Leg fractures are the most common traumatic injury of the long bones. They are evidence of a brutal and violent trauma, responsible for pluritissular lesions, including the opening of the fracture site, which can worsen the functional prognosis of the limb and, in extreme cases, the patient’s vital prognosis . Since 2007, the WHO (World Health Organization) has estimated the incidence of these fractures in the world at 26.9 per 100,000 inhabitants and a prevalence of 7.8% among road traffic injuries . In Burkina Faso, the 15 - 59 age group is the most affected. It constitutes the active population with a precarious standard of living, using two-wheeled vehicles involved in road traffic accidents   .
Regardless of the stage, open leg fractures remain a surgical emergency. It is necessary to know how to choose a therapeutic attitude, judge it on its results and adopt it. The management of open leg fractures has evolved remarkably over the last 30 years, with many controversies  . The use of central medullary nailing (CMN) has therefore been urgently required in order to meet the demands of time, trained teams and sufficient resources   . Its practice becomes difficult in developing countries where the patient’s under-equipment and precarious means limit the possibilities of emergency osteosynthesis . As a rule, for reasons of exceeding the time limit described (before six hours), the majority of our patients have always benefited from a two-stage treatment (trimming and plaster splinting followed by osteosynthesis after wound healing and normalisation of the infectious biological balance sheet). This type of procedure results in an extended hospital stay  .
Taking these realities into account and under the preconditions of a well-conducted medical treatment of the open fracture on admission, is it not possible to perform the CMN in emergency beyond the six hours recommended in the literature? In order to do this, we wish, through this study, to evaluate the results of CMN in emergencies in open leg fractures in adults in order to propose the adoption of this therapeutic method in countries with precarious health conditions.
2. Patients and Methods
This was a retrospective study with descriptive and analytical aims from 1 January 2014 to 31 December 2018, conducted in the Orthopaedics-Traumatology Department of the Yalgado Ouédraogo University Hospital in Ouagadougou, Burkina Faso. The aim of this study is to evaluate the anatomical and functional results of CMN in emergencies in adult open leg fractures (even beyond six hours) in order to propose the adoption of this therapeutic method in countries with precarious health conditions.. All patients aged at least 17 years, who presented with an open leg fracture, who were urgently treated by CMN within less than 48 hours and who were regularly followed up in our department, were included in this study.
The indication of CMN in emergencies was retained in the following cases:
Open fractures type I and II, according to the Cauchoix and Duparc classification;
Fractures of both leg bones and/or isolated fractures of the tibia ;
Simple, complex and comminuted fractures.
As soon as the patient arrived in the emergency department, after recovery of his hemodynamic state by infusion of solutes, macromolecule and analgesics, we proceeded to an abundant lavage of the wound, followed by a protective dressing and a temporary restraint by splint. Next, we administered amoxicillin-clavulanic acid intravenously. Following this medical treatment, X-ray images (face and profile) of the fractured leg segment, ipsilateral knee and ankle were taken.
After confirmation of the diagnosis of an open fracture, the osteosynthesis equipment and various consumables were prescribed to the patient. The wound dressing was renewed every 12 hours, and the patient was admitted to the operating room as soon as he had honoured his medical prescription.
Clinical and para-clinical follow-up made it possible to assess post-treatment progress with a view to detecting and treating complications. Thus, in case of suppuration, a bacteriological examination coupled with the antibiogram was systematically carried out.
Consolidation was assessed by standard radiography of the leg and functional results by the Lysholm functional score . Lysholm score’s criteria are presented in Table 1.
Data were collected on an individual form based on emergency and operating room records, medical records and patient interviews.
These data were analyzed with Epi Info 7 software.
Thus, over the five years, 91 patients were operated, either an annual frequency of 18 cases per year. The mean age of these patients was 34.1 years (17 - 77 years). The age range of 17 - 30 years predominated (51.4%). We noted more males than females, with a sex ratio of 2.2. Road traffic accidents (RTA) were the predominant etiology (93.4%), followed by sports accidents (2.2%).
According to Cauchoix and Duparc’s classification of open fractures  (Figure 1), we identified 59 type I open fractures (64.8%) and 32 type II open fractures (35.2%).
Table 1. Lysholm functional score .
Figure 1. Classification of open leg fractures according to Cauchoix and Duparc. (a) Open fracture type I of Cauchoix and Duparc in a 25-year-old patient following an RTA; (b) Cauchoix and Duparc open fracture type II in a 39-year-old patient following an RTA.
Seventy-six fractures of both leg bones (83.5%) and 15 isolated fractures of the tibia (16.5%) were recorded. Fracture patterns were simple in 46 cases (50.5%), complex in 27 cases (29.7%) and comminuted in 18 cases (19.8%). Figure 2 shows the types of fracture lines.
Almost a third of our patients (27%) were operated on between the 12th and 18th hour. The distribution of patients according to the time of surgical invention is shown in Figure 3.
