Abstract
Purpose
Surgical site infection (SSI) with methicillin-resistant Staphylococcus aureus (MRSA) is a serious post-operative complication, with head and neck cancer patients at greater risk due to the nature of their disease. Infection with MRSA has been shown to be costly and impart worse outcomes on patients who are affected. This study investigates incidence and risks for MRSA SSIs at a tertiary medical institution.
Materials and methods
This study reviewed 577 head and neck procedures from 2008 to 2013. Twenty-one variables (i.e. tumor characteristics, patient demographics, operative course, cultures) were analyzed with SPSS to identify trends. A multivariate analysis controlled for confounders (age, BMI, ASA class, length of stay) was completed.
Results
We identified 113 SSIs of 577 procedures, 24 (21.23%) of which were MRSA. Of all analyzed variables, hospital exposure within the preceding year was a significant risk factor for MRSA SSI development (OR 2.665, 95% CI: 1.06–6.69, z statistic 2.086, p = 0.0369). Immunosuppressed patients were more prone to MRSA infections (OR 14.1250, 95%CI: 3.8133–52.3217, p < 0.001), and patients with a history of chemotherapy (OR 3.0268, 95% CI: 1.1750–7.7968, p = 0.0218). Furthermore, MRSA SSI resulted in extended post-operative hospital stays (20.8 ± 4.72 days, p = 0.031).
Conclusions
Patients who have a history of chemotherapy, immunosuppression, or recent hospital exposure prior to their surgery are at higher risk of developing MRSA-specific SSI and may benefit from prophylactic antibiotic therapy with appropriate coverage. Additionally, patients who develop MRSA SSIs are likely to have an extended postoperative inpatient stay.
1
Introduction
Surgical site infections (SSIs) are a significant source of morbidity and mortality in surgical patients, and the second most common type of health care-associated infection . SSI is defined as an infection occurring at or adjacent to the incision within thirty days of the procedure, or within ninety days if a prosthetic was implanted . SSIs occur in 2–5% of surgical patients in the United States, resulting in a total of 300,000–500,000 cases of SSIs per year . Methicillin resistant Staphylococcus aureus (MRSA) is a major cause of surgical site infections (SSIs), and due to its resistance to many first-line antibiotics, MRSA SSIs have been shown to be associated with increased length of stay, mortality, readmission rates, and hospital charges .
MRSA has been implicated in head and neck post-operative infections worldwide . Specifically, MRSA SSIs have been shown to cause skin breakdown, cellulitis, and fistula formation in these patients, resulting in a complicated hospital course and subsequent increased length of stay, mortality, and hospital costs . Two primary reasons why MRSA is often the identified pathogen in these patients are the surgical site and the greater use of antibiotics in this subset of patients. The nature of head and neck surgical procedures may increase the risk for MRSA infections simply because the operating sites of these patients, the nares and cheeks, are the most common areas of colonization, and serve as the origin of MRSA SSIs . Because of this, the location of the operation may predispose these patients to post-operative infections. Additionally, head and neck surgical patients generally have had a greater exposure to antibiotics, whether for recurrent head and neck infections (i.e. tonsillitis, otitis media, and sinusitis) or peri-operative antibiotic use in clean-contaminated procedures, and this has been shown to correlate with increased MRSA infections .
Furthermore, head and neck cancer patients undergoing major surgery have other risk factors that further increase their chances of post-operative MRSA infections. First, these patients often have weakened host defenses due to their malignancy and/or chemoradiation . Second, head and neck oncological surgeries generally have longer operating times and prolonged post-operative hospital care compared to other ENT procedures, which may increase the likelihood of acquiring nosocomial MRSA . Finally, many head and neck cancer procedures require flap reconstruction, which has also been associated with increased rates of infection .
The purpose of this study was to identify patients with MRSA SSIs following any head and neck procedure. To our knowledge, no study has examined the incidence of post-operative MRSA surgical site infections in head and neck surgeries at a major institution in the United States. The few studies that have looked at the incidence of post-operative MRSA infection in ENT focused on specific procedures and were done in countries outside of the United States . Because the complications of post-operative MRSA infections can be severe, it is important to elucidate the risk factors and complications as well as create a foundation for future studies investigating prevention and treatment of post-operative MRSA infections in otolaryngology.
