Arch Orthop Trauma Surg DOI 10.1007/s00402-016-2418-8
TRAUMA SURGERY
Anterior augmentation plating of aseptic humeral shaft nonunions after intramedullary nailing Jan Gessmann1 • Matthias Ko¨nigshausen1 • Marlon Osman Coulibaly1 Thomas Armin Schildhauer1 • Dominik Seybold1
•
Received: 19 November 2015 Ó Springer-Verlag Berlin Heidelberg 2016
Abstract Introduction Humeral shaft nonunion after intramedullary nailing is a rare but serious complication. Treatment options include implant removal, open plating, exchange nailing and external fixation. The objective of this retrospective study was to determine whether augmentation plating without nail removal is feasible for treating a humeral shaft nonunion. Materials and methods Between 2002 and 2014, 37 patients (mean age 51, range 20–84 years) with aseptic humeral shaft nonunions prior to intramedullary nailing were treated with augmentation plating. The initial fractures had been fixed with retrograde nails (10 cases) or anterograde nails (27 cases). There were 34 atrophic nonunions and 3 hypertrophic nonunions. Nonunion treatment of all patients consisted of local debridement through an anterior approach to the humerus and anterior placement of the augmentation plates. Supplemental bone grafting was performed in all atrophic nonunion cases. All patients were followed until union was radiologically confirmed.
Results Union was achieved in 36 patients (97 %) after a mean of 6 months (range 3–24 months). There was one case of iatrogenic median nerve palsy that showed complete spontaneous recovery 6 weeks postoperatively. One patient sustained a peri-implant stress fracture that was treated successfully by exchanging the augmentation plate to bridge the nonunion and the fracture. No infections or wound healing complications developed. At a mean follow-up of 14 months, all patients showed free shoulder and elbow motion and no restrictions in daily or working life. Conclusions The results indicate that augmentation plating using an anterior approach is a safe and reliable option for humeral shaft nonunions after failed nailing, and the treatment has no substantial complications. Because the healing rates are similar to the standard technique of nail removal and fixation by compression or locking plates, we consider this technique to be an alternative choice for treatment. Keywords Humeral shaft nonunion Augmentation plating Intramedullary nailing Bone grafting Anterior approach
& Jan Gessmann [email protected] Matthias Ko¨nigshausen [email protected]
Introduction
Marlon Osman Coulibaly [email protected]
Intramedullary (IM) nails are frequently used in the treatment of humeral shaft fractures, especially since the development of newer nailing techniques and locking mechanisms. Recently published meta-analyses comparing the results of IM nailing with the gold standard of open reduction and plating found comparable high union rates [1, 2]. Still, nonunions of the humeral shaft occur, depending on fracture type, osteoporosis, and soft tissue damage as well as inadequate reduction, distraction of
Thomas Armin Schildhauer [email protected] Dominik Seybold [email protected] 1
Department of Trauma Surgery, BG University Hospital Bergmannsheil, Ruhr University Bochum, Bu¨rkle-de-laCamp-Platz 1, 44789 Bochum, Germany
123
Arch Orthop Trauma Surg
fragments and unstable fixation [3, 4]. Humeral nonunions after prior intramedullary nailing can be difficult to address due to possible bone loss around a loose nail or locking bolts, which complicates the reconstructive procedure and leads to poorer operative results compared to prior nonoperative treatment or plate fixation [5]. If untreated, osteolysis and bone loss around the nail and locking screws increase the risk of secondary fractures [6]. A permanent nonunion leads to severe restrictions of shoulder and elbow function and may lead to a major disability, especially in elderly patients who depend on their arm support to use canes, crutches or wheeled walkers [6]. Different treatment techniques have been described for the specific reconstructive problems, including open reduction and plating with removal of the nail, exchange intramedullary nailing, Ilizarov external fixation and the application of a wave plate [5, 7–11]. Although the technique of augmentation plating for tibial and femoral shaft nonunions after intramedullary nailing without nail removal has been proven very successful [12–14], there are only a few reported cases in the literature where this biomechanical concept was applied to humeral shaft nonunions. Therefore, the aim of this study was to review our results of augmentation plating of humeral shaft nonunions related to healing rates, complications and function.
51 years (range 20–84 years). The initial fractures were treated with retrograde nailing (10 cases) or anterograde nailing (27 cases). Two of the fractures had been described as initially open. A fracture type classification was not possible because radiographs before and/or shortly after nailing were only available for a few cases. The mean time period until the treatment of the nonunion with augmentation plating was 16 months (range 4–120 months). One case with a delayed union after 4 month was included in the current series. The X-rays demonstrated signs of a predictable nonunion as described for tibial nonunions after intramedullary nailing [15]: There was a lack of callus formation, gap distraction at the fracture site and radiographic loosening of the nail and locking bolts (Fig. 1). The right humerus was affected in 17 cases and the left in 20 cases. All nonunions were located in the middle third of the humeral shaft, with 29 nonunions at the diaphyseal middle and 8 at the proximal end of the middle third. The nonunions were classified as atrophic/oligotrophic in 34 cases and as hypertrophic in 3 cases. All but one patient had been referred to our hospital for nonunion repair, and 10 of those patients referred had been treated elsewhere for diaphyseal nonunion with at least one additional surgical procedure, such as removal of locking screws (‘‘dynamization of the nail’’), iliac crest bone grafting or
Materials and methods The electronic medical database at the authors’ institution was searched for all aseptic humeral shaft nonunions after failed intramedullary nailing from January 2003 to December 2014. All patients treated with augmentation plating without removing of the nail in this time period were included in the study if adequate postsurgical radiological follow-up was available. A healed nonunion was defined as a bridging callus and/or at least three visible cortices on the radiographs. X-rays were also examined for signs of implant failure, such as progressive loosening of the nail/locking screws or of the augmentation plates and screws, as an indirect sign of inadequate bone healing. Clinical data preoperatively and at the final follow-up had to contain fully documented shoulder and elbow motion as well as any symptoms of pain or dysfunction. In total, 40 patients were identified in the database, of whom 37 met the inclusion criteria and were evaluated in the study. Of the three patients with incomplete data, only immediate postoperative X-rays (2 cases) or a 3 month follow-up (one case) were available. None of the three could be reached by phone or mail for a further clinical and radiological follow up. The average age of the 14 women and 23 men at the time of admittance to our clinic for nonunion treatment was
123
Fig. 1 a Female patient with a delayed union 4 month after retrograde nailing; there is a distraction gap at the fracture site, lack of callus formation and radiographic loosening of the nail and locking bolts. b Final follow-up at 26 month after augmentation plating and bone grafting showing bony union (union was radiologically confirmed after 6 month)
Arch Orthop Trauma Surg Table 1 General information on patients and results Age (years, range)
51 (20–84)
Sex (male/female)
23/14
Side (right/left)
17/20
Nail (anterograde/retrograde)
27/10
Initial open/closed fracture
2/35
Type of nonunion (atrophic/hypertrophic)
34/3
Time between fracture and augmentation plating (month, range)
16 (4–120)
Operation time (min, range)
84 (52–150)
Neurovascular injury
1/37 (2.7 %)
Union rate (%)
36/37 (97 %)
Time to union (month, range) Follow up (month, range)
6 (3–24) 14 (3–48)
