Skeletal Muscle Mass in Acromegaly Assessed by Magnetic Resonance Imaging and Dual-Photon X-Ray Absorptiometry

Abstract

Context: GH and IGF-I are nitrogen retaining and anabolic, but the impact of long-term exposure to supraphysiological GH and IGF-I, either from endogenous overproduction in acromegaly or exogenous sources, on skeletal muscle (SM) mass is not clear.

Objectives: The objectives of the study were to assess SM mass by whole-body magnetic resonance imaging (MRI) in acromegaly and test the hypothesis that dual-energy x-ray absorptiometry (DXA) lean tissue mass-derived estimates of SM accurately estimate true SM mass.

Design, Setting, and Patients: The design was a cross-sectional study in 27 acromegaly patients compared with predicted models developed in 315 nonacromegaly subjects and to matched controls.

Outcome Measures: Mass of SM from whole-body MRI and lean tissue from DXA were measured.

Results: SM mass did not differ from predicted or control values in active acromegaly: 31.75 ± 8.6 kg (acromegaly) vs. 33.06 ± 8.9 kg (predicted); SM was 95.6 ± 12.8% of predicted (range 66.7–122%) (P = 0.088). Lean tissue mass (DXA) was higher in acromegaly than controls: 65.91 ± 15.2 vs. 58.73 ± 13.5 kg (P < 0.0001). The difference between lean tissue mass (DXA) and SM in acromegaly patients was higher than that in controls (P < 0.0001) consistent with an enlarged non-SM lean compartment in acromegaly. SM mass predicted by DXA correlated highly with SM mass by MRI (r = 0.97, P < 0.0001). SM (MRI) to SM (DXA predicted) ratio was 1.018 (range 0.896–1.159), indicating high agreement of these measures of SM.

Conclusions: SM mass in active acromegaly patients did not differ from predicted values. SM mass estimated from DXA agreed highly with SM by MRI, supporting the validity of the DXA model in assessing SM in acromegaly and other disorders of GH/IGF-I secretion.

The DXA prediction model of skeletal muscle has the potential to be successfully applied to the study of skeletal muscle in GH disorders and during their therapy.

GH and IGF-I have important effects on body composition. With regard to skeletal muscle (SM), GH, and IGF-I together are anabolic and promote protein synthesis. When these hormones are deficient, as in adult GH deficiency (GHD), SM mass is reduced but increases with GH replacement. However, SM mass has not been convincingly shown to be enhanced when GH is given to non-GHD humans. In endogenous GH and IGF-I excess, the disease acromegaly, few direct data are available on SM mass. Because new potent therapies for acromegaly have the potential to induce a functional GHD in these patients (1), a greater understanding of SM mass under conditions of GH/IGF-I excess is needed. The ideal method for examining SM mass is total body magnetic resonance imaging (MRI), considered the reference method for assessing SM mass in vivo because it is precise, reliable, and ideally suited for measuring whole-body SM mass, its quantification, and distribution (2,3). Therefore, in the first part of our study, we examined, for the first time, SM mass in acromegaly by total-body MRI.

We also sought in the second part of our study to validate a model for the determination of SM from dual-energy x-ray absorptiometry (DXA) because DXA is less expensive and more accessible than MRI. Because DXA partitions body composition into fat, bone mineral, and lean tissue (Fig. 1), of which the latter is the combined mass of SM, organ and other soft tissues (including tissue water), DXA cannot alone directly assess SM mass. However, selected lean soft tissue estimates by DXA can be used to predict SM mass (4), but the validity of these models when applied in patients with acromegaly is presently unknown. Accordingly, we examined whether DXA could be used to accurately estimate SM mass in acromegaly and thus in disorders of GH secretion.

Figure 1.

MRI: ATFM, Adipose tissue-free mass. DXA: BMC, Bone mineral content; lean, total body lean tissue; ALT (includes SM and other soft tissues in arms and legs only). Other tissues include soft tissues and residuals. Adapted from Wang et al. Hydration of fat-free body mass: new physiological modeling approach. Am J Physiol 1999;27:6.

