| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
Interact CardioVasc Thorac Surg 2009;9:763-766. doi:10.1510/icvts.2009.214023 © 2009 European Association of Cardio-Thoracic Surgery
Primary sternal plating to prevent sternal wound complications after cardiac surgery: early experience and patterns of failure
a Department of Surgery, University of Alabama at Birmingham, 1922 7th Avenue South, KB 217, Birmingham, AL 35294-0016, USA Received 8 June 2009; received in revised form 3 August 2009; accepted 6 August 2009
*Corresponding author. Tel.: +1-205-934-9600; fax: +1-205-975-5867.
Sternal closure with rigid titanium plates (primary sternal plating) may reduce sternal wound complications in high-risk patients. We began performing primary sternal plating for the following indications: obesity, manual laborer, osteoporotic sternum, or intraoperative transverse sternal fracture. Patients receiving plate closure were compared to a risk-matched control group receiving wire closure. Outcomes of interest were postoperative length of stay and sternal wound complications [sterile dehiscence or deep sternal wound infection (DSWI)]. Wound complications were classified by time of occurrence as early (  30 days postoperation) or late (>30 days postoperation). Of 445 total cardiac cases during the 5-year study period, 129 (29%) met inclusion criteria. The plate group (n=30) and wire group (n=99) were generally well-matched in terms of risk factors. Postoperative length of stay was significantly shorter in the plate group (median 7 vs. 8 days, P=0.023). No early sternal wound complications occurred in the plate group, compared to 12 (12%) in the wire group (P=0.067). The incidence of late sternal wound complications was 10% in both groups (P=1.0). Primary sternal plating appears to provide benefits over wire closure during the early postoperative period, but may not prevent late wound complications in patients with osteoporosis or extreme obesity.
Key Words: Sternal dehiscence; Mediastinitis; Sternal plating
Sternal dehiscence and deep sternal wound infection (DSWI) are serious post-sternotomy wound complications with significant morbidity and mortality [1–3]. Risk factors for sternal wound complications include obesity, diabetes, and chronic pulmonary disease [4, 5]. Sternal dehiscence and DSWI often occur together, suggesting that instability in the early postoperative period predisposes to infection [2]. Several innovative closure techniques have been proposed for maximizing sternal stability in high-risk patients, but a reliable technique remains elusive [6, 7]. Some centers have begun using rigid titanium plates to close the sterna of high-risk patients, a technique known as primary sternal plating. Plate closure is more stable than wire circlage in vitro [8, 9], and previous studies demonstrate that DSWI rates are lower among plated patients [10]. This report reviews our early experience with primary sternal plating, comparing outcomes in plated patients to a wire-closed group with similar risk factors.
The Veteran's Affairs Medical Center in Birmingham, Alabama (BVAMC) provides surgical care for US military veterans. Selective primary sternal plating was initiated at BVAMC in 2005, and has been consistently performed since 2007 in patients deemed to be at high risk for sternal wound complications. Indications for primary sternal plating were as follows (one or more): body mass index (BMI) of 30 or greater, employment as a manual laborer, osteoporotic sternum, or intraoperative transverse sternal fracture. Primary sternal plating is performed after the cardiac procedure is completed and sternal hemostasis achieved. The sternal halves are measured for thickness and then approximated using five or six standard stainless steel wires. Soft tissue is cleared from the sternal surface to allow sufficient contact between plate and periosteum. Rigid titanium plates (SternaLock, Walter Lorenz Surgical, Jacksonville, FL, USA) are then positioned between the wires, ensuring that the wires and plates are not in contact in order to minimize the risk of galvanic corrosion. The configuration and number of plates is individualized to the patient. If necessary, plates are bent to conform to the sternal surface, using tools provided with the plating kit. Each plate is secured in place using self-tapping screws of appropriate length to engage the superficial and deep cortices but not to exceed the thickness of the sternum. No drilling is required. The overlying soft tissue and skin are closed in standard fashion. The plates can be cut rapidly with conventional wire cutters if emergent reentry is required. After discharge, BVAMC patients are routinely evaluated in the cardiovascular surgery clinic at two and six weeks. Long-term routine follow-up is performed by cardiologists and primary care physicians. All cardiac cases performed through a median sternotomy at BVAMC between 1 January 2004 and 31 December 2008 were considered for inclusion in this study. Approval for the study was obtained from the BVAMC Institutional Review Board. Patient