exchange nailing. Three patients had a documented radial palsy, one of whom had undergone open exploration of the radialis nerve before nonunion treatment. All three patients had a complete recovery of radial nerve function within 3 months of nailing or neurolysis. General information on the patients is summarized in Table 1. The contraindication for augmentation plating was a prominent nail causing restricted shoulder or elbow motion. Loosening of the nail with osteolysis around the nail, locking screws, and major gaps at the nonunion site were not contraindications for augmentation plating. However, none of the patients showed severe bone loss at the nonunion site that would have prevented a secure plate and screw fixation. The humeral nonunions were exposed via an anterior approach. A straight incision at the lateral border of the biceps muscle was made, and, after cutting the fascia, the biceps muscle was retracted medially, bringing the brachialis muscle into view. The brachialis muscle was split longitudinally throughout its lateral third and transected down to the bone. The nonunion zone was debrided by removing all fibrous tissue, interposed muscle and necrotic bone, followed by decortication, until bleeding bone was observed to provide a viable bed for the bone graft and to stimulate callus formation. The plates were placed on the anterior face of the humerus in all cases. A 4.5-millimetre limited contact dynamic compression plate (LCDCP) was used in most patients, including a broad plate with staggered holes in 11 cases and a narrow plate in 20 cases. In 2 cases, small fragment screws with washers were used in the 4.5-millimetre plates. This technique has been previously described for peri-prosthetic humeral fractures [16]. A 3.5-millimetre plate was used in five patients. A 4.5-millimetre locked compression plate (LCP) was used in one patient for initial nonunion treatment and in another for revision surgery with a periimplant fracture distal to the nail and a previously applied
Fig. 2 a Female patient with an atrophic nonunion 15 month after retrograde nailing; the proximal locking bolts had been removed 6 month after initial fracture treatment. b Final follow-up at 13 month after augmentation plating and bone grafting showing bony union
LCDCP (see ‘‘Results’’). The 4.5-millimetre LCDC plates ranged from 5 holes to 10 holes (mean 7 holes) with the use of 4–9 screws (mean 6 screws). The 3.5-mm LCDC plates ranged from 10 to 12 holes with a mean of 8 screws. The plates were, on average, 2–3 holes longer in oblique nonunions than in nonunions with a transverse fracture line. The length of the plate was chosen so that at least two screws could be applied proximal and distal to the nonunion site. Whenever possible, the screws were inserted bicortically and directed to either side of the nail. Additional interfragmentary compression screws were applied in four cases of oblique nonunion lines. Locking bolts were removed in three cases: in one the proximal screws and in two the distal screws were removed because of prominent screw heads. In three other cases, the locking screws had been removed elsewhere for ‘‘dynamization of the nail’’ without any other nonunion treatment (Fig. 2). With the exception of three cases, autologous cancellous bone grafts taken from the anterior iliac crest were used. In one case of a morbidly obese patient with chronic mycosis of the inguinal region, a demineralized bone matrix was applied because Hierholzer et al. [17] found no difference in the healing rates of humeral shaft nonunions using either autologous bone graft or demineralized bone matrix. The other two cases without additional bone grafting were hypertrophic nonunions (Fig. 3). The operations were performed by 16 different senior surgeons with a mean operation time of 84 min (range 52–150 min). Postoperatively, all patients were allowed early rehabilitation with free shoulder and elbow motion to avoid joint stiffness. Each patient was followed clinically and
123
Arch Orthop Trauma Surg
Fig. 3 a Male patient with a hypertrophic nonunion 12 month after anterograde nailing. b Fourteen month postoperative X-rays at final follow-up (union was radiologically confirmed after 3 month); the patient had been treated without additional bone grafting
radiologically at regular time intervals of 6 weeks and 3 months; after that, additional visits were scheduled at different intervals until bone union was radiologically confirmed.
Results In 36 patients (97 %), the nonunion healed both clinically and radiologically after a mean of 6 months (range 3–24 months). However, it should be noted that the exact healing interval could not be defined precisely due to the retrospective nature of the study. Cases with a prolonged healing time were observed at irregular intervals after the 3 month visit, so there is a possible bias concerning the maximum healing time. The mean clinical follow-up was 14 months (range 3–48 months). At the final follow-up, no restrictions of shoulder or elbow motion resulting from the surgery were documented. However, there was a relatively large group of patients who complained of a foreign body sensation around the anterior plate. Therefore, removal of both the intramedullary nail and the augmentation plate was performed in eight cases after a mean of 19 months (range 10–33 months). Still, clinical outcomes were good, as none of the patients had limitations on daily activities and all previously employed patients had returned to work by the final follow-up. There was one persistent hypertrophic nonunion that had been treated with a 5-hole, 4.5millimetre LCDCP with small fragment screws and washers (Fig. 4). Despite the hypertrophic character of the nonunion, iliac crest bone grafting had initially been applied. There had been severe osteolysis around the nail
123
Fig. 4 Female patient with a persistent nonunion after augmentation plating with a 4.5-ml LCDCP and small fragment screws with washers
and distal locking bolts. At the latest follow-up, 37 month postoperatively, X-rays as well as a CT scan showed a thin persistent nonunion line but without progression of the excessive loosening of the nail and locking bolts. The augmentation plate and screws did not show any signs of loosening. Because the patient was free of pain or a feeling of instability, had free shoulder and elbow function and refused revision surgery, the nonunion has been left in the current state. The patient is still in regular follow-up. Due to a fall, a 54-year-old patient sustained a horizontal peri-implant stress fracture at the distal end of the nail and plate with a newly occurred radial palsy 8 weeks after augmentation plating. Both the nail and augmentation plate ended at the same height, resulting in a predetermined breaking point of the humeral shaft. Revision surgery included exchanging the anterior LCDCP for a longer LCP to bridge the nonunion and the distal fracture and reapplying new bone graft. The nonunion and the peri-implant fracture healed within 10 months, and the radial palsy showed a complete spontaneous recovery after 3 months. We did not see any iatrogenic radial palsy in this study, but there was one case of transient median nerve palsy that also recovered spontaneously after 6 weeks without any permanent neurological deficit. No infections or wound healing complications developed.
Arch Orthop Trauma Surg
Discussion A shaft nonunion in long bone fractures after intramedullary nailing is a rare complication, especially since the introduction of locking nails. However, it has been indicated that the treatment of diaphyseal nonunions after failed locked nailing of the humerus is not analogous to similar problems observed in the tibia or femur [6]. While in the lower limb, body weight and smaller torsional forces contribute to dynamic loading, the humerus is subject to more rotatory forces and minor axial loading [18]. Humerus interlocking nails have been designed to provide better rotational stability than earlier nail designs [19], although they do not match the torsional stiffness of a plate fixation [18]. Still, rotational instability combined with gap distraction at the fracture site contributes to the majority of humeral shaft nonunions after intramedullary nailing [20]. Bone loss around the nail and locking screws caused by a loose nail complicates the reconstructive surgical procedure and increases the risk of secondary fractures. In addition, removal of humeral nails can be associated with iatrogenic damage to the rotator cuff and further bone loss if the nail has sunk deep below the cortex or is fitted tightly in the proximal or distal bone fragment. Considering these specific problems, techniques resulting in bone consolidation without removal of the IM nail might simplify reconstructive surgery of the humeral shaft. Traditionally, removal of the intramedullary device followed by open compression plating has been advocated and has shown high healing rates [5, 21]. Lin et al. [22] presented a large study of 86 patients with humeral shaft nonunions managed by removal of previous implants, open reduction and internal fixation with dynamic compression plating, supplemented with cancellous bone grafts. Of these patients, 38 were primarily treated with IM nailing. All nonunions healed within an average of 18 weeks. McKee et al. [5] reported the treatment results of nail removal and plating (9 patients) and exchange nailing (10 patients). Union was achieved in all patients with plating and bone grafting but in only 4 of 10 patients with exchange nailing. Exchange nailing alone appears to be ineffective, as Flinkkila et al. [6] found a similar low healing rate of nonunions in 6 of 13 patients. Better results have been reported when exchange nailing is combined with decortication, fibrous tissue removal and bone grafting. Lin et al. [23] reported on 23 patients treated by open exchange locked nailing with bone grafting, with additional compression wiring in 19 of those patients. All but one nonunion healed. A recently published study comparing interlocking nails and locking compression plates (20 patients in each group) found no significant differences in the treatment of humeral shaft nonunions [3]. There were 4