Subjects and Methods

Subjects

Acromegaly group

We studied 27 subjects with active acromegaly, 17 males and 10 females, of which 20 were Caucasian, six Hispanic, and one African-American, with a mean age 45.6 ± 8.06 yr and a mean body mass index of 30.8 ± 5.34 kg/m². Active disease was defined by a high serum IGF-I level, which had been elevated for at least 0.5–14 yr (mean 7.2 ± 3.97 yr) before the study. Levels of GH (immunoradiometric assay) were 8.7 ± 16 μg/liter (range 0.58–62 μg/liter) and IGF-I 635 ± 175 μg/liter (range 337–987) (Table 1). Twenty subjects had noncurative transsphenoidal surgery from 6 months to 14 yr previously (mean 4.2 ± 4.26 yr). Five had received prior radiotherapy (RT); two γ-knife RT (5 and 8 months previously) and two stereotactic fractionated RT (6 months, 9 yr previously). Six had prior medical therapy that did not normalize their IGF-I; three dopamine agonists, four long-acting octreotide, one short-acting octreotide and two pegvisomant, which were last taken a mean of 12.2 months (range 2–36 months) before the study. Seven had no therapy before the study and subsequently had surgery. All 27 had pathological confirmation of a GH-secreting pituitary tumor. Eight male subjects had secondary hypogonadism; five were on stable replacement doses of testosterone for more than 1 yr prior and three had untreated mildly low testosterone levels. Seven females had regular menses and three were postmenopausal and not on hormone replacement therapy. One subject had secondary adrenal and thyroid insufficiency and was on stable replacement doses of prednisone (5 mg/d) and Synthroid (100 μg/d). Two had type 2 diabetes mellitus treated with oral hypoglycemic agents and one had type 1 diabetes mellitus treated with an insulin pump; glycosylated hemoglobin levels were between 6 and 7% at the time of study. All were ambulatory outpatients with normal renal function and no liver disease.

Table 1.

Acromegaly subjects characteristics

Subject no.	Age/gender	Prior therapy	IGF-I (μg/liter)a	GH (μg/liter)b	BMI	W/H	SM (MRI) (kg)	Predicted SM by DXA (kg)c	Lean tissue (DXA) (kg)
1	48/M	S	500	1.8	26.1	0.997	34.32	34.69	68.79
2	55/M	S/C/P	580	1.2	34.1	0.972	37.96	32.76	68.35
3	37/M	S	551	2.2	30.6	0.900	43.47	41.71	80.28
4	43/M	S/RT	725	10.0	24.6	0.818	34.22	35.78	69.77
5	57/M	S/RT	855	3.7	30.0	0.840	26.94
6	40/M	S/SA	518	4.8	30.5	0.903	37.23	33.23	66.52
7	46/M	S/SA	387	2.3	40.8	1.053	38.48	38.30	82.97
8	35/M		799	62.0	32.0	0.843	43.58	45.18	87.76
9	47/M		646	3.2	29.5	0.910	41.51	39.89	83.61
10	43/M	S	337	0.7	30.2	0.967	33.97	32.91	73.04
11	37/M	S/RT	660	3.5	35.9	0.893	40.60	36.36	76.18
12	44/M	S	561	2.6	25.0	0.888	36.45	40.69	83.44
13	49/F	S/C/P	468	1.5	37.5	0.947	28.91	28.84	59.37
14	49/M	S/SA	508	1.8	29.5	0.97	34.61	34.59	69.86
15	56/M	S/BC	828	8	28.5	0.988	27.77	26.17	60.63
16	50/M		505	2.4	28.0	0.951	34.51	35.89	69.58
17	35/M		906	42	26.5	0.8477	40.04	41.51	73.81
18	44/M	S	937	7.1	36.9	1.088	43.39	40.36	88.23
19	51/F	S	572	1.9	24.7	0.818	20.80	20.06	48.10
20	40/F	S	671	3.3	28.2	0.783	23.61	21.27	45.21
21	50/F		987	3.8	31.2	0.810	24.30	24.12	49.86
22	50/F	S/C	486	2.7	23.2	0.867	18.82	17.41	39.57
23	41/F		829	52.0	23.9	0.651	23.40	25.12	51.46
24	68/F	S	540	5.6	32.0	0.907	14.89
25	30/F	S	792	2.2	43.5	0.950	34.47	37.49	68.08
26	42/F	S/RT	517	2	40.0	1.0	18.57	19.28	41.40
27	47/F		499	0.58	29.5	0.8918	20.48	20.08	42.10
	46 ± 1.67 17 M/10 F		635 ± 175	7.7 ± 3.1d

S, Transsphenoidal surgery; C, cabergoline; BC, bromocriptine; SA, somatostatin analog; P, pegvisomant; BMI, body mass index; W/H, waist to hip ratio. 

