Volume 134, Issue 1 , Pages 218-224, July 2007
A reinforced sternal wiring technique for transverse thoracosternotomy closure in bilateral lung transplantation: From biomechanical test to clinical application
Article Outline
Objectives
A high incidence of failure of transverse thoracosternotomy closure, involving the loops of wire cutting through the sternum, remains a significant morbidity after bilateral lung transplantation. We postulated that placing peristernal wires inside the usual longitudinal wires could prevent the longitudinal wires from cutting through the sternum. The aims of this study were to investigate the biomechanical and clinical efficacy of the proposed reinforced sternal closure technique.
Methods
In vitro, 24 artificial sternal models were wired with the reinforced or conventional wiring techniques and were tested either by means of longitudinal distraction or anterior–posterior shear (n = 6 per group). In vivo, the 6-month outcomes of 70 bilateral lung transplantations, including 27 reinforced and 43 conventional wiring techniques, were assessed.
Results
Reinforced wiring was stronger than conventional wiring for both longitudinal distraction (yield load: 585 ± 60 vs 334 ± 21 N [P = .03]; maximum load: 807 ± 60 vs 525 ± 34 N [P = .03]; postyield stiffness: 91.0 ± 22.0 vs 32.8 ± 11.8 N/mm [P = .04]) and anterior–posterior shear (yield load: 405 ± 9 vs 364 ± 16 N [P = .03]; postyield stiffness: 47.4 ± 6.1 vs 27.5 ± 5.1 N/mm [P = .04]). In multivariate analysis, the use of the conventional wiring technique (odds ratio, 5.38; P = .04) and osteoporosis (odds ratio, 18.31; P = .0005) were significant risk factors associated with sternal dehiscence. In the patients with osteoporosis (n = 25), the incidence of sternal dehiscence in the reinforced wiring group (4/16 [25%]) was significantly lower than that in the conventional wiring group (7/9 [78%], P = .02).
Conclusion
Osteoporosis is a significant risk factor for sternal dehiscence after bilateral lung transplantation. The new reinforced sternal wiring technique provides biomechanically superior fixation of the sternum and clinically reduces the incidence of sternal dehiscence in high-risk osteoporotic patients undergoing bilateral lung transplantation.
CTSNet classification: 12
Abbreviations and Acronyms: BLT, bilateral lung transplantation, pcf, per cubic foot
The transverse thoracosternotomy (clamshell incision) is a standard approach for bilateral lung transplantation (BLT).1, 2, 3 This technique allows optimal exposure of both pleural spaces in both pediatric and adult patients.1, 2, 3, 4 In addition, this technique allows easy access to the ascending aorta and the right atrium to initiate cardiopulmonary bypass when needed during BLT.1, 2, 3, 4, 5 However, a high incidence of poor sternal healing, ranging from 32% to 46% and often requiring surgical intervention, is a significant problem with this incision.3, 6, 7, 8, 9 Many factors contribute to the high incidence of sternal complications in patients undergoing transplantation: osteoporosis, poor nutritional status, diabetes mellitus, obesity, heavy smoking, preoperative chronic use of corticosteroids, and posttransplantation immunosupression.
The standard conventional method of closing the sternum is the use of 2 or 3 loops of stainless-steel wire. Failure of this closure technique usually involves the wire cutting through the bone, thus producing separation of the sternal edges with consequent wound dehiscence, overriding of the sternal edges, and infection.3, 6, 7, 8, 9 Several techniques to overcome this problem have been devised, including the use of intramedullary Steinmann pins or Kirschner wires, sternal plates, peristernal cables, or bilateral anterolateral thoracotomies without sternal division.3, 4, 8, 9, 10 Although these techniques have improved the results considerably, there are still potential disadvantages, including cost issues for the new materials, such as pins, plates, or cables, and migration for intramedullary pins. The bilateral thoracotomy technique provides limited access in patients who require concomitant cardiac surgery, cardiopulmonary bypass, or both, as well as in those who have restrictive pulmonary disease with small pleural cavities or pulmonary hypertension with cardiomegaly and in those with dense pleural adhesions.8, 10, 11, 12
The technique of reinforcement with parasternal wires inside the usual transverse peristernal wires has been successfully applied for median sternotomy closure in cardiac surgery.13, 14, 15, 16 We postulated that this modified median sternotomy closure technique could be applied to transverse sternotomy closure after lung transplantation by placing peristernal wires inside the usual longitudinal wires. These additional wires have the potential to prevent the longitudinal wires from cutting through the sternum, especially in osteoporotic patients (Figure 1). In this new technique the same wire used for standard sternotomy closure could be used with the advantages of cost-effectiveness, reliability, and technical familiarity. To verify the biomechanical integrity, we carried out mechanical tests of the proposed reinforced wiring technique using an artificial model of human sternum. The aims of this study were as follows: (1) to investigate the biomechanical efficacy of transverse sternotomy closure with peristernal reinforcement by means of mechanical testing in vitro with an artificial model of human sternum and (2) to introduce the new technique into our lung transplantation program in those patients having a transverse sternotomy for BLT, especially those with significant osteoporosis, and to assess the midterm outcomes.
