- Email: [email protected]
Complications after interlocking intramedullary nailing of humeral shaft fractures
Injury, Int. J. Care Injured 45S (2014) S9–S15
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Complications after interlocking intramedullary nailing of humeral shaft fractures Asen Baltov *, Rashkov Mihail, Enchev Dian Department of Trauma Surgery, Emergency Trauma Hospital ‘‘N.I.Pirogov’’, Sofia, Bulgaria
A R T I C L E I N F O
A B S T R A C T
Keywords: Humeral shaft fracture (HSF) Interlocking intramedullary (IM) nailing Complications of nailing
Introduction: IM nails have gained popularity for stabilization of humeral shaft fractures (HSF). The initial enthusiasm was tempered by a number of specific complications and thus indications need to be reevaluated. Patients and methods: This retrospective study includes 111 patients with HSF subjected to a treatment protocol of IM fixation with first and second generation of humeral nails. Antegrade approach was used in 105 (94.5%) and retrograde in 6 (5.5%) patients. Reaming was performed in 51 (45.9%) fractures. The study covers a period of 10 years. Mean follow-up time was 3.5 (1–6) years. This investigation is directed at technical errors and complications, especially those corrected by secondary surgery. Results: We registered 52 (46.85%) intra-operative complications in 40 (36.04%) patients, on average 1.3 per patient. The most common were: distraction n = 5 (4.5%), long proximal locking screws n = 9 (8.1%), additional diaphyseal fracture n = 7 (6.3%) and countersinking of the nail in the humeral head n = 8 (7.2%). The number of postoperative complications was 40 (36.0%) related to 19 (17.1%) patients. Technical errors, such as distraction, longer nail and additional fractures have affected time to union and resulted in chronic shoulder pain. 36 (32.5%) secondary surgeries were needed to address these problems. When first generation nails are used, the intra-operative complication related risk increases 1.58 times, and the postoperative complication related risk is 1.67 times higher compared to second generation nails. According to Constant–Murley score excellent and very good functional results were achieved in 93 (83.78%) patients. While reaming did not influence the clinical results for both nail generations, overall better results were achieved with second generation nails. Postoperative shoulder pain has been registered in 18 (16.2%) patients. Conclusion: We registered a number of technical errors and complications, which we consider technique specific. The analysis and avoidance of these complications, related only to IM nailing of the humerus, will allow IM nails to successfully bridge the gap between functional bracing and plating. ß 2013 Elsevier Ltd. All rights reserved.
Introduction
Patients and methods
Intramedullary nail (IM) fixation is an established method of treatment of high-energy long bone fractures especially in polytrauma setting, as well as osteoporotic, impending and pathological fractures. The technical errors and complications after interlocking IM nailing of humeral shaft fractures (HSF) are recorded and discussed in only a limited number of studies [1–3]. The aim of this investigation is to analyze the results and complications of humeral shaft fractures stabilized with interlocking IM nailing.
Over a 10 year period (January 2000 to November 2010), patients with diaphyseal humeral fractures were eligible to participate in this study. Patients where a humeral nail was used for delayed post reconstruction procedures were excluded. Fractures were classified according to the AO classification and according to Gustilo–Anderson [4] if they were open. First and second generation interlocking IM nails were used. Such details were documented as intra-operative, perioperative and post-operative complications including the development of superficial infection, deep infection, iatrogenic extension of fractures proximally and distally, insertion of longer proximal and distal locking screws, metal work failure, presence of fracture gap post-operatively (distraction at the fracture site >4 mm and angulation >208), development of non-union and malunion, the necessity of insertion of additional metal work, iatrogenic nerve
* Corresponding author at: Emergency Trauma Hospital ‘‘N.I.Pirogov’’, Ist Department of Trauma Surgery, E Totleben 21, Sofia, Bulgaria. Tel.: +359 888 874040. E-mail address: [email protected] (A. Baltov). 0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.10.044
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S10
Reaming of the canal was performed in 51 (45.9%) cases. Surgery on average was delayed 6 days (0–14). The mean length of surgery was 84.3 min. It was 70 min (range 40–195 min) in cases of closed reduction and 105 min (range 60–210 min) in cases of open reduction. The means of fixation utilized in this series was the first generation nail of R-T (Smith & Nephew) in 34 cases (30.6%) with only one option for (dynamic) interlocking proximally and distally, and the second generation nails with more interlocking options: UHN (Synthes) n = 25 (22.5%); ARHN (Sanatmetal) n = 29 (26.2%); EHN (Euro Instrument) n = 7 (6.3%) and Versa UHN (DePuy) n = 16 (14.4%), (Table 2). Polytrauma patients had mean hospital stay of 21 (10–60) days. In cases of isolated humeral fractures, the average period was 6 (4– 11) days. 109 fractures (98.2%) healed after the initial procedure with an average time to union of 3 months (range 1–8). Fixation was considered stable enough in 104 (93.7%) patients, who started early rehabilitation from 6th to 12th postoperative day. Secondary displacement with proximal protrusion and telescoping of the nail occurred in 4 (3.6%) patients, breakage of interlocking screws occurred in 3 (2.7%) and non-union in 2 cases (1.8%). Shoulder function was graded excellent and very good in 93 (83.8%) patients, according to Constant–Murley score.
damage, proximal protrusion of nails, and development of avascular necrosis of the humeral head. Bone healing was defined as the presence of bridging of fracture gap with callus in three cortices in two-plane X-rays, as well as the lack of pain on passive arm rotation. The functional status of the limb was assessed as follows: subjective complaints, presence of pain, range of movement of shoulder and elbow, muscle strength recovery. Sensitivity and motor activity in the areas innervated by radial and musculocutaneous nerves have been also recorded and analyzed. Examination of the function of shoulder was performed according to the Constant–Murley scale [5], based on age and gender at the last follow up. Deep infection was defined as osteomyelitis with presence of chronic suppuration and isolation of microbiological agent [6]. Follow-up was performed on the second and fourth week and then monthly until the sixth month. Thereafter patients were followed annually as necessary. The mean follow-up time was 3.5 years (1–6). Results Overall, 40 (36.0%) females, and 71 (64.0%) males, with an average age 49 years (range 17–83) met the inclusion criteria and formed the study group. Thirty-one (27.9%) patients sustained polytrauma with a mean Injury Severity Score of 22 (14–36). Closed fractures were seen in 101 cases (90.9%), and open fractures were seen in 10 cases (9.1%), (n = 5 of were of grade I, n = 2 of grade II, and n = 3 of grade III according to Gustilo–Anderson). Fracture distribution according to the AO classification is shown in Table 1. 20 (18%) fractures were in the proximal third, 65 (58.6%) in the middle third, 15 (13.5%) in the distal third, and 11 (9.9%) had segmental localization, engaging 2 or 3 levels. Concomitant injury of the radial nerve was found in 10 (9%) patients. It was combined motor and sensory in all cases. Antegrade surgical approach was used in 105 (94.5%) fractures and retrograde in 6 (5.5%). Closed reduction was achieved in 97 (87.4%) patients. 14 (12.6%) fractures were reduced in an open manner.