^aNormal IGF-I ranges: ages 26–30 yr, 117–329 ng/ml; 31–35 yr, 115–307 ng/ml; 36–40 yr, 109–284 ng/ml; 41–45 yr, 101–267 ng/ml; 46–50 yr, 94–252 ng/ml; 51–55 yr, 87–238 ng/ml; 56–60 yr, 81–225 ng/ml; 61–65 yr, 75–212 ng/ml; 66–70 yr, 69–200 ng/ml. 

^bNadir values after 100 g oral glucose in all patients except no. 3 and 24. 

^cSM predicted from DXA (see text). ALT is equivalent to ALST (Ref. 4). The equations are: males = 1.1932 + 0.0115 ALT (kilograms) − 0.0034 age (years) + 0.0001 ALT (kilograms) age (years) + 0.035 (African-American) − 0.0468 (Asian) − 0.0015 (African-American) ALT (kilograms) + 0.0011 (Asian) ALT; females = 1.1932 + 0.0115 ALT (kilograms) − 0.0034 age (years) + 0.0001 ALT age − 0.1866 + 0.0063 ALT + 0.0007 age (years) + 0.035 (African-American) − 0.0468 (Asian) − 0.0015 (African-American) ALT + 0.0011 ALT (Asian). 

^dMean ± se for acromegaly group. 

Nonacromegaly comparison group

A group of 185 females and 130 males, aged 18–84 yr, of different ethnicities were studied to develop a SM mass prediction model for comparisons to the acromegaly group. A subset of these subjects was also matched by gender, weight, and age in a ratio of 3–4:1 acromegaly subject for a secondary analysis comparing SM (MRI) and lean tissue mass (DXA). Subjects were ambulatory, nonsmoking, weight stable (±2 kg over the prior 6 months), and not heavy exercisers. Those with a history of untreated diabetes mellitus, malignant/catabolic conditions, or taking medications that could potentially influence body composition were excluded.

The study was approved by the Institutional Review boards of Columbia University Medical Center and St. Luke’s-Roosevelt Hospital Center. All subjects gave written informed consent before participation.

Study design

The 27 subjects with acromegaly underwent laboratory and body composition testing including anthropometric (n = 27), MRI (n = 27), and DXA (n = 25) studies. Control subjects underwent body composition testing.

Study data were analyzed in two parts. Part I compared the acromegaly subjects’ measured SM with those in the nonacromegaly population including to values predicted by a regression model. Part II of the analysis consisted of a methodological comparison of MRI determined SM mass to that predicted by DXA.

Laboratory evaluations

Subjects with acromegaly underwent blood sampling fasting and at 60, 90, and 120 min after a 100-g oral glucose tolerance test. Serum was frozen at −80 C in multiple aliquots. Fasting samples were assayed for IGF-I and all time points for GH. Each subjects’ samples were run in the same assay and in duplicate. On separate days, within 2 wk, subjects underwent laboratory and body composition testing.

Body composition evaluations

Anthropometric measurements

Body weight was measured with a digital scale to the nearest 0.01 kg and height with a stadiometer to the nearest 0.5 cm (Holtain Stadiometer, Crosswell, Wales). Skinfold thickness was measured on the right side of the body to the nearest 0.1 mm with a Lange caliper (Country Technology Inc., Gay Mills, WI) (5).

