The number of individuals who sustain a fracture in the United States is increasing1, particularly among the elderly2-4. Elderly patients with a fracture have a higher rate of complications compared with their younger counterparts, as well as a profound decrease in quality of life, with an increase in morbidity, mortality, and economic costs5-10. In studies aimed at determining major factors contributing to poor outcomes among fracture patients, it has been recognized that as many as 59% of elderly patients with a hip fracture have coexistent malnutrition11-14. Malnutrition is known to contribute to increased infection and mortality rates, delayed wound-healing, and impaired immunity15-20.
Recent data have indicated that dietary protein supplementation decreases systemic complications, hospital stay, morbidity, and even mortality in fracture patients21-25. Seven amino acids have been considered “conditionally essential,” in that they may be rate-limiting under certain conditions and can act to replace essential and nonessential amino acids when stores are diminished. They are glutamine, arginine, cysteine, histidine, proline, taurine, and tyrosine. Conditionally essential amino acids are rate-limiting factors in the anabolism process during times of stress and malnutrition26-29. Two of the essential amino acids, arginine and lysine, have been reported to stimulate osteoblast proliferation, activation, and differentiation30.
Relatively few investigations have described the effects of nutritional support on fracture-healing. Protein malnutrition has consistently been reported to result in negative effects on the quality of fracture callus22,31-33. The effects of supplementary protein on fracture-healing in well-nourished animal models have been highly variable. One study has indicated that malnutrition reversal increases fracture callus strength31. The effects of conditionally essential amino acids on fracture-healing and the effects of varying amounts of dietary protein on the associated injured soft tissues are unknown.
We examined the effects of protein malnutrition, protein supplementation, and conditionally essential amino acids on muscle and bone-healing in an in vivo fracture model in rats. Our hypothesis was that dietary supplementation with conditionally essential amino acids would improve the course and quality of fracture and soft-tissue healing.
Materials and Methods
Prior to undertaking this project, all procedures were approved by the institution's Animal Care and Use Committee. One hundred adult male Sprague-Dawley rats were studied. The animals were housed for five days to acclimate to the environment before receiving any intervention. Ten animals served as controls and received a standard 15% protein diet throughout the eleven-week study. The remaining ninety rats received a 6% protein diet for five weeks to induce a state of protein malnutrition as previously reported by Day and DeHeer31. Serial venous samples were obtained from tail veins to assess serum albumin levels as a serologic marker of malnutrition12,14,34. After the five-week malnutrition phase, all ninety animals in the treatment group were anesthetized with ketamine and maintained by means of inhaled isoflurane in oxygen. They underwent femoral intramedullary nailing with a 0.045-in (1.14-mm) Kirschner wire with use of a retrograde approach through the femoral intercondylar notch. A standardized closed midshaft femoral fracture was created after intramedullary nailing with use of a modified version of the device described by Bonnarens and Einhorn (Fig. 1)35.
After the femora were fractured, the rats were randomly divided into three groups of thirty rats each. Group P6 received a diet with 6% protein; Group P15, a diet with 15% protein; and Group P30, a diet with 30% protein (Purina Mills, St. Louis, Missouri). One-half of the 30% protein diet consisted of generic protein, and the other half contained equal amounts of the conditionally essential amino acids glutamine, arginine, and taurine. All three diets were isocaloric and identical in mineral content.
The rats were allowed to bear weight as tolerated and to consume ad libitum. At two, four, and six weeks after surgery, ten animals from each group were killed with CO2 inhalation in a closed chamber.