Complications We registered 52 (46.9%) intraoperative complications in 40 (36.0%) patients (an average of 1.3 in each patient), 71 (64.0%) patients showed no complication. 40 (36.0%) postoperative complications were recorded in 19 (17.1%) patients, (Tables 3 and 4). From a total of 7 (6.3%) diaphyseal intraoperative fractures, 5 fractures occurred with nails of the second generation. In four cases an unreamed nail was forcefully passed through a narrow canal. Although second generation nails did not cause a change in the method of fixation, they influenced the final functional result and resulted in delayed healing. In 2 (1.8%) patients treated with a first generation antegrade reamed IM nail a fracture of the greater
Table 1 Patient distribution according to the AO/OTA. Type A 38 (34.2%) A A A A A A A A A
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
Type B 46 (41.4%) 4 11 3 3 5 5 2 7 0
10.5% 28.9% 7.9% 7.9% 13.2% 7.9% 5.3% 18.4% 0%
B B B B B B B B B
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
Type C 27 (24.3%) 7 3 4 4 12 1 1 10 4
15.2% 6.5% 8.7% 8.7% 26.1% 2.2% 2.2% 21.7% 8.7%
C C C C C C C C C
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
2 6 1 3 7 2 0 5 1
7.4% 22.2% 3.7% 11.1% 25.9% 7.4% 0% 18.5% 3.7%
Table 2 Distribution according to the type of IM nail. Type of interlocking nail
Proximal interlocking
Distal interlocking
Dynamic
Static (one-plane)
Static (two-plane)
Dynamic
Static (one-plane)
Static (two-plane)
R-T/Smith & Nephew/ UHN/Synthes/ ARHN/Sanatmetal/ EHN/EuroInstrument/ Versa UHN/DePuy/
34 21 2 3 8
1 26 1 8
3 1 3
34 16 11 6 8
6 15 1 8
3 3
Total
68 (61.3%)
36 (32.4%)
7 (6.3%)
75 (67.6%)
30 (27%)
6 (5.4%)
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S11
Table 3 Intra-operative errors and complications in 44 patients. N (%)
Complications/Intraoperative Fracture of the greater tubercle of humerus Fracture of the diaphysis Supracondylar fracture Unstable fixation Missed locking Broken interosseal screw Long proximal interlocking screw Long distal interlocking screw Injury of radial nerve Iatrogenic injury of lateral antebrachial cutaneal nerve Distraction >5 mm Inadequate embedding Total
2 7 3 5 1 4 9 5 1 2 5 8
First generation 34
(1.8%) (6.3%) (2.7%) (4.5%) (0.9%) (3.6%) (8.1%) (4.5%) (0.9%) (1.8%) (4.5%) (7.2%)
2 2 1
52 (46.85%)
Second generation 77
With reaming 51
5 2 5
2 3 2 2 1
4 1 3
4 3 3 1
5 2
4 4 3 1
1 6 2
Without reaming
2 4
5 4
2 1 5
4 3
20
32
25
27
Table 4 Post-operative complications in 19 patients. Complications/Postoperative
Count (%)
Protrusion Impingement nail Impingement proximal screw Migration of the proximal screw Migration of the distal screw Debricolage Infection Delayed union Non-union Avascular necrosis Broken interlocking nail
5 12 5 4 1 2 1 3 2 2 3
Total
40 (36%)
(4.5%) (10.8%) (4.5%) (3.6%) (0.9%) (1.8%) (0.9%) (2.7%) (1.8%) (1.8%) (2.7%)
First generation 34 4 5 4
1 1
Second generation 77 1 7 1 4 1 2
1
2 2 2 2
16
24
tubercle occurred due to either lateral insertion of the nail or excessive widening of the entry portal. Supracondylar fractures were seen in n = 3 (2.7%) cases: 2 fractures in retrograde nails, when no additional fixation was needed and one case with anterograde technique, which required additional fixation with a plate (Fig. 1A). Excluded from this study is an osteoporotic supracondylar fracture of a 52-year-old female, which occurred with retrograde insertion of the nail. It required change of treatment modality; fixation was achieved with 4.5 AO compression plate (Fig. 1B). Additional fixation was performed in 5 (4.5%) cases due to residual rotational instability after locking with second generation
With reaming 51 2 5 3 1
1 1 1
Without reaming 60 3 7 2 3 1 2
1
2 1 2 2
15
25
IM nails (4 fractures were fixed with inter-fragmentary screws and in one case cerclage wires were applied). A missed proximal interlocking hole in an R-T nail was found in 1 (0.9%) case. This is a specific technical error also described by others [7]. In this case it had no consequences to bone healing and the final functional result (Fig. 2). Another technical error is intra-operative, intra-osseous breakage of distal locking screws in 4 (3.6%) cases. These screws have a small step of the thread and a large head and they break when inserted at an inaccurate angle. No fracture site instability or delay in bone healing was noted; however difficulties with nail removal are to be expected (Fig. 3).
Fig. 1. Iatrogenic supracondylar fracture: (A) fixation with additional plate, (B) conversion of the fixation after supracondylar fracture.
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S12
Fig. 3. Interosseous screw breakage. Fig. 2. Missed interlocking.
Proximal interlocking with a longer screw was considered a technical error when at least 4 mm protrusion through the medial or postero-medial cortex was found. This occurred in 9 (8.2%) cases and caused discomfort and limitation of shoulder motion in 6 patients: two cases with antero-posterior interlocking and four patients with latero-medial interlocking. This required screw removal. Three patients had oblique dynamic proximal interlocking, which did not cause complaints. Longer distal interlocking screws were found in 5 (4.5%) cases with anterograde interlocking intramedullary nailing. It caused tenderness on palpation over the posterior surface of the arm in slim patients. Their removal was required in two cases. Prominence of the nail (Fig. 4) was found on postoperative Xrays in 8 (7.2%) cases. It caused discomfort, hampered recovery of shoulder motion and delayed return of overall arm function. The type of complications leading to shoulder discomfort, besides proximal protrusion of nail, were various including: telescopic effect in 5 (4.5%) patients with osteoporosis, incomplete embedding of the nail at surgery, leading to subacromial impingement in 12 (10.8%) cases, and in 5 (4.5%) patients impingement of the head of proximal interlocking screw. This was significantly more frequent in interlocking IM nailing with first generation nails (Fisher exact test p < 0.01). Shoulder related complaints led to additional surgical procedures in 18 (16.2%) cases: 14 extractions of nail and 4 removals of the interlocking proximal screws. This led to improvement of the final functional result in 12 cases from good to very good, and in 6 cases from satisfactory to good. Damage to the radial nerve was reported postoperatively in 1 (0.9%) patient, who recovered spontaneously after 6 months. Two (1.8%) cases of injury of the lateral cutaneal nerve of the forearm with persistent paresthesia in the dermatome were detected and
Fig. 4. Protrusion of a nail and a proximal screw.
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S13
Fig. 5. Non-union due to: (A) distraction; (B) unstable interlocking intramedullary nailing; (C) revision with a 4.5 DCP.
were attributed to distal anteroposterior interlocking of the nail. Osteomyelitis was diagnosed in 1 patient. Fracture distraction was recorded in 14 (12.6%) cases and in 5 (4.5%) patients it measured more than 4 mm. In 3 (2.7%) patients this caused delayed healing and in 2 (1.8%) non-union. According to Fisher’s exact test the risk of non-union was significantly higher if the fractures were locked in distraction p < 0.01 (p-value = 0.000000164). In the cases of delayed union, within 6 months after fixation, we found 1 (0.9%) screw breakage and 4 (3.6%) migrations of proximal locking screws with antegrade IM nails, and 1 (0.9%) distal locking screw with retrograde nails. However no additional surgical interventions were performed. Non-union was diagnosed in 2 (1.8%) patients. The first case was a high-energy fracture of B 2.2 type, which was fixed in distraction of 10 mm with retrograde nail of second generation, insufficient uni-planar proximal and dynamic distal locking. This resulted in residual rotational instability and breakage of the proximal locking screw. After 18 months the nail was removed and fixation achieved with 4.5 mm DCP, (autologous bone graft was harvested locally from a consolidated butterfly fragment), (see Fig. 5). The second instance of non-union occurred after a high-energy A3.2 fracture, fixed with an antegrade interlocked IM nail, locked
dynamically, proximally and distally. No reaming was performed of a narrow canal (less than 8 mm) and fracture was locked in distraction (11 mm). At 12 months, due to a breakage of a distal locking screw and a secondary rotational instability, the nail was removed, fixation was revised with a 4.5 mm DCP and autologous bone grafting was performed. In both non-unions, healing occurred after a period of 6 months and full range of motion of adjacent joints was restored. One case (0.9%) of debricolage occurred with a fracture of type C1.1 and additional fracture of the proximal humerus with comminution of the greater tubercle. Fixation was performed with a Cloverleaf plate, without nail removal. Fracture healed in 3 months and good shoulder function was restored (Fig. 6). We diagnosed AVN of the humeral head as a late complication in 2 (1.8%) patients with associated fracture of the proximal humerus and of advanced age. This resulted in decreased range of shoulder movement. No further surgery was performed in these low-demand patients. Varus deformity in excess of 258 was recorded in 2 (1.8%) patients with limited range of shoulder abduction (1408 in both cases). In 2 (1.8%) cases with a floating elbow, a significantly decreased range of elbow extension was measured (Fisher’s exact test p < 0.01), (0.004586).