MRI

Total and regional body SM volumes were measured by whole-body multislice MRI on a 1.5 T scanner (6X Horizon; General Electric, Indianapolis, IN) in all comparison group subjects and in 16 acromegaly subjects and on a Philips 1.5 T Gyroscan (Philips Medical, Andover, MA) in the 11 remaining acromegaly subjects. Subjects were placed on the MRI platform with their arms extended above their heads, and about 40 axial images of 10 mm thickness at 40 mm intervals from head to toe were acquired. SM volumes reported in this study are those free of intermuscular adipose tissue in which intermuscular adipose tissue is defined as the adipose tissue (AT) located between muscle groups and beneath the muscle fascia as previously described (6,7). Adipose tissue volumes were also measured and were previously reported in 24 of these patients (8). Images were analyzed with SliceOmatic image analysis software (TomoVision) in the Image Reading Center of the New York Obesity Research Center. MRI volume estimates were converted to mass using the assumed density of 1.04 kg/liter for SM. The coefficient of variation for repeated measurements of the same scan by the same observer of MRI-derived SM volumes was 1.4% (4).

DXA

Whole-body and regional body composition were estimated by DXA (software version 11.4; Lunar DPX, Madison WI). The mass of lean tissue, fat, and bone mineral for the whole body and specific regions was provided by the software. Appendicular lean tissue mass (ALT) was considered equivalent to the sum of lean tissue in both the right and left arms and legs. The coefficient of variation of repeated daily measurements was 1.7% for leg lean tissue, 2.0% for arm lean tissue, and 2.6% for appendicular lean tissue (4).

Part I: SM prediction model development procedure

Prediction equations for the mass of total body SM were developed using generalized linear models from the comparison group data. Prediction models for SM mass were adjusted for factors known to influence SM including age, height, race, gender, and weight. For model development, the comparison group was randomly separated into two groups, model development (two thirds of subjects) and cross-validation (one third of subjects) whose characteristics were similar (Table 2). For the multiple regression analysis, the MRI-measured SM compartment was the dependent variable, gender was a fixed factor and age, body weight, height, and ethnicity were included as covariates. All main effects for covariates and possible two-way interactions were investigated. Covariates that contributed significantly to the model were initially included to find the best-fitting model with the lowest se. The developed models were then validated by the leave-one-out method (9). The prediction models were validated in the cross-validation group. SM values for each subject in the model cross-validation group were calculated using the developed prediction equations. Observed differences between estimated and actual SM mass were tested for significance by Student’s t tests, and the level of agreement was assessed by the Bland and Altman method (10). Predicted values (validation group) and measured values (development group) did not differ significantly (Table 2). Correlation coefficients of the mean measured and predicted SM mass with the difference between them were for females (SM r = 0.092, P = 0.38) and males (SM r = 0.19, P = 0.11), demonstrating good agreement between the models. The equations developed are shown in Table 2.

Table 2.

Characteristics of nonacromegaly subjects in model development and validation groups

Group characteristics	Model development		Model validation
	Women	Men	Women	Men
Sample no.	124	87	61	43
Age, yr	45.5 ± 17.2	36.6 ± 14.9	45.2 ± 16.4	40.0 ± 15.2
Weight, kg	68.7 ± 14.3	80.0 ± 10.2	66.5 ± 14.4	84 ± 12.3
Height, cm	162 ± 7.3	176 ± 6.8	162 ± 7.2	178 ± 6.7
SM, kg (actual)	20.6 ± 2.9	33 ± 4.9	20.7 ± 3.6	34.5 ± 5.4
SM, kg (predicted)			21.3 ± 2.7a	35.8 ± 4.8b
Developed models for SM
Female, SM (kg)	−18.35 + (0.034) age + (0.482) wt + (0.120) height − (0.195) ethnicity − (0.002) wt2 − (0.001) wt age
Males, SM (kg)	−30.276 − (0.158) age + (1.343) wt + (0.013) ht −(0.006) wt2

Models were developed for prediction of SM in male and female patients with active acromegaly. 

^aP = 0.14 vs. actual value. 

^bP = 0.11 vs. actual value; age, years; Ht, height (centimeters); wt, weight (kilograms). Ethnicity: Asian, one; African American, two; H, Hispanic, three; other, four; white, five. 