demographics, indications for plating, and operative details were collected by chart review. Patients receiving primary sternal plating were identified using implanted device records. The control group was defined as all patients who met at least one indication for sternal plating, but who received standard wire closure. Once the plate and wire groups were identified, additional chart review was performed to determine the following outcomes: postoperative length of stay, in-hospital mortality, and sternal wound complications of DSWI (by US Centers for Disease Control criteria) or sterile dehiscence or non-union. Sterile dehiscence was defined as sternal instability by physical exam (e.g. sternal click, instability with cough) or radiographic imaging, in the absence of infection. Sternal wound complications were classified as either early or late, depending on whether they occurred within or after 30 days of operation, respectively. Preoperative characteristics, operative details, and outcomes were compared between the plate and wire groups using Fisher's exact and Wilcoxon rank-sum tests for categorical and continuous variables, respectively. Plate patients who experienced a sternal wound complication were reviewed in greater detail to identify the complication details. Statistical analyses were performed with SAS 9.1.3 (SAS Institute, Cary, NC, USA).
During the study period, 445 patients underwent cardiac surgery through a median sternotomy at BVAMC. One hundred and twenty-nine patients (29%) met at least one high-risk criterion for sternal plating; 30 received primary sternal plating and 99 received standard wire closure. The overall median follow-up time was 33 months (range, 2–61 months). Patient characteristics, indications for plating, operative details, and clinical outcomes for the plate and wire groups are summarized in Table 1. All in-hospital deaths were unrelated to sternal wound complications.
The plate and wire groups were generally well-matched in terms of preoperative and intraoperative characteristics, although plate patients were more likely to be current smokers and have transverse sternal fractures. Total operative time was statistically equivalent for the plate and wire groups (median 258 vs. 243 min, P=0.366). Postoperative length of stay was significantly shorter in the plate group (median 7 vs. 8 days, P=0.023). Overall, 25 sternal wound complications occurred (19%). Early wound complications nearly all involved sternal wound infection (92% DSWI), whereas late complications were predominantly sterile (92% sterile dehiscence or non-union). There was a strong trend toward fewer early sternal wound complications in the plate group (0% vs. 12%, P=0.067). This trend achieved statistical significance when a one-sided hypothesis test was used (H1: early complication rate is lower in the plate group, P=0.035, Fisher's exact test). Late sternal wound complication rates were equivalent in the plate and wire groups (10% vs. 10%, P=1.0). Late sternal wound complications in the plate group followed two distinct patterns. The first complication occurred in an elderly non-obese male with chronic obstructive pulmonary disease. This patient's sternum had been stabilized with two plates (upper L-shaped, lower T-shaped) due to an intraoperative finding of sternal osteoporosis. His initial postoperative course was uneventful, but he presented 12 weeks after surgery with an unstable sternum and draining wound. Computed tomography of the chest confirmed sternal dehiscence in the setting of low-grade mediastinitis. Although the plates themselves were undamaged, they no longer provided any stabilization because the screws had pulled out of his osteoporotic sternum (Fig. 1). This patient was treated initially with hardware removal, debridement, and intravenous antibiotics, followed by radical sternectomy and bilateral pectoralis flap reconstruction. Findings at debridement were consistent with sternal osteomyelitis, although deep tissue cultures were negative.
The other two complications consisted of sterile dehiscences, diagnosed at 35 and 109 weeks postoperatively. Both patients had been plated due to extreme obesity (BMI >40). The first patient had received a single upper X-shaped plate; his imaging revealed broken wires and separation of the inferior sternal body. Although his sternal instability was clinically minimal, he complained of severe pain and was treated with operative removal of broken wires and placement of a 12-hole ladder-shaped plate on the lower body of the sternum. His recovery from the second plating was uneventful and his pain resolved. The second patient had received two plates (upper X-shaped, lower T-shaped). His chest radiograph demonstrated fracture of the lower plate and separation of the lower sternum (Fig. 2). He was offered re-plating of the lower sternum, but declined further intervention and chose conservative management with pain medication and activity limitation instead. These patients’ BMIs were significantly higher than the rest of the plated cohort (P=0.030, Wilcoxon rank-sum test).