Fig. 6. (A) Debricolage; (B) revision with the nail in situ.
S14
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
The intra- and postoperative complications required a total of 36 (32.4%) additional procedures in 33 patients, averaging 1.09 procedures in each case. Eventually, 6 (5.4%) patients were considered to have poor or satisfactory functional results. Discussion In an updated meta-analysis focusing on total complication rate (TCR), Heineman et al. [8], have found that plate fixation has a lower TCR compared to interlocking IM nailing. They suggested that although the difference is decreasing, there is still evidence to support the view that plates are more favourable than nails in terms of TCR. In an earlier version the same group had found that plate fixation decreases the risk of revision surgery by up to 74% and the likelihood of shoulder impingement by nearly 90%. It was demonstrated that plated fractures have a lower revision rate in comparison to those treated with an IM nail (6% versus 18%, p = 0.03) and there is a similar disparity between the incidence of shoulder complaints (1% versus 21%, p = 0.002). Moreover, delayed healing and non-union were more frequent in the group of IM nails (8 out of 73/11%), compared to those fixed by a plate (5 out of 83/ 6%) [8]. In the randomized study of Singisetti et al. [9], 65% of patients treated with first generation ante-grade interlocking IM nailing were found to have excellent and good shoulder function, while in cases of plate fixation, it was found in 95%. They also pointed out that only 50% of nailed fractures have achieved union at 4 months, compared to 75% of those fixed by a plate. The issue of the entry portal is a subject of multiple studies but is still a controversial point. The main problem in antegrade insertion of a nail lies in the fact that the axis of the medullary canal exits through the joint surface of the humeral head and the rotator cuff, which limits the choice of insertion sites [10,11]. There is still a debate about the incidence and severity of subacromial impingement and persistent pain, as well as about assessment methods and adequacy of follow-up regarding those complaints. Ajmal et al. [12] reported poor shoulder function in 41% of patients with an entry portal affecting the rotator cuff. Several randomized trials have raised awareness of the fact that subacromial impingement and persistent pain are found in one of every 5 patients with anterograde nailing. Heinsen and Hempel [13] have used ultrasound and found transient injuries of the rotator cuff prior to anterograde IM fixation in 10 of 36 cases. Postoperatively, 5 ruptures of the tendon of supraspinatus muscle were recorded, three of them resulted from nail protrusion, the rest, were a consequence of accompanying degenerative changes. They concluded that anterograde IM fixation does not increase the likelihood of rupture of the rotatory cuff, provided the surgical technique is followed correctly: complete embedding of the nail end in the humeral head and careful repair of the tendon of the supraspinatus muscle. Impingement was also described as a consequence of too proud proximal interlocking screws, as well as protrusion of the proximal end of nail [14,15]. A reliable strategy for prevention of shoulder pain is retrograde nail insertion through extra-articular entry portal, which in many cases may give good and excellent results [16,17]. The complications are also well-known: supracondylar fracture, pain and limited elbow motion, weakness in the triceps muscle, periarticular ectopic ossifications and adhesions in the region of elbow, with consequent limitation of function [18]. According to Farragos et al. [1], there is still an argument as to which technique more reliably preserves shoulder and elbow joint function. In this series of predominantly antegrade nails, there are several complications that have led to impairment of shoulder function.
Proximal protrusion of the nail with so called ‘‘telescopic effect’’ was noted in 4 (3.9%) patients, incomplete embedding of nail, which resulted in subacromial impingement recorded in a total of 12 (11.7%) patients, and impingement caused by the head of a proximal interlocking screw in 5 (4.9%) patients. 18 (17.5%) additional surgical procedures were needed to address those issues [1]. As a whole, 65 patients had no pain in the shoulder, 28 patients had mild and moderate pain. However, 13 had intermittent pain, 4 had moderate to severe pain and 1 was in constant pain. A possible reason for a decreased range of motion and lower functional scores in patients who underwent IM nailing was offered by Li et al. [19]. They measured rotational malalignment by postoperative CT and found that 27.2% of nailed patients had an internally rotated humeral head, which affected negatively shoulder motion. Insertion of locking screws in IM nailing of humerus can incur additional vascular and nerve injury, specific only to the applied technique of fixation. At risk are: axillary nerve, lateral cutaneal nerve of forearm, and median nerve. In cases of proximal interlocking in the frontal and sagittal plane, both branches of the axillary nerve can be damaged [20,21]. Screw insertion in the oblique position is considered potentially less hazardous. Notwithstanding the theoretically high risk, only one case of iatrogenic injury to these nerves in anterograde and one case of retrograde interlocking IM nailing has been described so far [22]. Antero-posterior distal locking is considered safer, but the risk of injury to the musculocutaneal nerve is well recognized. Two cases of this type are described in scientific literature [23,24]. Two types of iatrogenic fractures during interlocking IM nailing are to be distinguished. Stable fractures do not require a change in the treatment method or supplementary fixation, only occasional external protection. Unstable fractures demand additional fixation. This prolongs surgery time, increases blood loss and radiographic assessment [25,26]. The incidence of iatrogenic fractures varies between 2% and 22%, for both surgical techniques, but is higher in cases of retrograde interlocking IM nailing. According to Fernandes et al. [27], it is 10.7%, Verbruggen et al. [28] suggested the incidence is even higher up to 17%. Iatrogenic fracture of the greater tubercle during anterograde IM nailing (2–11%) occurs with a too lateral insertion of a rigid nail. When the end of the nail abuts the medial cortex it generates pressure on the lateral cortex of the head of humerus via a lever from the acromion [29]. Another common iatrogenic fracture is that of the diaphysis (incidence 1–5%). Insertion of a rigid nail through a narrow medullary canal with no prior reaming can result in shaft fragmentation when driving in the nail. Additional fractures can occur in the proximal third of the diaphysis or the supracondylar area. In the latter case the reason behind this is the straight, but narrow distal diaphyseal canal, most frequently in osteoporotic bone [30]. Garnavos and Kanakaris [31] reported a fracture of the diaphysis after eccentric drilling during distal locking of the nail. In this study of 111 fractures, there were 9 (8.7%) iatrogenic fractures with anterograde approach and 3 (50%) – with the retrograde technique. Angulations are registered in all methods of treatment. In cases of nail fixation, excessive angulation is quite rare and the observed deformities can be accepted [32]. Varus deformity is typical of fractures of the proximal one third of the diaphysis and is caused by deltoid muscle pull. If not taken into account, this results in a too lateral nail insertion point. The proximal curvature of nail (from 58 to 108), especially with first generation nails, which have only dynamic interlocking, will lead to fixation in varus angulation (up to 308).