Hormone assays

GH was measured by a two-site immunoradiometric assay (Diagnostic Systems Laboratories, Webster, TX). The standards contain 22K recombinant human GH calibrated to the World Health Organization International Reference Preparation of human GH 88/624. The intraassay coefficient of variation is 3.1% and the interassay coefficient of variation is 5.9%. Assay sensitivity in our laboratory is 0.05 μg/liter.

IGF-I was measured by chemiluminescent immunometric assay (Immulite; Diagnostic Products Corp., Los Angeles, CA). The standard is calibrated against World Health Organization first IRP 1988, IGF-I 87/518. The normal ranges are: ages 26–30 yr, 117–329 ng/ml; 31–35 yr, 115–307 ng/ml; 36–40 yr, 109–284 ng/ml; 41–45 yr, 101–267 ng/ml; 46–50 yr, 94–252 ng/ml; 51–55 yr, 87–238 ng/ml; 56–60 yr, 81–225 ng/ml; 61–65 yr, 75–212 ng/ml; and 66–70 yr, 69–200 ng/ml.

Statistical methods

Part I

Each acromegaly subject’s SM mass was compared with their respective predicted values by paired t test. Secondarily, each acromegaly subjects’ SM and total body lean tissue mass (DXA) (referred to as lean tissue mass) were compared with the mean of three to four controls matched for gender, age ± 5 yr, and weight ± 3 kg by paired t test. Each acromegaly subjects’ SM mass was also compared with values predicted for an adjusted weight (i.e. weight minus increase in lean mass) using otherwise the same prediction equation in Table 2 by paired t test.

Part II

The following calculations were made: fat-free mass (FFM) (DXA) (kilograms) = weight (kilograms) − total fat (DXA) (kilograms); adipose tissue free mass (MRI) (kilograms) = weight (kilograms) − total AT (MRI) (kilograms); ALT (DXA) (kilograms) = sum of lean tissue in arms and legs measured by DXA (kilograms). SM was predicted from DXA ALT measurements by published prediction equations (Table 1) and compared with subjects’ actual SM by paired t test. Pearson’s correlation tests were used to assess the relationships between the body composition measurements. Data are given as mean ± sd unless stated otherwise. P < 0.05 was considered significant.

Results

Part I: comparison of SM in acromegaly with predicted and nonacromegaly subjects

MRI-measured SM masses are shown in Table 1. Acromegaly subjects’ SM mass did not differ from predicted values in all combined [31.75 ± 8.6 (acromegaly) vs. 33.06 ± 8.9 kg (predicted) (P = 0.088)], males [37.0 ± 4.9 vs. 38.34 ± 3.4 kg (P = 0.142)], or females [22.83 ± 5.6 vs. 24.07 ± 2.4 kg (P = 0.397)] considered separately. On average, SM was 95.6 ± 12.8% of predicted values (range 66.7–122%) (Fig. 2). In males, SM did not differ significantly from predicted when groups with different gonadal steroid status were considered separately.

Figure 2.

Measured and predicted SM mass in patients with acromegaly. There were no significant between-group differences between measured and predicted SM mass.

SM did not differ from matched controls in acromegaly subjects overall (P = 0.871) or males (P = 0.821) or females (P = 0.577). Lean tissue mass (DXA) was higher in acromegaly subjects than matched controls overall [65.91 ± 15.2 vs. 58.73 ± 13.5 kg (P < 0.0001)] and in males [75.17 ± 8.2 vs. 67.7 ± 7.5 kg (P = 0.0004)] but only somewhat higher in females [49.46 ± 9.3 kg vs. 43.46 ± 2.7 kg (P = 0.0791)]. The difference between lean tissue mass (DXA) and SM in acromegaly patients was significantly higher than that difference in matched controls (P < 0.0001), consistent with an enlarged non-SM lean compartment in acromegaly patients. The nonlean SM compartment (Fig. 1) includes soft tissues (including tissue water) and organs.

SM mass is acromegaly subjects were also compared with values predicted from a weight that was adjusted down by the amount of increase (if present) in lean tissue over controls, as a crude estimation of preacromegaly weight. Measured SM mass did not differ from weight-adjusted predicted values in all subjects combined [31.75 ± 8.6 (acromegaly) vs. 32.49 ± 7.5 kg (predicted) (P = 0.36)] or in male [37.0 ± 4.9 vs. 36.9 ± 4.2 kg (P = 0.30)] or female [22.83 ± 5.6 vs. 23.0 ± 2.0 kg (P = 0.91)] subjects considered separately.