Animal weights were subsequently obtained. The quadriceps muscles and femora were harvested bilaterally, and the muscle tissue was weighed. Muscle tissue was preserved by immersion in isopentane and was stored at -80°C for subsequent analyses. Myosin heavy-chain concentrations in each quadriceps muscle specimen from the group killed at two weeks were determined with use of an electrophoretic separation method followed by a densitometric analysis (Eastman Kodak, Rochester, New York). Frozen muscles were minced in ice-cold homogenization buffer (250 mM of sucrose, 100 mM of KCl, 5 mM of EDTA, and 20 mM of tris(hydroxymethyl)aminomethane (Tris), pH 6.8), homogenized by hand in glass tissue grinders, and boiled in sample buffer for two minutes at a final protein concentration of 0.125 mg/mL. One microgram of total protein assayed by the Bradford method was run on the gel as previously described36. Total protein content (gram of protein per gram of muscle) was determined for a section of quadriceps from each muscle specimen with use of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). Total RNA was extracted with use of the TRIzol reagent (Invitrogen, Carlsbad, California) according to the manufacturer's protocol. Total RNA (1 μg) was reverse transcribed to cDNA with use of the StrataScript reverse transcriptase enzyme (Stratagene, La Jolla, California) in a 20-μL reaction mix containing reaction buffer, dNTP (deoxyribonucleotide triphosphate), and oligo d(T) primers. Relative expression levels for rat-specific insulin-like growth factor-1 (IGF-1) and IGF-2 along with their corresponding receptors, actin, myosin (subtypes 1b, 2a, 2b, and 2d), and vascular endothelial growth factor (VEGF) mRNA were quantified with use of a SYBR green real-time polymerase chain reaction assay (Qiagen, Valencia, California) and the housekeeping gene β-actin as the internal standard. Polymerase chain reactions were performed with use of a Rotor-Gene RG-3000 (Corbett Research, Sydney, Australia), and counts were determined by the Rotor-Gene software. Relative levels of gene expression were determined with use of Q-Gene, and significant differences in gene expression were determined with use of the REST-XL (Relative Expression Software Tool)37,38. Muscle protein content and mRNA analysis were only obtained in the rats killed at two weeks because of specimen loss in a freezer mishap.
The harvested femora were examined with dual x-ray absorptiometry (QDR-2000, Hologic, Waltham, Massachusetts). The area of callus on the healed femora was isolated, and bone mass density was measured. In addition, the total femur, including callus, as well as the contralateral, noninjured femur, were measured. Five sets of femora from each diet group at each time-interval were assigned to biomechanical testing, and the other five sets of femora were sent for histomorphometric analysis. Prior to torsional testing, digital calipers were used to measure the maximum and minimum periosteal diameters (Dp) of the callus and shaft at the original fracture cross section and at the midshaft for the control and contralateral femora. The maximum and minimum endosteal diameters (De) were determined from photographs of the fractured ends. A torsional holder was machined for each femur. Its long axis was visually aligned within and centered in the holder's window. Devcon 5 Minute Epoxy (Devcon, Riviera Beach, Florida) was used to pot the femur ends into the holder. The holder was then placed in a torsional test fixture mounted on a TA-HDi testing machine (Stable Micro Systems, Surrey, United Kingdom). The steel struts between the holder ends were then severed with a cut-off wheel (Dremel Tools, Racine, Wisconsin). The lower end was then rotated about its axis at a constant speed (0.75 rad/sec) while the upper end was held from axial rotation but free to move otherwise. The computer software controlling the test machine acquired and stored its crosshead position, cable force (F), and sample time at 100 Hz. A digital camera (DSC-S75; Sony, Tokyo, Japan) was used to take photographs of each specimen before and after testing and to make an audiovisual recording of each test. Force was plotted as a function of crosshead position to observe the relative peak force (fracture) occurrences. The absolute maximum applied force (Fmax) was multiplied by the fixture shaft radius and cable radius to determine the corresponding applied torque (Tmax). Assuming a hollow elliptical cross section, we used the Tmax and the callus-shaft cross-section diameters to calculate the maximum shear stress occurring in the plane of the cross section at the ends of the minor periosteal diameter39. The formula used to determine maximum shear stress is τmax = Tmax/R (N/mm2), where q = De/Dp and r = [πDpd p2 (1-q4)]/16 (mm3). For pure torsional load on the femur (assumed to be composed of isotropic-homogeneous bone), maximum shear stress is equal to the maximum normal stress, which occurs perpendicular to a plane that is oriented 45° to the cross section (or to the long axis of the femur).
Nondecalcified sections of bone were cut and stained with toluidine blue. Each section was cut to include the entire callus and 1 cm of femoral diaphysis proximal and distal to the fracture site. Each section was captured with use of a standardized image area and dots per inch with digital image capture software (Photoshop; Adobe Systems, San Jose, California) and analyzed with use of image analysis software (Image-Pro; Media Cybernetics, Silver Spring, Maryland). The fracture callus circumference and cortical bone were outlined, and the area was calculated to obtain the total bone area (mm2). With use of a thresholding technique on toluidine blue-stained images, all bone formation in a defined area was used to calculate the amount of bone per unit area to determine the relative bone density (Fig. 2).