2
Materials and methods
A retrospective chart review study was performed with approval from the University Hospitals Cleveland Medical Center (UH-CMC) Institutional Review Board. Specific inclusion criteria included any patient who was > 18 years of age and underwent major head and/or neck surgery at UH-CMC by the Head and Neck Surgery Department attending physicians between January 1, 2008 and January 1, 2014, and were found to have a documented surgical site infection during their post-operative hospitalization. Specific exclusion criteria included age < 18 years, undergoing surgery at any non-UHCMC institution, or procedure completed before or after the specified date range. Eligible patients were identified by cross-referencing lists generated from the hospital’s infection control and coding departments and were compiled into a comprehensive, de-identified patient list. Surgical site infections were defined as infections occurring within 30 days of the procedure in accordance with CDC parameters . For patients with multiple SSIs following separate surgeries, only the first incidence of SSI was included. SSIs were determined by the history and physical exam findings (i.e.: erythema, edema, purulence). After identifying all patients who developed SSIs, two groups were established comparing the qualities of patients with MRSA infections to those with non-MRSA infections. The population underwent a wide variety of operations involving the upper aerodigestive tract, skin, salivary glands, and endocrine glands.
Independent variables analyzed included gender, age, BMI, smoking status, alcohol and drug use, preoperative chemoradiation therapy, prior head and neck surgery, procedure duration, estimated operative blood loss, albumin, and prior hospital admissions. Additionally comorbidities like hypertension, diabetes mellitus with hemoglobin A1C, vascular disease, immunosuppression and nutritional status given by prealbumin and albumin levels were analyzed. Immunosuppression was defined by either the known diagnosis of HIV, chronic use of steroids, or prescription use of TNF-α inhibitors. The presence of vascular disease was given if there was documented diagnosis of peripheral vascular disease or coronary artery disease.
Dependent variables measured included length of stay, number of organisms, organism type, presentation of infection, and complications. Complications were defined as readmission or return to operating room within one year of initial procedure, flap failures, fistula formation, and bacteremia.
Data gathered from records was entered into a Microsoft Excel spreadsheet (Microsoft, Redmond, WA) and transferred to SPSS Statistics 22 (IBM, Armonk, NY). Associations of independent and dependent variables were analyzed using Pearson’s χ 2 test. Continuous variables were assessed using the Student’s t -test or the Mann-Whitney test (parametric and non-parametric respectively). A p value of ≤ 0.05 was considered statistically significant. A multivariate analysis was done to control for confounding factors.
2
Materials and methods
A retrospective chart review study was performed with approval from the University Hospitals Cleveland Medical Center (UH-CMC) Institutional Review Board. Specific inclusion criteria included any patient who was > 18 years of age and underwent major head and/or neck surgery at UH-CMC by the Head and Neck Surgery Department attending physicians between January 1, 2008 and January 1, 2014, and were found to have a documented surgical site infection during their post-operative hospitalization. Specific exclusion criteria included age < 18 years, undergoing surgery at any non-UHCMC institution, or procedure completed before or after the specified date range. Eligible patients were identified by cross-referencing lists generated from the hospital’s infection control and coding departments and were compiled into a comprehensive, de-identified patient list. Surgical site infections were defined as infections occurring within 30 days of the procedure in accordance with CDC parameters . For patients with multiple SSIs following separate surgeries, only the first incidence of SSI was included. SSIs were determined by the history and physical exam findings (i.e.: erythema, edema, purulence). After identifying all patients who developed SSIs, two groups were established comparing the qualities of patients with MRSA infections to those with non-MRSA infections. The population underwent a wide variety of operations involving the upper aerodigestive tract, skin, salivary glands, and endocrine glands.
