Internal fixation of shaft humerus fractures by dynamic compression plate or interlocking intramedullary nail: a prospective, randomised study

Humerus fracture is unique amongst the long bone fractures in its tolerance of less than anatomical reduction. Shortening up to 3 cm, rotation <30° and angulation up to 20° are considered acceptable [16]. Due to this fact, most of the humerus fractures are still managed conservatively and have good functional results. The most common indication of operative intervention is inability to achieve acceptable reduction, followed by associated vascular lesions, open fractures, radial nerve palsy, polytrauma patients, floating elbow and pathological fractures [17]. The preponderance of the fracture in young males, commonly in third and fourth decade of life, was seen in our series, as has been reported by other similar studies [18]. Road traffic accident is the most common mode of injury, especially in younger patients.

In the past, open reduction and plating was the preferred method of operative intervention, and still continues to remain the gold standard. However, conventional plating technique involves an extensive surgical approach for open reduction of fracture and is theoretically associated with increased risk of radial nerve injury and more blood loss. The excellent result of intramedullary interlocking nails in tibia and femur fractures has stimulated interest in applying the same methods in humerus fractures. Intramedullary nails are subjected to smaller bending loads than plates because they are closer to the mechanical axis than the usual plate position on the external surface of the bone. Intramedullary nails can also act as load-sharing devices in fractures with cortical contact. Moreover, the stress shielding commonly seen with plates and screws is minimized with intramedullary nails. Intramedullary nailing in humerus fractures is a less invasive procedure which maintains the biology and gives a good, stable fixation. It is also assumed to result in quicker union, less blood loss and less chances of radial nerve injury. However, controversy still exists over the best method of fixation. Ooyung et al. [19] in a meta-analysis of ten studies comparing the results of plating versus nailing concluded that both achieve similar results in humerus fractures, but plating was associated with reduced shoulder problems.

Union rates in our study were comparable between the two groups, non-union being seen in 8 % in ILN and 8 % in DCP group. Similar rates of non-union have been observed in most of the studies [20, 21]. Non-union in the DCP group is usually due to extensive soft tissue dissection or malreduction and is often associated with implant failure. Even though the effect of reaming might facilitate bone healing, non-union has been reported in 0–9 % of cases [22, 23] managed with reamed intramedullary nails. Non-union in ILN usually results from distraction at the fracture site as humerus seems to be less forgiving than tibia or femur in this aspect. Non-union in both groups was managed by open reduction and plating with bone graft, as has been suggested in the literature [24]. The time to achieve union in our study was less in ILN than in plating group, although not statistically significant. Denies et al. [25] also reported earlier union in nailing, probably due to the less invasive nature of the procedure and the maintenance of the fracture haematoma. Changulani et al. [26] also reported earlier union in nailing with a statistically significant difference.

Whether to use reamed or unreamed nails is still a controversial topic. The advantages of reaming include the significant increase in the blood flow to the muscles and surrounding soft tissues, and this increase persists for up to 6 weeks. The increase in soft tissue blood flow may increase the cortical blood flow and thus more chances of union. Further, the cost of reamed nails is less, which is an advantage for developing nations, like ours. The disadvantage of reaming is the chances of radial nerve injury, especially when there is a gap at the fracture site. Further, some studies have shown extensive heat necrosis from nailing in small diameter canals. Reaming in small diameter canals may lead to distraction at the fracture site.

Infection, iatrogenic radial nerve palsy and hardware failure are most important complications associated with plating. We had higher rates of infection and iatrogenic radial nerve palsy in the plating group, as has been seen by Changulani et al. [26]. However, in a meta-analysis, Bhandari et al. [27] did not show higher risks of infection or radial nerve palsy with plating. Radial nerve is at a definite risk in plating, with special precautions being taken to prevent nerve from coming beneath the plate. Although radial nerve injury after nailing is rare, the risk can be further minimized by ensuring accurate reduction of the fracture (no gap) before passage of the reamers or the nail and by avoiding reaming in areas of comminution where the nerve is closely apposed to the bone.