An a priori power analysis with use of a one-way analysis of variance sample-size test (Sigma Stat; Jandel Scientific, San Rafael, California) was performed. Bending strength (N/mm2) was used as the primary outcome variable. A minimum difference in means of 0.8 was set as the significant level, and a standard deviation of 0.3 was projected on the basis of previous work. The analysis revealed a minimum sample size of five animals per group to yield a power of 0.8 at p < 0.05. The Student t test was used to compare values for the control and malnourished rats. One-way analysis of variance was used to test for significant differences among treatment groups. Significance was set at p < 0.05.
Malnutrition was achieved by five weeks in all rats as indicated by the mean 41% decline in serum albumin levels (3.7 to 2.18 g/dL) and a 7% reduction in body weight of the protein-restricted animals compared with controls (p ≤ 0.001). The mean consumption (22 g/day) of each isocaloric diet was not significantly different among the groups, and food consumption patterns remained statistically similar across all four groups during the subsequent six-week study period (p > 0.23). At the conclusion of the six-week study period, the P30 group had higher levels of albumin than the P6 group (p < 0.02) but similar levels to the control and P15 groups (p > 0.31) (Fig. 3).
Two weeks after surgery, all three study groups had a significantly lower mean total body weight compared with the control group (p < 0.01). By four weeks after femoral nailing, the mean overall body weight in the P30 group was equivalent to that of the controls (p = 0.21), while the convalescent weight gain was significantly less in the P6 group (p = 0.0001) and the P15 group (p = 0.0019). At six weeks postoperatively, both the P15 and P30 protein diet groups demonstrated similar body weight compared with the control group (p > 0.31 and p > 0.84, respectively). However, the P6 group continued to be significantly lower than the P15, P30, and control groups with respect to body mass (p < 0.012) (Fig. 4).
At the end of the six-week study period, the mean quadriceps muscle weight from the fractured leg was significantly greater in the P30 diet group (4.90 g) compared with the P15 (4.23 g; p = 0.012) and P6 (3.92 g; p < 0.01) groups. Both the P15 and P30 groups had statistically similar muscle mass relative to the control animals at six weeks (p > 0.99 and p > 0.77, respectively), but the P6 rats still lagged in muscle mass recovery (p = 0.004). Interestingly, the noninjured thigh in the P30 group had a significantly greater muscle mass compared with that in all other groups, including the controls (p = 0.014), the P15 group (p = 0.0043), and the P6 group (p = 0.0017) (Fig. 5). There was no significant difference in quadriceps weight between fractured and uninjured limbs in the P30 group, whereas the fractured limb in both the P6 and P15 protein groups exhibited significantly less muscle mass compared with the contralateral limb (p = 0.04 and p = 0.012, respectively).
The total protein content of muscle in the fractured limb at two weeks was 23% greater in the P30 group compared with the P6 and P15 diet groups (p = 0.0026 and p = 0.016, respectively). There were no differences between the P15 and P6 groups with respect to total protein content at two weeks. The total protein content of muscle from the uninjured limb was 13% and 15% greater in the P30 group compared with the P6 and P15 diet groups, respectively (p = 0.0059 and p = 0.027). Again, no difference was detected between the P6 and P15 groups in terms of total protein content of the uninjured limb quadriceps muscle (Fig. 6). In addition to these macroscopic muscle differences, there was evidence of differences at the molecular level. Expression of mRNA for IGF-1 and IGF-2, IGF receptors, myosin, actin, and VEGF were all significantly decreased in the P30 group compared with all other groups (p < 0.045 for all). The P15 group showed significant decreases compared with the P6 group in mRNA expression of all of the measured proteins except actin (Table I). Myosin heavy-chain content at the two-week interval was significantly greater in the P30 group compared with the P6 group (p = 0.042) (Fig. 7).
