Given the human toll poststernotomy infection exacts, decisions that affect the care of patients with PSM must be made with due weight accorded to all valid and relevant clinical evidence. This synthesis of the integrated use of NPWT in preparing complex mediastinal wounds for closure via reconstructive surgery provides a sober picture of the quality of the evidence base. To date, the scientific community has been largely unreflective and uncritical of NPWT. In contrast, a surgical approach predicated on effective debridement combined with early transposition of well-vascularized tissue is a well-established treatment for deep space infections.9
To identify the relevant literature, a comprehensive search was performed in Ovid MEDLINE, PubMed, CINAHL, EMBASE, and the Cochrane Wounds Group database within the Cochrane Collaboration Library from January 2000 to June 2018. Investigators also searched the internet using meta-search engines for studies concerning the treatment of PSM. The keywords used for the literature search included “post-sternotomy mediastinitis,” “deep sternal wound infection,” “negative pressure wound therapy,” “vacuum assisted closure,” and “VAC.”
To supplement these literature searches, investigators hand-searched the bibliographies of key articles and used the “Related Citations in PubMed” function, which uses a word-weighted algorithm that compares words from the title, abstract, and medical subject headings of each citation to create a list of best matches. The authors examined all 434 abstracts of the articles identified in the initial search. Potentially relevant abstracts underwent full-text review. The authors ultimately selected 12 retrospective observational studies, 1 prospective observational study, and 1 case study.
This study was carried out using a protocol based on methods for rating evidence quality promulgated by the Agency for Healthcare Research and Quality (AHRQ).10 Under its Evidence-Based Practice Program, the AHRQ extensively analyzed 12 scales or checklists concerned with grading observational studies to arrive at a set of high-performing scales with which individual articles may be compared. According to the AHRQ, acceptable systems for assessing the quality of observational studies should consider the following five key domains: comparability of subjects, intervention, outcome measurement, statistical analysis, and funding or sponsorship. Of these five domains, the comparability of subjects is most significant to prevent selection bias, whereas funding or sponsorship is least significant.
Inclusion and Exclusion Criteria
The search was limited to English-language articles. Eligible articles were published between January 2000 and June 2018. No studies were excluded based on patient demographics. All 14 studies included participants diagnosed with PSM who underwent NPWT followed by omental and/or myocutaneous flap reconstruction. Any articles that examined NPWT as an intervention to prevent PSM or as the primary treatment for PSM were excluded.
Articles that diagnosed DSWI in accordance with the CDC criteria5 were included. Five of the 14 included articles used the CDC criteria5 and reported the PSM classification as described by El Oakley and Wright6 for all included subjects. Studies that did not define the criteria for a DSWI-based diagnosis were excluded to minimize the risk of definition bias.
The primary outcome measure was an objective assessment for resolution of DSWI. The measures included length of hospital stay and mortality. All included reference articles provided statistical analyses of their data.
Although all studies lacking a disclosure of manufacturer involvement were included, uniform requirements for declaration of conflicts of interest were frequently not reported. Table 3 shows which studies disclosed a commercial relationship.
Critical Appraisal of Methodological Quality
The second stage in identifying evidence most likely to influence the results involved assessing the quality of published studies. In essence, the study’s design, conduct, and analysis must have minimized potential biases and drawn honest conclusions based on the weight of the evidence.
All studies selected from the search were appraised for their methodological rigor using a common appraisal tool.11 The assessments were performed independently by two reviewers. Rating disagreements were solved by consulting senior faculty. The results of all reviewers’ evaluations were compiled into a bibliographic database.
