 radiographs of different energies was linearly related to the lean-tissue mass compartment. DXA measurements were performed on the same day as BMR analysis.

The differences between subjects with tetraplegia and paraplegia and between patients with complete and incomplete injury were statistically analyzed with the Mann-Whitney U test, and the correlations were performed with Spearman correlation analysis by using standard software.c Univariate analysis of variance was used to compare BMR and BMR/kg values within the osteoporotic group, osteopenic group, and group with normal bone density.

Results 

Thirty men with chronic traumatic SCI (mean age, 32±10y) were enrolled in this study. Nineteen (63.3%) were paraplegic, and 11 (36.7%) were tetraplegic. They had sustained injury an average of 30.2 months before the study.

BMR correlated significantly with BMD of the total femur, femur neck, trochanter, and shaft. However, there was no correlation between BMR and BMD of Ward’s triangle. These correlations were stronger in tetraplegic patients (table 1). There was a moderate correlation between BMR and lean-tissue mass (r=.66, P<.001), although femur BMD values did not correlate with lean-tissue mass in our study group. Fat-tissue mass did not correlate with hip BMD and BMR (table 2).

Table 1. Correlation Between Hip BMD and BMR Values

	BMD	BMR
		Patients With Paraplegia (n=19)		Patients With Tetraplegia (n=11)		All Patients (N=30)
		r	P	r	P	r	P
	Total femur	.33	.09	.53	.09	.41⁎	.02
	Femur neck	.39	.16	.67⁎	.02	.44⁎	.01
	Femur trochanter	.32	.17	.62⁎	.04	.43⁎	.02
	Femur shaft	.29	.22	.65⁎	.02	.37⁎	.04
	Femur Ward’s triangle	.26	.28	.52	.09	.30	.11

⁎ P<.05.

Table 2. Spearman Correlations Among BMR, Hip BMD, Lean-Tissue Mass, and Fat-Tissue Mass

	Variables	BMR		BMD		LTM		FTM
		r	P	r	P	r	P	r	P
	BMR	NA	NA	.41	.02	.66	.001	−.34	.06
	BMD⁎	NA	NA	NA	NA	.24	.18	.11	.57
	LTM	NA	NA	NA	NA	NA	NA	−.33	.07

Abbreviations: FTM, fat-tissue mass; LTM, lean-tissue mass; NA, not applicable.

⁎ Total hip BMD value of proximal femur.

When we compared patients with sensory complete and sensory incomplete injuries, we observed no significant difference in lean-tissue mass, BMR, and hip BMD values (table 3). Although we found no difference in hip BMD values between tetraplegic and paraplegic patients, BMR and lean-tissue mass values differed significantly for them (table 4).

Table 3. Comparison Between Patients With Sensory Complete and Patients With Sensory Incomplete Injuries

	Variables	AIS Grade A (n=22)	AIS Grade B (n=8)	P
	Age (y)	31.86±10.78	32.50±10.55	.72
	Fat-tissue mass (kg)	28.94±11.82	32.82±11.45	.48
	Lean-tissue mass (kg)	38.23±12.13	38.05±8.78	.43
	BMR (kcal/d)	1433.09±488.12	1170.00±393.82	.31
	BMR/kg	21.90±8.37	16.56±4.98	.12
	Femur neck BMD	0.88±0.21	0.81±0.20	.37
	Femur Ward’s triangle BMD	0.83±0.23	0.75±0.21	.48
	Femur trochanter BMD	0.71±0.18	0.68±0.16	.62
	Femur shaft BMD	1.03±0.24	0.93±0.24	.39
	Total femur BMD	0.88±0.19	0.82±0.20	.49

NOTE. Values are mean ± SD.

Table 4. Comparison Between Patients With Paraplegia and Tetraplegia

	Variables	Tetraplegia (n=11)	Paraplegia (n=19)	P
	Age (y)	28.6±10.4	34.0±10.3	.06
	Fat-tissue mass (kg)	33.85±10.44	27.73±11.99	.15
	Lean-tissue mass (kg)	33.23±9.56	41.05±11.27	.02⁎
	BMR (kcal/d)	1128.63±299.90	1498.57±507.85	.03⁎
	BMR/kg	16.84±3.66	22.58±8.97	.01⁎
	Femur neck BMD	0.88±0.14	0.85±0.24	.98
	Femur Ward’s triangle BMD	0.84±0.14	0.79±0.26	.51
	Femur trochanter BMD	0.69±0.12	0.70±0.20	.68
	Femur shaft BMD	1.01±0.19	1.00±0.27	.80
	Total femur BMD	0.86±0.14	0.86±0.22	.77

NOTE. Values are mean ± SD.

⁎ P<.05.

Patients who were divided into 3 groups according to T scores of the femur showed no difference in BMRs (table 5).

Table 5. Results of Univariate Analysis of Variance Between Groups

	Group	BMR (kcal/d)	P	BMR/kg	P
	Normal BMD (n=9)	1477.33±398.29	.71	21.42±5.10	.53
	Osteopenic (n=13)	1320.46±235.94		18.57±4.35
	Osteoporotic (n=8)	1303.25±791.78		22.51±13.58

NOTE. Values are mean ± SD.