Fracture Callus Structure and Strength
Bone mineral density of the fracture callus as determined by dual x-ray absorptiometry was 17% greater at six weeks in the P30 group (p = 0.0033) compared with corresponding bone windows in control animals, but it was not significantly different from the P6 or P15 groups. Interestingly, the uninjured limbs in the P30 group demonstrated bone mineral density that was statistically similar to the control rats by the four and six-week intervals (p = 0.07 and p = 0.81, respectively), whereas the bone mineral density in the P6 and P15 groups never became equivalent to the controls (p < 0.045 to p < 0.01). Comparable results were found on histomorphometric examination. The P30 group possessed 14% more bone mineral density in the fracture callus than the P6 group at six weeks (p < 0.05). The P6 group had significantly less relative bone density than all other groups at the two-week period (p < 0.001).
On biomechanical testing, there were no significant differences in maximum load to failure between each of the diet groups at the two, four, or six-week time-intervals (p > 0.18 for all). By the six-week testing interval, there were no differences between any of the diet groups and the controls (p > 0.43 for all). No differences were found in maximum shear stress between the three diet groups at each of the two, four, and six-week time-intervals (p > 0.060 for all). Testing of the contralateral, uninjured femora revealed no consistently significant differences or trends.
In this malnourished rat model, dietary supplementation with conditionally essential amino acids improved some aspects of recovery and healing of the bone and soft tissues after femoral fracture, although no biomechanical improvements were demonstrated. The animals in the supplemented protein diet group were able to restore body mass and serum albumin better than those with a “normal” or protein-deficient diet. They were better able to restore muscle mass and protein content in the injured limb and to form callus with increased mineral content. The levels of mRNA for important proteins were lower in the high-protein diet group, suggesting that correction of protein malnutrition affects gene transcription. Improvements in the strength of bone-healing, as judged with biomechanical testing, however, were not demonstrated in this model.
Dietary protein supplies the body with amino acids, which serve as the building blocks for the proteins necessary for healing after injury. Negative nitrogen balance due to decreased intake, loss of amino acid reserves through soft-tissue damage, increased metabolic expenditures, and increased urinary excretion of amino acids compromises the ability to heal efficiently after trauma16,20,40. There is abundant research demonstrating the detrimental effects of malnutrition on the healing environment of fractures in orthopaedic trauma patients. However, little data are available on the potential for nutritional support to ameliorate the effects of malnutrition on fractures and their surrounding soft tissues.
In this study, we created a stabilized femoral fracture in a malnourished rat model, as demonstrated by decreased serum albumin levels and reduced body weight. We showed that protein supplementation with conditionally essential amino acids resulted in reversal of these parameters of malnutrition more rapidly than with usual dietary protein levels. Not surprisingly, rats with continued dietary deficiencies in protein content continued to show signs of malnutrition throughout the study.
In addition to reduced body weight and decreased serum albumin, the P6 group, which had a continuing dietary protein deficit after fracture, showed a delayed and depressed healing response as evidenced by the lower quadriceps muscle mass, decreased muscle protein content, decreased myosin heavy-chain content, and reduced callus mineral density compared with the P15 and P30 groups.
Messenger RNA levels for muscle protein and growth factors in the injured limbs at two weeks showed a stepwise progression in accordance with dietary protein levels. We studied insulin-like growth factor 1 (IGF-1 and IGF-2) and its receptors, which are important in skeletal growth through cellular proliferation and differentiation of periosteal cells, osteoblasts, and chondrocytes41-43 actin and myosin, which are structural proteins for muscle, and vascular endothelial growth factor (VEGF), which is expressed during fracture-healing and skeletal muscle regeneration and promotes neoangiogenesis43-45. The significant step-wise increase in mRNA for these growth factors, matrix molecules, and receptors in the lower protein diet groups, occurring in a dose response manner, is interesting and the cause is uncertain. It may reflect an attempt to compensate for an inadequate healing response. The increase in mRNA combined with the decreased myosin heavy chain, total protein content, and muscle mass suggest that the lack of functional proteins present for healing may work by means of a feedback mechanism to induce upregulation of gene expression for these proteins at the time-point of measurement in this study. This conjecture is supported by the data from the P30 group that showed higher myosin heavy chain, total protein, and muscle mass with lower gene expression levels, suggesting a negative feedback mechanism in this group. On the basis of this speculation, we would expect gene expression for the growth factors, matrix molecules, and receptors to be higher in the P30 group at the early time-points of initial healing, and this may form the basis for further study.
