Abstract
Background:
Sternal wound infections (SWI) can be a devastating long-term complication with significant morbidity and health care cost. The purpose of this analysis was to evaluate the cost-effectiveness of negative pressure incision management systems (NPIMS) in cardiac surgery.
Materials and Methods:
All cardiac surgery cases at an academic hospital with risk scores available (2009–2017) were extracted from an institutional database (n=4,455). Patients were stratified by utilization of NPIMS and high-risk was defined as above the median. Costs included infection related readmissions and were adjusted for inflation. Multivariable regression models assessed the risk-adjusted cost of SWI and efficacy of NPIMS use. Cost-effectiveness was modeled using TreeAge Pro using institutional results.
Results:
The rate of deep SWI was 0.9% with an estimated cost of $111,175 (p<0.0001). The rate of superficial SWI was 0.8% at a cost of $7,981 (p=0.08). Risk-adjusted NPIMS use was not significantly associated with reduced SWI (OR 1.2, p=0.62) and thus not cost-effective. However, in the high-risk cohort with an OR 0.84 (p=0.72) and SWI rate of 2.3%, NPIMS use cost $205 per patient with an incremental cost-effectiveness ratio of $179,092. Therefore, NPIMS is estimated to be cost-effective with a deep SWI rate over 1.3% or improved efficacy (OR<0.83).
Conclusion:
Sternal wound infections are extremely expensive complications with estimates of $111,175 for deep yet only $7,981 for superficial. While NPIMS was not cost-effective for SWI prevention as currently utilized, a protocol for use on patients with a higher risk of sternal infection could be cost-effective.
Keywords: Cardiac surgery, sternal wound infection, quality improvement, cost-effectiveness
INTRODUCTION
The development of sternal wound infection (SWI) occurs despite optimal treatment of patients and can be a devastating complication. The rate of SWI ranges from 0.5% to 8% in the literature, with deep SWI (DSWI) accounting for approximately half of these infections.(1–4) The impact of DSWI can be extensive with increased risk for other complications, multiple reoperations, and decreased long-term survival.(1, 2, 5) The result is an approximately 50% reduction in postoperative quality of life compared to peers without infection.(6, 7) The resource utilization associated with DSWI is also remarkable with increase length of stays and cost.(4, 5, 8, 9) Cost estimates vary widely with a recent estimate as low as $56,000 while others have estimated DSWI can more than quintuple the cost of surgery.(4, 5, 8–12)
While the rate of superficial SWI (SSWI) has been reported to be between 1% and 6.4%, many of these occur after 30 days.(2, 3, 13) Information regarding the impact of SSWI on outcomes and resource utilization is extremely limited. In one series, the mortality rate for patients with SSWI was 0.5% at 30-days and 2.1% at 1-year (compared to 2.7% and 4.8% for patients with DSWI).(1) While some reports include SSWI as well as DSWI in their calculations demonstrating increased morbidity and resource utilization, only one provides an estimate showing a $29,400 increase in hospital cost within the very small sample size of 3 SSWIs.(8)
Numerous studies have evaluated risk factors for development of SWI and specific interventions have been introduced aiming to reduce the risk of infection beyond standard antibiotic prophylaxis and meticulous sterile technique.(1, 3, 14, 15) The method most widely adopted has been decontamination of nasal and oropharyngeal endogenous microorganisms.(16) Another promising option is the prophylactic placement of a sterile negative pressure incision management system (NPIMS) such as Prevena (Acelity, San Antonio, TX) to prevent development of an infection. Observational and randomized studies have demonstrated reduced rates of SWI with use of NPIMS.(17, 18) In an effort to reduce SWI in high-risk patients, a NPIMS was introduced at our institution in 2014. This affords an ideal opportunity to evaluate the cost and efficacy of this intervention. The purpose of this analysis was twofold. First, we aimed to fill in the knowledge gap regarding SSWI and estimate risk-adjusted incremental costs associated with this infection. Next, the cost-effectiveness of the NPIMS was evaluated using estimated cost of the device and infections, infection risk reduction with NPIMS, and quality of life estimates associated with SWI.