Figure 1.
The new reinforced wiring technique. Two peristernal stainless-steel wires are placed on the each side of the sternum at the level of the third and fifth intercostal spaces inside the conventional 2 longitudinal wires, which cross the sternotomy line.
Materials and Methods
Biomechanical Investigation
Model of human sternumBiomechanical analyses were performed at the Swinburne University Biomechanics Laboratory with an artificial sternal model constructed from rigid polyurethane foam. This material is commonly used as a test medium in the in vitro simulation of human bone (Sawbones; Pacific Research Laboratories, Inc, Vashon, Wash).17, 18 Pilot experiments were performed to identify the optimum model with appropriate material density, such that a no. 6 stainless-steel monofilament wire (Ethicon, Somerville, NJ) would cut through the material at a clinically relevant force of 150 N. The average tension of a wire in human median sternotomy closure is known to be 100 N.19 Various models, including cellular rigid polyurethane foam with a bone density of 0.12 g/mL (7.5 per cubic foot [pcf]) and 0.32 g/mL (20 pcf) and laminated blocks with or without an E-glass–filled epoxy sheet (1.7 g/mL, 106 pcf), were tested to optimize the arrangement for the simulation of human cortical and cancellous bone. A cellular rigid polyurethane foam block with a bone density of 0.32 g/mL (20 pcf) without lamination was selected for testing.
Sternal wiring techniquesFor the reinforced wiring group, a length of no. 6 stainless-steel wire was threaded horizontally around the upper half of the divided artificial sternum 1.5 cm from the sternotomy so as to form an encircling loop, which was then twisted tight (Figure 2). This procedure was repeated in the lower half of the sternum. The 2 standard vertical (longitudinal) wires, also no. 6 stainless steel, were then placed through the sternum on either side of the sternotomy, just beyond the 2 horizontal reinforcing wires. These vertical wires were then twisted tightly such that the strain was taken mainly by the reinforcing wires. For the conventional wiring group, 2 standard vertical wires were placed without peristernal wires.
Figure 2.
The appearance of the bone models before and after destructive testing. The parasternal wires in the reinforced wiring group prevented the longitudinal wires from cutting through the bone until the material itself parted. In contrast, failure of fixation in the conventional wiring group occurred because of the wires cutting through the bone.
Twenty-four sternal models wired either by using the reinforced or conventional technique were randomly allocated for testing (n = 6 per group) either by using longitudinal distraction or anterior–posterior shear. The half of the bone model in each side was reinforced with metal plates for attachment to the testing machine (MTS model 819; MTS System Co, Minneapolis, Minn). The testing machine was set in the displacement control mode and distracted the 2 sternal fragments at a rate of 1 mm/s. The material testing system digitally acquired load and displacement data every 0.5 seconds. Three biomechanical variables, including yield load (in newtons), maximum load (in newtons), and postyield stiffness (in newtons per millimeter), were used in this analysis (Figure 3). The yield load was defined as the point on the load-displacement curve where the curve became nonlinear at the initiation of the wire cutting through the bone. The maximum load was the ultimate strength of the construct.17 The postyield stiffness was the slope of the load-displacement curve over a 2-mm displacement beyond the yield point.
Figure 3.