Our data demonstrate that, compared to standard wire closure, primary sternal plating is associated with a shorter postoperative length of stay and no significant increase in operating time. Although the technique may reduce early wound complications, it appears to have no effect on late wound complications. The one-day reduction in length of stay may be attributable to the improved postoperative comfort, pulmonary toilet, and mobility provided by a more stable sternal closure. To ensure that this finding did not represent confounding due to changes in postoperative management over time, we investigated temporal trends in length of stay. The postoperative length of stay was not significantly associated with the date of surgery (r=0.05, P=0.61) or the calendar year (P=0.31), leading us to conclude that the length of stay was not confounded by time trends over the study period. We also ensured that the reduction in length of stay was not driven by a few high outliers in the wire group or by group differences in indications for sternal plating. Results were the same when patients with isolated transverse sternal fractures or postoperative stays longer than 30 days were excluded from analysis. Despite a trend toward a greater proportion of combination procedures in the plated group (17% vs. 8%, P=0.18), the use of sternal plates for closure did not significantly increase operative time. The analysis had 77% power to detect a difference of 30 min in overall operating time. Our findings regarding early sternal wound complications are very similar to those of Song et al. who demonstrated that the rate of DSWI in high-risk patients was nil among those closed with plates vs. 15% among those closed with wires [11]. Their study's ability to evaluate late wound complications was limited, since the median follow-up was only 15 weeks. However, similar to our experience, they observed plate fracture and sternal dehiscence in a patient with extreme obesity. The two sterile dehiscences that occurred in extremely obese patients were most pronounced at the distal aspect of the sternum. This observation lends clinical confirmation to in vitro studies concluding that the distal sternum is subjected to the greatest mechanical stress [12]. In a recent study using polyurethane sternal models, Pai et al. found that four plates (three X-shaped and one box-shaped) resulted in maximal stability of the distal sternum [13]. It is possible that the sterna of extremely obese patients would be better stabilized by adding additional plates. However, since the distal sternum is primarily cartilaginous, additional distal plates may not significantly improve stability. Also, the costs of the plates and screws and the potential for infected hardware must be considered. In our practice environment, we have found that the use of two plates plus five wires strikes an appropriate balance between sternal stability, amount of prosthetic material, and cost. Our practice has evolved based on our early experience. After observing that screws may fail to hold in osteoporotic bone, we no longer use plates for osteoporosis. Instead, we employ an alternate stabilization technique such as peristernal figure-of-eight or Robicsek wiring in these patients [7, 14]. Although previous studies listed osteoporosis as an indication for plating with no apparent adverse outcomes, the actual prevalence of osteoporosis in their plate cohort was unclear [10, 11]. We currently use a single plate (box-, L- or T-shaped) only for isolated transverse sternal fractures. For all other indications, we place two plates, usually an X-shaped plate on the manubrium and an L- or T-shaped plate on the lower body of the sternum, along with five standard stainless steel wires. Our study has several limitations. Although specific indications for sternal plating were used, no formal plating protocol was in place during the study period. The control group was primarily historical, consisting of high-risk patients closed with wires before the plating technique was introduced. For these reasons, selection bias and unmeasured confounding variables could have affected the results. Due to small sample size and a relatively low number of wound complication events, power was low and regression modeling could not be used to adjust for group differences. However, group differences such as the greater proportion of smokers in the plated group would be expected to produce bias toward the null, and results remained the same when isolated transverse sternal fractures were excluded from analysis. In conclusion, our early experience confirms previous studies demonstrating that primary sternal plating is beneficial in the early postoperative period. However, our data also suggest that plates may not provide adequate stabilization to prevent late sternal wound complications, especially among patients with sternal osteoporosis or extreme obesity. Our experience highlights the need for additional prospective clinical studies to refine the indications for primary sternal plating and further standardize the technique.
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
Top
Abstract
1. Background