The following sources of bias were assessed and ranked on a numeric scale:
- the method of allocation to study groups (random, 2; quasi-random, 1; and selected concurrent controls, 0)
- the data analysis and presentation of results (appropriate statistical analysis and clear presentation of results, 2; inappropriate statistical analysis or unclear presentation of results, 1; and inappropriate statistical analysis and unclear presentation of results, 0)
- the presence of baseline differences between groups that were potentially linked to study outcomes (of particular importance for observational studies: no baseline differences present or appropriate statistical adjustments made for differences, 2; baseline differences present and no statistical adjustments made, 1; and baseline characteristics not reported, 0)
- the objectivity of the outcome (objective outcomes or subjective outcomes with blinded assessment, 2; subjective outcomes with no blinding but clearly defined assessment criteria, 1; and subjective outcomes with no blinding and poorly defined, 0)
- the completeness of follow-up for the appropriate unit of analysis (90%, 2; 80%–90%, 1; and 80% or not described, 0).
The quality of evidence for each study was determined after considering each of the above elements and graded as follows:12
- High (8–10) = Additional research is not likely to impact our confidence in the study’s measured effect.
- Moderate (6–7) = Additional research is likely to impact our confidence in the study’s measured effect.
- Low (4–5) = Additional research is very likely to have a significant impact on our confidence of the study’s measured effect.
- Very low (<4) = The study’s measured effect is very uncertain.
Quality of the Evidence
Investigators considered limitations in study design and the risk of bias across all 14 studies that met the inclusion criteria.2,13–25 The quality of evidence was downgraded according to the explicit criteria outlined previously. The data show that 71% of the reports were nonrandomized. In general, appropriate statistical analyses were carried out, and results were clear (93%); 36% of the studies accurately accounted for baseline differences between severity, although 14% failed to report the diagnostic criteria. In all but one study, outcomes including length of stay and mortality were reported. Only one study reported follow-up results. Serious complications related to the use of NPWT, including right ventricular rupture, atrial fibrillation, respiratory arrest, recurrent infection when used for longer than 21 days, and retained sponge, were reported in this group of studies. Five of the 14 studies (36%) declared no conflict of interest, whereas the remainder failed to make a statement regarding manufacturer involvement.
In summary, there is a high risk of bias attributable to lack of randomization. Definition bias may have occurred in two-thirds of the studies, and therefore misdiagnosis and misclassification could have affected the accuracy of observations. Although a few of the authors commented on reinfection, only one study addresses follow-up. In the absence of information about late complications, doubt arises as to whether the overall results would have been altered if the missing information had been available.
The quality of the body of evidence is categorized into four levels, as outlined previously: high, moderate, low, and very low. In this analysis, 48% (n = 8) of the studies were of very low to low quality, and 36% (n = 5) of the studies were of moderate quality. One study was of high quality. The risk of bias and quality of evidence for all included studies are summarized in Table 4.
Detailed specifications for each study are provided in Table 3. This set of parameters captures key outcomes and measures that likely impact patient care. Among the cumulative 429 patients, following NPWT, 425 underwent pectoralis major or mental flap reconstruction, 3 were closed with pectoralis major and omentum, and 1 patient received a rectus abdominis muscle flap. Median length of stay was 29 ± 16 days. There were 41 deaths in this inpatient group, a mortality of 10%.
Treatment of PSM takes place in a care continuum that includes (1) early diagnosis and risk assessment, (2) active treatment including concerted efforts to eradicate infection and manage comorbidities known to increase the risk of wound sepsis, (3) stabilization of the sternum, and (4) restoration of soft tissue integrity. Surgical treatment is initiated by a thorough excision of all pathologic tissue including necrotic, infected, and poorly vascularized soft tissue. A large body of evidence supports closure by secondary intention using omental or myocutaneous flaps. This research synthesis investigates the interim between initial excision-debridement and flap reconstruction in which the options are either wound packing and passive drainage or NPWT.
Methodological limitations and biases exist in much of the evidence base for treating PSM. Imprecision, lack of clarity, bias, and insufficient attention to confounding variables are common. When interpreting observational studies, wound care practitioners should be mindful that they may be relying on poor methodology, and the quality of clinical decisions and confidence in their outcomes will never be better than the quality of information supporting those decisions.