Under local anaesthesia (100% spinal anaesthesia) and after careful trimming, tibial CMN osteosynthesis was performed at the same time as the operation. The universal tibial nail in static or dynamic assembly (Figure 4), or the Küntscher nail, were the main implants used. A universal tibial nail was used in 84 cases (92.3%) and a Küntscher nail in 7 cases (7.3%). Depending on the type of intramedullary nailing, we achieved dynamic locking in 6 cases (6.6%) and static locking in 05 cases (5.5%). However, in 80 cases (87.9%) no locking was performed, the nail was placed with hard friction.
Reduction was anatomical in 29 cases (31.87%); osteosynthesis was stable in 81 cases (89%).
The average hospital stay of our patients was 6 days (4 - 36 days). They all went out to their homes with close bandaging appointments on a case-by-case basis.
Wound healing was achieved within an average of six weeks (2 - 51 weeks). Longer delays were marked by the occurrence of suppuration. Thus, the evolution was favourable in 81 cases (89%), marked by effective healing within three weeks.
At a mean follow-up of 28 months, radiographic consolidation was effective in all of these 81 patients (89%).
Table 2 shows the distribution of 80 patients according to radiographic reduction and consolidation.
(a) (b) (c)
Figure 2. Radiographic images of the face and profile of the legs showing the types of fracture lines. (a) Isolated open fracture of the tibial diaphysis in a 25-year-old patient following an RTA; (b) Complex open fracture of both leg bones in a 39-year-old patient following an RTA; (c) Transverse open fracture of both leg bones in a 36-year-old patient following an RTA.
Figure 3. Distribution of patients by time to surgery.
Table 2. Distribution of fractures according to radiographic reduction and consolidation of our patients (n = 80).
(a) (b) (c)
Figure 4. Main types of centromedullary nailing. (a) Static locking tibial centromedullary nailing of a bifocal fracture in a 35-year-old patient; (b) Dynamic unipolar locking of a transverse fracture in a 34-year-old patient; (c) Unlocked centromedullary nailing of an isolated transverse middiaphyseal fracture of the tibia in a 26-year-old patient.
Consolidation was perfect on an anatomical reduction in 29 cases (36%) (Figure 5).
We found 10 cases (10.99%) of suppuration of the surgical wound. The time to onset of suppuration ranged from three days to 23 days, with an average of 11.3 days. The most frequently found germ was Escherichia Coli in 9 cases, followed by Staphylococcus aureus in 1 case. These patients were treated according to antibiotic susceptibility testing. These suppurations were mainly located in the middle third of the leg (50%). The onset of suppuration was not related to the time to management. Table 3 shows the distribution of suppuration according to the time to management.
There was no correlation between the onset of suppuration and the time to patient management (Khi2: X2 = 5.806 et P = 0.56).
In Figure 6, we have a good healing of the surgical wound in 2 weeks.
At the average 18-month follow-up, the evolution was favourable in 80 cases (87.9%) and with late complications in 11 cases (12.1%). These 80 patients had had regular clinical and radiographic follow-up until the removal of the osteosynthesis material. The mean time to consolidation in these patients was five months (2 - 9 months). The distribution of patients by time to consolidation is shown in Figure 7.
Figure 5. Consolidation on static locked centromedullary nailing in a 29-year-old patient.
Figure 6. Surgical wound healing in a 40-year-old patient.
Table 3. Surgical wound healing based on time to management.
The 11 patients who presented late complications (pseudarthrosis, delayed consolidation, osteitis and vicious callus), were retreated. Figure 8 shows the frequency of these late complications.
At the average 48-month decline, our functional results were satisfactory in 81.3%; these patients had returned to activity at a variable level of ability. The mean overall Lysholm score was 87.23 ± 11.06 (64 - 100). Table 4 summarizes the functional results according to the Lysholm score.
Our therapeutic results were good and excellent in 61 patients (81.3%).
Figure 7. Time to consolidation in follow-up patients (n = 80)
Figure 8. Distribution of late complications among the 11 patients.
Table 4. Functional outcomes of patients reviewed at 48 months (n = 80).
During our study, we were confronted with certain difficulties and shortcomings related to its retrospective nature and its size, which was not large enough. Nevertheless, we can make comments and lead a discussion after comparing them with the data in the literature.
The average hospital stay of our patients was five days with (4 - 36 days). Duirek et al.  in 2011 in Morocco, noted an average hospital stay of 10 days in their series. Ouedraogo et al.  in 2012 and Dakoure et al.  in 2013 in Burkina Faso, reported 21.5 and 32.1 days respectively in their study. This short stay of our patients is explained by our therapeutic approach: open fractures type I and II are treated in a single operation, unlike the other authors who treat open leg fractures type I, II, III in two or even three operations  . Emergency CMN is therefore an efficient process to cope with the lack of financial means and accommodation places in our hospital units.