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
In this series 4 patients were noted to have a varus deformity ranging between 168 and 258. Reduced shoulder abduction up to 308 was measured which affected overall shoulder function. Infection after interlocking IM nailing is a comparatively rare complication. It has been recorded in 0–6% of cases. Many series, though, do not distinguish between closed and open fractures. [33,34]. In this study there was only one (1.07%) patient (type 2 diabetes mellitus) with osteomyelitis after interlocking IM humeral nailing. The nail was removed after 10 months, when the fracture was healed. Infection (Staphylococcus aureus) was successfully treated by debridement, canal reaming and lavage. This did not seem to influence the functional result. In conclusion, humeral fractures can be treated with different operative techniques [35–40]. Specific technical errors and complications related only to intramedullary nails were registered in this retrospective case series. Strict adherence to proven indications (high-energy C-type fractures, polytrauma and impending fractures) and good surgical technique will allow the interlocking IM nailing to bridge the gap between functional bracing and the plating and to achieve better results compared to both of them. Conflict of interest statement No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study. References [1] Farragos AF, Schemitsch EH, McKee MD. Complications of intramedullary nailing for fractures of the humeral shaft: a review. J Orthop Trauma 1999;13:258–67. [2] Garnavos C. Intramedullary nailing for humeral shaft fractures: the misunderstood poor relative. Curr Orthop 2001;15:69–75. [3] Lin J, Shen PW, Hou SM. Complications of locked nailing in humeral shaft fractures. J Trauma 2003;54(5):943–9. [4] Gustilo RB, Merkow RL, Templeman D. Current concepts review. The management of open fractures. J Bone Joint Surg 1990;72A:299–304. [5] Constant CR, Murley AHG. A clinical method of functional assessment of the shoulder. Clin Orthop 1987;214:160–4. [6] McKee MD. Fracture of the shaft of the humerus. In: Bucholz RW, editor. Rockwood Green’s fractures in adults. 6th ed., Philadelphia: Lippincott Williams & Wilkins; 2006. p. 1117–59. [7] Asencio G, Buscayret F, Trabelsi A, Bertin R, Hammami R, Megy B, et al. Intramedullary interlocking nailing for humeral fractures: report of 38 cases treated by Russell and Taylor nail. Rev Chir Orthop 2001;87:749–57. [8] Heineman DJ, Poolman RW, Nork SE, Ponsen KJ, Bhandari M. Plate fixation or intramedullary fixation of humeral shaft fractures. Acta Orthop 2010;81(2):216–23. Updated in: Acta Orthop 2010; 81 (4) 517. Updated Acta Orthop 83 (3) 317–318. [9] Singisetti K, Ambedkar M. Nailing versus plating in humerus shaft fractures: a prospective comparative study. Int Orthop 2010;34:571–6. [10] Cox MA, Dolan M, Synnott K, McElwain JP. Closed interlocking nailing of humeral shaft fractures with the Russell-Taylor nail. J Orthop Trauma 2000;14:349–53. [11] Flinkkila T, Hyvo¨nen P, Lakovaara M, Linden T, Ristiniemi J, Ha¨ma¨la¨inen M. Intramedullary nailing of humeral shaft fractures. A retrospective study of 126 cases. Acta Orthop Scand 1999;70:133–6. [12] Ajmal M, O’Sullivan MO, McCabe J, Curtin W. Antegrade locked intramedullary nailing in humeral shaft fractures. Injury 2001;32:692–4. [13] Heinsen J, Hempel D. Praoperatve sonographische Untersuchung der Rotatorenmanchette vor antegrader Nagelung von Humerusschaftfracturen. Osteosynthese International 2001;9(Suppl 2):S85–7. [14] Dimakopoulos P, Papadopoulos AX, Papas M, Panagopoulos A, Lambiris E. Modified extra rotator-cuff entry point in antegrade humeral nailing. Arch Orthop Trauma Surg 2005;125:27–32.
S15
[15] Pestatodes G, Karataglis D, Papadopoulos P, Christoforides J, Gigis J, Pournaras J. Antegrade interlocking nailing of humeral shaft fractures. J Orthop Sci 2004;9:247–52. [16] Sanzana ES, Dummer RE, Castro JP, Diaz EA. Intramedullary nailing of humeral shaft fractures. Int Orthop 2002;26:211–3. [17] Vescei N, Kolonja A, Mousavi M, Ve´csei V. Intramedullary fixation of humerus shaft fractures. An analysis of complications of 2 implants with special reference to outcome after management with the unreamed humerus interlocking nail. Wien Klin Wochenschrift 2001;113:597–604. [18] Loitz D, Ko¨nnecker H, Illgner A, Reilmann H. Retrograde intramedullary nailing of humeral fractures with new implants. Analysis of 120 consecutive cases. Unfallchirurg 1998;101:543–50. [19] Li Y, Wang C, Wang M, Huang L, Huang Q. Postoperative malrotation of humeral shaft fracture after plating compared with intramedullary nailing. J Shoulder Elbow Surg 2011;20:947–54. [20] Albritton MJ, Barnes CJ, Basamania CJ, Karas SG. Relationship of the axillary nerve to the proximal screws of a Flexible Humeral Nail system: an anatomical study. J Orthop Trauma 2003;17:411–4. [21] Prince EJ, Breien KM, Fehringer EV, Mormino MA. The relationship of proximal locking screws to the axillary nerve during antegrade humeral nail insertion of four commercially available implants. J Orthop Trauma 2004;18:585–8. [22] Lo¨gters T, Wild M, Windolf J, Linhart W. Axillary nerve palsy after retrograde humeral nailing: clinical conformation of an anatomical fear. Arch Orthop Trauma Surg 2008;128:1431–5. [23] Blyth MJ, Macleod CM, Asante DK, Kinninmonth AW. Iatrogenic nerve injury with the Russell-Taylor humeral nail. Injury 2003;34:227–8. [24] Kesemenli CC, Subasi M, Arslan H, Necmioglu S, Kapukaya A. Comparison between the results of intramedullary nailing and compression plate fixation in the treatment of humerus fractures. Acta Orthop Traumatol Turc 2003;37(2):120–5. [25] Apard T, Lahogue JF, Prove S, Hubert L, Talha A, Cronier P, et al. Retrograde locked nailing of humeral shaft fractures: a prospective study of 58 cases. Rev Chir Reparatrice Appar Mot 2006;92:19–26. [26] Muckley T, Diefenbeck M, Sorkin AT, Beimel C, Goebel M, Bu¨hren V. Results o the T2 humeral nailing system with special focus on compression interlocking. Injury 2008;39:299–305. [27] Fernandez FF, Winkler H, Merkel G, Freudenberg S, Wentzensen A. Erste klinische Erfahrungen mit der unaufgebohrten Marknagelung bei Humerusschaftfracturen. AktTraumatol 1999;29:269–76. [28] Verbruggen JPAM, Adriaens S, Goessens ML, Stapert JWJL. Reamed nailing of humeral fractures: long-term results in 77 cases. Osteo Trauma Care 2002;10:192–6. [29] Herbst U, Ruettger K, Mockwitz J. Experiences with the Russell-Taylor nail in humeral shaft fractures – an analysis of postoperative results and complications. Osteo Trauma Care 2003;11:13–20. [30] Fernandez FF, Matschke S, Hu¨lsenbeck A, Egenolf M, Wentzensen A. Five years’ clinical experience with the unreamed humeral nail in the treatment of humeral shaft fractures. Injury 2004;35:264–71. [31] Garnavos C, Kanakaris N. Stress fracture due to unsuccessful targeting during intramedullary nailing. Eur J Trauma 2003;29:105–7. [32] Rommens PM, Kuechle R, Bord T, Lewens T, Engelmann R, Blum J. Humeral nailing revisited. Injury 2008;39:1319–28. [33] Blum J, Rommens PM. Proximal interlocking of humeral intramedullary nails and the risk of injury of the axillary nerve. Unfallchirurg 2002;105:9–13. [34] Chapman JR, Bradford Henley M, Agel, Benca PJ. Randomized prospective study of humeral shaft fracture fixation: intramedullary nails versus plates. J Orthop Trauma 2000;14:162–6. [35] Karataglis D, Stavridis SI, Petsatodis G, Papadopoulos P, Christodoulou A. New trends in fixation of proximal humeral fractures: a review. Injury 2011;42(April (4)):330–8. [36] Biber R, Zirngibl B, Bail HJ, Stedtfeld HW. An innovative technique of rear entry creation for retrograde humeral nailing: how to avoid iatrogenic comminution. Injury 2013;44(April (4)):514–7. [37] Matassi F, Angeloni R, Carulli C, Civinini R, Di Bella L, Redl B, et al. Locking plate and fibular allograft augmentation in unstable fractures of proximal humerus. Injury 2012;43(November (11)):1939–42. [38] Noda M, Saegusa Y, Maeda T. Does the location of the entry point affect the reduction of proximal humeral fractures? A cadaveric study. Injury 2011;42 (Suppl 4):S35–40. [39] Hardeman F, Bollars P, Donnelly M, Bellemans J, Nijs S. Predictive factors for functional outcome and failure in angular stable osteosynthesis of the proximal humerus. Injury 2012;43(February (2)):153–8. [40] Collis PN, Clegg TE, Seligson D. The invisible nail: a technique report of treatment of a pathological humerus fracture with a radiolucent intramedullary nail. Injury 2011;42(April (4)):424–6.