From this research synthesis, a concern emerged for the regular failure to disclose manufacturer relationships. Given the attention to the influence of manufacturer funding in clinical trials of NPWT, meeting uniform requirements for declaring conflicts of interest is material.26
In response to 77 reported injuries and 6 deaths over a 2-year period, the FDA issued public safety warnings to healthcare providers and patients regarding the use of NPWT.27 The FDA guidelines identified both the risks of using NPWT and the recommended mitigation measures. The FDA warns providers and patients to look for potentially life-threatening complications such as bleeding and infection. The FDA reminds every clinician of the contraindications to NPWT use, including exposed vasculature, organs, or nerves in the wound. Given the complexity of properly maintaining, testing, and using NPWT in the setting of PSM, providers should ask whether early adoption presents a concomitant risk for poorer outcomes compared with the alternative of aggressive debridement and early secondary closure. This review of the literature confirms that there is a paucity of reliable studies to elucidate the benefits of NPWT as a “bridge to reconstruction” following PSM.
The unusual complications of NPWT and FDA warnings underscore the possible dangers associated with early adoption. They also provide a sober reminder that appropriate patient screening and proper use are imperative. The endorsement of a new procedure such as NPWT as a “bridge to reconstruction” for which effectiveness and safety are not empirically supported represents an inefficient use of societal resources because the cost of treating device-associated complications may outweigh the potential benefit to patients.
This literature review was limited to articles authored in English as the primary language. Therefore, more promising research regarding the use of NPWT as a “bridge to reconstruction” following PSM may be available in another language.
This literature review failed to find ample support for routine use of NPWT as a “bridge to reconstruction.” In the absence of evidence that would stand up to appropriate scientific scrutiny, this treatment practice cannot be recommended. Were the body of evidence of moderate quality and the benefit-harm equilibrium balanced, NPWT might be considered in individual cases. However, these results demonstrate that this is not the case. The investigators conclude that rigorous evaluative studies to assess the true effectiveness of NPWT as a “bridge to reconstruction” must precede its adoption.
1. Deniz H, Gokaslan G, Arslanoglu Y, et al. Treatment outcomes of postoperative mediastinitis
in cardiac surgery; negative pressure wound therapy versus conventional treatment. J Cardiothorac Surg 2012;7:67.
2. Ennker IC, Malkoc A, Pietrowski D, Vogt PM, Ennker J, Albert A. The concept of negative pressure wound therapy (NPWT) after poststernotomy mediastinitis
—a single center experience with 54 patients. J Cardiothorac Surg 2009;4:5.
3. Petzina R, Malmsjo M, Stamm C, Hetzer R. Major complications
during negative pressure wound therapy in poststernotomy mediastinitis
after cardiac surgery. J Thorac Cardiovasc Surg 2010;140(5):1133–6.
4. Sjogren J, Gustafsson R, Nilsson J, Malmsjo M, Ingemansson R. Clinical outcome after poststernotomy mediastinitis
: vacuum-assisted closure
versus conventional treatment. Ann Thorac Surg 2005;79(6):2049–55.
5. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition of health care-associated infection and criteria for specific types of infections in the acute care setting. Am J Infect Control 2008;36(5):309–32.
6. El Oakley RM, Wright JE. Postoperative mediastinitis
: classification and management. Ann Thorac Surg 1996;61(3):1030–6.
7. Rand RP, Cochran RP, Aziz S, et al. Prospective trial of catheter irrigation and muscle flaps for sternal wound infection. Ann Thorac Surg 1998;65(4):1046–9.
8. Levi N, Olsen PS. Primary closure of deep sternal wound infection following open heart surgery: a safe operation? J Cardiovasc Surg (Torino) 2000;41(2):241–5.
9. Harlin SL, Willard LA, Rush KJ, Ghisletta LC, Meyers WC. Chronic wounds of the lower extremity: a preliminary performance measurement set. Plast Reconstr Surg 2008;121(1):142–74.
10. West S, King V, Carey TS, et al. Systems to rate the strength of scientific evidence. Evid Rep Technol Assess 2002;(47):1–11.