The P15 group also recovered from malnutrition by six weeks. However, there were several differences between the P15 group and the P30 group. The P30 group regained control levels of total body weight earlier, gained significantly more quadriceps muscle weight in both the injured and un-injured legs, and had higher total protein content in the quadriceps muscle at two weeks, and lower mRNA levels. The P30 group had the most relative bone mineral density at all time-intervals. The P30 group had quicker and more complete recovery from the effects of both malnutrition and fracture compared with animals fed a so-called normal protein diet, as well as compared with animals that were continuously malnourished. The significant improvement in quadriceps muscle mass in the contralateral uninjured limb of the P30 group suggests a more generalized anabolic effect of the dietary supplementation. Amino acids are known to be secretagogues for pituitary growth hormone, which has been used to accelerate fracture-healing in animals46. A hormonal response is a potential mechanism for the anabolic effects on soft tissue remote to the fracture. This systemic effect has important implications for the injured patient, but further study is needed.
Our research design does not allow us to clearly distinguish whether the beneficial effects we observed were due to the total amount of protein in the P30 diet or to the specific content of conditionally essential amino acids. Previous investigation has suggested that the simple addition of standard composition protein to the diet does not improve fracture-healing47. Einhorn et al. performed a biomechanical study of the healing of experimental fractures and found that “the supplementation of dietary protein or mineral in excess of the calculated requirements neither improved nor impaired fracture-healing.”48 On the basis of this information, we chose to use protein supplementation consisting entirely of conditionally essential amino acids for our P30 group, in order to optimize the chances of seeing a positive effect. The role of specific amino acids as potential metabolic response modulators to stress has received increased attention20,26,27,29. We selected the three most abundant and well-studied amino acids for this experiment. It may be worth noting that approximately 25% of the amino acids in bone morphogenetic protein-2 (BMP-2), a growth factor in clinical use to improve bone-healing, are in the conditionally essential category.
Our study does have some limitations. We were not able to assess overall limb function well. We were unable to evaluate any parameter that might translate into a patient-oriented outcome, such as pain or the ability to walk. Our biomechanical data did not demonstrate significant differences, which may be due to several factors, including the accelerated healing time-frame in rats, the time between measurement intervals, and the high variability between animals. Because of this limitation, the actual importance of the histological and chemical differences between the groups is unclear.
As with any animal study, extrapolation to human patients is hazardous. Obviously, different species have evolved different dietary requirements and responses. The use of previously healthy adult rats in this study may limit the implications of the findings for human subpopulations, for example, the elderly patients discussed in our introduction.
In conclusion, dietary supplementation with conditionally essential amino acids led to increased muscle mass in the body and leg, higher muscle protein content in the leg, and greater bone mineral density in the fracture callus in protein-malnourished rats with stabilized femoral fractures. Although caution is required in extrapolating the physiologic responses in rodents to patients, these findings suggest that dietary supplementation with specific proteins or amino acids in patients with a fracture could lead to quicker recovery and more effective rehabilitation. Clinical studies are indicated. ▪
In support of their research for or preparation of this manuscript, one or more of the authors received grants or outside funding from the Orthopaedic Trauma Association. None of the authors received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.
Investigation performed at the Comparative Orthopaedic Laboratory, Department of Orthopaedics, University of Missouri, Columbia, Missouri
1. , Furner S, Rice DP. Musculoskeletal conditions in the United States. 2nd ed. Rosemont, IL: American Academy of Orthopaedic Surgeons; 1999.
2. . Hip fractures: a worldwide problem today and tomorrow. Bone. 1993;14 Suppl 1: S1-8.
3. . Prevention of hip fractures. Am J Med. 1993;95: 75S-78S.
4. , Melton LJ 3rd. The worldwide problem of osteoporosis: insights afforded by epidemiology. Bone. 1995;17(5 Suppl): 505S-511S.
5. , Autier P, Barette M, Vanderschueren D, Lips P, Haentjens P. Functional outcome and quality of life following hip fracture in elderly women: a prospective controlled study. Osteoporos Int. 2004;15: 87-94.