Three biomechanical variables, including yield loading (in newtons), maximum loading (in newtons), and postyield stiffness (in newtons per millimeter), were used in this analysis. The yield load was defined as the point on the load-displacement curve at which the curve became nonlinear at the initiation of the wire cutting through the bone. The maximum load was the ultimate strength of the construct. The postyield stiffness was the slope of the load-displacement curve over a 2-mm displacement beyond the yield point.
Clinical Investigation
Patients and study periodFrom March 2004 through June 2006, a total of 72 BLTs were performed at The Alfred Hospital. All but 2 patients survived and completed the 6-month postoperative follow-up, and thus 70 BLTs were included in this study. Institutional ethics committee advice was sought, and written ethical exemption from informed consent to use and evaluate the modified technique was obtained because this was deemed a quality assurance activity.
Sternal wiring techniques and study groupPatients received either the new reinforced wiring technique (reinforced wiring group) or the conventional wiring technique (conventional wiring group) for transverse sternotomy wound closure after BLT. The sternal wiring technique was chosen on the basis of surgeon’s preference, and the reinforced technique was preferentially used in patients with significant osteoporosis. For the reinforced wiring group, a length of no. 6 stainless-steel wire was threaded horizontally around the upper half of the divided sternum (peristernal wire) at the level of the third intercostal space so as to form an encircling loop, which was then twisted tight (Figure 1). This procedure was repeated in the lower half of the sternum at the level of the fifth intercostal space. The 2 standard vertical junction wires (longitudinal wires), also no. 6 stainless-steel, were then placed through the sternum on either side of the sternotomy, just beyond the 2 peristernal reinforcing wires. For the conventional wiring group, 2 standard longitudinal wires were placed without peristernal wires. For the remainder of the chest closure, 2 figure-eight pericostal sutures were placed on either side of the chest in the 2 groups.
Transplantation protocol and surgical procedureDonor and recipient selection, operative technique, and posttransplantation management have been described elsewhere.5, 20, 21, 22, 23 Bone mineral densities were assessed by using dual-energy x-ray absorptiometry before BLT, and osteoporosis was defined as a lumbar spine T score of less than −2.5, as per the World Health Organization standard.24 All transplantation candidates with osteopenia/osteoporosis were treated with elemental calcium (1000-1500 mg/d) and vitamin D (400 IU/d) during the waiting period for transplantation,25 and these treatments were continued after BLT.
Data collectionData were retrieved from the institutional transplantation database and medical records. Sternal dehiscence was defined as sternal separation or override on the lateral chest radiograph requiring surgical intervention. As potential clinical risk factors for transverse sternotomy wound dehiscence, we used the well-described risk factors for median sternotomy wound dehiscence after cardiac surgery and for posttransplantation bone fracture (Table 1).6, 13, 14, 15, 16, 24, 25
TABLE 1. Clinical investigation: Pretransplantation recipient demographics in bilateral lung transplantations
| Recipient demographics | Bilateral lung transplantations (n = 70) |
|---|---|
| Age (y) | 45 ±2 |
| Sex (male/female) | 38/32 |
| Diagnosis | |
| Chronic obstructive pulmonary disease | 28(40%) |
| Cystic fibrosis | 23(33%) |
| Interstitial pulmonary fibrosis | 13(19%) |
| Pulmonary hypertension | 6(8%) |
| Bone mineral density | |
| Lumbar spine (g/cm2) | 1.03±0.02 |
| Lumbar spine (T score) | −1.7±0.2 |
| Femoral neck (g/cm2) | 0.86±0.02 |
| Femoral neck (T score) | −1.4±0.2 |
| Osteoprosis (T score <−2.5) | 25(36%) |
| Corticosteroid use | 24(34%) |
| Diabetes | 2(3%) |
| Body mass index | 22.3±0.5 |
| Smoking history >20 pack-years | 31(44%) |
Continuous data were expressed as means ± standard errors of the mean, and categoric data were reported as counts and proportions. Comparisons between the 2 groups were made with the unpaired t test for parametric variables, the Mann-Whitney U test for nonparametric variables, and the Fisher exact test for categoric variables. Univariate and multivariate risk factor analysis for sternal dehiscence was performed by using logistic regression. All possible prediction variables listed in Table 1 were entered into a univariate analysis. Risk factors with a level of significance defined as a P value of less than .1 in the univariate analysis were then entered into the multivariate model. The sternal wiring technique variable (P = .64 on the univariate analysis) was also entered into the multivariate model. Analysis was performed with the Statview 5.0 software package (SAS Institute, Inc, Cary, NC).
