top of page

Comparative Effects of Platelet Rich Plasma (PRP) and Placebo in Nonunion of Long Bone Fractures


Long bone nonunion is among the most devastating of traumatic fracture complications and its treatment method is still a major problem for Orthopedic Surgeons and Physical Therapy Doctors. Fracture nonunion rates of the humerus, femur, and tibia are 2-6%, 2-8%, and 3-6.4%, respectively. Although many methods have been proposed for the nonunion of long bone fractures, it still remains unresolved. These types of fractures pose a great economic burden on the healthcare system, with the reported financial cost of the humerus, femur and tibia being 31,132 USD – 34,440 USD – 32,660 USD, respectively. Taking measures to prevent such fracture nonunion is seen as the best method to combat this problem. This is accomplished by keeping the fracture site in the same direction using internal or external fixators. Necrotic debridement, keeping the infectious tissue and fracture sites in line with the intramodular nail are also among the successfully applied methods. Despite these surgical techniques, nonunion can still occur to a certain extent. Several treatment models with interesting results have also emerged recently, including electromagnetic application, biodegradable implants, and platelet-rich plasma (PRP). In the content of PRP, Platelet-derived growth factor (PDGF), vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), fibroblast growth factor (FGF), and transforming growth factor (TGF-β1, which is also responsible for tissue formation and repair in the human body). It is an autologous treatment method with a high concentration of activated platelets (about 1.2 million platelet cells in 1 ml plasma), which contains beneficial factors such as TGF-β2). PRP has been used successfully in musculoskeletal injuries. There have been many scientific studies reporting interesting results of PRP applications for the treatment of bone nonunion cases. Although some studies have proven that PRP increases the rate of union of the fracture site, some studies have suggested that PRP has no effect. Thus, this study aims to reveal the effects and healing rates of PRP application in nonunion cases of long bone fractures.

Material and Method

Study Participation Rate This study is a clinical study of 17 months (May 2011 to October 2012) with the contribution of Shiraz University Medical Sciences department at Shahid Rajee Trauma center (Level I Trauma Centre). The study protocol was approved by the Institutional Review Management (IRB) and the AJA University Health Sciences Ethics Committee, and written consent was obtained from all participants. This study has also been registered with the Iran Clinical Trial Institution (IRCT201208262445N1; In this study, 79 adult patients older than 18 years of age and suffering from long bone (Femur, Tibia, Humerus, Ulna) fracture nonunion for more than 6 months were included. Those under the age of 18, infected patients, vascular disease, extensive skin wounds at the fracture site, musculoskeletal disorders, pathological fractures, tumors, extensive fractures, and active immune system disorders were not included in this study. In addition, those who have a platelet count of less than 100,000 platelets/ml and more than 650,000 platelets/ml in their blood and those who have received growth factor, steroid and AIDS treatment in the last 1 year were not included in this study.

Randomness and Intervention

The patients included in this study were divided into 2 study groups based on complete randomness, and the registration number of each patient was randomly assigned by a computer-assisted digitizer. Those in the PRP group (37 patients) received 5 mL of PRP, and from the other placebo group (38 people), normal saline was taken from the fracture site or the open reduction and internal fixation site after intramodular nailing as placebo.


Study Protocol and Measurements

All of the patients included in the study underwent the same history and physical therapy the day before the operation, and important information such as age, sex, and clinical characteristics of the patients (fracture site, mechanism of injury, previous surgeries, duration of the fracture) were recorded. At the same time, 5mL of blood was taken from the patients to determine the hemoglobin and platelet count. Autolog PRP was prepared using the Gravitational Platelet Separation System according to the protocol defined by the manufacturer. As a result, a total of 54 mL of blood was drawn from the right cubital vein during the operation into tubes containing citric acid. To separate platelets from red blood cells and plasma, blood-filled tubes were rapidly centrifuged at 3200 rpm for 15 minutes. Approximately 5-6 mL of PRP was obtained and transferred to the syringe. It has been proven that the platelet concentrate in PRP is 4.2-4.8 times higher when compared to normal blood values. General anesthesia was administered to all patients using the standard protocol. As prophylaxis, 1 gram Cafezoline was injected preoperatively and low molecular weight heparin was administered postoperatively as deep vein thrombosis. All operations were performed by 2 orthopedic surgeons, and the surgical technique was purely based on clinical decisions and expert determinations. Under general anesthesia, all previously inserted (if any) implants were removed, and non-union long bone fracture ends were determined. For femur fractures, intramedullary nailing was applied while preserving the notch canal. An anterograde approach was used to insert the stainless steel fixation nail. Stainless plates and screws were used for humerus fractures, ulna and tibia. All necrotic tissues were debrided and bacterial culture was taken from the fracture site. Standard techniques of bone graft with iliac crest were applied to all patients. A membrane was formed using the muscular fascia and PRP or Placebo was injected into the fracture site in the periosteum. Surgeons were not informed about the contents of the syringe. After 48 hours, the patients were discharged from the hospital without any problems and were instructed to participate in the rehabilitation program against the risk of pseudothrosis. Rehabilitation practices were initiated 24 hours after the operation, and light joint work was allowed for patients with femur and tibia fractures after 4 weeks. Patients with humeral or ulnar fractures were put on plaster until 3 weeks before rehabilitation started. Patients were checked every 45 days for 270 days in day clinics. The patients were controlled clinically and radiologically (anteroposterior and lateral angles) at each visit, and pain at the fracture sites was measured manually on a visual analog pain meter (VAS) from 1 to 10. Painless full weighted joint movements were defined clinically for low extremity fracture nonunion and painless mild motions for high extremity fractures. Presence of callus bridge in at least ¾ cortis over the AP was defined as rasiological union and this was supported by lateral radiographic data. Clinical and radiological unions that occurred only 9 months after the application were defined as the inadequacy of the application. Again, at each visit of the patient, limb length was measured compared to pre-application. Those with limb shortening greater than 4 mm were recorded. The success rates of post-operative injections were recorded by comparing the groups according to insufficient union and nonunion. All patients and patient visitors were never informed about this study after the operation.