6. , Lee EJ, Christmas C, Wenz JF. The economic impact of medical complications in geriatric patients with hip fracture. Orthopedics. 2003;26: 49-53.
7. , Aharonoff GB, Zuckerman JD, Koval KJ. Mortality risk after hip fracture. J Orthop Trauma. 2003;17: 53-6.
8. , Cameron ID, Cumming RG, Easter S, Seymour J, Kurrle SE, Quine S. Quality of life related to fear of falling and hip fracture in older women: a time trade off study. BMJ. 2000;320: 341-6.
9. , Goeree R, Papaioannou A, Adachi JD, Papadimitropoulos E. Economic implications of hip fracture: health service use, institutional care and cost in Canada. Osteoporos Int. 2001;12: 271-8.
10. , Keinanen-Kiukaaniemi S, Jalovaara P. Mortality and quality of life after trochanteric hip fracture. Public Health. 2001;115: 323-7.
11. , Blackburn GL, Vitale J, Cochran D, Naylor J. Prevalence of malnutrition in general medical patients. JAMA. 1976;235: 1567-70.
12. , Jensen TG, Smith TK, Johnston DA, Dudrick SJ. Nutrition in orthopaedic surgery. J Bone Joint Surg Am. 1982;64: 1263-72.
13. , Maurer SG, Su ET, Aharonoff GB, Zuckerman JD. The effects of nutritional status on outcome after hip fracture. J Orthop Trauma. 1999;13: 164-9.
14. , Cornell CN, Carbone B, Levine B, Chapman D. Protein depletion and metabolic stress in elderly patients who have a fracture of the hip. J Bone Joint Surg Am. 1992;74: 251-60.
15. , Ammann P, Chevalley T, Rizzoli R. Protein intake and bone growth. Can J Appl Physiol. 2001;26 Suppl: S153-66.
16. , Souba WW. Amino acids in surgical nutrition. Principles and practice. Surg Clin North Am. 1991;71: 459-76.
17. , Heppenstall RB, Friedenberg ZB, Hozack WJ. A prospective assessment of nutritional status and complications in patients with fractures of the hip. J Orthop Trauma. 1990;4: 49-57.
18. , Hill GL. Impaired wound healing in surgical patients with varying degrees of malnutrition. JPEN J Parenter Enteral Nutr. 1986;10: 550-4.
19. , Dionigi R, Comodi I, Dionigi P, Campani M. Preoperative prediction and quantification of septic risk caused by malnutrition. Arch Surg. 1982;117: 266-74.
20. , Kobayashi H, Sheffield-Moore M, Mittendorfer B, Wolfe RR. Essential amino acids are primarily responsible for the amino acid stimulation of muscle protein anabolism in healthy elderly adults. Am J Clin Nutr. 2003;78: 250-8.
21. , Handoll HH. Nutritional supplementation for hip fracture aftercare in the elderly. Cochrane Database Syst Rev. 2004;1: CD001880.
22. , Handoll HH. A systematic review of protein and energy supplementation for hip fracture aftercare in older people. Eur J Clin Nutr. 2003;57: 895-903.
23. , Rapin CH, Bengoa JM, Delmas PD, Vasey H, Bonjour JP. Dietary supplementation in elderly patients with fractured neck of the femur. Lancet. 1990;335: 1013-6.
24. , Guyer H, Diaz-Escriu F, Mellado-Navas J, Castells M, Pladevall M. Nutritional supplementation of elderly hip fracture patients. A randomized,double-blind, placebo-controlled trial. Age Ageing. 2000;29: 425-31.
25. , Nelson CL, Bopp MM, Puskarich-May CL, Walls RC. Nightly enteral nutrition support of elderly hip fracture patients: a phase I trial. J Am Coll Nutr. 1998;17: 155-61.
26. , Cynober LA. Amino acids with anabolic properties. Curr Opin Clin Nutr Metab Care. 1998;1: 263-72.
27. , Haymond MW, Darmaun D. Effect of glutamine on leucine metabolism in humans. Am J Physiol. 1996;271: E748-54.
28. , Wilmore DW. Is glutamine a conditionally essential amino acid? Nutr Rev. 1990;48: 297-309.