MATERIALS AND METHODS
Patient Data
The University of Virginia Institutional Review Board approved this study with a waiver of patient consent due to its retrospective nature (IRB Protocol #19247). The records for cardiac surgery patients from January 2009 through September 2017 were extracted from an institutional Society of Thoracic Surgery (STS) Adult Cardiac Surgery Database. Inclusion criteria were cardiac surgery cases with predicted risk of deep sternal wound infection (PDSWI) available; categorized as isolated cases by the STS coronary bypass, valve or combination. Patients were stratified by utilization of NPIMS, which was introduced in 2014. Use of the device was at the discretion of the surgeon, but utilized for patients at elevated risk for SWI. The NPIMS was placed over the incision after suture closure while sterile and left in place for 5–7 days or until discharge. The risk score includes 14 variables including age, sex, body surface area, lung disease, kidney function, diabetes, and race among others and was derived from national STS data.(19) STS patient level data was merged with a quality improvement database containing all cardiac surgery infection data, which was verified by chart review and readmission information was obtained through 90 days after surgery.
Cost Data
Patient level STS data was matched to hospital finance records obtained from the Clinical Data Repository. Every hospital charge code has an associated cost valuation that is derived from direct and indirect costs. Therefore, charges are converted to calculated costs and individual transactions are summed for the entire admission. Hospital costs were adjusted to 2017 equivalent dollars using the market basket for the Center for Medicare and Medicaid Services (CMS) Inpatient Prospective Payment System that captures medical related inflation. Total hospital costs were calculated for the index hospitalization except for patients who were readmitted for a surgical site infection, in which case the readmission was included in the total cost.
Statistical Analysis
Continuous variables are presented as median [interquartile range as 25th, 75th percentiles] and categorical variables as counts (%). All cost and risk score data are presented as mean ± SD regardless of normality to more fully represent total values. Patient demographics, operative characteristics and short-term outcomes were assessed by nonparametric univariate analysis, Mann-Whitney U Test or Chi-Square Test as appropriate. Multivariable logistic regression assessed the association between NPIMS on rates of infection with risk adjustment using STS predicted deep sternal wound infection and year. Additionally, generalized linear regression with a gamma distribution and log link identified the risk-adjusted costs associated with DSWI and SSWI. A subgroup analysis was performed on high risk patients, defined as above the median PDSWI. All statistical analyses were performed using SAS Version 9.4 (SAS Institute, Cary, NC).
Cost-Effectiveness Model
A model was developed to analyze the financial impact of negative pressure incision management on sternal wound infections in a cohort of patients at high risk for sternal wound infection (Figure 1). All cost-effectiveness modeling was performed using TreeAge Pro Version 2017 (TreeAge Software, Inc., Williamstown, MA). Institutional rates of DSWI and SSWI were applied to both arms of the analysis. For patients who received NPIMS treatment these rates were multiplied by the risk reduction identified by logistic regression analysis. The institutional risk-adjusted cost for both DSWI and SSWI identified by generalized linear regression were also included. The cost of the incision management system was known to be $435 and applied to all patients in the treatment arm. The utility (quality of life) was set to 1 for patients without SWI. Although data is extremely limited, best estimates indicate a 50% reduction in utility with DSWI.(6, 7) No estimates were available for SSWI, but based on the relative small impacts on length of stay and mortality compared to DSWI, a 10% reduction in utility was assigned for SSWI. A second model was run using the risk reduction estimates from the single available randomized trial holding all other institutional variables constant.
Figure 1:
Decision tree with estimated costs, infection rates, and utilities.