Part II: comparison of MRI and DXA measurements

Lean tissue mass measured by DXA correlated highly with SM mass estimates made by MRI (r = 0.97, P < 0.0001) and the ratio of SM (MRI) to lean (DXA) was 0.493 ± 0.035 (range 0.432–0.560), within the expected range. ALT (DXA) correlated highly with SM (MRI) (r = 0.976, P < 0.0001) (Fig. 3).

Figure 3.

ALT by DXA correlated highly with SM by MRI (r = 0.958, P < 0.001).

FFM by DXA correlated with adipose tissue free mass by MRI (r = 0.986, P < 0.0001) and SM by MRI (r = 0.835, P < 0.0001). The ratio of SM (MRI) to FFM (DXA) was 0.48 ± 0.082 (range 0.332–0.777).

Predicted SM (DXA) was calculated based on a published prediction equation in healthy subjects (1) (Table 1). Measured SM (MRI) (mean 31.25 ± 8.6 kg) and predicted SM (DXA) (mean 32.15 ± 8.2 kg) did not differ (P = 0.213). Predicted SM (DXA) correlated highly with SM (MRI) (r = 0.949, P < 0.0001). Bland and Altman analysis (10) of the level of agreement showed that the correlation coefficient of the mean of measured SM (MRI) and predicted SM mass (DXA) with the difference between them was r = 0.049 (P = 0.81), demonstrating good agreement (Fig. 4). SM (MRI) to SM (DXA predicted) ratio was 1.018 (range 0.896–1.16), also demonstrating high agreement between the DXA predicted and MRI measured SM masses.

Figure 4.

Bland-Altman plot of the relationship between SM measured by MRI and SM estimated by DXA in patients with acromegaly. The correlation coefficient of the mean of measured SM (MRI) and predicted SM mass (DXA) with the difference between them was r = 0.049 (P = 0.81), demonstrating good agreement.

Discussion

GH and IGF-I have important anabolic effects on SM. GH induces a positive nitrogen balance and protein synthesis (11). IGF-I increases protein synthesis as well as inhibits proteolysis and is important for SM formation (11). In rodent models, IGF-I receptor deficiency leads to SM hypoplasia and when IGF-I is overexpressed, SM mass increases and hypertrophies (11,12). Although debated, it is generally believed that the effects of GH on SM are completely mediated by IGF-I (11). The clinical pictures, either of GH deficiency or excess, which are accompanied by reduction or increase, respectively, of both GH and IGF-I should reflect the combined net effect of both hormones on SM. A number of studies have examined the effect of GH/IGF-I deficiency on SM. In hypophysectomized rodents (11) and GHD humans (13), protein synthesis is reduced and increases with GH administration (14). SM mass is decreased in patients with GHD (13) and physiological GH replacement increases lean body mass (by DXA) (15), SM mass (by MRI) (16), muscle strength (16,17), exercise capacity (17), and physical performance (18). Although muscle fiber size has been reported to increase (18), by P³¹ magnetic resonance spectroscopy no change in muscle energy stores or muscle fiber type distribution was detected with GH replacement (16). The beneficial effects of GH replacement on SM were shown to plateau at 16 months of therapy (17). The effect on SM of supraphysiological levels of GH and IGF-I, either due to exogenous or endogenous sources, has not been clearly established.