29. , van de Poll MC, van Gemert WG, Dejong CH. Amino acid adequacy in pathophysiological states. J Nutr. 2004;134(6 Suppl): 1575S-1582S.
30. , Fini M, Giavaresi G, Giardino R. Human osteopenic bone-derived osteoblasts: essential amino acids treatment effects. Artif Cells Blood Substit Immobil Biotechnol. 2003;31: 35-46.
31. , DeHeer DH. Reversal of the detrimental effects of chronic protein malnutrition on long bone fracture healing. J Orthop Trauma. 2001;15: 47-53.
32. , Aldini NN, Cane V, Zaffe D, Giavaresi G, Rocca M, Guzzardella GA, Giardino R. Effects of essential amino acids and lactose on bony fractures and defects in rabbits: a preliminary histomorphometric study. Arch Orthop Trauma Surg. 1999;119: 39-45.
33. , Floman Y, Simkin A, Avinezer A, Freund HR. The effect of protein malnutrition and nutritional support on the mechanical properties of fracture healing in the injured rat. JPEN J Parenter Enteral Nutr. 1986;10: 564-7.
34. , Tolo VT, McAfee PC, Burest P. Nutritional deficiencies after staged anterior and posterior spinal reconstructive surgery. Clin Orthop Relat Res. 1988;234: 5-11.
35. , Einhorn TA. Production of a standard closed fracture in laboratory animal bone. J Orthop Res. 1984;2: 97-101.
36. , Roy RR. Electrophoretic separation of rat skeletal muscle myosin heavy-chain isoforms. J Appl Physiol. 1993;75: 2337-40.
37. , Janovjak H, Miserez AR, Dobbie Z. Processing of gene expression data generated by quantitative real-time RT-PCR. Biotechniques. 2002;32: 1372-4, 1376, 1378-9. Erratum in: Biotechniques. 2002;33:514.
38. , Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 2002;30: e36.
39. , Young WC. Formulas for stress and strain. 5th ed. New York: McGraw-Hill; 1975. p 292.
40. , Young DH, Ramias L, Schiller WR. Effect of major trauma on plasma free amino acid concentrations in geriatric patients. Am J Clin Nutr. 1990;51: 1040-5.
41. , Zoidis E, Zapf J, Hassa P, Hottiger MO, Auer JA, Schneider R, Gander B, Luginbeuhl V, Bettschart-Wolfisberger R, Illi OE, Merkle HP, von Rechenberg B. Localized insulin-like growth factor I delivery to enhance new bone formation. Bone. 2003;33: 660-72.
42. , Jingushi S, Ikenoue T, Urabe K, Sakai H, Iwamoto Y. Expression of parathyroid hormone-related peptide and insulin-like growth factor I during rat fracture healing. J Orthop Res. 2003;21: 511-20.
43. , Visger JM, Balian G, Hurwitz SR, Diduch DR. Systemically administered mesenchymal stromal cells transduced with insulin-like growth factor-I localize to a fracture site and potentiate healing. J Orthop Trauma. 2002;16: 651-9.
44. , Zacchigna S, Zentilin L, Ramirez-Correa G, Pattarini L, Salvi A, Sinagra G, Giacca M. Vascular endothelial growth factor stimulates skeletal muscle regeneration in vivo. Mol Ther. 2004;10: 844-54.
45. , Bao M, de Guzman L, Bunting S, Peale FV Jr, Ferrara N, Steinmetz H, Hoeffel J, Cleland JL, Daughterty A, van Bruggen N, Redmond HP, Carano RA, Filvaroff EH. Vascular endothelial growth factor stimulates bone repair by promoting angiogenesis and bone turnover. Proc Natl Acad Sci USA. 2002;99: 9656-61.
46. , Gatti R, Cappellin E, Schiraldi C, De Palo CB, Spinella P. Plasma lactate, GH and GH-binding protein levels in exercise following BCAA supplementation in athletes. Amino Acids. 2001;20: 1-11.
47. , Oxlund H. Local anabolic effects of growth hormone on intact bone and healing fractures in rats. Calcif Tissue Int. 2003;73: 258-64.
48. , Bonnarens F, Burstein AH. The contributions of dietary protein and mineral to the healing of experimental fractures. A biomechanical study. J Bone Joint Surg Am. 1986;68: 1389-95.