Sensitivity analyses were performed by allowing the model to range between the 95% confidence intervals for institutionally derived data (DSWI and SSWI costs, NPIMS effect size) or the range published in the literature (infection rates). For the infection rate, the ratio of SSWI to DSWI was held constant for the analysis of overall SWI rate. Cost-effectiveness was determined using a willingness-to-pay threshold of three times the 2016 GDP per capita of the United States ($156,585).(20, 21)
RESULTS
A total of 4,455 patients underwent cardiac surgery from 2009–2017 with a predicted risk score available. The overall sternal infection rate was 1.9%, with 0.9% classified as deep and 0.8% superficial. The median predicted risk of deep sternal wound infection was 0.29%, the cutoff used to define high risk. A total of 911 (20.5%) patients received treatment with NPIMS, with 63.5% of patients in 2017 using a NPIMS. There were numerous baseline differences between patients with versus without NPIMS treatment identified on univariate analysis (Table 1). Unsurprisingly, the mean predicted risk of DSWI was significantly greater in patients receiving NPIMS treatment compared to those who did not (0.5% vs 0.4%; p<0.0001). Patients who received NPIMS treatment were more likely to have undergone CABG (73% vs 65%; p<0.0001).
Table 1:
Baseline and operative patient characteristics
| NPIMS | No NPIMS | ||
|---|---|---|---|
| Baseline Characteristics | (n = 647) | (n = 2226) | p value |
| Age (y) | 67 (58–74) | 69 (60–76) | 0.005 |
| Body mass index (kg/m2) | 30.6 (26.9–35.5) | 28.1 (24.8–32.4) | <0.0001 |
| Female | 212 (32.8%) | 665 (29.9%) | 0.160 |
| Prior cerebrovascular accident | 70 (10.9%) | 204 (9.2%) | 0.199 |
| Lung disease (moderate/severe) | 73 (11.4%) | 223 (10.0%) | 0.310 |
| Smoker | 405 (62.7%) | 458 (20.6%) | <0.0001 |
| Diabetes | 383 (59.2%) | 828 (37.2%) | <0.0001 |
| Insulin dependent diabetes | 152 (40.0%) | 252 (30.6%) | 0.001 |
| Dialysis dependent renal failure | 29 (4.5%) | 44 (2.0%) | 0.0004 |
| Hypertension | 571 (88.3%) | 1752 (78.7%) | <0.0001 |
| Coronary artery disease | 512 (80.0%) | 1630 (73.6%) | 0.001 |
| Prior myocardial infarction | 298 (46.2%) | 906 (40.7%) | 0.013 |
| Heart failure within 2 weeks | 329 (50.9%) | 1043 (46.9%) | 0.073 |
| Peripheral arterial disease | 344 (15.5%) | 120 (18.6%) | 0.060 |
| Ejection Fraction (%) | 58 (45–63) | 57 (47–63) | 0.053 |
| Endocarditis | 24 (3.7%) | 84 (3.8%) | 0.940 |
| Aortic insufficiency (moderate/severe) | 50 (7.7%) | 168 (11.5%) | 0.008 |
| Mitral regurgitation (moderate/severe) | 122 (18.9%) | 518 (29.6%) | <0.0001 |
| Prior valve surgery | 34 (5.3%) | 113 (5.1%) | 0.850 |
| Prior CABG | 34 (5.3%) | 122 (5.5%) | 0.830 |
| STS PROM (%) | 1.8% (0.9%-4.1%) | 1.8% (0.9%-3.8%) | 0.092 |
| STS PDSWI (%) | 0.4% (0.2%-0.6%) | 0.3% (0.2%-0.5%) | <0.0001 |
| Operative Characteristics | NPIMS | No NPIMS | |
| Reoperative Surgery | 58 (9.0%) | 218 (9.8%) | 0.529 |
| Elective | 354 (54.7%) | 1317 (59.2%) | 0.043 |
| CABG | 465 (71.9%) | 1409 (63.3%) | <0.0001 |
| Valve | 276 (42.7%) | 1162 (52.2%) | <0.0001 |
| CABG/Valve | 94 (14.5%) | 345 (15.5%) | 0.546 |
| Cross clamp time (min) | 74 (59–92) | 72 (58–90) | 0.084 |
| Cardiopulmonary bypass time (min) | 95 (76–121) | 98 (79–122) | 0.250 |
CABG = coronary artery bypass grafting; STS = Society of Thoracic Surgeons; PROM = predicted risk of mortality
Table 2 demonstrates the unadjusted outcomes comparing patients with versus without NPIMS treatment. The rates of DSWI were similar between groups (0.9% vs 0.9%; p=0.93) as were the rates of SSWI (1.0% vs 0.9%; p=0.56). The NPIMS treatment group did have higher rates of prolonged ventilation, readmission, health care cost and length of stay. After risk adjustment, NPIMS was not associated with a significant reduction in DSWI and/or SSWI (Table 3). The point estimate for high-risk patients was lower at 0.84 (0.33–2.15) but remained not statistically significant (p=0.717).