Results
Biomechanical Investigation
Longitudinal distractionThe mean load-displacement curve of longitudinal distraction is shown in Figure 4, A. In the reinforced wiring group yield load, maximum load, and postyield stiffness were all significantly greater than those in the conventional wiring group (Table 2). The appearance of the bone models before and after failure is shown in Figure 2. Breakage in the conventional wiring group occurred because of the longitudinal wires cutting through the bone. In contrast, the peristernal wires in the reinforced wiring group prevented the longitudinal wires from cutting through the bone, and breakage did not occur until the material itself parted.
Figure 4.
Load-displacement curve of biomechanical testing. A, Longitudinal distraction. In the reinforced wiring group yield load (P = .03), maximum load (P = .03), and postyield stiffness (P = .04) were significantly greater than in the conventional wiring group. B, Anterior–posterior shear. Yield load (P = .03) and postyield stiffness (P = .04) in the reinforced wiring group were significantly greater than in the conventional wiring group.
TABLE 2. Biomechanical investigation: Load-displacement tests
| Variable | Reinforced wiring | Conventional wiring | P value |
|---|---|---|---|
| Longitudinal distraction | n = 6 | n = 6 | |
| Yield load (N) | 585±60 | 334±21 | .03 |
| Maximum load (N) | 807±60 | 525±34 | .03 |
| Postyield stiffness (N/mm) | 91.0±22.0 | 32.8±11.8 | .04 |
| Anterior–posterior shear | n = 6 | n = 6 | |
| Yield load (N) | 405±9 | 364±16 | .03 |
| Postyield stiffness (N/mm) | 47.4±6.1 | 27.5±5.1 | .04 |
The load-displacement curve of anterior–posterior shear is shown in Figure 4, B. Yield load and postyield stiffness in the reinforced wiring group was significantly greater than that in the conventional wiring group (Table 2). Bones were destroyed before reaching the maximum load not at the level of wiring but at the level of the end of the metallic plates reinforcing the bone for attachment to the measuring system. Therefore the maximum load was not measurable in this experiment.
Clinical Investigation
Recipient demographicsRecipient pretransplantation demographics are shown in Table 1. Twenty-five (36%) patients were given diagnoses of osteoporosis, and 24 (34%) patients were receiving corticosteroids for their underlying disease before BLT. Twenty-seven patients received the reinforced wiring technique and 43 recipients received the conventional wiring technique for transverse sternotomy wound closure.
Incidence of sternal dehiscenceA sternal dehiscence was seen in 15 (21%) of 70 patients after BLT. There was no significant difference in the incidence of sternal dehiscence between the reinforced and the conventional wiring groups (19% and 23%, respectively; P = .77). Of the 15 patients, failure of sternal fixation was due to the longitudinal wires cutting into the sternum in 14 patients (in both group), and in 1 patient it was caused by breakage of the longitudinal wire (in the reinforced wiring group). All failures occurred within 4 months after BLT (range, 1-4 months). Significant bacterial isolates, including methicillin-resistant Staphylococcus aureus, Pseudomonas aeruginosa, and Candida albicans, were detected by means of wound culture in 5 patients (3 in the reinforced wiring group and 2 in the conventional wiring group). These patients were treated with antibiotics and surgical intervention, including rewiring and debridement. One patient with persistent deep sternal infection required sternal resection and muscle flap construction. All 15 patients survived to discharge from the hospital, and no mortality was seen during the 6-month follow-up period.
Risk factor analysisUnivariate and multivariate analyses for risk factors of sternal dehiscence are shown in Table 3. In the univariate analysis osteoporosis was the only factor significantly associated with sternal dehiscence (P = .002). In the multivariate analysis use of the conventional wiring technique (odds ratio, 5.38; P = .04) and osteoporosis (odds ratio, 18.31; P = .0005) were significant risk factors associated with the development of sternal dehiscence.