patients in nail group and 3 patients in the LCP group who had first undergone intramedullary nailing. All of these nonunions healed uneventfully with bone grafting and either exchange nailing or LCP fixation [3]. One persistent nonunion appeared in the nail group. The nonunion healed after revision surgery with nail removal, followed by open locked compression plating and bone grafting [3]. Different techniques for nonunion treatment without removal of the IM nail have been described. Apart et al. [24] treated seven hypertrophic humeral nonunions using a compression device to apply isolated interfragmentary compression over the initially used nail. Neither debridement nor additional bone grafting was performed, but an initial distraction gap at the fracture site was noted in each case [24]. A technically demanding treatment option is the application of an Ilizarov external frame. The technique relies on gradual compression at the nonunion site and can be achieved after removal either of the IM nail itself [25, 26] or of the distal locking screws while leaving the nail in situ [9, 27]. For both procedures, high union rates have been reported without the need for additional bone grafting, even in cases of bone loss and atrophic nonunions. However, disadvantages are the bulky construction of the circular frame and the high incidence of local pin site infections. In contrast to the above-described techniques, which rely on adding compression to the nonunion site, the use of a wave-like contoured plate and bone grafting without removal of the IM nail has also been reported [8, 11, 21]. The wave plate spans the nonunion site and is fixed far off proximally and distally to the humerus. Marti et al. treated two patients successfully with wave plates when removal of the nail during the nonunion repair was considered to be too hazardous [21]. Gerber et al. [11] used wave plating in 6 patients and reported healing at a mean of 16 weeks. They documented one case of radial nerve palsy as the only complication in their study [11]. Relatively long plates are needed for flexible bridging of the nonunion, resulting in healing by callus, and the wave configuration results in a very prominent implant. However, no adverse effects have been described in the above-cited case studies [11, 21]. Reliable results have been reported for the technique of augmentation plating after IM nailing for nonunions in the lower extremities [12, 14, 28, 29]. The technique takes advantage of the high axial stiffness resulting from in situ nails and uses short plates to minimize adverse rotational micromovements without compressing the nonunion site. Nadkarni et al. [14] used locked compression plates with bone grafting without removing the nails in 2 patients with humeral nonunions. Both cases achieved union within 6 months. Wu et al. [30] demonstrated that sufficient rotational stability can be achieved even with smaller
123
Arch Orthop Trauma Surg
implants and without the need for a locking mechanism. The authors augmented the IM nails with small staples at the nonunion site, combined with bone grafting, in two patients and reported uneventful healing after 5 and 7 months [30]. Concurrent to these very short staple implants, the short plates that were used in our study showed sufficient supplemental stability with the nail left in situ. Larger plates with at least 10–12 holes that bridge or compress the nonunion site and with at least 8 cortices engaged on both sides of the nonunion with additional interfragmentary screws, are recommended in cases using only plate fixation for humeral shaft nonunions [3, 8, 31– 35]. In cases of doubtful stability, a second plate in a 90° plane or, alternatively, strong onlay cortical bone grafts or intramedullary fibular allografts have been described to enhance stability at the nonunion site [32, 36, 37]. Rubel et al. [32] demonstrated the highest axial and rotational in vitro stiffness using the two-plate construct, with a healing rate of 89 % in 18 patients. No negative effects on the healing results have been reported due to the possible greater amount of bone and soft tissue exposure with the addition of a second plate [32]. Similarly, soft-tissue management and devascularization of the nonunion site are theoretical concerns of augmentation plating. However, the combination of intramedullary and extramedullary fixation devices has shown no negative effects on bone healing in the lower extremities [12, 13, 28, 29]. Furthermore, unlike in the treatment of acute fractures, we believe that extensive debridement of the atrophic nonunion sites with bone grafting is one of the key factors of successful treatment and, therefore, makes opening the nonunion site necessary. A meta-analysis by Kontakis et al. [31] found bone grafting to be a crucial factor for successful treatment, independent of the type of nonunion (atrophic or hypertrophic). However, the hypertrophic nonunions in our series responded to stability alone without the need of additional bone grafting. By using short augmentation plates, extensive soft tissue exposure can be avoided. In addition, the anterior approach that was used on all patients in this study avoids the radial nerve, which is often embedded in scar tissue at the posterior site of the humerus, without the need for visualization and exploration of the nerve [34]. Another major concern may be the risk of stress riser fractures when combining IM nails and plates. This specific problem has been particularly described for femoral fractures in elderly patients [38–40]. There was one patient with a peri-implant fracture after a low energy fall in our series. Critically stated, in this patient the plate and nail tip acted as a stress riser because both implants ended at the same level. To minimize the stress rising effect, ending of the augmentation plate near the IM nail tip need to be avoided.
123
It should be noted that none of our patients had a prominent nail that restricted shoulder or elbow motion. In these cases, nail removal is indicated to facilitate free joint function, and open plating is recommended [11, 41]. We did not detect any adverse effects from anterior plating on shoulder or elbow function in our patients. These functional results are concurrent with the report by Livani et al. [34], who treated 15 humeral nonunions by anterior plating (12 patients were initially treated nonoperatively, and 3 were treated with plate osteosynthesis). Furthermore, leaving the nail in situ eliminates the risk of iatrogenic damage, especially to the shoulder, and may reduce operation time compared to revision of humeral nonunions by removal of previous implants and plates [22]. Another constraint that should be noted is that none of the patients in this study showed severely compromised bone or osteolysis at the nonunion site, which would have made stable fixation with short plates impossible. In these cases, the use of long wave plates, fibular allografts, vascularized autografts, bone morphogenic proteins and mesenchymal stem cells have been proposed [8, 36, 42, 43]. The main limitations of the current study are the retrospective evaluation without a defined follow-up protocol and the evaluation of operations performed by many different surgeons with heterogeneous experience. The final follow-up of most patients ended with a diagnosis of a healed nonunion. As we were able to contact only a few patients, the number of implant removals due to a foreign body sensation of the anterior plate may be even higher, and other possible late complications might have been overlooked. Still, there is little literature on humeral nonunion treatment after intramedullary nailing, especially using a standardized, uniform surgical technique. The fact that the number of surgeons and their level of experience did not affect the outcome in respect to healing, function and complications underlines the safety of the procedure and the reproducibility of the results, rather than being a limitation when interpreting the study. In conclusion, our results indicate that the technique of augmentation plating using an anterior approach is a safe and reliable option for humeral shaft nonunions after failed nailing. Because the healing rates are similar to those of the standard technique of nail removal and fixation by compression or locking plates, we consider the technique as an alternative choice of treatment.
References 1. Ouyang H, Xiong J, Xiang P, Cui Z, Chen L, Yu B (2013) Plate versus intramedullary nail fixation in the treatment of humeral shaft fractures: an updated meta-analysis. J Shoulder Elbow Surg
Arch Orthop Trauma Surg
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