The early complications in our patients were limited to suppuration of the surgical wounds observed in 10 cases (10.99%). Our results are similar to those of Soudre et al.  in Burkina Faso and Peter et al.  in England, who reported a frequency of suppuration of 12% and 11% respectively. Ndayisaba et al.  in Burundi, Ouedraogo et al.  in Burkina Faso and Tscherne et al.  in Germany, found lower infection rates than ours, respectively 7.8%, 8.7% and 3.9%. In contrast, Dakoure et al.  in Burkina Faso, Moyikoua et al.  in Congo Brazzaville, Ditengou et al.  in Mali and Ikem et al.  in Nigeria, found a higher frequency of suppuration than ours, respectively in 34.7%; 35%; 35.7% and 39.3%.
In our series, there was no correlation between the onset of suppuration and time of treatment (P = 0.56). This suppuration could be explained either by initial massive contamination, or by the poor quality of the external dressings, or by non-adherence to antibiotic therapy, thus a probable secondary superinfection.
The mean time to consolidation in our study was 5 months (2 - 9 months). Duirek et al.  in 2011 in Morocco, in their series of leg fractures, noted a mean time of consolidation lower than ours, with 3.5 months (03 - 04 months). Our study was limited to open fractures. The loss of the fracture hematoma through the skin opening in our series contributed to slowing down the consolidation process . This may explain the long consolidation time we observed.
At a mean follow-up of 28 months, our radiographic results were favourable, marked by effective consolidation in all patients with visible callus bone on the radiographic images, and an anatomical reduction in 36% and satisfactory in 64%; which corroborates the results of Bonnevialle et al.  and those of Duirek et al. .
At an average follow-up of 48 months, our functional results were excellent and good in 61 patients (81.3%); this rate is similar to those of Dakoure et al.  in 2013, who noted 87.9% good to excellent results.
These anatomical and functional results are compatible with an adequate socio-professional reintegration; which testifies to the reliability of our therapeutic attitude which proves to be effective and to be recommended for countries with limited resources.
Open leg fractures are very common injuries and have a major socio-economic impact, as they occur most often in adult males. Emergency CMN of open leg fractures appears to be an effective solution. It favours a considerable reduction in hospital stays, while providing satisfactory results, thus allowing an early and adequate socio-professional reintegration of the patient.
Therefore, it imposes basic principles, namely the medicalisation of the wound upon admission of the patient. Therefore, postoperative follow-up must be rigorous in order to detect and treat complications early and efficiently.
 Zucman, J. and Maurer, P. (1969) Two-Level Fractures of the Tibia: Results in Thirty-Six Cases Treated by Blind Nailing. Journal of Bone and Joint Surgery, 51B, 686-693.
 Dakoure, P.W.H., Da, S.C. and Sanou, B.G. (2013) Aspects épidémiologique et lésionnel des fractures ouvertes récentes de jambe au CHU de Bobo-Dioulasso (Burkina-Faso). Médecine d’Afrique Noire, 60, 286-290.
 Girard, D., Pfeffer, F., Galois, L., Traversari, R., Mainard, D. and Delagoutte, J.-P. (2003) Enclouage centromédullaire non alésé dans les fractures de jambe: A propos de 106 cas. Revue de Chirurgie Orthopédique, 89, 59.
 Ikem, I.C., Oginni, L.P. and Ogunlusi, L.I. (2006) Determinants of Management Outcome in Open Tibia Fractures in Ile-Ife. Nigerian Journal of Surgical Research, 8, 81-85.
 Laffargue, P., Mulliez, A., Mielcarek, P., Baudson, H. and Bachour, F. (2005) Complications de l’enclouage centromédullaire du tibia en fonction du matériel. Revue de Chirurgie Orthopédique, 91, 168-170.
 Ribault, L., Vergos, M. and Konan, P. (1990) Les fractures ouvertes de jambe. Indication thérapeutique à propos de 47 cas traités dans un centre Hospitalier Régional en Zone Sub-Sahélienne de l’Afrique de l’Ouest. Médecine d’Afrique Noire, 37, 304-307.
 Tegner, Y. and Lysholm, J. (1985) Rating Systems in the Evaluation of Knee Ligament Injuries. Clinical Orthopaedics and Related Research, No. 198, 43-49.
 Ouedaogo, A. (2012) Aspect épidémiologiques, diagnostiques, thérapeutiques et évolutives des fractures ouvertes de jambe au CHU-YO: A propos de 126 cas. Thèse de doctorat en médecine ouagadougou, UFR/SDS, No. 2121, 57-66.
 Soudre, R., Yilboudo, J., Ouedraogo, K. and Ouiminga, R.M. (1987) Aspects étiologiques des ostéites post-opératoires à l’hôpital Yalgado Ouedraogo de Ouagadougou, Burkina-Faso. Interface Afrique, 29-32.
 Ndayisaba, G., Bazira, L. and Gahongano, G. (1992) Place de l’antibiothérapie préventive en chirurgie osseuse en milieu tropical. A propos de 59 complications septiques. Médecine d’Afrique Noire, 39, 597-598.
 Moyikoua, A., Ngaste-Oko, Boutity-Buang, Ondzoto, J.M., Kaya, J.M. and Pena-Pitra, B. (1992) Résultat du traitement initial des fractures ouvertes récentes des membres à propos de 150 cas traités au C.H.U. de Brazzaville. Médecine d’Afrique Noire, 39, 755-762.