Independent variables analyzed included gender, age, BMI, smoking status, alcohol and drug use, preoperative chemoradiation therapy, prior head and neck surgery, procedure duration, estimated operative blood loss, albumin, and prior hospital admissions. Additionally comorbidities like hypertension, diabetes mellitus with hemoglobin A1C, vascular disease, immunosuppression and nutritional status given by prealbumin and albumin levels were analyzed. Immunosuppression was defined by either the known diagnosis of HIV, chronic use of steroids, or prescription use of TNF-α inhibitors. The presence of vascular disease was given if there was documented diagnosis of peripheral vascular disease or coronary artery disease.
Dependent variables measured included length of stay, number of organisms, organism type, presentation of infection, and complications. Complications were defined as readmission or return to operating room within one year of initial procedure, flap failures, fistula formation, and bacteremia.
Data gathered from records was entered into a Microsoft Excel spreadsheet (Microsoft, Redmond, WA) and transferred to SPSS Statistics 22 (IBM, Armonk, NY). Associations of independent and dependent variables were analyzed using Pearson’s χ 2 test. Continuous variables were assessed using the Student’s t -test or the Mann-Whitney test (parametric and non-parametric respectively). A p value of ≤ 0.05 was considered statistically significant. A multivariate analysis was done to control for confounding factors.
3
Results
3.1
Demographics
We determined no significant differences in age, gender distribution, smoking status, alcohol and drug use, past medical history of hypertension, obesity (as given by BMI), diabetes (by hemoglobin A1C level), immunosuppression or vascular disease between patients who acquired MRSA SSIs and patients with non-MRSA SSIs (all p > 0.1) ( Table 1 ). Smoking status and alcohol use were determined by history of daily use. Vascular disease was defined by a diagnosis of peripheral vascular disease and/or coronary artery disease. Immunosuppression defined by use of chronic steroids or immunosuppressant medications. Patients with immunosuppression were more prone to MRSA infections over no infections (OR 14.1250, 95% CI: 3.8133–52.3217, p < 0.001). Additionally patients with lower preoperative albumin were more likely to develop a MRSA SSIs compared to non-MRSA SSIs (p = 0.039).
| Variable | MRSA | Non-MRSA | p Value |
|---|---|---|---|
| Mean age (years) | 62.2 | 59.2 | 0.140 |
| Gender (% female) | 50% | 45.6% | 0.985 |
| Smoking status (%) | 83.3 | 71.1 | 0.692 |
| Alcohol use (%) | 41.6 | 27.8 | 0.633 |
| Illicit drug use (%) | 25% | 12.2% | 0.487 |
| Hypertension | 54.2 | 57.3 | 0.995 |
| BMI | 24.0 | 25.6 | 0.140 |
| Diabetes mellitus | 33.3 | 18.9 | 0.511 |
| HbA1C a | 6.6 | 7.5 | 0.124 |
| Vascular disease | 45.8 | 26.7 | 0.352 |
| Prealbumin (mg/dL) | 20.5 | 22.7 | 0.198 |
| Albumin (g/dL) | 3.4 | 3.7 | 0.039 |
| Prior inpatient stay (days) | 4 | 3.53 | 0.037 |
| Immunosuppression | 33.3% | 14.2% | 0.202 |
| Prior head/neck radiation | 45.8 | 47.7 | 0.925 |
| Prior chemotherapy | 37.5 | 34.4 | 0.781 |
| Prior head/neck surgery | 33.3% | 28.9% | 0.673 |
| Postoperative hospital stay (days) | 20.8 | 11.3 | 0.031 |
| Estimated blood loss (cc) | 216.14 | 238.7 | 0.329 |
| Procedure duration (min) | 575.5 | 463.5 | 0.049 |
| Interim before SSI identified (days) | 20.0 | 18.8 | 0.436 |
| Interim before return to OR b | 35.8 | 19.8 | 0.165 |
| Number of operative revisions b | 0.7 | 0.5 | 0.214 |
| Interim until readmission (days) c | 37.3 | 21.1 | 0.148 |
| Length of readmission c | 6.8 | 5.9 | 0.226 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