22(3):387–395. doi:10.1016/j.jse.2012.06.007 (S10582746(12)00270-4 [pii]) Kurup H, Hossain M, Andrew JG (2011) Dynamic compression plating versus locked intramedullary nailing for humeral shaft fractures in adults. Cochrane Database Syst Rev 6:CD005959. doi:10.1002/14651858.CD005959.pub2 Singh AK, Arun GR, Narsaria N, Srivastava A (2014) Treatment of non-union of humerus diaphyseal fractures: a prospective study comparing interlocking nail and locking compression plate. Arch Orthop Trauma Surg 134(7):947–953. doi:10.1007/s00402014-1973-0 Volgas DA, Stannard JP, Alonso JE (2004) Nonunions of the humerus. Clin Orthop Relat Res 419:46–50 (00003086200402000-00008 [pii]) McKee MD, Miranda MA, Riemer BL, Blasier RB, Redmond BJ, Sims SH, Waddell JP, Jupiter JB (1996) Management of humeral nonunion after the failure of locking intramedullary nails. J Orthop Trauma 10(7):492–499 Flinkkila T, Ristiniemi J, Hamalainen M (2001) Nonunion after intramedullary nailing of humeral shaft fractures. J Trauma 50(3):540–544 Catagni MA, Guerreschi F, Probe RA (1991) Treatment of humeral nonunions with the Ilizarov technique. Bull Hosp Jt Dis Orthop Inst 51(1):74–83 Ring D, Jupiter JB, Quintero J, Sanders RA, Marti RK (2000) Atrophic ununited diaphyseal fractures of the humerus with a bony defect: treatment by wave-plate osteosynthesis. J Bone Joint Surg Br 82(6):867–871 Patel VR, Menon DK, Pool RD, Simonis RB (2000) Nonunion of the humerus after failure of surgical treatment. Management using the Ilizarov circular fixator. J Bone Joint Surg Br 82(7):977–983 Kontakis GM, Papadokostakis GM, Alpantaki K, Chlouverakis G, Hadjipavlou AG, Giannoudis PV (2006) Intramedullary nailing for non-union of the humeral diaphysis: a review. Injury 37(10):953–960. doi:10.1016/j.injury.2006.02.050 (S00201383(06)00120-3 [pii]) Gerber A, Marti R, Jupiter J (2003) Surgical management of diaphyseal humeral nonunion after intramedullary nailing: waveplate fixation and autologous bone grafting without nail removal. J Shoulder Elbow Surg 12(4):309–313. doi:10.1016/mse.2003. S1058274603000442 (S1058274603000442 [pii]) Roetman B, Scholz N, Muhr G, Mollenhoff G (2008) Augmentive plate fixation in femoral non-unions after intramedullary nailing. Strategy after unsuccessful intramedullary nailing of the femur. Z Orthop Unfall 146(5):586–590. doi:10.1055/s-20081038526 Birjandinejad A, Ebrahimzadeh MH, Ahmadzadeh-Chabock H (2009) Augmentation plate fixation for the treatment of femoral and tibial nonunion after intramedullary nailing. Orthopedics 32(6):409. doi:10.3928/01477447-20090511-12 (orthopedics.40080 [pii]) Nadkarni B, Srivastav S, Mittal V, Agarwal S (2008) Use of locking compression plates for long bone nonunions without removing existing intramedullary nail: review of literature and our experience. J Trauma 65(2):482–486. doi:10.1097/TA. 0b013e31817c9905 (00005373-200808000-00041 [pii]) Yang JS, Otero J, McAndrew CM, Ricci WM, Gardner MJ (2013) Can tibial nonunion be predicted at 3 months after intramedullary nailing? J Orthop Trauma 27(11):599–603. doi:10.1097/BOT. 0b013e31828f5821 Seybold D, Citak M, Konigshausen M, Gessmann J, Schildhauer TA (2011) Combining of small fragment screws and large fragment plates for open reduction and internal fixation of periprosthetic humeral fractures. Int J Shoulder Surg 5(4):105–107. doi:10.4103/0973-6042.91004
17. Hierholzer C, Sama D, Toro JB, Peterson M, Helfet DL (2006) Plate fixation of ununited humeral shaft fractures: effect of type of bone graft on healing. J Bone Joint Surg Am 88(7):1442–1447. doi:10.2106/JBJS.E.00332 (88/7/1442 [pii]) 18. Blum J, Machemer H, Baumgart F, Schlegel U, Wahl D, Rommens PM (1999) Biomechanical comparison of bending and torsional properties in retrograde intramedullary nailing of humeral shaft fractures. J Orthop Trauma 13(5):344–350 19. Rosenbaum A, Uhl R (2012) Nonunion of humeral shaft fractures following flexible nailing fixation. Orthopedics 35(6):512–515. doi:10.3928/01477447-20120525-08 20. Baltov A, Mihail R, Dian E (2014) Complications after interlocking intramedullary nailing of humeral shaft fractures. Injury 45(Suppl 1):S9–S15. doi:10.1016/j.injury.2013.10.044 21. Marti RK, Verheyen CC, Besselaar PP (2002) Humeral shaft nonunion: evaluation of uniform surgical repair in fifty-one patients. J Orthop Trauma 16(2):108–115 22. Lin CL, Fang CK, Chiu FY, Chen CM, Chen TH (2009) Revision with dynamic compression plate and cancellous bone graft for aseptic nonunion after surgical treatment of humeral shaft fracture. J Trauma 67(6):1393–1396. doi:10.1097/TA. 0b013e31818c1595 (00005373-200912000-00043 [pii]) 23. Lin J, Chiang H, Hou SM (2003) Open exchange locked nailing in humeral nonunions after intramedullary nailing. Clin Orthop Relat Res 411:260–268. doi:10.1097/01.blo.0000069898.31220. 9e 24. Apard T, Ducellier F, Hubert L, Talha A, Cronier P, Bizot P (2010) Isolated interfragmentary compression for nonunion of humeral shaft fractures initially treated by nailing: a preliminary report of seven cases. Injury 41(12):1262–1265. doi:10.1016/j. injury.2010.05.023 (S0020-1383(10)00319-0 [pii]) 25. Raschke M, Khodadadyan C, Maitino PD, Hoffmann R, Sudkamp NP (1998) Nonunion of the humerus following intramedullary nailing treated by Ilizarov hybrid fixation. J Orthop Trauma 12(2):138–141 26. Tomic S, Bumbasirevic M, Lesic A, Mitkovic M, Atkinson HD (2007) Ilizarov frame fixation without bone graft for atrophic humeral shaft nonunion: 28 patients with a minimum 2-year follow-up. J Orthop Trauma 21(8):549–556. doi:10.1097/BOT. 0b013e31814612c8 (00005131-200709000-00006 [pii]) 27. Menon DK, Dougall TW, Pool RD, Simonis RB (2002) Augmentative Ilizarov external fixation after failure of diaphyseal union with intramedullary nailing. J Orthop Trauma 16(7):491–497 28. Ueng SW, Chao EK, Lee SS, Shih CH (1997) Augmentative plate fixation for the management of femoral nonunion after intramedullary nailing. J Trauma 43(4):640–644 29. Choi YS, Kim KS (2005) Plate augmentation leaving the nail in situ and bone grafting for non-union of femoral shaft fractures. Int Orthop 29(5):287–290. doi:10.1007/s00264-005-0668-0 30. Wu CC, Shih CH, Chen WJ, Tai CL (1998) Staple augmentation to treat a humeral nonunion after failed Seidel locked nailing. Arch Orthop Trauma Surg 118(1–2):42–44 31. Kontakis GM, Tosounidis T, Pagkalos J (2007) Humeral diaphyseal aseptic non-unions: an algorithm of management. Injury 38(Suppl 2):S39–S49 (doi:S0020-1383(07)80008-8 [pii]) 32. Rubel IF, Kloen P, Campbell D, Schwartz M, Liew A, Myers E, Helfet DL (2002) Open reduction and internal fixation of humeral nonunions: a biomechanical and clinical study. J Bone Joint Surg Am 84-A(8):1315–1322 33. Cadet ER, Yin B, Schulz B, Ahmad CS, Rosenwasser MP (2013) Proximal humerus and humeral shaft nonunions. J Am Acad Orthop Surg 21(9):538–547. doi:10.5435/JAAOS-21-09-538 (21/ 9/538 [pii]) 34. Livani B, Belangero W, Medina G, Pimenta C, Zogaib R, Mongon M (2010) Anterior plating as a surgical alternative in the
123
Arch Orthop Trauma Surg
35.
36.
37.
38.
treatment of humeral shaft non-union. Int Orthop 34(7):1025–1031. doi:10.1007/s00264-009-0863-5 Martinez AA, Cuenca J, Herrera A (2009) Two-plate fixation for humeral shaft non-unions. J Orthop Surg (Hong Kong) 17(2):135–138 Willis MP, Brooks JP, Badman BL, Gaines RJ, Mighell MA, Sanders RW (2013) Treatment of atrophic diaphyseal humeral nonunions with compressive locked plating and augmented with an intramedullary strut allograft. J Orthop Trauma 27(2):77–81. doi:10.1097/BOT.0b013e31825360fa Hornicek FJ, Zych GA, Hutson JJ, Malinin TI (2001) Salvage of humeral nonunions with onlay bone plate allograft augmentation. Clin Orthop Relat Res 386:203–209 Leibner ED, Mosheiff R, Safran O, Abu-Snieneh K, Liebergall M (1999) Femoral fracture at the proximal end of an intramedullary supracondylar nail: a case report. Am J Orthop (Belle Mead NJ) 28(1):53–55
123
39. Harris T, Ruth JT, Szivek J, Haywood B (2003) The effect of implant overlap on the mechanical properties of the femur. J Trauma 54(5):930–935. doi:10.1097/01.TA.0000060999.54287. 39 40. Kim HK, Noh KC, Chung KJ, Hwang JH (2012) Use of Huckstep nail in the periimplant femoral shaft fracture. Indian J Orthop 46(6):718–720. doi:10.4103/0019-5413.104240 41. Ring D, Jupiter J (2003) Humerus nonunion after intramedullary rod fixation. Tech Orthop 18(4):356–359 42. Wright TW (1997) Treatment of humeral diaphyseal nonunions in patients with severely compromised bone. J South Orthop Assoc 6(1):1–7 43. Murena L, Canton G, Vulcano E, Surace MF, Cherubino P (2014) Treatment of humeral shaft aseptic nonunions in elderly patients with opposite structural allograft, BMP-7, and mesenchymal stem cells. Orthopedics 37(2):e201–e206. doi:10.3928/0147744720140124-26
TRAUMA SURGERY
Anterior augmentation plating of aseptic humeral shaft nonunions after intramedullary nailing Jan Gessmann1 • Matthias Ko¨nigshausen1 • Marlon Osman Coulibaly1 Thomas Armin Schildhauer1 • Dominik Seybold1
•
Received: 19 November 2015 Ó Springer-Verlag Berlin Heidelberg 2016
Abstract Introduction Humeral shaft nonunion after intramedullary nailing is a rare but serious complication. Treatment options include implant removal, open plating, exchange nailing and external fixation. The objective of this retrospective study was to determine whether augmentation plating without nail removal is feasible for treating a humeral shaft nonunion. Materials and methods Between 2002 and 2014, 37 patients (mean age 51, range 20–84 years) with aseptic humeral shaft nonunions prior to intramedullary nailing were treated with augmentation plating. The initial fractures had been fixed with retrograde nails (10 cases) or anterograde nails (27 cases). There were 34 atrophic nonunions and 3 hypertrophic nonunions. Nonunion treatment of all patients consisted of local debridement through an anterior approach to the humerus and anterior placement of the augmentation plates. Supplemental bone grafting was performed in all atrophic nonunion cases. All patients were followed until union was radiologically confirmed.