Contents lists available at ScienceDirect
Injury journal homepage: www.elsevier.com/locate/injury
Complications after interlocking intramedullary nailing of humeral shaft fractures Asen Baltov *, Rashkov Mihail, Enchev Dian Department of Trauma Surgery, Emergency Trauma Hospital ‘‘N.I.Pirogov’’, Sofia, Bulgaria
A R T I C L E I N F O
A B S T R A C T
Keywords: Humeral shaft fracture (HSF) Interlocking intramedullary (IM) nailing Complications of nailing
Introduction: IM nails have gained popularity for stabilization of humeral shaft fractures (HSF). The initial enthusiasm was tempered by a number of specific complications and thus indications need to be reevaluated. Patients and methods: This retrospective study includes 111 patients with HSF subjected to a treatment protocol of IM fixation with first and second generation of humeral nails. Antegrade approach was used in 105 (94.5%) and retrograde in 6 (5.5%) patients. Reaming was performed in 51 (45.9%) fractures. The study covers a period of 10 years. Mean follow-up time was 3.5 (1–6) years. This investigation is directed at technical errors and complications, especially those corrected by secondary surgery. Results: We registered 52 (46.85%) intra-operative complications in 40 (36.04%) patients, on average 1.3 per patient. The most common were: distraction n = 5 (4.5%), long proximal locking screws n = 9 (8.1%), additional diaphyseal fracture n = 7 (6.3%) and countersinking of the nail in the humeral head n = 8 (7.2%). The number of postoperative complications was 40 (36.0%) related to 19 (17.1%) patients. Technical errors, such as distraction, longer nail and additional fractures have affected time to union and resulted in chronic shoulder pain. 36 (32.5%) secondary surgeries were needed to address these problems. When first generation nails are used, the intra-operative complication related risk increases 1.58 times, and the postoperative complication related risk is 1.67 times higher compared to second generation nails. According to Constant–Murley score excellent and very good functional results were achieved in 93 (83.78%) patients. While reaming did not influence the clinical results for both nail generations, overall better results were achieved with second generation nails. Postoperative shoulder pain has been registered in 18 (16.2%) patients. Conclusion: We registered a number of technical errors and complications, which we consider technique specific. The analysis and avoidance of these complications, related only to IM nailing of the humerus, will allow IM nails to successfully bridge the gap between functional bracing and plating. ß 2013 Elsevier Ltd. All rights reserved.
Introduction
Patients and methods
Intramedullary nail (IM) fixation is an established method of treatment of high-energy long bone fractures especially in polytrauma setting, as well as osteoporotic, impending and pathological fractures. The technical errors and complications after interlocking IM nailing of humeral shaft fractures (HSF) are recorded and discussed in only a limited number of studies [1–3]. The aim of this investigation is to analyze the results and complications of humeral shaft fractures stabilized with interlocking IM nailing.
Over a 10 year period (January 2000 to November 2010), patients with diaphyseal humeral fractures were eligible to participate in this study. Patients where a humeral nail was used for delayed post reconstruction procedures were excluded. Fractures were classified according to the AO classification and according to Gustilo–Anderson [4] if they were open. First and second generation interlocking IM nails were used. Such details were documented as intra-operative, perioperative and post-operative complications including the development of superficial infection, deep infection, iatrogenic extension of fractures proximally and distally, insertion of longer proximal and distal locking screws, metal work failure, presence of fracture gap post-operatively (distraction at the fracture site >4 mm and angulation >208), development of non-union and malunion, the necessity of insertion of additional metal work, iatrogenic nerve
* Corresponding author at: Emergency Trauma Hospital ‘‘N.I.Pirogov’’, Ist Department of Trauma Surgery, E Totleben 21, Sofia, Bulgaria. Tel.: +359 888 874040. E-mail address: [email protected] (A. Baltov). 0020–1383/$ – see front matter ß 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.injury.2013.10.044
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S10
Reaming of the canal was performed in 51 (45.9%) cases. Surgery on average was delayed 6 days (0–14). The mean length of surgery was 84.3 min. It was 70 min (range 40–195 min) in cases of closed reduction and 105 min (range 60–210 min) in cases of open reduction. The means of fixation utilized in this series was the first generation nail of R-T (Smith & Nephew) in 34 cases (30.6%) with only one option for (dynamic) interlocking proximally and distally, and the second generation nails with more interlocking options: UHN (Synthes) n = 25 (22.5%); ARHN (Sanatmetal) n = 29 (26.2%); EHN (Euro Instrument) n = 7 (6.3%) and Versa UHN (DePuy) n = 16 (14.4%), (Table 2). Polytrauma patients had mean hospital stay of 21 (10–60) days. In cases of isolated humeral fractures, the average period was 6 (4– 11) days. 109 fractures (98.2%) healed after the initial procedure with an average time to union of 3 months (range 1–8). Fixation was considered stable enough in 104 (93.7%) patients, who started early rehabilitation from 6th to 12th postoperative day. Secondary displacement with proximal protrusion and telescoping of the nail occurred in 4 (3.6%) patients, breakage of interlocking screws occurred in 3 (2.7%) and non-union in 2 cases (1.8%). Shoulder function was graded excellent and very good in 93 (83.8%) patients, according to Constant–Murley score.
damage, proximal protrusion of nails, and development of avascular necrosis of the humeral head. Bone healing was defined as the presence of bridging of fracture gap with callus in three cortices in two-plane X-rays, as well as the lack of pain on passive arm rotation. The functional status of the limb was assessed as follows: subjective complaints, presence of pain, range of movement of shoulder and elbow, muscle strength recovery. Sensitivity and motor activity in the areas innervated by radial and musculocutaneous nerves have been also recorded and analyzed. Examination of the function of shoulder was performed according to the Constant–Murley scale [5], based on age and gender at the last follow up. Deep infection was defined as osteomyelitis with presence of chronic suppuration and isolation of microbiological agent [6]. Follow-up was performed on the second and fourth week and then monthly until the sixth month. Thereafter patients were followed annually as necessary. The mean follow-up time was 3.5 years (1–6). Results Overall, 40 (36.0%) females, and 71 (64.0%) males, with an average age 49 years (range 17–83) met the inclusion criteria and formed the study group. Thirty-one (27.9%) patients sustained polytrauma with a mean Injury Severity Score of 22 (14–36). Closed fractures were seen in 101 cases (90.9%), and open fractures were seen in 10 cases (9.1%), (n = 5 of were of grade I, n = 2 of grade II, and n = 3 of grade III according to Gustilo–Anderson). Fracture distribution according to the AO classification is shown in Table 1. 20 (18%) fractures were in the proximal third, 65 (58.6%) in the middle third, 15 (13.5%) in the distal third, and 11 (9.9%) had segmental localization, engaging 2 or 3 levels. Concomitant injury of the radial nerve was found in 10 (9%) patients. It was combined motor and sensory in all cases. Antegrade surgical approach was used in 105 (94.5%) fractures and retrograde in 6 (5.5%). Closed reduction was achieved in 97 (87.4%) patients. 14 (12.6%) fractures were reduced in an open manner.