Table 2:
Operative outcomes
| NPIMS | No NPIMS | ||
|---|---|---|---|
| Characteristics | (n = 647) | (n = 2226) | p value |
| DSWI | 2 (0.3%) | 4 (0.2%) | 0.534 |
| SSWI | 4 (0.6%) | 19 (0.9%) | 0.554 |
| Either SWI | 6 (0.9%) | 23 (1.0%) | 0.804 |
| Operative mortality | 11 (1.7%) | 46 (2.1%) | 0.556 |
| Major morbidity† | 98 (15.2%) | 264 (12.0%) | 0.031 |
| Permanent stroke | 11 (1.7%) | 33 (1.5%) | 0.692 |
| Cardiac arrest | 12 (1.9%) | 35 (1.6%) | 0.618 |
| Prolonged ventilation | 69 (10.7%) | 162 (7.3%) | 0.005 |
| Renal failure requiring dialysis | 15 (2.3%) | 46 (2.1%) | 0.696 |
| Reoperation for any reason | 28 (4.3%) | 96 (4.3%) | 0.987 |
| Readmission | 83 (13.0%) | 202 (9.2%) | 0.005 |
| Discharge to facility | 212 (33.3%) | 589 (27.0%) | 0.002 |
| Total cost | $60,603 ± 46,367 | $49,890 ± 37,846 | <0.0001 |
| Length of stay (days) | 6(5–9) | 6 (5–7) | <0.0001 |
| Intensive care unit stay (hours) | 66 (28–115.8) | 47.5 (26–90) | <0.0001 |
Major morbidity includes: permanent stroke, cardiac arrest, renal failure, deep sternal wound infection, prolonged ventilation, reoperation for any reason
Table 3:
Risk-adjusted effect of NPIMS by infection classification for all comers and for the high-risk cohort.
| Infection type for all comers | OR | CI (95%) | p value | c-statistic |
|---|---|---|---|---|
| SSWI | 1.08 | 0.41–2.84 | 0.875 | 0.580 |
| DSWI | 1.45 | 0.50–4.21 | 0.496 | 0.649 |
| Either SWI | 1.23 | 0.60–2.52 | 0.581 | 0.615 |
| Infection type for high-risk | OR | CI (95%) | p value | c-statistic |
| SSWI | 1.19 | 0.32–4.37 | 0.798 | 0.629 |
| DSWI | 0.59 | 0.15–2.30 | 0.447 | 0.657 |
| Either SWI | 0.84 | 0.33–2.15 | 0.717 | 0.624 |
PDSWI=Predicted Deep Sternal Wound Infection
Patients who had NPIMS use had significantly higher mean total hospital costs ($59,554 vs $49,541, p<0.0001). After risk adjustment, the additional cost attributable to the DSWI was $111,175 (p<0.0001; Table 4). The cost estimate for SSWI was $7,981, although the association only trended towards statistical significance (p=0.08).