TABLE 3. Clinical investigation: Logistic regression analysis for risk factors of sternal dehiscence
| Variable | Odds ratio (95% CI) | P value |
|---|---|---|
| Univariate analysis | ||
| Conventional sternal wiring technique | 1.33(0.40–4.43) | .64 |
| Osteoporosis | 8.05(2.21–29.42) | .002 |
| Multivariate analysis | ||
| Conventional sternal wiring technique | 5.38(1.05–27.50) | .04 |
| Osteoporosis | 18.31(3.58–93.69) | .0005 |
A subanalysis of the patients with osteoporosis (n = 25) showed that the incidence of sternal dehiscence in the reinforced wiring group (4/16 [25%]) was significantly lower than that in the conventional wiring group (7/9 [78%], P = .02). In the patients without osteoporosis (n = 45), there was no significant difference in the incidence of sternal dehiscence between the reinforced wiring (1/11 [9%]) and conventional wiring (3/34 [9%], P > .99) groups.
Discussion
Failure of transverse sternotomy wound closure remains a significant morbidity after BLT.3, 6, 7, 8, 9 Osteoporosis is one of the major risk factors for posttransplantation fragility of the bones. Osteoporosis is increased and wound healing delayed by the use of corticosteroids before and after lung transplantation.24, 25 In this study osteoporosis was found to be a significant risk factor for transverse sternotomy wound dehiscence after BLT. A simple modification of the conventional wiring technique with additional peristernal wiring inside the longitudinal wires was designed and tested biomechanically and clinically to prevent posttransplantation sternal dehiscence. This new reinforced technique was shown to have superior fixation compared with the conventional technique by means of biomechanical tests. Moreover, this new technique significantly reduced sternal dehiscence in high-risk osteoporotic patients undergoing clinical lung transplantation.
Several techniques have been tried to overcome this problem, including the use of intramedullary Steinmann pins or Kirschner wires, sternal plates, peristernal cables, or ultimately 2 anterolateral thoracotomies without sternal division.3, 4, 8, 9, 10 Although these techniques have decreased the incidence of sternal dehiscence, there are potential disadvantages of their routine application.8, 10, 12 Sternal pins, plates, or cables are foreign bodies, and thus removal of these materials is necessary when sternal wound infection occurs. There also is a cost issue for these new materials. Intramedullary pins can migrate and thus require extra operations.
From the view of sternal wound healing, the technique with 2 anterolateral thoracotomies without sternal division might be an optimal method of surgical access. However, there are technical difficulties with this approach because of inferior exposure of hilar structures compared with transverse thoracosternotomy. Switchover from this technique to sternal division is occasionally necessary because of limited access, especially in the following patient groups: those with restrictive pulmonary disease with small pleural cavities, those with pulmonary hypertension with cardiomegaly, those with dense pleural adhesions, and those who require concomitant cardiac surgery, cardiopulmonary bypass, or both.8, 9, 10, 12 Although Varela and colleagues11 described a transthoracic wall cannulation technique for this specific circumstance, the technique with 2 anterolateral thoracotomies without sternal division is not a common approach for BLT in most transplantation centers.
The new technique applied for transverse sternotomy closure in this study potentially requires only 2 additional wires of the same material as normally used, and therefore it is cost-effective and technically familiar to any transplantation surgeon. The efficacy of this new technique seen in both biomechanical and clinical investigations in this study is achieved by the additional wires preventing the longitudinal wires cutting through the sternum.
The retrospective and observational nature of the analysis represents the main limitation in the current study. There was a patient selection bias toward using this new technique more frequently in osteoporotic patients. Infection is another likely risk factor for sternal dehiscence. Five (33%) of 15 of the patients with sternal dehiscence had sternal wound infection. Although potential confounding variables were considered and adjusted by means of mutivariate analysis, the number of patients in each group limited the statistical power of the study. However, further investigation with a larger number of patients in a prospective randomized manner would be necessary to delineate further the role of infection in the causation of dehiscence. Other wiring techniques (eg, the figure-eight wiring technique)17 and suturing materials (eg, heavy polydioxanone monofilament sutures)26 would have a potential to be applied for transverse sternotomy closure, and further biomechanical and clinical investigation of those might be useful.
In conclusion, sternal dehiscence represents a major complication after BLT. Osteoporosis is a significant risk factor for sternal dehiscence after BLT. The new reinforced sternal wiring technique provides biomechanically superior fixation of the sternum and clinically reduces the incidence of sternal dehiscence in high-risk osteoporotic patients. Early screening of bone mass loss and the institution of prophylactic therapy might also be important. Although further investigation is needed, this new reinforced wiring technique shows promise in preventing sternal complications after BLT, especially in osteoporotic patients.