Discussion 

In this study, we investigated the relationship between BMR, body composition, and femur BMD values in men with SCI. We found that BMR correlated strongly with lean-tissue mass, whereas there was no correlation between BMR and fat-tissue mass. In other words, BMR was higher among those with more lean-tissue mass. This finding was consistent with the previous literature.15 On the other hand, our results show that there is a positive correlation between BMR and hip BMD in patients with SCI, especially in subjects with tetraplegia. It is known that resting energy expenditure correlates strongly with spine, total hip, and total body bone mineral content and BMD in normative populations.16 To our knowledge, this is the first study examining the association between hip BMD and BMR in patients with SCI.

Osteoporosis in patients with SCI has some different features from those seen in normative populations. The pattern of bone loss in SCI is partly different from that in the other endocrine causes of osteoporosis (eg, chronic steroid use, chronic neuroleptic use, prolactinoma, hypogonadism, hyperthyroidism, idiopathic osteoporosis).17 Osteoporosis is mainly regional and occurs below the level of injury. After SCI, an early increase of osteoclastic bone resorption is associated with a pronounced decreased osteoblastic bone formation.18 Demineralization begins with losing large amounts of calcium and other minerals in the urine.18 The demineralization rate seems to be affected directly by the removal of load from bone after injury.

BMD at different sites declines with increasing age and is inversely related to the time after injury, which indicates continuous bone loss beyond the first 2 years postinjury.19 A study6 that examined BMD in physically active subjects with SCI and able-bodied subjects showed that BMDs of both the hips and legs were significantly lower in the SCI group. However, there was no difference between the 2 groups in the BMD of the spine and arms.6 In patients with SCI, osteoporosis occurs exclusively in the sublesional areas and predominantly in weight-bearing skeletal sites such as the distal femur and proximal tibia, which are trabecular-rich sites, whereas the diaphyseal areas of the femur and the tibia, which are cortical-rich sites, are relatively spared.20 The predominance of bone demineralization at the distal femur or proximal tibia could explain why these areas are preferential fracture sites.21 Although the most common site of demineralization reported in the literature for SCI patients is the knee, we used hip BMD values in our study. The reason for this was the lack of defined fracture thresholds and T scores for knee osteoporosis in patients with SCI.

Demirel et al22 reported that BMD values of the lower extremities were similar in subjects with paraplegia and tetraplegia. We also found no difference in hip BMD values between subjects with tetraplegia and paraplegia. Nevertheless, lean-tissue mass and BMR were significantly lower in the tetraplegic group. Injury level does not affect hip BMD scores according to our results. Similarly, in another study17 that was performed to evaluate supralesional and sublesional BMD in SCI patients, the researchers claimed that the degrees of demineralization for the lumbar spine, pelvis, and lower limbs were independent of the neurologic level.

In our study there was no difference in BMR, hip BMD, or lean-tissue mass between patients with sensory complete and patients with sensory incomplete injuries. We thought that sensory preservation has no effect on these parameters, and it seems that the effect of injury level on BMR is solely by way of its effect on lean-tissue mass in SCI. On the other hand, we checked whether there was any difference among BMR values when we grouped patients according to WHO criteria for osteoporosis. There was no significant difference among groups; this showed us that this grouping, according to WHO criteria, has no discriminative value.

There are many factors that affect BMD, but physical activity and body composition are the best known positive factors. Lean- and fat-tissue mass are 2 components that form body composition. Although both soft-tissue components have some impact on bone, lean-tissue mass has a greater effect than fat-tissue mass on bone density per kilogram of tissue mass.23 Generally, it is known that muscle mass correlates with the magnitude of bone.24 Makovey et al25 reported that central adiposity has a positive relationship with BMD in men over 50 years and women under 50 years, although lean-tissue mass had stronger relationships with bone measures than fat-tissue mass in both sexes at all ages with the measures of volumetric density. However, our results show that there is no correlation between hip BMD and fat- or lean-tissue mass. In patients with SCI, body composition is severely modified. In this respect, some differences may be seen in the relationship between body composition and BMD. Total fat-tissue mass is higher in patients with SCI than in healthy control subjects, whereas lean-tissue mass and BMD are significantly lower.26

Choi and Pai8 presented a study that described for the first time that BMR correlated more strongly with bone density in elderly people than lean-tissue mass, total body fat, and body mass index. Similarly, we found that hip BMD correlated more strongly with BMR than with lean-tissue mass or injury level in patients with SCI. These results indicate that bone metabolism is a major component of BMR; however, the clinical importance of this relationship is not established. Identifying the markers of bone metabolism can be of help to have a better understanding. Future studies should focus on the relationship between BMR and the markers of bone turnover and also how BMR relates to the changes in those markers during different stages of SCI such as acute, subacute, and chronic.

Conclusions 

There is a strong correlation between hip BMD and BMR in men with SCI. Patients with paraplegia had higher BMR values than patients with tetraplegia. Hip BMD values did not differ significantly between patients with paraplegia and tetraplegia and those with sensory complete and incomplete injury.