Table 4.
Risk-adjusted cost estimates for each infection classification
| SSWI Covariates | Estimate | 95% CI | p value |
|---|---|---|---|
| SSWI | $7,981 | (-923–18,455) | 0.081 |
| DSWI | $111,175 | (88,843–136,798) | <0.0001 |
| Either SWI | $63,0906 | (51,424–76,128) | <0.0001 |
Using the results listed above, because NPIMS use was not found to be effective at reducing infections (OR >1) it is not cost-effective as currently utilized. Examining the high-risk cohort (PDSWI > 0.29%), the 90-day SWI rate was 50/2222 (2.25)%, deep SWI 27/2222 (1.22%) and superficial SWI 23/2222 (1.04%). The incremental cost per patient is $205 and the incremental cost-effectiveness ratio is $179,092 per utility adjusted unit gained. One-way sensitivity analyses demonstrate the model is sensitive to the infection rate (>2.4%), DSWI cost (>$124,345), NPIMS cost (<$409), and NPIMS effectiveness (OR<0.83) (Figure 2). The model is sensitive to efficacy of NPIMS for preventing DSWI alone (cost effective with OR <0.81) and SSWI efficacy (OR<0.74) when the other OR is held constant at 0.84. When considering DSWI and SSWI rates separately, NPIMS is sensitive to DSWI rate (>1.3%) but not the SSWI rate. The model is not sensitive to the SSWI cost. The interaction between the infection rate and NPIMS cost (both of which can be manipulated based on policy) with regards to cost-effectiveness can be seen in Figure 3 with blue denoting when NPIMS is cost-effective and orange when NPIMS is not cost effective. As the cost of NPIMS increases (moves to the right on the x-axis) the infection rate needed to make its use cost-effective increases as well.
Figure 2:
One-way sensitivity analyses for infection rate (A), DSWI cost (B), NPIMS cost (C), and NPIMS effectiveness (D) with the threshold for cost-effectiveness demonstrated with the vertical dashed lines.
Figure 3:
Two-way sensitivity analysis by infection rate and NPIMS cost with the blue area representing scenarios where NPIMS is cost-effective, while the orange area is not cost-effective.
Using the randomized controlled trial data the efficacy can be estimated with an OR of 0.50 for DSWI and 0.13 for SSWI. In this model NIPMS has a cost savings of $316 per patient. This model is still not sensitive to either DSWI or SSWI cost, but remains sensitive to the infection rate (cost effective with overall infection rate >1.0% or DSWI rate >0.46%).
DISCUSSION
The incremental cost estimate for deep SWI including infection related readmission was $111,175. This dwarfed the estimate for superficial SWI at $7,981, which did not reach statistical significance with a p-value of 0.08. Further, this study demonstrates that use of NPIMS devices can be cost-effective, although was not as currently utilized. NPIMS did not appear to be efficacious in all patients, but instead may be useful only in those at higher risk of SWI. When examining these high risk patients, NPIMS use had a net cost of $205 per patient with a cost-effectiveness ratio of $179,092 indicating low potential value. Cost-effectiveness is dependent upon the infection rate, DSWI cost, cost of the device, and efficacy of the NPIMS system. Utilization would be cost-effective with in patients with an overall SWI rate >2.4%, DSWI rate >1.3% or improved efficacy (OR <0.83), both of which may be found by selecting on higher risk patients. At institutions where the cost of DSWI is higher (>$124,345) or the cost of the NPIMS device is lower could also justify use as cost-effective. Figure 3 is a useful way to assess the potential cost-effectiveness of NPIMS at any institution based on the cost of the device and the risk level of patients intended for use.