Results Union was achieved in 36 patients (97 %) after a mean of 6 months (range 3–24 months). There was one case of iatrogenic median nerve palsy that showed complete spontaneous recovery 6 weeks postoperatively. One patient sustained a peri-implant stress fracture that was treated successfully by exchanging the augmentation plate to bridge the nonunion and the fracture. No infections or wound healing complications developed. At a mean follow-up of 14 months, all patients showed free shoulder and elbow motion and no restrictions in daily or working life. Conclusions The results indicate that augmentation plating using an anterior approach is a safe and reliable option for humeral shaft nonunions after failed nailing, and the treatment has no substantial complications. Because the healing rates are similar to the standard technique of nail removal and fixation by compression or locking plates, we consider this technique to be an alternative choice for treatment. Keywords Humeral shaft nonunion Augmentation plating Intramedullary nailing Bone grafting Anterior approach
& Jan Gessmann [email protected] Matthias Ko¨nigshausen [email protected]
Introduction
Marlon Osman Coulibaly [email protected]
Intramedullary (IM) nails are frequently used in the treatment of humeral shaft fractures, especially since the development of newer nailing techniques and locking mechanisms. Recently published meta-analyses comparing the results of IM nailing with the gold standard of open reduction and plating found comparable high union rates [1, 2]. Still, nonunions of the humeral shaft occur, depending on fracture type, osteoporosis, and soft tissue damage as well as inadequate reduction, distraction of
Thomas Armin Schildhauer [email protected] Dominik Seybold [email protected] 1
Department of Trauma Surgery, BG University Hospital Bergmannsheil, Ruhr University Bochum, Bu¨rkle-de-laCamp-Platz 1, 44789 Bochum, Germany
123
Arch Orthop Trauma Surg
fragments and unstable fixation [3, 4]. Humeral nonunions after prior intramedullary nailing can be difficult to address due to possible bone loss around a loose nail or locking bolts, which complicates the reconstructive procedure and leads to poorer operative results compared to prior nonoperative treatment or plate fixation [5]. If untreated, osteolysis and bone loss around the nail and locking screws increase the risk of secondary fractures [6]. A permanent nonunion leads to severe restrictions of shoulder and elbow function and may lead to a major disability, especially in elderly patients who depend on their arm support to use canes, crutches or wheeled walkers [6]. Different treatment techniques have been described for the specific reconstructive problems, including open reduction and plating with removal of the nail, exchange intramedullary nailing, Ilizarov external fixation and the application of a wave plate [5, 7–11]. Although the technique of augmentation plating for tibial and femoral shaft nonunions after intramedullary nailing without nail removal has been proven very successful [12–14], there are only a few reported cases in the literature where this biomechanical concept was applied to humeral shaft nonunions. Therefore, the aim of this study was to review our results of augmentation plating of humeral shaft nonunions related to healing rates, complications and function.
51 years (range 20–84 years). The initial fractures were treated with retrograde nailing (10 cases) or anterograde nailing (27 cases). Two of the fractures had been described as initially open. A fracture type classification was not possible because radiographs before and/or shortly after nailing were only available for a few cases. The mean time period until the treatment of the nonunion with augmentation plating was 16 months (range 4–120 months). One case with a delayed union after 4 month was included in the current series. The X-rays demonstrated signs of a predictable nonunion as described for tibial nonunions after intramedullary nailing [15]: There was a lack of callus formation, gap distraction at the fracture site and radiographic loosening of the nail and locking bolts (Fig. 1). The right humerus was affected in 17 cases and the left in 20 cases. All nonunions were located in the middle third of the humeral shaft, with 29 nonunions at the diaphyseal middle and 8 at the proximal end of the middle third. The nonunions were classified as atrophic/oligotrophic in 34 cases and as hypertrophic in 3 cases. All but one patient had been referred to our hospital for nonunion repair, and 10 of those patients referred had been treated elsewhere for diaphyseal nonunion with at least one additional surgical procedure, such as removal of locking screws (‘‘dynamization of the nail’’), iliac crest bone grafting or
Materials and methods The electronic medical database at the authors’ institution was searched for all aseptic humeral shaft nonunions after failed intramedullary nailing from January 2003 to December 2014. All patients treated with augmentation plating without removing of the nail in this time period were included in the study if adequate postsurgical radiological follow-up was available. A healed nonunion was defined as a bridging callus and/or at least three visible cortices on the radiographs. X-rays were also examined for signs of implant failure, such as progressive loosening of the nail/locking screws or of the augmentation plates and screws, as an indirect sign of inadequate bone healing. Clinical data preoperatively and at the final follow-up had to contain fully documented shoulder and elbow motion as well as any symptoms of pain or dysfunction. In total, 40 patients were identified in the database, of whom 37 met the inclusion criteria and were evaluated in the study. Of the three patients with incomplete data, only immediate postoperative X-rays (2 cases) or a 3 month follow-up (one case) were available. None of the three could be reached by phone or mail for a further clinical and radiological follow up. The average age of the 14 women and 23 men at the time of admittance to our clinic for nonunion treatment was
123
Fig. 1 a Female patient with a delayed union 4 month after retrograde nailing; there is a distraction gap at the fracture site, lack of callus formation and radiographic loosening of the nail and locking bolts. b Final follow-up at 26 month after augmentation plating and bone grafting showing bony union (union was radiologically confirmed after 6 month)
Arch Orthop Trauma Surg Table 1 General information on patients and results Age (years, range)
51 (20–84)
Sex (male/female)
23/14
Side (right/left)
17/20
Nail (anterograde/retrograde)
27/10
Initial open/closed fracture
2/35
Type of nonunion (atrophic/hypertrophic)
34/3
Time between fracture and augmentation plating (month, range)
16 (4–120)
Operation time (min, range)
84 (52–150)
Neurovascular injury
1/37 (2.7 %)
Union rate (%)
36/37 (97 %)
Time to union (month, range) Follow up (month, range)
6 (3–24) 14 (3–48)
exchange nailing. Three patients had a documented radial palsy, one of whom had undergone open exploration of the radialis nerve before nonunion treatment. All three patients had a complete recovery of radial nerve function within 3 months of nailing or neurolysis. General information on the patients is summarized in Table 1. The contraindication for augmentation plating was a prominent nail causing restricted shoulder or elbow motion. Loosening of the nail with osteolysis around the nail, locking screws, and major gaps at the nonunion site were not contraindications for augmentation plating. However, none of the patients showed severe bone loss at the nonunion site that would have prevented a secure plate and screw fixation. The humeral nonunions were exposed via an anterior approach. A straight incision at the lateral border of the biceps muscle was made, and, after cutting the fascia, the biceps muscle was retracted medially, bringing the brachialis muscle into view. The brachialis muscle was split longitudinally throughout its lateral third and transected down to the bone. The nonunion zone was debrided by removing all fibrous tissue, interposed muscle and necrotic bone, followed by decortication, until bleeding bone was observed to provide a viable bed for the bone graft and to stimulate callus formation. The plates were placed on the anterior face of the humerus in all cases. A 4.5-millimetre limited contact dynamic compression plate (LCDCP) was used in most patients, including a broad plate with staggered holes in 11 cases and a narrow plate in 20 cases. In 2 cases, small fragment screws with washers were used in the 4.5-millimetre plates. This technique has been previously described for peri-prosthetic humeral fractures [16]. A 3.5-millimetre plate was used in five patients. A 4.5-millimetre locked compression plate (LCP) was used in one patient for initial nonunion treatment and in another for revision surgery with a periimplant fracture distal to the nail and a previously applied