In part I of this study, we examined, for the first time, SM mass by MRI in patients with active acromegaly and found it not to differ from predicted values. Controversy exists as to whether lean body mass (LBM) or SM are increased in acromegaly. Only one prior study directly examined SM mass in acromegaly (19). In this study, the combined mass of SM and skin, as measured by computed tomography (CT) scan, decreased after surgical therapy, suggesting that their mass had been increased in active acromegaly (19). However, SM values were not compared with those of controls (19). Other studies, using a variety of models, indirectly estimated lean tissue mass in acromegaly. In acromegaly total-body nitrogen is increased (14,20), but both protein breakdown and synthesis are increased and protein oxidation is unchanged (21), which would in net favor no change in muscle mass. When examined by muscle biopsy, hypertrophy of type 1 and atrophy of type 2 muscle fibers (22) were found. Estimates of body cell mass (BCM), derived from those of total-body potassium (TBK), were increased in some (20,23) but similar to controls (24) in other acromegaly patients. BCM and LBM estimates derived from TBK and total body water (25) or bioelectrical impedance analysis (BIA) have been reported to decrease with treatment (26). The validity of BCM estimates derived from TBK in patients with acromegaly has been questioned because they rely on the assumption that the intracellular K concentration is unaltered in acromegaly when the relationship between BCM and K counting has been established only in normal subjects (27). LBM, as estimated by DXA, was increased in acromegaly patients (27), but in that report extracellular water, as determined by sodium dilution, was also increased, leading these authors to conclude that increased soft tissue mass hydration and not SM was the cause of increased LBM (27). Reductions in LBM estimated by DXA during octreotide treatment have also been attributed to reductions in soft tissue fluid (15). The apparent increases in LBM in acromegaly may be attributed to increases in extracellular protein (collagen) with little effect on tissue hyperplasia (27). We found increases in DXA-measured total body lean tissue in acromegaly patients, which are likely accounted for by the expected increases of soft tissues and organs, the major components other than SM of total body lean tissue mass as measured by DXA.

Interpretation of our results needs to consider that we compared SM in our acromegaly subjects with that expected for their weight; SM cannot be accurately predicted without factoring in weight. Thus, although our acromegaly subjects had no change in SM mass for their current weight, we cannot exclude that their weight had not increased over time due in part to an absolute increase in SM mass. This cannot be determined without the preacromegaly body composition. However, our results were unchanged, even when we performed our analyses using subjects’ weights that were adjusted down by lean tissue (DXA) increases. Interpretation of our results should also consider that although we did not find SM to differ from predicted overall in males or within subgroups with different gonadal steroid status, we cannot exclude that in some males hypogonadism prevented an increase in SM. In a prior study, some patients with active acromegaly had BCM measures lower than expected, which was suggested to possibly be due to coexistent gonadal insufficiency or reduced physical activity (23). Larger numbers of subjects should be studied to further examine the role of gonadal steroid status on SM mass in acromegaly.

Our data in acromegaly are in accord with data showing no clear effect of supraphysiological GH administration on SM mass or strength in non-GHD subjects, such as the elderly (28) or healthy volunteers (29). In young healthy men and women, supraphysiological GH for a month increased FFM (by DXA and BIA), but this was explained by increases in TBW and extracellular water (29). These authors (29) and others (30) concluded that the reported anabolic effect of GH should be questioned because despite the nitrogen-retaining properties of GH, available evidence shows that supraphysiological GH administration increases fluid retention but not muscle strength, mass, or function.

Elucidation of the true nature of SM mass changes in acromegaly has been somewhat hindered by the fact that the necessary direct measurement of SM by MRI or CT cannot practicably be applied on a large scale. We therefore, in part II of our study, assessed the validity of a model for predicting SM mass from DXA measurements of ALT in patients with acromegaly. Only one prior study compared SM measured by CT in acromegaly patients with estimates of FFM by BIA, TBW, and TBK (31), and these were found to correlate well with those obtained by CT (31). In our study, we have shown in a large group of patients with acromegaly that SM can be predicted with a high degree of accuracy from DXA ALT measurements. Our data suggest that the model can be successfully applied to SM assessment in patients with GH disorders. Investigation of possible changes in SM during therapy of acromegaly is warranted because although SM may not be altered in patients with active acromegaly, GHD is clearly associated with reduced SM, and therapy of acromegaly with new potent medical therapies can potentially induce a state of functional GHD. Further validation of this DXA prediction model across changes in GH status such as before and after therapy of acromegaly or GHD should be undertaken.

In conclusion, SM mass measured by MRI in acromegaly patients did not differ from predicted or control values. SM predicted from DXA ALT agreed highly with SM by MRI, supporting the validity of the developed DXA prediction equations in assessing SM in acromegaly. This DXA prediction model of SM has the potential to be successfully applied to the study of SM in GH disorders and during their therapy.