The elimination of reimbursement by the CMS for treatment of SWI has aligned hospital, surgeon and patient incentives to prevent SWI, even if the prevention measures increase index operation costs.(22) Considering the incentives and interested parties, we sought to quantify the impact of a recent quality improvement project at our institution that utilizes NPIMS devices in patients at high-risk for SWI. The decision to utilize the device was at the discretion of the surgeon. Some risk factors that were commonly considered include body mass index, diabetes, tobacco use, prior cardiac surgery, COPD, and CABG among others.(4, 15, 23, 24) Despite not having a standardized protocol, all of these risk factors were more prevalent in the cohort that received a NPIMS.
The reduction in infection risk association within our population were much more modest than previously published literature, with an OR of 1.23 in the entire cohort and 0.84 for the high-risk cohort, neither of which were statistically significant. Previous research includes a retrospective analysis that utilized historic controls and demonstrated a much larger risk reduction (OR 0.28, p<0.05).(18) However, this industry-funded study was not risk-adjusted and had a very large time bias. Another study by the same group, and also industry funded, was a randomized trial in high-risk obese patients.(17) This demonstrated an even larger reduction in infection risk (OR 0.22, p=0.027). This estimate appears overly optimistic as the NPIMS reduced the infection rate to 4% compared to 16% in the controls, numbers that are outside the standards reported in the literature. Another consideration is that NPIMS may increase in efficacy the higher the infection risk. If this is true then limiting use to higher risk patients would both increase the efficacy and event rate, both of which increase the cost-effectiveness in our model. If the results from the randomized trial represent the true estimated risk reduction, then NPIMS use would save money. However, incredibly few patients have a predicted risk score this high (upper quartile in our cohort had a PDSWI score of 0.48%).
The cost of DSWI has been well studied, but the cost estimates vary widely. The most recent estimate was risk-adjusted with $56,003 attributable to DSWI, with a range of $27,000 to $129,000.(4, 5, 9, 10, 12) Our estimate of $111,175 directly attributable to DSWI is higher than many estimates, but includes infection related readmissions out to 90 days. Furthermore, the estimate is based on any deep DSWI and not just the STS captured infections within 30 days. Unlike DSWI, the cost associated with SSWI is relatively unknown. There are certainly costs associated with development of SSWI including the those of antibiotics, wound management, additional time spent in the hospital and for outpatient care. However, the only estimate in the literature ($29,400) is an unadjusted derivation from three patients in Japan.(8) Our analysis provides a risk-adjusted estimate that places the cost far lower at $7,981. However, due to the size of the cost estimate this estimate was not statistically significant. It is important to estimate SWI costs that include post-discharge data as this made up approximately 90% of SSWI in a Norwegian report.(13)
This analysis is limited by its retrospective nature with potential for selection bias. While this is a time-based quality improvement project, this is mitigated by selective use of the NPIMS device and a risk-adjusted analysis. The surgeries were all performed at a single large academic practice and experiences at other hospitals may vary. Finally, the methodology to estimate cost associated with SWI are not based on care provided specifically for their treatment, but instead calculated using linear regression. However, the estimates are risk-adjusted and for DSWI compare favorably to previous calculations.
CONCLUSIONS
Sternal wound infections are expensive complications that can result in a high level of morbidity and mortality. Superficial sternal wound infections are estimated to increase cost by almost $8,000 while DSWI increased costs by more than $100,000. The use of NPIMS was not cost-effective as currently utilized. While industry sponsored trials appear to over estimate the efficacy of NPIMS in reducing rates of sternal wound infection, protocol driven utilization could be cost-effective. The protocol should target higher risk patients, for whom the device may be more effective and the event rate makes utilization cost-effective. Alternatively, a decrease in the price of the device would allow for use in more patients while remaining cost-effective. Quality improvement for rare complications is extremely difficult to accomplish in a cost-effective manner. Quantitative analysis such as this help to guide appropriate resource utilization in an era emphasizing value.
ACKNOWLEDGEMENTS:
We would like to thank Rochelle Jobes, RN our Infection Prevention nurse for all her diligent work on this and all other infection related quality improvement projects. This work was supported by the National Institutes of Health, grants T32 HL07849 and UM1 HL088925.