Complications We registered 52 (46.9%) intraoperative complications in 40 (36.0%) patients (an average of 1.3 in each patient), 71 (64.0%) patients showed no complication. 40 (36.0%) postoperative complications were recorded in 19 (17.1%) patients, (Tables 3 and 4). From a total of 7 (6.3%) diaphyseal intraoperative fractures, 5 fractures occurred with nails of the second generation. In four cases an unreamed nail was forcefully passed through a narrow canal. Although second generation nails did not cause a change in the method of fixation, they influenced the final functional result and resulted in delayed healing. In 2 (1.8%) patients treated with a first generation antegrade reamed IM nail a fracture of the greater
Table 1 Patient distribution according to the AO/OTA. Type A 38 (34.2%) A A A A A A A A A
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
Type B 46 (41.4%) 4 11 3 3 5 5 2 7 0
10.5% 28.9% 7.9% 7.9% 13.2% 7.9% 5.3% 18.4% 0%
B B B B B B B B B
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
Type C 27 (24.3%) 7 3 4 4 12 1 1 10 4
15.2% 6.5% 8.7% 8.7% 26.1% 2.2% 2.2% 21.7% 8.7%
C C C C C C C C C
1.1 1.2 1.3 2.1 2.2 2.3 3.1 3.2 3.3
2 6 1 3 7 2 0 5 1
7.4% 22.2% 3.7% 11.1% 25.9% 7.4% 0% 18.5% 3.7%
Table 2 Distribution according to the type of IM nail. Type of interlocking nail
Proximal interlocking
Distal interlocking
Dynamic
Static (one-plane)
Static (two-plane)
Dynamic
Static (one-plane)
Static (two-plane)
R-T/Smith & Nephew/ UHN/Synthes/ ARHN/Sanatmetal/ EHN/EuroInstrument/ Versa UHN/DePuy/
34 21 2 3 8
1 26 1 8
3 1 3
34 16 11 6 8
6 15 1 8
3 3
Total
68 (61.3%)
36 (32.4%)
7 (6.3%)
75 (67.6%)
30 (27%)
6 (5.4%)
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S11
Table 3 Intra-operative errors and complications in 44 patients. N (%)
Complications/Intraoperative Fracture of the greater tubercle of humerus Fracture of the diaphysis Supracondylar fracture Unstable fixation Missed locking Broken interosseal screw Long proximal interlocking screw Long distal interlocking screw Injury of radial nerve Iatrogenic injury of lateral antebrachial cutaneal nerve Distraction >5 mm Inadequate embedding Total
2 7 3 5 1 4 9 5 1 2 5 8
First generation 34
(1.8%) (6.3%) (2.7%) (4.5%) (0.9%) (3.6%) (8.1%) (4.5%) (0.9%) (1.8%) (4.5%) (7.2%)
2 2 1
52 (46.85%)
Second generation 77
With reaming 51
5 2 5
2 3 2 2 1
4 1 3
4 3 3 1
5 2
4 4 3 1
1 6 2
Without reaming
2 4
5 4
2 1 5
4 3
20
32
25
27
Table 4 Post-operative complications in 19 patients. Complications/Postoperative
Count (%)
Protrusion Impingement nail Impingement proximal screw Migration of the proximal screw Migration of the distal screw Debricolage Infection Delayed union Non-union Avascular necrosis Broken interlocking nail
5 12 5 4 1 2 1 3 2 2 3
Total
40 (36%)
(4.5%) (10.8%) (4.5%) (3.6%) (0.9%) (1.8%) (0.9%) (2.7%) (1.8%) (1.8%) (2.7%)
First generation 34 4 5 4
1 1
Second generation 77 1 7 1 4 1 2
1
2 2 2 2
16
24
tubercle occurred due to either lateral insertion of the nail or excessive widening of the entry portal. Supracondylar fractures were seen in n = 3 (2.7%) cases: 2 fractures in retrograde nails, when no additional fixation was needed and one case with anterograde technique, which required additional fixation with a plate (Fig. 1A). Excluded from this study is an osteoporotic supracondylar fracture of a 52-year-old female, which occurred with retrograde insertion of the nail. It required change of treatment modality; fixation was achieved with 4.5 AO compression plate (Fig. 1B). Additional fixation was performed in 5 (4.5%) cases due to residual rotational instability after locking with second generation
With reaming 51 2 5 3 1
1 1 1
Without reaming 60 3 7 2 3 1 2
1
2 1 2 2
15
25
IM nails (4 fractures were fixed with inter-fragmentary screws and in one case cerclage wires were applied). A missed proximal interlocking hole in an R-T nail was found in 1 (0.9%) case. This is a specific technical error also described by others [7]. In this case it had no consequences to bone healing and the final functional result (Fig. 2). Another technical error is intra-operative, intra-osseous breakage of distal locking screws in 4 (3.6%) cases. These screws have a small step of the thread and a large head and they break when inserted at an inaccurate angle. No fracture site instability or delay in bone healing was noted; however difficulties with nail removal are to be expected (Fig. 3).
Fig. 1. Iatrogenic supracondylar fracture: (A) fixation with additional plate, (B) conversion of the fixation after supracondylar fracture.
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S12
Fig. 3. Interosseous screw breakage. Fig. 2. Missed interlocking.
Proximal interlocking with a longer screw was considered a technical error when at least 4 mm protrusion through the medial or postero-medial cortex was found. This occurred in 9 (8.2%) cases and caused discomfort and limitation of shoulder motion in 6 patients: two cases with antero-posterior interlocking and four patients with latero-medial interlocking. This required screw removal. Three patients had oblique dynamic proximal interlocking, which did not cause complaints. Longer distal interlocking screws were found in 5 (4.5%) cases with anterograde interlocking intramedullary nailing. It caused tenderness on palpation over the posterior surface of the arm in slim patients. Their removal was required in two cases. Prominence of the nail (Fig. 4) was found on postoperative Xrays in 8 (7.2%) cases. It caused discomfort, hampered recovery of shoulder motion and delayed return of overall arm function. The type of complications leading to shoulder discomfort, besides proximal protrusion of nail, were various including: telescopic effect in 5 (4.5%) patients with osteoporosis, incomplete embedding of the nail at surgery, leading to subacromial impingement in 12 (10.8%) cases, and in 5 (4.5%) patients impingement of the head of proximal interlocking screw. This was significantly more frequent in interlocking IM nailing with first generation nails (Fisher exact test p < 0.01). Shoulder related complaints led to additional surgical procedures in 18 (16.2%) cases: 14 extractions of nail and 4 removals of the interlocking proximal screws. This led to improvement of the final functional result in 12 cases from good to very good, and in 6 cases from satisfactory to good. Damage to the radial nerve was reported postoperatively in 1 (0.9%) patient, who recovered spontaneously after 6 months. Two (1.8%) cases of injury of the lateral cutaneal nerve of the forearm with persistent paresthesia in the dermatome were detected and
Fig. 4. Protrusion of a nail and a proximal screw.
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
S13
Fig. 5. Non-union due to: (A) distraction; (B) unstable interlocking intramedullary nailing; (C) revision with a 4.5 DCP.
were attributed to distal anteroposterior interlocking of the nail. Osteomyelitis was diagnosed in 1 patient. Fracture distraction was recorded in 14 (12.6%) cases and in 5 (4.5%) patients it measured more than 4 mm. In 3 (2.7%) patients this caused delayed healing and in 2 (1.8%) non-union. According to Fisher’s exact test the risk of non-union was significantly higher if the fractures were locked in distraction p < 0.01 (p-value = 0.000000164). In the cases of delayed union, within 6 months after fixation, we found 1 (0.9%) screw breakage and 4 (3.6%) migrations of proximal locking screws with antegrade IM nails, and 1 (0.9%) distal locking screw with retrograde nails. However no additional surgical interventions were performed. Non-union was diagnosed in 2 (1.8%) patients. The first case was a high-energy fracture of B 2.2 type, which was fixed in distraction of 10 mm with retrograde nail of second generation, insufficient uni-planar proximal and dynamic distal locking. This resulted in residual rotational instability and breakage of the proximal locking screw. After 18 months the nail was removed and fixation achieved with 4.5 mm DCP, (autologous bone graft was harvested locally from a consolidated butterfly fragment), (see Fig. 5). The second instance of non-union occurred after a high-energy A3.2 fracture, fixed with an antegrade interlocked IM nail, locked
dynamically, proximally and distally. No reaming was performed of a narrow canal (less than 8 mm) and fracture was locked in distraction (11 mm). At 12 months, due to a breakage of a distal locking screw and a secondary rotational instability, the nail was removed, fixation was revised with a 4.5 mm DCP and autologous bone grafting was performed. In both non-unions, healing occurred after a period of 6 months and full range of motion of adjacent joints was restored. One case (0.9%) of debricolage occurred with a fracture of type C1.1 and additional fracture of the proximal humerus with comminution of the greater tubercle. Fixation was performed with a Cloverleaf plate, without nail removal. Fracture healed in 3 months and good shoulder function was restored (Fig. 6). We diagnosed AVN of the humeral head as a late complication in 2 (1.8%) patients with associated fracture of the proximal humerus and of advanced age. This resulted in decreased range of shoulder movement. No further surgery was performed in these low-demand patients. Varus deformity in excess of 258 was recorded in 2 (1.8%) patients with limited range of shoulder abduction (1408 in both cases). In 2 (1.8%) cases with a floating elbow, a significantly decreased range of elbow extension was measured (Fisher’s exact test p < 0.01), (0.004586).