Impairment of shoulder function is the main drawback of interlocking nailing. Shoulder pain in these patients may be related to violation of the rotator cuff, prominent nail end, adhesive capsulitis or unknown causes [27, 28]. We had shoulder problems in 20 % of our patients. One patient with protruding nail required a second surgery for the removal of implant. Chao et al. [28] also reported three patients with proximal protrusion of the nail. This usually arises from not pushing the nail distal enough, possibly from fear of producing a distal fracture, or from migration of an unlocked nail. We suggest assessing the length accurately before passing the nail and using C-arm till the procedure ends. Similar findings have been reported by many studies [28, 30, 31]. However, Flinkilla in an analysis of shoulder from different studies reported similar shoulder scores in both nailing and plating groups, with plating having better abduction and flexion [32]. We had intra-operative comminution at fracture site in one patient where the medullary canal was narrow which resulted in distraction at fracture site and eventually resulting in non-union. Introduction of large reamers or nail into a narrow canal can result in comminution at fracture site which are usually undisplaced fissures not requiring fixation. The incidence of this complication has decreased from 6 to 1.8 % due to introduction of newer nail designs [33]. Re-operation rates were similar between the two groups as has been seen by Denies et al. [25]. However, Bhandari et al. [27] reported more re-operations in the nailing group. Indications of re-operation in plating are union problems, hardware failure and revision for radial nerve palsy while the nailing patients it is for union problems, removal of protruding implant and management of preoperative fracture. A disadvantage of nailing is the need of fluoroscopy for the procedure and the associated risks to the surgeon and the theatre personnel. The fluoroscopy time reported by us is comparable to that reported in the literature and seen for other long bone fractures. To overcome this, expandable nails have come into the market which require less extensive use of fluoroscopy.

In our study, nailing was superior to plating with respect to the average post-operative stay of the patients and operating time. The main reason for the longer stay in the plating group in our study was because of the longer delay in surgery in the plating patients. This was due to our tendency to do DCP only when the swelling had completely subsided. In contrast, ILN does not need rigorous subsidence of the swelling. Most of our patients belonged to far-off places where sterile-dressing facilities were not readily available and thus tended to stay till the operative wound was deemed clean. ILN patients had an edge due to their smaller surgical wounds and so were discharged earlier. The shorter stay, with a less invasive method, such as closed nailing, is of great advantage in developing countries where the orthopaedic hospital beds are limited and resources are scarce. For the same reason, the less operating time reported by us is also advantageous. Chao et al. also had shorter operative time in ILN, although the difference was not significant, while Chaudhary et al. had shorter operative time in the plating group [29, 34]. ILN was also associated with significantly decreased blood loss than plating, as has been seen in most of the studies [29]. Although this difference is statistically significant, but in the clinical settings, this difference is marginal.

The ASES score and the final outcome in our series did not show any significant advantage of one method over the other. Some studies have shown plating to be more effective, while others have found better results with nailing [35, 36]. However, the meta-analysis of different randomized and quasi-randomised controlled trials comparing the two failed to find any significant difference between the two with respect to ASES scores [37].

The most important factors in obtaining fracture healing are anatomical reduction, stable fixation and adequate blood supply. Although internal fixation with DCP may result in a better reduction, it also carries a more extensive soft tissue dissection with risk of radial nerve lesion and infection. ILN provides secure and rigid fixation with less soft tissue damage and maintaining the biology. Although ILN is associated with relatively increased incidence of shoulder complications, it has definite advantages in terms of shorter hospital stay, less blood loss and shorter operative time, which are of immense importance in the developing countries with limited resources.

We conclude that antegrade locked intramedullary nailing is an effective alternative to plating in shaft humerus fractures as it has comparable results in terms of union rate and complications. In addition, it has the added advantage of lesser operative time and shorter hospital stay, both of which have a distinct advantage in those centres in developing countries which have the facility of fluoroscopy.