Funding: This work was supported by the National Institutes of Health (T32 HL07849, UM1 HL088925)
ABBREVIATIONS
- CMS
Center for Medicare and Medicaid Services
- DSWI
deep sternal wound infection
- NPIMS
negative pressure incision management system
- PDSWI
predicted risk of deep sternal wound infection
- STS
Society of Thoracic Surgery
- SSWI
superficial sternal wound infection
- SWI
sternal wound infection
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Presentation: American College of Surgeons Scientific Forum, October 21–25, 2018
Disclosures: Authors report no conflicts of interest
REFERENCES
- 1.Risnes I, Abdelnoor M, Almdahl SM, Svennevig JL Mediastinitis after coronary artery bypass grafting risk factors and long-term survival. Ann Thorac Surg 2010:89:1502–1509. [DOI] [PubMed] [Google Scholar]
- 2.Ridderstolpe L, Gill H, Granfeldt H, Ahlfeldt H, Rutberg H Superficial and deep sternal wound complications: incidence, risk factors and mortality. Eur J Cardiothorac Surg 2001:20:1168–1175. [DOI] [PubMed] [Google Scholar]
- 3.Meszaros K, Fuehrer U, Grogg S, Sodeck G, Czerny M, et al. Risk Factors for Sternal Wound Infection After Open Heart Operations Vary According to Type of Operation. Ann Thorac Surg 2016:101:1418–1425. [DOI] [PubMed] [Google Scholar]
- 4.Loop FD, Lytle BW, Cosgrove DM, Mahfood S, McHenry MC, et al. J. Maxwell Chamberlain memorial paper. Sternal wound complications after isolated coronary artery bypass grafting: early and late mortality, morbidity, and cost of care. Ann Thorac Surg 1990:49:179–186; discussion 186–177. [DOI] [PubMed] [Google Scholar]
- 5.Sears ED, Wu L, Waljee JF, Momoh AO, Zhong L, et al. The Impact of Deep Sternal Wound Infection on Mortality and Resource Utilization: A Population-based Study. World J Surg 2016:40:2673–2680. [DOI] [PubMed] [Google Scholar]
- 6.Jideus L, Liss A, Stahle E Patients with sternal wound infection after cardiac surgery do not improve their quality of life. Scand Cardiovasc J 2009:43:194–200. [DOI] [PubMed] [Google Scholar]
- 7.Falcoz PE, Chocron S, Stoica L, Kaili D, Puyraveau M, et al. Open heart surgery: one-year self-assessment of quality of life and functional outcome. Ann Thorac Surg 2003:76:1598–1604; discussion 1604. [DOI] [PubMed] [Google Scholar]
- 8.Kobayashi J, Kusachi S, Sawa Y, Motomura N, Imoto Y, et al. Socioeconomic effects of surgical site infection after cardiac surgery in Japan. Surg Today 2015:45:422–428. [DOI] [PubMed] [Google Scholar]
- 9.Graf K, Ott E, Vonberg RP, Kuehn C, Haverich A, et al. Economic aspects of deep sternal wound infections. Eur J Cardiothorac Surg 2010:37:893–896. [DOI] [PubMed] [Google Scholar]
- 10.LaPar DJ, Crosby IK, Rich JB, Fonner E Jr., Kron IL, et al. A contemporary cost analysis of postoperative morbidity after coronary artery bypass grafting with and without concomitant aortic valve replacement to improve patient quality and cost-effective care. Ann Thorac Surg 2013:96:1621–1627. [DOI] [PubMed] [Google Scholar]
- 11.Jenney AW, Harrington GA, Russo PL, Spelman DW Cost of surgical site infections following coronary artery bypass surgery. Anz J Surg 2001:71:662–664. [DOI] [PubMed] [Google Scholar]