11. Damiani G, Pinnarelli L, Sommella L, et al. Vacuum-assisted closure
therapy for patients with infected sternal wounds
: a meta-analysis of current evidence. J Plast Reconstr Aesthet Surg 2011;64(9):1119–23.
12. Atkins D, Best D, Briss PA, et al. Grading quality of evidence and strength of recommendations. BMJ 2004;328(7454):1490.
13. Agarwal JP, Ogilvie M, Wu LC, et al. Vacuum-assisted closure
for sternal wounds
: a first-line therapeutic management approach. Plast Reconstr Surg 2005;116(4):1035–40.
14. Bapat V, El-Muttardi N, Young C, Venn G, Roxburgh J. Experience with vacuum-assisted closure
of sternal wound infections following cardiac surgery and evaluation of chronic complications
associated with its use. J Card Surg 2008;23(3):227–33.
15. De Feo M, Della Corte A, Vicchio M, Pirozzi F, Nappi G, Cotrufo M. Is post-sternotomy mediastinitis
still devastating after the advent of negative-pressure wound therapy
? Tex Heart Inst J 2011;38(4):375–80.
16. De Feo M, Vicchio M, Nappi G, Cotrufo M. Role of vacuum in methicillin-resistant deep sternal wound infection. Asian Cardiovasc Thorac Ann 2010;18(4):360–3.
17. De Feo M, Vicchio M, Sante P, Cerasuolo F, Nappi G. Evolution in the treatment of mediastinitis
: single-center experience. Asian Cardiovasc Thorac Ann 2011;19(1):39–43.
18. Domkowski PW, Smith ML, Gonyon DL Jr, et al. Evaluation of vacuum-assisted closure
in the treatment of poststernotomy mediastinitis
. J Thorac Cardiovasc Surg 2003;126(2):386–90.
19. Karian L, Granick M. Sternal wound reconstruction with omental flap for poststernotomy mediastinitis
. Eplasty 2013;13:ic33.
20. Luckraz H, Murphy F, Bryant S, Charman SC, Ritchie AJ. Vacuum-assisted closure
as a treatment modality for infections after cardiac surgery. J Thorac Cardiovasc Surg 2003;125(2):301–5.
21. Morisaki A, Hosono M, Sasaki Y, et al. Evaluation of risk factors for hospital mortality and current treatment for poststernotomy mediastinitis
. Gen Thorac Cardiovasc Surg 2011;59(4):261–7.
22. Petzina R, Hoffmann J, Navasardyan A, et al. Negative pressure wound therapy for post-sternotomy mediastinitis
reduces mortality rate and sternal re-infection rate compared to conventional treatment. Eur J Cardiothorac Surg 2010;38(1):110–3.
23. Salica A, Weltert L, Scaffa R, et al. Negative pressure wound treatment improves Acute Physiology and Chronic Health Evaluation II score in mediastinitis
allowing a successful elective pectoralis muscle flap closure: six-year experience of a single protocol. J Thorac Cardiovasc Surg 2014;148(5):2397–403.
24. Segers P, de Jong AP, Kloek JJ, de Mol BA. Poststernotomy mediastinitis
: comparison of two treatment modalities. Interact Cardiovasc Thorac Surg 2005;4(6):555–60.
25. Vos RJ, Yilmaz A, Sonker U, Kelder JC, Kloppenburg GT. Vacuum-assisted closure
of post-sternotomy mediastinitis
as compared to open packing. Interact Cardiovasc Thorac Surg 2012;14(1):17–21.
26. Kairinos N, Pillay K, Solomons M, Hudson DA, Kahn D. The influence manufacturers have on negative-pressure wound therapy
research. Plast Reconstr Surg 2014;133(5):1178–83.
27. Food and Drug Administration. UPDATE on Serious Complications
Associated with Negative Pressure Wound Therapy Systems: FDA Safety Communication. Silver Spring, MD: US Department of Health and Human Services; 2014.
Keywords:Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.
complications; flap reconstruction; mediastinitis; negative-pressure wound therapy; sternal wounds; sternotomy; vacuum-assisted closure