We thank Meredith Jewson for technical assistance, Anne Griffits and Sharon Daly for assembling and verifying the clinical data, and members of the Heart and Lung Transplant Unit, The Alfred Hospital, for their assistance.
References
- . An improved technique for bilateral lung transplantation: rationale and initial clinical experience. Ann Thorac Surg. 1990;49:785–791
- . The clamshell approach for the surgical treatment of complex cardiopulmonary pathology in infants and children. Eur J Cardiothorac Surg. 1997;11:298–306
- . Improved sternal fixation in the transsternal bilateral thoracotomy incision. J Thorac Cardiovasc Surg. 1996;112:137–141
- The clamshell incision: an improved approach to bilateral pulmonary and mediastinal tumor. Ann Thorac Surg. 1994;58:30–32
- . Extended pericardiotomy avoids cardiopulmonary bypass during bilateral sequential lung transplantation. J Thorac Cardiovasc Surg. 2005;129:466–467
- Sternal wound infection: our experience with 200 cases. J Cardiovasc Surg. 1994;35:103–104
- . Clamshell or sternotomy for double or heart-lung transplantation?. Eur J Cardiothorac Surg. 1999;15:333–339
- . Birateral sequential lung transplantation without sternal division eliminates posttransplantation sternal complications. J Thorac Cardiovasc Surg. 1999;117:358–364
- Bilateral sequential lung transplantation without sternal division. Eur J Cardiothorac Surg. 2003;23:894–897
- Sternal approximation for bilateral anterolateral transsternal thoracotomy for lung transplantation. Ann Thorac Surg. 2005;79:e19–e20
- Transcutaneous extracorporeal cannulation for bilateral lung transplantation without splitting the sternum. J Thorac Cardiovasc Surg. 2000;119:402–403
- . Technical considerations in adult lung transplantation. Semin Thorac Cardiovasc Surg. 2005;16:322–331
- . Wound complications after median sternotomy: relationship to internal mammary grafting. J Thorac Cardiovasc Surg. 1989;98:1096–1099
- . Mechanical analysis of midline sternotomy wound closure. J Thorac Cardiovasc Surg. 1999;117:1144–1150
- . The prevention and treatment of sternum separation following open-heart surgery. J Thorac Cardiovasc Surg. 1977;73:267–268
- . Modified Robicsek technique for complicated sternal closure. Ann Thorac Surg. 1991;52:1179–1180
- . A biomechanical comparison of three sternotomy closure techniques. Ann Thorac Surg. 2002;73:563–568
- . Validation of a bone analog model for studies of sternal closure. Ann Thorac Surg. 2002;74:739–745
- Biomechanical comparison of median sternotomy closures. Ann Thorac Surg. 2004;77:203–209
- Feasibility and utility of a lung donor score: correlation with early post-transplant outcomes. Ann Thorac Surg. 2007;83:257–263
- International guidelines for the selection of lung transplant candidates: 2006 update—a consensus report from the pulmonary scientific council of the International Society for Hart and Lung Transplantation. J Heart Lung Transplant. 2006;25:745–755
- . Definitions of primary graft dysfunction after lung transplantation: differences between bilateral and single lung transplantation. J Thorac Cardiovasc Surg. 2006;132:140–147
- Maximizing the utilization of donor organs offered for lung transplantation. Am J Respir Crit Care Med. 1999;160:265–271
- . Perspective: the diagnosis of osteoprosis. J Bone Miner Res. 1994;9:1137–1141
- . Osteoprosis and lung transplantation. Chest. 2000;117:476–481
- Polydioxanone sternal sutures for prevention of sternal dehiscence. J Card Surg. 2006;21:580–584
Dr Oto, Meredith Jewson, and Mr Venkatachalan (left to right)
PII: S0022-5223(07)00526-0
doi:10.1016/j.jtcvs.2007.03.003
© 2007 The American Association for Thoracic Surgery. Published by Elsevier Inc. All rights reserved.
Volume 134, Issue 1 , Pages 218-224, July 2007