Fig. 2 a Female patient with an atrophic nonunion 15 month after retrograde nailing; the proximal locking bolts had been removed 6 month after initial fracture treatment. b Final follow-up at 13 month after augmentation plating and bone grafting showing bony union
LCDCP (see ‘‘Results’’). The 4.5-millimetre LCDC plates ranged from 5 holes to 10 holes (mean 7 holes) with the use of 4–9 screws (mean 6 screws). The 3.5-mm LCDC plates ranged from 10 to 12 holes with a mean of 8 screws. The plates were, on average, 2–3 holes longer in oblique nonunions than in nonunions with a transverse fracture line. The length of the plate was chosen so that at least two screws could be applied proximal and distal to the nonunion site. Whenever possible, the screws were inserted bicortically and directed to either side of the nail. Additional interfragmentary compression screws were applied in four cases of oblique nonunion lines. Locking bolts were removed in three cases: in one the proximal screws and in two the distal screws were removed because of prominent screw heads. In three other cases, the locking screws had been removed elsewhere for ‘‘dynamization of the nail’’ without any other nonunion treatment (Fig. 2). With the exception of three cases, autologous cancellous bone grafts taken from the anterior iliac crest were used. In one case of a morbidly obese patient with chronic mycosis of the inguinal region, a demineralized bone matrix was applied because Hierholzer et al. [17] found no difference in the healing rates of humeral shaft nonunions using either autologous bone graft or demineralized bone matrix. The other two cases without additional bone grafting were hypertrophic nonunions (Fig. 3). The operations were performed by 16 different senior surgeons with a mean operation time of 84 min (range 52–150 min). Postoperatively, all patients were allowed early rehabilitation with free shoulder and elbow motion to avoid joint stiffness. Each patient was followed clinically and
123
Arch Orthop Trauma Surg
Fig. 3 a Male patient with a hypertrophic nonunion 12 month after anterograde nailing. b Fourteen month postoperative X-rays at final follow-up (union was radiologically confirmed after 3 month); the patient had been treated without additional bone grafting
radiologically at regular time intervals of 6 weeks and 3 months; after that, additional visits were scheduled at different intervals until bone union was radiologically confirmed.
Results In 36 patients (97 %), the nonunion healed both clinically and radiologically after a mean of 6 months (range 3–24 months). However, it should be noted that the exact healing interval could not be defined precisely due to the retrospective nature of the study. Cases with a prolonged healing time were observed at irregular intervals after the 3 month visit, so there is a possible bias concerning the maximum healing time. The mean clinical follow-up was 14 months (range 3–48 months). At the final follow-up, no restrictions of shoulder or elbow motion resulting from the surgery were documented. However, there was a relatively large group of patients who complained of a foreign body sensation around the anterior plate. Therefore, removal of both the intramedullary nail and the augmentation plate was performed in eight cases after a mean of 19 months (range 10–33 months). Still, clinical outcomes were good, as none of the patients had limitations on daily activities and all previously employed patients had returned to work by the final follow-up. There was one persistent hypertrophic nonunion that had been treated with a 5-hole, 4.5millimetre LCDCP with small fragment screws and washers (Fig. 4). Despite the hypertrophic character of the nonunion, iliac crest bone grafting had initially been applied. There had been severe osteolysis around the nail
123
Fig. 4 Female patient with a persistent nonunion after augmentation plating with a 4.5-ml LCDCP and small fragment screws with washers
and distal locking bolts. At the latest follow-up, 37 month postoperatively, X-rays as well as a CT scan showed a thin persistent nonunion line but without progression of the excessive loosening of the nail and locking bolts. The augmentation plate and screws did not show any signs of loosening. Because the patient was free of pain or a feeling of instability, had free shoulder and elbow function and refused revision surgery, the nonunion has been left in the current state. The patient is still in regular follow-up. Due to a fall, a 54-year-old patient sustained a horizontal peri-implant stress fracture at the distal end of the nail and plate with a newly occurred radial palsy 8 weeks after augmentation plating. Both the nail and augmentation plate ended at the same height, resulting in a predetermined breaking point of the humeral shaft. Revision surgery included exchanging the anterior LCDCP for a longer LCP to bridge the nonunion and the distal fracture and reapplying new bone graft. The nonunion and the peri-implant fracture healed within 10 months, and the radial palsy showed a complete spontaneous recovery after 3 months. We did not see any iatrogenic radial palsy in this study, but there was one case of transient median nerve palsy that also recovered spontaneously after 6 weeks without any permanent neurological deficit. No infections or wound healing complications developed.
Arch Orthop Trauma Surg
Discussion A shaft nonunion in long bone fractures after intramedullary nailing is a rare complication, especially since the introduction of locking nails. However, it has been indicated that the treatment of diaphyseal nonunions after failed locked nailing of the humerus is not analogous to similar problems observed in the tibia or femur [6]. While in the lower limb, body weight and smaller torsional forces contribute to dynamic loading, the humerus is subject to more rotatory forces and minor axial loading [18]. Humerus interlocking nails have been designed to provide better rotational stability than earlier nail designs [19], although they do not match the torsional stiffness of a plate fixation [18]. Still, rotational instability combined with gap distraction at the fracture site contributes to the majority of humeral shaft nonunions after intramedullary nailing [20]. Bone loss around the nail and locking screws caused by a loose nail complicates the reconstructive surgical procedure and increases the risk of secondary fractures. In addition, removal of humeral nails can be associated with iatrogenic damage to the rotator cuff and further bone loss if the nail has sunk deep below the cortex or is fitted tightly in the proximal or distal bone fragment. Considering these specific problems, techniques resulting in bone consolidation without removal of the IM nail might simplify reconstructive surgery of the humeral shaft. Traditionally, removal of the intramedullary device followed by open compression plating has been advocated and has shown high healing rates [5, 21]. Lin et al. [22] presented a large study of 86 patients with humeral shaft nonunions managed by removal of previous implants, open reduction and internal fixation with dynamic compression plating, supplemented with cancellous bone grafts. Of these patients, 38 were primarily treated with IM nailing. All nonunions healed within an average of 18 weeks. McKee et al. [5] reported the treatment results of nail removal and plating (9 patients) and exchange nailing (10 patients). Union was achieved in all patients with plating and bone grafting but in only 4 of 10 patients with exchange nailing. Exchange nailing alone appears to be ineffective, as Flinkkila et al. [6] found a similar low healing rate of nonunions in 6 of 13 patients. Better results have been reported when exchange nailing is combined with decortication, fibrous tissue removal and bone grafting. Lin et al. [23] reported on 23 patients treated by open exchange locked nailing with bone grafting, with additional compression wiring in 19 of those patients. All but one nonunion healed. A recently published study comparing interlocking nails and locking compression plates (20 patients in each group) found no significant differences in the treatment of humeral shaft nonunions [3]. There were 4