Fig. 6. (A) Debricolage; (B) revision with the nail in situ.
S14
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
The intra- and postoperative complications required a total of 36 (32.4%) additional procedures in 33 patients, averaging 1.09 procedures in each case. Eventually, 6 (5.4%) patients were considered to have poor or satisfactory functional results. Discussion In an updated meta-analysis focusing on total complication rate (TCR), Heineman et al. [8], have found that plate fixation has a lower TCR compared to interlocking IM nailing. They suggested that although the difference is decreasing, there is still evidence to support the view that plates are more favourable than nails in terms of TCR. In an earlier version the same group had found that plate fixation decreases the risk of revision surgery by up to 74% and the likelihood of shoulder impingement by nearly 90%. It was demonstrated that plated fractures have a lower revision rate in comparison to those treated with an IM nail (6% versus 18%, p = 0.03) and there is a similar disparity between the incidence of shoulder complaints (1% versus 21%, p = 0.002). Moreover, delayed healing and non-union were more frequent in the group of IM nails (8 out of 73/11%), compared to those fixed by a plate (5 out of 83/ 6%) [8]. In the randomized study of Singisetti et al. [9], 65% of patients treated with first generation ante-grade interlocking IM nailing were found to have excellent and good shoulder function, while in cases of plate fixation, it was found in 95%. They also pointed out that only 50% of nailed fractures have achieved union at 4 months, compared to 75% of those fixed by a plate. The issue of the entry portal is a subject of multiple studies but is still a controversial point. The main problem in antegrade insertion of a nail lies in the fact that the axis of the medullary canal exits through the joint surface of the humeral head and the rotator cuff, which limits the choice of insertion sites [10,11]. There is still a debate about the incidence and severity of subacromial impingement and persistent pain, as well as about assessment methods and adequacy of follow-up regarding those complaints. Ajmal et al. [12] reported poor shoulder function in 41% of patients with an entry portal affecting the rotator cuff. Several randomized trials have raised awareness of the fact that subacromial impingement and persistent pain are found in one of every 5 patients with anterograde nailing. Heinsen and Hempel [13] have used ultrasound and found transient injuries of the rotator cuff prior to anterograde IM fixation in 10 of 36 cases. Postoperatively, 5 ruptures of the tendon of supraspinatus muscle were recorded, three of them resulted from nail protrusion, the rest, were a consequence of accompanying degenerative changes. They concluded that anterograde IM fixation does not increase the likelihood of rupture of the rotatory cuff, provided the surgical technique is followed correctly: complete embedding of the nail end in the humeral head and careful repair of the tendon of the supraspinatus muscle. Impingement was also described as a consequence of too proud proximal interlocking screws, as well as protrusion of the proximal end of nail [14,15]. A reliable strategy for prevention of shoulder pain is retrograde nail insertion through extra-articular entry portal, which in many cases may give good and excellent results [16,17]. The complications are also well-known: supracondylar fracture, pain and limited elbow motion, weakness in the triceps muscle, periarticular ectopic ossifications and adhesions in the region of elbow, with consequent limitation of function [18]. According to Farragos et al. [1], there is still an argument as to which technique more reliably preserves shoulder and elbow joint function. In this series of predominantly antegrade nails, there are several complications that have led to impairment of shoulder function.
Proximal protrusion of the nail with so called ‘‘telescopic effect’’ was noted in 4 (3.9%) patients, incomplete embedding of nail, which resulted in subacromial impingement recorded in a total of 12 (11.7%) patients, and impingement caused by the head of a proximal interlocking screw in 5 (4.9%) patients. 18 (17.5%) additional surgical procedures were needed to address those issues [1]. As a whole, 65 patients had no pain in the shoulder, 28 patients had mild and moderate pain. However, 13 had intermittent pain, 4 had moderate to severe pain and 1 was in constant pain. A possible reason for a decreased range of motion and lower functional scores in patients who underwent IM nailing was offered by Li et al. [19]. They measured rotational malalignment by postoperative CT and found that 27.2% of nailed patients had an internally rotated humeral head, which affected negatively shoulder motion. Insertion of locking screws in IM nailing of humerus can incur additional vascular and nerve injury, specific only to the applied technique of fixation. At risk are: axillary nerve, lateral cutaneal nerve of forearm, and median nerve. In cases of proximal interlocking in the frontal and sagittal plane, both branches of the axillary nerve can be damaged [20,21]. Screw insertion in the oblique position is considered potentially less hazardous. Notwithstanding the theoretically high risk, only one case of iatrogenic injury to these nerves in anterograde and one case of retrograde interlocking IM nailing has been described so far [22]. Antero-posterior distal locking is considered safer, but the risk of injury to the musculocutaneal nerve is well recognized. Two cases of this type are described in scientific literature [23,24]. Two types of iatrogenic fractures during interlocking IM nailing are to be distinguished. Stable fractures do not require a change in the treatment method or supplementary fixation, only occasional external protection. Unstable fractures demand additional fixation. This prolongs surgery time, increases blood loss and radiographic assessment [25,26]. The incidence of iatrogenic fractures varies between 2% and 22%, for both surgical techniques, but is higher in cases of retrograde interlocking IM nailing. According to Fernandes et al. [27], it is 10.7%, Verbruggen et al. [28] suggested the incidence is even higher up to 17%. Iatrogenic fracture of the greater tubercle during anterograde IM nailing (2–11%) occurs with a too lateral insertion of a rigid nail. When the end of the nail abuts the medial cortex it generates pressure on the lateral cortex of the head of humerus via a lever from the acromion [29]. Another common iatrogenic fracture is that of the diaphysis (incidence 1–5%). Insertion of a rigid nail through a narrow medullary canal with no prior reaming can result in shaft fragmentation when driving in the nail. Additional fractures can occur in the proximal third of the diaphysis or the supracondylar area. In the latter case the reason behind this is the straight, but narrow distal diaphyseal canal, most frequently in osteoporotic bone [30]. Garnavos and Kanakaris [31] reported a fracture of the diaphysis after eccentric drilling during distal locking of the nail. In this study of 111 fractures, there were 9 (8.7%) iatrogenic fractures with anterograde approach and 3 (50%) – with the retrograde technique. Angulations are registered in all methods of treatment. In cases of nail fixation, excessive angulation is quite rare and the observed deformities can be accepted [32]. Varus deformity is typical of fractures of the proximal one third of the diaphysis and is caused by deltoid muscle pull. If not taken into account, this results in a too lateral nail insertion point. The proximal curvature of nail (from 58 to 108), especially with first generation nails, which have only dynamic interlocking, will lead to fixation in varus angulation (up to 308).