- 12.Mokhtari A, Sjogren J, Nilsson J, Gustafsson R, Malmsjo M, et al. The cost of vacuum-assisted closure therapy in treatment of deep sternal wound infection. Scand Cardiovasc J 2008:42:85–89. [DOI] [PubMed] [Google Scholar]
- 13.Berg TC, Kjorstad KE, Akselsen PE, Seim BE, Lower HL, et al. National surveillance of surgical site infections after coronary artery bypass grafting in Norway: incidence and risk factors. Eur J Cardiothorac Surg 2011:40:1291–1297. [DOI] [PubMed] [Google Scholar]
- 14.Gummert JF, Barten MJ, Hans C, Kluge M, Doll N, et al. Mediastinitis and cardiac surgery--an updated risk factor analysis in 10,373 consecutive adult patients. Thorac Cardiovasc Surg 2002:50:87–91. [DOI] [PubMed] [Google Scholar]
- 15.Milano CA, Kesler K, Archibald N, Sexton DJ, Jones RH Mediastinitis after coronary artery bypass graft surgery. Risk factors and long-term survival. Circulation 1995:92:2245–2251. [DOI] [PubMed] [Google Scholar]
- 16.Segers P, Speekenbrink RG, Ubbink DT, van Ogtrop ML, de Mol BA Prevention of nosocomial infection in cardiac surgery by decontamination of the nasopharynx and oropharynx with chlorhexidine gluconate: a randomized controlled trial. JAMA 2006:296:2460–2466. [DOI] [PubMed] [Google Scholar]
- 17.Grauhan O, Navasardyan A, Hofmann M, Muller P, Stein J, et al. Prevention of poststernotomy wound infections in obese patients by negative pressure wound therapy. J Thorac Cardiovasc Surg 2013:145:1387–1392. [DOI] [PubMed] [Google Scholar]
- 18.Grauhan O, Navasardyan A, Tutkun B, Hennig F, Muller P, et al. Effect of surgical incision management on wound infections in a poststernotomy patient population. Int Wound J 2014:11 Suppl 1:6–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Shahian DM, O’Brien SM, Filardo G, Ferraris VA, Haan CK, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 1--coronary artery bypass grafting surgery. Ann Thorac Surg 2009:88:S2–22. [DOI] [PubMed] [Google Scholar]
- 20.Robinson LA, Hammitt JK, Chang AY, Resch S Understanding and improving the one and three times GDP per capita cost-effectiveness thresholds. Health Policy Plan 2017:32:141–145. [DOI] [PubMed] [Google Scholar]
- 21.Anderson JL, Heidenreich PA, Barnett PG, Creager MA, Fonarow GC, et al. ACC/AHA statement on cost/value methodology in clinical practice guidelines and performance measures: a report of the American College of Cardiology/American Heart Association Task Force on Performance Measures and Task Force on Practice Guidelines. J Am Coll Cardiol 2014:63:2304–2322. [DOI] [PubMed] [Google Scholar]
- 22.Calderwood MS, Kleinman K, Soumerai SB, Jin R, Gay C, et al. Impact of Medicare’s payment policy on mediastinitis following coronary artery bypass graft surgery in US hospitals. Infect Control Hosp Epidemiol 2014:35:144–151. [DOI] [PubMed] [Google Scholar]
- 23.Ottino G, De Paulis R, Pansini S, Rocca G, Tallone MV, et al. Major sternal wound infection after open-heart surgery: a multivariate analysis of risk factors in 2,579 consecutive operative procedures. Ann Thorac Surg 1987:44:173–179. [DOI] [PubMed] [Google Scholar]
- 24.Grossi EA, Esposito R, Harris LJ, Crooke GA, Galloway AC, et al. Sternal wound infections and use of internal mammary artery grafts. J Thorac Cardiovasc Surg 1991:102:342–346; discussion 346–347. [PubMed] [Google Scholar]