patients in nail group and 3 patients in the LCP group who had first undergone intramedullary nailing. All of these nonunions healed uneventfully with bone grafting and either exchange nailing or LCP fixation [3]. One persistent nonunion appeared in the nail group. The nonunion healed after revision surgery with nail removal, followed by open locked compression plating and bone grafting [3]. Different techniques for nonunion treatment without removal of the IM nail have been described. Apart et al. [24] treated seven hypertrophic humeral nonunions using a compression device to apply isolated interfragmentary compression over the initially used nail. Neither debridement nor additional bone grafting was performed, but an initial distraction gap at the fracture site was noted in each case [24]. A technically demanding treatment option is the application of an Ilizarov external frame. The technique relies on gradual compression at the nonunion site and can be achieved after removal either of the IM nail itself [25, 26] or of the distal locking screws while leaving the nail in situ [9, 27]. For both procedures, high union rates have been reported without the need for additional bone grafting, even in cases of bone loss and atrophic nonunions. However, disadvantages are the bulky construction of the circular frame and the high incidence of local pin site infections. In contrast to the above-described techniques, which rely on adding compression to the nonunion site, the use of a wave-like contoured plate and bone grafting without removal of the IM nail has also been reported [8, 11, 21]. The wave plate spans the nonunion site and is fixed far off proximally and distally to the humerus. Marti et al. treated two patients successfully with wave plates when removal of the nail during the nonunion repair was considered to be too hazardous [21]. Gerber et al. [11] used wave plating in 6 patients and reported healing at a mean of 16 weeks. They documented one case of radial nerve palsy as the only complication in their study [11]. Relatively long plates are needed for flexible bridging of the nonunion, resulting in healing by callus, and the wave configuration results in a very prominent implant. However, no adverse effects have been described in the above-cited case studies [11, 21]. Reliable results have been reported for the technique of augmentation plating after IM nailing for nonunions in the lower extremities [12, 14, 28, 29]. The technique takes advantage of the high axial stiffness resulting from in situ nails and uses short plates to minimize adverse rotational micromovements without compressing the nonunion site. Nadkarni et al. [14] used locked compression plates with bone grafting without removing the nails in 2 patients with humeral nonunions. Both cases achieved union within 6 months. Wu et al. [30] demonstrated that sufficient rotational stability can be achieved even with smaller
123
Arch Orthop Trauma Surg
implants and without the need for a locking mechanism. The authors augmented the IM nails with small staples at the nonunion site, combined with bone grafting, in two patients and reported uneventful healing after 5 and 7 months [30]. Concurrent to these very short staple implants, the short plates that were used in our study showed sufficient supplemental stability with the nail left in situ. Larger plates with at least 10–12 holes that bridge or compress the nonunion site and with at least 8 cortices engaged on both sides of the nonunion with additional interfragmentary screws, are recommended in cases using only plate fixation for humeral shaft nonunions [3, 8, 31– 35]. In cases of doubtful stability, a second plate in a 90° plane or, alternatively, strong onlay cortical bone grafts or intramedullary fibular allografts have been described to enhance stability at the nonunion site [32, 36, 37]. Rubel et al. [32] demonstrated the highest axial and rotational in vitro stiffness using the two-plate construct, with a healing rate of 89 % in 18 patients. No negative effects on the healing results have been reported due to the possible greater amount of bone and soft tissue exposure with the addition of a second plate [32]. Similarly, soft-tissue management and devascularization of the nonunion site are theoretical concerns of augmentation plating. However, the combination of intramedullary and extramedullary fixation devices has shown no negative effects on bone healing in the lower extremities [12, 13, 28, 29]. Furthermore, unlike in the treatment of acute fractures, we believe that extensive debridement of the atrophic nonunion sites with bone grafting is one of the key factors of successful treatment and, therefore, makes opening the nonunion site necessary. A meta-analysis by Kontakis et al. [31] found bone grafting to be a crucial factor for successful treatment, independent of the type of nonunion (atrophic or hypertrophic). However, the hypertrophic nonunions in our series responded to stability alone without the need of additional bone grafting. By using short augmentation plates, extensive soft tissue exposure can be avoided. In addition, the anterior approach that was used on all patients in this study avoids the radial nerve, which is often embedded in scar tissue at the posterior site of the humerus, without the need for visualization and exploration of the nerve [34]. Another major concern may be the risk of stress riser fractures when combining IM nails and plates. This specific problem has been particularly described for femoral fractures in elderly patients [38–40]. There was one patient with a peri-implant fracture after a low energy fall in our series. Critically stated, in this patient the plate and nail tip acted as a stress riser because both implants ended at the same level. To minimize the stress rising effect, ending of the augmentation plate near the IM nail tip need to be avoided.
123
It should be noted that none of our patients had a prominent nail that restricted shoulder or elbow motion. In these cases, nail removal is indicated to facilitate free joint function, and open plating is recommended [11, 41]. We did not detect any adverse effects from anterior plating on shoulder or elbow function in our patients. These functional results are concurrent with the report by Livani et al. [34], who treated 15 humeral nonunions by anterior plating (12 patients were initially treated nonoperatively, and 3 were treated with plate osteosynthesis). Furthermore, leaving the nail in situ eliminates the risk of iatrogenic damage, especially to the shoulder, and may reduce operation time compared to revision of humeral nonunions by removal of previous implants and plates [22]. Another constraint that should be noted is that none of the patients in this study showed severely compromised bone or osteolysis at the nonunion site, which would have made stable fixation with short plates impossible. In these cases, the use of long wave plates, fibular allografts, vascularized autografts, bone morphogenic proteins and mesenchymal stem cells have been proposed [8, 36, 42, 43]. The main limitations of the current study are the retrospective evaluation without a defined follow-up protocol and the evaluation of operations performed by many different surgeons with heterogeneous experience. The final follow-up of most patients ended with a diagnosis of a healed nonunion. As we were able to contact only a few patients, the number of implant removals due to a foreign body sensation of the anterior plate may be even higher, and other possible late complications might have been overlooked. Still, there is little literature on humeral nonunion treatment after intramedullary nailing, especially using a standardized, uniform surgical technique. The fact that the number of surgeons and their level of experience did not affect the outcome in respect to healing, function and complications underlines the safety of the procedure and the reproducibility of the results, rather than being a limitation when interpreting the study. In conclusion, our results indicate that the technique of augmentation plating using an anterior approach is a safe and reliable option for humeral shaft nonunions after failed nailing. Because the healing rates are similar to those of the standard technique of nail removal and fixation by compression or locking plates, we consider the technique as an alternative choice of treatment.
References 1. Ouyang H, Xiong J, Xiang P, Cui Z, Chen L, Yu B (2013) Plate versus intramedullary nail fixation in the treatment of humeral shaft fractures: an updated meta-analysis. J Shoulder Elbow Surg
Arch Orthop Trauma Surg
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