A. Baltov et al. / Injury, Int. J. Care Injured 45S (2014) S9–S15
In this series 4 patients were noted to have a varus deformity ranging between 168 and 258. Reduced shoulder abduction up to 308 was measured which affected overall shoulder function. Infection after interlocking IM nailing is a comparatively rare complication. It has been recorded in 0–6% of cases. Many series, though, do not distinguish between closed and open fractures. [33,34]. In this study there was only one (1.07%) patient (type 2 diabetes mellitus) with osteomyelitis after interlocking IM humeral nailing. The nail was removed after 10 months, when the fracture was healed. Infection (Staphylococcus aureus) was successfully treated by debridement, canal reaming and lavage. This did not seem to influence the functional result. In conclusion, humeral fractures can be treated with different operative techniques [35–40]. Specific technical errors and complications related only to intramedullary nails were registered in this retrospective case series. Strict adherence to proven indications (high-energy C-type fractures, polytrauma and impending fractures) and good surgical technique will allow the interlocking IM nailing to bridge the gap between functional bracing and the plating and to achieve better results compared to both of them. Conflict of interest statement No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study. References [1] Farragos AF, Schemitsch EH, McKee MD. Complications of intramedullary nailing for fractures of the humeral shaft: a review. J Orthop Trauma 1999;13:258–67. [2] Garnavos C. Intramedullary nailing for humeral shaft fractures: the misunderstood poor relative. Curr Orthop 2001;15:69–75. [3] Lin J, Shen PW, Hou SM. Complications of locked nailing in humeral shaft fractures. J Trauma 2003;54(5):943–9. [4] Gustilo RB, Merkow RL, Templeman D. Current concepts review. The management of open fractures. J Bone Joint Surg 1990;72A:299–304. [5] Constant CR, Murley AHG. A clinical method of functional assessment of the shoulder. Clin Orthop 1987;214:160–4. [6] McKee MD. Fracture of the shaft of the humerus. In: Bucholz RW, editor. Rockwood Green’s fractures in adults. 6th ed., Philadelphia: Lippincott Williams & Wilkins; 2006. p. 1117–59. [7] Asencio G, Buscayret F, Trabelsi A, Bertin R, Hammami R, Megy B, et al. Intramedullary interlocking nailing for humeral fractures: report of 38 cases treated by Russell and Taylor nail. Rev Chir Orthop 2001;87:749–57. [8] Heineman DJ, Poolman RW, Nork SE, Ponsen KJ, Bhandari M. Plate fixation or intramedullary fixation of humeral shaft fractures. Acta Orthop 2010;81(2):216–23. Updated in: Acta Orthop 2010; 81 (4) 517. Updated Acta Orthop 83 (3) 317–318. [9] Singisetti K, Ambedkar M. Nailing versus plating in humerus shaft fractures: a prospective comparative study. Int Orthop 2010;34:571–6. [10] Cox MA, Dolan M, Synnott K, McElwain JP. Closed interlocking nailing of humeral shaft fractures with the Russell-Taylor nail. J Orthop Trauma 2000;14:349–53. [11] Flinkkila T, Hyvo¨nen P, Lakovaara M, Linden T, Ristiniemi J, Ha¨ma¨la¨inen M. Intramedullary nailing of humeral shaft fractures. A retrospective study of 126 cases. Acta Orthop Scand 1999;70:133–6. [12] Ajmal M, O’Sullivan MO, McCabe J, Curtin W. Antegrade locked intramedullary nailing in humeral shaft fractures. Injury 2001;32:692–4. [13] Heinsen J, Hempel D. Praoperatve sonographische Untersuchung der Rotatorenmanchette vor antegrader Nagelung von Humerusschaftfracturen. Osteosynthese International 2001;9(Suppl 2):S85–7. [14] Dimakopoulos P, Papadopoulos AX, Papas M, Panagopoulos A, Lambiris E. Modified extra rotator-cuff entry point in antegrade humeral nailing. Arch Orthop Trauma Surg 2005;125:27–32.
S15
[15] Pestatodes G, Karataglis D, Papadopoulos P, Christoforides J, Gigis J, Pournaras J. Antegrade interlocking nailing of humeral shaft fractures. J Orthop Sci 2004;9:247–52. [16] Sanzana ES, Dummer RE, Castro JP, Diaz EA. Intramedullary nailing of humeral shaft fractures. Int Orthop 2002;26:211–3. [17] Vescei N, Kolonja A, Mousavi M, Ve´csei V. Intramedullary fixation of humerus shaft fractures. An analysis of complications of 2 implants with special reference to outcome after management with the unreamed humerus interlocking nail. Wien Klin Wochenschrift 2001;113:597–604. [18] Loitz D, Ko¨nnecker H, Illgner A, Reilmann H. Retrograde intramedullary nailing of humeral fractures with new implants. Analysis of 120 consecutive cases. Unfallchirurg 1998;101:543–50. [19] Li Y, Wang C, Wang M, Huang L, Huang Q. Postoperative malrotation of humeral shaft fracture after plating compared with intramedullary nailing. J Shoulder Elbow Surg 2011;20:947–54. [20] Albritton MJ, Barnes CJ, Basamania CJ, Karas SG. Relationship of the axillary nerve to the proximal screws of a Flexible Humeral Nail system: an anatomical study. J Orthop Trauma 2003;17:411–4. [21] Prince EJ, Breien KM, Fehringer EV, Mormino MA. The relationship of proximal locking screws to the axillary nerve during antegrade humeral nail insertion of four commercially available implants. J Orthop Trauma 2004;18:585–8. [22] Lo¨gters T, Wild M, Windolf J, Linhart W. Axillary nerve palsy after retrograde humeral nailing: clinical conformation of an anatomical fear. Arch Orthop Trauma Surg 2008;128:1431–5. [23] Blyth MJ, Macleod CM, Asante DK, Kinninmonth AW. Iatrogenic nerve injury with the Russell-Taylor humeral nail. Injury 2003;34:227–8. [24] Kesemenli CC, Subasi M, Arslan H, Necmioglu S, Kapukaya A. Comparison between the results of intramedullary nailing and compression plate fixation in the treatment of humerus fractures. Acta Orthop Traumatol Turc 2003;37(2):120–5. [25] Apard T, Lahogue JF, Prove S, Hubert L, Talha A, Cronier P, et al. Retrograde locked nailing of humeral shaft fractures: a prospective study of 58 cases. Rev Chir Reparatrice Appar Mot 2006;92:19–26. [26] Muckley T, Diefenbeck M, Sorkin AT, Beimel C, Goebel M, Bu¨hren V. Results o the T2 humeral nailing system with special focus on compression interlocking. Injury 2008;39:299–305. [27] Fernandez FF, Winkler H, Merkel G, Freudenberg S, Wentzensen A. Erste klinische Erfahrungen mit der unaufgebohrten Marknagelung bei Humerusschaftfracturen. AktTraumatol 1999;29:269–76. [28] Verbruggen JPAM, Adriaens S, Goessens ML, Stapert JWJL. Reamed nailing of humeral fractures: long-term results in 77 cases. Osteo Trauma Care 2002;10:192–6. [29] Herbst U, Ruettger K, Mockwitz J. Experiences with the Russell-Taylor nail in humeral shaft fractures – an analysis of postoperative results and complications. Osteo Trauma Care 2003;11:13–20. [30] Fernandez FF, Matschke S, Hu¨lsenbeck A, Egenolf M, Wentzensen A. Five years’ clinical experience with the unreamed humeral nail in the treatment of humeral shaft fractures. Injury 2004;35:264–71. [31] Garnavos C, Kanakaris N. Stress fracture due to unsuccessful targeting during intramedullary nailing. Eur J Trauma 2003;29:105–7. [32] Rommens PM, Kuechle R, Bord T, Lewens T, Engelmann R, Blum J. Humeral nailing revisited. Injury 2008;39:1319–28. [33] Blum J, Rommens PM. Proximal interlocking of humeral intramedullary nails and the risk of injury of the axillary nerve. Unfallchirurg 2002;105:9–13. [34] Chapman JR, Bradford Henley M, Agel, Benca PJ. Randomized prospective study of humeral shaft fracture fixation: intramedullary nails versus plates. J Orthop Trauma 2000;14:162–6. [35] Karataglis D, Stavridis SI, Petsatodis G, Papadopoulos P, Christodoulou A. New trends in fixation of proximal humeral fractures: a review. Injury 2011;42(April (4)):330–8. [36] Biber R, Zirngibl B, Bail HJ, Stedtfeld HW. An innovative technique of rear entry creation for retrograde humeral nailing: how to avoid iatrogenic comminution. Injury 2013;44(April (4)):514–7. [37] Matassi F, Angeloni R, Carulli C, Civinini R, Di Bella L, Redl B, et al. Locking plate and fibular allograft augmentation in unstable fractures of proximal humerus. Injury 2012;43(November (11)):1939–42. [38] Noda M, Saegusa Y, Maeda T. Does the location of the entry point affect the reduction of proximal humeral fractures? A cadaveric study. Injury 2011;42 (Suppl 4):S35–40. [39] Hardeman F, Bollars P, Donnelly M, Bellemans J, Nijs S. Predictive factors for functional outcome and failure in angular stable osteosynthesis of the proximal humerus. Injury 2012;43(February (2)):153–8. [40] Collis PN, Clegg TE, Seligson D. The invisible nail: a technique report of treatment of a pathological humerus fracture with a radiolucent intramedullary nail. Injury 2011;42(April (4)):424–6.