This study was approved by Uludag University Animal Experiments Local Ethics Committee and was carried out on 20 New Zealand white rabbits, aged 3 months, weighing between 2.5 and 3.0 kg, obtained from the Experimental Animal Breeding and Research Centre with the permission of the Deanery of Uludag University Medical Faculty.

In the first week of the study, two rabbits in the immunonutrition group aspirated the dipeptiven solution given by gastric catheter. One of the rabbits died within minutes of aspiration and the other on the following day. In the control group, in the first week, one rabbit trapped its right lower extremity in the cage while it was being cleaned and sustained a femoral fracture. A pelvipedal plaster was applied but the rabbit died 3 days later. In the second week, one rabbit in the control group was found dead in its cage of unknown causes. The study continued with eight rabbits in the immunonutrition group and eight rabbits in the control group. Throughout the study no surgical site infection developed in any of the rabbits. Apart from the rabbit that sustained a femoral fracture, no functional problems were observed in any rabbit other than the created fibula fracture.

In recent years several scientific studies have focussed on glutamine as a non-essential amino acid as it has been determined to have significant physiological roles. Glutamine facilitates the process of rapid cell reproduction as it is a leading source of both energy and biosynthesis (13, 14). Particularly in critical cases such as multiple injuries, sepsis, or burns, glutamine is used as an essential amino acid as it undertakes the precursor task for nucleic acids, nucleotides, amino sugars, and proteins (3). Under catabolic conditions, glutamine takes on the aspects of an amino acid (3, 15). When muscle tissue is exposed to excessive stress, the most important of the amino acids which provide a repository of some amino acids, is glutamine (16). A fall in the level of glutamine is associated with poor clinical results, as starting with a decreased amount in the process of muscle destruction, morbidity and mortality rates may increase. Parenteral or enteral administration of glutamine to critical patients leads to decreased complications associated with sepsis and better functional results (17, 18). Glutamine has the essential properties of an amino acid in respect to affecting lymphocyte proliferation in the most appropriate way by neutrophil and macrophage functions 19–(22) . By eliminating a lack of glutamine, it is possible to restore nitrogen balance and improve immunosuppression (17).

To the best of our knowledge there are no studies on the effect of orally administered glutamine on fracture healing. In this study, taking various beneficial effects into consideration, it was thought that glutamine could make a positive contribution to fracture healing, and in this context, the similarity of wound healing to fracture healing was the guide.

As a single indeterminate structure, glutamine quickly becomes denatured in water. The dipeptide structure formed when glutamine is combined with alanine may remain stable in aqueous solutions. L-amino acids are more easily absorbed than d-amino acids (16). Therefore, in this study an l-alanine l-glutamine solution was used to achieve absorption with active transport of the glutamine and to prevent rapid denaturing.

In an experimental study of 50 rabbits with fibula fracture and femoral condyle defects, and orally nourished with essential amino acids containing lactose, Fini et al. (23) reported a positive effect on fracture healing. Although the current study investigated the effect on rabbit fibula fracture of glutamine as a different essential amino acid, both studies showed a similarity in respect to the chosen animal model and the choice of amino acid given. In the fracture model created in the rabbits, cartilage islands started to form on the sixth day of fracture healing, on the 12th day the centre was filled with cartilage tissue, and in the third and fourth week bone tissue replaced the cartilage showing complete union. Thus, in the current study, the experiment was terminated in the fourth week as a sufficient time for fracture healing in rabbits.

Although various techniques are used for the radiological evaluation of fracture healing, most of these have been shown to be inadequate as they are not objective and vary from person to person. In some studies, the bridging status between the fractured ends has been evaluated from direct radiographs (9, 24). In the current study, the widely used Lane and Sandhu (9) scoring system was used in the radiological evaluation by more than one orthopaedic, traumatology, and radiology specialist 25–(28) . We are of the opinion that this provides a more objective evaluation.

In clinical application, bone union is monitored with both physical examinations and radiological tests. Similarly, in the current study, the presence of bone union was investigated by forced movements to the fractured fibula in two planes, and radiology provided an additional evaluation. Dimar et al. (10) emphasised clinical examination in two planes for stability evaluation and the approach in the current study supports this.

Some histopathological evaluations related to fracture healing have given qualitative results rather than numerical values (29, 30). Allen et al. 31–(33) reported a 5-stage histopathological evaluation. However, Huo et al. (12) recommended a more comprehensive method, which was used in the current study. By examination of 10 stages in this system, more detailed and accurate results are obtained (28, 34) (35).

Physiological and pathological events in living organisms occur as a result of substances which may have harmful effects. The most important of these are free oxygen radicals which are balanced by the body's antioxidant system (36). The effects of antioxidant molecules on fracture healing have been examined with particular focus on some vitamins and essential amino acids (4, 31) (33, 37–39) . As part of the antioxidant system, glutamine being a precursor of the glutathione-peroxidase enzyme, also serves as an antioxidant. Studies by Durak et al. (40) and Sarisözen et al. (31) emphasised that the effect of antioxidants in reducing free oxygen radicals found in fracture haematoma may have a positive effect on fracture healing. Therefore, in the current study it was thought that as well as other effects, glutamine as an antioxidant could positively influence fracture healing by reducing free oxygen radicals found in the fracture area, but indicators related to this were not examined.

In a study on rabbits by Sinha and Goal (41), where the amino acids lysine and arginine were administered orally, increased callus vascularisation and mineralisation was reported and the healing period was accelerated to 2 weeks. In an experimental study by Kdolsky et al. (42) a femoral diaphysis defect was created in guinea pigs and intramedullary fixation was applied to the femur with a k-wire. The findings were evaluated histologically, mechanically, and radiologically and it was concluded that the animals which had received L-Arginine had better fracture healing and mechanical stability. Comparing the current study to that one, the amino acid used was glutamine, the fibula was used rather than the femur, and there was no mechanical evaluation. Therefore, although the idea of the effect of amino acids on fracture healing is similar, from a technical aspect there is no similarity to the current study.

In the histological evaluation of the current study, all the rabbits that had received glutamine were seen to be at stage 9 full recovery, whereas half of the control group were at stage 8 recovery with cartilage and immature bone together and the others were at stage 9 recovery. This finding justifies the view that glutamine has a positive effect on fracture healing. However, no statistically significant difference was determined between the two groups. Similarly, in the radiological evaluation, the immunonutrition group was determined as mean 3.42 points from a total of 5 by the orthopaedic and traumatology specialists and mean 3.35 points by the radiology specialists whereas the control group points were low at mean 2.5 and 2.65 respectively. Despite the difference between them, there was no statistically significant difference. This is thought to be because of the low number of animals used in this study. It is also thought that histological evaluation at an earlier stage of healing may yield significant results.

In conclusion, despite the beneficial effects of glutamine on the speed of wound healing and reducing morbidity and mortality rates in sepsis, and the positive effect observed on bone healing in the immunonutrition group clinically, radiologically, and histopathologically, no statistically significant difference was determined compared to the control group in this study. Additional light will be shed on this subject in literature by subsequent research considering different parameters.