Statistical Analysis

In order to achieve 90% success in determining the 5% difference between the healing rates of fractures with an α value of 0.01, the healing coefficient was taken as 1.96 and at least 32 patients were included in each study group. An additional 79 participants were added to the study to compensate for non-evaluable patients and dropouts. The Social Science Statistics Package for Windows (SPSS Windows), version 16.0, was used to analyze the data. Paired T-Tests were used to compare within-group results, independent T-Tests were used to compare between-group results, Chi-Square or Fisher tests were used to compare categorical variables and success rates. Data were reported as ± SD values. Those with a two-sided p value less than 0.05 were evaluated in the statistically significant category.

PRP nin kırık tedavisinde etkileri
PRP Kemik grefti Kırık Tedavisi

In this study, we determined how effective PRP (Platelet Rich Plasma) used with bone graft for the healing of long bone fractures was against the randomly selected placebo group. We have clinically demonstrated that PRP application significantly increases the healing process. In addition, PRP application provided a shorter recovery period, lower limb shortening, and less post-operative pain complaints. The results of this study clearly showed that PRP treatment together with autologous bone graft is the most rational method to follow in the healing of long bone fractures.
To our knowledge, this study is among the most important randomized clinical trials for this disease. Recently, Prof. Malhotra R. also conducted this study on 94 patients with long bone fractures and achieved a success rate of 87.2%. He kept the bone union status under clinical and radiological regular monthly controls for 4 months. At the end of a 4-month period, he observed that the bones of 82 patients were completely fused. 32 patients showed trabecular bridging on X-Ray at the end of the 2nd month, and 41 patients had this result at the end of the 3rd month. 12 patients did not show any improvement. In another clinical study, Prof. Memeo used PRP in 7 previously unsuccessful patients with anterior arm post-traumatic bone fractures. All patients showed excellent clinical outcome and all recovered completely. Dr. Samy Daniels investigated the effect of PRP in femoral neck fractures with internal fixator, and published a report on fast recovery time, low post-operative pain complaints and an increase in the definitive healing rate, which also supports our study. However, Dr. Say suggested that PRP did not have a serious effect on the lower extremity healing process. When all these are evaluated together, considering the results obtained, it should be clearly emphasized that the application of PRP has given excellent results in the treatment of long bone fractures.
Some veterinarians have also tried PRP application on animals. Dr. Hakimi and his teammates, in their study on 6-week-old small pigs, showed that PRP accelerated bone regeneration. Dr Gerard conducted a study on the effect of PRP on the healing of bone fractures in dogs. It has been demonstrated that PRP provides very fast and early healing in the maxillary bone. Alongside the results of our study, Dr. Kanthan Stevens revealed in his study of tibial fractures of rabbits that PRP mixed with bone graft provides much faster and permanent healing compared to bone grafting alone. However, Dr. Kanthan also stated that applying PRP alone without bone grafting did not contribute much. Dr Galasso and his colleagues also carried out a study using a locked intramodular nail and PRP together. As a result, their success rate is higher than 72%.
They reported that it increased to 91%. However, they did not confuse PRP with bone graft. Some other studies have concluded that PRP is beneficial in the first stages, but its effect decreases completely in the later stages. The reason for this is that the Growth Factors do their job in the first stage and lose their effectiveness in the later stages, which is quite natural. In fact, PRP is very successful in removing non-viable autologous bones, but its effect is not sufficient in the formation of new bones. In addition, PRP significantly increases the early recovery rate. We also observed that PRP had a pain-reducing effect in the first phase (up to 12 weeks) in our study. This result can be explained by the end of the platelet life approximately 5 days after the injection. The infection rate in our study was slightly higher than normal international values, because the density of people in our hospital was slightly higher than normal conditions.
We were also exposed to some limitations during our study, the first of which was the low participation in the study. However, our study showed a success rate of 90% in determining the difference between PRP and placebo method. As a result, the results of our study are reliable and can be evaluated. Another limitation is that the femur, tibia, humerus and ulna bones must be evaluated together. Considering them separately and conducting a different and comprehensive clinical study will show us more specific and definitive results.
In conclusion, mixed with autologous bone graftPRP application, resulted in a remarkably high recovery rate, short recovery time, low limb loss, and low post-operative pain complaints in the healing of bones after intramodular nail or plate&screw applications.

bottom of page