22(3):387–395. doi:10.1016/j.jse.2012.06.007 (S10582746(12)00270-4 [pii]) Kurup H, Hossain M, Andrew JG (2011) Dynamic compression plating versus locked intramedullary nailing for humeral shaft fractures in adults. Cochrane Database Syst Rev 6:CD005959. doi:10.1002/14651858.CD005959.pub2 Singh AK, Arun GR, Narsaria N, Srivastava A (2014) Treatment of non-union of humerus diaphyseal fractures: a prospective study comparing interlocking nail and locking compression plate. Arch Orthop Trauma Surg 134(7):947–953. doi:10.1007/s00402014-1973-0 Volgas DA, Stannard JP, Alonso JE (2004) Nonunions of the humerus. Clin Orthop Relat Res 419:46–50 (00003086200402000-00008 [pii]) McKee MD, Miranda MA, Riemer BL, Blasier RB, Redmond BJ, Sims SH, Waddell JP, Jupiter JB (1996) Management of humeral nonunion after the failure of locking intramedullary nails. J Orthop Trauma 10(7):492–499 Flinkkila T, Ristiniemi J, Hamalainen M (2001) Nonunion after intramedullary nailing of humeral shaft fractures. J Trauma 50(3):540–544 Catagni MA, Guerreschi F, Probe RA (1991) Treatment of humeral nonunions with the Ilizarov technique. Bull Hosp Jt Dis Orthop Inst 51(1):74–83 Ring D, Jupiter JB, Quintero J, Sanders RA, Marti RK (2000) Atrophic ununited diaphyseal fractures of the humerus with a bony defect: treatment by wave-plate osteosynthesis. J Bone Joint Surg Br 82(6):867–871 Patel VR, Menon DK, Pool RD, Simonis RB (2000) Nonunion of the humerus after failure of surgical treatment. Management using the Ilizarov circular fixator. J Bone Joint Surg Br 82(7):977–983 Kontakis GM, Papadokostakis GM, Alpantaki K, Chlouverakis G, Hadjipavlou AG, Giannoudis PV (2006) Intramedullary nailing for non-union of the humeral diaphysis: a review. Injury 37(10):953–960. doi:10.1016/j.injury.2006.02.050 (S00201383(06)00120-3 [pii]) Gerber A, Marti R, Jupiter J (2003) Surgical management of diaphyseal humeral nonunion after intramedullary nailing: waveplate fixation and autologous bone grafting without nail removal. J Shoulder Elbow Surg 12(4):309–313. doi:10.1016/mse.2003. S1058274603000442 (S1058274603000442 [pii]) Roetman B, Scholz N, Muhr G, Mollenhoff G (2008) Augmentive plate fixation in femoral non-unions after intramedullary nailing. Strategy after unsuccessful intramedullary nailing of the femur. Z Orthop Unfall 146(5):586–590. doi:10.1055/s-20081038526 Birjandinejad A, Ebrahimzadeh MH, Ahmadzadeh-Chabock H (2009) Augmentation plate fixation for the treatment of femoral and tibial nonunion after intramedullary nailing. Orthopedics 32(6):409. doi:10.3928/01477447-20090511-12 (orthopedics.40080 [pii]) Nadkarni B, Srivastav S, Mittal V, Agarwal S (2008) Use of locking compression plates for long bone nonunions without removing existing intramedullary nail: review of literature and our experience. J Trauma 65(2):482–486. doi:10.1097/TA. 0b013e31817c9905 (00005373-200808000-00041 [pii]) Yang JS, Otero J, McAndrew CM, Ricci WM, Gardner MJ (2013) Can tibial nonunion be predicted at 3 months after intramedullary nailing? J Orthop Trauma 27(11):599–603. doi:10.1097/BOT. 0b013e31828f5821 Seybold D, Citak M, Konigshausen M, Gessmann J, Schildhauer TA (2011) Combining of small fragment screws and large fragment plates for open reduction and internal fixation of periprosthetic humeral fractures. Int J Shoulder Surg 5(4):105–107. doi:10.4103/0973-6042.91004
17. Hierholzer C, Sama D, Toro JB, Peterson M, Helfet DL (2006) Plate fixation of ununited humeral shaft fractures: effect of type of bone graft on healing. J Bone Joint Surg Am 88(7):1442–1447. doi:10.2106/JBJS.E.00332 (88/7/1442 [pii]) 18. Blum J, Machemer H, Baumgart F, Schlegel U, Wahl D, Rommens PM (1999) Biomechanical comparison of bending and torsional properties in retrograde intramedullary nailing of humeral shaft fractures. J Orthop Trauma 13(5):344–350 19. Rosenbaum A, Uhl R (2012) Nonunion of humeral shaft fractures following flexible nailing fixation. Orthopedics 35(6):512–515. doi:10.3928/01477447-20120525-08 20. Baltov A, Mihail R, Dian E (2014) Complications after interlocking intramedullary nailing of humeral shaft fractures. Injury 45(Suppl 1):S9–S15. doi:10.1016/j.injury.2013.10.044 21. Marti RK, Verheyen CC, Besselaar PP (2002) Humeral shaft nonunion: evaluation of uniform surgical repair in fifty-one patients. J Orthop Trauma 16(2):108–115 22. Lin CL, Fang CK, Chiu FY, Chen CM, Chen TH (2009) Revision with dynamic compression plate and cancellous bone graft for aseptic nonunion after surgical treatment of humeral shaft fracture. J Trauma 67(6):1393–1396. doi:10.1097/TA. 0b013e31818c1595 (00005373-200912000-00043 [pii]) 23. Lin J, Chiang H, Hou SM (2003) Open exchange locked nailing in humeral nonunions after intramedullary nailing. Clin Orthop Relat Res 411:260–268. doi:10.1097/01.blo.0000069898.31220. 9e 24. Apard T, Ducellier F, Hubert L, Talha A, Cronier P, Bizot P (2010) Isolated interfragmentary compression for nonunion of humeral shaft fractures initially treated by nailing: a preliminary report of seven cases. Injury 41(12):1262–1265. doi:10.1016/j. injury.2010.05.023 (S0020-1383(10)00319-0 [pii]) 25. Raschke M, Khodadadyan C, Maitino PD, Hoffmann R, Sudkamp NP (1998) Nonunion of the humerus following intramedullary nailing treated by Ilizarov hybrid fixation. J Orthop Trauma 12(2):138–141 26. Tomic S, Bumbasirevic M, Lesic A, Mitkovic M, Atkinson HD (2007) Ilizarov frame fixation without bone graft for atrophic humeral shaft nonunion: 28 patients with a minimum 2-year follow-up. J Orthop Trauma 21(8):549–556. doi:10.1097/BOT. 0b013e31814612c8 (00005131-200709000-00006 [pii]) 27. Menon DK, Dougall TW, Pool RD, Simonis RB (2002) Augmentative Ilizarov external fixation after failure of diaphyseal union with intramedullary nailing. J Orthop Trauma 16(7):491–497 28. Ueng SW, Chao EK, Lee SS, Shih CH (1997) Augmentative plate fixation for the management of femoral nonunion after intramedullary nailing. J Trauma 43(4):640–644 29. Choi YS, Kim KS (2005) Plate augmentation leaving the nail in situ and bone grafting for non-union of femoral shaft fractures. Int Orthop 29(5):287–290. doi:10.1007/s00264-005-0668-0 30. Wu CC, Shih CH, Chen WJ, Tai CL (1998) Staple augmentation to treat a humeral nonunion after failed Seidel locked nailing. Arch Orthop Trauma Surg 118(1–2):42–44 31. Kontakis GM, Tosounidis T, Pagkalos J (2007) Humeral diaphyseal aseptic non-unions: an algorithm of management. Injury 38(Suppl 2):S39–S49 (doi:S0020-1383(07)80008-8 [pii]) 32. Rubel IF, Kloen P, Campbell D, Schwartz M, Liew A, Myers E, Helfet DL (2002) Open reduction and internal fixation of humeral nonunions: a biomechanical and clinical study. J Bone Joint Surg Am 84-A(8):1315–1322 33. Cadet ER, Yin B, Schulz B, Ahmad CS, Rosenwasser MP (2013) Proximal humerus and humeral shaft nonunions. J Am Acad Orthop Surg 21(9):538–547. doi:10.5435/JAAOS-21-09-538 (21/ 9/538 [pii]) 34. Livani B, Belangero W, Medina G, Pimenta C, Zogaib R, Mongon M (2010) Anterior plating as a surgical alternative in the
123
Arch Orthop Trauma Surg
35.
36.
37.
38.
treatment of humeral shaft non-union. Int Orthop 34(7):1025–1031. doi:10.1007/s00264-009-0863-5 Martinez AA, Cuenca J, Herrera A (2009) Two-plate fixation for humeral shaft non-unions. J Orthop Surg (Hong Kong) 17(2):135–138 Willis MP, Brooks JP, Badman BL, Gaines RJ, Mighell MA, Sanders RW (2013) Treatment of atrophic diaphyseal humeral nonunions with compressive locked plating and augmented with an intramedullary strut allograft. J Orthop Trauma 27(2):77–81. doi:10.1097/BOT.0b013e31825360fa Hornicek FJ, Zych GA, Hutson JJ, Malinin TI (2001) Salvage of humeral nonunions with onlay bone plate allograft augmentation. Clin Orthop Relat Res 386:203–209 Leibner ED, Mosheiff R, Safran O, Abu-Snieneh K, Liebergall M (1999) Femoral fracture at the proximal end of an intramedullary supracondylar nail: a case report. Am J Orthop (Belle Mead NJ) 28(1):53–55
123
39. Harris T, Ruth JT, Szivek J, Haywood B (2003) The effect of implant overlap on the mechanical properties of the femur. J Trauma 54(5):930–935. doi:10.1097/01.TA.0000060999.54287. 39 40. Kim HK, Noh KC, Chung KJ, Hwang JH (2012) Use of Huckstep nail in the periimplant femoral shaft fracture. Indian J Orthop 46(6):718–720. doi:10.4103/0019-5413.104240 41. Ring D, Jupiter J (2003) Humerus nonunion after intramedullary rod fixation. Tech Orthop 18(4):356–359 42. Wright TW (1997) Treatment of humeral diaphyseal nonunions in patients with severely compromised bone. J South Orthop Assoc 6(1):1–7 43. Murena L, Canton G, Vulcano E, Surace MF, Cherubino P (2014) Treatment of humeral shaft aseptic nonunions in elderly patients with opposite structural allograft, BMP-7, and mesenchymal stem cells. Orthopedics 37(2):e201–e206. doi:10.3928/0147744720140124-26
